Wikidata:Edit groups/QSv2T/1579418272014

Edit group QSv2T/1579418272014

Summary	#quickstatements;	Author	Lewoniewski
Number of edits	2,889 (more statistics)	Example edit	Q445216

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

The conclusion seems to revert the batch as described--Ymblanter (talk) 20:09, 3 February 2020 (UTC)[reply]

@Ymblanter: Precisation: not only the single batch, but also, as I wrote, revert all the additions of occupation (P106)scientist (Q901) made on 18/1, 19/1, 27/1, 28/1. --Epìdosis 22:31, 3 February 2020 (UTC)[reply]

Yes, indeed--Ymblanter (talk) 06:33, 4 February 2020 (UTC)[reply]

Discussion[edit]

@Lewoniewski: I propose to revert this batch (2889 edits), the previous batch 1579368136777 (100 edits) and, in general, to revert all the additions of occupation (P106)scientist (Q901) made starting yesterday and today (nearly 17k edits in total, at the moment I write) for the following reasons, which are mainly those pointed out by @Sapphorain, Ghuron: here:

it hasn't any source and the criterium of addition isn't clear
it is clearly wrong in some cases (e.g. Quintus Aelius Tubero (Q2123334), Ctesias (Q296325), Franz Bücheler (Q72739))
it is redundant in most cases (most of the edited items already have more precise values of occupation (P106), in particular values which are subclass of (P279)scientist (Q901))
1. this redundancy can be badly reflected on infoboxes on Wikipedias and other Wikimedia project

If an item has a value of occupation (P106) which has subclass of (P279)scientist (Q901), it can be found as a scientist by this query:

SELECT ?person { ?person wdt:P31 wd:Q5; wdt:P106/wdt:P279* wd:Q901 }

Try it!

I request also the opinion of

WikiProject Ontology has more than 50 participants and couldn't be pinged. Please post on the WikiProject's talk page instead. and

Notified participants of WikiProject Data Quality. --Epìdosis 13:27, 19 January 2020 (UTC)[reply]

I don't agree. Statements was added based on related Wikipedia articles in Polish, Russian and other language editions. As I mentioned before in the discussion, the query above doesn't give clear list of the related items. Additionally, I didn't find the list of "general" classes that can be not added to Wikidata items when there is some specific subclass.--Lewoniewski (talk) 20:19, 19 January 2020 (UTC)[reply]
- @Lewoniewski: My home wiki is Russian, and I'm not sure I understand your explanation. Yes, people that were both politician (Q82955) and scientist (Q901) will be listed, which is completely ok. Since economist (Q188094)subclass of (P279)social scientist (Q15319501) and social scientist (Q15319501)subclass of (P279)scientist (Q901), economists will also be listed here. Anyway, if you will find discrepancy in hierarchy, it is better for the project to fix rather than artificially add hundred thousands redundant statements. And yes, we do not have list of "forbidden" actions, we suppose to use common sense instead. Ghuron (talk) 04:05, 20 January 2020 (UTC)[reply]
- @Lewoniewski: I think I understand the source of confusion now. In SPARQL when you write wdt:P279* wd:Q901(note asterisk) it will give you not only direct sub-class of scientist (Q901) but also sub-sub-classes, sub-sub-sub-classes, etc (see [1]). So as long as we maintain Q901 hierarchy, my query should give you complete list of scientists Ghuron (talk) 07:19, 20 January 2020 (UTC)[reply]
  - @Ghuron: Exactly. Actually Wikidata ontology not perfect and can be improved. Therefore when I will use the SPARQL query that you proposed (with wdt:P279* wd:Q901) I will find among scientists also Q7747 (Vladimir Putin) because economist (Q188094) is a "sub-sub-class" of scientist (Q901). Even more, if somebody is aligned to active user (Q22075476), Wikipedian (Q23038345), black hat (Q1332289), Jesuit (Q63173086), naturalist (Q18805), ecologist (Q15839134), command module pilot (Q20114066), pseudoscientist (Q22082749), cracker (Q32554) - they will be also classified as scientist (Q901). The full list of the subclasses (over 600) you can find by using this query. I am not sure if in the near future it will be possible to solve the similar problems with Wikidata ontology. Therefore I decided to enrich the Wikidata items by providing more unambiguous statements, so users will have access to data with higher quality. --Lewoniewski (talk) 21:04, 27 January 2020 (UTC)[reply]
    - @Lewoniewski: ontology is far from perfect, but I do not see how does it prevent one from fixing a small (<1000 items) subset of scientist (Q901) subclasses. This is collaborative project, we are supposed to work together to improve quality of data in a consensus way rather than invent a new ontology to fix very specific query. And since you are using "higher quality" data from Category:Scientists in alphabetical order (Q6620777), you are going to assign Gleb Nosovsky (Q356003)occupation (P106)scientist (Q901) which would be funny, since he is clearly pseudoscientist (Q22082749) and you do not seem to consider them as true scientists.
    - Ok, since we are clearly in disagreement here, may be there is a way to fix your query without adding too many occupation (P106)scientist (Q901)? For instance, you can exclude pseudoscientists like:
      SELECT ?person { ?person wdt:P31 wd:Q5; wdt:P106 ?occupation FILTER (?occupation != wd:Q22082749) ?occupation wdt:P279* wd:Q901 }
      Try it!
      alternatively you can request all items, that has at least one article, that is included in Category:Scientists in alphabetical order (Q6620777):
      PREFIX mw: <http://tools.wmflabs.org/mw2sparql/ontology#> SELECT DISTINCT ?item { hint:Query hint:optimizer "None" . ?cat schema:about wd:Q6620777 SERVICE <http://tools.wmflabs.org/mw2sparql/sparql> { ?page mw:inCategory ?cat } ?page schema:about ?item . }
      Try it!
      (see mw:MW2SPARQL for more details) Ghuron (talk) 07:37, 28 January 2020 (UTC)[reply]
Support reversion. Infovarius (talk) 09:12, 24 January 2020 (UTC)[reply]
- @Lewoniewski: Please do stop new unsourced additions of occupation (P106)scientist (Q901) until the discussion reaches a conclusion. Sapphorain (talk) 23:45, 27 January 2020 (UTC)[reply]
  - I blocked new additions for one day. I edit my initial proposal in this way: revert all the additions of occupation (P106)scientist (Q901) made on 18/1, 19/1, 27/1, 28/1 - at the moment there are four users supporting this revert (Sapphorain, Ghuron, Infovarius and I) and only one against (Lewoniewski). --Epìdosis 23:52, 27 January 2020 (UTC)[reply]
    - Ok, great. When can we expect the final decision? --Lewoniewski (talk) 07:09, 28 January 2020 (UTC)[reply]
      - @Lewoniewski: I think we can close the discussion with a decision on the 2nd of February, after two weeks from the start. --Epìdosis 09:12, 28 January 2020 (UTC)[reply]

The above discussion is preserved as an archive. Please do not modify it. Subsequent comments should be made in a new section.

Wikidata:Edit groups/QSv2T/1579418272014

Edit group QSv2T/1579418272014

Discussion[edit]

Navigation menu

Search