Jump to content

Wikipedia talk:Manual of Style/Film/Archive 15

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 10←Archive 13Archive 14Archive 15Archive 16Archive 17→Archive 20

RFC on proposed production section guidelines

If we were to update the Manual of Style for productions sections of film articles what should it say. --Deathawk (talk) 04:26, 5 October 2017 (UTC)

For a couple years now I've been dissatisfied with the production sections for a variety of film articles. In talking with other editors who dabble in film articles, I tend to find that I'm not alone in this thought. Looking at the production guidelines here on the MOS it seems that it does not provide adequate instructions for making what I'd call a "good" production section.

I'm gonna quote myself from a discussion I had with @NinjaRobotPirate:

"I feel like a lot of times Production sections are A) Either added when the article is created because editors feel like that's what needs to be done or B) Are overly long and contain an excessive amount of detail. When it's A, you can usually tell because the news is really generic and almost has like a template to it at this point. Both of these types of sections, I feel harm Wikipedia as they are not really readers I feel will turn away from these sections after encountering too many of them"

I would really like us to nail down, what we think, if anything, the MOS should say for production sections. For reference here is the link to a previous (failed) attempt to update it. --Deathawk (talk) 04:26, 5 October 2017 (UTC)

  • I agree that many "Production" sections are essentially a string of events with the constant announcement of the precise date and what periodical reported the information. I am not sure what exactly to suggest that would be applicable across all such sections. My preferred approach is to have a summary sentence at the beginning because too often we start in the middle of the production "action". The lead section summarizes the article body, so I believe a section should stand alone and be able to state, "So-and-so was directed by so-and-so based on so-and-so whatever." But I'm not sure how many others would agree with that. Other suggestions I have to improve the quality of such sections:
    • Wherever uncontroversial, leave out the source as unnecessary in stating a fact like when a director was hired for the film.
    • Avoid including specific dates since these are when events are reported, not necessarily when they actually took place. Stating the month and the year, or more broadly "early" or "late" in the year, suffices. Avoid repeating months and years and create better flow, e.g., "The following November, <event took place>."
    • Exclude information that may be extraneous from an encyclopedic perspective. Not all casting announcements have to be reported equally and can be trimmed depending on the actors' final billing.
Thoughts on these points? Anything we want to talk about in regard to "crew" lists or sub-sectioning? I don't know if crew lists are popular enough to be included in the MOS, but I think they should exist like cast lists exist even when the infobox states the names. It's sometimes the only opportunity to mention other credits like the costume designer, production designer, art director, etc. wherever applicable. Erik (talk | contrib) (ping me) 13:35, 5 October 2017 (UTC)
  • I agree with some of the points being made here like reducing repetitive formulaic proseline writing but disagree with others. For example, a director being hired for a film not matter how uncontroversial is a crucial step in the filmmaking process and a good article show provide some information on how and when he/she came on board.--TriiipleThreat (talk) 20:41, 5 October 2017 (UTC)
  • TriiipleThreat, my apologies, I did not word the first bullet point well enough. I mean to say to write "The studio hired so-and-so in May 2017 to direct the film," as opposed to, "On May 21, 2017, Variety announced that the studio hired so-and-so to direct the film." Hope that means we agree there? Erik (talk | contrib) (ping me) 21:13, 5 October 2017 (UTC)
  • I think we should have something stating that the wording should be from a production standpoint and not from an audience standpoint. In other words,report how it happened in the production process not when it was revealed to the audience, or just that it was. Does that at all make any sense? --Deathawk (talk) 06:48, 6 October 2017 (UTC)
  • Procedural close of RfC. "What should it say?" is an awfully vague and broad question to start an RfC for. Please come up with a proposal, preferably with preliminary consensus, and only then open an RfC. – Finnusertop (talk ⋅ contribs) 16:21, 5 October 2017 (UTC)
    Unless I'm mistaken there's nothing at RFC prohibiting broad questions, and I want to nail down a consensus that all editors are happy with. Furthermore in my message below the initial RFC I bring up the specific issues I want to solved, The alternative would be to create a draft version that may not appeal to everyone. which would be detrimental to the cause. --Deathawk (talk) 20:12, 5 October 2017 (UTC)
    I'm a bit confused as to how the question you're asking is either short or simple. DonIago (talk) 20:41, 5 October 2017 (UTC)
    Well, it's fairly common to have a "gather input" discussion like this, it's just usually not an RfC. When it is, it's best to state up front that the purpose of the RfC is to help identify what the issue really is, or to get suggestions to use as options in a later RfC, or whatever the goal is; and that the current RfC isn't intended to arrive at final language. (And this RfC could be edited to do that.) We've had a least two of those on other pages in the last couple of weeks, and WP has a long history of them, even if they're not the common format/intent of RfCs. Vague/broad ones are a bad idea mainly when the nature of the question or issue isn't clear, e.g. "what should we do about PoV in this article?", etc. Anyway, in this case I thin we're trying to ID what the actual issues to address really are.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  23:49, 6 October 2017 (UTC)
  • Generally agreed that "many 'Production' sections are essentially a string of events with the constant announcement of the precise date and what periodical reported the information'; and with 'wording should be from a production standpoint and not from an audience standpoint'. These are severable concerns, however. It would probably be best to identify what passages in this MoS page, relevant to what we're talking about, are unclear, and propose specific revisions to them. From such a discussion would probably emerge some kind of final change proposal, which could be a separate RfC to approve or reject that clarification.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  23:55, 6 October 2017 (UTC)
@SMcCandlish:: The production section barely says anything about how to create content for it, only given that it can be structured into subsections. I was curious so I looked back and it appear that prior to 2009 there was wording stating that it should be written in prose, but as a result a revamp this wording was deleted. You can find the last revision that had the blurb in it, here. From what I can tell this was an oversight on the editors part. --Deathawk (talk) 04:19, 7 October 2017 (UTC)
Well, we should clearly restore that, since it's a general MoS provision about sections that tend to attract "lists of trivia" (MOS:TRIVIA), which applies here. Beyond that, people should suggest specific wording or at least "points to cover" they want to flesh out this section with, or we'll just wring our hands a lot and get nothing done. Maybe start a bullet list. I'll start one below.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  05:28, 7 October 2017 (UTC)

List of points to cover

I'll start a draft list of points to cover. I'm trying to write this in "guideline-ese" already, so this list could work even if simply copy-pasted into the guideline. Feel free to directly copy-edit in this post, the purpose of which is to serve as a discussion draft, not my personal proposal. Just add something like "Revised by ~~~~" after my sig, I guess. Please keep it generically worded enough it will make sense in MOS:FICT as a general guideline for production of creative works.

  • A production section – like those on plot and critical reception – should be written as prose, not a bullet list.
  • Material should generally be in chronological order, but should not provide a litany of exact dates.
    • Reserve full dates for key facts, like commencement and completion of a production, release in primary market, and other events noteworthy within the entire context of the production.
    • Dates important for one production may not be for another. The date a production was announced to the public is usually not, unless there is some context to it (e.g. the announcement of a new work in a franchise, because of demand for more in the series; this does not apply to the average stand-alone work, for which there is no anticipation).
    • Avoid repetitive sentence introductions like "On 19 January 2017, ..." or "In May 2018, ...".
    • Some information may be better grouped rather than strictly chronological, especially if the production material may grow to the point of division into subsections.
  • Write with a production focus, not an audience or news one.
    • Include the fact that something important happened in the production process (and how, if relevant), not when it was revealed to the public or in what publication (that's already indicated by the source citations).
  • Do not include trivial factoids likely of interest only to serious fans or only to a select few readers. Facts should be of encyclopedic significance to a broad readership. Examples of trivia:
    • Name-dropping of non-notable people involved; coverage of day-to-day activity and minor business arrangements; what finance companies are backing the production; equipment brands used; and many others.
  • For works not already released, frequently update information, remove information that is no longer relevant or correct, compress material that has become less important over time, and avoid recentisms, as production proceeds. Remember that when something was announced is much less important than what was announced.
    • Do not include tentative or speculative information; Wikipedia does not attempt to predict the future. Do not state future-scheduled events as if certain.
    • Using the {{As of}} template is strongly recommended for any information that includes a projected future date. This puts the article in a category of pages needing updates by a certain year or month.
    • Generally do not include at all any material that will soon be redundant, irrelevant, or obsolete. For example, the announcement of a pre-production decision is usually rather meaningless after the production has already begun, as most production decision announcements are after release. Inclusion of such tidbits is often a maintenance problem.
    • Avoid hiring-related news for anything except key roles. Context should be provided for those that are included. Group them together into a date range when possible. See above about removing such details when they become irrelevant.

That's what I've got off the top of my head. Others may think of obvious missing points, or want to refine one with a sub-bullet to address concerns about undesirable things they frequently encounter in production sections

PS: I've written this in generalized enough terms that it can, with little if any rewriting, be applied to other media like TV, comics, video games, novel series, musicals, etc. We might eventually want this in MOS:FICT instead, with just a short WP:SUMMARY-style pointer to it from MOS:FILM, etc. .
 — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  05:28, 7 October 2017 (UTC)
Update: Have notified the talk pages of the guidelines on fiction, TV, comics, video games, novels (that there are so many suggests some additional merging), plus music (production stuff also applies to albums, musicals, etc.).  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  05:39, 7 October 2017 (UTC)
Revised to try to work in points by Deathawk, below.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  07:58, 7 October 2017 (UTC); revised 22:12, 8 October 2017 (UTC)

  • I guess I'll give my wishlist of things to cover. . For the most part I agree with @SMcCandlish: points, and I sought not to repeat too many of his, however since this has been on my mind a lot, I do want to repeat some of the main ones. I would disagree with just posting this to the MOS:FICT and simply referincing that in this MOS. We should highlight key points in both.
    • Avoid recentisims, especially if they'll become redundant in the near future (IE: Saying a film will take place in XXX is unhelpful because when the film comes out this will be common knowledge) instead focus on how these developments came about (IE Why did the production choose to set the story there).
    • Avoid coverage of facts that may only be of interest to a very select few (IE: what finance companies are financing the film, when the executive producer came on board)
    • Avoid the date that the film was announced to the public unless there is some context to it (IE: Fantastic Beasts and Where to find them can have such a section because it made after it was assumed the Harry Potter series was dead, however there is very little reason to provide this coverage for your average indie drama or comedy.
    • Avoid casting news for anything except the leads and group them together when possible. Ensemble films would be an exception but these should provide a range of dates rather than listing each individual actor. Context should be provided for those that are included.
That would I think cover most of my concerns. I may think of more later. --Deathawk (talk) 06:17, 7 October 2017 (UTC)
I integrated your points into the list above (hopefully did it well, and kept it worded broadly enough to apply to more than films). I had trouble with the first one, since it doesn't appear to be recentism but trivia (giving a factoid without giving context/relevance); and where a story is set is plot info not production info, anyway. I think it's covered generally now, though. However, the bit about ensembles seems like WP:CREEP; the fact that we want to require encyclopedic context in order to include a casting choice date pretty much rules out creating a list of casting dates for a whole ensemble cast, since there can't be an encyclopedically noteworthy story with regard to every such hire. Not dwelling on casting choice news also applies to crew and production (i.e., your executive producer point fits into the same "rule").

Yes, the key points should be WP:SUMMARY-style treated in each of the relevant topical guidelines (I clarified that, above). But we should not have every point repeated in each of them, because we centralize guideline material. When we don't, we end up with WP:POVFORKs between guidelines. Having PoV-forking going on in WP:P&G material is pretty much as bad as it gets on Wikipedia. It can sometimes cause years of disruptive conflict.
 — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  07:58, 7 October 2017 (UTC)

@Finnusertop, Erik, and TriiipleThreat: pinging previous RfC commenters. What do you think of the draft above? Anything wrong-headed? Obviously missing? Confusingly worded?  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  08:21, 7 October 2017 (UTC)

@@SMcCandlish: . It's starting to take shape, however I do not like the wording " For example, the announcement of a pre-production decision is usually rather meaningless after the production has already begun, as most production decision announcements are after release I understand what we are trying to get at here, but, I do think we could better word it. I guess my main concern, is I do a lot of my work in like how ideas are formed, how certain plot points came about, and broadly speaking that could be determined to be "Pre production decisions" I understand that's not what we're getting at, but could we better word it, so we make sure we don't kill these parts of the article. --Deathawk (talk) 08:40, 7 October 2017 (UTC)

Thus "the announcement of a pre-production decision" wording; it's the announcement and its date that eventually become obsolete trivia; it's probably the #1 production-section problem I see in articles begun before a production's release. The actual decisions, as you say, may sometimes be important, but what press release they were announced in on what day is not. If you still think it's unclear, have at a rewrite!  :-)  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  22:12, 8 October 2017 (UTC)
  • Comment I would like to see MOS more explicitly discourage a practice I've noticed on many TV articles, and a few film articles, where "Production" is euphemistic code for critical analysis of the final product that happens to come from primary sources, usually in the form of massive quote farms -- "Writing" means commentary by the writers on the final product, "Casting" means commentary by the cast members (or about how the cast members appeared) in the final producet, and so on. This is the case with literally all of our articles on the individual episodes of season 7 of Game of Thrones, and to a certain extent with the previous season, all of which are currently GAs. Maybe we wouldn't have so many bogus GAs if MOS:FILM and MOS:TV explicitly told editors not to do this and GA reviewers could quickly check the MOS to find that out. Hijiri 88 (聖やや) 09:11, 7 October 2017 (UTC)
@Hijiri88: I am not sure I fully understand what you are saying. In many cases interviews with writers, casts, and filmmakers are the best way of covering the issues. To ask us to exclude that information seems crazy, and would, I think, be detrimental to the project. --Deathawk (talk) 09:50, 7 October 2017 (UTC)
"Production" and "subjective analysis of the final product" are not the same thing, by any stretch of the imagination, and in the case of the individual articles I mentioned we actually can't write anything at all about "casting" since the actors were all cast in the production of previous episodes years earlier. Writing long sections of articles (in the case of the articles I mentioned, basically the whole articles) based on primary sources is not a good idea (in fact basing the whole article on primary sources is unacceptable). There's nothing "crazy" or "detrimental" about it. Hijiri 88 (聖やや) 13:16, 7 October 2017 (UTC)
@Hijiri88: Again, I'm not sure I understand what you're saying. Are you stating that we should avoid using quotes from interviews because they are primary sources? There is no, way to actually do that, and still cover the information reliably. That sounds like a bizarre stipulation to have. I almost think like we may not be talking about the same thing. --Deathawk (talk) 18:19, 7 October 2017 (UTC)
We don't turn our articles into quote farms. We should maintain at most a 50:50 ratio of quotation to original prose, and the examples I linked currently do not do that. Similarly, we should not base entire sections of our articles on primary sources like interviews, and we definitely are not allowed base entire articles based on primary sources -- the articles I alluded to are currently based entirely on primary sources. I don't frankly care what you are talking about, so if we are not talking about the same thing it doesn't really matter. I am just saying that if we are restructuring the MOS instructions about "Production" sections, we should summarize these core policies in a clearer fashion than we currently do. That said, it is entirely possible that we are talking about the same thing, in which case I would highly encourage you to familiarize yourself more thoroughly with our core policies, before suggesting changes to MOS subpages. Hijiri 88 (聖やや) 23:34, 7 October 2017 (UTC)
Butting in here, Primary sources refer to the author of the source, not the subject. This is the same with interviews. If the author/interviewer isn't affiliated with the subject then it is not considered a primary source.--TriiipleThreat (talk) 23:49, 7 October 2017 (UTC)
Nope. Please familiarize yourself with the policy. If the cited source is an original article incorporating critical analysis of the content of an interview, then maybe it could be argued as secondary, but if all the source is doing is printing original claims by cast and crew members, then it is a primary source. Hijiri 88 (聖やや) 23:55, 7 October 2017 (UTC)
That's not what the policy says. It specifically refers to accounts written by people close to the event.--TriiipleThreat (talk) 00:03, 8 October 2017 (UTC)
This will hopefully be my last comment here, as I've long since grown sick of arguing over this page, but this is so cut-and-dry that I might as well reply: the policy explicitly states Further examples of primary sources include [...] (depending on context) interviews. "depending on context" here clearly refers to the distinction I made in my previous comment, not that interviews conducted and transcribed by third-parties and printed in ostensibly independent fora are somehow "not primary sources". Hijiri 88 (聖やや) 06:51, 8 October 2017 (UTC)
@Hijiri88: This is getting wildly off track. The link you linked to is the "No original research Policy" where the guideline for Primary source exists to stop people from misinterpreting text. For instance, by writing about the bible and then citing the bible as their only source. This is not to state that quotes from primary sources should not be used. Elsewhere in , WP:PRIMARYNOTBAD the text says "sometimes, a primary source is even the best possible source, such as when you are supporting a direct quotation. In such cases, the original document is the best source because the original document will be free of any errors or misquotations introduced by subsequent sources." Given that the production section does not provide analyst or interpretations then this is acceptable. --Deathawk (talk) 03:37, 8 October 2017 (UTC)
You don't appear to have looked at the articles I referred to, as "the production section does not provide analyst or interpretations" is totally wrong. Maybe it shouldn't, but it does.
Also, NOR clearly says we are not allowed base entire articles on primary sources (which is currently the case with every single article on an episode of GOT season 7) and should probably not base large sections of articles on primary sources. And when all the significant content of the production section is in the form of quotes from said primary sources, that's a problem too.
Hijiri 88 (聖やや) 22:27, 9 October 2017 (UTC)
There is some disagreement that information from someone involved in production commenting on production is necessarily a primary source. We should consider these as strong sources to be use, particularly if it is in some type of retrospective (such as explaining why they used a certain actor, or why a certain scene was cut). This is a type of transformation of information about the production process, and regardless if you want to consider it primary or secondary, is strong information we want to include. (waaaay back when trying to establish the notability for fictional elements, I came up with the term WP:1.5 sources to describe these high-value but seen as primary sources that aren't quite secondary but their presence from reliably-published sources would make them have the weight of secondary sources and so people wouldn't write these off as primary ones) Rarely are you going to get a "true" secondary source on the production of a film unless that film is of exceptional weight (such as most of Kubrick's body of work). Mind you, I stress the need for this production information to come by the way of an RS (such as an article in Variety or Hollywood Reporter), the fact that the work covered it showed the production aspect was important; we need to reduce the reliance on Twitter and social media to bolster a production section artificially (as those are SPS).
The quotefarm issue is obviously still an issue which way you dissect the sources; no issue there. --MASEM (t) 22:52, 9 October 2017 (UTC)
Agree that a reduction in primary-sourced quote farming would be good, but no particular wording is coming to my mind, so someone else can add it.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  22:12, 8 October 2017 (UTC)
  • Comment This is turning into a list of rules, which I think we should avoid. The MOS should be about providing advice on how to write better articles, not rules and regulations. For instance, no actors beside the lead? Most films wouldn't work without supporting roles. They are even awards for best supporting roles. Also depending on the article structure grouping them together may not make the best sense. We should leave some editorial discretion for some of these choices. Furthermore the announcement date may give a general idea of when a production decision was made. For example, instead of writing "In October 2017 it was reported that so and so was cast" you could write "By October 2017, so and so was cast."--TriiipleThreat (talk) 10:36, 7 October 2017 (UTC)
    I agree with TriiipleThreat. We should not be heavy-handed in dictating to do this and that. We should be precise about the truly recurring problems in such sections, regardless of a film's context. I disagree with what has been called trivia, such as naming production companies or equipment brands. Certainly there are films where the companies or brands are not relevant to mention in prose, but it may be relevant for other films. For example, if a distributor acquires rights to a film that screened at a festival, it would be relevant to know how that film got backed in the first place. And some films will have certain equipment used for the first time or in a novel way. If such elements are being discussed in running prose in more than a passing way, especially by general-audience sources, then we should be open to the material's encyclopedic qualification. Erik (talk | contrib) (ping me) 19:23, 7 October 2017 (UTC)
I agree with TriiipleThreat and Erik. The guideline having leeway is why we state at the top in bold: "There is no defined order of the sections." Flyer22 Reborn (talk) 19:29, 7 October 2017 (UTC)
I also agree with TriiipleThreat and Erik. Though I wouldn't be opposed to adding in a link to WP:PROSELINE somewhere. Otherwise, we shouldn't be precisely dictating what can and cannot appear in these sections. - Favre1fan93 (talk) 01:08, 8 October 2017 (UTC)
@TriiipleThreat, Erik, and Favre1fan93: I think the goal of the bullet points was to list what we find problematic with production sections now, rather than to simply list soloutions. That said I've been thinking about a better way we could incorporate some of the "Don'ts" to express that it is ok for some things, and I think I've come up with an idea.
The production section should concentrate on the unique aspects of the production over tasks inherent to all films. Events which are of interest to only a very select few shoud be avoided unless these significantly impacted the work. For instance, while 90% of the time it's not worth mentioning anything about finance companies involved, if they forced significant revisions in the script, than it's worth noting. In cases such as these the articles should provide ample context as to why it's worth being covered, at the same time care should be taken to make sure that we aren't merely cover day to day business news.
The above paragraph A) Explains what not to do B) Explains what to do and C) provides exceptions, where it would be appropriate to ignore this rule. It could use work I'm not really in love with the word "Exception" as I feel these elements come up to much for that to be the right word, but I'm hoping that we can use this to start a discussion --Deathawk (talk) 02:19, 8 October 2017 (UTC)
This still seems like the same exact thing, just not in a list form. - Favre1fan93 (talk) 03:34, 8 October 2017 (UTC)
There are some major differences I would say The biggest would be that it does not explicitly state what to excludes but provides context of how one may be able to decide if something should be. True, it does tell people that " Events which are of interest to only a very select few should be avoided" although what that covers is open to interpretation. --Deathawk (talk) 06:19, 8 October 2017 (UTC)
Seems good to me, but should be generalized to be part of what's eventually in MOS:FICT, e.g. with "tasks inherent to most productions of a particular kind". And guideline language wouldn't say "90% of the time", since no one's done a measurement; we'd use "most of the time", "usually", etc. Change "it's" to "it is". Generalize "revisions in the script" to "changes". "In cases such as these" can be compressed to "In such cases," (and a comma is needed there).  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  22:18, 8 October 2017 (UTC)
Good suggestions and I'll incorporate them, I would like to avoid generalizing too much. I understand that you eventually want these guidelines to apply to MOS:Fiction in general, but I think for specific media, specific wording is needed. --Deathawk (talk) 06:02, 9 October 2017 (UTC)
All in due course. It's important that this be handled consistently across media. The draft below is pretty much exactly what I mean by doing a WP:SUMMARY style compression on a per-medium basis.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  09:38, 11 October 2017 (UTC)
  • I added a note to my original draft that people can just directly copy-edit it. I'm not trying to control the text, just wanted to get a draft started for development.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  22:06, 8 October 2017 (UTC)
I agree with TriiipleThreat, Erik, Favre1fan93, and Flyer22. However, I think Deathawk is on the right track in developing a more prose style version that avoids the bullet list of do's and don't's. This is a far better direction to head in. Huggums537 (talk) 23:14, 9 October 2017 (UTC)
Yep. Each fiction-related MoS can have a précis like this, with MOS:FICT hosting the the generalized version.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  09:38, 11 October 2017 (UTC)

Draft

Although I'm not sure we're all exactyl on the same page yet, I couldn't help but figure I'd try my hand at writing a draft, incorproating the points we've discussed.

Production sections should not be a indiscriminate list of news and trivia associated with the making of the film, but rather should instead be focused on how the film got from point A to the screen. As such the section should focus on events that impacted it as opposed to merely day to day business operation.

Although announcements to the public can be notable, production sections should primarly focus on how things unfolded from a production stndpoin. Similarly revaluations about when plot elements or settings were revealed should, most of the time, not be included, instead focus on how these elements came about.

Readability should be of a high importance. Sentences should avoid repetitive phrasing and information should be presented in a clear manner. Exact dates, are usually not needed unless there is a significance to it. '

I'd still like to say something about problematic casting sections, maybe in the first paragraph, but that's something I still feel we need to come to a consensus on. Any feedback would be appreciated. --Deathawk (talk) 04:40, 11 October 2017 (UTC)

It's been about five days and I would like to revive this discussion so I'm going to ping @TriiipleThreat, Erik, Favre1fan93, and Flyer22 Reborn:. Do you guys think this solves the problem you had with the earlier proposal? --Deathawk (talk) 03:53, 16 October 2017 (UTC)
Seems reasonable to me.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  19:10, 16 October 2017 (UTC)
I'll wait for others to weigh in. All of the recent suggestions on this page are sort of overwhelming. Flyer22 Reborn (talk) 19:48, 16 October 2017 (UTC)

Original research about box office and reception

I've seen many editors adding their own personal interpretation of box office reception and what the review aggregators have said – for example, "Despite being a box office success, film was universally panned for its derivative plot and poor acting." I've been trying to clean this up lately, but it's sometimes difficult to get my changes to stick. Editors seem to think that if they can find two or three reviews that criticize the acting, it means there's a critical consensus that the film was panned for the acting. For example, this edit by Drown Soda (talk · contribs) adds a Rotten Tomatoes-style critical consensus that says, "The film received mixed reviews from critics, with many criticizing its narrative inconsistencies and overt violence, while some, such as Roger Ebert, praised the film's aesthetic influences from German impressionism". There's no citation in the article that many critics criticized its narrative inconsistencies – this seems to be synthesis. Similarly, Fourlaxers (talk · contribs) wants to add in this edit that a film was "universally panned" – presumably because it has a low rating on the aggregators. How do we know it was "universally" panned, however? Rotten Tomatoes does have several positive reviews listed, including from its so-called "top critics". How can a film be "universally panned" when there are positive reviews? This edit speculates on both the box office performance ("major box office success") and the reception ("heavily panned for blah blah blah"). And there was yet another brief edit war over such synthesis at The Emperor's New Groove when TheDisneyGamer (talk · contribs) wanted to add that a film "performed disappointingly" but got "positive reviews". Since I'm facing some pushback on this issue, I think maybe we should update the MOS to discuss this. Perhaps we should explicitly say to just use what the review aggregators and box office performance websites say, not to add our own analysis of such. NinjaRobotPirate (talk) 10:19, 16 October 2017 (UTC)

This ongoing WP:SYNTHESIS and WP:EDITORIALIZING is endemic in film reception sections. Personally I think every single claim should be explicitly attributed to somebody. If Metacritic proclaims a film as "universally acclaimed" then our reception section needs to actually explicitly state that as Metacritic's own conclusions (based on the reviews they actually surveyed), rather than "The Dark Knight was acclaimed by [presumably all] critics" (to use an example I have corrected on multiple occasions). The key point here is that editors must summarize the film's reception, not interpret aggregator statistics and financial data for themselves. Betty Logan (talk) 18:11, 16 October 2017 (UTC)
Yep. And the examples NinjaRobotPirate provides aren't directly comparable. The "universal" claim is impermissible (it would require that no reviewer anywhere disagreed), while the "The film received mixed reviews from critics, with many criticizing its narrative inconsistencies and overt violence, while some, such as Roger Ebert, praised the film's aesthetic influences from German impressionism" is fixable by replacing the "many" weasel-wording with specific attributed assessments.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  19:06, 16 October 2017 (UTC)
NinjaRobotPirate and Betty Logan, as you know, we've had discussions about whether or not to include, and how to include, summary statements time and time again. For example, Wikipedia talk:WikiProject Film/Archive 52#Summary statement for "Reception" section and Wikipedia talk:WikiProject Film/Archive 55#Critical response summary - was consensus ever achieved?. Some of us prefer summary statements; some of us don't. But we are always against WP:Synthesis. I prefer summary statements, but, as seen at Talk:Star Wars: The Force Awakens/Archive 5#Reception, I've noted that I dislike "According to" wording for summary statements because it's a misleading form of WP:In-text attribution; it suggests that it's only according to that source that the film did well, badly, or was mediocre. We should not be stating "According to [so and so], Schindler's List received acclaim from both film critics and audiences." We should simply state the following: "Schindler's List received acclaim from both film critics and audiences." That is what we currently do at that article. Per WP:Due weight, if most sources state that a film did well, badly, or received mixed reviews, we should not be attributing the matter to one source as if it's simply an opinion and is not the fact that it is. That stated, some type of in-text attribution may be needed in some cases GoneIn60 seemed to be arguing that in the aforementioned Star Wars: The Force Awakens discussion when pointing to different films.
As for NinjaRobotPirate's proposal, we already have the #"secondary sources" for the film's reception are NOT the same thing as what many editors are likely to read "secondary sources" as and #Proposed clarification of reviews' relation to WP:PSTS and MOS:TONE sections above, and nothing has been resolved in those sections. We all agree not to engage in WP:Synthesis, but we have editors disagreeing on what a secondary source or review is and how it is to be used when summarizing a film's critical reception. Flyer22 Reborn (talk) 20:12, 16 October 2017 (UTC)
There are always going to be examples of improper synthesis, editorializing, and lack of attribution. A MOS guideline isn't likely going to put much of dent into the problem, and attempting to limit what is acceptable could be counter-productive in some situations. As Flyer22 points out, there have been discussions in the past at various article talk pages where a solution was reached. The potential proposal above to limit summary statements to claims made by "review aggregators and box office performance websites" could bring some of those past solutions into question, particularly when there was overwhelming consensus to have a summary statement (usually in situations when review aggregators are in clear agreement with one another). Some of those discussions simply focused on tweaking the phrasing used (i.e. removing peacock adjectives/adverbs). I think its best to leave a preference like this within the article's jurisdiction, but I'd be willing to consider a specific proposal on what would be added or changed in the MOS. We already have policies and guidelines in place to address improper statements and claims as we come across them. --GoneIn60 (talk) 06:43, 17 October 2017 (UTC)
Agreed, especially since (for older films) more reliable secondary sources exist, like film and arts journals. And box office performance site don't tell us anything about critical reception at all, only "uncritical reception". ;-)  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  09:39, 17 October 2017 (UTC)

WP:SYNTHESIS but have you read Wikipedia:What SYNTH is not, specifically §SYNTH is not just any synthesis? An encyclopedia is almost certainly going to be synthesis, the question is whether this synthesis is OR or not. Emir of Wikipedia (talk) 10:02, 17 October 2017 (UTC)

See earlier discussion. Recap: The last thing we should do is refer to what we do in summarizing sources as "synthesis" when that word has taken on a very special and very negative meaning here. The essay already has this covered for anyone who wants to philosophize on that, but it's important that we do not confuse new (or old but insufficiently clued) editors that OR's meaning of synthesis has anything to do with our normal editing practices.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  10:54, 17 October 2017 (UTC)
I'm not sure what you're defending, Emir. In a nutshell, an editor cannot combine multiple individual reviews and claim a general trend (or multiple general trends). We leave that to sources that look at these reviews and make such high-level assessments themselves. Erik (talk | contrib) (ping me) 14:06, 17 October 2017 (UTC)

Rotten Tomatoes' consensus

RT usually (when there are more than a few reviews) includes a consensus statement, often along the lines of "Engaging action sequences and a likable cast aren't enough to save Actionthriller Man from a weak script and predictable, cliched plot."

From what I have seen, we seem to quote this fairly often. That said, Binksternet has challenged the use at Movie 43 (which sounds horrible enough to haunt a few A-listers for years to come). The consensus statement is "Unsigned...not a reliable source."[1] While I don't always agree with anyone, Binksternet doesn't seem to be one to float trial balloons to challenge a consensus, so I checked here. I don't see anything for or against the use of this material. Yes, it is "unsigned". No, I don't think Binksternet is part of the cabal established to protect the stars of Movie 43.

This page does offer a link to Wikipedia:Review aggregators, an essay, which says, "...Rotten Tomatoes's reported 'consensus' and Metacritic's 'metascore' description are prose that may help readers understand a film's reception." While that wording is pretty definitive, an essay shouting "Respect my authoriti!" doesn't sell it for me.

Do we have anything more substantial on this? - SummerPhDv2.0 16:05, 4 September 2017 (UTC)

Oh, I've floated plenty of trial balloons in my time.
I respect the essay Wikipedia:Review aggregators which tells us that the consensus text is acceptable. Erik wrote that in 2010, and I'm sure it's been the practice ever since. But who writes the consensus text at Rotten Tomatoes? At Wikipedia:Identifying reliable sources we are told that "Content from a collaboratively created website may be acceptable if the content was authored by, and is credited to, credentialed members of the site's editorial staff." The consensus statement at Rotten Tomatoes is not credited, which is the main problem here. Binksternet (talk) 16:27, 4 September 2017 (UTC)
That section, WP:UGC, seems to be talking about user generated content, "Sites with user-generated content include personal websites, personal blogs, group blogs, internet forums, the Internet Movie Database (IMDb), the Comic Book Database (CBDB.com), content farms, most wikis including Wikipedia, and other collaboratively created websites" with your quote discussing an exception to that rule. Is RT's consensus user generated? - SummerPhDv2.0 18:58, 4 September 2017 (UTC)
I'm pretty sure it's written by the site's editorial team, but this isn't explicitly stated anywhere I can find. NinjaRobotPirate (talk) 21:15, 4 September 2017 (UTC)
SummerPhDv2.0, Binksternet, NinjaRobotPirate: I'm positive that the critics' consensus is not user-generated. The Rotten Tomatoes staff adds a summary, especially after the number of reviews hit a certain threshold. I think if there is any individual who pens the consensus, it would be Tim Ryan. He wrote "Critics Consensus" articles on Rotten Tomatoes that were around years ago (though not anymore). Might be that it was simplified to penning a few dozen words for each film. Maybe someone can tweet to him about who does it? I don't see any reason to exclude the consensus on this basis; it is akin to a periodical giving their own nutshell on what critics thought. And sometimes there may not be such a nutshell available elsewhere, and I hate to have summary recaps of the critical reception be so numerical. A reason not to include it would be if the wording was essentially useless. For example, I did not include it for Gods of Egypt because it was a silly bit: "Look on Gods of Egypt, ye filmgoers, and despair! Nothing beside remains. Round the decay of this colossal wreck, boundless and bare. The lone and level sands stretch far away. (Apologies to Shelley.)" It does not say anything about why the film was bad. I take a similar approach with reviews if a critic is just funning around with slamming a film and not getting into why it was bad for them. Erik (talk | contrib) (ping me) 13:50, 5 September 2017 (UTC)
I agree that the RT consensus sometimes doesn't add anything useful. In general, though, I think it's OK. For classics, infamous flops, and cult films, you can often find better sources that analyze the film's reception in depth. NinjaRobotPirate (talk) 20:34, 5 September 2017 (UTC)

SummerPhDv2.0, Binksternet, NinjaRobotPirate: This is too funny, the answer just emerged here: "Jeff Giles, a 12-year Rotten Tomatoes veteran and the author of books like Llanview in the Afternoon: An Oral History of 'One Life to Live', writes what the site calls Critics Consensus, a one-sentence summary of the response to each film." Erik (talk | contrib) (ping me) 15:25, 7 September 2017 (UTC)

Excellent. It seems Erik's shadow-career at NYT is finally paying some dividends for us.
At the risk of totally stepping in it, I propose adding a brief statement to this style guide clarifying the authorship and general reliability of the "consensus" statements, with clear wiggle room to allow us to leave it out when the statement is not ... um ... clear (to those who need to brushed up on their Shelley) or is better replaced by a clear summary from another source. Thoughts? - SummerPhDv2.0 16:17, 7 September 2017 (UTC)
Timing! My complaint is answered: Jeff Giles is the author. Binksternet (talk) 16:20, 7 September 2017 (UTC)
It caused me to do a double-take. I read that as "Jeff Giles, a 12-year-old Rotten Tomatoes veteran and the author...". Betty Logan (talk) 17:01, 7 September 2017 (UTC)
Yeah, I was wondering why a preteen was writing the critical consensus statements. That NYT article was interesting because it mentions RT does fact checking and verifies whether reviews are positive/negative with journalists. I suppose we could add a line in the MOS about this. I might also include advise against adding one's one synthesis-tinged consensus based on the RT score. This is popular among some editors, who seem to believe that their insights are just as legitimate as the RT consensus. NinjaRobotPirate (talk) 17:57, 7 September 2017 (UTC)
The biggest problem is editors extrapolating the RT scores to a wider critical consensus. RT only speaks for the reviews it has surveyed and there is no attempt to ensure they are representative. For a good example of how much difference this makes compare the scores for Ballerina and Leap, and take note that they are the same film, but just with a different voice cast... Betty Logan (talk) 18:26, 7 September 2017 (UTC)
Adding the Rotten Tomatoes' consensus and Metacritic score to the Critical reception section has become standard practice. I'm fine with it, except for the case of films that don't really have a consensus at these sites because they just aren't well-known, and in the case of some really old films, which came out before these review aggregation sites were available. Using the sites at the beginning of the Critical reception section in these cases can be misleading. It's because of this that, in the case of really old films (or just some that came out before the sites existed), it's better to put the Rotten Tomatoes' consensus and Metacritic score later on in the Critical reception section. Our guideline also currently states, "For older films, it is important to distinguish between contemporary critical reception (from reviews published around the time of initial release) and subsequent reception (from reviews made at later dates). Use secondary sources to determine if a film's initial critical reception varies from the reputation it has today." We can try to not include the Rotten Tomatoes' consensus and Metacritic score for some articles, but editors like consistency, and they will want to add them anyway because they will not understand why whatever article in question should deviate from the rest. Flyer22 Reborn (talk) 01:07, 8 September 2017 (UTC)

RT for older films

Flyer's comment above regarding inclusion of RT scores in older films recalls a recent discussion at Talk:The Shining (film). Based on that discussion I walked away with an understanding that the implicit consensus was not to include such scores for films that were released substantially before RT existed. Of course, that begs several questions, such as:

  1. For older films, should the RT score be included at all if no other sites have commented upon it?
  2. For older films, can the RT score be included as long as it's made explicitly clear that the score reflects a current view?
  3. When should the cut-off be for including RT scores?

In my ideal world the MoS would be updated to address this concern, perhaps not just for RT but also for other sites, but I admit I don't have high hopes without this at least going through an RfC. DonIago (talk) 16:26, 8 September 2017 (UTC)

Although only an essay, WP:AGG states:

Aggregator scores are most effective and accurate for films released in the 2000s and beyond. This is because more reviews are available online and as a result contemporary critical reception is more clearly defined. Prior to the 2000s, Rotten Tomatoes and Metacritic did not exist, and reviews were typically not online. Sources besides Rotten Tomatoes and Metacritic should be sought out for films released before the 2000s; reports of critical consensus will likely exist in print sources. E.g., Alien, released in 1979, has a score of 97% on Rotten Tomatoes, but the critical reception at the time of release was mixed.

I am of the general view that aggregators should not be used for pre-2000 films, for the main reason that they mix retrospective and contemporary reviews and therefore do not give a clear representation of the contemporary reception or how opinion has evolved over time. This issue recently came up at 2001: A Space Odyssey (film), which incidentally adequately sums up the contemporary reception and the film's modern day reverance so I don't see how the article is improved by adding the RT score. I would say that for older films—especially classics that have been the subject of substantial commentary—we should be looking beyond aggregators for summaries of their critical standing. There is a valid point that in cases where opinion has not changed much then the aggregator score does not misrepresent the reception, but in cases where there has been a noticeable shift over time they are misleading and should be excluded. Granted, this is difficult because many editors see them as a regular feature and often you get "drive-bys" installing them on every article, but that doesn't mean we have to accept them or we can't discourage them. I am not going to launch a huge campaign to remove them but if they pop up on articles on my watchlist and I disagree with their inclusion I reserve the right to remove them. Betty Logan (talk) 16:47, 8 September 2017 (UTC)
I think Betty has summed up my opinion fairly well. THe other flip side to this is that those older films with current RT scores frequently don't have a lot of RT reviews listed. So, then you get into statistical significance of the data. I think if you want to discuss "modern" views, then just look to RT as a source for providing individual reviews that you can summarize. I would stay away from using the aggregate as a "summary whole" and look more to third party sources that discuss it.  BIGNOLE  (Contact me) 19:38, 8 September 2017 (UTC)
Rotten Tomatoes has flaws sure but I don't think the age of a film makes the Rotten Tomatoes score any less helpful or unhelpful depending on your opinion of Rotten Tomatoes. It still serves the same blunt overview. There is a tendency towards older films to get better scores because more of the positive reviews remain available for Rotten Tomatoes to catalog. For films before ~1990 I will try to point out that "Rotten Tomatoes retrospectively collected reviews from N critics ..." to give readers some context, and as you say in point 2. more context would be better. I wouldn't be in favor of removing them but for older films I would de-emphasize the score by putting it at the end of the Critical response section instead of at the start. If you are looking for a cutoff point why allow any Rotten Tomatoes scores from before 1998 when it was started? Any reasons for not including old Rotten Tomatoes scores come down to the same complaints everyone has about the flaws of Rotten Tomatoes and the general need to give those scores some context. -- 109.76.249.25 (talk) 19:43, 8 September 2017 (UTC)
Except for the "aggregators should not be used for pre-2000 films," I agree with what Betty stated in her "I am of the general view" paragraph. I would obviously support explicit caution against using Rotten Tomatoes for very old films. I just don't think we should state that they shouldn't be used for pre-2000 films. Really, the WP:AGG essay should changed in that regard. If the older film's reception has generally remained consistent and the Rotten Tomatoes view reflects that, I don't see an issue with including Rotten Tomatoes, but I wouldn't place it at the beginning for very old films. And if the film's reception has changed, and the Rotten Tomatoes score/consensus is included in the part of the paragraph making the change clear, I think it's fine in that case as well. Flyer22 Reborn (talk) 20:12, 8 September 2017 (UTC)
The problem though is that it is difficult to know for sure if a film's reputation has remained consistent unless you obtain independent evidence. With post-2000 films most reviews will be contemporary and if there is a shift over time the RT score can be double checked either through the Wayback machine or through the article's own page history. Pre-2000 that becomes impossible. Betty Logan (talk) 22:13, 8 September 2017 (UTC)
I really can't support the guideline stating that we should not use Rotten Tomatoes for pre-2000 films. I support cautioning against use of Rotten Tomatoes for very old films, and approaching this matter on the case-by-case basis I noted above. As this is a guideline, we should be guiding more than restricting anyway. Flyer22 Reborn (talk) 22:21, 8 September 2017 (UTC)
There's nothing wrong with using RT for older films PROVIDING that it's made clear that it is a contemporary take on the film AND it definitely should not be at the head of the section, but at the bottom, following the reviews of the time. You can also add the proviso that to be used there must be an adequate number of reviews to be aggregated. Darkwarriorblake / SEXY ACTION TALK PAGE! 22:30, 8 September 2017 (UTC)
It's not a contemporary take though. If you take the entry for 2001: A Space Odyssey it has reviews dating from 1968 up to modern day. It mixes contemporary reviews with retrospective reviews, so doesn't accurately relay either standing. The only films where you can be sure that Rotten TOmatoes is accurately summarizing the contemporary receeption is post-2000, or thereabouts, when Rotten TOmatoes came into existence. Betty Logan (talk) 22:43, 8 September 2017 (UTC)
Rotten Tomatoes seems to create a new page for films on their re-release, judging by Terminator 2 3D and Phantasm Remastered. However, if you check the Phantasm reviews, you'll see there's one listed by Vincent Canby, who died in 2000. I checked a big budget film from 15 years ago, LOTR 3, and that seems to have accumulated quite a few retrospective reviews. I don't think there's any way to be sure that the RT score doesn't mix old and new reviews. NinjaRobotPirate (talk) 01:52, 9 September 2017 (UTC)
You have to be careful to not be misled by the dates on the reviews: for example, the Sean Nelson review for Return of the King was aggregated last month but if you click through to the review you can see it is indeed from 2003. However you can also verify through Wayback that the RT rating was 94% so has hardly changed in the last five years; indeed, going back through the page history you can see it was 94% a decade ago. The salient point here is that we have a score trajectory for anything that was released while RT has been in existence. If the reception changes noticeably then Wayback can be used to source the contemporary reception. This is not possible for older films. Betty Logan (talk) 02:25, 9 September 2017 (UTC)

Betty, looking at the exchanges involving you, Kevskerr, WikiPedant and MarnetteD at the 2001: A Space Odyssey (film) article, as seen here, here, here and here, I don't see an issue with Rotten Tomatoes being included so late in the section. And I had seen it there before; it was there for sometime. It echoes what is stated by other sources; so I understand the argument that it's redundant. But I don't think that this particular case is a case we should be concerned about. I do question WikiPedant adding it as a reference for the following line, though: "2001: A Space Odyssey is now considered one of the major artistic works of the 20th century, with many critics and filmmakers considering it Kubrick's masterpiece." This line appears to be a topic sentence, and therefore doesn't need sourcing. If it does need sourcing, it should be attributed to a source that explicitly supports the line. Flyer22 Reborn (talk) 20:37, 11 September 2017 (UTC)

Proposed text

Given that it's been over a week since anyone's commented on this, I'd like to suggest the Critical response section be modified from

Current text

Review aggregation websites such as Rotten Tomatoes or Metacritic are citable for statistics pertaining to the ratio of positive to negative reviews; caution should be exercised when using aggregator scores that combine original reviews with reviews from later dates.

to

New text

Review aggregation websites such as Rotten Tomatoes or Metacritic are citable for statistics pertaining to the ratio of positive to negative reviews; caution should be exercised when using aggregator scores that combine original reviews with reviews from later dates. When including statistics from these sites for films that were released prior to the year the site was created, the statistics should be de-emphasized by placing them at the bottom of the section, as their statistics are potentially less accurate prior to that year. It should also be made clear when the statistics are being captured ("As of May 2015, Rotten Tomatoes reported that 50% of critics gave the film positive write-ups, based on a sample of 68, with an average score of 5.2/10."). See WP:AGG for more information about concerns regarding aggregators' accuracy for earlier films.

New text in bold. Apologies if my quotebox formatting sucks; I haven't used them very often. DonIago (talk) 14:37, 20 September 2017 (UTC)

Oh, wow!! Huggums537 (talk) 20:33, 21 September 2017 (UTC)
I hope that's a "This sounds great!!!" oh wow and not a "OMG what do you think you're doing???!!!???" oh wow. :p DonIago (talk) 20:59, 21 September 2017 (UTC)
The "Wow" was not related to your proposal. Sorry for the confusion. It looks pretty good except one sentence is a bit complicated. I might change:
  • "When including statistics from these sites for films that were released prior to the year the site was created, the statistics should be de-emphasized by placing them at the bottom of the section, as their statistics are potentially less accurate prior to that year."

To something more simple like:

  • "The statistics from these websites are potentially less accurate for films that were released before the websites actually existed, therefore they should be placed at the bottom of the section in those cases."

Although, I'm not entirely sure adding this in is a good idea or not. However, if it is added, I'd simplify it in some way like I mentioned. Huggums537 (talk) 23:42, 21 September 2017 (UTC)

Good catch. I think the addition captures the consensus of the above discussion, but I'm certainly open to hearing from other editors. DonIago (talk) 03:40, 22 September 2017 (UTC)
Betty, I'm actually glad you said that because one of my first instincts was to suggest eliminating everything in the proposal except the last sentence, with the idea that linking to WP:AGG is probably sufficient enough. I just wanted to avoid offending Doniago... Huggums537 (talk) 04:28, 22 September 2017 (UTC) But, linking to Wikipedia:Review_aggregators#Limitations would be even better maybe? I say this because I feel that linking to the relevant content would be far easier, and less complicated (in appearance) than duplicating it at MOS. I also like Doniago's idea to make sure that the captured date of the statistics is included too, so I guess I'm really just talking about removing that first sentence, which is the one that does the "inappropriate encouraging" Betty was talking about...
I'm curious though as a newcomer, how common is it to take this step to make the transition of material from essays into the guidelines and how commonplace is it for guidelines to link to essays? Is it more common for policy to usually link to policy, or can guidelines, policies, and essays freely link between each other with fairly common frequency? Huggums537 (talk) 05:24, 22 September 2017 (UTC)
Eh, the text doesn't always need to be placed at the bottom. In some cases, it should be, though. Flyer22 Reborn (talk) 05:29, 22 September 2017 (UTC)
It sounds like there's substantive reservations about my proposed text. Fair enough! Flyer, would you be willing to create an alternate version? I'm asking you as I'm not sure how you'd define "some cases". Thanks! DonIago (talk) 13:06, 22 September 2017 (UTC)

This is a good start and I commend your for trying to make this improvement and bring greater consistency to this. I'm glad the wording is cautious, there are already editors who will happily delete Rotten Tomatoes from older film articles, and it wouldn't take much encouragement to start a cull. I do still have some reservations about the specifics of proposed text though. For starters even using the word statistics feels like it is giving undue weight to the scores created by review aggregators, and for similar reasons I'm wary of using the word "report" since they look at review and subjective interpret that review in various ways (quantize) to create the scores, unlike a most sporting sporting event where the scores aren't an interpretation and actually are objectively reporting.
While I welcome a more consistent wording I think it is rarely necessary to specify "As of" (better than sloppy editors who write "currently" but it shouldn't be needed). "As of" is only of limited use. If you find that article text no longer matches what the source says (as I recently found with Popeye which seems to have drifted from negative to mixed) what are you to do? The article text should match the source, shoudn't it? Unless editors also include an archived copy of the page the date context doens't help much. It seems like a sensible idea but I fear editors will follow that idea to absurdity and recommendation and write "As of" absolutely every time.
I don't think pointing pointing editors to WP:AGG/WP:RTMC is good enough unfortunately. The page starts with a big disclaimer that it is an essay not a rule which doesn't inspire confidence and again makes it look like we are falling back on an essay and that a proper consensus still hasn't formed.
It is a good start though, please keep trying. -- 109.79.119.230 (talk) 22:06, 22 September 2017 (UTC)

I appreciate your expressing your concerns, but it seems like you're expressing what you don't like about the text but not suggesting alternatives in the process. I'm happy to come up with a revised text block (though I'd also like to hear back from @Flyer22 Reborn: as noted above), but I don't want to get into a ping-pong match of my coming up with text only to have the editors who expressed concerns say that I didn't address them either. I hope that's understandable.
Regarding the "as-of" issue, I feel like if there's nothing to indicate that than "currently", which you also opposed, is strongly implied. The dating can also give editors who may not bother to check the citation a reason to refresh the data in any case.
As for whether to or not to point to the essay...I don't think there's anything wrong with doing so as long as we make it clear that we're doing it in a "for more information on this subject" context.
Still...is this better (changes from existing text in bold)?
Proposed text #2

Review aggregation websites such as Rotten Tomatoes or Metacritic are citable for data pertaining to the ratio of positive to negative reviews; caution should be exercised when using aggregator scores that combine original reviews with reviews from later dates. The data from these websites is potentially less accurate for films released before the websites existed, therefore it should be placed at the bottom of the section in those cases. It should also be made clear when the information is being captured ("As of May 2015, Rotten Tomatoes reported that 50% of critics gave the film positive write-ups, based on a sample of 68, with an average score of 5.2/10."). See this summary of limitations regarding aggregators' accuracy for earlier films for more information on this subject.

DonIago (talk) 12:45, 25 September 2017 (UTC)

Please don't encourage people to use ", with an average score". It's grammatically incorrect, and it's been an uphill battle to convince people to stop doing this. One doesn't say, "I bought one apple yesterday, with another apple being bought today". Instead, you would say, "I bought one apple yesterday and another today." With is a preposition, not a conjunction. NinjaRobotPirate (talk) 17:30, 25 September 2017 (UTC)
Fixed? DonIago (talk) 19:46, 25 September 2017 (UTC)
Doniago, you're making good efforts here. However, I noticed your new proposal doesn't address Betty or Flyer22's concerns. I like the last sentence very much that points exactly to the specific place in the essay it needs to. Great job on that. Also, as you know, I'm in favor of adding the date at the time of inclusion. However, I think we should reword it in such a way that it appears to be more of a suggestion, and less of a mandate. In other words, maybe it could be strongly encouraged, but not an absolute requirement. As far as the concerns of the IP user about sourcing goes for including the dates, well, if you are updating the statistics then you should be using updated sources to do so anyway. So, what about the sources? Update them of course. It should be done anyway. My biggest concern though, is still that first sentence, which is also the one that is debatable with Betty and Flyer22. Betty's concern was that this sentence encourages editors to always include the data, but it would be inappropriate to do so in some cases. Flyer22's concern was that the data doesn't always need to be placed at the bottom. It would have to be carefully reworded to address these issues. I do like how you simplified that first sentence though. I also like your usage of the word, "data". Good job on that. Perhaps something like this:
Example

Review aggregation websites such as Rotten Tomatoes or Metacritic are citable for data pertaining to the ratio of positive to negative reviews; caution should be exercised when using aggregator scores that combine original reviews with reviews from later dates. The data from these websites is potentially less accurate for films released before the websites existed, therefore caution should also be used in determining if they are citable in those cases. In some cases, the data should usually be placed at the bottom of the section with a notification of the possible discrepancies. See this summary of limitations regarding aggregators' accuracy for earlier films for more information on this subject. It's also recommended to include the date the data was captured: ("As of May 2015, Rotten Tomatoes reported that 50% of critics gave the film positive write-ups, based on a sample of 68, and an average score of 5.2/10.").

Huggums537 (talk) 21:33, 25 September 2017 (UTC)

I don't have a suggestion at the moment. Flyer22 Reborn (talk) 22:09, 25 September 2017 (UTC)

I really dislike "In some cases", because it begs the question of which cases. I also feel that sentence is awkwardly constructed, and "a notification" is, to me, unclear. Additionally, I think that if we're going to include a "see also", as we're doing, then it should be at the end of the paragraph. Lastly, this edit re-adds the "average score" wording that NRP explicitly took issue with. Sorry for the stream of criticism there. DonIago (talk) 02:39, 26 September 2017 (UTC)
Regarding the RT statement...An alternative way to phrase that (with less commas I might add) would be:
As of May 2015, 50% of the 68 reviews compiled by Rotten Tomatoes are positive and have an average score of 5.2 out of 10.
I am a fan of less commas, the use of present tense, and avoiding fractions. I am not sure how I feel about the use of "write-ups"; I think "reviews" is adequate and allows us to remove the self-explanatory "critics". Also, "based on" can be misleading if we want to mention the average score last. The percentage of positive reviews is not based on the average score, but the previous examples seem to imply that it is. --GoneIn60 (talk) 05:20, 26 September 2017 (UTC)
Doniago, No worries! Critical peer review is motion forward. I knew some parts might be problematic, and would likely need to be ironed out anyway. Besides, I agree with you on most points, but not all. Agree with dropping "In some cases". Agree with sentence restructure, and clarity on notification. Agree with putting "See also" back at the end. Lastly, disagree about the edit regarding "average score", because it was my understanding that NRP took an issue with the grammar, not the score itself. So, simply changing "with" to "and", or using GoneIn60's suggestion should solve the issue unless I'm missing something. If you look carefully at how I reintroduced "average scores", you'll see the single word change I made earlier. I didn't just put it right back in without any thought, although the change was so unnoticeable I can understand how you might have missed it. I think maybe everyone might like this though:
Revision

Review aggregation websites such as Rotten Tomatoes or Metacritic are citable for data pertaining to the ratio of positive to negative reviews; caution should be exercised when using aggregator scores that combine original reviews with reviews from later dates. The data from these websites is potentially less accurate for films released before the websites existed, therefore caution should also be used in determining if they are citable in those cases. In such cases, it is suggested to place the data at the bottom of the section with a disclosure explaining why there may be inaccuracies. It's also recommended to include the date the data was captured: ("As of May 2015, 50% of the 68 reviews compiled by Rotten Tomatoes are positive, and have an average score of 5.2 out of 10."). See this summary of limitations regarding aggregators' accuracy on earlier films for more information on this subject.

Huggums537 (talk) 06:11, 26 September 2017 (UTC)

P.S. If we really do come to a consensus on this, does anybody care if I get to be the one to do the edit that implements it??? I think it's really exciting the thought that somebody like me with only slightly more than 600 edits gets to participate in collaboration to improve the guidelines and possibly be the one who does the actual edit that makes the change. Me! Yes, me! Huggums537 (talk) 07:01, 26 September 2017 (UTC)

Making a couple of tweaks for WP:TONE and flow. Hopefully nothing problematic...
Revision #2

Review aggregation websites such as Rotten Tomatoes or Metacritic are citable for data pertaining to the ratio of positive to negative reviews; caution should be exercised when using aggregator scores that combine original reviews with reviews from later dates. The data from these websites is potentially less accurate for films released before the websites existed, therefore care should be exercised in determining whether to refer to them. Consider placing the data at the bottom of the section with an explanation of why there may be inaccuracies. It is also recommended to include the date the data was captured: ("As of May 2015, 50% of the 68 reviews compiled by Rotten Tomatoes are positive, and have an average score of 5.2 out of 10."). See this summary of limitations regarding aggregators' accuracy on earlier films for more information on this subject.

Huggums, I'm fine with you implementing the final form of this, if you really want to put your name on the firing line. :p DonIago (talk) 14:15, 26 September 2017 (UTC)

Excellent tweaks! The wording flows very well in my opinion. Also, thanks for allowing me to "pull the trigger"! So, I guess now we just wait to see if anybody else has any objections? Huggums537 (talk) 22:35, 26 September 2017 (UTC)
Yep. I think if nobody speaks up within a week, we can consider that implied consensus. If someone has an issue at that point they can always revert and speak up here at that point. DonIago (talk) 13:08, 27 September 2017 (UTC)
I'm iffy on the addition. "Consider placing the data at the bottom of the section" may be used by some to start moving the content to the bottom when it really is not needed and disrupts the flow. I would change "Consider placing the data at the bottom" to "Consider whether the data is best placed at the bottom of the section." Better yet, maybe we should go with "lower in the section" instead of "at the bottom of the section" since there will be some cases where the data flows better a little higher up. Or we could state "Consider whether the data is best placed lower in, or at the bottom of, the section." As for "with an explanation of why there may be inaccuracies," is this referring to the edit summary? That is what I assumed. If it's referring to the section, there usually will not be sources explaining why there may be inaccuracies. As for "include the date the data was captured," I don't feel strongly about it, but I usually prefer to forgo "As of." Flyer22 Reborn (talk) 04:16, 28 September 2017 (UTC)
Flyer, I introduced the idea of explaining the inaccuracies because I realized that the first few proposals were essentially instructing editors to go ahead and include possibly inaccurate information "as long as you place it at the bottom". I could not find a way to reconcile this until I had the idea to include some kind of disclaimer somewhere in the body of the article to give readers a notice of explanation for it. Also, I think any one of your ideas that begin with the word "consider" sound great to me. Huggums537 (talk) 11:16, 28 September 2017 (UTC)
Revision #3.14159

Review aggregation websites such as Rotten Tomatoes or Metacritic are citable for data pertaining to the ratio of positive to negative reviews; caution should be exercised when using aggregator scores that combine original reviews with reviews from later dates. The data from these websites is potentially less accurate for films released before the websites existed, therefore care should be exercised in determining whether to refer to them. To avoid giving these sites undue weight in such circumstances, consider placing the data lower in the section, with an explanation of why there may be inaccuracies. It is also recommended to include the date the data was captured: ("As of May 2015, 50% of the 68 reviews compiled by Rotten Tomatoes are positive, and have an average score of 5.2 out of 10."). See this summary of limitations regarding aggregators' accuracy on earlier films for more information on this subject.

How's this? Huggums, I know you're keen on the explanatory text, but I agree with Flyer that sources are unlikely to provide that, and it can be reasonably inferred that the earlier a film came out prior to the site existing, the greater the potential risk of inaccurate data. I've incorporated Flyer's suggestion that the data be placed lower down but not necessarily at the bottom, and I've thrown in a reference to WP:UNDUE to explain why the data should be de-emphasized. At this point, if we're going to revise the text further, could I ask that a dissenting editor provide the next version? DonIago (talk) 13:12, 28 September 2017 (UTC)
Thanks for the piece of pi! I hope it's not an omen... I'm not so sure the inference is reasonable, but the possibility that it could be makes me a little more flexible about parting with the explanation. So, I'm glad you mentioned it. Huggums537 (talk) 14:44, 28 September 2017 (UTC)
I'm fine with the latest revised text. Flyer22 Reborn (talk) 20:03, 30 September 2017 (UTC)
Not to undercut myself, but is "caution should be exercised when using aggregator scores that combine original reviews with reviews from later dates" so similar to "care should be exercised in determining whether to refer to them" that the text should be refactored? Maybe strike the first instance and make it "However, the data from these websites is potentially..."? DonIago (talk) 00:54, 1 October 2017 (UTC)
I think it's probably ok either way. I'm with Flyer that it's fine the way it is. In fact, I think the texts should be similar since they are expressing very similar ideas. Also, I'm not clear as to what the change you're proposing is exactly, so it's hard to agree or disagree. I know you spelled it out above, but that could be interpreted a couple different ways. Huggums537 (talk) 07:56, 1 October 2017 (UTC)
Also, if it's true that the inaccuracies can be reasonably inferred by readers who are most likely not nearly as familiar with the aggregator websites as editors, then why are we having to explain the possibility of inaccuracies to editors to begin with? If one rationale makes sense, then the other doesn't. Do you see the conflict there? That's kind of why I'm still for the explanation to readers also, because if editors need an explanation, then readers probably do as well, and if readers don't need one, then why do editors need one? See the catch there? Huggums537 (talk) 12:46, 1 October 2017 (UTC)
Not really? And what about Flyer's concern as to how one could provide sourcing for such an explanation? DonIago (talk) 14:59, 5 October 2017 (UTC)
Sorry. I've been out of pocket for a while. I think the concerns about sourcing are irrelevant since sourcing isn't really needed in order to make the general inference more obvious to the reader/editors that the data from existing sourcing could be inaccurate. The need for an explanation rests in it's ability to transform the "possibility" of making the inference into a "certainty". If the possibility of the inference exists in the first place, and sourcing is no problem for the "possible" inference, then sourcing should also be no problem for making the inference a "certainty". What I was trying to express earlier is that if we need to make certain that editors understand this guideline (inaccuracies issue) then we should also make certain the readers understand the issue as well. It's sort of double standard to say that readers can make the inference on their own without any help, but editors can't make the inference, so they need an explicit guideline for help. It makes more sense to say that we can can help make certain that both editors/readers will make the inference. Also, if readers actually really could easily make the inference on their own without any help, then so could editors, and an explicit guideline explaining possible inaccuracies would not be needed for editors either. That's why I said it seemed like a conflict to me. Does it make more sense explained in that way? Maybe I'm making too big a deal of it, but it just doesn't make sense to me why we feel it's so important to warn editors about these inaccuracies, but don't feel obligated to warn readers in any way. Weird. Huggums537 (talk) 22:12, 9 October 2017 (UTC)
I think that the "caution should be exercised when using aggregator scores that combine original reviews with reviews from later dates" part, which is currently in the guideline, is Betty Logan's addition. It's there per what is stated in the section immediately above this one. Flyer22 Reborn (talk) 17:40, 1 October 2017 (UTC)
DonIago recently contacted me to advise that nobody besides myself seems to have any opposition to the last changes to the proposed text for over a week now. So, I'll incorporate the text as it is if nobody has any objections. I'm not married to the inclusion of an explanation in articles, and the worries that Flyer22 Reborn/Doniago had about sourcing will be averted if we leave that part stricken. Besides, I got to thinking about it, and I'm not too fond of forcing editors to plaster notices on articles anyway. Maybe it can be introduced as a suggested option to editors somewhere down the line, but for now, I'm happy as is if everyone else is. Huggums537 (talk) 01:41, 18 October 2017 (UTC)
I hadn't noticed that the latest proposals included a pointer to Wikipedia:Review aggregators#Limitations. I removed it, per arguments made above. It's best not to link to essays in our policies and guidelines anyway, unless the essay is included in the See also section. Furthermore, they are commonly removed from our policies and guidelines unless included in the See also section. I also changed "consider placing the data lower in the section" to "consider whether it is best to place the data lower in the section," per an argument I made above about not making it seem like the aggregator(s) must be placed lower. I know that we revised the text with softer language, but I think that "consider whether it is best to place the data lower" is even softer. Flyer22 Reborn (talk) 00:06, 21 October 2017 (UTC)
Flyer22 Reborn, I like your edit that revises the text with even softer language, and I think "consider whether it is best to place the data lower" sounds just fine to me. However, I'm confused about your objection to pointing to that part of the essay since the critical reception section already points to the same essay right below the section header within it's own "See also sub section" (of sorts). Also, I was unsure which statements you referenced "per arguments made above" about the pointer. If you were referring to the statement Betty Logan made about the first draft which suggested that "it would be better if we simply transferred the "2000s and beyond" bullet point from Wikipedia:Review_aggregators#Limitations into MOS:FILM" then maybe we should have waited another week or two for her to respond. Perhaps we could revisit Betty's idea as a possible solution since you disagree with pointing to essays. I guess I'm a bit puzzled as to why the pointer has existed through half a dozen iterations of the proposal over the past month and no real objections have been made about that part of it for some time now until the actual implementation of it just now. Weird. Anyway, I actually don't have a problem with you removing the proposed pointer since a pointer already exists anyway, only the pointer that DonIago proposed links to a specific section of the essay, whereas the pointer under the section header links to the essay in general. Lastly, please accept my apologies for previously altering your comment. I'm just going to have to get over the fact that typos exist, and I'm not good enough at mind-reading what other people's intentions are to attempt fixing them. Huggums537 (talk) 08:25, 21 October 2017 (UTC)
Hi, Huggums537. No need to ping me to the page since it's on my watchlist. I'm not sure that I agree with the Critical response section stating "See also: Wikipedia:Review aggregators," but at least it's listed as a see also. And I did mention including such links in a "see also" way above. Pointing to essays directly in the text, as though they carry any weight, is something we commonly avoid in our policies and guidelines, as recently as this edit at WP:NOT. As for where I disagree with what Wikipedia:Review aggregator states, see my "20:12, 8 September 2017 (UTC)" post in the #RT for older films section above. I stated, "Except for the 'aggregators should not be used for pre-2000 films,' I agree with what Betty stated in her 'I am of the general view' paragraph. I would obviously support explicit caution against using Rotten Tomatoes for very old films. I just don't think we should state that they shouldn't be used for pre-2000 films. Really, the WP:AGG essay should changed in that regard. If the older film's reception has generally remained consistent and the Rotten Tomatoes view reflects that, I don't see an issue with including Rotten Tomatoes, but I wouldn't place it at the beginning for very old films. And if the film's reception has changed, and the Rotten Tomatoes score/consensus is included in the part of the paragraph making the change clear, I think it's fine in that case as well." Betty had a response for that, but I maintained that "I really can't support the guideline stating that we should not use Rotten Tomatoes for pre-2000 films. I support cautioning against use of Rotten Tomatoes for very old films, and approaching this matter on the case-by-case basis I noted above. As this is a guideline, we should be guiding more than restricting anyway." The wording at Wikipedia:Review aggregators is currently stricter with regard to pre-2000 films. As for missing that the proposal included a pointer to the essay, I noted in a followup edit that "I wasn't paying good enough attention." That's all it was -- I somehow overlooked that part of the text each time. Flyer22 Reborn (talk) 17:46, 22 October 2017 (UTC)
Regarding this edit you made to my comment, it was fine. Your removal was correct. But, as seen by the edit I made after that, more fixing was needed. I hate typos very much, and your edit brought attention to what I needed to fix on that matter. Flyer22 Reborn (talk) 18:00, 22 October 2017 (UTC)
Flyer22, sorry about the unnecessary ping, and thanks for clarifying things with quotes. I clearly see your disagreements now. I guess I missed them because I wasn't looking beyond the already somewhat lengthy "proposed text" subsection of this mega-discussion. Also, thanks for your understanding about the typo edit. Huggums537 (talk) 04:41, 23 October 2017 (UTC)

Proposed clarification of reviews' relation to WP:PSTS and MOS:TONE

I'm frequently seeing over-broad assertions like "A film review is a secondary source for a film [novel, TV show, etc.]" (and several variants of this sentiment). This strikes me as misleading, confusing, and a gross overgeneralization.

To make a specific recommendation of draft wording, a more accurate version might be something like:

A review of a work can be – by Wikipedia's definitions and the very nature of a review – a secondary source for the plot of a film and some other self-evident objective facts about it, where the review lacks speculation on these matters. Reguritative material, such as the cast and crew, and other details simply pulled from the credits, are tertiary sourcing when they come from a review.

This type of source remains a primary source for anything subjective about the film, if this came directly from the author(s) of the review. A review can sometimes be secondary for a specific subjective assessment, but only if it explicitly draws on the views in previously published materials by others about that work, that director/producer, etc., in coming to the assessment.

Generally, a secondary source for something subjective about a film is essentially a review of reviews, involving analysis, evaluation, interpretation, and/or synthesis of various reviewers', critics', or academic analysts' views. Examples include non-automated meta-reviews, and articles in arts and media journals that examine critical reception.

Because it remains subjective opinion, such an synthesized assessment is always on the cusp of the primary/secondary distinction, and thus should generally be directly attributed, and often directly quoted, especially if paraphrasing may raise neutrality or tone concerns. In short, you are not allowed to say that a work is good or bad, tedious or exciting, classic or a ripoff, in Wikipedia's own voice no matter how many reviewers have said so.

I think the guidelines should spell this out in wording fairly similar to this, in some format, whether it be paragraphs like this, or a bullet list of points, or whatever. This will be consistent with our WP:V, WP:NPOV, and WP:NOR policies, with the WP:RS guideline, and with WP:Manual of Style#Point of view, which specifically and intentionally (I would know – I wrote that part of it) uses subjective review material in its examples. This is written generically enough that whatever final version emerges can be put into the main WP:MOS (or MOS:FICT though this applies to artistic and nonfiction/documentary works, too, not just fiction) and simply summarized and linked to from MOSFILM and other fiction-related MOSes, and from RS or NOR. Or the material could live at RS or NOR and be linked to from MoS. I don't care. This is both "what you can say based on what" content guideline and "what you can say, and how, in an encyclopedic voice" style guideline material (as are various other MoS rules and even MOSFILM ones).

PS: This is a discussion draft, not a !vote; this would likely require an RfC to implement in final form.
 — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  15:19, 22 September 2017 (UTC)

While correct, I think it might be better to start that a goal of a Reception section is to summarize the individual reviews of the film, and ideally that would be through a secondary/tertiary source that evaluates the collective of reviews, which are individually primary sources for their critiques. This type of source, short of MC/RT's short summary, doesn't typically exist, so to that end we do allow careful construction of a summary of these primary sources for the film's reception, the individual reviews, for a Reception, as long as synthesis of new thoughts is avoided and due weight is kept to in considering the top-tier review sources, stressing the need for attribution, citations, and quotations. Then examples of what is not synthesis, and what is synthesis in these types of areas. For example, one type of acceptable synthesis would be the case of if many of the key reviews praise an actor's role in a film, then it is not synthesis to say "(Actor)'s performance was highlighted by reviews. (add few choice quotes here)." It would be synthesis to take reviews from when the film was released, and from X years in the future, note a difference in tone, and say "The film's reception changed over time", generally. There's a handle of other representative situations that could be added here too. --MASEM (t) 15:38, 22 September 2017 (UTC)
No disagreement Masem, but you're talking about the narrow goal of the big discussion above, and I'm addressing a broader one across fiction coverage and how we source it. So, the two ideas are not at cross purposes.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  15:57, 22 September 2017 (UTC)
We probably need something outside of this MOS, dealing with any type of reception related to published works, and the howtos of doing these "right" that avoid OR, which would by necessity need to explain the primary/secondary relationship of reviews to films and the like that you have outlines. Possibly, this could go into WP:WAF but this needs to extend to non-fiction works too. I would think of developing this as a separate guiideline so that you can start with definitions and concepts , and then get to the core of the matter that combines that, and then provide a do / do not type list or provide strong examples where it is done right. --MASEM (t) 16:04, 22 September 2017 (UTC)
Yar. I was just thinking of how much of MOS:WAF actually applies to writing about works of fine art, and about documentaries and non-fiction books. That whole page may need to be broadened a bit. I think it's already understood by most editors who care about such things that the parts of it that generalize beyond fiction per se are to be treated as applicable (the way we write articles on such works demonstrates it), but either it could say so explicitly, or even be renamed to, e.g. WP:Manual of style/Writing about works. The separate guideline idea no longer seems to fly; proposals to create them almost always fail in the 2010s, and it's generally only feasible to introduce new guideline material into existing GL pages. Not impossible to have a new one, but surely an uphill battle.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  21:49, 22 September 2017 (UTC)
I agree, and since WP:CONTEXTMATTERS and WP:ANALYSIS already cover this adequately, I don't see the change you're proposing as controversial. Bright☀ 15:47, 22 September 2017 (UTC)

I'm still not seeing why a film review is not a secondary source in regard to the primary source, the film. First, WP:PRIMARY says, " A primary source may only be used on Wikipedia to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the primary source but without further, specialized knowledge. For example, an article about a novel may cite passages to describe the plot, but any interpretation needs a secondary source." Per WP:SECONDARY, "A secondary source provides an author's own thinking based on primary sources, generally at least one step removed from an event. It contains an author's analysis, evaluation, interpretation, or synthesis of the facts, evidence, concepts, and ideas taken from primary sources." Are you trying to say that a film review does not count as an analysis, evaluation, interpretation, or synthesis of the film? Erik (talk | contrib) (ping me) 16:57, 22 September 2017 (UTC)

You are correct. The issue extends thought to writing summary statements that represent a collective number of reviews. Saying "Many reviewers praised X's performance." (for example) is a type of synthesis of primary sources with respect to reviews' opinions on the film. As to how much that synthesis is a problem (going beyond what allowances we do give for summarizing sources as a necessity of writing articles, and instead entering into novel interpretation, against NOR) is the issue. --MASEM (t) 18:02, 22 September 2017 (UTC)
Another issue is failing to distinguish between secondary and primary material in a one-author review of a film's content. To just copy-paste the example I gave at another page: The whole problem with reviews as sources is that they can veer between secondary and primary right in the same sentence, multiple times: "The Fisher King-like character of Jake is subjected to unreasonable demands and expectations from friends, family, and neighbors upon returning to his hometown, an idyllic fantasy village reminiscent of those in Big Fish and Edward Scissorhands; he struggles with depression as he lets everyone down, until he reaches out to his Percival – the 'manic pixie dreamgirl' love interest Jennifer." Quite a lot of that is primary and right from the reviewer's own head, unless he/she is explicitly drawing on prior analysis of this film. We can't say in WP's own voice that this movie is based on the Fisher King of Arthurian legend, though we can attribute this assessment to the reviewer; same goes for the potentially controversial assessment of the nature of Jennifer's character, or the subjective comparison of the town to those in other films. We don't need to attribute the bare summary parts, e.g. that Jake went home, had trouble coping with the demands placed on him, and reached out to Jennifer who became a love interest; that's secondary sourcing. A WP plot summary is on shaky grounds when it weaves in one-reviewer assumptions and opinions, even with citations.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  21:49, 22 September 2017 (UTC)
I get the impression that you are highly passionate about the issues and wish to express your views. However, I feel that your proposal is somewhat opinionated and overly complex. Also, there are some technical issues that may be confusing. For example:
  • "Reguritative material, such as the cast and crew, and other details simply pulled from the credits, are tertiary sourcing when they come from a review."
Well, a review could gather those details from other places besides the credits. So this entry is somewhat misleading and could confuse the reader as to what quantifies tertiary sourcing.
  • "This type of source remains a primary source for anything subjective about the film, if this came directly from the author(s) of the review. A review can sometimes be secondary for a specific subjective assessment, but only if it explicitly draws on the views in previously published materials by others about that work, that director/producer, etc., in coming to the assessment."
This whole passage is obviously opinionated, as indicated by the disagreement of other editors regarding the dispute of what is/n't a primary/secondary source.
  • "Generally, a secondary source for something subjective about a film is essentially a review of reviews, involving analysis, evaluation, interpretation, and/or synthesis of various reviewers', critics', or academic analysts' views. Examples include non-automated meta-reviews, and articles in arts and media journals that examine critical reception."
This passage seems kind of unnecessary since most everybody knows this, as evidenced by the fact that they are already using "reviews of reviews" as sources.
  • " Because it remains subjective opinion, such an synthesized assessment is always on the cusp of the primary/secondary distinction, and thus should generally be directly attributed, and often directly quoted, especially if paraphrasing may raise neutrality or tone concerns. In short, you are not allowed to say that a work is good or bad, tedious or exciting, classic or a ripoff, in Wikipedia's own voice no matter how many reviewers have said so."
Again, you are correct. However, these policies and guidelines have adequate coverage, and it seems redundant to repeat it here. Also, at the risk of being nit picky, "Wikipedia's own voice" seems like confusing and over-generalised terminology. Please don't kill me for being so critical. Huggums537 (talk) 01:08, 23 September 2017 (UTC)
I think you misunderstood why I said it's misleading. Many details that will be found in film credits are available to reviewers even while the film is still in production. So, analysis of the same information found in the credits could also come other sources. Your analogy below with the monument makes the misleading implication that the information just had to have come from the credits through a "chain of custody", (as you called it). Talking about films is very different from talking about monuments for this purpose. I understand the point you are trying to make, but it's still misleading in that respect. Huggums537 (talk) 01:45, 25 September 2017 (UTC)
Then they're copying-pasting details from somewhere else; it's the same situation. We deal with this all the time when "journalism" just regurgitates as a press release; it's not actually secondary sourcing. But this is off-topic side trivia; whether Angelina Jolie was really in the film Hackers isn't controversial in any way, so a tertiary source for it is sufficient, and so is the film itself as a primary source; it doesn't actually matter in any way whether a review that say she was in it is primary or secondary. This isn't the kind of fact under discussion here. The proposal at issue is primarily about WP:AEIS claims about the content, while the earlier, longer thread is primarily about a SYNTH claims in particular, of one sort: claims that critical reception has changed over time. PS, on "Wikipedia's own voice" – it's a phrase MOS and some other pages use more than once, and people seem to get it. I'm not introducing it.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  05:48, 25 September 2017 (UTC)
Even if we were talking about monuments, it would still demonstrate that your argument is just as opinionated as your proposal due to putting a bias of undue weight on tertiary sources (all of the weight, in fact). In your analogy, you're presenting only examples of tertiary sources covering monuments. This misguides the reader into forming their own false conclusions that all sources covering monuments must be tertiary sources. That would be just as absurd as me providing a few examples of secondary sources covering monuments and expecting readers to conclude that all sources covering monuments are secondary. A fair and impartial argument would say that tertiary sources covering monuments are tertiary sources, and secondary sources covering monuments are secondary sources. I could also argue that a secondary source copied from a secondary source, copied from a secondary source, is still a secondary source, which doesn't transform it into a tertiary source, and that would be just as opinionated (not to mention unreasonable). Huggums537 (talk) 01:09, 26 September 2017 (UTC)
Briefly:
  • Regurgitative material: It's still tertiary, because it's direct repetition without any analysis, evaluation, interpretation, or synthesis. If Enc. Britannica got the date on a monument from Collier's who got it from another encyclopedia, who got it from another, who got it from another, who got it from a photograph of the dedication plate on the monument with the date on it, Enc. Brit. remains a tertiary source when we cite them for the date, even if the entire "chain of custody" of the information is clearly visible in citations from one work to another. There is no magical transubstantiation that happens to turn tertiary material into secondary material when it is rote repetition. This principle is why WP cannot use WP:UGC material as secondary sourcing even if it cites its own sources; otherwise WP:CIRCULAR would not exist, and we could use one WP article as a source for another one.
  • "if it explicitly draws on the views in previously published materials" – Opinionated? It's a summary of the relevant aspects of WP:PSTS. And it could be copyedited to tweak the wording of the summary. "You don't agree with me" and "you are wrong" are hardly synonymous. "You care a lot" and "you are wrong" certainly aren't either. I believe you're imagining a tone problem, though feel free to actually identify one and propose a revision that rectifies it.
  • "This passage seems kind of unnecessary since most everybody knows this" and "these policies and guidelines have adequate coverage, and it seems redundant to repeat it here" – Except this thread unmistakably disproves these assertions. See comment by Jclemens immediately below as just one example among others I've already pointed out in the larger thread. I would agree that MOSFILM is not the place to codify this material, which is why I've suggested it be done somewhere else, and just summarized here and at other fiction MOSes as a reminder.
 — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  10:44, 24 September 2017 (UTC)
  • Simply incorrect Any plot summary, any interpretation, any subjective impression... these are all secondary. Direct quotes or bare facts such as run time, rating, etc. aren't, but that's not the point of the review. It is simply not possible to summarize a work without editorial decision on what is and is not important, and thus any review is secondary. Jclemens (talk) 05:42, 23 September 2017 (UTC)
  • Everyone arguing with SMcC on this is wrong, and it's incredibly disturbing that so many experienced users could hold such an incorrect understanding of WP:PSTS. How on earth Jclemens can square [A]ny subjective impression ... [is] secondary. [...] [A]ny review is secondary. with the following, more accommodating, admission that the review is itself a new primary document about the review itself is beyond me. This whole affair frankly reminds me quite a bit of a quite-serious incident some years back when someone claimed that an ancient ancient chronicle was not a primary source for a supposed battle it reported on because its compilers had not been personally present at the battle, even though it was literally the only extant source to mention the battle. Or, to give a more recent and content-relevant example, when this and this happened. Hijiri 88 (聖やや) 06:54, 23 September 2017 (UTC)
    • It's pretty easy, actually: I don't redefine secondary sources to suit what I think they ought to be. If a secondary source exists about an ancient battle and no primary sources exist... well, then you have no primary sources. Original content is primary. Quotes of said original content are primary. Commentary, summary, reviews... secondary until the cows come home. You simply cannot do a plot summary without analysis (of the primary source) and interpretation (deciding what to keep). Please, feel free to refute this, as I'm as boggled by those of you who think reviews of fictional media aren't secondary. Jclemens (talk) 07:51, 23 September 2017 (UTC)
If a secondary source exists about an ancient battle and no primary sources exist... well, then you have no primary sources. That's not how it works. If there's an ancient chronicle from which all information on a particular event is subsequently taken, that ancient chronicle is the primary source for that event. It doesn't matter whether the compilers of the chronicle were direct personal witnesses of the event. One of numerous fallacious assumptions that user made a few years back was that the compilers composed their own original account rather than (metaphorically) "copy-pasting" an earlier account written closer to the event, possibly by an eyewitness (something that is demonstrably the case with other historical works, such as Luke-Acts, which is also recognized project-wide as a primary source). Ancient chronicles can never be used as secondary sources to demonstrate notability, and are by definition subject to pretty much all aspects of WP:PRIMARY -- we are not allowed use them as the only sources for articles, we need to take care when using them for extensive passages therein, and so on.
You simply cannot do a plot summary without analysis (of the primary source) and interpretation (deciding what to keep). So ... you think MOS:FILM's allowing us to create our own summaries without reference to secondary sources is out of line with the spirit of NOR? It's good that we agree on that point, I guess, but I think it puts you firmly in the minority of editors of this talk page.
I'm as boggled by those of you who think reviews of fictional media aren't secondary You are assuming (despite your own admission otherwise further up) that certain types of sources just "are" primary or secondary, regadless of context. If you think that the tags I placed on those Game of Thrones articles were inappropriate because the articles contained extensive passages cited to "secondary sources" ... well, pretty much everyone who replied on those article talk pages appeared to agree with me, so I would advise against simply removing the tags, but if you wanted to I guess you could reopen the discussions in place of finding more secondary sources that analyzed the episodes' critical reception.
Hijiri 88 (聖やや) 08:18, 23 September 2017 (UTC)
WP uses the idea that if there is transformation of primary or secondary information into a work about a topic, then that work is secondary to the topic, period. Keep in mind that means a recollection of an event by a person but which does not contain any analysis or transformation of information short of summarizing is a primary source for the event. (In film, this would be a recap of a film without any review). --MASEM (t) 08:25, 23 September 2017 (UTC)
@Masem: Please define "to the topic", and explain how it differs, or does not differ, from "to the content under discussion". Also, your parenthetical clause at the end seems to indicate that you actually agree with SMcC and me. We are not talking about simple plot recaps; what we are talking about is opinionated film reviews. Hijiri 88 (聖やや) 08:42, 23 September 2017 (UTC)
It probably means the same thing; basically just because we have a secondary source of a review for a film does not mean it is necessarily a secondary source for all other aspects of a film (here, the point in question being the broad describe of the overall reception of the film). My concern is your hypothetical example which puts more weight on "steps-removed" approach to primary/secondary distinction, rather than the transformative element we use for the same. "Steps-removed" is necessary to judge third-party and/or independence but not the primary/secondary distinction. --MASEM (t) 13:06, 23 September 2017 (UTC)
Hijiri88, It's pretty clear that the use of documents on ancient battles is confusing. A source that analyzes other primary sources is a secondary source, full stop. Now, if it's ancient and the primary sources are lost, it's still a secondary source... but 1) we study it, rather than the events depicted, because 2) it's not conducted with the same standards of documentation we expect in the modern era. As such, I don't disagree with the cautions about using ancient secondary sources you're citing--just that being under the same restrictions as primary sources and treated like primary sources doesn't make them such. But at any rate, we're not talking about ancient secondary sources that are de facto primary witnesses to their era; we're talking about reviews.
I'm trying to respond in good faith here, but attacking plot summaries sourced to the fictional work itself as OR, or implying that I do, is either a red herring or a straw man. Wikipedia's plot summaries are secondary sources, but are not reliable because of our collaborativist nature.
I haven't done a thing to your GoT episode taggings, and hadn't planned to get involved in that discussion. I came here because I was notified on a page I watch, MOS Film isn't on my watchlist. Maybe I should, but I'd rather do other things with my time. I appreciate editors who try and bring discipline and rigor to fictional coverage, when such is available, but I am very much of the opinion that we use the best we have available, and those who desire additional rigor have the responsibility to 'upgrade' the coverage by replacing poorer sources with better ones, rather than simply complaining about the paucity of acceptable sources. Cheers, Jclemens (talk) 00:53, 24 September 2017 (UTC)
Well, with the ancient source, we don't know if it's "analyzing" other, primary, sources, of if it's quoting them verbatim. To give a topic-relevant comparison, if a review contains a verbatim quote from the film under review, and for whatever reason does not provide secondary analysis of the quote (actually whether it does or not, but that's immaterial), that portion of the review that is a quotation is still the primary source.
To go back to the ancient chronicle analogy, such chronicles frequently do present information on events to which their compilers were (or probably were) direct eyewitnesses, mixed in sometimes with material they clearly got from other sources. The Jewish War, for instance, includes information on the conquest of Jerusalem by Antiochus IV 200 years before its author was born, but is primarily focused on a later conflict in which its author himself was a combattant, although he certainly wasn't an eyewitness to everything. Sources are primary for some kinds of information and secondary for others, so A source that [does foo] is a secondary source, full stop. is an oversimplification -- not one that I think reflects a poor understanding of proper sourcing on your part, mind, just one that I think would be unhelpful if enshrined in MOS, because most Wikipedians can't read between the lines as good as you can.
With film reviews: all film reviews are, by definition, primary sources for their own authors' opinions; they are usually secondary sources for the content of the films they review (although, as opinion pieces that are not subject to strict editorial oversight or fact-checking, they are frequently unreliable as such); they are almost always primary sources for the films' critical reception, with the rare exception that retrospectively analyzes both the content of the film and the critical consensus.
This means that if we have an article on a film (or an episode of a TV show) that includes (a) a lead section, (b) a plot summary, (c) a section discussing "development", and (d) a section discussing "reception", (a) and (b) contain no explicit citations, (c) is sourced to six interviews with members of the cast and crew, and (d) cites Rotten Tomatoes, MetaCritic, and six individual reviews, then out of fourteen inline citations, only two are to non-primary sources. (This leaves aside the question of whether RT's sometimes-dubious assessment of whether this or that reviews is positive or negative skews their rating and makes it essentially just an opinion that needs inline attribution like a primary source.) When we have editors engaging in long debates on talk pages over their claim that "reviews are secondary sources, full stop" (and, though it's off-topic, "The Hollywood Reporter is a secondary source, full stop"), I think it would be helpful if MOS clarified proactively that this is not the case. (Yes, the overreliance on primary sources is more a problem with articles on TV shows than films, where films that have standalone article generally have to be discussed in secondary sources before they get those standalone articles, but the same general concern applies here.)
Hijiri 88 (聖やや) 03:06, 24 September 2017 (UTC)
Ah, but the fact that a review is primary for its author's opinion, which I do not dispute, is irrelevant when we're talking about coverage of the film: the author's primary opinion is secondary about the film, and in an article on the film, it's entirely appropriate. If we were talking about the reviewer's work, for sake of argument, then we'd need sources that commented on that reviewer's work, because his own reviews are not suitable secondary sources for an analysis of his work. It's not any more difficult than pointer arithmetic in C, I assure you. :-) Jclemens (talk) 05:29, 24 September 2017 (UTC)
Yes, but individual reviews are still primary for a film's overall critical reception. Hijiri 88 (聖やや) 06:43, 24 September 2017 (UTC)
No they're not--they're not even that: They are individual opinions about the film, from which our coverage is constructed. Multiple viewpoints are represented per WP:YESPOV--or at least that's the way it's supposed to work. Jclemens (talk) 00:34, 25 September 2017 (UTC)
Keep in mind that the fact a work gets reviewed by multiple RSes is why they are considered notable, because those reviews are considered secondary sources for the published work. As Jclemens points out, they are primary about the reviewer's opinion but secondary to the film. --MASEM (t) 05:33, 24 September 2017 (UTC)
...and? No one here is arguing for a change to NFILM to make film reviews unacceptable as secondary sources to demonstrate notability. I'm talking about basing large chunks of articles on topics that are notable on primary sources. The vast, vast majority of film reviews cited on Wikipedia are not cited as secondary sources for information on the films themselves: they are being cited as primary sources for the reviewers' opinions and for the films' overall critical reception. Hijiri 88 (聖やや) 06:43, 24 September 2017 (UTC)
Actually, some clarification might be necessary regarding that, too. Many city newspapers make a point of doing at least a basic review of every movie shown in town that week, so there are no mainstream films that will not have numerous reviews written about them. But they're not in-depth coverage. I don't presently see confusion about this on this talk page, but given that people are seriously confused about basic PSTS sourcing, I would not be surprised.
From the standpoint of what qualifies for a good WP article, incorporation of a number of "leading" film review's comments and opinions about a film show that there is indeed secondary sources about the film (via these reviews) and it is essential as a published work to describe some of this reception. The more subtle issue is how one assembles that reception section without introducing unallowed synthesis. Assembling of multiple ideas (primary or secondary) from multiple sources (primary, secondary, tertiary or whatever) is a necessary bit of synthesis for any part of writing WP, not just reception sections. As long as there is a reasonable defined practice for writing film receptions that uses the most respected reviewers and keeps focused on DUE, then there is usually no problem in the weak synthesis needed to provide narrative flow within a reception section from the multiple primary works about the individual reviewers' thoughts. (We don't want to be troweling local paper reviews to find the one that praises the film when everyone else ranted about it, for example, or vice versa). And I do disagree that "The vast, vast majority of film reviews cited on Wikipedia are not cited as secondary sources for information on the films themselves". A very common example is that reviews will praise or pan an actor, director, or some other key personnel relative to either their past works or to the industry as a whole (eg The Dark Knight (film), in discussing Ledger or Nolan's work). Those are clearly incorporating secondary sources about the film (it's transformative information, in this case constructive criticism of the primary source), even though they are primary sources for the reviewers' comments. I think we're getting far too hung up on the nature of "primary" or "secondary" and need to simply keep in mind that reception sections, like any other article topic, film or otherwise, is generated from summarizing and paraphrasing multiple sources, and as such, editors have to be careful of keeping the necessary amount of synthesis needed to summarize (allowed by NOR) to a minimum, follow DUE, and avoid crossing the line into true NOR problems. --MASEM (t) 13:39, 24 September 2017 (UTC)
And I do disagree that "The vast, vast majority of film reviews cited on Wikipedia are not cited as secondary sources for information on the films themselves". A very common example is that reviews will praise or pan an actor, director, or some other key personnel relative to either their past works or to the industry as a whole (eg The Dark Knight (film), in discussing Ledger or Nolan's work). If a reviewer praised or panned an actor, director, or some such, do we cite their opinions on these points as facts in Wikipedia's voice? No: we almost always attribute such opinions inline because the reviews are primary sources for said opinions. We do this in the context of discussing the films' critical reception, for which, yes, individual film reviews are primary sources. The Dark Knight article does not appear to be an exception. I have no idea what you are talking about with Those are clearly incorporating secondary sources about the film (it's transformative information, in this case constructive criticism of the primary source), even though they are primary sources for the reviewers' comments., since it would appear to directly contradict your claim immediately above that we are citing them for their praising or panning some aspect of the film. You say they are secondary sources for the content of the film and primary sources for their own opinion, but you don't back up your claim that any more than a tiny minority of such citations are for the content of the film. Hijiri 88 (聖やや) 21:54, 24 September 2017 (UTC)
Critiquing a specific actor's role, a director's vision, a composer's songs within the context of a film are all content about the film. Yes, as opinions, they must have inline attribution to the reviewer, but when incorporating all this, this is adding to content about the film. If you're talking about the actual plot of the film, and the themes it conveys, that's a totally different matter, as most reviews do not cover the film in that amount of depth to use that. But that's where you start going to film historians and experts that can work that out for us, those are still secondary sources for us. --MASEM (t) 22:28, 24 September 2017 (UTC)
They are not being cited for factual information about actor's roles or director's visions (film reviews hardly ever give that type of information, and they would be unreliable sources for it anyway as they are not subject to fact-checking or significant editorial oversight); they are being cited for their authors' opinions of such. You actually have things the wrong way around -- many reviews will give a brief and largely accurate outline of a film's plot (usually but not always avoiding spoilers, mind you), while not giving independent, factual coverage to behind-the-scenes details. But it's pretty clear I'm not going to convince you of this point. If this comes to an RFC I'll make a statement their summarizing my main points in all the above comments. I'm confident the vast majority of the Wikipedia community would be in basic agreement with me on all of them. Hijiri 88 (聖やや) 23:23, 24 September 2017 (UTC)
You're arguing there are two types of secondary sources for a film, the difference between a review and, say, production details. Both are secondary sources for the film, as they are both transformative of the primary source; a review is going to be based on what a reviewer saw come to the screen, production details are going to be insights of what happened before it got to the screen (and argubly this is closer to primary information, but I consider it secondary for the most part particularly in light of notability concerns). Secondary information is not required to be factual, just transformative that you cannot get by a simple summary of the primary source, the film itself. --MASEM (t) 23:40, 24 September 2017 (UTC)
An opinion can be an interpretation or evaluation, both of which are supported in the definition of a secondary source. Its author has analyzed elements of a primary source, including concepts and ideas, and formed a conclusion on how well those were executed. I guess I'm not understanding why a film review doesn't qualify as a secondary source for that analysis. If opinions weren't allowed, then surely "interpretation" wouldn't exist in the definition. A book review is even listed as an example of a secondary source. --GoneIn60 (talk) 11:01, 25 September 2017 (UTC)
In all the above, I have to concur with Hijiri88, as does pretty much everyone else on WP for the last 16 years; these are basic sourcing principles. It really is alarming how many people don't get it and stubbornly refuse to do so when it comes to writing about fiction.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  10:44, 24 September 2017 (UTC)
I'm afraid you're simply not correct in your interpretation of how folks have understood secondary sourcing throughout the years. I'm not bothered by this, but confused as to how you haven't understood that others see things differently. I get what you're saying, I disagree, but the only thing that concerns me--no, not even really concerns, just puzzles me--is that you believe you have a longstanding universal consensus behind you. You don't. Most people don't care one way or the other, and some of us who do care, disagree with your interpretation. Jclemens (talk) 00:38, 25 September 2017 (UTC)
Thing is, what you have is "me and some other film/fiction editors don't agree with you". What I have is policy agrees with me, and source usage (and rejection of incorrect source usage) on every other topic on the system agrees with me, and even some of the other film/fiction editors agree with me. Any rational outside observer who was not vested and involved in Wikipedia's internal policy wrestling would know which way to bet. The long version of that bet: that some film (and other fiction) editors haven't quite been doing a particular kind of sourcing properly, due to a very slight misunderstanding (due in turn to unclear policy wording, plus the insularity of wikiprojects and their tendency to PoV fork their own WP:PROJPAGE "rules" until the community gets around to correcting them, plus the bad habit of using farcically over-broad language when talking about the subject, e.g. "reviews are secondary sources for films" when the truth is "reviews are sometimes secondary sources for particular facts about films"). The not-quite-right sourcing usually just slides by because the V/NOR/NPOV "enforcers" pay no attention to these articles, most of the facts at issue are pretty trivial, other and better sources are often added later at the articles where the topic is of high notability, and we all have bigger fish to fry on this project, like BLP cleanup and disruptive parties' intensive PoV pushing on socio-political topics.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  05:48, 25 September 2017 (UTC)
If policy is truly on your side, convince me that I'm wrong, arguing from the text of policy and from that foundation explaining your interpretation thereof. Again, I've been in a thousand (actually, that's probably not an exaggeration) policy disagreements on Wikipedia, and an editor who is so thoroughly wedded to their differing interpretation that they believe everyone else has always agreed with them (well, within reason, obviously. Everyone never agrees upon anything in Wikipedia) is rare indeed. If you went through from the ground up, we might find more commonality then you simply articulating conclusions based on assumptions we don't share. Jclemens (talk) 00:39, 26 September 2017 (UTC)
Comment – Is this really being proposed for MOS:FILM? This level of detail would likely result in instruction creep and the discussion might be better suited for WT:NOR or WT:IRS before attempting to make any significant changes here. Besides, the editors watching those talk pages are likely to have a lot more experience dealing with matters like this one (and probably wouldn't favor an extensive interpretation of WP:IRS in a MOS style guideline). Just sayin'. --GoneIn60 (talk) 07:41, 24 September 2017 (UTC)
Please read the proposal carefully. It's being proposed as a clarification that will live somewhere (possibly at NOR or RS) and which will have a WP:SUMMARY-style pointer to it from MOSFILM and other fiction-, art-, work-related MOSes, as a reminder. This is why it's not written in film-specific wording. It's being discussed here because this is where people have been discussing it, in a long-running thread immediately above the proposal. We have a good pool of editorial eyeballs on the issue right here, right now, and WP:CONSENSUS can form anywhere. This is also discussion-draft language. It's not like a block of material is going to be inserted into policy without additional discussion and revision. :-) The talk pages of NOR, V, and RS have already been notified of this discussion and asked for input.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  10:44, 24 September 2017 (UTC)
I was going to try to stay out of this, but I'm with User:Jclemens, User:Masem, User:GoneIn60, and User:Erik on this one. They make the most sense to me. Huggums537 (talk) 22:07, 25 September 2017 (UTC)
@Huggums537: I don't understand: Erik posted in this thread once, 95 minutes after it was opened, while (apart from the above) you posted three times, all after Erik. How is that "staying out of this" but agreeing with Erik? Anyway, you and those agreeing with you that "film reviews are secondary sources, full stop" are in the clear minority here -- virtually everyone on Wikipedia agrees with SMcC (in essentially saying that context matters), and has done so for more than a decade. Hijiri 88 (聖やや) 08:50, 26 September 2017 (UTC)
I agree that context matters. My problem is that a film review is not being called a secondary source in context of writing about a film. In covering a film's overall reception, an individual review would be a primary source. You would need an analysis of a set of film reviews as a secondary source to write about the overall reception. WP:NOR#Notes #3 says that examples of primary sources include "fictional works such as poems, scripts, screenplays, novels, motion pictures, videos and television programs". It says nothing about reviews being primary sources. This linked under that note says about secondary sources, "Secondary sources describe, discuss, interpret, comment upon, analyze, evaluate, summarize, and process primary sources. Secondary source materials can be articles in newspapers or popular magazines, book or movie reviews, or articles found in scholarly journals that discuss or evaluate someone else's original research." Erik (talk | contrib) (ping me) 15:03, 26 September 2017 (UTC)
Oh, we do agree that individual reviews are primary sources for a film's overall reception. See, my problem with how we use them has nothing to do with WP:YESPOV, but rather that a lot of our film articles (and, admittedly to a much greater extent, TV episode articles) contain in their "reception" sections nothing but citations of individual reviews and the (oftentimes skewed, due to the flawed binary approach to assessing an individual critic's reaction) RT score, which is in violation of NOR's proscription of basing large sections of our articles on primary sources: most of our readers will assume -- correctly -- that a section labelled "Critical response" in an article on this or that film is meant to address the overall reception, for which (we agree) individual reviews are primary sources.
WP:NOR#Notes is not meant to give a definitive list of all sources that, regardless of context, "are primary", and your citing it as you do appears to undermine your claim to agreeing that context matters. If it wasn't for WP:CREEP, and the incredible power that it has over the community, I would not disapprove of adding to that NOR footnote the clarification that Any source can be primary depending on the context, for example as a source of information on itself. Film reviews, as opinion pieces that are almost always used on Wikipedia as primary sources for either their authors' opinions or for a film's overall critical reception, are a good example., and I can guarantee you that the majority of the community would agree with me on the substance, but just wouldn't want that clarification added because of CREEP (and this thread not being sufficient evidence of this being an ongoing problem to warrant the clarification). This exact thing happened in May 2016 the last time I tried to significantly edit WP:V: virtually everyone agreed with me on the substance (which amounted to "If the cited source doesn't verify the content, then the content is unsourced"), but the three or four examples I gave of my clarification not being explicitly enshrined in the policy having led to disruption were not seen as sufficient to warrant the change.
Hijiri 88 (聖やや) 00:21, 27 September 2017 (UTC)
Stating a film's overall reception is distinct from sampling reviews writing about a film. One can easily have a critical reception section for which there is no aggregate score or overall summary, meaning that we sample whatever reviews are available (with no claim about overall reception). These are secondary sources in regard to the film, the primary source. That does not abruptly change when an aggregate score or overall summary is actually available. To think of it another way, the film reviews address the main topic, the film itself. The score or summary address the sub-topic, the film's overall reception. As previously stated, film reviews "describe, discuss, interpret, comment upon, analyze, evaluate, summarize, and process" the film. Erik (talk | contrib) (ping me) 00:48, 27 September 2017 (UTC)
@Hijiri: I think I have a better understanding of where you're coming from. From your perspective, most readers will assume the critical response section in a film article is addressing overall reception, and as a result, the individual film reviews listed could be viewed as primary sources. So even if a claim about overall reception isn't being made (as Erik noted above), your argument is that readers will think overall reception is the main focus of that section anyway. I can empathize with that viewpoint, but here's another to consider. WP:NOR bans the practice of basing an entire article on primary sources. However, it only says to "be cautious about basing large passages on them". Also, WP:PRIMARYNOTBAD makes some relevant points as well. So while context matters in determining primary vs secondary, we should also be taking into consideration the quality and reputation of the cited source, as well as the claim it was cited for (often a quote). In most well-written critical response sections, the first paragraph focuses on overall reception, while the subsequent paragraphs list examples of positive and negative reception. When listing an example, you're not necessarily making the claim that it represents overall reception, because otherwise you wouldn't bother listing negative examples for films that were well-received. Because we list examples that DO NOT represent overall reception, it is clear that the focus is shifting throughout that section.
Maybe this helps or doesn't, but I thought I'd talk through it, because I'm not sure we necessarily disagree for the most part. Perhaps we should pick apart an actual film article that has passed GA or FA nomination to help explain where our viewpoints diverge. --GoneIn60 (talk) 07:55, 27 September 2017 (UTC)
As I've admitted to SMcC on his talk page, I don't think this discussion is going anywhere, so I'm happy to just let it die out, but I'm just going to say that I find all the points in the above comment reasonable, except that (and I admit that this is a completely peripheral point) I don't think we have a good standard for what sources have a good "quality and reputation". I was actually going to open a discussion either here or on WT:FILM some weeks ago because I was reading a film article (I think it was one of the GAs) that devoted more than a third of its "critical response" to a quotation from Roeper that didn't seem to be saying anything unique or interesting about the film, and it felt like we were only quoting him because of his loose association with Roger Ebert (whom we, understandably, quote for just about everything). Roeper is a good critic (with a good quality and reputation), but he is not Ebert, and automatically treating him as "the new Ebert" for every article on a film that was released since the latter's death seems somewhat arbitrary (we don't cite whoever teaches Japanese at Harvard for every Japanese literature article just because they share an institution and title with Reischauer, and Roeper didn't so much inherit an academic institution and title from Ebert so much as a TV show and newspaper column). Even if we are not literally doing that for every film, it certainly seemed like we were doing it for that one and quite possibly others. I just wish I could remember the article. I decided not to open the discussion because I didn't want to go to WhatLinksHere and click through dozens of pages trying to find particularly blatant offenders. Hijiri 88 (聖やや) 08:32, 27 September 2017 (UTC)
Ultimately, I think a better solution is to focus less on whether reviews are primary or secondary (though recognize this does matter for avoiding OR), and instead discuss the appropriate ways to write a reception section which 99% of the time is going to be sourced to individual reviews. This should account for the "allowable" synthesis of sources that is necessary for summarizing and organizing a legible/readable summary as a tertiary source, and explain where the line is crossed when that goes into unallowable synthesis that fails NOR. --MASEM (t) 15:42, 27 September 2017 (UTC)
Agreed. I don't think anyone here really understood what I was getting at with the proposed wording (it had nothing in particular to do with films, or with assessing an overall critical reception; it's way more meta than that). But that's what I get for going "rogue meta" on the heels of a discussion like the sprawl above. There's also still an ongoing confusion about "secondary" here. A secondary source is one that is doing WP:AEIS stuff from material from other sources, about topic X. So, the above claims that a review is a secondary source for the review writer's personal analysis of a film (e.g., "A freudian romp through the post-Obama American mental landscape – tinged with dark humor, side references to The Purge, parody of Scientology and other cults of personality, self-references to the show itself – the series is fun but ultimately has the plausibility level of a comic book") is patent nonsense. It's entirely primary for all of that. It's secondary for material that involved other sources (as is often the case with magazine reviews and others written on a long enough time-scale, compared to those in daily newspapers, that they draw on previously published critical response), and in academic analyses in film journals and such (which draw on prior work even more so, and cite it explicitly). A review is also secondary for basic plot points, like which character did what, since in that case the work itself is a primary source for what it contains, and the review is selectively summarizing them for us. The cognitive dissonance is failing to understand that the film or other work is only a source, at all, of any kind, for its own content [e.g. plot points]; it cannot be treated as "the primary source" that makes the review "secondary" for anything of any other kind – only what is objectively factual about the film's content (including the credits). This becomes much clearer when you place "film" with something simpler, like a piece of abstract art that consists of nothing but a big red dot on a black canvas.

But it's become clear to me that people laboring under the "reviews are always secondary for everything they say about a film because the film is the primary source" fallacy are deeply buried in that fallacy and will not be pulled out of it in the course of this discussion. By all means, "instead discuss the appropriate ways to write a reception section which 99% of the time is going to be sourced to individual reviews." That's a small target and an easier task. Baby steps, baby steps ....
 — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  04:53, 28 September 2017 (UTC)

Yes, there is obviously a misunderstanding of the context in some cases. However, it is important to acknowledge that regardless if a source is primary or secondary, we just need to make sure we are avoiding WP:SYNTH violations and utilizing high-quality sources. As long as that is being accomplished, then the P v S debate can take a back seat. Further clarification at WP:PSTS is certainly warranted if the consensus agrees with the interpretation you've laid out. I think you've made some valid points here. --GoneIn60 (talk) 11:27, 28 September 2017 (UTC)
@GoneIn60: [As long as] we just [...] make sure we are avoiding WP:SYNTH violations [...] then the P v S debate can take a back seat Look at the thread above this one (from which I believe this one spun out). There is significant opposition to replacing a vague and clearly widely misunderstood use of the phrase "secondary sources" with a proscription of SYNTH. I have not been following recent developments there, but at least one user explicitly opposed the clarification with a CREEP-esque rationale here, and it's self-evident that after three weeks the change still has not been implemented. Hijiri 88 (聖やや) 12:09, 28 September 2017 (UTC)
Given what is presently at MOS/FILM, the right approach would be to create a wholly separate guideline or essay that applies to any type of creative work (film, book, music, video game, art, etc.) to describe how to construct non-SYNTH reception sections. In that, as a justification, we can get into details of primary/secondary, but keeping that the ultimate point is that having a reception section is a key element for notability of a work and reflected some type of secondary sourcing for the work, but writing some conclusions or analysis of those reviews may bring SYNTH and there are steps to avoid that. Then we can stress that the highest-and-best sources for such sections are academics and historians in the field that are providing retrospective, then more contemporary reviewers, and so forth. Then there doesn't need to be a change to MOS/FILM here outside of referencing that guideline/essay. --MASEM (t) 12:59, 28 September 2017 (UTC)
Create a separate guideline or essay? Because MOS:FILM already implicitly encourages the violation of one of our core content policies? You've got it entirely the wrong way around. If 80% (heck, even 20%) of readers of MOS:FILM think film reviews "are" secondary sources, then we can't tell people that contrasting a film's initial reception with it's retrospective reputation using "secondary sources" is acceptable. And given that the comment I linked above explicitly said that expanding the text beyond what it says now would be too much detail, creating a whole separate page would no doubt encounter even more backlash. (This assumes that said commenter would not support a proposal solely because I wasn't the one who proposed it, although their recent actions on a certain article talk page may indicate that this assumption is wrong.) Hijiri 88 (聖やや) 21:42, 28 September 2017 (UTC)

No, it's not as "simple" as that. A separate guideline or essay that applies to covering any creative work (not just film) would be beyond the scope of just the FILM project. Second, this guideline has to establish (as most are in agreement here) that a review is a secondary work for the film/book/whatnot, but a primary source on the film's reception, and the last thing we want to do is synthesis off primary sources in describing the film's reception. And thus to that point, there are certain types of synthesis that are allowed as part of standard summarizing, paraphrasing, and cohesive article writing, and there are certain types that are not. I would totally characterize the act of saying the film's reception over time changed just because later reviews were lower/higher than the early is a SYNTH violation. On the other hand, if I have the ten "more respected" reviews, and 6 of those spoke highly of an actor's role and none of the other reviews mentioned it, then there is allowed synthesis in saying "Reviewers praised (actor)'s performance." followed by appropriate statements or quotes. There are lines here in terms of allowed synthesis that is readily accepted in practice but not written down aynwhere, and this proposed guideline would serve that function. --MASEM (t) 21:52, 28 September 2017 (UTC)

I kind of agree with Hijiri here in a way because even though an editor might have to perform a certain amount of what you might call "ordinary synthesis" (which I wouldn't call it that either) in order to assemble information to include in an article, it's fundamentally different from the SYNTHESIS forbidden in policy. So, the creation of another MOS guideline or essay on "acceptable synthesis" could be viewed as a conflict that could very well be confusing for editors, and could possibly be abused to undermine the SYNTHESIS policy. The only way I could see supporting such an essay is if it intentionally avoided wording such as; "allowed synthesis". The creation of a guide that instructs well without saying in any way it's ok to synthesize would be preferable to one that outlines how synthesis is allowable. Would that make better sense? Huggums537 (talk) 17:19, 29 September 2017 (UTC)
I wouldn't think in this hypothetical guideline/essay that we would say "you are allowed to break SYNTH to write reception sections". There does need to be an explanation of primary and secondary sources, and that some means of assembling those can lead to SYNTH violations. But that's to provide justification for the guideline/essay, the rest being the actual advice of what are dos and donts of reception sections, providing a type of casebook or guidance to try to put lines in an otherwise blurry area of what's acceptable to do and what's not. For instance, this should specifically call out that one cannot use the fact that there are reviews that just happened to be published a year later that others to describe that the film's reception changed over time. (a point that started this whole thing). We shouldn't call these out (where we do allow for reasonable summation) as "acceptable synthesis" to misled people, just that internally we know that's the case. Instead, just present it as where the SYNTH line is crossed if you don't have secondary sources for the reception of a film. --MASEM (t) 17:35, 29 September 2017 (UTC)
I can see the possible benefit of the "actual advice" you're talking about, and think such a guide could be useful. I'm glad you see my meaning, and I like your idea to present the details with advisories warning editors when SYNTH lines are crossed. Huggums537 (talk) 00:15, 30 September 2017 (UTC)
I don't think creating a separate page is necessary, and it feels like a way of sidestepping the proper way to address the core problem. If editors need a whole separate page telling them what a primary and secondary source is, and don't know not to engage in unambiguous SYNTH (presumably having already read WP:SYNTH) when writing reception sections (and collecting individual reviews and comparing them to each other is unambiguous SYNTH), then they simply should not be writing reception sections. Linking WP:SYNTH would be enough. Trying to explain the difference between primary and secondary sources in different contexts brings us back to square one and my original disagreement with F22R -- if we aren't allowed explain the difference on this page, then the only solution is to link directly to the relevant policy, and the relevant policy is WP:SYNTH, not WP:SECONDARY (which I recognize is on the same page).
BTW, I still haven't gotten around to reading the 2015 RFC, but has anyone else? If so, did it address any of this? It feels like the purpose was to remove the direct invitation to SYNTH that had existed before that point, but by saying "secondary sources" rather than "not individual reviews" they just replaced a direct invitation to SYNTH with an indirect one.
Hijiri 88 (聖やや) 07:53, 30 September 2017 (UTC)
Hijiri could have a point that another manual might possibly run into CREEP concerns. So, I'm unsure about it really. I haven't read the RFC from 2015, but it sounds like they made a small improvement if they went from a direct to an indirect invitation to SYNTH. (The lesser of two evils is better if I just had to pick between the two.) Progress is progress, and there's always room for improvement, I suppose. Huggums537 (talk) 12:31, 30 September 2017 (UTC)
Saying "you are allowed to break SYNTH to write reception sections" is not something we should contemplate. Rather (and regardless where the material ended up) it should distinguish between WP:SYNTHESIS and summarizing what sources say, without confusingly referring to the latter as a form of synthesis, or heads will asplode.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  23:04, 1 October 2017 (UTC)
I'd like to make an addition to this discussion because it's recently come to my attention that anybody can make the claim that a source is a primary source since WP:ALLPRIMARY proves that all sources are primary for something. However, it also states, "More importantly, many high-quality sources contain both primary and secondary material." So, the earlier debate about the sources being primary is a moot point since ALLPRIMARY already proves that all sources are primary. This notion of it's [always] either/or primary/secondary is just ridiculous. Just because you can prove a source is primary doesn't disprove it as also being secondary. Huggums537 (talk) 01:26, 3 November 2017 (UTC)
Seems like kind of a non-argument to me. Few if any participants here are ever laboring under the assumption that any sources is always and forever, in every context, either primary or secondary. Such an assumption is not inherent or latent in this dicussion; it's "understood" that when we say "primary" or "secondary" we mean "contextually, for a particular claim". Otherwise ALLPRIMARY would effectively invalidate any discussion of general guideline treatement of primary and secondary sources.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  04:27, 3 November 2017 (UTC)
I get a different impression about the assumptions of editors. Even if participants in this discussion were not operating with the understanding that primary/secondary means contextually for a particular claim, it still wouldn't allow ALLPRIMARY to invalidate the guideline treatment of sources since there are primary sources that are strictly primary, with no secondary or tertiary characteristics even if all secondary and tertiary sources have some kind of primary characteristics. So, while it would appear at first that ALLPRIMARY could possibly invalidate that distinction, it would be a hasty assessment since the distinction between primary and secondary is still there when primary sources that have no secondary or tertiary characteristics exist. Does that even make any sense? Huggums537 (talk) 07:36, 3 November 2017 (UTC)
Well, sure, I was overgeneralizing a bit.  — SMcCandlish ☏ ¢ >ʌⱷ҅ᴥⱷʌ<  08:50, 3 November 2017 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Manual_of_Style/Film/Archive_15&oldid=1077444252"