Shortcuts: WD:PC, WD:CHAT, WD:?

Wikidata:Project chat

From Wikidata
Jump to navigation Jump to search

Wikidata project chat
A place to discuss any and all aspects of Wikidata: the project itself, policy and proposals, individual data items, technical issues, etc.

Please use {{Q}} or {{P}} the first time you mention an item or property, respectively.
Other places to find help

On this page, old discussions are archived after 7 days. An overview of all archives can be found at this page's archive index. The current archive is located at 2021/09.

Q108524859 vs Q12798037[edit]

Please have a look: Osypenko (Q108524859) vs Osipenko (Q12798037). Thx, HarryNº2 (talk) 05:50, 13 September 2021 (UTC)[]

They have the same spelling indeed, but different pronunciation. More comment needed from User:Бучач-Львів. --Infovarius (talk) 07:45, 16 September 2021 (UTC)[]
HarryNº2. Hi! Sorry for google-translate. To the words Infovarius it is also necessary to add that these surnames are written in different alphabets - Ukrainian and Russian, respectively. --Бучач-Львів (talk) 07:56, 16 September 2021 (UTC)[]
For surnames, we generally just add the script. --- Jura 09:59, 16 September 2021 (UTC)[]

Is not being able to see or query labels of deleted items a problem?[edit]

I dive in and out of editing Wikidata as a user, and recently came across some entities that I wanted to check to see if they had existed before. My thinking was that perhaps a user had created them, they got deleted for some reason at some point, potentially the user blocked and then later down the line a new user recreated them. Is there any existing tooling around this problem? Is this a problem? Or am I just imagining thing? ·addshore· talk to me! 19:51, 13 September 2021 (UTC)[]

@Addshore: It is a bit of a problem, compared to the language wikipedias where the "label" is also the way the deleted item is identified. Wikidata can't do that obviously since our item identifiers are not labels. I believe admins often add a label to their message when deleting, but I don't think there's any tooling around it. Might be useful, yes. ArthurPSmith (talk) 16:57, 14 September 2021 (UTC)[]
Depending on the admin deleting it (or the script the admin is using?), this is visible on the deletion log. At least for items one creates, this is sufficient. --- Jura 10:01, 16 September 2021 (UTC)[]
This idea came to my mind once in the past. Indeed, there is no way (known to me) to search for data of deleted items (in fact, any wiki pages in general). --Matěj Suchánek (talk) 16:10, 16 September 2021 (UTC)[]
Yeah, I don't think there is any way of doing this right now. In the future it might be possible to generate a dataset every now an again that could be helpful though. If this becomes available I'll have a play around and see what I can come up with. ·addshore· talk to me! 10:03, 20 September 2021 (UTC)[]

A few question concerning Wikidata[edit]


I just started editing wikidata (I mean, in addition of a few small edit I made a while ago). I mainly added credit for dub in The Legend of Spyro: A New Beginning (Q2578009).

1. Is there a recommended way to handle inverse property (P1696) ? This could usually have things that could be automated. For exemple, I both added Cyril (Q108537541) with the property characters (P674) to The Legend of Spyro: A New Beginning (Q2578009) and The Legend of Spyro: A New Beginning (Q2578009) with the property present in work (P1441) to Cyril (Q108537541). In addition, Cyril (Q108537541) is present as character role (P453) of Jeff Bennett (Q311103), voice actor (P725) of The Legend of Spyro: A New Beginning (Q2578009). That quite a bunch of duplicate entry. Is there a recommended way to handle them.

2. All of that made quite a bunch of data... I wonder if there is some simple way to generate a table from it, using wiki syntax (it would probably need some condition and stuff like that). But if I understand correctly, it will be one of the feature of Abstract Wikipedia (Q96807071)... So just need to wait {{Q:Q105576416}} implement this, and I'll probably be able to do something on not wikilambda or self-hosted wikilambda (or write a python script).

3. After some editing, I received welcome notification from many wikipedia. I figured that was because my edit made it to their wikipedia (like [1], where I added the game engine). Just wanted to made sure you are aware of what I consider an issue.

4. Is it okay to put the game as a source for it's dubber, when present in the credit ( via stated in (P248) ) ?

5. Is it okay to use imported from Wikimedia project (P143) when manually imported (I was unable to find official credit, but the french wikipedia give a list of voice actor. They don't have any source, and found conflicting other research on a few actor. I wasn't able to verify or deny these via data-mining either).

6. Is there some tool to make semi-automated edit/faster edit. For example, having to add template model that apply to every character present in the work is a bit long when they are the same for every new one.

Thank -- Marius851000 (talk) 20:55, 13 September 2021 (UTC)[]

@Marius851000 To answer your questions in order:
  1. I'm not aware of anything that handles this, though someone else might be. Part of the issue is that there are some inverse properties that don't really need copying because they're unbalanced and copying would put a huge number of statements on some items, or because they have a complex situation (such as child (P40), which could have either father (P22) or mother (P25) as its inverse)
  2. If you're making a table, you're going to want to use the query service. If you know SPARQL (Q54871), you can use the interface at; if you don't, you can follow the button at the top to the brand-new query builder. If you want to have the results on a continuously-updated wiki-page, you can check out User:ListeriaBot.
  3. Yes, that's typical.
  4. Yes, assuming that the information is verifiable by looking at the game itself (e.g. it's in the credits)
  5. Yes, though it's not good practice. If you have conflicting information from multiple sources, you can indicate this with the ranking system with appropriate qualifiers.
  6. QuickStatements is commonly used for semi-automated editing. If you know the edits you're going to make, you should be able to translate them into the QuickStatements input format, which handles the API mechanics of actually making the edits. Vahurzpu (talk) 00:38, 14 September 2021 (UTC)[]
  • For (4), I'd avoid doing that. --- Jura 10:40, 14 September 2021 (UTC)[]
  • About (5) / imported from Wikimedia project (P143): the property can be used when a statement is manually imported. --- Jura 10:32, 16 September 2021 (UTC)[]

Discourage adding descriptions to scientific articles?[edit]

Hi, I just saw this QS batch. It adds descriptions to a number of scientific articles in ast. This increases the number of triples BlazeGraph (BG) has to handle while adding no knowledge at all. Given our current situation with bots being opposed (see 1) despite them adding actual knowledge to the graph I think we should either discourage this kind of batches or simply exclude the descriptions from WDQS at this point to lower the total number of triples BG has to handle until a replacement is in place. WDYT?--So9q (talk) 09:40, 15 September 2021 (UTC)[]

  • I don't think we should restrict our editorial efforts because of such technical constraints. If the infrastructure is not capable enough to handle the load, WMF/WMDE should invest some money and make it fit; in order to do so, there needs to be some pressure sometimes… :-) —MisterSynergy (talk) 10:33, 15 September 2021 (UTC)[]
Still, it's such a waste. Why not simply browser-display the P31 value in the resp. language as description whenever it is not set? --SCIdude (talk) 15:19, 15 September 2021 (UTC)[]
BTW. I'm curious as to how well optimised the wikibase use is on mediawiki. Some entities appears to get prefetched, does this include «Cite Q» citations as well? Or is it just the entities that are linked via sitelinks that get prefetched? And how does the software identify what to prefetch? --Infrastruktur wdt:P31 wd:Q5 (T | C) 11:50, 15 September 2021 (UTC)[]
lots of QS edits add (close to) zero knowledge. it's hard to draw a line to decide which are worthless. Should we stop this current running job: ? BrokenSegue (talk) 16:06, 15 September 2021 (UTC)[]
Just because it's hard to draw a line in any continuous value doesn't mean you cannot say anything about the extremes. Example: we cannot say which exact grain of sand makes it a beach. But we can say it's not a (sand) beach when there is absolutely no sand, and it definitely is such a beach if all you see is water, sky, and sand. If someone were to, say, add and remove the same statement from the same entity over and over again, in multiple independent QS sessions, I'm sure there would be little disagreement to nicely asking them to find a new hobby. The above isn't that, and blocking well-intentioned contributions sounds like bad policy. But, maybe, one could nudge people to more meaningful endeavours. It isn't just server time that's wasted, it's the contributors' as well. One idea would be to either just hide edit counts beyond what's needed to categorise users into three or four experience levels. Or, maybe, there are alternative measures that include some notion of the usefulness by, for example, giving more gold starts for rarely-used properties. Or adding up the contributions' cumulative improvement in that item score I saw somewhere (Recoin?). Karl Oblique (talk) 20:24, 15 September 2021 (UTC)[]
What exactly is your point, Karl Oblique? --Succu (talk) 20:55, 15 September 2021 (UTC)[]
It shouldn't be *banned*, but discouraged. If people are motivated by their stats (possibly a wrong assumption on my part), changing the stats offered to better align with project interests might work. And as far as I can tell, the performance problems couldn't be immediately solved even if money were no concern. Therefore, and also just by general principle, the attitude of actively not caring about performance problems, or even welcoming them to create pressure, is surprising. Karl Oblique (talk) 21:41, 15 September 2021 (UTC)[]
I see no evidence people do this to up their edit count. It is very hard to make a rule about what we should be adding and what we shouldn't. BrokenSegue (talk) 22:21, 15 September 2021 (UTC)[]
I think it would be good to abolish the batch tool. Too many false information is added by those batches and most of it probably will never get corrected. It is also a great mistake multiplier. E.g. many Korean names items on wikidata are wrong. Most Korean names are gender neutral. Only a few have an associated gender. However, on wikidata, many of those actually unisex names are either categorized as female or male name. So, sometimes batches are running that at a gender to items of people that carry the name. And by this, some men become women on wikidata while some women become men. The mistakes only become more and a bot or batch cannot correct all these mistakes. Those batches lead to more mistakes and thus more work for real people. But the amount of mistakes is already so big, even when getting rid of those batches and have active users trying to fix it, it will take decades. --Christian140 (talk) 06:23, 16 September 2021 (UTC)[]
abolishing the batch tool is going too far. there's lots of good uses of it. I argue we just need to require references for things added with it. If someone adds a gender based on a first name we can later know that it was done incorrectly and undo it. or at the very least know not to trust it. BrokenSegue (talk) 18:53, 16 September 2021 (UTC)[]
Given the current systems, I suppose there isn't really a way around it.
I do find the label in "ast" (copy of "en") more problematic. --- Jura 10:03, 16 September 2021 (UTC)[]

Power outages documented?[edit]

If an area has a bunch of people who get affected by power outages do the outages and the length of time of the outages get added on Wikidata or does an invidiual interested in this start their own Wikibase instance and run it privately? Oduci (talk) 14:06, 15 September 2021 (UTC)[]

This looks like quite complex dataset. Every year, there will be a new number for each region. Additionally, you could have several data values each month, for example: 30% of inhabitants of region A had power outage for 2 days or more, 15% of inhabitants had outage for 1 day or more... Therefore, in my humble opinion, it's better to store this data elsewhere, such as in Tabular Data on Commons Vojtěch Dostál (talk) 13:12, 16 September 2021 (UTC)[]

Inappropriate alias?[edit]

Could you read this thread, please? Do you think that "Prince Firouz Nosrat-ed-Dowleh III" is a valid alias or not? Thanks. Emijrp (talk) 17:57, 15 September 2021 (UTC)[]

That must be some misunderstanding... A simple search for "Prince" returns >37,000 results. And while there are many where it's not a person's label on the first pages, it quickly turns into a *Who's Who* of European Royalty. My SPARQL attempts keep timing out with more a LIMIT > 5000, but I'd estimate there is a 6-digit number of labels and aliases that include *Prince*, and we haven't even started with the Viceroys. Apart from the normative power of the factual, it just makes sense to include the title when it is commonly used when referring to the person. A random article about the royal family I just found includes the title when the name is first mentioned, in image captions, and in tabular data. For someone such as Queen Elizabeth The Queen Mother (Q10633), little is left of their commonly-used label/name if you take out all titles. "Elizabeth The"? Karl Oblique (talk) 20:08, 15 September 2021 (UTC)[]
Any string (subject to BLP considerations) that reasonably might be supposed to be used to refer to an items subject, should be considered valid as an alias, since the purpose of the alias is to facilitate discovery of the item from such strings. I note the thread Emijrp pointed to: the OP there has notions about what is & is not a proper, valid alias, built on preconceptions rather than on any evidence of understanding the functions of aliases in the WD environment. --Tagishsimon (talk) 20:22, 15 September 2021 (UTC)[]
  • name (P2561) seems to be more suitable for the form of qualified alternate label Sapphorain is trying to add or reference. (I had a similar discussion with the same user, @Sapphorain: in case you haven't been notified). --- Jura 09:28, 16 September 2021 (UTC)[]

[notification] All village of Taiwan is imported from Taiwan government dataset[edit]

Hi all, all village of Taiwan (Q7930614) in Taiwan (Q865) including the future one, which use Code of Household Registration and Conscription Information System (Taiwan) (P5020) is all on Wikidata. Some Wikidata Taiwan (Q65555605) member linked Village item to Taiwanese aborigines (Q216839) Q55747241. Maybe some duplicated item still not merged due to someone imported from Chinese Wikipedia, the other one imported from government dataset. Maybe some day will deal with the abolished village which exist before digital record document, like 大豐里, 麗山里 Nangang District (Q271073), Taipei (Q1867). Supaplex (talk) 10:08, 16 September 2021 (UTC)[]

Is there a way to get the total number of references to a website on Wikidata? (Query service times out)[edit]

Hi all

I'm encouraging an organisation to share data on Wikidata, they have a huge amount of data which could be really useful. I want to collect some metrics for them before and after on how many references there are on Wikidata to their website. NavinoEvans has tried to make Wikidata queries for it but it just times out.

I've also tried using the same trick I use on Wikipedia which uses a search tool here.

Does anyone know of any other approaches we could use or if I'm doing something wrong? I just want a count of all the references on Wikidata to

Thanks very much

--John Cummings (talk) 15:51, 16 September 2021 (UTC)[]

@John Cummings: You can use the list=exturlusage (eu) API call to list all items (i.e. pages in namespace 0) with links to the website. It lists all different URLs to the website from each item, but it doesn't tell how many links there is from each item to the same URL. Neither does it tell if the link is from a reference, qualifier or statement value.
If the number of results is less than 10.000, you can get the count with a SPARQL query using MWAPI.
You can also make same search from the special page Special:LinkSearch. Dipsacus fullonum (talk) 20:23, 16 September 2021 (UTC)[]
@Dipsacus fullonum: Would you mind putting together an illustrative SPARQL query using MWAPI, Df? I think we have too few examples SPARQL/MWAPI examples to study, & I'm afraid I find the documentation opaque. thx. --Tagishsimon (talk) 20:28, 16 September 2021 (UTC)[]
Dipsacus fullonum thanks very much, I don't have a technical background so I don't understand how to use these tools, I tried using Special:LinkSearch before but what works on Wikipedia (linked above) doesn't work on Wikidata, would you be able to create the search term for and I'll write some instructions somewhere so people can do it in future? John Cummings (talk) 21:43, 16 September 2021 (UTC)[]
@John Cummings, Tagishsimon: John, you linked to a normal search (Special:Search using CirrusSearch), not to Special:LinkSearch which works fine for Wikidata items. Here is a query finding statements in items that have references containing URLs to I only included the http and https protocols, but other protocols can be added by adding more MWAPI calls to the SPARQL UNION. The query will give incomplete results for websites where the API calls would give more 10.000 results.
# Find item statements with references containing external links to using http and https protocols
SELECT ?item ?itemLabel ?property ?propertyLabel ?value ?valueLabel ?reference_URL
      # Find items with https URLs
      SERVICE wikibase:mwapi
        bd:serviceParam wikibase:endpoint "" .
        bd:serviceParam wikibase:api "Generator" .
        bd:serviceParam mwapi:generator "exturlusage" .
        bd:serviceParam mwapi:geuprop "title" .
        bd:serviceParam mwapi:geunamespace "0" .
        bd:serviceParam mwapi:geuprotocol "https" .
        bd:serviceParam mwapi:geuquery "*" .
        bd:serviceParam mwapi:geulimit "max" .
        ?item wikibase:apiOutputItem mwapi:title .
      # Find items with http URLs
      SERVICE wikibase:mwapi
        bd:serviceParam wikibase:endpoint "" .
        bd:serviceParam wikibase:api "Generator" .
        bd:serviceParam mwapi:generator "exturlusage" .
        bd:serviceParam mwapi:geuprop "title" .
        bd:serviceParam mwapi:geunamespace "0" .
        bd:serviceParam mwapi:geuprotocol "http" .
        bd:serviceParam mwapi:geuquery "*" .
        bd:serviceParam mwapi:geulimit "max" .
        ?item wikibase:apiOutputItem mwapi:title .
} AS %get_items
  INCLUDE %get_items
  ?item ?prop ?statement .
  ?property wikibase:claim ?prop .
  ?property wikibase:statementProperty ?ps .
  ?statement ?ps ?value .
  ?statement prov:wasDerivedFrom ?reference .
  ?reference ?refprop ?reference_URL .
  FILTER CONTAINS(STR(?reference_URL), "")
  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en" . }
Try it!
--Dipsacus fullonum (talk) 07:50, 17 September 2021 (UTC)[]
Update: I changed the query, moving the MWAPI service calls to a named subquery using the DISTINCT modifier in order to get each statement only once. --Dipsacus fullonum (talk) 09:57, 17 September 2021 (UTC)[]
Very nice! I agree with @Tagishsimon: that the mwapi documentation is lacking. A while back I created a section for illustrative queries at Wikidata:SPARQL_query_service/queries/examples/advanced#Mediawiki_API. --Azertus (talk) 17:18, 17 September 2021 (UTC)[]
I wrote some documentation and explanation for reference searching in Wikidata:SPARQL_query_service/query_optimization#A_query_that_has_difficulties. --Lockal (talk) 07:40, 20 September 2021 (UTC)[]

Item fix request[edit]

Is there someone that can kindly change official website (P856) from to on Mexico national football team (Q164089)? Many thanks in advance!!! -- 16:26, 16 September 2021 (UTC)[]

New Streaming Updater for Wikidata Query Service in production 18 Oct 2021[edit]

Hello all,

Thank you again for your all recent thoughts and feedback with regard to the recent Wikidata: Query Service (WDQS) scaling update Aug 2021, and for everyone who has responded to the WDQS user survey. As part of our ongoing work to scale WDQS, we will begin shipping the new Flink-based Streaming Updater from test servers to production on 11 October 2021, with the entire data transfer process expected to finish by 18 October 2021.

The primary goal of this new Streaming Updater is to reduce update lag, and throttling, of edits to Wikidata: going from an average of 10 edits/second to an average of 88 edits/second. We are excited that this is almost a 9x improvement in our ability to make sure that Wikidata Query Service has the freshest updates from Wikidata, a priority that many of you ranked highly in the recent survey. Additionally, the new update process will lessen the impact on Blazegraph itself by moving diff reconciliation away from the service. This update process will be more stable as a result, with more use cases like un/deletes handled correctly.

In order to minimize risks of failure during the rollout of the Streaming Updater, we will be moving individual servers over one at a time. During the anticipated 7 days this data transfer process will happen (11-18 Oct), it is possible that some users will see inconsistent behavior or other bugs while querying. While we encourage these bugs to be filed, please note that it may be difficult for the Search team to accurately diagnose the source of these errors due to the nature of the process. We hope for a seamless transfer, of course, in which users will not notice any errors during the switchover.

We previously announced the new Streaming Updater being released to test servers in March 2021; the changes announced there will now be effective for all WDQS users effective 18 Oct 2021.

The changes that allow the new Streaming Updater to reduce update lag comes with two notable changes, which have the potential to break current usage and workflows:

  1. Blank nodes in Wikidata have been skolemized. From a user perspective, (1) queries using isBlank() will need to be rewritten; (2) queries using isIRI/isURI will need to be verified; (3) WDQS results will no longer include blank nodes.
  2. Constraint Fetching -- specifically wikibase:hasViolationForConstraint -- will be temporarily disabled until we are able to expose constraint violations in a more production-ready way.

For more details on these changes, please refer again to our prior announcement.

We find it encouraging that the new Streaming Updater has not caused major relevant issues in the last six months while it has been on We understand these changes may not be optimal for everyone. However, we believe the ability to greatly reduce Wikidata’s edit lag will be a beneficial improvement for all editors.

We’re excited to ship the new Flink-based Streaming Updater to production, and believe this is a significant step in scaling Wikidata and WDQS. As always, we encourage you to report technical problems and/or leave general comments/feedback here in Project Chat.

Best, WMF Search & WMDE – The preceding unsigned comment was added by MPham (WMF) (talk • contribs).

  • It would be good to have some documentation page about "wikibase:hasViolationForConstraint". Can you add to (or link if it already exists from) Wikidata:SPARQL_query_service/Wikidata_Query_Help? --- Jura 14:43, 19 September 2021 (UTC)[]
    • This page provides documentation for how wikibase:hasViolationForConstraint is currently in use. Does that help? I'm not sure linking it to the Query Help page will be helpful in the near future because the new Streaming Updater will be disabling that particular function. Better documentation will be provided once constraint violations are exposed in a more production-ready way MPham (WMF) (talk) 17:08, 21 September 2021 (UTC)[]
    • Personally, I wont need it, but there is some demand for it. See the question at Wikidata:Project_chat/Archive/2021/09#Guidance_for_checks_after_upload. If it's not production ready yet, I suppose my answer there is sufficient. --- Jura 17:18, 21 September 2021 (UTC)[]
  • Thanks for getting this done - almost 9x performance boost, that's really impressive!! ArthurPSmith (talk) 16:49, 20 September 2021 (UTC)[]


Is Q29383381 a duplicate of Q58398507? Or do they both serve a different purpose? --Mbrickn (talk) 20:22, 16 September 2021 (UTC)[]

I was almost completely certain that they should be merged, but found Talk:Q4167836#Usage and, after checking, found a great many of such corresponding pairs of items. They are linked via category's main topic (P301) and topic's main category (P910). I believe commons categories were moved into the items at some point, connected by Commons category (P373)? Categories, and maybe lists as well, would seem to work well as a second (/third) set of sitelinks. But whoever came up with the status quo probably spend more time thinking about it, and there's some snag I'm missing. Karl Oblique (talk) 21:55, 16 September 2021 (UTC)[]
I think they can be merged in this case. The zh sitelink should be on the main item, and a separate category item isn't required when there's only a Commons cat. Ghouston (talk) 10:38, 17 September 2021 (UTC)[]

Question from Wikimedia Foundation Legal about Wikidata BLP email requests[edit]


On behalf of the Wikimedia Foundation's Legal department, I am asking for input on our current process for handling BLP-related requests that are sent by email to

I have asked on the talk page of the BLP policy.

I do not think this question quite qualifies as an RFC or an administrator's noticeboard post, but please spread the word to anyone who may be interested in the matter. Thank you! BChoo (WMF) (talk) 22:02, 16 September 2021 (UTC)[]

Middle initial property?[edit]

Hi. If there's a person whose middle initial is known but not their middle name, is there a property that should be used to store the initial? Thanks DemonDays64 | Talk to me 00:24, 17 September 2021 (UTC) (please ping on reply)[]

  • Some people use letters with dots as "given name", to me it looks ridiculous. --E4024 (talk) 01:04, 17 September 2021 (UTC)[]

Item fix request[edit]

Is there someone that can kindly change official website (P856) from to on Mexico national football team (Q164089)? Many thanks in advance!!! --2001:B07:6442:8903:7824:6311:54B2:93FD 07:14, 17 September 2021 (UTC)[]

Looks like User:Tagishsimon took care of it for you. ArthurPSmith (talk) 16:43, 20 September 2021 (UTC)[]

Import IEEE Xplore author ID matching existing ORCID?[edit]

Many author profiles on IEEE Xplore include their ORCID (such as in this one). Could we have a bot import IEEE Xplore author ID (P6479) when it matches profiles with ORCID iD (P496) that we already have? --Bender235 (talk) 13:25, 17 September 2021 (UTC)[]

P26 (spouse) and preferred rank and no value[edit]

If I currently query WDQS for the spouse (P26) of Brad Pitt (Q35332), I will get two results. However, Brad Pitt is divorced and NOT married at the moment. Therefore, the query should return zero results. In my opinion, there should be another spouse with 'no value' and preferred rank. That way, all the information is there and the aforementioned query gives the correct result (namely, that Brad Pitt is not currently married). After manually checking some items it seems to me that it is not currently handled like this. The problem occurs with other items in the same way. The spouse property seems to mean 'everybody who this person was ever married to', even for truthy/direct Wikidata. Can anybody please explain why it is not handled the way I propose? --Graue70 (talk) 13:59, 17 September 2021 (UTC)[]

I tend to agree that the item is missing a spouse of <no value> which is of preferred rank. I've done the same for the heritage designation of items that have had a heritage listing removed: normal rank with start and end dates for the listing claim, preferred rank for <no value> with a start time, such that a truthy query yields the currently valid result. --Tagishsimon (talk) 14:03, 17 September 2021 (UTC)[]
I see a problem with using "no value" for spouse. How will you prove the statement? A person could have gotten married in secret, and how to prove that it didn't happen? It is more reliably to just see that no statements for spouse have time qualifiers for the present time, and then conclude that noting is known about the present state. Dipsacus fullonum (talk) 15:04, 17 September 2021 (UTC)[]
Perhaps. Claims with no reference are what they are; users can rely on them to the extent they wish. They're suboptimal, but so too is a query result which suggests a truth which is false. For spouse, there may well be sources that explicity state the subject is single; and if not, then a reference inferred from (P3452), for instance, inferred from timeline of events (Q81204096), will throw some more light on the situation. Meanwhile, many statements are friable, and just as we do not know the subject has not got himself remarried in secret, nor do we know one way or another whether the the status of millions of other claims might without our knowledge have changed. --Tagishsimon (talk) 15:28, 17 September 2021 (UTC)[]
You do have a valid point. But using ranks may also give unexpected query results. Many of the questions in Wikidata:Request a query over time have been from users who didn't get the results they excepted because they didn't consider ranks (usually using the wdt: prefix when they really wanted all non-deprecated values). Dipsacus fullonum (talk) 15:47, 17 September 2021 (UTC)[]
By the way, is the current marriage status of a person really more important than previous values? If you have been married for a number of years but are no longer, does it not say as much about the person as the current single status? I have no particular objections to a "no value" value, but do not think it should have preferred rank. Dipsacus fullonum (talk) 16:31, 17 September 2021 (UTC)[]
Just a note: setting any current value (or "no value") as a "preferred rank" could have the unfortunate effect of excluding all "normal-ranked" values (e.g. spouses) from queries (according to Help talk:Ranking#Rank_and_sparql), and infoboxes such as Template:Infobox Wikidata (Q108290953) (deployed on millions of Wikimedia Commons categories). Granted, this latter may be more of a problem with Commons than Wikidata, and the Wikidata Infobox template has plenty of problems, but if queries or infoboxes only provide highest ranked values, then care should be taken to ensure other data isn't needlessly hidden, such as the many husbands of Elizabeth Taylor (Q34851). -Animalparty (talk) 20:27, 17 September 2021 (UTC)[]
Sure, but ideally the care will be taken by template authors looking at p:/ps: property paths rather than wdt: paths, if they want all the values. I guess it comes down to whether or not we think a currently true value should be awarded preferred rank when other historic values exist at normal rank, such that truthy=contemporary --Tagishsimon (talk) 20:39, 17 September 2021 (UTC)[]
Please do not use no value for a property that contains other values. This is never correct. --Quesotiotyo (talk) 21:41, 17 September 2021 (UTC)[]
It can be correct with appropriate qualifiers for when "no value" is valid. Dipsacus fullonum (talk) 21:46, 17 September 2021 (UTC)[]
Could you please point to an example? I am unable to think of any case where this would be appropriate. --Quesotiotyo (talk) 22:30, 17 September 2021 (UTC)[]
If a person was married to person A from 2003 to 2007, was single from 2007 to 2011, was married to person B from 2011 to 2020, and is single from 2020 to now, you could have the statements:
to give a complete picture of the person. However I think they would all be equally important to the description of the person so I wouldn't give the any of them preferred rank. If someone wants to know the current spouse, they should query for statements without a end time (P582) qualifier. Dipsacus fullonum (talk) 05:24, 18 September 2021 (UTC)[]
While I do not believe that this was the original intent of no value (see mw:Wikibase/DataModel#PropertyNoValueSnak and the last comment at Wikidata:Project_chat/Archive/2017/01#Never_married_persons), it does indeed seem to be the best way to model a definitive negative value for a specified subset. I also realized that it might be necessary to use both no value along with another value when modeling conflicting sources, e.g. if some sources state that a person was never married in his or her lifetime but other sources indicate that the person was married at some point. I retract my claim about it never being correct, so long as the no value statement is differentiated by either a qualifier or ranking and that there is a source attached which definitively states the negation.
--Quesotiotyo (talk) 18:12, 18 September 2021 (UTC)[]
This speaks to a fundamental unresolved (and probably unsolvable) philosophical question of Wikidata: when does it ever end? In theory, items can never stop accumulating data, as statements can be endlessly added, refined, and subdivided to reflect status and attributes at any identifiable point in time (e.g. number of children on 1 Jan 1900 = 1. Number of children on 2 Jan 1900, also = 1. Number of children on 1 Jan 1901= 2. But following a death the following week, Number of children on 10 Jan 1901= 1, etc. ad infinitum). In practical terms, well, what's practical for the pedant is impractical cruft for the common user. A discussion about the "end game" or future utility of Wikidata (will it be a place accessible to humans or just robots?) needs to be had elsewhere. -Animalparty (talk) 22:18, 18 September 2021 (UTC)[]
  • spouse (P26) has some specific aspects other properties may not have:
    • It may be desirable to differentiate living people from others.
    • The information may not be available for living people. So the presence or absence of P26 might not be that useful.
    • Once a person is dead, preferred rank isn't of much use.
    • In general, "no value" is more useful for deceased persons as it's easier to make a definite statement.
    • I suppose we could have a statement: Brad Pitt (Q35332) P26 = no value, qualified with point in time (P585) = 1999 using a reference: "10 most [..] bachelors of the year 1999".
--- Jura 14:02, 19 September 2021 (UTC)[]
I suppose if someone is really ambitious and into tedious cruft, one could make a bot that goes and adds "no value" of spouse to every human who died before the age of 1. How about that? That would be technically true (save for perhaps some odd instances where an infant is culturally married before they can walk), but functionally redundant and meaningless. One could also add number of children (P1971) = 0 and occupation (P106) ="no value" to every dead baby in royal genealogies. One could also indicate that Hammurabi (Q36359) has "no value" for LinkedIn personal profile ID (P6634), as well as every human who died before the year 2000. But why? When does it stop? -Animalparty (talk) 05:36, 20 September 2021 (UTC)[]
For deceased people worth checking for "novalue" or custom value to add, I think Wikidata:Database reports/top missing properties by number of sitelinks/P26 would be the better list to start with. --- Jura 05:43, 20 September 2021 (UTC)[]
I updated the report (it hadn't been running in years). It now stops at 85 sitelinks minimum and date of death (P570) after 1500. --- Jura 23:20, 21 September 2021 (UTC)[]

Item fix request[edit]

Is there someone that can kindly change official website (P856) from to on Mexico national football team (Q164089)? Many thanks in advance!!! --2001:B07:6442:8903:A98F:701C:9EEA:74D1 15:30, 17 September 2021 (UTC)[]

Done. --Tagishsimon (talk) 19:25, 17 September 2021 (UTC)[]

Züleyha Tan (Q19950682)[edit]

Alguién con toda buena voluntad pero poca atención hizo su art en ES:WP como Züleyhan Tan (sic). Su nombre es Züleyha Tan (Q19950682). Como me expulsaron de esa Wikipedia por solo demostrar sus errores -aunque con otro pretexto- yo no lo puedo cambiar. ES:WP no me interesa mucho, pero la gente de mi querido país, Turquía, sí. Gracias de antemano. --E4024 (talk) 15:53, 17 September 2021 (UTC)[]

  • Please use Wikidata:Café for Spanish language posts. --- Jura 13:52, 19 September 2021 (UTC)[]
✓ Done, E4024. I moved it to Züleyha Tan.--Esp1986 (talk) 16:21, 19 September 2021 (UTC)[]
Muchas gracias Esp1986! (Note: Muchas gracias means thank you very much; in Spanish.) Saludos cordiales. --E4024 (talk) 22:18, 19 September 2021 (UTC)[]
I think that this discussion is resolved and can be archived. If you disagree, don't hesitate to replace this template with your comment. E4024


For Agen Warehouse (Q28035251), the geocoordinates are sourced from en-wiki (which has them correct) but when I use the coordinates from Wikidata they are off by several blocks. Is there a way to fix this, or are geocoordinates in Wikidata only accurate to some small number of decimal places? (Please ping me if responding, I'm rarely on Wikidata these days.) - Jmabel (talk) 18:51, 17 September 2021 (UTC)[]

@Jmabel: Both WD and WP coordinates seemed to resolve to the same place for me, both in terms of the parameters listed on the geohack page, and the pushpin on ACME mapper. WD coordinates are stored with a precision value, the use or effects of which I'm not familiar. In my experience P625 values default to Precision: ±0.00001° but in the case of Agen Warehouse (Q28035251), the precision was specified at, from memory, ±0.015° ish. Long story short, I've deleted the existing coords and added them again to ±0.00001° precision. Does this solve the issue you see? --Tagishsimon (talk) 19:22, 17 September 2021 (UTC)[]
@Tagishsimon: Yes. So it was simply a matter of someone (User:GZWDer (flood)) explicitly stating an insufficient precision. Got it. - Jmabel (talk) 22:16, 17 September 2021 (UTC)[]
And that's apparently a bot. I wonder why it assumed such a low precision? - Jmabel (talk) 22:17, 17 September 2021 (UTC)[]

Capitalization of common names of organisms[edit]

Is there a policy somewhere on recording common names of animals, plants, and other organisms in labels? I ask because there is a wide divergence of practice, and when I recently changed some labels to remove uppercase, one insistent editor reverted all of my changes. Commonly used style manuals in English say not to capitalize common/vernacular names of organisms unless there is a proper noun in the name.

For example, The Chicago Manual of Style says:

For the correct capitalization and spelling of common names of plants and animals, consult a dictionary or the authoritative guides to nomenclature, the ICN and the ICZN, mentioned in 8.119. In general, Chicago recommends capitalizing only proper nouns and adjectives, as in the following examples, which conform to Merriam-Webster's Collegiate Dictionary:

  Cooper's hawk
  rhesus monkey
  Rocky Mountain sheep

The International Committee on Taxonomy of Viruses (ICTV) says: The first letters of words in a virus name, including the first word, should only begin with a capital when these words are proper nouns (including host genus names but not virus genus names) or start a sentence. Single letters in virus names, including alphanumerical strain designations, may be capitalized. ... Examples:

  Isolates of dengue virus 2 were obtained ....
  Detection of West Nile virus in human serum ....
  Salmonella phage SE1 was isolated ....
  Sida ciliaris golden mosaic virus (SCGMV) causes ....
  Aphids transmit potato virus Y (PVY)

Numerous other websites for English names give the same advice, for example,,, and

Is there, or could/should there be a written policy on capitalization of common names? UWashPrincipalCataloger (talk) 21:24, 17 September 2021 (UTC)[]

This is a common mistake by German-language people that think they have a good grasp on English (I'm one myself but I think I've passed this lesson). --SCIdude (talk) 09:41, 18 September 2021 (UTC)[]

Q108571821, “items on Wikidata” option does not appear in the article sidebar.[edit]

Why Q108571821 in IDWP, the “items on Wikidata” option does not appear in the article sidebar? --NengDorla (talk) 22:59, 17 September 2021 (UTC)[]

Possibly a cache issue somewhere. The link is there now. --Tagishsimon (talk) 23:17, 17 September 2021 (UTC)[]

Not sure whether to merge two items[edit]

There are two items, Template:Incomplete (Q5873067) and Template:Incomplete (Q17392254), which I'm not sure whether to merge. They seem to be items for the same template, and I know s:es:Plantilla:Incompleto (linked to Template:Incomplete (Q5873067)) has the same functionality as s:en:Template:Incomplete (linked to Template:Incomplete (Q17392254)), but there are also some cases of duplication, e.g. b:zh:Template:Incomplete and b:zh:Template:未完成. —CalendulaAsteraceae (talk | contribs) 23:39, 17 September 2021 (UTC)[]

Machinetranslated labels and descriptions[edit]

It is pretty unlikely that Wikidata would ever have good coverage with translations so would it be possible to pre-populate titles and descriptions at least for the most common languages as a failback? It may be feasible to generate those for other languages as well but for usability English would be most useful.

Technically this could be implemented so that there would be additional information if the label/description is machine translated and what is the source language and when it is created. Same kind of data could be used for storing information if the label/description is just blindly copied from other languages or generated. --Zache (talk) 06:34, 18 September 2021 (UTC)[]

I agree. It might well be that empty labels / descriptions in the browser UI cause some people to make impulsive bot runs to fill the fields. If P31 values are present they might be better candidates for the description to show, because they are manually translated. --SCIdude (talk) 09:30, 18 September 2021 (UTC)[]
No. Once Abstract Wikipedia is a thing every item may have a "multilingual" label and description.--GZWDer (talk) 15:28, 18 September 2021 (UTC)[]
The default description of an item would generally be the label of its P31 value. So I don't really see a need to try to machine translate something. The bot for Dutch descriptions works fairly well with that.
I'm not really sure what machine translations you would want to insert as English labels. --- Jura 13:51, 19 September 2021 (UTC)[]

linking airplane crashes to victims[edit]

Do we have a way to link airplane crashes to victims? plane crash near Formia of April 30, 1938 (Q108595147) had Jakob Friedrich Bollschweiler (Q1679021), Xhafer Villa (Q29863774) and Othmar Krainz (Q55684069) and others among its victims, but I couldn’t find a good way to link them. I know that there are still many modeling problems for death, but maybe there’s some sort of workaround until we resolve those? --Emu (talk) 21:41, 18 September 2021 (UTC)[]

participant (P710) and victim (P8032) are both used:
SELECT ?instance_of_aircraft_crash ?instance_of_aircraft_crashLabel ?property ?propertyLabel ?person
  ?instance_of_aircraft_crash ?p ?person.
  ?property wikibase:directClaim ?p .
  ?instance_of_aircraft_crash wdt:P31/wdt:P279* wd:Q3002150 .
  ?person wdt:P31 wd:Q5 .
  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en" . }
Try it!
Also different ways are used link the victim to the airplane crash. The most common are
SELECT ?person  ?property ?propertyLabel ?qualifier ?qualifierLabel ?instance_of_aircraft_crash ?instance_of_aircraft_crashLabel
  ?person wdt:P31 wd:Q5 .
  ?person ?p ?statement .
  ?statement ?pq ?instance_of_aircraft_crash .
  ?property wikibase:claim ?p .
  ?qualifier wikibase:qualifier ?pq .
  ?instance_of_aircraft_crash wdt:P31/wdt:P279* wd:Q3002150 .
  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en" . }
Try it!
SELECT ?person  ?property ?propertyLabel ?instance_of_aircraft_crash ?instance_of_aircraft_crashLabel
  ?person wdt:P31 wd:Q5 .
  ?person ?p ?instance_of_aircraft_crash .
  ?property wikibase:directClaim ?p .
  ?instance_of_aircraft_crash wdt:P31/wdt:P279* wd:Q3002150 .
  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en" . }
Try it!
I cannot say what is best or recommended. --Dipsacus fullonum (talk) 01:17, 19 September 2021 (UTC)[]
Thank you, that helped a lot! --Emu (talk) 10:27, 19 September 2021 (UTC)[]

Merge Syzygy pages Syzygy (Q54364) and Syzygie (Q2377241)[edit]

Syzygy (Q54364) and Syzygie (Q2377241) should both refer to the same disambiguation pages. A possible conflict is that they both have a different Wikipedia entry in Deutsch:

- Syzygy (Q54364) refers to while

- Syzygie (Q2377241) refers to

The correct Deutsch translation is Syzygie. 14:06, 19 September 2021 (UTC)[]

  • It's a Wikidata item for disambiguation pages of Wikipedia versions with a given title.
No translation is actually being done.
See Wikidata:WikiProject Disambiguation pages/guidelines for details. --- Jura 14:17, 19 September 2021 (UTC)[]
My remark on the translation was solely meant to advise on the compatibility conflict. « Disambiguation items in Wikidata exist only to provide interlanguage links »: this is exactly the problem here as the interlanguage link of the disambiguation pages are incorrect/missing. 15:28, 19 September 2021 (UTC)[]
The two items don't refer to the same disambiguation pages.
If you have a look at Wikidata:WikiProject Disambiguation pages/guidelines, you will notice that Q54364 needs some fixes to follow it.
If Wikipedia should display other interwiki links, this needs to be resolved at Wikipedia. They can do that based on said to be the same as (P460) statements. --- Jura 15:33, 19 September 2021 (UTC)[]

How to improve precision of dates (sample: died 20th century)[edit]

Please see Help_talk:Dates#How_to_improve_precision_of_dates. --- Jura 08:31, 20 September 2021 (UTC)[]

Owen Pierce Thompson at Wikipedia[edit]

Should w:Owen Pierce Thompson be fast deleted without debate? --RAN (talk) 12:22, 20 September 2021 (UTC)[]

That is no concern for Wikidata. Please keep Wikipedia discussions at Wikipedia. --Dipsacus fullonum (talk) 13:10, 20 September 2021 (UTC)[]
Deletion from English Wikipedia usually ends up with deletion from Wikidata, so a concern for me. --RAN (talk) 13:26, 20 September 2021 (UTC)[]
In this case, I don't think the item should be deleted. If it does, you should complain loudly, because it seems to be well-linked to other items, has a Commons category and several interesting identifiers. Vojtěch Dostál (talk) 13:33, 20 September 2021 (UTC)[]
Q7114601 shouldn't be problematic. --- Jura 13:40, 20 September 2021 (UTC)[]
I "converted" the obituary available at Commons to statements. Not sure though what's the best way to quote it though. Probably make an item for it and use that. --- Jura 14:49, 21 September 2021 (UTC)[]

copyright status (P6216)[edit]

I would like to add ineligible for copyright protection (Q61005058) to copyright status (P6216) because I think that none of the existing values (copyrighted; not yet determined; unknown value; public domain; copyrighted, dedicated to the public domain by copyright holder; orphan work; no known copyright restrictions) really represent this. There is "public domain", but in my opinion this value covers too many possible situations to be an accurate description. "copyrighted, dedicated to the public domain by copyright holder" is also listed in here even you could argue that "public domain" would represent the same thing. Can I just add it or is there any procedural I should follow?
Link to the discussion on the talk page: Property_talk:P6216#Adding_ineligible_for_copyright_protection
--D-Kuru (talk) 14:04, 20 September 2021 (UTC)[]

I think this would be described by
Specifically, "ineligible for copyright" is a route to being in the public domain (a copyright status), but it is not, in itself, a copyright status.
I feel your pain in trying to get a solid answer on the schema for this though, since I have basically very similar questions that I can't get an answer to. Inductiveload (talk) 15:08, 20 September 2021 (UTC)[]
@Inductiveload: Would also work for me. But wouldn't this apply to copyrighted, dedicated to the public domain by copyright holder (Q88088423) as well? --D-Kuru (talk) 09:40, 21 September 2021 (UTC)[]
Quite possibly. That said, both copyrighted, dedicated to the public domain by copyright holder (Q88088423) and ineligible for copyright protection (Q61005058) are subclasses of public domain (Q19652) (see the graph, so actually your proposal sounds correct (or at least...consistent with the incomplete schema currently used). Inductiveload (talk) 09:47, 21 September 2021 (UTC)[]

Wikidata weekly summary #486[edit]

F-droid Wikidata project? - Can I add similar applications to Vanilla Music (Q104403685) from the F-Droid repository?[edit]

Can I add similar applications to Vanilla Music (Q104403685) from the F-Droid repository?(I noticed no Wikidata items linking to it other than a list about games though I know Vanilla Music is not a game, it's a music player).

If yes, I might want to add(in no particular order):

  • "Raag", music player
  • "Music Player GO"
  • "Apollo", music player
  • "Simple Music Player"
  • "Odyssey Music Player"
  • "Auxio", music player
  • "BlackHole", music player
  • "Finamp", music player
  • "Odeon", music player

I did not find any of these in Wikidata. Also who here is interested in adding data related to F-Droid apps or knows any Wikidata project related to either F-Droid or Android apps? Oduci (talk) 19:24, 21 September 2021 (UTC)[]

@Oduci: those all seem like fine additions to wikidata. BrokenSegue (talk) 19:51, 21 September 2021 (UTC)[]

False constraint violation[edit]

The properties of Stephanie Devillé (Q9345692) are showing type constraint violations despite the item being of the correct type (instance of (P31) human (Q5)). What's happening there? --Sod25 (talk) 18:52, 21 September 2021 (UTC)[]

There are several statements with "deprecated" rank that probably should not be deprecated. You can read more about ranks at Help:Ranking. --Lydia Pintscher (WMDE) (talk) 19:07, 21 September 2021 (UTC)[]
Weird; deprecated ranks, as created in 2013 [3]. I've set them all to normal rank - Sod25, that should sort the issue out. Let us know if not. --Tagishsimon (talk) 20:18, 21 September 2021 (UTC)[]
Yes now fixed, thanks to you both. It would be good if properties with a deprecated ranking had "(deprecated)", linked to Help:Ranking, next to their values, which would make it easier to spot them when skimming a page and also make the ranking system more discoverable for new users. --Sod25 (talk) 04:27, 22 September 2021 (UTC)[]

Nominate your favorite Wikidata projects for the WikidataCon community awards[edit]

WikidataCon Community Awards 2021 banner.png

Hello all,

The Wikidata ecosystem is a huge galaxy of exciting content, tools, projects, powered by the communities as well as organizations working with the software and the data. For seven years, people are gathering, starting projects, developing tools, improving the editors' workflows, filling various gaps, working all together to give more people more access to more knowledge.

In the frame of the WikidataCon 2021, we are organizing the WikidataCon community awards to celebrate the work of people and groups involved in Wikidata, and highlight some projects nominated by the community.

Until October 10th, you can participate and nominate one or several Wikidata-related projects that you like, that are useful for you or for the community. Such a project can be for example: a community gathering or other initiative that led to great results (WikiProject, event, editathon…), a tool (gadget, script, external tool…) or any other action that led to improving Wikidata’s data, the workflow of its editors or the outreach.

The nomination process is taking place publicly and collaboratively on this talk page. You can also help improving the description of projects that are already nominated. After October 10th, the awards committee will select a few projects that particularly caught their attention, and will present them during the Wikidata community awards ceremony taking place during the first day of the WikidataCon.

In order to reach out to the broader audience as possible, feel free to share this message on the talk pages, social media groups or other channels where you are active and people from various groups and experiences on Wikidata can participate. Thanks in advance for your help! Cheers, Lea Lacroix (WMDE) (talk) 08:53, 22 September 2021 (UTC)[]

Make a wiki item for an institute[edit]

Greetings to everyone, I'm here to know the truth (wikipedia item) is against the law to do the institute as a subject? So.. Can I make an item for an institute? If the answer is yes. Will you help me with the ready-made template.. --Ahmadkurdi44 (talk) 18:54, 22 September 2021 (UTC)[]

Is there a good reason such edits are technically even possible?[edit]

[4]. This whether the page is protected or not. --- Jura 20:14, 22 September 2021 (UTC)[]

@Jura1: No she is not. If I see correctly in the side protection log, the last protection ended on February 21, 2020. --Gymnicus (talk) 20:43, 22 September 2021 (UTC)[]
Would it be possible to prevent edits like this with abuse filters? --Marsupium (talk) 09:24, 23 September 2021 (UTC)[]

Wikidata birthday: participate in a birthday present![edit]

Hello all,

As you may know, Wikidata was launched on October 29, 2012, and every year in October, we celebrate the anniversary of the project with birthday wishes, events and cake. As we will celebrate the 9th birthday of Wikidata this year, we wanted to share with you how you can get involved in this event!

1. Prepare a birthday present

Every year for the birthday, people prepare some presents for the Wikidata community. These presents can be useful, fun or interesting: a new Wikidata tool, a new WikiProject, a logo or another piece of art, a blog post, an important community discussion… they can be worked on alone or in collaboration with other people.

If you have ideas for a Wikidata birthday, now is the perfect time to start thinking about it, and finding other people to work with you! Once the present is ready to be announced, you can add it to this page. If you’re looking for inspiration, you can check what was done for the previous anniversaries in 2015, 2016, 2017, 2018, 2019, 2020.

If you would like to present your gift to the community, you will also have the opportunity to present it during the birthday celebration at WikidataCon 2021. More information will follow soon.

2. Attend the online WikidataCon 2021

The WikidataCon 2021 will take place online on October 29-31, and gather the Wikidata community around its theme, a sustainable future for Wikidata.

Feel free to check the program, the side events, the onsite gatherings, the registration, how to get involved, and the program Reimagining Wikidata from the margins.

3. Join or organize other events

Other events will take place onsite or online around the WikidataCon. For example, you can check:

  • The side events of the conference (events related to Wikidata and Wikibase that are not necessarily organized by the WikidataCon organizers and take place before or after the conference);
  • The preconference Transbordados, a series of events and trainings that will take place online, in Spanish and Portuguese;
  • If the local health measures allow it, some people will organize onsite gatherings to attend the WikidataCon together;
  • You can also organize a birthday event with your community or in your area: if so, feel free to add details on this page;
  • To connect with other people organizing Wikidata-related events, feel free to join the Wikidata Events Telegram group.

If you have any questions or need support, feel free to contact me or to write a message on this talk page.

Looking forward to interacting with you all around the Wikidata birthday!

Best regards, -Mohammed Sadat (WMDE) (talk) 08:25, 23 September 2021 (UTC)[]


Q6408072 and Q55616660 are same. please Merging these page. thanks.M.Nadian (talk) 10:00, 23 September 2021 (UTC)[]

done BrokenSegue (talk) 14:12, 23 September 2021 (UTC)[]

26 million articles to go...[edit]

I created a new tool for improving our scientific articles with main subjects 2 weeks ago. Since then, over a million articles have now got their first main subject!

We only got ~26 mil. left (out of a total of 73 mil. articles).

The tool uses some bandwidth, so recently I improved it, so it can run in the WMF Cloud Services Kubernetes platform also.

Feel free to help out :)--So9q (talk) 11:29, 23 September 2021 (UTC)[]

Don't many papers submitted come with "tags" e.g. this paper is tagged as "randomized algorithms", "rejection method" and "Random number generation". I wonder if we could import those in a sensible way. BrokenSegue (talk) 13:57, 23 September 2021 (UTC)[]

Constraint for "qualifiers should co-exist"?[edit]

relevant date for copyright (P9905) has just been created (hooray!). This property is used a qualifier on copyright status (P6216) statements. However, it is only meaningful when a determination method (P459) is also given.

The following query shows items where relevant date for copyright (P9905) is used, but at least one of determination method (P459), applies to jurisdiction (P1001) or public domain date (P3893) is not given:

# Editions with a copyright status that has a relevant date but not a determination method
SELECT DISTINCT ?work ?workLabel ?methodLabel ?whereLabel ?relevantDate ?pdDate WHERE {
  ?work wdt:P31/wdt:P279* wd:Q3331189 .
  hint:Prior hint:gearing "forward".

  ?work p:P6216 ?statusStatement .
  ?statusStatement pq:P9905 ?relevantDate.
    ?statusStatement pq:P459 ?method .
    ?statusStatement pq:P1001 ?where .
    ?statusStatement pq:P3893 ?pdDate.
    ?statusStatement pq:P459 ?where .
    ?statusStatement pq:P1001 ?method .
    ?statusStatement pq:P3893 ?pdDate .
  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
Show results

But, is there are way to formally specify such a constraint with property constraint (P2302)? Inductiveload (talk) 14:01, 23 September 2021 (UTC)[]