Property talk:P8345

Documentation

media franchise
this creative work belongs to this media franchise

Represents

media franchise (Q196600)

Data type

Domain

work (Q386724), virtual band (Q3736859), fictional entity (Q14897293), class of fictional entities (Q15831596), cast (Q15267437), amusement park (Q194195) or model series (Q811701)

Example

Super Mario Bros. (Q11168) → Mario (Q4803535)
Super Mario Bros. (Q535550) → Mario (Q4803535)
Dragon Ball Z: Budokai Tenkaichi 2 (Q2712053) → Dragon Ball (Q2020)
Dragon Ball GT (Q735669) → Dragon Ball (Q2020)

Tracking: usage

Category:Pages using Wikidata property P8345 (Q124317758)

Vague scope

Latest comment: 4 years ago16 comments4 people in discussion

Please for what type of items should this property be used? For characters and other in-world entities? For books/films/other works? What to do if some item is about fictional universe and franchise simultaneously? Or a game series and franchise? --Infovarius (talk) 09:37, 19 June 2020 (UTC)Reply

Er, it’s filed as Wikidata property related to creative works (Q18618644) and has a -property constraint (P2302)subject type constraint (Q21503250)relation (P2309)instance or subclass of (Q30208840)class (P2308)creative work (Q17537576)… What’s vague about that? Jean-Fred (talk) 10:11, 19 June 2020 (UTC)Reply

If an item mixes franchise, fictional universe and film/book/game/etc. series, it should probably be divided into separate properties. I think in most cases, someone simply added additional instance of (P31) statements to an existing franchise/universe/series item because they weren't aware that these concepts need separate items. Of course there are probably lots of cases where we only have items about the films/games/books in the series and nothing else from that franchise, in which case the need for a separate franchise item isn't quite as obvious.

As to the scope, I think fictional entities should indeed get a franchise statement. I think Jean-Fred's property proposal showed that other properties aren't all that well suited for this kind of statement. But there are other cases which we might want to discuss. For example, I think both individual books/games/films and the respective book/game/film series should get franchise statements. But I've hesitated to add the statement to seasons and individual episodes of television series. --Kam Solusar (talk) 14:46, 19 June 2020 (UTC)Reply

I think there might be some cases where it’s fine to have items with both instance of (P31)=fictional universe and media franchise − although I can’t think of an example.

Re: seasons and episodes: that’s a very good point, I did not think of it. I agree with you that the franchise on an individual TV series episode feels a bit weird. My first though was, let’s not add it if it’s part part of (P361) another work which already has the franchise statement − and then remembered that I certainly added it to both games and game series so that rule of thumb does not really work… Jean-Fred (talk) 17:43, 19 June 2020 (UTC)Reply

"it’s filed as Wikidata property related to creative works (Q18618644) and has a -property constraint (P2302)subject type constraint (Q21503250)relation (P2309)instance or subclass of (Q30208840)class (P2308)creative work (Q17537576)… What’s vague about that?" That's quite different from how it's currently used. --Trade (talk) 01:24, 12 July 2020 (UTC)Reply

At the time I wrote that, this was not the case − fairly sure the property was (almost) exclusively used on works, and not on in-universe entities (like characters − which I assume is what you hint at) :-) Jean-Fred (talk) 20:48, 13 July 2020 (UTC)Reply

Thanks to Kam, you were the closest to answer my question. I'll try to conclude. We can(should) add P8345 to characters and books/films/works and games series and fictional universe itself and probably to individual seasons and episodes. --Infovarius (talk) 15:17, 13 July 2020 (UTC)Reply
@Infovarius: I see now that my answer was curt and somewhat unhelpful, apologies − I guess I got a bit annoyed because I spent hours penning Wikidata:Property_proposal/media_franchise, and while of course it may not have been perfect (I definitely had not foreseen the TV episode situation for example), I did believe that the initial scope was clear enough to not be deemed "vague". Sorry about that. :-) Jean-Fred (talk) 10:27, 15 July 2020 (UTC)Reply

I do not know of any online database that uses "franchise" to describe fictional characters and entities. --Trade (talk) 15:44, 13 July 2020 (UTC)Reply

Well, there aren't all that many online databases focusing on characters to begin with (if we leave Fandom/Wikia wikis aside, which vary wily in quality, structure and scope). en:Template:Infobox character does have a "franchise" parameter, which makes this property useful to populate that field in the infobox.

Giant Bomb (Q1657282) apparently only lists the first appearance in its infobox and then has a long list of all appearances - so they don't seem to have a way of connecting characters to the respective franchises that they are part of. I.e. you can't tell just from the info in the infobox that Pikachu is part of the Pokemon franchise or that Luke Skywalker is a Star Wars character. Comic Vine (Q24688914) lists characters' first appearances and the "publisher". I guess this makes sense for a site focused on comic books, where publishers like DC/Marvel/etc. usually have big, shared universes where characters may appear anywhere. Though it doesn't seem to work too well for characters from other media that appeared in publications by several publishers. Sites like marvel.com (Q24258919) or Star Wars Databank (Q3968343) are solely focused on their own franchises anyways, so there's no need to state the obvious in their entries.

It also seems that other databases also don't have structured information about the fictional universes/continuities/canons that the characters/entities are a part of. But I think that from narrative universe (P1080) is valuable data we should keep, even if other databases don't store the same information. Just like this property. --Kam Solusar (talk) 16:42,13 July 2020 (UTC)

Do you have any example where a character's fictional universe belongs to a different media franchise than the character?

"where publishers like DC/Marvel/etc. usually have big, shared universes where characters may appear anywhere." you forgog to menion the numerous Marvel "subuniverses"

So how many media franchises do you propose we make for DC, Marvel, Disney, Cartoon Network and Nickelodeon?--Trade (talk) 17:10, 13 July 2020 (UTC)Reply

Sorry, I'm not entirely sure about what kind of data model you are argueing for. Do I understand you correctly that you mean media franchise (P8345) is unnecessary on items of fictional entities if they have a from narrative universe (P1080) statement and that universe item already has a media franchise (P8345) statement?

As to the number of potential franchises: I don't know, but I'd say we go with whatever external sources define as franchises. These things aren't an exact science and sometimes franchises can overlap or be part of a bigger franchise. --Kam Solusar (talk) 17:13, 14 July 2020 (UTC)Reply

I would be happy with adding media franchise (P8345) to fictional entities, regardless of whether they have from narrative universe (P1080): if I want to get all Gundam (Q732120) characters, jumping through Cosmic Era (Q2755142) and the likes is just a hassle. Also, while I see the point of fictional universes for say, Harry Potter or Dune, or Lord of the Rings, I see little point in creating such universe items for say, "James Bond" or "Hitman" or "Splinter Cell" − any franchise where there is not that much to say about ("it’s a universe more or less like ours") (but I know I’m somewhat biased against this universe concept, as I hinted in the original proposal). Jean-Fred (talk) 20:08, 14 July 2020 (UTC)Reply

9/10 times the media franchise would be of roughly the same value as the fictional universe (Harry Potter belongs to the Hary Potter franchise and the Harry Potter universe). My point was that using franchise on fictional characters would be of limited usefulness. --Trade (talk) 20:34, 14 July 2020 (UTC)Reply

Yes, if assuming there is such a fictional universe item, which I don’t think is the case right now − nor do I believe there necessarily should be: consistent with en:Fictional_universe#Universe_vs_setting, not every setting can be considered a fictional universe. Also, numbers-wise, there are currently 417 instances of fictional universe (Q559618) (even though there arguably has been work on this since at least 2014, when from narrative universe (P1080) ; while there are 491 media franchise (Q196600).

So while I do see your point and agree there might be some/a lot of overlap, I think that it’s better to keep the modeling consistent.

Jean-Fred (talk) 10:27, 15 July 2020 (UTC)Reply

"Also, numbers-wise, there are currently 417 instances of fictional universe (Q559618) (even though there arguably has been work on this since at least 2014, when from narrative universe (P1080)" It would seem that the people using this property are only interested in writing about a few selected universes. Can't say that's surprising.--Trade (talk) 10:59, 15 July 2020 (UTC)Reply

Universal Videogame List (Q64447538) appears to − see https://www.uvlist.net/characters/ where each character belongs to a Group, which sometimes maps to franchises more than to series.

Also, as I had written in Wikidata:Property proposal/media franchise, Wikidata does implicitly, with items like Mario franchise character (Q33093124), Q29006203 or Q17478752. Jean-Fred (talk) 20:58, 13 July 2020 (UTC)Reply

Criteria

Latest comment: 4 years ago9 comments3 people in discussion

I believe we need some better guidelines for what does and does not count as a media franchise

Does a video game with a digital-only soundtrack album release count? (Our Darker Purpose)
Does a television show with multiple physical album release count? (Panty & Stocking with Garterbelt)
Does a video game series with a tie-in comic series count? (Carmageddon & Call of Duty)

I'm sure others can think of more examples. --Trade (talk) 07:20, 17 September 2020 (UTC)Reply

@Jean-Frédéric:--Trade (talk) 08:11, 18 September 2020 (UTC)Reply

Media franchise isn't a term with an exact definition that we can use to verify if something is a franchise or not. When in doubt, it would probably be best to check whether any external sources call it a franchise. Allthough I guess we don't have to be too strict, since it's just a statement we use to state that works belong together. But one work plus a accompanying soundtrack or novelization seems a bit low. Personally, I wouldn't create items for "franchises" with less than, say, eight or ten works/things or so. --Kam Solusar (talk) 06:50, 21 September 2020 (UTC)Reply

While we are at it at what point do you think something should be considered a video game series? --Trade (talk) 07:44, 21 September 2020 (UTC)Reply

Hmm, I'd say usually 3 games or more. Two games don't really seem like a series to me, though there are probably some game databases that have entries for some series with just 2 games. When there's only two games, or one game plus an addon, remaster, different version etc., I'd only create an item for a series if there's already an external ID to justify it. --Kam Solusar (talk) 17:06, 21 September 2020 (UTC)Reply

Yeah, I tend to agree − I’ve been guilty of creating series items quite liberally, including for just two games (like “Foobar” and “Foobar 2”) and I now start to regret this. We have follows (P155)/followed by (P156) and they’re probably fine to use without a series item.

I would lean towards opposing a series item for a game + a remaster (“Foobar” & “Foobar Remastered”) ; and for a game + a DLC − the relationship between the two can also just be modeled directly (using, respectively, based on (P144) and our upcoming “expansion of”) without needing what boils down to a “container item”.

And that’s why I’d start being a bit wary of using external IDs to justify these series items: often enough, it seems that the sole reason for it is that their data model just does not allow relationships between games. Take MobyGames: they have no way to link together a game and its DLC, or a game and its remaster, or a game and its sequel ; so they fall back to grouping them (using the game group entity). It’s not only MobyGames: Gamekult and Giant-Bomb only have their “Franchise“ object too). Surprisingly, I have found GameFAQ more advanced there: they are able to express that The Only Traitor (Q91333442)expansion forThe Final Station (Q40888481) or Night Trap - 25th Anniversary Edition (Q77265302)remaster forNight Trap (Q1323428), and they don’t need create a franchise “container entry” to group them − while Mobygames has groups `night-trap-series` and `final-station-series` which make little sense to me.

Jean-Fred (talk) 10:02, 22 September 2020 (UTC)Reply

English Wikipedia’s criteria for inclusion in lists of multimedia franchises (Q17089095) seem reasonable enough to just reuse them for now: a franchise must have works in at least three forms of media, and must have two or more separate works in at least two of those forms of media.

And indeed, we should heed external sources.

It is indeed about grouping works together ; but I would say that is not necessary when these works are all linked together anyhow: a soundtrack album and the work via soundtrack release (P406) ; a tie-in comic with based on (P144)/plot expanded in (P5940), etc.

Jean-Fred (talk) 13:32, 21 September 2020 (UTC)Reply

The item you listed is a list of multimedia franchises that are notable enough for their own Wikipedia article. As WD have different notability criteria from WP we should be less strict about it. --Trade (talk) 12:03, 22 September 2020 (UTC)Reply

Thoughts on SCP (Q68466327) and Slender (Q68468406)?@Jean-Frédéric:--Trade (talk) 22:15, 19 October 2020 (UTC)Reply

Suggestion to expand the property to "media franchise or series"

Latest comment: 2 months ago10 comments3 people in discussion

Looking at the discussions above and the proposal page, where we use franchise and series interchangeably, it seems to me that the only way to make this property work as proposed is to change the label to a broader variant. If you look at how it was presented in the proposal from the beginning - the reason given there is that fictional characters are non-schematically filled in with 'part of' or 'part of the series'. But just adding media franchises doesn't solve the problem. Not always a cycle of works is called a franchise, but sometimes these cycles of works are gigantic compared to some tiny franchises. I think the right thing to do is to rename this to 'media franchise or series' with some aliases and start moving the series values from 'presented in' to this property.

I'm currently running into this problem on Wikipedia when I try to add a parameter to the fictional character infobox to specify a series of works in which the character is presented. Many characters have this property empty because they're not part of the media franchise, but part of a series within a medium - television or literature or cinema or video game only. So I have to link param with 'presented in work' instead, but there it looks like this: Q3244512#P1441. I only want to get series of work, not each work specifically. So I think the right thing to do would be to modify this to include both media franchises and cycles/series of works.

Btw, people on enwiki are currently also changing the video game categories from 'franchise' to 'series' [1], which just proves that the difference between the terms is too blurred to maintain both as we do now. Otherwise we'd have to go back to part of the series (P179) or create another property. Solidest (talk) 22:01, 2 September 2024 (UTC)Reply

Just to make sure I understand, this is specifically about the use of that property on characters (and maybe other in-universe, fictional concepts)? Asking because this P8345 was never formally expanded to cover fictional stuff. I initially meant to cover works, ie books, shows, games, movies etc. (I later discovered that also included things like amusement park theme rides ;-þ), and while I did have characters in my cross-hairs, it was effectively left out. Jean-Fred (talk) 07:22, 3 September 2024 (UTC)Reply

Yeah, it's specifically about the fictional entities and characters that are the problem. That it doesn't suit them to be labelled "part of the series" or "part of" of series/franchise, and "presented in work" is also too cluttered, but this property could suit them better I suppose. But I didn't have thought that for specific releases both "part of the series" and "media franchise" are used at the same time, making the renaming difficult. And enwiki's decision to replace word in categories doesn't make things easier either. The issue is that the difference between single-media series and cross-media series sometimes insignificant, but we have them filled in completely different and unrelated properties. Solidest (talk) 12:06, 3 September 2024 (UTC)Reply

Just so I understand: You would like the Wikipedia Infobox on characters to have a field "where they come from". In the case of Harry Potter (Q3244512), what would be the desired value? Harry Potter (Q8337) (the book series) ; Harry Potter (Q216930) the film series ; Harry Potter universe (Q5410773) ; Wizarding World (Q30739117) ? Jean-Fred (talk) 12:13, 3 September 2024 (UTC)Reply

I originally thought that anything that could be positioned as a 'franchise or group/series of works' should be placed there in chronological order. Or either with an order based on size, with the first entity being a cross-media franchise (or just use a preferred rank on a bigger one). Although now it seems to me that a chronological order with series type qualifiers would still work better, if this variant turns out to be viable. Solidest (talk) 12:26, 3 September 2024 (UTC)Reply

Another solution would be to rename this property to 'part of the media franchise' to prevent it from being freely used anywhere - since seeing what your ExLudo.js script generates in Wizarding World (Q30739117), I doubt it was intended to be filled that way, and characters would probably clutter it up even more. Perhaps a new property for characters would be needed - 'presented in series of works'. Solidest (talk) 12:12, 3 September 2024 (UTC)Reply

Yes, the current usage of P8345 is far wider than what I anticipated (which was essentially works and maybe characters in the future). It is widely used for:

creative works I had not had in mind, like collectible card game (Q734698) or album (Q482994) (mostly soundtrack albums?)
non-creative works, like toyline (Q57663626) or amusement ride (Q1144661)
creative works that are sub-parts of a work, like television series episode (Q21191270)
creative works where it is debatable whether part of a franchise (fanfics or fan-films)
fictional entities, like places or concepts (eg Quidditch position (Q21130120))
fictional "meta-entities", like level (Q1046315)

I don’t think 1/ is a problem., just a sign that media franchise are bigger than I thought :) I think I am also okay with 2/.

I don’t think 3/ makes a lot of sense, and just makes queries messy. I would consider deprecating that use, but I don’t know if we can be consistent on what is "enough of a work" to belong to a franchise

I think 4/ is problematic in terms of definition because “media franchise” implies intellectual property ownership, which by definition does not fit fanfiction. I also think 4/ and 5/ are confused with the use of either takes place in fictional universe (P1434) or from narrative universe (P1080). 6/ is the same beast, should be deprecated.

The funny thing is, I still think it makes sense to have Hermione Granger (Q174009)media franchise (P8345)Wizarding World (Q30739117), but it bothers me to have Godric's Hollow (Q3288068)media franchise (P8345)Wizarding World (Q30739117). I don‘t know how to explain it. I think the difference, perhaps, is that characters move between franchises in crossover fiction? Not sure. :) Jean-Fred (talk) 12:43, 3 September 2024 (UTC)Reply

Yes, and the main question under discussion is in which of these cases it makes sense to use media franchise and series of works at the same time, and where to fill them in:

creative works - ✔ widely used. standard filling scheme media franchise (P8345) + part of the series (P179)}
non-creative works - ? may exist, but I don't know of any cases where such a thing has a series of works.
creative works that are sub-parts of a work - ✔ sort of, since the franchise can be filled (I don't like this either, and support the addition of constraints), and part of the series (P179) - is a standard property for TV series
fan creative works - ? usually fanfic series don't exist, but potentially they could, and then it's filled in via the same properties.
fictional entities - ✔ yes, and the big question here is how to avoid putting series in present in work (P1441)
fictional 'meta-entities' - ? sort of, specific examples are needed as I don't really understand the scheme of filling this in, but potentially both a franchise and a video game series could be filled in, so I guess the filling in is the same as with fictional entities.

This makes two groups - (1) creative works, where the pattern is now media franchise (P8345) + part of the series (P179), and (2) fictional entities presented in creative works, where the pattern is predominantly present in work (P1441), with rare instances of P8345 and P179.

By the way, regarding the parallel to franchises - fictional universes - we also have a division into two groups for them: takes place in fictional universe (P1434) for creative works and from narrative universe (P1080) for fictional entities. So maybe the "present in franchise or series of works" property for fictional entities is really what we need to resolve this. Solidest (talk) 15:28, 3 September 2024 (UTC)Reply

Sorry I don't understand the problem except a "long" list of works in Q3244512#P1441 which looks quite normal yet. Also I see a problem in expanding the scope. Now we have 'some series' P8345 'franchise'. How this relation would look after expanding? --Infovarius (talk) 12:44, 4 September 2024 (UTC)Reply

The problem of the long list is more a problem of the intersection between the concepts of 'creative work' property and 'fictional entity' property. For fictional universes we have the two separate properties mentioned above. In 'present in work' you will only list creative works, not fictional entities. Therefore, such a division makes sense here as well. But now we only have a generalised Wikidata property related to works of fiction (Q22984512) descriptor, rather than the two kinds (that would be useful for properties like takes place in fictional universe (P1434) and from narrative universe (P1080) and present in work (P1441) and others), so perhaps this can be skipped for now and the separation left for later, if it's ever be needed.

Another problem with not dividing properties into for-creative-work and for-fictional-entities is staying with the 'video game entity - part of the series - video game series entity' scheme when we have a second property which will also be called a 'media franchise or series'. Presumably 'video game series' would have to be filled in both 'part of the series' and in 'media franchise or series', which is not really the best solution. Solidest (talk) 13:25, 4 September 2024 (UTC)Reply

Add topic