Wikidata talk:WikiProject Fictional universes

Move to Wikidata:WikiProject Fictional universes[edit]

I propose to move the task force to Wikidata:WikiProject Fictional universes per Wikidata:Project chat/Archive/2013/07#Rename "task forces" to "project". The fewer pages and backlinks it has, the simpler can it be moved. --Marsupium (talk) 07:37, 24 December 2013 (UTC)[reply]

@Marsupium:.

Done

First task[edit]

I am not sure that it is really the best solution to make 'fictional equivalents of real things', like 'fictional lion', 'fictional creative work', 'fictional jewellery' etc. I think the first task here would be to make a list and/or summary of previous discussions on this topic, for example Zolo's proposals for properties 'truth value' and 'is fictional'. Zolo's proposals were not really rejected, but there is no consensus for any solution brought up until now. Another solution would be to make fictional equivalents of the properties {{P|31|| (fictional instance of) and perhaps subclass of (P279). When there is such an overview, we could hopefully put a step forward and really build good practices. Bever (talk) 23:22, 11 January 2014 (UTC)[reply]

@Bever: I'm afraid this horse is dead, the property and the item are here. Truth value would mean we have to use a boolean on every query, this solution has not this problem and is clean. TomT0m (talk) 00:53, 19 January 2014 (UTC)[reply]

use of "fictional" items[edit]

It seems that the latest discussions was in favor of using items like Martinique Macaw (Q74575) to describe fictional items, but to what extent ? If a fictional character is said to be born in London, should we use "fictional London" ? Or should we only use fictional Xs for some key properties like instance of (P31) and subclass of (P279). Note that we cannot use them in non-item properties anyway. --Zolo (talk) 06:02, 19 January 2014 (UTC)[reply]

I'm in favor to keep things consistent, it's easier to always create the fictional item imho, no question to ask yourself, easy answer to newbies, no particular cases when we want to automate something. Maybe things are a little more complex in the case of biographie or romanced true stories though. TomT0m (talk) 13:32, 19 January 2014 (UTC)[reply]

To ease the creation of a fictional item, maybe we could code a gadget ? TomT0m (talk)

(edit conflict) If I understand the property fictional or mythical analog of (P1074) properly we have to create a second item <fictional foo> for every item <foo> and add this property??? Double the 15.000.000 items by bot? The property documentation gives "classes item" as "suggested values". I do not understand why we should handle fictional classes and fictional instances differently. But I think that we will run into weird trouble even with classes only:

I shall not say

~~<cup of tea of John Doe in the story "Foo tales"> instance of (P31) <cup of tea>.~~

but

<cup of tea of John Doe in the story "Foo tales"> instance of (P31) <fictional cup of tea>.

<fictional cup of tea> fictional or mythical analog of (P1074) <cup of tea>. ?

In guess that it makes much more sense to state that each fictional cup of tea is an instance of a cup of tea/that the class of cups of tea has the real and fictional ones as tokens. I do not think this property is a good idea. It is even worse regarding the fact that fictional is relative. What about predictions about the future? What about historical novels? Those with only historical facts? Those with mostly historical facts? What about a satire about Barack Obama (Q76) with (naturally) false(=unreal=fictional) facts? Is it about a <fictional analog of Barack Obama>? The articles bundled by possible world (Q1088088) might be a good starting point for reading to discuss these questions. Using that expression I think our classes should be usable for all possible worlds including the fictional ones. For the moment I am quite sure that we should use our current classes as real and fictional ones unless somebody brings good arguments for the opposite. Regards, --Marsupium (talk) 13:46, 19 January 2014 (UTC)[reply]

Your way of talking seems familiar, do we know each other ? Of course you are overdoing it kind of a lot, we will have only a fraction of fictional items, almost all of them won't have fictional analogs. The advantage of this solution is that we maintain a clear distinction beetween real world and fictional world object, so that fictional worls items will appear in queries only if someones want them. For the possible worlds question, please be clearer. TomT0m (talk) 14:26, 19 January 2014 (UTC)[reply]

I apologise. I might have learned English too much on the street or use too much pluralis auctoris (Q280177) and did not plan to address especially you, but had an edit conflict.

I fear I might not overdo. One would not only need such <fictional foo> classes for subclass of (P279)- and instance of (P31)-statements, but also for depicts (P180)-statements and similar ones for literature which will force require to double every class used for those statements. The more the Wikidata project and especially WD:WPVA will go on, the nearer will the ratio of needed <fictional foo> items get to 0.5 of the Wikidata items.
That is necessary, of course, but I think that there is a better solution then: subclass of (P279) <fictional entity (Q14897293)> should be enough. It is then necessary to exclude them explicitly.

I would like to explain my thoughts better, though I only came here interrupting my Wikidata break I am currently on since I feared that fictional or mythical analog of (P1074) will lead to much work that might not be constructive. Every possible world is sort of a coherent (cf. coherentism (Q1778809)) description of a way the world could possibly be. Every coherent "fictional system" is a possible world. It might have some rules. Some stories by Thomas Mann (Q37030) might be in accordance with all physical law (Q214070). The "Harry Potter fictional system" might only be in accordance with parts of all physical law (Q214070). Regards, --Marsupium (talk) 15:51, 19 January 2014 (UTC)[reply]

What I find a little strange (well, not really) is that you are using (in a bad context) a concept I recently used to deal with inconsistencies in Wikidata. But I'll assume this is totally unrelated :) No problem wath ya talking Yo. The meaning is more confusing. TomT0m (talk) 16:03, 19 January 2014 (UTC)[reply]

About real people in fictional works, I have just seen Julius Caesar (Q3188509) about Julius Casesar as depicted in Asterix. He is clearly assimilated with the real Julius Casear though depicted in a humorous way. So should he be a "fictional analog" of Julius Caesar. In other works, the cutoff between fictional characters and real characters may indeed be more problematic, unless we decide that everything that appears in a fictional work is fictional. But then that does not sound really satisfying. For visual arts, should a 19th century painting of Alexander the Great be said to depict a fictional Alexander ? What about an official portrait of Napoleon ? --Zolo (talk) 20:48, 19 January 2014 (UTC)[reply]

I suggest this way:

<Julius Caesar in Asterix>: instance of (P31) -> fictional character (Q95074) and fictional or mythical analog of (P1074) -> Julius Caesar (Q1048)
<cup of tea of John Doe in the story "Foo tales">: instance of (P31) -> <fictional object> and fictional or mythical analog of (P1074) -> <cup of tea>
<19th century painting of Alexander the Great>: instance of (P31) <artwork> (or something similar) and a property that tell us that describes <Alexander the Great>

--Paperoastro (talk) 23:33, 19 January 2014 (UTC)[reply]

Fictional person and fictional human[edit]

So I think fictional person comprises fictional humans and various sorts of mythological creatures with a sort of human-like behavior (deities, hobbits, etc.), right ? There seems to be some edge cases though, like squib (Q6133514). --Zolo (talk) 06:06, 19 January 2014 (UTC)[reply]

We could use the instance instance of (P31) -> fictional character (Q95074) for all fictional character (with Q95074 subclass of fictional entity (Q14897293)) and use fictional or mythical analog of (P1074) with human for human characters to maintain the connection with the "real world". Imho, squib (Q6133514) is a class and could be used in a hierarchy to classify the characters of the universe of Harry Potter. --Paperoastro (talk) 15:24, 19 January 2014 (UTC)[reply]

I suppose the hierachy should be the same for fiction and non-fiction, but I think the main practical difference is that many fictional characters are not really humans (gods, monsters, etc.) So would that be ok to:

have fictional human (Q15632617) fictional or mythical analog of (P1074): human (Q5)
have person (Q215627) fictional or mythical analog of (P1074): person (Q215627)
move a good part of p31: person (Q215627) to p31: fictional human (Q15632617) ?

--Zolo (talk) 20:23, 19 January 2014 (UTC)[reply]

To avoid the creation of several items, I suggest for fictional people:

P31 -> fictional character (Q95074)
fictional or mythical analog of (P1074) -> human (Q5) (as principal property, not qualifier).

The same for other kinds of characters (animals, objects, and so on...). In this manner, imho, we do not need to create items like fictional human (Q15632617). --Paperoastro (talk) 23:15, 19 January 2014 (UTC)[reply]

This would not be really satisfying, in fictions a lot of characters are non human and could be a person in broader sense. I think that fictional human is just a subclass of fictional characters. I think a few more items are really not a problem, one day you will wake up and they are likely to be there anyway :). TomT0m (talk) 10:40, 20 January 2014 (UTC)[reply]

What is the problem for other kind of characters? Mickey Mouse (Q11934): P31-> fictional character (Q95074) and fictional or mythical analog of (P1074) -> house mouse (Q83310); Vulcan (Q1088708): P31 -> fictional object and fictional or mythical analog of (P1074) -> planet (Q634). --Paperoastro (talk) 22:10, 20 January 2014 (UTC)[reply]

I would tend to think that Donald is a Toon :)

There are two fictional human items -> fictional human (Q15632617) and fictional human (Q15403010)

Proposal: Fictional Taxons[edit]

(I’m not sure if this is the right place to post this proposal, so please tell me where I should go if I’m wrong here.) I’ve thought a bit about how to map fictional characters, especially if they belong to fictional species, and I think the most “structured” way would be to create “fictional taxons” analog to the “real”, biological taxons (taxon (Q16521)). So, for example, giraffe (Q15083) has the following properties:

And if there ever where a real, relevant giraffe, it would be tagged as instance of (P31) giraffe (Q15083). We do this already for humans: Albert Einstein (Q937) instance of (P31) human (Q5).

Now the proposal: We create a data object “fictional taxon” and tag it as follows:

“fictional taxon” fictional or mythical analog of (P1074): taxon (Q16521)
“fictional taxon” subclass of (P279): fictional entity (Q14897293)

And then we create our fictional taxonomy, so to tag Donald Duck (Q6550), we could create the taxon “fictional duck in the Disney universe” and set:

“fictional duck in the Disney universe” instance of (P31): “fictional taxon”
“fictional duck in the Disney universe” fictional or mythical analog of (P1074): duck (Q3736439) (debatable, since real ducks don’t speak, hoard money, etc.)
Donald Duck (Q6550) instance of (P31): “fictional duck in the Disney universe”

Another example would be:

muggle (Q154224) instance of (P31): “fictional taxon”
muggle (Q154224) fictional or mythical analog of (P1074): human (Q5)
muggle (Q154224) parent taxon (P171): “humans in the Harry Potter universe”
wizard in the Harry Potter universe (Q15298259) instance of (P31): “fictional taxon”
wizard in the Harry Potter universe (Q15298259) parent taxon (P171): “humans in the Harry Potter universe”

and then

Problems/unsolved questions I have with this model:

You get a lot of fictional taxons.
A lot of universes include ducks. When do we create a new “fictional duck in universe X”-object and when does reuse of an existing “fictional duck”-object suffice?
Where do we draw the line between fictional and real taxons? Should we build up a whole fictional taxon tree for every fictional universe and then create parent taxon (P171)-relations between taxons as fit (which I think works best for large fictional taxons with a great independence from reality). Or should we try to directly refer fictional taxons to analog “real” taxons (which I think would be best for universes who only introduce a few new taxons).
What should the “root taxon” of a fictional taxonomy be? “lifeforms in the Harry Potter universe”?
In stories that don't introduce a fictional taxonomy at all, how should we map characters?

I’d love to hear your feedback on this! --mxmerz talk 14:43, 2 February 2014 (UTC)[reply]

“fictional taxon” subclass of (P279): taxon (Q16521) just about this statement, don't use subclass of, use fictional analog of. We created this property such as we don't mix classification of fictional stuffs and real world things. Apart from that I don't think an analog of real world taxonomy is really insteresting. TomT0m (talk) 17:03, 2 February 2014 (UTC)[reply]

Thanks, I changed it. Could you define “interesting”? I agree that it wouldn’t make much of a difference for fictional universes which don’t introduce their own species or change much of our reality, but especially for large universes (Star Wars, Middle Earth,…), couldn’t fictional taxons be useful to describe the in-universe relations between species and their relation to “real” species? --mxmerz talk 17:50, 2 February 2014 (UTC)[reply]

I'm not aware of a real universe with a real philogeny as in biological taxonomy, but why not. TomT0m (talk) 18:37, 2 February 2014 (UTC)[reply]

Tree of fictional creatures/persons[edit]

We should work with this tree for some time. It seems like some nodes are superfluous, some are in wrong place (at least from Russian labels view). Infovarius (talk) 22:29, 4 June 2014 (UTC)[reply]

Yep, items like list of Star Trek races (Q1700422)  

are just silly with their french label Star Trek Ethnographia, which is not a class of fictional character, it's a classification by itself :) the english label is extraterrestrial from Star Trek, something totally different, with the label speces of humanoid alien from Star Trek which is yet something totally different.

I would propose to rename this class extraterrestrial creatures of Star Trek, such that Spoke is an instance of it, and to create a class Star Trek Species, such that

⟨ Spock (Q16341) ⟩ instance of (P31) ⟨ Vulcan (Q6497384)  

 ⟩

and

⟨ Vulcan (Q6497384) ⟩ instance of (P31) ⟨ Star Trek Species ⟩

And

⟨ Vulcan (Q6497384) ⟩ subclass of (P279) ⟨ extraterrestrial creatures of Star Trek ⟩

Same goes from character race (Q2607197)  

which should not be in the subclass tree but more

⟨ Star Trek Species ⟩ subclass of (P279) ⟨ character race (Q2607197)  

 ⟩

All right ? TomT0m ^(talk) 10:15, 5 June 2014 (UTC)[reply]

Sounds good. --Infovarius (talk) 20:37, 5 June 2014 (UTC)[reply]

What property to set the fictional universe for a work?[edit]

Hello,

I thought from narrative universe (P1080) would be the right property to indicate that a work is in a fictional universe (say for example, Harry Potter and the Philosopher's Stone (Q43361) is a work in the universe Harry Potter universe (Q5410773)) but it seems that it is only for the fictional elements inside the book, so what would be the right property for the work itself? -Ash Crow (talk) 14:04, 19 July 2014 (UTC)[reply]

MaEr suggests to use part of (P361).

As I said on his talk page, to distinguish between the franchise elements and the fictional universe items could be interesting.

That would allow to generate a list of each items of a fictional universe, from an internal point of view.

But a quick review of the pertinence of part of should be done, to identify the cases were it's a little bit problematic to have several part of properties for a franchise element. --Dereckson (talk) 14:22, 19 July 2014 (UTC)[reply]

"Part of" is also supposed to be used for the magazine in which a short story is published, or so I guess if we want to be consistent with Help:Sources. It seems that this could cause some confusion. --Zolo (talk) 17:50, 19 July 2014 (UTC)[reply]

You use also "Part of" for a brother or sister of a sibship… "Part of" is a very generic property. --Harmonia Amanda ({https://www.wikidata.org/wiki/Wikidata_talk:WikiProject_Fictional_universes#Need_another_property{int:Talkpagelinktext}}) 23:29, 19 July 2014 (UTC)

Zolo and Harmonia Amanda — indeed, "part of" is very generic. According to Property talk:P361, it can link anything to anything.
Property talk:P1080 says that from narrative universe (P1080) links a fictional entity to a fictional universe. The property documentation says: domain = any under the subclass tree of fictional entity (Q14897293). This is also what the smaller blue box in Property talk:P1080 suggests. The language-dependant labels and descriptions, however, vary a bit, and aren't very clear. In some translations, they omit the information that both entity and universe should be fictional. --MaEr (talk) 09:50, 20 July 2014 (UTC)[reply]

So, I guess we should have a new property "Fictional universe of the work" (in French "univers de fiction de l'œuvre") to tell that a work uses this fictional universe? -Ash Crow (talk) 10:58, 20 July 2014 (UTC)[reply]

Alternatively, we can change the constraints on from narrative universe (P1080) so that we can use it, but I am not sure it is a good idea because Harry Potte volume 1 p1043: Harry Potter seems to mean something like "Harrry Potter volume 1 is a book that exists only in the Harry Potter Wikiverse. @Dereckson: if you want to get a list of objects from the Harry Potter universe, you should add the condition that the item should be about a ficitonal entity anyway, so that you do not get items like "list of Harry Potter characters ([1]). -Zolo (talk) 14:58, 20 July 2014 (UTC)[reply]

We clearly need either a new property (the suggested "Fictional universe of the work") or to change the constraint on from narrative universe (P1080). I can understand the need for a purely in-universe property so I would support a new property which would be used only for the works in a fictional universe. The same property, but with an out-universe point of view. And we could then clarify the description of both properties. What do you think? --Harmonia Amanda (talk) 22:18, 20 July 2014 (UTC)[reply]

Rather create a new property than change the existing one. The new property could be something like: (real-world literary work) describes (fictional universe). In the opposite direction, there could be the property: (fictional universe) is described by (real-world literary work). --MaEr (talk) 17:16, 21 July 2014 (UTC)[reply]

It seems good to me. (but the name you suggest seam a bit too generic for such a specific use, and « described by » would in French translate to « décrit par », which is already used by described by source (P1343)) -Ash Crow (talk) 20:08, 21 July 2014 (UTC)[reply]

Thinking about the name: would "describes the fictional universe" and "described in the work" would fit ? It seems clear to me, not confusing with any other property. -Ash Crow (talk) 23:37, 22 July 2014 (UTC)[reply]

~~I agree with these suggestions.--Harmonia Amanda (talk) 00:07, 23 July 2014 (UTC)~~. Actually this link with the next section. We need these properties:[reply]

to link a fictional element to his fictional universe : from narrative universe (P1080)
to link a fictional element to the work in which it appears : "Part of"? A new one? (discussion next section)
to link a work with the fictional universe it describes : a new one (discussion here)

So if we speak of 3 here, I suggest "Fictional universe of the work" for this property, instead of "is described by" (like MaEr proposed; too generic) and of "described in the work" (like Ash Crow; which would fit better for 2). --Harmonia Amanda (talk) 17:43, 23 July 2014 (UTC)[reply]

Actually, my proposal in this section is to create two new properties:

to link a work with the fictional universe it describes: <work> "describes the fictional universe:" <universe>
conversely, to link a fictional universe with a work that describes it: <universe> "described in the work:" <work>

"Fictional universe of the work" is not a really good property name as is does not fit in a <x> <property> <y> pattern. -Ash Crow (talk) 17:56, 23 July 2014 (UTC)[reply]

So it would be:

to link a fictional element to his fictional universe : from narrative universe (P1080)
to link a fictional element to the work in which it appears : "Part of"? A new one? "present in the work" / "mentionned in the work"? (discussion next section)
to link a work with the fictional universe it describes : a new one : <work> "describes the fictional universe:" <universe>
to link a fictional universe with a work that describes it: a new one : <universe> "described in the work:" <work>

3 and 4 being the reverse of one another. Ok, I approve the name suggestion.--Harmonia Amanda (talk) 18:04, 23 July 2014 (UTC)[reply]

@MaEr, Zolo, Dereckson, Harmonia_Amanda:: I made the proposals for the two properties discussed here. -Ash Crow (talk) 11:55, 24 July 2014 (UTC)[reply]

(The properties takes place in fictional universe (P1434) and fictional universe described in (P1445) were created shortly after this discussion, and the issue was resolved.) --Yair rand (talk) 21:04, 9 July 2017 (UTC)[reply]

Need another property[edit]

Hello, I think we need another property “present in the (fictional) work” (in French: « apparaissant dans l'œuvre »). This property would indicate in which work of the item is present. It would be like Property:P1080 but for out-universe information. For exemple Elessar (Q3721872) <present in> The Lord of the Rings (Q15228), or Basilisk in the Chamber of Secrets (Q15720826) <present in> Harry Potter and the Chamber of Secrets (Q47209).

Currently we use Property:P361 but it's so imprecise many contributors delete these, yet we need it for generating lists from a specific work and not a fictional universe. What do you think? --Harmonia Amanda (talk) 17:27, 23 July 2014 (UTC)[reply]

I'd say <fictional element> "present in the work" <work> for characters actually present in the work, and <fictional element> "mentioned in the work" <work> for characters simply named (yes, I want two new properties here as well. Could help with long and complicated stories like A Song of Ice and Fire (Q45875), where we have characters that are mentioned in one novel and really present in the next one.) -Ash Crow (talk) 18:10, 23 July 2014 (UTC)[reply]

I did the first demand ("present in work") Wikidata:Property_proposal/Creative_work. Should I make the second one too, @Ash Crow:?

Fictional stars, protostars, planets, moons, asteroids and such[edit]

Hello,

I'm trying to sort all the fictional astronomical object in the Serenityverse (Q17413155), and I'm not always sure about what instance of (P31) to set. Here are some comments/questions:

For fictional stars, there was an item star and-or planetary system described fictionally (Q3972970) with articles on en: and it:. I splitted it in two sub-items : fictional star (Q17424291) and fictional planetary system (Q17424312).
For fictional protostars, I'm not sure what to do? Either use fictional star (Q17424291) or create a fictional analog of protostar (Q204903)... Same question for fictional brown dwarves.
For fictional planets, no problem: fictional planet (Q2775969) already exists and is already in use (e.g. on the Star Wars planets)
For fictional moons: Endor (Q832100) uses fictional planet (Q2775969), but it looks like there is a mess in interwikis, as some Wikipedia articles are about the planet and not it's moon (at last es: who has Endor (Q12180673) for the moon). I guess it would be better to create a new "Fictional moon" item instead of using fictional planet (Q2775969). Am I right?
For everything else: I guess we have to create fictional analogs to comets, asteroids, galaxies. I wasn't able to find them if they already exist...

-Ash Crow (talk) 00:11, 27 July 2014 (UTC)[reply]

(Comment) How many items about non-existant stuff from Serenity... Are they all notable? --Infovarius (talk) 21:12, 12 August 2014 (UTC)[reply]

When to create fictional universe items?[edit]

I would guess that most fictional universes that have items for their component characters/entities/concepts do not have items themselves. Should all of them have items? If not, is there any other way to associate the fictional entity with the book/story/series? --Yair rand (talk) 13:37, 4 May 2015 (UTC)[reply]

Good question. My answer would be when you need it. I guess any work who have a character item deserves a corresponding item anyway for the universe he belongs to for structural reasons per WD:N. TomT0m ^(talk) 14:03, 4 May 2015 (UTC)[reply]

To link with the book/series you use present in work (P1441), not from narrative universe (P1080) and you don't necessarily need a fictional universe item. And when you need it (several different works sharing the same universe for example), you create it. --Harmonia Amanda (talk) 14:21, 4 May 2015 (UTC)[reply]

Join conversation at Property_talk:P144#Splitting_.22adapted_from.22_from_this_property[edit]

Hey All, I started a conversation about adaptation vs "based on" on P144's talk page please join the discussion, Sadads (talk)

Multiple timelines[edit]

The Star Trek universe (Q18043309) consists of three main timelines Prime Universe (Q47163964), Mirror Universe (Q674612), Kelvin Timeline (Q47164023) and several others that are less relevant.

how should we express in which of these timelines a narrative is set. I used narrative location (P840) but that should be used for locations. and different works can be set at the same location, but in different timelines and I don't like the idea of creating fictional locations for each timeline. I thought narrative location should be qualified with the timeline and I considered set in period (P2408) and that sounds appropriate at first, but then again, thats meant time periods.

I can't think of anther fictional universe that has multiple timelines (universe of The Legend of Zelda (Q1399781)?) so a new property is not justified. Is there a more general property we could use? any ideas? --Shisma (talk) 10:47, 7 October 2018 (UTC)[reply]

@Shisma: Could takes place in fictional universe (P1434) be an option for works? There remains the question how to qualify a certain statement to the timeline it applies to (like I've seen at Spock (Q16341)Q16341#P746). To use takes place in fictional universe (P1434) as a qualifier here seems odd to me. Maybe one could simply use of (P642). - Valentina.Anitnelav (talk) 16:51, 7 October 2018 (UTC)[reply]

@Valentina.Anitnelav: I like the idea of simply using of (P642) in combination with a timeline. There should maybe be a timeline class --Shisma (talk) 16:45, 9 October 2018 (UTC)[reply]

Items that are not only fictional universes[edit]

Running this query I come across a few problems. here is one:

Warcraft (Q815739)

instance of (P31)

fictional universe (Q559618)

video game series (Q7058673)

A thing cannot be a fictional universe and something else. Statements like Warcraft (Q815739) is fictional universe (Q559618) that runs on Microsoft Windows (Q1406) make no sense what so ever 😅. Let us resolve all those issues. Can someone query for things that are not only fictional universes? --Loominade (talk) 16:31, 26 November 2018 (UTC)[reply]

Here is a query giving all fictional universes that are also instances of something else: query – The preceding unsigned comment was added by Valentina.Anitnelav (talk • contribs).

I agree. In that particular case, we'd need 3 different items IMO: one for the video game series (Q7058673), one for the fictional universe (Q559618) that the games and related works take place in, as well as one for the media franchise (Q196600) which would include games, books, comics, the movie, etc. --Kam Solusar (talk) 12:44, 29 November 2018 (UTC)[reply]

And now I see that we already have an item for the franchise: Warcraft (Q6156907). And we have Warcraft universe (Q26181639). I'm not too familiar with Warcraft, but I thought the content of World of Warcraft was part of the overall Warcraft lore. And we have places in Warcraft (Q849468), which also has the statements instance of (P31)fictional universe (Q559618) part of (P361)Warcraft universe (Q26181639) and from narrative universe (P1080)Warcraft (Q815739). Warcraft universe (Q26181639) on the other hand also has the circular statement part of (P361)places in Warcraft (Q849468). That looks pretty chaotic. --Kam Solusar (talk) 13:13, 29 November 2018 (UTC)[reply]

Mythologies and Fictional Universes[edit]

is there anyone here who thinks these should be considered as fictional universes, linked in from narrative universe (P1080)? Lets make a new property, along the lines of from mythical canon and takes place in mythical canon. Thoughts?--Loominade (talk) 16:43, 26 November 2018 (UTC)[reply]

@Loominade: According to the constraints from narrative universe (P1080) can be used for mythologies, too, but the label does not express it and it is removed from mythical entities quite often (which I understand). There is a proposal on Property_talk:P1080 by Pharos to relabel from narrative universe (P1080) to be more inclusive, but this would need to be done in all languages (this would take some time and would probably cause some confusion). Maybe a new property would be better. We could name it "from mythical canon", like you proposed, but we could also name it (more inclusive) "from story cycle", like proposed by Pharos (from narrative universe (P1080) could be a subproperty of this new property, then). - Valentina.Anitnelav (talk) 08:23, 29 November 2018 (UTC)[reply]

Create/Import statements from Wikia[edit]

Notified participants of WikiProject Fictional universes

I created a little tool that makes it easy to create statements from Wikia pages: wikia-data.user.js.

Installation and Usage-Instructions here. I hope you find it as useful as I do. Please give me feedback --Shisma (talk) 18:34, 7 November 2019 (UTC)[reply]

Occupations of fictional characters[edit]

On Wikidata:WikiProject_Narration Wikidata_talk:WikiProject_Narration#fictional_occupations_that_are_real_occupations a discussion was started whether fictional characters' occupations should be indicated via fictional analogues of real occupations, if the occupation exists in the real world. (E.g. secretary (Q76451097) vs. secretary (Q80687)) - Valentina.Anitnelav (talk) 07:55, 11 December 2019 (UTC)[reply]

Canon, Inconsistencies and Retconing[edit]

I personally find this a fascinating aspect of fictional universes and i'd like to represent aspects like the following in wikidata.

Sherlock Holmes Retcon[edit]

In 1893 Arthur Conan Doyle became sick of writing about Sherlock Holmess so he decided to kill him off in the short and unappealing story The Final Problem (Q228119) just to bring him back years later in The Adventure of the Empty House (Q1171427). Nicholas Meyer – a Sherlock Holmes Nerd – apparently did not like these stories and declared them deliberate fabrications produced by Dr Watson in his story The Seven-Per-Cent Solution (Q7763420). Here Sherlock decided to leave the world stage for personal reasons. So he moved Dr Watson to publish The Final Problem (Q228119) as a cover story for his disappearance.

The Seven-Per-Cent Solution (Q7763420) is of course not considered to be a part of the canon of Sherlock Holmes (Q2316684).

Star Trek[edit]

Only Star Trek film (Q61283808) and Star Trek episode (Q61220733) are considered to be part of the Star Trek canon (Q3500963).

In fact there are so many Star Trek novel (Q72996181)s: no writer could possibly keep track of everything that ever happened in them and still could come up with a coherent story.

Among the first Star Trek novels with an original story was Spock Must Die! (Q7578767) from 1970. Here the Klingon Empire (Q10861987) is denied all future space travel by an omnipotent entity. This implicitly contradicts hundreds of books and episodes created later, where the klingons are still space faring. There is no explanation given and nobody ever felt the need to justify it. A subset of works published later could as well happen within the same universe because no Klingons show up in them or the narrative takes place earlier chronologically 🤷‍♂️

If you have more examples or similar cases feel free to summarize them here. My questions is how can we map this?

While the Nicolas Mayer universe can be basically described as a detour on the Holmes canon:

        "Detour"
     ╭────────────╮
━━━━━┻━━━━━━━━━━━━┻━━━━━━ Canon
            The Final Prblem
              happens here

Where Spock must Die is basically a dead end. Abandoned.

           "Dead end"
     ╭────────────
━━━━━┻━━━━━━━━━━━━━━━━━━━ Canon
           Everything else
            happens here

If you map the situation like this, you could create an entity for every Timeline and connect this entity takes place in fictional universe (P1434) with the work and qualify every statement that is only valid in that universe with from narrative universe (P1080) more generally valid in place (P3005). if the statement not correct in the generally accepted canon, it should be set to deprecated.

Do you have a better idea on how to map this? Can you imagine situation where this model could not be applied?--Shisma (talk) 22:12, 26 January 2020 (UTC)[reply]

Indicating animal breeds of fictional animals[edit]

Hello, I just stumbled on Harriet (Q92306814), a fictional poodle, that has both statements animal breed (P4743): poodle (Q38904) and instance of (P31): anthropomorphic dog (Q94167374). instance of (P31) was the property used to indicate the breed of a (fictional) animal before animal breed (P4743) existed. Now I'm thinking about using only animal breed (P4743) for the breed and using instance of (P31) for the species. This would reduce the number of (potentially needed) subclasses of fictional domesticated animal species, like fictional (anthropomorphic) cats and dogs.

I would propose to model the breeds of fictional animals as follows (using Trude as an example):

instance of (P31)anthropomorphic dog (Q88552181)

animal breed (P4743)poodle (Q38904)

If this is acceptable, I would start to clean up the items in a month or so, requesting the deletion of the fictional breed items (e.g. these and the anthropomorphic equivalents). Many of these items were created by me, but some of them also by Trade and Loominade: What do you think about this approach? – The preceding unsigned comment was added by Valentina.Anitnelav (talk • contribs) at 18:47, 28. Jul. 2020 (UTC).

I forgot to sign so probably the people mentioned did not get a notification: @Trade, Loominade: (sorry for the noise if you already saw this section) - Valentina.Anitnelav (talk) 11:51, 11 August 2020 (UTC)[reply]

Fine by me. Then it would also become easier to clean up anthropomorphic animal (Q2369882)--Trade (talk) 12:12, 11 August 2020 (UTC)[reply]

Done (but I only checked cats, dogs and horses) - Valentina.Anitnelav (talk) 14:32, 3 November 2021 (UTC)[reply]

Siamese twins[edit]

What would be the correct way to model a fictional siamese twin where each 'head' have their own characteristics (given name, hair/eye color, voice actor, love interest) and etc. @Valentina.Anitnelav:--Trade (talk) 21:06, 5 September 2020 (UTC)[reply]

I think we should just follow the model for real siamese twins (e.g. Chang and Eng Bunker (Q660581)); you can create a new item <fictional siamese twins> (i'm surprised that there is none, yet), one for the couple and one for each character and link them via has part(s) (P527)/part of (P361)/sibling (P3373). - Valentina.Anitnelav (talk) 15:12, 7 September 2020 (UTC)[reply]

@Valentina.Anitnelav: I looked at the Bunkers' items and I'm not sure they are yet a good exemplar; most of the individuals' biographical details should be on their respective items, with the item for the pair of them mostly just serving to link to / collect their own entities. Arlo Barnes (talk) 20:15, 17 November 2021 (UTC)[reply]

Mythical Creatures — Lexemes Party 🎉[edit]

Notified participants of WikiProject Fictional universes

WikiProject Fictional universes has participants who speak English, Spanish, French and German (more?). Let us complete Lexemes about Fictional universes in those languages!

Mythical Creatures — Lexemes Party

As a showcase lexeme I somewhat enriched dwarf (L22936). Translate these words into your favorite language! Ask, if you don't know how. Have fun! ✊ --Shisma (talk) 15:49, 16 November 2021 (UTC)[reply]

The conlang -- Láadan (Q35757) -- which I've been adding lexemes for as part of the weekly Lexeme Challenge also hosted on Dicare Tools (Q108275558) does not have words for any of those things; however, I added dedidewoth (L619219), "myth", in the spirit of your Party. Perhaps it could be added to the list for others to check their languages against? Arlo Barnes (talk) 20:12, 17 November 2021 (UTC)[reply]

better with subclasses (I was bold to change the link). --user:Infovarius 11:35, 18 November 2021 (UTC)[reply]

Duplicates or incorrectly classified in Pokémons?[edit]

What is the difference between items water (Q25588649) and water-type Pokémon (Q25930495)? I want to say: water-type Pokémon (Q25930495) seems only the description of water (Q25588649). I think that all this pair of items should be merged with each other, where water (Q25588649) is the label and water-type Pokémon (Q25930495) is the description. The objection would be that perhaps one of the two refers to a Pokémon concept, in which case one of the two is not classified correctly.

The point is the same for all types of Pokémons. --Fantastoria (talk) 17:36, 14 March 2022 (UTC)[reply]

Notified participants of WikiProject Pokémon - Valentina.Anitnelav (talk) 09:52, 15 March 2022 (UTC)[reply]

I disagree. Not only pokemon, but also moves are classified by type. Therefore, I have adjusted the statements accordingly. Of course, this would need to be done for all types. See also water-type Pokémon move (Q26001210). --MGChecker (talk) 13:13, 15 March 2022 (UTC)[reply]

P5102 and player choices[edit]

Let's say a game have two endings, one where a character dies and one where he lives. Is nature of statement (P5102) > optional (Q59864995) the right choice for death-related statements? I can't see what else to use. @Valentina.Anitnelav: --Trade (talk) 21:20, 8 April 2022 (UTC)[reply]

@Trade: I like nature of statement (P5102) > optional (Q59864995). - Valentina.Anitnelav (talk) 08:55, 9 April 2022 (UTC)[reply]

Exhaustive list and subclasses[edit]

1.- I try to have an exhaustive list of fictional character (Q95074) according to the type of work:

historical character (Q3375731) (or only for literature?)
literary character (Q3658341)
- comics character (Q1114461)
  - manga character (Q87576284)
  - webcomic character (Q111011320)
- theatrical character (Q3375722)
  - musical theatre character (Q63975020)
    - operatic character (Q50386450)
animated character (Q15711870)
- anime character (Q80447738)
puppet character (Q89349965)
audiovisual work character (pending)
- video game character (Q1569167)
- television character (Q15773317)
- film character (Q15773347)
  - horror film character (Q15780329)
  - superhero film character (Q63998451)
  - adventure film character (Q66808903)
  - musical film character (pending)
- music video character (Q65115256)
musical character (pending or equal to fictional character in a musical work (Q100708514))
- musical theatre character (Q63975020)
  - operatic character (Q50386450)
- fictional character in a musical work (Q100708514) (or song character => pending)
- musical film character (pending)

Would any be missing?

2.- On the other hand:

Would theatrical character (Q3375722) be a subclass of literary character (Q3658341)? Currently operatic character (Q50386450) is a subclass of musical theatre character (Q63975020), itself a subclass of theatrical character (Q3375722).
Would video game character (Q1569167), television character (Q15773317) and film character (Q15773347) be subclasses of «audiovisual work character», itself a subclass of fictional character (Q95074)?

Thanks and regards.

Pd. I don't know if I have been able to explain the points.

Just one short comment, for now: I think instead of theatrical character (Q16830339) you mean theatrical character (Q3375722) (it seems that the English label of the former item was misleading - for this reason I deleted it. I'm not sure about the right English term for de:Fach/fr:emploi...) theatrical character (Q16830339) is rather about a set of roles, not about specific fictional characters. It is closer to stock character (Q636497) - Valentina.Anitnelav (talk) 19:05, 5 June 2022 (UTC)[reply]

Yes, thanks. I have already made the changes in my comment above. I have also taken the opportunity to add those that I have discovered since then. --Fantastoria (talk) 08:43, 6 June 2022 (UTC)[reply]

Bluerasberry (talk) 19:16, 6 June 2022 (UTC)[reply]

We may lack one for puppets like Guignol (Q4179)  

or The Muppets (Q108273055)  

. author TomT0m / talk page 19:24, 6 June 2022 (UTC)[reply]

Comments:

Currently, legendary figure (Q13002315) is a subclass of human whose existence is disputed (Q21070568), which is in turn a subclass of hypothetical entity (Q18706315). From the description, it seems that it transcends the idea of a fictional character, I dare to say as a mythical character (they are now siblings and direct descendants of human whose existence is disputed (Q21070568)). I don't think we could put them as subclasses of fictional character (Q95074), properly.
In the same idea, as an idiomatic doubt, are there differences between the meanings of «fictional» and «fictitious»? In Spanish, both are translated the same the vast majority of the time, so I'm not sure. I think that legendary character/figure (and mythical character) is closer to a fictitious character than a fictional character, from what I am able to deduce from the descriptions and labels.
Guignol (Q4179) or puppetry character (not puppet itself): character from puppetry (Q588750). Perhaps a subclass of theater character? The Muppets seem to me to be a particular case of puppetry character (in line with other parallel cases, e.g. The Avengers/comics character?).

Thanks to both and regards. --Fantastoria (talk) 12:20, 7 June 2022 (UTC)[reply]

There already exists puppet character (Q89349965), we have also animated character (Q15711870) - Valentina.Anitnelav (talk) 13:24, 7 June 2022 (UTC)[reply]

We lacked a metaclass for those classes, so I created type of fictional character (Q112265308). author TomT0m / talk page 14:04, 7 June 2022 (UTC)[reply]

Is it different from stock character (Q636497)? --Infovarius (talk) 19:04, 7 June 2022 (UTC)[reply]

@Infovarius Yes. Your item is about stereotyped characters, like « the good guy », « the bad guy » or « the cowboy » in a fiction. « Ernst Stravo Blofeld » would be « The bad guy » in James bond. It would be a movie or a novel character with my item. author TomT0m / talk page 19:56, 7 June 2022 (UTC)[reply]

@TomT0m: How should we use this class?:

--Fantastoria (talk) 16:24, 9 June 2022 (UTC)[reply]

Comments:

The description of puppet character (Q89349965) doesn't make it clear whether it refers only to puppets in works of puppets (as expected) or to all puppets in general, because there is puppet (Q1913135) that does refer to all puppets (no matter where they appear), so I am able to deduce from the instances of each one. Instead, it emphasizes its difference from anthropomorphic puppet (Q88594927), which appears to be a subclass of puppet (Q1913135), rather than fictional puppet (Q37146376).
I add puppet character (Q89349965) and animated character (Q15711870) to the list.

Thanks and regards. --Fantastoria (talk) 08:22, 9 June 2022 (UTC)[reply]

puppet character (Q89349965) is for fictional characters, not for real puppets embodying a character (which could be found in a museum, for example). E.g. Kermit the Frog (Q1107971) is about the fictional character which could be embodied by several (real) puppets controlled by a puppeteer. - Valentina.Anitnelav (talk) 14:16, 9 June 2022 (UTC)[reply]

I think that's what I wanted to say, but I said it in that weird way. What happens is that I continue to emphasize that the descriptions are as precise and clear as possible to the idea of what the item is. Simplifying, puppet character (Q89349965) is the one I have added to the list at hand. --Fantastoria (talk) 16:24, 9 June 2022 (UTC)[reply]

PS:

historical character (Q3375731), music video character (Q65115256), fictional character in a musical work (Q100708514)?
At one or two lower levels?: manga character (Q87576284), webcomic character (Q111011320), anime character (Q80447738), horror film character (Q15780329), superhero film character (Q63998451), adventure film character (Q66808903).
In parallel to musical theatre character (Q63975020), there is no musical film character (e.g. Mary Poppins (Q3735317)).

--Fantastoria (talk) 12:08, 9 June 2022 (UTC)[reply]

At this point, I think I have the exhaustive list and the classification. I list it in my first comment and implement it. Thank you all. --Fantastoria (talk) 09:54, 23 June 2022 (UTC)[reply]

Multiple items for the same objects?[edit]

I ask because I don't know the Ninja Turtles universe. We have the following:

Aren't they each the same except for their appearance in different series of the same franchise? Merge? --Fantastoria (talk) 07:45, 6 August 2022 (UTC)[reply]

@J_1982: What do you think? - Valentina.Anitnelav (talk) 13:12, 9 August 2022 (UTC)[reply]

USS Enterprise (Q834003) and USS Enterprise (Q3547394) are the same thing at different times. The french Wikipedia has an article for both 🤷 – Shisma (talk) 17:06, 16 October 2022 (UTC)[reply]

plus: USS Enterprise-D (Q3547392) vs. USS Enterprise-D (Q845684). This time it is french and korean – Shisma (talk) 17:07, 16 October 2022 (UTC)[reply]

Could we use permanent duplicated item (P2959) and the values of instance of (P31): permanent duplicated item (Q107183688) or Wikimedia duplicated page (Q17362920) or Wikimedia permanent duplicate item (Q21286738)? But, in the cases I have found, there are not even Wikipedia articles or they are not duplicates. --Fantastoria (talk) 07:59, 27 October 2022 (UTC)[reply]

Customizable values[edit]

Considering that many open world games allows you to customize the sex, hair/eye color, ethnicity and etc. of player characters how do we indicate this? And how do we differentiate these values from the default appearance? Trade (talk) 17:35, 7 November 2022 (UTC)[reply]

Hi, @Trade:

There is default value (Q281782). Perhaps it could be used as a qualifier. For instance: applies to part (P518) ==> default value (Q281782). We have modding (Q1087043) too, although I am not sure if it is what you are looking for. --Fantastoria (talk) 09:28, 8 November 2022 (UTC)[reply]

Hi Trade, hi Fantastoria, I like using default value (Q281782) as a qualifier value. As qualifier property one could also use object has role (P3831). For optional values one could use nature of statement (P5102) ==> optional (Q59864995). Maybe one could give the default appearance a preferred rank. - Valentina.Anitnelav (talk) 09:58, 8 November 2022 (UTC)[reply]

Talking about video games with customizable player characters, not modding. @Fantastoria:--Trade (talk) 18:03, 8 November 2022 (UTC)[reply]

@Trade: That is why I said I was not sure!

I agree with Valentina.Anitnelav, although it seems that optional (Q59864995) does not have the sense of customizable. Perhaps we need a new item to give the meaning that is sought.

If preferred rank is used, the qualifier could be reason for preferred rank (P7452) ==> default value (Q281782).

--Fantastoria (talk) 09:25, 9 November 2022 (UTC)[reply]

The problem is that characters like The Last Dragonborn (Q114962186) does not show up in queries looking for female video game characters. --Trade (talk) 13:51, 9 November 2022 (UTC)[reply]

The Last Dragonborn (Q114962186) is missing the statement that it is (possibly) female. With "normal" queries you only find the best ("truthy") statements - in this case that it is a Nords (Q112597485). Other statements (e.g. that it is a game character) are omitted.

You could 1) add the statement subclass of (P279) video game character (Q1569167) to the Nords (Q112597485) character or 2) give also the statement that it is a video game character (Q1569167) a preferred rank

To find all characters that are video game characters (also those, that don't have instance of (P31) video game character (Q1569167) (or any subclass thereof) as their best value) you should replace wdt:P31 by p:P31/ps:P31 (this finds all statements).

example query:

SELECT ?item ?itemLabel WHERE {
    #does only find one character
    #?item wdt:P31 wd:Q112597485; wdt:P31 wd:Q1569167
  
    #finds all (two) characters, including the Last Dragonborn
    #?item wdt:P31 wd:Q112597485; p:P31/ps:P31 wd:Q1569167
    
    #finds all characteres, excludes those with a deprecated rank
    ?item wdt:P31 wd:Q112597485; p:P31 ?st. ?st ps:P31 wd:Q1569167.
    MINUS{?st wikibase:rank wikibase:DeprecatedRank}
    SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
    }

Try it!

- Valentina.Anitnelav (talk) 08:59, 10 November 2022 (UTC)[reply]

Tried a quick solution on The Last Dragonborn (Q114962186)--Trade (talk) 23:56, 14 November 2022 (UTC)[reply]

I'm not sure about the label - could we label it something like "custom value", corresponding to "default value"? - Valentina.Anitnelav (talk) 15:13, 15 November 2022 (UTC)[reply]

Enemy Countries[edit]

Beyond the characters, fictional countries can also be enemies. What property or qualifier could indicate that? Thanks. --Fantastoria (talk) 09:04, 26 November 2022 (UTC)[reply]

Hi Fantastoria, I think we can just use enemy (P7047) unless someone disagrees (if not I would just broaden the constraints to include fictional countries). - Valentina.Anitnelav (talk) 08:41, 1 December 2022 (UTC)[reply]

It's the first thing I used, but since the "warning" said «it couldn't be done», I thought there would be another way to do it. I will then apply the constrain to be able to use it. Saludos. --Fantastoria (talk) 09:29, 1 December 2022 (UTC)[reply]

Archive of Our Own not interested in working with Wikidata[edit]

I contacted the Open Doors project of this hugh fan fiction site, and after 2 weeks got a brush off. See https://www.wikidata.org/wiki/Talk:Q27764876#Not_interested_in_working_with_wikidata Vicarage (talk) 05:58, 1 December 2022 (UTC)[reply]

Just out of curiosity: what did you ask them, exactly? The last sentence "Documentation on the GitHub site also contains instructions for [...] proposing new features, which would include any wikidata-related suggestions" implies to me that they are asking for more specific suggestions (after you or we looked into the possibilities a bit closer). - Valentina.Anitnelav (talk) 08:43, 1 December 2022 (UTC)[reply]

I asked them (through a web form, so no copy of exact text) if they were interested working with wikidata. The 2 week delay and form response suggested no enthusiasm on their part, which would be required for me to work with them Vicarage (talk) 08:58, 1 December 2022 (UTC)[reply]

I understand your disappointment but I would not take their lack of enthusiasm too seriously or personally. They are also just people with a certain workload and maybe they have currently no idea how they could make use of Wikidata's data (hence they ask for more specific suggestions). If there should be a concrete idea how they could integrate the data in Wikidata we could reach out, again, in my opinion. Anyway: thank you for your initiative! - Valentina.Anitnelav (talk) 09:10, 1 December 2022 (UTC)[reply]

classes of imaginary entities as classes undetermined as to if their instances are mythical or fictional[edit]

During the last month classes of imaginary characters / animals were introduced. These are meant as intermediary classes above the fictional and the mythological (e.g. imaginary horse (Q115252411) and imaginary mammal (Q115257598)) and may be used to serve as "umbrella classes" (see also the related talk at Talk:Q115252411). They should differ from classes like "mythical horse" and "fictional horse" in that they make no claim as to their status in a community's or societies belief system (e.g. Harry Potter is purely fictional - this character is introduced in a work called a novel without claiming that any such person really existed - while Zeus is mythological - there were quite probably people who believed in him / this concept was used to relate certain natural and social phenomena). Besides these two there are many levels - there are folklore characters that people may or may not (have) believe(d) in (e.g. Santa Claus or the Easter Bunny), there are hoaxes, there are entities in conspiracy theories (e.g. reptilian (Q13114416)) and similar things. All these things could also be subsumed under the imaginary.

The use of classes of imaginary entities seems to stem from the French Wikipedia, if I understand it correctly.

Now I ask myself if we should embrace the use of these classes (there are advantages) or if we should discourage it (there are also disadvantages).

What would be the advantages of having this kind of class?

1) Potentially fewer classes: One advantage of having a class that is undetermined if its instances are purely fictional or mythical / legendary / religious / a hoax is that we don't have to create two items, one for the mythical entity and one for the fictional - we can just create one class for the imaginary. E.g. there would be only one class <imaginary duck> instead of one class <mythical duck> and one class <fictional duck>. Now we can determine if an entity is fictional or mythical or legendary or a hoax on an instance level.

2) More flexibility: There could be more flexibility to express the status of an entity on an instance level - we don't have to decide if this character from German or French folklore is fictional or mythological via the subclass and can just make it "imaginary x" + "folklore character". We don't have to shoehorn every character in one of two classes (fictional or mythical) just by calling it a dragon.

3) Possibly easier queries: It may be easier to query to all items about duck characters (from myth and fiction) with an umbrella class.

Example: monster (Q276453)

To have an example I created imaginary character (Q115537581). I made it a superclass of fictional character (Q95074), mythical character (Q4271324), legendary figure (Q13002315), imaginary mammal (Q115257598) (this already existed) and monster (Q276453) (if you disagree with using the concept of imaginary entities it may be still deleted).

monster (Q276453) has been an item used a lot on items for fictional characters but that was defined as mythical. I deleted the corresponding statement and made it a subclass of imaginary character (Q115537581). Now, if an item is made an instance of monster (Q276453) it is left unclear if this entity is actually mythical or fictional.

To make explicit that this item is fictional you have to add instance of (P31) fictional character (Q95074) (see King Kong (Q216810)).

To make explicit that this item is legendary you have to add mythical creature (Q2239243) (e.g. Mongolian death worm (Q272149))

What would be the disadvantages of having this kind of class?

1) Potentially a lot of cleanup and disruption of the current model: We already have a lot of classes for both fictional and mythical entities. To introduce this model consistently this would include a lot of cleanup and connected decisions: E.g. I created a lot of classes for mythical entites in fictional works (e.g. fictional dragon (Q30170627)). Should we merge them, again, into the original item and make this a subclass of "imaginary"? How should we express that this concept stems from myth without making it a subclass of the mythological / mythical? Apart from this arises the question if we should declare all fictional character classes as "imaginary" (as many of them are already used on mythical characters) to make clear that they are meant to be used both for mythical and fictional (and legendary and folklore and ...) entities?

2) Potentially more classes: Arising from the first part of point 1); We already have a lot of classes for both fictional and mythical entities. If we don't merge them we end up with three classes for fictional / mythical entities + imaginary entities.

3) Ensuring consistent use: One of the first questions that might arise is probably "What is the difference to just calling it "fictional""? Even the distinction between the fictional and the mythical is sometimes difficult to maintain. Different notions exist of what it means to be "fictional" or "mythical", these seem to differ from language to language and probably also within languages between fields of interest. While some people see no issue with categorizing a god or some person from mythology as a "fictional x" others find it inaccurate. It may be even more difficult to relate the difference between "imaginary" and "fictional" and to ensure the consistent use of items.

4) May be superfluous as an umbrella class: As to queries one could get all duck characters from myth, folklore and fiction by utilizing fictional or mythical analog of (P1074) duck (Q3736439) if both the class of mythical ducks and the class of fictional ducks link to duck (Q3736439) via fictional or mythical analog of (P1074). So an umbrella class would not be needed.

Even though I created an item imaginary character (Q115537581) as an example I'm really not sure as to my opion towards these items. With respect to monster (Q276453) it seemed like a good idea. It seemed also like a good idea to have a superclass above the mythical, the fictional, the legendary and folklore without being simplistic. I also like the idea to be more flexible as to if a character is mythical, fictional or legendary and not to have to shoehorn all in one of two classes ("fictional" or "mythical") just by declaring it a bird or a dragon. But I'm not sure if it is a good idea in the long run or if we just produce more noise that would be even harder to maintain and comprehend.

@Infovarius, Tsaag Valren: pinging you as participants in the previous discussion. I'm really sorry for the long read - thanks for any opinion about this matter. :) Valentina.Anitnelav (talk) 19:13, 1 December 2022 (UTC)[reply]

See https://www.wikidata.org/wiki/Wikidata:Property_proposal/subclass_of_with_uncertain_existance Vicarage (talk) 22:50, 1 December 2022 (UTC)[reply]

I answered there but I will also put it here a bit more lengthy and explicitly with respect to fictional / mythical / legendary things. The property proposal is not really about fictional entities so I don't want to overload the property proposal discussion:

Some time ago there has been a proposal for a property Wikidata:Property_proposal/Fictional_instance_of_real_object. One could phrase it something like "described as instance of" and use it for any kind of thing - scientific hypotheses, mythical entities, hoaxes, fictional entities, etc. For fictional entities and mythical entities one could even think of having a property "described as subclass of" to express specialization (for our winged lions, unicorns and other mythical beings that might be described as a type of lion or horse). I think there are advantages of this approach. But this would mean a complete overhaul of the model that is now (successfully and quite consistently) used for many years here. It would be quite disruptive and come up with a completely different model. We have to make sure that people follow the new model and stop using the old one. There are also some open questions that would need to be discussed with people involved in actually modelling fictional universes. E.g. should the use of subclass of (P279) be generally discouraged for mythical / fictional legendary entities and should they use only <described as subclass of>? Should we delete all existing classes of fictional analogs of real world entities (unless there is a Wikipedia article or this is used somehow else)? If we still allow subclass of (P279) for classes of fictional entities: should fictional characters also use <described as instance of> even if the class is already a class of fictional things (e.g. anthropomorphic duck (Q88569700)? We really need to make sure that everybody involved is aware of the change and most approve more or less. And probably we would also need to take some well modeled domain with many classes of fictional entities (Harry Potter or Star Trek or Tolkien's universe) and analyze which problems and unclarities might arise.

This is a lot of work given that the modelling of fiction actually works quite well in Wikidata (at least the last time a looked closely :) ). What gets unclear are the borders (mythical, legendary, hypothetical, alleged, etc.). I think it is likely that people involved in modelling fiction will stay with the current model (creating new classes for fictional analogs). - Valentina.Anitnelav (talk) 11:57, 2 December 2022 (UTC)[reply]

is an individual of taxon (P10241) and our current approach to modelling fictional instances of species[edit]

Now there exists individual of taxon (P10241) and may be used to model fictional entities. There are currently 70 instances using this property query. How should we use it and how should we proceed with the already existing items for fictional analogues of taxa (e.g. fictional tiger (Q27229763))? As a first idea I think we could replace fictional analogue items for concrete species (species (Q7432)) like tiger (Q19939) (fictional tiger (Q27229763)), lion (Q140) (fictional lion (Q27267085)), etc., but we should keep the items for rather general or vague ideas of animals (e.g. if a character is presented as a spider but it is not clear which species of spider it is we should use fictional spider (Q1056850)).

What are your thoughts about this property and how we should include it in the current approach of modelling fictional animal characters? - Valentina.Anitnelav (talk) 13:19, 26 December 2022 (UTC)[reply]

The way I see it, we should keep fictional items separate from real ones. A fictional tiger will belong to the fictional tiger taxon and not to the tiger taxon. The less interference there is between one set of items and another, the more orderly and coherent the system will be.

I would withdraw all its appearances for now until an agreement was reached on what to do (interference vs. non-interference), here or, if there were not enough contributions, in the Project Chat. Although it seems to me a rather unnecessary property in a fictional item. What does it really contribute? Isn't it enough to say that it is an instance of a fictional tiger?

--Fantastoria (talk) 09:01, 30 December 2022 (UTC)[reply]

I thinkg it is a continuation of policy that real organisms are not P31 but P102451 of a particular taxon. Except Q5. --Infovarius (talk) 17:48, 1 January 2023 (UTC)[reply]

Date in fictional universes[edit]

I once saw that there was a way to add a date when it was a fictional universe item, but I don't remember now where I saw it. Could you tell me how it was done? For example, to give dates on items in Tolkien's universe. Thank you. --Fantastoria (talk) 18:34, 25 April 2023 (UTC)[reply]

Hi Fantastoria, time index (P4895) is the property you are looking for, I think. You just need an appropriate time unit (there is Valian years (Q2467002), for instance). Kind regards, - Valentina.Anitnelav (talk) 18:49, 27 April 2023 (UTC)[reply]

I have used it at Eorl (Q2475930). Would you be so kind as to review the item? I'm trying to complete it with as much information as I can (including the withdrawal of notices) for reference. --Fantastoria (talk) 07:25, 28 April 2023 (UTC)[reply]

Thank you! In my opinion the dates (birth date, death date and others) should use <unknown value> (some value) instead of <no value>; <no value> implies that he was never born (and never died) but I think you want to express that there is a date but we are not able to express it ("Unknown value may also mean the value is actually a known object, but there's currently no Wikidata item about the object.", Help:Statements#Unknown_or_no_values).

I'm also not sure if English is the appropriate language tag for his name in native language as his native language is said to be Rohanese (or Rohirric). There exists the mis language tag for cases in which there is no language code (I found it here: Help:Wikimedia_language_codes/lists/all). You could add native label (P1705) for the name in English as English is the original language of Tolkien's books, if you think it is useful. - Valentina.Anitnelav (talk) 18:08, 28 April 2023 (UTC)[reply]

Some additional thoughts: I found Q2475930#P6249, you could also express this using age of subject at event (P3629) as a qualifier to his death date. And if you know the work in which he was 60 years old you could add this via applies to work (P10663).

I find your use of time index (P4895) as a full statement interesting (Q2475930#P4895). I have to think about this - but I have no better idea, currently :). - Valentina.Anitnelav (talk) 18:15, 28 April 2023 (UTC)[reply]

I have corrected the dates. Actually, I used those properties because it suggested them to me when adding cause of death (P509). It was not mandatory, but my idea is that there were no notices or warnings. One thing leads to another and, thanks to that, there is now more complete information.
I added the age at death before knowing how the dates of birth and death could be entered, so perhaps it is a piece of information that was not needed now that we have a satisfactory solution. However, I understand age of subject at event (P3629) and significant event (P793) as something 'external' to the subject, so I would not understand death itself as an event. Yes, if it were the death of another subject: at the death of A, B was so old. I also don't know if there is an alternative to indicate the age someone was when they died when their birth and death dates are not known (or if this case is possible in reality, not in fiction).
There are dates that cannot be put in any other way, it seems to me. In this case, those related to the beginnings (and endings) of a mandate. My least favorite thing is the qualifiers, but I couldn't think of better ones.

Regarding name in native language (P1559), I wanted to see how to express the name in different languages: the fictional ones and the real one (the latter being English, which is the original language of the work). Perhaps language of work or name (P407) could be used for the fictional native language, but it would be incomplete because you would have to say what that name was; that is, a way of saying that it refers to 'Eorl' in its fictional context. In fact, 'Eorl' is not the best example to illustrate the idea. Of course, I didn't know the unknown language code existed. This is more interesting because we can add a qualifier that specifies what language it is. I have already changed the data.
What would be missing would be to specify different names in different languages and the aliases. It seems to me that 'Eorl' had neither one nor the other ('Young' is not exactly an alias or a nickname).

There are other properties that I have not added that could be done. The most obvious cases are the places of birth and death. Putting 'unknown' seems to me like it would make little sense because that's what happens for most characters, although in some it could be determined in a general way (most or all Númenoréans were born in Númenór; the same for Gondorians). Others like mother and spouse seem more remarkable to me because the names of the mothers and female spouses are rarely known, although they 'had to exist'.

'Eorl' is not suitable for film or television counterparts, but, in that case, is all the information added in the same item or is it split into different items?

I would like that eventually we could have some tables on the main page of the project with the properties and qualifiers for the types of characters, places and objects, both recommended and optional. Different possibilities for the values and example items according to the cases. Ambitious? Yes, but there is no delivery date :). --Fantastoria (talk) 09:39, 29 April 2023 (UTC)[reply]

I very much support your plan to expand the property and qualifier tables for this project (but no pressure, of course :) ).

I had a look with which properties age of subject at event (P3629) is currently used and date of death (P570) is with 11502 items the second leading property: query. Denisova 11 (Q56233289) is one example (query for all items with date of death qualified with age of subject at event).
~~As to "Young": I think it is a generational suffix (P8017). (along with the Younger (Q19838173)~~ (here is a list of all properties related to names: query). see additional comment below
If you set statements about places of birth and death to <unknown> you make it clear that it is really unknown, e.g. you express that Tolkien never specified it (as far as you know, of course). No statements could also mean that simply nobody bothered to add the places.
I think it is useful to add a statement spouse <unknown value> (some value) - it indicates that he was married.
It is up to you if you split up Eorl into several items or add all information to one. I expressed some thoughts about the two options here: Wikidata:WikiProject_Narration/Questions#Characters_appearing_in_multiple_works_(adaptations,_spin-offs)_with_different_characteristics. If you keep all versions of Eorl in one item you can use applies to work (P10663) to make clear to which works a statement applies. But if versions significantly differ I would advise you to split up the item (otherwise it may get quite messy). - Valentina.Anitnelav (talk) 18:33, 29 April 2023 (UTC)[reply]
One could express time index (P4895)2,510 Third Age yearrelative to (P2210)king of Rohan (Q7468941) using significant event (P793): significant event (P793)royal inauguration (Q1905393)time index (P4895)2,510 Third Age year. - Valentina.Anitnelav (talk) 18:52, 29 April 2023 (UTC)[reply]
Maybe "the Young" is not really a generational suffix, but rather a byname. There is epithet (P8822) to indicate these - but we would need a suiting item. There is no "the Young", currently (I only found the Younger (Q19838173)). - Valentina.Anitnelav (talk) 19:26, 29 April 2023 (UTC)[reply]

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ As many ideas began to accumulate, I created a subpage to organize them there. The original table seems insufficient to me, although I only have a preliminary idea of what to show in the table: the property, how to use it, (mandatory, recommended, optional)?, qualifiers, comments. I have also opened the talk page with some general ideas. --Fantastoria (talk) 08:56, 30 April 2023 (UTC)[reply]

Fictional government forms[edit]

Trade pointed out at my talk page that there are some fictional-analog-items (fictional republic (Q106485190), fictional dictatorship (Q120200559) and fictional transitional government (Q120200577) as examples) that may be superfluous: we could just use basic form of government (P122) along with "real" item (republic (Q7270), dictatorship (Q317), transitional government (Q59281)) to indicate that a certain fictional country is a republic, e.g. Sao Rico (Q3091514)instance of (P31)fictional country (Q1145276), basic form of government (P122)republic (Q7270) (instead of Sao Rico (Q3091514)instance of (P31)fictional republic (Q106485190)).

Did I miss something why it is still useful to have these? I noticed that some of the fictional republic items (also?) represent a government body, not (only?) a country (e.g. Galactic Republic (Q733195)). I'm not sure if this could be modeled using basic form of government (P122). Pinging contributors to these and related items: @J 1982, Fantastoria, BB-9E Droid:. Thanks for your input! - Valentina.Anitnelav (talk) 14:38, 4 July 2023 (UTC)[reply]

Although we are talking about fictional countries, I think we can understand that terms like 'republic' or 'kingdom' in the name of a country are only part of its name (a part that describes its internal functioning). The same country can go from being a republic to a kingdom or a dictatorship without ceasing to be the same country. I agree that we shouldn't 'bring' the instance value below 'fictional country'.

As for Galactic Republic (Q733195), I seem to remember (I speak from memory) that it is more of a confederation (Q170156) of countries (or planets, I'm not sure) that includes countries with different types of government. For example, there is Naboo that seems to me to be a principality (at least it is ruled by a princess).

--Fantastoria (talk) 08:37, 12 July 2023 (UTC)[reply]

I have started by substituting fictional kingdom (Q96196475). --Fantastoria (talk) 12:05, 11 August 2023 (UTC)[reply]

In the event that another user recovers the value, what reason could be given to discontinue it? --Fantastoria (talk) 12:14, 11 August 2023 (UTC)[reply]

I'd argue that we want to avoid the creation of unnecessary items and redundancy. Probably we should set up a page for the modelling of fictional administrative entities (just like we have the page Properties for fictional characters for characters). This should follow the general model for administrative entities closely, but there may be peculiarities of fictional administrative entities that may be discussed/documented there. -- Valentina.Anitnelav (talk) 15:27, 15 August 2023 (UTC)[reply]

It seems to me that I have already replaced most of them. In many cases, it was redundant because the items were already instances of fictional country (Q1145276) and had basic form of government (P122). I've also created a warning to discourage the use of those values as instances. --Fantastoria (talk) 06:59, 21 August 2023 (UTC)[reply]

Ps. I have created the value applies to entities of another class (Q121462401) as reason for deprecated rank (P2241) because I have seen that from time to time there are uses of fictional country (Q1145276) that do not conform to works of fiction (see Bookland (Q669741) and Belgorod People's Republic (Q118629252)). --Fantastoria (talk) 10:45, 28 August 2023 (UTC)[reply]

There is also magocracy (Q3843396), which is an actual fictional form of government. It is a currently an instance of form of government (Q1307214), which is obviously not right but is it worth making a "fictional form of government" item and so on, or is there a better way? —Xezbeth (talk) 11:39, 29 August 2023 (UTC)[reply]

In a first quick approach to the problem, we could consider magocracy (Q3843396) as a particular case (or subclass) of oligarchy (Q79751) (on the level of fiction, of course), so that we would deinstantiate the former, and use the latter as the value of basic form of government (P122) for the cases involved and look for a way to relate it to magocracy (Q3843396) through qualifiers. --Fantastoria (talk) 10:53, 30 August 2023 (UTC)[reply]

character from a certain fictional universe[edit]

There is character from a certain fictional universe (Q112242467) as a super class for classes of entities restricted to a certain fictional universe (e.g. werewolf in Twilight (Q3831861)). Some of its sub classes are classes in the form of "character from work/universe/franchise x", used to express that a character is from a certain work or universe. This use is often redundant: we have present in work (P1441), media franchise (P8345) and from narrative universe (P1080) for this. Simba (Q1649583)instance of (P31)character from The Lion King (Q111889741) does not carry any more information than Simba (Q1649583)present in work (P1441)The Lion King (Q36479), media franchise (P8345)The Lion King (Q2901452), from narrative universe (P1080)The Lion King universe (Q73544451).

I propose to

Discourage the use of P31 to express that a character appears in a certain work / is part of a certain fictional universe or media franchise
Delete all "character from x" classes, after the information has been moved to other properties like present in work (P1441) or from narrative universe (P1080), media franchise (P8345) unless there is a) a sitelink or b) it is already in wide use in Wikidata or by a dedicated Wikidata project (Star Wars character, A Song of Ice and Fire character, Star Trek character)

I'd start with all smaller "character from x" classes (<50 instances) that are only about characters from a certain film (series), book or franchise: character from The Lion King (Q111889741), Toy Story character (Q111836280), Q119767999, Q115672082 and similar. Does somebody disagree? - Valentina.Anitnelav (talk) 12:03, 26 July 2023 (UTC)[reply]

@Valentina.Anitnelav: It seems quite reasonable. However, I have a doubt: for the exceptions that you propose, will it be possible to continue using those classes with P31 or will their use also be discouraged as indicated in point 1? --Fantastoria (talk) 12:13, 29 July 2023 (UTC)[reply]

I think in these (limited number of) cases they may continue to use P31 <character from x> as an established practice. There are 8 <character from x> classes with more than 100 instances, currently (Teenage Mutant Ninja Turtles character (Q73502689), character from Star Wars (Q33125444), A Song of Ice and Fire character (Q20086260), Star Trek character (Q73502569), list of James Bond characters (Q4359230), Scooby-Doo character (Q111819966), Game of Thrones character (Q20086263), Mario franchise character (Q33093124)). I would leave them alone, even though I support the discontinuation of P31 <character from x> also in these cases. If people editing these universes decide to discontinue the use of P31 <character from x> themselves, they may do so, of course. - Valentina.Anitnelav (talk) 15:17, 30 July 2023 (UTC)[reply]

Thus, we would have an asymmetric system in which, depending on the number of instances, the "characters of X" could be maintained (who says characters, says werewolf, elf, etc.)... for now, depending on the interest of a particular Wikiproject. I am also in favor of the most strict position, but I support that we start with the cases with fewer items. Could we have a query that would provide us with the list of the instances in question? Greetings. --Fantastoria (talk) 12:03, 11 August 2023 (UTC)[reply]

Here is a query for direct subclasses of character from a certain fictional universe (Q112242467), their number of sitelinks and (direct and indirect) instances: query. Not all of them follow the pattern <character from ...>.

Of course, I only consider the subclasses following the patterns <character from ...>, <... character> and similar for deletion; Paragon (Q96187632) is ok as this carries more information than just to express that a certain character appears in a certain universe/work/franchise. - Valentina.Anitnelav (talk) 15:22, 15 August 2023 (UTC)[reply]

Parallelism: As with characters, so with locations (eg Teenage Mutant Ninja Turtles location (Q108059638), Star Trek location (Q99281788), Star Wars location (Q108059698); going down in classification: astronomical location from the Star Wars universe (Q108059912), Star Trek planet (Q108059718), wizard school in Harry Potter universe (Q117379143)). What do these items contribute that do not present in work (P1441) or from narrative universe (P1080) with the corresponding generic fictional places as instance? --Fantastoria (talk) 10:41, 29 August 2023 (UTC)[reply]

I started with deletion requests part 1: All items for characters from a certain universe without any sitelinks and not in use (Wikidata:Requests_for_deletions#Q3375883 and the 3 following proposals). Regards, - Valentina.Anitnelav (talk) 11:35, 4 September 2023 (UTC)[reply]

Only seeing this section now. I fully agree with this, items like Q17478752 or Mario franchise character (Q33093124) have been bothering me for a long time now. Thank you for opening up the discussion. Jean-Fred (talk) 12:30, 5 September 2023 (UTC)[reply]

I agree, I think its a poor trend to have where all franchises need to have their own character items.StarTrekker (talk) 20:02, 5 September 2023 (UTC)[reply]

Fictional races and P31[edit]

Quick slightly tangential question: as I noticed the P31 change on Zelda characters, I noticed that we have Midna (Q1200192)instance of (P31)Twili (Q29586504). My understanding was to better change that to -ethnic group (P172)Twili (Q29586504) and tried instead -instance of (P31)fictional character (Q95074), however this triggered a constraint violation as "too broad". I went for fictional creature (Q2593744) but that’s hardly less broad… Usually I use fictional human (Q15632617), but these are not humans so not sure what to use there (I guess there is video game character (Q1569167) but I dislike such items, these characters might show up in half-a-dozen work type (manga, anime, game, TV show, board game…), and I think this is better expressed with present in work (P1441). Jean-Fred (talk) 10:38, 6 September 2023 (UTC)[reply]

A fictional race, especially one that is non-human, is not an ethnic group or a culture or even a "race", despite using the same word. In my opinion Midna (Q1200192) should be an instance of Twili (Q29586504) and video game character (Q1569167), with Twili (Q29586504) being a subclass of fictional humanoid (Q28020127) or a similar item that is more correct. —Xezbeth (talk) 07:47, 7 September 2023 (UTC)[reply]

Use of individual of taxon (P10241) is also possible.StarTrekker (talk) 20:48, 7 September 2023 (UTC)[reply]

I'm quite sympathetic towards this idea, but the current practice is to use instance of (P31) for taxa of fictional characters and there is no consensus to use individual of taxon (P10241), which would be quite a change to the current model (see see discussion from December 2022). Before endorsing the use of individual of taxon (P10241) we should revise our model, in my opinion, to avoid inconsistent modelling. A lot of items we keep due to "structural need" would be superfluous, then: That a character is a fictional lion could be expressed using instance of (P31)fictional character (Q95074) individual of taxon (P10241)lion (Q140), no need for fictional lion (Q27267085). It would be nice to have more people interested in this topic joining the discussion mentioned above. - Valentina.Anitnelav (talk) 09:09, 8 September 2023 (UTC)[reply]

Thanks all for chiming in. Clearly I was wrong to conclude that fictional races go to P172 rather than P31, and will not edit in that direction anymore. My general sense of this is that I don’t like much any modeling scheme that involves drilling down the P31/P279 tree (as in, Twili is so far down the tree that some querying is made really hard) ; but that’s not a debate for today :) Jean-Fred (talk) 09:32, 8 September 2023 (UTC)[reply]

Doubt with classification[edit]

We currently have the following:

fictional human settlement (Q108078133)

fictional neighborhood (Q60644691)

fictional ghost town (Q63040328)

fictional farm (Q70130710)

fictional suburb (Q96270074)

fictional locality (Q108059277)

fictional city (Q1964689)

fictional lost city (Q17004377)

fictional city-state (Q65630486)

fictional underground city (Q80196698)

fictional capital city (Q96146366)

fictional planned capital city (Q91338528)

Middle-earth locality (Q20671864)

fictional village (Q82551957)

fictional municipality (Q101497842)

fictional commune (Q106541075)

fictional town (Q106921111)

fictional hamlet (Q107642546)

fictional settlement (Q112117312)

It strikes me that fictional settlement (Q112117312) is a subclass of fictional human settlement (Q108078133). Wouldn't it be more appropriate for fictional settlement (Q112117312) to be the parent class of this classification and for fictional human settlement (Q108078133) to be just a subclass of fictional settlement (Q112117312)? --Fantastoria (talk) 16:00, 11 August 2023 (UTC)[reply]

Yes, of course. Infovarius (talk) 22:49, 12 August 2023 (UTC)[reply]

Done --Fantastoria (talk) 07:15, 18 August 2023 (UTC)[reply]

Fictional Entity vs Fictional Object[edit]

Are there fictional entities that are not fictional objects? From the definitions of its real analogues, I understand that the separation into two items is unnecessary in fiction: one is everything and the other everything that something can interact with. What can't something interact with in fiction? I propose that we merge them. --Fantastoria (talk) 12:32, 3 October 2023 (UTC)[reply]

That would depend on what universe something is in. I wouldn't support a merg personally.StarTrekker (talk) 15:52, 3 October 2023 (UTC)[reply]

What about subjects? Or essences from parallel universe? Or a MacGuffin (Q488733)? --Infovarius (talk) 20:30, 5 October 2023 (UTC)[reply]

Are there fictional entities that are not fictional objects? Certainly. Offhand I can think of fictional religions, fictional folk stories, fictional colors... there are plenty of intangible entities in fiction. Omphalographer (talk) 20:31, 21 October 2023 (UTC)[reply]

kinship to subject[edit]

How should i model the kinship between Mojo Jojo (Q3253091) and his "children" Butch (Q99603780), Brick (Q99604021) and Boomer (Q99603515)? I know adopted son (Q20746725) is a thing but i'm not sure if that's appropriate in this case Trade (talk) 18:17, 17 January 2024 (UTC)[reply]

For anyone who haven't watched the show, read here for context--Trade (talk) 18:17, 17 January 2024 (UTC)[reply]

I had a look at Pinocchio (Q6502703): here we used relative (P1038)Mister Geppetto (Q1428120)kinship to subject (P1039)foster father (Q20747105) and significant person (P3342)Mister Geppetto (Q1428120)object has role (P3831)creator (Q2500638). Maybe the relationship between Mojo Jojo and his "children" may be modelled similarly. - Valentina.Anitnelav (talk) 14:19, 20 January 2024 (UTC)[reply]

Personally I'm inclined to use the usual parent properties in cases like these. A qualifier like non-biological father (Q108759243) is what I would use.StarTrekker (talk) 21:12, 20 January 2024 (UTC)[reply]

Characters hundreds or thousands of years old[edit]

Some characters die at hundreds or thousands of years old. However, there is a restriction that warns that the value of age of subject at event is greater than 135. How could we set an exception for these cases? E.g. Elros. --Fantastoria (talk) 09:56, 22 January 2024 (UTC)[reply]

Good question. Is this propert restricted to humans? Because there are quite a few animals that can live longer than that.StarTrekker (talk) 12:49, 22 January 2024 (UTC)[reply]

I added it to Project chat. --Fantastoria (talk) 09:08, 5 February 2024 (UTC)[reply]

I asked in Project chat and they told me that exceptions had to be made one by one. However, I don't know how to do that (my attempt was unsuccessful). Could you tell me how I can do it? Thank you. --Fantastoria (talk) 10:33, 9 February 2024 (UTC)[reply]

I had a look at date of death (P570) and I think you added the exception correctly. Not sure why it does not have any effect. Maybe because this property is used as a qualifier? To be honest: I would not worry about it too much. The constraint is marked as a "suggestion constraint".

Btw: how did you get the link to a specific claim? - Valentina.Anitnelav (talk) 16:28, 9 February 2024 (UTC)[reply]

@Valentina.Anitnelav: Sorry, I don't understand your question.

Otherwise, that was the only way to implement the exception that occurred to me, but it didn't work. I know it's a suggestion, but I'm trying to find a way to 'overcome' them, both the required and the suggested ones, so that no warning appears in the item. --Fantastoria (talk) 11:22, 10 February 2024 (UTC)[reply]

@Fantastoria: I mean: How did you get this link: https://www.wikidata.org/wiki/Property:P3629#P3629$0F8B83BF-7132-464A-9675-1DC2B6DC6EA6 (a link to a specific statement). Or: how did you get the "identifier" for the specific statement (0F8B83BF-7132-464A-9675-1DC2B6DC6EA6, in this case). - Valentina.Anitnelav (talk) 12:51, 10 February 2024 (UTC)[reply]

@Valentina.Anitnelav: Ok. In the Elros item I clicked on the warning flag. That caused a notice (pop-up?) to appear. I clicked where it says 'range constraint' and copied the link :). --Fantastoria (talk) 09:23, 11 February 2024 (UTC)[reply]

Real vs Fictional[edit]

I am with the case of monarchs, but I think it can be extended to all cases. For the value of occupation (P106), do we use the real value (monarch (Q116)) or the fictional value (fictional monarch (Q112245172))? Do we use real or fictional values? --Fantastoria (talk) 10:14, 22 January 2024 (UTC)[reply]

I don't know, maybe it depends on if the fictional person is based on a real figure or rules a real country?StarTrekker (talk) 13:54, 22 January 2024 (UTC)[reply]

In my opinion we should use monarch (Q116). fictional monarch (Q112245172) may be useful to group titles in fiction (e.g. King of Númenor (Q18729869), Emperor of Cragmites (Q116807366)) but it should not be used in statements about individual fictional characters. For a related discussion see Wikidata_talk:WikiProject_Narration#fictional_occupations_that_are_real_occupations. Kind regards, - Valentina.Anitnelav (talk) 21:26, 23 January 2024 (UTC)[reply]

According to the discussion, I think there is some consensus to use real items as values when they exist instead of the fictional analogues. I have added it to the discussion of Table of characters. --Fantastoria (talk) 09:05, 5 February 2024 (UTC)[reply]

Whose idea was it to add dozens of "fictional version of occupation" items to Wikidata? Because i would frankly rather see them gone Trade (talk) 23:51, 6 February 2024 (UTC)[reply]

@Trade:, without going any further, I added fictional monarch (Q112245172) and fictional emperor (Q112245130) and I don't think they should go anywhere because they serve to structure the tree of fictional entities. I agree with you that we should delete any fictional profession analogous to a real one that does not have a structuring function in the class tree. --Fantastoria (talk) 12:05, 7 February 2024 (UTC)[reply]

I’m not sure I understand your answer ; you’re saying that fictional monarch (Q112245172) is useful because we can have King of Númenor (Q18729869)subclass of (P279)fictional king (Q16658189)subclass of (P279)fictional monarch (Q112245172)? But surely, we could skip that one and go straight to fictional head of state (Q70716623)? Jean-Fred (talk) 21:25, 8 February 2024 (UTC)[reply]

@Jean-Frédéric: actually, I haven't say exactly what you mentioned. I initially asked which was preferable: real values or fictional values. Established that real ones are preferable, I justified the existence of these 'fictional values' in structural utility, not as simple links between classes or between classes and items. Here is the classification of fictional head of state (Q70716623) as it stands now:

fictional head of state (Q70716623)

fictional president (Q112245149)

fictional President of the United States (Q73374532)

fictional President of the Confederate States of America (Q96377891)

fictional monarch (Q112245172)

fictional king (Q16658189)

King under the Mountain (Q3930941)

king of Rohan (Q7468941)

King of the Reunited Kingdom (Q16594772)

King of Númenor (Q18729869)

King of Arthedain (Q33134464)

King of Arnor (Q33134656)

king of Gondor (Q35623773)

King of the Andals, the Rhoynar, and the First Men (Q61811086)

fictional queen (Q16658194)

fictional emperor (Q112245130)

Padishah Emperor (Q4342084)

Emperor of the Reigning Dynasty (Q108456873)

Emperor of Cragmites (Q116807366)

I admit that I didn't explain myself in depth, but I justify myself by saying that it was just a simple response to a simple statement. I don't want to go into too much detail either because it wasn't the reason for this chat. I will say that, of course we can go to fictional head of state (Q70716623), but we can also go over it since it is still a fictional profession (regardless of whether we call it king or president, for example). Therefore, my explanation is that these classes group other classes of the same type and level and have structural utility as seen above. --Fantastoria (talk) 10:15, 9 February 2024 (UTC)[reply]

Personification[edit]

I have come across personification (Q207174) which links the classifications of fictional entities and mythological entities. I think we should have both classifications separate. However, after a while the only thing I can think of is that this item is not a subclass of fictional humanoid (Q28020127), since there are deities (thus, mythological entities) that are instances of personification (Q207174). On the other hand, there are fictional characters that are personifications. Would the latter justify the creation of 'fictional personification'? In my opinion, yes. What do you think? --Fantastoria (talk) 13:20, 7 February 2024 (UTC)[reply]

Does a personification have to be humanoid?StarTrekker (talk) 20:01, 8 February 2024 (UTC)[reply]

@StarTrekker: a personification must be (I would say by definition) a humanoid, but that is not what I am suggesting. The idea is whether the classification of what is real and the classification of what is fictional have to be separated from each other or not. In fact and indirectly, establish this position as one of the objectives of this Wkiproject. Hence I propose the creation of this 'fictional personification' item. And, if I did it, it is because I had some doubts, but I have now cleared them up and I will create it in the next few days when I review the classification. --Fantastoria (talk) 10:27, 9 February 2024 (UTC)[reply]

I agree to separate the concept of personification as a device from the concept of a personification as a class of characters. I would suggest to make it a direct subclass of imaginary character (Q115537581) and use it for characters from all types of origin, be it myth, folk belief, legend, fantasy novels, etc. To mark a character as fictional or mythical you can add an additional statement instance of (P31)fictional character (Q95074). To make own items <fictional personification>, <mythical personification>, <legendary personification> ... does not scale well. - Valentina.Anitnelav (talk) 16:40, 9 February 2024 (UTC)[reply]

But that solution depends on us implementing or developing the idea of 'imaginary entity' or 'fictitious entity' (to generalize beyond characters). I read your presentation on this matter in a previous chat and, although I agree with much of what you say there, I fear that in the long run we would have an overclassification. I think, for the moment, that it is preferable to maintain independent classifications for fictional or mythological entities rather than having a mixed, half-constructed system. (For characters yes, but places?; depending on the type of characters.) However, I am open to developing the idea because there are entities that are simply imaginary: neither fictional nor mythological. --Fantastoria (talk) 11:47, 10 February 2024 (UTC)[reply]

Finally I have depreciated fictional humanoid (Q28020127) as value of subclass of (P279) into personification (Q207174) and have not created a fictional analogue of this because there is no direct subclass related to fictional. --Fantastoria (talk) 09:44, 25 February 2024 (UTC)[reply]

Superhero film characters et al[edit]

instance of (P31) currently has a none-of constraint (Q52558054) for superhero film character (Q63998451) and similar items, but is there a discussion of that somewhere? I don't have anything to link to upon being reverted besides offering my own opinion, and I remember reading a discussion about this or something similar. —Xezbeth (talk) 05:11, 26 March 2024 (UTC)[reply]

The constraint suggests to use character type (P9071) instead. I can't remember a serious discussion about using character type (P9071) over instance of (P31), but I'm in favour to use instance of (P31) only for basic classes (e.g. fictional, literary, mythical etc. character) and use other properties if possible. Do you see an issue with using character type (P9071)superhero film character (Q63998451)? Kind regards, - Valentina.Anitnelav (talk) 16:07, 26 March 2024 (UTC)[reply]

BTW: Do we need superhero film character (Q63998451) at all? Isn't instance of (P31)film character (Q15773347) character type (P9071)superhero (Q188784) enough? - Valentina.Anitnelav (talk) 16:10, 26 March 2024 (UTC)[reply]

Just realised that I didn't explain myself properly. I was changing it to film character (Q15773347) on various items and was reverted on Kingpin (Q1753322). My preference is to not use them at all. character type (P9071) works for most superhero film characters but it wouldn't for say action film character (Q116984649). I don't think any of them are useful, Jack O'Neill (Q719661) for example just looks wrong. —Xezbeth (talk) 17:56, 26 March 2024 (UTC)[reply]

I agree. Those items "action film character", "adventure film character" etc. without sitelinks should be deleted in my opinion (see also my comment at Wikidata_talk:WikiProject_Narration#adventure_film_character_(Q66808903)). We can preserve information by using genre (P136)action fiction (Q1762165). @Infovarius: Is it ok with you, too, or do you disagree? - Valentina.Anitnelav (talk) 10:09, 27 March 2024 (UTC)[reply]

Are you sure that P136=Q1762165 would not violate some constraint? --Infovarius (talk) 21:44, 27 March 2024 (UTC)[reply]

If fictional characters are not allowed subjects of P136, yet, I'd broaden its domain to include them. - Valentina.Anitnelav (talk) 10:18, 7 April 2024 (UTC)[reply]

I agree with the solutions proposed in this topic. Nothing would stop us from continuing to refine the system with Marvel superhero movie character and the like as is already done with literary works. Perhaps we need firm guidelines (not guidelines chiseled in stone) that guide us and recommend optimal ways to orderly disaggregate the same information present in those "compact" items.

Kingpin (Q1753322) is film character (Q15773347) and fictional human (Q15632617), appears in superhero film (Q1535153) and is part of Marvel Cinematic Universe (Q642878), limiting ourselves to the proposed example. I think the same can be done with all fictional characters. --Fantastoria (talk) 09:55, 29 March 2024 (UTC)[reply]

I didn't realize there was also Jack O'Neill. Jack O'Neill (Q719661) is film character (Q15773347) and fictional human (Q15632617), appears in action film (Q188473) and adventure film (Q319221) and is part of Stargate (Q11927). --Fantastoria (talk) 10:08, 29 March 2024 (UTC)[reply]

I agree. Just to be sure I'd point out that to express that Kingpin (Q1753322) appears in superhero film (Q1535153) we should use genre (P136), not present in work (P1441). - Valentina.Anitnelav (talk) 10:22, 7 April 2024 (UTC)[reply]

present in work (P1441) is ruled out because superhero film (Q1535153) is not a work but a genre. I agree. On the other hand, we could add the new usage to use genre (P136), as you said... or propose the creation of a new specific property for this case: present in works of genre. Whichever is most feasible. --Fantastoria (talk) 08:05, 12 April 2024 (UTC)[reply]