Wikidata weekly summary #659

edit

Weekly Summary #660

edit

Wikidata weekly summary #661

edit

Wikidata weekly summary #662

edit

Confusing edits

edit

What is the source for these claims? —Justin (koavf)TCM 17:49, 15 January 2025 (UTC)Reply

Okay, this explains it. Thanks. —Justin (koavf)TCM 17:50, 15 January 2025 (UTC)Reply
An occupational surname (Q829026) is supposedly derived from (named after (P138)) a job title, so where there is an existing claim [name]instance of (P31)occupational surname (Q829026)of (DEPRECATED) (P642)[occupation], I have assumed named after (P138) is the appropriate replacement for of (DEPRECATED) (P642). And as this qualifier applies to the subject item rather than just the P31 statement, I have added it as a main statement. My script copies references from the original statement to this new main statement, but in this case there were no references to copy. Swpb (talk) 18:00, 15 January 2025 (UTC)Reply

Wikidata weekly summary #663

edit

Your edits of Bronisław Kononowicz

edit

Hello, friend.

About half a year ago, you made some edits to the entry about Bronisław Kononowicz. I think I understand the reasoning behind your removal of the "invasion of Poland" entry he participated in - which is obviously your attempt at expressing respect to the subject of the article, who obviously defended Poland during said invasion rather than supported the perpetrators. I understand your edit was made in good faith and that you wanted to avoid this unfortunate implication that Kononowicz was an invader.

The thing is, this event is known by different names in different languages. "Invasion of Poland" is the most popular international name and, unfortunately (from this entry's point of view), is invader-centric - I believe you noticed this unfortunate and misleading implication that Kononowicz was an active party during this event, rather than reactive, and wanted to correct this by editing the entry. In Polish, it is known as "September campaign", which most (if not all) Poles understand as an attempt to defend the country. To clarify this (because the more popular worldwide perspective is attacker-centric rather than defender-centric), I specifically underscored the entry's subject's allegiance to Poland during that time to leave no room for any doubt. Polish-language readers of Wikidata would see that Kononowicz participated in a defensive war, even without this extra information.

If it were up to me, I would create two entries for this single event just to avoid this confusion - invasion of Poland and September campaign - but I think it would be redundant and also cause quite a mayhem in the entirety of this project, where many entries link to the invasion. So my proposition is to simply leave this event as it is - that Bronisław Kononowicz participated in the invasion of Poland as a defender who swore allegiance to the Second Polish Republic.

In case you want to exchange ideas about how to solve this issue or I am mistaken about your reasoning behind those edits, feel free to reach out to me. Thanks and have a good day! BunsBuggy (talk) 17:13, 21 January 2025 (UTC)Reply

I defer to your subject-area knowledge on this. Swpb (talk) 20:24, 21 January 2025 (UTC)Reply

Wikidata weekly summary #664

edit

Deprecated qualifier

edit

Hello, but just out of curiosity, where is the discussion where the property p642 has been made deprecated? Blackcat   15:40, 29 January 2025 (UTC)Reply

The discussion on getting away from P642 started nearly four years ago with this essay and the responses to it, and has continued here, here, here, and on Telegram. To be clear, the property is currently partially deprecated: its use is discouraged where more appropriate properties have been identified, but it is not yet universally obsolete. P642 is unique in having literally hundreds of use cases, and it is necessary to identify and clear out as many of these as possible before making it formally deprecated, to avoid being left with statements using a deprecated property with no clearly identified replacement. This is an ongoing, years-long process, but I see formal deprecation happening in the next few months. Swpb (talk) 16:19, 29 January 2025 (UTC)Reply
Interesting. Thus in the mid-long term p642 is going to disappear, isn't it? BTW, I asked about that because some queries and templates are based on of (DEPRECATED) (P642) as qualificator of instance of (P31) -- Blackcat   19:39, 30 January 2025 (UTC)Reply
Yes, P642 is eventually going to disappear. Do you know of any specific templates/modules/etc. still relying on it? The only one I've encountered so far was Module:Cycling race, which has since been updated. Swpb (talk) 19:50, 30 January 2025 (UTC)Reply
I am talking about some templates on it.wiki, but they have been fixed meanwhile. -- Blackcat   21:38, 30 January 2025 (UTC)Reply

taxon synonym of

edit

Hello, thanks you very much for your work but this property was intended to be used as main value, e.g. this was sufficiant, and at the opposite this seems not adequat and redundant. It was even one of the purpose of that property: to avoid the use of "of" (and consequently of "sublect has role"), thing originally not possible with "sublect has role". When the property "taxon synonym of" is added then "subject has role: synonym" should simply be removed. Same thing when the item is "instance of: synponym" , e.g. there, once the new property added the "instance of: synonym" should simply be removed. Christian Ferrer (talk) 20:55, 30 January 2025 (UTC)Reply

Agreed. A mistake on my part, apologies. I looked for other examples to fix, but only found one remaining. Swpb (talk) 14:31, 31 January 2025 (UTC)Reply
As per this talk page it is used 7788 time as qualifier... Christian Ferrer (talk) 14:46, 31 January 2025 (UTC)Reply
Fair enough; I was only looking at where its use as a qualifier was redundant to a main statement P2868 = synonym (Q1040689) already on the item. I'm happy to convert qualifiers to main statements where the main statements don't already exist. Swpb (talk) 14:59, 31 January 2025 (UTC)Reply
Thanks you for that. I have a question, this is a lot of items, do you proceed in an automated manner? Christian Ferrer (talk) 15:50, 31 January 2025 (UTC)Reply
Yes, with this. The script is running now, but it takes a while to process the inputs before starting batch edits. A batch this size will take about 4 hours to process inputs and 2 hours to make the edits. Swpb (talk) 16:58, 31 January 2025 (UTC)Reply
Thanks for the answer, best wishes. Christian Ferrer (talk) 21:23, 31 January 2025 (UTC)Reply

Weekly Summary #665

edit

Wikidata weekly summary #666

edit

Wikidata weekly summary #667

edit

P39 - 642

edit

Hi! While I support your idea of eliminating of (DEPRECATED) (P642), I think it is premature to withdraw the edit that allows P642 in position held (P39).


I currently add a lot of this, typically to indicate the location of a position and for posts where organization directed by the office or position (P2389) is not applicable. Pallor (talk) 14:03, 23 February 2025 (UTC)Reply

For the relation you are talking about, there is work location (P937). You see, I've put in the effort to make sure appropriate replacements are available for each usage. Is there any other usage of P642 on P39 that you think isn't covered by one of the documented use cases? Cheers, Swpb (talk) 17:29, 23 February 2025 (UTC)Reply
There are many situations where the P937 is not useful. The most well-known example is when an ambassador is accredited to a country where there is no diplomatic mission of the sending country (but don't stop there, it is common for someone not to work in the administrative unit that is their responsibility - think war, revolutionary situations, for example).
Where I still use P642 is as a function of the position of someone who is not at the top of the hierarchy: this could be a deputy minister, state secretary, party secretary, deputy director, etc.
Another problem is the range of infoboxes on Wikipedia. Currently, in all infoboxes, it is enough to call up the P642 data if you want to display the data explaining the position. The more properties we break down the "of" parameter into, the harder it is for the infobox programmer. Even now, we don't see much of the data that we have taken out of P642, but it would be nice to have a finite number of variations, otherwise the infoboxes become unusable. Pallor (talk) 18:23, 23 February 2025 (UTC)Reply
So, again, I think the appropriate replacements are available and documented. Let me link you to the specific table. If there's no official mission, but the ambassador is still located in their assigned country, P937 still applies. If they aren't, applies to jurisdiction (P1001) would be the fallback specified in the table (since ambassadors are officials). If someone works for but does not lead an organization, employer (P108) is what's needed. These assignments can be changed if they are wrong. The variations are quite finite though, and generally very few in number for any given domain, such as would apply to an infobox. I'm also not aware of any infoboxes still relying on P642 – when they do, and get broken by migrations, someone generally comes quicky to complain, and we get the reliance fixed before doing those migrations again. That has happened, to my knowledge, only twice. If you're aware of others, let me know and they'll be top priority. Swpb (talk) 21:28, 23 February 2025 (UTC)Reply
@Pallor: I would value your response to this, as I'm weighing when to ask for full deprecation (see this draft of an RfC). Swpb (talk) 16:04, 25 February 2025 (UTC)Reply
Dear Swpb! I have to discuss this with the Hungarian community, because it would certainly be a loss for the Hungarian Wikipedia Person infobox (hu:Sablon:Személy infobox) to completely eliminate P642. I don't know exactly how the infoboxes were checked, but I strongly suspect that several Wikipedia infoboxes will not display the data correctly, but obviously I don't know the French or Romanian infoboxes in such depth to be able to comment on this responsibly. Unfortunately, by the way, in your reply you have described exactly what I feared: instead of "of", several other (even if finite) qualifiers should appear in the infoboxes.
I will provide a link to the Hungarian discussion later (but the discussion there will obviously be in Hungarian) Pallor (talk) 16:19, 25 February 2025 (UTC)Reply
Thanks so much. There is actually no way (that I know of) to preemptively find external uses, so we have relied on external users noticing when their infoboxes etc. break, and coming to complain. I will take a look at the person infoboxes. Now, as to P642 having multiple replacements, I'd like to know why you see that as something to be feared. In fact, that is the whole point of the deprecation effort – when one property can mean so many different things, it is often impossible for people (and moreso for machines) to tell what it means in any given context. In the complex Module:Cycling race, we dealt with a case where one set of infoboxes was using in P642 in ten different ways. Even in that extreme example, once it was agreed what the replacement properties should be, updating the module (and its dependents) was pretty straightforward. We were easily able to have infoboxes check both the old and new properties, to avoid disruption; in the same way, parameters can check more than one property permanently, if e.g. the needed values may use employer (P108) or organization directed by the office or position (P2389). I see the Hungarian person infobox uses P642 in only two places, both of which are already looking at multiple properties, so fixing that will be super easy. Swpb (talk) 16:45, 25 February 2025 (UTC)Reply
First of all, I'll show you where I raised the issue, but of course you don't need to follow that, because the pollemma is in our language: hu:Wikipédia-vita:Sablonműhely#Jelentős_változás_a_Wikidatán_(Személy_infobox)
I will show you with concrete examples what I mean by loss, and I want you to see that this is not a Hungarian specific problem. If someone holds an office in Hungary like "alispán" (subprefect/vicecomes), it should be indicated in which county he/she holds it. Indicating this with P642 gives this result: hu:Ghyczy Dénes (Dénes Ghyczy (Q115113257)) (Now it is good)
If someone was a state secretary, we use P642 to indicate in which ministry: hu:Szakasits Antal (Antal Szakasits (Q1294013)) (Now it is good)
In the infobox it is shown that he was a library director, but not where or which library he was director of. This is what the infobox looks like without the library hu:Mader Béla (Béla Mader (Q16521633)) (P2389 No longer good)
In the infoboxes, these data appear under the same parameter (in the Hungarian infobox: "tisztség"), but now there is not one qualifier to specify the data, but 3, 4, 5 or 6, i.e. many. The question (which needs to be discussed with the Hungarian community) is whether the infobox can display these?
Currently on Wikidata, there are more than 3000 items in P642 that have a classification qualifier/explanatory data, 1500 of these items have a article, i.e. the infobox is involved (if every Wikidata item had a glossary, this number would be 3000). Pallor (talk) 17:32, 25 February 2025 (UTC)Reply
Regarding the number of affected languages: I have just manually checked all versions of Template:Infobox person (Q6249834). Of 218 Wikipedias and 24 other wikis, 6 5 (besides Hungarian) read the property, namely: Abkhazian, Greek, Irish, Hebrew, Norwegian, and Occitan. These infoboxes can very easily display the contents of multiple (or all) qualifiers in place of one; most, including Hungarian, already do. Swpb (talk) 18:10, 25 February 2025 (UTC)Reply

@Pallor: Having seen how easily the Hungarian template was fixed, and how few other templates are affected (I've given notice on the talk pages of each of them), does the removal of P642 as an allowed qualifier of P39 still cause you fear? Not trying to be glib; your opinion matters to me. Swpb (talk) 19:23, 25 February 2025 (UTC)Reply

Please wait a little longer, there is still some work to be done on the infobox, so discussions are still ongoing. Thx. Pallor (talk) 22:41, 25 February 2025 (UTC)Reply
Fair enough, holding off. Let me know when you're happy with the template, or if you'd like some help with it. Swpb (talk) 00:24, 26 February 2025 (UTC)Reply

Hi Swpb! Thanks for the link in the Hungarian Wikipedia, the interface is very interesting.

I have a question about applies to jurisdiction (P1001): according to the description, this property cannot be assigned to a person, only to the item of the positionss. It would have been nice to have been able to replace the P642 qualifier in the element of a thousand persons, but it cannot be used for this purpose. It would also be better to take it out of your table, because it is ambiguous, now it looks like it can be used as a qualifier for people element.

What to use instead? We would need a property that can be used as a qualifier, can be used for people element (P31:Q5), and can be used in case the person geographically did not do their job in the place they were assigned to manage. Pallor (talk) 12:14, 5 March 2025 (UTC)Reply

You are misunderstanding how P1001, and indeed qualifiers in general, are used. When used as a qualifier, "the item (institution, law, public office, public register...)" in the description applies to the main value of the qualified claim, not the subject of the page. Qualifiers in general take as their subject either the claim they are on or its value, not the subject item of the page as a whole. It is absolutely legitimate, for example, to qualify Cecília Müller (Q88571018)position held (P39)Chief Medical Officer (Q89199480) with applies to jurisdiction (P1001) = Hungary (Q28), because we are giving information about the position Chief Medical Officer (Q89199480) held by Cecília Müller (Q88571018), not about Müller directly. (We would never say Cecília Müller (Q88571018)applies to jurisdiction (P1001)Hungary (Q28); that's what the description and subject type constraint are trying to prevent.) Swpb (talk) 14:40, 5 March 2025 (UTC)Reply

Wikidata weekly summary #668

edit

Launch sites

edit

Hey there, I noticed you switched the use of of (DEPRECATED) (P642) to product, material, or service produced or provided (P1056) on launch sites and pads, e.g. this. Would you mind running back through and change these uses to vehicle normally used (P3438)? P1056 since the rockets/missiles aren't produced there and aren't services provided, so P3438 makes more logical sense. Thanks! Huntster (t @ c) 02:43, 28 February 2025 (UTC)Reply

Yes, I see what happened: on those items, of (DEPRECATED) (P642) was being used incorrectly, to qualify another qualifier (has use (P366) = missile launch facility (Q853409)), instead of the main statement (significant event (P793) = construction (Q385378)). When the value of a main statement is construction (Q385378), it's ordinarily safe to assume the value of of (DEPRECATED) (P642) is the thing constructed. With hundreds of thousands of uses to migrate, there is unfortunately no good way to detect this sort of erroneous use at scale. Changed these to P3438. Swpb (talk) 05:23, 28 February 2025 (UTC)Reply
Oh, completely agree, migrating away from P642 is a monumental task. Thank you for being responsive. Huntster (t @ c) 05:54, 28 February 2025 (UTC)Reply

Wikidata weekly summary #669

edit

Wikidata weekly summary #670

edit

Wikidata weekly summary #671

edit

Wikidata weekly summary #672

edit

Wikidata weekly summary #672 (correct version!)

edit

UK

edit

I don't see how United Kingdom of Great Britain and Ireland (Q174193) as the predecessor state to United Kingdom (Q145) can be the country (P17) of it? Vicarage (talk) 16:09, 27 March 2025 (UTC)Reply

Fixed. Swpb (talk) 16:57, 27 March 2025 (UTC)Reply

Wikidata weekly summary #673

edit

Wikidata weekly summary #674

edit

Wikidata weekly summary #675

edit

P642 bot

edit

Dear Swpb! Could you please use your bot to change the following schedules?

Only for country of citizenship (P27): Hungary (Q28) or Kingdom of Hungary (Q171150):

position held (P39): ispán (Q17393320) and lord-lieutenant of a county (Q29865383) alispán (Q5669049) chief magistrate (Q123023456)

of (DEPRECATED) (P642) -> applies to jurisdiction (P1001)

Thanks in advance! Pallor (talk) 10:31, 17 April 2025 (UTC)Reply

Running those batches now. For what it's worth, you don't need a bot to perform this migration: I've tried to make the migration process accessible to all, though its not as simple as I'd like. If you happen to try it yourself, I would be extremely grateful for any feedback on how it could be easier; we're never going to get rid of P642 if I'm the only one doing batch migrations. Swpb (talk) 14:16, 17 April 2025 (UTC)Reply
Thank you very much for how you did it. Unfortunately, I don't know how to use this tool. Pallor (talk) 23:16, 17 April 2025 (UTC)Reply

neighbourhood of

edit

Please check before adding located in the administrative territorial entity (P131). For at least two of your recent edits,

St Annes (Q19832995)located in the administrative territorial entity (P131)Lytham St Annes (Q1757473) and
Kelvinbridge (Q113261121)located in the administrative territorial entity (P131)Glasgow (Q4093)

the new statement targets are in fact not administrative entities; and adding these statements has confused these items' place in the chain of items that are administrative entities.

I don't know how many more of your new statements are similarly wrong, but that was two out of two in the two that I checked. Jheald (talk) 22:45, 23 April 2025 (UTC)Reply

Sorry, I'm undoing that batch. Should probably be location (P276); will redo with that property tomorrow. Swpb (talk) 00:33, 24 April 2025 (UTC)Reply
@Jheald: It looks like the statement values in your two examples are, in fact, administrative territorial entities, by transitivity of subclass of (P279):
If you don't have any further concern, I will re-apply the migration to all such statements. Swpb (talk) 14:42, 24 April 2025 (UTC)Reply
location (P276) would be better. Glasgow is not the directly relevant administrative territorial entity, that would be the Glasgow City council area; and Lytham St Anne's is no longer a civil parish. Jheald (talk) 11:29, 2 May 2025 (UTC)Reply
Changed to location (P276) on those two. In general, though, 1) it doesn't look like located in the administrative territorial entity (P131) restricts its values to the most direct administrative entity, and 2) statements are considered valid if they were true at some point, not necessarily the present. If all P131 statements where the value is a former administrative territorial entity (Q56061) were considered invalid, we would be looking at a lot of statements, very few of which I had anything to do with: here's a limited sample. Swpb (talk) 14:38, 2 May 2025 (UTC)Reply

P12765

edit

Hello, as the last time with "synonym of", it seems that you did mechanically an additional unwanted edit. E.g. this step is usefull, but the following one was not necessary. If we created the property it is especillay to avoid/replace the use of "subject has role". I have fixed some manually and I noticed that there are a lot of them [1]. I have restablished the constraint "as main value". We don't need anymore "subject has role" here, so is there a way that you use an automated tool to remove them? Christian Ferrer (talk) 06:27, 26 April 2025 (UTC)Reply

Thanks for catching this Christian. I can easily remove all these on Monday (I'm on mobile until then so I don't have access to my tools). By the way, these aren't recent edits – they predate our January conversation about taxon synonym of (P12763). I just didn't think at the time to apply that conversation to related properties. I understand that all these properties, including the new one, are intended as main only. Swpb (talk) 23:10, 26 April 2025 (UTC)Reply
Great, many thanks! Christian Ferrer (talk) 05:33, 27 April 2025 (UTC)Reply
I believe this is complete. Swpb (talk) 17:38, 28 April 2025 (UTC)Reply

@Christian Ferrer: I have found five statements on two items where I think it would be valid to use nomenclatural type of (P13478) as a qualifier: query. nomenclatural type of (P13478) can't be a main property on these items, since the subjects are neither type nor taxon, but scholarly papers. The alternative to adding those two exceptions would be to use a qualifier like relative to (P2210) or characteristic of (P13044), but that's obviously not as specific. Ok to add those papers as constraint exceptions? Swpb (talk) 19:53, 28 April 2025 (UTC)Reply

Thanks for asking. I removed the values stored with "of" for those cases, furthermore the concerned taxa are already quoted as main subjects. The new property is intended to be used for taxonomic types not for the articles dealing with that stuff. Christian Ferrer (talk) 20:03, 28 April 2025 (UTC)Reply

Wikidata weekly summary #677

edit

Wikidata weekly summary #678

edit

Migration of isotopes

edit

Hu Swpb, I saw those kind of edits. https://www.wikidata.org/w/index.php?title=Q466603&oldid=2346553250

I think the relative to (P2210)   as a qualifier to link isotope of hydrogen (Q466603)      to "Hydrogen" is inappropriate. The link should be done with is metaclass for (P8225)   as a main statement, it's made for this.

"Isotopes of hydrogen" is a metaclass, a class for isotope-classes like "deuterium" to be instance of. Any instance of "Isotope of hydrogen" is a subclass of "hydrogen". This is the definition for "is a metaclass for", exactly. author  TomT0m / talk page 15:42, 12 May 2025 (UTC)Reply

You're right, and I've migrated all such statements accordingly. Thanks for point it out! Swpb (talk) 17:46, 12 May 2025 (UTC)Reply

Wikidata weekly summary #679

edit

P642 migration for antipopes

edit

Hello, I think there was a small mistake in case like this. Perhaps in opposition to (P5004) is more appropriate? Thanks, --Jahl de Vautban (talk) 18:22, 16 May 2025 (UTC)Reply

Yes, good catch, I'll fix immediately. Swpb (talk) 18:28, 16 May 2025 (UTC)Reply

Wikidata weekly summary #680

edit

Wikidata weekly summary #681

edit

Modelling Subject has role refugee

edit

I have Norwegian war refugees register ID (P5871) for a person with subject has role (P2868) refugee (Q131572) and qualifier of (DEPRECATED) (P642) with Sweden (Q34). In other words, a person fleeing from Norway to Sweden during World War II. What can be used as qualifier here?. Also notifying The Project WikiProject Deprecate P642 WikiProject does not exist. Please correct the name. Pmt (talk) 06:30, 29 May 2025 (UTC)Reply

Possibly destination point (P1444) 217.211.178.17 12:41, 29 May 2025 (UTC)Reply
Yes, destination point (P1444) is the way to go. Swpb (talk) 16:06, 29 May 2025 (UTC)Reply
@Swpb Can you also implement this solution into «Project WikiProject Deprecate P642» Pmt (talk) 17:11, 29 May 2025 (UTC)Reply
Wikidata:WikiProject_Deprecate_P642/Use_cases#refugee. Swpb (talk) 18:19, 29 May 2025 (UTC)Reply
Thats fine for me. Thank you! Pmt (talk) 17:09, 29 May 2025 (UTC)Reply

Wikidata weekly summary #682

edit

Wikidata weekly summary #683

edit