User talk:ArthurPSmith/Archive/1

Welcome to Wikidata, ArthurPSmith!

Wikidata is a free knowledge base that you can edit! It can be read and edited by humans and machines alike and you can go to any item page now and add to this ever-growing database!

Need some help getting started? Here are some pages you can familiarize yourself with:

• Introduction – An introduction to the project.
• Wikidata tours – Interactive tutorials to show you how Wikidata works.
• Community portal – The portal for community members.
• User options – including the 'Babel' extension, to set your language preferences.
• Contents – The main help page for editing and using the site.
• Project chat – Discussions about the project.
• Tools – A collection of user-developed tools to allow for easier completion of some tasks.

Please remember to sign your messages on talk pages by typing four tildes (~~~~); this will automatically insert your username and the date.

If you have any questions, don't hesitate to ask on Project chat. If you want to try out editing, you can use the sandbox to try. Once again, welcome, and I hope you quickly feel comfortable here, and become an active editor for Wikidata.

Best regards! Liuxinyu970226 (talk) 09:07, 25 August 2015 (UTC)

Pywikibot for half-life claims

Hi! I am just writing a tutorial script on how to add quantities and units to Wikidata. It would probably be easy to adapt the script to add the NNDC values. Will post it a little later today to Wikidata:Pywikibot - Python 3 Tutorial. --Tobias1984 (talk) 14:52, 14 October 2015 (UTC)

I posted the example, but there might still be a problem with the pywikibot-api (Wikidata:Contact_the_development_team#Pywikibot:_Float_numbers). Will add more text and explanations to the example tomorrow. --Tobias1984 (talk) 21:23, 14 October 2015 (UTC)

Thanks for that example - I had been working on the other end (automated extraction of the data from the nndc page) so this will tie it together nicely. Does this pywikibot version require Python 3 or would it work in 2.7 also? ArthurPSmith (talk) 13:24, 15 October 2015 (UTC)

I ran the script against pywikibot core (See: Wikidata:Pywikibot - Python 3 Tutorial/Setting up Shop). According to this page https://www.mediawiki.org/wiki/Manual:Pywikibot/Installation#Initial_setup it should work with Python 2.7, but I think you need to import from __future__ import print_function, unicode_literals in the first line of the script so you don't have to change the print-functions and don't need to put u"something" in front of every string. --Tobias1984 (talk) 14:13, 15 October 2015 (UTC)

But I think we should wait for the (Wikidata:Contact_the_development_team#Pywikibot:_Float_numbers) issue to be resolved. It causes some really human-unfriendly numbers and diffs. --Tobias1984 (talk) 16:58, 15 October 2015 (UTC)

Noble gas

You reverted my edit on Q1307. You are right, but the czech translation is wrong. How could I edit czech translation of noble gas in this section?

Hi @Dvorapa: - looks like you figured out what to do, it's just a matter of editing the language labels on the target item, noble gases (Q19609). ArthurPSmith (talk) 01:54, 27 November 2015 (UTC)

Yeah, I figured it out finally. --Dvorapa (talk) 21:43, 5 December 2015 (UTC)

Property creator

You've just been granted the right of Property Creator. Good luck with it. And I hope you will use it wisely. New properties can be created at Special:NewProperty. Most important step is to choose the right datatype as it can't be altered afterwards. If you have any questions feel free to as another property creator or an admin with experience on creating properties. Mbch331 (talk) 21:27, 26 November 2015 (UTC)

Thanks! I really wasn't sure if I had enough experience for this, so I'm grateful for the trust. I will definitely be very cautious to start with while I'm still learning how things work here! ArthurPSmith (talk) 02:05, 27 November 2015 (UTC)

GRID mappings

Hi,

Missed the completion of adding GRID ids to wikidata and you'd asked me a question which was then cleared out I think, sorry about that. Great to see it added!

Yeah, mapping things by name is a real pain, it's part of why we built a tool for mapping unstructured text to ids. There's some good examples of issues here: http://symplectic.co.uk/product-news/grid-and-elements/

All the added data to the db is found or checked manually, so we might be able to add some but it'll depend on time/speed/cost. A list of potential things to add (e.g. things marked as universities in wikidata) would be a great start.

Although not all the metadata is filled in (wikipedia URLs, ISNI, etc) we've focused on the institutes which have the most scientific output as much as we can, so hopefully the top 1/4 with more automatically mappable IDs should represent the most 'important' ones to link up.

I'm always up for more of a chat about this, if you drop me a line at i.calvert@digital-science.com and we'll see how we can help out.

Ok, I'll probably be in touch! I have started adding the grid id's for institutions in your data with wikipedia references via the Quick Statements tool. Note there's a few wikipedia entries in your data that don't look right: some have an additional '#' linking to a portion of the wikipedia page (not helpful for identifying!) - for example 'Nobel Foundation'. And at least one ('American Medical Association') has an extra space at the end of the wikipedia string link. This is from the 12-14-2015 dataset. ArthurPSmith (talk) 17:08, 6 January 2016 (UTC)

Great! I'll have a look at the wikipedia page urls with anchors in. Thanks for the note about trailing spaces, it's an issue that pops up in some of the fields occasionally, I'll look at getting that fixed for the next release. If you have any lists of problems or possible issues then we can turn that into review tasks to check the data is right. For things without a wikipedia url I think there might be some more we can help with on our side rather than doing some fuzzy matching on names. IanCalvertDsci (talk) 10:59, 8 January 2016 (UTC)

By the way I completed the import from the latest dump based on wikipedia URL's - almost 10,500 entries. I'm looking at the ISNI values now. Those are also a little inconsistent - most are in standard ISNI format (4 groups of 4 digits separated by spaces) but a few of them are missing the spaces, some are missing leading zeros (I assume) and some have some extra space characters. So I need to normalize a little before doing a search. ArthurPSmith (talk) 16:46, 8 January 2016 (UTC)

Ha! I did a bunch of work on ISNI, then thought to run the query to see how many organizations actually have ISNI entries in wikidata: see http:// tinyurl.com/jnds3kr - only 245! So it will actually not help much with matching. Oh well. It could help with a bit of cross-checking between the two cases though. Also I'm starting to look at constraint issues - there are a bunch of duplicates in GRID! Will email you. ArthurPSmith (talk) 21:53, 8 January 2016 (UTC)

oh - actually there were a few more, see this query - tinyurl.com/jkmmkff - including subclasses just for university there are 440. Still it seems to be a small fraction of the actual assigned ISNI's for organizations. The vast majority of the ones in your list don't seem to match (by ISNI) anything currently in wikidata. ArthurPSmith (talk) 22:02, 8 January 2016 (UTC)

is a list of (P360) for categories

Thank you for belatedly returning to that RfC.

However, having thought about it since, given that there was (some) opposition to using is a list of (P360) on categories, perhaps a neater solution might be instead to introduce a parallel property to be specifically for categories, eg "category contains", but using the same syntax.

That way people could query specifically either for categories or lists with this kind of specification, without having to then filter their results as to whether the corresponding item was a category or a list.

What do you think? Jheald (talk) 18:58, 15 April 2016 (UTC)

Hi Jheald - A separate property would be ok with me, but we'd have to fix a lot of existing uses. I don't understand your example use case though - if you were doing a query to find all items that had the property is a list of (P360) xxx item with yyyy qualifier, wouldn't you want to see both categories and lists? If it was really important to limit to one or the other type you can always do that with an additional instance of (P31) criterion. As far as the label on the property being confusing in the case of categories, maybe we should add a "wikidata usage instructions" statement to explain what it means in the category case? ArthurPSmith (talk) 19:39, 15 April 2016 (UTC)

p2888

exact match (P2888) is ready. --Tobias1984 (talk) 19:30, 5 June 2016 (UTC)