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

Wikidata:Project chat

From Wikidata
(Redirected from Wikidata:Project Chat)
Jump to navigation Jump to search

Wikidata project chat
Place used to discuss any and all aspects of Wikidata: the project itself, policy and proposals, individual data items, technical issues, etc.
Please take a look at the frequently asked questions to see if your question has already been answered.
Please use {{Q}} or {{P}}, the first time you mention an item, or property, respectively.
Requests for deletions can be made here. Merging instructions can be found here.
IRC channel: #wikidata connect
On this page, old discussions are archived. An overview of all archives can be found at this page's archive index. The current archive is located at 2018/09.






a query

for deletions

for comment


for permissions


for deletion

and imports



Denisova 11[edit]

Is Denisova 11 (Q56233289) to be treated as humanoid (Q502931)? --Succu (talk) 21:30, 22 August 2018 (UTC)

As you know, I just created that, Yet you have not pinged me here. Why is that? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:33, 22 August 2018 (UTC)
Oh, I pinged you. So why? --Succu (talk) 21:38, 22 August 2018 (UTC)
Here we present the genome of ‘Denisova 11’, a bone fragment” --Succu (talk) 21:59, 22 August 2018 (UTC)
Are not those folks simply archaic humans (Q284851)?Bvatant (talk) 21:55, 22 August 2018 (UTC)
So how this item could be remodeled, Bvatant? --Succu (talk) 22:08, 22 August 2018 (UTC)
archaic humans (Q284851): A subclass of human evolution (Q83944)?!? Has part Homo heidelbergensis (Q105784)?!? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:27, 22 August 2018 (UTC)
Succu Andy Mabbett Indeed I don't know, because we stumble here on the fact that archaic humans (Q284851) is a subclass of human evolution (Q83944), has part(s)(?) elements such as Neanderthal (Q40171), itself instance of taxon (Q16521). I can't make sense for the moment of such a mess. The way taxons conflate with classes in WD is just a mystery for me, I'm currently trying to make sense of it and clarify constraints on parenthood relations mother (P25) father (P22) child (P40), in order to enable race horses, giant pandas and other pets to have proper genealogy, along with deities, dragons etc. I'm pretty sure that if Denisova 11 (Q56233289) is declared as type archaic humans (Q284851), she will not be allowed by the current constraints to have a father and mother like you and me, more than she is with her current types: neither humanoid (Q502931) nor hybrid (Q42621) are allowed to bear mother (P25) or father (P22) in the current state of affairs. Will think about it, but a quick and dirty way to shunt the taxonomic jungle would be to declare those folks as instance of person (Q215627), which would be more respectful to them than humanoid (Q502931) or hybrid (Q42621). But just my 0.02€ :-) Bvatant (talk) 22:51, 22 August 2018 (UTC)
I think archaic humans (Q284851) are all the members of Homo (Q171283) (Human) which aren't anatomically modern human (Q5891007). Taxons vs subclasses are a bit of a mystery. I thought Wikidata items needed to be a subclass of something if they could have instances. E.g., requires that Canis lupus (Q18498) be a subclass of something, presumably Canis (Q149892). But I only get reverted if I do that, most recently with "it's already a subclass". Ghouston (talk) 03:27, 23 August 2018 (UTC)
We could use instance of Q171283; notwithstanding that that is but an instance-of-an-instance-of first-order metaclass (Q24017414). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:42, 23 August 2018 (UTC)
Any comments to this solution? --Succu (talk) 18:41, 24 August 2018 (UTC)
@Succu: Now I'm a bit lost, sorry. Which problem does this change solve? We have two different problems here, #1 consensus on the type(s) of Denisova 11 (Q56233289) and #2 allow at least one of those types to bear mother (P25) and father (P22). Do you mean your proposal solve both, supposing the type archaic humans (Q284851) is added to Denisova 11 (Q56233289)? I don't want to nitpick, again, but those things are really tricky. In any case, seems to me your solution conflates the person/organism who was alive here and there, and its fossil remnant. A fossil does not have mother and father, for example. See my other comment below.Bvatant (talk) 20:50, 24 August 2018 (UTC)
I responded to this remark. --Succu (talk) 21:02, 24 August 2018 (UTC)
I'd like to discuss Denisova 11 (Q56233289) nothing else. I don't think the current English description (girl with a Neanderthal mother and a Denisovan father) is correct. „Denisova 11“ is the name of a bone fragment. I created minimalistic items for earlier findings: Denisova 3 (Q56240560), Denisova 4 (Q56240584), Denisova 5 (Q56240616) and Denisova 8 (Q56240592). I doubt the usage of place of burial (P119) and related properties is correct. All fragments were found at Denisova Cave (Q1029322). I can not find any evidence that the girl - which the bone fragment was part of - was buried there. I don't think the usage of hybrid (Q42621) for a human is correct. --Succu (talk) 19:56, 23 August 2018 (UTC)
Succu +1 ! I pretty much like your minimalist descriptions, as they elegantly kick off the taxonomic issue. And indeed, those entities are fragmentary fossils, not persons (whatever their classification).Bvatant (talk) 20:45, 23 August 2018 (UTC)
You appear to be labouring under a misapprehension. The item is about the humanoid, not the bone. If you wish to create an item about the bone, no-one is stopping you. Readers will also note that P119 has many aliases, not all of which imply burial per se; and that "Hybrid" is exactly the term used by one of the cited sources - in its title no less. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:20, 23 August 2018 (UTC)
Further: "The tiny arm or leg fragment belonged to Denisova 11, a 13-year-old hybrid hominin" - [1]. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 22:33, 23 August 2018 (UTC)
Your latest argument, Andy Mabbett, led me to read through en:Denisovan#Fossils and indeed "Denisova 2 and Denisova 3 are prepubescent or adolescent females, while Denisova 4 and Denisova 8 are adult males". If I get correctly the meaning of are in this sentence, the names "Denisova n" denote individual persons, and not their fossil remains. So much for your interpretation, Succu, even if I was tempted to buy it yesterday. I would be tempted to extend parenthood properties used on human (Q5) to the whole genus Homo (Q171283). But what is the best way to do that?  – The preceding unsigned comment was added by Bvatant (talk • contribs) at 16:47, 24 August 2018‎ (UTC).
„Denisova n“ refers to the label given to the specimen. See e.g. „Scientists exploring Denisova Cave in the Altai Mountains discovered the worn baby tooth in 1984 and labeled it ‘Denisova 2.’“. --Succu (talk) 18:32, 24 August 2018 (UTC)
@Succu:Sorry for nitpicking :-) ... the source you quote is not convincing at all. First it's not a primary research paper, but an article in general press, less precise on vocabulary in general, and the referent of "it" in "labeled it" can be either the tooth or the baby. Regarding Denisova 11, this article at Smithsonian's is clearly using the name several times as the name of the girl, not of her bone, e.g., Denisova 11’s mother was more closely related to Neanderthals dwelling in western Europe than those residing in the Siberian cave some 120,000 years ago. Denisova’s paternal relatives, on the other hand, stuck to the region surrounding the cave—Denisova 3, the hominin whose pinky toe first led scientists to the species, lived in the area a few thousand years after Denisova 11.Bvatant (talk) 19:29, 24 August 2018 (UTC)
Why are people assuming there is no ambiguity here and that those who use it one way or the other are simply wrong? - Jmabel (talk) 19:33, 24 August 2018 (UTC)
Please see above, where I wrote "The item is about the humanoid, not the bone. If you wish to create an item about the bone, no-one is stopping you.". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:16, 24 August 2018 (UTC)
So this is item not about a member of Hominidae (Q635162), but only the reseamblace (humanoid) with a „human“ (=Q5?) --Succu (talk)
I presume Andy meant to write "hominid" rather than "humanoid". And I suspect you are trying to pick a fight rather than to reach consensus. - Jmabel (talk) 02:58, 25 August 2018 (UTC)
Did he? He referenced humanoid (Q502931) [2], [3]. --Succu (talk) 20:57, 25 August 2018 (UTC)
From the original publication (first sentence of the abstract): „Neanderthals and Denisovans are extinct groups of hominins that separated from each other more than 390,000 years ago.“. What consensus do you have in mind, Jmabel? --Succu (talk) 20:23, 27 August 2018 (UTC)
The one most of the people here seem to be trying to work toward, rather than see how close to their own original view we can end up. - Jmabel (talk) 22:28, 27 August 2018 (UTC)
From en:Hominid (disambiguation): „A hominid is any primate in the family Hominidae“. From en:Humanoid: „is something that has an appearance resembling a human without actually being one“. From your point of view: are these references well done? --Succu (talk) 20:47, 28 August 2018 (UTC)
To cite from a publication I referenced today multiple times Genera of the human lineage (Q28177674): „What Is a Hominid?“ They are not wondering about to be humanoid (Q502931), Mr. Pigsonthewing. --Succu (talk) 20:00, 1 September 2018 (UTC)
Try Nuclear and mitochondrial DNA sequences from two Denisovan individuals (Q29013840). Of course these fragments belog to a particular individual, but they are not same as these individuals. --Succu (talk) 20:35, 24 August 2018 (UTC)
@Jmabel:. Yes there is ambiguity, that's why we are discussing :-). For the sake of Wikidata consistency (at least local consistency) we must come to a consensus, or agree the name is ambiguous and create two elements, one for the person, and one for the fossil.Bvatant (talk) 20:52, 24 August 2018 (UTC)
Only the secondary sources (you added) use the term „hybrid“ not the primary source: The genome of the offspring of a Neanderthal mother and a Denisovan father (Q56234568). --Succu (talk) 20:09, 27 August 2018 (UTC)
Neanderthal (Q40171) == Extinct species of the genus Homo"; Denisova hominin (Q151055) == "Paleolithic-era species of the genus Homo"; ;hybrid (Q42621) == "offspring of cross-species reproduction". Though why the former two items are not marked up as having, for instance, parent taxa, is left as an exercise for the reader. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:58, 28 August 2018 (UTC)
Could you please explain this cryptic answer. Thanks in advance. --Succu (talk) 21:00, 28 August 2018 (UTC)
wikt:exercise for the reader. But never mind, I found out why Q151055 has no parent taxon Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:13, 28 August 2018 (UTC)
Oh, it would be very interesting to have scientific reference for the treatment of Denisova hominin (Q151055) as a species (Q7432). --Succu (talk) 21:06, 28 August 2018 (UTC)
Perhaps you could ask an editor who labelled it as such. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:13, 28 August 2018 (UTC)
That's the way you argue: Version vom 11. März 2014, an early Bot expierence?
BTW The evolution of Homo sapiens denisova and Homo sapiens neanderthalensis miRNA targeting genes in the prenatal and postnatal brain (Q28603765) is a source which treated Denisova hominin (Q151055) as subspecies (Q68947)... --Succu (talk) 21:29, 28 August 2018 (UTC)
If this item is about the girl then why this item is not an instance of human (Q5)? --Succu (talk) 21:14, 24 August 2018 (UTC)
The description of Q5 is "common name of Homo sapiens (Q15978631), unique extant species of the genus Homo" (though I shall now remove the QID from that)). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:18, 24 August 2018 (UTC)
I think the usage of female (Q6581072) implies human (Q5). You added the references. --Succu (talk) 21:26, 24 August 2018 (UTC)
I would gladly extend the definition of Q5 to include the whole genus Homo.Bvatant (talk) 21:36, 24 August 2018 (UTC)
I don't think that's OK. Please see population (Q15840798). --Succu (talk) 21:29, 25 August 2018 (UTC)
That is a subclass of population (Q2625603) (which you added on 25 August) and of Homo (Q171283) (which you added on 26 August). Aside from that issue; what is its relevance here? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:20, 1 September 2018 (UTC)
I'm aware of what I added to population (Q15840798), and I think you are aware about what I meant above. --Succu (talk) 20:08, 1 September 2018 (UTC)
My question - which you have not answered - was "what is its relevance here?". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:25, 1 September 2018 (UTC)
„relevance“ of what. Please don't be vague. Thx. --Succu (talk) 21:13, 1 September 2018 (UTC)

I removed two wrong references to The genome of the offspring of a Neanderthal mother and a Denisovan father (Q56234568). --Succu (talk) 20:14, 1 September 2018 (UTC)

Promptly reverted of course... --Succu (talk) 20:20, 1 September 2018 (UTC)
[ec] I've restored the source. I note that Succu gives no indication of why they believe it to be "wrong". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:25, 1 September 2018 (UTC)
What about a full text search? --Succu (talk) 20:30, 1 September 2018 (UTC)
Maybe you find this change more „entertaining“. --Succu (talk) 20:52, 1 September 2018 (UTC)
entertaining. Not for me! --Succu (talk) 21:01, 1 September 2018 (UTC)

Is there any reason why this girl should not be treated as an instance of Denisova hominin (Q151055)? --Succu (talk) 05:48, 2 September 2018 (UTC)

Yes: The subject is "...a hybrid between Neanderthals and Denisovans.", as quoted from one of the cited sources; had a "Neanderthal mother" according to a second cited, and quoted, source; and was "...the offspring of a Neanderthal mother and a Denisovan father" according to the quoted title of a further cited source, which you yourself created as an item - which is why the reason I gave in the edit summary for the diff you provide was "per cited & quoted sources". Conversely, your claim that she was a straightforward "instance of Denisova hominin" was uncited. Your repeated, and unexplained, attempts to make this item about anything other than what it actually is about are becoming tendentious. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:15, 2 September 2018 (UTC)
So if you insist on the exact wording of your secondary sources: I can not find the word humanoid (Q502931) in your reference Remains of hybrid human girl with Neanderthal mother discovered in Siberian cave. What I'm missing? --Succu (talk) 15:07, 4 September 2018 (UTC)
What are you missing? My post with the time stamp of "22:42, 23 August 2018", it seems. "if you insist on the exact wording of your secondary sources" I do not. HTH. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:57, 4 September 2018 (UTC)
I deleted the wrongly sourced statement. --Succu (talk) 20:00, 4 September 2018 (UTC)
And I reverted you, because it is a sourced statement. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 05:47, 5 September 2018 (UTC)
It's not the first time you reverted claiming „it is a sourced statement“ (aka as before). Could you please explain how your sourced statement supports the claim you made? Thx. --Succu (talk) 21:17, 5 September 2018 (UTC)
BTW: Most of your claims raise a „type constraint“. As father of Denisova 11 (Q56464945) and mother of Denisova 11 (Q56464406), created by User:Yair rand (thx) do. --Succu (talk) 20:41, 4 September 2018 (UTC)
As I have said previously, bad constraints can and should be fixed. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 05:47, 5 September 2018 (UTC)
For hybrids we have parent of this hybrid, breed, or cultivar (P1531), so I've made an adjustment. --Succu (talk) 20:48, 11 September 2018 (UTC)
Reverted three times one, two, three with the edit comment as before by Mr. Mabbett. What exactly is wrong with this addition? --Succu (talk) 19:13, 17 September 2018 (UTC)

Maybe Archaic human admixture with modern Homo sapiens (Q4785541) helps for a more general understanding... --Succu (talk) 21:41, 4 September 2018 (UTC)

Subclass of Gene flow (Q143089)? Involving H. sapiens? How does that help? Oh, and since you are concerned with „type constraints“, that has one. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 05:48, 5 September 2018 (UTC)
Done. --Succu (talk) 21:17, 5 September 2018 (UTC)

Two Wikipedias have now an article. Maybe this helps to get this settled. --Succu (talk) 18:26, 16 September 2018 (UTC)

Make Wikidata P and Q templates universal in Wikimedia projects[edit]

I am writing to propose making P and Q templates in Wikimedia projects outside Wikidata, in every language, to redirect to Wikidata. Any comment here in support or criticism of this idea will help future planning.

Is anyone aware of this being discussed before? Can anyone say where they saw this, even if it was another language or project?

Consider {{P}} and {{Q}}. In Wikidata these templates are fundamental to quick discussion of Wikidata items and properties because they convert Wikidata identifiers to human readable text for conversations. Outside of Wikidata in other Wikimedia projects there are often P and Q templates which link to Wikidata. For example, see Commons:Template:P and Commons:Template:Q. Having these templates in place means that irrespective of language or project, editors can use P and Q templates to make connections to Wikidata.

Check out Template:Property (Q19694638) and Template:Wikidata entity link (Q17280715), which are Wikidata's own records of which Wikimedia projects have equivalent templates by any name. Each of these are currently in about 40 Wikimedia projects. In about 20 cases the name is P or Q. In other cases perhaps there is another name, but P and Q are redirects. For some of the cases P and Q go to non-Wikidata uses. I think there are about 800 Wikimedia projects total. If it would be useful to make P and Q templates universal then we should make some plan to set that up or reserve it.

In English Wikipedia en:Template:P currently goes to en:Template:Smiley to make an emoji. In German it goes to en:Template:P for "priority" or the number 1. In French fr:template:P goes to the portal system. It causes big social problems to change templates which get established in Wikimedia community projects. Wherever possible we should propagate templates out now so that in the future we do not have to negotiate for this in every language and every project. Currently I am talking about this on English Wikipedia at en:Template talk:P where there is resistance to the idea.

Sorry - I have a major deficiency in this, in that I do not know the best way to migrate templates across languages and projects. If anyone can point to documentation on how to do this or outline the workflow of how this should go down then speak up. Also, if anyone has an objection to making P and Q templates universal then please say why. Blue Rasberry (talk) 14:04, 5 September 2018 (UTC)

The simplest solution is to do this as parser functions. I already wrote those parser functions when I was at WMDE, but it was turned down as it would interfere with community work on templates. In my opinion parser functions are the right way to do this. Jeblad (talk) 18:42, 5 September 2018 (UTC)
@Jeblad: Thanks, I would not have known to talk about that. What do you mean you wrote those parser functions? You mean for migrating arbitrary templates, or for this case specifically? Blue Rasberry (talk) 01:29, 6 September 2018 (UTC)
@Bluerasberry: As I recall they were parser functions for label and description, but in effect nearly the same. Jeblad (talk) 02:25, 6 September 2018 (UTC)
Most projects have little or no need for such a template, and occupying two of the single-letter template names is really a bad idea, probably disrupting dozens of local workflows and making everyone there memorize new names. I would advise against trying to make these universal. --Yair rand (talk) 00:20, 6 September 2018 (UTC)
@Yair rand: How would you feel about doing this for languages which do not have a P or Q and instead use non-Latin characters? What about for languages which have these letters, but where P and Q templates do not exist? Blue Rasberry (talk) 01:29, 6 September 2018 (UTC)

This is not something that Wikidata can decide - it is exclusively a matter for the community on each individual project. I guess that it is unlikely to be a significant issue to create them on wikis that do not currently have templates by this name (unless it conflicts with any local policies, e.g. I have a very hazy recollection that one of the Indic language wikis doesn't (or didn't) like templates other than in that language's script). On wikis where templates by these names do already exist you need to discuss any proposals to change with the community on that project. Thryduulf (talk) 16:30, 6 September 2018 (UTC)

Any project that want to use Wikidata, or really that enable the client part of the Wikibase extension, will at some point need tools to point to items, properties, and a number of other entities. Now those projects has to implement those tools themselves. In fact, when any project must implement a module to make a template work it is a clear indication that some basic tool are missing. To not implement a tool just because it could be implemented as a Lua module is not very wise, and it blocks smaller projects from progressing.
There are several other modules that are reimplemented again and again, and those should also be replaced by generalized tools and libraries. Now many modules are copied between projects, and more often than not they should not be used at all because they fail even quite basic quality measures (not to say tests, they have none).
Please do request the devs to implement basic tools for accessing entities, it is the best way to help smaller communities to access and use Wikidata. Jeblad (talk) 18:17, 6 September 2018 (UTC)
Whether the tools are or are not needed or wanted is irrelevant to this discussion. This is about only the name for templates that refer to Wikidata properties and items (i.e. the equivalent of {{P}} and {{Q}}). This is something that can only be decided by the individual wikis. It is not possible for a discussion here to determine this for any project other than Wikidata. Thryduulf (talk) 20:50, 6 September 2018 (UTC)
Quite frankly, this does not make sense. Jeblad (talk) 11:52, 10 September 2018 (UTC)
@Jeblad: What doesn't make sense to you? Why do you think it doesn't make sense? Thryduulf (talk) 01:30, 17 September 2018 (UTC)

Country and city attribution of geolocated item[edit]

I am trying to make use of GeoData API to perform aforementioned task.

I found GeoData API very clean, simple and user friendly in retrieving various data according to geo location of the item. But there are difficulties with retrieving country/city affiliation of the item. While country can theoretically be get in a single request (also not always but only if being specified and not in name format but rather by its alphabetic designation), the city is possible to be get only for items which are cities by themselves. Let's imagine I want to determine in one request providing the coordinates of the Sagrada Familia temple the name of the item and that it is located in Barcelona, Spain. As far as I understood there is no way to do that. From the second hand this information does exist for every geo tagged item and is available for example through WikiData SPARQL query service. But then I'll need to perform a second request to WikiData which I would have liked to avoid by all means.

Looking on advice on the optimal strategy to perform the desired. If not is it possible to add the country/city attributes to GeoData list=geosearch attributes? -  – The preceding unsigned comment was added by Chainastole (talk • contribs).

Q741775 - "Enemy of the people" and "class enemy"[edit]

After a discussion concerning the term enemy of the people (as used by Donald Trump) I have seen that the Wikidata item Q741775 confounds two different concepts: the enemy of the people (stemming from Roman hostis publicus, used time and again, most notably in Nazi terminology as de:Volksfeind and in the Soviet Union mainly in the 1930s, Russian: ru:Враг народа. There is also the more specific term class enemy (German: de:Klassenfeind, Russian ru:Классовый враг which redirects to ru:Классовая борьба = class conflict/de:Klassenkampf) which was more widely used in the 1920s after the Revolution. There are some articles that deal with the class enemy such as da:Klassefjenden, hr:Klasni neprijatelj, nl:Klassenvijand, no:Klassefienden, and some that deal with the enemy of the people such als en:Enemy of the people, de:Volksfeind, ca:Enemic del poble, pl:Wróg ludu etc. I would like to separate the two (which means that the English language Wikipedia would have a redirect for en:Class enemy only). Any thoughts on this? --Andropov (talk) 08:47, 9 September 2018 (UTC)

  • As a native English speaker, I would definitely agree that these are two distinct concepts, and should be separated, even if they have sometimes been used in a similar manner as political epithets. - Jmabel (talk) 04:17, 10 September 2018 (UTC)
@Jmabel: Thank you for your input! Could you, speaking of muddled concepts, tell me what the difference between en:enemy of the people and en:public enemy is? Kind regards, --Andropov (talk) 10:14, 15 September 2018 (UTC)
In the U.S., at least, "public enemy" has a connotation of actual criminality. The FBI used to put out a list of the "Ten Most Wanted" criminals, and the person at the head of the list was referred to "Public Enemy Number One." "Enemy of the people" is less specific; the phrase echoes the Ibsen play of that name, where it is used more or less ironically, in that the person who is accused of being an "enemy of the people" is the doctor who correctly diagnoses that the waters of the town's famous baths are contaminated, and his publicizing this hurts the local economy. That irony may or may not be there in any particular usage. - Jmabel (talk) 16:25, 15 September 2018 (UTC)

The Dog[edit]

Thanks for merging that thing yesterday. I still don't understand Wikidata but I'll learn more as I get exposed to it. Poking around I discovered that a couple years ago I started the Wikipedia article about a documentary The_Dog_(2013_film) and here on Wikidata I changed The Dog (film) into The Dog (1992 film) which is now The Dog (Q7730510), a 1992 short film by Todd Field. I don't know if they are connected in any way. They shouldn't be if they are. Feel free to fix my links and the articles above as you see fit. Thanks for your attention. ~ JasonCarswell (talk)

All seems ok. There is The Dog (Q28136699) for 2013 film. --Infovarius (talk) 16:45, 21 September 2018 (UTC)

TV anime described as film[edit]

Why are there many TV anime entry on wikidata being mistakenly described as other things like film or TV drama or whatever other things? C933103 (talk) 13:31, 13 September 2018 (UTC)

Hi C933103. In general, this is because Wikipedia articles cover all media of a given franchise together, and this was used to seed Wikidata initially. For example, Black Lagoon covers 5 media (manga, both anime seasons, a light novel and an OAV) or City Hunter 12 (!) media.
I tend to think the correct way to do that is to have the "main item" about the media franchise, and separate item for each different medium. Some examples: Dragon Ball (Q2020) Naruto (Q642), Ghost in the Shell (Q92580), Cardcaptor Sakura (Q49182)
That’s a lot of work though :)
Jean-Fred (talk) 14:48, 13 September 2018 (UTC)
That would be hard to do interwiki link if wikidata entries are broken down this small...C933103 (talk) 08:03, 14 September 2018 (UTC)
Why so? Most of the interwiki links would be on the franchise item. If Wikipedia articles happen to exist dedicated to a movie or an OAV, then they can be linked to the Wikidata item. This is exactly what happens for Ghost in the Shell (Q92580):
That way, we have a 'proper' data model and reasonable interlinking, no? Jean-Fred (talk) 10:10, 14 September 2018 (UTC)
It would be good to fix these at some point. I tend to avoid them. In general, statements about any of the elements end up combined on some item. Maybe the situation has gotten better recently. --- Jura 10:20, 14 September 2018 (UTC)
If Wikipedia language version A create only an article for a novel version of a series, then Wikipedia language version B create only an article for comic version of a series, and then Wikipedia language version C create both articles then user would not be able to jump between language version A and language version C article on the same series directly. Unlike the current situation that all of them would be linked to same wikidata entry for the series and then only link to individual works when such wikipedia article exists.C933103 (talk) 03:06, 15 September 2018 (UTC)
@C933103: If there are article groups which have this problem, then for each item new placeholder pages should be created on each wiki, added as sitelinks to the items, and then turned into redirects. The alternative is that the concepts are conflated into one item, which is convenient from the WP perspective but incorrect from the WD perspective. Jc86035 (talk) 14:21, 15 September 2018 (UTC)
@C933103: That’s a valid concern, but I think in practice the pattern of linking together the main Franchise item will probably work in many cases, in particular the most complex ones. No Wikipedia links to Naruto (Q26971382) or Naruto (Q26971391), because it is likely that no Wikipedia will ever want to create a separate article for manga and anime.
Where I agree with you and am a bit worried about are cases where « what is the "main" medium » is debatable. For example, the manga Miss Hokusai (Q29940472) vs. its movie adaptation Sarusuberi (Q11580762). French Wikipedia is the only one having an article for both ; and for some language versions is is unclear which one they 'mainly' or 'primarily' talk about. A third item about a franchise would feel like overkill to me.
Jean-Fred (talk) 10:16, 18 September 2018 (UTC)

P17 country and disputed claims[edit]

Just a heads up that there is a discussion currently taking place on Commons about how best to handle country (P17) claims on places located in places that have disputed country claims, specifically as applied to Judea and Samaria Area (Q513200), which may or may not be part of Israel (Q801). This affects the display of infoboxes on that project. Bovlb (talk) 00:13, 14 September 2018 (UTC)

In the case of Crimean Peninsula (Q7835), it has been added to both Russia and Ukraine. Can the same be done for Judea and Samaria Area (Q513200)? The additional country would be State of Palestine (Q219060) or via Palestinian territories (Q407199). Ghouston (talk) 00:49, 14 September 2018 (UTC)
Ghouston Thanks for the suggestion. That is a reasonable way to handle multiple claims of ownership over a region. This specific discussion was about how to handle P17 claims on places (e.g. cities) within disputed areas. There are two obvious options:
  1. Add the seven dispute qualifiers onto the P17 claim for every city. If the qualification ever changes, we must maintain the information in hundreds of places.
  2. Remove the P17 claims on cities within disputed regions.
In addition, we need to consider whether project infoboxes can and should check for these and other qualifications (e.g. end time) whenever they try to indicate the country of an item. Cheers, Bovlb (talk) 15:50, 14 September 2018 (UTC)
No, it cannot be treated like the Crimea situation (In Crimea, 2 countries actually both claim the same land), as Israel has never formally claimed the West Bank as Israeli territory (except for a part of the Jerusalem municipality). Take a look at what started this, namely Category:Shim'a, then see how that is handled on en.wp: for region there it is stated "West Bank" (and not "Israel"). --Huldra (talk) 21:33, 14 September 2018 (UTC)
In that case, the country (P17) can just be set to State of Palestine (Q219060) or whatever, with optional dispute qualifiers. It's similar to Guantanamo Bay Naval Base (Q762570) having country (P17) of Cuba (Q241). Locations within the region should have the same country (P17) as the region itself. Ghouston (talk) 21:43, 14 September 2018 (UTC)
It's not that simple. Israel does administer the territory, various cities and settlements within are treated (for most intents and purposes) as Israeli. The State of Palestine's claim to the land (as well as the recognition of this entity) is in dispute. While Israel hasn't annexed the land, it hasn't said it does not belong to Israel - as this is former Mandate territory the legalities are rather murky, and Israel has been pursuing an agenda of ambiguity (as it does in other areas of policy).Icewhiz (talk) 21:51, 14 September 2018 (UTC)
I don't see the problem. If Israel doesn't claim it, then country (P17) shouldn't be set to Israel. If the State of Palestine does claim it, but it's in dispute, just add a disputed qualifier. Ghouston (talk) 23:48, 14 September 2018 (UTC)
Israel does not claim that it is part of Israel *now*, but it (along with most of the world) does claim that the status of the territories is to be determined by negotiations. It is assumed by many that most (or all) of the territories will end up being part of SoP, but this is not a fact. WarKosign (talk) 05:41, 15 September 2018 (UTC)
Not correct. All of the world, except Israel, consider these territories as occupied. Israel, and only Israel, consider them to be "disputed".--Huldra (talk) 20:03, 15 September 2018 (UTC)
I did not say anything about occupation. Whatever the status is now, the final status of the territories is yet to be determined. The world calls these "Palestinian Territories", not "State of Palestine", because even SoP says that it is prepared to negotiate and thus would probably end up with some subset of the territories (and maybe some extras as a result of territory exchange). WarKosign (talk) 21:22, 15 September 2018 (UTC)
Well, I think the infobox should reflect todays status, and not what some people dream about becoming the status tomorrow. --Huldra (talk) 22:24, 15 September 2018 (UTC)
It comes down to the details of its current status and also how country (P17) is used. We have non-administering countries in country (P17) on Guantanamo Bay Naval Base (Q762570) and Crimean Peninsula (Q7835). According to en:State of Palestine, it's recognized by 136 UN member states, so it has a lot of recognition. On the other hand, we also have a lot of items with country (P17) set to Taiwan (Q865), without bothering to mention that Taiwan (Q865) has much less recognition, and People's Republic of China (Q148) also claims the entire area. Ghouston (talk) 23:59, 15 September 2018 (UTC)
Recognition of SoP and of SoP's claim on the land are two separate issues. SoP is somewhat recognized, it owning the land is merely "what some people dream about". In fact while Israel doesn't make an official claim for the territories (except East Jerusalem), it has de-facto control of all of them (allowing civilian Palestinian control at some places), and has de-facto claim on the territory of the settlements. WarKosign (talk) 07:28, 16 September 2018 (UTC)
Again, this cannot be compared with the Crimea, or the Taiwan cases. In both of those case you have two countries officially claiming a piece of land belonging to them. This is not the case with the West Bank: Israel has never claimed the West Bank as part of Israel (except for a part of the Jerusalem municipality.) There is a good reason for this: if they did so, they either have to give the local Palestinians equal rights, (which will never happen, IMO,) or they would be officially instituting an apartheid system (unlike now, where they have unofficially instituted such a system), (There is a third way, of course, which is to make life as unliveable as possible for the Palestinians; forcing them to emigrate...but that is a long term strategy, and one which doesn't seem to be working very well,) --Huldra (talk) 20:37, 16 September 2018 (UTC)
The questions seem to be, for Judea and Samaria Area (Q513200), a) whether it makes any sense to set country (P17) to Israel, as User:WarKosign seems to want, when Isreal hasn't annexed the area but is occupying it and is the de facto government, and b) whether it makes any sense to set country (P17) to State of Palestine, when that state has never controlled that area and it's still unknown if they ever will. If neither makes sense, then I suppose country (P17) would be set to <no value>. Ghouston (talk) 21:56, 16 September 2018 (UTC)
Setting it to <no value> is definitely an improvement of present situation (..which is setting it to Israel, which is clearly wrong), --Huldra (talk) 22:28, 16 September 2018 (UTC)
Both Israel and SoP are associated with the territory. I'm not familiar with the technical side, if country (P17) attribute is exclusive it would be equally incorrect to set it to either. Both Israel and SoP claim and control the area to some degree; Israel claims less and controls more than SoP. If country (P17) indicates "belongs to" that neither would be correct. If it means "controlled by" then Israel would be correct (except areas A and maybe B); if it means "claimed by" then SoP would be correct (except East Jerusalem). If this tag has no intrinsic meaning and is used to indicate that there is some kind of association between land and a state (or a state-like entity), then it would be most correct to apply both and add some note explaining the complex situation. WarKosign (talk) 09:50, 17 September 2018 (UTC)
We have statement disputed by (P1310) for such a note, see Q39760#P17 --Pasleim (talk) 10:00, 17 September 2018 (UTC)
I think territory claimed by (P1336) describes SoP's relations to the territory accurately. For Israel, maybe located in the administrative territorial entity (P131) ? country (P17) seems wrong for either, because neither is "sovereign state of" Judea and Samaria Area (Q513200). WarKosign (talk) 10:11, 17 September 2018 (UTC)
No it doesn' Israel does NOT oficially claim the West Bank, --Huldra (talk) 20:39, 17 September 2018 (UTC)
I'm not sure which part of the above you're disputing. Please clarify? --Yair rand (talk) 20:45, 17 September 2018 (UTC)
The "claimed by" statement. In the above statement one have Senkaku Islands as an example. Now, that is clearcut: AFAIK both Japan and China claim them. Alas, Israel does not claim (officially) the West Bank, it administer certain parts, where they have criminally, according to international law, transferred some of their own population. --Huldra (talk) 22:45, 17 September 2018 (UTC)
So are you saing that SoP *does not* claim the territory because Israel doesn't claim it ? WarKosign (talk) 08:33, 18 September 2018 (UTC)
Huh?? How on earth do you come to that conclusion? --Huldra (talk) 20:27, 18 September 2018 (UTC)
I suggested to write that the area is claimed by SoP and apparently you argue against that. WarKosign (talk) 21:20, 18 September 2018 (UTC)
Plenty of countries have administrative territorial entities for areas they don't actually administer, so individual cities and whatnot can be simultaneously within administrative districts of different countries, only one of which is of the administration. However, a district that structurally only exists within one country's system (regardless of the geographic area it represents) clearly has more association to one country than the other. I think it would go in its P131 tree and probably also P17. I think we'd agree that even Taiwan's old districts for mainland areas it hasn't controlled for a long time would still have P17: RoC. --Yair rand (talk) 20:54, 17 September 2018 (UTC)
A relevant point here is that Republic of Crimea (Q15966495) only has country: Russia and Autonomous Republic of Crimea (Q756294) only has country: Ukraine.
There are a lot of issues to unpack here, and the decisions for each of them will affect dozens of items in similar situations. We need to figure these out systematically, and not try to just figure out an outcome for a single item. --Yair rand (talk) 21:02, 17 September 2018 (UTC)
Related: Wikidata:Project_chat/Archive/2017/07#What_country_are_Israeli_Jewish_settlements_in?. Anyone know how occupations (without territorial claims) are handled in other situations? The Occupied Enemy Territory Administration (Q7075820) seems to be considered as though it were a country in its own right, as is Allied-occupied Germany (Q2415901) (and not as, say, territories of the respective occupying powers), historical American unannexed territories I think are considered American, Guam (Q16635) has country (P17) United States of America (Q30) despite being officially specifically not part of the US. What do we do with United States Civil Administration of the Ryukyu Islands (Q55526)? We don't seem to be completely consistent about these things. Is the deciding factor which power administers the area? Which claims it? Which is recognized to have it? It's complicated, but I think we should try for as much consistency as possible. --Yair rand (talk) 22:54, 16 September 2018 (UTC)

I would think the correct data handling of Israel's relation to the West Bank involves using occupied territory (Q2577883) as a qualifier on something, though I don't know what. Thoughts? - Jmabel (talk) 23:50, 17 September 2018 (UTC)

Yes, good idea, I think that could cover it. --Huldra (talk) 20:27, 18 September 2018 (UTC)

I'm not sure that we have solved every aspect of this issue, but I believe the originally-reported problem is now resolved. Bovlb (talk) 02:20, 21 September 2018 (UTC)

P values to use for gene structure[edit]

Gene structure eukaryote 2 annotated.svg

Hello all. I'd like to try to encode the hierarchical organisation of gene components as summarised in the diagram (right). What are the most logical P properties to use? Is there something more specific than "is part of" / "is instance of", e.g. for the relationship between promotor and regulatory sequence? Evolution and evolvability (talk) 02:53, 15 September 2018 (UTC)

@Evolution and evolvability: The relationship between a regulatory sequence and a promoter in it can be described using part of (P361) and has part (P527). But I think there is currently no way to link between a gene (which has genomic start (P644) and genomic end (P645)) and its regulatory sequences. In addition, genomic start (P644) and genomic end (P645) of genes are currently not only including open reading frames, but also including untranslated regions. So we may have to create some new properties. --Okkn (talk) 06:08, 16 September 2018 (UTC)

Two new principles introduced by Andy Mabbett[edit]

At Q27652812 there is another instance of the by now customary edit war by Andy Mabbett, in which he reverts time after time, without providing any kind of argument for the "content" he added, or answering any argument about content. In this instance there is an additional twist; he introduces two new principles into Wikidata:

  • As an excuse for editing he postulated that if a concept can possibly be regarded as a lexeme this overrides the basic Wikidata model of having an item when there is a sitelink.
  • He wants to have an item devoted to the state of being new. That is, a separate item, for an already existing item, devoted to the moment when the concept was new. This is a revolution within Wikidata. Anything man-made is "new" at some point, any book, movie, car, plane or what have you. By Andy Mabbett's reasoning there should be items "new book", "new movie", " new car", "new plane", etc (for most existing items). Presumably to be followed by items "one-year-old book", "one-year-old movie", "one-year-old car", "one-year-old plane", etc. In turn to be followed by "two-year-old book", "two-year-old movie", "two-year-old car", "two-year-old plane", etc.

Brya (talk) 04:20, 15 September 2018 (UTC)

First, please Brya, notify the relevant user. It is rude to attack somebody in public without letting them know.
Second, if the scope of an item is unclear, it helps to look how the item is used on the project. species nova (Q27652812) seems mainly be used as a value of main subject (P921) on scholarly articles. Is the main subject of these articles a latin phrase or a species? --Pasleim (talk) 08:30, 15 September 2018 (UTC)
As to the first, he is all over the place anyway.
        As to how it is used on Wikidata, this is not saying much as these links were made by a bot, a bot, QuickStatementsBot, which is notorious for the many errors it makes. Putting a bot to work will run up numbers quickly, no matter the "value" of the content. These links are weird: if there is a paper on "The new model Volkswagen Tiguan", the main subject is not "new model car" but the Volkswagen Tiguan. Not to mention those minerals that supposedly are "new species of organisms". - Brya (talk) 09:44, 15 September 2018 (UTC)
Hey Daniel: I think most of the additions are made by you.
Hey User:Chris.urs-o: I think most of the (misplaced) mineralogical usages of species nova (Q27652812) as main subject (P921) are done by you.
--Succu (talk) 19:19, 15 September 2018 (UTC)
I agree that species nova (Q27652812) should not be about the newness/ novelty (which is ephemeral) but about the taxonomic term that has a specific meaning in biological nomenclature. My reasoning behind using it in main subject (P921) statements is that this basically flags the existence of at least one taxon treatment (Q32945461) in and/or at least one nomenclatural act (Q56027914) associated with the publication in a more consistent manner than the various variants of "new species"/ "sp. nov." / "spp. nov." etc. On that basis, it is then simpler, for example, to create lists of works which do not have ZooBank ID for name or act (P1746) or Plazi ID (P1992) statements but probably should have some. As for "new species" of minerals, I think that should be a separate item. --Daniel Mietchen (talk) 22:14, 15 September 2018 (UTC)
I need help. I need a list of haswbstatement:P921=Q83353 AND haswbstatement:P921=Q27652812. "New species" of minerals: type description (earth sciences) (Q56241591).
Type specimen, type locality and type description are valid concepts. --Chris.urs-o (talk) 03:15, 16 September 2018 (UTC)
@Chris.urs-o: Something like this query? --Succu (talk) 15:17, 16 September 2018 (UTC)
Thank you. --Chris.urs-o (talk) 15:45, 16 September 2018 (UTC)
As to papers that introduce a new species of organism, these should be linked to the new species they describe (for example Cherax snowden). Then, preferably the item of the species should be linked to the paper. I guess this might be done by bot, but probably not easily. - Brya (talk) 05:55, 16 September 2018 (UTC)
Or it could be a statement instead of a qualifier. - Brya (talk) 07:38, 16 September 2018 (UTC)

Looks like that Mr. Mabbett is avoiding this place. No reason givin in this and other matters.--Succu (talk) 05:39, 19 September 2018 (UTC)

I'm not "avoiding" this place at all. The fact that I am travelling (and will be for a further two weeks) - to speak about Wikidata at conferences and seminars - is widely known and has been previously (and at the time of writing is still) reported on this very page. Note also Brya's refusal to ping me when discussing me (a failure about which I have complained previously) and refuses deliberately to do so even when reminded to do so by other editors. Brya's opening post in this section contains a number of falsehoods in which - like your "avoiding" and "no reason given" allegations - he ascribes to me actions and beliefs which are not mine. I do not intend to dignify such abuse with a response, but I do note that Pasleim eloquently addressed the core issue. Absent a cogent rebuttal of that point, there is no need for further comment, and your tag-team edit warring to try to make the item about something which it is not (just as table (Q14748) is not about "a term used to describe an item of furniture with a flat top and one or more legs") is increasingly disruptive. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 07:38, 20 September 2018 (UTC)
I hope you do not encourage new users to use as before or as previously as an edit comment, Mr. Mabbett. --Succu (talk) 21:45, 20 September 2018 (UTC)
Species novum is an attribute of literature. It is where a species is introduced for the first time. It is definitely not an item in its own right. Where Andy is right is that it needs its own place however having it in a separate item is wrong. In the same way, species, subspecies, genus et al change over time. They change in literature. Almost by definition, what is meant is different with the new documentation. For this reason it is not realistic to have our current taxons exactly identify the described species novum and what follows. This is particularly clear because the full name of a taxon includes not only the author but also the publication.
As for the continuous bickering, I have largely stopped being involved by the bullying behaviour that is oh so easy to observe. Thanks, GerardM (talk) 05:34, 21 September 2018 (UTC)

Achim Raschka (talk)
Brya (talk)
Dan Koehl (talk)
Daniel Mietchen (talk)
FelixReimann (talk)
Infovarius (talk)
Joel Sachs
Josve05a (talk)
Klortho (talk)
Lymantria (talk)
Mellis (talk)
Michael Goodyear
Mr. Fulano (talk)
Nis Jørgensen
Peter Coxhead
Andy Mabbett (talk)
Prot D
Rod Page
Soulkeeper (talk)
Strobilomyces (talk)
Tommy Kronkvist (talk)
Pictogram voting comment.svg Notified participants of WikiProject Taxonomy

Noobie Questions[edit]

Now I'm curious about this WikiData thing. I found something needing merging, then came back about "The Dog" ( ). Seeing that no one had touched it I looked into it. The information does not overlap. Then I wondering about why I'm unable to add more details to that article.

For example, The Dog has two co-directors and I can't even add one, much less worry about duality.

I find it strange that I can add "terrorism" as one of the main subject qualifiers but I can't add "theft" "robbery" "bank" etc. (I pulled up Iron Man (2008 film) to use as a template/reference for adding details.)

Turns out I can still add a lot in "depicts" and "main subject". Is there a difference? I added much of the same to both.

I stumbled upon a glitch: Adding the film premiere info (URL, title, date, retrieved date, author), all was fine until I tried to add unknown authors, then it would freeze and no longer "publish" - even when I removed the 'author' part. The first time I'd filled out all the things without publish-saving. The second time I'd filled out each item and published one at a time - until I had to refresh Firefox. Similarly, I discovered you have to add the co-directors names and some details before you can even reference them. Empty non-links are not allowed yet people stubs are. Weird.

I also wanted to add the link to my original article (almost identical because I wrote them at the same time - but IG has an image that was removed on WP (who censor everything)).

I have no dog in maintaining "The Dog" or it's unimportant article, but I thought it'd be a good enough place to learn about WikiData.

I kinda see the potential but there's sooooooo far to go - and then there's the faaaaaaar bigger problem of the corporatocracy censorship, distortion of events, evasive truth, and intolerance of anti-establishment ideas and authentic freedom. Any A.I. that may use this is already crippled by all the corrupted systems, including legacy media the limited and limiting so-called source of "facts" allowed on Wikipedia, etc.

Suggestions for good tutorial videos or a short overview would be greatly appreciated. ~ JasonCarswell (talk)


Can anyone clarify what the difference between cirque (Q184368) and cirque (Q388227) is supposed to be? One is supposedly a subclass of the other, but they have the same English label, they're both described as valleys caused by glaciers, they have the same Commons category and essentially the same GeoNames code (one is singular the other is plural), so I have no idea which one to use. - Nikki (talk) 09:54, 15 September 2018 (UTC)

Sorry, I do not remember. Chaoborus (talk) 12:14, 15 September 2018 (UTC)
Both have dewiki sitelinks, and the dewiki sitelink for cirque (Q388227) seems to describe it as a general geological shape, possibly but not necessarily caused by glaciers, with cirque (Q184368) being a special form of it (i. e. the reverse of the subclass of (P279) situation on Wikidata). --Galaktos (talk) 23:14, 15 September 2018 (UTC)

How to propose renaming a property?[edit]

Is there a way to propose renaming a property, or propose replacing one with another? If so, how? Rua (talk) 17:53, 15 September 2018 (UTC)

If the name of the property is incorrect, you can just change the label. If the you want to change the purpose of the label, start with a discussion on the talk page of the property. It might result in a PFD procedure and new property request. Mbch331 (talk) 18:25, 15 September 2018 (UTC)

Q25056785 and Q56537023 refer to the same real life object[edit]

  • Q25056785 ("F.A. Brockhaus AG" – without blank between "F." and "A.") consists only of three statements and two interwikis links. This item belongs to w:en:F.A. Brockhaus AG. There is said:
"Friedrich Arnold Brockhaus founded his publishing house in 1805 in Amsterdam, [...]."


  • Q56537023 ["F. A. Brockhaus (Verlag; ab 1817: Leipzig)" with blank between "F." and "A."] consists of few more information and belongs to GND 65191-6. There is said:
"1805 in Amsterdam unter Rohloff & Compagnie gegründet; 1811 Umsiedlung nach Altenburg in Thüringen; 1817 nach Leipzig".

Therefore I guess: This two item should be fused. - However, Q464633 (Bibliographisches Institut & F. A. Brockhaus) belongs to a different real object, because it came in existence not before 1984 - due to an fusion of two distinct companies. The latter item belongs to w:de:Bibliographisches Institut & F. A. Brockhaus. There we can read:

"Als Beginn des Verlages F. A. Brockhaus wird die Gründung des Verlages 'Rohloff & Co.' am 15. Oktober 1805 in Amsterdam durch den Kaufmann Friedrich Arnold Brockhaus angesehen, der zuvor mit einem Handel für Wollstoffe gescheitert war. Da Brockhaus kein Mitglied der örtlichen Buchhändlergilde war, erfolgte die Gründung über einen Strohmann, den Buchdrucker J. G. Rohloff."

--Villa loga-WB (talk) 18:08, 15 September 2018 (UTC)

Gee fizz[edit]

Can someone protect Rick and Morty (Q15659308). It's currently my entire watchlist. GMGtalk 00:20, 16 September 2018 (UTC)

@GreenMeansGo: it has been protected, but requests that need sysop actions are placed on the administrators' noticeboard. Esteban16 (talk) 02:35, 16 September 2018 (UTC)


Is there a property to use to mark a battle in a war, and in relation to fronts and other battles? The part of (P361) piles everything together, and does not structure the information very well. Or can conflict (P607) be used for battles in war? –Susanna Ånäs (Susannaanas) (talk) 11:22, 16 September 2018 (UTC)

What about significant event (P793) per significant or notable events associated with the subject. Pmt (talk) 11:51, 16 September 2018 (UTC)
Thanks, I think it's much better. I realise that this may be the smallest of problems, it seems that most everything about wars is hard to model to the level of Wikipedia infoboxes: Who were the fighting parties, who lead them, how many casualties and prisoners were there on each side in each battle etc. – Susanna Ånäs (Susannaanas) (talk) 11:59, 16 September 2018 (UTC)

Geographical coordinates from image metadata[edit]

If we add geographical coordinates to a listing based on the metadata of an image pertaining to that listing, would we add a source for this statement, and, if so, what kind? I originally asked in Help talk: Sources. ARR8 (talk) 14:39, 16 September 2018 (UTC)

  • Not an answer to your question, but from my experience at Commons, a lot of images have inaccurate geocoord metadata in their EXIFs. Before trusting that, always look at a map and see if it is at least highly plausible. This should never be done by a bot and simply trusted. - Jmabel (talk) 17:17, 16 September 2018 (UTC)
Thanks for the reminder. In this specific instance, I have done so. The specific use case here is landmark listings on Wikivoyage which are missing coordinates but have pictures in Commons, so this will be a manual and infrequent process. ARR8 (talk) 02:29, 17 September 2018 (UTC)

wikibase and authentication methods[edit]

Hello, I'm new with wikidata and I don't know if it's the right place to ask about technical question about wikibase. I wonder if it's possible to have several authentication methods (ex: LDAP, oAuth, SAML, OpenID ...) when installing and configuring a standalone wikibase instance?  – The preceding unsigned comment was added by 2a01:cb1d:89d9:d000:886:6d0f:7907:f52c (talk • contribs).

You probably want to talk to the developers - Wikibase Installation has some info, also the project page outside of wikimedia. I can't find anything directly about authentication, but I expect it's something that could be configured as you like. ArthurPSmith (talk) 18:27, 18 September 2018 (UTC)

deactiveing article preview[edit]

How can I hide article preview pop-up permanently?Yamaha5 (talk) 02:44, 17 September 2018 (UTC)

@Yamaha5: disable the “Preview” gadget? or do you still want to have the gadget, but without the preview opening by default? --Lucas Werkmeister (talk) 10:43, 17 September 2018 (UTC)
@Lucas Werkmeister: This preview artcile window should be optional or should be disabled by gadget or prefrencess.
This Window
Yamaha5 (talk) 13:38, 17 September 2018 (UTC)
As Lucas wrote, you need to disable the "Preview" gadget. You can do this at Special:Preferences#mw-prefsection-gadgets --Pasleim (talk) 14:41, 17 September 2018 (UTC)

Wikidata weekly summary #330[edit]

Is there a simple description of lexemes and how they might be used?[edit]

Hi all

I'm writing a proposal for a project that includes a Wikidata element, I'm aware that lexemes are going to be available soon. Is there a description of what lexemes are somewhere that could be understood by a non technical person, and also some information on how they might be used or any future additions to them?


--John Cummings (talk) 08:13, 18 September 2018 (UTC)

What little I understand of it is that basically Wikidata is a general structured ontology (i.e. it's about concepts), whereas lexeme are structured lexicographical data (i.e. it's about words). It may or may not be accurate (as I said, I'm not very knowledgeable about lexemes) to say that lexemes structure Wiktionary data instead of Wikipedia data. Circeus (talk) 14:04, 18 September 2018 (UTC)
Hello John,
Lexemes are available since May 2018. Did you already browse Wikidata:Lexicographical data and its subpages? If something is missing in the existing documentation, let us know. Lea Lacroix (WMDE) (talk) 14:44, 18 September 2018 (UTC)
Naively, a lexeme is a word; it's just that some things longer than one word function as lexemes (e.g. phrases that have a meanings little related to their components, as in Cockney rhyming slang; legal phrases like "sub judice" or "ipso facto"). - Jmabel (talk) 15:54, 18 September 2018 (UTC)

@John Cummings: Lexeme is label (lemma) + set of forms (inflexion) + meaning (senses). For example for "run" you have label "run" which represents set of forms ("run, runs, running" ) and meaning ("To move forward quickly upon two feet by alternately making a short jump off either foot" source en.wikt). If you have more lexemes oriented questions you can find more discussions about it at Wikidata talk:Lexicographical data. KaMan (talk) 09:35, 19 September 2018 (UTC)

British/Canadian English[edit]

Is there a database report of labels and descriptions which differ between en, en-gb (British) and en-ca (Canadian), possibly excluding normal spelling differences like color/colour? This isn't usually an issue but it gets quite annoying when property labels are changed in en and not updated for en-gb and en-ca (I use en-gb). It would also help to fix these for items (I think in general if a page move for a sitelinked page occurs then the label should be kept on a report until it's updated or checked). Jc86035 (talk) 12:03, 18 September 2018 (UTC)

#SELECT DISTINCT ?property ?en_label ?engb_label ?enca_label ?en_desc ?engb_desc ?enca_desc
    ?property rdf:type wikibase:Property .
    optional { ?property rdfs:label ?en_label. FILTER(LANG(?en_label) = "en") }
    optional { ?property rdfs:label ?engb_label. FILTER(LANG(?engb_label) = "en-gb") }
    optional { ?property rdfs:label ?enca_label. FILTER(LANG(?enca_label) = "en-ca") }  
    optional { ?property schema:description ?en_desc . FILTER(LANG(?en_desc) = "en") }
    optional { ?property schema:description ?engb_desc . FILTER(LANG(?engb_desc) = "en-gb") }
    optional { ?property schema:description ?enca_desc . FILTER(LANG(?enca_desc) = "en-ca") }
Try it!
As you can see, the majority of properties have no engb or enca label (which at least stops them from getting out of sync). There are only 120 where all labels/descriptions line up:
SELECT DISTINCT ?property ?en_label ?engb_label ?enca_label ?en_desc ?engb_desc ?enca_desc
    ?property rdf:type wikibase:Property .
    optional { ?property rdfs:label ?en_label. FILTER(LANG(?en_label) = "en") }
    optional { ?property rdfs:label ?engb_label. FILTER(LANG(?engb_label) = "en-gb") }
    optional { ?property rdfs:label ?enca_label. FILTER(LANG(?enca_label) = "en-ca") }  
    optional { ?property schema:description ?en_desc . FILTER(LANG(?en_desc) = "en") }
    optional { ?property schema:description ?engb_desc . FILTER(LANG(?engb_desc) = "en-gb") }
    optional { ?property schema:description ?enca_desc . FILTER(LANG(?enca_desc) = "en-ca") }
    filter(str(?en_label) = str(?engb_label)). filter(str(?en_label) = str(?enca_label)).
    filter(str(?en_desc) = str(?engb_desc)). filter(str(?en_desc) = str(?enca_desc)).
Try it!
Personally I feel we should just deprecate en-gb and en-ca - the incredibly small number of cases where they're useful don't seem to make up for the hassle of chasing discrepancies and synchronising labels, and it is also a little weird that it's implicitly defining "en" to be en-us. But that's another issue, and probably something we should have thought about more five years ago. Andrew Gray (talk) 19:49, 18 September 2018 (UTC)
I'd assume plain "en" can use any widely-used variant spelling, and that en-gb would only be set in cases where there's a specific variant to be used. If there's no en-us, that's surprising. Ghouston (talk) 00:55, 19 September 2018 (UTC)
Yeah, it's not an ideal situation. In general the en versions mostly (but not universally) use en-us spellings for cases where the two differ. Thankfully Wikidata seems to be mostly free of edit wars about the correct English labels for items - we've avoided that traditional enwiki pastime! Maybe it's because we can't make snippy comments in the edit summaries :-)
We retained en-gb and en-ca after a fairly short discussion in May 2013 (Wikidata:Requests for comment/Labels and descriptions in language variants). The UI issues I mentioned then have improved a lot over the past five years; back then it was a real mess because you'd just get Q-numbers and P-numbers in links everwhere rather than falling back to en. So it's no longer as much of a problem as it used to be, just a bit of a maintenance hassle.
Going back to the original question, for items rather than properties, see User:Pasleim/Language statistics for items, which identifies about 75% of items as having an en label, but only ~2.5% with en-ca or en-gb. Numbers are a bit better for descriptions where a little over 30% of en-ca or en-gb have them, but I wonder how many of those are bot-filled boilerplate like "disambiguation page", "journal article", etc. Aliases for items are rare in both (10-20k items with aliases as against a few million for en)
By comparison, simple has been aggressively cleaned of localised terms, with virtually no items having labels, descriptions, or aliases. I guess the idea here is that the names of things are unlikely to be simplified so we won't need to have diferent labels? Andrew Gray (talk) 17:11, 19 September 2018 (UTC)
It seems silly to set the en_gb and en_ca labels and description to the same values as en: better to just leave them blank. Why should bots do that? If I change the en values, I typically won't notice if en_gb and en_ca differ, because they aren't shown by default, just the two languages I've set in a babel box. Ghouston (talk) 21:59, 19 September 2018 (UTC)
I definitely agree with this part. Copying en labels/descriptions to en_gb and en_ca seems like a sure-fire way to get outdated labels/descriptions there in the future. Just let the language fallback system do its job, IMHO. —Galaktos (talk) 10:18, 20 September 2018 (UTC)

How to enter this property?[edit]

SO I'm working with herbaria (because of the newly approved Index Herbariorum code (P5858)) and at Oakes Ames Orchid Herbarium (Q55828891) I am trying to enter the collection or exhibition size (P1436).

Entering a number? It's a no go. The property pretends the "units" subproperty/qualifier is optional, but clearly it is not. In fact, it won't save on older entries that have no unit if you attempt editing the property (e.g. Maksim Bahdanovič Literary Museum (Q12337378)). Alright, so I try to enter biological specimen (Q2075980) (I mean, what else am I supposed to use?!)... but it still won't save.

What am I doing wrong? Can I adjust the property so this nonsense will stop without manually adding exceptions for every collection in the world? Circeus (talk) 13:02, 18 September 2018 (UTC)

@Circeus: This should work with or without a unit, see the examples on collection or exhibition size (P1436). What exactly is the error message you are seeing? "won't save" suggests maybe there's something else wrong - if it's a constraint violation it will still let you save but then show a warning flag. ArthurPSmith (talk) 18:30, 18 September 2018 (UTC)
GAH. Turns out it was far more stupid than I thought: I pasted the number as found on IH with the space and all, and what goes on is that numbers (unlike dates) don't have any wiggle room whatsoever when you enter them. Circeus (talk) 19:40, 18 September 2018 (UTC)

Distinction of title of authority (Q7810129) and position (Q4164871)[edit]

Despite those items are said to be distinct, I don't see when to use which. There are also some Item that claim to be Instance of both (list). Can someone help me out? -- Dr.üsenfieber (talk) 13:07, 18 September 2018 (UTC)

@Dr.üsenfieber: title of authority (Q7810129) refers to how a person would be addressed - for example "Professor X", "Dr. Y", "President Z" etc.. position (Q4164871) is the label for the job they are in, whether or not they would be addressed that way, and it is usually much more specific, for example "President of Singapore", etc. ArthurPSmith (talk) 18:35, 18 September 2018 (UTC)

What do you think is the main benefit of Mix'n'match?[edit]

Hi, I'm preparing for a presentation on Mix'n'match at Wikiconference North America. I have a little experience with Mix'n'match, but I know that I don't know very much compared to what everyone else knows. If you are going to Wikiconference North America and would like to collaborate on the presentation, please e-mail me or leave a talk page message. If you're not going to the conference but have knowledge about it, you can tell me here and I'll incorporate it into my presentation.

On Mix'n'match, I've imported a database there and I've done some matching. As I see it, right now links to external identifiers like those found in library databases are nice to have, but I don't think other sites use them (I'm aware that SNAC pulls from Wikidata, but I don't think it picks up non-authority file links). I think of it like investing in Wikidata without knowing what the payoff will be. Are there sites that currently use links from Wikidata like those matched through Mix'n'match?

I know Magnus created Mix'n'match as a way to create redlink lists--is the Creation candidates part of the site currently used for making new Wikidata items? I haven't heard of many people using it, but what do you think? From Wikidata, the Listeriabot can make redlink lists, like those for the Women in Red project--are there other projects using Wikidata to make redlink lists?

Are there other things about Mix'n'match that you think other Wikipedians and library people should know?

Thanks, Rachel Helps (BYU) (talk) 16:00, 18 September 2018 (UTC)

  • Wikidata:BEIC used mix'n'match to identify missing authors and then import/create items for them. --Nemo 17:28, 18 September 2018 (UTC)
  • Oh, thanks, I'll definitely add that to my presentation. Rachel Helps (BYU) (talk) 18:21, 18 September 2018 (UTC)
The biggest benefit is to find external id's for people, but more recently I have been using it for paintings. It's great to match up Qids for datasets that connect to some external database already (like any of the catalogs in there already, you can easily upload your own dataset, but sometimes you just want the Qids). Jane023 (talk) 18:48, 18 September 2018 (UTC)
Hmm, when would you just want the Qids? Rachel Helps (BYU) (talk) 19:45, 18 September 2018 (UTC)
I always explain it like this to Wikimedians: In Wikipedia you get the basic information about a topic and links to further literature. In Wikidata it's the same thing. You can basic data about a topic and then humans and machines get links to other databases/catalogs/etc to find out more. One of the things Wikidata is good at is being this identifier hub that brings all these different identifiers together. When you have one identifier you can easily find all the others be it as a human in the website or as a machine via the API. --Lydia Pintscher (WMDE) (talk) 18:52, 18 September 2018 (UTC)
okay, so mix'n'match links are useful because the page they link to could provide more information about a subject, which can then be used to improve Wikidata/Wikipedia? Rachel Helps (BYU) (talk) 19:13, 18 September 2018 (UTC)
Yes that is one really important aspect. The other part is what Jane was referring to I believe: When you are building a third party app or are matching catalogs then Wikidata can be very helpful by being the glue between many catalogs. If you have a list of IDs from the German National Library and want say the equivalent IDs in the French National Library. The German National Library might not have the link but you can go to Wikidata to find it because they are all collected in the same item here. (Made up example! They probably link to each other but there are enough combinations where this connection is useful but doesn't exist.) --Lydia Pintscher (WMDE) (talk) 19:21, 18 September 2018 (UTC)
Oh, I can see how that would be useful. Do you know of third party apps or websites that are currently using Wikidata for this kind of thing? Rachel Helps (BYU) (talk) 19:45, 18 September 2018 (UTC)
  • More specifically about "Mix'n'Match": it's mainly a tool to avoid creating duplicates. Let's say you have some identifier about 100 martians that are notable and should be added to Wikidata, you can create the items directly as you know nobody else created them yet. However, if you want add some identifier about 100 British PMs, you should avoid creating new items as they mostly likely already have items. The identifier should be added to the existing item. Many datasets are somewhere between these two.
    There are few other tools to achieve the same. --- Jura 19:24, 18 September 2018 (UTC)
oh, definitely! It's such a basic convenience of Mix'n'match that I didn't even think of it. Rachel Helps (BYU) (talk) 19:45, 18 September 2018 (UTC)

Prefix wdtn: not working for quantities?[edit]

The following query should retrieve the normalized elevation of mountains, right?

select * where {
  ?mountain wdt:P31 wd:Q8502 .
  ?mountain wdtn:P2044 ?elevation .
} limit 3

Try it!

However, there are no results. Changing wdtn:P2044 to wdt:P2044 gives results, but then ?elevation is not guaranteed to be in meters.

The list of prefixes lists wdtn: (and further above explains normalization of quantities) but I can't find any example of this prefix being used. Toni 001 (talk) 14:07, 19 September 2018 (UTC)

Seems so indeed, but no idea why (I never used this prefix). Here is a workaround:
  BIND(wd:Q59805 AS ?mountain) .
  ?mountain p:P2044 ?statement .
  ?statement psv:P2044/wikibase:quantityAmount ?elevation .
  ?statement psn:P2044/wikibase:quantityAmount ?elevationNormalized .
Try it! MisterSynergy (talk) 14:18, 19 September 2018 (UTC)
@Toni 001, MisterSynergy: The wdtn: prefix is (currently) only used for normalized external identifiers (linking to the full URI according to formatter URI for RDF resource (P1921) – see the announcement), not for normalized quantities. --Lucas Werkmeister (talk) 15:00, 19 September 2018 (UTC)

Can properties contain a pair of values?[edit]

I don't mean a property that has more than one value, but one where the value itself is a pair. I'm dealing with URLs in an external database like:

I want to turn the link to these into a property, like the many other ID properties that already exist, but as you can see the URL has two parts, both of which are needed to identify a given entry. The first identifies the dictionary, the second the entry within that dictionary. Is there a way to capture this in a single value, in a way that the software can still piece together the link? Since the number of dictionaries is limited, these could also in theory all become separate properties; then the wdb= part of the URL is hardcoded into the property and only the id= is needed. What do others think? —Rua (mew) 16:03, 19 September 2018 (UTC)

There can only be one value, so you would need separate properties. I think separate properties would make sense anyway if they're separate dictionaries. - Nikki (talk) 16:56, 19 September 2018 (UTC)
They're separate, but they are stored in a common database with a common URL format. So kind of a grey area? —Rua (mew) 17:32, 19 September 2018 (UTC)
So you could just store the id in the form 'wdb=ONW&id=ID1000' and use the single property and URL, that works though it doesn't look pretty. If there's a standard form format for the pair it might be possible also to use that, for example ONW:ID1000 or ONW/ID1000 (if it is displayed in that way on the site for example, that might be a good standard to follow to really call it an id). We do have some properties like that where we go through a third-party reformatter (on the wmflabs tools server) to fix the url's. So it's possible to do this as a single property, it depends on what people agree looks right here. ArthurPSmith (talk) 18:03, 19 September 2018 (UTC)
I just went with multiple properties. See Wikidata:Property proposal/GTB database IDs. —Rua (mew) 18:16, 19 September 2018 (UTC)
But conceptually they're different, even if they're part of a single website. There are a few cases where we still have a single property for multiple types of identifiers (IMDb being the main one), but experience showed that it's generally a bad idea (e.g. it often leads to problems with constraints or formatter URLs, it makes using the data more complicated), so now we tend to avoid doing that and have split most of the old combined ones. - Nikki (talk) 09:55, 20 September 2018 (UTC)

Property creation for structured data on Commons[edit]

Hello all,

As you may now, the project to connect Commons and Wikidata in order to provide structured data for Commons is moving forward. While some heavy technical changes are made, both communities are discussing about the best way to organize the data in the future. You can follow the state of the project and discussions here. The team expects the first deployments of statements on Commons (e.g. “depicts”) in the first months of 2019.

I wanted to share with you a few suggestions in order to move forward on the Wikidata side and be ready to welcome these changes.

  • For now, most of the discussions take place on Commons, which make senses, since the changes will have a big impact on them, but what about having also a place to discuss on Wikidata? Should we use the existing Wikidata:WikiProject Commons, create a dedicated one, or just keep discussing on the Project Chat?
  • In the next months, some new property proposals will be made by editors working on this project. In order to make them more visible, what do you think about creating a special section, just like it was done for Lexemes?
  • In general, do you have any questions about the project, any concern you would like to raise? The team working on SDoC, and especially the Community Liaison Keegan (WMF) will be happy to help.

Thanks, Lea Lacroix (WMDE) (talk) 07:22, 20 September 2018 (UTC)

  • Maybe the following points should be born in mind:
  1. Commons will introduce a new entity type (M-entities for media)
  2. the plan seems to be that properties are only created on Wikidata even if they are used mainly on Commons
  3. items used as values would be created on Wikidata (an exception might be some individual media creators such as photographers)
  4. Commons may have different usage rules and constraints for properties compared to Wikidata
  5. Commons might end up with a complicated use of "depicts"-statements
  6. Commons might generate some statements directly from image metadata
  7. it's possible to limit use of properties to Commons only (just as some properties only apply to L-entities or P-entities).
Hope I got that right. Personally I think mainly #4 could lead to confusion. --- Jura 08:01, 20 September 2018 (UTC)
@Lea Lacroix (WMDE): Will this have any impact on existing Wikidata properties that have Commons media files as values (for example image (P18) which is "Commons media file" data type)? ArthurPSmith (talk) 14:39, 20 September 2018 (UTC)

Hoe verdwenen straten in Haarlem op te voeren[edit]

The Haarlem Erfgoed & Data WikiProject does not exist. Please correct the name.

Hoe kun je een verdwenen straat in Haarlem opvoeren? De huidige link werkt via een bot en verwijdert handmatige toevoegingen.

Zo heb ik de Achterstraat in de Burgwalbuurt gevonden. In Kurtz/Temminck 1999 via lemma's achterhaald. Per 14 juni 1876 verdeeld: hernoemd tot Spaarnwouderstraat (van Hagestraat tot Korte Brug) en Antoniestraat (van Hagestraat tot Antoniebrug)  – The preceding unsigned comment was added by Arjan Vink (talk • contribs).

I'm not sure what you are saying about bots and links, does this query belong on the Dutch Wikipedia? It's possible to make a separate item on Wikidata for an old street if it was subdivided into two new streets. In this case, Achterstraat divided into (Spaarnwouderstraat (Q17286899) and Antoniestraat (Q17286360) on 14 June 1876. In that case, the inception date of the last two streets would be 14 June 1876, and Achterstraat would be disestablished on the same date. Ghouston (talk) 11:54, 20 September 2018 (UTC)
Some context: Members of Historische Vereniging Haerlem (Q3238760) are active on Wikidata working on current and historic streets. See for some background information (in Dutch).
@Arjan Vink, Hannolans, Ghouston: the suggestion is also how I would probably do it. Multichill (talk) 15:42, 21 September 2018 (UTC)

Trying to edit two entries[edit]

I've been trying, unsuccessfully to edit two items:

Isabel's husband does not have an article or item. His name was Martín López de Henestrosa y Córdoba, not Martín López de Córdoba (Q5407465) Martín (Q5407465) married 2x: First wife was Sancha Alfonso Carrillo and his seconds wife was Teresa Álvarez de Haro. I tried to fix both Isabel and Martín but was unable to do so. I could add references when both have been fixed.--Maragm (talk) 13:42, 20 September 2018 (UTC)

@Maragm: unsourced and contested statements removed. strakhov (talk) 14:23, 20 September 2018 (UTC)
@Strakhov:, thanks. Will edit articles when I have time.--Maragm (talk) 14:46, 20 September 2018 (UTC)

new items in one window are not available in another[edit]

At this time the responsiveness of the system is so bad that you cannot work on lists where you have to create items that are to be used in the next. This is highly demotivating. The same is true with adding publications to authors.. The new tool by Magnus does not get enough space so that single researchers can be added within the timeframe of multiple days.

This is in my opinion a major handicap to keep people involved. Thanks, GerardM (talk) 17:46, 20 September 2018 (UTC)

Currently investigating this with others. Sjoerd de Bruin (talk) 18:56, 20 September 2018 (UTC)

Commons category[edit]

Just wondering (sorry if all these have already answered):

1) Why we need Commons category (P373) since we have commonswiki?

2) Why Commons category (P373) is not a link like commonswiki?

3) Is there a bot that checks:

  • a) if P373 value is different from commonswiki link?
  • b) if there is a value for P373 but not for commonswiki and add it?
  • c) if there is a value for commonswiki but not for P373 and it?

Xaris333 (talk) 12:08, 21 September 2018 (UTC)

@Xaris333: multiple items (category item, list item, article item) often link to the same category. Downstream users are using it. Mike is running some bots to keep things sane and we have several reports. Mike can probably tell a bit more about it. Multichill (talk) 15:45, 21 September 2018 (UTC)
@Xaris333: I'll let others answer (1) and (2), as I mostly work with the commonswiki links. With (3), Pi bot (talkcontribslogs) does (a) and (b), and DeltaBot (talkcontribslogs) does (c), as much as they can. There are quite a few cases where the bot can't resolve the differences / add the links, though, that need human checking. If you're interested in helping clean these up, then Wikidata:Database reports/Constraint violations/P373 is a good place to start (and if you spot any patterns that a bot could sort through, let me know and I can see if I can code something up). Thanks. Mike Peel (talk) 16:51, 21 September 2018 (UTC)
Why Commons category (P373) is not a link like commonswiki? is a question I have been asking since my first week here. What is the holdup? I think it had to do with whether the field should hold the category or the page. --RAN (talk) 01:24, 22 September 2018 (UTC)
The arguments over whether to use the Category or the Gallery (main-space) page were entirely about the interwiki link. Commons category (P373) is the Commons category, regardless of whether the interwiki link points to the Category or the Gallery. - Jmabel (talk) 02:31, 22 September 2018 (UTC)

Predecessor and successor state[edit]

For Germany (Q183) is there a field for predecessor state names and successor state names? I get an error message now for citizenship unless I have the proper name of the country for the years a person was living in that country. --RAN (talk) 01:19, 22 September 2018 (UTC)