Wikidata:Project chat/Archive/2019/05

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

qualifies for event and winner

Hello. Do you have any solution for the constraint in Q29937708#P3085? Xaris333 (talk) 00:05, 22 April 2019 (UTC)

together with (P1706) should do it. Circeus (talk) 00:53, 22 April 2019 (UTC)
I don't think is a good solution. Xaris333 (talk) 01:00, 22 April 2019 (UTC)
You do realize that property literally exists for exactly that use case? Circeus (talk) 15:26, 22 April 2019 (UTC)
OK. But this is not so helpful. The first team has the qualifier winner and the other the qualifier together with. Infobox templates will need some changes... Xaris333 (talk) 17:06, 22 April 2019 (UTC)

@Circeus: Problem: "allowed qualifiers constraint" "together with is not a valid qualifier for qualifies for event – the only valid qualifiers are:..." Xaris333 (talk) 21:53, 22 April 2019 (UTC)

Easy enough to fix. May take a little bit to propagate properly though. Circeus (talk) 00:44, 24 April 2019 (UTC)
  Question @Circeus, Xaris333: Why having all this redundancy ? We have already an item for the competition in which there is already the winner of the competition, for a start. Second, if there is an item for the team, there is already the club somewhere in the item of the team, is’nt it ? Results, we have the same information stored potentially for each competition the team won, for each qualifying competition … Maybe some people like to enter the same information several time by hand but as many time as necessary but … isn’t it smarter to try to make the infobox code more able to retrieve itself the information stored in other items ? author  TomT0m / talk page 08:28, 30 April 2019 (UTC)
That would be the way to go if there was a SPARQL interface available from infobox/module code. Unfortunely, we cannot use SPARQL queries in that context, which makes it really difficult to access data which is not stored in a connected item directly, or in items which are (forward) linked from such an item. If there is just a backlink as you indicate, one can barely use such data in infoboxes; redundancy is thus necessary to enable data use in Wikipedias. —MisterSynergy (talk) 08:49, 30 April 2019 (UTC)
@TomT0m: What do you mean? See Q24529775#P3085. Where else we have the information that Valencia CF (Q10333) qualified for 2018–19 UEFA Champions League (Q30032467) through 2017–18 La Liga (Q24529775)? Xaris333 (talk) 10:34, 30 April 2019 (UTC)
@Circeus, Xaris333: (answering both) Oh, sorry I misinterpreted the « winner » qualifier as « winner of the competition », not as « winner of the qualification for the competition». For some reason it would have been clearer if « winner » qualified a « reward » statement, but somewhat more indirect if that requires items for qualifications for competitions like « reward : qualification for the Champion’s league ». Maybe including « reward » in the naming of the property would avoid the confusion, like « reward as qualification for ». But it’s not that important. author  TomT0m / talk page 08:16, 1 May 2019 (UTC)

Grafana API for data retrieval

I am looking for a way to retrieve data from some of the Wikidata-related Grafana charts. Does Grafana have a suitable API for such a task, and does anyone already have experience how to use it? —MisterSynergy (talk) 07:55, 30 April 2019 (UTC)

@MisterSynergy: Grafana works just like Wikidata: API + a ton of json. So no HTML pages begin generated. If you just open and hit F12 (Firefox) you'll see things like [1] being loaded. Have a look at to see how the API is supposed to work. Multichill (talk) 18:55, 1 May 2019 (UTC)

Jean-Francois Aubry (Q57172378)

I came across this item with the repeatedly wrongly added ORCID-number, and I noticed that it is wrongly linked from a lot of scholarly articles. Could somebody please fix this issue? And besides this specific case, are there maybe more cases like this? I think it should be possible to write a query to find such cases. A contemporary constraint for scholarly articles and the stated authors might also be useful. 12:48, 30 April 2019 (UTC)

@GerardM:, who added the ORCID iD (which I since now removed) in this edit. As a more general point, such cases (ORCID iDs added to items about people who died before 2012, when ORCID came into existence) are already trapped, at Wikidata:Database reports/Complex constraint violations/P496. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:54, 30 April 2019 (UTC)
I know, thats how I found it. But dead people being linked as authors in scholarly items are not trapped yet as far as I know. 12:56, 30 April 2019 (UTC)
Well actually a perfectly fine situation existed and you assumed that there is only one person by that name.. Check this out and revert your changes.. Thanks, GerardM (talk) 13:58, 30 April 2019 (UTC)
I assumed no such thing. And the item as you left it, with an ORCID and an 1802 date of death, was far from "perfectly fine". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:58, 30 April 2019 (UTC)
@Pigsonthewing: Actually that item was ORIGINALLY created with an ORCID and links to this person as author of scholarly articles, but then Reinheitsgebot added birth and death dates based on this page, which appears to be a completely different person (the name is even different - just "Francois", not "Jean-Francois"). I'm not sure how to fix this to prevent bot actions to break it again, but GerardM is correct that removing the ORCID is not the right solution here. ArthurPSmith (talk) 17:16, 30 April 2019 (UTC)
Actually it is not hard. Remove the stuff that does not belong with the original author and add a new item for the one who is dead. Thanks, GerardM (talk) 17:36, 30 April 2019 (UTC)
But Reinheitsgebot seemed to be working of the ORCID id which is weird. Maybe that would be fixed now? ArthurPSmith (talk) 20:42, 30 April 2019 (UTC)
Anyway, at least we can do the first bit of cleanup - see François Aubry (Q63400224). ArthurPSmith (talk) 20:48, 30 April 2019 (UTC)
Hoi, people make mistakes. It was a case of mistaken identity. Perfectly normal. Remember people make 6% errors on average cherish this for the other 94%. Thanks, GerardM (talk) 05:25, 1 May 2019 (UTC)

Catalan help with kraft paper

Can someone look at the Catalan articles for Kraft paper (Q1755609) and Kraft paper (Q11940064) and tell me how they should be distinguished in their descriptions? It's possible one of them should be made equivalent to co:Category:Brown paper, but the two concepts look pretty much identical to me. - PKM (talk) 20:35, 1 May 2019 (UTC)

expert talk

How to set a talk to an expert as a reference? Regards, Conny (talk) 21:09, 30 April 2019 (UTC).

One way would be to put the details of the talk in a user subpage, or some other web page, and use it's URL. That wouldn't be a very good source though, according to Help:Sources. Ghouston (talk) 00:31, 1 May 2019 (UTC)
@Conny: No original work like in WP, only fact. If your expert is good enough, he can provide you a reference describing what is correct according to its knowledge. I don't support the proposition of Ghouston which breaks all rules of references. Snipre (talk) 14:18, 1 May 2019 (UTC)
@Snipre: Thank you very much! Regards, Conny (talk) 14:29, 1 May 2019 (UTC).
It would be nice if Wikidata (and Wikimedia projects in general) did have a method of interviewing people, preferably with audio and video recordings. It could be just as reliable as a journalist doing the same thing, and there are many people who journalists never get around to, or when they do, they don't necessarily ask about certain details (like citizenship and year of birth). Ghouston (talk) 22:15, 1 May 2019 (UTC)
This is basically the "wiki-affidavit" idea I proposed last year at the 2018 Wikimedia Conference. - Jmabel (talk) 23:02, 1 May 2019 (UTC)
This is dangerous because WD or Wikimedia start to produce some references, this implies some responsibility in term of publication especially if that media are used as sources for articles or statements. I find really strange that an expert can't provide any documents supporting his affirmation: nothing prevents him to mention one of its own document. Without any document, this is an original work, so no valid in WP at least perhaps even in WD. Snipre (talk) 14:46, 2 May 2019 (UTC)
you could always upload your oral history and primary material to commons and use at wikinews. i would be happy to transcribe oral histories at wikisource.
the danger is in editors who proscribe how to create encyclopedic content. we typically link to oral histories under external links at wikipedia. Slowking4 (talk) 15:24, 2 May 2019 (UTC)

Series of property in an item

Hello. Just asking: was there a change about the series the properties are appearing in the items? I have noticed some changes. And how often that is happening and who is decide about that? Thanks. Xaris333 (talk) 23:45, 1 May 2019 (UTC)

@Xaris333: See MediaWiki:Wikibase-SortedProperties. --Marsupium (talk) 13:14, 2 May 2019 (UTC)

Megan Leroy

I am an agent trying to create a wikipedia page for a client. I have her websites, videos, articles, etc. How can I go about writing this? --Virginia Actors (talk) 02:18, 2 May 2019 (UTC)

Hello, Virginiaactors1 (talkcontribslogs). It looks like you've created the Wikidata item Megan Leroy (Q63431046). Wikidata is for machine-readable information, not the articles you're probably familiar with. If you want to write a Wikipedia article, you're going to need to go to Wikipedia's Your First Article page to create it.
While this isn't strictly a matter for Wikidata, I should caution you that you might run into problems on Wikipedia. In particular, since you're an agent you need to follow the paid-contribution policy. You also need to verify that your client is considered notable under relevant notability policy; if she isn't, any page you create will likely be deleted.
If you need more help, you can ask over on the Wikipedia Teahouse. ―Vahurzpu (talk) 04:33, 2 May 2019 (UTC)

Wikidata:ScienceSource project, Facto Post 23

There is now some reasonably complete documentation for ScienceSource (Q55439927) at Wikidata:ScienceSource project. Still work in progress, certainly. One of the links there was mentioned in Facto Post 23, out on Monday: Wikidata:ScienceSource project/NCBI2wikidata dashboard gives the basic story about metadata additions here by the project.

Wikidata:ScienceSource project/Project tools is about the software for the project, now complete in terms of tools. In the last row is, the new front end for the ScienceSource wiki. I'm not going to explain all about that here, but please prompt me about it.

Please create an account on and test the front end, anyway. Charles Matthews (talk) 11:17, 2 May 2019 (UTC)

growth of wikidata

why growth of wikidata is slow but many items in the world examplethis address or in world have 100000000 scholar articles but in wikidat have 18000000  – The preceding unsigned comment was added by Amirh123 (talk • contribs).

In my opinion WD is growing too fast. A lot of areas need some consolidation. This includes scholarly articles. --Succu (talk) 18:17, 19 April 2019 (UTC)
I often work with scholarly articles relative to taxa, and an issue I see is that the author(s) are very often quoted as text, even when the corresponding items, for the author(s), exists. Christian Ferrer (talk) 08:31, 20 April 2019 (UTC)
Depends on the purpose we have into having the scholarly articles on Wikidata. One of the purposes is to ease the reference work of statements, it then makes sense to already have the article you want to cite on Wikidata than to have to create an item or add many snaks for this. Of course this poses the problem to know if and when the datas will be cleaned up and the string statements will be translated into their structured counterpart. I think it depends, there is easy cleaning that will clean a lot of datas at once and a long tail of very hard cases with barely known or hard to identify authors that may not be done. It may not be a real problem. author  TomT0m / talk page 09:38, 20 April 2019 (UTC)
Agree. Scholarly article items are generally complete and rarely include duplicates. There might be one or the other aspects in the way these are created that could be improved (read: improve the bot script). It would be more of an issue if one created a separate item for every author in these and someone else would be expected to merge them afterwards.
Also, there is a proposal to store all of them in a separate Wikibase. This might allow optimized maintenance.
BTW has some promising features.
In other fields, even a several groups of 1000 items for the same concepts with hardly any statements can be much more burdensome. One needs to identify the concept, add statements, and merge duplicates. --- Jura 09:51, 20 April 2019 (UTC)
For me are bad formatted titles, attribution to the wrong language and missing end pages a serious problem. Do we have a clue of how many statements are referenced by scholarly articles? --Succu (talk) 18:51, 20 April 2019 (UTC)
Bad titles are a problem. I am regularly spotting instances where a paper is given a title that is clearly and unambiguously wrong to anyone who looks at the pdf, but that's what the publisher hands out as faulty data ([2], [3], [4], [5]). Don't even get me started on book reviews where the author of the reviewed work ends up in the author field of the article! Circeus (talk) 12:59, 29 April 2019 (UTC)
Imports solely based on BHL bibliography ID (P4327) (via BHL-API) are often wrong. I doubt we have processes to fix them other then by hand. --Succu (talk) 19:13, 29 April 2019 (UTC)

Meta question: why have scholarly papers and authors - are we a Wiki

Hoi, at this time I have a long andgrowing list of authors I want to process with SourceMD. All these people came to my attention as a result of the "real world", this includes twitter, new awards and the like. These people are all notable on a Wikipedia level, so are their subjects and often so are their papers. At this time all duplicate authors because of duplicate ORCiD identifiers are merged. I am quite happy to keep it that way depending on the availability of reports.

The meta question is, when we include authors and their papers there is no cut off point. A person is an author in Crosref, ORCiD and is therefore notable in Wikidata. When this is agreed upon, we need to use tooling like SourceMD. As a follow up when we want to use Wikidata for references, all the references and the meta data needs to find a place in Wikidata. There is no helping that. Linking papers to subjects does not need to rely on external sources, the fact that it is linked as a source to an article suffices. Our practice has been that we include any and all authors and their papers. The personal position that we should consolidate is then to be framed in words like "what does it take to get our scholarly information in a state where it is functionally complete" the alternative of having no scholarly papers is not feasible imho.

We are a Wiki so not being complete is part of how we do our thing. Getting out scholarly information to a higher level takes effort and collaboration. Thanks to SourceMD we have imported much from ORCiD for many disciplines. Were we to collaborate with ORCiD, we could invite scholars who have been authenticated in ORCiD, to update their profile in Wikidata, maybe for their co-authors as well. When references are linked to their papers, we signal this and invite comments. Collaboration is part of the Wikimedia 2030 strategy so why not just do this?

Thanks, GerardM (talk) 12:22, 21 April 2019 (UTC)
E.g., ORCID ID 0000-0003-2016-4208 is not notable, it's self-published. – 04:39, 23 April 2019 (UTC)
Why would this not be notable.. self published means that the origin is the person identified this in ORCiD. It does not follow that it is non-notable. Thanks, GerardM (talk) 08:32, 23 April 2019 (UTC)
Nobody can be sure of the real name, Google / IETF / OEIS / ORCID / etc. took it on my say-so, and the two former + mostly inactive WikiMedia accounts are worse. – 12:40, 29 April 2019 (UTC)
I have no idea what this is trying to say. Please make complete sentences. Circeus (talk) 12:50, 29 April 2019 (UTC)
It says ORCID ID is self-published shit with a notability below IMDb in other words. – 00:53, 3 May 2019 (UTC)
If it is saying that, then it is making itself look very foolish. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:35, 4 May 2019 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────"Were we to collaborate with ORCiD":

-- Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:34, 4 May 2019 (UTC)

African games 2007 boxing

My name is Ahmed Samir super heavyweight boxer I won a gold medal in African games 2007 in Algeria and they made a mistake with my name and they wrote my friend the heavyweight instead of mine  – The preceding unsigned comment was added by (talk • contribs).

Something changed with auto-suggest

Over the last two or tree days, I've noticed that when I type the name of an item as the value of a property, and the name is a full match for an item (not necessarily the item I want), the auto-suggest drop-down list disappears.

For example if I type "Birmingha", as a value for located in the administrative territorial entity (P131), I see a drop-down whose first entry is "Birmingham major city in England", but as soon as I type the final character, "m", that drop-down disappears, meaning that I cannot choose between Birmingham, England, Birmingham, Alabama, or Birmingham, Michigan. This is particularly unhelpful if I cut and paste the full term, as I never see the drop-down.

It also causes problems if the value I'm typing contains another word, for example the English word "Parisian" ceases to display a drop-down as soon as the string "Paris" is typed. It does not reappear when I type the "i" or the "a" (but, oddly, does when I type the "n").

Is it just me, or has something changed in the code? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:28, 4 May 2019 (UTC)

It's not just you. Very frustrating, the bug has been reported a few places, including here: and here Moebeus (talk) 11:07, 4 May 2019 (UTC)
Hitting the down arrow key seems to make the missing suggestion/search results data appear. On the iPad, clicking the field a second time works. - PKM (talk) 11:46, 4 May 2019 (UTC)

"Deprecated" geological periods

Does anyone have any ideas for how we should model geological periods that are part of "deprecated" or regional geological period classification systems? For an example of the problem, see en:Ordovician#Subdivisions. Many of the redlinks here have articles in other language Wikis and WD items (example Ashgillian (Q2474301)).

I'm thinking we need an item for each regional classification system and then make these items part of that classification system to avoid complicating our "main" ontology which follows the current international standard. But perhaps all of the immediate subdivisions of the Ordovician should be <part of> Ordovician, qualified with the relevant classification scheme. I'd like advice from experts in this field (which I am not!). - PKM (talk) 22:36, 3 May 2019 (UTC)
Pretty much, because the property meant exactly for this (criterion used (P1013)) require items to link to. Circeus (talk) 03:06, 4 May 2019 (UTC)
Metaclassification seems the way to go to me. See User:TomT0m/Classification (especially User:TomT0m/Classification#Classifying classes). criterion used (P1013) seems a little out of scope here, it would fit to link a class like
⟨ human child ⟩ subclass of (P279)   ⟨ human ⟩
criterion used (P1013)   ⟨ age ⟩
, that is link a class to the creteria used to discriminate with other member of the criteria, not so much to link a class to a « system » of classification which is not a criteria by itself, in my opinion. Or I did not get your point? @PKM, Circeus: author  TomT0m / talk page 08:48, 4 May 2019 (UTC)
PKM is saying there are multiple exclusive systems to subdivide Ordovician and asking how to properly take that into account when linking the various proposed eras to Ordovician (Q62100) or, I would assume, each others (i.e. through follows (P155) or followed by (P156)). Circeus (talk) 11:11, 4 May 2019 (UTC)
@Circeus: Yes, that's what I am looking for. In any case, it seems we need items for each classification system. - (Updated: Created ICS Standard Global Chronostratigraphic (Geochronologic) Scale (Q63463770), chronostratigraphic classification scheme (Q63463337), and various related items.) PKM (talk) 19:19, 4 May 2019 (UTC)

Research projects

Hi all, I came across a few items which were about researches (proposals). It is for example about PARTHENOS (Q48802992), an european mission; Q48802622, a specific German project to identify Slowaki handwritings from the 18th century. My question is: should we keep these kind of items, or should we make a clear line what we accept and not accept? Can someone help me out? Q.Zanden questions? 17:14, 4 May 2019 (UTC)

I think we should keep these, based on Notability Criterion #2: "It refers to an instance of a clearly identifiable conceptual or material entity. The entity must be notable, in the sense that it can be described using serious and publicly available references." - PKM (talk) 22:18, 4 May 2019 (UTC)

Help needed with sport properties

Hi, I have been asked to help with writing a module to create infoboxes for football players in dawiki from data in Wikidata (Template:Infobox football biography (Q5616966) is used now). I have identified some information needed for the infobox which not yet is in Wikidata:

  1. member of sports team (P54) is used for a player's teams. But I see no way to indicate if the player was a youth player or senior player when they were on the team. That is needed because the infobox have different sections for youth career and senior career. Can any existing property be used a qualifier to indicate that?
  2. coach of sports team (P6087) is used when a person is head coach for a team. But all types of coach jobs are in the football player infoboxes. What is the best way to indicate when a person is some other type of coach than head coach, like assistant coach, goal keeper coach etc.?

Besides, number of matches played/races/starts (P1350) and number of points/goals/set scored (P1351) are used as qualifiers for member of sports team (P54). It seems that they now only count matches and goals in the team's domestic league (as is required for the infoboxes), but as far as I know it isn't specified anywhere that cup matches, international matches or others cannot be counted with these qualifiers which could be a problem. Do you have ideas how to make sure to get counts for the domestic league only?

I have alreay asked my questions in WikiProject Association football, but got no useful replies to the questions. So I hope you can help. Thank you. --Dipsacus fullonum (talk) 11:58, 4 May 2019 (UTC)

For your former question a new item should be created, as needed, for, say "Manchester United Youth Team". For the latter we may need new properties; but it seems like the data model is sub-optimal. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:53, 4 May 2019 (UTC)
MisterSynergy (talk) 18:48, 4 May 2019 (UTC)
@MisterSynergy: @Pigsonthewing: @Dipsacus fullonum: May I ask an additional question then? I now have head coach (P286) , coach of sports team (P6087) for head coach (P286) as far as I can see are used 50/50 for association football club (Q476028) and persons having sport (P641). now, from dec 2018 is used for teams. Should then tems linking to head coach (P286) be corrected, and the English version describing P286 as on-field manager or head coach of a sports club (not to be confused with a general manager P505, which is not a coaching position) or person changed to an atlethic wich I think also will be the same as the german and French description. Breg Pmt (talk) 08:40, 5 May 2019 (UTC)
@Pmt: Sorry, I don't understand what you are asking. coach of sports team (P6087) is used for persons to indicate for what team they are head coach. head coach (P286) is used for teams to indicate who their head coach is. --Dipsacus fullonum (talk) 14:25, 5 May 2019 (UTC)
Ooops! Sorry! was thinking the wrong way somewhere?. Breg Pmt (talk) 14:32, 5 May 2019 (UTC)
@MisterSynergy: Thanks for your suggestions. However I cannot see how participant in (P1344) can solve the problem, because then you will miss for which team the goals was scored. I also fear that it may be difficult to find only the coach jobs using employer (P108). --Dipsacus fullonum (talk) 14:33, 5 May 2019 (UTC)
The participant in (P1344) claims would need to be equipped with member of sports team (P54) qualfiers (or something similar), besides the statistical value qualifiers. —MisterSynergy (talk) 14:36, 5 May 2019 (UTC)
I remind you that this way is wrong for loading data into infobox. Different language sections of Wikipedia may have different agreements for the counting of apps (goals) in domestic leagues. For example, with or without Football League Championship play-offs (Q4202932) or MLS Cup Playoffs (Q6716635). So, if we created an external database for this purpose, then there would be a table, each record of which would correspond to a single line of the infobox. In turn, player statistics in each season of the tournament (national league, national super cup, intertoto cup, etc.) would be associated with the records of that table. In any case, we will have to do this, because it is more reasonable to load statistics from Wikidata not only into infoboxes, but also into the tables for all player's career tournaments. Maybe we should use series ordinal (P1545) for member of sports team (P54)? Then we can do the following:

But you can notice how differently this is filled in infoboxes in ruwiki (2016—2018|2019—present), enwiki (2016—2019|2019—present) and dewiki (2016—present). You may ask why this is "2016—2018" if he left Borussia Dortmund (Q41420) on January 2, 2019? But I remind you case of Denis Boyarintsev (Q945319) who signed a contract with Spartak Moscow (Q29112) at December 25, 2008 and left club at January 8, 2010, so he played for Spartak Moscow (Q29112) only in 2009 Russian Premier League (Q1084791) and we should indicate this as "2009 | Spartak Moscow (Q29112) | 19 (0)" in infobox. Christian Pulisic (Q22279773) didn't play for Borussia Dortmund (Q41420) with #1 in 2019 as the team did not play a single match during this period of time, therefore we consider the date of the end of the #1 period as 2018. At least two problems remain: agree that the series ordinal (P1545) in participant in (P1344) is the reference to series ordinal (P1545) in member of sports team (P54) and suggest qualifiers for dates that correspond to the dates of the last and first matches, when a player was eligible to play for the team. At the same time, if someone wants to use the dates of signing contracts and making transfers in infoboxes, they must also be specified using other qualifiers. And it’s not quite clear to me how to separate club teams and national teams so that they fall into different sections of infoboxes. Сидик из ПТУ (talk) 20:18, 5 May 2019 (UTC)

Time of an event

2018 FIFA World Cup Final (Q31043671). Is there a way to show exactly the time the game started or that is an information we don't need to store in Wikidata? Xaris333 (talk) 14:02, 5 May 2019 (UTC)

I don't think it is possible so as long as the time datatype have day as it's highest precision. --Dipsacus fullonum (talk) 14:17, 5 May 2019 (UTC)
Special HH:MM:SS-formatted qualifier needed for it. Сидик из ПТУ (talk) 05:49, 6 May 2019 (UTC)

90+3 and not 93 minute

Hello. France 4-3 Argentina (Q16270333) -> points/goal scored by (P1363).

I thinks something is wrong there. The last goal (by Sergio Agüero (Q119562)) scored at the 3 minute of the additional time. After the 45 of the two-halves the time is counting like: 45+1, 45+2, 90+1, 90+2, 90+3 etc. We only count after 90 (91, 92, 93..., 120) only in overtime (Q186982). But in Wikidata we can write only 93 and not 90+3. In France 4-3 Argentina (Q16270333) the goal scored at 90+3 and not in 93 minute. Xaris333 (talk) 14:32, 5 May 2019 (UTC)

@Xaris333: I think that's what Wikidata:Property proposal/match interval was intended to address. Mahir256 (talk) 06:12, 6 May 2019 (UTC)

Julian Calendar - ERROR

Could you please check — date of birth in Julian calendar from Wikidata became wrong for Wikipedia (if the date passes after (over) a year):

  • 31 December 1894Julian (from Wikidata) — shown in Russian Wikipedia as: 31 декабря 1894 " .. bracket1 .. "12 января 1895" .. bracket2 .. " (example 1 and 2) --Ivtorov (talk) 09:37, 5 May 2019 (UTC)
  • It's a question for ruwiki. Maybe this change. --- Jura 10:55, 5 May 2019 (UTC)
    • Thank You very much Jura! --Ksc~ruwiki (talk) 17:31, 5 May 2019 (UTC)
    • We fixed it. The topic is over. --Ksc~ruwiki (talk) 18:30, 5 May 2019 (UTC)
  • Thank you! --Ivtorov (talk) 15:55, 6 May 2019 (UTC)

New status of constraint: "suggestion"

Hello all,

In April, we announced a new level of constraint, "suggestion", based on community request. The change is now deployed. You can now add suggestion constraint (Q62026391) as constraint status (P2316) on constraint statements. The notification icon will be a flag, unless a mandatory or ordinary constraint is already listed, in that case its icon will appear.

Please note that this is the first roll-out of the feature, some issues may appear. You are also very welcome to give feedback, if some things are not working as you wish they would, feel free to ping me or raise it in the ticket. Thanks, Lea Lacroix (WMDE) (talk) 12:17, 6 May 2019 (UTC)

Wikidata weekly summary #363

Categories versus Categories by language

Hello from el.witkionary, Hello User:Infovarius. In many languages a Grammatical category X and a Grammatical category X by languages is the same thing. e.g.

We try to add to the category where the CONTENT is relevant.The distinction at wikidata is not clear to me. I would understand it thus (hypothetically)

  • 0. Prespositions
  • 0.1. Prepositions by type
  • 0.2. Prepositions by language

But still, most wiktionaries, only have Cats 'by language' and place them under the title of the general Category.

Another example:

The problem is that we cannot link to en.wiktionary at all. Thank you. --Sarri.greek (talk) 22:49, 3 May 2019 (UTC) (and Sarri.greek@el.wiktionary - talk)

Maybe I'm an idiot, but manual Interwiki links are not, by and large, forbidden or deactivated. I'm pretty sure such links can (and should) still be used precisely in such instances where the distinction that exist at the Wikidata level is simply not relevant to the specific combinations of wikis at issue. Circeus (talk) 03:12, 4 May 2019 (UTC)
Sorry, my english is not that good. I do not understand the policy of wikidata as applied at the recent changes. But never mind. --Sarri.greek (talk) 08:09, 4 May 2019 (UTC)
It's not a wikidata policy. What I mean is if the Wikidata items cannot allow the el and en categories to be on the same item, you can have the manual link on the el:Κατηγορία:Προθέσεις wikipage to en:Category:Prepositions by language.
... you do know there is wikitext to make a manual interwiki sidebar link identical to the one generated through wikidata, right? Circeus (talk) 02:32, 5 May 2019 (UTC)
A, ok Circeus. I thought: because your people changed it, I have violated some policy of yours. Thank you for clarifying. --Sarri.greek (talk) 06:47, 7 May 2019 (UTC)

Integrate Wikidata in external Wikis

Hello, I asked this question here a few years ago, and it was not the moment, but maybe now there is a solution?
At Vikidia (Children Encyclopedia) based on Mediawiki, we would like to use Wikidata, like we use InstantCommons. Two reasons : use the data in LUA templates, and second, to handle our internal interwiki, and also the external interwiki (towards Wikipedia mainly). Is there a way?
Some pushes to install our own Wikibase and start a local Vikidata, but that would be a copy, instead of filling the same common database. WDYT? Plyd (talk) 14:46, 7 May 2019 (UTC)

Device models

There are quite a few items for devices, such as iPhone (Q621427), which use manufacturer (P176) to link to the name of a company, in this case Apple Inc. (Q312). This is a bit lazy, and inaccurate, since many products these days are manufactured by OEM manufacturers. E.g., iPhones are generally made by Foxconn (Q463094), but often such details won't be publicly available. A better method, which doesn't require extensive research, is to use brand (P1716) to link to an item for the brand, much like humans are linked to surnames. The brand can be linked to its owner using owned by (P127). E.g., Canon AE-1 (Q55674) -> Canon (Q63554165) -> Canon Inc. (Q62621). I think it would be necessary to create a new item for each brand, since conflations such as I made on Arri (Q11207), as both a company and a brand, are not really valid, and it relies on them having the same label, which may not be true in future or in all languages. To find all the devices for a particular company then requires a) finding the brands that it owns b) finding the devices using the brand. Ghouston (talk) 03:58, 8 May 2019 (UTC)


Can someone help me clear the error message in NC-4 (Q13458934) under Herbert C. Rodd? It seems that his occupation is already a subclass of military personnel, so it should not be giving an error message. --RAN (talk) 23:49, 8 May 2019 (UTC)

I believe only the subclass of and instance of property can "look up the tree" (through the relation (P2309) qualifier to a constraint), and other property (here the occupation (P106) on the crew member items) must use one of the items listed in crew member (P1029). Circeus (talk) 01:11, 9 May 2019 (UTC)

On behalf of…

Stumbled across this problem. When someone gets a award received (P166) on behalf of someone else, then which property should be used as qualifier? I can't find anyone, but it should be a quite common problem? Jeblad (talk) 20:04, 4 May 2019 (UTC)

Can you give an example? I can think of someone collecting an award that was issued to somebody else, who couldn't attend a ceremony in person, but the award isn't going to the person who collects it. Ghouston (talk) 01:50, 5 May 2019 (UTC)
that could also be the case for a widow collecting a medal after her husband's death, for example, if I understand the question correctly. --Hsarrazin (talk) 18:16, 8 May 2019 (UTC)
Some kind of organization gets an award, but the award itself is collected by the CEO of the org. If the org. itself has no office the award stays at the CEO. It is a quite common practice. Jeblad (talk) 10:29, 9 May 2019 (UTC)

Interwiki conflict

Interwiki conflict for example items = Olavo de Carvalho (Q729048)/Category:Olavo de Carvalho (Q32398192)

This is just an example but problem is general: placing 'Commons site links' (other sites) from topic (WD page) to topic main 'category' (WD page) results in discrepancy on Wikimedia Commons category page with the wikidata link on the left column leading to category page instead of main topic (while main topic WD page is otherwise stated on that page). So please fix or review permission Wikidata:Requests for permissions/Bot/Pi bot 6. Thanks --DDupard (talk) 07:59, 8 May 2019 (UTC)

That link goes to the the Wikidata item that the category is linked with. There's another Wikidata link in the infobox which goes to the main item, if you want that one. I'm not sure that it's really a problem. Ghouston (talk) 09:02, 8 May 2019 (UTC)
At least in terms of coherence it's an 'Interwiki conflict'.--DDupard (talk) 11:39, 8 May 2019 (UTC)
Link on the same Commons category can't be at the same time at two different items. In such conditions better use Commons category (P373) in items like Olavo de Carvalho (Q729048) (not categories). It's done in your example. I don't see the problem too. --Ksc~ruwiki (talk) 19:24, 8 May 2019 (UTC)
@DDupard: There is a standard settled approach here. If a category item like Category:Olavo de Carvalho (Q32398192) exists, then that is what the Commons category c:Category:Olavo_de_Carvalho should be site-linked to. In turn Category:Olavo de Carvalho (Q32398192) should be related to Olavo de Carvalho (Q729048) via the statement Category:Olavo de Carvalho (Q32398192) category's main topic (P301) Olavo de Carvalho (Q729048).
It may not be beautiful, but it is agreed, and it works.
One the other hand, if a category item like Category:Olavo de Carvalho (Q32398192) does not exist, then there is no need to create it, and in that case c:Category:Olavo_de_Carvalho may be site-linked directly to the main item Olavo de Carvalho (Q729048).
As User:Ksc~ruwiki says, in each case the property Commons category (P373) can be used to give the link back to Commons. Jheald (talk) 00:34, 9 May 2019 (UTC)
Well, what I am saying is that on Commons page, there are TWO different links to TWO different WD entries.--DDupard (talk) 11:54, 9 May 2019 (UTC)
The sidebar is giving you the sitelink for the category page, the infobox is giving you the sitelink for the subject of the infobox. It's not a bug, it's a feature. Jheald (talk) 12:15, 9 May 2019 (UTC)

Eight duplicates

There is one paper, but eight items for it:

How can this happen? 08:59, 9 May 2019 (UTC)

  • You could ask Pubmed about it. --- Jura 09:57, 9 May 2019 (UTC)
  • 2 of your 8 are the same item. But if you follow the DOI links on each of them (which are all different) I believe you will find that despite the identical titles and authors, publication dates, etc. they are actually different - at least I spot-checked a few and they were distinct corrections. ArthurPSmith (talk) 17:46, 9 May 2019 (UTC)

Edition War Q63256315

There is a edition war currently with the refered object. At the Location of the object when I write the name of the city where is the chapel with the painting, the editor Nono364 imediately revert the edition. Would you please tell him to stop this kind of action. Thank you, GualdimG (talk) 21:18, 3 May 2019 (UTC)

Made the same action with Q63431692. GualdimG (talk) 21:27, 3 May 2019 (UTC)
@GualdimG: [6] is a correct edit by user:Nono314, adding a less specific location (by you) isn't. I already left a message about this on User talk:GualdimG more than a month ago.
Please add the most specific location. It's quite disrespectful by the way to put in an incorrect username. Multichill (talk) 23:56, 3 May 2019 (UTC)
@Multichill: The question is not insert a more specific location. The question, and the action which should be censored, is to delete a more abrangent location. What is the problem, or the dificulty, if, for instance, we insert about an art item that beside being in a Chapel it is at the Town, or the city, where it is? The name of the Town, or the city, permits the further readers, or users, to have imediately a more knowledgable idea of the location of the item. Furthermore, is only acceptable one Location to any item? Is problematic to have two locations for the same item? Why is it possible to insert more then one place in the statement Location? Is it forbiden to insert a second place in Location?
About the the wrong designation of the user Nono314, that was not intended, and result only by a mine lack of memory. I am sorry about that. GualdimG (talk) 03:31, 4 May 2019 (UTC)
I really appreciate the work you're doing with paintings, but having your own system different from the standard convention isn't going to work. Wikidata, unlike Wikipedia, isn't primarily aimed at direct readers, but to be a machine readable repository of structured data. Wikidata will always have a bit of redundancy, but we try to avoid it. Based on the location (P276) it's easy to derive the located in the administrative territorial entity (P131) where it's located. It took me minimal effort to create Wikidata:WikiProject sum of all paintings/Top administrative territorial entities
At Wikidata:WikiProject sum of all paintings/Imprecise location I created an overview of paintings with imprecise locations. It's a bit over a thousand paintings that need to be fixed. Might sound like a lot, but we have over 300.000 paintings with location so it's less than 0,5% of these.
If I understand correctly, your primary concern is lists and you want to see the location in them. I took the liberty to update one of your lists. Does this fix your problem? Multichill (talk) 11:59, 4 May 2019 (UTC)
Edition war? I reminded you the rules, and you didn't bother to come on my talk page discuss or drop any comment. So, you may be waging it, @GualdimG:. Now it appears you were told the same at least twice by Multichill and you didn't listen either. This is a community project with rules established by consensus, not your private sandbox. You can't just go by your own rule, show contempt for the established ones, and now even attack those that follow them and waste their time fixing your damaging edits.
Indeed you can have several values to record the location history for an object as in Madonna and child with saints (Q3213754). It has however been established long ago that duplicating information is a bad thing, and moreover the description for location (P276) explicitly says it should not be used to designate an administrative entity. If a painting hangs in a church, it is always safe to put that in location (P276), while using it as collection (P195) is a bet on artwork management of which you have absolutely no clue. If you want to generate a list displaying the city, why not take 30 seconds to improve your query so that it displays it, instead of disorganizing everyone's data in order to fit your own narrow goal?
I see you also insist on putting Visitation (Q691810) in depicts (P180) instead of main subject (P921) where most people would use it. While this is not really wrong it also fails to integrate with common uses and will cause your items not to show when people look for paintings on such themes. The items you create are very welcome, but will only be useful to the community if you fill them according to common rules instead of perverting existing items so that they fit with your self-established lists criteria --Nono314 (talk) 12:30, 4 May 2019 (UTC)
To much talk....You said: "Indeed you can have several values to record the location history for an object ..", so, why are you deleting the name of the city/town in the Location of several art items I have created, or changing it for the name of the chapel/ church/museum in which the painting is? Please, @Multichill: @Nono314:, I ask you to revert the several deletions/changing of locations of cities/towns. There is much more interesting tasks to do than changing which is well done. I am sorry, I also call @Lea Lacroix (WMDE): to clarify this issue. Thank you, GualdimG (talk) 17:04, 4 May 2019 (UTC)
Yes, too much talk, this has been explained in length above. You started to alter each and every item you encountered to match your fantasies and now we are left with cleaning this up. But at least the outcome of this discussion is that there is now an official project to-do list to revert your misdeeds. --Nono314 (talk) 17:18, 4 May 2019 (UTC)

@GualdimG: Hi! I hope you are doing fine. Can you just explain the rationale for why one'd need a broader location on an item with a specific location on it? My sense is that if we add redundant information this could create problems with querying. A solution --if there is a case for bringing up a broader and a specific location onto an item-- might be to include a qualifier. More importantly to us all here: thanks for the hard work and for caring for this project, I am sure we'll find a simple solution to the matter at hand. Greetings! --Joalpe (talk) 21:30, 4 May 2019 (UTC)

Hi @Joalpe:! and thank you for your concern about this issue. I made some entrys in Wikipedia using data from Wikidata about artworks. In general, for a sinthesis of one artwork, beside obviously his Author and date of creation, is important to a reader know the Museum where it is and the city where the museum is. For the most famous museums, like Louvre, Prado, Ufizzi, is almost dispiciend to add the localization, and also to others museums which have the name of the city in his name. But for many others less known museums, or Chapels, or Churches, etc, if we put the name of the collection at the Location, you have the same name two times at an wikipedian entry, and so we are shrinking the informatiom that a reader could imediately obtain. This is the main concern that leads me to have the name of the city/town in the Location, which the wikidata system permits and we can do. I put, also, the question at the contrary. What do we lost, or which is the dificulty to a query, the problems that arise from having 2 or more items in the Location statement, like in the statement Depicts, or even in Instance? And if there are any kind of problems/dificulties why not to limit the positions of Location to 1 (one) only possibility? Thank you. Greetings, GualdimG (talk) 08:08, 5 May 2019 (UTC)
@Joalpe, GualdimG: this is just a data retrieval problem. I updated pt:A Visitação na pintura to retrieve the higher up location. I also did this with the test, but I think you missed that because I never got a response about that. Let's try again: Does this fix your problem?
Besides that I switched your list to main subject (P921). To explain the difference:
Could you please also use main subject (P921)? I have some work in progress documentation at Wikidata:WikiProject sum of all paintings/Christian religious art and you can see the most used ones at Wikidata:WikiProject sum of all paintings/Top main subjects. Multichill (talk) 10:31, 5 May 2019 (UTC)
@Joalpe, Multichill: To the question "Does this fix your problem?", obviously it doesn´t. At the location are responses like (in portuguese): Ucrânia, Grande Londres, 1.º arrondissement de Paris, Distrito de Viseu (for Viseu), Alta Baviera (for Munich), Distrito de Évora (for Évora), Distrito de Lisboa (for Lisbon), Innere Stadt (for Viena), Ilha de França (for Paris), Província de Sevilha (for Sevilha), and so on. And I not checked every one. As I said previously, you are changing (having troble/work), with something that is well done. Only because someone don´t follow your personal strict roule that is not mandatory by the Wikidata system.
Also with Depicts. Definitions, in english, "depicted entity (see also P921: main subject)", and in french, "personne, lieu, objet ou événement représenté par l’œuvre" (same in spanish, portuguese..). But by your rule we can´nt insert the event/événement at Depicts. This is the matter. And like I had ask earlier, you should "revert the several deletions/changing of locations of cities/towns" at the several artworks items I have created. And the same with the article at Wikipédia. Is worst then previously. Thank you! GualdimG (talk) 11:38, 5 May 2019 (UTC)
@GualdimG: Most of the unexpected results can be attributed to items still using the wrong location or the query service not up to speed yet or a fun layered administrative system in the country. I updated the query to look for cities. All the cases you mentioned are fixed.
You should really read my previous message more careful, nowhere in there I say "But by your rule we can´nt insert the event/événement at Depicts", on the contrary, I mention "Often the main subject gets included too." Multichill (talk) 12:04, 5 May 2019 (UTC)

So far, you have not demonstrated that insert the name of the city/town in the Location statement is wrong, or that insert the name of the event in the Depicts statement is wrong. Further more, if any of those was wrong the Wikidata would generate a "!" warning which never happened. You managed, in the pt:A Visitação na pintura entry to present the name of the city in the Location column. Good. But this was another way of achieving the same goal, goal that was already achieved. You were wasting time, re-doing something that was already done. And there are so many paintings, even in Wikimedia Commons, that are waiting for someone create the correct Wikidata items for them. So, returning to beginning, please, "revert the several deletions/changing of locations of cities/towns made by you at the several artworks items I have created". Thank you. GualdimG (talk) 14:39, 5 May 2019 (UTC)

@GualdimG: Up until now I've tried to be helpful and friendly. I ignored your strong words like "edit war", "censorship" and other accusations. But my patience with you is running low, especially when you start doing edits like [7].
location (P276) is for the most specific location, that's also in the description "location of the item, physical object or event is within. In case of an administrative entity use P131. In case of a distinct terrain feature use P706." both me and Nono314 told you that several times, but you seem to be unwilling to listen. Multichill (talk) 20:18, 5 May 2019 (UTC)
@GualdimG, Multichill: Aren't there two issues being conflated here? The first issue is the query on the specific list GualdimG has generated on Wikipedia in Portuguese. This has been solved, thanks to Multichill. The second issue is around a confusion between location (P276) and located in the administrative territorial entity (P131) for artwork. Perhaps a note could be added to Wikidata:WikiProject_Visual_arts/Item_structure, so that this confusion is less likely to arise. Would this make sense to you, GualdimG? PS: On a side note, GualdimG, thanks for the hard work on paintings, but we should talk --perhaps with User:DarwIn-- on how to something like Quick Statements could help you out. It'd also be nice if references were added to statements. --Joalpe (talk) 20:02, 6 May 2019 (UTC)
GualdimG, I think you may have a fundamental misconception about the purpose of Wikidata. This is not Wikipedia. The primary purpose here is not to be useful or convenient to human readers. The primary purpose here is to be a uniform, consistent database that is convenient for automated querying and consistency checking by machines.
You keep asserting that it's not a problem to list two locations (one specific, one more general) for a painting. You keep asserting that this might be more useful for a human reader. But, I'm sorry, you're wrong on both counts. It is a problem to list unnecessary, redundant information in a database like this, and that problem is not offset by the allegedly greater potential utility to a human reader, because utility to a human reader is not what we're striving for here.
If the database lists specific locations for paintings, and if you want to display more general locations to human readers, the right way to do that is to write a query that automatically extracts the more-general location by referring to property P131. This way, the more-general location is visible to users who are looking up any painting, not just the handful for which you've (manually, tediously, unnecessarily) added it to the painting's explicit properties. (And if there's something wrong with the more-general label that P131 gets you to, the right thing to do is to fix that label.)
Yes, this may seem like more work at first, but it's the kind of work we simply have to do if the goal is to create a uniform, fine-grained, general-purpose, structured database. --Scs (talk) 20:55, 5 May 2019 (UTC)

@Scs: I inserted the P131 at the item Q63256315 and appeared the warning "item requires statement constraint An entity with located in the administrative territorial entity should also have a statement country." Now we should identify the country for the artwork. Which one? Would´nt be more simple admit (which we can do, as I have proved for so long) the name of the city/town/village etc. at the Location statement? If we can use several lines for Depicts, or for the Instance statement, etc. (and this is no obstacle to a "uniform, fine-grained, general-purpose, structured database") why we can not use several lines to the Location statement, one of them could be the city/town/village? GualdimG (talk) 00:00, 6 May 2019 (UTC)

@Scs: If, in fact, P276 is intended to one sole location (not an administrative one) for the item at any given moment, I suggest a more precise re-definition of P276, like this: "Only the more precise location of the item, physical object or event is within, at any period of time. To indicate an administrative location use P131. To indicate a distinct terrain feature use P706." GualdimG (talk) 09:53, 6 May 2019 (UTC)

@Joalpe, Scs: I supposed that the question were already solved, but now @Multichill: reverted [8] the edition I had made inserting a Administrativ location on one painting justifying the reversion because "Redundant with location, shouldn't be used on movable objects". So, now, is redundant having in the item the chapel/church/museum were the painting is and the village/town/city where the Chapel/church/museum is? But at the same time is not redundant having in the Colletion and at the Location the same value? (Thank you Joalpe for your words above). GualdimG (talk) 20:34, 6 May 2019 (UTC)

@Joalpe, Scs, Multichill: One interesting case I found today: [9]. I really don´t know what should do about Location. I am waiting for guidelines. GualdimG (talk) 18:56, 7 May 2019 (UTC)

Why would it be something other than Pinacoteca Nazionale di Bologna (Q1103550)? - Jmabel (talk) 22:40, 7 May 2019 (UTC)
For guidelines: Add location (P276) to paintings, never add located in the administrative territorial entity (P131) to paintings. Make sure the target of location (P276) has located in the administrative territorial entity (P131) and country (P17). Multichill (talk) 15:45, 8 May 2019 (UTC)

Wikidata:Administrators' noticeboard#user:GualdimG. Multichill (talk) 16:13, 10 May 2019 (UTC)

Automatically create lexemes based on item aliases

What about automatically creating lexemes from item aliases and linking the lexeme to the item?--Malore (talk) 19:14, 6 May 2019 (UTC)

Hi Malore, the idea sounds interesting but I guess (maybe wrongly) it would create rather poor lexemes with few data (the lemma, the language and the sense but I see not much more to gather from item). More importantly, I'm not sure how to get the lexical category which is needed for creating a lexeme. So I like the idea but maybe in the other way around, a tool that if an already existing lexeme has no sense, then look up if an item has the same label. Cheers, VIGNERON (talk) 11:51, 7 May 2019 (UTC)
  • Maybe start with labels first? Some aliases are suitable as lexemes, others not necessarily. --- Jura 10:03, 9 May 2019 (UTC)
You're right. Maybe it would be better to create a tool and make the users check if from a particular alias can be created a lexeme, if that lexeme already exists, and which lexical category it belongs. It would be a lot slower than do it automatically but without doubt quicker than the current process. --Malore (talk) 15:18, 10 May 2019 (UTC)
In English I think you could pretty reliably do it by looking for labels that are single lower-case words not ending in 'ing', 'ly', or another indicator of being a different part of speech - these are almost always nouns. I wonder if a simple WDQS query can generate a list - seems like something that would time out though ArthurPSmith (talk) 17:32, 10 May 2019 (UTC)
This query gives a list of such words in English labels that don't currently have lemmas - I enforced a "subclass" relation to ensure these were not proper nouns of some sort that just happened to be lower-case, so that might work in other languages that don't lower-case labels too. Could be useful! ArthurPSmith (talk) 17:48, 10 May 2019 (UTC)


This is really stupid, but can someone add نوجين مصطفى as the Arabic name for Nujeen Mustafa (Q58918236). I'm not really seeing how I pull up the option to add it. GMGtalk 23:06, 7 May 2019 (UTC)

I did it by adding ar-0 to the Babel codes on my userpage, but it seems like there should be a better way to do it. Vahurzpu (talk) 00:42, 8 May 2019 (UTC)
Yeah...I don't speak the language at all, but I reached out to a steward who does, which is now I know what it is. Also thanks. GMGtalk 01:17, 8 May 2019 (UTC)
In Preferences, you can enable the LabelLister gadget which will let you add a label in any language when you need to. - PKM (talk) 01:30, 8 May 2019 (UTC)
A more temporary solution would be to view the page in Arabic (, then you will also be able to edit the Arabic label/description/aliases. (Of course, you’ll have to find your way around an Arabic user interface…) --Lucas Werkmeister (talk) 16:39, 10 May 2019 (UTC)

Interpreting the Grafana dashboard for Query Service

Hoi, I have followed for the past few days what the dashboard shows. What I noticed is that two servers, the wdqs1004 and wdqs1005 are the ones that are the ones with the biggest problems. I have seen such similar issues on the mini/mainframes I am familiar with. The solution, provided by the OS was a process where the data of all drives was actively moved around. Stagnant data was moved to newer drives and dynamic data to the drives that had most of the stagnant data. As a consequence issues like we see at this time were mitigated by spreading the load. Is this the case and, is a solution like this feasible at this time? Thanks, GerardM (talk) 13:24, 8 May 2019 (UTC)

@GerardM: As I understand it, query load is only actively load-balanced between the wdqs100* servers; the 200* servers are at a backup location; they receive all updates but none of the queries. Why there seem to be persistent differences between the different 100* servers I don't know; maybe the load-balancer for queries needs to be tweaked somehow? ArthurPSmith (talk) 17:49, 8 May 2019 (UTC)
Load balancing and keeping performance up is a dark art.. When new servers are added and all the new data goes there, it makes for the behaviour we see. It may also be that the servers are not equal and are expected to behave as equal. Thanks, GerardM (talk) 18:21, 8 May 2019 (UTC)
@ArthurPSmith: according to wikitech:Wikidata query service#Hardware, traffic is sent to both… but… the primary cluster [i. e. 100*] sees most of the traffic. --Lucas Werkmeister (WMDE) (talk) 10:44, 10 May 2019 (UTC)
Ah, I stand corrected. I think I had actually read that, but my brain switched "most" to "all". I wonder if there's a better way to load-balance that would help avoid these long lag times we sometimes get? ArthurPSmith (talk) 15:01, 10 May 2019 (UTC)
For what we look at, the problem is in the diverse behaviour of the different servers. Why is there such an uneven responsetime? Thanks, GerardM (talk) 15:15, 10 May 2019 (UTC)

Why is MediaWiki:Special-newitem deleted?

It looks like MediaWiki:Special-newitem is deleted, yet this is what I understand to be used for the sidebar. What's going on? Pdehaye (talk) 23:27, 8 May 2019 (UTC)

  • Special:NewItem not Special-newitem - Jmabel (talk) 02:44, 9 May 2019 (UTC)
@Pdehaye: Interface messages (pages in the MediaWiki: namespace) fall back to the copy of the message in the software source code (defined here) – that’s also what you see when you open the page (below the deletion notice). Unless you want to change the message from the default value, there’s no need to have it as an actual page on the wiki, and when such an override was previously necessary but then became unnecessary, it’s customary to delete the page again, so that future changes to the message in the source code (in the case of translations, coming from take effect immediately instead of being overridden by the now outdated wiki page. --Lucas Werkmeister (WMDE) (talk) 10:32, 10 May 2019 (UTC)
@Lucas Werkmeister (WMDE): Thanks!

Le Diable à Paris

Hi, can someone take a look at Le Diable à Paris Q3222291 and Le Diable à Paris (Q42367359)? Merging perhaps? Thank you for your time. Lotje (talk) 14:01, 9 May 2019 (UTC)

Hi Lotje,
no, for cataloguing books, there is a "work" item Le Diable à Paris (Q3222291), and a specific edition item Le Diable à Paris (1868 ed.) (Q42367359). If there were other editions to be added to wikidata, each would be linked to the work item through edition or translation of (P629)/has edition or translation (P747). In fact, there are already 3 different edition items linked on Le Diable à Paris (Q3222291).
this is the standard procedure to catalog Books. See Wikidata:WikiProject Books to learn more about this :) --Hsarrazin (talk) 17:28, 9 May 2019 (UTC)
Thank you Hsarrazin Lotje (talk) 04:33, 10 May 2019 (UTC)
@Hsarrazin: only me again. I followed your advise and had a look at the Wikidata:WikiProject Books, going through the names of the Participants I came accross arashtitan. He is not active since 2016, does this mean the list should be updated? Thank you for your time.   Lotje (talk) 04:57, 10 May 2019 (UTC)
the list is for notification purpose... even if not active, this user could wish to be notified... I wouldn't remove the name without contacting them :) --Hsarrazin (talk) 08:35, 10 May 2019 (UTC)

Proposal to create Showcase Queries

I've written a proposal to create Wikidata:Showcase Queries.
A community recognition of a high quality query which brings together high quality content across Wikidata.

We have showcase items and example queries, but not a process to identify, praise, and share Wikidata's true superpower: the combination of statements from many different items.

I've received positive feedback on my draft of this concept on the Wikidata telegram and facebook groups, so now I've moved the draft out from my userspace and am listing it here. Included in the proposal page is suggested criteria for how such a recognition would be judged. In summary, they are "comprehensive, of high quality, clean, and useful". I'd appreciate your thoughts. On the talkpage of Wikidata:Showcase Queries there are also some existing discussions on the nuances of some of the critieria too. Sincerely, Wittylama (talk) 10:44, 10 May 2019 (UTC)

Liste der größten Verkehrsflughäfen

Auf der Seite "Liste der größten Verkehrsflughäfen" fehlt Flughafen Dammam (IATA Code DMM) welcher offensichtlich wesentlich größer ist als Denver. Dies steht sogar auf der Seite vom Denver Airport.

Auszug aus Denver international Airport: Der Denver International Airport (IATA: DEN, ICAO KDEN, kurz DIA) ist ein internationaler Verkehrsflughafen knapp 40 Kilometer nordöstlich der Innenstadt von Denver, Colorado, in den USA. Er ist, gemessen an der Grundfläche, nach dem Flughafen Dammam der zweitgrößte Flughafen weltweit  – The preceding unsigned comment was added by (talk • contribs) at 12:38, 10 May 2019‎ (UTC).

Du bist aus irgendeinem Grund im Wikidata-Projekt gelandet, obwohl Deine Anfrage ausschließlich die deutschsprachige Wikipedia betrifft. Bitte frage unter de:Wikipedia:Fragen zur Wikipedia oder de:Diskussion:Liste der größten Verkehrsflughäfen. --MisterSynergy (talk) 12:41, 10 May 2019 (UTC)

Constraint violation

Why am I seeing a "Commons link should exist" constraint violation on Q982067#P1472, when the Commons creator link does exist ? I've tried deleting the statement and re-creating it, but the warning won't go away. Jheald (talk) 09:07, 5 May 2019 (UTC)

I have now seen the same on Q42247#P1472 and Q7931687#P1472 (though not some others). Some systematic problem with the Commons Creator page (P1472) constraint check? Jheald (talk) 18:00, 5 May 2019 (UTC)
@Jheald: I’ve seen this on newly-created Creator templates and assumed it was just sone sort of database lag. Are you seeing it on templates that have been around for some time? - PKM (talk) 20:23, 10 May 2019 (UTC)
@PKM: I suspect something's gone wrong with the check, and it may be that now whenever such a statement is updated or refreshed, it's getting a violation flag regardless of the reality of the situation. My first suspicion would be the constraint checker is no longer properly taking account of the templates being in their own Creator namespace, but that is just a guess. Jheald (talk) 20:39, 10 May 2019 (UTC)

Penalty shoot-out

Hello. 2015 UEFA Super Cup (Q4630327). points/goal scored by (P1363). Is there a way to show the penalties? (penalty shoot-out (Q2691960)) Xaris333 (talk) 02:09, 5 May 2019 (UTC)

applies to part (P518) + penalty shoot-out (Q2691960). Circeus (talk) 02:25, 5 May 2019 (UTC)

@Circeus:. For points/goal scored by (P1363) you have to insert match time of event (P1390) as a qualifier. But for penalty shoot-out (Q2691960) the time is not counting like additional time or extra time. Xaris333 (talk) 14:55, 5 May 2019 (UTC)

@Xaris333, Circeus: I boldly added allowed qualifiers constraint (Q21510851) with that qualifier property on points/goal scored by (P1363) page. --Liuxinyu970226 (talk) 05:01, 7 May 2019 (UTC)

I have removed required qualifier constraint (Q21510856) because is wrong for penalty shoot-out (Q2691960). Xaris333 (talk) 14:50, 11 May 2019 (UTC)

How to sitelink en:William Barnard Clarke

How to sitelink en:William Barnard Clarke, which is not a disambiguation page, but describes five (or possibly six) different people of this name?

The lead paragraphs are about William Barnard Clarke (Q18159784); but it also has considerable content about William Barnard Clarke (Q63638656) in particular. Is it appropriate for it to be sitelinked to the first; in which case should there be some sort of other reference from the second Qid ? Or is there another arrangement that would be preferable? Jheald (talk) 10:14, 9 May 2019 (UTC)

One way to do it is to make the article instance of = Wikipedia article covering multiple topics and then use main subject to point to each of the individuals that are covered in the article. Moebeus (talk) 15:24, 9 May 2019 (UTC)
It strikes me as an example of what some people call the "Bonnie and Clyde problem". —Scs (talk) 10:39, 10 May 2019 (UTC)
The Wikipedia article is poorly crafted. Articles shouldn't be a mishmash of people with the same name. Clarke the physician and naturalist (and his family members with the same name) should be split into a new article, distinct from the architect and archaeologist. -Animalparty (talk) 22:57, 10 May 2019 (UTC)
I went ahead and split off en:William Barnard Clarke (physician), which is now linked to the corresponding Wikidata item. And you may already know this, but for other, more legitimate joint biographies, Wikipedia redirects could be made, then temporarily broken, to allow linking (e.g. en:George W. Peckham or en:Clyde Barrow). -Animalparty (talk) 00:16, 11 May 2019 (UTC)

no property for JSTOR book ID

We have JSTOR article ID (P888) which is intended for articles on JSTOR, but JSTOR also has thousands of digital books, see for instance [10]. Entering the unique id (which I'm calling the "JSTOR book ID", but might also apply to other publication types) in JSTOR article ID (P888) produces a working link, but raises a format constraint issue for not matching the regex. See j.ctt2tv93j at Q60791206 for an example. I don't know if it's better to create a new property or merely expand the scope and format constraints of JSTOR article ID (P888). I'll leave proposing or modifying properties in the code-savvy hands of data wizards here. -Animalparty (talk) 20:58, 10 May 2019 (UTC)

We also have JSTOR topic ID (P3827) and JSTOR journal ID (P1230). I think we need a separate property “JSTOR book ID”. - PKM (talk) 12:18, 11 May 2019 (UTC)

Question on reliability of sourcing

Where can I find info on the reliability of sourcing with regards to using it to cite here? On Wikipedia:Notability of reliable sources you understand what is a good source. Do the same rules apply here?  – The preceding unsigned comment was added by 2600:6c56:6f08:1cf:0:464:3322:362b (talk • contribs) at 21:40, 10 May 2019‎ UTC (UTC).

  • In general, yes; obviously, no particular preference here for English-language sources, and when it comes to things like what IDs a particular external database uses, the primary source (that database) is preferred to a secondary source. I'm sure there are other small differences, but should be nothing major. - Jmabel (talk) 22:48, 10 May 2019 (UTC)
  • Wikidata does work a bit differently then Wikipedia. When we for example see a wrong claim that has a source (which is wrong) we often deprecate a claim instead of removing it like you would in Wikipedia. We also don't have policy to remove many sources that would be removed on Wikidata Wikipedia. ChristianKl❫ 10:02, 11 May 2019 (UTC)
    • @ChristianKl: I have no idea what you mean by "We also don't have policy to remove many sources that would be removed on Wikidata." - Jmabel (talk) 15:10, 11 May 2019 (UTC)
      • I meant to say "would be removed on Wikipedia". ChristianKl❫ 16:28, 11 May 2019 (UTC)

John Cotton's Notebook

How can we improve John Cotton's Notebook (Q63687248)? It represents an artist's sketchbook, with a mixture of sketches and hand-written text - often more than one sketch per page. Do we need an item for each page, for the "depicts" statements, or shall we add the all to the one item, using page numbers as qualifiers? Is there a similar item that shows a good data model? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:34, 11 May 2019 (UTC)

Memorandum Book Showing Colored Sketches Mostly of European Helmets (Q63213640) is a similar item. I use “collection/inventory number” rather than “owned by/catalog code” for museum items. - PKM (talk) 12:10, 11 May 2019 (UTC)
Though I do use “owned by” for provenance ... - 12:47, 11 May 2019 (UTC)
@Pigsonthewing: see Très Riches Heures du duc de Berry (Q211062) for individual illuminations with their own items modeled as “part of”. - PKM (talk) 19:50, 11 May 2019 (UTC)

Wikidata list and array

Hello. User:Xaris333/SuperCup. Why I get "Array" and "QArray"? Xaris333 (talk) 20:36, 11 May 2019 (UTC)

location of discovery (P189)

Hello, how to allow that this property accept also a value that is a subclass of geographic entity (Q27096213)? example of logic potential use in Rausch 572 (Q19359611), a specific referenced biologic specimen can have been discovered in a place that is recorded here as a subclass of a geographic entity. I tried but this don't work properly. Christian Ferrer (talk) 22:13, 11 May 2019 (UTC)

Christian Ferrer, it wasn't necessary adding an exception, instead, I've added geographic entity (Q27096213) to the constraint. --Tinker Bell 00:46, 12 May 2019 (UTC)
Great, thanks you. Christian Ferrer (talk) 05:00, 12 May 2019 (UTC)

Slave Coast

Sorry for the mess I created on 'Slave Coast'. I only wanted to change the meta link at the Frisian article to Dutch Slave Coast instead of Slave Coast. Can someone check if I correctly cleaned up? Thanks. Muijz (talk) 08:36, 12 May 2019 (UTC)

Note: relevant articles Slave Coast (Q1360692), Dutch Slave Coast (Q16209924). I have merged Slave Coast (Q63751379) back into Slave Coast (Q1360692). Somebody else will need to check the others, I'm out of time. Jheald (talk) 09:35, 12 May 2019 (UTC)

Any way to identify top 500 computers?

Hi there. Is there currently any way to identify computers that have been part of TOP500 (Q633765)? For example Sunway TaihuLight (Q24702533) is marked as a supercomputer (Q121117), while Titan (Q44070) is one-of-a-kind computer (Q28542014). Should we have a property like "place in top 500" or is part of (P361) enough? Strainu (talk) 08:50, 12 May 2019 (UTC)

How about described by source (P1343) = TOP500 (Q633765), with qualifiers publication date (P577) and ranking (P1352) ?
I would personally prefer to restrict part of (P361) to objects that are physically part of a bigger object, as far as possible. Jheald (talk) 09:25, 12 May 2019 (UTC)

Tool to swap citations to items needed

Several of the references on Alexander McMillan Shields (Q15627317), to the same source, have eight statements:

  • published in
  • volume
  • issue
  • page(s)
  • publication date
  • title
  • author name string
  • reference URL

I have now created an item for the source: Prominent Californian Ornithologists. III. A. M. Shields (Q63758656) and need to swap over each of the citations. Is there a tool that can do this simply and quickly? (QuickStatements seems a sledgehammer to crack a nut, in this case.) Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:02, 12 May 2019 (UTC)

@Pigsonthewing: There's a gadget you can turn on in user preferences, which lets you copy a reference (including all its bits and pieces) from one statement, and then add it to another. I only activated it recently, and have been finding it very useful. Jheald (talk) 15:02, 12 May 2019 (UTC)
Thanks. I use that, but am looking for something that will do all the changes in one action, incuding deleting the obsolete statements. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:44, 12 May 2019 (UTC)

Performance query

Hoi, today the performance on the query service is particularly bad. A record hour and a half. None of my jobs are running, that is my contribution to not making it worse. However, the question remains it is obvious that we do not cope, what are the plans? Thanks, GerardM (talk) 12:31, 12 May 2019 (UTC)

Can't create item

I can't create an item for voy:de:Frankfurt am Main/Museen, it always claims the page does not exist. The item needs to be instance of -> Wikimedia list page and country -> Germany. -- 16:32, 12 May 2019 (UTC)

I created it for you, Q63762371, please fill it in.--Ymblanter (talk) 17:54, 12 May 2019 (UTC)

Fusion 'applies to other entity' and 'refers to different subject'

Moved from WD:BN. --Matěj Suchánek (talk) 16:00, 2 May 2019 (UTC)

Those two qualificators have the same meaning?!, they are on the list of Wikidata reasons for deprecation).

--MHM55 (talk) 07:54, 1 May 2019 (UTC)

  • @Jura1:, as creator of refers to different subject (Q28091153), would you agree that these are the same? - PKM (talk) 19:47, 3 May 2019 (UTC)
    • Probably. If so, "subject" might be more precise than "entity" as it refers to a specific part of a statement. OTH, it couldn't really be "object". --- Jura 10:57, 5 May 2019 (UTC)
      •   Support merge. - PKM (talk) 19:30, 12 May 2019 (UTC)

cleaning our statements sourcing...


Now, a lot of items, especially authors have on same statements an old source, like imported from Wikimedia project (P143) svwp and regular sourcing with stated in (P248) + an authoritative source like an official library database...

in some cases, there are 10 P248 + & or 2 P143...

Why not have a bot cleaning those unreliable sources (P143), when there is a clean valid P248 ? What do you think ? It would remove a lot of junk info, without losing any real data. --Hsarrazin (talk) 18:13, 8 May 2019 (UTC)

+1, imported from Wikimedia project (P143) was only a temporary solution to populate WD. By cleaning these sources, we will show that WD is moving forward in the direction of quality and is working to avoid circular referencing with WP. Snipre (talk) 18:24, 8 May 2019 (UTC)
agreed. P143 is only a temporary solution. It would be great to seek out specific references in a Wiki(m|p)edia page that P143 had been based on. It'd greatly improve Wikidata's data quality. A while ago I tried this approach importing references of death dates from Wikipedia PL Deaths in 2019 equiv pages into WD (e.g [11]). Kpjas (talk) 19:03, 8 May 2019 (UTC)
Depends on the context. For example, when I add an image, I also add a ref using imported from Wikimedia project (P143) and Wikimedia Commons, and I do not see this solution as temporary.--Ymblanter (talk) 19:21, 8 May 2019 (UTC)
I am not convinced. What do we win if we remove such imported from Wikimedia project (P143) references? I find them often quite useful, even if "real" references with stated in (P248) and/or reference URL (P854) are present on the same claim. I'd say we let data users decide which references they want to use and which ones not—whoever ignores imported from Wikimedia project (P143) would likely also ignore inferred from (P3452) and based on heuristic (P887).
That said, I think we should overhaul Help:Sources and add separate sections for at least imported from Wikimedia project (P143)+Wikimedia import URL (P4656), inferred from (P3452), and based on heuristic (P887), including a description how they work and that some users possibly might want to ignore them. —MisterSynergy (talk) 19:23, 8 May 2019 (UTC)
I don't think that's a good idea. I saw many times statements based on id was incorrect. For example, there are two different dates (day, month and year) in the person's date of birth (P569). One of the dates has imported from Wikimedia project (P143) used as a source. We can always watch the article in Wikipedia and check correctness of the date. The another date had two source date of birth (P569) and id. But if Wikipedia's article really has such date (day, month and year), then id has only year. Removing imported from Wikimedia project (P143) as a source of second date would mean removing real valid source of statement. --Ksc~ruwiki (talk) 19:49, 8 May 2019 (UTC)
@Ksc~ruwiki, MisterSynergy: Please consider the mentioned case only: a statement where several references are present and some of them are imported from Wikimedia project (P143).
Then please remember that most Wikipedias have as rule that Wikipedia statement can't be used as reference for other Wikipedia statement. By not respecting that rule, we get a bad image and we have some difficulties to convince Wikipedias to use our data. We can always argue that we can filter the data according to the references, but for Wikipedians we are a mix of good and bad quality data, and they prefer to work with databases offering good quality data only (less work for them and no need of edit complex lua functions to filter what they want).
Then most importations from Wikipedias were performed without any consideration for referencing: no date, no link to the article version. As Wikipedia articles are changing due to correction and improvement, there is an increasing trend that WP articles don't mention any more the data we have in WD and so the reference link is lost. We can only say that once in the past one WP article (which one ? sitelink are changing too, so there is no guarantee that the Wikipedia article currently linked to the item contained once the data present in WD) mentioned the data.
So your examples are stucked with that situation and this definitively creates a bad impression of WD: even if I want to check the origin of a WD statement by using the sitelink, I can find no mention of the data because the WP article was changed in the midtime or worse because the sitelink was changed. How can you convince people that you are doing a good job in WD with that ?
imported from Wikimedia project (P143) is a static link, and WP are dynamic data compilation. That's why the property has to be deleted and replaced by normal references (static ones) or at least the link between the WD statement and WP has to be stronger by including the page version of the article in the reference. Snipre (talk) 07:30, 9 May 2019 (UTC)
I completely agree with the initial proposal of Hsarrazin per Snipre. --Epìdosis 09:03, 9 May 2019 (UTC)
Well, Wikipedias have to process references anyways, even if there was no P143 at all. There are a bunch of reference qualifers per reference, and the module/template coder has to arrange them to something which makes sense as a useful Wikipedia reference. It is pretty simple to either ignore references which contain P143, or to ignore a statement completely in case no other valid reference is there.
The lack of versioning for many (older) P143 references is a convenience problem, but as Jura points out, practically any editor with basic Wikimedia editing experience is able to read page histories. I do use this occasionally and find it very useful to deal with obscure statements which I find when doing various consistency checks; in fact, there are only few cases which I came across where the imported value has already been removed from the Wikipedia page. Anyways, for quite some time now we have the additional Wikimedia import URL (P4656) reference qualifier, which points to a specific version of the import source and makes this even less of a concern. —MisterSynergy (talk) 20:09, 9 May 2019 (UTC)
My English is far from perfect but I described the situation you have meant (I have realized your idea very well). At least I hope I could described example with possible problem of removing imported from Wikimedia project (P143) from sources. Of course I remember The Wikipedia is not quite reliable source, but articles in Wikipedia should be based on reliable sources. If it is so, noone will remove them from the article. So it's not a problem to check the information. (About the property Wikimedia import URL (P4656) previously they have wrote. And by the way ids too often contain wrong information in coparison with Wikipedia). Besides I really forgot about problem they have mentioned bellow. Some of statements use several qualifiers. For example, some populated place was located in several located in the administrative territorial entity (P131) during it's history. We have a sourse like OKTMO 179/2016 (Q26833494) (in Russia) but for modern administrative territorial entity, but it doesn't contains information about start time, we take it from Wikipedia and make another source imported from Wikimedia project (P143). So we get total comlex of sources for all components of statement. If remove one of the sourses we'll make original research. Of course I don't mind if any user remove imported from Wikimedia project (P143) from item, if he or she see reliable sources for all components of statement, but I disagree the idea to do it by bot. P.S. The text I have wrote is just the result of prsonal experience of removing vandalism both in Wikidata and Wikipedia, so I really think imported from Wikimedia project (P143) is usefull.--Ksc~ruwiki (talk) 20:03, 11 May 2019 (UTC)
  • It's tempting, but it's always better to have a way to find where the statement came from. --- Jura 09:58, 9 May 2019 (UTC)
@Jura1: Do you really want to dig into the history of a WP article to find where the data present in WD was once mentioned ? Because imported from Wikimedia project (P143) is not a guarantee that the data is really displayed in the WP article. And do you really want to check the history of a WD item to be sure that the link between the item and the WP mentioned as reference was not changed after several years ? Snipre (talk) 13:23, 9 May 2019 (UTC)
  • I think most users with some experience of Wikipedia and Wikidata can find it fairly easily. --- Jura 13:27, 9 May 2019 (UTC)
  •   Support for the specific case mentionned (which is already a lot of cases), it would clearly improve the situation. Cdlt, VIGNERON (talk) 12:05, 11 May 2019 (UTC)
  • If this should be cleaned up by bot, claims with qualifiers should be probably excluded. If the list of references should support a claim (with each reference possibly supporting a different part (e.g. start time (P580)/end time (P582))) it is not clear if the authoritative source supports the whole claim (with all qualifiers) or just part of it. - Valentina.Anitnelav (talk) 15:43, 11 May 2019 (UTC)
  • I think in general we should try to find primary or secondary sources. A library catalog, another encyclopedia or tertiary source doesn't really get us closer to that. The advantage of the "imported from" reference is that it might actually help us find one. BTW, we still need a bot operator to clean up the last library catalog import: Wikidata:Bot_requests#Cleanup_VIAF_dates. --- Jura 19:38, 12 May 2019 (UTC)

constraint for "instance of" in property "Twitch channel ID"

Property: Twitch channel ID (P5797)

Currently data objects need to be (instance of (P31)) human (Q5), video game developer (Q210167), video game publisher (Q1137109), software company (Q1058914) or clan (Q989470). However there's a lot of twitch channels out there that don't match either of those. Examples would be but are not limited to groups of people that stream on one channel (e.g. the yogscast), events (e.g the German gaming convention gamescom), as well as non-Gaming twitch channel (e.g. the official NASA twitch channel). I believe there are a lot more examples. We could either try to add all these possible values for the constraint or remove the constraint altogether. I am suggestion the latter. --Madmaurice (talk) 19:56, 9 May 2019 (UTC)

Any thoughts? Maybe from the creator Kissa21782? --Madmaurice (talk) 10:35, 11 May 2019 (UTC)
Hi. Do note, that I am certainly not the creator of anything – I've just edited the property in the past. But I do agree that the constraint is a bit unnecessary, and the best solution would be as you said to remove it altogether. YouTube channel ID (P2397) for example does not have the constraint at all. Instagram username (P2003) does, but I think that one should be deleted as well. --Kissa21782 (talk) 15:11, 11 May 2019 (UTC)
Oh, sorry, you're right. I misread the edit history. Pintoch created the property with the constraint. And subsequent edits added more possiblities for instance of. Maybe they have any thoughts about it. Apart from that I'm a bit new to the whole discussion process on Wikidata. I obviously shouldn't just go ahead and change constraints because I want to... so how does this usually work? --Madmaurice (talk) 23:37, 11 May 2019 (UTC)
I think that you should go ahead and delete the constraint – there's very good reason to do so. If someone were to disagree with good reasoning, they'll start a chat about it. Doing smart and pro-Wikidata edits are always welcome. --Kissa21782 (talk) 20:28, 12 May 2019 (UTC)
Someone removed it. Thanks eitherway! --Madmaurice (talk) 16:25, 13 May 2019 (UTC)

How to indicate precision in Date up to two month?

Very often we can find in the sources "... someone died in April-May 1940 ..." How can we write this in wikidata? Example: Q2615881 --Glovacki (talk) 13:02, 10 May 2019 (UTC)

date of death (P570) 1940, quailifiers: earliest date (P1319) April 1940, latest date (P1326) May 1940. --Dipsacus fullonum (talk) 13:54, 10 May 2019 (UTC)
Dipsacus fullonum, thank you! And what should I write if "... someone died between May 1940 and May 1941 ..."? I can write date of death (P570) unknown, quailifiers: earliest date (P1319) May 1940, latest date (P1326) May 1941, but it looks ugly, because date of death is not unknown, it is just known with some precision. --Glovacki (talk) 10:29, 13 May 2019 (UTC)
You could write "date of death (P570) 1940s" with qualifiers "earliest date (P1319) May 1940, latest date (P1326) May 1941". And if the decade isn't known, you can use century as the precision for date of death (P570) etc. --Dipsacus fullonum (talk) 10:44, 13 May 2019 (UTC)


At Thomas Alva Edison (Q8743), one of Edison's children is listed as generic William (Q12344159), is it proper to use the generic name until an entry is created for that child? --RAN (talk) 01:46, 12 May 2019 (UTC)

I'd say heck no. A name is not a person, and denoting it as such only confuses things. We have number of children (P1971) (which I added) to list total number, including non-notable members. -Animalparty (talk) 02:54, 12 May 2019 (UTC)
No, it violates the constraints and data-model. The proper way is to create a new item. If you want to avoid that, you could also use unknown value with stated as (P1932). ChristianKl❫ 09:18, 13 May 2019 (UTC)
@Mekala Harika: who added the value. ChristianKl❫ 09:20, 13 May 2019 (UTC)

WD not being automatically updated when moving pages on local project?

Anybody encountered this before? GMGtalk 14:36, 13 May 2019 (UTC)

Nevermind. After doing a few of them by hand it seems to be working again. GMGtalk 17:47, 13 May 2019 (UTC)

Wikidata weekly summary #364

Encoding errors at author string

Many statements of author name string (P2093) have wrong enconding (an - incomplete - list is at Wikidata:Database reports/Constraint violations/P2093). An example is "M Sánchez-Pérez" at Q59570196. The question is, if a bot could correct most of these encoding errors. 11:22, 13 May 2019 (UTC)

Is Q461012 dead or alive?

Hi all. I would like to request that a native French speaker reviews the french article about Dorothy Head Knode (Q461012) (fr:Dorothy Head) as it is mentioned that she died in October 2015 (categorized in fr:Catégorie:Décès en octobre 2015) while the English article (en:Dorothy Head Knode) does not mentioned that she died. The problem is that her death date gets imported here from the French Wikipedia repeatedly. So, we need to resolve this conflict between enwiki and frwiki. (Let's ping the last editor of the frwiki article @Jmax: and the last editor of the Wikidata item @Pichpich:) Thanks. --Meno25 (talk) 01:17, 14 May 2019 (UTC)

  • The French article has an English reference for the date. Anyways, until it's added to the statement, I set the statement to deprecated rank. --- Jura 04:10, 14 May 2019 (UTC)

Should P31 only list the most precise subclass?

A Danish hill had this statement with a reference:

I added this more precise statement with a reference:

We also have:

Now my question is if the first statement should be deleted so only the most precise case of instance of (P31) is left? --Dipsacus fullonum (talk) 06:31, 14 May 2019 (UTC)

  • Not necessarily. Here it's not entirely clear if Q12302769 covers just one aspect of Yllebjerg (Q23732972). Also, sometimes references disagree about the exact nature of an item, so various P31 are useful. --- Jura 06:41, 14 May 2019 (UTC)
I would say that in this case there is no aspect of Q23732972 which is covered by Q54050 but not also covered by Q12302769. --Dipsacus fullonum (talk) 07:15, 14 May 2019 (UTC)

[Breaking Change] wbeditentity including empty alias set will remove all aliases

Hello all,

This change is relevant for everyone using the wbeditentity endpoint of Wikidata’s API.

While working on editing the termbox from mobile, we discovered a bug in our code of the wbeditentity endpoint, that does not conform with the implicit interpretation of the documentation.

A request including {"aliases":{"en":[]}} should, according to the implicit interpretation of its documentation, replace all aliases in English by an empty string, meaning removing all aliases. However, at the moment this action is not actually performed, meaning that this request would leave the aliases untouched.

We want to fix this bug, because we need this request to work in order to be able to remove all aliases also in the new termbox on mobile. We are treating this bug fix as a breaking change because the documentation was ambiguous, and there may be some tools currently sending requests with empty alias arrays when nothing need to be touched, intentionally or not.

If you are maintaining a tool, please inspect your tool usage of wbeditentity endpoint, and make sure that no calls with empty alias arrays are sent unless the intention is to remove these aliases.

According to our breaking change policy, this bug fix will be first deployed on later on May 28th, then on on June 12th.

If you have any question or issue, feel free to discuss in the related ticket. Cheers, Lea Lacroix (WMDE) (talk) 10:10, 14 May 2019 (UTC)

Urgent admin action needed

It appears that no admins are watching Wikidata:Administrators' noticeboard.

Urgent action is needed at Michael Moates (Q63245258), to remove defamatory statements as described at Wikidata:Administrators' noticeboard#Sock-puppet IP's and Vandals, almost twelve hours ago. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 08:55, 13 May 2019 (UTC)

@Pigsonthewing:   Done the actions requested at 22:12, 12 May 2019 (UTC). Mahir256 (talk) 09:03, 13 May 2019 (UTC)
@Mahir256: I have one more request. Would you consider removing Men's Rights Movement and alt-right? Both are not based in fact. Alt-right has a citation of a tweet from someone making a claim that doesn't know the subject and is just a random individual. It is not a news source but a biased Twitter user posting on the internet. Datamaster1 (talk) 10:32, 13 May 2019 (UTC)
I removed those two claims. ChristianKl❫ 10:41, 13 May 2019 (UTC)


I want to understand something. A friend of mine 2600:6c56:6f08:1cf::/64 was blocked by an admin for "Edit warring." I wanna help him to understand what he did wrong. There was an individual who was vandalizing a page Q63245258 and he reverted the edits multiple times. Now I understand the policy. The problem I am struggling with is according to Wikidata:Edit warring "Reverting vandalism is not edit warring" and "Reverting clear-cut violations of Wikidata:Living people is not edit warring" so why was he blocked? At this point, admins have removed the vandals content and he was right because everything removed was against policy. Can the block be lifted and if not can someone explain what he did wrong? Datamaster1 (talk) 11:10, 13 May 2019 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Relevant links are:

I, too, would be interested to know why - given the above quote from Wikidata:Edit warring - an IP editor was blocked by User:Jasper Deng for removing egregious BLP violations. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:32, 13 May 2019 (UTC)


@Mahir256: @Pigsonthewing: @ChristianKl: @Jasper Deng: - The user is now vandalizing WikiCommons on the same individual RyanForTrump which was banned by Jasper is now uploading images that are claiming things that were removed by admins here for not following policy. The image is no based in fact and goes against policy because it is not notable and will never be used in an article. I have nominated the image for deletion you can see that here commons:Commons:Deletion requests/File:Michael-Moates-as-Hannah-Thompson.jpg. The user is clearly vandalizing all properties regarding Michael Moates across all Wikimedia projects. Datamaster1 (talk) 03:34, 15 May 2019 (UTC)

I am sorry I corrected the link Datamaster1 (talk) 04:24, 15 May 2019 (UTC)
Admins on wiki commons have blocked RyanForTrump indefenetly. 2600:6C56:6F08:1CF:0:464:3322:362B 07:08, 15 May 2019 (UTC)


Request immediate block of IP they are back on Michael Moates (Q63245258) vandalizing again. They are clearly vandalizing at this point. 2600:6C56:6F08:1CF:0:464:3322:362B 18:54, 19 May 2019 (UTC)

Request immediate block of 2600:6C56:6F08:1CF:0:464:3322:362B as a sockpuppet of Mmoates. See [12] 19:01, 19 May 2019 (UTC)
The indvidual above has been found to be harassigng me and others. He is a sock of RyanForTrump which has already been banned here and at Wikicommons... you can see by his editing history that he has not edits other than his vandalism to Michael Moates (Q63245258). 2600:6C56:6F08:1CF:0:464:3322:362B 19:07, 19 May 2019 (UTC)
This section was archived on a request by: already mentioned on relevant admin board. --- Jura 10:27, 21 May 2019 (UTC)

Ideas sought for "real and measurable progress in solving a critical problem of our time"

The McArthur Foundation is offering US$ 100 million "to fund a single proposal that promises real and measurable progress in solving a critical problem of our time." Has anyone looked into that from a Wikidata/ Wikimedia perspective? --Daniel Mietchen (talk) 09:39, 13 May 2019 (UTC)

Currently, it seems like most of the work on Wikidata isn't metrics driven and most of the development is more speculative in nature then evidence-based. I would expect that there are many grants for which we are a better match. ChristianKl❫ 11:56, 13 May 2019 (UTC)
you could propose a wikidata driven sum of all knowledge translation + internet in a box. maybe we should undelete their images (negotiate about the meaning of CC licenses) before asking them for money. Slowking4 (talk) 13:03, 13 May 2019 (UTC)
Wikidata mainly exists to help Google, Amazon, Wikimedia, et al. get better results faster, i.e. results based on existing data. Not knowing what position Joe Blow played on his high school baseball team or the year in which a painting was painted isn't a critical issue of our time. I say save the money for curing childhood cancer, solving climate change, and generating new data for data enthusiasts to curate. -Animalparty (talk) 21:40, 13 May 2019 (UTC)
@Animalparty: I think Wikidata potential is broader. Wikidata does have the potential to help with biomedical research by allowing data to be accessible in new ways but that potential is experimental in it's nature and doesn't have clear metrics. ChristianKl❫ 16:28, 14 May 2019 (UTC)
"the year in which a painting was painted isn't a critical issue of our time" speak for yourself. the use of wikidata to link open knowledge and get it translated to local languages for non-English speakers can scale the sum of all knowledge. the same techniques used to correct painting dates, can be used on PubMed dates. in the medical field it can save lives. Slowking4 (talk) 02:22, 16 May 2019 (UTC)

Update: deployment dates for wb_terms redesign

Hello all,

This is an update regarding the dates of test environment and migration of wb_terms table replacement solution.

Due to various complications that the developers in the Wikidata team have been working on solving over the last few weeks, we unfortunately will have to push the dates for when a test environment for tools builders will be ready, which was supposed to be ready today, and the following dates for starting migration of wb_terms data into the new schema in production.

The new dates are:

  • 29th of May: Test environment for tool builders will be ready
  • 12th of June: Property Terms migration starts
  • 19th of June: Read property terms from new schema on Wikidata
  • 26th of June: Item terms migration begins
  • 3rd of July: Read item terms from one of the two schemas (as explained in this task)

You can find more information regarding those dates and how to prepare for them in this task, and we have dedicated a board to receive and help with any questions from tool builders that need to update their tools accordingly.

In order to keep all discussions in one place, we kindly ask you to react or ask your questions on Phabricator.

As a reminder, if you want to discuss with the developers, ask questions and get help in order to update your tools, you can join the IRC Mediawiki meeting, today at 15:00 UTC on #wikimedia-tech.

Thanks, Lea Lacroix (WMDE) (talk) 12:51, 15 May 2019 (UTC)

New museum digitalisation grant (pilot)

Grants:Project/Rapid/Rodrigo Tetsuo Argenton/Pilot project at Geoscience Museum

Hi guys, my name is Rodrigo Tetsuo Argenton (talk), I'm start a new GLAM digitalisation pilot, as in this start I'll focus a lot in Wikidata, it will be nice you check this proposal. To summary, the idea that touch here is check with the museum what's the best databases around geoscience, if possible import it, and what are the patterns (statements) used internationally to properly describe items related to the museum, minerals, gems, meteorites...

Let me know if you have any questions, and check out the project, if possible endorse it. Rodrigo Tetsuo Argenton (talk) 20:43, 15 May 2019 (UTC)

Urgent Admin Help Needed

Can someone block IP they are a sock of RyanForTrump and are back on Michael Moates (Q63245258) vandalizing again. They are making libelous statements. They are engaged in edit wars. See

They have already been banned from Commons as a vandal see. Commons:Administrators' noticeboard#Cross Wiki Vandalism

Also request protection of Michael Moates (Q63245258)

Datamaster1 (talk) 19:28, 19 May 2019 (UTC)

This section was archived on a request by: already mentioned on relevant admin board. --- Jura 10:26, 21 May 2019 (UTC)

Adding Data About Video Game Statistics?

Hello! I have been a longtime lurker on the Wiki project in general and I saw the Wikidata project as a way to get my feet wet in contributing to this. I also apologize if the way I word this isn't that straightforward or if my formatting isn't that tidy! I am still trying to understand a bit.

My understanding is that as long as the information is freely available that we can add it to data points? If so, is it possible to add the concept of statistics of video game characters to their entries if it comes from something like a fandom wiki?

An example I can give is for example Q59457373 (Brand from the game League of Legends) who according to the game they are from, have statistics like attack, magic, etc. These stats are archived via the fandom wiki page on this character. Is it possible to do this? I can possibly see the limitations (Source of data or the attributes of the data in question even) but I wanted to make sure.

Cheers, --Senator mailman (talk) 17:33, 16 May 2019 (UTC)

Three items / one person

Q58958272, Q60476647 and Q63979034 are one and the same person: Piotr Popik. Francesco 13 (talk) 23:40, 21 May 2019 (UTC)

@Francesco 13:   Merged Lαδδo chat ;) 01:51, 22 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 12:13, 22 May 2019 (UTC)

Merge Q797424 and Q3372150

Hello, there are two data objects for the same Chinese deity Nüba/Ba/Hanba. Could someone please merge them? Thank you, --Herr Klugbeisser (talk) 19:44, 13 May 2019 (UTC)

@Herr Klugbeisser: I only find Nuba (Q3372150). What is the other item? LaddΩ chat ;) 00:51, 14 May 2019 (UTC)
@Laddo:: There are two objects: Nuba (Q3372150) and Nuba (Q797424). Both talk about the same deity. One contains the links to the English language article in Wikipedia, plus a few others, and the other contains the links to the German and Estonian articles.--Herr Klugbeisser (talk) 18:56, 19 May 2019 (UTC)
@Herr Klugbeisser:   Done, both merged in Nuba (Q797424). Lαδδo chat ;) 11:26, 20 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 08:50, 23 May 2019 (UTC)

P279 loop needs attention

Have a look at the loop in this graph Here is a snapshot.

There is loop in the subclass of (P279) property for the following five objects:

I guess at least one of the listed subclass of (P279) relations should be broken or reversed, but which one? Is it mental representation (Q2145290) that should be subclass of (P279) of concept (Q151885) rather than the opposite? Note that mental representation (Q2145290) is also stated as instance of (P31) concept (Q151885).

Here are some diffs related to this loop:

--Larske (talk) 01:23, 14 May 2019 (UTC)

@Larske: I've been tracking this for a couple of years now; there's only one other subclass loop at present (caused by a disease ontology mismatch issue, hopefully this will be resolved by the people who work on those - several similar earlier ones were resolved that way). In this case these concepts are at such an abstract/meta level it's never been clear to me either (A) if it really matters to fix this, or (B) what the correct solution would be. Some examination of their context (P31 and P279 properties referencing them, or other relations) might help to disentangle this one - feel free to go ahead and give it a try! ArthurPSmith (talk) 12:27, 16 May 2019 (UTC)
It's not easy, for example en:Concept says "In contemporary philosophy, there are at least three prevailing ways to understand what a concept is" and I imagine that each one would need to represented as its own item with its own subclasses. Ghouston (talk) 02:10, 17 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 09:01, 23 May 2019 (UTC)

Best way to show in Wikidata that someone appears in a list in another project without creating a new category

I am adding Wikidata entries for the first aviation deaths found in this image: File:Aviator_deaths_in_Je_Sais_Tout_on_15_August_1912.jpg. I can link from the Commons file to Wikidata and Wikipedia, but is their an easier way to display FROM Wikidata that a person appears in this document, other than creating a category for each person mentioned. --RAN (talk) 14:18, 16 May 2019 (UTC)

  • Eventually, it should be possible to find from Wikidata edits like [16]. --- Jura 14:26, 16 May 2019 (UTC)
  • In this particular case, I'm pretty sure depicted by (P1299) is the property you want. The problem, of course, is that something cannot be at the same time a list article and a multimedia file. This could conceivably be created as an published article item with the file as illustration, though. Circeus (talk) 18:53, 16 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 09:10, 23 May 2019 (UTC)


Jasper Deng
Egon Willighagen
Denise Slenter
Daniel Mietchen
Emily Temple-Wood
Pablo Busatto (Almondega)
Antony Williams (EPA)
Devon Fyson
Samuel Clark
Tris T7
Robert Giessmann
Cord Wiljes
Jonathan Bisson
Charles Tapley Hoyt
Peter Murray-Rust
  Notified participants of WikiProject Chemistry

Currently, cinquasia red (Q418071) is <instance of> chemical compound and pigment, and <subclass of> red and color. Should the color “cinquasia red” be its own item? Can a substance be both a chemical compound and a pigment, or would we prefer separate items? - PKM (talk) 01:44, 17 May 2019 (UTC)

One thing I'd say is that a pigment should not be a subclass of a colour. A pigment is a chemical compound, and a colour is a type of light. Ghouston (talk) 01:59, 17 May 2019 (UTC)
I'm with Ghouston on this. A pigment has a property of having a colour. So, a pigment *is* a chemical compound and *has* a colour. --Egon Willighagen (talk) 13:10, 17 May 2019 (UTC)
Oh good that's what I thought too. - PKM (talk) 22:01, 17 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 09:11, 23 May 2019 (UTC)

Bug when units redirected

I recently noticed that when an item that has been used as a unit is redirected, it shows up as an unlinked Q-code in the interface. See this example: Q63862967#P1436. I just wanted to report this in case it's not a known issue. I think there are several thousand items of my past imports that would be affected by same item from the example being merged and redirected. There may be other cases live in Wikidata where this has happened as well. Dominic (talk) 15:33, 17 May 2019 (UTC)

There are bot which usually resolve all redirects a few days after a merge. However, units are a rather remote place of use which might be overlooked. There are currently ~100 statements with unit series (Q59221354). —MisterSynergy (talk) 20:42, 17 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 09:13, 23 May 2019 (UTC)

Talk pages consultation: Phase 2

List items that have certain property

How can I list "railroad lines" (Q728937) that have "OSM relation ID" property (P402)? Like for example this one - Q802995. Possibly limit to country: CZ too. How can I show results to map (in case the item contains coordtinates)? Thanks.--Kozuch (talk) 16:17, 18 May 2019 (UTC)

@Kozuch: Best place for a question like this is WD:RAQ Jheald (talk) 16:29, 18 May 2019 (UTC)
Thanks, asked there.--Kozuch (talk) 16:42, 18 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 09:18, 23 May 2019 (UTC)

Unable to add en.WS link item

I cannot add s:en:Author:Robin F. Wynne to Robin F. Wynne (Q18749301). Each time I try to publish I see an error message:

A page "Author:Robin F. Wynne" could not be found on "enwikisource".
The external client site "enwikisource" did not provide page information for page "Author:Robin F. Wynne".

I have tried adding the link to a newly created data item Robin F. Wynne (Q63930093), with the intent to merge, but I get the same error message from that page as well. Can someone determine why this error is occurring? --EncycloPetey (talk) 20:09, 18 May 2019 (UTC)

Same for me. Trying to add it:Categoria:Dolci a base di pasta di mandorle in Q63930113, but the same message appears. --Superchilum(talk to me!) 20:12, 18 May 2019 (UTC)
Same problem trying to link to Commons categories. Joostik (talk) 20:13, 18 May 2019 (UTC)

Sounds as though this is a general problem then. --EncycloPetey (talk) 20:16, 18 May 2019 (UTC)

Tried again, and now seems to be working. Still not sure why it wasn't working before, but seems to be corrected now. --EncycloPetey (talk) 20:26, 18 May 2019 (UTC)
apparently WMF outage - -- Slowking4 (talk) 16:45, 19 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 09:23, 23 May 2019 (UTC)

Strange wdtn: forms for some VIAFs ?

I never usually use the wdtn: forms of external-ID properties to get the full URLs, but I did today and got these rather odd results for some VIAFs:, where the identifier part seems to have been collapsed to just a couple of digits at most, eg Arsène Alancourt (Q2288101) -> <>.

For comparison, wdtn: values for Library of Congress authority ID (P244) seem to be as expected:

Data mis-written? Or something else? Jheald (talk) 22:27, 19 May 2019 (UTC)

Moved to Phab:T223946. --Tagishsimon (talk) 00:51, 21 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 09:29, 23 May 2019 (UTC)

how to represent name changes in academic departments?

Hi all - I'm trying to figure out how to represent name changes of academic departments. For instance, the MIT Department of Psychology is now known as Massachusetts Institute of Technology Department of Brain and Cognitive Sciences, (Q42307047) - this name change history is not yet represented in the item. (ref: [17]). For this and many similar cases, I would like to figure out how to represent "known as" with time qualifiers, since academic sub-units (and even universities themselves) change names all the time. Should it just be official name (P1448) with a start time and end time? -- Phoebe (talk) 17:31, 20 May 2019 (UTC)

Assuming it's just a name change, and not a more significant merger or reorganization, then attaching start time/end time qualifiers to the name properties would be the right thing - also the previous name should probably be in the term box under "Also known as". ArthurPSmith (talk) 17:44, 20 May 2019 (UTC)
@ArthurPSmith: thanks! I also need to figure out how to represent mergers, eg Laboratory for Computer Science & the AI Lab (neither of which have items afaik) became MIT Computer Science and Artificial Intelligence Laboratory (Q1354917) -- Phoebe (talk) 17:49, 20 May 2019 (UTC)
@Phoebe: I believe in most cases you want separate items for the items before and after, the "before" items should have replaced by (P1366) statements linking to the "after" items, and conversely the "after" items should have inverse replaces (P1365) statements. ArthurPSmith (talk) 18:07, 20 May 2019 (UTC)
What?! I'm pretty sure that's a job for official name (P1448) with appropriate time subproperties, especially if the name is all that changes. Circeus (talk) 20:16, 20 May 2019 (UTC)
Well in this particular case we are talking about two independent departments that merge into one. It matters because other items might have relationships (like authorship, affiliation) that attach to the individual component departments whether or not they have items. -- Phoebe (talk) 20:19, 20 May 2019 (UTC)
@Phoebe: See Faculty of Arts, Humanities and Cultures (Q58610834) for an example of a new faculty formed from the merger of Faculty of Arts, University of Leeds (Q24531913) and Faculty of Performance, Visual Arts and Communications (Q26215521). Simon Cobb (User:Sic19 ; talk page) 22:20, 20 May 2019 (UTC)

@Sic19: thanks!

This section was archived on a request by: Regards, ZI Jony (Talk) 09:33, 23 May 2019 (UTC)


Cluebot patrols en.Wikipedia and reverts obvious vandalism. I've enquired about having it run on this project; initially to patrol descriptions. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:11, 18 May 2019 (UTC)

Q1624024 vs Q63928711

Should R38 class (Q1624024) and ZR-2 (Q63928711) be merged? R38 class (Q1624024) links to a Wikipedia article on this entire class of airship, even though only one was built. The bulk of the article is on the the aircraft accident of the one built. ZR-2 (Q63928711) is the one that was actually built and crashed. --RAN (talk) 17:10, 18 May 2019 (UTC)

Metrics on use of Wikidata in other projects

I seem to recall seeing a report or tool, showing statistics on the use of Wikidata in our sister project. Where is it, please? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:09, 18 May 2019 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── Thank you. The Usage Dashboard is what I had in mind, but the rest are also very useful; so I've made WD:Metrics. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:50, 18 May 2019 (UTC)

Beware what the statistics are not telling. Some Mediawiki projects may use only the main value from some statements, while other may also show qualificator values, references, and additional information like for instance chains of administrative entities until the country for places, convert units to the local preferences etc. but both will count as a page using Wikidata. So some statistics about how much and what is used per page could also be interesting. --Dipsacus fullonum (talk) 20:25, 18 May 2019 (UTC)

Template:Item documentation - bumping requests from 2017

I love {{Item documentation}} and use it all the time. However, there are two bugs/requests outstanding from 2017 on the template's Talk page that I'd like to see improved.

  • The links to Taxonomy Browser don't work (possibly because {{Taxonomy Browser}} doesn't work), though the tool itself still works at this link). @JakobVoss:, can you help?
  • It would be great if we could add a link to "explore this property with SQID". Can someone here help with that?

Thanks, all! - PKM (talk) 20:02, 18 May 2019 (UTC)


Hello everyone, I would like to know if there is a substitute in Kasparbot for authority control between wikis and WD? Thank you for your replies.   --Eihel (talk) 08:07, 19 May 2019 (UTC)

  • Is this still needed? --- Jura 08:44, 19 May 2019 (UTC)


  1. How I can have more than 46 languages loaded on property page to have labels and descriptions already in table for faster edits? Eurohunter (talk) 14:57, 19 May 2019 (UTC)
  2. Why if I have more than 46 languages in my babel table these languages not even shows in "All entered languages" at property page? Eurohunter (talk) 14:57, 19 May 2019 (UTC)
  3. There should be bookmark with settings in the preferences which let me choose what langauges I want to use instead of set them in my userpage. Eurohunter (talk) 14:57, 19 May 2019 (UTC)
  4. Saving changes to these 46 items takes very long time, too long in 2019. Eurohunter (talk) 14:57, 19 May 2019 (UTC)
  • You could just load the labels in a spreadsheet and edit them there? --- Jura 15:58, 19 May 2019 (UTC)
    • @Jura1: What do you mean? Eurohunter (talk) 17:17, 19 May 2019 (UTC)
    • You can download the labels from Query Service, edit them in a spreadsheet and then upload the edited version with QuickStatements. --- Jura 17:40, 19 May 2019 (UTC)


Property:P1220 appears to have the url changed so the links do not work, anyone want to change the url formatter? Click on any of the Wikidata property examples and see the error message. --RAN (talk) 20:03, 24 May 2019 (UTC)

I hate to be that guy but i don't see any error message. All the example links works on my computer. --Trade (talk) 00:40, 25 May 2019 (UTC)
Vahurzpu did that. //Matěj Suchánek (talk) 11:07, 25 May 2019 (UTC)
Yeah, that was me. I didn't post here about it because the change didn't take effect immediately, and I wasn't sure how long it would take for it to do so. Vahurzpu (talk) 14:34, 25 May 2019 (UTC)
This section was archived on a request by: Matěj Suchánek (talk) 11:07, 25 May 2019 (UTC)

Talk pages consultation 2019 – phase 2

The Wikimedia Foundation has invited the various Wikimedia communities, including the English Wikipedia, to participate in a consultation on improving communication methods within the Wikimedia projects.

Phase 2 of the consultation has now begun; as such, a request for comment has been created at Wikidata:Requests for comment/Talk pages consultation 2019, phase 2. All users are invited to express their views. Individual WikiProjects, user groups and other communities may also consider creating their own requests for comment; instructions are at mw:Talk pages consultation 2019/Participant group sign-up. (To keep discussion in one place, please don't reply to this comment.) Jc86035 (talk) 16:22, 20 May 2019 (UTC)

item for home matches of a team

Hello. A long time ago I created items for home games of football clubs (like home matches of APOEL FC (Q27832699)) to use them with broadcast by (P3301). See Q52151351#P3301. That was the suggestion on Project Chat Wikidata:Project chat/Archive/2016/11#broadcast by. But now, with the constraints, I realized that there are some problems. I wasn't able to find an example of teams of others countries. Can you help me to add the appropriate properties on the items? Xaris333 (talk) 18:34, 20 May 2019 (UTC)

Does Wikidata lack a Quality mindset? Sometimes it feels more like a playground for python exercises...?

I start to see more and more edits done because it's there and I can do some Python instead of seeing that references adds quality to Wikidata!!

Should every source uploaded to Wikidata confirm all possible statements like a reference is a garbage can of everything we have found or should Wikidata have some understanding that a reference should add value and that not every source has the same quality? As Wikipedia starts using Wikidata driven templates things added to Wikidata also will appear in the articles if no one filter low quality additions out? E.g. Batch 12808 is adding references for gender by Marsupium. the source is a thesaurus Union List of Artist Names (Q2494649) and if we check Agnes Branting (Q4940450) we have for this person both en:primary sources the birthbok and Dictionary of Swedish National Biography ID (P3217) that is written by fully employed historians at the Swedish National Archives... what added value does a thesaurus add? In my book of quality it should not be added.

Question: is this a problem? Do we need to rank sources or should the consumer(every WD driven template) /Wikipedia reader be the person filtering what adds value? Or is more references always better - Salgo60 (talk) 19:35, 14 May 2019 (UTC)

@Salgo60: Concerning the specific task of that batch: It adds a reference to sex or gender (P21) statements only if there is no other reference at all. That is against Help:Sources and in best case they shouldn't have been added without a reference in the first place. sex or gender (P21) isn't as trivial as it seems. I stumbled upon many cases for which it was wrong here or in other databases. That's why I consider it valuable to have a source for the statements. Dictionary of Swedish National Biography ID (P3217) doesn't seem to state sex or gender (P21) explicitly in [18] for Agnes Branting (Q4940450), does it? In art history, Union List of Artist Names (Q2494649) and RKDartists (Q17299517) are probably the freely available person databases of their dimension with the highest quality. The majority of printed sources won't state sex or gender (P21) explicitly, neither seems Dictionary of Swedish National Biography ID (P3217) to do so. To me it seems for the majority of those statements we will either use databases like Union List of Artist Names (Q2494649) as a reference or none at all. I prefer the first. Union List of Artist Names (Q2494649) is issued by a renowned research institute and is a reliable source. The information that it backs a statement is the value added in my eyes. Until it's discussed if it should run at all, I've stopped said batch together with another running one that was adding Union List of Artist Names (Q2494649) references to occupation (P106) statements. (No Python used BTW, only federated SPARQLing.)
Concerning the general questions: should Wikidata have some understanding that a reference should add value and that not every source has the same quality? Yes. Do we need to rank sources or should the consumer(every WD driven template) /Wikipedia reader be the person filtering what adds value? Both, we have to provide information so that templates can filter the references and pick the best. At the moment Wikidata does a very bad job in this, I'm very interested in improving this situation. Or is more references always better? No. (And that's why I'm adding the references only to statements which don't have any other reference. In many cases existing ones are worse I think, but I haven't figured out a good way to rank them, apart from filtering by imported from Wikimedia project (P143) which I planned to do.)
Cheers, --Marsupium (talk) 20:39, 14 May 2019 (UTC)
Thanks for fast answer
  1. Swedish church books shows male/female. For Agnes Branting (Q4940450) you have on Page line 10 column 7 indicates female and column 6 is male.
  2. My point is that in 10 more months we will have 10 more sources like this that someone feels is good
  3. The quality of Getty as a source without links to a primary sources I think is also worth a discussion
  4. Having a Wikipedia with +300 languages and then let all templates do the filtering I feel is bad design. I believe in DRY (Don't repeat yourself)
+ for using federated SPARQL see my test
- Regards Salgo60 (talk) 21:21, 14 May 2019 (UTC)
For reference, ULAN does list the sources for its statements: see Branting, Agnes which cites "Allgemeines Künstlerlexikon / Artists of the World (2009-) accessed 1 March 2010" as well as the LOC authority file. - PKM (talk) 19:50, 15 May 2019 (UTC)
"Does Wikidata lack a Quality mindset" yes, it has a tool mindset; it lacks a quality circle process to curate references. but merely adding superfluous references is not a quality problem, per se, rather the lack of a quality improvement process to assess those references is. importing low quality data is not necessarily a problem, but the lack of quality process is. improvement is getting done on an ad hoc basis. there is not one best reference. Slowking4 (talk) 02:15, 16 May 2019 (UTC)
@Slowking4: and how do we move this forward? In the video above we also have the problem that we have an excellent source but in Wikidata we get people of +200 languages reading an article with this very good source. How do we communicate the "TRUST" of a source used? - Salgo60 (talk) 08:41, 16 May 2019 (UTC)
  • Personally, I think multiple sources are better, especially if they are of different levels: primary, secondary, and compilations of these. If some Wikipedia edition prefers, e.g., only sources with "authority" in the name, it can easily filter for that. --- Jura 08:05, 16 May 2019 (UTC)
@Jura: I think we will see the number of references will explode I can already see on e.g. Selma Lagerlöf (Q44519) has 12 references for birth dates and then a reader has difficult which source can you trust or not. When doing a federated search between Wikidata and the we have +20 mismatches and then its nearly impossible to say if Wikidata is correct because I dont know the quality of the Wikidata sources
I miss that we better describe if a source is a primary source or not, the quality of the source. And this Quality statement should be machine readable. I asked Denny Vrandečić how we better should model quality inside WD see tweet - Salgo60 (talk) 08:37, 16 May 2019 (UTC)
I think the GUI isn't really working for multiple sources at Q44519#P569. At least in the default skin, it's unclear where one reference ends and the next starts. Other than that, what is the problem with the statement? Do you think it's incorrect or incorrectly supported by some of the references included?
You could add type of reference (P3865) to some of the references, but the some extent the same information is included in the item for the reference.
Personally, I think the quality of a given reference can vary for each statement. It's a mistaken assumption by some Wikipedias that only references with the word "authority" in it can be trusted or should always be trusted. --- Jura 08:53, 16 May 2019 (UTC)
Yes this is complex, but I think Wikipedia/Wikidata should help the reader understand the quality/experience other have of using this source and also explain what sources are prefered using
I think Wikipedia/Wikidata could be an excellent place to gather opinions/warnings about a source that wuld help me understand what I can trust or should not trust. Today I feel its just up to me - Salgo60 (talk) 10:30, 16 May 2019 (UTC)
  • So the question actually is about a perceived lack of (expression of) a ranking between external sources referenced by Wikidata. This has nothing to do with "quality". There's also nothing new about it, in that usually references on Wikipedia are all grouped in a single section even if they are of differing "ranks"/reliability. The only difference is that Wikidata gravitates towards comprehensiveness when it comes to external identifiers, so there is some significant overlap in the functions performed by the many identifiers one item may have. Nemo 09:03, 16 May 2019 (UTC)
@Nemo: it's more about TRUST and communicate if a source has a track record of TRUST. Ranking sources of low quality makes no major change. Compare how Academic awards are ranked by external organisation - by IREG Observatory on Academic Ranking and Excellence. I guess that is an interesting approach if we could get people experts in the field share the experience they have of sources in a machine readable way.
My guess is that Wikidata/Wikipedia is an excellent platform for sharing knowledge of quality of sources. We have articles in 70 languages about the latest Nobel prize winner Nadia Murad Basee (Q22007112) but we have no exact birth date in Wikidata. I guess some of those people writing the articles has more or less good sources about her birth- Salgo60 (talk) 10:30, 16 May 2019 (UTC)
birth dates are controversial as personal identifiable information. you tend to see them for the deceased and athletes for disambiguation. and wikidata is a work in progress, you should not expect to find every fact that you want to find. you will have to build a consensus about how to curate references. maybe a discussion at Wikidata talk:WikiProject Source MetaData? your reference trust appears to be original; you might want to bring some academic citations to the table. Slowking4 (talk) 13:15, 16 May 2019 (UTC)

I do think Wikidata lacks a quality mindset, but this is by design. Wikidata leaves quality discussions historically to Wikipedia projects and concentrates on enabling those projects to provide additional information within the scope of that project. That said, I do believe Wikidata needs to document and better reference itself in this ongoing process of enabling quality for all other projects, not just Wikipedia. For example, we have this great "pot of gold" at the bottom of Wikidata items with all external ids. These ids have been criticized by LOD people for "not being full urls" but anyone who has hung around here since 2012 or 2013 knows external websites tend to change their base urls from time to time and thus the "id part" in our external id process has proved extremely useful to have. That said, I think there should be a better way of distributing the information hidden in these urls as references in an item for its various statements without having to copy the entire url with reference URL (P854) on each statement. Jane023 (talk) 12:11, 17 May 2019 (UTC)

@Jane023: There is. Just give the identifiers as the citations. Cf eg all the citations to the BL online catalogue on an item like A Residence on the Shores of the Baltic (1842 edition) (Q63314374) Jheald (talk) 22:23, 17 May 2019 (UTC)
Still messy, because that is actually more difficult to do than adding the url. Plus this increases the total number of property statements for that property. Not sure which is the lesser of two evils. Jane023 (talk) 22:43, 17 May 2019 (UTC)
The way described by Jheald is current consensus per Help:Sources#Databases. I've spent quite some time in getting stated in (P248)s in references where people didn't put them when creating the references. Their absence makes querying and reuse of references way harder if not impossible. I don't think the number of reference snaks should be an argument here, best way to keep entities small is not to add content at all – but yes, I often wish we had an equivalent for w:WP:NAMEDREFS. --Marsupium (talk) 20:12, 19 May 2019 (UTC)
  • Correct me if I'm wrong, but there seems some consensus that more quality related data about the sources we use as references is desirable. Should we try to move forward to some actable steps? Without much consideration I'd be interested in:
    1. Do we have a way to indicate if a source (or its different levels) is peer-reviewed?
    2. … to indicate if a tertiary source gives sources (like ULAN, see above) or not?
    3. … to easily get the level of a data source: researcher, institutions: research institute with discipline that of the content ---------> crowd-sourced sources, etc.?
    4. … (what else?)
I think I've stumbled upon an ontology for this kind of stuff, but don't remember where. What exactly can do to make judging easier? --Marsupium (talk) 20:12, 19 May 2019 (UTC)
@Marsupium: From #wmhack in Prague earlier today, User:Astinson (WMF) re-posted a tweet [19] about m:Cite_Unseen, which would seem to be sort of in this area. There was also this [20] from User:MartinPoulter about MonumentalGLAM [21], aiming to produce the Sum of All GLAMs, based on the existing Monumental tool. But I don't know if either of those were what you were thinking of. Jheald (talk) 21:12, 19 May 2019 (UTC)
Thanks for the links! I think m:Cite Unseen#Classifying sources is very interesting. We should be able to classify into those and more classes here with queries, but unlike Cite Unseen without heuristics and URL matching. Prerequisite probably that references use stated in (P248). --Marsupium (talk) 03:34, 20 May 2019 (UTC)
Agree @Marsupium: m:Cite Unseen#Classifying sources combined with a process like we have for featured articles but for featured sources see my tweet - Salgo60 (talk) 06:05, 20 May 2019 (UTC)

Here another link to a related discussion: Wikidata:Project chat/Archive/2017/10#Proposal on citation overkill
And to bring in a real world example for filtering on Wikidata: ru:Модуль:Wikidata and uk:Модуль:Wikidata use deprecatedSources and preferredSources lists which black- and whitelist some sources. It doesn't seem to work 100% tho (cf. here), but could be one way. --Marsupium (talk) 03:34, 20 May 2019 (UTC)

I did a draft for Wikidata:WikidataCon_2019 see T222142 "WikidataCon 2019: We need a better model communicating quality/relevance of sources in Wikidata / Provenance". If anyone will be in Stockholm for Wikimania2019 and would like to discuss this don't hesitate to contact me I live there ;-) - Salgo60 (talk) 05:47, 23 May 2019 (UTC)
My vision during Q & A Wikidataconf 2017 link at 42 min - Salgo60 (talk) 10:17, 27 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Adding many properties for existing items

I compared a local dataset related to Cultural Opposition (Understanding the Cultural Heritage of Dissent in the Former Socialist Countries) with Wikidata regarding people, organizations and groups. The comparison was about extending Wikidata with new properties for existing items and adding Same As property, so the local dataset will be connected to Wikidata.

Please your advice how to add this data (like providing you with a RDF file in a specific syntax, or...)

Thank You in advance, Ghazalgf (talk) 18:23, 18 May 2019 (UTC)

@Ghazalgf: If the properties you would need do not yet exist, the normal procedure is to propose them at Wikidata:Property proposal. What sort of data are you proposing to add? Is the dataset freely licensed, and is it already online and publicly viewable? Jc86035 (talk) 11:12, 19 May 2019 (UTC)
@Jc86035: First, the properties are existed but for some people and organizations they are missing, so we will extend Wikidata data. For example, Date of birth, Place of birth...

The dataset is published at Zenodo under, the license is there. There is also the SPARQL endpoint and the public browsing UI:

This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Best way to link to people profile on organisation websites?

These are not, typically "official websites", and they sometimes will have more than one (cf. Carlos A. Camargo (Q59765406)). This property is more connected more closely to affiliation (P1416) or employer (P108), and I feel relegating it to references is unhelpful. Circeus (talk) 06:43, 19 May 2019 (UTC)

I recommend described at URL (P973). - PKM (talk) 21:57, 19 May 2019 (UTC)
Oh! Good thinking. I wouldn't have considered that! Circeus (talk) 14:28, 20 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Shape Expressions arrive on Wikidata on May 28th

Hello all,

After several months of development and testing together with the WikiProject ShEx, Shape Expressions are about to be enabled on Wikidata.

First of all, what are Shape Expressions?

Shape Expressions (Q29377880) is a concise, formal modeling and validation language for RDF structures. Shape Expressions can be used to define shapes within the RDF graph. In the case of Wikidata, this would be sets of properties, qualifiers and references that describe the domain being modeled.

See also:

What can it be used for?

On Wikidata, the main goal of Shape Expressions would be to describe what the basic structure of an item would be. For example, for a human, we probably want to have a date of birth, a place of birth, and many other important statements. But we would also like to make sure that if a statement with the property “children” exists, the value(s) of this property should be humans as well. Schemas will describe in detail what is expected in the structure of items, statements and values of these statements.

Once Schemas are created for various types of items, it is possible to test some existing items against the Schema, and highlight possible errors or lack of information. Subsets of the Wikidata graph can be tested to see whether or not they conform to a specific shape through the use of validation tools. Therefore, Schemas will be very useful to help the editors improving the data quality. We imagine this to be especially useful for wiki projects to more easily discuss and ensure the modeling of items in their domain. In the spirit of Wikidata not restricting the world, Shape Expressions are a tool to highlight, not prevent, errors.

On top of this, one could imagine other uses of Schemas in the future, for example building a tool that would suggest, when creating a new item, what would be the basic structure for this item, and helping adding statements or values. A bit like this existing tool, Cradle, that is currently not based on ShEx.

What is going to change on Wikidata?
  • A new extension will be added to Wikidata: EntitySchema, defining the Schema namespace and its behavior as well as special pages related to it.
  • A new entity type, EntitySchema, will be enabled to store Shape Expressions. Schemas will be identified with the letter E.
  • The Schemas will have multilingual labels, descriptions and aliases (quite similar to the termbox on Items), and the schema text one can fill with a syntax called ShEx Compact Syntax (ShExC). You can see an example here.
  • The external tool shex-simple is directly linked from the Schema pages in order to check entities of your choice against the schema.
When is this happening?

Schemas will be enabled on on on May 21st and on on May 28th. After this release, they will be integrated to the regular maintenance just like the rest of Wikidata’s features.

How can you help?
  • Before the release, you can try to edit or create Shape Expressions on our test system
  • If you find any issue or feature you’d like to have, feel free to create a new task on Phabricator with the tag shape-expressions
  • Once Schemas are enabled, you can discuss about it on your favorite wikiprojects: for example, what types of items would you like to model?
  • You can also get more information about how to create a Schema

See also:

If you have any questions, feel free to reach me. Cheers, Lea Lacroix (WMDE) (talk) 13:28, 19 May 2019 (UTC)

  • congratulations to all the team. sounds like a lot of work leading up to this moment. will this affect the wikibase extension directly? the docker image? Pdehaye (talk) 13:41, 19 May 2019 (UTC)
  • @Lea Lacroix (WMDE): So presumably these are going to be important for Wikidata items that defines classes, and there will be a property like properties for this type (P1963) but on steroids, that will connect a class-item to a shape expression, which will record behaviour, properties etc that may be expected for instance-items of that class (and its subclasses), and for eg the sorts of values expected for particular properties for such instances?
So for example on a class like ceremonial county of England (Q180673) one might identify a shape expression to say that instances of the class should have located in the administrative territorial entity (P131) with a value that is instance of (P31) region of England (Q48091)
Is that the general sort of idea? Jheald (talk) 14:57, 19 May 2019 (UTC)
@Jheald: That is indeed one of the use-cases. I created a series of schemas on ceremonial county of England (Q180673) and we captured a possible scenario in a screen cast. The schemas is on github. Hopefully, this clarifies things? --Andrawaag (talk) 14:28, 24 May 2019 (UTC)
@Pdehaye: it's not decided yet. Seeing all the extensions and tools that we have around Wikidata, we may decide to have different "packages", like a data quality package, a data import package... For now, EntitySchemas, will not be included in the extension or the docker image.
@Jheald: You got the idea right :) Lea Lacroix (WMDE) (talk) 18:16, 19 May 2019 (UTC)
@Lea Lacroix (WMDE): So will something like a constraint violation show up on statements which do not conform? And wht sort of queries will be possible eg in WDQS? Jheald (talk) 18:44, 19 May 2019 (UTC)
  • @Lea Lacroix (WMDE): is there a plan for a natural language interface to generate a shape expression by picking classes, properties, and qualifiers (and importing their various constraints)? - PKM (talk) 20:02, 20 May 2019 (UTC)
@Jheald: At this point there is no interaction between the shape expressions and the interface of the entities. In general, we'll wait until the new namespace is populated, that the community had discussions, decisions and ideas about shape expressions, before moving on with new development. But if the community comes up with requests, we'll definitely investigate on them.
Querying the schemas is also not possible at the moment.
@PKM: There is no natural language interface for now, but did you check this tool that helps infering schemas from existing items? Lea Lacroix (WMDE) (talk) 16:16, 21 May 2019 (UTC)
@Lea Lacroix (WMDE): thanks, that will be very helpful! - PKM (talk) 20:18, 21 May 2019 (UTC)

Update: it's now live! You can for example have a look at E10, the shape for human, or create a new EntitySchema. Lea Lacroix (WMDE) (talk) 16:15, 28 May 2019 (UTC)

This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

get wikidata ID in Wikipedia article template

I need to construct an external link to which contains wikidata ID as a parameter (like in a template on Wikipedia. Is it possible to get current article's wikidata ID somehow? To be more concrete I deal with WIWOSM.--Kozuch (talk) 11:26, 20 May 2019 (UTC)

You need a Lua module. It should contain a code like this:
local p = {}

p.getCurrentId = mw.wikibase.getEntityIdForCurrentPage

return p
Then in the template, you will use {{#invoke:Name of the module|getCurrentId}}. If the article is connected to an item, it will return the item's ID. Matěj Suchánek (talk) 17:40, 20 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Items for non-notable film extras and non-notable fictional characters

I stumbled on the item Pixel Perfect (Q731789) and before I take an axe to it, I'd like to know if I'm the only one thinking that its cast member list is an aberration. The first thing you notice is that it has 35 entries which sounds like a really long list for a television film. Clearly, somebody (it seems like Escudero (talkcontribslogs) is the main editor here) took the whole cast list from IMDb and created items for the actors that appeared in the film but did not yet have a WIkidata entry. This created at least 25 if not 30 new entries for actors who do not meet notability requirements and only had bit parts or were even what appears like background extras. Moreover, the same user created items for all fictional characters of the film, again regardless of their notability. The result is items for fictional characters such as Falon Grace (Q42723221), Hapless Teacher (Q42889551), Audience Member #3 (Q42889557) and my favorite: Moist Towelette #3 (Q42889583). The problem is that if I submit any of these actors or any of the corresponding fictional characters, DeltaBot complains that the item is linked to the item for the non-notable actor and vice-versa. Could we agree that these items should be deleted? Note that I'm all for keeping the notable actors in the cast member list or even an actor that does not yet have a Wikipedia entry but does have an important role in the film. I am however for the deletion of the rest of the cast list (which in any case is most likely taken from a source of dubious reliability) and for the deletion of the items for fictional characters that do not exist in other works beyond that specific film. At the very least, all bit parts should be deleted as they serve no structural purpose. Pichpich (talk) 20:17, 20 May 2019 (UTC)

  • No, items are required for fictional characters as the datatype of the property to add them is item, not string. Bear in mind that this is not Wikipedia, we can't built lists without making items for every element.
    If you need to build a list that only displays leading actors, please filter the query accordingly. --- Jura 20:23, 20 May 2019 (UTC)
  • These fall under WD:N#3, items created for the purpose of improving statements in other items. Creating items for each otherwise-non-notable role in films is standard practice. The items should be kept. --Yair rand (talk) 01:51, 21 May 2019 (UTC)
In short, Wikidata largely doesn't care at all about notability. It feeds on data. It must have data. If you have data, Wikidata wants the data. All data. If you have a blog or a twitter handle or an entry in the phone book, you too can be in Wikidata! You are data! All hail data. -Animalparty (talk) 03:07, 21 May 2019 (UTC)
Nope. See WD:N. --Yair rand (talk) 03:23, 21 May 2019 (UTC)
It isn't that Wikidata doesn't care about notability: it's that the unit of notability isn't always the individual Q-item. The contents of Louis Armstrong's bedroom in his house museum may well be notable. The contents of your bedroom probably are not, even if they are very similar. -Jmabel (talk) 04:42, 21 May 2019 (UTC)
  • I'm the user that created a lot of these pages, yes, but was a question about completeness of the film item. I don't know if an actor is notable, but he/she appears in the end credits of a film, so I suppose he/she should be connected to the correspondent item. Escudero (talk) 08:05, 21 May 2019 (UTC)
  • Note that we do have the qualifier name of the character role (P4633), which IMO can (and should) be used for such cases instead of creating items for each and every background character. I'm perfectly fine with creating items for fictional characters if they have a decent-sized role in a work, appear in several works and/or have external IDs or reliable sources talking about them. I've created various myself and expanded a lot of existing items for fictional entities. But every background character listed in IMDb? That's a bit much, even for Wikidata's low barrier of entry. Especially since even the items for fictional characters that we already have (even those notable by Wikipedia standards) are mostly in a bad shape (especially references-wise) and only very rarely see any improvements after their creation. --Kam Solusar (talk) 09:47, 21 May 2019 (UTC)
    • That property was mainly done for theatre performances to allow recording the name of the character in the way it's stated on a theatre program. --- Jura 10:30, 21 May 2019 (UTC)
      • But could be used for other works like films, tv series of computer games as well without problems. --Kam Solusar (talk) 11:45, 21 May 2019 (UTC)
        • The problem for films would be that the qualifier would need to be repeated countless times. Theatrical performances are different as they are generally a specific performance of a production in a given language of a play. --- Jura 09:50, 23 May 2019 (UTC)
    • P4633 is intended to be supplementary, not a replacement for the actual item. --Yair rand (talk) 17:12, 21 May 2019 (UTC)
  • When entries like these get added -- I'm thinking about all the minor actors -- I'm curious to know what mechanisms exist to ensure we don't get too many duplicates, and how well these mechanisms apply in practice. (The same issue applies to authors of scientific papers.) It seems to me it's probably pretty easy to end up with duplicates, and potentially a lot of work to eliminate them. Obviously if external identifiers are assigned, this helps a lot, as their uniqueness constraints apply -- but looking up and assigning lots of external identifiers is work, too. Scs (talk) 10:41, 21 May 2019 (UTC)
    • Usually, such items should have a statement "present in work" that links back to the work. The problem with fictional characters is often the opposite: should we really use the same item for a character named the same way in works of different creators. --- Jura 11:07, 21 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Class of exceptions for a property constraint ?

Lucas Werkmeister (WMDE)
Jarekt - mostly interested in properties related to Commons
John Samuel
Yair rand
Jon Harald Søby
Was a bee
Peter F. Patel-Schneider
ZI Jony
  Notified participants of WikiProject property constraints

Is it possible to indicate that a whole class of items should be excepted from a property constraint? The qualifier exception to constraint (P2303) I think only excepts a single item.

For example, on image (P18) there is a format constraint (Q21502404) that if the label or the description of the item contains the word "map" or "carte" then locator map image (P242) should be used. But if the item is a map, then this message is inappropriate, and P18 is indeed the right property to use.

Is there a way to express this? Jheald (talk) 06:56, 21 May 2019 (UTC)

@Jheald: Not at the moment. There are plenty of constraints where the author apparently intended some interpretation like “no violation if the value is this item” (random example) or “no violation if the subject has this class” (random example), but the only interpretation that the WikibaseQualityConstraints extension actually follows is “no violation if the subject is this item”, i. e. “on that page, skip checking this constraint completely”. (Though I have no idea what KrBot does.) --Lucas Werkmeister (WMDE) (talk) 11:34, 21 May 2019 (UTC)
@Lucas Werkmeister (WMDE): Thanks, Lucas. So what's the best way forward here? If a property were to be proposed to capture this information, would it be possible to implement it? Jheald (talk) 10:35, 22 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Scientific Work

Hoi, I have added Q63974451 it is not published in a publication. Consequently, it is a "scientific work". The issue is that it does trigger warnings about the publishers. How to improve this? Thanks, GerardM (talk) 10:24, 21 May 2019 (UTC)

But it is published in some fashion? I think you need to add another appropriate instance of (P31) value... ArthurPSmith (talk) 15:12, 21 May 2019 (UTC)
I would additionally call that an example of a white paper (Q223729) or a technical report, though it looks like the technical report items need to be sorted out to differentiate them from one another...sigh. -- Phoebe (talk) 20:10, 21 May 2019 (UTC)
You will find the paper online as a PDF (check the item). The pedigree of organisations and people is impeccable. Thanks, GerardM (talk) 11:34, 22 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Items for multi-intersection Categories at Commons?

Naive question, since I'm still getting to know wikidata: Is something like monoplane with 1 tractor-piston-propeller engine (Q33120898) a good way of describing an aircraft in the wikidata sense of things? Wouldn't it make much more sense to tag something like LFU 205 (Q1724337) directly with wing configuration (P1654) = monoplane (Q627537) etc.? --El Grafo (talk) 12:27, 21 May 2019 (UTC)

These are different use cases. The first (though subclass of (P279)) embeds the item into a broad "classification tree", the second just defines details about it. A Boeing 737 (Q6387) is also a monoplane (as are most modern planes, in fact), so trying to sort models by that criterion alone is neither particularly efficient nor intuitive. Circeus (talk) 15:39, 21 May 2019 (UTC)
Yes, but wouldn't getting a list of all monoplanes be really cumbersome if items are tagged with this kind of pre-made intersection only? I guess the thing i didn't realize is that LFU 205 (Q1724337) should have both monoplane with 1 tractor-piston-propeller engine (Q33120898) and monoplane (Q627537) – something that is not possible at Commons with Categories because of "overcategorization". --El Grafo (talk) 08:48, 22 May 2019 (UTC)
Not sure I fully get the question ; but as monoplane with 1 tractor-piston-propeller engine (Q33120898) is subclass of (P279) monoplane (Q627537), tagging an item only with the former is sufficient, as this item is implicitly the latter. See for example Wikidata Graph Builder to explore that tree. Jean-Fred (talk) 11:26, 23 May 2019 (UTC)
(ie yes, we do have the equivalent presumptions to c:COM:OVERCAT here. Jheald (talk) 12:52, 23 May 2019 (UTC) )
There's a similar discussion going on over at Wikidata:Property proposal/Rigging, where the suggestion to add a new property 'Rigging' was countered with the argument that it's already possible for any ship to be instance of (P31) a specific subtype of ship (Q11446) (such as full-rigged ship (Q1581130)) that describes the rigging.
I don't know what the generally right approach is. (Personally I prefer finer-grained properties to finer-grained types, but I'm no expert.) Are there any general properties of data modeling that give any guidance here? Scs (talk) 12:04, 23 May 2019 (UTC)
The subclass-or-attribute question is a perennial, though in some ways it's a surprise it doesn't come up more than it does. I'm sure people must have written essays and guidance and help-pages on this, and logged past discussions, but I am not sure I'd know where such content would be -- or even where to go looking for it. Jheald (talk) 12:52, 23 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

« Fixing » constraints

This query shows that a few classes have

⟨ subject ⟩ subclass of (P279)   ⟨ unknown value Help ⟩

statements. I suspect this is the doing of peoply who wanted to « fix » constraint such as « values used in an instance of (P31) statement should have an instance of (P31) statement, so people just added such a statement. The problem with this is that it just does not really help to have those statements … (just putting this here, it may start a discussion about the correct use/fix of constraints and their usefulness, but I don’t really have a smart thing to say about this) author  TomT0m / talk page 08:28, 22 May 2019 (UTC)

  • I would guess they are trying to say the item is a class, but they are not sure where to put it in the class hierarchy. - Jmabel (talk) 15:29, 22 May 2019 (UTC)
    • Probably something like that, but this is sweeping things under the rug. It in effect make disappear the constraint from constraint report with keeping the class out of the class tree. Maybe it could be useful to add a « not unknown value Help » constraint on subclass of (P279). author  TomT0m / talk page 19:57, 22 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Properties for Community acquis (Q256414)-Items

The Publications Office of the European Union (Q480222) issues for every regulation of the European Union (Q240141), directive of the European Union (Q326124) and decision (Q732833) like European General Data Protection Regulation (Q1172506) or Regulation (EU, EURATOM) No. 883/2013 (Q63830649) a (nearly) unique number. The scheme has changed a few times in the history and is language specific. Are there properties for the numbering of EU secondary law? --Bcoh (talk) 09:37, 22 May 2019 (UTC)

@Bcoh: legal citation of this text (P1031) would seem to be a good approach for this - so European General Data Protection Regulation (Q1172506) would be legal citation of this text (P1031):"2016/679". Andrew Gray (talk) 18:43, 22 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Use of P973 as qualifier

described at URL (P973) is listed in allowed qualifiers constraint of archives at (P485). However, P973 has property scope constraint "as main value". Can P973 be used as a qualifier or only as main value?

I would not use described at URL (P973) as a qualifier - I would use plain old URL (P2699) or put a reference with reference URL (P854). - PKM (talk) 20:20, 22 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

IP banning prevents interwiki linking

Hello! Yesterday I had an interesting situation: I was doing a Wikipedia course in a school, and every students had its own account, so they could edit despite the IP being blocked in many projects. I searched a little bit about the IP and it seems that is a shared IP for lots of high schools, given by the government, so we can't blame a specific school or classroom. Students were uploading their articles when we noticed we couldn't add interwiki links, as the IP itself was banned in Wikidata. It seemed that there wasn't a global log-in happening in this case, so I had to add the links myself, what I could do.

I don't have a solution or a specific proposal, but I wanted to know if this issue is widespread or it is very punctual... and also if we can't prevent IPs from vandalizing Wikidata but, at the same time, allow adding interwikis. -Theklan (talk) 14:45, 22 May 2019 (UTC)

@Theklan: More granular blocking has been on the wishlist for a long time but I haven't heard of any recent progress on this... ArthurPSmith (talk) 15:38, 22 May 2019 (UTC)
welcome to the aggressive ip blocking (anyone can edit). even worse with account creation block, and blocks on one project affecting others. the admins don't think false positives are a problem, "because vandalism". even when more granular tools are available, admins do not see the need.[22] i recently saw this at a wikidata event, where a ip filter throttle prevented librarians from saving wikidata edits, because "new users at library ip." a culture change would be required, but until then editathons will continued to be impacted.Slowking4 (talk) 13:36, 24 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

How can we link between the Bain collection at Wiki Commons and the same image at Flickr Commons via Wikidata?

This is a data question, so posed here, instead of Commons, but is relevant to Wikidata. We migrated the entire Bain Collection from the Library of Congress to Wiki Commons and those images also appear in Flickr Commons with crowd sourced biographical information on the people pictured. Can anyone see a way that the identifiers we store here at Wikidata or at Wikimedia Commons can match any identifier at Flickr Commons so we can migrate the Flickr link to Wikidata automatically? Here is an example where I migrated one by hand: At Walter Simpson Dickey (Q7966031), the entry for the person in the image, I added the link to Flickr which has unique biographical information on the person. Can anyone see a way to automate the linkage? The Wikimedia Commons image uses an LCCN identifier (LCCN2014702600) File:Walter_S._Dickey_LCCN2014702600.tif and the Flickr Commons entry has an identifier: (ggbain.22649). You can see how I added the link from Flickr to Wikidata as an entry for "described at URL". Is it possible to automate the process? Is there anyplace where "LCCN2014702600" links to "ggbain.22649"? --RAN (talk) 15:04, 22 May 2019 (UTC)

  • I think it's a question for Commons: which properties to use to link from images to their source. Try c:Commons talk:Structured data. For the remainder, please see the answers to your previous question. It's above, you might want to read it and follow up on it. --- Jura 06:47, 23 May 2019 (UTC)
  • @Jura1: Structured data issues on Commons have been driven largely from the Wikidata side. Traditionally, Commmons' solution to this would simply be a mediawiki template that both puts the image in Commons:Category:George Grantham Bain Collection and has a parameter for the Flickr ID, and which automatically generates the URL on Flickr. - Jmabel (talk) 16:30, 23 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Request for merging to objects

Object Vetulani (Q7923845) should be merged with object Vetulani (Q21476612). Both refer to one family name. Francesco 13 (talk) 12:49, 23 May 2019 (UTC)

No, one is about the family name and the other is a disambiguation page. That's why it's tagged with family name has to use a different item than disambiguation pages (Q27924673). ChristianKl❫ 16:14, 23 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

No reference

Is there way to indicate that certain property needs reference? Eurohunter (talk) 16:18, 23 May 2019 (UTC)

Yes, there is. Use citation needed constraint (Q54554025). See sexual orientation (P91) for an example. Description of this constraint can be found at Help:Property constraints portal/Citation needed. --Shinnin (talk) 16:29, 23 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Incompatible types?

Sorry for the silly question, I was looking around a bit in the constraints and descriptions of some relevant items - is there a way to capture that an item shouldn't be instantiating two other items at the same time? I.e. to capture all people that are also planets, which would be likely an error, etc. Do we have a respective property? Cheers, --Denny (talk) 18:28, 23 May 2019 (UTC)

That would have to be a property constraint on instance of (P31) itself, but I don't think we would want to generally forbid multiple values (it's ok for something to be generally an instance of several different classes, that really does happen). Maybe the new ShEx support could be used to better specify what human (Q5) instances should look like, if that's what you're after? ArthurPSmith (talk) 19:07, 23 May 2019 (UTC)
the constraint for that is conflicts-with constraint (Q21502838), see Help:Property constraints portal/Conflicts with for details on using it. What you REALLY want to do, though, seems to be doing some queries instead. Circeus (talk) 19:10, 23 May 2019 (UTC)
What seems frequent is that P31 includes a few related types that would ideally be on separate items .. I think WikiProjects generally try to monitor this. --- Jura 08:39, 24 May 2019 (UTC)

Thanks for the pointers, that's super useful. Is there a way to make a property constraint trigger only if a specific value is given? I like and understand the examples on author, but I would like to say on P31, if this points to book (Q571) then P31 pointing to human (Q5) should be flagged. Does this make sense? --Denny (talk) 16:48, 24 May 2019 (UTC)

  • Currently one could only define: (e.g.) if an item has ISBN, it shouldn't be P31=Q5; if it has IMBD, it shouldn't be a book .. --- Jura 00:01, 25 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

preventing two properties from having the same value

is there a constraint that can prevent a property from having the same value as another property on the item? conflicts-with constraint (Q21502838) comes close, but you can't feed a "variable" in item of property constraint (P2305).

The base case use is preventing described at URL (P973) from being the same as official website (P856). Circeus (talk) 19:14, 23 May 2019 (UTC)

  • You could just query them once in a while and delete whatever shouldn't be there. --- Jura 10:04, 25 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Q477016 have new logo, but now you see the old logo. Please update. -- 18:30, 24 May 2019 (UTC)

en:file:Creative Technology company logo.svg is not available on Wikicommons. First the logo has to be uploaded on Wikicommons. --Kolja21 (talk) 18:56, 24 May 2019 (UTC)
Do you think we can use PD ineligible? --Trade (talk) 00:42, 25 May 2019 (UTC)
We could use PD-textlogo. Same licensing as commons:file:Creative Technology logo.svg. --Kolja21 (talk) 01:56, 25 May 2019 (UTC)
And don't replace the logo on Creative Technology (Q477016), but add it with preferred rank. Add start time (P580) to the new logo and end time (P582) to the old logo. Multichill (talk) 09:56, 25 May 2019 (UTC)
  Done. commons:file:Creative Technology company logo.svg.--Vulphere 06:25, 26 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)


On Arabic Wikipedia - and others - we use non-Gregorian date systems, and the linking of info from Wikidata has been disruptive to countless pages. How do we input and display non-Gregorian dates on here? If it's impossible, why's it been rolled out without consulting us?--عبد المؤمن (talk) 23:41, 24 May 2019 (UTC)

  • It's really up to arwiki to decide if they want to make use of dates from Wikidata or not. Supposedly conversion isn't much different than for MediaWiki in general. --- Jura 10:07, 25 May 2019 (UTC)
  • Wikidata does not support for language specific dates (including non-Gregorian dates), the bug is at phab:T63958. You could wrap the wikidata date with {{#time: xmj xmF xmY| insert wikidata date here }}.--Snaevar (talk) 20:38, 25 May 2019 (UTC)
This section was archived on a request by: Regards, ZI Jony (Talk) 10:01, 26 May 2019 (UTC)

Wikidata weekly summary #365

Adding a picture to a Wikidata profile

Hi, folks. A very "dumb" question, but how can I add a picture to a Wikidata profile? SirEdimon (talk) 21:00, 26 May 2019 (UTC)

The basic image property is image (P18) with a value that is a Commons image (without the "File:" prefix). There are several special kinds of images you can use as well - see London (Q84) for examples of maps, coat of arms image, nighttime image, etc. - PKM (talk) 22:38, 26 May 2019 (UTC)
What I mean is: look this profile: Q16735578. It has a image. I want to put a image on this profile: Q512265. There is a available image on Commons. So, how can I add and/or modify images on Wikidata profiles?--SirEdimon (talk) 23:06, 26 May 2019 (UTC)
I found it out. Thanks--SirEdimon (talk) 02:58, 27 May 2019 (UTC)
This section was archived on a request by: Vulphere 12:25, 27 May 2019 (UTC)

Incorrect usage of P248 instead of P143

Sometimes users are adding references to Wikipedia using property stated in (P248) instead of imported from Wikimedia project (P143). Right now query shows 5159 occurrences of this problem. Maybe it is possible to add none of constraint (Q52558054) restrictions with most popular wikis to stated in (P248)?

SELECT ?item ?itemLabel ?statement ?wpLabel {
  ?wp wdt:P31 wd:Q10876391 .
  ?item ?prop ?statement .
  ?statement prov:wasDerivedFrom ?refnode .
  ?refnode pr:P248 ?wp
  SERVICE wikibase:label { bd:serviceParam wikibase:language "en" }

Try it!Vort (talk) 10:25, 23 May 2019 (UTC)

It was suggested in this previous discussion that "the long term plan was to move it to 'stated in'". See also this more recent discussion. Scs (talk) 11:23, 23 May 2019 (UTC)
I am talking about opposite situation, where stated in (P248) is incorrect (instead of imported from Wikimedia project (P143)). Also often there is only single reference exists and just P248 → P143 replacement is needed. — Vort (talk) 11:38, 23 May 2019 (UTC)
Vort, if the very idea is that ultimately all imported from Wikimedia project (P143) ARE SUPPOSED TO BECOME stated in (P248) in the first place, how could there possibly be an "opposite situation"? Circeus (talk) 15:27, 23 May 2019 (UTC)
Nevertheless, there shouldn’t be references such as stated in (P248): English Wikipedia (Q328). I can offer to move (most of) them to imported from Wikimedia project (P143) as usual. Any objections? —MisterSynergy (talk) 16:10, 23 May 2019 (UTC)
I'm not sure I object (I don't know which way the wind is blowing), but it seems we've got a contradiction here, a lack of consensus. We've got people saying that stated in (P248) both should and should not be used for Wikipedia sources. We've got people saying that imported from Wikimedia project (P143) is both an obsolete placeholder, and still the right thing to use today. I don't know who to listen to. Scs (talk) 10:03, 25 May 2019 (UTC)
The move from P143 to stated in (P248) is only for non-Wikipedia sources, that is values like "English Wikipedia" should remain there.
P248 generally shouldn't be used for "English Wikipedia" etc.
In the meantime, I add the suggested none-of-contraint to stated in (P248). I used the most frequent values. --- Jura 10:28, 25 May 2019 (UTC)
(edit conflict) There seems to be some misunderstanding here. It was never suggested that all sources should be moved from imported from Wikimedia project (P143) to stated in (P248). In fact, the following is what makes P143 and P248 different from each other. P143 should be used with Wikipedia etc. and not with other sources, like GeoNames (Q830106), MusicBrainz (Q14005). That's what that previous discussion was about. And now Vort is making good point that there are also cases where P248 is used with sources that P143 is supposed to hold.
P143 isn't the right thing to use. It has been used bots and tools to indicate which Wikimedia wiki the information comes from. But it isn't considered a real source, unlike P248, and it shouldn't be used by humans. --Matěj Suchánek (talk) 10:35, 25 May 2019 (UTC)

All languages list

Is somewhere a list with all languages served by Wikidata (including languages like Austrain German which has no own Wikipedia)? Eurohunter (talk) 16:16, 23 May 2019 (UTC)

@Eurohunter: Here is a good proxy for this list, which hasn't been updated in over a week (@Pasleim: any idea what happened?). Mahir256 (talk) 18:37, 23 May 2019 (UTC)
@Mahir256: Thanks. Some of them are not "recognized by MediaWiki" in beta version. When they will be added? Eurohunter (talk) 18:52, 23 May 2019 (UTC)
@Eurohunter: See Special:APIHelp/query+wbcontentlanguages. --Lucas Werkmeister (WMDE) (talk) 14:35, 24 May 2019 (UTC)
@Mahir256: @Lucas Werkmeister (WMDE): any idea why PLbot is removing German (formal)? He is not removing for example Spanish (formal). Why there is formal version only of few languages? Why there is formal version of certain languages while we use here formal version instead of common language? Eurohunter (talk) 13:27, 25 May 2019 (UTC)

Gratulations re performance query

Hoi, the performance of query as seen in this display has improved importantly. It makes for improved usability of Wikidata particularly for those who edit. Thank you! GerardM (talk) 07:23, 25 May 2019 (UTC)


Why Special:MostInterwikis doesn't works? Eurohunter (talk) 13:30, 25 May 2019 (UTC)

The speciel page counts language links to other languages of the same type of project (i.a. links to other Wikipedias from a Wikipedia, links to other Wikivoyages from a Wikivoyage etc.). Because there is no different language versions of Wikidata, the count is always 0 here, and the special page is empty. --Dipsacus fullonum (talk) 15:51, 26 May 2019 (UTC)

Add commons sitelinks from P373

Is a bot tasked with adding commons sitelinks for items with Commons category (P373), e.g. special:permalink/943531568? (The opposite is being done slowly by DeltaBot.)--Roy17 (talk) 19:17, 26 May 2019 (UTC)

Pi bot is doing its best. Thanks. Mike Peel (talk) 20:13, 26 May 2019 (UTC)


Hans Bänziger (Q22114017) and Hans Bänziger (Q33663126)? --Magnus Manske (talk) 21:43, 30 May 2019 (UTC)

Bänziger the botanist wrote in Natural History Bulletin of the Siam Society (Q17623366), which is fully digitized. One of the articles lists as his affiliation "Department of Entomology, Faculty of Agriculture, Chiang Mai University", which is the same as what you'll find in this article by B^¨anziger the entomologist. Circeus (talk) 00:17, 31 May 2019 (UTC)
I think they are the same too, both appear at the same university, and only one VIAF entry. Most of his publications are for butterflies and moths and bees, and only one or two articles on plants that butterflies prefer. He is alive and we can write him. His most recent entries are on bees that drink tears, which were back in the news recently. --RAN (talk) 19:42, 1 June 2019 (UTC)
This section was archived on a request by: Liuxinyu970226 (talk) 21:38, 1 June 2019 (UTC)

I have data to contribute and would like help/advice please

Hi, I have statistical and factual data for a number of airports and related services (60+ airports and rising) that I would like to contribute here. Data includes (but not limited to), passenger statistics, on-time/delayed/cancelled stats, destinations, countries, airlines, local public transport timetables/destinations, public transport prices, types+numbers of services on-site at the airports (duty-free shopping, car rental providers, foreign exchange etc). All the metrics can be/are updated monthly (where applicable), and I can organise and format as required. I am not an experienced wiki editor by any means, so apologies for any stupid questions in advance. I would appreciate if anyone could help with the following: Which datasets would be appropriate for inclusion? Would I be able to submit the relevant data for 1 airport manually initially, and then progress to automating via the API for the rest? With regards to the data itself - what should I be aware of with respects to how the data is "credited" to it's source? (i.e. what must/must not be used/included?).

Many thanks in advance!Pete J Hall (talk) 08:35, 22 May 2019 (UTC)

@Pete J Hall: the first question is whether the data is provided under the right sort of license. Wikidata's main data space is CC0, so data that's provided by a government, public domain, or otherwise compatible is usually OK, but other data we may only legally be allowed to enter on a one-off sort of basis rather than systematically. The second question I would have is how much data are we talking about per item? You can take a look at John F. Kennedy International Airport (Q8685), for example the patronage (P3872) property which has been provided with one value per year; this sort of order of magnitude of data is ok here in Wikidata, but anything more detailed for a single item probably belongs as a data table in Wikimedia Commons, rather than here in Wikidata. If it looks like your data still can be placed here, then I would point you to Wikidata:Data donation if you haven't seen that already, for an outline of how to proceed and next steps. Good luck! ArthurPSmith (talk) 15:21, 22 May 2019 (UTC)

@ArthurPSmith thanks for the reply, although it's raised more questions - Your example of JKF airport - I am able to provide monthly passenger data, would/is it appropriate for me to add that in addition to the year records under patronage that are already there? If not, and as you suggest, I were to add it as a data table to wikimedia commons, where/how would that be done? Also, I see that some pages on wikipedia, for example Catania Airport [23] - there is no flight data on the page (in table format anyway), which does however exist according to the wikidata page Q540273 which also has a few multiple entries for the yearly data. If I were to add more passenger data to the wikidata page, and then use that data to embed a table in the wikipedia page (as opposed to the data already there) - is that permitted?

I assume I am going in the right direction here according to this: Wikidata:Data donation#How to add data to Wikidata but am a bit lost to be honest

Pete J Hall (talk) 17:55, 22 May 2019 (UTC)

@Pete J Hall: Yes, monthly data would likely be too detailed for Wikidata. Wikidata items don't behave well in the UI when there are thousands of statements. You would probably have to propose new properties to store airport data in Commons, unless we already have something suitable. An example of a tabular data property is tabular population (P4179), with an example file here:
@Bouzinac: you have been importing a lot of airport data - do you think you could give Pete J Hall a hand on this? − Pintoch (talk) 08:05, 23 May 2019 (UTC)
Hi there, yes, monthly data would be very welcome as SPARQL is able to mix and calculate the right annual sum (be it 12 months or one year). Months (YYYY-MM) may be interesting to compare seasonality between airports. And months data are compulsory because sometimes airports communicate on their fiscal year data (which does not coincide to year JAN-DEC)

See for instance the result for Lima airport [24] ; same phenomenon for Catania Fontanarossa/graph available in french/italian page, thanks to the wikidata, which helps data be translanguages. How to achieve this is : The simplest way to the complicated way :

If you still have difficulties, please send me the files and I'll openrefine-upload it onto Wikidata, provided there are labels or IATA/ICAO code. Bouzinac (talk) 08:40, 23 May 2019 (UTC)

@Bouzinac - Many thanks for the advice (and to the others for their comments also) Have installed openrefine and will see how I get on with that, have used similar in the past. Will be back if/when I get stuck Pete J Hall (talk) 15:37, 23 May 2019 (UTC) @Bouzinac - Ok, Am lost, have got monthly data for Catania airport in CSV format - how/where do I get it to you? I'm sure if I can see an import example from the data I have I can figure this out for the rest of the airports I have, I also have other related data that might be suitable that I can provide examples of also Pete J Hall (talk)

Where exactly are you stuck ? If you have monthly data, make sure, in Excel, you have 1 column with month (like "2019-05" : no day precision means it's a whole month value), 1 column with the passenger value. Then, you have to copy paste from Excel into your OpenRefine webpage. Then, you edit the scheme, and you search for Catania airport (or Q540273), and then you declare a new statement of P3872 and then with your mouse, you move your "passenger" value to the case, and you declare a date and you move the "month" value to this new case. And then you must justify your data with a source (reference URL (P854) or Wikimedia import URL (P4656). And then you hit upload to Wikidata. Bouzinac (talk) 08:39, 27 May 2019 (UTC)

@Bouzinac Hi again, I have attempted to follow your instructions, (I am confused by the references to "case") and managed to get as far as (I think) correctly adding the new statements etc, but then get issues highlighted with the following notification in OR "patronage (P3872) with invalid units Units such as Q5151 used on Catania-Fontanarossa Airport (Q540273) are invalid for patronage (P3872)." - Is this being caused by me? or does wikidata not handle this type of data? A series of screenshots or casts on how to do what I believe should be a simple process would be incredibly helpful if that is at all possible - either on here or elsewhere Pete J Hall (talk) 13:06, 27 May 2019 (UTC)

patronage (P3872) does not accept units, only values+date+refs. By case, I was mistaken as it's a french Template:Lang-fr/false friend (sorry) please understand with "case" by "box/field" in the openrefine form.

This form/schema should look somewhat like this

Example of schema into openrefine

Bouzinac (talk) 14:08, 27 May 2019 (UTC)

European Parliament candidates

We're in the middle of 2019 European Parliament election (Q16999180). Did anyone do some work to get the candidates documented? Multichill (talk) 09:53, 25 May 2019 (UTC)

@Multichill: yes, see for instance by Jacksonj04. − Pintoch (talk) 15:10, 27 May 2019 (UTC)
@Multichill: @Pintoch: There's a Listeria report of anybody with the appropriate candidate statements in the UK at User:Jacksonj04/2019_EU_Elections/Candidates which could easily be extended to the whole EU.

Wikidata weekly summary #366


Should I use publication date (P577) for the date a design was registered for trademark purposes, or is there a better date? - PKM (talk) 22:04, 27 May 2019 (UTC)

Immune response

Polish Wiki site "Odpowiedź odpornościowa" is equivalent to English "immune response"; therefore object immune response (Q25416241) should be merged into immune response (Q25554706). Francesco 13 (talk) 13:35, 30 May 2019 (UTC)

Q25416241 is current tagged as a disambiguation page. It should not be interwiki'd to a full article as long as it is, but I don't know enough Polish to tell whether it's just a stub or an actual disambiguation page. Circeus (talk) 15:44, 30 May 2019 (UTC)
Just   Merged, I don't see any reason within that Polish article to support me to believe that as "disambiguation page". --Liuxinyu970226 (talk) 21:37, 1 June 2019 (UTC)
This section was archived on a request by: Matěj Suchánek (talk) 19:04, 2 June 2019 (UTC)

Requesting item title change

Hi Wikidata regulars. While creating a new item, I mixed up two lines, so now the item I intended to be "Revúboé River" is now named "River" at [26]. Can someone with the proper rights move the item to "Revúboé River"? Sorry for the trouble.--SamHolt6 (talk) 03:05, 28 May 2019 (UTC)

@SamHolt6: I edited the item: no special rights are needed. Ghouston (talk) 03:34, 28 May 2019 (UTC)

WNeurodiversity and Canemousse usernames

I've realized I can't use either username on wikidata. Why is that? Am I under investigation or something? If you come to any conclusion whether any of these usernames are going to be allowed to enter or edit data in wikidata or not to be allowed to do so, please let me know. Canemousse (talk) 03:29, 28 May 2019 (UTC)

  • You seem to be using Canemousse on Wikidata right here as you made this comment. - Jmabel (talk) 03:44, 28 May 2019 (UTC)
Thank you for letting me edit. Now I'm gonna edit the item(with this account) I wanted to edit in the first place Canemousse (talk) 04:51, 28 May 2019 (UTC)
Now I realized that the reason I couldn't edit wasn't because I was barred from editing. The item was locked and that's fine. I'm happy to see that items are being locked on wikidata too as it might be necessary to prevent vandalism. ie. rose (Q102231) has instance of and meme as an example. Canemousse (talk) 05:00, 28 May 2019 (UTC)

Profession uniformity

There are multiple words, which have same meaning but different pages in some wikis like king, monarch or advocate, lawyer, etc. But in the profession section, only one of them should be used as value and rest should be converted to the said q item. Capankajsmilyo (talk) 09:04, 22 May 2019 (UTC)

@Capankajsmilyo: If they really have the same meaning the corresponding items should be merged (and any links will be automatically corrected to point to the merged item). Items in Wikidata generally represent only one concept, whether it is called by one or many different names. However your examples of "king" and "monarch" are indeed distinct ("monarch" is a more general term) and indeed here king (Q12097) is stated to be a subclass of monarch (Q116). If you are trying to do something with the query service, you can structure a SPARQL query so it looks for both instances of monarch (Q116) and any subclasses, so that really shouldn't be a problem. On the other hand perhaps there is some need to clean up the way the profession property is used here; I'm not sure there's any particular community group that has taken that on - there is Wikidata:WikiProject Q5 which is the closed wikiproject I could find here, but it doesn't seem very active. ArthurPSmith (talk) 15:36, 22 May 2019 (UTC)
Profession property does need some cleanup. How can I connect with those interested? Capankajsmilyo (talk) 05:13, 25 May 2019 (UTC)
@Capankajsmilyo: Probably best to start a discussion on Property talk:P106 (if that's the property you are interested in) and ping people who have previously commented on that page. ArthurPSmith (talk) 15:48, 28 May 2019 (UTC)

Want to know about creating a own blog about the biography

Kindly let me know the procedure of publishing biographies at Wikipedia.  – The preceding unsigned comment was added by Nazia kousar (talk • contribs).

Nazia kousar: This is Wikidata, not Wikipedia. You may want to read en:Wikipedia:Biographies of living persons. Esteban16 (talk) 11:22, 28 May 2019 (UTC)

Shape expressions are live on Wikidata

Hello all,

As announced here, we just released shape expressions on Wikidata. You can for example have a look at E10, the shape for human, or create a new EntitySchema.

A few useful links:

If you have any question or encounter issues, feel free to ping me. Cheers, Lea Lacroix (WMDE) (talk) 16:03, 28 May 2019 (UTC)

Property proposed: Shape Expression for class - Jheald (talk) 16:58, 28 May 2019 (UTC)

municipality as organization and geographical space

For WikiLovesMonuments organization it would be very usefull to distinguish a municipality as an organization and a municipality as geographical space. Property:P131 what means exactly? Thanks --Susanna Giaccai (talk) 08:07, 24 May 2019 (UTC)

According to its constraint, it's value should be a administrative territorial entity (Q56061), which seems to be a class of geographical area. Ghouston (talk) 08:53, 24 May 2019 (UTC)
OK, P131 is a geographical subdivision; but the Q15284 Municipality is not a geografhical subivision, it a administrative subdivision. So may i have 2 item?: Q2044 for Firenze as geographical space where are located some historical buildings like Palazzo Vecchio Q271928 and Firenze Q64007357 as the owner of Biblioteca delle Oblate Q3639664 and governor of municipality. --Susanna Giaccai (talk) 14:30, 24 May 2019 (UTC)
municipality (Q15284) has been defined as both a geographical area administrative territorial entity (Q56061), and a type of government organization municipal corporation (Q2097994), so it's a conflation between these two different concepts. It should be avoided if you want to be more precise in distinguishing these. I.e., if an item is for a government body, it could be an instance of municipal corporation (Q2097994) or a subclass of it, and if an item is for a geographical region, it can be an instance of administrative territorial entity (Q56061) or a subclass of it. The problem you've got is that Florence (Q2044) is an instance of comune of Italy (Q747074), which is a subclass of municipality (Q15284), which is saying that it's both a government body and a geographic area. Ghouston (talk) 05:44, 25 May 2019 (UTC)
Trying to split up something that is actually the same isn't going to work. I merged the duplicate. Multichill (talk) 10:01, 25 May 2019 (UTC)
The distinction is attempted elsewhere, e.g., City of Bristol (Q21693433) and Bristol City Council (Q16953796). The statements are bad there too. A government body, as an organisation, is hardly the same entity as the geographical area that it administrates. Or City of Westminster (Q179351) and Westminster City Council (Q7989145). Ghouston (talk) 10:20, 25 May 2019 (UTC)
I agree with @Ghouston: a governement body is absolutely not the same thing as the geographical area of a Municiality, so it is correct to have 2 different items. --Susanna Giaccai (talk) 14:10, 26 May 2019 (UTC)
One has population and area, the other has employees, a headquarters and can own things. The same thing occurs at higher levels of government, e.g., separate items for Italy (Q38) and Government of Italy (Q3773971). Ghouston (talk) 00:28, 27 May 2019 (UTC)
I think most of the time in Wikidata, these items are used to reference a geographical area. I'd suggest using that for existing "municipality" items, and create new items for the local governments themselves when required, e.g., in cases of employees or ownership, or when they have Wikipedia articles. Ghouston (talk) 00:33, 27 May 2019 (UTC)
In the case of Italy, that would mean changing comune of Italy (Q747074) so that it's purely a class of geographical entity, not a class of organization. Then if you were to try stating that an item was owned by Florence (Q2044), or was its employee, you'd probably get a constraint violation because it's not an organisation, which would give a hint to create a new item for its government. Ghouston (talk) 00:47, 27 May 2019 (UTC)
The metropolitan territory seems to be Metropolitan City of Florence (Q18288148) in this case. It can get confusing when you have several items for closely related concepts. Ghouston (talk) 00:51, 27 May 2019 (UTC)

@Ghouston: following your suggestions I created 3 adminstrative units: Comune di Firenze Q64066820, Comune di Sesto Fiorentino Q64072064 and Comune di Prato Q64070992 and two of there are part of città metropolitana di Firenze Q18288148 which is the administrative higher level created on 2015. I hope they are not deleted ;-). Thanks --Susanna Giaccai (talk) 13:20, 28 May 2019 (UTC)

I think these items are fine, I linked them with authority (P797)/applies to jurisdiction (P1001) to their areas. Ghouston (talk) 22:43, 28 May 2019 (UTC)

Nesesito ayuda urgente

Bueno, trabajo actualmente en un articulo llamado Anexo:Lista Roja de la UICN de especies extintas, aunque en realidad es un anexo.

Debido a que hubo una confusión con el nombre que originalmente tenia Anexo:Lista Roja de la UICN de especies extintas, es mas yo mismo se lo puse, fue una equivocacion de mi parte porque se creo un elemento "Q" totalmente diferente, osea propio de esta pagina y ahora que le cambie el nombre correctamente no lo puedo enlazar con la versión inglesa IUCN Red List of extinct species.

Para que entiendan mejor quiero eliminar el elemento "Q" que tiene actualmente Anexo:Lista Roja de la UICN de especies extintas.

Y fusionarla con el elemento "Q" de la version inglesa IUCN Red List of extinct species.

Erickespinal26 (talk) 20:46, 28 May 2019 (UTC)

Well, I am currently working on an article called Anexo:Lista Roja de la UICN de especies extintas, although in reality it is an annex.
Because there was a confusion with the name that originally had Anexo:Lista Roja de la UICN de especies extintas, it is more I put it myself, it was a mistake on my part because it created a totally different element "Q", that is proper of this page and now that I change the name correctly, I can not link it with the English version IUCN Red List of extinct species.
To better understand, I want to eliminate the element "Q" that currently has Anexo:Lista Roja de la UICN de especies extintas.
And merge it with the element "Q" of the English version IUCN Red List of extinct species.
Erickespinal26 (talk) 20:58, 28 May 2019 (UTC)
@Erickespinal26: See Help:Merge. - Jmabel (talk) 22:39, 28 May 2019 (UTC)

RFC on P973

FYI, a new RFC on described at URL (P973): Property talk:P973#Should URLs for bibliographies be used here?. Visite fortuitement prolongée (talk) 21:40, 28 May 2019 (UTC)

Languages in which page is not there

In the Wikidata UI, links for languages in which page is available is shown. However, it would be great if we can see a list of languages in which such page is not available. Capankajsmilyo (talk) 17:36, 29 May 2019 (UTC)

  • Even restricted to languages that have Wikipedias, that would be hundreds of languages for most items. - Jmabel (talk) 21:42, 29 May 2019 (UTC)

A wikidata solution to a table of wikiprojects?

Hello all, Below is a table that appears over at the WikiClassics User Group to document the relevant WikiProjects within a field. Something similar is planned for a STEM Wiki User Group.

It feels as though there should be a wikidata solution to generating such a table so that it spots new projects that start in other languages. At a minimum something is surely possible by supplying the row names, but possibly there's a way to tag the relevant wikiprojects (e.g. Q6723002) as part of the user group? Any implementation ideas? T.Shafee(evo&evo) (talk) 02:13, 29 May 2019 (UTC)

  Main Projects
  ar ca da de el en es fa fi fr hu it ja ka ko lt mk no pl pt ro ru sv tr zh data
Antiquity or Ancient history W W W W W W
Ancient Greece W W W W W W W
Ancient Rome W W W W W W
Roman Republic W
Roman Empire W
Greek mythology W W W W W W
Roman mythology W W
Latin Language W W W

There is {{#invoke:LanguageMatrix|Top20|Q1|Q2|Q3|Q4|Q5}} that could be adapted (see samples below). Initially done for lists like m:Wikimedia_CEE_Spring_2016/Structure/Esperantujo as {{#invoke:WikimediaCEETable|table|Q1|Q2|Q3|Q4|Q5}} --- Jura 06:46, 29 May 2019 (UTC)

Sample format 1
Article en zh hi es fr ar ru bn pt id ur de ja sw pnb jv wuu te tr ko sv Σ Wikidata st.
1 WikiProject Ancient History W W W 3 Q6301624 2
2 WikiProject Ancient Greece W W 2 Q6497946 2
3 WikiProject Ancient Rome W W W 3 Q6337458 2
4 WikiProject Roman Republic W 1 Q58207489 2
5 WikiProject Roman Empire W 1 Q15885059 2
6 WikiProject Mythology/Greek mythology W W W W 4 Q10859476 2
7 WikiProject Roman mythology W 1 Q18437016 2
8 WikiProject Latin W 1 Q15298323 1
Σ → Autolist 2 1 0 5 5 0 0 0 0 0 0 2 1 0 0 0 0 0 0 0 0 16 avg: 2\1\10% 15  

Sample format 2
Article   az   ba   be  
  bs   bg   crh   de   el
  et   hr   hsb   hu   hy   ka   kk   lt   lv   mk   myv   pl   ro   ru
  sq   sr   tr   tt   uk Σ
Wikidata st.
1 WikiProject Ancient History W W 2 Q6301624 2
2 WikiProject Ancient Greece W 1 Q6497946 2
3 WikiProject Ancient Rome W 1 Q6337458 2
4 WikiProject Roman Republic 0 Q58207489 2
5 WikiProject Roman Empire 0 Q15885059 2
6 WikiProject Mythology/Greek mythology W W 2 Q10859476 2
7 WikiProject Roman mythology W 1 Q18437016 2
8 WikiProject Latin 0 Q15298323 1
Σ → Autolist 0 0 0 0 0 0 0 2 1 0 0 0 0 1 0 0 0 1 0 0 0 2 0 0 0 0 0 0 0 0 7 avg: 1\0\3% 15  

Sample format 3
Article Nordic languages Neighboring languages Σ
Wikidata st.
  sv   fi   no   da   nn   is   fo   se   kl   en   de   ru   fr   pl   nl   lt   et   lv
1 WikiProject Ancient History W W W W 4 Q6301624 2
2 WikiProject Ancient Greece W W 2 Q6497946 2
3 WikiProject Ancient Rome W W W 3 Q6337458 2
4 WikiProject Roman Republic 0 Q58207489 2
5 WikiProject Roman Empire 0 Q15885059 2
6 WikiProject Mythology/Greek mythology W W W W 4 Q10859476 2
7 WikiProject Roman mythology W W 2 Q18437016 2
8 WikiProject Latin W 1 Q15298323 1
Σ → Autolist 0 1 1 1 0 0 0 0 0 2 2 0 5 2 1 1 0 0 16 avg: 2\1\11% 15  

@Jura1: Wikidata is magic. Now implemented at meta:STEM_Wiki_User_Group#WikiProjects. There are some formatting changes that I'd like to make (e.g. centering the '+' icons, omitting '-' icons as unnecessary. Maybe linking the "WikiProject Chemistry" to the wikidata item to omit the need for a separate "wikidata" column, and make optional the "commons" column, "statements", and flags etc. Having a return character before the pipe also seems to break the wikidata column. I've no LUA experience. Is it possible to add |parameters= in the same way? I'll have a go experimenting and try not to break anything. T.Shafee(evo&evo) (talk) 07:13, 29 May 2019 (UTC)
I'm not really a LUA expert either. I added the first change. It's a wiki, so you shouldn't really be able to break it by adding more parameters. Alternatively, you can copy it to a different module name. --- Jura 08:39, 29 May 2019 (UTC)
Running Wikidata queries from Lua is an interesting alternative to Listeria. I assume the module output is cached, so that the query doesn't need to be run every time the page is viewed? If so, how often would be the cache be updated, and could the page load much slower in these instances? Ghouston (talk) 03:11, 30 May 2019 (UTC)

Proposed, under construction, operational, etc.

I'm looking to add a property to a wind farm, to indicate that it is under construction.

I've found many items that could possibly relate to this:

But I was hoping there would be a property which could apply to any building, structure or otherwise, which refers to the "operational status" of something. e.g. "proposed", "under construction", "operational", "unfinished", "abandoned", "demolished" etc. Then we would only need the basic items of "building", "bridge", "wind farm", "structure", etc. and would have more flexibility that items which are basically overlaps of these properties, e.g. "wind farm under construction" "proposed tunnel" "demolished hospital" and so on.

I figure this is a basic idea which has been refused for some reason in the past, but looking through the project chat archives I can't find where this has been discussed. I've found where people have suggested using "significant dates" with "construction started", "construction finished", which is reasonable when the timeline is known - since the data doesn't get dated. But sometimes people don't know these dates or they are in the future. An example, in 2015 the item Garden Bridge (Q14629010) had "date of official opening" added as "2018". It appears an opening date in the future was used to indicate a non-open/non-operational structure - but this date lapsed and now the data is incorrect.

Any thoughts on the usefulness of such a property? Or am I totally on the wrong track? -- Chuq (talk) 04:43, 29 May 2019 (UTC)

I don't think such a property exists, and I agree it's a reasonable one. Best way to know what people think would be to head over to Wikidata:Property proposal and ask. Worst come to worst, they will tell you what property you actually do want.
P.S. Your list could be completed (at least as far as physical infrastructure goes) with destroyed building or structure (Q19860854) and abandoned (Q64030927). Circeus (talk) 05:29, 29 May 2019 (UTC)
Don't limit it to physical structures. Nakhon Suvarnabhumi Province (Q840866) is about a province which was proposed to be created, but which never materialized - but as I so far had no idea yet how to model this fact, that item (and a few similar one) have no instance of (P31) at all. Ahoerstemeier (talk) 08:40, 29 May 2019 (UTC)
state of use (P5817) and state of conservation (P5816) Ayack (talk) 08:41, 29 May 2019 (UTC)

Hello Chuq. Not sure if this helps, but have a look at en:Template:Infobox power station#Parameters. We managed to derive the status based on other parameters (such as construction date, etc). For dates which are unknown, setting the value as "unknown value" also works. Cheers, Rehman 10:37, 29 May 2019 (UTC)

The best solution is to use significant event (P793). Please look at the talk's page of the property (Property_talk:P793#How_do_we_use_this_property?) to see the parameters to use to define the different phases of a construction. Snipre (talk) 11:55, 29 May 2019 (UTC)
Hello Snipre. Please correct me if I'm wrong, but that property is used to only store dates of significant events right? It does not have a way of storing the current status of the entity? For example, on an item relating to Building XYZ, how do we state that the building is currently under construction? Note that the building had a proposed date, a topped-out date (that could be passed, even though the building is still under construction), and also a planned completion date. Rehman 02:49, 30 May 2019 (UTC)
Circeus (talk) 15:33, 30 May 2019 (UTC)

Swedish Church parishes and the church archive

We have done a project for Swedish Church parishes in Wikidata see Github and as Church archives are free available in Sweden we have also the Archive ID in Swedish National Archive ID (P5324)

Question: as a Churcharchive is not a parish but an archive with churchbooks gathered over some 100 years of people living in the parish how is this described in Wikidata?

As we also try to work tight with the National Archives in Sweden Linked data project TORA "Topographical register at the Swedish National Archives" it would be excellent to have a relation that is also described in other ontologies like DCTERMS SKOS.... - Salgo60 (talk) 10:43, 30 May 2019 (UTC)

I... don't see any problems whatsoever? I mean, there's no need for a solution in so far as nothing here conflicts with anything on the Swedish National Archive ID (P5324) property page... Circeus (talk) 15:47, 30 May 2019 (UTC)

problem with syntax of specific permalink


I would like to propose a property for Albertina (Q371908) museum ID, but the permalink to paintings includes [ ], which totally prevents to paste them as url on paintings items - see c:File_talk:Paul_Cézanne_-_Farm_in_Normandy,_c._1885-86_-_Google_Art_Project.jpg#reversed_picture_? . how could this be solved, to add IDs on wikidata, and link to the online collection ? --Hsarrazin (talk) 16:07, 30 May 2019 (UTC)

@Hsarrazin: property for this museum doesn't seem to add a lot. What they call a "permalink" seems to be just a link to their search engine, not sure how permanent that is going to be!
You can just urlencode the brackets so you get something like . That's what I just added to The Banks of the Loing at Moret (Q18683089). Multichill (talk) 16:50, 30 May 2019 (UTC)
thanks Multichill
whether the permalink is a link to search engine or directly to work, it is usefully giving info about the work :)
--Hsarrazin (talk) 16:54, 30 May 2019 (UTC)
In this case it doesn't really adds anything compared to described at URL (P973). It does need some more links, see Wikidata:WikiProject sum of all paintings/Collection/Albertina. Multichill (talk) 17:00, 30 May 2019 (UTC)
Also, looks like inventory number (P217) is available to fit this "id". Circeus (talk) 17:19, 30 May 2019 (UTC)
Sorry, I thought that was obvious, so I didn't mention it, but the "Inventarnummer" in the url means inventory number. So is a query work the work that has inventory number GE120DL. It's quite common for museums to use the inventory number in the url. You can play around with it, for exampleälde%5D will give you all the paintings on their website ("Objektbezeichnung" is object type and "Gemälde" means paintings). Multichill (talk) 20:08, 30 May 2019 (UTC)


Hoi, I am about to add a preprint for a scholarly article. I am no confident that the current use of "official website" with "preprint" as a qualifier is ok. The published paper is not necessarily the same. In my opinion a property "preprint available at" is to be preferred. Thanks, GerardM (talk) 20:02, 30 May 2019 (UTC)

Why would full work available at URL (P953) not be appropriate? That would be my first go-to. Circeus (talk) 00:09, 31 May 2019 (UTC)
Because a preprint is not the same thing. When someone makes the published version available, it is not necessarily the same as the preprint. GerardM (talk) 01:17, 31 May 2019 (UTC)
Why? I mean, I'm assuming what you're (sensibly, I'd think) linking to is the preprint you are referencing, with the preprint and final articles being two distinct items. Circeus (talk) 02:48, 31 May 2019 (UTC)

More than one SNAC Ark ID (P3430) for Guthrie Cortland Medical Center (Q64160636)

This item has more than one SNAC Ark ID (P3430), and not incorrectly. Yes, sometimes SNAC has duplicates, but this isn't the case. I was informed by them that when an organization has more than one name in its history, they will typically link the two in their archive but will not merge them, thereby leaving the single organization with more than one ID. The organization above remained the same but changed names. This must have come up before. Given that it will shoot off a single value constraint warning, is it okay to just leave it like this? Thanks, Hazmat2 (talk) 22:41, 30 May 2019 (UTC)

You could add an exception to constraint on the property page, or some time tag (i.e. valid in period (P1264)) to distinguish the IDs. Circeus (talk) 00:20, 31 May 2019 (UTC)
It looks like the SNAC id is still valid, so valid in period (P1264)) doesn't apply. How about applies to name of item (P5168), and adding this as a separator on the constraint? Ghouston (talk) 01:59, 31 May 2019 (UTC)
The only exception listed is for Doris Day, at [27] and [28]. That could also be a multiple name situation, since the items link to different VIAF ids under different names, but it's not very clear. Actors with stage names and women with maiden and married names are pretty common. Ghouston (talk) 02:08, 31 May 2019 (UTC)
I think the applies to name of item (P5168) idea makes the most sense at this point. I can add the exception, but at some point these may pile up if SNAC does what they say they do. Thank you, Hazmat2 (talk) 11:52, 31 May 2019 (UTC)


Hi, can bots create items? If you look into my edits there’s a series of repeated item creations that can easily be handled by bots. It’s too annoying to do it one by one.--Épine (talk) 08:27, 31 May 2019 (UTC)

I believe they surely can but you could alternatively also use Quickstatements tool (see Help:QuickStatements). Kpjas (talk) 09:05, 31 May 2019 (UTC)
@Kpjas: seems a bit complicated for me. Where can I place bot requests?--Épine (talk) 10:54, 31 May 2019 (UTC)
Wikidata:Bot requests or you can also create items using Petscan Kpjas (talk) 11:00, 31 May 2019 (UTC)

Internet Broadway Database show ID

Property:P1219 Last week I mentioned that the url formatter had to be changed for IBDB people. I forgot to mention that it needs to be done for the shows too. Thank you to the person that made the change last week. --RAN (talk) 15:49, 29 May 2019 (UTC)

Fixed, thanks! --RAN (talk) 21:47, 31 May 2019 (UTC)

family name qualifiers

Could someone please help fix Anson Chan (Q50838)? The info is correct but I dont know what qualifiers to use. Thanks in advance!--Roy17 (talk) 20:30, 31 May 2019 (UTC)

@Roy17: fixed it, but I had to use a kludge for her Chinese given name, as no item exists corresponding to it currently exists (the linked one, Anson (Q27179450), is an English male name that happens to be transcribed with the same hanzi in Chinese). Circeus (talk) 23:49, 31 May 2019 (UTC)


commons:Category:Montages is linked from montage (Q239298) but that seems wrong to me. Q239298 appears to be about montage as used in film or video, using rapid cuts among varied shots. commons:Category:Montages is about a still image composed of multiple smaller images. I don't see any appropriate item to which to attach that, but this one seems wrong. {{ping}Sadads}} you made this connection, do you disagree? - Jmabel (talk) 22:39, 31 May 2019 (UTC)

The correct wikidata item is really photomontage (Q828107). It's just easy to overlook because a typical photomontage is not as, erm, crude as the sort of mosaic montage involved here. Circeus (talk) 23:55, 31 May 2019 (UTC)
Not quite. photomontage (Q828107) is about creating a single image, not combining multiple images in a grid as found in the Commons category c:Category:Montages. It's already correctly linked to a different Commons category, c:Category:Photomontages. There's a Wikidata property collage image (P2716) for these grid montages, but there doesn't seem to be an item for it. Ghouston (talk) 02:34, 1 June 2019 (UTC)
I concur with Ghouston on this being distinct from photomontage (Q828107). Also, doesn't have to be a grid. Consider c:File:Seattle Schools - 1900.jpg; this style was very popular circa 1900. - Jmabel (talk) 05:06, 1 June 2019 (UTC)
I still think it's a subcategory of photomontage, it's just not a technically advanced one (or, in a way, one that is frequently in need of being referred to). A new item seems appropriate. 10:05, 1 June 2019 (UTC)
Commons has it the other way around, Photomontages‎ a subcategory of Montages. Well, there can be montages/collages of non-photographic images, like c:File:Art-portrait-collage 2.jpg. But then Commons also has c:Category:Collages as a subcategory of Montages, with subcategory c:Category:Photocollages linked to Photocollages (Q15810027). Who knows. Ghouston (talk) 10:44, 1 June 2019 (UTC)
I removed the bad Wikidata link from c:Category:Montages, in any case. Ghouston (talk) 01:25, 2 June 2019 (UTC)

Farmers and agriculturers?

I'm trying to make sense of some changes from 2018 on farmer (Q131512) - I think someone tried to change this into two different meanings by renaming the existing item and creating a new one for the old sense, but left all the inbound statements in place, so it's all a bit messed up - we now have 16000 statements pointing to the new meaning that were mostly created with the old meaning. See notes on Talk:Q131512. I can't quite work out what the difference between farmer (Q131512) [was "farmer"] and farmer (Q12323650) is - it's not clear in English but might be more obvious in another language.

Any thoughts? If need be I can just revert the items to where they were in July 2018 and work out changes from there, but I'd prefer not to do that unless necessary. Andrew Gray (talk) 21:05, 29 May 2019 (UTC)

(@Blackcat, Herrinsa:, who were involved in the changes, and @Esquilo: who merged the split item into Q12323650 and might have some thoughts. Andrew Gray (talk) 21:07, 29 May 2019 (UTC))
@Andrew Gray:, maybe I can help. In Italian we have two different terms, "Fattore" and "Agricoltore", which can be both translated in English as "Farmer". The problem is, that we consider "Agricoltore" someone generically involved in farming (or agriculture), whereas "Fattore" as bonus is also owner or manager of a farm (not always, but the term is used mainly for such figure). Hence the ambiguity, I suppose. -- Blackcat (talk) 21:26, 29 May 2019 (UTC)
I see that farmer (Q12323650) was created for da:Landmand (one who works in agriculture) i 2013. farmer (Q131512) is linked to da:Bonde (one who works in agriculture, and owns or leases the farm where they work and also live) --Dipsacus fullonum (talk) 07:34, 30 May 2019 (UTC)
We also have farmworker (Q5060555) and agricultural worker (Q19261760). Ghouston (talk) 07:36, 30 May 2019 (UTC)
The article sv:Bonde says the terms are synonymous. However, if the two terms are distinguished by farmer (Q12323650) actually owning or renting a farm or land and farmer (Q131512) only works there for salary (even though that description fitts better to farmworker (Q5060555)), then the swedish description and label for the two items should be swapped. In Swedish the main distinction is between "jordbruk" (which means growing crops and cattleraising) and "lantbruk" which also includes forestry. /ℇsquilo 07:54, 30 May 2019 (UTC)
At least in the U.S., "farmworker" tends to suggest a manual laborer, and often itinerant/migrant/seasonal; "agriculturalist" tends to suggest at least some formal education in agriculture at a post-secondary level. - Jmabel (talk) 15:17, 30 May 2019 (UTC)
I consulted a couple of Danish speakers (including our own @Fnielsen:) to try and work out what was happening with those two pages. If I understood right, it seems they are mostly synonymous, and the other Nordic languages have effectively merged them, but in Danish "bonde" has more of an old-fashioned tone to it, while you'd mostly use "landmand" for a modern farmer. (Note that the "bonde" article has a Middle Ages navbox, giving it a historical context). Andrew Gray (talk) 20:03, 30 May 2019 (UTC)
After thinking about it for a while, I think what we're going to end up needing to do is:
  • have a parent item to describe "farmer" in its broadest possible sense ("a person who does agriculture") without any specific details as to whether they're a farm-owner, family farmer, tenant, labourer, etc, or what type of farming it is (crops, animals...). We should probably use farmer (Q131512) for this as it has all of the P106 links pointing to it.
  • put all the generic sitelinks onto that item and ensure it has the label used by the sitelink in each language
  • set all the different "specific kinds of farmer" as subclasses (there are about 45 at the moment), creating any new ones as needed - there seem to be quite a few which are language specific.
  • move any specific sitelinks onto those subclass items
Does this sound reasonable? I'm happy to do the legwork for this over the weekend and come back to you all for confirmation it looks right. It's a bit of a mess but I think we can fix this Andrew Gray (talk) 20:03, 30 May 2019 (UTC)
  • Certainly something like that. There are a lot of subtleties here, and we'll definitely need clear descriptions of each item in as many languages as possible if people are going to get it right, though. For example, even just in English: a "farmer" can be anything from a "gentleman farmer" like George Washington to a "tenant farmer" including even a sharecropper, but it doesn't include a farm laborer who works on someone else's farm without having a plot of his or her own. And someone who herds cattle on the Western range in the U.S. in the 1900s (or a gaucho in Argentina) would certainly never be called a "farmer", though someone who raises dairy cows would, and even someone who raises beef cattle in more confined circumstances might be. Items don't need to line up with words in English or any other particular language, but it's always harder to explain concepts that don't. - Jmabel (talk) 20:17, 30 May 2019 (UTC)
Definitely agree. At the moment, we seem to be having things fall into a few natural categories:
  • subclasses of herder (Q12059906), including a few groups that are more "draft/pack animal drivers"
  • plus subclasses of rancher (Q1524582), which has been repurposed from "rancher" in a similar way to "farmer" and may also need fixing up
So it looks like we've got three emerging groups, one focused around crops, one around animals, and one around "peasantry". A very generic "person who practices any kind of farming" could sit on top of that and be our fallback for "is a farmer but I don't know any more details", and then we can migrate people down to subcategories as appropriate. I'll keep chipping away at this. Andrew Gray (talk) 21:39, 1 June 2019 (UTC)
Regardless of how the subtypes are currently set up, "agriculturalist" would be a sufficiently broad term in English to cover both crops and livestock, and farmer (Q131512) is linked to en:Farmer which also covers both. Ghouston (talk) 01:09, 3 June 2019 (UTC)

A little help

I was looking this animal Sambar deer (Q229337), and I notice another entry for the same animal Sambar deer (Q50139104).

If we search for Cervus unicolor, the best entry would be Q229337, not the Q50139104, that do not have any information.

How can we improve this, and make readers go to the best entry?

Rodrigo Tetsuo Argenton (talk) 04:19, 25 May 2019 (UTC)

You've struck at the heart of a Wikidata dilemma. One of those items is really about the name Cervus unicolor, which was the original combination for the species that represents the Sambar Deer, which is now know as Rusa unicolor. The correct property to connect them (although it makes no sense to connect two "instance of:taxon" with it, bad decisions with regards to constraints have lead us to that point) would be original combination (P1403), not taxon synonym (P1420). Circeus (talk) 15:30, 25 May 2019 (UTC)
See previous discussion, for example, at Wikidata:Project chat/Archive/2018/08#What heart rate does your name have?. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:13, 27 May 2019 (UTC)

I am sure this is the kind of stupid question that relative newcomers ask, but an hour of searching down the rabbit hole of project chats, property helper pages etc got me no further to an answer, and just made my head hurt. Why then is Q50139104 still considered an instance of a taxon and not something else ? Or why is there not a qualifier that can be added to the taxon name (P225) that can indicate it is no longer an accepted combination ?  – The preceding unsigned comment was added by CanadianCodhead (talk • contribs) at 31 May 2019‎ 19:04 (UTC) (UTC).

@CanadianCodhead: Currently the system, as imperfect as it is, goes as follow:
I think that covers everything. I personally loathe this system, but it is the one we're currently working with on Wikidata. We do not currently have a "this name is not in use" property (i.e. the opposite of taxon synonym (P1420)), but hopefully we will once this proposal is acted upon. Circeus (talk) 23:40, 31 May 2019 (UTC)

Just working with some data, I found this, which has an entry under the Instance of Q1040689 called Synonym of. I'm unclear if it is intended for this. Of course it also raises the issue of circular data, if one resource says A is the accepted name and B is the synonym, and a 2nd resource says the opposite. In this case I'm interpreting the data to mean they are trying to say Halimocnemis pilifera is the accepted name, and Halotis the synonym, when many other sources reverse it.

Sadly, wikidata doesn't handle splits that are about 50-50 well at all. But then, neither do any taxonomical database, as they are also required, ultimately, to pick a side. said to be the same as (P460) is probably the only solution in the interval. Circeus (talk) 02:11, 4 June 2019 (UTC)


I'm new to Wikidata and would appreciate guidance. I'd like to contribute data about women mayors in the U.S. state of Oregon. So far I am creating an item for the person if needed. On the item for the city, I'm adding head of government (P6) with the Q number for the person and any known qualifiers. On the item for the person, I'm also adding position held (P39) and here is my question. I started using mayor (Q30185) as the value, but then noticed that for many cities (larger ones especially) there is a specific item for the office of mayor of that city. The one example I've found for a city in Oregon is Mayor of Portland (Q20641416). My question is: should I always create an item for the office of mayor of a city (Eugene, Springfield, Salem, etc. -- all relatively small cities)? When would it be appropriate to use mayor (Q30185) as the value of position held (P39)? Thanks for any guidance. -- Lrobare (talk) 22:45, 30 May 2019 (UTC)

I have been adding the "Mayor of X" entries, I started at Wikipedia where they are slowly getting deleted, but I continue them here. My take on it is, if I have a large list I create the entry position:Mayor of X. But if it is a one-off situation for a current mayor, I just use the way you described it, position:mayor. The advantage of position:Mayor of X is that I can click on "what links here" and see a list of all the mayors from the location. --RAN (talk) 00:09, 31 May 2019 (UTC)
There's always the option of making it with a subproperty: mayor (Q30185) of (P642) Salem (Q43919). That's what of (P642) is for. Circeus (talk) 00:23, 31 May 2019 (UTC)
But we really don't want to have two or three, different, optional ways of doing this (or anything), right?
We don't want people writing queries to have to take two or three options into account (or to suffer wrong answers if they miss an option), right? —Scs (talk) 02:22, 31 May 2019 (UTC)
A single method is ideal. But this applies to any kind of position, not just mayors. Should you always need to create an item for the position before you can assign somebody to it, even for obscure positions with no other notable occupants? Some positions automatically have items because they have Wikipedia articles. Ghouston (talk) 02:34, 31 May 2019 (UTC)
An example is Maimunah Mohd Sharif (Q47526032), with two position held (P39) statements, both done using of (P642) to indicate the organisation. She was apparently the mayor of a restaurant. Ghouston (talk) 02:44, 31 May 2019 (UTC)
Sort of like Dave Van Ronk (Q1173376) was Mayor of Macdougall Street? - Jmabel (talk) 03:50, 31 May 2019 (UTC)
That would be a nickname, this one looks like a strange mistake. She was apparently the mayor of George Town (Q61092). Ghouston (talk) 03:56, 31 May 2019 (UTC)
I found an item for her position: Mayor of Penang Island (Q55622550). So now there's a mixture of methods for her positions. Ghouston (talk) 04:12, 31 May 2019 (UTC)
I think using an item for a position is logically better. A statement like Maimunah Mohd Sharif (Q47526032) position held (P39) executive director (Q267936) seems dubious, because executive director (Q267936) is really a class of positions, not a position itself. Such positions may vary greatly depending on the organization, and the organization is only mentioned in a qualifier (which also makes it a little harder to write queries). Ghouston (talk) 04:57, 31 May 2019 (UTC)
I also have been making "subclass of:mayor of a place in New York" and "subclass of:mayor of a place in New Jersey" so that I can see if I have all the cities by clicking "what links here". --RAN (talk) 20:01, 1 June 2019 (UTC)
These seem redundant, since the location is already implied by the location item. We don't have items like "place in New York state" after all? Ghouston (talk) 01:32, 2 June 2019 (UTC)
Yeah, that really doesn't make sense. Use a SPARQL query. --Yair rand (talk) 23:31, 2 June 2019 (UTC)
Then why do we list Chicago as instance of:city of the United States? That can be derived from country:United States. There are many examples of these combined fields. --RAN (talk) 01:51, 5 June 2019 (UTC)
I'm not really sure about city of the United States (Q1093829). A "city" in the US can be some kind of official designation, but the definition varies per state. There are also officially designated cities in the UK and Australia, among other places, so perhaps the "officially designated city" item should be more inclusive than US-only. Maybe also to city-equivalent terms in other languages. The item city of the United States (Q1093829) does need to exist of course because of Wikipedia articles. Ghouston (talk) 02:18, 5 June 2019 (UTC)

given names

How does wikidata plan to support say Chinese, Japanese or Burmese given names with given name (P735)? Or if we should just stick to name in native language (P1559)?--Roy17 (talk) 19:26, 24 May 2019 (UTC)

  • There are few items for Japanese names. If you are interested in formulating a plan, please do. --- Jura 10:03, 25 May 2019 (UTC)
Either create millions of given name items, most of which would only be used for a handful of times, or stop using item as its datatype which should have happened right at the start: Wikidata:Requests for comment/Personal names Property_talk:P735#Mess. Problems raised in its proposal and vote are still yet to be addressed.--Roy17 (talk) 15:36, 26 May 2019 (UTC)
Are yall sure, that your suggestion is supported by a community consensus? Right now there are less than 50k items that are instances of or subclass of given names. Only seven are Chinese given names. If we were really creating Chinese given name items, there will be millions more. Not to mention some people have assumed multiple names at different stages of life, e.g. en:Chiang_Kai-shek#Names. Most items (maybe 30-50%) would only be used for one person. On the other hand, two other kinds of Chinese names, courtesy name (P1782) and art-name (P1787) have data types of monolingual text: Wikidata:Property_proposal/Archive/32#East-Asian_names.--Roy17 (talk) 01:48, 30 May 2019 (UTC)
Fairly confident, this has been heavily debated over many years. It is also not at all problematic or unusual for an item to be only used in one location. Note that unless we specifically import a dataset of Chinese names, it is very unlikely that there would be millions of Chinese names on Wikidata, as there appear to only be ~260,000 items for Chinese people on Wikidata at all right now. --Yair rand (talk) 03:49, 30 May 2019 (UTC)
  • Coverage for family names is better. It seems that there is just no interest in Chinese given names. --- Jura 07:56, 30 May 2019 (UTC)
East Asian names have extra complications (homonymns separated by hanzi spelling), and we have relatively few Wikidata users literate in entering hanzi and/or chinese/japanese in general. Circeus (talk) 15:14, 30 May 2019 (UTC)
@Yair rand: I ran a sample query. There are ~95k biography articles under zh:Category:漢字姓氏. Eliminating idential given names returns ~60k. This means as much as 60% has a unique given name. Of course, as more people are added, many will share a given name so the rate will go down, so my estimate is about 40-50% of all items of Chinese people would have a unique given name. Jura1 ran an analysis on Sep 2015 and concluded ~70% of all items have a given name. So, 2~3 million items of Chinese people = 1 million Chinese given names items, if this community is willing to host such items. Is that a distant figure? Considering wikidata imports all sorts of researchers, non-notable mentions in databases, etc. that may not even have a wiki article, the answer is no.
That was just the problem of unique Chinese given names.
I do not speak Japanese, but I have heard Japanese given names are often difficult to pronounce for strangers. Quoted from en:Japanese name: given names are much more diverse in pronunciation and character usage. While many common names can easily be spelled or pronounced, many parents choose names with unusual characters or pronunciations, and such names cannot in general be spelled or pronounced unless both the spelling and pronunciation are given. Unusual pronunciations have especially become common, with this trend having increased significantly since the 1990s. So, how is wikidata going to cope with this? Do you create items based on kanjis? Transliteration and labels are gonna be problematic. So based on kanas? But many people are not named using kanas per se. I don't know whether Japanese people consider all the variations of for example en:Toshio the same given name, or different given names. (TLDR: in Japanese, the same kanjis may be pronounced differently; the same pronunciation may be written with different kanjis.)
ja:special:search/intitle:秀夫 is a pretty popular Japanese given name, but it is also the name of en:Lu Xiufu. 俊雄 is quite popular in Japanese and Chinese. Should two items be created for these cases?--Roy17 (talk) 20:30, 31 May 2019 (UTC)
"Should two items be created for these cases?" I'd say yes because there is no connection at all between the two names in pronunciation between Chinese and Japanese. Most western wikis seem to group Japanese names by their pronounciations, but I can't quite tell what systems (assuming the wiki has pages for names) that ja:wp uses, much less how Wikidata should go about it: at least some kanjis can even be used to spell completely different names! Circeus (talk) 23:18, 31 May 2019 (UTC)
I think if the same kanji has different pronunciations, they are technically different names, and we should create separate items for them. Similarly, when several kanjis have the same pronunciation, we should also have separate items. Those items can be linked with different from (P1889), and perhaps we could use the qualifier criterion used (P1013) with new criterion items explaining the differences.--Stevenliuyi (talk) 02:18, 1 June 2019 (UTC)
  • As there is some interest in Japanese names, we already have more than 2000 items for these. --- Jura 12:17, 1 June 2019 (UTC)
So, does this community confirm that it is willing to host such given name items? If yes, I will start massively populating Chinese people items with new given name items next year. I have no comment on names in Japanese, Burmese or other languages.--Roy17 (talk) 21:29, 8 June 2019 (UTC)

Wind speed tiers of a wind turbine

Hello. Each wind turbine model (example: Enercon E-126 (Q114672)) has multiple wind speed tiers. The minimum cut-in wind speed at which any power is generated, the rated wind speed at which the turbine generates 100% of capacity, and the cut-out wind speed at which the turbine speedbreaks are applied, and a maximum survivable wind speed before which the turbine structure will face a catastrophic failure. What is the best way to save these wind speed tier information on a particular wind turbine model item? Rehman 08:50, 26 May 2019 (UTC)

This is tricky. As a starting idea: significant event (P793) Q:"cut-in"/Q:"cut-out"/catastrophic failure (Q5051574) with qualifiers of wind speed (P5065) and P:"turbine angular speed" as a new property to link between wind speed and the angular speed of the turbine. Dhx1 (talk) 14:45, 26 May 2019 (UTC)
Unless the turbine did catastrophically fail, significant event (P793) is entirely improper for this. Circeus (talk) 19:14, 26 May 2019 (UTC)
Yes, P793 refers to a particular incident. I think we need new properties here, but I'm not sure if that is the best way, as the uses would most probably be limited to wind turbines? Rehman 13:27, 27 May 2019 (UTC)
It's a generic problem with Wikidata, User:Rehman. The database is only being used for a fixed number of properties (i.e. relationships between keys and values). That's understandable if the community wishes to impose a fixed ontology onto the database, but doesn't give niche applications the opportunity to create relationships like <Enercon E-126><has full power wind speed><30 km/hr>, etc. Perhaps the freer relationships available using structured data on Commons may represent a solution for you if/when we get its API exposed to Lua. --RexxS (talk) 14:21, 27 May 2019 (UTC)
We do not have properties for that, but maybe:
⟨ Enercon E-126 (Q114672)      ⟩ wind speed (P5065)   ⟨ value ⟩
criterion used (P1013)   ⟨ cut-in wind speed ⟩
That will require only to create items for cut-in wind speed, cut-out wind speed, rated wind speed and maximum survivable wind speed. But note that wind speed (P5065) despite its generic name is used only in relation to sport. That it would mean to broaden a scope of wind speed (P5065) or create a new property (and change a label of wind speed (P5065)).--Jklamo (talk) 15:42, 27 May 2019 (UTC)
Thanks Jklamo. This seems to be a good approach. I'll leave this thread open for a bit longer, and then go ahead with the implementation. I'd like to see if anyone has concerns about creating the separate Q items, if any. Cheers, Rehman 02:40, 29 May 2019 (UTC)
Alright, seems like there are no more concerns. User:Sillyfolkboy and User:Pintoch (who were involved in the creation of wind speed (P5065)), or anyone for that matter; are you able to help me expand wind speed (P5065)'s scope please? I' haven't done so before, and don't want to break anything. Rehman 03:48, 3 June 2019 (UTC)
@Rehman: The wind speed property is currently limited to qualifying usage and a m/s usage. I don't think it's much use for your use case as to apply wind speed types for a turbine even if we unrestricted wind speed to non-qualifying usages you would still need to request new properties to qualify each of the cut-in, cut-out, rated and maximum survivable wind speed types. As this is an unavoidable requirement, a simpler solution would be to simply create each of these as new, top level properties which can be applied without qualification to the turbine items. I can see we've already got similar properties like maximum sustained winds (P2895) for hurricanes. I'm happy to support a proposal for that if the four properties you mention are standard measures for turbines. Given the specificity of the terms, I think it's better to have specialised properties – I can't think of a way to genericize the "cut-in"/"cut-out"/"rated" properties, though possibly "maximum survivable wind speed" could find usage on other infrastructure items. Sillyfolkboy (talk) 17:58, 3 June 2019 (UTC)
@Sillyfolkboy: There's no need for new properties, criterion used (P1013) works as suggested by Jklamo. ArthurPSmith (talk) 14:53, 4 June 2019 (UTC)
@ArthurPSmith: Sorry, I missed that part. That is one versatile qualifier property! I suppose if you state the wind speed type item in Criterion Used then plain old speed (P2052) would be just as clear and avoids the issue of redefining every aspect of the sport wind speed property. Sillyfolkboy (talk) 19:24, 4 June 2019 (UTC)
User:Sillyfolkboy, based on the above, are we able to proceed with expanding the scope of wind speed (P5065)? Rehman 12:23, 6 June 2019 (UTC)

cut-in wind speed (Q64441403), rated wind speed (Q64441440), cut-out wind speed (Q64441513), survival wind speed (Q64441515), are created. An example of those in use can be found in Enercon E-126 (Q114672). Before I continue adding such data for other wind turbine models, appreciate any thoughts if this is the ideal way. Ping Jklamo, Sillyfolkboy, ArthurPSmith. Cheers, Rehman 08:46, 7 June 2019 (UTC)

@Rehman: Looks good to me. I've modified the wind speed wikidata property to make it more generalised. Sillyfolkboy (talk) 20:59, 7 June 2019 (UTC)
Thank you Sillyfolkboy! Rehman 06:11, 9 June 2019 (UTC)