Wikidata:Project chat

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

• Frequently asked questions
• Requests for deletions
• Merging instructions
• Report a technical problem
• Requests for protections, undeletions and other matters requiring administrator attention. To report vandalism, click here.

For realtime chat rooms about Wikidata, see Wikidata:IRC.

Start a new discussion

• Afrikaans
• العربية
• беларуская
• беларуская (тарашкевіца)
• български
• Banjar
• বাংলা
• brezhoneg
• bosanski
• català
• کوردی
• čeština
• словѣньскъ / ⰔⰎⰑⰂⰡⰐⰠⰔⰍⰟ
• dansk
• Deutsch
• Zazaki
• dolnoserbski
• Ελληνικά
• English
• Esperanto
• español
• eesti
• فارسی
• suomi
• føroyskt
• français
• Nordfriisk
• galego
• Alemannisch
• ગુજરાતી
• עברית
• हिन्दी
• hrvatski
• hornjoserbsce
• magyar
• հայերեն
• Bahasa Indonesia
• interlingua
• Ilokano
• íslenska
• italiano
• 日本語
• Jawa
• ქართული
• қазақша
• ಕನ್ನಡ
• 한국어
• kurdî
• Latina
• lietuvių
• latviešu
• Malagasy
• Minangkabau
• македонски
• മലയാളം
• मराठी
• Bahasa Melayu
• Mirandés
• مازِرونی
• Nedersaksies
• नेपाली
• Nederlands
• norsk bokmål
• norsk nynorsk
• occitan
• ଓଡ଼ିଆ
• ਪੰਜਾਬੀ
• polski
• پنجابی
• português
• Runa Simi
• română
• русский
• Scots
• davvisámegiella
• srpskohrvatski / српскохрватски
• සිංහල
• Simple English
• slovenčina
• slovenščina
• shqip
• српски / srpski
• svenska
• ꠍꠤꠟꠐꠤ
• ślůnski
• தமிழ்
• తెలుగు
• ไทย
• Tagalog
• Türkçe
• українська
• اردو
• oʻzbekcha / ўзбекча
• Tiếng Việt
• Yorùbá
• 中文

Edit

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 2026/07.

SpBot archives all sections tagged with {{Section resolved|1=~~~~}} after 1 day and sections whose oldest comment is older than 7 days.

Need of a process to become a mentor

Since just recently, we have the possibility to use the Mentorship features like many Wikipedias already have. However, currently we have self enrollment disabled. So either we need to turn that on, or we should add how to request to become a mentor on Wikidata:Requests for permissions. I suggest that we do the first one, and the current limit of a 365 days old account with at least 500 edits seems like a reasonable number to start with. Ainali (talk) 13:29, 23 June 2026 (UTC)

For comparison Event Organizers have no formal requirements for experience on Wikidata whatsoever. One is bound to have instances of the blind leading the deaf, to use a metaphor.

I see no use-case for having people request mentor rights. Since edits are easy to farm on Wikidata, the emphasis should be on account age, not edit count. It's in the spirit of Wikipedia to give users a lot of freedom even if that leaves room for messing up, so if an enthusiastic user with a half year old account wants to have a go at this, that seems ok to me.

One of the worries on Wikidata is new users making edits by the thousands, and not reading through all of the documentation or asking others for advice. Establishing a feedback cycle early can only be a good thing. Infrastruktur (talk) 14:26, 25 June 2026 (UTC)

I don't understand what you really want. Are you totally against users getting mentor rights, or against it being manually handed out? Ainali (talk) 15:43, 26 June 2026 (UTC)

The latter. People can self-enroll. I just find it strange to expect that mentors should have Wikidata experience when there currently is no such expectations of Event Organizers. Infrastruktur (talk) 15:57, 26 June 2026 (UTC)

Thanks, I agree. So would you rather go with even longer time than 365 days? Ainali (talk) 20:18, 26 June 2026 (UTC)

Nah. They don't need to be experts. I would be ok with as low as 6 months of experience for either role. Infrastruktur (talk) 20:39, 26 June 2026 (UTC)

Great! Then all we need is an admin checking the Editors who meet all eligibility criteria are automatically eligible to enroll as mentors on Special:CommunityConfiguration/Mentorship. Ainali (talk) 10:00, 27 June 2026 (UTC)

I'm supportive of this. We can always change the configuration later if need be, but let's give it a try with the default settings. Yirba (talk) 10:04, 27 June 2026 (UTC)

Alright, discussion has been open for 5 days, so I guess it's time. Mentorship has been enabled. Settings are still TBD. Currently configured to allow let accounts over 365 days old with 500 edits to self-assign themselves the mentor role, so give it a try. Mentees are currently automatically removed from mentorship after reaching either 500 edits or having an account registered for 30 days. Infrastruktur (talk) 18:05, 28 June 2026 (UTC)

Thanks for doing that. I won't be participating myself, but I'd be curious to hear what mentors and mentees on Wikidata make of this feature. Yirba (talk) 18:12, 28 June 2026 (UTC)

Thoughts on transcluding Special:ManageMentors as described at [1] so it can be added to Wikipedia:Growth Team features/Mentor list (Q14339834)? IntensionalLogician (talk) 21:38, 3 July 2026 (UTC)

Sounds like a good idea. Ainali (talk) 06:42, 4 July 2026 (UTC)

This RFC seems relevant. Bovlb (talk) 02:12, 9 July 2026 (UTC)

What are your thoughts regarding Trainers vs Mentors? It seems like the requirements are similar but they would serve different purposes: Trainers being more organization and group facing while mentors are more individual-new-user facing. They could probably be combined into a single thing that does both as well, but the workloads seem distinct enough that they probably shouldn't. -- IntensionalLogician (talk) 13:13, 12 July 2026 (UTC)

Asparagus fallax (Q1587417) vs. Asparagus fallax (Q2866764)

Hi Guys! This is the same species. Can somebody sort this out? I am not to familiar with Wikidata. Regards. Oesjaar (talk) 17:14, 5 July 2026 (UTC)

Hi @Oesjaar. Is Asparagus acutifolius (Q1587417) not actually Asparagus acutifolius, which is a different species to Asparagus fallax? Certainly, taxon databases seem to distinguish between the two. This is way out of my area of expertise though, so you might like to ask at Wikidata talk:WikiProject Taxonomy. --Yirba (talk) 17:38, 5 July 2026 (UTC)

@Oesjaar: What evidence is there that these are the same species? You added Asparagus fallax to Asparagus acutifolius (Q1587417) as the Afrikaans label, but that's the only instance of that name across any of the languages present. The Afrikaans article you linked to is even titled Asparagus acutifolius. What's going on here? — Huntster (t @ c) 18:21, 5 July 2026 (UTC)

Huntster & Yirba, both entries in Wikidata have the same name Asparagus fallax which does not make sense. One cannot use the same name for different species. Either the two articles are joined as one or one article's name changed to another species name. I am not a Wikidata expert. Regards. Oesjaar (talk) 06:16, 6 July 2026 (UTC)

@Oesjaar: I'm very confused by your statement. Asparagus fallax (Q2866764) is Asparagus fallax, Asparagus acutifolius (Q1587417) is Asparagus acutifolius. A. fallax only appears in the latter due to this edit where you added it as a label. Did you not realize you did this? — Huntster (t @ c) 07:02, 6 July 2026 (UTC)

They have different articles in several languages so they appear to be different species. Secretlondon (talk) 07:27, 6 July 2026 (UTC)

Thanks. I have reverted the Afrikaans label edit. Yirba (talk) 07:32, 6 July 2026 (UTC)

Thanks guys. It is sorted, appreciated. Oesjaar (talk) 16:27, 6 July 2026 (UTC)

Clarification on "structural need" for an author item

Hi,

I have a question about how the "structural need" criterion in Help:Notability should be interpreted.

Suppose an author's book already has a Wikidata item, but the author does not. The author has an official ISNI assigned through an authorised ISNI Registration Agency. Without an author item, neither the ISNI nor the existing book item can be linked through a single author entity.

In a situation like this, would creating an item for the author normally be considered sufficient to satisfy the "structural need" criterion, or would independent biographical coverage still be expected?

I'm only interested in understanding how this part of the policy is generally interpreted.

Thanks.

NewCosmicLibrarian (talk) 23:17, 5 July 2026 (UTC)

Yes, you absolutely should create an item for the author in this case. This is probably the prime example of "structural need". Dexxor (talk) 14:52, 6 July 2026 (UTC)

@Dexxor:, the question was asked generally, but would your answer be different if the questioner was the author and they had created the item for the book? TSventon (talk) 16:58, 6 July 2026 (UTC)

Yes, this changes things. If the book wasn't notable in the first place, then you can't use it to justify creating an item for the author. Dexxor (talk) 09:35, 7 July 2026 (UTC)

Presuming the item concerned is Las puertas del abismo (Q137333080) Secretlondon (talk) 15:11, 7 July 2026 (UTC)

Difference between instances of sourcing circumstance (Q104637420)

What are the differences between presumably (Q18122778), possibly (Q30230067), and probably (Q56644435), and implied (Q130436107)? Their definitions overlap, and it's not apparent why I should choose one over the other. Ewpa (talk) 17:55, 6 July 2026 (UTC)

I looked at the English labels and they say use presumably (Q18122778) if the source says presumed, that seems fairly clear. possibly (Q30230067) is less explicit, I would expect you use it when the source says possibly. The labels are

• presumably (Q18122778), the source specified value and explicitly stated that value is presumed. For use with qualifier P1480.
• possibly (Q30230067), for use with qualifier P1480 or P5102
• probably (Q56644435), the source specified value and explicitly stated that value is probable. For use with qualifier P1480 or P5102.
• implied (Q130436107), implied – should be used with qualifier P1480 to indicate the source specified a value and explicitly stated that value or connection is implied, but not confirmed

TSventon (talk) 18:15, 6 July 2026 (UTC)

Wikidata weekly summary #739

Here's your quick overview of what has been happening around Wikidata in the
week leading up to 2026-07-06. Missed the previous one? See issue #738.

Discussions

• Closed request for adminship:
  • 1Veertje - Closed as Successful
  • Ziv - Closed as Successful
• New requests for permissions/Bot:
  • gov-wikidata-match 1 - Task/s: Link existing GOV (historical gazetteer) building-like objects to existing Wikidata building-like objects.
  • AgenticCommonsBot 1 - Task/s: Add Open Library ID (P648) (OpenLibrary author ID) to Wikidata items about people, only where the OpenLibrary author and the Wikidata person have been confirmed to be the same individual, with a source on every statement.
• Closed request for permissions/Bot:
  • JigildikBot - Approved
  • SonalBot - Approved
  • ias-kbase - Approved

Events

• Upcoming events:
  • The Wikidata Platform (WDP) monthly office hour takes place tomorrow, 7th July at 16:00 UTC. These office hours are focus on the Blazegraph migration work, and the team is inviting all community members to join the discussion. Register via the event page
  • The next Wikidata+Wikibase office hours will take place on Wednesday, 18:00 CEST (in your timezone), 08th July 2026 in the Wikidata Telegram group. The Wikidata and Wikibase office hours are online events where the development team presents what they have been working on over the past quarter, and the community is welcome to ask questions and discuss important issues related to the development of Wikidata and Wikibase.
  • Sandbox sessions: Wikidata Reference Validator Tool - improving Wikidata references - Aug 15, 17:00 - 18:00 UTC.
  • Wiki Education + AASLH: Documenting America with Wikidata, a collaboration between Wiki Education and the American Association for State and Local History is offering six hour-long meetings held weekly to introduce attendees to Wikidata and improve content for Wikidata and Wikipedia on American history. Registration ends: July 17, 23:59 Pacific Time (PT)

Press, articles, blog posts, videos

• Blogs: (fr) Le point sur la diversité de genre dans les articles in *Regards sur l'actualité du mouvement Wikimedia*
• Papers:
  • Spatiotemporal-Heuristic A* Search for Pruning - Historical Influence and Succession Paths in Wikidata - Paper, by Michael James Liman (2026).
  • Wikontic: Constructing Wikidata-Aligned, Ontology-Aware Knowledge Graphs with Large Language Models - Paper by Chepurova et al., (2026).
• Videos: (Español) Introduction to Wikidata - Dr. Claudia De Souza and Prof. Dinah Wilson Fraites introduce Wikidata with Spanish-slides.

Tool of the week

• Zelph: Offers powerful analysis capabilities for the complete Wikidata knowledge graph while remaining adaptable for any semantic domain.
• World Cup Heights: An interactive data‑viz app that shows the heights of FIFA World Cup players, arranged so you can quickly compare teams, positions, and overall height distributions.

Other Noteworthy Stuff

• Call for local partners: WikidataCon 2027: Planning is underway for the next hybrid WikidataCon, and Wikimedia Deutschland is seeking a local affiliate to co-organize the on-site event. Expressions of Interest are due by 13 July.
• The Wikidata Platform team has published the 8th issue of its monthly newsletter. This edition includes updates on the migration of WDQS backend away from Blazegraph, including a transition into the next phase Initial Implementation (pilot) phase following completion of the installation phase. This update also include some Documentation on Query Rewrites the team will like the Wikidata community and all WDQS users to help review and leave feedbacks. see WDP monthly newsletter for more details

Newest properties and property proposals to review

• Newest General datatypes:
  • prisoner number (number or identifier assigned to a prisoner, to be used as a qualifier for P2632 (place of detention))
  • photographer (person who took this photograph)
  • ‎Gmedia rating (video games) (video game ratings of the Saudi Arabian rating system)
  • visual style (the visual style used in a video game (or other creative work))
  • garden (garden of this subject)
  • objects (category theory) (the precise objects of the category (category theory))
  • morphisms (the precise morphisms of the category (category theory))
  • terminal object (the precise terminal object of the category (category theory))
  • initial object (the precise initial object of the category (category theory))
  • products (category theory) (the precise products of the category (category theory))
  • coproducts (category theory) (the precise coproducts of the category (category theory))
  • isomorphisms (the precise isomorphisms of the category (category theory))
  • monomorphisms (the precise monomorphisms of the category (category theory))
  • epimorphisms (the precise epimorphisms of the category (category theory))
  • regular monomorphisms (the precise regular monomorphisms of the category (category theory))
  • regular epimorphisms (the precise isomorphisms of the category (category theory))
  • left adjoint (the left adjoint of the functor)
  • enriched over (the base monoidal category, bicategory, etc. of the enriched category)
  • monoidal category operation (the “tensor product” bifunctor of the monoidal category)
  • 2-morphisms (the precise 2-morphisms of the 2-category)
  • taxon footprint (image of a footprint of this taxon)
• Newest External identifiers: Sri Lankan Parliament member ID, ‎Taiwan Central News Agency news tag ID, Model-Kartei.de ID, Wildenstein Plattner Institute ID, Web The Television work ID, Web The Television person ID, U-NEXT person ID, U-NEXT video ID, Sokmil director ID, Sokmil idol ID, socrealizm.com.ua person ID, Répertoire des films classés film ID, ‎regjeringen.no ministry ID, ‎regjeringen.no politician ID, Objectif plumes work ID, National Operator Code (UK), ‎Extended Defence of Britain Database ID, Pacelli edition person ID, Myludo game ID, MyDirtyHobby ID, Maoyan film ID, ‎Los Angeles Historic-Cultural Monument number, Letterboxd writer ID, KSW organization ID, KSW person ID, KSW place ID, 1SCREEN movie ID, 1SCREEN Person ID, ‎Artistas Visuales Chilenos propertyvalue ID, ‎Canadian Trademark Registration Number, occitanica.eu ID, Hancinema company ID, ZicTrad person ID, H-NEXT Person ID, FOD video ID, Farzonagon ID, d Hits artist ID, ABEMA video ID, 17LIVE ID, ‎regjeringen.no government ID, Pure Adult actress ID, Pure Adult studio ID, ‎GBIF taxon ID, Digital portrait index person (GND) ID, torial username, Pontifical Urbaniana University ID, CatDat category ID, CatDat functor ID, CatDat category property ID, CatDat functor property ID, ‎Samtrafiken-vehicle ID, GIGA actress ID, Zen Pictures actress ID, Olyrix artist ID, AutorÒc 16-18 author ID
• New General datatypes property proposals to review::
  • dielectric strenght (maximum electric field a material can withstand before electrical breakdown occurs)
  • about page URL (URL to the official about page)
• New External identifier property proposals to review: Rakuten Books author ID, Rakuten TV adult Person ID, Identifiant sur le Western Orthodox Wiki, Le GRoG ID, NHK On Demand Program ID, Centre for Computing History exhibit reference ID, GCD character ID, Identificador de agente econômico na ANCINE, 1SCREEN Program ID, Architectenregister-id, LUDAP ID, TakaWiki ID, Terceiro Tempo Que Fim Levou ID, BiDünyaFilm film ID, CNUC code, Steady username, NHK ONE Program ID, SongLyrics.com song ID, Braunschweiger ProfKat ID, ClayArena shooter profile ID, Open Collective ID, Anobii author ID

You can comment on all open property proposals!

Development

• The development team are at an offsite retreat this week.

You can see all open tickets related to Wikidata here. If you want to help, you can also have a look at the tasks needing a volunteer.

Weekly Tasks

• Add labels, in your own language(s), for the new properties listed above.
• Contribute to the showcase Item and Lexeme above.
• Govdirectory weekly focus country: France
• Summarize your WikiProject's ongoing activities in one or two sentences.
• Help translate or proofread the interface and documentation pages, in your own language!
• Help merge identical items across Wikimedia projects.
• Help write the next summary!

· Previous issue · Full report · Unsubscribe . JosefAnthony talk ·MediaWiki message delivery (talk) 18:04, 6 July 2026 (UTC)

tabular population (P4179)

Hello! At the Basque Wikimedians User Group we are working on a project to upload data from the World Bank to Commons, and describe it at Wikidata. However, there's no direct property for linking a tabular data (WB-WDI-SP.DYN.LE00.IN data series (Q140402222)). We have some very partial tabular data properties, and the only one partially fitting is tabular population (P4179). Should we proceed with it, or should we ask for a more general property? Is there anoter way to do this? Theklan (talk) 18:32, 6 July 2026 (UTC)

I have been thinking about this too. Perhaps a general "tabular qualifier" that we could use on any property would be a better solution than trying to create an alternative for them all. Ainali (talk) 19:00, 6 July 2026 (UTC)

A general "tabular data" property isn't just a better option? Theklan (talk) 20:20, 6 July 2026 (UTC)

I think perhaps we mean the same thing. This is how I imagine it:

nominal GDP	123,456,789 dollar edit point in time 2025-12-31 Tabular data Data:some-GDP-file.tab 0 references add reference
		add value

Human Development Index	0.781 edit point in time 2025 Tabular data Data:some-HDI-file.tab 0 references add reference
		add value

By using it as a qualifier on a property, we know exactly what kind of tabular data that file holds. (Expansion on my thinking.) Ainali (talk) 08:32, 8 July 2026 (UTC)

Yes, I would support a property like this. Please propose it at wikidata:Property proposal — Martin (MSGJ · talk) 16:46, 8 July 2026 (UTC)

Extract coords from osm properties

a bot should extract coords from osm and add it to coordinate location (P625) or similar coord properties. for example, embassy of Denmark in Ireland (Q104764355): no description has OpenStreetMap node ID (P11693) set for over 3 years but coord is still missing. RoyZuo (talk) 14:54, 7 July 2026 (UTC)

@RoyZuo: Hi, I think it's a good idea in theory, but unfortunately not allowed. Yirba (talk) 15:22, 7 July 2026 (UTC)

@Yirba copying from osm to wd or the other way around, manually or automatically, is all not allowed. that sounds bloody ridiculous.

then i have an idea: make a simulator (that simulates human browsing internet):

1. it clicks osm properties from wd, which opens osm, and wait until the url becomes this format https://www.openstreetmap.org/node/1250409636#map=19/53.332582/-6.262032
2. it then jumps from that to (53.332582, -6.262032) on google maps.
3. it then jumps from (53.332582, -6.262032) on google maps to (53.33258, -6.26203) (rounded to 5 dp, which is precision of 1.11 m) on google maps.
4. it then copies the coords to wd.

so the simulator copies not from osm but google maps to wd.

i cant imagine how people can obey osm's rules without doing this, since i technically cannot copy it from osm even by hand. RoyZuo (talk) 15:42, 7 July 2026 (UTC)

it seems the most relevant jurisdictions that protect database rights are only the eu and countries that align with them, and eu database rights last only 15 years.

so data older than 15 years can be copied? RoyZuo (talk) 16:00, 7 July 2026 (UTC)

Hi there! As others have said, you cannot directly automatically add OSM ID to WD nor use OSM data to add to WD. For instance, OSM may have "outdoor_seating=yes". That doesnt mean you can then add "has amenity" "outdoor seating" to the wikidata item using OSM as the source.

What ive done with the coordinate location field in Data2Go (ios app for Wikidata) is to query OSM with where you choose the coordinate to be for enclosing elements to add to the Wikidata item. You need to manually confirm both coordinate location and osm id but it makes it easy on one screen. GA Kevin (talk) 23:56, 7 July 2026 (UTC)

Extract coords from Google Maps Customer ID

Google Maps Customer ID (P3749)

is there a bot that does this?

i've been adding osm properties for years, thinking a bot would then harvest plenty of properties based on that relation. now i learnt that didnt work. i see no point in linking to osm anymore.

trying to find similar map properties, i found p3749. is a bot extracting coords and possibly other info from that?--RoyZuo (talk) 07:30, 9 July 2026 (UTC)

Date error

James Meadows Rendel (Q6139585): British civil engineer (1799–1856) cannot be accepted as an occupant of 10, Kensington Palace Gardens W8 (Q27083395): building in Kensington and Chelsea, London, England, UK as the earliest start date is the 19.century, but surely the century should encompass people living within that 100 year period. Sorry if this doesn't make sense! No Swan So Fine (talk) 21:54, 7 July 2026 (UTC)

Tracked in Phabricator
Task T168379

Hi, this is a known issue. Not great, I agree. Yirba (talk) 22:13, 7 July 2026 (UTC)

The contempory constraint has been broken for a long time it does not throw to start or end of periods when comparing them. I've seen it for years, you are seeing it for centuries. Bugs have been raised and ignored https://phabricator.wikimedia.org/T168379 is 9 years old Vicarage (talk) 22:16, 7 July 2026 (UTC)

It's been marked as "medium priority" for 9 years! @Lydia Pintscher (WMDE), is there anything was can do to progress this ticket? — Martin (MSGJ · talk) 09:38, 8 July 2026 (UTC)

The team that would need to work on this is currently busy with mobile editing, improving the visibility of WikiProjects and adapting the Query UI and constraints to the upcoming query service migration. But I will bump this up and we'll look into it once they have time to put into constraints again. Lydia Pintscher (WMDE) (talk) 12:01, 10 July 2026 (UTC)

Thank you — Martin (MSGJ · talk) 17:20, 10 July 2026 (UTC)

Gistly AI generated YouTube video summarizer

Any complaints if we remove this URL from items?

https://gist.ly/ Trade (talk) 07:20, 8 July 2026 (UTC)

Do you have an example on an item? Ainali (talk) 08:15, 8 July 2026 (UTC)

The only item I could find it on was Volodymyr Zelenskyy: Ukraine, War, Peace, Putin, Trump, NATO, and Freedom | Lex Fridman Podcast #456 (Q131681281). I don't think it's appropriate to list AI-generated text in the context of source material. Yet another candidate for the abuse filter that warns about poor references? Infrastruktur (talk) 08:29, 8 July 2026 (UTC)

I do not care about accuracy. My problem is that anyone with access to a LLM can generate these summaries on demand meaning it has no real uniqueness or value

It's the same reason i oppose the use of speech synthesis (Q16346) files for spoken text audio (P989) Trade (talk) 08:59, 8 July 2026 (UTC)

Notability of businesses and business chains

Hi all,

@WikiBayer: recently deleted Q138763371, without going through an RfD and notifying the item creator (me). I noticed this because I had the item on my watchlist. This item described Bingz, a minor fast-food restaurant chain which operates 7 locations in Canada.

I had created this Wikidata item so it can be referred to in OpenStreetMap so shops can be associated with their chain brand (e.g. [2] and [3]). I appreciate that OpenStreetMap isn't a Wikimedia project, however I was under the impression that Wikidata items can be created to hold data about OSM objects. This is also encouraged by OSM's "Name Suggestion Index" (although Bingz specifically is not yet in NSI).

I would like to get some clarification on notability for businesses so I can avoid putting work into items that would be deleted.

For this deletion, the admin cited Wikidata:Notability, which states that an item must fulfill one of three main criteria, one of which is "It refers to an instance of a clearly identifiable conceptual or material entity that can be described using serious and publicly available references."

It seems to me that basically any chain restaurant can be described by public references, as there's usually a "business opening" puff story in local media. Are these to be considered "serious" references? Examples for Bingz specifically could be [4] and [5] - obviously there is little critical reporting there, it is a reprint of what PR served them, though it confirms this place exists/existed at the time.

Previously, Q109377032 was deleted as "not notable" (brief RfD). I haven't seen the item before it was deleted, I only know of it because it was linked in OSM briefly. The store's website has a list of press references although again not much critical reporting - is this below the threshold for inclusion in Wikidata?

I am assuming places like Steve's Music are in the clear as their existence (although especially their closing) has been widely reported beyond reprinting PR releases. How about small chains like Craig's Cookies which was in the news for its workers unionizing? Is a former pharmacy important to a local community operated by a person with links to obituaries okay? If someone's going to delete those in the future, I would like to know so I can host the data I've researched elsewhere instead.

Is there a better place to ask this?

Thank you, Qviri (talk) 15:46, 8 July 2026 (UTC)

@Qviri Items need a reputable not selfcreated source Steve's Music doesn't it either. Even though OpenStreetMap has similarities to Wikidata, they are not the same. WikiBayer (talk) 16:09, 8 July 2026 (UTC)

Sorry, I can't understand "Steve's Music doesn't it either.", is a word missing? --Qviri (talk) 16:40, 8 July 2026 (UTC)

Steve's Music doesn't have the required sourcing either. Secretlondon (talk) 11:52, 9 July 2026 (UTC)

Steve's Music Store (Q138253811) has references to two online articles in CBC News (Q2931014) with reporting on people's reactions to its closing. Is further sourcing required? I am rather surprised this doesn't meet Wikidata's notability requirements, can you create an RfD so this can be discussed? I will adjust my Wikidata contributions if this turns out to be the case. --Qviri (talk) 13:57, 9 July 2026 (UTC)

@WikiBayer: can you respond to my questions about Bingz and Steve's please? I have provided references. Are you saying BlogTO and CBC are not reputable? What criteria are we using for "reputable"? Please note that I have no affiliation with these businesses other than having eaten at Bingz exactly three times and having edited the two locations on OSM, so WD:SELF doesn't apply. Qviri (talk) 23:23, 9 July 2026 (UTC)

@Qviri take some photos of the chain, upload them to commons using c:special:uploadwizard, add them all to a c:com:category, then link that category back to the wikidata item, and set c:com:depicts = the wikidata item of the chain.

this doesnt guarantee protection from wikidata deletionists, but if they still delete it, that's gonna piss off more people because that deletion messes up commons data. RoyZuo (talk) 19:03, 9 July 2026 (UTC)

"@WikiBayer recently deleted Q138763371, without going through an RfD and notifying the item creator (me)."

if this is true, this sysop should face consequences.

because his action already messes up commons.

out of curiosity, i checked https://commons.wikimedia.org/w/index.php?search=bingz&title=Special%3AMediaSearch&type=image . strangely, it finds 1 file, even though bingz doesnt appear anywhere on its file page. i guess it might be because someone added 西少爷 bingz's chinese name to the wd item so commons somehow was able to find it.

with that i found http://industry.people.com.cn/n1/2018/0326/c413883-29889936.html https://vibes.8world.com/food/bingz-chinese-crispy-burger-singapore-2313456 https://cn.chinadaily.com.cn/a/202404/26/WS662b2287a3109f7860ddb1b4.html ...

deleting without going through an RfD is serious abuse of sysop tools. it wastes other users' time and effort. RoyZuo (talk) 19:12, 9 July 2026 (UTC)

I just want to leave a note of support, as I'm someone else who sometimes puts data in Wikidata as a way to store information about items in OpenStreetMap. I thought that having a place of storing structured data (especially about real-world entities) to be used across many projects was the whole point of Wikidata. I totally get that not every little company out there might be "notable" and that the notability criteria are under discussion to maybe tighten them up a bit. But there needs to (1) be clear guidance about how to establish whatever that baseline level of notability is, especially for common things like stores and local chains, and (2) a process that at the very least notifies creators before deleting their work when it looks like they're trying to add data in good faith. — PeterCooperJr (talk) 19:27, 9 July 2026 (UTC)

and i just found Q133829650: no description was also deleted by User:MisterSynergy out of procedure. how is a publicly listed company not notable? what's all this out-of-procedure deletion here? RoyZuo (talk) 19:33, 9 July 2026 (UTC)

Scale back your aggression. There is no "out of procedure" deletion here at Wikidata per Wikidata:Deletion policy. The vast majority of deletions is not being discussed at WD:RfD or elsewhere, and there is no policy which would prescribe a discussion. —MisterSynergy (talk) 19:56, 9 July 2026 (UTC)

vibes.8world.com and co. are reputable sources are you really going to claim that here? WikiBayer (talk) 22:40, 9 July 2026 (UTC)

@WikiBayer YES.

Do you know what you are talking about when you try to deflect the blame to a handful of random news articles I found in 5 minutes?

1. I dont see the equivalent of Wikipedia:Reliable sources/Perennial sources (Q59821108) on wikidata. Who decided these websites to be unreliable?
2. Channel 8 News (Q15940701): Chinese-language news department based in Singapore, rated by zh:Wikipedia:來源評級/8視界新聞 to be "generally reliable", including its "vibes" subdomain, quote: "旗下娛樂網站亦為通常可靠".
3. people.cn (Q5967721), which is internet branch of People's Daily (Q54340).
4. China Daily (Q851422).

When you're done discrediting those 3 news agencies, you can continue doing the same for the following:

1. https://www.news.cn/food/20230531/218edfe737844a228e5ce78a0d643de8/c.html
2. http://m.cnwest.com/xian/a/2021/06/27/19768232.html
3. https://app.dahecube.com/nweb/news/20240427/198292n70d2303b3b3.htm
4. https://www.bjnews.com.cn/detail/162401019614610.html
5. https://finance.sina.cn/china/gncj/2023-07-26/detail-imzcymix7231277.d.html
6. https://zqb.cyol.com/html/2014-11/17/nw.D110000zgqnb_20141117_1-09.htm
7. https://i.ifeng.com/c/89OJnLiMcb1

RoyZuo (talk) 08:26, 10 July 2026 (UTC)

Merge?

Should Q98273564 and Q1013867 be merged? ~2026-26390-87 (talk) 16:14, 8 July 2026 (UTC)

Yes, I think so.  Done — Martin (MSGJ · talk) 16:41, 8 July 2026 (UTC)

speech processing Q3358061

I speech processing (Q3358061) should be an instance of a new class call voice processing for which singing processing is also an instance. QueueSevenM (talk) 07:38, 9 July 2026 (UTC)

Sounds reasonable, go ahead — Martin (MSGJ · talk) 11:19, 9 July 2026 (UTC)

Okay, I have done that. QueueSevenM (talk) 19:22, 9 July 2026 (UTC)

Lexeme:L12213

The fifth sense in bid (L12213) has been given an English gloss that seems inappropriate in tone: "some weird causative usage, not unlike "call"". The use of the word "weird" makes me believe that the tone is inappropriate, and the gloss does not even clarify what sense of "call" it is referring to. How should this gloss be fixed? Intolerable situation (talk) 08:10, 9 July 2026 (UTC)

@Intolerable situation: The example sentences given at this page under "bid.02" could be enlightening; it seems like this sense is within the realm of finance/investing, and indeed there is a possibly helpful definition of wikt:bid up, if it gives you an idea of what to adjust the gloss to. Mahir256 (talk) 09:55, 9 July 2026 (UTC)

I am still not sure how to change the gloss. Intolerable situation (talk) 22:20, 9 July 2026 (UTC)

Merge

Can Q113989189 be merged into Q106157103? Thanks! --Like the windows (talk) 21:11, 9 July 2026 (UTC)

 Done Yirba (talk) 22:50, 9 July 2026 (UTC)

Proposal: Advisory message for new accounts creating many items

Related discussion:

• Wikidata:Administrators'_noticeboard/Archive/2023/02#New_items_creation_to_be_restricted_for_anonyms_and_for_not_autoconfirmed_users?
• Wikidata:Project_chat/Archive/2022/02#Stop_allowing_unregistered_edits
• Wikidata:Project_chat/Archive/2021/12#Allowing_IP's_to_add_data,_not_change_data
• Wikidata:Project_chat/Archive/2017/08#Restrict_merging_rights_to_autoconfirmed_users
• User:Bovlb/Throttle item creation by anonymous users
• User talk:Bovlb/Throttle item creation by anonymous users

I propose creating an AbuseFilter that displays a custom warning message when an unconfirmed account attempts to create a large number of new items within a short period. The filter would not prevent item creation, but would ask the user to confirm that they wish to continue.

Motivation

Creating a new Wikidata item requires making a series of good judgements about many different things. These are skills that are often easier to develop by first reviewing and improving existing items.

Some new contributors begin editing by creating large numbers of new items with little or no prior experience of improving existing ones. This frequently leads to avoidable problems, and many of these items end up being deleted.

New contributors who misunderstand an aspect of item creation often repeat the same mistake across multiple items before receiving feedback. A brief reminder when a newcomer begins creating many items could help to encourage a more gradual learning path.

This intervention is intended as guidance rather than a restriction. Good-faith contributors would still be able to continue creating items after acknowledging the warning.

I believe this is a modest intervention with a low cost to legitimate contributors and the potential to improve the quality of new items, provide a better onboarding experience for new users, and reduce volunteer time spent rescuing or deleting items.

Threshold

For the purposes of discussion, I suggest that the warning be triggered when an unconfirmed account (including temporary accounts) attempts to create its 10th new item within a 24-hour period. I welcome suggestions for different thresholds.

Message

You have created several new items in a short period. Creating new items requires making good judgements about whether an item belongs in Wikidata and how it should be represented. We recommend that new contributors spend some time reviewing and improving existing items to develop these skills before creating many new items. Please consider taking a break from item creation.

Discussion

Bovlb (talk) 23:57, 9 July 2026 (UTC)

• I'd like to see statistics on how big the problem is, but gererally I'm in favour of hard rather than advisory limits, and a longer timeframe than 24 hours. Genuine new users should be encouraged to flesh out their new entries, and there is little urgency in creating items here, and I suspect many large blocks of items are being done by sockpuppets who we already have problems withVicarage (talk) 04:16, 10 July 2026 (UTC)

  Looking at accounts registered during the first six months of this year, there were 327 users who created 10 or more items within the first 24 hours after registration, an average of about 1.8 users per day.

  A warning triggered on the 10th creation would have been shown before 2,943 item creations (about 16 per day). Approximately 40% of those items have since been deleted.

  A list of the affected users is available at User:Bovlb/unconfirmed users with 10 creations. Bovlb (talk) 06:47, 10 July 2026 (UTC)

  Setting the threshold to 5 would have affected 1,162 users (~6 per day). Bovlb (talk) 06:55, 10 July 2026 (UTC)

  47 of those users (~14%) are currently blocked. Bovlb (talk) 14:56, 10 July 2026 (UTC)

  It's interesting that 40% doesn't give a good picture. It's mostly 100% kept or 100% deleted. Secretlondon (talk) 17:19, 11 July 2026 (UTC)

  Yes. With bulk item creation, it's very much "make mistakes with one" -> "make mistakes with all".

  Also, many of the items not deleted are actually still at risk of deletion, and a few were rescued by interventions. From my experience, most of these bulk item creators never return to the project. Bovlb (talk) 17:59, 11 July 2026 (UTC)

  To clarify, the bimodal deletion rate is probably a real phenomenon, but it's also affected by how administrators respond. Bovlb (talk) 18:16, 11 July 2026 (UTC)

Very useful overview. Any chance of posting a 24 hour histogram for some eyecandy? Infrastruktur (talk) 18:03, 11 July 2026 (UTC)

Like this? Bovlb (talk) 19:55, 11 July 2026 (UTC)

That will do nicely, thanks. Infrastruktur (talk) 20:15, 11 July 2026 (UTC)

I like this and think we should try it. Is there some helpful link we could provide as well in the message to get the user to learn more? Ainali (talk) 06:57, 10 July 2026 (UTC)

Well, there’s WD:QUALITY, but I was trying to avoid linking to one of my essays today. Bovlb (talk) 07:10, 10 July 2026 (UTC)

+1 LydiaPintscher (talk) 11:57, 10 July 2026 (UTC)

I added a link to notability policy. Bovlb (talk) 14:57, 10 July 2026 (UTC)

How would it work with OpenRefine, or with whatever it is that creates items such as Q139806853? Peter James (talk) 16:02, 10 July 2026 (UTC)

That’s a good question. Items created by unconfirmed users via OpenRefine would be treated the same. OpenRefine can submit edits in a number of ways, and, as far as I know, it is not generally possible to determine from the edit itself that the tool was used.

My understanding is that the MediaWiki API requires a retry after an AbuseFilter warning, so I suspect OpenRefine would either fail or alert the user, although I have not tested that.

Do you have any thoughts on how this ought to affect unconfirmed OpenRefine users? To your knowledge, are a lot of new editors starting with OpenRefine?

For comparison, this would not affect QuickStatements users, since QuickStatements already requires users to be confirmed. Bovlb (talk) 16:16, 10 July 2026 (UTC)

Automated detection and warnings are a brilliant alternative to more heavy handed measures. Note that temporary accounts are never autoconfirmed and lasts at most 90 days, so maybe the filter should only apply when the number of edits is below what would have granted autoconfirmed status for a registered account? Infrastruktur (talk) 21:03, 10 July 2026 (UTC)

Do you have an example of such a filter? Bovlb (talk) 21:44, 10 July 2026 (UTC)

No, and I don't know that much about abuse filters. Infrastruktur (talk) 08:07, 11 July 2026 (UTC)

 Support with hard limit rather than soft limit per Vicarage. IntensionalLogician (talk) 22:36, 10 July 2026 (UTC)

I feel like it might be helpful to get some feedback from event organizers, but I don't know where I could usefully leave a pointer to this discussion. Bovlb (talk) 23:38, 10 July 2026 (UTC)

As one holding the event organizer rights, I think this is a good proposal. Admittedly, I haven't organized any events for newcomers since getting the rights, but I think it is a good practice to learn by improving existing items rather than creating new ones. Ainali (talk) 08:27, 11 July 2026 (UTC)

Property proposals from website operators

In the past week we've had a few external ID property proposals created by the operators of those websites:

• Wikidata:Property proposal/BiDünyaFilm film ID
• Wikidata:Property proposal/SongLyrics.com song ID
• Wikidata:Property proposal/AnimeOshi anime ID
• Wikidata:Property proposal/Mizan al-Arab poet ID

Even though these users have declared their affiliations, this doesn't exactly sit right with me. Especially from user accounts that have no prior activity on Wikidata. What does everyone else think? --Yirba (talk) 12:51, 10 July 2026 (UTC)

I think we can accept their suggestions, at least they will understand their subject matter, and are probably prepared to put in the hard work implementing them, but I'd expect strong support from our community, including WP articles that put their sites in context. Vicarage (talk) 13:30, 10 July 2026 (UTC)

This question ties in well with some concerns I've had with the notability policy reform RFC regarding what constitutes an "independent authority." While obviously not all external identifiers confer notability, tools like mix-n-match use the existence of multiple external IDs as a proxy for notability for their creation candidates feature, and some users at RFD use the number of (non social media) external IDs as an imperfect heuristic for notability. Thus, having non-notable external IDs negatively impacts the item creation and RFD process. Per Vicarage, minimally the database itself the external ID refers to should be notable enough for an item. IntensionalLogician (talk) 22:56, 10 July 2026 (UTC)

The recent case of Catholic Encyclopedia Wikitólica ID (P14483) seems relevant here: Wikidata:Property proposal/Enciclopedia Católica Wikitólica ID, Wikidata:Administrators'_noticeboard/Archive/2026/06#Spam, Wikidata:Properties_for_deletion#Property:P14483 Bovlb (talk) 18:07, 11 July 2026 (UTC)

Wikidata item for Quantumize need Q######## identifier

I am trying to create a Wikidata item for Quantumize, a cybersecurity software company focused on post-quantum cryptography. The item creation was blocked by the abuse filter as “specific spam items.” I am directly associated with Quantumize and want to disclose that conflict of interest. My intent is to create a neutral, factual Wikidata item using verifiable public sources and not promotional content. Could an experienced editor or administrator advise whether Quantumize currently meets Wikidata's notability requirements and, if so, assist with creating the item correctly? Eaquiroz (talk) 05:21, 11 July 2026 (UTC)

Hello. Please do not create items about yourself or your business. A quick Google News search for Quantumize returned no results, suggesting a lack of notability. Yirba (talk) 06:17, 11 July 2026 (UTC)

Help panel question on Q9084673 (15:38, 11 July 2026)

how do i add this en item on the list of languages? --Ssemmanda will (talk) 15:38, 11 July 2026 (UTC)

It links to the en: article, so not sure what you mean. Secretlondon (talk) 17:21, 11 July 2026 (UTC)

possible to search eng label with instance of (P31) and Ranga Reddy (Q16344056) using wikidata search

is it possible to search eng label with instance of (P31) and Ranga Reddy (Q16344056) using wikidata search on https://www.wikidata.org/w/index.php?search= কল্কি (talk) 00:01, 12 July 2026 (UTC)

@কল্কি: Hi, I'm not sure what you mean. Can you give an example of a search result you'd expect to be returned? Yirba (talk) 00:17, 12 July 2026 (UTC)

@Yirba for example i want to search disambugiation (Q16344056) for english label "peace" কল্কি (talk) 00:26, 12 July 2026 (UTC)

result must show only Peace (Q12981369) - কল্কি (talk) 00:38, 12 July 2026 (UTC)

Please note that Ranga Reddy (Q16344056) is a specific Wikimedia disambiguation page. You might be looking for Wikimedia disambiguation page (Q4167410) which is the class of Wikimedia disambiguation pages in general. Also, I think you might be better off using a SPARQL query for something like this. Yirba (talk) 00:49, 12 July 2026 (UTC)

@কল্কি: You can do it like so. [6] I'll throw in a little bonus tip too. If you want to search for something but you don't want results that include scholarly papers you can exclude those hits like so. [7] Or maybe you want only search results for humans? [8] This are two very useful things I use all the time. Infrastruktur (talk) 19:31, 12 July 2026 (UTC)

@Infrastruktur thnx. it resolved the issue ! কল্কি (talk) 00:14, 13 July 2026 (UTC)

Instagram username (P2003) and property constraint (P2302), subject type constraint (Q21503250)

Everything can be considered a profile, so what is the point? Eurohunter (talk) 09:15, 12 July 2026 (UTC)

Would the creator care to comment on this? @WG0uMSn0dJm5lfrWbgoi

@Eurohunter I’ve undone the changes; if there’s a valid reason for them (though I can’t see one), someone can add them back in :-) RVA2869 (talk) 15:40, 12 July 2026 (UTC)

@Eurohunter, @RVA2869: I think the idea is that items like LGBTQ community (Q51393) shouldn't have an Instagram username because there is nobody 'in charge' of the LGBTQ community so there can't be an official LGBTQ Instagram account. Bundeli (Q56399) is a language and languages generally can't have official Instagram accounts (conlangs aside, perhaps). Although there might be organisations involved in promoting or standardising a language, languages also develop naturally and aren't bound to a specific nation. And I have no idea why the Instagram account for Novuss (Q178505) is that of the Hong Kong Chess Association. So this property constraint is intended to catch things like that. Yirba (talk) 19:04, 12 July 2026 (UTC)

@Yirba: It makes sense, but isn't it possible to do it in other way like add other description or something? Eurohunter (talk) 20:36, 12 July 2026 (UTC)

@Eurohunter: Sorry, I don't understand. Is there a specific problem you've encountered with this constraint? Yirba (talk) 20:42, 12 July 2026 (UTC)

@Yirba: In this way we has to add every next possible type of thing which can have a Intagram profile, instead of just exclude incompatible items. Eurohunter (talk) 20:46, 12 July 2026 (UTC)

What do you think about a "Simple Wikidata" project?

Hi everyone, ​I’ve been thinking lately about how powerful but complex Wikidata is, especially for small applications, light web tools, or new developers who just want quick, straightforward facts without digging through deep ontologies, qualifiers, or multiple conflicting values.

What would you think if there was a "Simple Wikidata" project?

The goal wouldn't be to replace the main site, but to act as a lightweight, human-readable layer (similar to Simple English Wikipedia). It would focus on a single, preferred value per property and use clean, clear text/slugs instead of just strict Q/P IDs for everyday queries. It could be fully automated and synced via bots from the main project.

Do you think this is a feasible and useful idea for the ecosystem, or is it unnecessary given what we already have? Would love to hear your honest thoughts! -- Aws ^{💬 Talk} 09:28, 12 July 2026 (UTC)

Something like toolforge:reasonator? Ainali (talk) 09:45, 12 July 2026 (UTC)

@Ainali Not exactly. Reasonator is a great frontend for human readers, but it still queries the massive Wikidata live in the background, which means it doesn't solve the infrastructure strain or SPARQL timeouts.

Simple Wikidata would be a pre-filtered, lightweight backend database for developers and AI models. It pre-computes single preferred values and strips qualifiers, providing instant API responses and a small file size that can be downloaded for offline use.

Reasonator is for reading; this is for building. -- Aws ^{💬 Talk} 09:52, 12 July 2026 (UTC)

I don't think you should offer an editing interface if you cannot see all statements. That is bound to make the data messier. Ainali (talk) 13:06, 12 July 2026 (UTC)

The architecture is fully flexible. We can deploy it as a strictly read-only layer managed by automated bots, or allow restricted editing using strict data validation, This is an implementation detail we can easily decide later—it shouldn't overshadow the core utility of the project. -- Aws ^{💬 Talk} 13:15, 12 July 2026 (UTC)

Sorry, but I really don't understand the use case. Real data is often messy. wdt: predicates already get you the best value. In many cases you need multiple values and the data won't make sense if you only get one value. I feel like there's a danger that oversimplifying the data would result in misleading data being returned, and risk undermining trust in the project. Also, "human-readable" suggests natural language. But it also needs to be machine-readable so developers can integrate with it. I suppose one way to do that might be to use some sort of Large Language Model to translate natural language queries into SPARQL. That doesn't require single, preferred values or textual slugs, though. Yirba (talk) 11:08, 12 July 2026 (UTC)

@Yirba, we are looking at two entirely different paradigms here.

Using wdt: truthy predicates still requires the client to fetch, parse, and process complex RDF graphs with qualifiers and references. For a lightweight app running on a smartwatch, a low-bandwidth mobile network, or a local offline system, that overhead is a dealbreaker.

As for human-readable I don't mean natural language prose—I mean replacing dry property IDs like P36 with predictable JSON keys like "capital". Translating natural language to SPARQL via LLMs is an interesting idea, but it’s incredibly expensive, slow, and unreliable for real-time applications. It doesn't solve the core infrastructure strain or prevent SPARQL timeouts.

Sipmle Wikidata isn't undermining trust by hiding complexity; it's expanding accessibility by offering a fast, predictable backend layer for developers who just need clean facts instantly." -- Aws ^{💬 Talk} 11:19, 12 July 2026 (UTC)

Right now, if a developer wants to fetch a simple fact like 'What is the capital of Libya?', they have to write a complex SPARQL query, handle dozens of property IDs (like P36 and Q1016), and filter out multiple conflicting values or historical data manually in their code.

​Simple Wikidata flattens all of this. It gives developers a clean, context-aware API with human-readable keys and a single, pre-filtered 'preferred value' ready to use.

​Instead of spending hours decoding ontologies and dodging query timeouts, they get a direct answer in milliseconds. It turns a complex data-science task into a standard, lightweight API call. -- Aws ^{💬 Talk} 11:20, 12 July 2026 (UTC)

I just did SELECT ?capital WHERE { wd:Q1016 wdt:P36 ?capital. } and found out the capital of Libya in milliseconds. Didn't have to decode ontologies or dodge query timeouts. Sure, maybe there could be a simplified interface that means you don't need to use SPARQL. In a sense, that's sort of what the Wikidata Query Builder tries to do (though from a GUI perspective). But I don't see how you'd translate the complexity of Wikidata into what you're proposing. For one thing, where do the textual keys come from? The good thing about QIDs and PIDs are that they are persistent. Labels change all the time, but the IDs stay the same. Yirba (talk) 11:38, 12 July 2026 (UTC)

@Yirba Your query worked for one country because it's a micro-task. Now try scaling that. Querying a simple fact for 200 countries, 10,000 cities, or every 19th-century author in one go is exactly where SPARQL times out and the payload breaks mobile apps.

​As for the IDs vs. Labels, nobody is suggesting we abandon QIDs and PIDs internally; they are the ultimate source of truth. The human-readable keys (like "capital") would be a static, community-curated API mapping layer bound directly to those persistent IDs. If a label changes on Wikidata, the API key remains "capital".

​The Query Builder is just a visual skin that still spits out heavy SPARQL and complex JSON. Simple Wikidata isn't changing how data is stored—it's a high-scale serialization and caching layer for production apps. Big difference. -- Aws ^{💬 Talk} 11:57, 12 July 2026 (UTC)

I don't think I've ever seen WDQS time out on a query for 200 items. Timeouts happen when you hit millions of triples to sort through. And the new QLever endpoint for searching (coming soon!) should be even better for this. ArthurPSmith (talk) 17:33, 13 July 2026 (UTC)

Agreed. I tried replacing the wd:Q1016 in my query above with ?country and WDQS returned 132,394 results in 1,472 ms. Not bad, I think. Certainly nowhere near close to timing out. Yirba (talk) 18:12, 13 July 2026 (UTC)

This kind of sounds like what Abstract Wikipedia aims to be; basically creating Wikipedia pages out of data from Wikidata, but with proper grammar. Infrastruktur (talk) 20:36, 12 July 2026 (UTC)

This is actually the exact opposite of Abstract Wikipedia, Abstract Wikipedia takes structured data to generate natural language prose for humans, My proposal takes complex semantic graphs and flattens them into a high-performance, predictable JSON/API layer for machines and AI. -- Aws ^{💬 Talk} 20:49, 12 July 2026 (UTC)

 Oppose this seems to completely misunderstand the purpose of Wikidata. For example replacing the P*/Q* with a specific language loses Wikidata's essential language neutrality. If something like this were built it seems to me it would have little or no relation with what we have created here. ArthurPSmith (talk) 02:02, 13 July 2026 (UTC)

Q140520828

Hello, i created item on Q140520828 for son ha agency , but it was deleted before i could add reference and the statement ~2026-39488-21 (talk) 13:03, 12 July 2026 (UTC)

@~2026-39488-21: Hi, could you please share some references that demonstrates the subject's notability? (cc the deleting admin @WikiBayer) Yirba (talk) 13:43, 12 July 2026 (UTC)

A reputable not self created source is needed. Wikidata is not for marketing. WikiBayer (talk) 13:49, 12 July 2026 (UTC)

Relevant items: Q140520828 (~2026-39488-21), Q135914771 (Kermitmoore), Q131310668 (Michaelllee), Q131456942 (Williamboyda). CC deleting admins @Jianhui67, Ymblanter Bovlb (talk) 23:19, 12 July 2026 (UTC)

Given that I am no longer an administrator here, I do not have access to deleted pages. Nevertheless, I did a quick check on the entity in question. I do not think it is notable enough to warrant undeletion, though I could be proved wrong. The final decision still lies in the hands of the current administrator team. Jianhui67 ^{talk★contribs} 05:40, 13 July 2026 (UTC)

Lexeme:L1888

can (L1888) has to be separated from other English lexemes spelled "can", but that is not being done properly.

First, the meanings listed on the Merriam-Webster online dictionary entry (P11130) value link of can (L1888) do not match with those listed on the Oxford English Dictionary object ID (post-July 2023) (P12510) value link of the lexeme page, as the former includes definitions that are currently widely used like "be physically or mentally able to", "know how to" and "have permission to", while the latter is for an obsolete verb that is spelled "can". Perhaps we have to change the Oxford English Dictionary object ID (post-July 2023) (P12510) value so that it links to this OED item, which does include definitions like "Expressing physical or mental ability: [...] ".

Second, the lexeme page has L1888-S3 ("put into tins") and L1888-S4 ("fire (metphorical extension of 'throw away')"). However, since the lexeme page has modal verb (Q560570) as a value of instance of (P31), has forms like "could", "can’t", and "couldn’t", and does not have forms that are "canned", these two senses do not belong here, and may have to be moved to can (L30365). However, if these senses have to be moved, I do not know how they should be moved with proper attribution.

How should these problems be fixed? Intolerable situation (talk) 21:53, 12 July 2026 (UTC)

I believe senses cannot be "moved" or copied, you will need to create the new senses on can (L30365) and then you should be able to move the additional statements from the corresponding senses on can (L1888) to the new ones, and finally delete the incorrect senses there. Unless you think there's more needed to be done here? ArthurPSmith (talk) 02:06, 13 July 2026 (UTC)

Should the users who wrote the senses in can (L1888) be acknowledged when the senses are created in can (L30365)? If so, how should they be acknowledged? Intolerable situation (talk) 04:40, 13 July 2026 (UTC)

There is no need for that, we don't attribute factual data. The glosses are written text that maybe could be attributed, but I think that is preserved in the copy process? Not something we generally worry about here though. ArthurPSmith (talk) 15:04, 13 July 2026 (UTC)

Citation overkill

Hello! On places like enwiki we have en:WP:CITATIONOVERKILL. As far as I could find, we don't have that here. However, when I use Wikidata in infoboxes on nlwiki, I sometimes get ten references for an uncontroversial birth date, which is simply ugly. Is limiting the number of references something we want do here? Or is this something I should solve on the Wikipedia itself? Thanks for your thoughts! Dajasj (talk) 01:00, 13 July 2026 (UTC)

It's quite possible that we have excessive citations in places, but my general recommendation is that you solve it on the client side. We don't like to remove potentially useful information here just to make a specific template on a specific project look better. Bovlb (talk) 05:54, 13 July 2026 (UTC)

I think and fear so too ;) Although I do wonder whether it is really useful here? And in the end, making Wikidata more useful for Wikipedia hopefully makes it more used and more reliable. Dajasj (talk) 06:09, 13 July 2026 (UTC)

10 references is a bit extreme, but it can be useful to have redundant references in cases where there are multiple conflicting claims, or in case some references stop being accessible (due to link rot or otherwise). Some people run tools that remove low quality references if there are also high quality references. I wouldn't mind having some sort of guideline for how many references a statement should generally have, but I think it really varies from case to case so I wouldn't want it to be a hard limit. Yirba (talk) 11:39, 13 July 2026 (UTC)

July 2026 Wikimedia Café meetups regarding Wikimedia governance and options for reform

Hello! There will be two Wikimedia Café discussion opportunities in July. Both sessions will focus on Wikimedia governance, including possible follow-ups to the Movement Charter and options for reform. Participants may attend either or both Café sessions.

This month, to deconflict the Café meetups from Wikimania, the meetups will be held one day later than usual.

1. 26 July 2026 15:00 UTC (timestamp converter), at a time friendly to the Americas, Africa, and Europe
2. 27 July 2026 03:00 UTC (timestamp converter), at a time friendly to Asia and the Pacific

Please see the Café page for more information, including how to register!

↠Pine (✉) 03:34, 13 July 2026 (UTC)

Stale bot request

I've had a bot request open for 3 weeks with no response. Could someone take a look? Wikidata:Requests for permissions/Bot/RupumpedBot

This is my first bot, so I apologize if this is the wrong forum for such a request. I completely understand that the reviewers are volunteers, and I don't mind waiting in line for longer if 3+ week reviews are normal. Rupumped (talk) 09:01, 13 July 2026 (UTC)

@Rupumped: Hi. As per the instructions at Wikidata:Requests for permissions/Bot, you'll need to transclude your RfP page onto that page. Yirba (talk) 11:40, 13 July 2026 (UTC)

Done. Thank you! Rupumped (talk) 11:48, 13 July 2026 (UTC)

Request for independent assessment of a potential author/researcher item

Hello. I am Nohil Kodiyatar, the person concerned, and I would like to clearly disclose my conflict of interest.

I am seeking an independent assessment from experienced Wikidata editors regarding whether a Wikidata item about me would meet Wikidata's notability criteria. I have not created an item about myself because I am the subject and would prefer an independent editor to assess the available public records.

My ORCID iD is:

0000-0001-8430-1641 https://orcid.org/0000-0001-8430-1641

My journal publications include:

1. A Comprehensive Analysis of Student Development: Integrating IQ, EQ, SQ, and AQ for Enhanced Learning Outcomes

Journal of Ethics in Higher Education, No. 7.2, pp. 205–221.

DOI: https://doi.org/10.26034/fr.jehe.2025.6193

2. Educational Disparities: A Global Comparative Analysis of Social and Family Problems in Developing and Developed Nations

Academia Lasalliana Journal of Education and Humanities, Volume 6, Issue 2, pp. 275–281, 2025.

DOI: https://doi.org/10.55902/DYUA1273

3. The Impact of Time on Parallel Universes: A Study on Choice and Determinism

DS Journal of Multidisciplinary, Volume 1, Issue 2, pp. 11–15, 2024.

DOI: https://doi.org/10.59232/DSM-V1I2P102

My Google Scholar author profile is:

https://scholar.google.com/citations?hl=en&user=xj6ns2MAAAAJ

I am also an author or co-author of published books with ISBN records. Selected books include:

1. Scientific Understanding of 3I/ATLAS (C/2025 N1): Authentic Data, Observational Insights, and Information Ethics

ISBN-13: 979-8-2726-1381-3

2. Contemporary Advances in Artificial Intelligence Applications to Theoretical and Computational Chemistry

ISBN-13: 979-8-285-13304-9

3. Intelligence Beyond Brains: Chemical, Animal, and Artificial Minds in the 21st Century

ISBN-13: 979-8-2411-3143-0

4. Timeless Structural Ontology: Existence and Access

ISBN-13: 979-8-2460-9633-8

5. Nurturing Minds: The Educator's Essential Toolkit

ISBN-13: 979-8-3127-2399-1

My scholarly and publication records are publicly discoverable through journal publication records, ORCID, Google Scholar, Google Books, and ISBN-linked book records.

Could an independent Wikidata editor please assess whether these public records are sufficient for a Wikidata item about me under Wikidata's notability criteria?

I am happy to provide additional bibliographic information, journal records, ISBN records, or other public sources if required.

Thank you for your time and guidance. NohilKodiyatar99 (talk) 11:27, 13 July 2026 (UTC)

Hello, thank you for the message. To answer shortly, no. Technically, due to scarcity of independent verifiable record. I saw your ORCID and Google Scholar profiles, I saw that the majority of the records are self published through Zenodo, no substantial citation and no publication published collaboratively, a practice common in academia. Also, you could read our community stance regarding editing items about oneself here. Thank you. Yamato Shiya • (Hello!) 11:42, 13 July 2026 (UTC)

Thank you for your assessment. I would like to clarify one point, if I may. The three journal articles I listed are published in journals and have DOI records; they are not self-published through Zenodo.

Would these journal publications count as independent verifiable records for Wikidata purposes, even if the citation count is currently limited? Or are they still insufficient for an item about the author?

I am asking only to better understand Wikidata's criteria. Thank you for your guidance. NohilKodiyatar99 (talk) 13:50, 13 July 2026 (UTC)

At the current moment, still insufficient. Yamato Shiya • (Hello!) 14:03, 13 July 2026 (UTC)

Welcome. You can add your ORCID to your user page User:NohilKodiyatar99 using the {{Authority control|ORCID=...}} template. You should also feel welcome to edit items or even create new ones where you have knowledge but not a personal conflict of interest. If your works are recognized and cited then they will likely eventually be added here, and somebody will likely add an item for you as well. It doesn't help to try to rush things. ArthurPSmith (talk) 15:20, 13 July 2026 (UTC)

Contradicting requirements for properties

Marvel character ID (P5621) requires Marvel Universe (Q931597) for the property from narrative universe (P1080), but said property has it as a constraint that Marvel Universe (Q931597) should not be used for it. StarTrekker (talk) 11:31, 13 July 2026 (UTC)

Pinging @OmegaFallon who added the none-of constraint to P1080 and @Shisma who added the item-requires-statement constraint to P5621. Yirba (talk) 11:47, 13 July 2026 (UTC)

By the way, I've created a SPARQL query to find similar cases where one property's item-requires-statement constraint conflicts with another property's none-of constraint: https://w.wiki/SJYM Yirba (talk) 12:00, 13 July 2026 (UTC)

Add languages

Hi, could someone please add the other languages to my article? https://es.wikipedia.org/wiki/Berlinische_Galerie I don't have the rights to do that. The Wiki data id is: Q700222. The item is semi-protected and my account is not yet autoconfirmed here. Thanks! Buceador Viejo (talk) 15:21, 13 July 2026 (UTC)

 Done Yamato Shiya • (Hello!) 15:22, 13 July 2026 (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. Yamato Shiya • (Hello!) 15:22, 13 July 2026 (UTC)

My new entry, Q140529499, was deleted

Dear Wikidate Team and Administrators

I tried to create a listing for our company. Unfortunately, this listing has already been deleted twice, and I don't understand why. Our company actually exists. I've added several independent press releases, and even our company ID. Why was this listing deleted by @Yamato Shiya?

What can I do differently? Or what information is missing from a post so that it won't be deleted?

Thanks so much for your help, and see you soon, Tobi

TobiTasteMatch (talk) 16:08, 13 July 2026 (UTC)

@TobiTasteMatch: Hi, Wikidata has a notability policy. Simply put, we don't create items for every company. The company must be notable in some way. In this case, notability was not demonstrated. It doesn't matter if you add press releases or IDs; we do not create items for companies (or anything else) unless they meet the notability criteria. It may be the case that the company becomes more notable in the future and someone else will create the item. But we don't recommend creating an item for your own company. Yirba (talk) 17:19, 13 July 2026 (UTC)

Thanks a lot @Yirba.

I understand the point, and this is the same as with the other wiki pages. I thought a startup would have a better chance here. After all, a startup is notable in its own way and is always in the spotlight. These days, companies—including startups—need credibility, but it’s not possible to gain it this way because they’re completely excluded. Is there no chance in this case? No matter what links I add? Have a nice evening, Tobi TobiTasteMatch (talk) 18:24, 13 July 2026 (UTC)

@TobiTasteMatch: It is not Wikidata's role to lend credibility to a company. In fact, the idea that a company might be considered credible just because it has a Wikidata item is rather concerning. There are plenty of companies that are credible yet don't have Wikidata items. Wikidata also has items for disreputable companies. Anyway, if you are able to share some links to serious sources that describe your company (e.g. articles from major news publications), then we can consider the notability. But people should not be creating items for their own company. Yirba (talk) 18:58, 13 July 2026 (UTC)

thanks @Yirba I know that's not Wikidata's job. We've added various media reports, including two from Switzerland's largest media companies. One media company even wrote about us twice. Should I send them to you again? TobiTasteMatch (talk) 19:42, 13 July 2026 (UTC)

Hello, thank you for the message. Yirba had already stated the main reason. As for the reports, I read what you attached in the deleted items, one is a press release, and two are short reviews. I see that the reports has indications of advertorial, even reusing same photos. The subject needs more coverage than these two, independently and covering various aspects: what you submitted only covered the founding and some review, the coverage should be more than that, and perhaps could be undeleted in the future if such coverage exists.

In the meantime, please avoid recreating deleted items.

Thanks. Yamato Shiya • (Hello!) 23:55, 13 July 2026 (UTC)

Wikidata weekly summary #740

Here's your quick overview of what has been happening around Wikidata in the
week leading up to 2026-07-13. Missed the previous one? See issue #739.

Discussions

• Open request for adminship: GA Kevin - RfP scheduled to end after 14 July 2026 18:40 (UTC)
• New requests for permissions/Bot:
  • Nabbot 2 - Task: Translating English descriptions to Turkish using a predefined dictionary mapping.
  • AgenticCommonsBot 1 - Task: Add Open Library ID (P648) (OpenLibrary author ID) to items about people, where the OpenLibrary author and Wikidata person are the same individual, when every statement is sourced.
  • EquipboardBot - Task: Add Equipboard artist ID (P6912) to Wikidata items for musicians, sourced from Equipboard's own database of public artist identifiers.
• Call for Participation in Improving Constraints
  • Just about any aspect of constraints is in scope, from rewriting existing bots to expanding the power of property constraints to adding new kinds of constraint. If you are interested, sign up in WikiProject Property Constraints/improving.

Events

• Upcoming events:
  • Wikimania 2026 is fast approaching. Taking place 21 - 25 July in Paris, France. Can't attend? Register for a ticket to the virtual event and watch sessions online.
  • Interested in Wikibase Ontology Federation? Join the July 14 call, see the Cleaning Task Force page for more details.
• Previous events:
  • The 3rd Quarter Wikidata + Wikibase office hour was held. The transcript is available to read:
    Wikidata:Events/Telegram office hour 2026 07 08

Press, articles, blog posts, videos

• Blogs
  • Behind Better Data: Inside Wikidata’s First WikiProjects Days
  • Africa Wiki Challenge 2026 in the Ghanaian Pidgin Community: Documenting “Water for Life” Through Wikimedia Projects
  • Reflecting on progress and planning ahead: Dagbani Wikimedians User Group holds its last quarter team meeting
  • EduWiki Workshop: Combining PETScan and Generative AI for Smarter Wikipedia Editing
  • Why Computational Historians need Wikidata - a substack article by Jim Clifford
• Papers
  • A Demonstration of WikiRAG: An Evidence-based Link Prediction for Wikidata with Retrieval Augmented Generation - WikiRAG is an interactive framework that combines automated link prediction with retrieval‑augmented generation and human validation to reliably complete knowledge graphs like Wikidata by generating candidate links, retrieving supporting evidence, and enabling users to confirm and export validated triples. By Sabu, et al., (2026)
• Videos
  • Using Open Refine by Wikimedia Australia, Paul Duchesne presents "Using OpenRefine" during WikiCon Australia 2026, (National Film and Sound Archive, Canberra).
  • Let’s Get Mappy - making maps on Wikimedia platforms by Wikimedia Australia, Alex Lum presents how location (pushpin) maps, locator maps, and interactive "slippy" local maps end up on Wikipedia, and how infoboxes, Wikidata and OpenStreetMap works together to generate them
  • (繁體中文/zh-TW) Tutorial on Editing Temple Data for Wikidata and Wikimedia Commons - recording of a workshop (July 5) on Tutorial on Editing Temple Data for Wikidata and Wikimedia Commons, held by Wikimedia Taiwan Community and OSM Taiwan.
  • Open data tools for cultural heritage 2nd session in the workshop series “Open Data to Strengthen Online Cultural Heritage,” held March 24, 2026, and organized by the National Library of Uruguay (BNU), Wikimedians of Uruguay, Datysoc, and Data Uruguay.
  • GLAMs and Citizen Science Encouraging and Enabling Participation - Siobhan Leachman presents "citizen science workflows" (BHL, iNaturalist, Bionomia, Wikidata) to show how GLAM's can empower citizen scientists to reuse their collections.

Tool of the week

• rudof can be used as a command line tool or can be embedded as a library using cargo. It can be used to validate RDF data represented with different syntaxes like Turtle, NTriples, to name a few. As well as RDF data available through SPARQL endpoints like Wikidata

Other Noteworthy Stuff

• A Project Chat discussion on the new Mentorship features. Self-enrollment has been enabled with the following requirements - account age: 365 days, edits: 500.
• There is also a proposal to add an advisory message to new accounts that create multiple items.

Newest properties and property proposals to review

• Newest properties:
  • General datatypes: none
  • External identifiers: Rakuten TV adult person ID, KBS News topic ID, Rakuten Books author ID, PTS News Network topic ID, Nakamura Kikaku ID, ‎La Grange person ID, ‎Fontaines de Belgique ID, NHK On Demand program ID, 1SCREEN program ID, LUDAP ID, Bianco keyword ID, Architectenregister ID, ‎ClayArena shooter profile ID, ‎Sundsvall Municipality opendata beach ID, Terceiro Tempo Que Fim Levou ID
• New property proposals to review:
  • General datatypes:
    • relocated from (Object is the original or former {{Q|17334923}} of the permanently relocated {{Q|16222597}})
    • OpenType script tag (identifier for a script in OpenType)
    • transforms from (this fictional or mythical entity transforms from that form)
    • Korean Classification of Administrative Districts (code assigned to an administrative district of South Korea in the Korean Classification of Administrative Districts)
  • External identifiers: Anobii book ID, Google Maps Place ID, Viberate Artist ID, Brussels Inventory of Architectural Heritage style ID, Brussels Inventory of Architectural Heritage designer ID, TuneCore Japan artist ID, Deltarune Wiki article ID, INAPP OA author ID, CNR-IRIS ID, AnimeOshi anime ID, Guide Japan Person ID, CD Journal Artist ID, CIDIM Artist ID, Operabook Artist ID, Mizan al-Arab poet ID, Norton Simon Museum object ID, OpenStreetMap data item ID, Identifiant sur le Western Orthodox Wiki, A World Made by Travel person ID, Scent Base notes ID, radio.net IDs, Scent Base perfumer ID

You can comment on all open property proposals!

Did you know?

• Query examples: Items only connected to Telugu Wikipedia (source) - change the language code 'te' to make it work with any other Wikipedia
• Schema examples: French historic monument (E269) - With Wikimania taking place in the French Capital, you can use this entity schema if you feel inspired to or expand items on French monuments.
• Newest WikiProjects:
  • Project Pleasant Goat and Big Big Wolf - a Chinese-animated television franchise.
  • (Français) Project Orthodox Christianity
  • Project SUNY Libraries EMCO
• WikiProject Highlights:
  • Project Govdirectory Poland added several new subpages detailing coverage and progress on State, Local Agencies and Marshall's Offices.
  • Project Property Constraints added a subpage: Improving - a call for participation, as a result of discussions during WikiProjects Days 2026.
• Newest database reports: Most sitelinked Items (with statements)

Development

• Blazegraph migration: We are continuing to prepare the Query UI, Query Builder, constraint checks, etc for the migration to QLever (phab:T425618)
• Wikidata integration in the Wikimedia projects: We are investigating some Lua modules to find ones responsible for a lot of unnecessary entries in Watchlist and Recent changes on Wikipedia and co (phab:T429366)
• GraphQL: Added support for negation in property value pair filters (phab:T429580)
• REST API: We are working on the first pieces of support for Lexemes: retrieving a full Lexeme (phab:T429332)

You can see all open tickets related to Wikidata here. If you want to help, you can also have a look at the tasks needing a volunteer.

Weekly Tasks

• Add labels, in your own language(s), for the new properties listed above.
• Contribute to the showcase Item and Lexeme above.
• Govdirectory weekly focus country:
• Summarize your WikiProject's ongoing activities in one or two sentences.
• Help translate or proofread the interface and documentation pages, in your own language!
• Help merge identical items across Wikimedia projects.
• Help write the next summary!

· Previous issue · Help translate · Danny Benjafield (WMDE) talk · MediaWiki message delivery (talk) 18:00, 13 July 2026 (UTC)