User talk:ArthurPSmith

See User talk:ArthurPSmith/Archive for older discussions.

Jewish Museum Berlin object ID[edit]

Thanks for fixing my typo at Jewish Museum Berlin object ID and not excoriating me for it! --RAN (talk) 21:35, 29 January 2021 (UTC)Reply[reply]

I want my talk page to look like User talk:Belteshassar's talk page[edit]

Hi. I'm asking for your help(totally out of the blue) if my talk page can look like User talk:Belteshassar's talk page. LotsofTheories (talk) 21:03, 9 February 2021 (UTC)Reply[reply]

@LotsofTheories: I believe this is a user preference called "Structured Discussions" - see Special:Preferences under "Beta features", check the box "Structured Discussions on user talk". This page maybe helpful also. ArthurPSmith (talk) 22:03, 9 February 2021 (UTC)Reply[reply]

L14698 and L184508[edit]

Hi! I just came across the two lexemes bow (L14698) and bow (L184508) I asked myself why these two lexemes were not combined into one lexeme. Because you are the creator of both lexemes, what your intention is behind them? --Gymnicus (talk) 22:07, 20 February 2021 (UTC)Reply[reply]

@Gymnicus: Because they are two distinct lexemes, pronounced differently even though they are spelled the same and have the same forms. See the differing senses on the two lexemes. ArthurPSmith (talk) 13:20, 22 February 2021 (UTC)Reply[reply]

In my opinion, the different senses are no justification for separate lexemes. As a counter-example to this thesis, I would cite the lexeme Bienenstich (L10226) In the German language, this lexeme describes both the sting of a bee and a sheet cake. These two senses are also completely different and yet it is a lexeme. There must be other differences. For example, by the lexemes Bank (L34723) and Bank (L34791) the reason for the separation is the different grammatical forms. In this case of bow (L14698) and bow (L184508) the different pronunciation you mentioned could be the reason for the separation into two lexemes. But then, from my point of view, the different pronunciation should also be specified. There are various properties that can be used for this in Wikidata. You can see an overview in the template Lexicographical properties. Perhaps you can incorporate at least one of the three properties into each of the two lexemes so that one can see the difference between them. --Gymnicus (talk) 14:44, 22 February 2021 (UTC)Reply[reply]

@Gymnicus: I think any English speaker would understand the difference by looking at the senses, that was my implication. Yes hopefully pronunciation properties will be added, but I've added thousands of lexemes without those statements, and working on other things right now, so it's unlikely I will get to it. ArthurPSmith (talk) 16:05, 22 February 2021 (UTC)Reply[reply]

One reference on this. ArthurPSmith (talk) 16:07, 22 February 2021 (UTC)Reply[reply]

or more humorously... ArthurPSmith (talk) 16:11, 22 February 2021 (UTC)Reply[reply]

“I think any English speaker would understand the difference by looking at the senses” - This assumption has already been refuted by me, as you have already noticed. I speak English, but I didn't know the difference between bow (L14698) and bow (L184508). That is why it could have been that I thoughtlessly put the two lexemes together because I saw no difference between them, except for the different meanings. But as already shown on the lexeme Bienenstich (L10226), this is no reason not to merge the two lexemes. I think it's a shame that you don't add the pronunciation. Then one must hope that the two lexemes are not carelessly put together by someone. Unfortunately, I cannot add the pronunciation either, because I have no idea about it. --Gymnicus (talk) 21:29, 22 February 2021 (UTC)Reply[reply]

Strange merge[edit]

Hi Arthur, you've merged de:Universidad Santa María (Ecuador) (founded 1996) with en:Federico Santa María Technical University (founded 1931). [1] --Kolja21 (talk) 16:23, 26 February 2021 (UTC)Reply[reply]

@Kolja21: It is the same institution - however the inception date from GRID would appear to be incorrect (based on the dewiki text). It seems to be a branch of the Chilean university, Federico Santa María Technical University (Q457793); I'll link them together. ArthurPSmith (talk) 16:31, 26 February 2021 (UTC)Reply[reply]

Oh, wait, now I don't understand your comment. I didn't merge the de entry with the en entry, the de entry is for the university in Ecuador, the en one is from Chile. ArthurPSmith (talk) 16:35, 26 February 2021 (UTC)Reply[reply]

Apps[edit]

Should Android and iOS versions of apps have seperate items? I cannot for the life of me find anything regarding the modeling of apps --Trade (talk) 00:04, 1 March 2021 (UTC)Reply[reply]

@Trade: No, I think a single item is generally the right thing for a single piece of software - see Wikidata:WikiProject Informatics/Software/Properties and related pages. ArthurPSmith (talk) 18:53, 1 March 2021 (UTC)Reply[reply]

@Trade: @ArthurPSmith: You should create separate item for iOS and Android otherwise it will be messed. Eurohunter (talk) 20:44, 15 March 2021 (UTC)Reply[reply]

Messaggio[edit]

Ho ricevuto un messaggio tuo ma non ho capito se c'è un testo.... Ho fatto qualcosa?... – The preceding unsigned comment was added by Gabriele.badii (talk • contribs) at 18:41, April 16, 2021‎ (UTC).

@Gabriele.badii: It's just a standard welcome template - I try to add it when I see somebody commenting on here who seems new. Nothing you did was a problem at all! ArthurPSmith (talk) 12:27, 19 April 2021 (UTC)Reply[reply]

Thank you Gabriele.badii (talk) 12:34, 19 April 2021 (UTC)Reply[reply]

Call for participation in the interview study with Wikidata editors[edit]

Dear ArthurPSmith,

I hope you are doing good,

I am Kholoud, a researcher at the King’s College London, and I work on a project as part of my PhD research that develops a personalized recommendation system to suggest Wikidata items for the editors based on their interests and preferences. I am collaborating on this project with Elena Simperl and Miaojing Shi.

I would love to talk with you to know about your current ways to choose the items you work on in Wikidata and understand the factors that might influence such a decision. Your cooperation will give us valuable insights into building a recommender system that can help improve your editing experience.

Participation is completely voluntary. You have the option to withdraw at any time. Your data will be processed under the terms of UK data protection law (including the UK General Data Protection Regulation (UK GDPR) and the Data Protection Act 2018). The information and data that you provide will remain confidential; it will only be stored on the password-protected computer of the researchers. We will use the results anonymized (?) to provide insights into the practices of the editors in item selection processes for editing and publish the results of the study to a research venue. If you decide to take part, we will ask you to sign a consent form, and you will be given a copy of this consent form to keep.

If you’re interested in participating and have 15-20 minutes to chat (I promise to keep the time!), please either contact me on kholoudsaa@gmail.com or use this form https://docs.google.com/forms/d/e/1FAIpQLSdmmFHaiB20nK14wrQJgfrA18PtmdagyeRib3xGtvzkdn3Lgw/viewform?usp=sf_link with your choice of the times that work for you.

I’ll follow up with you to figure out what method is the best way for us to connect.

Please contact me using the email mentioned above if you have any questions or require more information about this project.

Thank you for considering taking part in this research.

Regards

Kholoud

judge (L4623)[edit]

Apologies for the incorrect edit. The context is linking the particular form in Sphinx of black quartz, judge my vow! (L501732). I meant imperative (Q22716) but was confused; 'case' versus 'mood' should have tipped me off. Sorry again and thanks for catching it. Arlo Barnes (talk) 18:12, 11 June 2021 (UTC)Reply[reply]

No problem - English is in some ways a much simpler language than some others... ArthurPSmith (talk) 18:21, 11 June 2021 (UTC)Reply[reply]

Property proposal: Tax Identification Number (Belarus)[edit]

Hi Arthur,

Could you take a look at Wikidata:Property proposal/Tax identification number when you have a chance? Aestrum (talk) 17:23, 3 August 2021 (UTC)Reply[reply]

Thank you! Aestrum (talk) 08:47, 4 August 2021 (UTC)Reply[reply]

Reverting edits by Kinvidia[edit]

Is there a chance you could revert all changes to existing items yesterday by Kinvidia (talk • contribs • logs)? See my comments on his user page. --Hjart (talk) 06:46, 27 August 2021 (UTC)Reply[reply]

@Hjart: (1) I'm not an administrator so I don't think I can do that easily, and (2) there are thousands of changes from that user dated August 26 (almost all Quickstatements), many of them creating new items, and the ones I checked look ok to me. However, the user is not responding to comments on their talk page so maybe something should be brought up on the Administrators Noticeboard about this? ArthurPSmith (talk) 12:32, 27 August 2021 (UTC)Reply[reply]
I posted a note on the noticeboard. We'll see what comes of it... ArthurPSmith (talk) 12:41, 27 August 2021 (UTC)Reply[reply]

music created for[edit]

Hi, thanks for marking "created for" ready for creation. Any chance that the related proposal music created for could also be ready? There's only been one objection and many supports, and I had hoped I had answered the objections. UWashPrincipalCataloger (talk) 18:59, 1 September 2021 (UTC)Reply[reply]

Author Disambiguator stalled?[edit]

Hi Arthur, I'm not being impatient, but I noticed that the last few batches haven't been executing. There even seem to be some that are still at READY, but are no longer queued [2]. Maybe you could take a look? Thanks! --Azertus (talk) 17:38, 9 September 2021 (UTC)Reply[reply]

@Azertus: Thanks for letting me know! The background job runner had definitely crashed. I restarted the server - however it won't automatically restart those queued batches, you need to manually restart at least one of them (or submit a new batch) and then I think it will run the others; it doesn't save OAuth credentials in a way that it can use them after restart. ArthurPSmith (talk) 20:35, 9 September 2021 (UTC)Reply[reply]
@Alatalojm, GerardM, BugWarp, Daniel Mietchen, Infovarius: See above - sorry for the problem, but you need to check your batches on Author Disambiguator and restart at least one of them. Thanks! ArthurPSmith (talk) 20:46, 9 September 2021 (UTC)Reply[reply]

Wikidata:WikiProject Energy - Are power outages relevant?[edit]

Do you know if it would be relevant to add power outages to Wikidata? I know of a power outage where around 10 thousand people had a power outage for about an hour. Would this be relevant to Wikidata or should I consider making my own Wikibase instance where I can add such information? Only asking for advice since I find Wikidata scary.(the politics of Wikidata, not the database software itself) Oduci (talk) 14:03, 15 September 2021 (UTC)Reply[reply]

@Oduci: It may depend on the size of the dataset you are planning, and also whether the outages are notable in the sense that they are documented by some third party source (a newspaper article for instance). We already have things like 2009 Brazil and Paraguay blackout (Q32375), 2006 European blackout (Q1324840), and 2006 Queens blackout (Q4606840) - along with many more instances of power outage (Q828827). If you plan to add at most a few thousand such things here it's probably fine. If you have millions of them, then a separate wikibase would likely be a better choice, just for capacity reasons. ArthurPSmith (talk) 20:15, 15 September 2021 (UTC)Reply[reply]

dissertation program ready but not appearing on the list of proposals ready for creation[edit]

Hi, thanks for marking the proposal for dissertation program ready for creation. But do you know why it is not appearing on the list of proposals that are ready for creation? UWashPrincipalCataloger (talk) 16:55, 16 September 2021 (UTC)Reply[reply]

@UWashPrincipalCataloger: It seems to be there now - maybe there's some delay? Not sure how that works. ArthurPSmith (talk) 17:57, 16 September 2021 (UTC)Reply[reply]

Or maybe I am just going blind! Anyway, thanks again! UWashPrincipalCataloger (talk) 19:27, 16 September 2021 (UTC)Reply[reply]

Wikidata:Property proposal/Smotrim.ru film ID, Wikidata:Property proposal/Channel One Russia show ID[edit]

Good evening. Could you summarize my suggested properties, which have been ready for a few weeks now? MasterRus21thCentury (talk) 20:54, 11 December 2021 (UTC)Reply[reply]

Thank you for your recent merges[edit]

I guess I created some items based on redirecting DOIs. Thanks for merging them. I was aware that was happening, but I thought I repaired them. I'm not surprised I missed some, but I regret missing so many. Trilotat (talk) 01:45, 14 December 2021 (UTC)Reply[reply]

@Trilotat: I don't think you were the worst offender! I've been going through the list at User:Ivan A. Krestinin/To merge/Scholarly articles and done what I can so far; waiting on the next update to that list to see if there's a lot still left. Unfortunately there are also many non-duplicates in that list due to some journals listing several distinct articles under the same DOI, so just merging based on DOI isn't sufficiently reliable. ArthurPSmith (talk) 16:55, 14 December 2021 (UTC)Reply[reply]

Wikidata:Property proposal/original catalog description[edit]

Hi Arthur,

Any chance you could take a look again at Wikidata:Property proposal/original catalog description? Other than Jura, there seems to be mostly support, and no additional comments in almost a month now. It would be great to be able to add these statements soon. Thanks! Dominic (talk) 15:39, 25 January 2022 (UTC)Reply[reply]

@Dominic: I am really not familiar with Commons or SDC at all; I'd prefer that a property creator more familiar with the situation do this. Check out the list on Wikidata:Property creators and see if you recognize anybody who might be more involved with Commons. ArthurPSmith (talk) 16:14, 25 January 2022 (UTC)Reply[reply]

Sure, no worries! I just saw you'd commented on it earlier. Thanks! Dominic (talk) 16:16, 25 January 2022 (UTC)Reply[reply]

pole positions (P10640)[edit]

Hello. Would you help with this property? MasterRus21thCentury (talk) 15:43, 16 April 2022 (UTC)Reply[reply]

@MasterRus21thCentury: It looks ok to me? I guess somebody else helped? ArthurPSmith (talk) 13:32, 18 April 2022 (UTC)Reply[reply]

Querying cross-wiki links via wikidata[edit]

Hello Arthur, I hope this is an appropriate and non-burdensome question. I saw you named on the [Wikiproject] and that you have developed many wikidata tools. In the way of due diligence, I have browsed resources like [[3]] and [[4]] (and I am a comfortable programmer and API-figure-outer) but I cannot find an answer to my question.

In service of an academic research project, I must compile all Wikipedia namepsace articles in all language editions that have 10 or more cross-wiki links. From each I will need it's language, title, link, article ID, and creation date. I am not asking you to build any queries for me, but I would be very appreciate if you could point me to the tool, link, or even jargon that I should be using to collect this dataset. Enfascination (talk) 00:54, 26 June 2022 (UTC)Reply[reply]

@Enfascination: Wikipedia articles use wikidata for cross-wiki links, but they may also use internal link text which would not be known to wikidata. But assuming you are ok with just the links provided by wikidata, then all you should need to do for this is to find all wikidata items that have 10 or more sitelinks. If you go to the query service (https://query.wikidata.org/) and click on "Examples" you can find several examples that count and rank items by number of sitelinks, for example try the "Authors, writers and poets ranked by sitelink" query. Let me know if you need further help on this. ArthurPSmith (talk) 13:04, 27 June 2022 (UTC)Reply[reply]
@ArthurPSmith this is tremendously helpful, thank you! That's a surprise that the sitelink system is still hybrid, with cross-wiki links stored either in wikidata or wikitext. I'm OK missing some links as long as I feel like I'm getting perhaps 90%, or all but the most recent. Where could I learn more about how these two systems coexist? If I can root around more I'll know what mental model to rely on (whether they are coequal representations, or one is trying (however imperfectly) to be the comprehensive store, etc). Knowing absolutely nothing, I guess my prior mental model would be that editors tend to add sitelinks as wikitext and those get detected, transferred to wikidata, and then removed by bots. Do you know who could tell me how far that is from the truth? Enfascination (talk) 20:58, 27 June 2022 (UTC)Reply[reply]
So the wikitext versions predate Wikidata, and when Wikidata got started (around 2013) there was an effort to import the sitelinks from wikitext, replacing it, so the vast majority of interlinking at that time was from Wikidata, not from the text. Since then the natural way to add links to other language versions goes through Wikidata, so I think it is still that way. But the wikitext option remains as a workaround for cases that don't work for some reason (for example only one sitelink is allowed per language in Wikidata, but some languages have multiple articles on almost the same topic, so only one of them can be linked via Wikidata and others would need direct links). Help pages like Help:Sitelinks may give you a bit more background on this, though I don't know if there's a general place it's all recorded. Wikidata:Project Chat is also of course a good place to ask general questions. ArthurPSmith (talk) 21:20, 27 June 2022 (UTC)Reply[reply]
Ah, I'll use the project chat for further questions. Thank you again! Enfascination (talk) 02:53, 29 June 2022 (UTC)Reply[reply]

Survey of Scottish Witchcraft[edit]

Hi Arthur, sorry for message out of the blue but I recall seeing your talk on using OpenRefine at WikidataCon and wondered whether you, or someone you knew, would be able to advise/train my student intern Maggie Lin to work with OpenRefine in terms of aiding her work taking very rich historical data from the Ms Access database for the Survey of Scottish Witchcraft and modelling it correctly/reconciling it to wikidata. There is a great deal we can do to improve/enhance the data already added and a great many insights and visualisations we can glean once the data is linked data format. Really not intending to ask for too much of your time or effort beyond making sure Maggie is cognisant of how to work with OpenRefine to get the best out of it. Any thoughts do let me know. Best, Stinglehammer (talk) 21:06, 27 June 2022 (UTC)Reply[reply]

Hi Stinglehammer, glad to hear from you, and I've enjoyed hearing about your project. I do still use OpenRefine regularly but I think I'm kind of behind-the-times on the best ways to use it with Wikidata; it's evolved a lot since I first got familiar with it and I'm a bit stuck on my initial use cases. If Maggie hasn't already done this, I do strongly recommend the tutorial videos at https://openrefine.org - also the documentation there generally is quite good. But there are parts I'm not very familiar with - particularly creating new Wikidata items, which is covered here for example: https://docs.openrefine.org/manual/wikibase/new-entities . It might be best to reach out the openrefine community on their mailing list to find somebody more up-to-date on how to use it? ArthurPSmith (talk) 13:41, 28 June 2022 (UTC)Reply[reply]

Merging of unordered author and ordered author[edit]

You'd better make new tools for merging ordered author and unordered author in Author Disambiguator. For Jelle Kaastra (Q58250692), we should merge more than 100 articles like Anatomy of the AGN in NGC 5548 (Q59973269). Sharouser (talk) 15:56, 7 July 2022 (UTC)Reply[reply]

@Sharouser: Hi - there's already some support for this - if you go to the work page and click "Auto-match unordered authors?" at the bottom it will suggest a number to set for the unordered author. But then there's another step to actually merge the two entries. If you have a suggestion for a better workflow for this please let me know! ArthurPSmith (talk) 16:03, 7 July 2022 (UTC)Reply[reply]

@ArthurPSmith: I want to merge all unordered authorlink in author (P50) and all ordered authorlink in author (P50) at one time. This way is more convenient and fast. You'd better upgrade author_item_oauth. Sharouser (talk) 16:15, 7 July 2022 (UTC)Reply[reply]

Ah, I see what you're saying. I'll check into it. ArthurPSmith (talk) 16:29, 7 July 2022 (UTC)Reply[reply]

OrcBot has a lot of edits adding authors without removing the author name string. If you find an article where this has happened, you’re very likely able to find that author appears in other articles with their presence as an author AND an author name string. I’m on board to help with a more effective cleanup of this bot’s good faith but certainly incomplete work. Trilotat (talk) 14:37, 8 July 2022 (UTC)Reply[reply]

Wikidata:Property proposal/title match pattern[edit]

Hey, Wikidata:Property proposal/title match pattern has been marked ready for two weeks ago. Is there something I can do to get this created? Do you have any more questions? – Shisma (talk) 06:30, 28 July 2022 (UTC)Reply[reply]

@Shisma: Could you respond to the comment from Dhx1 there about the label? That may seem an unresolved issue for property creators looking at this. ArthurPSmith (talk) 13:44, 28 July 2022 (UTC)Reply[reply]

i agreed with Dhx1 from the start. I just through the label can be changed at any time 😂. – Shisma (talk) 15:24, 28 July 2022 (UTC)Reply[reply]

another month has passed. Can I do something? Shisma (talk) 16:18, 26 August 2022 (UTC)Reply[reply]

@Shisma: I created it but it's bare-bones - can you add the examples and constraints? ArthurPSmith (talk) 16:57, 26 August 2022 (UTC)Reply[reply]

Sure — Shisma (talk) 19:14, 26 August 2022 (UTC)Reply[reply]

Could you please review my changes at website title extract pattern (P10999). Thanks – Shisma (talk) 19:56, 26 August 2022 (UTC)Reply[reply]

Looks good. I fixed a couple of minor issues. I don't think the formatter URL will actually do anything (at least not with the regular Wikidata UI - it's applied only on external id's, not strings). Otherwise it seems fine. ArthurPSmith (talk) 20:05, 26 August 2022 (UTC)Reply[reply]

off talkpage logging?[edit]

As you know, I am working on the isotopes. To be open to those interested, I am logging at talk:WP Chemistry#Charting the isotopes. But this could be too tiresome, and ineffective, for those following that Project page. Is there an option to do the logging elsewhere, while still posting the main issues (eg modeling proposals) at WDtalk:WP Chemistry? Could be a dedicated projecttalk subpage, or a page in my userspace. Ideas? -DePiep (talk) 08:17, 2 August 2022 (UTC)Reply[reply]

@DePiep: Either a dedicated project subpage or a userspace page would be fine - userspace probably better if you expect to be the only one editing it, project subpage if you're inviting or encouraging others to work on it too. ArthurPSmith (talk) 13:20, 2 August 2022 (UTC)Reply[reply]

GRID ID[edit]

My mistake Palapa (talk) 21:01, 2 August 2022 (UTC)Reply[reply]

Wikidata:Property proposal/Sports[edit]

Hello! This Wikidata property proposal page is currently in a difficult situation - the pages are no longer displayed, and there has been no consensus on 11 properties for 1.5 years. Could you solve this problem? MasterRus21thCentury (talk) 13:59, 4 August 2022 (UTC)Reply[reply]

@MasterRus21thCentury: I am not personally knowledgeable about sports at all so I haven't worked on those and don't really think I should. Also I'm not sure what you mean by "the pages are no longer displayed" - when I follow the link to the Sports property proposal page, all the proposals seem to be listed? ArthurPSmith (talk) 18:18, 4 August 2022 (UTC)Reply[reply]

I have a specific list about this and something needs to be done with it.

Need to create:

Demand immediate summing up due to lack of consensus since January 2021:

Need approval:

This is because I've increased activity on this thread lately and Wikidata members aren't upvoting like they've been in the past months. However, the queue may continue to grow, which will be very unpleasant for the community. MasterRus21thCentury (talk) 18:40, 4 August 2022 (UTC)Reply[reply]

@MasterRus21thCentury: I see you've also commented on Project Chat - that would be the best place to recruit some people to help with this. But you can go ahead and retire the "no consensus" ones yourself; see my comments on Project Chat for how to do that. ArthurPSmith (talk) 19:06, 4 August 2022 (UTC)Reply[reply]

@ArthurPSmith: So I can close threads that don't have consensus without having the property creator flag? MasterRus21thCentury (talk) 18:41, 5 August 2022 (UTC)Reply[reply]

Don't you have that flag? Technically it's certainly not required to have the flag to close proposals, but I guess it's better for people with the flag to do it to avoid controversy here. ArthurPSmith (talk) 19:51, 5 August 2022 (UTC)Reply[reply]

@ArthurPSmith Yes, I was stripped of my flag three months ago as a result of a discussion on the admin forum. I was advised to reapply in six months, that is, in November. However, we need to strengthen action on Wikidata properties. MasterRus21thCentury (talk) 08:07, 7 August 2022 (UTC)Reply[reply]

Properties ready[edit]

Hello! Would you be able to queue these properties for creation?

MasterRus21thCentury (talk) 10:09, 22 August 2022 (UTC)Reply[reply]

@MasterRus21thCentury: I took a look and ok'd some. However, as I've mentioned before, I have no expertise regarding sports, which is what most of these are, and (1) I look for at least one other supporting person (aside from me and proposer) when setting to "ready" status, and many of these seem to have no other supporters at all, and (2) there have been some complaints about non-English descriptions - I don't personally have a strong feeling on that and I guess I could add an English translation of your Russian descriptions if I had the time, but I don't right now; if you can add English descriptions where missing then setting them to "ready" would be straightforward after that. ArthurPSmith (talk) 14:07, 22 August 2022 (UTC)Reply[reply]
@ArthurPSmith: Currently, I have put English descriptions everywhere. Therefore, you can continue your work on setting properties for creation. MasterRus21thCentury (talk) 17:17, 22 August 2022 (UTC)Reply[reply]

Cancel a batch in authordisambiguator[edit]

Hi,

This batch is full of mistakes : https://author-disambiguator.toolforge.org/batches_oauth.php?id=f802b40f

Do it is possible to revert an entire batch ? Simon Villeneuve (talk) 23:38, 31 August 2022 (UTC)Reply[reply]

@Simon Villeneuve: Hi - I happened to catch it as it was happening, and I believe I've fixed all the problems (I've also been working on ATLAS author lists). Maybe you can check to confirm it looks ok now? In general though yes it is possible to revert a batch by going to the "details" link in the edit summary for one of the edits - that takes you to the "edit groups" app, which then has an option to "Undo entire group". ArthurPSmith (talk) 15:16, 1 September 2022 (UTC)Reply[reply]

Ok, thank you. It seems ok now.
How the hell did you do that ? Simon Villeneuve (talk) 15:44, 1 September 2022 (UTC)Reply[reply]

@Simon Villeneuve: Good question! I've been playing with the "lists" feature - I think you should be able to see https://author-disambiguator.toolforge.org/author_lists.php?list_id=55 ? Then for an individual work I can try to match the authors with that list. However, it's a bit time-consuming as there are lots of cases of duplicate or ambiguous name strings (for ATLAS in 2017 I have 41 duplicates to be checked - still 41 out of 2900 or so is pretty good). So it's coming from the angle of replacing lots of author name strings on a single manuscript, rather than doing it on lots of manuscripts for a single author. I think this way will end up being somewhat faster in the end, but it's still not super fast. ArthurPSmith (talk) 15:59, 1 September 2022 (UTC)Reply[reply]

Another big advantage of this is to have considerably less historic versions for each article element.
I'll let you work on ATLAS authors and focus on CMS. Let me know if you want to work on these too. Simon Villeneuve (talk) 17:47, 1 September 2022 (UTC)Reply[reply]

I haven't attempted CMS yet. I don't know if it'll be easier or harder - CMS groups the author list by affiliation, so hopefully it's clearer when you have duplicate name strings... Anyway, no plans to do CMS for a long time yet! Thanks for all you've done already, it's been a big help for ATLAS! ArthurPSmith (talk) 17:51, 1 September 2022 (UTC)Reply[reply]

The English noun "in"[edit]

This question probably belongs in the Lexicographic project, but as I'm still a bit uncertain what I'm looking at, and I find your username in the edit histories, I'd like to begin by asking you what in (L322367) is all about. The English word "in" as a noun? How come? It was created in October of 2020, but you created the preposition in (L2987) yourself already in 2018.

I have been looking at the adpositions and tried to figure out how to define their senses; please take a look at User:SM5POR/Languages for a view of my approach to the problem. The other day I stumbled upon Q90219924, the practical purpose of which I don't understand, as it appears to represent the spatial relation of "in/within" and "into" (it's mapped to both senses) for English as well as Russian.

According to the Concise Oxford Dictionary mentioned in my references, the word "in" exists as a preposition with around 23 different senses, as an adverb with 14, and as an adjective with three. But a noun? As English is not my native language, I hope you can explain this to me. SM5POR (talk) 08:33, 26 September 2022 (UTC)Reply[reply]

@SM5POR: Huh, an effect of several people using bots or scripts without looking carefully at what they are doing. Nevertheless, the reference to "WordData" allowed me to trace the source, which is the use of "in" as an abbreviation for the length unit "inch". It does appear that way (singular and plural) in written technical language, so I guess it's legitimate. But maybe there's a better way to represent this. I replaced the incorrect sense with a corrected one. ArthurPSmith (talk) 15:45, 26 September 2022 (UTC)Reply[reply]

Thanks, that cleared things up for me. To further clarify the sense, I added an item for this sense (P5137) link to inch (Q218593), just like the one for inch (L10963). Your solution looks fine to me.

What may be missing is a direct link between the two lexemes, indicating that one is an abbreviation of the other. For Main namespace items there is the property unit symbol (P5061) which spells out an abbreviation of the subject unit, but as its data type is a monolingual string, it can't be reused for linking to a lexeme sense. Maybe a sense-to-sense property could be defined to serve this purpose?

I actually dislike the proliferation of language-specific properties in the Main namespace, as it makes the item display difficult to read through massive use of the language of work or name (P407) qualifier, and I would like to see all those claims transferred to the lexeme namespace instead.

I don't use robots myself, precisely because I don't want to risk messing up thousands of items without seeing how they are actually used. I even think there is something fundamentally wrong with your database design approach if it's versatile enough to allow representing any kind of human knowledge, yet uniform enough that you can make robots perform millions of edits automatically. SM5POR (talk) 07:16, 27 September 2022 (UTC)Reply[reply]

@SM5POR: Thanks for the edits and ideas. Yes I don't use bots very much and largely share your philosophy there (most of my lexeme edits have been one by one). However, I do see their usefulness when importing or correlating info in Wikidata with external databases, I don't think that's a database design issue in that case. ArthurPSmith (talk) 12:52, 27 September 2022 (UTC)Reply[reply]

I must admit I have a tendency to generalize just to make a point, so please let me qualify my statement about robots a bit. Robots are ok to do really trivial things, such as searching the database for instances of a particular error, or even fixing some of them in items when all their properties are defined according to the expectations of the robot maintainer. My issue is with an overall design that in effect depends on systematic use of robots for database development and regular maintenance.

You may remember the discussion on time zone representation two years ago, in which both of us took part? I wasn't convinced that adding individual located in time zone (P421) claims for millions of localities when there are only some 30 time zones in the world, most of them geographically contiguous, is a sensible solution even in the short-term perspective.

As I understood it, current infobox implementations are ill-prepared to apply the heuristics to make use of property inheritance and transitivity. But they shouldn't! They should rely on standardized property retrieval library routines, rather than encode the rules of transitivity in each and every infobox. This is the design approach I'm concerned about.

The aforementioned discussion therefore inspired me to start coding in Lua, to see if I could somehow help improving on the situation. I haven't written much code yet, not even a working prototype, but you may want to take a look at my initial attempts at documentation to see what I'm aiming at.

Have I overlooked something, do you think (besides embarking on a project that will probably not be completed in my lifetime)? SM5POR (talk) 07:36, 28 September 2022 (UTC)Reply[reply]

Good luck! I haven't ventured into Lua programming myself but I can see the need. Maybe wikifunctions will be a new way to do this (better)? ArthurPSmith (talk) 21:13, 28 September 2022 (UTC)Reply[reply]

Kvant-Telecom[edit]

Dear Arthur,

Quanttelecom LLC (Q115769956) is a very generic business name. https://kvant-telecom.ru/ seems to be much larger (AS43727) and bears the same name.

Kind regards Perkerdansk (talk) 11:58, 20 December 2022 (UTC)Reply[reply]

@Perkerdansk: Hmm, I can see the confusion. I updated the main label with the one from their English web page here; this does seem to be the name of this company. We don't seem to have an item for https://kvant-telecom.ru/ so it probably should be added also. ArthurPSmith (talk) 13:48, 20 December 2022 (UTC)Reply[reply]

Wikidata external URL redirector - support for Statistics Canada Geographic code (P3012)[edit]

Is it possible that you could add P3012 support in your URL redirector? I am trying to expand its functionality but am running into limitations in making this work. I have detailed how I would like to see linking from P3012 values implemented here. -- Denelson83 (talk) 00:28, 6 January 2023 (UTC)Reply[reply]

@Denelson83: I could set up something based on the id length, but I don't see how to include language code as that's not a property of the ID but of the user. I'm not sure what you have in mind here. ArthurPSmith (talk) 15:00, 6 January 2023 (UTC)Reply[reply]

At this point, we will have to just default to the English version. And the ID length should suffice as a differentiator in this case. In this table, the ID is bolded:

Geography type	ID length	Example redirector URL	Resulting destination URL
province or territory of Canada (Q2879)	2 digits	https://wikidata-externalid-url.toolforge.org/?p=3012&id=35	https://www12.statcan.gc.ca/census-recensement/2021/dp-pd/prof/details/page.cfm?Lang=E&DGUIDlist=2021A000235
census division of Canada (Q18810091)	4 digits	https://wikidata-externalid-url.toolforge.org/?p=3012&id=3506	https://www12.statcan.gc.ca/census-recensement/2021/dp-pd/prof/details/page.cfm?Lang=E&DGUIDlist=2021A00033506
federal electoral district of Canada (Q17202187)	5 digits	https://wikidata-externalid-url.toolforge.org/?p=3012&id=35075	https://www12.statcan.gc.ca/census-recensement/2021/dp-pd/prof/details/page.cfm?Lang=E&DGUIDlist=2013A000435075
designated place of Canada (Q956318)	6 digits	https://wikidata-externalid-url.toolforge.org/?p=3012&id=350066	https://www12.statcan.gc.ca/census-recensement/2021/dp-pd/prof/details/page.cfm?Lang=E&DGUIDlist=2021A0006350066
census subdivision (Q17457753)	7 digits	https://wikidata-externalid-url.toolforge.org/?p=3012&id=3506008	https://www12.statcan.gc.ca/census-recensement/2021/dp-pd/prof/details/page.cfm?Lang=E&DGUIDlist=2021A00053506008

-- Denelson83 (talk) 20:18, 6 January 2023 (UTC)Reply[reply]

@Denelson83: Ok, this is implemented and I updated the formatter URL on P3012 to use this. ArthurPSmith (talk) 17:18, 9 January 2023 (UTC)Reply[reply]

Good stuff. Thank you. -- Denelson83 (talk) 17:22, 9 January 2023 (UTC)Reply[reply]

Maud Olofsson "bad merge"[edit]

Hello, you reverted my merge of Maud Olofsson (Q253640) and Maud Olofsson (Q100348520). I spent quite some time trying to find any indications that A roadmap for the implementation of mHealth innovations for image-based diagnostic support in clinical and public-health settings: a focus on front-line health workers and health-system organizations. (Q38368576) was written by any other than the former Swedish deputy prime minister, Maud Olofsson (Q253640). The article is the outcome of a round table discussion that took place during an event organized at STIAS with people from Sweden and South Africa. STIAS have on their own websites documented the former politician's participation workshops and roundtables during the 2010s. The booklet for the roundtable discussion that is described in the article, lists Maud Olofsson as "Wallenberg Foundation Ambassador" with an e-mail address belonging to a company (Romo Norr) where the former politician is a board member. An article published in one of Swedens largest newspapers three days before the event took place mentions both that the company in question is the former politician's own and that she and her husband will land in South Africa the same day to participate in a round table discussion in Stellenbosch. I find it highly unlikely that a namesake of the politician would be listed as an author of an article that is the outcome of the very same roundtable the politician were part of.

So I would say there is enough sources to disprove that my merge was bad in any way. (One might also argue that giving stating occupation (P106) as someone in their role as researcher (Q1650915) might be a bit of a stretch as well.)

[5]Former Swedish Deputy Prime Minister Maud Olofsson visits STIAS
[6]Investment and technology choices in the SA power sector: economic, social and environmental trade-offs

Moonhouse (talk) 17:57, 23 January 2023 (UTC)Reply[reply]

@Moonhouse: Ok, thanks for looking into this - but the paper does list her affiliation as Stellenbosch (I added that as a reference after unmerging). If you can add your sources appropriately on the item or perhaps on the talk page for Maud Olofsson (Q253640) that would probably be helpful. But if you are sure yes go ahead and re-merge them then. ArthurPSmith (talk) 21:26, 23 January 2023 (UTC)Reply[reply]

I don't see how the given affiliation directly contradicts that the two objects should be merged since the identity and the affiliation are two different questions. Whether or not she was affiliated directly with Stellenbosch (as per the article) or in the capacity of an ambassador to/representative of Marianne and Marcus Wallenberg Foundation (as the conference booklet as well as presentation of core organising committee state) I would say it is safe to assume that the two objects describe the same physical person. The aforementioned booklet and presentation of the core organising committee all include photos of Maud Olofsson (Q253640) and all the other authors of the article.

When it comes to the affiliation I can't say it is as clear-cut. She is either described as affiliated with both Stellenbosch and Wallenberg Foundation, as the article indicates, or just tasked by the latter to assist the former (as described in e.g. "The STIAS concept of a Roundtable").

So I would suggest that I merge the two objects and then add affiliation (P1416) (since I can find no proof in the article or elsewhere that she was in fact employed by the university) set to Stellenbosch Institute for Advanced Study (Q55829926) (with, at least, the article as a reference) and also add Marianne and Marcus Wallenberg Foundation (Q10577257) as an affiliation (supported by other documents as reference). -- Moonhouse (talk) 23:16, 23 January 2023 (UTC)Reply[reply]

@Moonhouse: Ok, yes, that sounds great, please go ahead with that then. Thanks. ArthurPSmith (talk) 14:14, 24 January 2023 (UTC)Reply[reply]

Property proposal[edit]

Hi there,

could'nt we restart Wikidata:Property proposal/Geneanet genealogist ID? Taking care of adding more examples, and pinging related projects. Thanks, 2A01:CB1D:8CC3:6500:2D8E:560:2681:4BAD 06:47, 2 June 2023 (UTC)Reply[reply]

Best to create a new proposal from scratch rather than reopen a very old one. That way it shows up at the top of lists for people who track those things. ArthurPSmith (talk) 19:05, 2 June 2023 (UTC)Reply[reply]

P11439[edit]

Thank you for your (encouraging) comment! But what am I to do exactly? May I change the property formatter URL by myself...?

Best regards, 2A01:CB1D:8CC3:6500:2D8E:560:2681:4BAD 19:56, 2 June 2023 (UTC)Reply[reply]

Just make an exact copy of the old proposal template, add a "2" to the page title, and make a note of whatever other things you have changed such as more examples. ArthurPSmith (talk) 20:34, 2 June 2023 (UTC)Reply[reply]

But, is it really necessary to make a new proposait? It would be the same project, with just a more accurate formatter URL. (And moreover, everyone already seems to agree...) 92.184.112.49 05:12, 3 June 2023 (UTC)Reply[reply]

? 92.184.112.216 10:53, 5 June 2023 (UTC)Reply[reply]

It's not necessary to create a new one, but it helps make it more noticeable to people who check property proposal lists. Up to you what to do. ArthurPSmith (talk) 13:25, 5 June 2023 (UTC)Reply[reply]

Sorry for my confusion, I had somehow thought you were the same anonymous user as the one just previous and had merged the two conversations. Yes of course you can just change the formatter URL. ArthurPSmith (talk) 14:14, 6 June 2023 (UTC)Reply[reply]