Wikidata:Project chat/Archive/2017/05

This page is an archive. Please do not modify it. Use the current page, even to continue an old discussion.

Server switch - wikis in read-only mode today

Reminder: the wikis will be in read-only mode today for 30 minutes, starting at 14:00 UTC. Trizek (WMF) (talk) 10:10, 3 May 2017 (UTC)

This section was archived on a request by: Matěj Suchánek (talk) 16:38, 6 May 2017 (UTC)

Changing the data type of properties

Is it possible to change the data type of properties after creation? I think ISIL (P791) should be classified as an "External Identifier", not as a "String".

When I became property creator, I learned that it is not possible. However, things may have changed in the last two years. Jonathan Groß (talk) 09:02, 21 April 2017 (UTC)

When the external identifiers were introduced, many string properties were converted to external ids, so this it is technically possible. However, on WD:Identifier migration/0, P791 is listed under Properties with serious objections to conversion. Matěj Suchánek (talk) 11:52, 21 April 2017 (UTC)

Indeed. However, no valid objection is raised there. As I said over a year ago "The first three letters of 'ISIL ID' stand for 'International Standard Identifier'". It's high time this was converted. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:21, 26 April 2017 (UTC)

@ArthurPSmith, Jura1: Comments? I've looked up the discussion from last year but I don't understand what the problem is. Jonathan Groß (talk) 09:30, 29 April 2017 (UTC)

You need to delete the formatter URL otherwise you end up with lots of dead links after conversion.
--- Jura 17:09, 29 April 2017 (UTC)

Yeah, the problem seems to be different ISIL id's actually are owned by different libraries, and there is no consistency in the way they are resolved - there is no universal resolver for them. So there is (at least when I last checked) no formatter URL to use. I was thinking of writing something that would try to send the different iSIL's to their different library resolvers but when I reviewed them they seemed to be too different to make this useful. ArthurPSmith (talk) 14:40, 1 May 2017 (UTC)

Kingdom property

A ruler is ruler when he rules some land. Do we have a property to mention the kingdom of a ruler? Please guide me. Capankajsmilyo (talk) 06:33, 24 April 2017 (UTC)

Take a look at George V (Q269412). It uses position held (P39) with qualifiers. ChristianKl (talk) 07:37, 24 April 2017 (UTC)

Thanks Capankajsmilyo (talk) 09:27, 24 April 2017 (UTC)

There is a duplicate claim. It’s both a « emperor of india » and an « emperor » of « india ». author  TomT0m / talk page 11:54, 1 May 2017 (UTC)

Anniversary

Is there a property to indicate the "anniversary" of a subject? I am looking for it for military units, for example most Canadian regiments have a day in the year that they celebrate as their anniversaries. But it could be used by other types of subjects as well. The Infobox military unit on English Wikipedia has a parameter for "anniversaries". Thanks, Amqui (talk) 13:31, 24 April 2017 (UTC)

Can I suggest annual commemoration Q18574943 (or awareness day Q422695 if appropriate). MassiveEartha (talk) 14:03, 24 April 2017 (UTC)

Those are not properties. Amqui (talk) 15:33, 24 April 2017 (UTC)

my mistake, I misread the question. MassiveEartha (talk) 15:43, 24 April 2017 (UTC)

(edit conflict) Like significant event (P793): annual commemoration (Q18574943); qualifier day in year for periodic occurrence (P837)? Matěj Suchánek (talk) 15:45, 24 April 2017 (UTC)

I thought significant event (P793) was only for past events, not recurring events, but reading the description and the constraints, I guess nothing refrain that. All examples given are for past "one-time" events, but, if that's the case, I will add one example that shows a recurrent annual event as an example in significant event (P793). Also, if that's the case, I added public holiday (P832) as a subproperty of significant event (P793). Thanks, Amqui (talk) 16:38, 24 April 2017 (UTC)

I meant just as a suggestion, otherwise a new property for "recurring event" should be created. Matěj Suchánek (talk) 17:45, 24 April 2017 (UTC)

Indeed « key event » is made for specific events, not for a class of events that occurs every year. author  TomT0m / talk page 11:57, 1 May 2017 (UTC)

unlink pages with lua errors

Half of the page Wikidata:List of properties/all is covered with lua errors, I suggest to use Wikidata:Properties instead.

Links can be found at prominient locations e.g. Help:Statements (section #Properties) d1g (talk) 06:57, 28 April 2017 (UTC)

Substitute the subpages then transclude to one. Have a process to build the properties on a daily basis. The page is useful (when it works) for doing a keyword find rather than trying to manipulate search.  — billinghurst sDrewth 09:02, 28 April 2017 (UTC)

There are two issues with Wikidata:List of properties/all:

1. Lua errors: substituting will make the page only accessible in English
2. The list has to be manually maintained. I count currently only 967 properties in the list, so around 70% of all properties are missing. --Pasleim (talk) 09:49, 28 April 2017 (UTC)

I  Support removing it - link to the search page with Property selected instead. ArthurPSmith (talk) 12:39, 28 April 2017 (UTC)

I trimmed text at Help:Statements, please check if Wikidata:Properties misses something. d1g (talk) 16:18, 1 May 2017 (UTC)

Property for the head of the Catholic Church

I'm trying to add the current and previous popes to Catholic Church (Q9592). I've added Property:P2388 with the value "pope" to describe the role position. Now I want to add the name of the popes [1]: is chairperson appropriate for that? or is a different property like "head of the organisation" needed to be created? --Grabado (talk) 18:54, 30 April 2017 (UTC)

Wondering whether head of state (P35) may be better, especially with the protocol components. Hard to separate the Vatican from the Church in this situation.  — billinghurst sDrewth 03:29, 1 May 2017 (UTC)

Thanks, @Billinghurst: I've already used head of state (P35) in Q237, but I don't think it would be suitable for Q9592. The Vatican City exists as a state only from 1929, but the papacy goes back 2000 years. The head of the church is a different role, not only for the Catholic Church. Q193312, Q483889, Q42504... Most organised religions have leaders. --Grabado (talk) 07:25, 1 May 2017 (UTC)

@Grabado: Do you really think this the best idea to put the name of the pope in the item of the Catholic church ? Because we can put the long list of the 266 popes and this will just create a heavy item. The best is to use the existing link between pope and Catolic church and look for the person item having the property pope with the more recent start date. Snipre (talk) 08:52, 1 May 2017 (UTC)

@Snipre: I don't know if is the best idea or not, but I've been using Q30 as an example to follow. There you can find a list with the presidents of the United States. --Grabado (talk) 09:02, 1 May 2017 (UTC)

I would say that the approach used on United States of America (Q30) is a bit unnecessary - if it has office held by head of government (P1313)/office held by head of state (P1906), then it should be possible for a reuser to get all the presidents through a query on their position held (P39) values. I don't think we should recommend people do this for all countries, as it'd get very unwieldy. Andrew Gray (talk) 16:53, 1 May 2017 (UTC)

Pope is a position. It does not matter what it is exactly at a given time. So on the human you add "position held" - "Pope" with qualifiers for start and enddate and with predessesor and successor. Thanks, GerardM (talk) 11:32, 1 May 2017 (UTC)

Two joined islands - what property?

I have two islands that are joined together loosely. Stor-Krån (Q29639237) and Lill-Krån (Q29639231). Is there any good way to specify they are "joined" or is it maybe unnecessary? --Fringilla (talk) 01:45, 1 May 2017 (UTC)

I would use connects with (P2789). Shinnin (talk) 01:48, 1 May 2017 (UTC)

@Fringilla: In addition to Shinnin's solution, create a parent item and use "has parts". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:16, 1 May 2017 (UTC)

This is what we do for Lewis and Harris (Q363778). Andrew Gray (talk) 11:59, 1 May 2017 (UTC)

Infoboxes

Is there any initiative going on to link infoboxes across various language wikis? I have tried linking a few, some existing material on it will serve as a good guidance.  – The preceding unsigned comment was added by Capankajsmilyo (talk • contribs) at 08:34, 30 April 2017 (UTC).

@Capankajsmilyo: Would you please clarify/restate your question. Do you mean linking the templates for the infoboxes trough interwikis? Or do you mean infoboxes on the articles themselves?  — billinghurst sDrewth 03:32, 1 May 2017 (UTC)

I mean the linking of Infobox template pages through Wikidata. Using wikipedia links in languages. Capankajsmilyo (talk) 03:55, 1 May 2017 (UTC)

there has been some progress at english w:Template:Infobox person/Wikidata and french Modèle:Infobox Biographie2, but the english pushback was substantial. if you know lua, it is possible, but i do not see much help. there is also a wikimania talk https://wikimania2017.wikimedia.org/wiki/Submissions/Automatic_infoboxes_with_Wikidata - Slowking4 (talk) 00:43, 2 May 2017 (UTC)

Special:Watchlist

In the 'Current property proposals' section, someone might want to add the new sports category to the hidden template that displays links to thematic lists of proposals. My thanks to the one who'll know how to do this. Thierry Caro (talk) 07:14, 2 May 2017 (UTC)

 Done --Pasleim (talk) 09:04, 2 May 2017 (UTC)

Requesting community review of place translations

As part of my work at Mapbox, we're sourcing professional translations of world place names and donating the data to Wikidata. Our goal would be translate major places in the world to all major world languages and over 20,000 missing place label translations have already been sourced and uploaded over the last 2 months using the special account user:planemad_mapbox. While we have an internal process to check every translation for quality by a native language speaker, it would be amazing to ensure they run through another layer of review by expert Wikidata translators for a final check.

As a first step, it would be great to get interested translators to add their name to User:Planemad mapbox/World places translation project#Community reviewers. There are 32 languages in the list as of now, and there are around 200-4000 translations in each that would need to be checked, most of them in Asian languages. What other channels could be used to reach out to the translator community? What would be the best mechanism to organise such an effort, would a spreadsheet work? --Planemad (talk) 08:13, 2 May 2017 (UTC)

Significant change: new data type for tabular data files

Tracked in Phabricator
Task T164207

Hello all,

We’ve been working on a new data type that allows you to link to the tabular data files that are now stored on Commons. This data type will be deployed on Wikidata on May 15th.

The property creators will be able to create properties with this tabular data type by selecting “tabular data” in the data type list.

When the property is created, you can use it in statements, and when filling the value, if you start typing a string, you can choose the name of a file in the list of what exists on Commons.

Before the deployment, you can test it on http://test.wikidata.org (example).

One thing to note: We currently do not export statements that use this datatype to RDF. They can therefore not be queried in the Wikidata Query Service. The reason is that we are still waiting for tabular data files to get stable URIs. This is handled in this ticket.

If you have any question, feel free to ask! Lea Lacroix (WMDE) (talk) 08:35, 2 May 2017 (UTC)

Harvest Templates: sharing/automation platform?

I love Harvest Templates, and have been running jobs for most of the day.

Each job will probably need to be run again next year, as new articles appear and more information is added.

PROBLEM: Next year I will probably have switched to another pet project, or have forgotten about these jobs altogether.

QUESTION: Is there a place where people post their Harvest Templates permalinks, and the date at which each was last run?

That "place" could even run the jobs automatically once in a while, if that proves safe enough (maybe after review by a few admins?). The advantages would be triple: Not forget running them, not rely on peoples' flimsy web browsers, and generate a success/failure log visible by anyone.

Thanks! Syced (talk) 10:06, 2 May 2017 (UTC)

If those edits fall within an existing project, you might want to post some information there. See: Wikidata:WikiProjects. Of course, you could possibly start a project for harvest templates permalinks. Danrok (talk) 15:40, 2 May 2017 (UTC)

Wikidata weekly summary #258

Here's your quick overview of what has been happening around Wikidata over the last week.

• Events/Press/Blogs
  • Past: FOSSnorth (Gothenburg), where Lydia gave a keynote about Wikidata
  • Past: Wikidata meetup in Berlin
  • Past: Datensummit in Berlin, organized by OKFN, where we introduced Wikidata and the query service
  • Combining Forces with the Wikipedia Universe - by Songkick, about their data donation
  • Introducing Wikimedian in Residence, Mr Andy Mabbett - by The History of Modern Biomedicine Research Group

• Other Noteworthy Stuff
  • Attention: Editing will be disabled for an hour on May 3rd, to allow a server switch
  • New filters for the Recent Changes will be deployed on May 9th. (more info)
  • new data type for tabular data to be deployed on May 15th
  • Try our new script that displays constraint violations on the item pages
  • Get involved in the movement strategy for Wikidata
  • Query the Mediawiki API with the Query Service? Give us your feedbacks
  • Documentation on using OpenRefine to match from a data set you have to items in Wikidata

• Did you know?
  • Newest properties: Canadian Coastguard Lighthouse ID, contributed to published work, Répertoire national des associations identifier, IPA number order, UNESCO Thesaurus ID, Athletics Australia athlete ID, GuideStar Israel organization ID, MobyGames developer ID, newspaper format, STW Thesaurus for Economics ID, Bollywood Hungama ID, last words, Reprezentacija ID, LoJ peak ID, Ishim ID, GINCO ID, VIVC grape variety ID, column, had as last meal, ADAGP artist ID, CONICET person ID, Medium username
  • Query examples:
    • 15415 people who died in 2016
    • Maps of statues of Jeanne d'Arc on a horse (source)
    • Families who’ve had the same occupation for five generations (source)
    • Couples of capitals more than 19,000 km away from each other (source)
    • A ten-persons “surname = given name” chain on Wikidata (source)
  • Newest database reports: list of newspapers by newspaper format

• Development
  • Working on Lexeme support - specifically grammatical features of Forms of a Lexeme and statements on Forms
  • Finishing touches on a new notification type for Echo. This will notify the creator of an article when it has been connected to an item on Wikidata.
  • The new d:Special:AvailableBadges list all badges that can be used for sitelinks.
  • Discussed implementation plan for Multi Content Revisions which are a building block for structured data on Commons (phabricator:T107595)
  • Worked on feedback for the constraints API and user script
  • Worked on database and access improvements to take load off a much-used database table

You can see all open tickets related to Wikidata here.

• Monthly Tasks
  • Add labels, in your own language(s), for the new properties listed above.
  • Comment on property proposals: all open proposals - proposals needing attention
  • Suggested and open tasks!
  • Contribute to a Showcase item.
  • Help translate or proofread the interface and documentation pages, in your own language!
  • Help merge identical items across Wikimedia projects.
  • Help write the next summary!

Read the full report · Unsubscribe · Lea Lacroix (WMDE) 14:09, 2 May 2017 (UTC)

how to insert omim ID's in external ID's

 Resolved

Hello. This is my first time here at wikidata, although i've done a fair amount on enwiki. Is there some way of inserting omim ID's into the infoboxes? Thank you. JeanOhm (talk) 19:04, 6 May 2017 (UTC)

@JeanOhm: Do you refer to what we call Online Mendelian Inheritance in Man (Q241953)? You can use OMIM ID (P492) for that. You may want to create a property for this; see Wikidata:Property proposal. Ping me if you need help. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:16, 6 May 2017 (UTC)

@Pigsonthewing: Wow! I thought that learning enwiki was difficult! I'm not sure I can even compose coherent questions about wd! Yes, you got what I meant by omim. I looked at User:Joshbaumgartner/property available summary/400-499 and see that P492 exists as a property (as you indicated above), so why would I need to propose it as a property??? I see GORASP1 (Q18045706) which is, I think, what generates the infobox on en:GRASP65 The info box has a section "identifiers". Under that heading, there are "aliases" and "External IDs". There are 3 entries in the External IDs "MGI 1921748", "HomoloGene 49916" and "GeneCards GORASP1". There should, IMHO, be a fourth "OMIM 606867".

I see on Q18045706 that the HomoloGene number appears under "Statements" "HomoloGene ID". I see that GORASP1 appears under "Identifiers" "HGNC gene symbol" and several other places. However, I can't see how/where the MGI number appears. So, 3 diffefrent numbers, one appears by magic, and the other 2 appear in 2 different locations. Needless to say, I'm completely unsure how to incorporate the omim entry! I suspect that I am missing something incredibly basic, but can't figure out what. Please help this lost soul if you can! Thank you. JeanOhm (talk) 22:25, 6 May 2017 (UTC)

@Pigsonthewing: Wow²!!! This is looking more and more like magic! I turned my attention to the next page on my list, en:GRASP55 a "sister" of the one I wrote about above. I decided to be bold and tried to stumble around the wd page for the info box at GORASP2 (Q18037865) until I figured out how to insert an omim. After failing multiple ways, I finally stumbled upon the solution. I clicked on the "add" after the list of "statements" and before the list of "identifiers". That allowed me to enter the P492 and correct omim number. Clicked save, looked at the GRASP55 page, and, there it is! The omim entry in the external IDs "OMIM: 608693 MGI: 2135962 HomoloGene: 9180 o GeneCards: GORASP2"

Whoopee!

But, why the omim appears first, when I added it last is beyond me. Also, why I now don't see the omim on the wd page is beyond me. Finally, note that in my initial flailing around, I've managed to insert an "o" between "9180" and "GeneCards", which doesn't ruin the function of the infobox but screams out "some newbie didn't know what they were doing!" So, if you can figure out how to remove that "o" I'd appreciate it, and promise not to do it again! On to the the GRASP65 wd page! Be afraid, be very afraid....8-) JeanOhm (talk) 03:30, 7 May 2017 (UTC)

@Pigsonthewing: YES!!! I just added the omim number to the GRASP65/GORASP1 wd page without adding an "o"!!!!!! However, it still appears first in the list, which doesn't bother me at all, except for the appearane of magic.

So, this thread has been simplified to how to remove the "o". Thank you. JeanOhm (talk) 03:40, 7 May 2017 (UTC)

@JeanOhm: Where are you seeing the "o"? Can you check it's still there, please? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:49, 7 May 2017 (UTC)

Sorry; I meant to say "Otherwise you may want to create a property..." Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:43, 7 May 2017 (UTC)

@pigsonthewing: The "o" is no longer there! More magic??? 8-) Thanks, JeanOhm (talk) 18:35, 7 May 2017 (UTC)

This section was archived on a request by: Matěj Suchánek (talk) 11:11, 8 May 2017 (UTC)

Taxonomy and authors

Hoi, in taxonomy an author is not by his or her name but by a unique string of text. There is a property for this. In principle to describe an taxon (species, genus) properly, the author is a mandatory part. The name of the person is typically not really well known but any publication does include the abbreviation for the name. Would it be possible to only show the abbreviation when we add the author (for a taxon) property and can we please only show this abbreviation in Wikidata? Thanks, GerardM (talk) 05:17, 24 April 2017 (UTC)

If a biologist has their abbreviation, then it makes him notable for Wikidata, doesn't it? A Wikidata item can hold more information then a plain string. Matěj Suchánek (talk) 10:01, 24 April 2017 (UTC)

I know. I do not want to find him though his label but for the value that is the author information. Thanks, GerardM (talk) 11:16, 24 April 2017 (UTC)

Assuming that by "taxonomy" is meant "taxonomy of algae, fungi, and plants", there do exist "standard forms" which are recommended to be used when author citation is desired. Author citation is found in scientific works but not elsewhere (it is not "mandatory"). As things are, it is possible, when entering data, to put in the relevant "standard form" in the data field of "taxon author", and just about always the system will find the item of this taxonomist. - Brya (talk) 11:21, 24 April 2017 (UTC)

When you want to indicate a particular taxon, without author and publication (including date) there is no way that your identification suffices. The taxonomy approach at Wikidata is cutting corners, that is not really relevant here. What is relevant is how we can easily identify the author(s) associated with a specific taxon. To do that it is easiest to search by author and not by the label for the name of the person as you know. Thanks, GerardM (talk) 14:43, 24 April 2017 (UTC)

Sorry, but I do not understand what you want to achive. Could you give an example please. --Succu (talk) 15:24, 24 April 2017 (UTC)

I am sorry to hear you disagree with the International Code of Nomenclature for algae, fungi, and plants. The whole purpose of a Code of nomenclature is to make sure that a name with a particular spelling can only be used for one particular taxon (each taxon name is unique). This taxon may vary considerably in circumscription and it usually is more important to know which taxonomic viewpoint is used, rather than who published the name. - Brya (talk) 16:45, 24 April 2017 (UTC)

When a description for a taxon exists, it varies widely in what it is, what is included etc. Often subspecies, varieties change their existence and up as species or the other way around. So what makes that a taxon is not cast in stone. Consequently the name of the author and the publication are part of any description of a taxon. You could know it is part and parcel of the nomenclature. With Wikidata we intent to have an immutable identifier for a concept. Consequently as you lot have picked fixed positions that are problematic in and of itself, author and publication data are not nice to have, I understand.

As to what I want to achieve is that when I know the abbreviation for the author of a taxon, I can find him by his abbreviation and not by the Wikidata label. Thanks, GerardM (talk) 05:03, 25 April 2017 (UTC)

You are mixing nomenclatural with taxonomic issues. A circumscription of a taxon reflects the opinion of one or multiple authors and is called taxon concept. An example is the taxon labeled as Dionysia Fenzl (1843). The circumscription of this genus started with a single species Dionysia odora (Q15339763) (according to Eduard Fenzl (Q113291) in 1843 - Dionysia sec. Fenzl (1843)). In 2007 Magnus Lidén (Q5989947) revised the taxon in Dionysia sec. Lidén (2007) in which he accepted 49 species. Lidén is the author of the currently accepted taxon concept of Dionysia and Fenzl is the author who published the name Dionysia for the first time (in accordance with the rules of International Code of Nomenclature for algae, fungi, and plants (Q693148)). For the first kind of authorship we have author (P50) for the second one taxon author (P405) and ex taxon author (P697). To find an author by his abbreviation you can use the searchbox (try "Britton" or "L.") or use a SPARQL query for botanist author abbreviation (P428) or author citation (zoology) (P835), but be aware that only names regulated by International Code of Nomenclature for algae, fungi, and plants (Q693148) are more or less standardized. Seaching for a zoological author e.g. "Li" could be funny. I remeber a taxon name which was published by three different persons named Li. --Succu (talk) 15:27, 25 April 2017 (UTC)

Really.. What is it that we have in an item? Is it unique or is it not? If it is not, we need items for every permutation. As we currently have one specific outlook on taxonomy it should be obvious that an author, a publication and a publication date uniquely describe what it is we refer to. I do accept that given current assumptions we do without.

When you consider author designations, Britton often goes with Rose.. But never mind. Thanks, GerardM (talk) 17:51, 25 April 2017 (UTC)

They were coworkers and published - as you know - The Cactaceae (Q1388502) and some other works together. So what? How is this related to your question? We don't talk about math, so your term permutation (Q161519) is misplaced here. Our items represent taxon names (assumed to be unique within a Code), and apparently not taxon concepts (labeled with a subjective taxon name). We haven't „one specific outlook on taxonomy“ as you stated it. E.g. we have implemented the different views of Angiosperm Phylogeny Group (Q539818) (APG I-IV). You refused do give an example for your concerns requested above. So I can only guess what this thread is about: „Would it be possible to only show the abbreviation when we add the author (for a taxon) property and can we please only show this abbreviation in Wikidata?“ So again what do you expect, GerardM?! --Succu (talk) 21:01, 25 April 2017 (UTC)

You are wrong that "taxon names" are unique within a code as what such a name implies varies widely. Thanks, GerardM (talk) 05:10, 3 May 2017 (UTC)

I didn't wrote that. --Succu (talk) 06:53, 3 May 2017 (UTC)

Any reason you still don’t want to use « subclass of » for those classifications ? It’s definitely relevant. --author  TomT0m / talk page 11:49, 1 May 2017 (UTC)

This is a little bit off-topic, TomT0m, but I think you know the arguments against this kind of modelling very well. Nothing changed. --Succu (talk) 20:27, 2 May 2017 (UTC)

The strategy discussion. The Cycle 2 will start on May 5

The first cycle of the Wikimedia movement strategy process recently concluded. During that period, we were discussing the main directions for the Wikimedia movement over the next 15 years. There are more than 1500 summary statements collected from the various communities, but only 3 from the Wikidata discussion. The strategy facilitators and many volunteers have summarized the discussions of the previous month. A quantitative analysis of the statements will be posted on Meta for translation this week, alongside the report from the Berlin conference.

The second cycle will begin soon. It's set to begin on May 5 and run until May 31. During that period, you will be invited to dive into the main topics that emerged in the first cycle, discuss what they mean, which ones are the most important and why, and what their practical implications are. This work will be informed and complemented by research involving new voices that haven’t traditionally been included in strategy discussions, like readers, partners, and experts. Together, we will begin to make sense of all this information and organize it into a meaningful guiding document, which we will all collectively refine during the third and last cycle in June−July.

We want to help your community to be more engaged with the discussions in the next cycle. Now, we are looking for volunteers who could

• tell us where to announce the start of the Cycle 2, and how to do that, so we could be sure the majority of your community is informed and has a chance to feel committed, and
• facilitate the Cycle 2 discussions here, on Wikidata.

We are looking forward to your feedback!

Base (WMF) and SGrabarczuk (WMF) (talk) 08:47, 27 April 2017 (UTC)

Thanks for this post. While there was not much input from wikidata on this so far, I would recommend that all wikidata-interested folks look at the summary linked above - wikidata appears very frequently (usually positively) in the strategy suggestions from our sister projects. I think it behooves us to think about what that means for wikidata in the coming years. ArthurPSmith (talk) 14:22, 27 April 2017 (UTC)

Sure, think and share the thoughts, and the best place to do that is here, on Wikidata talk:Strategy 2017 :) SGrabarczuk (WMF) (talk) 19:53, 27 April 2017 (UTC)

Seriously, SGrabarczuk (WMF): „What role do we want to play in the world in 2030?“ Sounds to me like hubris (Q735766). Something like the question what a role fullfills a modern mobile phone (Q17517) produced around the last Q1678709 fifteen years later. I think currently Wikidata has some impact on linked open data (Q18692990), but we are far away from modelling a ontology (Q324254) around that data. Supporting the last is more important to me than fabulating about fast evolving technologies. --Succu (talk) 20:38, 27 April 2017 (UTC)

@Succu: this is the movement strategy process, so we're focusing on the broadest directions. Of course we don't know what exactly happens in 15 years, but we can discuss e.g. how to increase the base of potential Wikidata users (including re-users), whether Wikidata should stay under Wikimedia umbrella, what role should Wikidata fulfill and what shouldn't within our movement, anything broad and abstract enough, anything that could be applicable to / cover 15 years. Please have a look at the summary of 1500 summaries linked above. Or, should I actually write, we could have discussed that, because now, there's going to be the Cycle 2, when we focus on 5 out of all those ideas. So again, please have a look at the links. SGrabarczuk (WMF) (talk) 20:51, 27 April 2017 (UTC)

No, SGrabarczuk (WMF), it's a WMF request, not something the „movement“ initialised as your answer states. Why should I'm interested in the „summary of 1500 summaries linked above“. Repeating strategy headlines is far away from something I would call a strategic planning (Q932522), but ignorance (Q815577) about the current state of Wikidata. --Succu (talk) 21:26, 27 April 2017 (UTC)

A year ago, in Esino Lario, Wikimania attendees were heavily interested in a broad strategy. Only a part of them were WMF staff. WMF is organizing it, but I doubt if this is really by WMF request (by the way, who is WMF in this case?). SGrabarczuk (WMF) (talk) 21:34, 27 April 2017 (UTC)

Not sure why people are objecting to this - WMF may be facilitating, but they are asking the community for direction here, this is important. ArthurPSmith (talk) 14:58, 28 April 2017 (UTC)

Probably I'm suffering on some kind of marketing allergy. --Succu (talk) 21:57, 28 April 2017 (UTC)

Dear Base (WMF) and SGrabarczuk (WMF), regarding your questions. As for where to announce, here and the mailing list are the right place. You already did that, but that doesn't guarantee commitment. As for facilitating cycle 2 discussions here, don't you think that it would be more interesting to gather more cycle 1 statements first? I mean, what is the point of having cycle 2 if almost nobody has cared so far to participate in cycle 1? Please do not follow your plan blindly, that is what bureaucrats do and we end up with no real impact. If the approach to cycle 1 has failed (and looking at the numbers it obviously has), then it is better to rethink the approach for the Wikidata community. Maybe it can be an independent track? The Wikidata:WikidataCon 2017 is in Oct 28th, it might be a good time to discuss it there.--Micru (talk) 06:41, 29 April 2017 (UTC)

@Micru:, thanks for your answer. The team focused on the multilingual and English projects is smaller and weaker than planned (there's 2 of us instead of 3). Thus, it was a bit more difficult to gain as much commitment here as we'd like to gain. I'll pass your request to WMF staff, but I can't promise anything. As for the number of statements, there are over 1800 of them globally, and certainly many of them are related to Wikidata. Perhaps some Wikidata contributors, who are active also elsewhere, participated in the discussions e.g. on their home wikis, or on Meta. (Sure, I'd prefer all big communities to have their local discussions). So, I agree, the situation is difficult. But not tragic. SGrabarczuk (WMF) (talk) 13:33, 29 April 2017 (UTC)

i sympathize with the grousing about the strategy process. but the WMF is making an effort to listen, and so the burden is on the community to communicate in action oriented steps, rather than the perpetual - WMF = frenemy. constructive feedback has a small hope; non-constructive has none. 98.163.68.171 22:03, 30 April 2017 (UTC)

Thank you for your patience, @Micru: Hi. Many Wikidatans, as SGrabarczuk notes, are likely to have participated elsewhere, such as metawiki, or in sister projects in their primary language. The analysis of the discussions in cycle 1 is recently completed (m:Strategy/Wikimedia movement/2017/Cycle 1/Report) and it would be impractical to have one community continue brainstorming raw ideas while most other people have already moved on to discussing, modifying, and spelling out in more detail the results of the initial idea collection that ended last month.

That said, Cycle 1 was a relatively abstract process of idea formation, in reaction to purposefully open-ended questions; and that's not an approach that everyone enjoys working with. Cycle 2 will be a lot more concrete and probably simpler to participate in for more people, as it will feature clear statements that we can debate in depth, toss around (and out), or further refine into something that is generally agreeable. Cheers, Quiddity (WMF) (talk) 00:53, 4 May 2017 (UTC)

seal (Q162919)

A seal (Q162919) is said to be at least two things according to the wikis I've looked at, and the Oxford Dictionary defintion.

So, it is either the embossed wax seal (or similar item), or the device used to make that embossed seal.

The Wikidata item for seal (Q162919) is not currently claimed to be a device of any sort, which is a problem when trying to state what the Butuan Ivory Seal (Q29634488) is.

What to do? Create a new item for the device, or something else? Danrok (talk) 18:25, 30 April 2017 (UTC)

enWP's article is for the device, and the article says that the result of the use of the device is called a seal impression or a sealing neither term having an item. Great Seal of the Realm (Q3114830) links to its use, and would seem to align with what you were wishing to do. I would suggest that if there are mislinking at Special:WhatLinksHere/Q162919 that there needs to be some correcting to identified items/terms, and some clarification on the descriptions.  — billinghurst sDrewth 03:44, 1 May 2017 (UTC)

Thanks! I can see what may need to be done, within the scope of English, however it's difficult to be certain of how this would work out in all languages. Danrok (talk) 23:54, 3 May 2017 (UTC)

@Danrok: The p+ve and n-ve of Wikidata and a large volunteer base! Other languages may have distinct words and it isn't even an issue; whereas some languages the whole matter will be conceptually foreign. In the end the answer here is always "it is a wiki". Which is always right, but not necessarily helpful.  — billinghurst sDrewth 04:21, 4 May 2017 (UTC)

Article (grammar)

Do we have an w:en:Article (grammar) property? For example, I speak Greek and some words have "ο" as an article (masculine singular), some "η" (feminine singular), some "το" (neuter singular), some "οι" (masculine–feminine plural) and some "τα" (neuter plural). For example,

• ο Καναδάς (Canada (Q16))
• η Γαλλία (France (Q142))
• το Ηνωμένο Βασίλειο (United Kingdom (Q145))

• οι Άλπεις (Alps (Q1286))
• τα Ηνωμένα Αραβικά Εμιράτα (United Arab Emirates (Q878))

If we don't have, do you think we need such a property? To show what is the article of the name of the item in each language?

Xaris333 (talk) 22:35, 30 April 2017 (UTC)

@Xaris333: I think it will be relevant for Lexeme that should be coming soon. Pamputt (talk) 23:16, 30 April 2017 (UTC)

Items in the Q namespace are not words. They are concept and a concept can be represented by different words that have different articles. The WMDE team currently works on the integration of Wikidictionary so that we have in the future a new namespace for words/lexeme that can store attributes such as the gender of a word. ChristianKl (talk) 10:33, 1 May 2017 (UTC)

Thanks! Xaris333 (talk) 22:01, 2 May 2017 (UTC)

IPA number order (P3917) has wrong datatype

IPA number order (BEING DELETED) (P3917) has data type "quantity", it should be "external-id" (I and others missed this during its proposal discussion). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:15, 1 May 2017 (UTC)

As far as I understand it's only possible to change "string" into "external-id" and vice versa after the property is created. That means we need a request for deletion for the existing property. ChristianKl (talk) 13:35, 1 May 2017 (UTC)

Why should it be "external-id"? What is this external ressource? Pamputt (talk) 20:37, 1 May 2017 (UTC)

Because it's an identifier, minted by an external authority. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:49, 3 May 2017 (UTC)

Just confirming that yes, it's only possible to do this from string to extID. --Edgars2007 (talk) 04:55, 4 May 2017 (UTC)

Mean height

Hello. Can elevation above sea level (P2044) be used as a mean height of the item (geographical object) as measured relative to sea level? Some places (like villages) have lowest and highest heights. Xaris333 (talk) 14:56, 2 May 2017 (UTC)

I stumbled upon that problem before, see Property talk:P2044#Altitude of a city?. Either we'd need additional properties for a minimum and maximum value, or we'd need a new datatype to encode a range, or use the error margins of the quantity type to encode minimum and maximum. But though the datatype allows an upper and lower bound to be set, the current UI only allows to set a single ± value. Ahoerstemeier (talk) 10:51, 3 May 2017 (UTC)

Quickstatements2

Hello. I am using quickstatements2

Q353925 P1082 450 P459 Q39825 P585 2001 S248 Q16331384

Is not adding P585 with 2001 [2]. Anyone one knows why?

Xaris333 (talk) 16:17, 2 May 2017 (UTC)

May be you can try +2001-01-01T00:00:00Z/9 instead of 2001? --Ghuron (talk) 19:15, 2 May 2017 (UTC)

Thanks! Its working! Xaris333 (talk) 20:27, 2 May 2017 (UTC)

Q1479935 - Game warden / Park ranger

Hi, we noticed an issue on de-WP. There exist different articles to the topic of Game wardens de:Wildhüter and Park rangers de:Ranger (Schutzgebietsbetreuer). In addition to that we also have de:Jagdaufseher, which covers a assistant to the main licensed hunter of an area. On en-WP en:Park ranger covers the whole history from the early beginnings to todays game wardens in most countries, while en:National Park Service Ranger describes only the occupation within the US National Park Service. The problems don't stop here. In Germany, some Rangers are volunteers and many don't have any law enforcement capacity, while in the anglophone world, law enforcement is the core activity and main topic of education. How can we deal with these differences with regard to Q1479935 and/or any other object? --H-stt (talk) 18:22, 2 May 2017 (UTC)

Not sure on the specifics here, but as a general rule if things are sufficiently different between languages, for wikidata the right thing to do is to find or create a "common" item that all of them can be grouped under, and then add subclass of (P279) statements to organize them. If there are interwiki links you want to retain you can still do it the old way (on the specific language wiki pages). ArthurPSmith (talk) 19:47, 2 May 2017 (UTC)

THX. I feared that answer, because it means that I have to do the work, instead of just pointing you to the issue and someone somehow magically resolves it ... *g* --H-stt (talk) 19:14, 3 May 2017 (UTC)

Quering multiple categories simultaneously

Hello, I'm trying to query PetScan with multiple categories. I tested comma, semicolon, new line and other separations between categories but it doesn't do the right thing. So the question is how to put multi categories as an input and get a sorted (and hopefully hyperlinked e.g. in the csv format) output of all inserted categories (their wiki articles)? It would save a lot of time instead of doing it for each single category. Here an example PetScan
Any tip would be appreciated. --Sky xe (talk) 19:07, 2 May 2017 (UTC)

Below the field where you enter categories, there is an option to select either 'subset' (left) or 'union' (right). To get all articles from all the listed categories, you need to select union. Shinnin (talk) 20:19, 2 May 2017 (UTC)

Thank you for your answer Shinnin, but isn't it possible to sort the output by category? Since i get a mixed output this way. Or is there eventually any bot which can does that so I might ask its owner to use it?

Wikidata API GetEntities with Property names

Hi,

Wikidata api's wbgetentities method returns properties of entities in the claims sub-object. If this property points to another wikidata object, it just returns the id of that object rather than returning also its name. For example this link returns 'Renaissance music'.

https://www.wikidata.org/w/api.php?action=wbgetentities&ids=Q201405&languages=en&format=json

And the first claim it returns is 'P279' -> Q1583807 (wikibase-item)

Here, P279 is the property 'subclass of' and the item with id Q1583807 is 'art music'. We need another service call to get the name of the property Q1583807 which is not very efficient when you think that there are at least 10 properties in average like this in Wikidata.

Is there any other way that makes it possible to access directly to property names, or would developers think about simply adding 'name' property for claims which have a datatype of 'wikidata-item'?

Thank you in advance.

Bluecun (talk) 11:19, 3 May 2017 (UTC)

New notification when a page is connected to Wikidata

Hello all,

For your information, during the next months, we will deploy a new notification type on the Wikimedia projects where the sitelinks are provided by Wikidata. When an editor creates a new page (Wikipedia article, Wikivoyage page, etc.) and this page is connected later to a Wikidata item, the creator of the page will receive a notification to inform him about that, with a message like “The page X was connected to a Wikidata item”, plus the link to the item.

This will hopefully help people to be more aware of Wikidata and the help we provide to the other projects. This feature will be disable by default for existing editors, and enabled by default for new editors.

You can find the plan of deployment on the different wikis here. Thanks go to Matěj Suchánek who suggested and developed this feature! Lea Lacroix (WMDE) (talk) 08:54, 3 May 2017 (UTC)

@Lea Lacroix (WMDE): Wouldn't be it more useful to notify when a page has been disconnected from an item? It could also be a form of vandalism.--Micru (talk) 11:33, 3 May 2017 (UTC)

That's interesting! I think we can think about adding this feature later. I created a ticket about this. Lea Lacroix (WMDE) (talk) 16:09, 3 May 2017 (UTC)

Stamp series

Is there any difference between stamp series (Q673383) and stamp edition (Q16937116)? Both seem to mean "a group of stamps with a similar design or theme". – Einstein2 (talk) 11:42, 3 May 2017 (UTC)

User:Zolo, could you please give an answer. --Succu (talk) 19:54, 3 May 2017 (UTC)

A stamp series (Q673383) contains several different stamp models (= stamp edition (Q16937116)). See for example de:Olympische Spiele 1896 (Briefmarkenserie).

Penny Black (Q107701) is a instance stamp edition (Q16937116). Some would simply tag it with Penny Black (Q107701)instance of (P31)postage stamp (Q37930), but that would be incorrect. An instance of stamp is one individual copy of a stamp (for example the Penny Black in the collection of museum X). --Zolo (talk) 08:58, 4 May 2017 (UTC)

@Zolo: Thanks for the answer. I got confused because items of a few series (such as Epaulettes (Q2531180) and Machin series (Q497949)) are tagged with instance of (P31)stamp edition (Q16937116). Can these be changed to instance of (P31)stamp series (Q673383)? – Einstein2 (talk) 11:49, 4 May 2017 (UTC)

@Einstein2: yes some items need to be fixed. --Zolo (talk) 12:52, 4 May 2017 (UTC)

Oh actually I see I had originally mixed things up in the description and labels.--Zolo (talk) 14:29, 4 May 2017 (UTC)

Track the number of identifier properties

Tracked in Phabricator
Task T114617

Hello all,

It is now possible to track the number of identifier properties for an item, in addition to the total number of claims.

You can access this information in Labs, directly via shell or using Quarry. Here you can find an example and the result will look like this. The row "wb-identifiers" corresponds to the number of identifier properties in the item.

This feature is already deployed, but the database is currently being populated ; some errors may occur during the next days.

If you have questions or suggestions, feel free to let a comment on the ticket. Thanks, Lea Lacroix (WMDE) (talk) 14:53, 3 May 2017 (UTC)

Voting has begun in 2017 Wikimedia Foundation Board of Trustees elections

This is a message from the Wikimedia Foundation Elections Committee. Translations are available.

Voting has begun for eligible voters in the 2017 elections for the Wikimedia Foundation Board of Trustees.

The Wikimedia Foundation Board of Trustees is the ultimate governing authority of the Wikimedia Foundation, a 501(c)(3) non-profit organization registered in the United States. The Wikimedia Foundation manages many diverse projects such as Wikipedia and Commons.

The voting phase lasts from 00:00 UTC May 1 to 23:59 UTC May 14. Click here to vote. More information on the candidates and the elections can be found on the 2017 Board of Trustees election page on Meta-Wiki.

On behalf of the Elections Committee,
Katie Chan, Chair, Wikimedia Foundation Elections Committee
Joe Sutherland, Community Advocate, Wikimedia Foundation

Posted by the MediaWiki message delivery • Translate • Get help 19:14, 3 May 2017 (UTC)

Help with Q6545954

Hi there! I'm hoping someone here might be able to help with a question I have about the wikidata entry for Ligado Networks, formerly LightSquared. Last year, the ENWP article for LightSquared was moved to en:Ligado Networks and the content of the article updated based on the company's name change. However, the wikidata entry that is linked from the Wikipedia article—Ligado Networks (Q6545954)—is titled "LightSquared" and still has the old company name and website etc. I'm not sure what the appropriate steps are to take here: should a new wikidata entry be created for Ligado Networks? Or is it possible to update the existing LightSquared entry (Q6545954) and change it to Ligado Networks?

In full disclosure: I was working on behalf of Ligado Networks via The Glover Park Group as part of my work at Beutler Ink last year when I requested the move and updates to the ENWP article, but am no longer under contract to them. I'm following up on this out of interest and as a favor to colleagues at The Glover Park Group. Thanks in advance for any help and advice! 16912 Rhiannon (talk) 19:49, 3 May 2017 (UTC)

if it is just a name change, then just use the existing wikidata item and change the name and the enwiki link. ArthurPSmith (talk) 20:56, 3 May 2017 (UTC)

Hi ArthurPSmith, thanks for your reply! Sorry if I'm being a bit dense here but the existing wikidata entry (Q6545954) already links to the updated ENWP article—is there something else that needs to be done with regards to that link? The main issue is that the wikidata has the wrong name and outdated details (e.g. website). Are you saying it's ok to just change the name of the wikidata entry from LightSquared to Ligado Networks? Again, apologies, I'm not very familiar with how wikidata works and I'm eager not to mess anything up. 16912 Rhiannon (talk) 21:13, 3 May 2017 (UTC)

Oh I didn't realize the link was fixed already. Yes just change the name and website. If you have some sort of third party source for the new website please add that as a reference. Also you should probably keep the old name as one of the "Also known as" entries so people can find it. ArthurPSmith (talk) 21:20, 3 May 2017 (UTC)

Another thought - you can keep the old "official website" entry and attach a end time (P582) qualifier. Similarly you can enter the old name under official name (P1448) with the same qualifier. ArthurPSmith (talk) 21:22, 3 May 2017 (UTC)

Thanks so much, ArthurPSmith! I've edited the entry but couldn't figure out the end time (P582) qualifier, so I just put the old name into "Also known as" for now and replaced the website. Appreciate your help. 16912 Rhiannon (talk) 13:41, 4 May 2017 (UTC)

Articles in Wiki category without label in X language

I have w:pl:Kategoria:Polscy lekkoatleci (Category:Polish athletes (Q7025791)) with(-out) subcategories (currently that doesn't matter). Can I have (in some sane way) a list of articles that doesn't have en label? Not using SPARQL (occupation (P106)=subclass of athletics competitor (Q11513337)). This time I want to have wikicategories based search. Petscan seemingly doesn't offer that or I wasn't able to put thing together. --Edgars2007 (talk) 05:03, 4 May 2017 (UTC)

If I understood correctly, you are looking for how to fill a category like Category:Articles with missing Wikidata information (Q16737095), right? Pamputt (talk) 05:32, 4 May 2017 (UTC)

Probably yes (if I understood what you said :D ). I want to add English labels to those pages (in w:pl:Kategoria:Polscy lekkoatleci), which doesn't have it. --Edgars2007 (talk) 05:59, 4 May 2017 (UTC)

So no, the category I mentionned is the opposite. It contains pages of a given linguistic version of Wikipedia that use label from Wikidata that are not translated into the language used by this linguistic version of Wikipedia. So, I do not think it is what you are looking for. Pamputt (talk) 06:08, 4 May 2017 (UTC)

You will see all the entries for categories in Reasonator when they have been defined with "is a list of" "whatever" and qualifiers that define what the category is specifically about. You can add labels in Reasonator. Thanks, GerardM (talk) 08:45, 4 May 2017 (UTC)

New: truthy dumps

Hello all,

As you may know, we've been providing dumps of the database as a beta feature, in several formats. You can find some documentation here.

We've recently added a new dump format: nt truthy dump. This dump is made of n-triples and contains only "truthy" information: statements with value. It doesn't contain qualifiers, references, full statements. That means this format is smaller and easier to process.

You can find an example of these dumps here. These dumps will be generated weekly.

If you have any question, feel free to reach hoo_man or me. Cheers, Lea Lacroix (WMDE) (talk) 07:59, 4 May 2017 (UTC)

Update of the constraint table

Hello all,

Since we deployed the constraint API and its user script, several of you noticed some errors in the constraints and fixed them. Thanks for all your feedbacks!

On May 11th, we will update again our constraint table, so the API and script are up-to-date. If you would like some modifications of the constraints to appear on this new version, please make your changes on the properties before May 11th.

Cheers, Lea Lacroix (WMDE) (talk) 08:11, 4 May 2017 (UTC)

named after

Hello. Can we use named after (P138) if there are more than one that may inspired the subject's name? Xaris333 (talk) 13:21, 4 May 2017 (UTC)

If you have two sources that disagree on this, sure, and link the references on the two or more distinct claims. If it's just your own speculation I wouldn't do that. ArthurPSmith (talk) 15:28, 4 May 2017 (UTC)

One source with 2 or 3 possibilities that may inspired the subject's name. Xaris333 (talk) 16:03, 4 May 2017 (UTC)

Section qualifier for Scientific or Academic Society membership

Hello,

I'm wondering if there is any qualifier that might be used to indicate that a certain person is in a section being member (member of (P463)) of a scientific or academic society (i. e., https://www.rsm.ac.uk/sections/sections-and-networks-list.aspx ) Thanks! --Toniher (talk) 15:04, 4 May 2017 (UTC)

Create an item for the section (as part of (P361) the parent organisation); use that as the value for P463. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:21, 4 May 2017 (UTC)

Trio

musical trio (Q281643) is described in English as "group of three musicians or composition for three". Does it need to be split, or is the English description wrong? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:42, 29 April 2017 (UTC)

@Pigsonthewing: A trio is not always 3 musicians, so it is safe to remove the music part. PokestarFan • Drink some tea and talk with me • Stalk my edits • I'm not shouting, I just like this font! 10:44, 29 April 2017 (UTC)

You're thinking of trio (Q16145172). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:47, 29 April 2017 (UTC)

I was wrong. It works. PokestarFan • Drink some tea and talk with me • Stalk my edits • I'm not shouting, I just like this font! 10:48, 29 April 2017 (UTC)

Yes trio should be split into music and people. Very odd that the two could be interchangeable. Jane023 (talk) 14:20, 30 April 2017 (UTC)

Note the same issue for quartet (Q1135557) and quintet (Q389888) - both are subclasses of musical ensemble and of musical composition. It is true the same word is used for both contexts, but the meanings seem distinct enough! ArthurPSmith (talk) 14:47, 1 May 2017 (UTC)

set string trio (Q29787068) = : 1) group of three string instruments (superset: "combination/configuration (group) of string instruments", instrumentation (Q617028)?), 2) group of musicians (superset: musical group (Q215380)), 3) composition/piece for group of three musicians (composed musical work (Q207628))? --Fractaler (talk) 12:11, 5 May 2017 (UTC)

Editing multiple language labels/descriptions/aliases at once?

In the past, I’ve been able to edit the labels section for all of the languages that I have listed with Babel. It worked most of the time, but I would occasionally get sent to a separate page where I could only edit the information one language at a time, and I would have to change the URL in order to edit the label information in a different language. But now when I try, every single time, I get sent to that other page for editing, and I can never edit all languages at once. Did something change in preferences that I’m not aware of? Or is there something that I need to do to change this back? Cheers, -- Irn (talk) 00:08, 2 May 2017 (UTC)

This sounds like your java script is broken or disabled. Did you recently add a new user script or gadget? Or did you enable a java script blocker? --Lydia Pintscher (WMDE) (talk) 15:52, 2 May 2017 (UTC)

Hm, no, I haven't changed or added anything. But since you mention that, I'm having other technical issues on en.wiki (and having that same problem here), and I wonder if they might not have the same cause? Also, for what it's worth, I'm having the issue with both Firefox and Chrome. -- Irn (talk) 00:55, 3 May 2017 (UTC)

So the other issue got resolved, and I've realized that I can't edit statements, either. And when I'm not logged in, I can edit statements. Javascript not working sounds like a plausible explanation. But how do I fix that? -- Irn (talk) 03:30, 4 May 2017 (UTC)

You can check what the issue is in the console of your browser. It should open when pressing F12. Can you check if there are any error messages? --Lydia Pintscher (WMDE) (talk) 07:22, 5 May 2017 (UTC)

Wow, yeah, there's a lot there:

Use of "importScriptURI" is deprecated. Use mw.loader instead.

This page is using the deprecated ResourceLoader module "jquery.ui.position".

This page is using the deprecated ResourceLoader module "jquery.ui.widget".

This page is using the deprecated ResourceLoader module "jquery.ui.core".

Please use "mediawiki.ui.button" or "oojs-ui" instead.

Use of "wgUserGroups" is deprecated. Use mw.config instead.

This page is using the deprecated ResourceLoader module "jquery.jStorage".

Please use "mediawiki.storage" instead.

Cheers, -- Irn (talk) 12:34, 5 May 2017 (UTC)

Connect Projects from Wikipedia to Wikidata

Question

We have projects on the Danish language Wikipedia. Normally we have the templates on the discussion page (like on ENWP). There has been a vote though to remove them from the discussion pages, and we're looking for a way to use alternatives.

The question is

Is it possible to connect pages in a database on Wikidata. For example: Project-properties as "P" with the articles as "Q" and qualifiers (like Stub, start, good, promising and excellent)

Why?

Se for example London which has 5 templates that say that the article is a "Good Article". But it falls under 5 projects. It should be possible for Wikidata to host the status of each article of a project on the various Wiki's.

For example

London falls under the scope of several projects. So The Wikidata article of London should get a property: Project - The Qualifier should then be Olympic Games, London, Cycling, Hanse, etc. The Danish Wikipedia would then place that article under a Qualifier like Status: "Good Article" and the Dutch Wikipedia maybe would place this as "Excellent Article". The LUA-module would do the rest.

Could this be achieved? I know that each wiki has it's own projects, but it would be beneficial for each wiki to compare. I hope to receive some positive thoughts. Kind regards,  Rodejong  💬 ✉️  12:55, 4 May 2017 (UTC)

@Rodejong: As far as I know this is being tackled by the project Multi-Content Revisions (phab:T107595). It has been in discussion for a long time, but it seems that finally it will be done, as it is also necessary for the Wikidata for Commons project. Once it is ready, theoretically you will be able to store and query that information.--Micru (talk) 15:08, 4 May 2017 (UTC)

@Micru: Thanks for the feedback. We'll have to wait then. :) Kind regards,  Rodejong  💬 ✉️  21:55, 4 May 2017 (UTC)

@Rodejong: A property for projects has been considered before and rejected because different wikis will likely have different projects associated with differing items, and we try not to get wiki-specific. As for the task feature you want, that's probably not MCR (though it might eventually plug in to MCR) but instead PageAssessments. (Ping Micru too.) --Izno (talk) 03:11, 5 May 2017 (UTC)

@Izno: Okay, that is unfortunate. Then we need to look for other possibilities. Thanks though for your feedback. Kind regards,  Rodejong  💬 ✉️  05:24, 5 May 2017 (UTC)

Schema.org and Wikidata

Schema.org provides a vocabulary for structured data markup on the Web. Structured markup using this markup can be found on more than 30% of pages in a sample of 10 billion. Schema.org is used by a wide range of organizations: from the New York Times to the WWF, whether global organizations like Greenpeace or UNESCO, or establishments like many local cinemas or pubs, Schema.org is used by more than 10 Million organizations worldwide to publish data on the Web. You can read more about Schema.org in CACM or on Wikipedia.

A major cost factor for applications using this data is in aggregating data about a given entity from different sources. Whereas the vocabulary is standardized - Schema.org defines properties and types - the identifiers for the individual items were not. This was done by design, to make it easier for publishers (see the aforementioned CACM paper for details).

In order to reduce the cost for applications consuming Schema.org markup, in particular smaller organizations and individual developers, to aggregate fragments of Schema.org markup from different sources, Schema.org is considering to encourage the use of Wikidata as a common entity base for the target of the sameAs relation.

There is also a class of entities, that are intermediate in generality, between very high general terms such as Person and birthDate and very specific concepts such as individual persons or movies, that may be standardized. This includes lists such as the list of languages and countries. The idea is to use SPARQL queries in order to produce and publish easy to use URIs for those items, e.g. https://lists.schema.org/Country/France. These would be published by Schema.org with a mapping to Wikidata as part of the normal Schema.org release process. The necessity for these arise from the fact that they will be easier to use and reuse than the Q-ID based Wikidata URIs.

This will allow anyone to grab a bunch of data from different sites, and integrate them with much less effort than currently. To name just one example: IMDB publishes data about movies using Schema.org. Wikidata uses these pages as references. By having IMDB using Wikidata identifiers, scripts like the one developed by Adam Shoreland, will be able to much easier compare the existing data in Wikidata with such external data sources - on many more sites.

Schema.org would like to discuss this step with the Wikidata community before implementing it, in order to discuss potential issues early and prepare for them. So I am here to open this discussion. --Denny (talk) 17:57, (wearing his Google hat) 4 May 2017 (UTC)

May I take the liberty of moving detailed discussion to Wikidata:Schema.org and its talk page? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:25, 4 May 2017 (UTC)

You may :) --Denny (talk) 20:05, 4 May 2017 (UTC)

Discussion moved to Wikidata:Schema.org / Wikidata talk:Schema.org. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:44, 5 May 2017 (UTC)

TTL of Constraints?

So I use check_constraints.js by User:Jonas_Kress_(WMDE), and I noticed one of the constraints beeing wrong, so I fixed it. But the tool still shows the old constraint. This was 3 days ago. How long are the contraints cached? -- MichaelSchoenitzer (talk) 14:12, 5 May 2017 (UTC)

#Update of the constraint table Matěj Suchánek (talk) 14:35, 5 May 2017 (UTC)

my novel

Ask questions about using the Wikidata's items to write in the my novel (this page), can will do that? and contrary to Wikidata policy? (details see in the link) --Nakare✝ (talk) 15:00, 5 May 2017 (UTC)

It is fine regarding Wikidata's data licensing policy. I am not sure whether user subpages are the best place to write a novel, but the idea is intriguingly novel. Curious to see what comes out :) --Denny (talk) 16:10, 5 May 2017 (UTC)

Error while trying to change preferences

Tracked in Phabricator
Task T165011
Resolved

Not sure where this should go but I will inform about this issue here. When I tried to change my preferences I get the following error message [WRQqWgpAMFgAAAIOqm4AAAAJ] 2017-05-11 09:09:46: Fatal exception of type "MWException". This does not happen on any other wikiproject that I have tried. --MiPe (talk) 09:11, 11 May 2017 (UTC)

same here ! Hsarrazin (talk) 09:15, 11 May 2017 (UTC)

Confirming. Seems to be reported as phab:T165011 (and it's caused by ORES). Matěj Suchánek (talk) 09:35, 11 May 2017 (UTC)

Thanks for reporting. We're going to have a look at this as soon as possible. Lea Lacroix (WMDE) (talk) 09:51, 11 May 2017 (UTC)

I can't access my preferences at all! Jared Preston (talk) 15:04, 11 May 2017 (UTC)

Jared Preston, MiPe, Hsarrazin, Matěj Suchánek and Lea Lacroix (WMDE) - Special:Preferences are now accessible. Trizek (WMF) (talk) 16:00, 11 May 2017 (UTC)

They certainly are, thanks for the quick fix! Jared Preston (talk) 16:29, 11 May 2017 (UTC)

This section was archived on a request by: Matěj Suchánek (talk) 18:40, 11 May 2017 (UTC)

QS question

How can I add <no value> with QuickStatements? Queryzo (talk) 05:39, 3 May 2017 (UTC)

@Magnus Manske: ? Queryzo (talk) 12:15, 5 May 2017 (UTC)

Seems impossible yet. Is there a particular set of items you would like to modify in bulk? Matěj Suchánek (talk) 16:40, 6 May 2017 (UTC)

Yes, all episodes of SpongeBob SquarePants (Q83279) with instance of (P31) = segment of a television episode (Q29555881) like Reef Blower (Q999929) should have IMDb ID (P345) = <no value>. Queryzo (talk) 17:30, 6 May 2017 (UTC)

Merging problem

Could somebody help me with this issue before it peters out in the end? Thanks a lot in advance! Best--Hubon (talk) 20:46, 5 May 2017 (UTC)

@Hubon: Q7315979 and responsibility (Q1274115) should not be merged; one is a concept, the other a disambiguation page. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:12, 5 May 2017 (UTC)

Okay, thank you!--Hubon (talk) 21:15, 5 May 2017 (UTC)

Why the limitation to search functionality? Makes finding people/things more difficult

We used to be able to search broadly through item data, whereas more recently the search limitation seems to be more related to label and description. Such limitations are regressive IMNSHO. I used to be able to search for something like "John Reading 1692" and it would find items like John Reading (Q6254393), however, no longer. Now I have to go to English Wikipedia and find the article, then progress back to the article. <aaaagh> This is simply going to lead more duplicates being created if one cannot search by name and the year of event to find people within the system. Similarly if one is searching for family names, one now has to hope that there is a description, rather than rely on the instance being present (and the instance is way more important).  — billinghurst sDrewth 10:26, 6 May 2017 (UTC)

Wikidata in VIAF

Question to those, who spent more time with VIAF and other like than me :) Is Wikidata link (WKP) in VIAF is reliable? For those VIAF identifiers, which aren't already mapped to Wikidata entries here, on Wikidata. If it is, then maybe it would be good idea to use VIAF dump to at least populate VIAF, and maybe other authority data? --Edgars2007 (talk) 16:42, 29 April 2017 (UTC)

Every month the VIAF entries in Wikidata are exported to VIAF and as a consequence the new entries for Wikidata appear. There are also bots that seek VIAF identifiers based on other identifiers. A good example is are the Getty Identifiers.

So my question is what do you mean with reliable. VIAF is a way to bring all the entries for many authorities together in the same way Wikidata links Wikimedia articles about the same topic. Both are best efforts and both rely on people to improve the data. Over time both projects gained many more connections. What VIAF and consequently WorldCat enables us to do is link our readers to their library and read. Thanks, GerardM (talk) 16:54, 29 April 2017 (UTC)

• There are some mismatches. It's interesting to follow how it evolves: try to find people with several identifiers and/or clusters that mix different people.
  If you look through the history of the property talk page, gymel was concerned about corrections that keep feeding back and forth if people would do a regular upload. You might already have noticed that multichill's bot re-adds the same incorrect VIAF identifier several times ..
  --- Jura 17:14, 29 April 2017 (UTC)

• @Edgars2007: It is as accurate as we the users make it. I occasionally see and correct errors at or end, and, so apart from GerardM's commentary I have no idea how long it takes to fix it at their end. It would be useful to track it, however, I do so many I can never remember what I have removed and fixed to go back and check.  — billinghurst sDrewth 05:27, 30 April 2017 (UTC)

There seems to be some misunderstanding about what Wikidata items I'm talking about. I'm talking about those items, which DOESN'T have VIAF (or any other) identifier, but VIAF profile has link to Wikidata. What I mean with 'reliable'. Well, can this Wikidata link (in VIAF) be used to map unconnected VIAFs with Wikidata items, without creating a huge mess? Of course, there will be errors, but... About not making the same mistakes multiple times, see related issue. --Edgars2007 (talk) 09:19, 30 April 2017 (UTC)

I believe that you are mistaken Edgars2007 … Ivars Ijabs (Q28966945) as per the VIAF item. Have you other examples that you would like checked?  — billinghurst sDrewth 12:07, 30 April 2017 (UTC)

Ivars Ijabs (Q28966945) didn't had VIAF two days ago, when I added it. Then VIAF profile had Wikidata link. --Edgars2007 (talk) 15:17, 30 April 2017 (UTC)

• The main risk is using QID from VIAF identifiers that mix several people. If the VIAF just relates to another person, one can easily move it to the right one, but if VIAF mixes several people, there might not be a place on Wikidata for that identifier. If you limit yourself to a subset of people that you follow regularly and avoid adding twice the same identifier, I think it can work out. Personally, I stay away from VIAF for locations. There are a few other type of items that can have identifiers, but most don't add QIDs.
  --- Jura 12:16, 30 April 2017 (UTC)
  • I personally would do it only for people. --Edgars2007 (talk) 15:17, 30 April 2017 (UTC)

It's not reliable, afaik:

• Wikidata:Project_chat/Archive/2015/10#Remember the bot-import of a quarter million dutch streets? Now guess what...
• Wikidata:Bot_requests/Archive/2015/12#VIAF Identifiers for Dutch streets

Wikidata bots importing bot-created bullshit. Ping User:Gymel for current VIAF situation? --Atlasowa (talk) 10:08, 4 May 2017 (UTC)

To be honest, I take issue at your calling bullshit. First, what is it that you achieve. Second, do you appreciate what VIAF represents? Third, there is nothing that provides an avenue to progress. Now that is something others are trying very hard to achieve. Please dim it. Thanks, GerardM (talk) 16:43, 4 May 2017 (UTC

I'm not following those issues closely enough any more to speak authoritatively. There was an one-time import of Wikidata-VIAF mappings taken from what's listed in VIAF back in mid-2015 which was o.k. for persons but had to be stopped for geographic items &c. because it became a real mess (also many VIAF ids here had been imported from en:WP where in turn they had been bulk-imported by a mapping donated by VIAF). Since then VIAF has taken up importing mappings recorded here on a regular schedule, which should stabilize the situation and improve the reliability of Wikidata links contained in VIAF. However I assume that VIAF is also performing associations of Wikidata items with existing clusters on their own, i.e. by purely automatic means.

I'm not aware of any bot currently importing data (especially VIAF IDs themselves) from VIAF and would still consider that a bad idea. The same caution also must be applied to VIAF numbers found at places (BNF, GND, ...) known to be data providers for VIAF: These sources normally just reciprocate the automatic mapping of their identifiers performed by VIAF. Entering 'novalue' for VIAF ID (P214) instead of simply removing incorrect VIAF IDs is considered a good strategy since it provokes VIAF to unhinge the Wikidata item in question from a cluster it is associated with and also might prevent or unveil some cases of overly dull insertions here. -- Gymel (talk) 21:57, 6 May 2017 (UTC)

So offensive language is ok? Not answering any of the question is ok? It is all about you.. Wonderful. The WMF has a project to engage librarians and you need librarians to make a difference. A lot of engagement is going on and we need a better way to communicate with them and seek resolution about the issues there are. You do not achieve this by burning down the house. Your "good strategy" is atrocious if you want them to engage with us. Why do we need it because it is how we can people to go to libraries and read. We are closing in on having *your* library shown in WorldCat.

So please think again and help out with better ways to make a positive difference.. NB the qualiy of VIAF is certainly as good as Wikidata. Thanks, GerardM (talk) 06:36, 7 May 2017 (UTC)

HarvestTemplates-like tool that also moves image to Commons if appropriate

I am using HarvestTemplates to add images to many items, mostly from the Japanese Wikipedia.

PROBLEM: Rather than being on Commons, most images are stored on the local Wikipedia, most with no valid reason I believe. Example

I dream of a tool that would harvest templates, and would move images to Commons each time it needs, after checking the file's license/etc, and do all of the appropriate things like adding a NowCommons banner and copying author/metadata information.

Of course I would still have to fix categories manually, I guess, except if the tool comes up with a solution to this, which would be wonderful but probably too much to ask for.

If you know anything like this, please let me know, thanks :-) Syced (talk) 02:12, 2 May 2017 (UTC)

Take a look at en:w:Wikipedia:Moving files to Commons#Tools. None of these tools is perfect, but it saves some time.--Jklamo (talk) 10:50, 2 May 2017 (UTC)

I am using CommonsHelper which is the first listed there, but it takes ages to move a thousand pictures, as you have many clicks and page loads for every single picture. So I am looking for a HarvestTemplate that would perform this step automatically. Syced (talk) 05:53, 8 May 2017 (UTC)

Wikipedia article-draft-generator for Met Museum project

I'd like to share with you a Wikipedia article-draft-generator we're working on for the Met Open Access Artworks Challenge (May 15 - June 30) with the Metropolitan Museum of Art in New York City.

For an example of its use, click on any of the redlinks on this page, applied to a selection of artworks at the museum through Listeria.

We're also trying to internationalize the "Museum of Babel" functionality (w:en:Template:Mbabel, and also utilizing w:en:Template:Mbabel/MET) that this works from.

I've translated the most basic functionality from English to French (w:fr:Template:Mbabel), but that could be improved, and also would be great in other languages.

The main issue with translating the template to other Wikipedias is the variety of different Lua modules installed — I used Module:Wd to to create the original Mbabel template on English, but that one is only on 8 languages. If anyone is experienced with Wikidata2 (on 45 languages), I had created the rudimentary French version using that, and maybe the best course of action would be to recode the complete English version of Mbabel using Wikidata2, then we could use that with small modifications on the other languages.

It would be fantastic to have this in many languages, as the Met is an encyclopedic museum covering global cultures over millennia, and Wikimedians around the world will find some of their country's great artworks there.--Pharos (talk) 19:26, 6 May 2017 (UTC)

After going through some of the draft articles generated by the above module (without saving them), e.g., Madonna and Child (Q19912040)), I feel that it's promising. I don't know whether you are still planning to auto-generated more text (for e.g., dimensions, genre etc.) in the article section. Another suggestion is to link the generated infobox directly to the Wikidata and not just generating the infobox content for every article. Direct linking to Wikidata ensures up-to date information (on all wikipedias). You can check a recent conversation on this topic here. What I am suggesting is to ensure Infobox artwork... loads recent data from Wikidata. Jsamwrites (talk) 21:00, 6 May 2017 (UTC)

The draft articles do currently have genre (where it exists in Wikidata) in the main article body, but dimensions are only in the infobox, which seems to be the standard with most artwork articles. My dimensions implementation is a bit weak since dealing with units of measurement is difficult, and I would appreciate help there. The current set-up does actually have live representation of Wikidata in the infobox (though not in the article body, which is subst'ed), though perhaps there are ideas of coding that better as well.--Pharos (talk) 17:00, 7 May 2017 (UTC)

The template has been adapted in ar-WP. Any further action regarding generating the links list there? --Zack (talk) 18:29, 7 May 2017 (UTC)

@زكريا: The list is generated by , if you've used that before, it seems to be pretty popular on ar.wikipedia. An example of SPARQL code is at w:User:Pharos/Mbabel.--Pharos (talk) 17:27, 8 May 2017 (UTC)

I think I translated the Spanish version correctly. I worry about the advanced wiki coding, and the template matching so someone may want to check. --LauraHale (talk) 09:02, 8 May 2017 (UTC)

Deleted by a Spanish Wikipedia admin. They really don't take well to people with few edits creating templates. And they frown heaily on the automatic creation of articles. So yeah, it probably won't survive unless an admin creates it with the pull to keep it. --LauraHale (talk) 09:17, 8 May 2017 (UTC)

"Deleted by a Spanish Wikipedia admin. They really don't take well to people with few edits creating templates." That's awful. Do we have any experienced es.Wikipedia editors where, who can help? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:15, 8 May 2017 (UTC)

@LauraHale: Looks like you had some success in the end! Do you think we can get it working with on Spanish? Feel free to copy/adapt the SPARQL from English.--Pharos (talk) 19:22, 8 May 2017 (UTC)

Very cool! I would be very useful for readers of minor languages to be able to access all of this information in their own language (be it via article generation or off-wiki dynamic rendering, or even on-wiki dynamic rendering if that can be implemented). Syced (talk) 11:33, 8 May 2017 (UTC)

@Syced: Great to see your interest! Do you think you could help with improving the implementation on French, and maybe starting it on Japanese?--Pharos (talk) 19:18, 8 May 2017 (UTC)

Being busy on other projects I won't have much time to help, but here is a small correction: "Bachi Bouzouk noir est Peinture (d) par Jean-Léon Gérôme de 1869, en collecion de Metropolitan Museum of Art." -> "Bachi Bouzouk noir est un tableau de 1869 par Jean-Léon Gérôme, il fait partie de la collection du Metropolitan Museum of Art." Replace "du" with "de la" when starting word is a feminine noun like "Galerie". Syced (talk) 05:50, 9 May 2017 (UTC)

General "identifier" property?

I'm currently in the process of editing Braunschweigs urban districts (Stadtbezirke), which each also have a number that is refered to in official documents as "Amtliche Nummer" ("official number"). The numbers are not successive, and structured after electoral districts. Some examples: (for a full map, see [3]).

• 120: Östliches Ringgebiet
• 131: Innenstadt
• 321: Lehndorf-Watenbüttel.

However, I could not find an appropriate property to record this numbers in Wikidata. Also, I could not find any other reference to "Amtliche Nummer" in German cities, so I guess it could be a local invention. A short survey of other cities in Lower Saxony shows that either such numbers doesn't exist for other cities, or they are not recorded in Wikidata.

What is the best way to record such identifers? I would think of a general "identifier" property, but the search only results in many specialized identifiers like "FANTOIR code (FANTOIR code (P3182))", "GNS Unique Feature ID (GNS Unique Feature ID (P2326))", "UMLS CUI (UMLS CUI (P2892))" or the like.

--Rohieb (talk) 20:25, 6 May 2017 (UTC)

@Rohieb: catalog code (P528), with qualifier catalog (P972). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:53, 6 May 2017 (UTC)

Thanks, that will do! --Rohieb (talk) 13:42, 7 May 2017 (UTC)

Interwiki linking of French "Vermeil" and English "Silver-gilt" articles

Don't know if this is the right forum for this question or not, do not use Wikidata much, but happened upon a small problem: I attempted to create a new link from the English article "Silver-gilt" to the corresponding French article "Vermeil": the current French article has a link to the English article "Vermeil", which is now a redirect to "Silver-gilt" (as apparently they are the same thing, exactly). The Interwiki-ness of it all doesn't seem to be able to detect the redirect I created yesterday from the former English Vermeil article, and so when I try to link from Silver-gilt to the French article on Vermeil, I get errors. The error message I am given tells me that I am welcome to merge the topics, but performing a merge on the English Wikipedia, with which I am very familiar, is difficult enough, and the error message gives no links or information on how one might perform said merge. I tried to do it, I failed, I don't think I am going to be spending the time to find out the entire process on my own, but it needs doing and moreover the difficulty I was faced with as a well-intentioned editor should probably be looked at by a developer somewhere so that in the future when someone tries to do what I did they don't end up lost in the w-data. Thanks!! KDS4444 (talk) 13:17, 7 May 2017 (UTC)

I deleted the redirect link and merged the two items. Now fr:Vermeil links to en:Silver-gilt. Q15221792 is now a redirect to silver-gilt (Q390584). Mbch331 (talk) 15:10, 7 May 2017 (UTC)

Datatype for SPARQL query property

Should the recently created Wikidata SPARQL query equivalent (P3921) use the "URL" datatype? It's currently "string", which means a formatter URL (P1630) won't work. Please see Property talk:P3921#Data type. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:17, 7 May 2017 (UTC)

Interesting property, but I have not heard about it yet. I need some input:

• What's the intended use according to the property proposal, briefly summarized?
• How exactly would an example with formatter URL (P1630) look like, if it had URL datatype? Would the Wikidata SPARQL query equivalent (P3921) values be the same? Please provide an example.
• On the other hand, what's the advantage of the current situation with String datatype?

Thanks, MisterSynergy (talk) 18:28, 7 May 2017 (UTC)

The property proposal is at Wikidata:Property proposal/Wikidata SPARQL query equivalent; it includes an example. With the formatter URL which I recently removed, that would resolve to https://query.wikidata.org/#?item wdt:P31 wd:Q5. ?item wdt:P27 wd:Q1036 - which is an invalid query URL, Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:41, 7 May 2017 (UTC)

Okay thanks. If I get it right, we have two options:

• String datatype (as it is): value is ?item wdt:P31 wd:Q5. ?item wdt:P27 wd:Q1036; a possible formatter URL could be: https://query.wikidata.org/#SELECT%20%3Fitem%20%7B%20$1%20%7D (not sure whether this would work, not tested; maybe URL encoding is a problem)
• URL: value would have to be a full link to the query service, thus https://query.wikidata.org/#SELECT%20%3Fitem%20%7B%20%3Fitem%20wdt%3AP31%20wd%3AQ5.%20%3Fitem%20wdt%3AP27%20wd%3AQ1036%20%7D

If I had to choose between these two, I would prefer the String solution, just as it is. This is much more flexible e.g. for use in Wikipedia templates, although template programmers have to build the queries by themselves. (If I overlooked anything important here, please let me know; Thanks!) —MisterSynergy (talk) 19:23, 7 May 2017 (UTC)

Link to Query Service, as it could be generated by a Wikipedia template: Query Service (see source for details). Works with this simple example. Experienced template programmers can probably still improve it a lot. —MisterSynergy (talk) 19:32, 7 May 2017 (UTC)

As I noted above; formatter URLs don't work with "sting" data type. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:03, 7 May 2017 (UTC)

• There are 50 properties with String data type and a formatter URL.
• Even if it did not work: at least Wikipedias can create links with String data type, as it appears, and this seems to be the main motivation for this property. With URL datatype we would have to set full query service links as values, which makes it effectively impossible to use the values of this property with any other query tool than our Query Service.

I am still in favor of the current situation. —MisterSynergy (talk) 20:18, 7 May 2017 (UTC)

Using this formatter URL, the demo item Category:Ugandan writers (Q7439502) has a functional link to the query service provided by the formatter URL (P1630) property. —MisterSynergy (talk) 20:27, 7 May 2017 (UTC)

• P1630 was made for string datatype, so obviously it works.
  --- Jura 05:17, 8 May 2017 (UTC)
  • Jura refers to formatter URL (P1630). As can be seen from the absence of a link on Category:Ugandan writers (Q7439502), his claim is false. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:07, 8 May 2017 (UTC)
    • If you see a link on Category:Ugandan writers (Q7439502) or not depends if you have activated the authority control gadget in your account settings. --Pasleim (talk) 14:19, 8 May 2017 (UTC)
    • Which link are you missing on Category:Ugandan writers (Q7439502)? In my browser, the P3921 value of this item links to the Query Service, and the link is generated based on the P1630 value in Property:P3921. You should check your Wikidata settings, as well as your activated javascript files for errors. —MisterSynergy (talk) 14:21, 8 May 2017 (UTC)
      • That's odd; I see a link when logged out, but not when logged in. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:23, 8 May 2017 (UTC)
        • Javascript stuff was recently updated, and they finally dropped support for some deprecated functions. You need to check your javascript user scripts for those. Your browser console might help, but this is always a little tricky. —MisterSynergy (talk) 19:43, 8 May 2017 (UTC)

What property to use to list previous names for a parliamentary/legislative committee?

Is there an ideal property for a parliamentary or congressional committee to capture all the previous names it's been known as over time? For example, a US congressional committee might have been created in 1795 [4] under the name "Committee on Commerce and Manufactures", but its name has changed a few times to "Committee on Commerce," and "Committee on Interstate and Foreign Commerce," to "Committee on Energy and Commerce" today. The item right now United States House Committee on Energy and Commerce (Q1560114) doesn't capture the name and the time spans for those previous titles. I cannot locate a Wikidata item that models something quite like this. Properties nickname (P1449) and replaces (P1365) don't seem like the right solution. Any pointers or recommendations appreciated. Thanks. -- Fuzheado (talk) 00:00, 8 May 2017 (UTC)

@Fuzheado: How about adding official name (P1448) with qualifiers start time (P580) and end time (P582)? Mahir256 (talk) 01:36, 8 May 2017 (UTC)

@Fuzheado: I agree with Mahir256 if it is a close match for the committee, however, often you can find the scope or terms of reference has significant difference between the old and committee, so there I would create separate items, and then create one as preceding, and the other as succeeding each other, and give start and end dates for the structures.  — billinghurst sDrewth 04:09, 8 May 2017 (UTC)

@Mahir256: @billinghurst: - Thanks for the ideas, and it looks like the official name (P1448) with start and end times looks like the ideal solution. For this particular case, where the committee description for United States House Committee on Energy and Commerce (Q1560114) claims it is the longest one in existence, it makes sense to have one Wikidata item representing the continuous, but relabeled, committee. But I agree if it is clear it is a different committee in scope and not a continuation of a previous committee, it should have a different item. Thanks. -- Fuzheado (talk) 04:52, 8 May 2017 (UTC)

FYI, here's an example of an item I've modeled using the recommendation above - United States Senate Committee on Small Business and Entrepreneurship (Q499076) -- Fuzheado (talk) 06:25, 8 May 2017 (UTC)

Matching entities from DBpedia to Wikidata

I need to match the entities between DBpedia to Wikidata that is, say I have retrieved a list of Musicians name with some additional information about them in DBpedia. How can i use that DBpedia data to retrieve other information about the same list of Musicians in Wikidata?  – The preceding unsigned comment was added by Psashwini (talk • contribs) at 09:17, 8 May 2017‎ (UTC).

DBpedia identifiers are based on the Wikipedia sitelinks, which are recorded in Wikidata. Also, DBpedia contains mappings to Wikidata. So from either side, it should be straightforward to access the other. Where are you struggling? --Denny (talk) 15:40, 8 May 2017 (UTC)

@Denny: Thanks for the reply. Can you tell me an example to access DBpedia from Wikidata? And also how to get the additional information in WIKIDATA for the same list of musicians which i have retrieved from DBPedia (Using SPARQL query) ?

INIMATI

Hi i want to create a page but i dont know how to use

 – The preceding unsigned comment was added by StudioInimati (talk • contribs) at 13:31, 8 May 2017 (UTC).

@StudioInimati: Special:NewItem lets you define the label (= "name") of the subject item and a description, both in your preferred language which you can edit via Special:Preferences. Be sure to check, however, that the page you want to create meets the notability standards. Jonathan Groß (talk) 11:57, 8 May 2017 (UTC)

Beta Feature Two Column Edit Conflict View

From May 9, the Two Column Edit Conflict View will be available as a beta feature on all wikis. The Two Column Edit Conflict View is a new interface for the edit conflict resolution page. It highlights differences between the editor's and the conflicting changes to make it easy to copy and paste pieces of the text and resolve the conflict. The feature fulfils a request for a more user-friendly edit conflict resolution from the German Community’s Technical Wishlist. Everyone is invited to test the feature and we hope that it will serve you well!

Birgit Müller (WMDE) 14:41, 8 May 2017 (UTC)

Wikidata weekly summary #259

Here's your quick overview of what has been happening around Wikidata over the last week.

• Discussions
  • New request for comments: Updating References for External Data

• Events/Press/Blogs
  • Registration is open for the Celtic Knot Conference 2017 (July 6th, Edinburgh)
  • Pre-hackathon of the Wikimedia Hackathon, May 12th-14th in Prague
  • Wikimedia Hackathon 2017, May 19th-21st in Vienna. On site or remotely, you can join the Wikidata documentation sprint

• Other Noteworthy Stuff
  • Vote in the board election (see also Voter guides and rationales
  • Sitelinks for the new Doteli Wikipedia (dtywiki) can be added
  • You can now track the number of identifier properties for an item
  • We provide weekly a new type of database dumps: nt truthy dumps
  • Wikipedia article-draft-generator developed by Pharos for Met Museum project
  • Schema.org proposes to encourage the use of Wikidata as a common entity base for the target of the sameAs relation

• Did you know?
  • Newest properties: light sector, Wikidata SPARQL query equivalent, Canadian Coastguard Lighthouse ID, contributed to published work
  • Query examples:
    • All places (with coordinates) that gave its name to something (source)
    • Paintings depicting monkeys (source)
    • Artworks with presumably depicted persons (source)
  • Newest database reports: list of oldest lighthouses

• Development
  • Worked more on support for lexicographical data (specifically support for lexicographic category and statements on forms)
  • Dealt with issues after the deployment of automatic interwiki links for Wiktionary
  • Worked on constraint violation user script based on your feedback (improving messages to make it more understandable, layout, etc)
  • Add support for the following new language codes for monolingual text properties: brx, chn, cop, gez, quc, kjh, nr (will become available in a few days)

You can see all open tickets related to Wikidata here.

• Monthly Tasks
  • Add labels, in your own language(s), for the new properties listed above.
  • Comment on property proposals: all open proposals - proposals needing attention
  • Suggested and open tasks!
  • Contribute to a Showcase item.
  • Help translate or proofread the interface and documentation pages, in your own language!
  • Help merge identical items across Wikimedia projects.
  • Help write the next summary!

Read the full report · Unsubscribe · Lea Lacroix (WMDE) 14:57, 8 May 2017 (UTC)

biologically motion (Q29373587) and locomotion (Q22269718)

@Fractaler:

I think that biological process (Q2996394) is correct: covers both individuals and populations.

Therefore item Q22269718 represented both individuals and groups.

The most extreme case is when swarming animals (e.g. bees or ants) behave individually or as group.

If we are about to change classification, we should start with Q2996394 and remove link to Q2625603 IMO. d1g (talk) 21:15, 25 April 2017 (UTC)

About biological process (Q2996394), description: "collection of molecular events ...". So, description of biological process (Q2996394) is: 1) "process...within the individual" or 2) "process...within the individual + between individuals"? --Fractaler (talk) 09:58, 26 April 2017 (UTC)

Probably some sitelinks should be moved from animal locomotion (Q925958) (treated as animal locomotion) to locomotion (Q22269718) or both should be merged and a new item for animal locomotion should be created. --Succu (talk) 13:38, 26 April 2017 (UTC)

Wikidata isn't ready yet to describe "between individuals" part in detail. We need participants property at very lest to describe processes.

My suggestion is to use 1. one item for individuals and 2. other for groups.

Herding information can be very complex and not well-studied as processes within one organism.

Movement information belongs to individuals for the most part, other part is rather complex or specific (kangaroo inside pouch, a parasite within animals) d1g (talk) 04:59, 28 April 2017 (UTC)

Yes, we have individual and group (of individuals). But what is individual? superorganism (Q916139) is group (of individuals)? Consist of individuals, =individual? Of course, "herding step" is not same as "cattle (Q830)'s step". relocation (Q2918584)? About parasite+animals: it is a passive moving. --Fractaler (talk) 10:41, 28 April 2017 (UTC)

I'm not sure what biologically motion (Q29373587) is meant to cover, since "biologically motion" is not even a grammatical phrase in English. Is the item intended to cover any kind of movement associated with a biological origin (such as blinking, the unfolding of flowers), or is the item concerned only with movements that change the location of an individual, or only movements initiated by an individual to change its location, or only movements initiated by an individual to change its location in response to specific stimuli? All of these have different biomechanical and physiological terminology applied to them. --EncycloPetey (talk) 21:18, 26 April 2017 (UTC)

physical model (Q12149837): linear motion (Q2035846). For organism (Q7239) we have: active motion (Q29374366)/passive motion (Q29374520) --Fractaler (talk) 14:16, 27 April 2017 (UTC)

We also have a travel (Q61509) "travel" "movement of people between relatively distant geographical locations" as subclasses of Q79782 and Q1914636. d1g (talk) 04:35, 28 April 2017 (UTC)

Tele-tourism/tele-travel, three-dimensional virtual tourism (Q7797256), virtual tour (Q2915546), etc. are "travel"? --Fractaler (talk) 08:12, 28 April 2017 (UTC)

I would define uses in telecommunication, entertainment, education, but not as active/passive movement.

I see them as alternatives to traditional travel, not as subclasses. d1g (talk) 16:11, 1 May 2017 (UTC)

also: human migration (Q177626) - "movement by people from one place to another" --Fractaler (talk) 13:31, 4 May 2017 (UTC)

We have the change (active or passive) of a location of an individual biological object. opposite of (P461) - localization (like localization (Q22288298) for "microtransporting", "micro passive moving")? Fractaler (talk) 13:31, 9 May 2017 (UTC)

Proposing properties for highly-used catalogues

This query shows the most-used catalogues used as qualifiers in catalog code (P528), with qualifier catalog (P972). We have:

• ten with over 5,000 uses
• another 14 with over 2,000 uses
• another 7 with over 1,000 uses

I am minded to propose properties for each of these, then ask for a bot to migrate the IDs. That way, a formatter URL (P1630) can be applied. Any thoughts, before I do? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:25, 7 May 2017 (UTC)

Please don't do the painting ones. These tend to be very paper and no place to link to. No added value and having some special cases makes it harder to work with. Multichill (talk) 20:26, 7 May 2017 (UTC)

There is no requirement that external-id properties be linkable. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:11, 8 May 2017 (UTC)

Is there a purpose served in having unlinkable external ids rather than catalog codes? ArthurPSmith (talk) 15:52, 8 May 2017 (UTC)

The same purposes for which we create such properties from new. For example, easier lookup; such as with tools like Resolver. Also being able to run constraint reports to trap malformed, multiple and duplicate values. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:51, 9 May 2017 (UTC)

Interesting idea to make a property! I never thought of that. I create artist catalogues and you can see most of the ones I created and completed showing up in your query. I would be interested in two properties, one for John Smith (the original inventor of the catalog or catalogue raisonné) and his updater a half-century later, Hofstede de Groot. I made an overview of Hofstede de Groot here: Wikidata:WikiProject sum of all paintings/Hofstede de Groot and en:User:Jane023/Hofstede de Groot. Jane023 (talk) 06:03, 8 May 2017 (UTC)

Here's the first: Wikidata:Property proposal/Principal Galaxies Catalogue ID. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:37, 9 May 2017 (UTC)

Female sports

We do have women's sports (Q920057) and women's association football (Q606060), but we kind of miss the male equivalent. In most cases (but not in all), association football (Q2736) will be the male representation, but it seems to me that we miss an overall item to have distinction between male, female and <all> soccer. The same might/will be the case for other sports. Would it make sense to quickly fix this gender-gap, as female sport is quickly gaining importance, or should we simply drop the female part and just use one item for all? For myself, I think it will make sense to separate the two, but only when we are doing it consequently. Edoderoo (talk) 15:42, 9 May 2017 (UTC)

• Use association football (Q2736) for all: it’s the same type of sport, with different competition classes (men's association football, women's association football, etc.). We typically don’t distinguish again between genders in claims. Example: actor (Q33999), but not actress (Q21169216) as occupation (P106) values.
• Those values are often used with sport (P641), and this leads to a lot of redundant information in sports-related items. I try to keep P641 usage in shape, but the vast amount of redundant claims makes this a tedious task, thus I plan to propose usage changes for P641 anyway (again).

—MisterSynergy (talk) 17:04, 9 May 2017 (UTC)

Averaged annual precipitation

Hello. Do we have a property for averaged annual precipitation? Is that the precipitation height (P3036)? And how to use it? Xaris333 (talk) 15:16, 2 May 2017 (UTC)

From what I can see that property should be used with events like storms or blizzards. I believe tabular data (to be deployed) will help us with this kind of data. Matěj Suchánek (talk) 16:53, 6 May 2017 (UTC)

I don't think it's a good idea to host this data as tabular data. Tabular data isn't queriable in the same way and statistics like these seem useful to query. ChristianKl (talk) 11:51, 10 May 2017 (UTC)

Software version

I think it is important to mention only the current release.version.update.patch See e.g. TP of FreeBASIC where I removed all old (prolly buggy) versions of this compiler. Does not make much sense, does it?  Klaas `Z4␟` V:  15:35, 2 May 2017 (UTC)

I reverted and let a message on Talk:Q195929. If someone remembers where is the discussion about the interest of specify "obsolete" values, vould you point to it? Thanks in advance. Pamputt (talk) 17:59, 2 May 2017 (UTC)

Please keep all versions. There are people interested to know when one specific version was released. Set the rank of the most recent version to "preferred" and the rank of all other versions to "normal". And please don't write 'obsolete' behind the version number. --Pasleim (talk) 20:20, 2 May 2017 (UTC)

Rather than reverting you better tell me before. BTW the property that was not used in any version (so I did not know) is called 'version type' and for the content one can use e.g. 'current' or 'deprecation' I found out. Perhaps 'beta' could be an ideaa to add as well. I think it is wise to standardize these words a bit. There are people who love to use them and sometimes they are mentioned on some Wikipedia language versions.  Klaas `Z4␟` V:  13:10, 3 May 2017 (UTC)

The property description seems clear to me: "version(s) of the software, current and past". Given that it explicitly includes past versions, there's no reason to expect it to only contain the current version. It would therefore be also wrong to deprecate old versions. It's the general habit in cases like this to use the "preferred rank". ChristianKl (talk) 22:24, 9 May 2017 (UTC)

Query service URLs

The Query service's URLs, viewed in Firefox, at least) have spaces in, for example:

https://query.wikidata.org/#SELECT %3Fitem %3FitemLabel %3Fvalue %3FvalueLabel%0A{%0A%09%3Fitem wdt%3AP3885 %3Fvalue .%0A%09SERVICE wikibase%3Alabel { bd%3AserviceParam wikibase%3Alanguage "en%2Cen" } %0A}%0ALIMIT 1000

making it difficult to use them in wiki links, email, etc.

1. Is there a simple work-around?
2. Can this be fixed; or an easily copyable version of the link be provided, in the same manner as the short URL pop-up?

-- Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:12, 4 May 2017 (UTC)

When I copied the code above and ran the query once, it changed to a link without spaces (in Google Chrome). It made https://query.wikidata.org/#SELECT%20%3Fitem%20%3FitemLabel%20%3Fvalue%20%3FvalueLabel%0A%7B%0A%09%3Fitem%20wdt%3AP3885%20%3Fvalue%20.%0A%09SERVICE%20wikibase%3Alabel%20%7B%20bd%3AserviceParam%20wikibase%3Alanguage%20%22en%2Cen%22%20%7D%20%0A%7D%0ALIMIT%201000 out of it. And made automatically a link of it, without any braces put around it... Q.Zanden questions? 14:38, 4 May 2017 (UTC)

Thank you, but the above is also how it looks in Firefox after it has run. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:48, 4 May 2017 (UTC)

Other than switching browsers, does anyone have a solution? What about the suggestion in my second bullet point? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:30, 10 May 2017 (UTC)

Using Wikidata in location maps on enwp?

Continuing my work with infoboxes, I've been trying to embed location maps in telescope articles by using coordinate location (P625) and country (P17) to identify both the location and the relevant map to use. It works quite nicely at my canonical example of South Pole Telescope (Q1513315). However, this doesn't work in some cases, such as Arecibo Radio Telescope (Q44547), where the "United States of America" map doesn't show Puerto Rico. Although P17 could be modified to say "Puerto Rico", that doesn't help as that isn't the correct information (since it's a territory, not a country). Using located in the administrative territorial entity (P131) is tricky, since location maps don't necessarily exist for each administrative area.

Following from discussion on enwp, it sounds like there may be structural issues here, hence this message. Comments on this issue would be appreciated. :-) Thanks. Mike Peel (talk) 03:01, 5 May 2017 (UTC)

can you make an exception for the United States, and use P17 everywhere else? ArthurPSmith (talk) 13:03, 5 May 2017 (UTC)

Could you code it to look for a map from P131 first, and if no map available fall back on P17? That would be something, at least. On a different note, what code are you using to select a map for Antarctic locations? I note the SPT article still has "continent: Antarctica", but P131:Antarctic Treaty area (Q21590062) is probably a more reliably used one. (Or just "anything with coordinate below 60S" would work... as a hardcode) Andrew Gray (talk) 11:31, 6 May 2017 (UTC)

@ArthurPSmith, Andrew Gray: Liverpool Telescope (Q6658589) has two values for P131, one covering the island and one covering the part of the island that the telescope is in. In that case it's not easy to automatically know which one is the correct one to show. Is there a good way to clearly say "Canary islands" or "Hawaii" etc. using another property? Thanks. Mike Peel (talk) 22:01, 9 May 2017 (UTC)

List

Is there any list of mainspace articles of Hindi Wikipedia which don't have links for English Wikipedia? Capankajsmilyo (talk) 06:19, 7 May 2017 (UTC)

@Capankajsmilyo:

SELECT ?item ?page ?url WHERE {
  ?item ^schema:about ?url .
  ?url schema:isPartOf <https://hi.wikipedia.org/>; schema:name ?page .
  MINUS { ?item ^schema:about/schema:isPartOf <https://en.wikipedia.org/> } .
} ORDER BY ?page

Try it!

Matěj Suchánek (talk) 12:46, 10 May 2017 (UTC)

Thanks a lot @Matěj Suchánek: Capankajsmilyo (talk) 13:13, 10 May 2017 (UTC)

How to add patent number to person

This person - Avner Halperin (Q29859312) was developer (or co-developer) of several patents see list. We have patent number (P1246) to use when it is an article about the patent but how cal I add the list of patent numbers to Avner Halperin (Q29859312)? does patent number (P1246) qulify also for persons like someone use in Reem Al Marzouqi (Q19648874)? Geagea (talk) 01:14, 8 May 2017 (UTC)

The patent as it applies to an object; or the patent as a written application. If the former, as the patent is for an object or a process, wouldn't that patent detail belong on the object alone, and utilise the creator (P170) tag for the person as creator of the object? I wouldn't have thought that a person is assigned a patent number, as a patent can be a commercial object, inheritable and all other components as a product. For the latter use as the written patent application, which would presumably have an author, or other detail which is related.  — billinghurst sDrewth 03:51, 8 May 2017 (UTC)

Also noting that it would seem reasonable to link to the object/process from the creator that you can use notable work (P800) which would seem applicable. We may need to look at the guidance for P800 to ensure that the explanatory text aligns.  — billinghurst sDrewth 04:06, 8 May 2017 (UTC)

We do have an identifier for invention of a person. So maybe it is a good idea to make a new property: inventor of patents(?) that will be fill with with the list of US patent inventions the person invent. The page www.google.com/patents/US????? includs a lot's of data about the patent.Geagea (talk) 23:06, 8 May 2017 (UTC)

@Geagea: Use Justia Patents inventor ID (P3874). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:45, 9 May 2017 (UTC)

Thanks @Pigsonthewing:. Looks ok. Geagea (talk) 05:16, 10 May 2017 (UTC)

Clicking Wikidata SPARQL query equivalent (P3921) on list of people with the family name Armstrong (Q20007426) not neat query result

Hi. The addition of queries to items is interesting, and when clicking the query link on list of people with the family name Armstrong (Q20007426) it wraps the query inside another query at WQS. Is that the actual intent? Or should the query be able to be run without modification? Or is there something amiss with the how query has been inserted? Yours cluelessly ...  — billinghurst sDrewth 22:31, 8 May 2017 (UTC)

Not to worry, I worked it out, someone had added SELECT to the query. Have we a constraint validation to check for queries starting with SELECT?  — billinghurst sDrewth 23:04, 8 May 2017 (UTC)

Correct-ish, though my solution fails. Still requires resolution.  — billinghurst sDrewth 23:06, 8 May 2017 (UTC)

There is a formatter URL (P1630) that triggers the query in a query. Mbch331 (talk) 11:15, 9 May 2017 (UTC)

Fixed; the query string given should not be a full SPARQL query for the query service, just the defining part which selects equivalent items with SPARQL. There’s an example on Property:P3921. —MisterSynergy (talk) 11:19, 9 May 2017 (UTC)

Umm :-/ pretty unhelpful result for 99% of the population. Those who know who to sparkle with SPARQL would have probably done it with their eyes closed; those who don't will presumably find the output of the item pretty dick useless if that is the result. Whatever rocks your boat.  — billinghurst sDrewth 11:31, 9 May 2017 (UTC)

AFAIK this property was meant to be used in Wikipedia category pages. One could easily create a template or even a gadget that includes all the relevant missing parts of the query, including localization, comments, etc. This way much more flexible queries can be generated from this data. —MisterSynergy (talk) 11:42, 9 May 2017 (UTC)

Maybe we can find a formatter url that produces a query that also shows the labels? ChristianKl (talk) 12:48, 9 May 2017 (UTC)

As people have said above, the idea is to get templates on the different projects to do most of the heavy lifting -- for example c:Template:category contains on Commons, which pulls up images, commons categories and co-ordinates based on a string like this. (Example pages using it).

Having just the Sparql specification on the wikidata item means that the query can be adapted to show relevant fields for the project, in relevant languages. (Also that the query can be worked on and improved, without having to re-write any of the data).

But it would be easy enough to add whatever fields anyone wants to the default URL-formatter. (Note that changes to the URL-formatter may take some time -- weeks even -- to propagate through to an item, if the item hasn't been edited in the meantime). Jheald (talk) 13:00, 9 May 2017 (UTC)

There are many categories and lists that have a human readable property that can be easily interpreted by software. It is used for instance in Reasonator to show the content of those lists/categories. They are based on "is a list of" and the qualifiers define what there is to be had. So when info does not work well like this does, get something with an equivalent functionality that does work. Thanks, GerardM (talk) 05:12, 10 May 2017 (UTC)

I’ve changed the formatter URL (P1630) to include labels and descriptions as well – good point. I’ve also added a comment and a {{Complex constraint}} to Property talk:P3921 to check whether the query strings given in items that use Wikidata SPARQL query equivalent (P3921) actually starts with ?item. Experienced users might want to look at this new constraint. —MisterSynergy (talk) 13:30, 9 May 2017 (UTC)

Thanks @MisterSynergy, Jheald: some of us are neanderthal man when it comes to SPARQL; and possibly learning how to rub sticks together to get fire with data extraction from WD. [I now understand what and why, though not how to implement it, or the clear benefits for the wiki.] So all help and explanation is valuable. For the sister and smaller wikis where there are less coders, and less critical mass, looking to parcel this up, as some sort of outreach, and how to reap the benefits is something that would be great to be considered.  – The preceding unsigned comment was added by Billinghurst (talk • contribs) at 01:32, 10 May 2017‎ (UTC).

The implementation in a wiki project basically consists of two parts: crafting a template (requires some SPARQL knowledge) and adding the template parameterless to a category page or list page (no knowledge required). Smaller projects without any editor to set the template up can ask here at Wikidata for help.

I also plan to have one of these Property Help Pages for P3921 soon, which explains how the values should look like, and how data users can finally use the data again. Unlike many others, this property is not self-explanatory and needs detailed instructions. However, there is a dispute right now on Property talk:P3921#?item about correct usage (full queries vs. query fragments) which I’d like to have resolved first. —MisterSynergy (talk) 05:25, 10 May 2017 (UTC)

Adding an LOC heading

I added the Library of Congress Subject Heading "financial statements" to https://www.wikidata.org/wiki/Q192907. Did I do it the right way? 208.95.51.38 14:16, 9 May 2017 (UTC)

No. I'm actually wondering why you think a "Library of Congress Subject Heading" is the same as a "Library of Congress ID for authority control for persons and organizations". Sjoerd de Bruin (talk) 14:59, 9 May 2017 (UTC)

Library of Congress authority ID (P244) is definitely wrong. There is also Library of Congress Classification (P1149) which may be closer to what you are looking for (but I don't think it would be the string "financial statements" either way). ArthurPSmith (talk) 16:02, 9 May 2017 (UTC)

At least according to the Wikidata property example (P1855) at the property page of Library of Congress authority ID (P244) (dialect (Q33384)) you can use Library of Congress authority ID (P244) for Library of Congress Subject Headings. But you have to use the control number (in this case sh85048313), not the subject heading itself. -Valentina.Anitnelav (talk) 21:16, 9 May 2017 (UTC)

I've been adding over time dozens of LOC Subject headings using Library of Congress authority ID (P244) and I don't see a reason to stop doing it. sh numbers are legit values according to the talk page. We should just change definition to "Library of Congress ID for authority control for persons, organizations and subject headings". DGtal (talk) 09:08, 10 May 2017 (UTC)

How important should Wikidata item of this property (P1629) considered to be?

There are currently a bunch of constraints on different properties that suppose that every property should have Wikidata item of this property (P1629). If we consider this as being of high importance, should we require property proposals in general to fill the corresponding information before we create the property? ChristianKl (talk) 09:38, 10 May 2017 (UTC)

No. Simply create an item representing the subject, at the time of property creation, if none already exists. Otherwise, we'd have items for rejected properties, which may not be sensible. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:08, 10 May 2017 (UTC)

In most cases the concept that a property is supposed to describe is well defined enough to be a clearly identifiable conceptual or material entity and thus notable enough to be it's own item even if we decide it's not important enought to have it's own property. ChristianKl (talk) 13:45, 10 May 2017 (UTC)

In most cases, perhaps, but you asked about requiring it in general. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:39, 10 May 2017 (UTC)

If there are 5 item created in the span of a year that aren't notable, I don't think that's a problem. On the other hand it's a benefit if the person who creates the subject item tries to create a subject item that would be notable even without the property. ChristianKl (talk) 17:09, 10 May 2017 (UTC)

Constraints for date of birth (P569)?

Can someone look at how David Ronfeldt (Q11855858) triggers constraints for date of birth (P569)? What's wrong, should the constraints be different? ChristianKl (talk) 13:32, 10 May 2017 (UTC)

Those are leftovers of experimental constraints. They have been removed, but the database will be updated soon. Sjoerd de Bruin (talk) 13:45, 10 May 2017 (UTC)

Merges of different classes of item

I've just undone a merge of Telipogon diabolicus (Orchidaceae, Oncidiinae), a new species from southern Colombia (Q27048467) and Telipogon diabolicus (Q25839916) - the former is a scholarly article (Q13442814), the latter a taxon (Q16521). Do we need a filter that prevents, or at least warns about, merges of items with incompatible instance of (P31) values? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:36, 5 May 2017 (UTC)

This sort of merge can easily spotted via the existing constraint violations. --Succu (talk) 19:57, 5 May 2017 (UTC)

Indeed it can - in some cases. However, that requires someone to fix it; leaves it broken in the interim; and does not educate the original editor. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:12, 5 May 2017 (UTC)

For this domain it's standard procedure to fix constraint violations as soon as possible. And I don't think an user, like Orchi, who makes a single mistake needs to be „educated“. --Succu (talk) 20:48, 5 May 2017 (UTC)

Constraint reports are not updated in real time. Where did I say anything about a single domain; or about educating Orchi, or any other named individual? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:53, 5 May 2017 (UTC)

You noticed an error regarding an item you created and made a generalization. I simply responded to your example. I don't think working along real-time (Q347066) is something Wikidata is designed for. --Succu (talk) 21:07, 5 May 2017 (UTC)

Why is the identity of the creator of an item relevant? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:10, 5 May 2017 (UTC)

It isn't, but it's how you noticed this error. So could you give please other examples to show this is a serious problem we should act on. --Succu (talk) 21:18, 5 May 2017 (UTC)

My question was "Do we need a filter that prevents, or at least warns about, merges of items with incompatible instance of (P31) values?". Anyone care to address that? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:00, 7 May 2017 (UTC)

what do you consider incompatible? We have a lot of items that are both a building and an organization (hospitals and museums, for instance) which seems to me a reasonable thing to do but they are quite different classes in principle. I think our constraint violation system is the way to handle this along with other problems - the new gadget to show violations should be helpful for merges too I would think. ArthurPSmith (talk) 15:40, 8 May 2017 (UTC)

If someone has taken the trouble to make separate items for a building and an organisation, I don't think think we should be merging them. Contsraint violations help us to fix errors after the fact, not to prevent them. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:29, 10 May 2017 (UTC)

I agree that it's preferable to distinguish a building from the organisation and that items like this shoudn't be merged. ChristianKl (talk) 22:11, 10 May 2017 (UTC)

When you consider an organisation that is designed by an architect then anything goes. This combination is a Wikipedia artefact and wrong in our context. Thanks, GerardM (talk) 05:37, 11 May 2017 (UTC)

Embedded queries

Given a query page like :

https://query.wikidata.org/#SELECT%20%3Fitem%20%3FitemLabel%20%3FitemDescription%20%7B%20%3Fitem%20wdt%3AP31%20wd%3AQ5%3B%20wdt%3AP27%20wd%3AQ1036%3B%20wdt%3AP106%20wd%3AQ36180.%20SERVICE%20wikibase%3Alabel%20%7B%20bd%3AserviceParam%20wikibase%3Alanguage%20%27en%2Cde%2Cru%2Cfr%2Ces%2Cit%2Cja%2Czh%27%20%7D%20%7D%0A

and the "embedded" version:

https://query.wikidata.org/embed.html#SELECT%20%3Fitem%20%3FitemLabel%20%3FitemDescription%20%7B%20%3Fitem%20wdt%3AP31%20wd%3AQ5%3B%20wdt%3AP27%20wd%3AQ1036%3B%20wdt%3AP106%20wd%3AQ36180.%20SERVICE%20wikibase%3Alabel%20%7B%20bd%3AserviceParam%20wikibase%3Alanguage%20'en%2Cde%2Cru%2Cfr%2Ces%2Cit%2Cja%2Czh'%20%7D%20%7D%0A

is there a tool to toggle between the two views (especially from the former to the latter), without hacking the URL (which may be daunting for novice users)? Could a button to do this be added? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:26, 10 May 2017 (UTC)

On the embedded version, there is an icon and a link in the bottom left of the page. — Envlh (talk) 18:02, 10 May 2017 (UTC)

So there is; thank you. What about in the other direction? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:32, 10 May 2017 (UTC)

Would be useful. Please request it on Wikidata:Contact the development team. —MisterSynergy (talk) 21:55, 10 May 2017 (UTC)

Tracked in Phabricator
Task T151057

See phab:T151057 for the Discovery team's thoughts on more systematic presentation of icons and buttons to switch between different output modes. Jheald (talk) 22:13, 10 May 2017 (UTC)

New filters for Recent Changes - Beta deployment schedule for your wiki

Hello!

(Sorry to write in English for non English-speakers. Please help translate to your language and also inform everyone about this change, thanks!)

The Collaboration team is going to launch a new Beta feature on your wiki, ⧼eri-rcfilters-beta-label⧽. This deployment would happen on May 9 (hour to be precised). This Beta feature is an improvement of the current ORES Beta feature. We have already announced those filters previously, but ORES predictions needed some ajustements for your wiki and we have preferred to reschedule to make them better.

What it this new feature?

This feature improves Special:RecentChanges and Special:RecentChangesLinked by adding new useful features that will ease vandalism tracking and support of newcomers:

• Filtering - filter recent changes with easy-to-use and powerful filters combinations.
• Highlighting - add a colored background to the different changes you are monitoring, to quickly identify the ones that matter to you.
• Quality and Intent Filters - user ORES predictions to identify real vandalism or good faith intent contributions that need assistance.

What will happen?

At the moment, the ORES beta feature is available in your Beta preferences. We the deployment will be done:

• ORES predictions highlighting is activated by default for all users (not in Beta anymore).
  • It is symbolized by a "r" in all RecentChanges pages, Special:Contributions, Special:Watchlist, for all users.
  • Users who want to change the accuracy level of ORES predictions or opt-out can do so in their preferences, in the "Watchlist" tab.
• Previous ORES Beta users will have a highlight of likely problem edits with colors. It will be possible to disable it in your preferences, in the "Watchlist" tab as well.
• Anyone who wants to have the new filters have to enable them in their preferences, in "Beta" tab. Other users have nothing to change.

How to prepare this change?

You can discover the purpose of this project by visiting the quick tour help page. Also, please check the documentation (and help to translate it). Please also have a look at the translations of the interface on translatewiki.net. The messages are prefixed as rcfilters- and ores-rcfilters-.

This feature is already available on most wikis. To see how the prediction filters based on ORES, you can enable them on English, Portuguese or Polish Wikipedia.

Please ping me if you have questions. :)

All the best, Trizek (WMF) (talk) 09:36, 3 May 2017 (UTC)

The deployment will be done on May 9 at 23:00 UTC. Trizek (WMF) (talk) 12:03, 4 May 2017 (UTC)

Click on "no thanks" on the invitation is not working

Tracked in Phabricator
Task T165048
Resolved

@Trizek (WMF): I've been getting this message all day, yet despite clicking "No, thanks" it comes back on every re-load of the page. This is terribly frustrating. Furthermore, I can't access my preferences. What's going on? Jared Preston (talk) 15:02, 11 May 2017 (UTC)

I'm reporting this bug concerning the message, Jared Preston - however, I haven't experienced it: I've clicked on "no thanks" and it is gone for good. Which browser do you use? It may be linked to the different problems related to the Preferences one you mention (already reported, see on the most recent section of this page). Sorry for the disturbance. Trizek (WMF) (talk) 15:05, 11 May 2017 (UTC)

Thanks for the quick reply; I guess it is related. I use Pale Moon. Jared Preston (talk) 15:08, 11 May 2017 (UTC)

I discover Pale Moon with your reply, Jared Preston! :)

I got the confirmation it is related to the Special:Preferences bug, and we are actively working on it. Trizek (WMF) (talk) 15:12, 11 May 2017 (UTC)

Can you retry, Jared Preston? Trizek (WMF) (talk) 15:59, 11 May 2017 (UTC)

Wonderful, it worked! My preferences are back too. Merci, Benoît! Jared Preston (talk) 16:27, 11 May 2017 (UTC)

I'm happy if you are happy, Jared! :) Trizek (WMF) (talk) 16:35, 11 May 2017 (UTC)

The Cycle 2 of the movement strategy discussion starts on May, 11

2 weeks ago, I posted that the Cycle 2 would start last week. It was postponed though, and the final official date is May, 11 (tomorrow). The end was also moved - to Friday, June 9. You can discuss 5 themes here. (You can also add and discuss your own themes). Any questions? SGrabarczuk (WMF) (talk) 13:02, 10 May 2017 (UTC)

Another update! The Cycle 2 ends on June, 12. SGrabarczuk (WMF) (talk) 07:41, 11 May 2017 (UTC)

Regex help, please

Can someone explain why all the items with History of Modern Biomedicine ID (P3885) are showing us as format violatons, after I made the [^\s\/]+ constraint mandatory in this edit, please? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:16, 11 May 2017 (UTC)

I've removed the added spaces, maybe that helps? Sjoerd de Bruin (talk) 10:23, 11 May 2017 (UTC)

AFAIR that should help. The trailing space after the nowiki-tag of the format constraint caused the problem. —MisterSynergy (talk) 10:29, 11 May 2017 (UTC)

Thank you. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:42, 11 May 2017 (UTC)

Query Service examples

Tracked in Phabricator
Task T165029
Resolved

Hi, for some reason for me the example list in Query Service is empty. There is no javascript error and it was same when i was logged out too. --Zache (talk) 11:13, 11 May 2017 (UTC)

Thanks for reporting, it's happening for me as well. We will have a look. Lea Lacroix (WMDE) (talk) 11:37, 11 May 2017 (UTC)

@Zache, Lea Lacroix (WMDE): Should be fixed now. --Lucas Werkmeister (WMDE) (talk) 14:13, 11 May 2017 (UTC)

A little slow to load, though works after waiting an extra two breaths.  — billinghurst sDrewth 14:23, 11 May 2017 (UTC)

Finding paintings

I wrote a tool that suggests items on Wikidata that might be paintings. You can find it at Wikidata:WikiProject sum of all paintings/Possible paintings. It currently reports about 2000 suggestions in 45 different languages. I could use some help with reducing this report especially in languages that don't use the Latin script. Multichill (talk) 20:30, 7 May 2017 (UTC)

What sort of help? Noli me tangere (Schongauer) is a painting and Balzac et le daguerréotype is not. Both these are good bets based on the title, but not sure things. All the best: Rich Farmbrough, 02:45, 8 May 2017 (UTC).

Hey Multichill, thanks a lot for writing that tool! I think it is wonderful and I am so glad that you spend so much time and dedication in tasks like this. It means a lot to me and to the community as well. As said, a heartfelt thank you.--Micru (talk) 06:58, 8 May 2017 (UTC)

@Rich Farmbrough: judging from your user page, you only speak English. I cleared English some time ago. I'm looking for people who can read the linked Wikipedia articles so they can determine what the item is about. In the case of la:Noli me tangere (Sartius), my Latin is good enough to determine this is a painting so I merged it into Noli me tangere (Q3877890). As for fr:Balzac et le daguerréotype, I'll leave that to a native French speaker to figure out. Multichill (talk) 09:37, 8 May 2017 (UTC)

@Micru: happy to see my work is appreciated! Multichill (talk) 09:37, 8 May 2017 (UTC)

Nice! I"m still on vacation and I'm still working on the this Met list. Should I keep on working on that...or...? Missvain (talk) 08:29, 8 May 2017 (UTC)

@Missvain: Unless you learned, Korean, Japanese or some other non-Latin script language, I would stick to the MET :-) Multichill (talk) 09:37, 8 May 2017 (UTC)

BWHAHA. that is what I get for not reading before coffee. Missvain (talk) 16:52, 8 May 2017 (UTC)

@Multichill:

for orthodox icons should I use "Painting" or "Icon" as p31 ? Hsarrazin (talk) 21:31, 11 May 2017 (UTC)

@Hsarrazin: I'm not sure, but leaning towards icon (Q132137). Multichill (talk) 10:28, 12 May 2017 (UTC)

Comments request

Please comment this proposition for centralizing most important community requests in the Wikidata:Community_portal. Thanks Snipre (talk) 20:21, 11 May 2017 (UTC)

Spacecraft - Instance and subclass issues

Wikidata ontology for gun, cleaned up - using Wikidata Graph Builder

Wikidata ontology for spacecraft, with instances and subclasses mixed together problematically

Help and advice needed: In cleaning up the ontology of spaceships and satellites, I'm coming into conflict with @Ivan A. Krestinin: regarding the basic understanding of instances and subclasses for this domain. Right now, it is a big mess of perhaps 20-30 well-thought-out subclasses (eg. unmanned spaceships, communications satellites, navigation satellite, et al) mixed with many items erroneously created as "subclass of" spaceships or satellites (examples: Dragon 2 (Q17122887), Geosynchronous Space Situational Awareness Program (Q26417718), SRET (Q18027603)). These are rather minor country-specific projects or initiatives, and I would argue are not inherently new "classes" of spacecraft. In cleaning things up, I went in and removed them as subclasses, and put in proper "instance of" statements pointing to relevant subclasses. However, @Ivan A. Krestinin: has massively reverted the changes with only "not single spacescraft" as the rationale for declaring them "subclass of." I think this is a pedestrian and improper understanding of classes and is not a justification to massively revert. Just because an effort has produced more than a "single spacecraft" does not mean a new "class" of vehicle has been created. I'm not a stranger to this – I've successfully cleaned up the firearms and guns ontology with good results, as you can see from the current snapshot of the resulting subclasses in Wikidata Graph Builder. The "spacecraft" ontology now is a mess, and the massive reverting is keeping it in that state. I'm seeking advice and experiences from others as to what has been done in the past. I've worked with database modeling and cleanup projects outside of Wikidata, so I wanted to check with veterans as to whether this is a valid distinction for instances and subclasses. -- Fuzheado (talk) 01:36, 12 May 2017 (UTC)

You just mix instance of (P31) and subclass of (P279). For example Kosmos 2471 (Q6433282) is instance of satellite. But GLONASS-K (Q2428375) is type/class of satellites. — Ivan A. Krestinin (talk) 04:05, 12 May 2017 (UTC)

You seem to be overaggressive in calling anything with n > 1 as a new "class" of satellites. I can believe your argument that GLONASS satellites, being a prominent geolocation initiative, might constitute a prominent subclass of satellites. But you massively reverted dozens of items of very minor significance for which there is no good argument for their being a new "class." In fact, most of those were short-lived, small scale projects that should simply be instances. What's your response to that? -- Fuzheado (talk) 04:31, 12 May 2017 (UTC)

My bot uses current ontology during data import. For example instance of satellite can have COSPAR ID (P247) property. Class can not have it. Also class/instance relations are checked by {{Constraint:Type}}. Your edits violate such constraints. — Ivan A. Krestinin (talk) 04:41, 12 May 2017 (UTC)

This discussion comes up occasionally. As far as I understand, Ivan's position seems to be pretty much what we usually do: use instance of (P31) only if the item is about a specific instance of something. Otherwise it is a “subclass” in our project, even if it just describes the set of identical instances of this “class” with their shared properties. Does this create any stress in the ontology? —MisterSynergy (talk) 05:56, 12 May 2017 (UTC)

This isn't quite right. A set of identical items may be a subclass of one wikidata item, and an instance of another (a metaclass - for example GLONASS could be an instance of "type of satellite" in the present case). But in general subclass of (P279) is the right relationship for most such cases. What Fuzheado has done with firearm (Q12796) (or what looks like a consensus in that area) may be something that could be applied generally to manufactured products, but I'm not sure if it's ontologically quite right - TEC-9 (Q1265013) (of which there have been 100's of thousands made) is given for example as an instance of short gun (Q7816242) which is a subclass of firearm (Q12796). So in fact what "gun" or "handgun" here means is essentially a class whose instances are specific models of weapons, but not the individual weapons themselves. As long as it's done consistently I think this is ok, but perhaps the names may be a little confusing as they stand. Perhaps something for discussion under Wikidata:WikiProject Ontology? ArthurPSmith (talk) 17:22, 12 May 2017 (UTC)

@Fuzheado:, TEC-9 (Q1265013) is unfinished.

If I can own a Intratec TEC-DC9, then Q1265013 should be a type... but I don't see a subclass of (P279) relation at this item (Q1265013).

<Intratec TEC-DC9 #123 owned by d1g> P31 <Q1265013> would be disconnected from weapon (Q728) d1g (talk) 19:25, 12 May 2017 (UTC)

@ArthurPSmith: What I've done is the most widely accepted way to organize firearms, not just in Wikidata but in the real world - specific models are "instances of" a class of weapons (semi-automatic pistols, assault rifle, automatic shotgun, et al) unless a specific make/model define a new paradigm. Prominent examples include the 1911 pistol (a Colt M1911), AR-style rifles (based on AR-15), AK-style rifles (based on AK-47) or even the Glock pistol. These are instances within their domain that became so influential that they defined a new subclass of weapon. However, those that do not rise to this level don't warrant a "subclass of" property. So how do we model the lineage of weapons otherwise? That brings us to... @D1gggg: While the TEC-9 certainly has notoriety, I'm not sure it rises to be a subclass. The TEC-DC9 is a variant, which I'm inclined to use based on (P144) to model, a I did with Spandau lMG 08 (Q15129149) and this diff. [5]. I'm open to discussing the merits of each approach. -- Fuzheado (talk) 00:29, 13 May 2017 (UTC)

It is nowhere written that item should should have one P31 or P279.

the P31 means "relation no longer propagates down to hierarchy"

a P279 means "relations WILL propagate down to its child's (P31)"

Open Australian 5-dollar note (Q28794069) to see that it describes a design (P31), but also about manifestation of Australian dollar (transitively comes with P279 property)

It is possible to say <his 5 AUD note> P31 Q28794069 (the note from pocket, specific and very new design) or <his 5 AUD note> P31 Q4209331 (the note from pocket $5 AUD, but exact design is unknown).

When you open Australian 5-dollar note (Q28794069)    under "classification" it would be linked to artificial physical object (Q8205328) - which means I did everything right with P31/P279

@ArthurPSmith: @Fuzheado: when you open TEC-9 (Q1265013)    or Spandau lMG 08 (Q15129149)    they don't link to an artifact or have a nice "classification" hierarchy. d1g (talk) 01:32, 13 May 2017 (UTC)

Have to agree with Ivan edits at Rubin (Q2171616) - they group 10 items, so relation is <subclass of>

Otherwise, if many types are not desirable, we can remove <subclass of> at <Rubin> item and place <P31> of <something> at <Rubin 4> and so on if they share some other property (location, manufacturer). d1g (talk) 06:13, 12 May 2017 (UTC)

I also think that subclass of (P279) is right in this context. Given the way our constraints are set it otherwise produces issues. ChristianKl (talk) 08:19, 12 May 2017 (UTC)

I totally disagree. Why would a small group mean a new subclass. It makes the translation of our information more problematic for no particular gain. In the past I did indicate that the whole of the classification is problematic as there is nobody able nor willing to explain it. So no.. ten items that share a peculiarity does not make a subclass.  – The preceding unsigned comment was added by GerardM (talk • contribs).

I was working on <Rubin>s, <Uran>s - how this isn't a type?

In order to disambiguate "Rubin" we must have a every type with "Rubin" name. d1g (talk) 18:46, 12 May 2017 (UTC)

Our disadvantage is based on a bit different understanding on "class" term. But your can introduce second parallel ontology with different classes if you need. For example GLONASS-K (Q2428375) subclass of (P279) list of GLONASS satellites (Q12646602), but also we can say: GLONASS-K (Q2428375) instance of (P31) "satellite type", GLONASS-K (Q2428375) instance of (P31) "navigation satellite type" or GLONASS-K (Q2428375) instance of (P31) "satellite series". You just need to create new item with label "satellite type/class/series". — Ivan A. Krestinin (talk) 19:14, 12 May 2017 (UTC)

Removal of aliases using Pywikibot

Is there a good way of removing all of the aliases on an item for a specific language (in my case, Bengali) using Pywikibot? Replacing the list of aliases for that language with the empty list throws an error of 'index out of range', and replacing it with [''] throws an error that aliases must be one character long. (Deleting key 'bn' from the alias dictionary, or pretty much anything else, results in a no-op.) Mahir256 (talk) 01:52, 12 May 2017 (UTC)

Wikibase has a comprehensive doc for this. Looking to the code, I think you can evade errors with this code:

aliases = {}
aliases['bn'] = list(map(lambda al: {'language': 'bn', 'value': al, 'remove': ''}, item.aliases['bn']))
item.editEntity(aliases)

Matěj Suchánek (talk) 06:32, 12 May 2017 (UTC)

FYI, I noticed [6] and blocked this unapproved bot. Multichill (talk) 10:24, 12 May 2017 (UTC)

@Multichill: Mahir256 seems to also have made a lot of nonbot edits. I'm not sure that blocking his whole account is the way to go. Maybe we also need a clearer policy about what's a bot. It seems like saying everything done via Python code is a bot while everything done via QuickStatements isn't, is a bad policy. ChristianKl (talk) 12:45, 12 May 2017 (UTC)

You seem to be a bit confused Christiaan. I didn't block Mahir256, I blocked User:Twofivesixbot. Twofivesixbot is an unapproved bot doing large amounts of edits, something that is not allowed in our bot policy. We often disable unapproved bots and enable them again when a request for approval gets done. Multichill (talk) 19:22, 12 May 2017 (UTC)

Okay, then it makes sense. ChristianKl (talk) 22:59, 12 May 2017 (UTC)

Could we have a property for "first"? Example: Grace Banu

Hi, I'm a noob to Wikidata, so forgive any vagaries. I created an article on Wikipedia Grace Banu and a related item in Wikidata: Grace Banu (Q29439153). Banu is a Dalit and Transgender activist, and her "claim to importance" on Wikipedia is that in addition to being a high-profile activist, she was the first transgender woman to earn an Engineering Degree in Tamil Nadu.

I was looking for a property that would mark such a "first", "first to", "pioneer in", etc. There doesn't seem to be a relational property which indicates this primary position. I understand such a factoid might emerge from a query in the category (earliest date of a woman to earn an Engineering Degree in Tamil Nadu), but it struck me that these watershed moments are both quite common and quite significant in history. Similarly, "last" is also a distinct and important property.

What are thoughts on this, if I am even thinking about this in the right way? Thanks, Jake Ocaasi (talk) 17:03, 12 May 2017 (UTC)

Easiest way is find another item about a human (Marie Curie (Q7186)):

educated at (P69) with qualifiers academic major (P812), end time (P582) and academic degree (P512)

When we fill this information, we would be able to answer questions like you said. Make sure that university is linked to Tamil Nadu. d1g (talk) 20:00, 12 May 2017 (UTC)

Supplement GND ids from VIAF

I've recently checked which human (Q5) items have VIAF ID (P214) but no GND ID (P227). Then I looked up the missing GND ids in a current custom-built VIAF endpoint with this query. That yielded 166,962 items in total and 11,958 items for "economists" (mentioned with their GND in EconBiz) for which their GND could be added in Wikidata.

I want to insert the GND ids for the "economists" subset into Wikidata. For now, the plan is to generate statements for QuickStatements2, like this:

Q72474|P227|"1055279814"|S248|Q54919|S813|+2017-04-01T00:00:00Z/10

in batches of 1,000.

Are there concerns about this approach - or perhaps hints to a tool more suited for a task of this size? Jneubert (talk) 09:47, 8 May 2017 (UTC)

Seems like a good idea to me. Do we know how confident we can be in VIAF’s matching? (I’ve noticed a few edge case errors in their data before, but they’re probably rare). We should also add VIAF as a reference for the statement? Frankieroberto (talk) 10:52, 8 May 2017 (UTC)

VIAF clusters may be wrong in some places. However, that can only be fixed one by one. For example, I accidentally found the fencer Sebastian Bachmann (Q65725) having a VIAF link to http://viaf.org/viaf/198515565, which in turn links to http://d-nb.info/gnd/1049485092 (somebody publishing at Augsburg university about compliance and qualification models). There is no indication these being the same person, and on the contrary, [7] shows that they are different. Therefore I set the VIAF ID to "unknown value". But this error would not have been introduced by the mapping, but was already in Wikidata through the VIAF external id. As you suggest, I plan to add VIAF as a source to each introduced statement, to make clear that the information has been derived from VIAF. Jneubert (talk) 12:04, 8 May 2017 (UTC)

Pay attention to the "undifferentiated names". These are unstable disambiguation pages by GND and should not be added to Wikidata. For example http://d-nb.info/gnd/155294156 is linked by https://viaf.org/viaf/111745991/ which is an entry in your data set. --Pasleim (talk) 11:03, 8 May 2017 (UTC)

@Pasleim:: Thanks for the hint. Did you find http://d-nb.info/gnd/155294156 in one of my result files? For the "economists" set from EconBiz, that should not occur since it includes only gndo:DifferentiatedPersons. Additionally, the VIAF source file seems to have no schema:sameAs statement for http://d-nb.info/gnd/155294156. Jneubert (talk) 11:47, 8 May 2017 (UTC)

If you only include gndo:DifferentiatedPersons then I'm okay with the import. --Pasleim (talk) 14:15, 8 May 2017 (UTC)

@Jneubert: http://d-nb.info/gnd/155294156 is a name identifier (not a person identifier) and thus deprecated. Please don't add any of this sort as GND statements on Wikidata! Jonathan Groß (talk) 11:53, 8 May 2017 (UTC)

I'm fully aware of the problem, and I will not introduce links to UndifferentedPersons (the whole approach is designed to preclude that). Jneubert (talk) 12:08, 8 May 2017 (UTC)

If you're using QuickStatements, apply for a bot flag for your 'automated' account - it will work more quickly. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:04, 8 May 2017 (UTC)

Thanks for the hint - I've filed a bot request now. Jneubert (talk) 14:46, 8 May 2017 (UTC)

With some helpful hints from Pintoch re. the exclusion of possible duplicates and an appropriate reference format (here), I've finished the task of including the GNDs of the almost 12,000 persons related to economics (e.g., H. William Scharling (Q6096410)). If somebody is interested to do the same thing for another domain, or perhaps for another authority linked in VIAF, I'd happily try to help. Jneubert (talk) 12:59, 13 May 2017 (UTC)

API throws 'badtoken'

I don't know if this is the right place to ask, but: whenever I request a csrf token with action=query&meta=tokens and use it in a PHP script, I get the 'badtoken' error. I thought it might have something to do with my script, but when I open https://www.wikidata.org/w/api.php?action=help&modules=checktoken in my browser I get the response 'invalid'. The token is URL-encoded. Can you help me?--Phipra (talk) 13:43, 13 May 2017 (UTC)

book (Q571)

Sorry if I revive an old unending discussions but we there has long been real problem with book (Q571), and it is a widely used item, and we should really solve it once and for all.

• Sitelinks like en:Book are essentially about physical books (sheets bound together)
• English Wikidata description states that Q571 is "medium for a collection of words and/or pictures to represent knowledge or a fictional story, often manifested in bound paper and ink, or in e-books". Not sure it makes any sense. How can a "medium" be "often manifested in bound paper and ink, or in e-books" ?
• book (Q571) has various subclass of (P279) values, some of them sound mutually incompatible.
• Help:Sources suggest that book (Q571) should be used as a P31 of some "work items" like Bible (Q1845). But book (Q571) is also used for items about individual physical books like Florentine Codex (Q1106019)

All this could be solved *really easily*. Let's just accept that book (Q571) is about physical books and use another value for items about text text (Q234460) or a "written work" item or whatever. But keeping the Q571 nonsense after all those years is really a shame.--Zolo (talk) 14:23, 7 May 2017 (UTC)

I agree. To be a book (Q571) is rather a characteristic of some individual books.

There is literary work (Q7725634) for works of (fictional/aesthetic) literature and non-fiction (Q213051) as a genre of written works, but I could not find something more generic. To use text (Q234460) for that purpose would probably work but I also like the idea of having a new "written work" item comprising both literary work (Q7725634) and non-fiction (Q213051) as this would make it clear (by its name) that an item is about a work and not about a manifestation/a physical book. Valentina.Anitnelav (talk) 16:11, 7 May 2017 (UTC)

If I remember well the problem was not to replace book by another definition, but to define which definition should be used. Work was too generic and nobody was able to provide a good alternative, alternative which should be integrated in a global classification for all works. Currently people are working on items and not on classes of items. Few persons try to have a more general view of the classification or ontology in WD. If we want to change some definitions on large number of items we have to define first what will be the global classification. Just to avoid new changes in the future each time someone comes with a new special case.

For that we can start from the scratch, but this always leads to endless discussions as the cultural background is very large in WD. So the second is to find an existing classification for work and start the discussion from that or thoses examples. Snipre (talk) 20:55, 7 May 2017 (UTC)

@Snipre: the ontological questions around texts has lead to lengthy discussion, but I think it has been mostly solved, more or less satisfyingly. Here, we just have a very simple problem about the use of an item that is undisputably incorrect, we only need to settle on another one to clean things up.

Can we just agree on using "written work". Or perhap "verbal work" as some people will point out that it can be materialized as audiobooks ? --Zolo (talk) 08:02, 8 May 2017 (UTC)

@Zolo: Sorry to be bold but when you look at the subclasses of creative work (Q17537576) (see here), you find literary work (Q7725634), text (Q234460), apocrypha (Q4068032), superwork (Q24261960) and anime and manga (Q10901350). We should have

* text (Q234460) subclass of literary work (Q7725634) or the inverse but at least not both concepts at the same level of the classification

* anime and manga (Q10901350) subclass of literary work (Q7725634)

* apocrypha (Q4068032) subclass of literary work (Q7725634)

* and I don't know how we can deal with superwork (Q24261960), creative work (Q17537576) and literary work (Q7725634). We should have somewhere an explanation giving the information how to classify works between these different concepts. I don't need a link to the discussion but a summary page where everyone can understand the classification rules.

Just changing once the classification in the Project chat won't be useful as next month most of contributors will forget what was said and everyone will start again to classify according to his vision and not according to a general classification scheme. Snipre (talk) 09:04, 9 May 2017 (UTC)

This kind of nonsense with instance of (P31) and subclass of (P279) has existed since the day these properties were created. Wikidata is mostly an unusable mess due to the way these properties have been used. Surely, instance of (P31) should only be used for singular items (an instance, one example), as in a single copy of a book in museum, or a single aircraft, etc. And, an instance should be an instance of class (subclass of (P279)), not an instance of an instance? Danrok (talk) 00:27, 9 May 2017 (UTC)

@Danrok: there's nothing wrong with a class being an instance of a metaclass, or even an instance of a class with some level of ambiguity as to level (think of a class as a set of items and potential items that are its instances). However, most of the relationships between generic things should be subclass of (P279), not instance of (P31) (if either) as you suggest. Singular physical things cannot be classes, but outside of people, places, objects, and organizations most items in wikidata can be thought of both as classes and as instances of other classes. Creative works in particular fall into this pattern. ArthurPSmith (talk) 20:57, 9 May 2017 (UTC)

@ArthurPSmith: Your idea may be valid, however I don't think that is how the properties instance of (P31) and subclass of (P279) are meant to be used. The way they are meant to be used is very simple, and is shown here: Classification in Wikidata. Danrok (talk) 23:20, 9 May 2017 (UTC)

@Danrok: not sure why you think that contradicts anything I said. I helped with significant updates to the Help:BMP page, and have been involved with the Wikidata:WikiProject Ontology group so I've looked at a lot of this in practice. A big problem area other than books is chemistry - I can't think of any case where it would be useful for wikidata to have an item corresponding to a single manifestation of a particular chemical compound or element etc so EVERYTHING in chemistry and physics is basically a subclass of entity. But specific chemical elements, for example, can be grouped into classes and so should have relations as instances of other wikidata items. But none of that is particularly simple - we've been arguing about it in the chemistry project for many months. ArthurPSmith (talk) 12:47, 10 May 2017 (UTC)

First of all, if the item is split, the one with the P31 relations should be at Q571, because there are 176435 statements that would otherwise need to be changed.

Second, a book is more specific than a text/written work. Even if "[physical] book" is split off, the label should remain "book". Some texts/written works are not books. For example, a short poem, a speech, a pamphlet, and a newspaper article are not books. We should not introduce additional ambiguity.

Regarding the relationship between individual physical printed copies of books and the books, see WD:Books#Exemplar_properties. A single physical book is an exemplar of (P1574) of the book. --Yair rand (talk) 19:09, 9 May 2017 (UTC)

@Yair rand: the thing is I do not see any satisfying definition of "book" that does not point to a physical object. Do you have a definition of book that would make it "more specific than a text/written work", i.e. make it a subclass of text ? To me the two concepts are disjoint. We can be more specific about that type of text by providing the genre like "novel" or "dictionary" (this is currently mostly done through genre (P136)). --Zolo (talk) 07:22, 13 May 2017 (UTC)

Oxford dictionary tries this way: "A literary composition that is published or intended for publication as a book." Cleary the very vast majority of entries with Q571 as P31 use "book" in this sense and not in the sense of a single physical object. ("Gone with the Wind.", "The old man and the Sea." ...) This use should be taken into account. 123 (talk) 19:04, 13 May 2017 (UTC)

Denying Bot's edits

Hi, everyone. In Wikidata, could we use the template or something to deny bots' editing? This bot keeps adding wrong information on this data many times. I've asked the owner to stop the edit, but s/he hasn't noticed. Most wikis have the template for allowing/denying bots' edits but I'm not sure Wikidata has the same template. I'm happy if someone tells me how to do. Thank you and best regards. --Ohtani tanya (talk) 00:20, 14 May 2017 (UTC)

Look into the source code of the Japanese article, and you will see such a link: [[En:afreecaTV|En:afreecaTV]] - it should be a wikilink probably, but due to the missing preceding colon actually it's interwiki link. This is why the bot keep trying to connect that article with the English one. --XXN, 00:58, 14 May 2017 (UTC)

Thank you, XXN. I asked sysops in Japanese Wikipedia to get rid of the words that you pointed. --Ohtani tanya (talk) 05:57, 14 May 2017 (UTC)

Missing properties

I was adding data on Vidyasagar (Q4673553), but there are no properties for adding religious sub-division (sect), intiation date, intiation place, initiator. Capankajsmilyo (talk) 08:25, 10 May 2017 (UTC)

@Capankajsmilyo: For all but the latter, you can use significant event (P793), with qualifiers. If you tell me the values, as Q IDs, I'm happy to demonstrate. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 08:43, 10 May 2017 (UTC)

@Pigsonthewing: Religion - Jainism (Q9232), Sect - digambara (Q9436), Subsect - Digambara Terapanth (Q5275549), Inintiationdate - 30/06/1968, initiationplace - Ajmer (Q200049), intiator - Gyansagar (Q4673522) Capankajsmilyo (talk) 11:40, 10 May 2017 (UTC)

@Capankajsmilyo: Something like this. Or this. (Other permutations are possible.) Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:14, 10 May 2017 (UTC)

@Pigsonthewing: thanks. Looks good for now. Can significant person be replaced with something better? Also Digambara Terapanth was to be added to religion as a qualifier in Jainism. I can't find a good method for that.

Further, it would be helpful if you could give some suggestions on how to add tree and symbol to Mahavira (Q9422), Parshvanatha (Q1400271) and Rishabhanatha (Q9429). Capankajsmilyo (talk) 13:10, 10 May 2017 (UTC)

@Capankajsmilyo: I suspect we need a new property to replace "significant person" in this context; perhaps "officiant", which would also cover weddings, funerals, presidential swearings-in, etc. I don't understand what you mean by "tree and symbol". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:40, 10 May 2017 (UTC)

@Pigsonthewing: Please see infobox at en:Mahavira for tree and symbol. Also intiator is neither officiant nor significant person. He is more like a mentor referred to as Acharya or guru. Capankajsmilyo (talk) 17:11, 10 May 2017 (UTC)

@Capankajsmilyo: At..? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:39, 10 May 2017 (UTC)

@Capankajsmilyo: how about student of (P1066) which is meant to cover mentor ?  — billinghurst sDrewth 23:34, 10 May 2017 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ Student of seems good for now. @billinghurst:, @pigsonthewing:, along with age and tree, need suggestion for mythological age of deities. See Ajitanatha (Q3351786) and its corresponding infobox in English wikipedia for reference. Capankajsmilyo (talk) 05:47, 11 May 2017 (UTC)

age of subject at event (P3629) seems the nearest property for putting down the age. ChristianKl (talk) 21:30, 11 May 2017 (UTC)

I'm afraid that won't serve the purpose. Is there anything else available? Why don't we have simple age? Capankajsmilyo (talk) 17:38, 14 May 2017 (UTC)

In general a human has a different age in 2017 than he had in 2015 and as such it doesn't make sense to have a stable property. Saving the year of birth is usually the best but sometimes we have a source that states that the human was of age X at a particular moment in time when we don't have their birth year, so we introduced age of subject at event (P3629). You can't reasonably specify that Alice is ten years old without saying when Alice was ten years old. As I remember from the proposal discussion this property is also supposed to be used for works of fiction. ChristianKl (talk) 20:51, 14 May 2017 (UTC)

When is relevant in case of actual people. It can't be used for religious mythologies whose age goes in billions and trillions of years. For example, see english wikipedia infobox for Rishabhanatha (Q9429). Capankajsmilyo (talk) 19:28, 15 May 2017 (UTC)

Mount

Is it possible to use mount (Q25397729) as a property? Hindu deities usually has an animal associated with them, they ride on. How to add that? Capankajsmilyo (talk) 17:38, 14 May 2017 (UTC)

Properties are the things that start with P. That means you can use mount (P3091) while mount (Q25397729) is an item. ChristianKl (talk) 17:43, 14 May 2017 (UTC)

Thanks a lot, exactly what I needed. Capankajsmilyo (talk) 19:15, 15 May 2017 (UTC)

Festivals

Need suggestion of a property which can be used for "festivals" in Rama (Q160213), as used in infobox on English Wikipedia. Capankajsmilyo (talk) 19:15, 15 May 2017 (UTC)

WikidataCon: join the program committee and other teams!

Hello all,

I'm glad to have more information to give about the WikidataCon event, that will take place in Berlin on October 28th-29th.

This conference will be built by and for the Wikidata community. In that regard, we are actively looking for volunteers to take part to the most important projects of the organization.

• Program committee: will coordinate and select the content of the event. → Learn more and apply before May 17th
• Scholarships committee: will define criteria and select people that will receive a scholarship. → Learn more and apply before May 17th

But also open groups on various topics:

• Accessibility: can be involved on any topic regarding the accessibility (on a large scale) of the event, the respect of the diversity, the inclusiveness.
• Social events: will help people connecting with each others during the conference.
• Communication: logos, pictures, posters, signs, goodies, and all the things.
• Logistics: about every practical questions.
• Catering: because food is an important topic.
• Documentation: keep track of everything happening.

If you don't find answers to your questions on the sub-pages of Wikidata:WikidataCon_2017, feel fee to ask on the talk page! At anytime you can reach me to talk about the event.

Thanks, Lea Lacroix (WMDE) (talk) 13:02, 10 May 2017 (UTC)

What kind of workload do you expect from committee members? ChristianKl (talk) 21:25, 11 May 2017 (UTC)

It depends of the time in the planning. The program committee will have a lot to do at the end of the submission process, the scholarship committee when they will have to select the recipients. But this should not be more than a few hours a week. Lea Lacroix (WMDE) (talk) 14:31, 15 May 2017 (UTC)

Merge of language properties

I have just made my ever biggest decision over Wikidata and closed discussion about merging two properties with 1,000,000+ uses. This is a really major change, so no actions should be taken until some other aspects are discussed (made a subsection for each one there). Matěj Suchánek (talk) 15:01, 13 May 2017 (UTC)

• It's regrettable that the WikiProject Books wasn't able to come up with a working solution for their field without adversely affecting the rest of Wikidata.
  --- Jura 16:11, 13 May 2017 (UTC)

The problem is not what are the rules for books and what are the rules for movies, the problem is how do we want to save data in WD and what are the recommendations for external people who want to extract data from WD. Data duplication should be avoided and data have to be stored in the more related item. Very simple rule which will allow code reusability (queries, infoboxes,...) for data extraction for books and movies. Snipre (talk) 20:12, 13 May 2017 (UTC)

You'd still need to come up with a working system for books based on the current input tools and users' capacity to use them. Reusers can easily code two properties, but the book project still needs to come up with a working system. "Working system" meaning actually being implemented and maintained in practice, not merely limited to thousands of government-sponsored, bot-created items for articles.
--- Jura 06:05, 14 May 2017 (UTC)

I must be stupid, it seems simple to me, for a original implementation of a concept (literary work, book, speech, film, ...) it is in a language (it doesn't need restating original to define, it is what it is). BANG. There are editions and translations and they are in their respective languages. The merge seems the right way to proceed. [If we can decipher male and female siblings from code, we can decipher original language on the base item of the literary work/original concept.]  — billinghurst sDrewth 09:20, 16 May 2017 (UTC)

Is the image importing not still running?

I had to recently add a lot of images from English Wikipedia infobox to Wikidata along with media legend. I saw a few cases where this work was done by some bot. Why has that bot been stopped? Will it run again?  – The preceding unsigned comment was added by Capankajsmilyo (talk • contribs).

There was some mass-import that had to be undone as it imported many images not suitable for Wikidata items.

You can use WikiData Free Image Search Tool (Q21283216) to add them for items that interest you.
--- Jura 06:36, 14 May 2017 (UTC)

Can I mass import images for english wikipedia pages of a specific english wikipedia category? I am not able to understand how to do that with the stated tool. Capankajsmilyo (talk) 19:32, 15 May 2017 (UTC)

Disable and remove king from occupation

I have initiated a discussion on this before and would like to resume the same. King seems to be a position hel rather than an occupation. So it should be disabled as a value in occupation. Suggestions and comments invited. Capankajsmilyo (talk) 18:29, 13 May 2017 (UTC)

There's no way in Wikidata to disable specific values. At the same time, the constraint should give a warning when someone uses the property this way because king (Q12097) doesn't subclass occupation (Q12737077). ChristianKl (talk) 20:56, 13 May 2017 (UTC)

As people are likely to re-add it, you could just set it to deprecated rank and use it to check if position held (P39) is filled in.
--- Jura 06:33, 14 May 2017 (UTC)

That is an open invitation to remove all of them. Easy enough. So no do not deprecate; it is plain wrong. Thanks, GerardM (talk) 06:41, 14 May 2017 (UTC)

Not sure what you thank for, but "deprecated" is a rank. It doesn't imply removal. If you want to keep them with normal or preferred rank, do you actually think "king" is a suitable value for occupation (P106)?
--- Jura 06:54, 14 May 2017 (UTC)

I do not thank you. Your argument is bad. King is not an occupation and seeking it wrong and needs to be removed not deprecated. Thanks, GerardM (talk) 07:09, 14 May 2017 (UTC)

The deprecated rank is designed for things that are wrong and avoids people readding it with a normal rank. Sjoerd de Bruin (talk) 08:04, 14 May 2017 (UTC)

I have always understood deprecated rank as a way of indicating that someone said it was true but we know it isn't. I find the interpretation that deprecated rank is for anything wrong (without having a source that tells the opposite) very dangerous. (In my opinion, if there is something to be deprecated, it would be that king is a valid occupation.) Matěj Suchánek (talk) 08:30, 14 May 2017 (UTC)

I agree with Matěj Suchánek. Not every possible wrong value should be included and in a case like this deprecating isn't the way to go. ChristianKl (talk) 09:34, 14 May 2017 (UTC)

Creating items by copying - mostly. Is there a tool?

Forgive what is probably a recurring question from a wikidata noob, but I'm frustrated and exasperated and looking for help.

While editing the en wikipedia page for the McMath Solar Telescope, I encountered Category:Articles using Infobox telescope using locally defined parameters and started digging.

MOST large telescopes are on that category list. So, I thought, well, I can kill this one. and came over to wikidata and started adding bits to the entry at Q56901 so that I could remove them from the Infobox telescope on the wikipedia page. This seemed like a good thing. But I quickly ran into issues. For example, the McMath is a "solar telescope" Ok, there's an item for that, and it's a "reflecting telescope" Ok, goody. But it's an "Unobstructed" "Off-Axis" reflecting telescope. I found items for a few types of reflectors; cassigrains, gregorians, etc, but really lots and lots of types of reflector just aren't there. Building them is a pain when they are all instances of "reflecting telescope." It would be lovely if I could pick something like a Gregorian reflector and just say "copy this except for the label and the description to a new item."

Similarly, it would be lovely if, knowing what the items in the Telescope infobox are, if there was a way to have a template, with the various elements already there and it would be possible to just fill them in. Honestly, otherwise, I see why almost every telescope out there is in that category list, it's a tools-to-make-tools-to-make-tools problem.

Doubtless, what I'm asking violates the spirit of the database at some level. I certainly understand the idea of each element being related in a giant tree of elements, and I've been reading the discussions here where opinons are strongly held about subclass vs instance of, etc. I get it. On the other hand, I'ld like to move a lot of information out of en.wikipedia infoboxes for stuff I'm interested in, and get them into wikidata, so I'm asking. Any tools to help with that? Rboatright (talk) 09:06, 14 May 2017 (UTC)

You can find list of many Wikidata-related tools and gadgets at Wikidata:Tools. In particular, PetScan can create items for articles that haven't been connected to Wikidata yet, and add the same statement(s) to a list of items generated from a list of articles that have something in common (mostly a category). HarvestTemplates can extract useful data from template parameters and add it to items. Finally, QuickStatements (or its legacy version with instructions) can create items and/or add arbitrary data using instructions from a text document. Moreover, some gadgets (example) to export data from infoboxes have been developed for Wikipedia communities. Matěj Suchánek (talk) 10:41, 14 May 2017 (UTC)

@Rboatright: QuickStatements is probably best in this case, but for one-off copies, there's User:Magnus Manske/duplicate item.js. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:02, 15 May 2017 (UTC)

Constraint table updated

Hello all,

Since we deployed the constraint API and its user script, several of you noticed some errors in the constraints and fixed them. We now have updated the constraint table to take your modifications in account.

Thanks, Lea Lacroix (WMDE) (talk) 14:34, 15 May 2017 (UTC)

Is there going to be a regular shedule of updating the table? Is there a good reason why it can't happen daily? ChristianKl (talk) 15:18, 15 May 2017 (UTC)

It just takes a long time to extract all the constraints from the talk pages. We hope to read constraints from statements on properties soon (using property constraint (P2302); see e. g. COSPAR ID (P247) for some existing constraint statements), and once we can do that updates should be processed instantaneously. --Lucas Werkmeister (WMDE) (talk) 15:45, 15 May 2017 (UTC)

What’s the expensive part? It sounds quite simple to extract all constraint templates from all property talk pages… Covi pages receive updates on a daily basis as well. —MisterSynergy (talk) 16:17, 15 May 2017 (UTC)

While I'm not completely sure why it should be that expensive to read 4000 talk pages and run a script of them, I think developing a solution that works with the constraints statements on properties is the optimal way to go, so I approve of this plan. ChristianKl (talk) 08:10, 16 May 2017 (UTC)

Should the description in the proposal documentation match the description of the property?

In https://www.wikidata.org/wiki/Property_talk:P1144 the description in the proposal documentation is "Library of Congress Control Number (Q620946) (LCCN) for a bibliographic record: a specific version, edition or translation (Q3331189) of a book or other publications in the U.S. Library of Congress Online Catalog (LCOC).}}" while the description of the property is "Library of Congress ID for bibliographic records (for books, serials, maps, music, video, etc only; for names use P244)". Till now I was operating under the assumption that we want a unified description that's the same in both places. Do you believe that it's worthwhile to have to a different one (and one that uses Wikitext on the proposal documentation? ChristianKl (talk) 15:58, 15 May 2017 (UTC)

The |description= parameter of the {{Property documentation}} template is for the proposal phase, and ideally the given descriptions (and many other parameters) should be transferred to the property once it is created, and subsequently deleted from the template to reduce redundancy. The property talk page fetches the description again from the property, it shows it in the header section of the green box.

However, property labels and descriptions are refined occasionally, even long time after the property was created. This is in fact somehwat dangerous, since it can lead to slightly different property scopes within different languages, which is a problem of course. But there is no rule to my knowledge which prohibits this process, and I would keep it like that. The property proposal process does not bring properties to perfection, it is merely a decision over their eligibility. Refinement by the actual property users it thus necessary. —MisterSynergy (talk) 16:26, 15 May 2017 (UTC)

Wikidata weekly summary #260

Here's your quick overview of what has been happening around Wikidata over the last week.

• Discussions
  • Development input: Client editing prototype

• Events/Press/Blogs
  • Past: Pre-hackathon of the Wikimedia Hackathon, May 12th-14th in Prague (see a video of the Wikidata and SPARQL training)
  • Upcoming: Wikimedia Hackathon 2017, May 19th-21st in Vienna. On site or remotely, you can join the Wikidata documentation sprint
  • Upcoming: WikiCite, May 23rd-25th in Vienna
  • Next Wikidata meetup in Berlin: June 7th in WikiBÄR
  • You can still register for the Wikidata/election data hackathon in Ulm, Germany (June 23-25)

• Other Noteworthy Stuff
  • Discuss the movement strategy on Wikidata. The Cycle 2 started on May, 11.
  • WikidataCon: join the program committee and other teams!
  • TwoColConflict is now enabled on Wikidata
  • OSMgadget is a tool to help Wikipedia editors find objects in OpenStreetMap, and tag them with their Wikidata IDs
  • The OSM ↔ Wikidata matcher tool
  • Two language properties will be merged but the discussion still needs to go on
  • Tabular data files datatype is now enabled
  • The constraint table has been updated

• Did you know?
  • Newest properties: NLL player ID, Italian Senate ID, ALPG golfer ID, Stereo Ve Mono artist ID, BioRxiv ID, narrower external class, Juwra.com ID, MLL player ID, Directorio Grierson ID, RANM member ID, Old Bailey Proceedings ID, Tumblr ID, Bmx-results.com rider ID, Israel Antiquities Authority ID, OlimpBase Chess Olympiad player ID, ESTC citation number, named by, Reactome pathway ID, NFF person profile ID, Chamber of Deputies of Italy ID, face value, Cinema ID, Digital Valencian Library author ID, copyright owner, Rallye-info.com driver or co-driver ID, V&A item ID, MotoGP racer ID, eWRC-results.com athlete ID, USATF athlete ID, Track and Field Statistics male athlete ID, Track and Field Statistics female athlete ID, Diamond League athlete ID
  • Query examples:
    • Paintings in the Louvre Museum, with heights and widths (source)
    • Chart of USA cities founded per year (source)
    • Country flags including a symbol of sun (source)
    • Birthplace of Spanish Queen Consorts (source)

• Development
  • More work on Special:NewLexeme page (Phab:T164582)
  • Correctly compare time values on constraints (Phab:T164279)
  • Check wording of messages/constraint check results (Phab:T164354)
  • Creating grammatical features for Forms (Phab:T160525)

You can see all open tickets related to Wikidata here.

• Monthly Tasks
  • Help adding missing statements to properties
  • Add labels, in your own language(s), for the new properties listed above.
  • Comment on property proposals: all open proposals - proposals needing attention
  • Suggested and open tasks!
  • Contribute to a Showcase item.
  • Help translate or proofread the interface and documentation pages, in your own language!
  • Help merge identical items across Wikimedia projects.
  • Help write the next summary!

Read the full report · Unsubscribe · Lea Lacroix (WMDE) 19:48, 15 May 2017 (UTC)

RevisionSlider

RevisionSlider will be available as a default feature for all users on all wikis from May 17. The RevisionSlider adds a slider view to the diff page so that you can easily move between revisions. The slider view is collapsed by default, and will load by clicking on it. It can also be turned off entirely in the user preferences. RevisionSlider has been a default feature on German, Arabic and Hebrew Wikipedia for 6 months and a beta feature on all wikis for 8 months. The feature fulfills a wish from the German Community’s Technical Wishlist. Thanks to everyone who tested RevisionSlider and gave valuable feedback to improve the feature! We hope that RevisionSlider will continue to serve you well in your work.

Birgit Müller (WMDE) 14:44, 16 May 2017 (UTC)

Category pages for properties with missing statements

With the new visibility of constraints and a few complaints that certain recently recreated properties don't have all the relevant statements filled, I have the desire to better understand which properties lack certain statement.

For that I think it would be good to have a few category pages:

• Properties without instance of (P31)
• Properties without Wikidata item of this property (P1629)
• Properties without Wikidata property example (P1855)
• Properties without property proposal discussion (P3254)
• Properties with value external-id but without formatter URL (P1630)

It would also be great to have the show the age of a property on such a page. ChristianKl (talk) 21:38, 10 May 2017 (UTC)

Not sure what you mean with “categories”. I suggest to use simple queries such as this one:

SELECT ?prop { ?prop a wikibase:Property . MINUS { ?prop wdt:P31 ?instance . } }

Try it!

This is for missing instance of (P31) claims, but you can easily insert any other “missing” property as well. —MisterSynergy (talk) 21:51, 10 May 2017 (UTC)

When I say category I mean pages like https://www.wikidata.org/wiki/Category:Properties_ready_for_creation . ChristianKl (talk) 22:15, 10 May 2017 (UTC)

I think it would be good to someone place the task of working through those lists on https://www.wikidata.org/wiki/Wikidata:Contribute/Suggested_and_open_tasks ChristianKl (talk) 10:20, 11 May 2017 (UTC)

We could add the tracking categories to {{Property documentation}} or Module:Property documentation, that shouldn't be problem. But adding properties or items directly to categories is impossible, given that they are not rendered from wikitext. Matěj Suchánek (talk) 11:37, 11 May 2017 (UTC)

Okay, than let's stay with a SPARQL solution. I created https://www.wikidata.org/wiki/Wikidata:WikiProject_Properties/Adding_missing_statements to have a page that lists queries about properties with missing statements. One issue with the current solution is that the list doesn't show property name and description. It would be great if someone with more SPARQL experience looks at it and makes any improvements he can think of. ChristianKl (talk) 12:18, 11 May 2017 (UTC)

I updated the queries and how they are displayed on your project page. —MisterSynergy (talk) 12:32, 11 May 2017 (UTC)

Thanks. Currently the constraints are set in a way that a property needs two property examples. Can you change the SPARQL code in a way that it requires two examples instead of just one?

It would also be great to have a query for properties that lack label or descriptions in a given language. ChristianKl (talk) 12:53, 11 May 2017 (UTC)

Also done. —MisterSynergy (talk) 13:20, 11 May 2017 (UTC)

Three is absolutely no requirement for more than one example, in most cases. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:23, 11 May 2017 (UTC)

Currently, there are constraints set on multiple properties that assert that properties should have at least two examples. I'm open to changing this fact but changing it likely needs some form of consensus. ChristianKl (talk) 19:13, 11 May 2017 (UTC)

You don't need consensus before making an edit, just like nobody bothered to obtain it before adding that bogus constraint. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:00, 17 May 2017 (UTC)

@Matěj Suchánek:I think it would be good to have an automatic category for properties for which no constraint is set. ChristianKl (talk) 14:48, 11 May 2017 (UTC)

Those in templates or those in statements? What should produce such a category? Matěj Suchánek (talk) 18:40, 11 May 2017 (UTC)

A simple Petscan query suffices here (diff between Category:All Properties and Category:Properties with constraints including sub categories). —MisterSynergy (talk) 18:59, 11 May 2017 (UTC)

Currently, the one's in the template produce the warnings that are served with the script. In a perfect world I would like that all constraint management is done via the statements but currently the working constraints are in the templates. I think there are some properties where nobody set the constraints via templates and it would be good to have a list of them as long. ChristianKl (talk) 20:35, 11 May 2017 (UTC)

Informal versions of given names; nicknames

Bill Thompson (Q4911143) officially has the given name "William", but is known to everyone by the short form of that name, "Bill" (just as my passport and my aged aunts call me "Andrew").

In Bill's case, I have found two ways to record this.

• given name (P735) = William (Q12344159); qualified nickname (P1449) = "Bill" (as an en-GB string)
• P735 = William (Q12344159), with a second P735, value Bill (Q18245781), ranked as referred, and qualified as instance of (P31) = nickname (Q49614)

Which is preferable, and why - or is there a better way to do it?

To me, the former is more logical, but the latter, using an item rather than a string, is more structured.

Another option would be the former, but with short name (P1813) as the qualifier. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:59, 11 May 2017 (UTC)

"Count" link at P1449 claims 30 times more uses as direct item-statements;

I would avoid values of qualifiers for basic information, this way queries would need only direct property relations (wdt:)

P735 is for passport name, nickname for anything else. d1g (talk) 06:44, 12 May 2017 (UTC)

What seems most logical is to use Andy Mabbett's first approach when both the formal name and the nickname are known. I would save the structure one given name (P735) for the nickname value ranked as preferred, and qualified as instance of (P31) = nickname (Q49614), for the case of a person who's nickname is known, and known to be a nickname, but the formal given name is not known. This, of course, would be rare because if you knew enough about a person to be sure the name was a nickname, you'd probably also know the formal name. Jc3s5h (talk) 11:04, 12 May 2017 (UTC)

Instance of nickname??? Wouldn't it be more appropriate to qualify this with has use (P366)? So the Bill is qualified as "Use" nickname. Noting that I can also see the deprecated P794 (P794) utilised.  — billinghurst sDrewth 23:47, 16 May 2017 (UTC)

Join the next cycle of Wikimedia movement strategy discussions (underway until June 12)

Message is available for translation on Meta-Wiki

The Wikimedia movement strategy core team and working groups have completed reviewing the more than 1800 thematic statements we received from the first discussion. They have identified 5 themes that were consistent across all the conversations - each with their own set of sub-themes. These are not the final themes, just an initial working draft of the core concepts.

You are invited to join the online and offline discussions taking place on these 5 themes. This round of discussions will take place between now and June 12th. You can discuss as many as you like; we ask you to participate in the ones that are most (or least) important to you.

Here are the five themes, each has a page on Meta-Wiki with more information about the theme and how to participate in that theme's discussion:

• Healthy, Inclusive Communities
• The Augmented Age
• A Truly Global Movement
• The Most Respected Source of Knowledge
• Engaging in the Knowledge Ecosystem

On the movement strategy portal on Meta-Wiki, you can find more information about each of these themes, their discussions, and how to participate.

Posted by MediaWiki message delivery on behalf of the Wikimedia Foundation • Please help translate to your language • Get help

21:08, 16 May 2017 (UTC)

Can anybody explain what "Auxiliary data matcher" is, and how it works?

I've seen seen this user making very helpful "manual matches" in Mix'n'match, obviously by checking which items already exist in Wikidata by looking the catalog id in the values of the according property. However, I cannot determine a pattern as to when the user/bot runs, and on which catalogs. Particularly, I cannot understand why it runs regularly on catalog #432 (and matched 4634 items), but less so on #431, where it matched only 3776 items. This is kind of sad, because #431 is the catalog I actively work on. #432 is an exact, stale copy (differing only in the descriptions), yet it has hundreds of matches #431 lacks. I could not find any documentation about User:Auxiliary data matcher. It would be great if somebody can shed some light? Perhaps @Magnus_Manske:? -- Jneubert (talk) 08:49, 17 May 2017 (UTC)

There are two "modes" here. One is if an item on Wikidata has been changed to use the respective property, and that property value exists in mix'n'match in the respective catalog, it will be auto-associated. A second mode is that, on mix'n'match import, a third-party property (e.g. VIAF) has been associated with a mix'n'match entry, and an item has that respective third-party property/value set; this will lead to a match as well. --Magnus Manske (talk) 09:57, 17 May 2017 (UTC)

Script and API module for constraint checks

Hello there,

In the past few months, the development team has mentored a student, Olga, to help us developing a user script that displays the constraints violations on the item pages.

To use the script, add the following line to your user/common.js:

mw.loader.load( '//www.wikidata.org/w/index.php?title=User:Jonas_Kress_(WMDE)/check_constraints.js&action=raw&ctype=text/javascript' );

You will see an icon next to violations. When you click it you will see the full report.

This script is based on a new API module for constraint checks that one can use to check constraints on items and statements. At the moment, the constraint checks are only derived from the property discussion page constraint templates, not directly from statements. They are then stored in a database table. We are running a script to update this table every now and then or when you ask for it. Also note that some constraint checks are disabled (for example the format check). In the future we will support adding and updating constraints on property statements and we will implement support for some constraints that are currently still missing.

If you try it, feel free to give us feedback! You can also add comments or subtasks on Phabricator (see the ticket for the API module and the user script).

If there is no major disagreement, we would like to turn this script into a gadget in the next days.

Thanks go to Olga and all the developers that helped her providing this new feature :)

See also:

• API Documentation
• Self-documentation for the module

Cheers, Lea Lacroix (WMDE) (talk) 12:45, 27 April 2017 (UTC)

Seems promising but it doesn't work for me. On Saint-François-de-Sales Church (Q962691) for example, I see no icon for messes.info Catholic church ID (P1644), while there is two values... — Ayack (talk) 21:11, 27 April 2017 (UTC)

Some bugs:

• Qualifier constraint violation icons are placed on every qualifier of any statement that has any violating qualifier, rather than just the particular violating qualifier itself. This is very confusing.
• "Target required claim" messages don't say what the missing statement is, "Qualifiers" messages don't mention which qualifiers are accepted or prohibited, "Range" messages don't say the minimum/maximum, and "Value type" messages don't say what value type is required.
• Some properties such as P570 have the text "⧼wbqc-violation-message-range-parameters-needed⧽" shown as the message.
• Upon clicking "edit" for a statement and then clicking "cancel", qualifier-level violations disappear but statement-level violations remain.
• The "Value only" constraint does not work.

--Yair rand (talk) 21:23, 27 April 2017 (UTC)

@Ayack: Still no icon for messes.info Catholic church ID (P1644)? I see one now.

@Yair rand:. Thank you very much for your reports. Do you have some examples for your points #1 and #5? About the content of the messages, we're currently working on it to improve them. Lea Lacroix (WMDE) (talk) 06:28, 28 April 2017 (UTC)

@Lea Lacroix (WMDE): The script doesn't work when the "Drag'n'drop" gadget is selected. When I deactivate it I see the icon. — Ayack (talk) 08:59, 28 April 2017 (UTC)

Thanks. This problem is tracked here. Lea Lacroix (WMDE) (talk) 08:41, 29 April 2017 (UTC)

https://www.wikidata.org/wiki/Q1364876 shows a type constraint for UBERON ID (P1554). This seems wrong. Maybe the tool doesn't check the subclass tree? ChristianKl (talk) 06:32, 29 April 2017 (UTC)

@ChristianKl: The tool does check the class hierarchy, but the type constraint demands that the item is a subclass of anatomical structure (Q4936952) or tissue (Q40397) – but median nerve (Q1364876) is an instance of nerve (Q9620). Either the constraint is incorrect, or the item should be changed to be a subclass of nerve (Q9620) instead – which one is right depends on how medical items are modeled in Wikidata, which I’m not familiar with. (Note that constraints are only imported from templates periodically, so if you update the constraint on the property talk page, it will take a while until it takes effect.) --Lucas Werkmeister (WMDE) (talk) 08:51, 29 April 2017 (UTC)

It would be nice to have a button to fix the issue that the constraint is about. For the Inverse Constraint there could be a one-click solution to create the inverse statement. ChristianKl (talk) 06:09, 30 April 2017 (UTC)

•  Comment Good work.
  • I just notice a problem that I hadn't noticed on Special:ConstraintReport, but was already present there. The text on constraint violations for required properties constraints is misleading. On property talk pages, the templates for Ivan A. Krestinin's bot it reads "Item: items with this property should also have .." while the script displays "This property must only be used when there is another statement using the property defined in the parameters.". We should keep to closer to the words of the templates.
  • I think the "required qualifier constraint" fails.
  • Maybe the color of the icon could be different depending on whether it's a mandatory constraint that is broken or not.
    A question mark for non-mandatory constraints and the (current) exclamation mark for mandatory constraints could do.
  • If the constraints have parameters (e.g. item constraints), maybe the parameters could be larger then the explanatory text and placed above that.
  • For single value constraints on identifiers, maybe need to re-word it: if an item does have several identifiers, these should obviously all be included. Presenting these as violations might lead people to remove one or the other.
  • BTW could you re-load the constraint definitions?
    --- Jura 03:56, 1 May 2017 (UTC)

I absolutely love this feature! There is only one thing that I find slightly odd: some constraints take parameters, and the value of these parameters is not integrated in the description of the conflict. These values are replaced by a phrase such as "the property defined in the parameters", and the actual value of the parameter is displayed underneath. I find this very hard to read. For instance, instead of:

Conflicts with

This property must not be used when there is another statement using the property defined in the parameters.

property: headquarters location

I would rather like to read:

Conflicts with

This property must not be used when there is another statement using the headquarters location property.

I do understand that it's harder for you to produce this output (this probably requires some changes in the API, and makes translation a bit harder), but I think this would be really worth the effort. − Pintoch (talk) 22:30, 2 May 2017 (UTC)

• Including it in the text might make localization and wordings based on mandatory/non mandatory status more difficult.
  --- Jura 05:20, 8 May 2017 (UTC)
  • @Pintoch: We’re working on the messages (here’s a checklist), it’ll just take a while to update all the messages and then for the changes to be deployed. Once that’s done, we plan to remove the parameters from the UI altogether, since they will no longer be necessary. --Lucas Werkmeister (WMDE) (talk) 14:44, 12 May 2017 (UTC)
    Tracked in Phabricator
    Task T164354
    

Single value: this property generally contains a single value.

Exceptions are possible as rare values may exist. Exceptions can be specified using exception to constraint (P2303).
List of violations of this constraint: Database reports/Constraint violations/Project chat/Archive/2017/05#Single value, SPARQL

Item: items with this property should also have coordinate location (P625)

Exceptions are possible as rare values may exist. Exceptions can be specified using exception to constraint (P2303).
List of violations of this constraint: Database reports/Constraint violations/Project chat/Archive/2017/05#Item P625, SPARQL. Count by value

• @Lucas Werkmeister (WMDE): To allow us to compare, could you link the equivalent messages from the above two constraints. In English, it's currently:
  1. "Single value: this property generally contains a single value."
  2. "Item: items with this property should also have coordinate location (P625)"

If there is different wording for mandatory constraints, please link that as well.
--- Jura 08:05, 14 May 2017 (UTC)

@Jura1: The messages are here and here, and currently read:

• This property must only have a single value. That is, there must only be one claim using this property.
• This property must only be used when there is another statement using the property defined in the parameters.
• This property must only be used when there is another statement using the property with one of the values defined in the parameters.

“Single value” doesn’t have any parameters, so I wasn’t going to touch it. “Item” has parameters, and I want to improve it, but I haven’t gotten to it yet. You can look at Change 353569 to see the proposed improvement to the “Conflicts with” message, which is similar.

We don’t distinguish between mandatory and non-mandatory constraints yet (see T164254), but once we do, I think the plan is to display the constraint differently, not to use a different message. --Lucas Werkmeister (WMDE) (talk) 08:44, 15 May 2017 (UTC)

@Lucas Werkmeister (WMDE): Ok. Sounds good. For consistency with the current approach, would you switch the texts to the ones at Module:I18n/constraints?
--- Jura 11:02, 15 May 2017 (UTC)

@Jura1: I hadn’t seen that before… where is it used? But as far as I can tell, those texts wouldn’t allow us to include the parameters in the message… --Lucas Werkmeister (WMDE) (talk) 12:23, 15 May 2017 (UTC)

@Lucas Werkmeister (WMDE): I'm sure you probably did as I think it's the text currently in use at Wikidata (on property talk pages).
--- Jura 17:34, 15 May 2017 (UTC)

@Jura1: ah, then I just didn’t know the connection :) --Lucas Werkmeister (WMDE) (talk) 09:09, 16 May 2017 (UTC)

@Lucas Werkmeister (WMDE): can you sync the descriptions in your code with these? Do you need help with this?
--- Jura 06:25, 18 May 2017 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ @Jura1: Well, there are several problems with them…

• The parameters for these messages are totally different. Sometimes, they contain %s, which can be replaced with an actual constraint parameter, but sometimes they instead specify “[item A]” or “[item B]”, which is impossible to replace (because it’s localized).
• In some cases, I don’t think the messages are suitable. For example, the “one of” constraint: “value must be one of the specified items. Please expand list if needed.” (emphasis added) – the last part makes sense in a template, but really doesn’t make sense to display in the constraint report, because it’s not at all obvious to the user where this list resides (and in fact that will soon change, once we import constraints from statements).
• As far as I understand, copying the existing translations (which would be the main benefit of using these messages) is impossible – if I add them to the i18n files in the code, translatewiki.net doesn’t import them, and they’re erased the next time translatewiki.net updates the i18n files.

I can compare the messages with the ones we currently have and try to carry over any improvements, but I don’t think using those messages directly is possible, sorry. --Lucas Werkmeister (WMDE) (talk) 08:52, 18 May 2017 (UTC)

@Lucas Werkmeister (WMDE): Sounds good, I think it's sufficient if you try to keep the (English) wording from Module:I18n/constraints and adapt it for variables where you want to use them.
--- Jura 18:04, 18 May 2017 (UTC)

@Jura1: Okay, thanks. I’ve updated all the messages now, by the way, so you can see them all here, and they’re also available for translation on translatewiki.net. I think they’re probably going to be deployed sometime next week. --Lucas Werkmeister (WMDE) (talk) 09:49, 19 May 2017 (UTC)

Properties for circumstances of death

I'm trying to work out how we can say that Richard Lyons (Q7327521) was killed during the Peasants' Revolt (Q498871). place of death (P20) is for geographical locations, not contextual ones; killed by (P157) is for people; cause of death (P509) for the "medical circumstances", manner of death (P1196) for a broad classification (accident, suicide, etc). There doesn't seem to be one we can use for events - or am I just missing something obvious? Andrew Gray (talk) 12:31, 17 May 2017 (UTC)

significant event (P793) with qualifiers? ArthurPSmith (talk) 14:39, 17 May 2017 (UTC)

significant event (P793) as a qualifier? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:56, 17 May 2017 (UTC)

What would that mean? What general meaning would P793 have as a qualifier? --Yair rand (talk) 20:28, 17 May 2017 (UTC)

I am with Pigsonthewing's suggestion. Adding it would indicate that the death was associated with an event. Seems similar to how someone would have won an award for valour, eg. a military cross awarded resulting from the Battle of Long Tan (Q259526). Having it unassociated with the death has two separate data points.  — billinghurst sDrewth 22:27, 17 May 2017 (UTC)

What kind of association does it indicate? How could one use this data? It seems really unstructured. --Yair rand (talk) 03:14, 18 May 2017 (UTC)

It is inter-relating a death with an event. Francis Sheehy-Skeffington (Q5482453) executed on the day directly related to Easter Rising (Q193689). Putting the deaths against the item for the uprising would be less relevant, however, show me deaths related to the event is relevant, especially where memorials exist listing names. Same for things like 9/11, Eureka Rebellion, +++ and the lists of events where an assignation is quite pertinent.  — billinghurst sDrewth 11:46, 18 May 2017 (UTC)

"is related to" is not structured data. Specific properties need to show specific relationships. --Yair rand (talk) 16:01, 18 May 2017 (UTC)

statement is subject of (P805) is another idea. Matěj Suchánek (talk) 07:18, 19 May 2017 (UTC)

Inscription (P1684) asks for a mandatory language which can only be a language with a language code?

screenshot when adding a language to an inscription

I tried to add the language Numidian (Q35761) to an inscription (P1684) which wasn't possible since it doesn't have a Wikipedia or a language code? I'm not sure what the criteria are. I don't think this restriction is very sensible since inscriptions are often in an old language (like Numibian or Etruscan) and will never have their own Wiki.

Couldn't we change it somehow, so a language with an item is sufficient? --Incabell (talk) 19:55, 17 May 2017 (UTC)

Having it's own Wiki isn't necessary for a language to be added to Wikidata but every language still has to be added individually. You can file a phabricator task to get the language added. Given that it has a ISO 639-3 code that should be possible. ChristianKl (talk) 22:17, 17 May 2017 (UTC)

@Incabell: For a monolingual texte, it's more simple You have just to follow Help:Monolingual text languages. --Fralambert (talk) 03:53, 18 May 2017 (UTC)

Cool! Thanks so much to both of you! I will do that :) --Incabell (talk) 09:18, 19 May 2017 (UTC)

SPARQL Code to find category items (instance of (P31)) that do not have sources that have an EN source

I know nothing about SPARQL. Please help. PokestarFan • Drink some tea and talk with me • Stalk my edits • I'm not shouting, I just like this font! 23:12, 17 May 2017 (UTC)

What do you mean by sources? If it is Help:Sitelinks, then:

Wikidata:SPARQL_query_service/queries/examples#Wikimedia_projects d1g (talk) 08:23, 18 May 2017 (UTC)

@D1gggg: By sources I mean references to items of properties. PokestarFan • Drink some tea and talk with me • Stalk my edits • I'm not shouting, I just like this font! 20:39, 18 May 2017 (UTC)

Use WD:RAQ and explain it more in detail there. Matěj Suchánek (talk) 07:25, 19 May 2017 (UTC)

How to say "located on $island" and/or "located in $X archipelago"?

I'd like to find a way of easily knowing which island/archipelago/state a telescope is located in for use in location maps in en:Template:Infobox telescope, so I'm trying to find a good way of saying "Telescope is on the island of X". It seems that setting country (P17) to, for example, Hawaii (Q782) is generally frowned upon (in favour of links to United States of America (Q30)), even though the aliases for that property include "state". The next level down is located in the administrative territorial entity (P131), but the definition of that seems to be quite a mess, with all sorts of administrative areas underneath it (I tend to go for the smallest one, which I think is the norm, but that's normally city-level). There is location (P276), which is a bit more confused (I thought this was the catch-all, but it actually looks like it's only for movable objects - in which case I've erroneously added quite a few locations using this property so far...).

As far as I can see, we don't have a property for "located on island", or even "located in state" - would it be worth proposing these, or is there a better way to do this? Thanks. Mike Peel (talk) 06:37, 19 May 2017 (UTC)

We have located in/on physical feature (P706). Shinnin (talk) 07:01, 19 May 2017 (UTC)

@Shinnin: Thanks! Isn't that more "on Mauna Kea (Q131230)" rather than "on Hawaii (Q68740)", though? Or is it intended for both cases? Thanks. Mike Peel (talk) 16:10, 19 May 2017 (UTC)

@Mike Peel: the reason that country includes state is I believe for the variation "nation state".

Administrative regions have been through many iterations of names, and it is cascading as best I can tell; something can reside in the unit of a state/county/... or something smaller council/borough/... it is not inaccurate. One could even say that places could fall under multiple administrative locations, as a local council, and a state government as both impose their respective administrative requirements. So I would have used P131.  — billinghurst sDrewth 17:04, 19 May 2017 (UTC)

Hackathon & documentation sprint

Hello all,

Just as a reminder, the Wikimedia hackathon takes place this weekend. A lot of people are currently hacking around the Mediawiki code and other stuff, in the venue in Vienna or remotely.

There is a track dedicated to documentation, where people will try to improve the documentation for our projects, make it updated, more understandable, more user-friendly. We plan to work especially on the beginner user doc for Wikidata, the Query Service and the constraint reports, Wikibase installation... and other stuff.

So, quite a lot of things should happen around the Wikidata: and Help: pages in the next days. If you want to join our taskforce, feel free to ping me! Lea Lacroix (WMDE) (talk) 10:35, 19 May 2017 (UTC)

Update about WikidataCon

Hello all,

Here are some new information about the WikidataCon event, that will take place in Berlin on October 28th-29th.

We just nominated the members of the two committees that will have an important role in the organization of the event. Thank you very much to all the applicants!

• Program committee: andrawaag, Spinster, Masssly
• Scholarship committee: Micru, Venuslui

For all the other topics, you can freely join the volunteer groups, discuss on the talk page, suggest ideas.

I'm also glad to give you an overview of the next steps for the WikidataCon:

• Scholarship application process: June 8th-July 31st
• Call for submissions for the program: June 12th-July 31st
• Registration opening: June 19th (two other ticket releases are planned in August and September)
• Selection of the scholarships recipients: August
• Review of the submissions for the program: August
• Announce of the program: September 1st

This dates can still evolve, I'll keep you informed. For more information, check Wikidata:WikidataCon 2017.

Last but not least: in order to be able to provide scholarships, and help people who can't afford the travel and accommodation to participate, we requested a grant from the WMF. This grant will be reviewed soon, and you can help by endorsing the project. For this, you can add a message at the bottom of the grant submission. This would be very useful for the project!

Thank you very much, Lea Lacroix (WMDE) (talk) 14:07, 19 May 2017 (UTC)

Grant proposal for WD editor for Wikipedia

My proposal about the Wikidata editor directly from Wikipedia is at the last stage of consideration. In this regard, I would like to ask you to express your ideas and concerns, if any, on the discussion page. The idea is very close to the prototype of the Wikipedia editor by Wikidata team, but it is possible that the result will be different. —putnik 20:41, 19 May 2017 (UTC)

Expertise needed - armor

Wikidata's Fashion project could use help from an expert on armor (ancient, medieval, Japanese, or modern). Our worklist of armor and helmet items is here. Please help if you can! - PKM (talk) 20:51, 19 May 2017 (UTC)

What is the property for the date when a ship collides, or when an object is destroyed?

Please answer and ping me as {{Ping|PokestarFan}}. Thanks. PokestarFan • Drink some tea and talk with me • Stalk my edits • I'm not shouting, I just like this font! 00:54, 20 May 2017 (UTC)

@PokestarFan: For ships look at Wikidata:WikiProject Ships/Properties #Significant events. - Kareyac (talk) 05:43, 20 May 2017 (UTC)

Property for pattern of a textile

Do we have a property that should be used to indicate the <pattern or motif> of a woven textile or other patterned object? Values might include things like paisley, palmette, gryphons <in> roundels, and so on (any instance of pattern (Q2083958) or one of its subclasses). - PKM (talk) 01:13, 20 May 2017 (UTC)

depicts (P180) looks to be the "nearest" one. - Kareyac (talk) 05:54, 20 May 2017 (UTC)

Q693514

Can somebody delete all descriptions or add new ones for the tv series? The film is now under Q3187345. Queryzo (talk) 07:53, 25 May 2017 (UTC)

 Done using MediaWiki:Gadget-dataDrainer.js. Matěj Suchánek (talk) 12:00, 26 May 2017 (UTC)

This section was archived on a request by: Matěj Suchánek (talk) 12:00, 26 May 2017 (UTC)

Russian social media sites: properties needed?

The two websites with the most values for website account on (P553), for which we as yet have no properties, are Odnoklassniki (Q1123836) and My World@Mail.Ru (Q4299858). Would a Russian speaker please look at whether we should have properties for them (i.e. how well-used they are on ru.Wikipedia and other projects), and if so how the formatter URLs should be constructed? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:05, 15 May 2017 (UTC)

Not a Russian-speaker, but Odnosklassniki is at least very popular in Russian speaking countries, so I would support creating a property for that site. Stryn (talk) 16:42, 16 May 2017 (UTC)

Yes, we should have properties for both of them.--Ymblanter (talk) 11:13, 20 May 2017 (UTC)

Websites with no URL for 'official website'

My thanks to MisterSynergy for this query which finds instances of website (Q35127) or its subclasses, with no official website (P856).

It currently returns 3118 items. These can be broken down into various types:

• Items that are wrongly described as websites
• Websites with no URL in P856 (which should have the URL added)
• Defunct websites, with no URL in P856

In the latter case, I suggest the URL should be added, and an "end date" qualifier, if necessary with an "unknown" value. How, then should, say, an archive.org URL be included?

Also, how can the addition of URLs best be automated? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:03, 20 May 2017 (UTC)

Movie title vandalism

Could somebody please undo all changes by User Arpyia[8]? The user added massively non-french titles as "french" lables to movies. This should be reverted as soon as possible, otherwise the complete lable section gets useless. --88.76.216.204 22:05, 16 May 2017 (UTC)

And by the way, the system should automatically prevent users from doing so many edits in such a short period of time unless they have special permission if Wikidata wants to be taken seriously. --88.76.216.204 22:12, 16 May 2017 (UTC)

If the movie doesn't have a French title, it is okay to use the movies original title as French label. All movie items edited by User:Arpyia seem to have no French title. --Pasleim (talk) 22:28, 16 May 2017 (UTC)

It is NOT okay! If a movie has no french title, then there should be no title entered. Otherwise you would have to add 5,000 "titles" for 5,000 languages on every movie page if the movie has no title in these languages. It would only be okay if the movie had been released under that title in France. But this is not the case. --88.76.216.204 22:38, 16 May 2017 (UTC)

Don't mistake labels for title (P1476) statements. Statements are the part on Wikidata which hold the referenceable information. If a movie was never released in a French speaking country, one should not add title (P1476) with language code "fr". In contrast, the label does not have to be the "official" name. It's main purpose together with alias and description is to find and identify an entity. Users with the French UI should also be allowed to search for German movies. --Pasleim (talk) 23:01, 16 May 2017 (UTC)

But where is this "title (P1476)" on movie pages for translated titles, like a "real" French title for a movie that has been released in France? I have never seen it. And why wouldn't you find a lable, when it is the same as something that already has been entered? And what is the use of lable if it contains the same value in hundreds or thousands of languages? And why says it "enter a lable in French" then? --88.76.216.204 23:13, 16 May 2017 (UTC)

For example at Q160060 the French title "Le Discours d'un roi" under which the movie actually was released in France is only added as "lable" not as "title". This is the case for all movie pages. --88.76.216.204 23:16, 16 May 2017 (UTC)

If that was the released title and the field is absent, then add it, and reference it, it is a wiki.  — billinghurst sDrewth 04:06, 17 May 2017 (UTC)

The film wasn't shot in France, so in Wikidata we would have a separate item for translations.

When you say "it never had a French title" you mean a "translation into French".

The King's Speech (Q160060) is a work item, not edition item; see Wikidata:WikiProject Books.

The biggest point is that translations aren't made by the same crew who shot a film.

Exact details about title (P1476) are under discussion now. d1g (talk) 04:16, 17 May 2017 (UTC)

"The film wasn't shot in France, so in Wikidata we would have a separate item for translations.": But you don't have it, do you? For years everybody put these translation into the lable fields

"When you say "it never had a French title" you mean a "translation into French".": No, that's not what I mean. User:Arpyia added the German title as French lable to German TV movies who never have been shown in France or any French speaking country. If this would become default behaviour, you would have thousands of identical lables for different languages for example for a German TV movie that never has been shown outside of Germany. If it where correct to add the German title as French lable, then it would also be correct to add the German title as Chinsese lable, Hindi lable, {name any language} lable. This would make no sense.

"Exact details about title (P1476) are under discussion now.": What does this mean? Where is this discussion taking place? --92.73.28.26 08:43, 17 May 2017 (UTC)

I would like to know how it is going to be now. Will there again be a way to find out the official title of movies in a different country/language or will there be just a big "lable" mess with anything in it? Until the mass edits of User:Arpyia the lables more or less contained these titles and Wikidata was useful to find them out. No this is not the case anymore and I would really like to know a solution. --92.73.28.26 22:06, 17 May 2017 (UTC)

I will try to clean-up the edits over the weekend.
--- Jura 06:36, 18 May 2017 (UTC)

Thank you, it seems to be fine again now. --92.213.10.115 15:30, 21 May 2017 (UTC)

Circular referencing

Is it possible for constraint violations to be used to discourage referencing Wikidata on items (e.g. references imported from Wikimedia project (P143) → Wikidata (Q2013), reference URL (P854) → https://www.wikidata.org/*), and should this generally be done? (See also this discussion on enwiki.) Jc86035 (talk) 10:18, 19 May 2017 (UTC)

I think inferred from (P3452) would have been the correct way to use this kind of Wikidata-reference. Just as imported from Wikimedia project (P143) that property is actually very useful to indicate provenance, although they are of course not an external source. —MisterSynergy (talk) 10:31, 19 May 2017 (UTC)

Bonus:

#claims referenced by P143:Q2013 (imported from: Wikidata)
SELECT ?item ?property ?value {
  ?property a wikibase:Property; wikibase:statementProperty ?statementProperty; wikibase:claim ?claim .
  ?item ?claim [ ?statementProperty ?value; prov:wasDerivedFrom [ pr:P143 wd:Q2013 ] ] .
}

Try it!

We have around 10.000 claims with imported from Wikimedia project (P143): Wikidata (Q2013) in references. My query for reference URL (P854) with Wikidata unfortunately times out. —MisterSynergy (talk) 10:42, 19 May 2017 (UTC)

Using simply Special:WhatLinksHere/Q2013 also yields results (but not the URL references). Steak (talk) 10:48, 19 May 2017 (UTC)

@MisterSynergy, Steak: (Also pinging Fram from the enwiki discussion, and VIGNERON, who added a lot of these to railway stations with QuickStatements.) Should the references be summarily removed, or should they be kept until they're referenced with other things? Jc86035 (talk) 11:14, 19 May 2017 (UTC)

• Parts of it might be moved from imported from Wikimedia project (P143) to inferred from (P3452), but this requires an individual manual check of each case (i.e. this is not a bot-able task).
• “No internal references” is a Wikipedia policy, not a Wikidata policy. Both policies are correct due to the inherent differences of both project, thus they are not a contradicting. We definitely keep those references.
• If you find external references for the claims in question, add them of course and still keep the internal ones. The “references” section indicates data provenance at Wikidata, not “external references” only.
• Wikipedias that use data from Wikidata can accept or refuse references or even entire claims based on the nature of a reference. If you only want to use “externally referenced” data, you compare provided references against Help:Sources and only use claims and/or references that you are happy with. If imported from Wikimedia project (P143)/inferred from (P3452) are not valid sources in your project (as typically the case for Wikipedias), you can just omit this claim entirely if you want to.
• Another common misunderstanding of Wikipedia users is about the data that is collected here: we want referenceable data, even if it is known to be incorrect. It is thus not a problem if we import “crappy” data from Wikipedias, as long as we tidy it up at some point (using ranks and references). It is very useful in this case to have information about data provenance (via imported from Wikimedia project (P143)), and tons of corrections flow back to Wikipedias which were otherwise not at all able to properly improve their “crappy” information.

—MisterSynergy (talk) 11:31, 19 May 2017 (UTC)

In my opinion the self-references should be removed. Wikidata cannot be a source for itself. If we would accept this, the door would be wide open for any kind of true unsourced data by simply stating "imported from Wikidata". It is not clear what the source is in the statements where there is already put Wikidata as reference. At least the exact item would then be required, but even then I would oppose to put Wikidata as a source. Steak (talk) 11:38, 19 May 2017 (UTC)

For the sake of data reusers I think it's good to have these references be made via inferred from (P3452). ChristianKl (talk) 12:13, 19 May 2017 (UTC)

Hi (thanks Jc86035 for the ping),

First of all, I *did* imported the data from Wikidata, so I don't really see the problem with saying it; would you rather prefer me importing data without saying where I took them from?

Then, I did it for a very specific case (adjacent station (P197) wich is a symetric property, a property where values are both trivial to find and uneasy to reference) and after some checking (for example check if the two stations are less than 100 km apart, then 50 km apart, etc. I found - and corrected - some errors but very few). Finally, the problem seems simple to me either: the data on the first item is correct, so reimporting it on other item is correct too, the data is wrong and reimporting give it twice more chance to be corrected.

I wonder if the problem is really with the property (imported from Wikimedia project (P143), or inferred from (P3452)) or with the value (Wikidata (Q2013)). Maybe I would be clearer and better to put:

⟨ Anieliny train stop (Q3144464)    ⟩ adjacent station (P197) ⟨ Samostrzel (Q303880)    ⟩
imported from Wikimedia project (P143) ⟨ Samostrzel (Q303880)    ⟩

). What do you think?

Cdlt, VIGNERON (talk) 13:45, 19 May 2017 (UTC)

I do think that would be clearer.

When Wikipedia imports values from Wikidata, they want to import only those values that have external references. As a result it's helpful for the purposes of Wikipedia when we use inferred from (P3452) for internal references. This allows us to have data provenance while at the same time allowing third-party data users like Wikipedia to see easily which of our claims have external references. ChristianKl (talk) 18:33, 19 May 2017 (UTC)

I agree; use of inferred from (P3452) instead of imported from Wikimedia project (P143) would be better here, and it should point to the specific item rather than to Wikidata (Q2013). Given the fact that the results set offered above is very homogeneous (almost exclusively P197 claims are affected), this might be a bot-able task after all. @VIGNERON: the way you used the {{Claim}} templated indicates a qualifier, but please use a reference. Just in case that’s not clear ;-) —MisterSynergy (talk) 07:09, 20 May 2017 (UTC)

Does everybody agree and can I start a Wikidata:Bot requests? VIGNERON (talk) 11:08, 20 May 2017 (UTC)

After reading this discussion, I asked this query to be built. It shows items that have claims supported by the English wikipedia, but the item itself doesn't link to the english wikipedia. What to do with these items? (1000+ items) --Q.Zanden questions? 17:55, 20 May 2017 (UTC)

An item doesn't need to have a link to the English Wikipedia to be sourced by the English Wikipedia. Information could be imported from some list. It could also have been imported from an item that's now deleted. There isn't something we have to do about those items. ChristianKl (talk) 18:36, 20 May 2017 (UTC)

Correct. For example, I imported the years where titles were approved from the grandmaster list in german wikipedia, but by far not all of the grandmasters have an article. Steak (talk) 16:19, 21 May 2017 (UTC)

@ChristianKl, Steak:, but then I would use reference URL (P854) or something like that... Wouldn't that somehow be more logical? Or imported from Wikimedia project (P143) with qualifier of (P642) Q(article from the wikipedia-page)? Q.Zanden questions? 17:05, 21 May 2017 (UTC)

@Q.Zanden: reference URL (P854) is not supposed to used for data from Wikimedia projects but for external sources. This allows Wikipedia to import all data that has an external source. Using "of" would be nice but currently that's not the practice for a lot of imported data. ChristianKl (talk) 20:20, 21 May 2017 (UTC)

Are montages acceptable as an item's image (P18)?

I just found out that some items have a montage (example) as their image (P18) field. Is it acceptable?

I would argue that a montage is not acceptable, it is like hard-coded fake multiplicity where a single image is expected.

My app shows Wikidata items on a map and in lists. I want to use images as map pins and list thumbnails, it looks great for normal images, but for montages it just looks like a grid of indiscernible things.

Thanks! Syced (talk) 08:26, 17 May 2017 (UTC)

For montages we have the property montage image (P2716). So I would say image (P18) should not be used for montages. --Pasleim (talk) 08:30, 17 May 2017 (UTC)

This seems sensible. I wonder if we can run some kind of a query to find all image (P18) which are in commons:Category:Montages and move them across? Andrew Gray (talk) 09:23, 17 May 2017 (UTC)

That would be great indeed! Anyone knows how to perform that? Thanks! Syced (talk) 10:32, 17 May 2017 (UTC)

Here's a start - after a bit of massaging of PetScan, this list is 1179 items which use an image in the Commons montages category (depth 7, not including "Before and after‎" or "Then and now"). It will include some that use it correctly in montage image (P2716), of course, but I couldn't figure out quite how to distinguish those. Hopefully a start for someone to work with! Andrew Gray (talk) 11:03, 17 May 2017 (UTC)

Bingo! This is a query for everything on that list which has P18 and does not have P2716. It's not perfect, but it gets 1095 results, so there's plenty to fix. Andrew Gray (talk) 11:14, 17 May 2017 (UTC)

Montages are often used in infoboxes, which rely on P18 - so please be careful when removing these, and check the cases where the montage is being used as an image on Wikipedias. Thanks. Mike Peel (talk) 06:39, 19 May 2017 (UTC)

If a community explicitly wants montages in their infoboxes, they should use montage image (P2716) and only fall back to image (P18) if there is none. In my opinion :-) Syced (talk) 06:51, 19 May 2017 (UTC)

Given that montage image (P2716) subclasses image (P18) I would expect that a template would still import the data. Does it work differently? ChristianKl (talk) 08:23, 22 May 2017 (UTC)

With the infobox templates I've worked on I've been linking specifically against P18. It is possible to change that to call P2716 first, and if that doesn't exist then it can fall back to P18. However, then there would need to be some way to say "the better image to illustrate this topic is not the montage", which is then getting complex ... it's easier just to let people set P18 to the best image to illustrate the topic, and just use that. Thanks. Mike Peel (talk) 08:27, 22 May 2017 (UTC)

Source statements for items syntesized from authorities - recommendations?

I'm going to import 1.3k items for economists, for which we have the GND ID (P227) and RePEc Short-ID (P2428) identifiers and information we can collect from a "join" of the authority files. (Of course we make sure that neither of the ids exist in WD so far.) The properties are:

• name (from GND, and, if different, alias from RePEc)
• English description ("economist" + the top level affiliations from RePEc)
• German description ("Wirtschaftswissenschaftler/in + biographic description field from GND)
• gender (from GND)
• occupation economist (Q188094) (from RePEc)
• birth and death data (from GND - not yet included in the examples)
• and of course the respective ID values

I'd like to make clear which information stems from which source and would be very grateful for hints how to approach this in compliance with established best practices. (Help:Sources seems not to deal with such tasks.) Please take a look at my first take (examples: Richard Donald Rogerson (Q29961563), Bruce D. Smith (Q29961579)). Since I could not find a way to source description field (which may contain important and possibly contradictionary information in German and English), I've added comments to the references of the respective external identifiers. Perhaps others have found a better way to provide that source information.

Comments on the quantity of information (too much? too less?) and on the wording are also very welcome. Jneubert (talk) 10:11, 17 May 2017 (UTC)

Vladimir Alexiev (talk) 11:59, 13 March 2017 (UTC) Jonathan Groß (talk) 17:52, 26 March 2017 (UTC) Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits Jneubert (talk) 13:47, 29 April 2017 (UTC) Sic19 (talk) 20:42, 12 July 2017 (UTC) Wikidelo (talk) 21:15, 8 May 2018 (UTC) ArthurPSmith (talk) 19:52, 22 August 2018 (UTC) PKM (talk) 19:40, 23 August 2018 (UTC) Ettorerizza (talk) 06:44, 8 October 2018 (UTC) Fuzheado (talk) 03:47, 19 December 2018 (UTC) Daniel Mietchen (talk) 16:30, 7 April 2019 (UTC) Iwan.Aucamp (talk) 21:48, 3 October 2019 (UTC) Epìdosis (talk) 23:49, 22 November 2019 (UTC) Sotho Tal Ker (talk) 00:52, 1 May 2020 (UTC) Bargioni (talk) 09:48, 02 May 2020 (UTC) Carlobia (talk) 14:34, 11 May 2020 (UTC) Pablo Busatto (talk) 03:22, 23 June 2020 (UTC) Matlin (talk) 10:53, 6 July 2020 (UTC) Msuicat (talk) 21:57, 27 August 2020 (UTC) Uomovariabile (talk) 10:04, 27 October 2020 (UTC) Silva Selva (talk) 17:21, 30 November 2020 (UTC) 1-Byte (talk) 15:52, 14 December 2020 (UTC) Alessandra.Moi (talk) 17:26, 16 February 2021 (UTC) CamelCaseNick (talk) 21:20, 20 February 2021 (UTC) Songceci (talk) 18:45, 24 February 2021 (UTC)]] moz (talk) 10:48, 8 March 2021 (UTC) AhavaCohen (talk) 14:41, 11 March 2021 (UTC) Kolja21 (talk) 17:37, 13 March 2021 (UTC) RShigapov (talk) 14:34, 19 September 2021 (UTC) Jason.nlw (talk) 15:15, 30 September 2021 (UTC) MasterRus21thCentury (talk) 20:22, 18 October 2021 (UTC) Newt713 (talk) 08:42, 13 March 2022 (UTC) Pierre Tribhou (talk) 08:00, 20 March 2022 (UTC) Powerek38 (talk) 17:21, 14 April 2022 (UTC) Ahatd (talk) 08:34, 4 August 2022 (UTC) JordanTimothyJames (talk) 00:54, 31 August 2022 (UTC) --Silviafanti (talk) 17:07, 14 September 2022 (UTC) Back ache (talk) 02:03, 1 November 2022 (UTC) AfricanLibrarian (talk) M.roszkowski (talk) 10:44, 4 January 2023 (UTC) Rhagfyr (talk) 19:36, 9 January 2023 (UTC) — Haseeb (talk) 13:10, 4 August 2023 (UTC) 沈澄心✉ 13:26, 15 November 2023 (UTC) MrBenjo (talk) 15:20, 23 April 2024 (UTC) S.v.Mering (talk)

Notified participants of WikiProject Authority control

Add complete references according to Help:Sources#Databases to claims. Labels and descriptions to not require references, they are just there to disambiguate. Whatever information you provide there, should also in some way appear in the statements. See Help:Label and Help:Description for details, also regarding style guidelines. —MisterSynergy (talk) 10:47, 17 May 2017 (UTC)

@MisterSynergy:: Thanks for the hints. "complete references" reccur on a retrieved (P813) with each stated in (P248), right? I'll happily add this. Jneubert (talk) 11:20, 17 May 2017 (UTC)

“Complete references” means that each and every (database) reference should contain stated in (P248), the database property with the identifier, title (P1476), and either publication date (P577) (if available) or retrieved (P813) (or even both). This is even the case when many “identical” references need to be added to the same item. I have edited Bruce D. Smith (Q29961579) to provide an example: Q29961579#P106. A publication date (P577) is not available for entries of this database, thus I omitted it. —MisterSynergy (talk) 11:35, 17 May 2017 (UTC)

Despite it would be technically easy to add all the statements you suggested, I hesitate to follow your advice here, for two reasons:

1. . it is perhaps information overload for people looking at this, as it duplicates information already provided and easy to look up in "External identifiers" section
2. . Storing this information redundantly may cause inconsistencies if the information at the external id property changes (think of: It turns out that a GND id falsly covers two distinct people, and the id value is split up). This should be updated eventually in the external identifier property, but very probably not in every reference, where it may be stored redundantly.

For these reasons, I would lean more to leaving it at stated in (P248) and retrieved (P813) in cases where the external identifier for the record is already given in the according property. Jneubert (talk) 13:48, 17 May 2017 (UTC)

• Information at Wikidata should be primarily machine-readable, and machines have a different view on things. We thus need to prepare our data accordingly, although this sometimes contradicts our intuition and Wikipedia experience. It is particularly important to understand that the Wikidata web frontend is not an interface for actual data users, unlike for Wikipedias with it’s human readership. The Wikidata web frontend is almost exclusively a tool for Wikidatians (editors), while machines “read” our data via the Query Service or Wikipedia templates.
• A typical scenario of data usage would be a Wikipedia infobox that pulls some data from the associated Wikidata item (say P106 of a person). It can directly use the information in the provided reference to deliver this reference to Wikipedia readers as well. Templates are designed to some extent along the guidelines of Help:Sources for reference processing. They do not look into other claims to find a reference, or parts of a reference. Wikidata was purposely designed to provide individual references for each and every individual claim.
• Inconsistencies among IDs etc. are being worked on by the use of constraint reports. Depending on the type of properties, we apply certain constraint on the values etc., and violations are reported on a daily basis by a bot. If for instance IDs change or are duplicated, we find the items using this ID listed on maintenance lists. It is thus not a problem to deal with this type of redundancy.

—MisterSynergy (talk) 14:19, 17 May 2017 (UTC)

Convinced! And thanks again for your kind explanations. Jneubert (talk) 15:10, 17 May 2017 (UTC)

The information in the description is different - providing only what is in the properties would mean to drop that information completely, because the correct wikidata items are not available to me in the automated process (and often would require intellectual research and perhaps creation of items for institutes etc.). On the other hand, the information is useful for disambiguation, in particular when the items are to be matched to other sources via Mix'n'Match. If we want to keep it, it seems appropriate to me to figure out a way to hint at its provenance. But of course, when we drop it, the issue does not occur. Jneubert (talk) 11:20, 17 May 2017 (UTC)

Labels are another issue - they may differ in the sources. The GND names follow a very strict set of rules (which sometimes leads to unfamiliar results). The RePEc names are provided by the authors themselves, so they may be closer to common use - but sometimes authors are sloppy, provide everything in lower or upper case, or whatever. So I've chosen to provide the GND name as label and the RePEc name (if different) as alias. I'd think that it may be helpful to provide this provincance information as well. Jneubert (talk) 11:20, 17 May 2017 (UTC)

There is no possibility to add a reference to labels and descriptions. An alternative would be to provide a useful edit summary which describes provenance. —MisterSynergy (talk) 11:35, 17 May 2017 (UTC)

Thanks for confirming my assumption that there are no references for labels and descriptions - I came up with using a comment as part of the reference in the external identifier statement for that reason. Using the edit summary would be another workarround. But is there any way to provide a custom edit summary with quickstatements2? (I've only seen log entries like Added [en] label: Bruce D. Smith, #quickstatements; batch #579 by User:JneubertAutomated) Jneubert (talk) 13:48, 17 May 2017 (UTC)

The comment (DEPRECATED) (P2315) property is highly disputed, as it holds unstructured data. It is currently nominated for deletion at Wikidata:Properties for deletion. There is also nobody looking for this piece of information in the qualifiers of the external-id property. I would just use the edit summary.

Custom edit summaries can be set by bot frameworks for sure. I did not work with QS2 yet, so I can’t help there. —MisterSynergy (talk) 14:27, 17 May 2017 (UTC)

@MisterSynergy: I've reformatted the threads a bit for better tracability - I hope that's ok.

Ok for me —MisterSynergy (talk) 14:27, 17 May 2017 (UTC)

New iteration: Example James R. Markusen (Q29963324) - with full references, removed comments (@MisterSynergy: Thanks for the hint at the deletion list!), and with dates of birth/death. Jneubert (talk) 16:31, 17 May 2017 (UTC)

Looks like an “okay” item for me, based on the fact that both databases do not provide more information (country of citizenship (P27) would have been nice). Regarding Wikidata:Notability I think that those items would be covered by #2 (“can be described using serious and publicly available references”, which both databases likely are). However, it would be great if someone else could confirm this. It would also be good if you can link the items to other items via claims. —MisterSynergy (talk) 20:08, 17 May 2017 (UTC)

I agree with the idea that both items are notable because both databases are serious and publicly available sources. It's also worth noting that there should be a bot approval request for the specific task of important those values. ChristianKl (talk) 10:18, 18 May 2017 (UTC)

Thanks for the hint, the bot request is already on it's way - support welcome :) Jneubert (talk) 18:21, 18 May 2017 (UTC)

For the example item James R. Markusen (Q29963324) GND has information on his country (P17) and given that country names are unique it should be possible to import them.

The GND field "Beziehungen zu Organisationen" can be mapped to affiliation (P1416). Given that GND identifies the organsiations with ID's and we have the ID's for many organisations in our database there should be no disambiguation problem. ChristianKl (talk) 10:18, 18 May 2017 (UTC)

I have considered importing country names and affiliations. However, that would require considerable effort (more than I can do now). Both come with issues: While the nationality is relatively reliable (changes seldom), it may be politically loaded in some cases (think Palestine). I'm quite sure, there are policies in Wikidata to deal with such cases, that may not be easy in an automated way, in particular if there should be subtle differences to the GND policies.

Affiliations are a very interesting use case (and extremely useful to have). The GND affiliations however may be out of date (they are not updated systematically). In most cases, the RePEc affiliations should be more current, because with RePEc's rankings for people and (indirectly) institutions, they have great incentives for authors to keep their data up to date. However, there are almost 14k of EDIRC institutions, three to four levels deep, and linked to RePEc authors and nothing else in the world. So that makes a full-fleged mapping project of its own.

The good news is, that any piece of data we skip now can be added easily later on. As a low hanging fruit, I plan to add GND date of birth and death to all "economists" in Wikidata which lack this information (and I will happily provide links to the code, so others can do the same for their domain of interest). Jneubert (talk) 18:21, 18 May 2017 (UTC)

While countries are politically loaded, it's not Wikidata's job to pass judgement of what happens to be a country. We report what other reliable sources say and if GND says that something is a country that's good enough for us.

Affiliation statements on Wikidata that aren't qualified with start time (P580) and/or end time (P582) don't "change". The statement means that there's was a time when the person was affiliated with the organisation, it's not a claim that they presently have that affiliation. ChristianKl

Hell NO! When we do not agree with the values of another source because they are wrong, we refuse them. We have done so in the past. When the GND has it wrong, we should not accept it. Exactly because we find that it is not reliable. Thanks, GerardM (talk) 19:10, 21 May 2017 (UTC)

(talk) 20:57, 18 May 2017 (UTC)

@ChristianKl: I've created an "todo" entry about supplementing items with properties from GND on the WikiProject Authority Control

I think GND's "Typ : Person (piz)" is a valid reference for claiming that the item is about a human. ChristianKl (talk) 10:18, 18 May 2017 (UTC)

@ChristianKl: Do you think I should add an explicit source statement to "human", too? Jneubert (talk) 18:21, 18 May 2017 (UTC)

Sources for "human" aren't as important as other sources but it's better if everything is sourced. Given that it isn't much effort and there are a lot of new entries that are created I would advocate sourcing it. ChristianKl (talk) 08:25, 22 May 2017 (UTC)

I copy the entries for the field "Andere Namen" into aliases. At least for German. ChristianKl (talk) 10:18, 18 May 2017 (UTC)

I was in doubt and finally decided against this, because GND has sometimes lots of alias names (e.g., Karl Marx (Q9061) - compare it's GND entry. Similar John Maynard Keynes (Q9317) / GND). Often variations are very slight, and in my eyes don't add much to findability, sometimes titles and other information are mixed in. What is worse, the variants don't come with a language tag (sometimes an indication in parantheses, appended in free text), so you would end up with, e.g., "German" aliases in all languages and scripts. I'm not aware of any method to filter these to some "relevant" subset. -- But perhaps that will be solved by somebody someday, and we could add such a subset of name variants later on. Jneubert (talk) 18:21, 18 May 2017 (UTC)

Yachtmaster (Q476246)

This item (sailor (Q476246)) is used as occupation for athletes competeting in the sport of sailing, but it's not a good choice. How can we fix this? --Cavernia (talk) 11:34, 20 May 2017 (UTC)

create a new item that's a better one to describe the occupation? If some of the language wiki's listed under sailor (Q476246) are better under the new item then you can move them too. ArthurPSmith (talk) 06:42, 23 May 2017 (UTC)

Sailor might appear as the correct item, but the meaning of this is seaman/mariner/seafarer, not an athlete competing in the sport of sailing. This confusion is also a case in Wikipedia, see Glen Foster. I've now created sailor, hopefully this will work better. --Cavernia (talk) 09:39, 23 May 2017 (UTC)

Height of actors and models

How to write 5 feet 6 inches? Capankajsmilyo (talk) 03:48, 21 May 2017 (UTC)

66 inches. You have to convert everything to use the smallest unit. --Edgars2007 (talk) 04:23, 21 May 2017 (UTC)

in measurements, 5 1 feet = 12 inches. MechQuester (talk) 13:46, 22 May 2017 (UTC)

MechQuester means 1 foot = 12 inches. For the height of people, you may also like to use metre (Q11573) or centimetre (Q174728). Jared Preston (talk) 15:52, 22 May 2017 (UTC)

I don't envy American, having to use these non-decimal conversions all the time ... but who would want to switch to the metric system anyway? You'd have to get a new set of tools! Jonathan Groß (talk) 06:07, 23 May 2017 (UTC)

Is Petscan working?

I tried logging into Widar. Afterwards, and after I refereshed the page, i still am ot "logged in". MechQuester (talk) 15:11, 22 May 2017 (UTC)

I logged out and logged in successfully You have authorized WiDaR to edit as Billinghurst. Congratulations! You can always log out here.  — billinghurst sDrewth 21:17, 22 May 2017 (UTC)

Same. You have authorized WiDaR to edit as PokestarFan. Congratulations! You can always log out here. PokestarFan • Drink some tea and talk with me • Stalk my edits • I'm not shouting, I just like this font! 21:51, 22 May 2017 (UTC)

Wikidata weekly summary #261

Here's your quick overview of what has been happening around Wikidata over the last week.

• Discussions
  • Open request for adminship: MisterSynergy, Queryzo, ChristianKl

• Events/Press/Blogs
  • Upcoming: Wahlsalon (event Codefor.de about election data): Kickoff on May 24th in Berlin
  • Upcoming: Open Science: the National Plan and you meeting for Dutch researchers: May 29th in Delft, The Netherlands
  • Current: WikiCite, May 23rd-25th in Vienna
  • Current: Met Open Access Artworks Challenge (May 15th-June 30th) help improving the data about the collection of The Metropolitan Museum of Art!
  • Past: Wikimedia Hackathon 2017, May 19th-21st in Vienna
  • Paper: Becoming Wikidatians: evolution of participation in a collaborative structured knowledge base (Alessandro Piscopo, Christopher Phethean, Elena Simperl)

• Other Noteworthy Stuff
  • The extension RevisionSlider is now available by default on Wikidata. Try it on a talk or project page!
  • Check out some updates about the WikidataCon
  • We reached Q30,000,000! Help us improve it and make a showcase item out of it.
  • An additional 10k article placeholders on Welsh Wikipedia will be made indexable in search engines

• Did you know?
  • Newest properties: Unifrance person ID, Unifrance company ID, IECIC 2015 ID, Songkick venue ID, BVMC work ID, secretary general, Nature preserve in North Rhine-Westphalia ID, PIM authority ID, PASE Domesday place, PASE Domesday person ID, channel number, signed form, CETS no., final event, programming paradigm, Bridgeman artist ID, BDCYL authority ID, Clochers de France ID, Global Trade Item Number, Unifrance film ID, Base biographique AUTOR ID, NNL work ID
  • Query examples:
    • Websites but without an official website
  • Plants with associated emoji (source)
    • Languages ordered by number of speakers (source)
    • HMDB identifiers with more than one item (source)
    • Painters with a lot of identifiers on Wikidata, but few Wikipedia articles (source)
    • Buildings with the greatest numbers of elevators (source)
    • Map of female writers on Wikidata by birthplace- generated by SPARQL query (source)
  • Newest WikiProjects: WikiProject Counter-Vandalism, WikiProject Documentation

• Development
  • Participating to the Wikimedia hackathon 2017, having a lot of interesting talks with users, designers, developers...
  • Helping improving the documentation for the property constraints, the beginner documentation about Wikidata, how to install Wikibase
  • Rewriting most constraint violation messages to include more useful information (phab:T164354, help update translations)
  • Working on a tree view for the Query Service (phab:T152676)
  • Setting up a constraints development/testing wiki
  • Moving PropertySuggester extension from github to gerrit (phab:T104309)
  • Working on performance of change dispatching so database of wikidata goes read-only in lower rate (phab:T162557)
  • Reducing number of duplicates rows in our database table for labels, descriptions and aliases to make it faster (phab:T163551)

You can see all open tickets related to Wikidata here.

• Monthly Tasks
  • Add labels, in your own language(s), for the new properties listed above.
  • Comment on property proposals: all open proposals - proposals needing attention
  • Suggested and open tasks!
  • Contribute to a Showcase item.
  • Help translate or proofread the interface and documentation pages, in your own language!
  • Help merge identical items across Wikimedia projects.
  • Help write the next summary!

Read the full report · Unsubscribe · Lea Lacroix (WMDE) 11:57, 23 May 2017 (UTC)

sports league level (P3983)

Please see sports league level (P3983). Property Wikidata property example (P1855). I have used part of the series (P179) but I don't think is correct. Do you have any suggestion? The number with out the corresponding item I thing is not enough. Xaris333 (talk) 14:21, 23 May 2017 (UTC)

Start of the 2017 Wikimedia Foundation Funds Dissemination Committee elections

Translations of this message are available on Meta-Wiki.

On behalf of the Wikimedia Foundation Elections Committee, we are pleased to announce that self-nominations are being accepted for the 2017 Wikimedia Foundation Funds Dissemination Committee and Funds Dissemination Committee Ombudsperson elections. Please read the letter from the Wikimedia Foundation calling for candidates at on the 2017 Wikimedia Foundation elections portal.

Funds Dissemination Committee
The Funds Dissemination Committee (FDC) makes recommendations about how to allocate Wikimedia movement funds to eligible entities. There are five positions being filled. More information about this role can be found at the FDC elections page.

Funds Dissemination Committee Ombudsperson
The Funds Dissemination Committee Ombudsperson receives complaints and feedback about the FDC process, investigates complaints at the request of the Board of Trustees, and summarizes the investigations and feedback for the Board of Trustees on an annual basis. One position is being filled. More information about this role can be found at the FDC Ombudsperson elections page.

The candidacy submission phase will last until May 28 (23:59 UTC).

We will also be accepting questions to ask the candidates until May 28. You can submit your questions on Meta-Wiki. Once the questions submission period has ended on May 28, the Elections Committee will then collate the questions for the candidates to respond to.

The goal of this process is to fill the five community-selected seats on the Wikimedia Foundation Funds Dissemination Committee and the community-selected ombudsperson. The election results will be used by the Board itself to make the appointments.

The full schedule for the FDC elections is as follows. All dates are inclusive, that is, from the beginning of the first day (UTC) to the end of the last.

• May 15 (00:00 UTC) – May 28 (23:59 UTC) – Nominations
• May 15 – May 28 – Candidates questions submission period
• May 29 – June 2 – Candidates answer questions
• June 3 – June 11 – Voting period
• June 12–14 – Vote checking
• June 15 – Goal date for announcing election results

More information on this year's elections can be found at the 2017 Wikimedia Foundation elections portal.

Please feel free to post a note about the election on your project's village pump. Any questions related to the election can be posted on the talk page on Meta-Wiki, or sent to the election committee's mailing list, board-electionswikimedia.org.

On behalf of the Election Committee,
Katie Chan, Chair, Wikimedia Foundation Elections Committee
Joe Sutherland, Community Advocate, Wikimedia Foundation

Posted by the MediaWiki message delivery • Translate • Get help

21:05, 23 May 2017 (UTC)

BREAKING CHANGE: sitelink encoding changing in RDF export/WDQS

In RDF exports of Wikidata and in Wikidata Query Service, sitelinks were always encoded by url-encoding the sitelink text - i.e. link to "Category:Stuffed animals" were encoded as /wiki/Category%3AStuffed%20animals.

While this encoding produces a working link, after some time we've arrived to a conclusion that such encoding is very inconvenient, due to mismatch with how titles are encoded in Mediawiki, and this mismatch makes it harder to look up the links. See more in phab:T131960.

We have decided to change the encoding, so that the encoding of the sitelink above would be /wiki/Category:Stuffed_animals. The encoding now should match how titles are encoded in Mediawiki codebase (non-ASCII characters that Mediawiki encodes will still be encoded as before).

Implementation of this change will require database reload, and during that time there might be inconsistent results returned for some time (some entities may have new sitelink encoding and some the old one). I apologize in advance for any inconvenience caused by that. I will announce additionally when the switch is process has started and when it is complete.

--Smalyshev (WMF) (talk) 16:32, 24 May 2017 (UTC)

Compare items

Hello. Is there a way to compare two items? I want to check what properties they have in common, what are their differences etc. Xaris333 (talk) 18:02, 24 May 2017 (UTC)

Special:ComparePages Matěj Suchánek (talk) 18:14, 24 May 2017 (UTC)

Hitting our 30 millionth item!

Congrats Wikidata! We hit 30 million items! Or at least an item with code 30000000. (BTW The item is The Synergistic Activity of Thyroid Transcription Factor 1 and Pax 8 Relies on the Promoter/Enhancer Interplay (Q30000000)). PokestarFan • Drink some tea and talk with me • Stalk my edits • I'm not shouting, I just like this font! 17:18, 20 May 2017 (UTC)

Congratulations, Harej. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:14, 20 May 2017 (UTC)

We have 26,242,154 content pages at the time of this writing. Saying we have hit 30 million item is misleading. In general I don't see it makes much sense to count items that we have delete in the amount of our items. ChristianKl (talk) 18:33, 20 May 2017 (UTC)

The way I've been putting it is that we hit "item number 30 million," not that it's the 30 millionth item. Harej (talk) 18:40, 20 May 2017 (UTC)

And have to delete or redirect over ten percent of our newly created items... It's simply a valueless number. We should have more important and quality related indicators to be proud of. --Succu (talk) 21:33, 21 May 2017 (UTC)

i look forward to your participation here Wikidata:Item quality campaign, since it is a topic of critical importance. Slowking4 (talk) 03:14, 22 May 2017 (UTC)

and for that I look forward to the guidance on how we handle primary sources as references, especially where they are not web-based items. Adding a reference that doesn't exist within the system, for a new subject is problematic. People flee from that discussion. (it becomes my lips are flapping)  – The preceding unsigned comment was added by Billinghurst (talk • contribs).

It's possible to add a reference via stated in (P248). It works but the UX isn't as nice as it could be. Deciding on a different system isn't trivial but there's a wish to integrate Wikipedia citations more with Wikidata. See https://www.linkedin.com/pulse/wikicite-2016-building-repository-all-wikipedia-open-data-dario . That desire is partly what drives WikiCite which is currently happening. See also https://meta.wikimedia.org/wiki/Grants:Project/Harej/Librarybase:_an_online_reference_library ChristianKl (talk) 16:13, 22 May 2017 (UTC)

The UX isn't nice? Gross understatement for adding reference detail.

I understand the how and what; ... I state again the complexity of adding references of primary sources. It is simplistic to say use "stated in" as it is not that simple for many primary references. One has to go and create each record prior to its use, and add sufficient detail for it to be recognised and utilised by yourself (by chance you utilise the source again), or suitably detailed so some other sucker stumbles across it with the right terms so they can use it. Be it the parish register for Kensington for the baptism (and specifying which volume, ...), the parish register for Westminster for the marriage (all the details); then the parish register for Lambeth for the burial (and all the detail). To add all that reference detail for one person when I am trying to just add that single person is a 'big call'. As I have noted previously, it is far far easier for me to copy and paste record data and detail onto an author talk page at Wikisource, with its contextual additional information, than to add bit-wise reference detail here.

I will now stop "carrying-on" here as I am not wanting to be a broken record; but until that is greatly revamped it is hugely problematic and will simply not happen when editing person by person, and this is from someone who has done many '0000s of person by person data entry here.  — billinghurst sDrewth 21:41, 22 May 2017 (UTC)

Beyound your personal user experience (Q1047808), billinghurst, do you have proposals how Help:Sources could be rendered more vivid? --Succu (talk) 20:10, 23 May 2017 (UTC)

I have no issue with the page about creating a source — and I note that the process for creating an item is something that can be aligned with something like Wikidata:Wikiproject Book as a useful guide. After that it becomes the quality of the reference and "stated in" is less pertinent for primary sources, eg. for Ashton Keynes parish, eg. this; or as useful for what I added for date and place of death in Edith Sophia Hooper (Q18910300). We also need to be able to separate the creation of a source, from the following creation of a good reference. We need to introduce concepts such as type of reference (P3865), age of subject at event (P3629), ...  — billinghurst sDrewth 11:35, 24 May 2017 (UTC)

Saying that every person who adds a source is supposed to add author (P50) is not helpful when the goal happens to be to reduce the workload. It's also interesting that the page currently doesn't say anything about ISBN numbers. I think it would be preferable if a human just has to put in the ISBN number and the rest is done by bots. GND data is explicitely CC0, so a lot of data can be sourced in many cases without any issues. :

In other matters the copy reference tool is still broken in the sense that it requires a page reload (it didn't before the changes). Given that it's a tool that has to explicitely activitated it's also not ideal for new users. It would be better to have a tool with deeper integration for the purpose than this gadget. It would be great if the tool could then also also copying from one item to another.ChristianKl (talk) 11:24, 24 May 2017 (UTC)

I am yet to find a primary source that has an ISBN. Similarly ISBNs are really a 'modern' concept and only available for the past 50 years. Many books don't have them. Your proposal for ISBN is useful though would be better equipped to just better populate WD with book data and its submission, either when creating a source, or at any time. I would welcome any tool that allows better input into WD. Something that inhales old book data from Internet Archive would be nice, and then allows the separation of work and edition data.

I would agree that sourcing has to be made easier, and I believe that is the challenge. Having to jump out of a page to create a source is a problem, not have a readily applicable framework for the source type slows things down. Some of the tool work done for the WE-framework is interesting in giving a ready form to complete, and I can see that there you could enter an ISBN and pre-populate a form to submit.  — billinghurst sDrewth 11:53, 24 May 2017 (UTC)

It's one of the more important things I'm doing here since the establishment of Help:Sources. --Succu (talk) 17:43, 22 May 2017 (UTC)

@Harej: Welcome to the club :-) Jonathan Groß (talk) 06:09, 23 May 2017 (UTC)

Sources as not a perfect solution

• Modern books, books that you find in your local library, tend to have an ISBN. Consequently identifying a book by their ISBN as a source is a major improvement. When you are out of luck and you have not, it is seriously relevant to have details for that book as its lack of visibility make the book questionable as to its notability.. Books have been republished in the last 50 years or whatever.
• Books without an ISBN that are in Wikisource / Open Library never mind, can be referenced to this source. Perfect solution.
• There are many lists on the web that are the reference for every list item. A good example are awards; they have their own websites but we also have Wikipedia articles with such lists. You may want individual references .. but hey sorry.

My personal point is that we need to be more relaxed. Often people take all kinds of things for granted that is not shared universally. As a rule of thumb something like 4% of a source like Wikidata is erroneous; that makes for 1,200,000 errors. Adding sources does not really make a difference, seeking differences with other Sources does. Thanks, GerardM (talk) 06:18, 25 May 2017 (UTC)

@MartinMalmsten: Is this something for KB to take a look at? -- Innocent bystander (talk) 07:03, 25 May 2017 (UTC)

Is it possible to add geographic areas on Wikidata yet?

Hi all

I remember a little while ago seeing something about a way of adding area files for locations on Wikidata, is this possible yet? If so and I have a .kml file with the data in how do I import it into Wikidata (I know someone who can use quickstatements if helpful).

Thanks

--John Cummings (talk) 14:42, 23 May 2017 (UTC)

There is a datatype now, but there is no support for KML yet. Sjoerd de Bruin (talk) 14:45, 23 May 2017 (UTC)

there is the property geoshape (P3896) - it links to a Commons file so you have to upload there and as Sjoerd mentions that doesn't support KML right now, but there are other map formats you can use. ArthurPSmith (talk) 14:47, 23 May 2017 (UTC)

Thanks Sjoerddebruin and ArthurPSmith, the current process is beyond me, I'll wait for it to be more developed. --John Cummings (talk) 19:22, 24 May 2017 (UTC)

See also KML file (P3096). --Marsupium (talk) 19:54, 24 May 2017 (UTC)

Any widget to convert an instance statement to a subclass statement?

Very often I find items that are stated as an instance of X, whereas they are actually a subclass of X.

Fixing such mistakes takes quite some time. Is there a browser widget that makes the switch with a single click? That would be wonderful :-)

If not, is there a collection of such Wikidata browser widgets? (so that I can steal the code)

Thanks! Syced (talk) 08:11, 25 May 2017 (UTC)

Remove "Item by title" from sidebar

This was already suggested two years ago, but was recently discussed again during the Wikidata Documentation sprint during the Wikimedia Hackathon 2017 in Vienna. The feature isn't being used a lot anymore. It will still be listed on Special:SpecialPages, but I don't think it deserves such a prominent position in our sidebar.

Opinions here please. Sjoerd de Bruin (talk) 14:24, 25 May 2017 (UTC)

I agree, having only one search bar is better UX.

In most modern websites a decision like this would be supported by running split tests. Is there a good reason why we can't run a split test for this?

There's also other information where I'm unsure about it's use:

• Print/export: Create a book/Download as PDF/Printable version
• Random item - I don't think that browsing random Wikidata items is useful for our users in the same way that browsing random Wikipedia pages can be fun. ChristianKl (talk) 16:29, 25 May 2017 (UTC)

Rabbit meat

Please merge rabbit meat (Q16386532) to rabbit meat (Q3556748).  – The preceding unsigned comment was added by 2001:2d8:e131:39ed::33d5:30b1 (talk • contribs) at 30 May 2017‎ (UTC).

This section was archived on a request by: Matěj Suchánek (talk) 07:02, 31 May 2017 (UTC)

Improving empty arwiki items

Can somebody improve this empty arwiki items?

Queryzo (talk) 08:31, 31 May 2017 (UTC)

I don't think it's very useful to put a list like this in the English Project Chat. I neither see a reason why those items are worth of special attention nor do I see the audience of the English Project Chat as well equipped to deal with the list. ChristianKl (talk) 11:47, 31 May 2017 (UTC)

Looks like an error, as all these items contain an ar.wp sitelink, but when you click on the wd-item on the sidebar at the ar.wp they go to a different Qid. I'm now merging them. Q.Zanden questions? 11:57, 31 May 2017 (UTC)

All  Merged --Q.Zanden questions? 12:08, 31 May 2017 (UTC)

@QZanden: Thank you for that! Queryzo (talk) 17:06, 31 May 2017 (UTC)

You're welcome! Q.Zanden questions? 17:51, 31 May 2017 (UTC)

This section was archived on a request by: Matěj Suchánek (talk) 18:13, 31 May 2017 (UTC)

Image uniqueness: are there tolerated exceptions?

There is a constraint stating that image (P18) has to be unique, one per item.

But when I see an item like lion (Q140) I can start to understand what people might want to include several images for a single item, each image with a specific qualifier (in this example, male/female for animals).

Are such exceptions tolerated? If yes, is there a list or policy describing in what cases it is OK?

Most other cases of multiplicity have zero value, with similar or even sub-par images, logos, etc.

Thanks! Syced (talk) 10:41, 17 May 2017 (UTC)

Right now we have two images for males/females in Q15978631.

I think that every textbook on anatomy presents both sexes, not just one.

My point is that we should allow multiple images, but only when it is meaningful.

In engineering 3-side/4-side images are standard.

See #Unique value constraint? at P18 talk paged1g (talk) 11:27, 18 May 2017 (UTC)

I never understood why we have a unique constraint for «image», but in the case of male / female animal it make perfect sense to create a male and a female subclass for lions. Actually there is probably different datas for each (size, weight, mean lifetime, …) author  TomT0m / talk page 20:45, 18 May 2017 (UTC)

I also think that the unique constraint should be removed given that there are multiple cases where it makes sense to have multiple entries. ChristianKl (talk) 06:59, 19 May 2017 (UTC)

I agree that the constraint should in principle be removed (also buildings: interior and exterior ext); however, we should be careful that Wikidata does not turn into a trash can by users adding hundreds of images using P18 to the same item.--Ymblanter (talk) 11:17, 20 May 2017 (UTC)

i see when items are merged, that there are multiple images. Slowking4 (talk) 03:21, 22 May 2017 (UTC)

When we remove the property we could write Wikidata usage instructions (P2559) to say that the amount of images should be kept small and each additional image is supposed to provide a clear benefit. Maybe we could even say that there shouldn't be two images with the same qualifiers. ChristianKl (talk) 21:30, 25 May 2017 (UTC)

Image: Most recent, or most recognizable?

Let's say we have two quality pictures for a famous person, which one do we want for image (P18)?

• Most recent picture, probably taken at an old age a bit before they died?
• Most recognizable picture, when they reached their peak popularity? Obviously that picture will be a bit older, and maybe of slightly lesser quality (ex: lower resolution, black and white)

I am in favour of option 2, but just checking whether there are objections, thanks! Syced (talk) 11:06, 17 May 2017 (UTC)

• A picture of a person in the prime of the activity they are known for seems more suitable than a picture of a person on their deathbed. Most of the time we don't have that much of a choice though. Pick a suitable one and add that.
  --- Jura 05:52, 18 May 2017 (UTC)

• • I agree, although I think for living persons we should have the most recent one (as long as the quality is appropriate). Sebari – aka Srittau (talk) 20:24, 18 May 2017 (UTC)

• • • best picture, that illustrates the subject the best. this will not be reducible to formula such as "most recent" or largest resolution. photo curation by selection for infobox is implied. Slowking4 (talk) 03:20, 22 May 2017 (UTC)s

• • It's interesting to have multiple images as being discussed here. Use of qualifiers (like point in time) can be used to find the relevant pictures. It is difficult to decide what's the best option since it is subjective. Jsamwrites (talk) 22:23, 25 May 2017 (UTC)

Should a country have itself as a country?

To me it does not make sense that France (Q142) country (P17) France (Q142).

It is redundant. Is there any case where a user or app would think "Oh, is that country in itself? Let's check that property to find out whether it is or not!"

Should I remove such cases? Cheers! Syced (talk) 03:17, 18 May 2017 (UTC)

That is circular, have a look at what has been done for England, Northern Ireland, Wales and Scotland (Q22) are each a constituent country of the United Kingdom (Q3336843).  — billinghurst sDrewth 05:50, 18 May 2017 (UTC)

Property talk:P17#Should countries have statements with country: themselves? --Jklamo (talk) 08:11, 18 May 2017 (UTC)

I can imaging someone writing a script that looks at place of birth (P19) and then takes country (P17) to find the country in which someone was born. A country is a valid value for place of birth (P19), so the statement could help a person writing such a script. ChristianKl (talk) 10:26, 18 May 2017 (UTC)

Allowing redundancy to make queries easier is a bad idea in my opinion. Should we add instance of (P31) musical ensemble (Q2088357) to all rock band (Q5741069) items because some developers don't know how to recurse (p:P31/ps:P31/wdt:P279* Q2088357) and thus might miss them? In the birthplace example, I just go up the administrative divisions until I get a country. Syced (talk) 15:54, 18 May 2017 (UTC)

If you think the proper way of finding the country of an town isn't to use the country property that's on the town but to recurse administrative divisions, why does the country property exist? ChristianKl (talk) 20:30, 18 May 2017 (UTC)

For things that are not administrative divisions, such as Afghan National Police (Q8568). You have much more Wikidata experience than me so I won't waste your time with this, but my personal wish is that one day all such redundancy will be avoided :-) Cheers! Syced (talk) 03:11, 19 May 2017 (UTC)

The notion that from a place or administrative entity you get a country is wrong. Everybody who died in what is now called India before the start of that republic is not an Indian (for instance).. the same applies for many other "countries". It is why we have former countries. Thanks, GerardM (talk) 03:54, 19 May 2017 (UTC)

Very true! So I guess we will have to admit that the example we were debating on ("find the country in which someone was born") actually has no easy solution. We will have to find a better example. Or move along, since I can live with this redundancy, and there is a lot of other tasks to work on before worrying too much about this :-) Syced (talk) 07:20, 19 May 2017 (UTC)

In a perfect would the country property would be qualified in a way that allows you to check whether a town was indeed at the time of the birth in the corresponding counrty ;) I admit that practically this often won't be the case in Wikidata in it's current state but I see no reason why it couldn't be in the future. ChristianKl (talk) 12:18, 19 May 2017 (UTC)

How should we do when the recurse spreads? Sápmi (Q62132) is located in four nations! Kiruna, Sweden is located in Sapmi, who is located in Russia! Does that mean that Kiruna is located in Russia? -- Innocent bystander (talk) 15:21, 26 May 2017 (UTC)

Common labels and descriptions

One of them major problems of this project, IMO, is the flooding of watchlist and recent changes page, and also item history cluttering. Premises and conditions for this exists, since there are many large series of items which could have an identical label in multiple languages (e.g. almost all those based on Latin script) for each item, and also common descriptions or with a similar pattern. There were granted many flood flags and bot flags for adding by one label/description per language for such series of items, but why not to add a big part of them at once, in one or two edits, using bots? For example, take a look at this item history - it's very well visible how in more than ten edits several users and bots have added few labels/descriptions for individual languages one by one; there are many other better examples with more cluttered histories spread on several pages and hudreds of entries in item history, but this is a fresh one with 5 such edits made yesterday (and it's a part of a series, multiply these edits to several thousands/tens of thousands which are in the same situation):

Some 65 languages based on Latin script in Wikidata almost always share the same label per item for some proper names: human names and settlement names originating from countries with official languages based on Latin script, etc. The descriptions for each language are either common for a series of items, or have common parts and follow a common pattern with an additional variable (e.g. a higher administrative territorial unit for populated places). Thus bots can add in a single edit even by 200 descriptions, and by 60+ labels (though is acceptable in another edit), instead of having several tens or hundreds of entries in item histories to achieve this. This practice is already applied to items of disambiguations/templates/categories, but it should be extended for other major series of items, for example humans: sportspeople (especially footballers, most commons), politicians; taxons; populated places; films, books, singles and other albums, etc, etc. For descriptions of the most large series of items, we could create and maintain several 'dictionaries' of terms, like those from MediaWiki:Gadget-autoEdit.js, which should be used by bot operators when are mass-editing items. To achieve a better and faster translation of the list of descriptions, there can be used the Mass Message delivery system to ask for help on local Village Pumps. The available mass editing tools on Labs should show a message which says that editing a label/description for a single language in thousands of items it's not recommended and this should be discussed in a proper place to add this task to the next bot run. Thus, I think there should be a Wikidata Etiquette regarding mass-editing large series of items (thousands/tens of thousands) both for humans and bots. --XXN, 11:29, 25 May 2017 (UTC)

I think the core problem is that when I browse an item history or the watchlist I don't care about it's 'bug' or 'jv' label and the optimal solution would be to filter it out.

As far as there's a Project Chat for languages besides English, the main language of the corresponding project chat isn't English and as a result, I don't think it's a good idea to do mass message delivery in English to that Project Chat. WikiProjects for the respective languages would be a more appropriate way but currently we don't have a lot of them for individual languages. Having a WikiProject to ping when a speaker of a particular language is needed would also be valuable. ChristianKl (talk) 12:14, 25 May 2017 (UTC)

It might be worth to create a new property that indicates whether instances of a given class can be expected to have the same name in all languages that share the Latin alphabet. ChristianKl (talk) 13:17, 25 May 2017 (UTC)

I created https://www.wikidata.org/wiki/Wikidata:Property_proposal/Generic#language-independence_of_instances as a proposal to store the information about which item's labels can simply be copied. ChristianKl (talk) 17:24, 25 May 2017 (UTC)

I agree, adding a lot of labels is not a problem (but maybe it may and should be optimized), lack of filtering on page history is. Yarl 💭  22:33, 25 May 2017 (UTC)

Create multiple items using quick statements

hello, I'm trying to create a list of new items using the QuickStatements tool. According to the instruction, I need to type the word "CREATE" in the first line and then statements and labels might be added easily after the word "LAST". Here an example that should have created two new items and added some information to them which unfortunately did not:
CREATE
LAST Lar "نقطة الغليان" Dar "كتاب" P52 Q968721 P31 Q571 P364 Q13955
CREATE
LAST Lar "الله والإنسان" Dar "كتاب" P54 Q968721 P31 Q571 P364 Q13955

Am I missing something? If my way was right but the tool is inactive in the moment, is there another possibility to create multiple items (for example from an excel list) and adding necessary statements and labels/description to them? Thanks in advance.. --Sky xe (talk) 21:21, 25 May 2017 (UTC)

@Sky xe: The old QuickStatements it's not working at the moment, try the new one - http://tools.wmflabs.org/quickstatements/ --XXN, 23:45, 25 May 2017 (UTC)

You'd need a newline and a "LAST" before each statement/description/label.
--- Jura 11:32, 26 May 2017 (UTC)

Like this:

CREATE
LAST Lar "الله والإنسان"
LAST Dar "كتاب"
LAST P54 Q968721
LAST P31 Q571
LAST P364 Q13955
CREATE 
LAST Lar "نقطة الغليان"
LAST Dar "كتاب"
LAST P31 Q571
LAST P364 Q13955

But then with the correct tabs. Q.Zanden questions? 12:13, 26 May 2017 (UTC)

Wikidata:WikiProject Counter-Vandalism (WD:CVN)

Hey folks, as you might have already seen in the most recent weekly newsletter, there is now a new WikiProject Counter-Vandalism. (Possible) scopes of this project include:

• Team up, streamline CVN processes & share experiences
• Introduce CVN tools; improve tools with feedback
• Act as a helpdesk for CVN related issues
• Indicate to data users that Wikidata actually has an active CVN unit
• … (your ideas here)

New members are welcomed in this collaborative community project! —MisterSynergy (talk) 09:43, 26 May 2017 (UTC)

Which one is the preferred way to describe the start and the end of TV shows?

I've realized that there are at least two ways to describe when a tv show started in the current entities and one of them generates a constraint report. The accepted one by the constraints is using publication date (P577) to tell when the show started and I don't know if there is any for when the show is finished that goes with this one. Others use start time (P580) and end time (P582) respectively, which seems more correct to me for this use case, but both, start time (P580) and end time (P582), trigger an error in the constraint reports because television series (Q5398426) is not an instance nor a subclass of occurrence (Q1190554). I don't think we should be fixing the error by setting television series (Q5398426) as a subclass of occurrence (Q1190554), because it isn't, but we could fix the constraint if this use case is applicable. Anyway, regardless of whether we change or not the constraint I would like to know which is the preferred way of specifying the start and the end of the shows to fix the data retrieval and automatic categorization of our infobox. -- Agabi10 (talk) 12:25, 26 May 2017 (UTC)

I would use start time (P580)/end time (P582). I could agree on adding publication date (P577) to episodes, but TV shows - no. --Edgars2007 (talk) 12:29, 26 May 2017 (UTC)

In which property is stored the item title showed on the wikidata website?

Hello, I'm new to Wikidata and I'm still trying to grasp the data model. So please apologize if my question is too dumb or if this is not the appropriate place to ask. I try to understand, when I access an item (say, wikidata.org/wiki/Q22), the wikidata website shows a title (in this case, Scotland). I have downloaded the dump, and when I access an item, is there a property in the claims that contains the 'default' title, or is Wikidata's website showing the title from the English Wikipedia? If so, what if there is no article for this item on enwiki? Thanks in advance, −PetaRZ (talk) 14:52, 26 May 2017 (UTC)

These titles are stored in "labels" (multi-language, they are not properties). The primary title shown in Wikidata depends on the language you selected in Special:Preferences. The language selected by default is the English one I guess. We have native label (P1705) for storing native [sic] labels though. Strakhov (talk) 15:09, 26 May 2017 (UTC)

Here you have an example for accessing these labels through Wikidata SPARQL query. Strakhov (talk) 15:11, 26 May 2017 (UTC)

Very clear, thanks! −PetaRZ (talk) 15:19, 26 May 2017 (UTC)

Pubs loaded as houses

Here is a list of pubs that have the instance of property set to house. I'd appreciate if somebody could update the instance of property to be pub (Q212198) instead.

The following query uses these:

• Items: house (Q3947)   

• Properties: coordinate location (P625)   , instance of (P31)   

  SELECT DISTINCT ?item ?itemLabel WHERE
  {
    ?item wdt:P625 ?location ;
          wdt:P31 wd:Q3947 ;
          rdfs:label ?itemLabel .
    FILTER(strEnds(?itemLabel, "Public House") && !CONTAINS(?itemLabel, "Outside") && !CONTAINS(?itemLabel, "Former"))
  }
  

Try it!

Thanks. Edward (talk) 20:02, 30 May 2017 (UTC)

I'm running with QuickStatements right now. Q.Zanden questions? 21:21, 30 May 2017 (UTC)

Batch 626 is deleting the house (Q3947) statements and batch 627 is adding the pub (Q212198) statements. Please stop the running if it is malfunctioning! Q.Zanden questions? 21:31, 30 May 2017 (UTC)

Thanks. Edward (talk) 21:40, 30 May 2017 (UTC)

 Done You're welcome Q.Zanden questions? 12:13, 31 May 2017 (UTC)

This section was archived on a request by: Q.Zanden questions? 19:08, 1 June 2017 (UTC)

QuickStatements

@Magnus Manske: Is QuickStatements down? Received this error message on pressing "Do it".

Warning: parse_ini_file(/data/project//replica.my.cnf): failed to open stream: No such file or directory in /data/project/magnustools/public_html/php/common.php on line 57

Warning: mysqli::mysqli(): (HY000/1045): Access denied for user ''@'10.68.23.53' (using password: NO) in /data/project/magnustools/public_html/php/common.php on line 126

Warning: parse_ini_file(/data/project//replica.my.cnf): failed to open stream: No such file or directory in /data/project/magnustools/public_html/php/common.php on line 57

Warning: mysqli::mysqli(): (HY000/1045): Access denied for user ''@'10.68.23.53' (using password: NO) in /data/project/magnustools/public_html/php/common.php on line 126

Fatal error: Call to a member function real_escape_string() on boolean in /data/project/wikidata-todo/public_html/quick_statements.php on line 32

Jc86035 (talk) 07:40, 31 May 2017 (UTC)

AFAIK the old QS is down for a couple of days now, and there are already opened issues in Magnus’ bitbucket code repository. Have you also tried the new version of QS? —MisterSynergy (talk) 07:54, 31 May 2017 (UTC)

@MisterSynergy: Does the new QuickStatements also convert article names to Wikidata item IDs? Jc86035 (talk) 08:00, 31 May 2017 (UTC)

I don’t know, I have barely used it. There are a couple of things which apparently don’t work with the new version right now. Please try by yourself: new QS. —MisterSynergy (talk) 08:04, 31 May 2017 (UTC)

No, sitelink conversion isn't available. It is (should be) available here. --Edgars2007 (talk) 12:08, 31 May 2017 (UTC)

I used the old QuickStatements yesterday evening (CEST) without any issues. – Máté (talk) 12:26, 31 May 2017 (UTC)

Labs had some hiccup, should work now. --Magnus Manske (talk) 13:38, 31 May 2017 (UTC)

This section was archived on a request by: Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:17, 1 June 2017 (UTC)

Moving statements from one item to another

Hi! The subject of József Lőrincz (Q1160734) was originally a poet born in 1947, having articles in the Esperanto and Hungarian Wikipedia. Last year, the English article about a football player with the same name was incorrectly added to the item, after which a lot of statements about the footballer were added. Currently, claims about the two persons are mixed in the item. I created a new item (József Lőrincz (Q30095490)) and added the English sitelink there. Is there an easy way to transfer the relevant claims to the new item? – Einstein2 (talk) 08:39, 31 May 2017 (UTC)

Duplicate Q1160734 using Magnus' user-script (User:Magnus_Manske/duplicate_item.js), without labels, as a new item. Remove the claims that you don't want from the new item (and later from the old one), then merge that into Q30095490. In future do the duplication instead of creating a new, bare, item. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:00, 31 May 2017 (UTC)

Thank you! – Einstein2 (talk) 12:56, 1 June 2017 (UTC)

This section was archived on a request by: Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:55, 1 June 2017 (UTC)

Global overview of the strategy discussion

Some of the strategy-related posts look like they were ignored. Maybe most users think there's no need to react here? Maybe I generate sentences or look like a mass message bot? :) Everything is clear to everybody? I doubt all of that, but, well. Anyway, I hope someone will notice and appreciate my messages:

• There's a global overview of discussions that took place in the first week of the Cycle 2. There's a lot of feedback, maybe someone somewhere posted something related to your thoughts, raised a similar issue, maybe even had the same idea. Check that out!
• In the first week of the Cycle 2, we've gathered 35k bytes of comments. For comparison, in the Cycle 1, which lasted a month, 50k were saved. That's a progress, but I'm still asking for more. More opinions, more concerns, more questions, any type of feedback. And remember, no such thing as a stupid question :)

SGrabarczuk (WMF) (talk) 18:09, 25 May 2017 (UTC)

@SGrabarczuk (WMF):The way your summary works, it's hard to go to the actual post that's behind a summary headline. Can you change your process in a way that the actual posts are linked? ChristianKl (talk) 19:00, 25 May 2017 (UTC)

@ChristianKl:, sure, I can do that, but my promise applies only to my work (Wikidata, English Wikipedia, and Meta-Wiki). I will also pass this request along to other coordinators. In particular, I'm not responsible for the overview, because I didn't write that. Thanks for your feedback (here and on the dedicated talk pages). SGrabarczuk (WMF) (talk) 13:06, 27 May 2017 (UTC)

Un-linking a name

Hello! Through this edit on Janaka (Q1500207) I happened to remove the name of spouse of subject Janaka. Although his wife's name is "Sunayana", it should not be linked to Sunayana (Q7638864) which is about a film named Sunayana. We do not have an article on wife in en.wiki. So is there an option of just keeping the name but not linking it anywhere? §§Dharmadhyaksha§§ {T/C} 07:00, 26 May 2017 (UTC)

There's no way to keep the name without a string but on Wikidata you can simply create a new item for the wife of a person. I created Sunayana (Q30059291). In case there's more information known about Sunayana, feel free to add it to the existing item but that's not necessary to warrant the item. ChristianKl (talk) 08:41, 26 May 2017 (UTC)

Thanks @ChristianKl: for the info. §§Dharmadhyaksha§§ {T/C} 06:59, 27 May 2017 (UTC)

Collaboration with VIAF.

Hoi, I noticed that Wikidata are included by librarians. I blogged about this. The consequence is that we should actively seek their collaboration. VIAF is like Wikidata actively maintained and both their and our content have its problems. Collaboration is a reason to rejoice because it means that we have shared objectives. All of us want people to read, not just Wikipedia (any Wikipedia) but also books and periodicals. Libraries and librarians have an enviable track record. Together we will achieve what we will not achieve alone. GerardM (talk) 20:26, 23 May 2017 (UTC)

They are fast. We should have a bot that removes redundant identifiers and corrects references. Sjoerd de Bruin (talk) 21:03, 27 May 2017 (UTC)

What can we do to make this happen effectively? Thanks, GerardM (talk) 12:08, 28 May 2017 (UTC)

Fill in empty fields

I have this query:

https://query.wikidata.org/#SELECT%20%3FgameLabel%20%3FdevLabel%20%3FpublisherLabel%20%3FdateLabel%20WHERE%20%7B%0A%09%3Fgame%20wdt%3AP31%20wd%3AQ7889%3B%0A%09wdt%3AP136%20wd%3AQ744038.%0A%20%20%09OPTIONAL%20%7B%0A%20%20%20%20%20%20%3Fgame%20wdt%3AP178%20%3Fdev.%0A%20%20%20%20%20%20%3Fgame%20wdt%3AP123%20%3Fpublisher.%0A%20%20%20%20%20%20%3Fgame%20wdt%3AP577%20%3Fdate.%0A%20%20%20%20%7D%0A%20%09SERVICE%20wikibase%3Alabel%20%7B%20bd%3AserviceParam%20wikibase%3Alanguage%20%22en%22.%20%7D%0A%7D

Is there a way I can fill in missing fields using an interface like this? I think in SQL Server and MySQL you can type in new values in a result set table. Thanks. SharkD (talk) 22:16, 25 May 2017 (UTC)

@SharkD: For SPARQL help, see Wikidata:Request a query. Yarl 💭  22:31, 25 May 2017 (UTC)

My question is not about SPARQL queries, but rather about whether I can type values into a result set table like in SQL Server and MySQL Workbench. SharkD (talk) 23:13, 25 May 2017 (UTC)

You could use Tabernacle. --Edgars2007 (talk) 06:51, 26 May 2017 (UTC)

I looked at Tabernacle, but it seems to only allow you to add values, not change or remove them. So it's kind of limited. SharkD (talk) 14:11, 28 May 2017 (UTC)

Adding sources with QuickStatements

Hi, I'm having a problem. I'm in the process of importing a database of Norwegian missionaries in collaboration with a partner of Wikimedia Norge, and am using QuickStatements for this. However, I would of course like to put sources on the statements I make, to make it clear where stuff is coming from. Anyways, my problem is that when I try to add two or more source properties to a statement, they are added as two different sources, even though they are just different aspects of the same source. You can see an example of what I mean here (scroll down to the gender property to see what it looks like). So my question is if anyone knows how it's best to solve this – is there perhaps another tool I should use instead, or...? Jon Harald Søby (WMNO) (talk) 22:22, 26 May 2017 (UTC)

The old QuickStatements never supported that functionality, but I am not sure about the new version. Complex references can be added by bots, but that is not as trivial to set up as the preparation of a QuickStatements input. —MisterSynergy (talk) 22:27, 26 May 2017 (UTC)

The new version does support that functionality. 123 (talk) 22:41, 26 May 2017 (UTC)

There we go, I had no idea there even was a new version. The Magnus train is too fast for me. It worked splendidly. Thank you very much! Now I just need to sort my data properly, and I'll be good to go. Jon Harald Søby (WMNO) (talk) 23:06, 26 May 2017 (UTC)

I have nothing to add to the question, but I love that we now have a database of something so specialised as Norwegian missionaries :-) Andrew Gray (talk) 10:35, 27 May 2017 (UTC)

Is there an example somewhere of adding a complex sourcing statement in the new QS (complex = stated in:Qxx, chapter/entry: string, author:Qxx, page(s):nnn)? "{datavalue_source}] is an array of datavalue of the sources, in Wikidata API notation" is opaque to me still, sorry. - PKM (talk) 19:53, 27 May 2017 (UTC)

Excel issue

In Excel v1609 when I create a hyperlink to a Wikidata record (for instance '=HYPERLINK("http://www.wikidata.org/entity/Q23044915")', it automatically takes me to the json page for the record instead of the regular page. How can I fix this? Thanks. SharkD (talk) 03:25, 28 May 2017 (UTC)

Use https://www.wikidata.org/wiki/Q23044915. And yes, I know, that SPARQL outputs http://www.wikidata.org/entity in JSON results. --Edgars2007 (talk) 03:34, 28 May 2017 (UTC)

Thanks. Is there a workaround? SharkD (talk) 06:22, 28 May 2017 (UTC)

The entity URL actually doesn’t always return JSON – it applies content negotiation to return the most appropriate format (see WD:Data access#Linked Data interface). So if you open http://www.wikidata.org/entity/Q42 in a browser, it redirects you to the entity HTML page, since the browser specified that it would like to see HTML. JSON is just the default format if the client doesn’t specify which formats it can accept.

You can override content negotiation and explicitly return a particular format by appending the format like a file name extension to the URL, e. g. http://www.wikidata.org/entity/Q42.html or http://www.wikidata.org/entity/Q42.ttl. —Galaktos (talk) 14:20, 28 May 2017 (UTC)

Very good, thank you. SharkD (talk) 15:51, 28 May 2017 (UTC)

Royal Governor of La Florida

The "Royal Governor of La Florida" is Spanish. The territory involved is more or less the current Florida. When I apply our current logic, Florida becomes part of "New Spain". I shudder to think of all the consequences.. Any ideas? Thanks, GerardM (talk) 12:29, 28 May 2017 (UTC)

Could you link to items when making a case like this? It makes it much easier to follow your reasoning. Are you talking about Royal Governor of La Florida (Q30077894) being about New Spain (Q170603)? ChristianKl (talk) 12:48, 28 May 2017 (UTC)

Hoi, yes. This is the current list of office holders. Thanks, GerardM (talk) 13:26, 28 May 2017 (UTC)

Byte order mark

Is there any way someone could please add a byte order mark (BOM) to the Wikidata Query's outputted CSV files? MS Excel does not parse unicode characters properly without it. It's not hard to fix, but it would eliminate an extra step in my workflow. Unless there's a good reason to omit it. Thanks! SharkD (talk) 14:09, 28 May 2017 (UTC)

Help:Sources#Books

Is Help:Sources#Books still up to date and the recommended way to add book sources? I’m especially dubious about the instruction to add P2439 (P2439) (instead of original language of film or TV show (P364) and language of work or name (P407)) and about the instance of (P31)book (Q571) (I think there was a discussion about this here recently). —Galaktos (talk) 13:11, 27 May 2017 (UTC)

Wikidata:WikiProject Books is the most up-to-date reference that I know. P407 is the one to use, the original language has a closed discussion to delete. Re the instance, there are a range of choices depending on what it is, and there are some that are more specific.  — billinghurst sDrewth 05:41, 28 May 2017 (UTC)

The discussion is ongoing to define what will be the final model in that page. This is no more updated page that this discussion about the question of which property to use for the language.

For the book/work discussion, there were some discussions, but no real decision. Snipre (talk) 19:44, 28 May 2017 (UTC)

Wikidata query: Recently URL include spaces, causing problems in MarkDown and wikicode

Tracked in Phabricator
Task T166491

Wikidata Query URLs is a good way to share queries to people not familiar with SPARQL (See for instance the "Query examples" section of any "Wikidata weekly").

Until recently, Wikidata Query URLs did not contain any space. But recently, they contain spaces (due to a Firefox change or due to a Wikidata Query change, I don't know).

This results in parsing errors when trying to share a query via either Markdown or wikicode.

Have you experienced the same? Any solution? Using a tool to URLencode the query URL does not work, as it is already half-URLencoded.

Thanks! Syced (talk) 09:00, 29 May 2017 (UTC)

Hello Syced, thanks for noticing. We will investigate as soon as possible. Lea Lacroix (WMDE) (talk) 09:11, 29 May 2017 (UTC)

How to apply a condition to items expanded from a wildcard?

I want all items that are in a country or one of its sub-divisions, so I wrote this query: ?item wdt:P131*/wdt:P17 ?countryId.

The query above works fine... except when a sub-division used to belong to another country, like for Pristina (Q25270).

Great news: in such cases we should only consider the unique located in the administrative territorial entity (P131) that has no end time (P582), and the problem is solved.

My question: how to alter my query above to only accept located in the administrative territorial entity (P131)s that have no end time (P582)?

Thanks! Syced (talk) 09:47, 29 May 2017 (UTC)

The current statement should have preferred rank, then the second statement would be ignored. I can't find any reasonable SPARQL solution now. Matěj Suchánek (talk) 10:59, 29 May 2017 (UTC)

Workflow for the case when a user removes Wikidata description and WikiLink from an item instead of merging

While going through items that might be deleted I noticed the case of https://www.wikidata.org/w/index.php?title=Q20725951&redirect=no . A user moved the fr-Wikilink to Terminal 21 (Q3278033) instead of merging the items. Is there a template that we use to inform a user in a case like this, that it would be better to merge the items the next time? ChristianKl (talk) 11:05, 29 May 2017 (UTC)

Have an abuse filter that prevents it, and customise the error message. Is there any requirement to move a page in ns:0?  — billinghurst sDrewth 11:26, 29 May 2017 (UTC)

I think he does not talk about “page moves” as in Wikipedias. The problem is that users manually remove sitelink(s) from one item and add them to another item, leaving the old one empty. If they had used the merge tool instead, the old item would have been converted to a redirect; now we are sitting on an un-quantifiable pile of “broken” residual items… —MisterSynergy (talk) 11:41, 29 May 2017 (UTC)

If you look at the Taxonomy bits, it is the right thing to do to accumulate the links on one away from the synonym, and not to merge, so you have a a conundrum. As a move of a sitelink has a set summary, you could set conditions in an abusefilter that tracks a move for users not in a group, with less than xxx edits, or some other criteria. I think that you are going to need to have a deeper analysis of what you want and what is the problem. Otherwise you may be wanting to see if you can track when an item moves from one to zero sitelinks.  — billinghurst sDrewth 12:41, 29 May 2017 (UTC)

Wikidata weekly summary #262

Here's your quick overview of what has been happening around Wikidata over the last week.

• Discussions
  • Closed request for adminship: MisterSynergy, Queryzo, ChristianKl (all successful)
  • New request for comments: Allow the creation of links to redirects in Wikidata

• Events/Press/Blogs
  • Past: WikiCite, May 23rd-25th in Vienna. Documentation will be available soon, in the meantime you can check on Twitter and the video streams to see what happened
  • Six degrees on Wikidata, by Andrew Gray
  • Some statistics on scholarly data in Wikidata, by Finn Årup Nielsen
  • Wikidata for librarians, by Dan Scott

• Other Noteworthy Stuff
  • Eurowings app uses Wikidata
  • Are you searching for geolocations of Wikipedia articles on the OSM map? Alex-7 created a web app.
  • Breaking change: sitelink encoding changing in RDF export/WDQS

• Did you know?
  • Newest properties: Filmweb.pl ID, racon signal, Vitaskrá ID, SSB urban settlement number, Austrian Textbook ID, EKATTE place ID, members have occupation, National Library Board Singapore ID, SHARE Catalogue author ID, Sequence Ontology ID, supports programming language, subreddit, sports league level, TA98 Latin term, Misjonsarkiv person ID, Unifrance person ID, Unifrance company ID, IECIC 2015 ID, Songkick venue ID, BVMC work ID, secretary general, nature preserve in North Rhine-Westphalia ID, PIM authority ID, PASE Domesday place, PASE Domesday person ID
  • Query examples:
    • Map of subreddits associated to a place on Earth (source)
    • Potential reviewers for papers in computer science (source)
    • Paintings with 'portrait' in the name that lack P180 / depicts (source)
    • Publications and presentations about WikiCite (source)
  • Newest WikiProjects: WikiProject Zika Corpus
  • Newest external tools: Wikidata to CSV
  • Newest database reports: Roger Moore filmography

• Development
  • Attending WikiCite
  • Showing badges on other projects (phab:T73887)
  • Improving label tracking (phab:T151717)
  • Working more on statements on forms in order to support lexicographical data

You can see all open tickets related to Wikidata here.

• Monthly Tasks
  • Add labels, in your own language(s), for the new properties listed above.
  • Comment on property proposals: all open proposals - proposals needing attention
  • Suggested and open tasks!
  • Contribute to a Showcase item.
  • Help translate or proofread the interface and documentation pages, in your own language!
  • Help merge identical items across Wikimedia projects.
  • Help write the next summary!

Read the full report · Unsubscribe · Lea Lacroix (WMDE) 15:22, 29 May 2017 (UTC)

QUDT unit ID (P2968) for prefixes and systems of units?

The only indicated domain for this property is unit of measurement (Q47574), but the QUDT Unit Ontology also has prefixes like giga- (Giga) and systems of units like SI (SystemOfUnits_SI). Is it OK to use this property for that? -- IvanP (talk) 18:30, 25 May 2017 (UTC)

Well, it says "unit ID". I'm afraid if you use it with non-units, that property won't work - URLs would be wrong, etc. - and it would be wrong semantically too. However, nothing prevents one from having another QUDT-related property. --Laboramus (talk) 18:30, 26 May 2017 (UTC)

@Laboramus Although non-units, Giga and SystemOfUnits_SI (for instance) belong to the QUDT Unit Ontology and the URL would be right: http://qudt.org/qudt/owl/1.0.0/unit/Instances.html#Giga, http://qudt.org/qudt/owl/1.0.0/unit/Instances.html#SystemOfUnits_SI. -- IvanP (talk) 19:06, 26 May 2017 (UTC)

if it doesn't break the formatter URL then I would favor keeping a single ID and changing the label to more accurately reflect what it covers (if necessary). ArthurPSmith (talk) 13:54, 30 May 2017 (UTC)

Using type of reference (P3865) and publication date (P577) when importing data from a Wikipedia template

Currently, when data is imported from Wikipedia we often just have imported from Wikimedia project (P143) filled. Unfortunately, that doesn't provide much information and it frequently doesn't allow tracking the exact location of the import. I think it would be great if publication date (P577) would get filled with the last edit of the Wikipedia page in question and type of reference (P3865) would be filled with the template from which the information comes. When a tool for example imports the date of birth (P569) of Angela Merkel from Template:Infobox officeholder it could specify type of reference (P3865) Template:Infobox officeholder (Q5830052). ChristianKl (talk) 13:02, 27 May 2017 (UTC)

I agree with your suggestion and it needs to be incorporated to currently available tools like 'harvesttemplates', 'petscan' etc. I hope there is a consensus on this topic since we must be able to distinguish the data coming from the infobox versus the main content of a Wikipedia page. Jsamwrites (talk) 13:23, 27 May 2017 (UTC)

Hi ChristianKl, Additionally, how about the ability to include the revision identifier of the Wikipedia page instead of publication date (P577)? However, I doubt whether we have a property tracking the revision identifier. Jsamwrites (talk) 13:28, 27 May 2017 (UTC)

I don't think we have currently a property for tracking a revision identifier but that doesn't mean we can't create one for this purpose. The use-case seems important enough to have it's own property. ChristianKl (talk) 13:55, 27 May 2017 (UTC)

Hi ChristianKl, I have created a property request Wikidata:Property proposal/Property metadata#wikimedia revision identifier. Jsamwrites (talk) 15:39, 27 May 2017 (UTC)

If we are recording a date, it should be with retrieved (P813)  — billinghurst sDrewth 05:45, 28 May 2017 (UTC)

Exactly. Don't forget imported from Wikimedia project (P143) isn't a real source. Just a hint where they data came from. It's temporary until we can replace it with a real reference to back up a statement. So please don't expand on this and focus on finding real references to support statements. Multichill (talk) 19:07, 28 May 2017 (UTC)

The term "real source" is a bit funny. It's the source where the data is coming from. There are multiple reasons to quote sources. One reason is to give a statement authority. imported from Wikimedia project (P143) doesn't give statements authority, but there are other reasons for sources. The ability of understanding where information comes from is important.

There are cases where different Wikipedia's have a different version of the truth. In those cases I think it can be valuable to have good information because that can make it easier to fix the errors.

Is your main concern that if we provide higher quality information about the provenance of the imports that this will demotivate people from adding additional sources? Especially as long we have Wikipedias that don't import fact from Wikidata that aren't sourced with stated in (P248) or reference URL (P854), I don't think that there's a sizable demotivation effect. ChristianKl (talk) 14:41, 29 May 2017 (UTC)

IMO you are arguing a circular case. If you are saying that citing WPs is problematic with "imported from" as the information may have been changed or updated, then you are stating that it is a reliable source. We want :wikipedia:reliable sources and a dynamic wiki(pedia) is not; and we state the case for what is suitable at Help:Sources. We should not be encouraging the use of "imported from", and continuing to legitimise it, and embedding it with more in-depth means of linking just seems counter-productive, we can generally identify the point of reference by the date of the item creation, and correlate it with the history of the linked article; to me that seems enough. Personally I think that "imported from" should be deprecated for general use, and only used for mass imports from new wikis.  — billinghurst sDrewth 21:54, 29 May 2017 (UTC)

@billinghurst:I haven't used the phrase "reliable source" at all. I don't think the providing provenance information is only valuable for those sources that are reliable. If provenance information is provided it allows the user to check himself where the information comes from and form his own judgements. It's even possible to argue that provenance information is more valuable in cases where it's not clear whether the information is actually true.

I don't think that whether or not tools like PetScan that add information with "imported from" should be legitimized is a productive discussion, or that we should work to deligitimize them. I believe in plurality and the idea that Wikidata can have a big tent where some people like to have more information (even when it is imported from Wikidata) and other people only care for those information that has external references. Those data users that only care about external references can ignore all the data that doesn't have them.

As far as circularity goes you are basically saying that you like it when it's harder for users of Wikidata to find the external source that a Wikipedia used to reference a claim that's "imported from", because you think that making interaction with "imported from" statements uncomfortable enough leads to delegitimization of them. ChristianKl (talk) 09:50, 30 May 2017 (UTC)

Error while merging items: A conflict detected on enwiki

Using a Wikidata query I found duplicate English Wikipedia articles, marked them to merge, and after waiting for 6 months I just actually merged them. Now I want to merge the Wikidata items as well, but I get an error:

Error while "Please wait...": A conflict detected on enwiki: Q16974727 with enwiki:Embassy of the United States, Dar es Salaam, Q16964539 with enwiki:Embassy of United States, Tanzania

I have merged items in the past, but it is the first time I see this error. What is happening? If someone could perform the merge that would be great, thanks :-) Syced (talk) 10:13, 29 May 2017 (UTC)

You need to remove the redirecting sitelink manually before you merge. —MisterSynergy (talk) 10:27, 29 May 2017 (UTC)

And still people want redirects as sitelinks... Sjoerd de Bruin (talk) 10:29, 29 May 2017 (UTC)

@Sjoerddebruin: If we would make a decision to have redirects, then it wouldn't be that hard to write the merge tool in a way that it wouldn't throw an error in this case. The core problem is that in the status quo we have links to redirects and those are poorly supported. ChristianKl (talk) 11:09, 29 May 2017 (UTC)

It worked, thanks! Syced (talk) 08:58, 30 May 2017 (UTC)

Ideas for new tools

At the recent WikiCite event, we had a hack-day, before and during which, people proposed tools or services which they either wanted to collaborate to create. This happens for other hack days too, of course. I have seen Phabricator used for this, but I fear that this is inaccessible to some of our editors, and that ideas posted there are not readily seen by the people who might collaborate to build them.

Do we have a better forum, or should we? Perhaps something like Wikidata:Bot requests?

Please see below for an example of a tool which could be proposed in this way. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:49, 29 May 2017 (UTC)

Some users have posted their ideas to Wikidata talk:Tools. But I still encourage using Phabricator. Matěj Suchánek (talk) 06:30, 30 May 2017 (UTC)

Suggestion: Tool to build single-field forms from queries

Further to the above question, I have an idea, and a use-case, for a tool. It would take any SPARQL query, for example:

#Enter employer to find living employees with no known ORCID iD 
SELECT DISTINCT ?person  ?personLabel WHERE {
  ?person wdt:P31 wd:Q5.
  ?person wdt:P108 wd:Q223429                   # employed at...
  FILTER NOT EXISTS { ?person wdt:P496 ?orcid } # no ORCID iD
  FILTER NOT EXISTS { ?person wdt:P570 ?date }  # not dead
  SERVICE wikibase:label { bd:serviceParam wikibase:language "en" }
}

Try it!

and allow a user to indicate that one specified item (or property) - for example, in this case, Q223429 - was to be treated as the input from a form. This might be written literally as, say, "Q$" or "Q$$$" or some other unlikely-elsewhere string by the query author, as (in the example):

?person wdt:P108 wd:Q$$$ # employed at...

The specified item (or property) could be used more than once in the query.

The tool would then generate a web page (on the toolserver?) that would use the text from the opening line's comment as its label, and allow a second user to either enter a QID, or have an auto-complete do that for them. It would then run the query and return the results for it, in "embed" format.

The use case for the above example is that I could email the URL for the form to contacts leading n ORCID implementation at several universities, who could then use it to find which of their colleagues' ORCID iDs we are missing; but others could use the tool for markedly different purposes.

Does anyone fancy making this? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:49, 29 May 2017 (UTC)

As far as I can see, this is possible via the widget in the query service. It doesn't have yet full support, but your example does work. I was able to quickly change Q223429 to another institution by typing its name or id. Matěj Suchánek (talk) 06:27, 30 May 2017 (UTC)

I think this is already supported via query templates:

#TEMPLATE={ "template": "Living employees of ?company with no known ORCID ID", "variables": { "?company": { "query": "SELECT ?id ?idLabel WHERE { ?id wdt:P1128 ?numEmployees. } ORDER BY DESC(?numEmployees) LIMIT 100" } } }
#Enter employer to find living employees with no known ORCID iD
SELECT DISTINCT ?person  ?personLabel WHERE {
  BIND(wd:Q223429 AS ?company).
  ?person wdt:P31 wd:Q5.
  ?person wdt:P108 ?company.                   # employed at...
  FILTER NOT EXISTS { ?person wdt:P496 ?orcid. } # no ORCID iD
  FILTER NOT EXISTS { ?person wdt:P570 ?date. }  # not dead
  SERVICE wikibase:label { bd:serviceParam wikibase:language "en" }
}

Try it!

– embed link. The template specifies the query description with the variables used (?company), as well as a query for suggestions for each variable (in this case, I chose the 100 companies with the most employees (P1128)). --TweetsFactsAndQueries (talk) 08:26, 30 May 2017 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ I'm grateful to both User:Matěj Suchánek and User:TweetsFactsAndQueries for their replies, but I fear each has missed my point. I want to send a URL to non-Wikidatans (perhaps via a mailing list where the recipients are not all known to me), which will present them with a form where they can enter the name of an institution (or its QID), and insert that onto and then execute the query I specified, without them ever seeing query code. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's editsl~

TweetsFactsAndQueries is close to that, you only have to click on the question mark icon though. Hope there will be more configurable display options soon. Sjoerd de Bruin (talk) 17:56, 30 May 2017 (UTC)

Property proposal template

There are several long-standing requests at Template talk:Property proposal which require the attention of someone with Lua skills. Can anyone assist, please? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 09:58, 30 May 2017 (UTC)

I will do my best. Matěj Suchánek (talk) 10:25, 30 May 2017 (UTC)

Internationalization

This change looked strange to me. But then I realized that in other languages than my own, you write things like this in other ways than I do. How do we internationalize things like this in a good way? -- Innocent bystander (talk) 16:00, 30 May 2017 (UTC)

See also Property talk:P281#Ranges. Apart from postal code (P281) are there more cases "like this"? --Marsupium (talk) 22:38, 30 May 2017 (UTC)

Well, almost any range (with some exceptions) uses a longer – instead of - in Swedish, that includes years like 1970–2017. One exception I am aware of is date-ranges like 1970-04-16--1974-04-14 which uses -- instead. The Polish postal codes looks like they includes a - makes it even more complicated than I first expected. -- Innocent bystander (talk) 06:55, 31 May 2017 (UTC)

Better link: Property talk:P281#Incorrect postal code change. --Edgars2007 (talk) 12:12, 31 May 2017 (UTC)

Template for user page that shows the Wikiprojects in which I'm active

Is there a template that I can put onto my user page that lists the Wikiprojects in which I'm a participant automatically? ChristianKl (talk) 17:48, 30 May 2017 (UTC)

@ChristianKl: Not that I know, but I use this link: https://meta.wikimedia.org/w/index.php?title=Special%3ACentralAuth&target=. It gives a neat summary of all wikis you have ever entered and it keeps track of your activities per wiki. You can make a list with this page as source. Q.Zanden questions? 17:55, 30 May 2017 (UTC)

That link doesn't seem to offer any information about WikiProjects. I think it would be useful if users could put a template on their user pages that show a list of their WikiProjects under their bable box to make the WikiProjects more discoverable. ChristianKl (talk) 18:27, 30 May 2017 (UTC)

Sorry, I misunderstood your question. Normally (on nl.wp), every WikiProject creates his own infobox on a page like Wikidata:WikiProject xxx/Infobox. That way every WikiProject is responsible for their own visibility. Q.Zanden questions? 22:50, 30 May 2017 (UTC)

Properties for deletion - closures needed

On 12 May, I posted to the Administrators' noticeboard, saying

The following discussions appear to have reached consensus, or stalemate, and are ready for closing and (where applicable) deletions:

• Wikidata:Properties for deletion#doubles record (P555) and singles record (P564)
• Wikidata:Properties for deletion#comment (P2315)
• Wikidata:Properties for deletion#Property:P546
• Wikidata:Properties for deletion#Property:P2837
• Wikidata:Properties for deletion#Yahoo! Japan Talent Database ID (P3284)
• Wikidata:Properties for deletion#Property:P3417
• Wikidata:Properties for deletion#IDEAS person ID (P3649)

Please can someone act?

The section has now been archived. Only one of the discussions I listed, struck through above, has been closed. Do we need more admins? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:28, 30 May 2017 (UTC)

Quantity is not the solution. We need admins that can make sense of these large complicated discussions and have the braveness of making a tough decision on them. It's not the most rewarding task. Sjoerd de Bruin (talk) 17:13, 31 May 2017 (UTC)

Sitelinks to redirects

I am trying to link this video game to this enwiki redirect, but Wikidata is ignoring the new link and using this as the sitelink instead. Is there a workaround? I plan on adding new records for other titles in the series, so the same problem will occur again. SharkD (talk) 23:45, 30 May 2017 (UTC)

@SharkD: Feel free to opine on this topic here. For the moment, though, I'm sure it is possible but I do not recall the exact hack used to make it happen. Mahir256 (talk) 02:29, 31 May 2017 (UTC)

The hack that works at the moment is to first make an edit that breaks the redirect. Then you set the Wikidata sitelink. Afterward you reset the redirect, so that it works properly. ChristianKl (talk) 11:25, 31 May 2017 (UTC)

DBLP data donation

Colleagues at dblp computer science bibliography (Q1224715), whom I met at WikiCite last week, have begun to donate data. My bot has just processed the first batch, with >4,800 values, including >1,300 DBLP author ID (P2456) (a few duplicates were ignored), plus assorted aliases, and values for VIAF ID (P214), GND ID (P227), ORCID iD (P496), ACM Digital Library author ID (P864), zbMATH author ID (P1556), & Google Scholar author ID (P1960). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:32, 31 May 2017 (UTC)

Extract items from article list

Is there any tool for extracting Wikidata items from a list of articles in a given language? I.e. I want to add a statement to all entries in a manually created list in Wikipedia, it's simple by using QuickStatements, but then I need to have know the item-ID for each article. Instead of manually retrieving each item-ID I want to upload a list of articles and then a list is generated containing items and article labels. --Cavernia (talk) 17:46, 31 May 2017 (UTC)

The old Quick Statements tool does this. I hope it is up again, after it had some downtimes recently. —MisterSynergy (talk) 17:52, 31 May 2017 (UTC)

How then? --Cavernia (talk) 18:02, 31 May 2017 (UTC)

In the very basic form you need to provide input tripels for each edit (subject item, property, value; separated by tabulators). The subject item is typically a Wikidata Q-ID, but the tool can also deal with Wikipedia page titles – you just need to give the corresponding wiki into the box left to the “Do it” button. Before the editing starts, the tool will lookup the Q-IDs and process the input. This also works with more complex input, e.g. if you add sources. —MisterSynergy (talk) 18:12, 31 May 2017 (UTC)

This will only work for items, not for values. If I have a list of 1000 items containing 7 parameters for each item, and I want to import 5 of them into Wikidata, but then I need to know the item IDs of the parameters which would of course take some hours to collect manually. I can write a command line script which does this operation, but I wanted to know if there already is a simpler solution. --Cavernia (talk) 19:16, 31 May 2017 (UTC)

Oh I see, you’d like to convert values, not subject items. PagePile would be another option then. Create a pile containing your wikipedia titles first; then, there is a button without text on this tool (between “Create a pile” and “Recent piles”) which guides you to a filter tool; select your page pile ID, and the “To Wikidata” filter will translate it to Q-IDs. —MisterSynergy (talk) 19:48, 31 May 2017 (UTC)

Excellent! --Cavernia (talk) 20:15, 31 May 2017 (UTC)

Lifestyle

The property proposal of lifestyle (P1576) listed values as vegetarianism (Q83364) and mysticism (Q45996) example, but the current use is mostly vegetarian (Q18338317) and mystic (Q12328016). I personally think the proposal was right and the current values should be converted. Please leave your thoughts here. Sjoerd de Bruin (talk) 20:08, 31 May 2017 (UTC)