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

Wikidata:Project chat

From Wikidata
Jump to navigation Jump to search

Edit

ProVe, a new tool to help with the quality of references

[edit]

Thanks to all of you you have so far tried ProVe, the new tool for checking the quality of references in Wikidata. It's greatly appreciated :-) (link to the archived discussion here).

@samoasambia thanks for your suggestions! We are updating the script import instructions in the documentation to avoid copy-pasting code, great idea. We're also adding ProVe to the tools catalog. Thanks for your help!

@Huntster thanks for letting us know, this was a bug. It's now been fixed, if you could try again and let us know if the new version works that would be great! Thanks

Just as a reminder for everyone else, ProVe provides information about the quality of the references of Wikidata items, based on techniques like large language models, triple verbalisation, and semantic similairty. We have also developed the **ProVe Gadget**, which visually presents ProVe's results as a widget at the top of a Wikidata item page. Any Wikidata user can easily turn this gadget on, see here for install instructions. You can use it to request the processing of references, showing reference scores, navigating problematic references, and quickly fix them with better ones.

If you're curious about this we'd greatly appreciate your feedback! :-) Albert.meronyo (talk) 10:03, 18 October 2024 (UTC)[reply]

I am trying to use it, should be a useful thing, thanks. For some reason, the gadget does not analyse web.archive.org correctly and is unable to 'read' the text from the web archive. So, it says about it 'Sentence in external URL to be checked, possibly not authoritative'. --Wolverène (talk) 12:23, 18 October 2024 (UTC)[reply]
Thanks! I think we may have to do additional checks to read text from the web archive, indeed. What item are you trying to analyse? Thanks for using ProVe! Albert.meronyo (talk) 13:39, 18 October 2024 (UTC)[reply]
Well, e.g. Q23648408. (It is also listing IGN there as not authoritative although this is quite a well-known video games-related media with the serious team...) --Wolverène (talk) 19:31, 19 October 2024 (UTC)[reply]
Thanks! We're looking into this Albert.meronyo (talk) 08:49, 23 October 2024 (UTC)[reply]
Hi @Wolverène, thank you very much for your input. I've reviewed the issues you mentioned. First, accessing web.archive.org takes a considerable amount of time, roughly 1–2 minutes. The current web crawler engine on the backend is set to wait 15 seconds per URL to prevent overloading the web crawling process. Second, certain web servers, such as IGN, may deny access to their pages by the web crawler to protect their rights. Due to this access policy, the tool is unable to access IGN pages automatically. Dignityc (talk) 07:14, 29 October 2024 (UTC)[reply]
I see. Thank you. --Wolverène (talk) 04:58, 30 October 2024 (UTC)[reply]

Permanent records removal

[edit]

I am writing on behalf of Mr. Morgan Hermand-Waiche to formally request the permanent removal of all records associated with him from your website. Mr. Hermand-Waiche believes that the presence of this information constitutes a violation of his rights, and he does not wish for any details pertaining to him to be displayed publicly. I would like to emphasize that I am connected from the company account to validate that this information is accurate. We appreciate your attention to this matter, and please note that the reference source, Who's Who, has already deleted his records. link: https://www.wikidata.org/wiki/Q20089624 ____________________________________________________________________________________________________________________________

Je vous écris au nom de M. Morgan Hermand-Waiche pour demander formellement la suppression permanente de tous les dossiers le concernant de votre site web. M. Hermand-Waiche estime que la présence de ces informations constitue une violation de ses droits et il ne souhaite pas que des détails le concernant soient affichés publiquement. Je tiens à souligner que je suis connecté depuis le compte de l'entreprise pour valider que ces informations sont correctes. Nous vous remercions de l'attention portée à cette demande, et nous tenons à préciser que la source de référence, Who's Who, a déjà supprimé ses dossiers. Merci pour votre assistance rapide. Angeleml (talk) 12:22, 23 October 2024 (UTC)[reply]

The French Wikipedia consider him notable enough to have a page at https://fr.wikipedia.org/wiki/Morgan_Hermand-Waiche , we generally do not delete items on Wikidata if there's a Wikipedia page associated to it and even if we would delete an item with an Wikipedia page, that item would get automatically recreated.
As far as I can see the French Wikipedia does not use Who's Who has the reference for their data. If the French Wikipedia does decide to delete the article please mention it here.
Note that both Wikidata and Wikipedia are part of the Wikimedia Foundation. If you want to make a formal legal request addressed at the Wikimedia Foundation, https://meta.wikimedia.org/wiki/Wikimedia_Foundation/Legal/Legal_Affairs describes how to correspond with the legal department. ChristianKl ❪✉❫ 12:40, 23 October 2024 (UTC)[reply]
Thank you! And thanks for deleting the birth date from here as it was his most important concern. Angeleml (talk) 16:56, 23 October 2024 (UTC)[reply]
@Angeleml: Unfortunately, if the information is already public elsewhere in a reliable source, it will not be removed.--Jasper Deng (talk) 06:53, 24 October 2024 (UTC)[reply]
@Jasper Deng: this information does not seem to be public, even if it were public, it's a personal data. Cheers, VIGNERON (talk) 10:39, 25 October 2024 (UTC)[reply]
Clearly it must have been gotten from somewhere unless Hermand-Waiche added the statements himself Trade (talk) 13:18, 30 October 2024 (UTC)[reply]
As the OP wrote, the statement was sourced to Who's Who which deleted their article. So while that public source existed in the past it currently doesn't exist.
Regardless, the standard our current policies suggest is "statements should generally not be supplied unless they can be considered widespread public knowledge or openly supplied by the individual themselves".
While, I personally, think that's a bit broad for date of birth (P569) that's the standard we currently have via WP:LP as date of birth (P569)living people protection class (P8274)property that may violate privacy (Q44601380). ChristianKl ❪✉❫ 16:23, 31 October 2024 (UTC)[reply]

Neighborhood has Facebook groups

[edit]

Hello everyone,

I'm a bit confused on whether I can attach Facebook user group to the wikidata entry of my neighborhood. The thing is that a lot of districts and neighborhoods in my city have Facebook and Telegram user groups, where people post about their problems, buy and sell, opening of new shops, discounts and so on. There are not "official", of course, but number of members reaches minimum 7 000. I think they are vital to uncover to whomever moves to a new district or neighborhood, I've already added them to OSM, but it's troubling for me to correctly include them into Wikidata. At this point of time, I've created this entry of my neighborhood Orkhevi (Q130437988) and not sure whether I included Facebook groups' info correctly.

Could you pls advise on this matter? David Osipov (talk) 06:45, 28 October 2024 (UTC)[reply]

They need a qualifier object of statement has role (P3831) community group (Q106464965), but I'm not sure WD should record this given the poor mapping of our districts and a morass of Facebook groups Vicarage (talk) 08:40, 28 October 2024 (UTC)[reply]
About the mapping - we're working on it with OSM community :) There is a lot to do, yeap, but we're progressing. Thank you for the piece of advice, I'll use it right now. David Osipov (talk) 11:45, 28 October 2024 (UTC)[reply]
Unofficial groups being added to items is fraught with potential problems and it's unclear to me what the scope of "I'll use that right now" means. I think it might be worth seeking some more consensus and caution before adding or changing a significant number of statements. William Graham (talk) 14:05, 28 October 2024 (UTC)[reply]
Yeap, that's what I want to get - some consensus on it. As for now, I've 100% aware of all the unofficial groups of my own neighborhood for now, so that I've edited only it for now. But I'm also aware of unofficial user groups of other neighborhoods and districts in the city of Tbilisi, Georgia. David Osipov (talk) 14:35, 28 October 2024 (UTC)[reply]
Currently, the property is for official pages and this is not an official page. Creating a discussion on https://www.wikidata.org/wiki/Property_talk:P2013 about whether to use the property also in nonofficial instances and if so under which circumstances would be the way forward. That talk page is the proper place for a discussion as anyone who wants to understand how the property can be used can look at it, while project chat discussions are not easily findable. ChristianKl ❪✉❫ 08:51, 31 October 2024 (UTC)[reply]

How to add a signature image?

[edit]

I see that many persons have their signatures (e. g. Ronald Reagan), while others do not (e. g. Sergey Aksakov). How do I add a person's signature, if there alredy exists an image at commons? — Monedula (talk) 15:26, 28 October 2024 (UTC)[reply]

use signature (P109) - Salgo60 (talk) 16:25, 28 October 2024 (UTC)[reply]
That's OK, but where is the button "Add property"? — Monedula (talk) 17:57, 28 October 2024 (UTC)[reply]
Be sure to use the desktop version of Wikidata, adding statements on mobile is sadly still not possible in 2024. Sjoerd de Bruin (talk) 08:55, 29 October 2024 (UTC)[reply]
Video how you change from the desktop version - Salgo60 (talk) 09:16, 29 October 2024 (UTC)[reply]

Now I see that is is called "add statement". But there are 2 of them on each page. Is there any difference? — Monedula (talk) 15:41, 29 October 2024 (UTC)[reply]

There is no difference, you can use any one of them and the result will be the same (after a page refresh). Samoasambia ✎ 17:05, 29 October 2024 (UTC)[reply]

withdraw a property proposal

[edit]

after some discussion it looks like the SWERIKS projects agree that they can use SWERIK ID (P12192) also for political parties in Sweden --> the Property proposal Wikidata:Property_proposal/SWERIK_Party_ID should be withdrawn.

  1. I changed the status to withdrawn
  2. what else is needed to be done ?

- Salgo60 (talk) 16:20, 28 October 2024 (UTC)[reply]

@Salgo60: 1. is all you need to do, thanks! ArthurPSmith (talk) 13:00, 29 October 2024 (UTC)[reply]

Automatic RFD notifications?

[edit]

Hi all, I have been watching from the sidelines for a few years now, but I'm back at it.

I have been spending quite some time on WD:RFD and wonder if it has ever been proposed that nominations to RFD should result in messages being sent to item creators? or recent editors? (probably not all, but perhaps in some cases?)

New user, no talk page, only contributor to an item that has been RFDed, send a welcome and a note about the deletion request? ·addshore· talk to me! 17:14, 28 October 2024 (UTC)[reply]

We have discussed such a bot several times, e.g. here: Wikidata:Project chat/Archive/2022/07#Deletion without discussion. There may be similar discussions from that time, but I don't fully remember where they are. —MisterSynergy (talk) 18:33, 28 October 2024 (UTC)[reply]
An interesting read, I can certainly see how 1 notification for each 1 RFD would be a bad idea.
Almost 1 notification per person maximum per day could be about right, or maybe even per week.
It would certainly make me as an admin feel better when trying to clean up WD:RFD / speed up the process as I wouldn't feel that I have to reach out to people as they will already have been informed.
Without having to rely on a bot posting messages to talk pages, we could always have the existing bot just @ mention the creator and or last editor.
Anyway, I'd love to hear some more and current thoughts! ·addshore· talk to me! 20:30, 28 October 2024 (UTC)[reply]
When using the RequestDeletion tool for creating RfD (can be activated in the settings) at least the creator gets notified (that does not work for bulk deletion requests). --Dorades (talk) 20:41, 28 October 2024 (UTC)[reply]
IMO the process should be split into nominations for deletion and requests for (speedy) deletion. Nominating would involve sending a message to the creator and letting them know what is wrong and what they should do in order to not have their item deleted (cf. User:Bovlb/How to create an item on Wikidata so that it won't get deleted). The discussion could be held on the user's talk page or the item's talk page (which can be categorized, so that there is a general overview of currently or previously nominated items). Having these discussions on WD:RfD is very unfriendly (it's a long page with many threads, it takes long to publish a comment there, it's sensitive to accidental structure changes because we have bots maintaining it, etc.). The process of nominating could also be automated which I think is desperately needed. --Matěj Suchánek (talk) 08:32, 29 October 2024 (UTC)[reply]
Sounds like a pretty nice idea! ·addshore· talk to me! 08:59, 29 October 2024 (UTC)[reply]
See User:ChristianKl/Draft:ProposeDeletion for a proposal for allowing PROD of items. GZWDer (talk) 12:04, 29 October 2024 (UTC)[reply]
That's also an interesting read, I wonder if it might be better facilitated by functionality from MediaWiki core or extensions? Delayed deletion would likely be a fairly easy one to code up. A user permission that allows you to submit a page for proposed deletion. Potentially some automatic notification to the author? A page to list what is pending? Automatic delete action if noone clicks object? Then it wouldn't have to rely on bots, more edits to Wikidata in general, and properties and statements etc? ·addshore· talk to me! 14:05, 29 October 2024 (UTC)[reply]
I think it's good if PROD uses statements. That's a feature not a bug. Adding statements means that the request for deletion shows up on watchlists. It shows up on the history of the item. It shows up on the contribution history of a user. It makes it possible to use Listeria to watch for statements in a given area of Wikidata being deleted. Each Wikiproject could have a Listeria that lists nomination for automatic deletions relevant to the Wikiproject.
Tools like QuickStatements would also make it easy to nominate a greater number of statements for deletion. Generally, it's good to reuse features within Wikidata to allow more tools to interact with it.
I would oppose a quickly-coded up way where users that are not administrators can propose to delete items without that being visible in watchlists/page history/user contributions.
The property/statement approach has the advantage that users can add new reasons for deletion and all reasons for deletion can be automatically translated (as items can have text in all languages). It makes it easier for non-English speakers to interact with the process. Wikidata is an international project and finding a good solution includes finding ways to automatically translate messages. Messages that are likely best written in item label/descriptions and templates and that can be edited by people outside of WMDE or a person who codes up an extension. ChristianKl ❪✉❫ 10:11, 31 October 2024 (UTC)[reply]

We did a workaround 2021 see T291659 - Salgo60 (talk) 09:27, 29 October 2024 (UTC)[reply]

Merge Piantoni (Q63041615) with Piantoni (Q63064860)

[edit]

Gentlemen, even if I enabled the box "merge" in my gadgets, it is impossible to merge these 2 items. Any idea please ? Ricercastorica (talk) 08:42, 29 October 2024 (UTC)[reply]

Because you are not supposed to merge disambugation items with family name items. Sjoerd de Bruin (talk) 08:53, 29 October 2024 (UTC)[reply]
precisely: Piantoni refers only to a family name, thank you Ricercastorica (talk) 09:34, 29 October 2024 (UTC)[reply]
We care about what template Wikipedia pages use. If they use the Template:Disambiguation (Q6148868) we see it as a disambguation page and not as a page about family names even if all the entries on the page at a particular time are family names. If something is about a family name, Template:Human name disambiguation (Q6817136) should be used on the Wikipedia page. ChristianKl ❪✉❫ 10:33, 31 October 2024 (UTC)[reply]

Duplicado

[edit]

In wikidata the people Q16190529 and Q15628981 is the same person. Is possible to merge them? Froin (talk) 18:44, 29 October 2024 (UTC)[reply]

Done. --Wolverène (talk) 04:57, 30 October 2024 (UTC)[reply]

Importing WP & WMC categories into Wikidata

[edit]

There are many categories with data that is not in the corresponding Wikidata item.

For example, list of unsolved problems in biology (Q2628887) did not have subclass of: open problem (it could also be list of: open problems) set despite that the ENWP article has the category Category:Lists of unsolved problems. Likewise, Vuze (Q3563863) does not have programmed in: Java set despite its WMC category being in c:Category:Free software programmed in Java. These are just a few examples, there are also other categories about country of origin, year of birth/death, year of film production, software license, etc etc. It would waste a lot of time to enter the data again redundantly manually and also would be very monotonous for many similar items (even just the small number of unsolved problems list items).

Is there some tool that automatically and/or semi-automatically imports data from ENWP & WMC categories? If not, such is needed and could increase the data in Wikidata a lot. It doesn't mean the categories would be redundant, they never will be – but the data already entered should I think also be in Wikidata and be reasonably in sync with it. Prototyperspective (talk) 18:19, 30 October 2024 (UTC)[reply]

A WD item corresponding to a WP list should not not be a subclass of the list element type. A query on P279 unsolved_problem should return a list of problems here, not WP pages with their own curated list. The WP list pages, and the disambiguation pages, have little purpose on WD which has its own way of recording the information, and should just be a bare bones record that a WP page exists. Vicarage (talk) 18:58, 30 October 2024 (UTC)[reply]
This is at most tangential to what this thread is about.
Nevertheless, I'm also interested in this. So what's the Wikidata list of open problems in biology? Is it nearly as complete as the Wikipedia list and what's the rationale of not making that item a subclass of open problems but just be a useless item that isn't showing up in relevant queries and just record[s] that a WP page exists? Prototyperspective (talk) 19:32, 30 October 2024 (UTC)[reply]
A list of castles is neither an instance nor a subclass of castle, its a subclass of list, but it would be pretty pointless to record it. I note that list of unsolved problems in biology (Q2628887) correctly reports a constraint violation over this. Making WD as comprehensive as the WP lists is a good idea, or ensuring the category statements become instance/subclass ones, but that would be done by annotating the list/category members, not the list/category items. Vicarage (talk) 20:37, 30 October 2024 (UTC)[reply]
So as far as I understand the correct way would be to have a new wikidata item for "unsolved problems in biology" with that subclass statement and a statement has list: Q2628887. If that's the case I think it would be better to have the item better findable and better interlinked, it's expectable that the WP list only lists a subset of these, this is already implied by it being linked as WP item so I think the way to add subclass of to List items would be better in many cases, including this one. Prototyperspective (talk) 21:37, 30 October 2024 (UTC)[reply]
If you make a proposal to automatically add information based on some Wikipedia category and the examples you bring are constraint violating, that's not tangential. Correct, ontology matters to Wikidata. is a list of (P360) works fine for stating what lists are about. There's no reason to misuse subclass of (P279) for it.
Data imports from Wikipedia should only happen if you have a way to add a lot of claims that are correct and not constraint violating.
When it comes to Vuze (Q3563863) programmed in (P277) Java (Q251) that's something someone could import from Wikipedia but it would be better to import that information from a reliable source outside of Wikimedia. Instead, of having a goal of "increasing data a lot" it's often better to have a goal like increasing quality data. ChristianKl ❪✉❫ 09:33, 31 October 2024 (UTC)[reply]
  • Sorry if it was unclear – you misunderstood. Of course it's relevant but it's tangential to what the actual subject of this thread is. I may have chosen those examples not very well. One could also consider the thread to be a dual-subject thread but the thread-title and far far more important subject is the other subject so I hope that doesn't drown in the discussion about this.
  • is a list of would work fine if there was a wikidata item for unsolved problems in biology but there is none and again I think it would be preferable to set this on the Wikidata item of the Wikipedia's list item because that item is not defined by the Wikipedia list – it only has that linked in the interlanguage links. Maybe it becomes clearer that way: the item could be renamed to "unsolved problems in biology" and it's just normal for Wikipedias to describe this with an incomplete list. Why would it not be better to set these things on the item that is linked on the Wikipedia list? If you want to somehow be able to have Wikipedia lists always have a list Wikidata item is somebody creating such open problems items? I'm interested in these in particular (and think such content aggregation is of special usefulness & interesting, e.g. when it comes to modern problem-solving) so I picked that example.
  • Yes, again it's not about the particular examples. It could spark or be done after discussion of some constraint violations that maybe should not be some. One could also make it so that cat-syncs that would result in constraint violations are resolved in predefined ways such as "create a new Wikidata item for open problems in {field/subject} when importing from cat Open problems" etc. It wouldn't import anything that is constraint-violating (there may even be a list of potential catsync imports that would be constraint-violating so one can go through them starting with those that would change most items and resolve what to do about the WD constraint-WP-cat-mismatch with one option being to simply not sync that and other options including changing the constraints or writing the data differently or adding it to different items etc).
  • Yes, importing this programmed in data from GitHub is another thing I wanted to ask about at some point. The goal is not to increase data a lot, it's to at least match the data already in Wikipedia or Wikimedia Commons (and my example was WMC not WP). For items that have categories, the WD item usually does not have as much data as the categories – so what's the point of the WD item...e.g. who would use that instead of the category if the latter has more data.
Prototyperspective (talk) 13:16, 31 October 2024 (UTC)[reply]
Having Wikipedia lists (pages with templates like Template:Dynamic list (Q11164797) be linked to Wikimedia list article (Q13406463) is a general policy we use. We do that because in general sometimes a Wikipedia has both a list and another article to the same topic. The decision of what templates to use is made by Wikipedia and not one that's made by Wikidata.
If you import Vuze (Q3563863)programmed in (P277)Java from Github, you have a well-sourced statement and thus the infobox for Vuze (Q3563863) could import that statement. If you import the same statement via Wikipedia categories, you can't as easily use it in Wikipedia infoboxes because Wikipedia cares about statements being sourced.
Note that the word sync has a different meaning than the word import. To do syncing you would also need to decide with what happens when data changes. Data might change on the Wikidata side and it might change on the Wikipedia side (and each Wikipedia has their own bot policies if a bot is supposed to do changes). We do have data imports from Wikipedia from ten years ago where the data changed on the Wikipedia side but we still host the same data on Wikidata as before. ChristianKl ❪✉❫ 14:35, 31 October 2024 (UTC)[reply]
Yes, I'll ask about such imports from GitHub & GitLab another time and I was surprised when I noticed this hasn't been done so far. 1. It was however only an example. 2. When it comes to these, one could still set that on the item and then only add the source whenever such a GitHub import is done and have it without source prior to that.
Yes, and that's the reason why I used the word sync – category changes should update the data on the item(s) and Wikidata items changes should e.g. suggest a category change on WMC & WP. Prototyperspective (talk) 15:28, 31 October 2024 (UTC)[reply]
Asking about GitHub imports is likely going to do very little. Bots are generally run on Wikidata because someone cares about important specific data. Wikidata is largely a place where people with different interests come together and people work on what they care about.
Speaking of Wikipedia in the singular seems a bit naive. You would need to get a bot approval in every project in which you would want to run a bot to do something. Convincing any of the major Wikipedias to allow that is not going to be easy.
If you would want to do that kind of interwiki politics you could create a bot and seek relevant approvals, but that's a complex project. There are a lot of complex projects that someone could do to help Wikidata have more and/or better data. ChristianKl ❪✉❫ 16:07, 31 October 2024 (UTC)[reply]
WikiProject software seems to care about software. I don't see an explanation why it would do very little but then this thread is not about that anyway. Good point, I was speaking of ENWP (for now at least) which has the most complete cat tree. From the Wikidata side, the sync to Wikidata is what matters and that is what the thread is about. Prototyperspective (talk) 17:31, 31 October 2024 (UTC)[reply]
I think there are few people who are able to do the kind of bot work that are constraint by ideas about what to do on Wikidata. Constraints are usually either doing the work or about finding consensus.
I think that most people on EnWiki consider unsourced claims pretty unimportant and won't be helpful with the your goal of adding more unsourced claims to Wikidata based on category data.
Policy-wise that still leaves the question of how to model data syncronization so that mistakes like applying list of unsolved problems in biology (Q2628887)subclass of (P279)open problem (Q1321906) don't happen open and then the question of how data removal is managed. After that it's still up to finding someone who's interested in creating a bot for the task that will run regularly. ChristianKl ❪✉❫ 18:18, 31 October 2024 (UTC)[reply]
Each category import like 'Make Category:Writers by nationality write to the country prop if none is set and list items that don't have anything that is a subclass of writer or writer set in occupation' would be configured separately in the imports system. That's better than doing it all by hand or having Wikidata be far less useful than just using the more complete WP categories. Enwiki people do not consider categories unimportant which is why they're relatively well-maintained and a huge tree. This is WD anyway so I don't know why you bring it up. After that it's still up to finding someone who's interested in creating a bot This isn't about a bot itself, it's about a tool and this thread is about that – this is the project chat isn't it; I hope interested people read the starting post. On the other hand, maybe Wikidata will always be useful only for identifiers, not for statements as Wikimedia Commons and ENWP have far more data in their categories while main statements are missing or are flawed on Wikidata even for items as fundamental/major as 'Present', 'Future', etc and also when the correct statements could already be inferred from the WP cats. Prototyperspective (talk) 19:07, 31 October 2024 (UTC)[reply]
Again, this is a proposal to add bad data. country (P17) is not supposed to be used for people and Wikidata does not have a "nationality" property (the last three proposals were rejected).
You used the word "sync" instead of the word "import". Import is a word that's about moving data in one direction while "sync" is a word to move it in two directions.
Anything, that regularly adds data is a bot.
It's worth noting that tools like PetScan exist and someone who wants can do get a list of items from a category with them and use QuickStatements to enter data. Syncronization data however is more complicated and requires you to have an ungoing process and for that you need a bot. ChristianKl ❪✉❫ 19:59, 31 October 2024 (UTC)[reply]
Again you focus on the specific example. I know there's many properties for countries and I don't know which is the correct one so I just said country.
Yes, again I said sync exactly because of that and I already clarified that.
A bot only executes, a tool like mix'n'match is more comprehensive than that.
PetScan exist and someone who wants can do get a list of items from a category with them and use QuickStatements this is the first thing you wrote (and this not a thread on your talk page) that actually relates to the thread subject, thanks. Maybe the tool could be be built around PS+QS. I didn't say a bot wasn't needed, the best ways to implement this would probably involve one (existing or new). Prototyperspective (talk) 21:20, 31 October 2024 (UTC)[reply]
I asked you for how you would think about preventing bad data to be added with the tool. That's an important question. Instead of answering it, you suggest an example that adds bad data.
As I said, we don't have a property for nationality. Nationality is a social construct over which wars are fought. We don't want to fight in Wikidata about whether people in Crimea are of Russian or Ukranian nationality. It's a construct for which we in the past decided not to add properties.
A tool like Mix&Match adds data when you tell it to add data. That's the permission it has. It doesn't add data a year from now. If you actually want to keep data up to date, you would need something that can add data in the future and that's a bot. ChristianKl ❪✉❫ 23:30, 31 October 2024 (UTC)[reply]
  1. Every change batch/query/rule would need to be discussed first.
  2. False, there are widely used props like Property:P27
  3. One could sync semi-manually. No, a tool is more comprehensive than a bot. A bot would likely be part of the system somehow.
Prototyperspective (talk) 14:05, 1 November 2024 (UTC)[reply]
country of citizenship (P27) is for the country of citizenship which is something different than nationality. If you take the example of Ukraine, for a lot of time Ukraine wasn't a country so nobody could have been a citizen of it but Ukrainian nationalists still consider those people who spoke Ukrainian at the time to be of Ukrainian nationality.
Citizenship is a well-defined concept. Nationality is a concept that's a lot more subjective. Putin has quite different ideas about who's of Russian nationality then Zelenskyy.
Given that the Wikipedia category is not about citizenship but about nationality you can't import it into the citizenship property.
Data-modeling matters. ChristianKl ❪✉❫ 19:05, 1 November 2024 (UTC)[reply]
Again it was an example and another field it can be imported into can be found if more appropriate. When it comes to your particular objections: a property can have multiple values and you can you can add disputed qualifiers. Prototyperspective (talk) 19:26, 1 November 2024 (UTC)[reply]
The key issue, is that if you want to effectively contribute to Wikidata, it takes understanding how Wikidata works. That especially goes for if you want to make more systematic changes.
It would be possible that this community decides to have a nationality property. There have been three times that one was proposed and every time our community decided against having one. Even if you personally would decide differently, that does not change the consensus we had when rejecting those proposals.
If someone has not put in the mental effort to understand how the examples the bring actually work, that usually means that also have no idea about the effects that what they propose will have for the more general case. ChristianKl ❪✉❫ 21:35, 3 November 2024 (UTC)[reply]
Makes sense, but it's still important to at least focus on the actual proposal rather than the examples and there are nationality properties. I already specified them multiple times for musicians and they're widely used. I did not know which particular country prop would be applicable here and didn't want to check and it's not about the subject anyway. Prototyperspective (talk) 21:43, 3 November 2024 (UTC)[reply]

Faruqolaitan123 creations

[edit]

The user @Faruqolaitan123 seems to have spent 2022 creating a lot of hard to identify individuals, many of the items they have created seem to have only a single name and the most basic statements. What should one do when coming across cases like these? Nominate each for deletion or something else? StarTrekker (talk) 20:25, 30 October 2024 (UTC)[reply]

I looked through the human items created by Faruqolaitan123, and didn't see any indicators of notability or see any structural need for them (they don't have any links to other items). I also tried to match details on a few, without luck. I would nominate for deletion. Iamcarbon (talk) 01:48, 1 November 2024 (UTC)[reply]

Accidentally created an item via typo not realizing it already existed elsewhere

[edit]

I created Cantopotamon zhuaiense (Q130745113) not realizing that Cantopotamon zhuhaiense (Q107084939) already exists. This came from an existing typo on a Commons image, and I was careless and also didn't notice the typo. Given I don't assume there are gaps in the serialized IDs, does this QID just get recycled and used for something else? Somewhat new here. TheTechnician27 (talk) 18:11, 31 October 2024 (UTC)[reply]

We don't recycle Q-IDs. Just merge the items. ChristianKl ❪✉❫ 18:47, 31 October 2024 (UTC)[reply]

Scope of official website (P856)

[edit]

There's a bit of a back-and-forth on Daltile (Q120633664) over whether a subdomain of an official website should be listed in official website (P856) when 1) the main domain/page is already listed and 2) the subdomain website is already listed in a more specific property. I read the talk page of P856, and I came away with the impression that P856 is reserved for URLs that point to pages that are, were, or can be easily mistaken for the home pages of a website. Third-party guidance on this issue is appreciated. Thanks. BrownCat1023 (talk) 05:20, 1 November 2024 (UTC)[reply]

There seems little point having 2 entries for P856, especially when P12454 is used. Using https://daltile.com and relying the site's server rules to add the www would also be sensible. Vicarage (talk) 06:36, 1 November 2024 (UTC)[reply]
Thanks for the feedback. I removed the second entry for P856 again with a link to this thread in the edit summary. BrownCat1023 (talk) 16:57, 1 November 2024 (UTC)[reply]

Soliciting SPARQL queries for inclusion in benchmark

[edit]

I am starting work in small project to benchmark query engine performance on useful Wikidata queries. I am asking for queries from the Wikidata user community to potentially be part of the benchmark. If you are a user of any Wikidata SPARQL service please send queries that you find useful to my email. Say what you used the query for and whether you would like to be noted as the source of the query. Queries that take considerable time or time out are especially welcome, particularly if the query caused you to switch from the official Wikidata Query Service to some other service. More information about the project is available in Wikidata:Scaling_Wikidata/Benchmarking. Peter F. Patel-Schneider (talk) 12:32, 1 November 2024 (UTC)[reply]

Items for humans without birthdate in the item and article in de-WP
Items for humans without german description in the item and article in de-WP:
Items for films without german description in the item and article in de-WP:
Geographical items without coordinates and article in de-WP:
  • query-main.wikidata.org (has to be split in two parts/runs regarding the list of objects, one single list/run would be preferred)
M2k~dewiki (talk) 13:41, 1 November 2024 (UTC)[reply]
Thanks. Peter F. Patel-Schneider (talk) 14:15, 1 November 2024 (UTC)[reply]
In particular, note that the last query (with all the parts active) runs in under 2 seconds on QLever. https://qlever.cs.uni-freiburg.de/wikidata/VK6CZi Peter F. Patel-Schneider (talk) 14:36, 1 November 2024 (UTC)[reply]
I know, but QLever only return 198 items, while WDS finds about 6.000 items in the first part and 1.500 in the second part. M2k~dewiki (talk) 14:41, 1 November 2024 (UTC)[reply]
Additional information: the missing information found by these queries can be extracted from templates and infoboxes from de-WP using some scripts from
and can be imported using QuickStatements. Also see
M2k~dewiki (talk) 14:58, 1 November 2024 (UTC)[reply]
Ah, that's because just removing the # produces a different query. Adding the required UNION results in https://qlever.cs.uni-freiburg.de/wikidata/lY46kD, which has about 7500 results. Peter F. Patel-Schneider (talk) 14:59, 1 November 2024 (UTC)[reply]
The advantage of WDS is in that case, that I can run it on a daily base, while QLever is updated only once a week. M2k~dewiki (talk) 15:03, 1 November 2024 (UTC)[reply]

Rock vs Rock formation

[edit]

Hi, can someone please briefly explain the semantic difference between rock (Q1404150) and rock formation (Q631305)? Thanks! Ping @Helena Dvořáková: with whom we've been discussing this. Vojtěch Dostál (talk) 10:30, 3 November 2024 (UTC)[reply]

I don't know much about geology. If we look at the list of Wikipedia articles the Spanish Wikipedia seems to have different articles for both. Asking ChatGPT for the difference between the two articles gives us:
"Peña" (from the first article) appears to be a specific type of rock formation characterized by:
An isolated rocky hill or outcrop
Typically formed by glacial action
Has a specific formation process where glaciers erode weaker surrounding material
Often features a characteristic "tail" (cola) on the leeward side
"Formación rocosa" (from the second article) is a broader geological concept that:
Encompasses all types of rock outcrops
Includes multiple formation processes (not just glacial)
Can be formed from any rock type (sedimentary, metamorphic, or igneous)
Is more of an umbrella term used in geomorphology
In Wikidata terms, "peña" would be a subclass of "formación rocosa". Think of it like this:
"Formación rocosa" is like the parent category "vehicle"
"Peña" is like a specific type of vehicle, say "pickup truck"
So if you're creating or editing Wikidata items, "peña" should probably have a "subclass of" (P279) relationship pointing to "formación rocosa" or a similar broader geological formation concept. ChristianKl ❪✉❫ 21:20, 3 November 2024 (UTC)[reply]
@ChristianKl Thanks Christian, we'll probably stick to rock formation (Q631305) as it's more likely to be well defined. From what you have written here, it seems that the Spanish meaning of rock (Q1404150) is quite distinct and difficult to translate across languages and I'm afraid the item's definition might still be subject to future changes. Vojtěch Dostál (talk) 09:59, 4 November 2024 (UTC)[reply]

Second edition with expanded title

[edit]

Unsure how to treat a WD item for a literary work whose second edition has an expanded title. Same publisher but obviously different publication dates, number of pages and ID number in the National Library catalogue:

And another query: what's the statement for the literary expression (Q7751154) that the title is based on?
Am awaiting advice on how to add the National Library of Israel catalogue number. -- Deborahjay (talk) 16:51, 4 November 2024 (UTC)[reply]

Wikidata weekly summary #652

[edit]
Here's your quick overview of what has been happening around Wikidata in the
week leading up to 2024-11-04. Missed the previous one? See issue #651

Discussions

Events

Press, articles, blog posts, videos

Tool of the week

Other Noteworthy Stuff

  • A small project on benchmarking query engine performance on useful Wikidata queries is asking for queries from the Wikidata user community to potentially be part of the benchmark. If you are a user of any Wikidata SPARQL service please send queries that you find useful to Peter F. Patel-Schneider. Say what you used the query for and whether you would like to be noted as the source of the query. Queries that take considerable time or time out are especially welcome, particularly if the query caused you to switch from the official Wikidata Query Service to some other service. More information about the project is available in Wikidata:Scaling_Wikidata/Benchmarking.

Newest properties and property proposals to review

You can comment on all open property proposals!

Did you know?

Development

  • Mobile statement editing: We are making progress on the technical investigation for how to make it easier to edit statements on mobile. A lot more work to be done after that though.
  • We fixed the sidebar link to the main page in many languages (phab:T184386)
  • Codex: We are continuing with the migration of the Query Builder to Codex, the new design system. The migration of Special:NewLexeme is almost finished.
  • Query Service: We have updated the list of languages for the language selector in the UI (phab:T358572)
  • Vector 2022: We are continuing to adress issues of the Item UI in dark mode (phab:T369385)
  • Wikibase REST API:
    • We are moving from v0 to v1.
    • We have finished the work on the new endpoint for creating Properties.
  • Action API: We’re improving the way the wbformatvalue API handles invalid options (phab:T323778)

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

Q5396050, Q1214130

[edit]

Hi. Can someone, please sort out the mess KFan3 has created by merging Q5396050 and Q1214130. Though the two persons have the same name, one of them is a pianist the other one is an actor. Ervin Nagy pianist has article on English Wikipedia only. Thanks. Csigabi (talk) 18:43, 4 November 2024 (UTC)[reply]

Seems to have been sorted out by M2k~dewiki. You simply need to use that "restore" button just before the mess was introduced. --Matěj Suchánek (talk) 12:41, 5 November 2024 (UTC)[reply]

Hello

[edit]

This wiki data page was made and now it has no title, how do I delete it thanks https://www.wikidata.org/wiki/Q130896158 Icecreamlips21 (talk) 22:08, 4 November 2024 (UTC)[reply]

Don't remove the statements beforehand and add it to WD:RfD, there is also a gadget that helps you nominate items for deletion. Sjoerd de Bruin (talk) 22:40, 4 November 2024 (UTC)[reply]
You don't delete it. You first restore the bad edits you made that removed the title and then you can nominate it for deletion by clicking More/Request for Deletion. ChristianKl ❪✉❫ 10:54, 5 November 2024 (UTC)[reply]

"fish ladder (Q606619)" and "fish passage" (Q12366145)

[edit]

Hi, I am pretty sure that the two WD objects "fish ladder (Q606619)" and "fish passage" (Q12366145)” represent the same thing. However, “fish ladder” seems to me to be the more common name, at least that is my gut feeling. It would be good if someone could merge the two objects. --Mister Pommeroy (talk) 07:54, 5 November 2024 (UTC)[reply]

What makes you think so, based on the description and duplicate sitelinks I doubt that? Sjoerd de Bruin (talk) 08:52, 5 November 2024 (UTC)[reply]
I came up with this because the descriptions in the articles in the different Wikipedias for both the items listed under “fish ladder” and the items listed under “fish passage” refer to the other name. A few examples are:
  • enWP: A fish ladder, also known as a fishway, fish pass, fish steps, or fish cannon, is a structure on or around artificial and natural barriers (such as dams, locks and waterfalls) to facilitate diadromous fishes' natural migration as well as movements of potamodromous species.
  • deWP: A fish ladder or fish pass [...] is a hydraulic engineering facility on watercourses to enable fish to overcome migration barriers in the form of weirs, hydropower plants and, if necessary, waterfalls during fish migration.
  • itWP: A fish passage is a hydraulic engineeringwork built along rivers or streams at artificial dams, such as weirs or dams. Alternative designations are “comeback ladder,” “upstream ladder,” or “fish ladder.”
In most Wikipedias, only one article exists under one name or another. In the Spanish Wikipedia, there are two articles under the different names, but it has been suggested that the articles be merged. --Mister Pommeroy (talk) 11:33, 5 November 2024 (UTC)[reply]
fish passage (Q12366145) is a general term. Currently, we have fish ladder (Q606619), fish ramp (Q12366148), and fish lift (Q28382647) as examples on it in Wikidata. Given that there are different types of fish passages than just fish ladders, they don't belong merged. If you want Interwikilinks, https://www.wikidata.org/wiki/Wikidata:Sitelinks_to_redirects#:~:text=The%20main%20goal%20of%20sitelinks,represented%20by%20separate%20Wikidata%20items. explains how to get them. ChristianKl ❪✉❫ 11:44, 5 November 2024 (UTC)[reply]
Retrieved from "https://www.wikidata.org/w/index.php?title=Wikidata:Project_chat&oldid=2270398034"