Wikidata talk:WikiProject Media Representation/Archive 1

This page is an archive. Please do not modify it. Use the current page, even to continue an old discussion.

Riz Test

I am planning to create the following items;

• has characteristic (P1552)=passes the Riz Test

• has characteristic (P1552)=fails the Riz Test

• has characteristic (P1552)=Riz Test not applicable (In the description: the creative work does not meet the initial criteria of having at least one Muslim character)

If the Riz Test is failed, the following can be used as qualifiers;

• criterion used (P1013)=Muslim character/s are talking about, the victim of, or the perpetrator of terrorism

• criterion used (P1013)=Muslim character/s are presented as irrationally angry

• criterion used (P1013)=Muslim character/s are presented as superstitious, culturally backwards or anti-modern

• criterion used (P1013)=Muslim character/s are presented as a threat to a Western way of life

• criterion used (P1013)=Male Muslim character/s are presented as misogynistic

• criterion used (P1013)=Female Muslim character/s are presented as oppressed by her male counterparts

Wallacegromit1, focus on media historiography and works from the Global South Valentina.Anitnelav (talk) 14:12, 16 November 2020 (UTC) Blue Rasberry (talk) 23:59, 21 December 2020 (UTC) Maxime Lijil (talk) 10:03, 4 February 2024 (UTC) PaperHuman (talk) 00:56, 7 March 2024 (UTC)

Notified participants of WikiProject Media Representation

1. Please let me know if the above is suitable or if any changes need to be made? (especially the wording)

2. Would determination method (P459) be better suited than criterion used (P1013), even though determination method is used mostly for technical specifications.

@Nomen_ad_hoc: Welcome to the Project! Wallacegromit1 (talk) 16:32, 21 November 2020 (UTC)

Yes - Please go ahead. I like criterion used (P1013) better than determination method (P459). From my point of view determination method (P459) would rather apply to the method used to determine if the criteria are fulfilled or not (e.g. an algorthm or software used, if this was the case). And thank you for setting up the Project page :) - Valentina.Anitnelav (talk) 16:40, 21 November 2020 (UTC)

@Wallacegromit1: I started to use the test and I changed the description of Q102182612 a bit: I changed it to "mayor muslim character" to reflect the wording at https://www.riztest.com/ ("If the film/show stars at least one character [...]")

I judged Hanna (Q170268) as passing the Riz test due to the Moroccan hotel owner but I'm in doubt: His religion is not thematized at all, but I thought the condition "identifiable by ethnicity" to apply here. I'm also a bit in doubt as this character is of significance to the plot, but it is not really a main character. What do you think? - Valentina.Anitnelav (talk) 19:28, 27 November 2020 (UTC)

@Valentina.Anitnelav: The below article lists Office Space as passing the Riz Test, even though the character is not a major or leading role. I do not believe the religion needs to be thematised. As long as the character is identifiably Muslim and has dialogue or some interaction with others, it should satisfy the initial requirement. I do not believe Q102182612 needs to have major in its description. An example of Not meeting the criteria, would be if Muslim characters were used as background or have non-muslim characters just passing by them.

Riz Test Pass Office Space

You may use nature of statement (P5102) > rarely (Q28962310) with Hanna (Q170268)! Wallacegromit1 (talk) 20:49, 27 November 2020 (UTC)

@Wallacegromit1: Thanks for the article! I deleted the restriction and I agree with "...and has dialogue or some interaction with others". I would additionally exclude very minor characters (e.g. a cashier who has some dialogue but only for the purpose of handling the payment). - Valentina.Anitnelav (talk) 21:13, 27 November 2020 (UTC)

@Valentina.Anitnelav: Agreed! Thanks for the clarifying question.

@Valentina.Anitnelav: Kindly requesting reading the Riz Test section of the project, as I have added many clarifications and some new options. Let me know if they are ok? Wallacegromit1 (talk) 07:17, 30 November 2020 (UTC)

Yes, I like it. - Valentina.Anitnelav (talk) 07:33, 30 November 2020 (UTC)

New Tests Queries

1. TV episodes and Book volumes can be tested individually. However, for TV and Book series, passing any test would require atleast 55% of the individual episodes/volumes to also pass the same test? Extra 5% for any odd numbers or errors.

2. Planning to add atleast two of the following: the Lena Waithe test, Shukla/Apu test, Rachel Feldman test/score, MacGyver Test, and Kimberly Peirce test, over the next month.

a. After the above, maybe a pause for adding new tests till Feb/Mar 2021, to allow for more coverage of the ones that already exist and better data modelling? Or should we not add any for a few weeks?

b. After adding the tests, I would be sectioning the tests in specific categories of Gender, Queer/LGBTQIA+, Ethnic Portrayals, Region Specific, and Format Specific. Let me know if you have any other ideas of sectioning the tests?

c. The Kimberly Peirce test seems too similar to the Mako Mori test. I do not believe a work could pass one, but not the other? [Peirce test]

d. Some of these tests may require different wording, such as meets or fulfills all criteria, rather than pass? Would love to know what others think.

e. The Rachel Feldman test/score could be interesting to model, as it has a pass criteria, as well as a scoring system. Would love your thoughts? [Feldman score]

Wallacegromit1, focus on media historiography and works from the Global South Valentina.Anitnelav (talk) 14:12, 16 November 2020 (UTC) Blue Rasberry (talk) 23:59, 21 December 2020 (UTC) Maxime Lijil (talk) 10:03, 4 February 2024 (UTC) PaperHuman (talk) 00:56, 7 March 2024 (UTC)

Notified participants of WikiProject Media Representation Wallacegromit1 (talk) 14:02, 26 November 2020 (UTC)

@Wallacegromit1:

1. I agree with this.

2. Yes, please.

2a I agree with a pause untill February/March 2021. This gives us some time to reflect the application of existing tests in Wikidata

2b I agree with the sectioning. We could do this in tabs, as many other projects do. When having these sections we could also work with other data available. (E.g. number of protagonistic/antagonist/supporting characters in wikidata of a certain ethnicity/gender for a certain region/genre, etc.) In my opinion we don't have to restrict ourselves to tests. Some time ago a wrote some queries about Disney characters and gender (e.g. Gender ratio by narrative role, Gender ratio by year/film) One could do similar things for ethnicities (e.g. Ethnicities in Disney films - but I guess the data are not complete, here)

2c The Kimberly Peirce test is a bit narrower than the Mako Mori test (it focuses on protagonists/antagonists - the Mako Mori test just has the condition of an own "narrative arc". Supporting characters may have a narrative arc, too). More films would pass the Mako Mori test than the Kimberly Peirce test. But I fear that the conditions "The female lead has dimension and exists authentically with needs and desires that she pursues through dramatic action" and "And the audience can empathize with or understand the female lead’s desires and actions" are actually a bit harder to judge and judgements may be a bit more controversial. I would probably postpone the inclusion of the test unless there already exists a bigger data collection so statements in Wikidata could be sourced with that and judgements/reasons given there could inform judgements here.

2d I guess so. Which test(s) do you have in mind?

2e Yes, this test is interesting. Along with the scoring we should probably include the criteria met. I'm just wondering if a film without any sex scene would meet the last criterium - I guess so, or not? - Valentina.Anitnelav (talk) 08:41, 30 November 2020 (UTC)

I added the tabs (I copied them from Wikidata:WikiProject_Video_games) and moved text from the main page to the sections (Bechdel-Wallace/Mako-Mori --> Gender; Vito-Russo --> Queer/LGBTQA+; Riz-Test --> Ethnicity) - I hope this is okay. If not, feel free to just revert! - Valentina.Anitnelav (talk) 12:44, 6 December 2020 (UTC)

@Valentina.Anitnelav:

1. Perfect!

2a. Agreed!

2b. Agreed! You can put those in the tabs as analysis queries. You may have to do the tab sectioning, as my experience with formatting is limited.

2c. Agree to the pause on the Kimberly Pierce Test.

2d. They may come up in the future.

2e. As the test seems to be sex positive, a film without a sex scene, may just not be counted for a score. Or it could be counted towards not being sexualised.

Qualifiers for nature of statement (P5102)

I introduced a new qualifier to use with media representation tests: dubious (Q104378399). I was always a bit unhappy with the rarely (Q28962310) and disputed (Q18912752) as it did not capture all problems that appeared when expressing the nature of a statement. Now we have:

• rarely (Q28962310): If the statement is accepted to be true but it is admitted that it is hardly the case. (E.g. if a film is said to pass the Bechdel Test but two women only talk for a very short time about something else than a man.)
• disputed (Q18912752): If the statement is disputed. (This may be applied to media critics, but also to the comment section of the Bechdel Test Movie List, for example)
• dubious (Q104378399): If an evaluation is dubious. (I mostly introduced it for the case that one editor has doubts about the statement of another editor but there is no decision, yet. Especially with respect to the application of Feldman criteria I see a lot of potential to apply this qualifier right away (e.g. if an editor decides that there is rather no stereotyping of women but is rather unsure about one character.)). - Valentina.Anitnelav (talk)

One example for the use of dubious (Q104378399) would be Lawrence of Arabia (Q228186): There is one source (Metzler Film Lexikon (Q98542844), a German film lexicon) that describes the film as also showing the homosexuality of the hero. I think it is still quite dubious if the first criterium "at least one character that is identifiably LGBTQIA+" is fulfilled - there are probably a lot of viewers that would not identify the protagonist as a homosexual. -Valentina.Anitnelav (talk) 11:54, 20 December 2020 (UTC)

@Valentina.Anitnelav: Agreed, and Love It! Do we put this in every test, or on top as instructions?

I would put it on top as instructions, or maybe in a new section "qualifiers" on the main site. - Valentina.Anitnelav (talk) 15:03, 20 December 2020 (UTC)

Replies & Discussions

@Valentina.Anitnelav:

1. Created all the tests. Have Feldman Test in Gender and Format Specific, as it would have to be modified significantly for Television.

2. Feldman Test: Co-writers and Co-directors counted as 2 Points? (Sorry, but did not understand your exact meaning above)

3. Fedman Test: Agree, that No sex scenes should be scored. Just wondering between 2 points (Criteria f) or 1 point (Criteria g)?

4. Fedman Test: Stereotyped may require further explanation. I was thinking of putting Manic Pixie Dream Girl (Q958161) in the bracket as an example?

5. Fedman Test: As time goes on new stereotypes might emerge, this criteria may become dynamic, and the result may change over time? Or, should we fix it to stereotypes present 5 years before and after the first release? (I prefer the dynamic result, as the old result will always be preserved in the wiki)

6. Waithe Test: Should it have a similar Not Applicable as the Riz Test. If no, the test could be region specific to the Majority Caucasian/White countries Only?

7. Landau Test: I added the additional criteria, only when passed, as it was not originally in the article. However, globally marriage/wedding is a more common alternative to pregnancy (which would be rare). Just thought it would help with data specificity when a giant dataset of thousands may be created? However, if you do find better ways in your discussions with other Wikiprojects to set tropes, like the TV Tropes ID (P6839), I am on board. Will NOT create Falling Sick or Kidnapped/Needs saving, as it may mostly fall under a "problem for the male protagonist".

8. Landau Test: "Primary" could include any woman, if only one woman exists in the work? In the article, The Jungle Book Passes the test, I do not believe any female character really exists in the film, other than a small scene with a female snake.

9. Landau & Shukla Test: Not Applicable maybe required if no women or ethnic characters are present in the work?

10. Mako Mori & Vito Russo Test: are applicable to all works, however may require specific criteria if they are not passed?

@Wallacegromit1:

2. I see that I probably misunderstood your wording. I thought you were thinking about creating just one of [ criterion used (P1013)= 2 points for a female writer or director] OR [ criterion used (P1013)= No female writer or director]. In the first case you would add the criteria only to passing films, in the latter only to failing films.

But I think that your proposal of adding

• criterion used (P1013)= 2 points for a female writer or director to films passing the first criterium and
• criterion used (P1013)= No female writer or director to films failing the first criterium might work better (there would be less misunderstanding).

3. They should be able to get all three points (2 for the 5th criterium and 1 for the 7th), in my opinon. My reasoning: The first four criteria apply to the diversity of the crew, the three last to the presentation of female characters. If considering films failing the first 4 crew-related criteria, films without a sex scene would have no chance to pass. I would find it a bit odd that a sex scene would be a prerequisite for films by a male dominated crew to pass the Feldman-test.

5. Interesting question. I would tend to use stereotypes present when the film was made. But I have to think about it.

6. I would use this also for films from non-caucasian/white-dominated countries, at least for Black women (I see that it might be a bit difficult for your modification "and people considered Dark in their respective individual culture"). It is always interesting to see how these tests score in different environments (this could be helpful to interpret the score in caucasian/white-dominated countries)

7. I also think this data is interesting. But it would be better to have a more generalized way to do it.

8. I did not think so... I've never seen the "live-action" Jungle Book but with respect to the traditionally animated version I would also be very surprised if it should pass - it is certainly only a minor character.

9. I think "Not applicable" would be required for the Landau test. A film without any ethnic characters would fail the Apu/Shukla test, as I understand it (criterium 1) - similar to the Bechdel test.

10. Yes, I also thought about this...

- Valentina.Anitnelav (talk) 23:36, 12 December 2020 (UTC)

@Valentina.Anitnelav:

Agree with all the above!

New Questions:

1. Should the criteria be labelled as the 2 points for Feldman Criteria a, 1 points for Feldman Criteria g or Landau Criteria 3, etc..., while keeping the full criteria in the alias. This will help with the spoiler issue, and will be cleaner and easier to understand?

2. Can you please create a Query for the Feldman Test, showing the qualifiers please? I have already added them to 5 movies.

@Wallacegromit1:

1. Yes, we could do this.

Related question: I noticed only now that the Feldman test requires three female producers (somehow I did not give any thought to this before). I don't think that this makes sense for films with only 1 producer. Do we need to amend this with a quota (e.g. three female producers or at least 1/2 of the producers female?).

2. I wrote three queries:

• a simple query which has the disadvantage of repeating the films for each criterium
• a grouped query where the criteria are concatenated and
• another query where each criterium gets its own column

For Feldman's 50-%-female-crew criterium I wrote a python script that could be useful for films with a larger crew (It is based on footnote 5, but I also took information in wikidata into account). I plan to share it during the next days here with some explanations so others could use it. Do we need a tools-section or should we put it in the section for the test? When using it, determination method (P459) could be useful (to make explicit that the result was approached algorithmically). - Valentina.Anitnelav (talk) 18:50, 13 December 2020 (UTC)

@Valentina.Anitnelav:

1. It is either 3 female producers or 3 female department heads, I just went to the department heads when only 1 or 2 producers existed. An option could be majority female producers (above 50%), with no more than one male producer? This is tricky, as it is based on Academy Award recognition criteria of max 3 producers. We may be able to ask the writers of the article or Rachel Feldman herself to clarify?

2.a. Love the queries. Thank You! Would decide between them soon.

2.b. Fantastic Idea! I was doing this manually. It could be put in the Tools Section in the Main Page, as well as create a sub-section within the test.

2.c. Agree with the determination method (P459)

Wallacegromit1 (talk) 23:00, 13 December 2020 (UTC)

@Wallacegromit1: 1. Very good idea. Would you do it :)? Then we could also ask about the sex scene issue. I'm thinking about creating an item "no sex scene or sexual advances"/"Feldman criteria g not applicable". But there is still the question if such films will get the point or not. - Valentina.Anitnelav (talk) 09:52, 14 December 2020 (UTC)

@Valentina.Anitnelav:

Can you create a list of questions to ask? I will find and send an email to the article writers and Rachel Feldman. Wallacegromit1 (talk) 11:00, 14 December 2020 (UTC)

@Wallacegromit1: Hm... I could just think about the two already mentioned:

We were thinking about the application of two of the critera of the Feldman test in special cases

1. 1 point for three female producers or three female department heads: There are a lot of films with less than three producers. Does a film with one or two producers, all of which are female, get the point (regardless of the amount of female department heads)?
2. 1 point if a sex scene shows foreplay before consummation, or if the female characters initiate or reciprocate sexual advances: There are films without any sex scene or sexual advances. Do these films fail this last criterium? It could be crucial for films with a male-dominated crew, in which case a sex scene/sexual advances would be a prerequisite for passing the test. - Valentina.Anitnelav (talk) 13:27, 14 December 2020 (UTC)

I created missing prerequisite for Feldman Criterion (h) (Q104176605) (maybe the wording could be improved). For the time being I will not count those films as meeting criterium g) (so they will get 0/1 point here). The wording suggests this rather than the other way round... We can change this later. With this interpretation The Hunger Games: Catching Fire (Q574583) fails the Feldman-Test. - Valentina.Anitnelav (talk) 14:52, 14 December 2020 (UTC)

@Wallacegromit1: When submitting this one question came to my mind: What do we count as "sexual advances"? Does kissing count? Are activities that would be rather framed as "romantic" counted, too (asking out for dinner, gifts, etc.)? I was looking for definitions - unfortunately most explanations I found were written from a sexual-harrassment-perspective, but they included those activities that would be counted as romantic in another setting/relationship. Maybe my perspective was too narrow when thinking of criterium g). - Valentina.Anitnelav (talk) 15:10, 14 December 2020 (UTC)

@Valentina.Anitnelav:

I like missing prerequisite for Feldman Criterion (h) (Q104176605) for the time being. Have emailed, and gotten a reply from one of the writers, liking the idea of "If three producers are not present, then there should be a majority of female producers (50% or above), with no more than one male producer". However, they suggested waiting for a reply from Mz. Feldman herself. Will keep you posted.

Kissing counts as a sexual advance. Asking out for dinner, gifts, etc... does not, unless the person is clearly/transparently doing these for a sexual favour. Rejecting a dinner would not be counted towards the criteria in a positive or negative.

@Wallacegromit1: Thanks a lot, I agree. Now I start to have questions about the Landau-Test :).

1. Do female antagonists count as "primary female character"? I would think so - but in this case no film with a male protagonist and female antagonist would pass.
2. As to "ending up pregnant":
   1. I would take this literally and would not apply this to female characters that are mothers/pregnant at the very beginning. (So I would rather not apply it Never Rarely Sometimes Always (Q63183583)).
   2. Does this include female characters that are shown as mothers in the end, without being shown as pregnant in the film at all (e.g. after a time lapse)?

I think I'm not completely sure about the reasoning behind the pregnancy criterium (as you observed, too, it does not strike me as particularly common to let a primary female character end up pregnant, in difference to the marriage plot.) I'm happy with any decision (I just want to be transparent about how we apply these criteria).

BTW: I found marriage plot (Q6772896). This could be used for cases were finding a partner is a central storyline. I would add it using has characteristic (P1552). - Valentina.Anitnelav (talk) 12:33, 16 December 2020 (UTC)

@Valentina.Anitnelav:

Firstly, the Feldman test might be getting a minor update, as Mz. Feldman mostly created the test for the 2017 article, and is now interested in slightly modifying it for 2021, for it to become a more practical future proofed test. Most of the criteria is similar, so continue adding them to items, as you have done so for the last 2 days. Thank You for that!

Do female antagonists count as "primary female character"? - Yes

no film with a male protagonist and female antagonist would pass - Yes, unless there is another female protagonist. (Example: Snow White and the Huntsman (Q624603)). Even Kill Bill: The Whole Bloody Affair (Q223132) passes, as every character is a villain, and the main character is a Anti-Villain. This could be a unique aspect of the test, nothing wrong with it.

female characters that are mothers/pregnant at the very beginning - Agreed and even have a baby bump, or get the news that they are pregnant at the begining. Unless the premise is a woman getting or ending up pregnant like Knocked Up (Q222800) or The Switch (Q911520). I think its movies like these that the criteria was created.

I wonder if the Landau Test should also be made into a "Film Only Test"?

I now admire the simplicity of the Bechdel, Mako and Russo Tests! Thanks for your Idea for this project, lets us discuss these in greater detail.

marriage plot (Q6772896) - Perfect! I just wonder if has characteristic (P1552) is the best place, maybe another property may need to be created similar to the IMDB Tags. But for now, its good.

I have selected grouped query where the criteria are concatenated. It is clean and easy to understand immediately. It is also the one I sent to the article writers and Mz. Feldman, who loved it. Will add this to the Quesry section in the test soon. I will also be adding the individual column option as well if someone wants to filter bey criteria. We may have to change the way we right the scores, from 8/10 to 8, as 10/10 is read as the lowest by the filter due to the "divide symbol". Let me know if it is a minor issue? Thanks Again!

I saw Newest WikiProjects section in the [Weekly Summaries]. Should I add Media Representation for the next weeks summary?

Finally, after your reply to the above, I will break down and rearrange this entire discussion into their respective topic headers, so it will be easier for people who want to see discussions regarding particular tests/topics, can easily view them and we can have clearer discussions in the future. Wallacegromit1 (talk) 06:31, 17 December 2020 (UTC)

@Wallacegromit1:

• Good News! I agree with all of the above. Please add our project to the Newest WikiProjects section, now that we already have some info. I agree with breaking this discussion down in topics.
• As to marriage plot (Q6772896) and has characteristic (P1552): Yes, after I had a look at the property page: to be honest we are generally close to "abusing" this property already by using it for test results (I started a discussion at Domain and scope of this item). As to the current use I think it is far from perfect but "appropriate" (to put it like this). The (type of) storyline may be constructed as an inherent characteristic of a narrative work (or its perception). I think an own property may be proposed if we find/create more types of storyline or related aspects. Currently our devices to model plots are generally not good.
• Related to this: I'm rather thinking about an own property for the test results as we are collecting more and more media diversity tests (I could not find the discussion but has characteristic (P1552) has rather been a kind of workaround as I remember it). I found those two property proposals related to the Bechdel Test (Wikidata:Property_proposal/Complies_with, Wikidata:Property_proposal/fails_compliance_with). But I actually like the simplicity of using just one property to indicate (non-)compliance. The issue was also discussed here Reasonably quick way to resolve "non-compliance property" issue?. I especially like Jura's idea of using something like assessment (P5021) and test score (P5022). I have not been aware of these discussion when they took place and I don't think we have to decide about it right now (or in the next weeks) but I think we should keep it in mind.
• One more thought that came to my mind when reading Reasonably quick way to resolve "non-compliance property" issue? is about the use of criterion used (P1013) for Waithe and Shukla: As one criterium is a precondition of the other we could just use a score. So 0/3 for the Waithe Test would mean there are no black women, 1/3 that there are black women, but no black women in power, 2/3 that there are black women in power, but no black women in power in a healthy relationship and 3/3 that all criteria are fulfilled (the film passes). We could also use the score for the Vito Russo Test, the Mako Mori Test and the Bechdel test. In this case I would propose to delete the Waithe/Shukla related criteria. What do you think?
• I think I would keep the 10/10 as it makes it clear that 10 is the maximum number of points. You can just sort the results in descending order (DESC(?score), for example). - Valentina.Anitnelav (talk) 13:09, 17 December 2020 (UTC)

@Valentina.Anitnelav:

Like the property proposal, thank you for starting it. Lets see it evolve.

The issue with a score compliance of 1/3 for the Waithe Test, is that we will lose the exact reason for not having compliance. We can do away with them completely?

Agreed, lets keep the 10/10 point writing system.

@Wallacegromit1: The Waithe Test, the Bechdel test, the Vito Russo test, the Mako Mori test and the Shukla test have criteria that build upon each other (unlike the Feldman Test, the Landau Test and the Riz-Test). So the exact reason is actually visible in the score:

That there is a black female character is a precondition of criteria 2 and 3 of the Waithe Test. That there is a black female character in power is a precondition of criterium 3. So 1/3 for the Waithe test would mean that there is a black female character, but she is not in power. 2/3 that there is a black female character in power, but she is not shown in a healthy relationship.

That there are two female character is a precondition of criteria 2 and 3 of the Bechdel Test. That they speak to each other is a precondition of criterium 3. So 1/3 for the Bechdel test would mean that there are two female characters, but they don't speak to each other. 2/3 that there are two female characters who speak to each other but not about something else than a man, etc. - Valentina.Anitnelav (talk) 12:08, 18 December 2020 (UTC)

@Valentina.Anitnelav:

Love the idea, makes sense. However, we must still include "Considering named characters only" as well? We will also have to go through all every other item already done. Big project, but good data.

I have made Waithe & Landau Tests Film Only as well, as the dynamic makes little sense for Television Episodes, but maybe Books?

"Considering named characters only" would be still needed. The reasoning behind this is that data got imported from Bechdel Test Movie List, which only considers named characters. The original Bechdel Test did not have this restriction. One could think about scraping the score from Bechdel Test Movie List, too, but their License is cc by-nc, data in wikidata are licensed as cc0, which is not compatible. Our data won't be complete for a very long time but at least we have a way of providing this information along with new contributions. - Valentina.Anitnelav (talk) 21:14, 19 December 2020 (UTC)

@Wallacegromit1: For the Shukla and the Waithe Test I will start moving the information from criterion used (P1013) to review score (P444), soon. If you are ok with it I will request the deletion of the following items after this, as they are not needed anymore:

• Q104144441,
• Q104144445,
• Q104144452,
• Q104144373,
• Q104144364,
• Q104144367. - Valentina.Anitnelav (talk) 08:58, 20 December 2020 (UTC)

Sorry for ignoring your "I have made Waithe & Landau Tests Film Only as well, as the dynamic makes little sense for Television Episodes, but maybe Books?"... I think it should be applicable to books, too. I kind of agree that the Landau Test does not seem useful for television series/episodes. But I would think the Waithe Test does, somehow? (I'm not that much into television series. This is probably the reason I miss your point, here. Sorry, if you already explained it somewhere else...) - Valentina.Anitnelav (talk) 12:23, 20 December 2020 (UTC)

@Valentina.Anitnelav:

Yes, "Considering named characters only" should continue as well.

Shukla and the Waithe Test I will start moving the information from criterion used (P1013) to review score (P444) - Agree, however, rename the items that already exist, rather than deleting and creating new ones, as we have been the only ones to have added these for now, and we can easily edit every item we have added the test, to be updated. Example: change Shukla Test (Q104144441) to 0/1 Shukla Score or, something like that.

Firstly, I will keep the Waithe test for Television as well, some shows can achieve these. Its tricky per episode, depends on the importance of the characters, but could be interesting, and even if the series may not pass some of the episodes may. Making creative writers to think about these, I now realise, that is the point of these tests.

Landau Test I will make for Film and Books Only. It may also work for Theatre?

I will also add a point in the Tasks section that most of these tests are for fictional works only, and documentary and non-fiction will be getting there own test.

To have an example of how I envision the new Shukla/Waithe Test statements: Q1591769#P1552. So the old items would not be needed, anymore. - Valentina.Anitnelav (talk) 15:31, 20 December 2020 (UTC)

@Valentina.Anitnelav: Perfect! I understand now. Yes, please delete the qualifiers. Wallacegromit1 (talk) 15:17, 21 December 2020 (UTC)

Script to estimate the gender ratio of a given film

I uploaded the script to estimate the gender ratio of the crew of a given film here: https://github.com/Valeninta/wikidata_media_representation. I hope this is helpful (let me know if something is unclear). - Valentina.Anitnelav (talk) 16:31, 16 December 2020 (UTC)

@Valentina.Anitnelav:

Thank You so much! Can I kindly request some screenshots of how to do the process as well, particularly because I am not a coder myself. It's Ok if not, Will give this a try anyway, its well written. We can add this to the Tools and Feldman Test Sections? Wallacegromit1 (talk) 06:30, 17 December 2020 (UTC)

Yes, of course. Just to make sure: Do you already have python/did you succeed to install python? Is it about the command line use? In these cases it would be actually good to know about your operating system (as the procedures vary between operating systems). - Valentina.Anitnelav (talk) 13:15, 17 December 2020 (UTC)

@Valentina.Anitnelav:

I already have python installed with Windows 8.1.

@Wallacegromit1: I updated the instructions supposing that Anaconda is installed/will be installed. I'm not a Windows user myself and this seems to me to be the easiest solution to work with python on Windows. Let me know if it worked for you! - Valentina.Anitnelav (talk) 16:05, 18 December 2020 (UTC)

@Valentina.Anitnelav: I tried all of the above, but it says the error - ModuleNotFound 'genderize'. Any ideas?

@Wallacegromit1: Does it work if you run pip install git+https://github.com/SteelPangolin/genderize#egg=genderize instead of pip install git+https://github.com/SteelPangolin/genderize? Do you get any errors when running this command? - Valentina.Anitnelav (talk) 16:23, 21 December 2020 (UTC)

@Wallacegromit1: I updated the script so that it does not need genderize anymore. Now there are less prerequirements. You would need to download it again. Does this work? - Valentina.Anitnelav (talk) 17:34, 21 December 2020 (UTC)

@Valentina.Anitnelav: Perfect! This is incredible and clean. Thank You! Wallacegromit1 (talk) 18:17, 21 December 2020 (UTC)

@Wallacegromit1: Unfortunately I discovered a small mistake in my script. It only shows if there are no names to look up in genderize.io. This can be the case if the gender of all persons could be determined using other sources. I corrected this. To prevent this mistake bothering you you would need to download the corrected script again... - Valentina.Anitnelav (talk) 19:06, 22 December 2020 (UTC)

Query Structure

@Valentina.Anitnelav:

I have just started to add the tests to Television Series.

Tiny Pretty Things (Q104153320) - The tests have been added to the series as a whole, as well as all 10 episodes. Please see if any improvements are required, including in the references?

Kindly requesting that, All queries need to be updated accordingly to Only count films and/or TV series, so the individual episodes do not crowd the numbers. We can also have a separate query for episodes as well.

Finally, all the summary queries can also be added to the first/home page of the project.

Thank You for these Queries! Wallacegromit1 (talk) 03:10, 29 December 2020 (UTC)

@Wallacegromit1: I like that you added the exact time index to the reference section for the episodes. I would rather use number of episodes (P1113) as a qualifier to indicate the number of episodes the test passes/has that score. We should create a section with all the qualifiers that are in use and how they are used. I also think about some notes about references (something in the line of "Normally the work itself is ok. If a work was rated in other sources these should be considered. They should be preferred unless there are errors that can be easily shown using the original work or a transcript, preferably using specifications like quotation (P1683), page(s) (P304) or time index (P4895) .")

I will update the queries. Good idea to have our summaries at our start page. Maybe as a link to a Listeria table (example: List of characters by Arthur Conan Doyle)? - Valentina.Anitnelav (talk) 11:52, 29 December 2020 (UTC)

@Valentina.Anitnelav:

qualifier to indicate the number of episodes - Should I use 6 or 6/10 or 6 out of 10? I prefer 6 or 6 out of 10, as it would not confuse with the review score. Or we could keep the same format, easier for new people joining in?

some notes about references - Please see point 5 of Task/Instructions in the Home Page.

Listeria Link - Perfect! Thank You for all your work.

section with all the qualifiers - Should this be in the Tasks/Instructions section or a new section for qualifiers only? Maybe the Task/Instructions and this Qualifier sections should be included in every sub-page as well?

@Wallacegromit1:

qualifier to indicate the number of episodes - I would just use "6". The number of episodes can be found using number of episodes (P1113).

some notes about references - I missed this. Thanks!

section with all the qualifiers - I would put this into a new section "Data model". There we could also present our modelling approach in general when we decided about the issue Wikidata_talk:WikiProject_Media_Representation#Fails_compliance_with (moving from the quite general has characteristic (P1552) to a dedicated property for test results). We can link to it from each other section. - Valentina.Anitnelav (talk) 13:25, 30 December 2020 (UTC)

Unfortunately Listeria does not seem to work with the queries needed to produce the summary. I just added the queries to a summary section. - Valentina.Anitnelav (talk) 15:29, 30 December 2020 (UTC)

Items in need of additional opinions or discussion - Gender related tests

In this section you can discuss items where you are unsure of the application of Gender-related tests. If there is already a statement/evaluation that you would like to challenge, please ping the originator of the statement/evaluation. You may add nature of statement (P5102) disputed (Q18912752) or dubious (Q104378399) as a qualifier to the statement in question. - Valentina.Anitnelav (talk) 12:27, 20 December 2020 (UTC)

Carnage (Q169564)

Start: Copied from Talk:Q169564

@Valentina.Anitnelav:

Its been a while since I have seen this film, however, I thought the missing prerequisite for Feldman Criterion (h) (Q104176605) will be more suited for this movie?

Also, as Yasmina Reza is a co-writer, the score should be 1 point for Feldman Criterion (a) (Q104144788)?

Wallacegromit1 (talk) 23:41, 27 December 2020 (UTC)

@Wallacegromit1: I started to use missing prerequisite for Feldman Criterion (h) (Q104176605) only if there is no major romantic/sexual/intimate relationship at all in the movie, without considering the film's subject or character constellation. I don't know in this case. Please just change it if you prefer missing prerequisite for Feldman Criterion (h) (Q104176605).

I thought that "co-writer" refers to the credit received (like "co-producer" or "co-director"). When credited as a screenwriter I just applied the 50%-rule. So every writer that wrote the script with somebody else should be considered a co-writer? Probably excluding women-only co-writing teams? - Valentina.Anitnelav (talk) 11:17, 28 December 2020 (UTC)

@Valentina.Anitnelav:

missing prerequisite for Feldman Criterion (h) (Q104176605) - Its fine!

So every writer that wrote the script with somebody else should be considered a co-writer? - I thought so, as it is still reducing the number of women and their impact behind the scenes. You can still keep the 2 points for this one, as the female writer also wrote the play this is based on, maybe this can be an exception.

Probably excluding women-only co-writing teams? - Ofcourse!  – The preceding unsigned comment was added by Wallacegromit1 (talk • contribs) at 12:53, 28. Dez. 2020 (UTC).

End: Copied from Talk:Q169564

@Wallacegromit1: While applying the Feldman test to more items I grew more and more convinced that Feldman is refering to the title "co-writer" and not to the fact that more than one person wrote the script when she is writing 2 points for a female writer or director (if co-writer reduced to 1 point) - No Co-director, as the are extremely rare and is not a professional title.

To have two examples of women explicitly credited as co-writers (not as "full" writers) that should only get 1 point:

• Inés Bortagaray (Q18275942) is credited as co-screenplay for The Invisible Life (Q64166529), not as screenplay imdb).
• Marina de Van (Q1443641) is credited as collaboration on screenplay for 8 Women (Q274895) imdb

When referring to co-director she was also not referring to any director working in a team but to the "credit" co-director (Nora Twomey (Q15438228), for example, is explicitly credited as a co-director of The Secret of Kells (Q908816), not as a director imdb)

Screenwriters and directors are given equal weight in the original clause (both 2 points). For this reason I think that we should treat them also equally in the case of mixed teams: If a work with 50% or more female directors should get 2 points, a work with 50% or more female screenwriters should get 2 points, too. Meaning

• Shadow of a Doubt (Q465979) should get 2 points (2/3 female screenwriters)
• The Hunger Games (Q212965) should get 1 point (1/3 female screenwriters)
• 8 Women (Q274895) should get 1 point (female co-writer)

What do you think about equaling the rules for director teams and screenwriter teams?- Valentina.Anitnelav (talk) 23:02, 3 January 2021 (UTC)

@Valentina.Anitnelav:

I think you are right! The above point system is perfect for writing and directing. I agree that the issue is the "co-" in the credit. Thanks for the clarification.

Promising Young Woman (Q62595383)

@Wallacegromit1: Hi Wallacegromit, I just had a look at Promising Young Woman (Q62595383) and I would let it pass both the Russ Test (due to the relationship between Cassie and Gail) and the Tauriel Test (under the "hobby"/"non-professional activity" clause: She does accomplish the goal set before her (make Al accountable). Maybe we could make it a dubious pass instead of a dubious fail?

I would actually even have let it pass the Waithe test - but this would have been a misunderstanding on my part: I always thought that "healthy relationship" would include all relationships - including friendship - but this does not seem to be the case. Now there are probably some false Waithe Test positives that I have to recheck...

BTW: I changed my opinion as to Carnage (Q169564): I agree with you that missing prerequisite for Feldman Criterion (h) (Q104176605) is more appropriate. I just started to think we should not see all depictions of married couples as depictions of a romantic/sexual relationship. - Valentina.Anitnelav (talk) 10:26, 11 August 2021 (UTC)

@Valentina.Anitnelav:

For Promising Young Woman (Q62595383) - Agreed on Russ Test Pass, and Tauriel Test as a dubious pass, as she is high functioning on the surface, and is able to accomplish her final goal in some capacity.

Waithe test - Close friendships do count for the test. It could be a dubious pass, but feels more like a dubious fail or just fail, as we do not see her socialising beyond being colleagues.

Agree with Carnage (Q169564) decision! Wallacegromit1 (talk) 05:50, 28 August 2021 (UTC)

Licorice Pizza (Q102036245)

@Koavf: Hello, Koavf! I see that you evaluated Licorice Pizza as both failing the Mako Mori Test and the reverse Mako Mori Test. I tend to disagree here. Both Alana and Gary have their own storylines that are not about supporting another person's storyline (they are both protagonists). Their storylines are strongly intertwined so that they are supporting each other's storylines at some points, but they still have their own storylines. As there are no clearly identified gender-neutral characters, this would be also a bit weird. Both have storylines that focus on self discovery: both are about finding or choosing the (right) partner and finding their place in life; Alana sets out or at least tries to become an actress, Gary a business man. I think the film should both pass the Mako Mori Test and the reverse Mako Mori test. What is your reasoning as to this?

Related to that: I set the result of the Riz-Test to not applicable (Q105773175) as there are no clearly identified Muslim characters. I also think that the film should pass the Shukla Test (Q104144383) due to it's Jewish characters. The whole Kane (Haim) family is Jewish, as is Joel Wachs (Benny Safdie) and they speak to each other (I did not count the exact time but on multiple occasion and for an extended amount of time). Even though it may be generally controversial if the Jewish population are one ethnic group (separate from a religious group) I think it should still qualify for this test (it was and is often enough treated and regarded as such). Kind regards, - Valentina.Anitnelav (talk) 11:47, 27 January 2022 (UTC)

@Valentina.Anitnelav: I think your suggestions are all perfectly valid. I had a hard time with the Mako Mori and reverse Mako Mori tests, as you could definitely argue that Alana exists as a fleshed out character and the star of the film, but everything she does ends up being with or in reference to Gary. I think ultimately, you are correct and I was wrong. —Justin (koavf)❤T☮C☺M☯ 16:17, 27 January 2022 (UTC)

@Valentina.Anitnelav: Sorry, look at me being so sloppy: as I understand the Shukla Test, this is for "persons of color" and while the Jewish people are certainly minorities in 1973 Southern California, I think that they would broadly be considered white or white presenting or having the kinds of social advantageous of whiteness, so I don't think it passes. —Justin (koavf)❤T☮C☺M☯ 16:20, 27 January 2022 (UTC)

@Koavf, Wallacegromit1: Oh, I had a look at The Shukla Test and you are right ("The Shukla Test, for books, films and television where a) two main characters who are people who of colour b) talk to each other without c) mentioning their race."). We seem to diverge in our definition quite a lot (see Wikidata:WikiProject_Media_Representation/Ethnic_Portrayals#Shukla_Test). How should we deal with that? Should we relabel the current Shukla Test item in "Shukla Test for ethnic minorities" and create a new item for the Shukla Test in its original form? The current data only reflects the ethnic minority focus, not the people of colour focus (so films may pass it due to ethnic minorities that are considered "white" and other may fail even thought there are people considered "of colour" talking to each other (if they are not an ethnic minority in the country of origin of the film)). - Valentina.Anitnelav (talk) 18:49, 27 January 2022 (UTC)

I don't think we should create new media tests that haven't already been established somewhere else "in the wild" so to speak. We could create an infinite number of tests of our own accord on Wikidata. —Justin (koavf)❤T☮C☺M☯ 18:52, 27 January 2022 (UTC)

Well, it already exists for some time with exactly this description ("test of the portrayal of ethnic minorities in creative works"). The only other alternative would be to just delete all data. I think the idea behind it was to make it less centered on the Anglo-American film industry and make it more applicable to films from other regions, too. What is an disadvantaged group and what not differs from region to region. Valentina.Anitnelav (talk) 20:36, 27 January 2022 (UTC)

I found the version of the test we are referring to: It is described here, by Nikesh Shukla: "I want to see a film where two ethnic minorities talk to each other for more than five minutes about something other than race". - Valentina.Anitnelav (talk) 20:51, 27 January 2022 (UTC)

@Koavf, Valentina.Anitnelav: The original intention was to broaden the scope of the test to include films outside the Anglo-American film industry. As an example - There will always be ethnic Indians in a Indian/Bollywood film, which is not really breaking any barriers for the industry it was produced in.

However, I am open to having the Shukla test be strictly specific to the Anglo-American media landscape only, this would help it be interpreted in a easier manner. But, the spirit of the test, seems to be global, and the article almost discusses the idea of watching things rarely seen in a global context. I will leave the decision up to you both, but we may be able to split the test - one could be the exact wording, the other could be for a global context. Wallacegromit1 (talk) 06:34, 1 February 2022 (UTC)

It seems like the Hollywood film industry (i.e. not the Nigerian, Bollywood, etc.) is what the test was made to measure, so I think we should stick to the wording and framing of the test as devised. —Justin (koavf)❤T☮C☺M☯ 12:50, 1 February 2022 (UTC)

The creator of the test is British and does not restrict this test to Hollywood films. So why should we? The sentence "I want to see a film where two ethnic minorities talk to each other for more than five minutes about something other than race" is universally applicable, not just to Hollywood film. And the representation of ethnic minorities is an issue everywhere, not just in the US. Yes, Hollywood films have the biggest audience worldwide so they are kind of the most prominent (and most often used as examples), but why should it not be useful to evaluate films from other regions?. What problem do you see with applying this to non-Hollywood films, exactly? - Valentina.Anitnelav (talk) 15:25, 1 February 2022 (UTC)

I almost forgot: If you only want to get the subset of Hollywood or US films, you can get them using SPARQL: films from the USA with a Shukla test result. Of course there may be disagreement if a certain group should be considered an ethnic minority in the spirit of the test (as a disadvantaged group). So if you think that Jews should not be considered an ethnic minority in the USA (in the spirit of this test) it is okay with me. But thinking of German films, for example, Jewish characters should be considered as ethnic minority characters (in the spirit of this test). -Valentina.Anitnelav (talk) 15:53, 1 February 2022 (UTC)

@Koavf, Wallacegromit1: I'm not sure if this is the issue at hand for Kovf, but I can imagine that the lack of specificity may be an issue (if any ethnic minority of any region may be considered, it makes it difficult to find all films that pass the Shukla test due to having people "of colour" talking to each other.)

One solution would be to have two items for each version of the test (both introduced by Niklesh Shukla):

• The current item would represent the wording used in his article After the Bechdel Test, I propose the Shukla Test for race in film (from October 2013) "I want to see a film where two ethnic minorities talk to each other for more than five minutes about something other than race"
• The (new) second item would represent the wording used in The Shukla Test (from August 2013) "The Shukla Test, for books, films and television where a) two main characters who are people who of colour b) talk to each other without c) mentioning their race". We could think about making the second item only applicable to films from producing countries with a "white" majority group. Maybe we could provide a list with ethnic groups that are considered "of colour" in the US (to me it is not always that clear). If we take the US as a reference point we could probably just take the groups mentioned here: en:Person_of_color (and in the equivalent articles, e.g. here: en:Asian_Americans - but then: are en:Middle_Eastern_Americans considered "of colour" in the US or not (as they are excluded from the group of Asian Americans)? en:Central Asians in the United States seem to be considered "white").

As another solution (or in addition to the first option) we could be even more specific and note the groups that were considered for this test using the qualifier including (P1012). I created one example at Waves (Q56205334) (Q56205334#P5021). - Valentina.Anitnelav (talk) 19:09, 1 February 2022 (UTC)

@Valentina.Anitnelav: Sorry if I made things unclear: I think we should use whatever standard the original person set. If it's "person of color" or "ethnic minority" or whatever else, then that should be exactly what our standard is locally. Now, there may be problems with the wording as such (e.g. an "ethnic minority" in one context is different than what constitutes a minority in a different context). Thanks for putting so much thought into this. —Justin (koavf)❤T☮C☺M☯ 19:49, 1 February 2022 (UTC)

Ok, then I would keep the item Shukla Test (Q104144383) for the "ethnic minority" wording. I added the source via described at URL (P973). If anybody wants to create an item for the "people of colour"-wording I'm okay with it (but I won't do it myself). - Valentina.Anitnelav (talk) 07:56, 2 February 2022 (UTC)

Mini Project or Mini Film Festival

@Valentina.Anitnelav:

I was wondering if we can do a "Mini Project or a Mini Film Festival", to increase the quality of Metadata related to all the movies curated in Women Make Film (Q77855044), BBC's The 100 greatest films directed by women (Q105406968) and on the AWFJ’s Top 100 Films List (Q95415779). This can help generate more data visibility of movies made by or about women.

We can include/ping WikiProject Movies and WikiProject Women to increase the number of participants, and use the Representation Tests, including the ones in the Suggest a Test section, and/or any other ideas of quantitative or qualitative analysis of the films.

We can eventually do similar Mini Projects to help increase metadata quality for African Diaspora Movies, Queer Cinema, Grey Cinema, Indigenous Cinema, Southeast Asian Cinema, or others. This will help different WikiProjects to also work together for common or overlapping goals, and have more eyes on these items. It can also be a fun exercise to watch new films. Let me know what you think? You can also disagree, and will only do it if enough (atleast 5) people agree. Wallacegromit1 (talk) 12:51, 9 February 2021 (UTC)

@Wallacegromit1:I like the idea very much and would be on board. I hope that it won't take too long until the assessment property is created as it would be nice to use the final data model for that. - 13:15, 9 February 2021 (UTC)

@Valentina.Anitnelav: Would you like to wait for the property to be created? or start and complete this mini project by March or, End of March? We could split the number of movies we need to watch between the number of people participating. Some movies on these lists, already have well modeled items on Wikidata with the representation tests as well, so we can wait. Wallacegromit1 (talk) 17:26, 9 February 2021 (UTC)

@Wallacegromit1: I would rather wait for the property to be created. I expect that it won't take long, now (maybe one week or so). I think we can start with preparations like pinging projects, making a list of the films, having a look at which are available online, making a page for this mini project etc. It could be an issue that libraries are currently closed (at least this is an issue where I live - I'm not sure about the situation elsewhere). This would make it harder to access films that are not available online.

If the property is not created in, say, two weeks or so we could start anyway :). - Valentina.Anitnelav (talk) 16:04, 10 February 2021 (UTC)

@Valentina.Anitnelav:

1. Agreed! We can start later, and I am already preparing a combined list of the films, and their availability. (Some of the films should be available for free on Kanopy (Q22058467) and the Internet Archive (Q461), especially the earlier ones)

2. I would also like to include most of the tests in the suggest a test, as soon as the property is ready. Some movies are documentaries, which can benefit from the Non-Fiction tests, and also having the new tests with more eyes can help find quirks faster.

3. We can run this from 1st March to 30th May 2021 (which would include International Women's Day (Q38964) and Women's Health Month), and have atleast 8 to 10 movies each for every participant over the 3 months (Depending on the number of participants, and what they are able to access). And every week could be based chronologically through the decades (Example: Week 1 - 1910 to 1930). These are just ideas, nothing is set yet.

4. The list may not be mandatory (due to access issues), as long as the participants can watch movies with similar themes and add metadata for them, or may watch movies on other similar or local lists. This can help movies that were not part of any of the three core lists, and can serve as a Parallel Section to the project/festival.

5. We may require a few "Model Items", as reference for everyone. Any suggestions?

6. Should the page for this project be in a "Tab" on WikiProject Media Representation? Should we Ping the other WikiProjects now to gauge interest, ask for guidance and help set up?, Or ask them later?

7. As an FYI, The Idea for this came from the Wikiquote #SheSaid drive, and interviews of female and male directors facing data visibility issues of their films on search engines and streaming. The mini project does not need to be as big or successful as the #SheSaid drive, but more a fun way to enrich Wikidata. Wallacegromit1 (talk) 14:26, 11 February 2021 (UTC)

@Valentina.Anitnelav:

a) There are 440 films and short films, 2 more are TV shows (which we can remove or ask to add info on the episodes the women have directed), with 286 unique directors, majority female.

b) Is there a way to share the list in an excel file? Have already had Wikidata Identifiers linked to the films. Wallacegromit1 (talk) 05:28, 14 February 2021 (UTC) @Wallacegromit1:

Thank you!

2. I agree.

3.-4. Nice ideas. Yes, people should definitely be allowed to include films not on the list and make their own suggestions.

5. I have no suggestions, yet, but I will look into it during the next days.

6. Yes, I would create an own tab, maybe "Task Forces" or something like this? For the project/festival we can make an own sub section or sub page with e.g. a listeria list of the films.

Not here. You could set up a google spreadsheet. With the help of the wikidata ids (thank you so much for your work!) we could create a listeria list (this would also be helpful to monitor completeness). - Valentina.Anitnelav (talk) 10:40, 14 February 2021 (UTC)

@Valentina.Anitnelav:

Please see the list in the below link. You may transfer it on to a google spreadsheet. My experience with listeria is limited, so you will have to create one, sorry. Wallacegromit1 (talk) 14:23, 14 February 2021 (UTC)

Link - https://ethercalc.net/pnrypi5wb1

@Wallacegromit1:

I created Wikidata:WikiProject_Media_Representation/Task_Forces and included there four listeria lists (one for the BBC list, one for the AWFJ list and two for Women Make Film; I had to split up the latter as wikidata's sparql endpoint does only accept a list of values up to 200). This is just to have a start - we can still change the pagename or we can change the sectioning. If we add the lists to the films (e.g. via described by source (P1343)) we could also make one big listeria list as we could query the films by that (e.g. described by source (P1343) BBC's The 100 greatest films directed by women (Q105406968)) and not by a list of qids (which has the limit of 200). Just let me know what you prefer. - Valentina.Anitnelav (talk) 21:49, 14 February 2021 (UTC)

Next Steps

@Valentina.Anitnelav:

1. Thank You for the lists and section, as well as filing in some more ids. "Task Forces" is ok.

2. Some movies are on multiple lists, is that ok?

3. For movies in the WMF list, we can use present in work (P1441)=Women Make Film (Q77855044). For the other two lists, is there a property for "On list"?, otherwise we can use part of (P361), published in (P1433) or described by source (P1343), with series ordinal (P1545). The BBC list has no description as it is just a list, so maybe not described by source (P1343)?

4. All the directors can also have the movies as being there notable work (P800) with the lists as the references, and this project can help improve the directors items as well.

5. Are there any other ways to measure progress? We can take a snapshot of the quality now and after to see how much has improved?

6. Should we Ping the other projects and let them know what we are planning? Or will there be too many questions (We could have more of an explanation above the lists in Task Forces Page)? I was hoping they can also guide us in setting this up, and any other properties or criteria we should be adding to the items, that we have missed.

7. Should we start creating items for the other tests? Atleast for the tests themselves, and the qualifiers can wait for the property to be created? Wallacegromit1 (talk) 06:57, 15 February 2021 (UTC)

@Wallacegromit1:

2. I think this is ok.

3. We should not use present in work (P1441)=Women Make Film (Q77855044) as present in work (P1441) is meant to be used for fictional or fictionalized entities (e.g. characters) that appear in a work. We can use described by source (P1343) for all cases.

5. Interesting idea. We should think about that.

6. Maybe we should formulate one small paragraph explaining this task force and then we could ping other projects.

7. I will ping a property creator (maybe tomorrow?) asking if they could create the property as we want to use it soon. - Valentina.Anitnelav (talk) 07:44, 15 February 2021 (UTC)

@Valentina.Anitnelav:

3. Thank You for the clarification! Yes, lets use described by source (P1343) in all the cases.

5. Maybe the basics of Director, Writer, Atleast one actor, Producer and Studio, should be added to all the items.

6. I will add a description on top of the lists in a few hours. Please feel free to edit it as you see fit.

7. Thanks! Wallacegromit1 (talk) 11:54, 16 February 2021 (UTC)

@Valentina.Anitnelav: What about the movies that did not have Wikidata Ids, we should let them be in another space, so people can create them? Wallacegromit1 (talk) 14:17, 16 February 2021 (UTC)

@Wallacegromit1: I already created the missing ones or I found them under another title. So the lists should be complete, now (I added the ids in the etherpad spreadsheet you shared). - Valentina.Anitnelav (talk) 15:01, 16 February 2021 (UTC)

@Valentina.Anitnelav: I have added a description of the Festival on the List Page, let me know if I should Ping the other WikiProjects or wait till the New Property is created (As it would avoid confusion)? Sorry to repeat myself. Wallacegromit1 (talk) 08:17, 17 February 2021 (UTC)

@Wallacegromit1: I made a ping at the proposal discussion page. Maybe we could wait until Saturday/Sunday? Anyway we should ping other projects on Sunday, I think. - Valentina.Anitnelav (talk) 13:40, 18 February 2021 (UTC)

@Wallacegromit1: I made some changes to the Task Force page and added it as a tab. It does not seem that the property is going to be created, soon. So I think we should just go ahead and ping projects. Would you like to inform the Women and Movie Projects? I would take care of improving the lists. I created a "interested?" section where people can add themselves (if they want). Should we add a passage that people may join for only part of the task force? - Valentina.Anitnelav (talk) 11:23, 23 February 2021 (UTC)

@Valentina.Anitnelav:

1. Thanks for the changes.
2. Yes we should add a passage for people who may join only part of the task force.
3. Should we create some of the tests from the Suggest section? I think the top three Reverse tests are ready, as well as the Tyrion Test, Russ Test, Tauriel Test, Raleigh Becket Test, AfroBubbleGum Test, DuVernay Test, Topside Test, May Test, Coates Test, Zimmerman Test & the Finkbeiner Test.
4. MacGyver test, Villarreal Test & Ko Test can be added later after more people see them. Villarreal Test - I never got a reply back from the writer, will try again.
5. Celia Imrie Test - Kindly requesting some explanations/clarity to be added as notes below some of the criteria please. And also, can you please answer the last question I asked in the discussion on that test. We can then add this test as well.
6. After adding these tests and maybe updating the Summary Query in each section/tab, I will Ping the other projects on the same page as the Task. Wallacegromit1 (talk) 12:47, 23 February 2021 (UTC)

@Wallacegromit1: I agree to create all the tests you mentioned in 3. I had to split up the Women Make Film list post 1990 in two lists as there were seemingly too many items on that list (there were problems when updating). I'm not sure about the limit, but it seems to be somewhere around 150 for our number of columns. By this occasion I also changed the default sorting order to sort by year for all lists. I tried to create a big list with all films (you find it here: Wikidata:WikiProject_Media_Representation/Task_Forces/First_Metadata_Film_Festival_Listeria), but unfortunately I could not add more than 5 columns before having problems with updating that list. - Valentina.Anitnelav (talk) 15:28, 23 February 2021 (UTC)

@Valentina.Anitnelav:

1. Thank You as always! The list seems fine, lets have others tell us what other information they would like to see.
2. I have already created the Topside Test (Q105640076) and the Coates Test (Q105641628), working on the Zimmerman Test and will continue to create more tomorrow. Can you please create a New Tab for Non-Fiction/Documentary please? (For some reason, I am unable to edit them)
3. I have already Pinged the other WikiProjects as well, so they can start familiarising themselves with the tests and the festival lists. I have also added the Listeria list at the bottom of the Festival Notes. Wallacegromit1 (talk) 15:41, 23 February 2021 (UTC)

@Wallacegromit1:

I added them (you would need to edit this page: Wikidata:WikiProject_Media_Representation/Tabs as this is the one included as a header). I added a "Non-Fiction" tab and a "Other" tab - here we could add the Tyrion test and the Celia Imrie test. I started to arrange the tabs in two rows. Does it look okay for you? - Valentina.Anitnelav (talk)

@Valentina.Anitnelav: Perfect, Thank You! Did you receive the Task Forces Ping? Wallacegromit1 (talk) 16:22, 23 February 2021 (UTC)

@Wallacegromit1: No, I did not receive that ping. Did you get any message (e.g. that there were too many members)? - Valentina.Anitnelav (talk) 16:25, 23 February 2021 (UTC)

@Valentina.Anitnelav: Could be because I did not sign it. Have signed, did you receive it now? Wallacegromit1 (talk) 16:47, 23 February 2021 (UTC)

@Wallacegromit1: No. It does not work if you just put your signature there. You have to add your signature and the pings in the same edit. Maybe this is the reason that it still did not work. - Valentina.Anitnelav (talk) 16:49, 23 February 2021 (UTC)

@Valentina.Anitnelav: How about now? Wallacegromit1 (talk) 16:56, 23 February 2021 (UTC)

@Wallacegromit1: No, unfortunately not :( - Valentina.Anitnelav (talk) 17:00, 23 February 2021 (UTC)

@Valentina.Anitnelav: One last time, has it been Pinged? If not, can you please do it? Wallacegromit1 (talk) 17:24, 23 February 2021 (UTC)

@Wallacegromit1: Yes, it worked - Valentina.Anitnelav (talk) 17:26, 23 February 2021 (UTC)

@Valentina.Anitnelav: I re-edited and Pinged everyone. Did you receive another Ping? There seems to have not been any change or communication from anyone. Wallacegromit1 (talk) 15:08, 24 February 2021 (UTC)

@Wallacegromit1: No, I did not receive a second ping. But the other ping worked. I would not worry too much if people don't respond the first day. - Valentina.Anitnelav (talk) 16:47, 24 February 2021 (UTC)

Further Queries

@Valentina.Anitnelav:

1. Has there been any further contacts from other WikiProjects?

2. I was thinking that the model we have of the tests with has characteristic (P1552) seems to be more appropriate than the new model we were thinking. I am fine either way, but just wanted to share the fact that we may not have to change it. What do you think? Wallacegromit1 (talk) 18:05, 27 February 2021 (UTC)

@Wallacegromit1:

1. No, unfortunately not.

2. The property has characteristic (P1552) is a bit too general, an own test related property is better, I think. I think that the new model is as expressive as the has characteristic (P1552) approach (and it will be actually a bit easier to write queries with the new property). Do you see any issues (maybe I missed a point)?- Valentina.Anitnelav (talk) 18:13, 27 February 2021 (UTC)

@Valentina.Anitnelav:

1. I will be watching some movies next week from 1900 to 1930 and filling out some of the tests. You may select movies on your own criteria and access if you like. We can re-ping the project a week or two later. Is there any other way to increase awareness, maybe Project Chat?

1.5. In any case this is meant to be fun, if more people join later its great, if not its fine too.

2. Makes Sense. I was just having minor doubts, all good! Wallacegromit1 (talk) 18:22, 27 February 2021 (UTC)

1. I'm thinking about a selection criterion for the next week(s) (maybe I will add a "watchlist" to my interested?/participant entry). Project chat may be a good place (I'm not that good with raising awareness/publicity, either).

1.5. Totally agree with this point :) - Valentina.Anitnelav (talk) 18:30, 27 February 2021 (UTC)

Increase diversity in our Wikidata integrations research

Hello WikiProject Media Representation: Gender editors!

Wikidata for Wikimedia Projects at Wikimedia Deutschland is focused on enhancing data reuse and functionality between Wikidata and its sibling projects. We are sorely lacking diversity in our research and are seeking fresh perspectives.

We are conducting 1-hour interview calls to understand editors' experiences (positive & negative) when integrating Wikidata with other Wikimedia projects (Wikipedia, Commons, Wikisource etc.). Your insights will help ensure the products we build are suitable for all editors, regardless of gender, sexual-orientation, language or experience level. We also invite you to write to us if you prefer to contribute in written form.

Interview Information:

• Interviews will be conducted in English.
• Compensation is available.
• Sessions will be scheduled soon and on a rolling basis.
• Sign up here if you are interested.
• Diversity: We are keen to hear from women editors; editors with a diverse gender or sexuality; beginner or casual editing experience and/or those who edit in non-latin or non-western languages.

Thank you for your interest in contributing to our research, Danny Benjafield (WMDE) (talk) 07:33, 19 February 2024 (UTC)

Items in need of additional opinions or discussion - LGBTQA+ related tests

In this section you can discuss items where you are unsure of the application of Queer+LGBTQA+-related tests. If there is already a statement/evaluation that you would like to challenge, please ping the originator of the statement/evaluation. You may add nature of statement (P5102) disputed (Q18912752) or dubious (Q104378399) as a qualifier to the statement in question. - Valentina.Anitnelav (talk) 10:40, 20 December 2020 (UTC)

AfroBubbleGum Test vs. BubbleGum Test

@Wallacegromit1: I'm thinking how to convert this into the new model. The easiest possibility seems to be to have two test items: AfroBubbleGum Test (Q105726358) and one (still to be created) <BubbleGum Test> (or something similar). Then we could have the following applications:

• passes the AfroBubbleGum Test (Q105726358) if the work is AfroBubbleGumist (corresponds to has quality: Q105726911)
• passes the <BubbleGum Test> if the work is BubbleGumist but not AfroBubbleGumist (corresponds to has quality: Q105727442)
• fails the <BubbleGum Test> if the work is neither AfroBubbleGumist nor Bubble Gumist (corresponds to has quality: Q105727461

What do you think? - Valentina.Anitnelav (talk) 12:55, 6 March 2021 (UTC)

@Valentina.Anitnelav:

1. Is there another terminology we can use? Maybe fullfils? AfroBubbleGum and DuVernay tests, purposefully do not want to include the terms "Pass or Fail". The DuVernay test should purely be a score. If no other terminology is available then lets use it.

2. In terms of creating a separate BubbleGum Test makes sense, as we can specify it a little better.

3. Thank You for all the modelling, will change the instruction in the first page to reflect these. Wallacegromit1 (talk) 16:15, 6 March 2021 (UTC)

@Wallacegromit1:

1. I'm not sure if the difference between "pass" and "fulfill" is striking enough to warrant an own item. To me to fulfill the AfroBubbleGum test would be the same as to pass the AfroBubbleGum test. As to the DuVernay test I would just omit the "assessment outcome" qualifier and only add the criteria/score qualifiers.

2.-3. Thanks! - Valentina.Anitnelav (talk)

@Valentina.Anitnelav: Agreed! Wallacegromit1 (talk) 16:42, 6 March 2021 (UTC)

Increase diversity in our Wikidata integrations research

Hello WikiProject Queer+LGBTQA+ editors!

Wikidata for Wikimedia Projects at Wikimedia Deutschland is focused on enhancing data reuse and functionality between Wikidata and its sibling projects. We are sorely lacking diversity in our research and are seeking fresh perspectives.

We are conducting 1-hour interview calls to understand editors' experiences (positive & negative) when integrating Wikidata with other Wikimedia projects (Wikipedia, Commons, Wikisource etc.). Your insights will help ensure the products we build are suitable for all editors, regardless of gender, sexual-orientation, language or experience level.

Interview Information:

• Interviews will be conducted in English.
• Compensation is available.
• Sessions will be scheduled soon and on a rolling basis.
• Sign up here if you are interested.
• Diversity: We are keen to hear from women editors; editors with a diverse gender or sexuality; beginner or casual editing experience and/or those who edit in non-latin or non-western languages.

Thank you for your interest in contributing to our research, Danny Benjafield (WMDE) (talk) 16:28, 13 February 2024 (UTC)

P.s. We also invite you to write to us if you prefer to contribute in written form. Danny Benjafield (WMDE) (talk) 16:28, 13 February 2024 (UTC)

Items in need of additional discussion or opinions - Ethnicity related tests

In this section you can discuss items where you are unsure of the application of Ethnic Portrayal-related tests. If there is already a statement/evaluation that you would like to challenge, please ping the originator of the statement/evaluation. You may add nature of statement (P5102) disputed (Q18912752) or dubious (Q104378399) as a qualifier to the statement in question. - Valentina.Anitnelav (talk) 12:13, 20 December 2020 (UTC)

Aladdin (Q215518)

@Wallacegromit1: First: Thank you for edit concerning the Riz-Test: I was also wondering about how to evaluate Jasmine, but I've not been sure (and I tend to judge in favour of the film in dubious cases). With the source you provided I think we can just delete my initial evaluation.

I'm thinking about the application of Shukla Test (Q104144383) to films that are set in "a faraway place": What should be an ethnic minority in these cases:

• The ethnic minority in the producing country (so if set in Arabia Arab characters would still be counted as ethnic minority characters if the film is produced in the US (for example))
• The ethnic minority of the place where the film is set (so if set in Arabia Arab characters would not be counted as ethnic minority characters)
• The ethnic minority in the film (so if a film is set in Arabia and the majority of characters is white, Arab characters would be counted as ethnic minority characters)

To be frank: my favourite is the first one. This seems to best capture the initial idea of the test (representation of ethnic minorities in (Hollywood) films). I don't think that the second is a good idea, as this would allow films with white main characters in Arab countries (for example) to pass without any of the Arab characters speaking to each other. (Maybe one could emphasize that they should not be from colonizing countries, but I don't know...). I'm a bit more sympathetic with the third one.

For now I let Aladdin pass: Q215518#P1552. Characters talk about love, dreams, their relationship and trust, power, wealth and poverty, making a living, administration, Aladdin and Jasmine, etc for at least 5 Minutes. What do you think? - Valentina.Anitnelav (talk) 10:31, 21 December 2020 (UTC)

@Valentina.Anitnelav:

Riz Test - Aladdin is a perfect case of nuance and subjective interpretation. Jasmine could be considered to be oppressed more by the patriarchy, rather than Islamic values alone. And many might consider her to be strong willed and full of agency regardless. This is one of those cases where the use of disputed (Q18912752) or dubious (Q104378399) is perfect, but we can put the passing as Deprecated Rank, as the passing is the interpretation. And if the film is taken at pure technicality alone, it does not pass.

Shukla Test - You are right! The first one is in keeping with the initial idea of the test. The ethnic minorities in the country/region of production will be counted towards the test. A film like The Farewell (Q60611804) passes the test, as it is produced in USA, even though it has a Chinese cast and language.

The second point will be incredible, as it would almost be a Double Pass Shukla Test. Let's take the example of The Last King of Scotland (Q176826), a British-German production taking place in Uganda, with a mostly Black cast, and some characters and actors that are also ethnic minorities in Uganda itself. However, the film passes the Shukla test regardless of the ethnic minorities in Uganda. (FYI: I do not remember if the movie actually passes the Shukla test, as most of the conversations happened with the White characters, may need to look it up.)

For the third point - A movie like The Darjeeling Limited (Q658944), an American produced movie, which takes place in India, with a majority Indian supporting cast and only has three white lead actors, would not pass the test unless the Indian characters talk to each other about something other than their own or the leads race. If it was flipped, an Indian production taking place in America, and white characters talk to each other about something other than race for more than 5 minutes, it would pass the test.

Overall, the first point surpasses the other points. Thanks for the clarification question! Wallacegromit1 (talk) 14:28, 21 December 2020 (UTC)

Thanks, I agree. I readded this claim with reason for deprecated rank (P2241) cannot be confirmed by other sources (Q25895909) (should there be other sources backing up this claim, we may set the claim to normal rank, again). - Valentina.Anitnelav (talk) 14:39, 21 December 2020 (UTC)

The Lion King (Q36479)

I currently rated this as having no black woman and less than two ethnic minority characters. As to anthropomorphic animals I would only take those into account that are explicitly characterized as belonging to an ethnic minority (by being said so, behaviour or accent). I also would only take those minorities into account that are related to a real-world minority (one could think of the Hyenas as belonging to an in-narrative ethnic minority). I identified ethnic minorities with respect to the producing country (in this case: the USA). With this in mind there is only one character that I would identify to belong to an ethnic minority: Rafiki (Q326075). There maybe some people who think about this differently - so I put this up here with my reasoning. - Valentina.Anitnelav (talk) 12:23, 21 December 2020 (UTC)

@Valentina.Anitnelav:

For anthropomorphic animals or non-human characters, the actors who voiced or played them will be counted towards the score. This is how they were counted here Next Bechdel, in the notes. As many non-minority actors could also do the accents and behaviour. Wallacegromit1 (talk) 14:34, 21 December 2020 (UTC)

@Wallacegromit1: Thanks. I agree with this for characters where the ethnicity is not expressed. Do we still count characters as ethnic-minority characters if they are presented as such, even if they are played/voiced by a white? I think, for example, of Prince Faisal in Lawrence of Arabia who is played by Alec Guiness. - Valentina.Anitnelav (talk) 15:37, 21 December 2020 (UTC)

@Valentina.Anitnelav:

For Live-Action & Animation - In the spirit of the test, a ethnic-majority actor/voice-actor, inhabiting/voicing a ethnic-minority character will not be considered for the test. Alec Guiness will not be considered in Lawrence of Arabia.

Exception - If the same actor/voice-actor, plays or voices multiple characters in the Film/TV, then it could be considered. We can call it the "Apu Clause", and add dubious (Q104378399). Examples include, characters in the TV series The Simpsons (Q886), or multiple genders in The Nutty Professor (Q936576), for gender based tests. A unique example would be Edna Mode (Q2490541) in the The Incredibles (Q213326), which will not be considered for ethnic or gender tests, but in Incredibles 2 (Q24832112) she will, as the voice actor did a few more voices in the second film.

Thanks. That makes sense. Now I have to recapitulate my edits for these tests... Should we make this clear in the test description? - Valentina.Anitnelav (talk) 14:13, 23 December 2020 (UTC)

@Wallacegromit1: I added the additional clause to the Waithe test and the Shukla test. Now I'm thinking about the Riz test and Aladdin (Q215518). Most (all?) the Muslim characters are voiced by white actors. If we don't consider them we would end up with Q102182612. But there is at least one source that considers them and rates this film as Q104145006. Should we exclude the Riz-Test from our additional clause? Should we use an additional qualifier that makes it explicit that also Muslim characters voiced by white actors where considered? Something along the lines of "considering also Muslim characters voiced by non-Muslim ethnic-majority actors"? -Valentina.Anitnelav (talk) 11:55, 27 December 2020 (UTC)

Caché (Q383173) and the Riz test

I find this film very hard to rate. The problem is Muslim character/s are talking about, the victim of, or the perpetrator of terrorism or extreme violence (Q102182821). One could argue that Majid is the victim of psychological violence (performed by French character Georges Laurent, both by lying to his parents about Majid in the past and by accusing him of terrorizing him) and that he predominantly talks about the psychological violence/terror he is accused of by Georges Laurent (terrorizing his family by sending violent drawings). But I'm not sure if this is the kind of terrorism/extreme violence covered by Muslim character/s are talking about, the victim of, or the perpetrator of terrorism or extreme violence (Q102182821).

The decision is complicated by the fact that the film's main topic is the collective guilt that comes from colonisalism and that it refers to Paris massacre of 1961 (Q200472) and Iraq War (Q545449). This puts the character into the context of the cases of extreme violence/terrorism I'm certain the test should cover without the character itself actually being involved with it.

Currently I rated it as passing but I change my mind every other day so I would be glad about a second opinion. - Valentina.Anitnelav (talk) 16:00, 28 December 2020 (UTC)

I just realized that Majid's parents disappeared during the Paris massacre of 1961 (Q200472) which would make him an indirect victim of extreme physical violence. I noticed only today that the description of Muslim character/s are talking about, the victim of, or the perpetrator of terrorism or extreme violence (Q102182821) was broadened to include extreme violence quite general. If violence triggered by anti-Muslim sentiment should be included, too, this film should probably fail. But is this actually the case? - Valentina.Anitnelav (talk) 16:26, 28 December 2020 (UTC)

@Valentina.Anitnelav:

I have never seen this film, but have heard about it. You have given the film, passes the Riz Test with Dubious, which I think is perfect. However, if you feel the level of psychological violence is too directly or indirectly linked to Islam, I would say you can add in fails the Riz Test with Dubious. The Dubious should remain. Please also see if the criteria Muslim character/s are presented as a threat to a Western or non-Islamic way of life (Q102183028), is also being impacted in some way?

If violence triggered by anti-Muslim sentiment should be included - Is this shown to be directly related to Islam and the characters? Is it is Indirect, it may not apply.

@Wallacegromit1: I did not read the psychological violence as being linked to Islam directly but rather to the repression of colonial guilt. Muslim character/s are presented as a threat to a Western or non-Islamic way of life (Q102183028) is also an interesting question; as the character of Majid evocates guilt in the main character of Georges Laurent, also with respect to his life style (Georges Laurent became rather successful while Majid stayed poor) one could read it as a kind of threat to the legitimacy of the lifestyle. I would tend to say that this is not covered by Muslim character/s are presented as a threat to a Western or non-Islamic way of life (Q102183028), but here, again, it is a bit dubious.

The events of Paris massacre of 1961 (Q200472) rather provide the background of the story, they are not a part of the actual plot. They are mainly shown as related to colonialism, in my reading. - Valentina.Anitnelav (talk) 13:46, 29 December 2020 (UTC)

I should add that the character of Georges Laurent does certainly feel directly threatened by the violent drawings and the surveillance tapes of his home. But the sender is never revealed in the film. Georges Laurent accuses Majid and his son of sending the pictures and tapes, but the film does not frame these accusations as fair or justified. In the end this would be a threat to one family, not to Western lifestyle quite generally. - Valentina.Anitnelav (talk) 14:00, 29 December 2020 (UTC)

I think I will keep passes the Riz Test with Dubious. Multilayered films that open a wide room of possible interpretations and associations (also of threats) are difficult to rate. If keeping directly to the plot without thinking about associations the film allows and evocates I think the film passes. But, of course, this discussion may be reopened later. - Valentina.Anitnelav (talk) 14:22, 29 December 2020 (UTC)

Waithe Test Good Relatioinship

@Valentina.Anitnelav: Is there a way to show that a Black Woman, is not in a position of power, but is in a healthy relationship? As the point system does not allow for this value to be entered. Wallacegromit1 (talk) 05:31, 14 February 2021 (UTC)

@Wallacegromit1: This is true. One could create an item for the statement "At least one Black Woman is shown in a healthy releationship" and add it as a qualifier with complies with (P5009). - Valentina.Anitnelav (talk) 10:47, 14 February 2021 (UTC)

Increase diversity in our Wikidata integrations research

Hello WikiProject Media Representation: Ethnic Portrayal editors!

Wikidata for Wikimedia Projects at Wikimedia Deutschland is focused on enhancing data reuse and functionality between Wikidata and its sibling projects. We are sorely lacking diversity in our research and are seeking fresh perspectives.

We are conducting 1-hour interview calls to understand editors' experiences (positive & negative) when integrating Wikidata with other Wikimedia projects (Wikipedia, Commons, Wikisource etc.). Your insights will help ensure the products we build are suitable for all editors, regardless of gender, sexual-orientation, language or experience level. We also invite you to write to us if you prefer to contribute in written form.

Interview Information:

• Interviews will be conducted in English.
• Compensation is available.
• Sessions will be scheduled soon and on a rolling basis.
• Sign up here if you are interested.
• Diversity: We are keen to hear from women editors; editors with a diverse gender or sexuality; beginner or casual editing experience and/or those who edit in non-latin or non-western languages.

Thank you for your interest in contributing to our research, Danny Benjafield (WMDE) (talk) 07:36, 19 February 2024 (UTC)

Propose renaming all tests after demographic rather than honored individual

I like these tests and I think Wikidata is a great place both to import data and create new data to report these evaluations.

Wikidata benefits from having these tests. These tests apply to all sorts of contexts.

I question and challenge the current trend of naming tests after honored individuals or creating new texts to honor more people. There are many minority demographics, any of whom might appreciate having a media representation test to sort portrayal. I anticipate the ongoing need for more tests.

At the same time, the current test names are not translatable or globally relevant. To understand a test a user would need to look up the test and perhaps the named person, whereas an alternative could be to use translatable, understandable terms for each test. Names could follow the model "(demographic) media representation test", in which case the tests are more accessible to people of more cultures as people could translate test names into their own languages to understand them.

I propose

1. give tests from here forward generic names which are translatable
2. consider renaming the current tests in circulation to be general translatable terms

Thoughts from anyone else?

Blue Rasberry (talk) 23:55, 21 December 2020 (UTC)

Maybe we could do something like "(proper name) test for the representation of (demographic) in the media".

So we would have "Bechdel test for the representation of women in the media", "Mako-Mori test for the representation of women in the media", "Landau test for the representation of women in the media", "Riz test for the representation of Muslims in the media", "Waithe test for the representation of black women in the media".

I would hesitate to leave out the name of the person that created it or that is honored by it, if this is part of the common name (one would not leave out "Beck" in Beck Depression Inventory (Q813655) or "Pimsleur" in Pimsleur Language Aptitude Battery (Q7194732)).

This way we would also avoid ambiguities (we have several tests for the representation of women in the media). - Valentina.Anitnelav (talk) 22:24, 22 December 2020 (UTC)

Wallacegromit1, focus on media historiography and works from the Global South Valentina.Anitnelav (talk) 14:12, 16 November 2020 (UTC) Blue Rasberry (talk) 23:59, 21 December 2020 (UTC) Maxime Lijil (talk) 10:03, 4 February 2024 (UTC) PaperHuman (talk) 00:56, 7 March 2024 (UTC)

Notified participants of WikiProject Media Representation

Welcome Blue Rasberry!

1. give tests from here forward generic names which are translatable - This can be accomplished in the Aliases section, tests related to a particular subject matter have already been added to some of the tests.
2. consider renaming the current tests in circulation to be general translatable terms - I thought about that as well, however, as Valentina has pointed out, there are many gender, ethnicity and LGBT+ related tests that are already there and more to come. It will become harder to differentiate.

Thanks for your clarification queries. Wallacegromit1 (talk) 02:59, 29 December 2020 (UTC)

• Help:Label explains how labels are determined in Wikidata. I don't really see how we could substitute our description to the actual name of a test. The description can always go into the description. Bear in mind that each test has its own POV. --- Jura 13:48, 31 December 2020 (UTC)

Propose to redirect all talk subpages to here

I propose to redirect all talk subpages to here.

Currently this project contains subpages each of which has its own talk page, such as with

• Wikidata talk:WikiProject Media Representation/Ethnic Portrayals

The disadvantage with this is that users would need to watch each talk page to see activity, rather than the one main talk page. I recommend centralization of talk pages as a general practice for all WikiProjects.

Thoughts from others? Blue Rasberry (talk) 23:58, 21 December 2020 (UTC)

I see. I think the distinct talk pages could be useful for people interested in only one topic/aspect and it would be easier to look up conversations about one topic (e.g. if there are doubts about the application of one test to look back if there already has been a similar case). Is there a way we could still keep relatively permanent topical sections (e.g. gender, ethnicity, lgbtqa+) with each topic having its own archive? - Valentina.Anitnelav (talk) 18:56, 22 December 2020 (UTC)

I included the talk pages from our subpages here. I hope this makes it easier to keep track of what is going on on related talk pages. - Valentina.Anitnelav (talk) 16:46, 31 December 2020 (UTC)

Fails compliance with

@Valentina.Anitnelav, Wallacegromit1: Please comment on the proposal for a "fails compliance with" property. Does this work for your tests, or does one of the other proposals (in the comments) fit better? Omegatron (talk) 14:50, 20 December 2020 (UTC)

In my opinion none of the proposals is perfect, currently. I will come up with some options or thoughts how we could make the proposals workable in the next weeks. Wallacegromit1 or anybody else who is interested in this: please feel free to do the same. I would prefer to first discuss them here and then comment at Wikidata:Property proposal/fails compliance with - Valentina.Anitnelav (talk) 15:20, 20 December 2020 (UTC)

It seems to me that the complies with (P5009)/fails compliance with model is not going to work for all of our tests. The cases which can not be covered are

• The DuVernay test which does not deliver a pass-or-fail result, but just a score
• Cases when a test is not applicable due to some missing non-obvious preliminaries. This may be the case with Riz Test (Q102180036), for example, as this only asks about the portrayal of Muslims, if there are any in the work. If there are no Muslims, this test is simply not applicable. It would be an advantage to be able to make it explicit that this test is not applicable (in difference to "not applied" - which would be indicated by a missing value).

I would favour a model with as few properties as possible (so to need a third property to capture tests with a score would be rather a disadvantage) and I would favour a model that offers as much flexibility as possible to capture new tests with their peculiarities.
This said I would prefer the assessment (P5021) strategy supported by Jura and Yair rand. assessment (P5021) is restricted to humans, currently. So we would either need a new property or we would need to broaden the domain of assessment (P5021). If we use this model, we would need an additional property outcome (or something similar). I will give some examples how I would model our current statements using the assessment (P5021)/test score (P5022) model (I will use assessment (P5021), for now, even if it is not clear if this should be used for other domains than humans):

assessment	Bechdel test edit outcome passes nature of statement rarely 0 references add reference Vito Russo test edit outcome fails test score 2/3 0 references add reference Riz Test edit outcome not applicable 0 references add reference Landau Test edit outcome fails criterion used third Landau or Reverse Landau Test criteria 0 references add reference DuVernay Test edit test score 1/5 criterion used none of the characters are whitewashed or played by an actor of a different ethnicity 0 references add reference
		add value

This would be my suggestion.
The following are just additional thoughts and ideas: the complies with (P5009) property could still be useful as a qualifier for the DuVernay test, for example: one could have a score along with a list of complies with (P5009) values, instead of using criterion used (P1013). To use criterion used (P1013) here seems a bit odd, to me, as "none of the characters are whitewashed or played by an actor of a different ethnicity" was not the criterium that was crucial for the outcome (failure or pass) ... Alternate example of the DuVernay test using fails compliance with.

assessment	DuVernay Test edit test score 1/5 complies with (P5009) none of the characters are whitewashed or played by an actor of a different ethnicity 0 references add reference
		add value

- Valentina.Anitnelav (talk) 11:12, 23 December 2020 (UTC)

• I like the approach with "test taken" (maybe the label could be improved to "assessment") and a qualifier "outcome". --- Jura 11:15, 23 December 2020 (UTC)

• All of this looks suitable for the project. I will differ to your judgments. Thanks! Wallacegromit1 (talk) 01:05, 4 January 2021 (UTC)

Feldman Test

I am planning to create the below:

The Feldman Score/Test

• has characteristic (P1552)=passes the Feldman Test

• has characteristic (P1552)=fails the Feldman Test

If the Feldman Test is Passed or Failed, the following can be used as qualifiers;

• criterion used (P1013)= 2 points for a female writer or director (Co-writer or co-director are also counted? Maybe an additional qualification with the score of 1, half of the original)

Or,

• criterion used (P1013)= No female writer or director

• criterion used (P1013)= 1 point for a female composer or director of photography

Or,

• criterion used (P1013)= No female composer or director of photography

• criterion used (P1013)= 1 point for three female producers or three female department heads

Or,

• criterion used (P1013)= Less than three female producers or three female department heads

Or,

• criterion used (P1013)= No female producers or female department heads

(In the notes section of the article, the following was said: A department head was defined as the person who could win the Academy Award for the position. Only those credited as “producer” — not “executive producer” or other adjective-modified producer — were counted toward the female producer quota. Should we have the same conditions? We should definitely change the Academy Award position?)

• criterion used (P1013)= 2 points if there’s a female protagonist who determines story outcomes

Or,

• criterion used (P1013)= No female protagonist who determines story outcomes

• criterion used (P1013)= 2 points if no female characters were victimized, stereotyped or sexualised

Or,

• criterion used (P1013)= Female characters were victimized, stereotyped or sexualised

• criterion used (P1013)= 1 point if a sex scene shows foreplay before consummation, or if the female characters initiate or reciprocate sexual advances

Or,

• criterion used (P1013)= No sex scene show foreplay, nor any female characters initiating or reciprocating sexual advances

• Use review score (P444) to provide the final score, for pass or fail!

@Wallacegromit1: I agree, just a few comments:

As to the Feldman Test I would prefer the second qualifier option (2 points for a female writer or director etc.).

As to the last point of the Feldman Test: I'm inclined to give it also to films without any sex scene. I had a look at the films judged to pass the Feldman test (https://projects.fivethirtyeight.com/next-bechdel) and these include Arrival and Finding Dory. So also films without sex scenes are scored. I don't think that films without sex scenes should be rated lower for not including a sex scene (sometimes sex is just nor relevant to plot, character development, topic or tone of the film) and it does not seem to be the intent of the test.

I would stick with excluding executive producers and other producers (one could make exceptions in cases where a female ...-producer is considered extraordinarily relevant for the film). One should probably generalize the "academy-award-clause" to include all film awards (or film-related awards).

@Valentina.Anitnelav:

Please see Feldman Test Update Below, let me know if any further questions exist, otherwise we can start working on the below till next year. The first criteria has an additional point clause, and the sexualised aspect has also been covered with the protagonist. If no questions then I will modify the test immediately.

A movie passes with a score of five or higher:

• 2 points for a female writer or director (if co-writer reduced to 1 point) - No Co-director, as the are extremely rare and is not a professional title.

• 1 point for a woman composer or director of photography

• 1 point for one woman full producer (not co, associate, or EP)

• 1 point for Four women department heads

• 1 point for 50 percent women crew

• 2 points if there’s a non-sexualized woman protagonist who determines story outcomes

• 1 point if female characters are not victimized or stereotyped

• 1 point if a sex scene depicts foreplay or sexual pleasuring of the woman, preceding or instead of intercourse, or if the woman character initiates sexual advances and expresses her desire in a mode that is not only about servicing the male character’s impulses

@Wallacegromit1: Agree with the first 7 points. I like the modification "if the woman character initiates sexual advances and expresses her desire in a mode that is not only about servicing the male character’s impulses". I would just replace "the male character's impulses" by "a male character's impulses" (to be inclusive to cases where the other character is not male - "the" somehow suggests that their is a male character). I encountered some films which received the last point and will loose it again, now. Does reciprocating sexual advances still count (with the same modification)?

And: do we keep it that way that kissing counts, even if it is not framed as linked to a sexual desire but rather to "romantic desire"? To include the expression of romantic desire would make it more inclusive to views on intimate/romantic relationships linked to asexuality (Q724351). I think I understand the point of the last criterium (to point out female sexual agency) and one could argue that it is underrepresented, at least in some countries. But, on the other hand, if only accepting advances motivated by and expressing sexual desire and not advances where this is left unclear or rather framed as romantic it may be read as viewing people not having and thus not expressing sexual desire as somehow deficient which would be somehow discriminatory against such kind of views. We could also create an own item for a modified version of this last criterium (including advances motivated by romantic desire) to be able to discriminate between both, later, if this should be desirable.

The last passage is rather a "train of thought", but I hope it is still understandable. - Valentina.Anitnelav (talk) 11:50, 18 December 2020 (UTC) Edit: I made some alterations to my train of thought. - Valentina.Anitnelav (talk) 13:12, 18 December 2020 (UTC)

To summarize my "train of thought" in one question: Do we only count advances (e.g. kissing, or other ways of expressing romantic affection to another person) if it is explicitly framed as sexual? Should we put every advance that is not framed as explicitly sexual into missing prerequisite for Feldman Criterion (h) (Q104176605)? - Valentina.Anitnelav (talk) 13:12, 18 December 2020 (UTC)

@Valentina.Anitnelav:

• Have asked to clarify the last point above.

• Do we only count advances (e.g. kissing, or other ways of expressing romantic affection to another person) if it is explicitly framed as sexual? - No. I believe many non-sexual advances can be very intrusive, as such can be counted towards reciprocation. There should be a score for when someone winks at a character, and the other winks back. Or, if someone wants to hold hands and the other one doesn't, AKA "Rey/Finn Clause", they will not get a point. Hope that makes sense.

• Should we put every advance that is not framed as explicitly sexual into missing prerequisite for Feldman Criterion (h) (Q104176605)? - Maybe we should change the wording to Feldman criteria (g) not applicable. This should be used when no foreplay or sexual/romantic advances exist in a work, where maybe purely platonic or family relationships exist. Gazing at someone will be not be applicable, staring into each others eyes will. Wallacegromit1 (talk) 17:08, 18 December 2020 (UTC)

@Wallacegromit1: Thanks for the answer! I was just at formulating a thought as to criterium 1, thus my quick reply: I think women credited as co-directors should be counted if their contributions are widely recognized as significant. This is for example the case with Ágnes Hranitzky (Q1056063) and Werckmeister Harmonies (Q509169) where she is often credited as director next to Béla Tarr (see Mubi entry: https://mubi.com/de/films/werckmeister-harmonies). I have the vague memory that she was only credited as a co-director as some awards only admit one director... But I could not find any sources, now, so this is not very accurate, probably :) -Valentina.Anitnelav (talk) 17:18, 18 December 2020 (UTC)

@Valentina.Anitnelav:

• Yes ofcourse! I think we should think of it in the way you wrote it. If only one person is allowed to get the award, will it be a woman? I think this could be our exception. Captain Marvel (Q23781129) has two directors, with Anna Boden (Q4766862) as the first billed and Ryan Fleck (Q3453777) as the next. If an award was given, with only one of them winning, I can say for certain it will go to Anna Boden (Q4766862). Even if not, there are ofcourse rare cases where the contribution is clear and apparent, and the point could still be given out, if enough clarity exists publicly. Example: the Wachowskis (Q195719)

• It is really only a vague memory, so it should not be taken at face value :). I'm not completely sure if I got you right: In my opinion Werckmeister Harmonies (Q509169) should get at least one point for being co-directed by a woman. The reason: Ágnes Hranitzky (Q1056063) is generally recognized as a director (without the co-) in secondary sources/databases/reference works, even though the screen credit was only co-director. When bringing up Ágnes Hranitzky (Q1056063) and (my vague memory of) the awarding procedure I was rather thinking of the following: We should also keep in mind that there are many women not fully credited for their work, especially if they worked together with men and especially for older works. In these cases we should also follow assessments by other authorities and not just look at the screen credit. - Valentina.Anitnelav (talk) 18:18, 18 December 2020 (UTC)

@Valentina.Anitnelav: Have asked Mz. Rachel for clarification with the Captain Marvel (Q23781129) example, as even the writers are two women and one man. Will keep you posted. I believe we are thinking too much about this, I personally believe the movie should be getting the point for a female director, and you are right about the experts opinion, as not always do women get credit. Wallacegromit1 (talk) 18:34, 18 December 2020 (UTC)

@Valentina.Anitnelav:

Mz. Feldman has asked us to decide for ourselves. As such, I believe, if a work has multiple directors, and 50% or more are women, then they should receive the 2 points. If less than 50%, then 1 point only. We can use nature of statement (P5102) = attribution, for works by women discovered or credited later.

And, as for the last criteria - 1 point if a sex scene depicts foreplay or sexual pleasuring of the woman, preceding or instead of intercourse, or if the woman character initiates sexual advances and expresses her desire in a mode that is not only about servicing a male character’s impulses. "the" has been changed to "a", and she designed the test for male-female dynamics only, so we can either use "Not Applicable" or, as they would automatically not be servicing male impulses in female queer only films, then maybe we can give them a point?

If all is alright with the above, I will update the test, and some of its options.

@Wallacegromit1: Yes, I agree. We may create several "not applicable" items - e.g. "criterium g) not applicable as there is no romantic or sexual relationship involving a female character" and "criterium g) not applicable as the only romantic or sexual relationship involving a woman is a same-sex relationship" (maybe we could think of something shorter). I'm not sure about giving the point by default (via review score (P444)) if this is not what Feldman intended. To me it seems reasonable to give the point. However we decide: When writing a query one could specify that every item with "criterium g) not applicable as the only romantic or sexual relationship involving a woman is a same-sex relationship" should receive one additional point. So data reusers could decide to circumvent the constriction to male-female dynamics (if we apply it). Here is a modification of our example query, where every film that did not meet criterium g) due to it not being applicable gets one point nevertheless: [query]...

BTW: Feldman dropped the "reciprocating sexual advances" clause, so we should, too, I think? Interestingly, films that don't stereotype or victimize female characters only get 1 point, now - so to pass a film definitely needs major female contributions, now. - Valentina.Anitnelav (talk) 20:09, 19 December 2020 (UTC)

@Valentina.Anitnelav:

• The query for alternative score is perfect, Thank You! Yes, lets not provide a point by default, and data re-users can add it with the query later.

• I am updating the Feldman Test, and lets see what Not Applicable item will be better suited. Also, please go back and update every item we have added the test to, some points may be increased or decreased.

• films that don't stereotype or victimize female characters only get 1 point, now - so to pass a film definitely needs major female contributions - Yes, which is in keeping with the tests original idea about the environment in which the work is created. However, 2 points if there’s a non-sexualized woman protagonist who determines story outcomes, has allowed us to add points for many animated films, which oddly have many men in the staff. Fun Test!

For the early years: If the women in the ink and paint department then would have gotten the same credit as men in the visual effects department today, the male/female ratio in the crew would probably look a bit more favourable. And one should not forget about the women in the story, concept and/or character development department, who did not always receive credit (some of the more well known: Mary Blair (Q231082), Retta Scott (Q4977804), Bianca Majolie (Q2900817), Sylvia Holland (Q3507221), Ethel Kulsar (Q28167908)) Film makers/producers then did not always take it too serious with giving credit to their crew - of course this did not only affect women, but men, too. I will stick with the imdb credits, for now, but if I feel like digging into Ink & Paint: The Women of Walt Disney's Animation I would probably take these into account, too. - Valentina.Anitnelav (talk) 14:48, 20 December 2020 (UTC)

@Valentina.Anitnelav: I think this is tricky. If we keep to the "people with real power and awards" element, they should not. However, they certainly made major contributions in shaping the film experience, but sadly not given their due. As long as there is an external source that re-instates their contribution, I would count it, as they would have been acknowledged today. You should absolutely use Ink & Paint: The Women of Walt Disney's Animation, as a source for film history and credit, and if they are credited in the book as department heads, or the only people in the department, count them in. This would also give notice and attention to women in the industry pre-1950s. I was hoping these tests would allow us to find hidden information like this. And these sources can help us re-instate their power and prestige. Wallacegromit1 (talk) 15:11, 21 December 2020 (UTC)

Feldman Test for Animated Films

We have to think of a good "replacement" for the director of photography in animated films (this position does often not exist there). I already have some ideas in mind: Adding "female animation director" as another possibility to get this one point, if there is one (this is not always the case...). Sometimes there is a group of supervising animators/animation supervisors (that supervise different aspects of the animation process - mostly the animation of different characters). In this case one could think of something like ">50% female supervising animators". One could also think about taking layout artists into account (they develop the "camera positions", besides others - but in most cases they are not viewed as that influential on the whole film as is the case with directors of photography). Until now no film came even close to the numbers I have in mind, but I wanted to put this issue on our agenda. - Valentina.Anitnelav (talk) 11:05, 21 December 2020 (UTC)

@Valentina.Anitnelav:

Are there Key Animators or Animation directors, someone as key as the director of photography? Maybe someone who sets the art style? I do not know enough about animation, maybe we could ask in Wikiproject Movies about an alternative to a cinematographer?

I like the 50% for supervising animators, but only if no other major credit could be found. Wallacegromit1 (talk) 14:58, 21 December 2020 (UTC)

@Wallacegromit1: With animators "key" does have a different meaning than in other departments (Key animators are those that draw the key frames - they are not necessarily those in authority). My suggestion would be the following:

• First have a look at the animation director
• if there is no animation director, have a look at the animation supervisors/supervising animators. If there were at least 50% female animation supervisors and at least half of the main characters were supervised by women this would be counted. "Supervising animator" is actually a major credit for the traditional Disney films. Disney's Nine Old Men (Q241920) where supervising animators and the supervising animators are sometimes also those mentioned in Animation award nominations, if there is no animation director
• If there are no animation supervisors/supervising animators as this is a smaller production look at key animators.

The art style is set by concept artists, not by the animators themselves...

My reasoning for thinking of animation directors/supervising animators: They decide how the characters are framed and are in that way similar to a director of photography. - Valentina.Anitnelav (talk) 15:11, 21 December 2020 (UTC)

@Valentina.Anitnelav: Perfect! Agree with the steps above. Can you also add this in the instructions for the test please. Wallacegromit1 (talk) 15:16, 21 December 2020 (UTC)

I just added the first two points. I'm not sure about the third one. Maybe other positions should be taken into account when there are no supervising animators. With supervising animators and animation directors we already covered a lot of cases, I think. Unfortunately the division of work and the titles assigned vary a lot in animation production and it is not always clear what is meant, especially on an international level. - Valentina.Anitnelav (talk) 16:02, 21 December 2020 (UTC)

@Valentina.Anitnelav: Lets say if the first two points are not met, then no point given, unless a similar or higher role exist in another industry. Wallacegromit1 (talk) 18:22, 21 December 2020 (UTC)

Ambiguity of "Protagonist", male gaze and sexualization

@Wallacegromit1:

1. Ambiguity of "Protagonist": Strictly speaking the protagonist is the central character of a story. There is usually only one, sometimes two - e.g. in the case of buddies or couples that share the protagonist role. But in everyday use this word tends to be used also for main characters (on the protagonist's site) that are not the central character but, for example, the female lead or the deuteragonist (Q2479076). I applied 2 points for Feldman Criterion (f) (Q104144812) in the strict sense, including only the most central character and excluding female leads (in the case of films with a male protagonist). I'm not sure if this is what is intended by the test. How do you read this criterium?
2. Male gaze and sexualization: With a rather wide reading of "sexualization" (e.g. as male gaze) criterium f) is actually hard to be met by any film in current film history. I guess that most of 20th century's famous female protagonists (in a strict reading) and female leads are sexualized in a male-gazey way at some point during the film (in some genres more, in some less). May I assume that "sexualization" refers to a narrower meaning of the word, something like "the female protagonist (or lead) is mainly defined by her sexual attractiveness or physical desirability"? - Valentina.Anitnelav (talk) 12:45, 2 January 2021 (UTC)

@Valentina.Anitnelav:

# Protagonist - I have been doing the same or basing it on deuteragonist (Q2479076). Unless, there is only one female character who has a significant screen time or has been integral to the plot, and is not considered a lead/protagonist. This would be on a case by case basis, but the stricter version should be adhered to, as much as possible.

# Male gaze and sexualization - This is super tricky! So I have not given a point to Wild Nights with Emily (Q55287256) for criterium f, as the whole premise is the sexualisation of Emily Dickinson (Q4441), but in a social way. However, there is no nudity and no sex scene, only some kissing. In terms of Male Gaze, there is none. I have given a point for criterium f to Band Aid (Q28173417), even though there is sex and nudity, but not in a way that is meant to be sexual for the audience. I have been uncertain about the sexualisation aspect, as I do beileve it is about Male Gaziness, but I went with a more literal interpretation, as it is really tough to apply Male Gaze objectively. What do you think? Wallacegromit1 (talk) 00:28, 4 January 2021 (UTC)

@Wallacegromit1:

Protagonist - Agreed

Male gaze and sexualization - Thanks for your example of Wild Nights with Emily (Q55287256). I agree with your decisions above. I would even admit some "transmission area" where some male gaziness (some sexualization) may show in a sex scene.

Since the "sexualization" clause was moved from now-criterium g) to now-criterium f) I have the following question in mind: when is there so much sexualization that it eradicates the other condition of the clause (that she determines story outcomes). For example: I think it is reasonable to call Ellen Ripley (Q988925) sexualized when she has to fight the Alien in her underwear (youtube). But I would still find it a bit too harsh to deny this film the 2 points for having a female protagonist that determines story outcome. I think that the sexualization has to appear throughout the film or in a significant amount of scenes.

I agree that "male gaze" is tough to apply objectively. I'm just thinking of collecting a couple of questions that may decrease the level of subjectivity in this criterium. Maybe something like this:

• Are her sexual attractiveness/desirability one of the most dominant feature of the female protagonist or the features that have the biggest impact on the plot?
• Is pleasing a man or men shown as the fulfillment of her role?
• Is she presented as an object of sexual desire in the majority of scenes she appears in?

One example of a film that would fail here, in my opinion, is Once Upon a Time in the West (Q168154). Jill is not a protagonist but the leading lady (one could argue that she is the tritagonist (Q3775957)). For the sake of argument I would assume we consider her here. One could actually argue that she determines plot outcomes (she is the one that builds the community there) but does not pass the "non-sexualized" condition: While she is definitely presented as having other qualities, too, I still think that her sexual attractiveness is the one that is central to the film. - Valentina.Anitnelav (talk) 09:41, 4 January 2021 (UTC)

We could also think about allowing a reduced number of points (1 point) for cases where there is a female protagonist (in the strict sense) that undisputedly determines story outcome but is also arguably sexualized to some extent during the film. This could apply, for example, to Alien (Q103569). (We should still consider giving no points if one of the questions given above can be answered with "yes"). This would make this kind of sexualization visible without making these films equal to films that don't have any female protagonist that determines story outcomes at all. - Valentina.Anitnelav (talk) 10:01, 4 January 2021 (UTC)

@Valentina.Anitnelav:

I Agree with most things above.

* I think your questions can be their own test in a way, and sound similar to the Sexy Lamp Test, which I will be suggesting. Regardless, I think for the Feldman Test Criteria, they are perfect in summarizing a "sexualised/exploitative" use of women. I think we should add your questions below the criteria, and would be interesting to see movies that fail this criteria and pass the Sexy Lamp Test, and Vice Versa. This would let us know the different interpretations of how women are used as plot devices, and provide a spectrum of nuance, which is the point of having different tests.

* I think Mz. Feldman changed it to non-sexualised in this criteria to allow animated movies to pass easily, and not directly about sexuality in general.

* I like the reduced point idea, as some movies are just too on the nose or the "Male Gaze" is not taken as a negative or offensive, but more a characteristic of the scene, especially in parodies.

@Wallacegromit1:

I have to admit that I'm a bit reserved regarding the Sexy Lamp test, as this certainly seems to be a a good conversation starter, but maybe a bit too blunt/subjective to be recorded in Wikidata. I'm generally a bit reserved regarding tests that may tend to be "offensive" towards characters - I had a look at a Huffpost article to have an impression of its actual application and I tend to find it a bit insulting and over the top - even though I get the gist of it :). It tends to overemphasize sexist characteristics of female characters and forgets about other qualities that do not fit into the scheme, which is absolutely ok as a satirical instrument but maybe not in Wikidata... (I also think that it would be difficult to source)

I think that Feldman criterion f) actually gets the point of the Sexy Lamp test with a better wording. We could add the Sexy Lamp test as one aid to decide about Feldman criterion f) (e.g. "Could the female protagonist be replaced by a sexy lamp for most part of the film?"). Would this be okay with you to utilize the Sexy Lamp test for the Feldman test and maybe not as a test on its own?

I thought about the reduced point. Let us take this approach, for now:

Central questions to help to decide as to Feldman criterion f):

1. Is her sexual attractiveness/desirability one of the most dominant features of the female protagonist or the feature that has the biggest impact on the plot?
2. Is pleasing a man or men shown as the fulfillment of her role?
3. Could she be replaced by a "sexy lamp" for a large part of the film?
4. Is she presented as an object of sexual desire in the majority of scenes she appears in?

• If a film fails one of the first three questions: 0 points for criterion f)
• If a film fails the fourth question and her impact on story outcomes is rather minor: 0 points for criterion f)
• If a film fails the fourth question and she has a major impact on story outcomes: 1 point for criterion f)
• If the number of scenes where the character is said to be sexualized as to question 4) is rather small and is of rather little relevance to the character's characterization: 2 points for criterion f) (this would be the Ellen Ripley case)

Some thoughts about sourcing: If a film is said to (partly) fail criterion f) this should be sourced, especially if it is a rather well known film. If a rather unknown film is said to (partly) fail criterion f) and there is no source the sexualization has to be particularly evident.

I will add this to the non-sexualized protagonist clause. This simply helps to determine how we apply it. We can expand on it as we find more and more examples/cases.

Just one question out of curiosity: "I think Mz. Feldman changed it to non-sexualised in this criteria to allow animated movies to pass easily". I'm not sure if I understand correctly: why do you think that it is easier for animated films to pass than for non-animated? I think that animated films provide a special challenge: one can argue that the character design itself is already sexualizing women; they are mostly designed as being particularly beautiful/attractive/cute - of cause, also Hollywood actresses need to be mostly beautiful to be chosen for a leading role, but at least you don't have the problem of deciding if her body is already sexualized :). I'm still at thinking about how to apply the criterion to animated films with this particularity in mind. - Valentina.Anitnelav (talk) 15:18, 5 January 2021 (UTC)

@Valentina.Anitnelav:

I had similar reservations with the Sexy Lamp Test. For now, I have removed it from the test suggestions.

Agree to all the questions! But in this regard Wild Nights with Emily (Q55287256) passes this criteria, should I change it, as I kept going back and forth with this one?

Animated movies - I meant to say children's or family movies in general. In terms of character design, if something is obvious like Jessica Rabbit (Q1631670) in Who Framed Roger Rabbit (Q379873), it would not pass the criteria. Ghost in the Shell (Q1066948) will pass barely or disputed, even though her body is heavily sexualised, she is presented as mostly melancholic or tough.

@Wallacegromit1: Sorry, I misread your Wild Nights with Emily decision: Yes, I think it should pass: Sexualization as a topic should be allowed as long as the protagonist is not sexualized on screen.

There is some debate as to the sexualization of Pocahontas (Q2625989), for example, based on her character design. I find this one tricky (especially as this is a much more sensitive topic). I just made it a disputed case. We should probably just stick to the sources, also for animated films - and in some cases a more "feminine" design may be more problematic than in other cases and may be seen as crossing the border where a similar design would be ok for a different character. - Valentina.Anitnelav (talk) 16:19, 5 January 2021 (UTC)

@Valentina.Anitnelav:

Wild Nights with Emily - Thank You!

Pocahontas (Q2625989) - This has been a subject of debate and using external sources for this would be better. Dubious is fine for now.

There is the Babs and Kara test, which looks at character design by examining whether specific characters would be recognisable even if they were wearing identical bathrobes and had their hair completely covered. Maybe a future test, not anytime soon though, harder to model and standardise.

Pocahontas (Q218894) Yes, I agree; I used external sources for that and both a disputed and a dubious. - Valentina.Anitnelav (talk) 17:48, 5 January 2021 (UTC)

@Wallacegromit1:: I created 1 point for Feldman Criterion (f) (Q104698048) that can be used for cases where the character design is arguably sexualized, but the character itself is not reduced to being sexually desirable. I think this could also be used for other genre films (like superhero films). - Valentina.Anitnelav (talk) 07:12, 6 January 2021 (UTC)

Incomplete assessments

@Wallacegromit1: In some cases a film clearly passes the Feldman test but the assessment is incomplete as to one or two criteria (e.g. Summer 1993 (Q29963333) clearly passes but I'm not sure as to criterion (h)). I currently just made no statement, which I think is correct, but I'm not sure about the score. I made it a 8/10 but this may be misleading. Should I make it a 8/9 to make clear that the assessment is incomplete? - Valentina.Anitnelav (talk) 07:36, 6 January 2021 (UTC)

@Valentina.Anitnelav: 8/9 is perfect! I was having similar issues with some films. I think this way others can see that some tests are not complete and can fill those criteria later. Brilliant!