Wikidata talk:WikiProject Movies
- 1 Please join the discussion
- 2 Modelling of hierarchical animation creation
- 3 Fast way of adding episode names to season items
- 4 Entering curated movie lists
- 5 Columbo (1): seasons
- 6 Columbo (2): character types
- 7 Bechdel test based on unreliable source
- 8 Random film items to improve
- 9 movies parts of a film festival edition (Q27787439)
Please join the discussion
Modelling of hierarchical animation creation
Please give me an advice how to better model Happy Merry-Go-Round (Q4109736)? It has 48 annual vypusks (issue? release?), each is published as a whole animation film (~10 min) but consists of 2-4 sujets. Each sujet is also distributed as a separated short film (example: Antoshka (Q7475124)). If I create item per each issue how should I connect sujet-issue-whole franchise? --Infovarius (talk) 15:53, 21 June 2019 (UTC)
- @Infovarius: I think this could be modelled as a anthology film (Q336144); to connect the sujets to the film you could use part of (P361) and there is also number of parts of this work of art (P2635) for the number of short films in a vypusk. (I did something similar with Fantasia (Q943192)) . Valentina.Anitnelav (talk) 07:30, 7 August 2019 (UTC)
Fast way of adding episode names to season items
- @Gonnym: You can use SPARQL to check for existing labels (https://w.wiki/5Ur or shorter: https://w.wiki/5Uu) and use QuickStatements to add episode names (with something like this
Q11096052|Len|"Pilot" Q10945181|Len|"Honor Thy Father" …
Entering curated movie lists
Is there a preferred way to enter lists of movies curated by institutions, e.g. AFI's 100 Years...100 Movies. This particular list has two editions (1998 and 2007). Citizen Kane is the top movie in both of them. Could we add "part of series" in CK's entity item, reference the Q719345 entity with "series ordinal" set to 1? Or would we rather add "part of" to Q719345 with 100 entries, each pointing to one of the list entries? How could the different editions be specified? Is there any type of film list entered into Wikidata that might serve as an example?--126.96.36.199 18:19, 6 August 2019 (UTC)
Columbo (1): seasons
Thanks to No Time to Die (Q66619706), I noticed that we missed out on that series.
It has a few interesting aspects, notably that there seem to be different ways of slicing it into seasons. I solved this by creating two "season 10" items:
- Columbo, season 10(Q66664246): season of television series (all episodes of 1990 to 2003)
- and Columbo, season 10(Q3965275): season of television series (3 episodes from 1990-1991)
and setting one to preferred.
- the 10th season is 3 episodes, the other ones are special episodes. It's the English Wikipedia that should split the page --Lombres (talk) 11:59, 28 August 2019 (UTC)
Columbo (2): character types
On the episode items, I think this could be added with the object has role (P3831)-qualifier to the cast list. This leaves room for the "character role" being used for an item with the actual character. --- Jura 11:14, 22 August 2019 (UTC)
Bechdel test based on unreliable source
A lot of films have the property has quality (P1552) with "fails the bechdel test" (Q45172088) or "passes the Bechdel Test" (Q45171911). For those of you who don't know what the Bechdel test is, see here. However, these statements are often based on only one reference, the website Bechdel Test Movie List (Q45150204, link), which is completely user-generated. It's actually a forum. The source is totally unreliable, so I would send a bot to delete all of the Bechdel test statements that cite that source, if we all agree --Lombres (talk) 12:05, 28 August 2019 (UTC)
- Is there any reliable source for this information? May be to delete the property completely? --Infovarius (talk) 14:46, 29 August 2019 (UTC)
Random film items to improve
Notified participants of WikiProject Movies
As one of the project goals is to improve items, I added a random selection at
It picks from items with the lowest number of statements and should update daily.
- Thank you @Jura1:. I created a version for Arabic Wikipedia since most of the Arabic items need improvement. Could you please explain what "?count" and "?rnd" refer to? --FShbib (talk) 17:30, 26 September 2019 (UTC)
- @FShbib: Cool. Thanks for doing that.
- ?count is the number of items the query selects from. It seems it times-out if there are more too many (>4000 or 5000). If it gets too high, we need to reduce the maximum number of statements.
- ?rnd is a sequential number from these items that the queries selects from.
- The approach for enwiki list follows the sample suggested by Matěj Suchánek at Wikidata:Contact_the_development_team#SPARQL_query_OFFSET_by_?var_(instead_of_number). It picks 20 sequential ones from the items starting at ?rnd (usually created all at the same time).
- WD:WikiProject Movies/reports/random/film works slightly differently: ?rnd is chosen within (?count/20). Then, within each 20 segments, one item is picked. Accordingly, the items should be spread out overtime. I just made Wikidata:WikiProject Movies/reports/random/film/itwiki using the same approach. Here ?offset combing ?rnd and ?count is used.
- BTW, there is no requirement to display the numbers in "columns=". --- Jura 18:16, 26 September 2019 (UTC)
- @FShbib: Cool. Thanks for doing that.
movies parts of a film festival edition (Q27787439)
I wanted to model the movies that are part of this year’s slash Filmfestival (Q27981131) (also have to create items for all these editions ^_^), but I have found neither guidelines nor example items where this is done. Would that be in scope? And how would we do it? Most obvious would be has part (P527) but if I recall correctly it has a mandatory inverse property, and I could not find any good subproperty of it (tracklist (P658) nor list of works (P1455) have different domains).