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

Wikidata:Project chat

From Wikidata
Jump to navigation Jump to search

Should we certify Wikidata trainers?

[edit]

Many well-intentioned new users arrive at Wikidata without understanding how to contribute effectively. Nevertheless, they often jump straight into creating items. Some have no notion of notability and create items about themselves and their friends. Some don't know how to write descriptions or use aliases. Some don't know how to add claims. Some don't realise that they're expected to respond to feedback. These new users can be coached and corrected individually, but it would be more efficient if we could somehow educate them before they started editing.

Many of these users are first introduced to Wikidata through training sessions and editathons, often funded by WMF grants. I have reviewed numerous grant applications on Meta, selecting those related to Wikidata, identifying the editors involved, and assessing their Wikidata edit history. Some trainers are clearly qualified, but many lack experience or have a history of problematic editing. The WMF does not appear to conduct a thorough review of grantee qualifications. In some cases, grants have been approved even when applicants demonstrate a lack of Wikidata experience or a pattern of problematic editing. A result of this is that WMF funds may inadvertently create extra work for project volunteers.

I have interacted with users who claim to be certified trainers, but there is no clear way to verify their credentials or determine the criteria. When I have been able to obtain details, it seems to mean that they sat through a slideshow. Wikidata trainers should be experienced editors who have demonstrated a mastery of best practices. I therefore propose a lightweight community process to certify Wikidata trainers, improving the quality of Wikidata training, while also assisting WMF staff in evaluating grant proposals.

  1. New user group "trainer" managed via Request for Permissions. This group need not be associated with any actual permissions.
  2. The criteria for "trainer" are based on their contributions here:
    1. Substantial (TBD) experience editing Wikidata
    2. Conformance to project policies and guidelines, especially notability
    3. Appropriate response to feedback
  3. To prevent unnecessary clean-up work for volunteers, the community expects the WMF to ensure that any grant-funded program introducing new editors to Wikidata includes at least one certified trainer. Grantees are expected either to recruit a certified trainer or apply for certification themselves.

Related discussions and resources: meta:Wikimedia Conference 2018/Program/41, Wikidata:Train the trainers, Wikidata:Project_chat/Archive/2023/12#Wikidata-related_grant_proposals, commons:Commons:OpenRefine/Train-the-trainer program 2023-24, Wikidata:Open Online Course, commons:File:2022-12-02 Formation Wikidata 3 heures.pdf, Wikidata:Administrators'_noticeboard/Archive/2018/07#Using_English_Wikipedia_as_a_reference_url, Wikidata:Administrators'_noticeboard/Archive/2024/02#All_the_participants_of_an_editathon_blocked, meta:Talk:Wikimedia_Foundation_Board_noticeboard#Misuse_of_grant_funds, Wikidata:Event Organizers

Bovlb (talk) 05:17, 30 January 2025 (UTC)[reply]

Thanks for taking valuable time to review grant applications for editathons and similar. I think you pretty much confirm what some of us have suspected that some arrangers are way better at writing the grant applications than teaching others the basics of Wikidata. The participants will have items deleted or worse, so they will have a bad first time experience and will likely never come back. And existing voluntairs will have to do the clean up.
I see two viable solutions. Stop giving grants that cover anything other than direct expenses for editathons to do away with the perverse incentive. A formal vetting procedure should be established. This should be a clear and simple set of requirements to make it fast to assess candidates. I basically agree with your suggestions. Experience could be boiled down to having 1000 manual edits on Wikidata. They don't have to read all the policies, just the most important ones and they should have read pretty much all of the help pages. When you say "Wikidata trainer" I'm assuming you mean event organizers should be explicitly told not to let participants edit on Wikidata unless they have demonstrated the necessary know how? This might include people who are organizing Wikipedia-only events.
Permissions can include the permission to bulk-create new users, as we want to avoid editathon participants editing with similar IP addresses and be perceived as socks. Infrastruktur (talk) 08:01, 30 January 2025 (UTC)[reply]
Thanks for your support.
While I share your concerns, I don't think it is within our power here to eliminate perverse incentives from WMF grants. I also don't think we can exercise fine-grained control over what happens in local meetings.
The recurring scenario seems to be when people who don't know how to contribute to Wikidata attempt to train others, they will reproduce their own problems. My hope is that simply by ensuring someone involved knows what they're doing, we will naturally eliminate many problems of this type. I realise that this proposal will not solve all problems in this area, but it seems like a small concrete step we can take, that could have a significant benefit. I raised this issue with the board of trustees, and was told that they are trying to do better at qualifying grant proposals. Providing a credential and making that a pre-requisite should make it easier for WMF staff to do that job.
I accept your point about bulk-creation of new users, and note the existing Wikidata:Account creators user group. Bovlb (talk) 15:34, 30 January 2025 (UTC)[reply]
Given that we do control what goes on on Wikidata and don't control what the WMF is doing when it comes to grant appplication solving the issue with a user right at Wikidata seems to be a good idea. That leaves the question open about whether to use an existing right or a new one.
Are there people who need the Event Organizer right and that we would trust with that right that we don't trust to be trainers? The same goes for the account creator right. ChristianKl04:16, 31 January 2025 (UTC)[reply]
I've no objection to reusing an existing user group, but Event Organizer has very different criteria. Looking at the 8 current holders, I see that one has only 646 edits, despite holding this right since November. I wouldn't want to make it harder to get permission to use the CampaignEvents extension because we had added three major criteria. The other candidate user group is Account Creator, which might also be needed by trainers. This group has no published criteria and no non-bot holders. I proposed a new group because I wanted the purpose of the group to be crystal clear. Bovlb (talk) 15:22, 31 January 2025 (UTC)[reply]
Update: It looks like that one low-edit rights holder may be an alt account for a WMDE staff member. All the others holders of Event Organizer have large numbers of edits. Bovlb (talk) 17:36, 31 January 2025 (UTC)[reply]
Responses have been few, but positive, with two specific suggestions:
  • Use 1,000 manual edits as a threshold for experience
  • Re-use an existing user group, such as Event Organizer.
Bovlb (talk) 20:47, 6 February 2025 (UTC)[reply]
I support the initial proposal Bovlb and also the last message (1000 manual edits as threshold and re-using an existing user group), assuring events don't add cleanup for the community is surely desirable. Epìdosis 19:18, 10 February 2025 (UTC)[reply]
The existing "Event Organizer" criteria include either receiving a grant or having experience in organizing an event. This would be an issue for this proposal as we would like to require trainer certification before doing either. Bovlb (talk) 20:38, 11 February 2025 (UTC)[reply]
In theory, we don't even need a specific user group, it could even be a list editable only by admins, like how enwiki handles AWB access, if trainers do not need specific user rights. I feel re-using an existing group can cause confusion in the future if we decide to use it more, and I'm not sure sysadmins will be too happy to add a user group that (from a technical/wiki access POV) does nothing.
I'm definitely in favour of having some certification process. KonstantinaG07 (talk) 10:01, 12 February 2025 (UTC)[reply]
Advantages of having it be a user group include that it shows up on the global accounts page, and there's a clear log of it being granted (and revoked). There's some precedent for having single-project user groups, but less for user groups with no associated interface rights. Bovlb (talk) 21:05, 12 February 2025 (UTC)[reply]
It doesn't sound like a user group that does nothing will be a sticking point. Bovlb (talk) 03:58, 13 February 2025 (UTC)[reply]
As long as there are support for it he seems willing Trade (talk) 00:30, 14 February 2025 (UTC)[reply]

Do we really want to include people's self-diagnosed conditions as part of Wikidata? I feel iffy about this--Trade (talk) 03:06, 5 February 2025 (UTC)[reply]

Maybe a different name like "claimed" and/or possibly aggressive removals of instances that are not sourced to a statement by the person? Circeus (talk) 17:27, 9 February 2025 (UTC)[reply]
Or just get rid of them entirely Trade (talk) 00:29, 14 February 2025 (UTC)[reply]

Modelling the banning of a book

[edit]

Various jurisdictions ban and restrict books, movies and similar for various temporal periods. I'm looking for an example of how to model this. I was expecting to find it in Lolita (Q127149), but that doesn't seem to mention it's banning, despite it being covered multiple times in w:en:List_of_books_banned_by_governments. Stuartyeates (talk) 06:42, 10 February 2025 (UTC)[reply]

I don't find many examples, but a naive approach would be something like:
significant event (P793)
Normal rank banned book (Q65770793)
start time (P580) 1945-06-06
valid in place (P3005) Austria (Q40)
laws applied (P3014) Verbotsgesetz 1947 (Q571913)
0 references
add reference
add value
While applies to jurisdiction (P1001) and legislated by (P467) could be used, I think it would be redundant with those statements being on the law. If a agency applies a law, it could possible be modeled with authority (P797). Ainali (talk) 07:55, 11 February 2025 (UTC)[reply]
Thank you Ainali, I have created One for the gods (Q132218611), which in New Zealand has been censored in two ways, a complete ban and and R18 restriction. I've managed to articulate everything but this difference between the two time ranges. I'm using censorship (Q543) rather than banned book (Q65770793) or something else, because at least in New Zealand censorship is media-independent. Thoughts? Stuartyeates (talk) 07:24, 12 February 2025 (UTC)[reply]
I think adding a qualifier object of statement has role (P3831) -> R18 (Q66813043) is at least a start even though it is not perfect. What were your thoughts on not going with valid in place (P3005) or authority (P797)? Ainali (talk) 08:44, 12 February 2025 (UTC)[reply]
applies to jurisdiction (P1001) seems better to me, because it's more specific than valid in place (P3005). I have in mind Index Librorum Prohibitorum (Q22765)/list of prohibited books by the Russian Orthodox Church (Q64620541) which apply to all members of the church, where ever they are. authority (P797) seems less clear in cases such as this than participant (P710), because in New Zealand (and I believe in the case of Index Librorum Prohibitorum (Q22765)) different parties decide on the censorship classification and enforce breaches. authority (P797) implies enforcement to me. Stuartyeates (talk) 09:00, 12 February 2025 (UTC)[reply]
I've just added a statement to Dialogue Concerning the Two Chief World Systems (Q1768303). Stuartyeates (talk) 09:44, 12 February 2025 (UTC)[reply]
I was thinking authority (P797) since we were talking about laws before. And if it is not enforced, is it really a ban or not more of a (non-)recommendation?
Separately, I thought censorship (Q543) was a term specifically used when made by a state, and thus wouldn't be the term to use when it is your random movement/organization trying to do a similar thing. Ainali (talk) 06:53, 14 February 2025 (UTC)[reply]

Register to the Wikidata Data Reuse Days, starting next week!

[edit]

Hello everyone,

"sign up for the online event”

Our upcoming online event, the Data Reuse Days 2025, is starting next week, on February 18th!

The event will be focusing on how people and organizations use Wikidata's data to build interesting applications and tools. It aims to bring together data reusers and editors to learn from each other and discover cool Wikidata-powered applications. We will also cover technical aspects of Wikidata and share research and ideas for the future of reusing Wikidata’s data.

The event is taking place online, we will have sessions taking place during periods of 2 hours through 8 days, from February 18 to 27, 2025. You will be able to join sessions live, or to watch the replay later on. The sessions take place in English, and are open to everyone interested in the topic of data reuse. You can look at the program to discover the many topics we will touch during the event: introduction sessions, livecoding, application showcase sessions, AI and LLMs, updates about APIs and Query Service…

For the first time in our events series, we are using the CampaignEvents Extension that allows us to have a registration list onwiki. If you plan to attend, we strongly encourage you to register on the event page so you can receive updates about the event and access the Jitsi room link.

If you have any questions or suggestions, feel free to write on the talk page or to reach out to me directly. We’re looking forward to holding this event with you! Best, Lea Lacroix (WMDE) (talk) 08:15, 10 February 2025 (UTC)[reply]

"author name string" => "author"

[edit]

Is there a quicker way to convert a heavily referenced "author name string" field into an "author" field other than just creating a new field for "author", copy-and-pasting the references over one by one, and then deleting the original? DS (talk) 17:34, 10 February 2025 (UTC)[reply]

Something like Author Disambiguator (Q76693569)? RVA2869 (talk) 17:59, 10 February 2025 (UTC)[reply]
Author Disambiguator is my preferred tool for that task. Works well and preserves all qualifiers and references. William Graham (talk) 20:48, 10 February 2025 (UTC)[reply]

Wikidata weekly summary #666

[edit]

Potential property (British Isles) - Wikishire - thoughts?

[edit]

Wikishire (Q22080966) seems to be a parallel wikipedia for "notable" places in the British Isles, including the Republic of Ireland (although to what extent I'm unsure). I can't find any property proposal or prior discussion of Wikishire and would appreciate some thoughts before jumping to proposing it. Tæppa (talk) 22:52, 10 February 2025 (UTC)[reply]

There are 3 instances of WD items using Wikishire as a P973 "Described at URL": Q21995548, Q17525861, Q2106793 Tæppa (talk) 22:58, 10 February 2025 (UTC)[reply]
Go ahead and make it Trade (talk) 04:27, 11 February 2025 (UTC)[reply]

Help with classifications

[edit]

Hello everyone, I'm reaching out for help on something I want to be working on. I'm a research chemist and I'd like to use wikidata to write a list of software used in chemistry. Moreover, it would be very nice to classify the software according to what it is specifically used for.

For example, some software is used to only visualize molecular structures, other software is used to edit them and other ones are used to execute simulations or to compute quantum properties. Some software may cover several of these functionalities.

My aim is to enter the information in a structured way so that it is later simple to run a query asking for all the software which can be used to do a specific thing.

I have been doing the following: I noticed the entries "Chemistry Software" (Q4497736) instance of "Software Category" and "Molecule Editor" (Q1963488) subclass of "editor" already existed.

I went through several chemistry related software and added the statement instance of "Chemistry Software".

In "Molecule Editor" I added subclass of "Chemistry Software" and I added the entry "Molecular Dynamics Engine" (Q132189548) subclass of "Chemistry Software".

This works fairly fine to retrieve categorized software as I can structure a query retrieving exactly what I want, I just want to make sure this is the correct approach to do this.

I don't know whether it could be a problem, but this could potentially lead to the same item being instance of several subclasses of "Chemistry Software".

Let me know what you think about this and whether there are better ways to do this. You can go through my edits history to see what I did specifically. Davidoskky (talk) 10:56, 11 February 2025 (UTC)[reply]

Itemviews?

[edit]

Is there a way to get "itemviews" like pageviews on Wikipedia? Abductive (talk) 11:32, 11 February 2025 (UTC)[reply]

Yes. Example. (And with m:User:Ainali/PreViewStats.js in your global.js you can get it in the page header on every item/article on any wiki.) Ainali (talk) 19:55, 11 February 2025 (UTC)[reply]

Modelling trade unions / 'linked' organisations

[edit]

Hello! I have a question occurring from creating an item for Bournville Works Men's and Women's Councils Q132197356 Does anyone have any suggestion on how to model the related industry and/or company? I see that the RMT Q6979153 have 'industry' P452 as transport. But they are not in the business of transport, if that makes sense, so it doesn't seem to quite fit for me?

Anyway, for now I have put 'industry' for the Bournville council, following the RMT model.

Another relationship to model would be the specific companies that unions are set up to work with. In the UK at least, there are some unions that are specific to just 1 company, whereas other unions may work with a list of selected companies (that is to say, some of these lists would be one related item, but others could go into the 10s). Is there a suitable property to express this relationship?

eg. Bournville Works Men's and Women's Councils Q132197356 = ?some property to express relationship to = Cadbury Q6677525

This does seem to be a problem across Trade Union / Students' Unions items, I just looked over a few as examples (it would be nice to be able to add this info to their items too!):

Thanks for any ideas / pointers! Medievalfran (talk) 12:21, 11 February 2025 (UTC)[reply]

According to the UN standard for nonprofits ICNPO (Q124022726) use labor union (Q178790) as industry (P452).
For linked industries, I would use field of work (P101), but there is no uniformity yet. Best NGOgo (WikiProject Nonprofits!) 14:23, 11 February 2025 (UTC)[reply]
Thanks very much, I'll use that for now! If anyone has any thoughts on how to link with relevant specific companies that the unions negotiate with, that would also be appreciated. Medievalfran (talk) 17:34, 11 February 2025 (UTC)[reply]

question re wikipedia list properties

[edit]

Does it make sense for this item "List of garden structures at West Wycombe Park" to have the P31= "folly" (type of structure) and the country / historic county properties describing the location of the objects in the list?

Is there a reason it wasn't marked with P31= Wikimedia list article by a bot? Tæppa (talk) 02:52, 12 February 2025 (UTC)[reply]

I think you are right, and we don't want these list items cluttered with other properties Vicarage (talk) 04:10, 12 February 2025 (UTC)[reply]
A user with a red username has since removed the property from that particular item, but I've just seen a couple more examples, these two being the only ones in Middlesex:
"list of residents of 10 Downing Street" (Revision that adds P7959)
"list of public art in Islington" (Revision that adds P7959)
Is there scope for requesting a bot edit to apply P31 Wikimedia list article to all wikipedia list items? Any other "ontology" changes like requiring lists to have P360 "Is a list of" and deprecating location properties? I ask this not to start a discussion in this thread but to understand the playing field before messing with items that appear in SPARQL results as being in historic counties of England Tæppa (talk) 06:39, 12 February 2025 (UTC)[reply]
Location properties tend to be WAY overused in my experience. Why does Pan American Games (Q230186) needs continent (P30)?? Why does 1951 Pan American Games (Q1203233) needs both location (P276) and country (P17)? I dare not ask, and dare even less Be bold. Circeus (talk) 21:42, 12 February 2025 (UTC)[reply]

Direction of duplicate item merges

[edit]

I came across the edit histories of Q123689179 and Q132191662 earlier tonight. The merge gadget has "merging into the older entity" selected by default, so I'm assuming that this is the preferred direction of duplicate item merges. Since this isn't what happened with Q123689179 and Q132191662, should the direction of the merge be reversed, or should things just be left as-is? BrownCat1023 (talk) 03:03, 12 February 2025 (UTC)[reply]

I'd personnaly redirect the merge, especially as it is quite recent (if it was an old merge of several years, maybe that would create more harm than good). --Jahl de Vautban (talk) 06:50, 12 February 2025 (UTC)[reply]
Thanks. I've reversed the merge target. BrownCat1023 (talk) 16:42, 12 February 2025 (UTC)[reply]

Third-party opinion requested for Q8023931

[edit]

An IP address made a series of strange edits to Freedom Mobile (Q8023931) on February 6; some were productive, some were not. I attempted to clean up the edits, only to have my efforts reverted by the IP with the summary "nonsense". I've reverted back with a detailed edit summary explaining why, but I'd feel more comfortable if another editor took a look at the situation. BrownCat1023 (talk) 03:20, 12 February 2025 (UTC)[reply]

That is a well-known LTA, the so-called "Vodafone vandal", making disruptive edits on Wikidata for over five years. Since they are not able or willing to collaborate constructively, the best way is to get the item semi-protected and ask for a rangeblock of the IP. --Dorades (talk) 19:52, 12 February 2025 (UTC)[reply]
I posted to the admin noticeboard at Wikidata:Administrators' noticeboard#Semi-protection request for Q8023931; however, no action was taken (probably because an earlier post reported the situation at Q8023931 as an "edit war", and the page was temporarily fully protected in response). The IP is apparently now on a new address based on recent edits to Q10711016. *shrug* BrownCat1023 (talk) 02:03, 16 February 2025 (UTC)[reply]
I share this experience from a few years back when I was more active in counter vandalism. It is tortuous to check all their edits and they don't seem to be problematic at a first glance. Their first registered account, as far as I now, was GM-27IT (see the related sock puppet category on it.wp), if that's of interest to you. --Dorades (talk) 13:25, 16 February 2025 (UTC)[reply]

Scope of P7213

[edit]

newspaper archive URL (P7213) is currently labeled "newspaper archive URL" (in English), and was structured as such to reflect newspapers, but I see no good reason why it shouldn't be expanded to include all periodicals such as magazines, newsletters, and scholarly journals, with a broader, more inclusive primary label such "periodical archive URL" or "newspaper/magazine archive". There is often a vague, indistinct line between what constitutes a "newspaper" versus a "newsletter" versus a "news magazine", versus a "magazine" versus a "journal" versus a "scholarly journal", and so forth, and such identities can change over time due to mergers, acquisitions, relaunches, rebranding, editorial changeover, editorial neglect, etc. The property is already used on non-newspaper items such as Computer Gaming World (Q707182), The Town Crier (Q123862477) and AMSTAT News (Q131940282) and from "What links here" there is evidence it is occasionally being incorrectly used in place of reference URL (P854) (especially on human items), i.e. a reference to a single URL source that happens to be a newspaper article. Full disclosure: I've already pro-actively added aliases to the property such as "magazine archive URL" and "back issues at", but wanted to have open discussion on the scope of the property for better consensus. Pinging @Jura1, Nomen ad hoc, TomT0m, Thierry Caro: as participants in the original property proposal discussion. Notified Wikidata:WikiProject Periodicals. -Animalparty (talk) 03:49, 12 February 2025 (UTC)[reply]

Good idea. You need to point to this discussion from the property page Vicarage (talk) 04:15, 12 February 2025 (UTC)[reply]
Ah yes, thank you. Done. -Animalparty (talk) 04:20, 12 February 2025 (UTC)[reply]

Save the Date! WikidataCon 2025 returns October 31 - 2 November

[edit]

Hello all,

I am excited to announce the next edition of the WikidataCon, taking place over three days from 31st October to 2nd November later this year.

Organised by Wikimedia Germany, this event celebrates the entire Wikidata Community in all of its aspects; editors, tool builders, patrollers, data reusers, members of GLAM and other organisations that use and contribute to Wikidata.This year, the event format will be an online-style conference with a full online-program that attendees can join and participate with program-content from anywhere in the world.

We also encourage local chapters and community user groups to organise in-person meetups alongside the event to bring people together and also celebrate Wikidata becoming a teenager for its 13th Birthday (October 29th).

What kind of sessions to expect at WikidataCon 2025?

  • Learn more about basic and advanced editing skills
  • Workshops, live-editing, games and discussions
  • Exchange tips and tricks with Wikidata experts
  • Discover community-built tools and gadgets
  • Hear about new WikiProjects and initiatives

In the coming weeks and months, we’ll be revealing more details on the program content, call for proposals, daily schedule, themes and program tracks. Please check Wikidata:WikidataCon 2025 regularly for more info.Please don’t hesitate to reach out to the organisers at info@wikidatacon.org or by leaving a message on the event talk page.

Hope to see you all at WikidataCon! -Danny Benjafield (WMDE) (talk) 10:17, 12 February 2025 (UTC) for Wikimedia Germany[reply]

Hotel Brand description

[edit]

Hotel Brand has the description "an identifiable of a hotel, usually as a part of the hotel name". Either this is an instance of wikidata employing strange syntaxes or the grammar needs fixing. The meaning of "hotel brand" is obvious but what the author intended to write as the description is beyond me and I don't want to replace it with something less intricate, so I would appreciate it if someone were to improve it. Tæppa (talk) 10:38, 12 February 2025 (UTC)[reply]

What is the difference with hotel chain (Q1631129)? Sjoerd de Bruin (talk) 10:53, 12 February 2025 (UTC)[reply]
Good question. I would have expected to find the distinction when looking at Marriott International (Q1141173) and it's various child organisations such as Marriott Hotels & Resorts (Q3918608) or Ritz-Carlton (Q782200) or Renaissance Hotels (Q2143252) but the use of P31 seems to be pretty haphazard. Tæppa (talk) 11:15, 12 February 2025 (UTC)[reply]
[edit]

A few of us are submitting a proposal for an ISWC 2025 Challenge related to Wikidata. An ISWC challenge involves several groups of people producing systems to solve a challenge problem. For more information on ISWC Challenges see the call for challenges at https://iswc2025.semanticweb.org/#/calls/challenges.

As opposed to most recent challenges, which have a particular task and fully quantitative evaluation, this challenge will be open to any system that identifies errors or missing information in Wikidata and, preferably, proposes updates to Wikidata that fix the errors or fill in the missing information. The system can be either fully automatic or semi-automatic. Submitted results be evaluated by an expect panel. Scoring will be influenced by size, plausibility, correctness, effect, and closeness to existing information. Systems that just add new individuals to Wikidata are not the target of the competition.

The page https://www.wikidata.org/wiki/User:Peter_F._Patel-Schneider/challenge will be used for information related to the challenge.

If you are interested in potentially submitting to a challenge of this sort please respond by email to pfpschneider@gmail.com and also put an entry in the Discussion page for that page. Expressions of interest received on or before 18 February 2025 will help improve the chances that the challenge will be part of ISWC 2025. Expressions of interest received after that date may help improve the chances that the challenge will be part of ISWC 2025.

Peter F. Patel-Schneider, Rosario Uceda-Sosa, Ege Atacan Doğan Peter F. Patel-Schneider (talk) 17:43, 12 February 2025 (UTC)[reply]

Wikipedia frontends?

[edit]

Should there be an item for a Wikipedia pendant to Wikidata front end (Q51139628), i.e. Wikipedia frontend? I think there are several frontends for Wikipedia, aren't there? I was looking for it when creating the item for the new site WikiTimeline (Q132230985) which uses Wikipedia to create a timeline (I suggested it could also use Wikidata and it has several issues like not showing many events in the article).

Another item that could have Wikipedia frontend set would be Wikiwand (Q18208301) or is mirrors data from (P3447)Wikipedia (Q52) actually correct and better? On a related note, shouldn't the Wikidata Query Timeline have an item too or be set somehow on the Histropedia item? Prototyperspective (talk) 02:43, 13 February 2025 (UTC)[reply]

Promoting a wikidata integration

[edit]

Alma (Q25498553) is a library software system mainly used by larger research libraries and consortia. British Library (Q23308) recently selected it as their new system. I recently proposed a new Alma feature, which basically boils down to an integration between their 'sets' functionality and wikidata. See https://ideas.exlibrisgroup.com/forums/308173-alma/suggestions/49470473-linked-data-powered-sets-and-collections and https://www.wikidata.org/wiki/User:Stuartyeates/PeopleForBookDisplays Note that much of this is built around Library of Congress authority ID (P244)/VIAF cluster ID (P214)/ORCID iD (P496) which are library-land identifiers for people. Hints, suggestions, corrections and work on the coverage of those properties greatly appreciated. Stuartyeates (talk) 23:00, 14 February 2025 (UTC)[reply]

Bunker website blocked by spam filter

[edit]

I was adding the URL for an entry for pictures.nl/ Bunkerpictures, a website of bunker pictures (lots of lovely concrete corridors and machinery rooms), and got the addition blocked with

The text you wanted to publish was blocked by the spam filter. This is probably caused by a link to a forbidden external site. The following text is what triggered our spam filter: bunker pictures.nl

I've not come across the spam filter before, and surprised its messages don't link to more information, Where is it documented, and why doesn't it like bunkers?

(I had to split the 2 words to even allow me to ask the question here) Vicarage (talk) 19:56, 15 February 2025 (UTC)[reply]

The website is globally blacklisted. A local admin on meta wiki can fix it if you report it there. It looks like a serious site so it's dumb to block it. I'm guessing some of the pictures might contain symbols from the nazi-era. Infrastruktur (talk) 21:02, 15 February 2025 (UTC)[reply]
Thanks. For the benefit of others, you can appeal these blocks at https://meta.wikimedia.org/wiki/Talk:Spam_blacklist, I wish they'd change their WD popups to make this clear, and that we'd add this to our help system. Vicarage (talk) 21:31, 15 February 2025 (UTC)[reply]

References for vernacular names of taxa

[edit]

I can add a taxon common name (P1843) to a taxon with a citation with either a reference to a listed website by using stated in (P248), or citing a URL with reference URL (P854). What I can't find, is any way of citing a published textbook. Is that possible, please? Would it be possible to have a listing "stated in ISBN x-xx-xxxx-xxxx-x" or similar? Thanks! - MPF (talk) 20:47, 15 February 2025 (UTC)[reply]

Seems to me like the best solution here is to create a new item for the textbook (if one does not exist already) and use the textbook's item as the value for stated in (P248) M2Ys4U (talk) 03:05, 16 February 2025 (UTC)[reply]
There is no string property on the model of author name string (P2093) or street address (P6375) that can be used for references. If an item doesn't exist, you have to manually "construct" the reference from the various properties used for describing published material: title (P1476), author (P50) etc. See the list of relevant properties. Circeus (talk) 06:03, 16 February 2025 (UTC)[reply]
ISBN and other properties can be used in a reference - Q17155#P825 and Q20063#P279. If there are likely to be several citations of a book, an item for the book would be useful then each reference would only need stated in (P248) and page(s) (P304). Peter James (talk) 12:46, 16 February 2025 (UTC)[reply]

P13304

[edit]

Hello everyone. This is my first time trying to create a property, but I don't quite understand why PWBA.com player ID (P13304) doesn't generate a direct URL link, even though formatter URL (P1630) is filled in correctly. Could you help me figure out the issue? Mitte27 (talk) 01:21, 17 February 2025 (UTC)[reply]

You need to wait for about a day, the servers only update these URLs about once a day. Midleading (talk) 04:10, 17 February 2025 (UTC)[reply]

Disagreement about proper pointing to Wikipedia articles

[edit]

See Q3432204 and Pitkävuori (Q132349139) as well as User_talk:Alabasterstein#Pitkävuori. Could we kindly ask for a third opinion. Thanks! --Rießler (talk) 14:40, 17 February 2025 (UTC)[reply]