Wikipedia talk:Notability (books)/Archive 6

Archive 1

←

Archive 4

Archive 5

Archive 6

Archive 7

Criteria 1 and inheritance

The first point in criteria states: "1.The book has been the subject[1] of two or more non-trivial[2] published works appearing in sources that are independent of the book itself". Now if the author has a previous genuinely award winning published book, or if the author is notable for other causes, lets say social work, singing, politics, etc. then it is quite obvious that the book written by such an author, lets say their biography, would receive these "two or more non-trivial" reviews. For authors notable in other popular cultures like films, TV and other mass-media, their books receiving reviews is quite natural howsoever lame the book might be or trashy the review be. Similarly, history of quality-books is bound to get reviews for new books. Such cases of inherited notability should be mentioned in the guideline. Currently the guideline says "at least one of the following criteria" should be met and this first criteria is damn easy to meet in case of inherited notability. §§Dharmadhyaksha§§ {Talk / Edits} 05:08, 3 November 2015 (UTC)

Unanswered since long I would like to ping some editors who have substantially edited this guideline page. @James500, Fuhghettaboutit, and Tokyogirl79:. §§Dharmadhyaksha§§ {Talk / Edits} 09:45, 23 November 2015 (UTC)

• Strongest possible oppose. This guideline is already far too restrictive. In view of our editor retention problems (because deletionism is killing the project by driving editors away), no further restrictions whatsoever should be introduced under any circumstances. This particular proposal is ludicrous. There is no such thing as an 'inherited' book review. 'Inherited notability' would be arguing that a book about, eg, tigers, which received no reviews, is notable because its subject matter (ie tigers) is notable, because Tiger is a valid article. Also, if a book is famous (or popular or influential), which it will be if it has prominent reviews, it does not matter if it is un-meritorious. What we should be doing here is reducing the number of book reviews required from two (which is pointless) to one. James500 (talk) 10:24, 23 November 2015 (UTC)

I don't understand what you are opposing if there is no proposal for change made. Anyways... I understand the gist of what you are saying but I guess you have misunderstood me. Lets take an example; a writer wins a Man Booker for some work and that would make his all books in future (and past also) to be reviewed. If the criteria is further reduced to one-review; which is ridiculous, you are actually putting many authors in the criteria 5 as "historically significant". And there are inherited book reviews in the sense that when pop icons start writing books or publish their biographies their books get so-called reviews; not because the writer is notable for writing but is notable for other non-literary stuff. That's what am calling here as "inherited notability"; as n notability inherited by writers from other non-literary works. §§Dharmadhyaksha§§ {Talk / Edits} 10:48, 23 November 2015 (UTC)

I don't understand what you are talking about. Are you proposing to modify criteria 1 so that it will not accept what you are calling 'inherited' reviews? James500 (talk) 11:14, 23 November 2015 (UTC)

Am not proposing anything yet, just discussing this issue. §§Dharmadhyaksha§§ {Talk / Edits} 11:19, 23 November 2015 (UTC)

I am not sure why we can't have a Wikipage on every book, just like we have a Wikipage on every film. Anybody that reads the book might want to find out what information exists about the book. - Kautilya3 (talk) 11:33, 23 November 2015 (UTC)

We don't have a wikipage on every film although the criteria there too is quite lenient and a topic all together different for a different venue. For example we don't always have articles about all A certified films or the "B Grade" films. Going by our statement every issue of Grihshobha will have separate article and I know you aren't proposing to do that. §§Dharmadhyaksha§§ {Talk / Edits} 11:54, 23 November 2015 (UTC)

We do have some television series where each episode or each season has a separate article. It looks like we are a liberal house, and the days of scholarly elitism are gone. Pretty much every book published will have thousands of readers. At least some of them might be interested to find out what third party information exists about the book. If we have an article that provides summary of what the reliable sources say, it serves a useful purpose. How much load this will put on the conscientious editors that care about the quality of the content is another matter. - Kautilya3 (talk) 14:00, 23 November 2015 (UTC)

In principle, the idea of a wikipage for every book ever written is seductive, a sort of virtual 'library of Alexandria' catalog. But, listing every possible thing (whether a book or something else) flies against the basic principle of this encyclopedia - that only notable works are included and that notability derives from independent sources. With books, the main problem is that there are many self published ones that sit unread on the shelves of family and friends (I have a few of those sitting on my self right now) with no external validity whatsoever. Authors then try to use Wikipedia articles as a device to promote themselves and their work and, in practical terms, cleaning up these articles takes time and is both frustrating as well as demotivating for the editors involved in the clean up. I'm not sure making a special notability exception for books is a workable idea. Re Dharmadhyaksha's question, I don't see why 'inherited notability' is an issue. A book doesn't have to be good to be notable, it just needs to be notable. If that notability is derived from the subject matter, then so be it. --regentspark (comment) 14:28, 23 November 2015 (UTC)

According notability to all books might be unworkable because of self publishing and vanity presses. On the other hand, it has been suggested that we should, in particular, accept as notable: all books with at least one review, all bestsellers, all books written/published before some date (suggested dates have included 1450 (introduction of printing), 1750, 1900 and 1923 (public domain in US)), all books by a notable author, all books published by a 'respectable' publisher, or at least the notable ones or the very big ones (Penguin Books is an example that comes to mind). We can also look at the number of editions, translations, library holdings, etc. None of these criteria are obviously out of the question. Notability is not the basic principle of Wikipedia, it is only a guideline, invented around 2006 or 2007. The core policies are V, NOR and NPOV. Non-notability is not an organic threat to those. James500 (talk) 02:25, 24 November 2015 (UTC)

• I'm not entirely sure what you're asking here. Are you asking for books to inherit notability from their author, if the author has published multiple notable works? Or are you asking for us to not use reviews as a sign of notability? I need you to explicitly state what it is that you want to change. I think that what you're arguing is this:

That if an author is particularly well known for whatever reason (popular author, award winner, celebrity, etc), then they will be more likely to gain coverage for a work because of their name value alone. Because of this, criteria should be made more strict in order for the site to be more exclusive and only cover works that are particularly noteworthy.

If that's the case, then I have to say that I do not support this. Reviews are an excellent way of showing notability for a work, especially as they're not always a guarantee. An author can be a household name, yet not gain enough coverage in RS for their individual works to pass notability guidelines. I can think of several authors who are extremely well known in their genres, yet their books never gained the needed amount of coverage to warrant an entry. Heck, I don't think that any of Snooki's books have articles and that most of them probably never will pass NBOOK, yet she's undeniably well known. My point here is that celebrity of any type is not a guarantee of book reviews. It can make it more likely, but it's not a guarantee.

Do I think that the guidelines are too loose? Maybe. Sometimes, when we have some decidedly non-notable works squeak in through the cracks. However at the same time, these guidelines also make it possible for a lot of books that should be notable pass as well. Making things more strict isn't really going to solve anything at this point in time because the book guidelines are a little strict as it is and there are a lot of books that should have articles but don't. The same guidelines that allow us to have books on seemingly fluffy, trite books also allow us to have books on more serious, academic topics as well. You also have to take into consideration that there are a lot of books that are extremely commonly used or well respected in their respective areas, yet they receive very little coverage because they're so niche. Tokyogirl79 (｡◕‿◕｡) 15:28, 23 November 2015 (UTC)

• Basically, notability for books isn't something that will be fixed simply by making guidelines more strict. Tokyogirl79 (｡◕‿◕｡) 15:28, 23 November 2015 (UTC)

• @Kautilya3: Just the way you are complaining here that every damn episode has an article, someone at other wikiprojects might be talking similar stuff about every damn silly book being included here.
  @RegentsPark and Tokyogirl79: My source of deriving notability was not from subject matter like a Tiger but from the author's other notable endeavors. Someday Rakhi Sawant and Sunny Leone are gonna write books on Indian Politics and Role model women of India and these are going to receive "reviews" from tabloids and page 3 columnists and we are going to end up keeping these articles here. Kim Kardashian seems to have published a book called Selfish that has her selfies and it has been "reviewed" making it notable for a stand-alone article.[1][2] Check what The Guardian has to say about the business of celebrity publishings. I guess you have now understood my concern with these examples, which I should have given before.
  Anyways... If altering the clause 1 would hinder in writing about books which Tokyogirl talks about, with academic topics, then it should not be altered. But something should also be added in the criteria which would hinder writing about books that I am talking about. §§Dharmadhyaksha§§ {Talk / Edits} 04:32, 24 November 2015 (UTC)

• While I hate to say this since like many, I don't particularly like Kim Kardashian, if her book has received enough coverage to warrant an article then it should have an article. (Of course if no one writes the article then that becomes a bit of a moot point.) We shouldn't pick and choose what gets an article based on the merits of the author or the lack thereof, as this runs the risk of coming across as censorship to incoming editors. Initially selecting what should or shouldn't warrant coverage in Wikipedia may seem like a no brainer, but then the problem here is that choices like this are highly subjective. What may seem like a throwaway book by a sensationalistic person might actually be important or noteworthy to someone else. Not only that, but who makes these decisions? One person? Many? Saying that we should judge notability for a book based on the worth and infamy of the person (which seems to be ultimately what this is about at its core) could set a very, very dangerous precedent. With this in mind, someone could easily argue that we shouldn't cover any more of X's books because their work is overly marketed tripe or that so-and-so is just looking for attention, so of course this book will be covered. That's a situation that can easily lend itself to bias and censorship - two things that can be poison to an encyclopedia known for covering a wide variety of subjects - especially since Wikipedia does not have the limits that other encyclopedias have. Bluntly put, it's not a good idea to create a policy that calls for us to disregard reviews if someone is a well known public persona, regardless of how that person gets that media attention. Tokyogirl79 (｡◕‿◕｡) 04:53, 24 November 2015 (UTC)
• That's not to say that I endorse us having a page for everything. If someone wrote something and it is already adequately covered in the author's article (like say, an autobiography) then there can be a good argument for not having a separate article. However even then that is something that should be done on a case-by-case basis and with common sense rather than more guidelines, as someone could easily argue that any biography or autobiography is fair game for being redirected into the main article. Tokyogirl79 (｡◕‿◕｡) 04:55, 24 November 2015 (UTC)

If common sense was so common we wouldn't need many guidelines. Our guideline here says that with 3 reviews a book's article is completely secured to not be merged or deleted and gains it's absolute notability and independent article space. Any case-by-case basis arguments in such merger/deletion discussions are instantly refuted by showing 3 reviews from gossip columns. And that's what I am hoping to get this changed somehow. I don't know how this can be done. Maybe the 3 reviews criteria can be moved out from the absolute-notable-guideline to a may-be-notable-guideline. Or maybe some other way can be worked out. §§Dharmadhyaksha§§ {Talk / Edits} 05:27, 24 November 2015 (UTC)

• However I again have to argue: if something has received enough sourcing to warrant an article then the article should remain. If something goes to AfD or is proposed for a merge/redirect then we could make an argument based on the depth of coverage in the sources, such as the sourcing only announcing that a book will release or that it exists. If the sourcing is something like book reviews and/or articles that go into depth about the work, then those would show notability. Basically, in these situations you'd have to argue that the author's article (or a series article) already substantially covers the work in question and that an extra article would be unnecessary, but this is difficult to argue in many situations unless there's really only enough notability for one single article rather than multiple small articles. This is something that's very, very hard to write effective rules for since it's just so easy for them to be misused because notability is something that's fairly subjective. Tokyogirl79 (｡◕‿◕｡) 05:39, 24 November 2015 (UTC)

• Well then maybe we should add "in-depth reviews" to our criteria. It does say "non-trivial" reviews but the footnote on non-trivial actually explains non-RS and non-independent and non-editorial-based-publications. It does not highlight a need for in-depth review. Also if we are adding in-depth as a clause criteria we would need to state that not all 3 reviews should be in-depth. Of course, addition of such a vague and volatile term of "in-depth" would also require a big footnote of explanation. §§Dharmadhyaksha§§ {Talk / Edits} 10:35, 24 November 2015 (UTC)

• The two non-trivial reliable sources criterion is a site-wide criterion for reliability. There is no way we can change it here. On the other hand, I think the footnote for "nontrivial" is not doing much. It is explaining reliability but not nontriviality. It should state things like the source should directly address and discuss the book, and not merely mention it. I am sure I have seen such explanations of notability in other places. - Kautilya3 (talk) 10:55, 24 November 2015 (UTC)

No, no, no, that's not right Dharmadhyaksha and Kautilya3. The guideline already contains a restriction relating to depth of coverage. Footnote 1 says "The "subject" of a work means non-trivial treatment and excludes mere mention of the book, its author or of its publication, price listings and other nonsubstantive detail treatment." So there is no need for an extra criteria relating to depth of coverage. It may be that the separate notion of a "non trivial work" (footnote 2) is confusing because it uses the same name for a different thing, but note that footnote 2 is an exclusive definition, and anyway the most that could be required is cosmetic changes to terminology. No one doubts that "non trivial treatment" in footnote 1 excludes coverage so wafer thin as to be useless. That said, bearing in mind the amount of space that a plot summary, and other straightforward bibliographic information, that can be sourced from the book itself, takes up, we don't need a huge amount of coverage in reviews to write a fairly detailed article. We could, in truth, write a very lengthy article using a book as a primary source for itself (and the preface of a textbook often looks a bit like a review and tends to be quoted by reviewers) though there are technical rules against that. I'm not opposed to all 'brief' coverage: inclusion of a book in a sufficiently important and selective bibliography should certainly suffice by analogy to the 'selective database' criteria of NJOURNALS. James500 (talk) 11:53, 25 November 2015 (UTC)

As an example of "wafer thin" coverage is "Books received" or "Recently published" or "Recommended library purchase" lists that some journals and newspapers have. While that's way more coverage than most books ever get, it's utterly useless for content. Like James500, I would not be opposed to allowing some of this.

Meanwhile, the last thing we want to do is to second-guess what reviewers had in mind. A book gets reviewed because the author is a celebrity? Or because the book received a blitz ad campaign? Or because the reviewer owes the author a favor? Or maybe, sometimes it happens, because the book is actually worth reading? None of our business, ever, unless the story behind the story makes it into RS, at which point this detail simply becomes more information and more coverage (and we don't negate the "poisoned" reviews).

I would also favor liberalizing in favor of some inherited notability, something like all non-self-published books of an author are notable once three, say, pass the current level. Or a wee bit tighter, require six current criteria notable books, but for an author with three to five, lower the standard to one non-trivial review. Or some game like this, but the simpler the better. I think it ludicrous that we can't include most of the works of Catherine Cookson, one of the best-selling novelists of all time (100 million), or any of the works of Nicholas Seare, a very obscure pen name of an incredibly famous best-selling novelist. Choor monster (talk) 16:13, 25 November 2015 (UTC)

• Trying to categorize what is in-depth in a review is pretty difficult because the definition of this can greatly differ depending on the publication. A review in some place like Booklist will seem short to many, however for its type of publication the review is pretty lengthy. Saying that we shouldn't use trade or short reviews isn't entirely a good solution either, since these publications do not review every book out there and some of them are very highly thought of like Booklist (the official publication of the ALA), School Library Journal /Library Journal (LJ was created by the guy who made the Dewey Decimal System), and the Horn Book Magazine (the oldest magazine of its type in the US). Heck, the content from the Bulletin of the Center for Children's Books could be considered fairly brief and that's a fairly well respected academic journal. Basically, it can be difficult to really argue what is or isn't in-depth, which is why some of the shorter reviews are still considered usable. Of course in many instances it depends on where something is published, as a short review in a newspaper will rarely hold the same weight as a short review in the SLJ. Tokyogirl79 (｡◕‿◕｡) 04:40, 26 November 2015 (UTC)

My idea of 'significant coverage' includes any decent sized paragraph or passage, so I suspect I would accept anything that could be reasonably be characterised as a review rather than a mere listing. My idea of a trivial mention is an entry in a telephone directory. In view of the relative difficulty of merging works that are anonymous or have multiple authors, and the fact that notability does not preclude merger, I am inclined to keep the minimum level of coverage as low as possible. James500 (talk) 11:06, 26 November 2015 (UTC)

Footnote 4 formulation

I proposed a slight rewording of footnote 4, as it can be misinterpreted that the listed sources are considered unreliable across the board, which contradicts WP:RS allowing inline citations "to e-commerce pages such as that of a book on a bookseller's page" under certain strict conditions (see WP:RS#E-commerce sources). proposed update

This resulted from a discussion at WP:RSN#Merchant sources as a source. In order to avoid fragmentation of the discussion, please contribute to the discussion there, not here. --Francis Schonken (talk) 08:50, 14 December 2015 (UTC)

Add something to the self-published section

Hi guys! I wanted to add something to the self-published section. I basically wanted to add something like as follows to the end of the section to make it a little more clear for incoming users.

"In short, books are not inherently notable for existing and self-published books are very unlikely to gain the coverage necessary to pass notability guidelines."

The section is somewhat clear already, but I think that putting it in slightly clearer language would probably help out a bit when pointing this out to people. There's always some confusion on this part and this would make it just a bit easier. I don't mean this to come across as a slight to self-published authors, but not everyone that tries to add their book is particularly savvy with the English language (for various reasons like it's a second language, they're particularly young, etc) and they might not otherwise really get the gist of the phrase "Self-publication and/or publication by a vanity press do not correlate with notability." Writing it like this would just help clear it up. Of course we can't guarantee that they will read this section, but at least this way we can ensure that it's as clear as possible.

Also, I was thinking that we could probably justify having a redirect to this specific section, given that we almost always have someone trying to add their self-published book to Wikipedia. Maybe we could have it as WP:SELFPUB-NBOOK and then have a note about this at WP:SELFPUB? Tokyogirl79 (｡◕‿◕｡) 04:02, 11 May 2016 (UTC)

• Support, excellent addition to make it clearer (possibly with an apostrophe after the word "existing"?). Coolabahapple (talk) 05:11, 13 May 2016 (UTC)

Speedy criteria for books

I've opened up a discussion about a speedy criteria for books here. The basic gist of this speedy criteria is that it would only cover works that were self-published, issued through a vanity press, fanfiction, or were released as a podcast novel post 2000. Works where there is a credible assertion of notability (akin to how A7 works) and where the author has an article would not qualify under this criteria (akin to how A9 works). I figured that this would help keep us from having to bring myriad obviously nn works to AfD or PROD, where it'd eat up more time than what really needs to be given. It's just so easy to release works via these venues nowadays and we've seen an increase of people adding works of this nature to Wikipedia that it seems reasonable to open up discussion on this again.

The main discussion will happen on the speedy page, so I hope to see some of you guys there. Tokyogirl79 (｡◕‿◕｡) 06:45, 12 August 2016 (UTC)

Audible bestseller lists as a sign of notability?

I was wondering if we can or should add Audible bestseller lists to the list of bestseller lists that would give notability per this guideline. This would be the only e-commerce site that would be allowed to exist under this criteria and would be specifically highlighted as such.

The reason it could be added is that notable media outlets reprint these lists, specifically the New York Times, SF Gate, and Star Tribune. I'm aware that some of this is Associated Press type stuff, but I think that this is one instance where the list being placed in so many newspapers would be a good sign that it's one of the more important lists. You also have lists like this one, which lists the top audio books, where the list is entirely taken from Audible. You also get places like Bustle and The Bookseller listing the site's 'best of' type lists for the year.

If we have it, the guideline will be changed to something like this:

Bestseller lists in retailer or e-commerce sources like Amazon or self-published sources like personal websites, blogs, bulletin boards, wikis, and similar media are not considered reliable. The only exception is Audible.com, whose bestseller lists are republished by multiple notable and reliable media outlets.

What do you guys think? This could open up the playing field for some of the books that might not get heavy sales in print or e-book formats, but sell well enough (or are at least downloaded enough) in audiobook format to be included in a bestseller list. The only thing I can say would make it slightly difficult is that the company is owned by Amazon, which could cause some to argue that Amazon bestseller lists should be included. I'm not too worried about that, as the Amazon lists aren't republished even nearly to the degree the Audible lists are and they're so loosely defined and so dynamic that it would be easy to say why they wouldn't be included. Tokyogirl79 (｡◕‿◕｡) 14:44, 4 January 2017 (UTC)

Discussion at Wikipedia:Articles_for_deletion/Growth_Hacker_Marketing

You are invited to join the discussion at Wikipedia:Articles_for_deletion/Growth_Hacker_Marketing. It may involve interpretation of NBOOK#4. Tigraan^{Click here to contact me} 17:12, 10 July 2017 (UTC)

RfC to amend this and related guidelines

Wikipedia:Village_pump_(proposals)#RfC:_Amending_WP:NMEDIA_and_related_guidelines_to_accord_with_WP:PSCI.2FWP:NFRINGE -- Jytdog (talk) 21:22, 18 September 2017 (UTC)

Library Journal and CHOICE

Library Journal and CHOICE run cursory "reviews" of books—a short paragraph, a few sentences. I'd call them blurbs. I've seen AfDs where they weren't "counted" towards notability since they weren't full reviews or really any more than a summary retread plus, hopefully, a sentence of analysis and recommendation. Do reviews from these sources count as "non-trivial" works for NBOOKS#1 or not? czar 15:57, 15 November 2016 (UTC)

• I'm kind of halfhearted on this. On one hand, they are pretty short and they're not really what I prefer for sourcing for articles. On the other hand, they do give opinions on books and they don't review everything that they're sent - they can be pretty discerning. For example, I've seen NYT bestsellers (main list, not the side lists) not get reviews from the LJ or other trade journals. I've asked similar questions here before and the general consensus tended to lean towards short reviews of this nature being usable since they're still technically reviews and the outlets themselves do have editorial oversight and are generally well respected, with the exception of Kirkus Reviews. There's also the potential of what not using these outlets could eliminate, such as Booklist, which is run by the ALA. Like I said, I'm pretty halfhearted on this. Tokyogirl79 (｡◕‿◕｡) 06:17, 17 November 2016 (UTC)

I've used them both as sources in articles before, and I have no issue with that, but I am asking about whether they "count" towards significant coverage (GNG). For instance, would blurbs in LJ and CHOICE and another similarly small paragraph in another journal together count as sig cov? Because I see no way we can write a detailed article on the subject with that as the sole source material. czar 18:58, 17 November 2016 (UTC)

• I'd actually include Publishers Weekly and Booklist with these two as well, though they tend to have a little more meat than LJ and CHOICE. Each of these periodicals tears through thousands of books a year as a matter for librarians and other collectors to decide what is worth purchasing, not necessarily with discretion for only including the most important books of the year. When I look through Book Review Digest & Index and think about which can have WP pages—it doesn't make sense for us to support WP articles about books that have only received coverage in two or even three of these four sources, and no outside coverage. We can't do the subject justice with that little material. czar 21:56, 18 November 2016 (UTC) For posterity, wanted to add that they are all trade publications: Booklist, Choice, LJ, PW, and with the possible exception of PW, they are usually less than a paragraph and indiscriminate in terms of the choice to review, hence why they shouldn't be used as a sign of significant coverage. PW/Kirkus are a bit longer than the others, but I still look to three major reviews in other reliable publications as an indicator that there is enough reliable, secondary source content to do encyclopedic justice to the topic. czar 19:10, 10 July 2017 (UTC)
• Perhaps I'm prejudiced about CHOICE, because I was a reviewer there for over 10 years, and probably reviewed 100 items. Choice is by design a selective review journal;the reviews are written by faculty experts (or librarian experts in the case of reference books), and the intention is to include only those books that are recommended for undergraduate libraries, or that are so strong advised against that a review is thought necessary. The reviews are intended to be critical, not just descriptive, and I would usually spend an inordinate amount of time on mine, trying to get it exactly right and to include as much as possible within the word limit. I've almost never reviewed for LJ-- Choice reviewers generally do not, because the intent is to avoid any one person having too much influence by reviewing in both places. I did write a few reviews there for material not covered by Choice, but Ifoundtheir format too brief to be appealing. But the intention in LJ is similar, to list only recommended books for public libraries. They both fulfill their purposes--most small academic libraries rely very heavily on Chocie reviews, and many public libraries on LJ (larger libraries will usually try to buy a book of potential interest before it is published and reviewed, because readers are likely to ask for it as soon as it is published or they see a review.)

I consider both usable to show notability, though not by themselves. They're at least as selective as other reviewing media. Choice, at least, is substantial. Another way of looking at it is as a surrogate--there being a review in either, means that there will be enough attention elsewhere to show likely notability.

PW and Booklist have a similar purpose, but their reviews are normally extremely brief, and I do not consider them substantial. When I taught librarianship, I made a point of showing my students the differences. DGG ( talk ) 04:41, 5 November 2017 (UTC)

Sure, they're useful (vital?) for librarians but the question is whether to use them as indicators of wider coverage for notability in a generalist encyclopedia. This discussion is prompted by many cases of individual book article AfDs that claim significant coverage based on reviews solely in CHOICE, PW, Kirkus, LJ, and the like: no additional, in-depth reviews with substance. Those library trade sources are then unreliable indicators and when taken their own are usually barely enough to get a summary of the book, nevertheless enough content with which to write a article on the book that does justice to the topic. No one is advocating barring them from articles (an issue of reputation/reliability), though there are stronger sources for almost all of their short review content, but just not "counting" those sources when considering book notability arguments, based on their meager content for writing an encyclopedia article. czar 16:07, 5 November 2017 (UTC)

they are vital to librarians because they're selective. If they;re selective they show editorial oversight and discrimination. The question is how substantial the content. That's why I say I wouldn't accept Booklist and and PW themselves, but only with other supporting substantial content. A single LJ review I consider also insusficient. A single Choice review I consider often sufficient, but that's because I know the effort htat goes into making the content susbstantial, though the length bried. DGG ( talk ) 22:08, 6 November 2017 (UTC)

BKCRIT#5

Is there a clear standard to evaluate whether an author is so historically significant that any of the author's written works may be considered notable? Based on some new pages I'm curious about P. G. Wodehouse, who seems to be on the borderline here; it's possible he's common subject of academic study but I have no way of knowing. He's not Fyodor Dostoyevsky or Charles Dickens, but he's also not Jonathan Ames or Richard Amory (to pick two random names I've never heard of from Category:20th-century American novelists). power~enwiki (π, ν) 06:23, 1 March 2018 (UTC)

• Similar examples of authors difficult to assess according to this criterion may include George Sand (list of works), Maurice Maeterlinck (list of works), George Bernard Shaw (list of works), G. K. Chesterton (list of works) and Graham Greene (list of works) – or is it more clear cut for any of these? --Francis Schonken (talk) 11:33, 1 March 2018 (UTC)
• Maybe we could add the requirement that the book itself must be available in a critical edition for being automatically eligible under criterion 5. Would make it more clear-cut, and also: a critical edition supposes a "critical", thus somewhat independent, editor's notes which could provide material, and reliable sourcing, for the content of the article on the book even if that particular book is only mentioned in passing in other scholarship on the author. --Francis Schonken (talk) 12:49, 1 March 2018 (UTC)
  • Based on those, perhaps "the author has been the subject of at least two full-length biographies" is a reasonable test here? (with some requirement that the biographies be about them as an author; the hypothetical of a novel by Phineas Gage was raised in 2009, the last time this was discussed). power~enwiki (π, ν) 20:18, 1 March 2018 (UTC)
    • Nah, totally wrong criterion. E.g. this former guru has two separately published biographies: one published in the seventies by devout followers, another, some decades later, by a hired copywriter who specialises in writing biographies of pop icons. A biography written about an author does not signify that that author's oeuvre as a whole is regularly discussed in scholarly literature. The former guru I mentioned above has published a lot: it's not Wikipedia's task to be a platform for promotion of that literature which receives little or no reviews ever, and even less in scholarly literature. A "two biographies" requirement would steer criterion 5 in the wrong direction, I'm afraid. --Francis Schonken (talk) 11:58, 2 March 2018 (UTC)
      • That bibliography you link is an utter disaster of an encyclopedia entry, but your point (that a biography-based test won't work) is valid. I can't access either of Llywrch's suggested reference works, which makes suggesting an explicit criterion difficult. power~enwiki (π, ν) 19:05, 9 March 2018 (UTC)
    • @Power~enwiki: and regarding the criterion I proposed itself? "Oeuvre regularly discussed in scholarly literature" (easy to check via Google Scholar) + "critical edition of the book in question available" is what I proposed (& getting away somewhat of the too bendable "author of exceptional significance" requirement): would you think that would work as a more efficient replacement of the current BKCRIT#5? --Francis Schonken (talk) 20:44, 9 March 2018 (UTC)
      • I think the combination is too strict; "critical edition available" should normally be enough on its own, and I doubt that every single one of Dickens or Dostoyevsky's books have critical editions. power~enwiki (π, ν) 22:34, 12 March 2018 (UTC)
    • This proposal sounds like a terrible idea. (1) I can't see any reason why the commentary needs to be in a critical edition as opposed to some other source, such as, for two random example, the "Studies in English Literature" series edited by David Daiches or "100 Must Read Science Fiction Novels" by Andrews and Rennison. Or in any other source whatsoever. (2) I can't see any evidence that the books we expect to satisfy this criteria always have critical editions. (3) The whole point of SNG is that the media in Britain and America is thought to have strong systematic biases against high-brow topics, the less recent past, poor countries, non-English speaking countries etc, meaning that GNG, or any criteria based on amount of coverage will tend to reproduce those biases. Consider for example the authors who have biographies in the second part of volume 4 of The Penguin Companion to Literature, which deals with oriental and african writers who are supposed to be the most important in those countries, and therefore should satisfy criteria 5 as presently written. This part is marketed as containing information less accessible to Europeans, and much of which, in the case of Africa, is not readily available elsewhere. Do all of those authors have critical editions of all their books published, preferably in the west. Do most of them? Do any of them? I suspect the answer to that question is "no". And if that, or anything like it, is the case, what you propose is out of the question. James500 (talk) 21:19, 10 March 2018 (UTC)
• I'm just passing thru, but I have to ask: is everyone concerned about the issue of "common subject of academic study" aware of the existence of such reference works like The MLA Bibliography & The Year's Work in English Studies? While these probably overlook certain genres (e.g. Romance novels, Science Fiction), one can be confident that if a serious academic study on an author or work has been published one or both of these will include it in their annual publication. Stating that X number of papers in one of these references would be an objective measure of a specific author's or book's notability. And preforming a search for the authors Francis Schonken listed above in those references would be a validating test whether that would be a good measure of notability. (And avoid making the standard for book notability even more complicated.) -- llywrch (talk) 22:08, 1 March 2018 (UTC)
  • Re. The MLA Bibliography (a US publication) & The Year's Work in English Studies (emphasis added): how does that work for Godfried Bomans, Annie M. G. Schmidt, Gerard Reve, Hugo Claus, Louis Paul Boon, Tom Lanoye, Herman Brusselmans, ... ? Would they show up in such lists? And if they don't, how would one know whether they show up in non-English-language scholarship? --Francis Schonken (talk) 12:27, 2 March 2018 (UTC) Two more: Louis Couperus, Willem Elsschot. --Francis Schonken (talk) 19:00, 9 March 2018 (UTC)
    • The MLA Bibliography is international in coverage, & covers periodicals in multiple languages. Of course, the problem is that I expect it to be weak in its coverage of non-Western publications -- e.g. those in Japanese & Chinese. As for Year's Work in English Studies, I mentioned that as an example; I expect that similar annual bibliographies exist for other literary languages. (I'd be very surprised if no such series exist for the major world languages, such as German, French, Italian, etc.) My point is that these resources exist, & that relying on them provides an objective standard that would minimize fights over notability. Which the way the current standard is written reflects happens on a chronic basis. Coming up with a standard like this obviously won't solve all notability problems, but it could end bickering over some of them. -- llywrch (talk) 20:26, 2 March 2018 (UTC)
      • So, could you determine with that source (or both sources) whether the examples of authors given above pass or don't pass BKCRIT#5? Just work out a few examples and let us see on which side of BKCRIT#5 they end. And whether or not there is an advantage in doing the determination with those lists as to compared with doing the determination with Google Scholar (see below): I'd guess it'd be more laborious than doing it with Google scholar anyhow. Why would we choose a more laborious method – or is there some other advantage? I don't suppose Google Scholar would be less objective (for the "across all languages" aspect). --Francis Schonken (talk) 19:00, 9 March 2018 (UTC)
  • I suppose Google Scholar would do the trick much more easily than the two publications cited by Llywrch, and for all languages, easily accessible, without needing to check two volumes for several consecutive years. It shows indeed, for example, that P. G. Wodehouse passes the current version of the 5th criterion with flying colours. At least the "author's life and body of written work would be a common subject of academic study" part... but how does one evaluate "the book's author is of exceptional significance" part, seemingly very prone to eternal discussion over variant interpretations, and sort of circular (why is the book significant? – while the author is significant – why is the author significant? – while his books are significant – repeat ad infinitum)? ...which leaves power~enwiki's question still sort of unanswered I suppose: doesn't the 5th criterion put the bar too low? Hence my proposal to add the "critical edition should be available" requirement. --Francis Schonken (talk) 13:58, 2 March 2018 (UTC)
    • Criteria 5 does not put the bar too low. Furthermore, a book that has gone through a critical edition is ipso facto notable even if it does not satisfy criteria 5. The introduction and notes in a critical edition will satisfy GNG. Furthermore, any book credibly published in one of the 'classics' type series (eg Penguin Classics, Oxford World's Classics, Everyman's Library etc) is notable. We should in fact introduce an additional criteria 6 for these sort of books. James500 (talk) 17:37, 8 March 2018 (UTC)
      • No, "The introduction and notes in a critical edition will satisfy GNG" is incorrect: in a critical edition, the editor's work is a single source: GNG requires multiple independent sources. Also, "credibly published in one of the 'classics' type series" does not prove notability in itself, neither in a GNG logic, nor in a NBOOK logic. Sorry, not helpful. --Francis Schonken (talk) 18:26, 9 March 2018 (UTC)
        • Looking at the List of Penguin Classics, almost all of the non-compilation titles are notable, but it's not a good notability guideline. power~enwiki (π, ν) 19:05, 9 March 2018 (UTC)
• (1) GNG does not require multiple sources in absolute terms. If the introduction and notes in a critical edition are long enough and detailed enough they will satisfy GNG. And they normally are. What GNG actually says is that multiple sources are normally expected on the assumption that notable topics are more likely to receive coverage in multiple smaller sources rather than in one big one. Put another way, GNG assumes that smaller sources are more common than bigger ones. This is an assumption that has never been statistically tested and might actually be wrong. (2) The whole point of a series like the Penguin Classics is that it is supposed to contain only the most important and famous books. Since the series is intended to contain only books that are notable, inclusion in it is an indicator of notability. (3) In practice, books don't get critical editions, or into series like the Penguin classics, unless there is a lot of other coverage available. So even if these things didn't satisfy GNG in of themselves, you could infer that it is a virtual certainty that there will be other coverage that does. (4) All of the 'non-compilation' books in the Penguin Classics are notable. If some of them are not blue linked in the list yet, it is simply a mistake. For example, "Servitude and Grandeur of Arms" by Alfred de Vigny is actually located at Servitude et grandeur militaires, its original French title. The editor who compiled the list of Penguin Classics simply was not aware of this. And we still have a lot of missing notable topics. In fact, the overwhelming vast majority of notable topics (certainly more than 95% of them) are missing. This is due to our editors ignorance of history, ignorance of scholarly literature and where to find it, ignorance of foreign languages, ignorance of any place outside America and Britain, and indeed ignorance of anything other than what is on television right now. James500 (talk) 15:05, 17 March 2018 (UTC)
• (1) Google Scholar is absolutely no good. It appears to have serious systematic bias against pre-twentieth century history and possibly later. There have been cases where important nineteenth century authors have literally dozens of books reviews and many citations, but GScholar does not know about any of these. GScholar gets the h-indexes of certain important nineteenth century scholars badly wrongly underestimated because it is not aware of most of their citations. GScholar seems to either not include a great deal of the scholarly periodical literature of the nineteenth century, or it does not index such literature properly, including some of the most important and famous periodicals. I was once told that GScholar was no good for anything before about 1995. You cannot use it to disprove the notability of certain historical topics. (2) On the subject of Google, their search engine does not work perfectly. If you put in the title of a book into GBooks you often find that many of the book reviews in GBooks don't come up in the search results. You have look up the volume index of each volume of periodicals that is likely to have reviews in order to find most of them. Add to that GBooks' infuriating habit of returning only one result for a particular periodical which may have hundreds of volumes, meaning that you have to search each volume of the periodical separately. James500 (talk) 15:30, 17 March 2018 (UTC)

Clarification of NBOOK

Presently our standards exclude "instruction manuals" from the inherent notability extended by NBOOK. Would there be support for a proposal to clarify that by adding a paranthetical "(including games and gaming rule books)"? I feel like this is in the general spirit of NBOOK's proscription of instruction manuals, but it's occasionally a source of contention. For instance, can Kids on Stage: The Charade Game for Kids [3] qualify under NBOOK's "two review = notability" criteria because it includes a bound rule book in its box? Looking at a different genre of game, can Heroes Unlimited Revised - 1st Edition Rules [4] qualify as notable on the basis of nothing more than a review on rpggamer.net and the "Gaming Gang" podcast? Chetsford (talk) 18:31, 18 August 2018 (UTC)

• Oppose The sort of instruction manuals NBOOK is talking about is the sort that come with a television or stereo system. It is reasonable to exclude such things as they generally do not receive coverage, are not distinct from the consumer electricals they are supplied with and are not entertainment, education or anything that would normally be described as literature. To try to treat games and their rulebooks as analogous would plainly be inappropriate. Something like the Warhammer franchise represents a large body of fantasy and science fiction literature stricto sensu. This is true of their 'rulebooks'. Likewise the books that come with Avalon Hill games contain detailed gazetteers of historical battlefields and detailed chronological accounts of what actually happened during historical battles that are works of history like any other history book. This sort of thing should be treated the same as any other book. James500 (talk) 05:23, 19 August 2018 (UTC) Similarly, the game Chronology really is a history book in the form of a set of cards each with details of an historical event printed on them, and The Revolution Game from learningames mostly consists of a very detailed history book of the places, battles and chronology of the revolution in loose leaf form. To say that these should not be treated in the same way as other books isn't feasible. James500 (talk) 05:34, 19 August 2018 (UTC)

Application of these guidelines

Does NBOOK apply to single panel editorial cartoons like this? Argento Surfer (talk) 15:45, 11 September 2018 (UTC)

Nope. Single panel comic taken from a magazine that has since appeared in books does not make NBOOK relevant at all. Looking forward to the wall of text wikilawyering its justification, though. — Rhododendrites ^talk \\ 16:33, 11 September 2018 (UTC)

• I wouldn't apply NBOOK to it for the same reason that we don't use it to show notability for say, a specific poem or painting. Sometimes being frequently re-published can help show notability for something, but in this situation you'd have to show where there has been some in-depth discussion about the artwork. Being listed in a schoolbook as an example without any sort of explanation as to why the specific artwork is more relevant than another with a similar tone or theme doesn't really accomplish that in my opinion. Something else to take into consideration with the use of this artwork is the publishers - if the same publishers are using the image then that kind of lessens any impact the frequent use could have. In the end though, the main thing that would keep this from using NBOOK's guidelines is the fact that it's not a book. Some of the signs of notability may be similar, but there are just as many that aren't. By NBOOK's guidelines this artwork would actually fail, since none of the book mentions are in-depth and there doesn't seem to be any sort of reviews for the work.

I do think that there's a need for detailed notability guidelines for artwork. Here are some of my criteria for artwork notability:

1. The artwork is held in the permanent collections of a notable museum or national archive.
   1. The implication here is that the museum would be extremely choosy about what it permanently holds, so only a small amount of art would be kept. Also, the museum would be something like the Smithsonian or Guggenheim that's notable itself.
2. The artwork has won a major award in the art world.
   1. IE, the top awards, something notable in its own right
3. The artwork has been considered by reliable sources to have made a significant contribution to a notable or significant motion picture, or other art form, or event or political or religious movement.
   1. Think Mona Lisa, Girl With a Pearl Earring, or the Obama Hope poster. This type of thing is actually par for the course with notability for creatives as a whole, as this is something that is part of both NBOOK and NFILM.
4. The artwork has received in-depth coverage in independent and reliable sources.
   1. Things in this avenue are stuff like reviews, articles about the artwork and its significance, books written about the artwork or cover the artwork heavily.
5. The artist is so overwhelmingly notable and significant that any of their artworks would be considered notable.
   1. Like NFILM and NBOOK, this would apply to less than 1% of artists out there and is more for people like Michelangelo, as the implication here is that since the artist is that notable there will absolutely be coverage out there.

TBH, if I were to apply anything to this I'd say that it would be more along the lines of NFILM. If you do want to start work on a notability guideline for artworks, feel free to use whatever I've written here. ReaderofthePack (｡◕‿◕｡) 23:16, 13 September 2018 (UTC)

• The more I think about this, the more I think that this needs to be a thing. Let me know if you guys are interested in creating this. I'm totally on board with helping to create this guideline, as we can't really judge artwork by the same standards that we would creative works like films and books because there are some pretty major differences. ReaderofthePack (｡◕‿◕｡) 13:45, 14 September 2018 (UTC)
• Certain parts of NBOOK are applicable to any creative work because they are universally applicable general principles. Criteria 4 (taught in schools) is one of them. I wasn't using the school books as evidence of coverage, I was using them as evidence of the curriculum taught in schools. There are other sources for that such as an Ofsted report. We do however need an SNG for artworks, and I would be happy to help. It would need, in addition to the above, to include a provision along the lines of OBK. James500 (talk) 04:03, 15 September 2018 (UTC)

• ... No. Obviously not. It even clarifies as much in the very first section. If part of it can also be said of how we evaluate other kinds of works, it must be in a guideline that applies to other kinds of works. This guideline is about books. Not comic books, not graphic novels, and certainly not single comic strips. — Rhododendrites ^talk \\ 16:10, 11 November 2018 (UTC)

In-universe sourcing and deletion?

Do we delete on the basis of in-universe sourcing?

The particular example I'm thinking of is in relation to one large (multi-book) section of the fictional world of a long series of fictional books. Notability is not really the issue: the series is very obviously notable, the author would evidently meet the "historically significant" test, not merely notability (two published biographies, countless magazine articles) or WP-article-presence and this "slice" of the fictional world is a substantial one within it, well suited to an article.

Yet the article is being pushed towards deletion by an editor. They've just removed the sources and tagged it as "unsourced".

There is indeed a shortage of non in-universe discussion about this topic. There is some (there are two books on the author, which overlap onto this). Yet most of that recognises that this world exists but doesn't attempt to describe it: they leave that up to the fictional books themselves. In particular, the whole series is a lightly veiled parody or pastiche of real world objects. There is much discussion within the fandom as to exactly what maps where, but this is the sort of discussion which WP looks down upon. Neither is it George Orwell. I'm unaware of any O level passnotes guides on "themes in this universe", as abound for Animal Farm.

Yet again too, this is a UK series (like Viz, like Redwall, like a similar world-context article for this same series already, like indeed the background articles to Animal Farm) where these articles would be unquestioned on WP if their subject was produced by DC Comics or Marvel, but if they're outside the US comics canon the rules are held to be quite different.

Thoughts? Andy Dingley (talk) 17:50, 30 June 2018 (UTC)

We cover topics in proportion to their reliable, secondary source coverage. If this topic is an important aspect of a fictional world, it should be covered summary style in its parent article and only need be split out when warranted by an overabundance of material in reliable, secondary sources. If the topic has only primary source coverage—or uses sparse secondary source coverage as a coat rack for reams of primary source summary—the best course of action is to merge somewhere more general or if that is impossible, delete. czar 19:52, 30 June 2018 (UTC)

We cover topics in proportion to their reliable, secondary source coverage.

What's the policy basis for that? Andy Dingley (talk) 20:51, 30 June 2018 (UTC)

WP:PROPORTION czar 18:27, 18 August 2018 (UTC)

There is no policy basis, and PROPORTION does not support deletion in such cases. James500 (talk) 03:28, 30 August 2018 (UTC)

The guideline WP:FAILN provides for merging when there is not sufficient information for a separate article. But in the statement above I would change " only need be split out when warranted by an overabundance of material" to "should be split when there is sufficient material ". If there is actually an "overamundance", the article is too long and must be split somehow. WP:Summary style is the best guide. In practice, the determining factor has often been whether the relevant aspect is discussed or even referred to independently of the main work. In the past, I have usually preferred separate articles, because merging can often lead to reduction of the material to such an extent that it eventually gets turned into a mention, and then sometimes removed from the article as insignificant. We have no enforceable guideline for article content, and covert deletion has in the past been used so destructively as to seem a deliberate way to do covert deletion. I would like to think that this no longer happens. DGG ( talk ) 17:31, 6 December 2018 (UTC)

fwiw @DGG, I've found the word choice of "overabundance" to be a brighter line than "sufficient" or "significant" czar 17:54, 6 December 2018 (UTC)

people keep using the term"bright line" around here. There are no bright lines, except with respect to a few lgeal issues. Everything needs to be interpreted, almost everything admits of exceptions, everything can be changed in general by consensus or similarly by consensus in any general case. There are some things that may sound like bright lines, but thet's only until you try to analyze or apply one. We absolutely do not host copyvio, but we do make exceptions for fair use, and our rules for exceptions are exceedingly complicated and a great many individual cases admit of argument. We don't print unsourced or poorly sourced gossip about living people, but the question arises about what is so poorly sourced to count as unsourced, and what is gossip rather than significant information in context. Certainly with respect to notability there's nothing even approaching a fuzzyline--I think of it as two opposite centersthat together make up half our proposed content, with a broad disputable zone in between. Sometimes I think of it like a 16 the century world map, where the actually known parts of the world make up a minority of the area.

but to the extent that the distinction is meaningful, I intended to specifically reject overabundance, as far too demanding a criterion under any reasonable interpretation. If it makes things easier to decide, it's because it puts the boundary where they will be very few split to consider. Im not arguing over wording, but where to make the distinction. DGG ( talk ) 04:47, 7 December 2018 (UTC) ~

Is a review of a single book sufficient to argue that a book series it is part of is notable?

Such an argument was recently made at a certain AfD, where NBOOK was mentioned. I don't see anything in NBOOK right now that addresses it one way or another, through in general WP:NOTINHERITED comes to mind. That said, I think an argument could be made then when most or all of a series books are notable, an article about a series might be warranted - but at the same time, GNG remains a concern, and the overall solution can be a navigational template. Thoughts? --_{Piotr Konieczny aka Prokonsul Piotrus| reply here} 03:55, 15 March 2020 (UTC)

A single review of a single book (or as in the case you link a single reliably-published review plus a self-published blog review) isn't even enough to argue that the book itself is notable. I've been making a fair number of articles on (technical) books lately and I think the bare minimum for GNG is two reliably-published in-depth reviews, but it's far preferable to have more than that. For a series, I'd want either multiple reviews of the series as a whole, or multiple books in the series that are independently notable and whose reviews say something nontrivial about the whole series. —David Eppstein (talk) 04:09, 15 March 2020 (UTC)

Government by Judiciary

An editor posted the following question on my talk page:

"I have a question about the article for Government by Judiciary. Would the fact that this book has been cited over 2,000 times enough to establish its notability? See this. Futurist110 (talk)"

Now, this guideline doesn't take this into account, so I was wondering what the consensus would be on this? Onel5969 ^{TT me} 17:18, 28 March 2019 (UTC)

Citation based notability is hard enough for academics that I would be mighty reluctant to add it to NBOOKS. However 2000 cites suggests that it clearly is notable and there should be sourcing to support such notability which I think is the case given coverage in the NYT and Washington Post (not a full review but given the appellation clearly suggests notability). Best, Barkeep49 (talk) 17:44, 28 March 2019 (UTC)

Thanks Barkeep49! Onel5969 ^{TT me} 11:25, 29 March 2019 (UTC)

That book has far more published reviews than are necessary for notability. High citation counts are suggestive but reviews are clearer. —David Eppstein (talk) 05:42, 15 March 2020 (UTC)

Meaning of literary award?

What is meant by literary award? Is it the narrow meaning of literary (writing considered to be an art form) or the vast meaning? If it is the narrow one, this criterion cannot be applied to political science or economics books for example. Right? Ali Pirhayati (talk) 09:39, 1 July 2020 (UTC)

Wikipedia:Notability (poems)

I just noticed this doesn't exist, should it redirect here? Should we add some section about poems here, or at least the very word somewhere? --_{Piotr Konieczny aka Prokonsul Piotrus| reply here} 06:35, 2 July 2020 (UTC)

SNGs and GNG

There is a discussion at Wikipedia talk:Notability on the relationship between SNGs and the GNG which might be of interest to editors who watch this page. Best, Barkeep49 (talk) 21:47, 7 November 2020 (UTC)

Gameable, over-inclusive criterion

This doesn't work:

"The book has been the subject of two or more non-trivial published works appearing in sources that are independent of the book itself. This includes ... reviews."

It isn't factual that every published review constitutes non-trivial coverage. Many reviews are perfunctory and, if the work being reviewed and the work publishing the review are both in the same narrow field, virtually automatic/guaranteed – the review publication will attempt to review every non-self-published book that came out recently in that field.

This assumption that 2+ reviews in any publication auto-confer notability will also over-include non-notable works of fiction, because there are multiple journals that churn out large numbers of reviews of novels. We do not apply a criterion like this to films, television, or other works, for good reason. It may be salvageable by rewording it to require an in-depth review in a publication that does not specialize in reviews or in the same topic as the work being reviewed, or something like that. Unless something changes, this page is basically in direct conflict with WP:GNG by redefining non-trivial to include the trivial.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  16:38, 16 January 2018 (UTC)

• Agreed czar 11:11, 1 March 2018 (UTC)
• Utter nonsense. What SMcCandlish says in the thread above is manifestly factually untrue manifest total nonsense from start to finish. Literally nothing he says is accurate. First he quote mines criteria 1 by omitting the footnotes to that criteria. This is completely misleading because footnote 1 to that criteria fully answers all the criticisms that he is making. What that footnote says is this: "The "subject" of a work means non-trivial treatment and excludes mere mention of the book, its author or of its publication, price listings and other nonsubstantive detail treatment." The key words there are "non-trivial treatment" and "nonsubstantive detail treatment". They are a paraphrase of the words "significant coverage" which appear in GNG. Moreover, they are an accurate paraphrase and don't need to be changed. The main idea behind criteria 1 is that two book reviews (or similar coverage) of normal length and detail always constitute significant coverage. This idea is correct and is merely an accurate restatement of GNG, because GNG says exactly the same thing. Footnote 1 prevents the use of sources which, even taken together, are so insubstantial as to be useless. Which is all that GNG actually does. So the criteria does not say that every published work constitutes non trivial coverage. Nor does it say anything close to that. Nor does it say anything within a parsec's distance of that. Next I am confronted by epic exaggeration of the number of book reviews available and spectacularly misleading statements about their general level of detail. At a minimum, the vast and overwhelming majority of book reviews are not "perfunctory". In fact, they are far from "perfunctory". In fact, they are massively lengthy and detailed. I am not aware of any book reviews that are "perfunctory" (not a single example is offered, never mind proof of the assertion that there are "many") and it is clear to me that, if any such reviews do actually exist, they must be exceptionally few in number both in absolute terms and as a proportion of the total, because I cannot recall ever seeing even one, and I have seen a huge number of reviews in a huge number and varied range of sources. Again, I have never heard of a single periodical that tries to review "every non-self-published book that came out recently in that field". Not one. In view of the number of books published, it is not apparent to me that it would even be physically possible for a periodical to review all non self published books in its particular field unless that field were narrower than the breadth of a postage stamp and presumably commercially non-viable (no one will buy a periodical if its scope is too narrow to be useful). If such periodicals exist at all (again, not a single example is offered, let alone proof that the number of such periodicals is large enough to be worth giving them a moment's thought in this context) I can only infer they must be exceptionally few in number for the same reasons. My experience of the number of reviews in scholarly journals, for example, is that the exact opposite is true. These periodicals typically contain a list of books received in addition to the reviews. And what you find, over and over again, is that a massive proportion of the books received, let alone published, are not reviewed at all. It can, indeed, often be very difficult to find reviews of works that one knows to be of exceptional importance and notability. Furthermore "reviews everything in its field" is not even a valid objection to a source. If a periodical did review every book published in a particular field, that would in fact be exceptionally strong evidence that all books published in that field are of exceptional importance and therefore notable. A periodical would not want to review books that were not worthy of notice. If it reviews those books, it does so because it thinks them notable. Nor does an objection based on "reviews everything in its field" have anything whatsoever to do with GNG. GNG is exclusively concerned with the length and detail of a source. GNG does not require that sources cover only a small fraction of the topics within their field for the sake of the sort of absurd notion that is presented as some kind of 'elitism' ('only the best is good enough') but is really an expression of extreme anti-intellectual philistinism (including in particular a complete lack of curiosity about, or interest in, anything above the level of the lowest common denominator, which comprises of, and only of, all those topics which are so obvious that even the least intelligent and most poorly educated people would know all about them; a lack of interest that typically develops into extreme hostility to all learning whatsoever, and into endlessly screaming and screaming and screaming for the deletion of ever increasing amounts of content, as we see all too often in the Wikipedia space and its talk pages). GNG is about "how much coverage does this topic have?". GNG is not about "other topics must have less coverage than this topic", which is a classic "other stuff exists" argument that is completely without merit and completely incompatible with GNG. Criteria 1 will not result in the over inclusion of fiction. There are no journals that review huge number of novels. In fact, I have conclusive positive evidence from reliable sources that conclusively proves beyond any possibility of doubt whatsoever that there is a serious paucity and unacceptably low number of reviews of novels which will result in serious under inclusion of fiction unless we adopt a test that does not require such books to have reviews. P N Furbank clearly and unequivocally states that "low brow" best selling books "frequently" receive no reviews at all (Source: P N Furbank, "The Twentieth Century Best Seller" in Boris Ford, The Pelican Guide to English Literature, Penguin Books, 1961, volume 7 ("The Modern Age"), page 430). (Notice also how I cite sources in support of what I say, unlike those who argue for deletion. And it is high time we apply WP:V to the Wikipedia space). We do apply a criterion exactly like this to all films, television, and other works, and that criterion is called GNG. The proposal to exclude all periodicals that "specialize in reviews in the same topic as the work being reviewed" is not only absurd. It is actually totally meaningless. How specialized? And what is a "topic"? A periodical whose scope extends to the whole of history, or to the eighteenth century, or to archaeology, or to French history, or to economic history, could not be regarded as "specialized" because those are vast subjects, and they produce vastly more books than any periodical is likely to be able to review, as far as I can see. But you can be absolutely sure that the moment we put a word like "specialised" in the guideline, deletionist mega-trolls would start insisting that absolutely every scholarly journal is specialised. Because deletionist mega-trolls hate any subject is too intelligent for them. Such a proposal would dumb down Wikipedia to the level of a children's encyclopedia for babies. Likewise the words "in depth" must always be avoided like the plague because deletionist mega-trolls always use things like that as a pretext for trolling of the "no matter how much coverage there is, I won't accept it is sufficiently in depth and will always demand more" variety. The idea that we should exclude "publications that specialize in reviews" irrespective of the number, length and quality of those reviews is utterly bizarre. Some of those publications are amongst the best sources possible, such as The New York Review of Books. I am not aware of any whose reviews are not in fact excellent and more than good enough for our purposes. It might also introduce needless additional systematic bias, for example against children's books because children do not read reviews. Quite apart from all of this, SNG do not have to comply with GNG. The whole point of SNG is to provide additional criteria for inclusion, because it is well known that GNG is seriously under inclusive of certain types of topics. So that would not be a valid argument against this guideline even if it was true. It is true that criteria 1, as presently written, can be gamed. But only by deletionist mega-trolls. The fact is that criteria 1 is actually seriously under inclusive. There is actually no possible justification for requiring that there be at least two reviews as is presently required. One book review of normal length is more than enough to conclusively prove the notability of a book. We should improve this guideline by replacing the number two with the number one. James500 (talk) 17:14, 8 March 2018 (UTC)

  Holy wall of text, Batman! Oh, and One book review of normal length is more than enough to conclusively prove the notability of a book is completely crazy. EEng 05:52, 12 November 2018 (UTC)

• Agree with SMcCandlish The question of Publisher's Weekly, which is designed for evaluation of books by bulk purchasers such as libraries, has recently been broached elsewhere. It publishes 10,000+ reviews annually many of which are of books that are clearly not in the spirit of NBOOK. We've also had recent cases about books being kept on the basis of a review on something like rpggamer.net which, again, I don't think is in the spirit of NBOOK, even if it is in the literal reading of it (maybe). Chetsford (talk) 18:23, 18 August 2018 (UTC)
  • Libraries are not bulk purchasers: They are not wholesalers. The Publishers Weekly website is obviously designed for use by the book reading general public and print copies held in university libraries etc will be read by professors and students of literature etc. The reviews are in the spirit of NBOOK. Their number is small by our standards (we have more than five million articles and the number of topics that satisfy GNG probably exceeds 100 million). The first part of this proposal has been archived now, but it met opposition from DGG and Tokyogirl79 (now called Reader of the Pack). The fact that WP:CREATIVE is being invoked elsewhere in support of reviews in Publishers Weekly is evidence that consensus supports the use of those reviews as evidence of notability. James500 (talk) 04:53, 19 August 2018 (UTC)
• Eh - I'm not sure what the problem is. It isn't factual that every published review constitutes non-trivial coverage. - which is why the criterion specifies non-trivial, right? 2+ reviews in any publication - But it's not in any publication. in-depth review in a publication that does not specialize in reviews - Well, we want them to be competent with reviews. I presume you mean e.g. Kirkus or Publisher's Weekly. I've not noticed a problem except when people try to pass off short bits or sponsored content as real reviews. Hence it should be independent, non-trivial, etc. This is not specific to books, fwiw. Video games, for example. In that case, the overwhelming majority of reviews come by way of popular blog-based websites that pump out a lot of content, so it's less challenging to satisfy a review threshold than some other kinds of media.

All that said, what I don't understand is why we're including bestseller lists among other types of media in NBOOK#1. That should be its own bullet, if it's going to be there. Also, a common problem with a subset of SNGs is here. It says "A book is notable if it ...meets... at least one of the following criteria." Something that satisfies one of those criteria but hasn't actually received in-depth coverage in multiple reliable sources independent of the subject should not be considered notable. — Rhododendrites ^talk \\ 05:45, 19 August 2018 (UTC)

• • (1) The purpose of including best seller lists in criteria 1 is to allow an article based on one best seller list plus one book review, instead of two book reviews. Its inclusion acheived strong consensus when it was discussed some time ago. (2) Removing the word "is" would mean that all books would have to satisfy GNG. That would defeat the purpose of having SNG. If you want everything to satisfy GNG there is no point in having SNG. James500 (talk) 06:00, 19 August 2018 (UTC)
    • Rather than get sucked into this "is something notable even if there aren't sources providing significant coverage" GNG vs. SNG wheel-spinning, I'll save it for when there's actually a proposal. — Rhododendrites ^talk \\ 18:23, 19 August 2018 (UTC)
• Agree with SMcC with the addition that reviews are also not generally "independent": if I read a book and write a review of it for my blog or monthly newspaper column or whatever, I haven't provided independent coverage of the book; it's likely that everything I know about it comes from the book itself. Hijiri 88 (聖やや) 16:38, 19 August 2018 (UTC)

• ?????! Whoa there. That a source is a review in no way affects its status as independent. If it's connected to the publisher, written by the author herself, or is a sponsored review, then that's one thing. But a review is not only typically independent but one of the best sources for a piece of media. A review is by definition not just a synopsis/summary but involves analysis, evaluation, interpretation, and/or judgment of the original work. That's the most basic basis for evaluating notability. By this logic the only reliable sources for books are not reviews, but, what, reviews of reviews? — Rhododendrites ^talk \\ 18:23, 19 August 2018 (UTC)

a review is [...] one of the best sources for a piece of media That depends. If a book is reviewed in a scholarly journal like Monumenta Nipponica that (long, detailed) review (article) is probably one of the best sources for the book in question; but a "did I like it, should you buy it"-type review -- what most editors probably think of when they see those words in the guideline -- is not really reliable for anything other than its author's opinion, and certainly doesn't provide "independent coverage" of any of the kind of content we would want to include in an encyclopedia article. Hijiri 88 (聖やや) 18:32, 19 August 2018 (UTC)

Sure. We can sort out what constitutes a good review. It's partially already specified in the guideline, but it could be further clarified, of course. But even "I liked it you should buy it" can be independent. It's just a lousy source is all. I was just responding to reviews are also not generally "independent". — Rhododendrites ^talk \\ 18:36, 19 August 2018 (UTC)

even "I liked it you should buy it" can be independent Again, that very much depends. It's not independent for any information about the book itself that it seems to have just parroted from the book itself. (This review, for example, clearly got all its information directly from [the reviewer's misinterpretation of?] the book itself, and so is completely useless for factual claims.) The only information we could cite such a review for is the reviewer's opinion of the book, which is not enough to build an article out of, which in turn is what notability guidelines are supposed to be about. Even two reviewers' "I liked it you should buy it"/"I didn't like it you should borrow it from the library at best" opinions are not enough to build an article out of, and I would argue that 100 such opinions would not be either. Hijiri 88 (聖やや) 18:50, 19 August 2018 (UTC)

No, a review is independent unless the reviewer is owned or otherwise controlled by the seller. I can't see anything wrong with the review in the Japan Times. One brief quote in a much longer article will not render a review useless. A review ought to be based on the reviewers' reading of the book. If he gets all of his information second hand, his review will be less reliable, not more reliable ("Chinese whispers"). Misinterpretation is relevant, but goes to reliability not independence. That said, what has he misinterpreted? James500 (talk) 06:14, 20 August 2018 (UTC)

I can't see anything wrong with the review in the Japan Times. [...] That said, what has he misinterpreted? Our article on the Man'yōshū, and articles such as waka (poetry) and Kaifūsō, should clarify that he was wrong to call the MYS Japan's first collection of "domestic poetry" ("Japanese language" does not mean the same thing as "domestic"), Susumu Nakanishi that he got the name of "Susumu Nakahashi" wrong, Interpretation of the title of the Man'yōshū that he is misinterpreting the title as literally referring to leaves, etc., etc. The errors in that review show its author was completely reliant on the content of the book he was reviewing, which means he was not "independent" of it for anything other than his own opinion. Hijiri 88 (聖やや) 08:38, 20 August 2018 (UTC)

• Oppose Wikipedia has almost 6 million articles. I don't feel that possibly getting 10000 articles/year on books is excessive or over-inclusive. Regarding "gameable", there are better ways of generating press coverage for any book with a non-trivial marketing budget. This should still keep out self-published books, obscure fringe books, and books that receive no attention from libraries or consumers, which is sufficient. power~enwiki (π, ν) 18:40, 19 August 2018 (UTC)
• Where's the beef? The OP doesn't provide any evidence that there's a problem that needs fixing – no examples; no statistics; no surveys; nothing. Andrew D. (talk) 18:43, 19 August 2018 (UTC)

Even one article on a fringe book that is reliant on credulous reviews, like this one was, is too much. Hijiri 88 (聖やや) 18:50, 19 August 2018 (UTC)

Fringe theories are already excluded by WP:FRINGE. Unreliable (including credulous) reviews are already excluded by this guideline because they are not reliable. What happened in that AfD was a disagreement over whether the sources in question were reliable. That depended at least in part on whether you were reading the book as zoology or as entertainment. Folklore etc can be read as if it was fiction. It is also worth bearing in mind that no actual evidence of unreliability was produced at the AfD. Probably FRINGE was misapplied, because a claim that a book is "enjoyable" or that certain groups of people who like to read nonsense would probably want to read that book is not a fringe theory, even if the book itself is total nonsense. Even if there was a problem, all that would be required is a brief mention of WP:FRINGE in the guideline, not tearing the guideline apart in a way that prevents perfectly reliable non-fringe sources being used to support the notability of non-fringe books. "Even one is too many" is a reason that should never be used to change a guideline (it would cause massive disruption to the project for very little gain) especially when the one in question got deleted because there is apparently already a policy. WP:FRINGE already exists, therefore no problem, QED. James500 (talk) 05:47, 20 August 2018 (UTC)

it would cause massive disruption to the project How so? It would seem to me the opposite is the case. Hijiri 88 (聖やや) 08:38, 20 August 2018 (UTC)

• Agree with SMcCandlish. It seems to me that the two-review standard is probably the weakest notability standard across all notability standards. It means that two reviews of the form "this book is only good as toilet paper" can justify an article. At the very least, reviews of the book should state that the book is notable. Zero^talk 06:40, 27 August 2018 (UTC)
  • The two reviews standard is no lower than GNG. Criteria 1 is GNG. Raise the standard for criteria 1 and editors will simply invoke GNG instead, and with good reason. And the article will be kept every time. GNG is not interested in whether coverage is positive or negative, and criteria 1 of NBOOK should not be either. Criteria 1 has stood for many years and no one has complained about it till very recently. The 'standards' of some of the other SNG should be lowered. The hurdle they set is far too high, much higher than GNG, and they are driving editors away from the project. According to WP:Size, our rate of article creation is exceptionally low (687 per day compared to more than 1800 per day several years ago). The last thing we want to be doing now is to delete more articles. If we increase deletion even further, the project is in real danger of collapsing from lack of editors. To be honest talking about "standards" and using words like "high" or "low" in this context is misleading, because increased deletion of accurate verifiable content results in a reduction of quality. Deleting more articles on books will not result in a "higher standard of notability" in any meaningful sense of those words, but it will actually result a loss of utility for our readers. Our readers are the ones who suffer when accurate content is deleted. James500 (talk) 03:19, 30 August 2018 (UTC) And if two reviews say that a book is awful, having a Wikipedia article is a good thing because it will put a 'health warning' on the book that will stop our readers getting ripped off or misled by a bad or inaccurate book. James500 (talk) 03:38, 30 August 2018 (UTC)

You are arguing in circles. GNG requires "significant coverage" and you are simply assuming that two reviews is significant coverage when that is what you need to establish. I contend that two reviews is piss-weak coverage, almost none in fact. Your predictions of doom don't strike me as realistic. Personally I'm an inclusionist, but this guideline provides hardly any standard at all. If a few articles on rubbish books disappear, I won't shed any tears. Zero^talk 04:01, 30 August 2018 (UTC)

(1) Two reviews of normal length is significant coverage. (2) If those articles disappear, our readers will shed tears when they get ripped off by the sellers of those rubbish books, because we failed to warn those readers the books were rubbish. This encyclopedia exists for those readers. James500 (talk) 04:30, 30 August 2018 (UTC)

Two reviews of normal length is by no standard significant coverage by the general notability guideline. SNG are designed as a shortcut for cases where the GNG standard is met obviously, not edge cases where one person can argue "two sources is enough" and others are left asking how we can do justice to the topic based on two normal-length articles alone. czar 15:42, 2 September 2018 (UTC)

• As written, I think the criterion is rather weak. Not necessarily the "two reviews" part but rather the "sources" part. Ideally, we would like the sources to not just be independent but also have a record of dispassionate reviewing. Publishers Weekly, for example, doesn't fit that standard since it is a trade journal and is in the business of promoting the book industry as a whole. I'm not exactly sure how to fix this in the wording but, yes, the wording does need to be changed. --regentspark (comment) 16:02, 2 September 2018 (UTC)
  • Publishers Weekly is not in the business of promoting the books it reviews. No evidence has ever been offered of that. It is not a trade journal. Its book reviews are are aimed at librarians. University, school and public librarians are academics, scholars and educationalists. They are not tradesmen and there is no 'library trade'. Its reviews also aimed at bibliographers, and professors and students in the field of literature. They are also aimed at the book reading general public. None of those people are tradesmen either. A requirement that sources be "dispassionate" would be a terrible idea, because the effect would be that any source that expressed an opinion of any kind, whether positive or negative, about a book would be immediately smeared as "biased". WP:BIASED states that we do not require sources to be biased, objective or neutral, and GNG has never required sources to be neutral as well as independent. And with good reason. Because it would be a charter for mass deletion that would allow any source that expressed an opinion of any kind to be smeared as biased. In fact, sources that expressed no opinion at all would be smeared as expressing one. It would make it absolutely impossible to write an article about anything. That is why GNG does not have such a requirement. There is no reason to think that the reviews in Publishers Weekly are any less dispassionate than reviews anywhere else. If the reviews were promotional, if they were mere adverts, no one would read Publishers Weekly. The reviews have to be tolerably objective to keep the librarian etc audience happy, otherwise they would vote with their feet and read reviews elsewhere. Think about it. If Publishers Weekly was rubbish, librarians etc would buy something else. It can't afford to be promotional rubbish, because it would go out of business if it was. If Publishers Weekly promoted the books it reviewed, that would be a case of the tail wagging the dog. The reality is that a commercially published book review has to tell readers what they want to know, not what the reviewer would like them to think. James500 (talk) 06:57, 3 September 2018 (UTC)

  Publishers Weekly (PW) is an American weekly trade news magazine targeted at publishers, librarians, booksellers and literary agents (emphasis mine). Definitely a trade journal. And, given its mandate, it stretches credibility to think that publishers will have no influence on what it chooses to review. --regentspark (comment) 17:27, 3 September 2018 (UTC)

  Our Wikipedia article on Publishers Weekly is not a reliable source, so there is not point in quoting that (WP:CIRCULAR). The source that our article cites in support of that claim does not, as far as I can see, say that it is a trade journal. Nor does it give exactly the same list as our article. In a 2005 interview, Sara Nelson said something to the effect that she wanted Publishers Weekly read by the general public. Librarians are not part of any trade. Booksellers are not part of the publishing trade and would appear to be independent (provided they are not owned by the publisher). On the face of it they are retailers and they can turn round to the producers (ie publishers) and say "we don't want to sell your book in our shop, so we won't buy copies of it from you". As far as I am aware, the vast majority of subscribers are not publishers. I think it stretches credibility that they control what gets reviewed because they seem to be heavily outnumbered by other readers. I would need to see positive evidence of such control before I could support that argument. The publishers who do read it are presumably reading it for news and not for reviews. The book reviews appear to be aimed at people who would want objective reviews. As far as I am aware they started as forecasts the purpose of which was to accurately predict popularity, not influence it. A forecast would be useless to a bookseller (retailer) if it was not objective (it would in fact hurt him to act on faulty predictions). And librarians will obviously want dispassionate reviews. So if the reviews were not 'dispassionate' as you put it, the audience would stop reading them, most of the readership of the periodical would be gone, and Publishers Weekly would go out of business. I would need to see more convincing evidence before I could agree with what you are saying. James500 (talk) 07:11, 4 September 2018 (UTC)

  Do you realise that your long walls of text are read by almost nobody except yourself? I started to read this one but stopped at "As far as I am aware, the vast majority of subscribers are not publishers." since you have absolutely no way to have such information and in any case it is completely irrelevant to RegentsPark's point. Zero^talk 10:42, 4 September 2018 (UTC)

  Yeah. I'm not at all known for brevity, and I tried to wade through all that stuff in your huge post up top, James500, but too much of it's just ranty invective, opinion-mongering, and off-topic stuff that isn't responsive to the issues raised. I gave up after a while. At very least, use some paragraph breaks, and try to arrange such material in cohesive blocks.  — SMcCandlish ☏ ¢ 😼  16:01, 8 November 2018 (UTC)

  That is total nonsense, SMcCandlish. James500 (talk) 22:47, 8 November 2018 (UTC)

  You seem to like that word a lot without having any idea what it actually means.  — SMcCandlish ☏ ¢ 😼  08:35, 11 November 2018 (UTC)

  Our article on PW gives a breakdown of who subscribes to them. It says that 19,000 out of 25,000 readers in 2004 were not publishers, if I read the figures correctly. It has been there since the first revision of 6 December 2004 and might come from one of the three sources given there. James500 (talk) 13:31, 4 September 2018 (UTC) This source gives a similar breakdown of PW's subscriptions, and states that the majority of subscribers are not publishers: [5]. James500 (talk) 08:30, 6 September 2018 (UTC)

  Regardless of who it is aimed at, this is a publication of the publication trade and the reviews are very brief paragraphs (a uniform 200–220 words) written by anonymous staff members. The claim that such reviews add significantly to the notability of a book is completely untenable. Zero^talk 02:30, 4 September 2018 (UTC)

  200 words is plenty and is not particularly brief. I think 100 words would generally be significant. There are many professionally published printed encyclopedias with lots of articles 100 words long or shorter, and I see no reason why our articles should absolutely need to be longer than those in other encyclopedias. In any event WP:NBOOK does not presently say in express words that 220 words is enough, so you are free to argue that at AfD, and even if you were right (which is denied), there is nothing in the guideline to correct. Further, if we were to put "220 words is not enough" in NBOOK, that would bring NBOOK into direct conflict with GNG, which has never said that. The example GNG gives for insignificant coverage is a 14 word passage, and if you want to raise that bar, the place to start is GNG, not NBOOK. James500 (talk) 07:38, 4 September 2018 (UTC)

  If I'd written "20-22 words" you would have replied "20 words is plenty and not particularly brief". Bye now. Zero^talk 10:42, 4 September 2018 (UTC)

  I'm back because I can't resist this. Press release by Publishers Weekly: Publishers Weekly, long the leading trade magazine for the book publishing and bookselling industry. Bye again. Zero^talk 11:13, 4 September 2018 (UTC)

  NYT says the reviews in PW can be up to 500 words long, not 220 words, so your number wasn't accurate. James500 (talk) 13:31, 4 September 2018 (UTC)

  That 2005 article says 200–500 but my numbers were based on actual word counts in the most recent issue. There is one review about 250 words, two about 230 words, and the others all range from 150 to 220 words. Compare it to how many words you have needed in this thread to make very simple points. Zero^talk 00:24, 5 September 2018 (UTC)

• @Rhododendrites: Re, "I'm not sure what the problem is. It isn't factual that every published review constitutes non-trivial coverage. - which is why the criterion specifies non-trivial, right?" – You have to look at the wording and analyze it a bit more carefully. It does not at all say that non-trivial reviews are required. It defines reviews as non-trivial, across the board: "The book has been the subject of two or more non-trivial published works appearing in sources that are independent of the book itself. This includes ... reviews." Similarly, someone else claimed that there can't be a problem with fringe books and credulous reviews because a credulous review would not be independent. But it doesn't actually say anything like that at all. The review - already auto-labeled "non-trivial" simply for being a published review at all – does not have to have any particular character, or any particular relationship (rather, lack of one) to the subject; only the publication has to be independent. So, a biased, credulous, even directly interest-conflicted review or three constitute grounds for establishment of notability, as long as the publications that ran the so-called reviews aren't connected directly to the subject. There's not even a requirement that they be reputable publications. This is not okay. It's obviously not what we want, and this is simply a wording problem. It should be pretty easy to fix.  — SMcCandlish ☏ ¢ 😼  16:01, 8 November 2018 (UTC)
  • That is nonsense from start to finish, SMcCandlish. The guideline does not come close to defining book reviews as non-trivial per se. It defines book reviews as "published works" which may or may not be trivial. Furthermore, the guideline clearly does require that publications be reputable. Footnote 2 to criteria 1 clearly says that "Non-trivial" excludes media that are not reliable. Once again you are ignoring the footnotes and asserting that the guideline says things it clearly doesn't. James500 (talk) 21:58, 8 November 2018 (UTC) I have copy edited the guideline in such a way that I think it will make it wholly impossible to wikilawyer the definition of "published works" by claiming that one does not know what the word "this" refers to: [6]. The other criticisms you make appear to be the result of your continued failure to read the footnotes to the guideline. James500 (talk) 22:29, 8 November 2018 (UTC)

It defines book reviews as "published works" which may or may not be trivial. Yeah, that may be the case, but the fact is that as this guideline has been used, you and others have been using clearly trivial reviews to prop up whole articles.[7][8][9][10] (Note that I am not commenting on whether those specific topics are notable or whether non-trivial sources do exist, just that a significant portion of the reviews cited as being non-trivial clearly were anything but. I'm also not including cases like this where it seems the reviews were in fact of the non-trivial kind.) So it doesn't actually matter that the guideline could be read as being more careful/distinguishing than it needs to be, since it actually is read (and cited in AFD discussions) as being more permissive/inclusive than it should be. Hijiri 88 (聖やや) 23:50, 8 November 2018 (UTC)

The reviews in question were not trivial and would have satisfied GNG. If you do not agree with that, you will have to change GNG as well, and the place to do that is WT:N, not here. I think you have WP:SNOW chance of making GNG less inclusive, since that has been tried many times before. I certainly cannot support any change to this guideline that would subject books to notability criteria that are less inclusive than GNG. Further, the fact that articles have survived AfD proves that the guideline reflects the consensus established at AfD, that it reflects consensus established by long standing and always followed precedent. James500 (talk) 03:26, 9 November 2018 (UTC)

Some of them may be non-trivial and may satisfy GNG, but it's clear that was not the criterion applied. Most reviews of novels that are printed in daily or weekly newspapers are just "I liked / didn't like this book, and here's why". Currently, as in the above linked diffs (all of which cited, perhaps sometimes implicitly, NBOOK, not just GNG), NBOOK is being used to allow for X number of such primary sources / opinion pieces to be used to give a book its own article: this is not acceptable. Hijiri 88 (聖やや) 05:00, 9 November 2018 (UTC)

I disagree with everything that you say. A book is a primary source for itself. A book review is therefore a secondary because it is based on a primary source and evaluates the primary source. So the type of sources that satisfy criteria 1 are ipso facto secondary by reason of being what they are. (Quite apart from the fact that a requirement for secondary sources (as opposed to independent sources) serves no useful purpose, is not included in other SNG, and is not supported by real scholars who urge the use of primary sources in preference to secondary sources). James500 (talk) 05:33, 9 November 2018 (UTC)

I explained this to you three months ago: are we doomed to repeat this dance forever? X number of "independent secondary sources", when such independent secondary sources just amount to "I like it you should read it", is not enough to build an article. Hijiri 88 (聖やや) 05:42, 9 November 2018 (UTC)

It is clearly possible to build an article out of book reviews that contain significant coverage. Book reviews are typically a lot more detailed than a single seven word sentence. I would have thought that seven word sentence would certainly be considered "non-substative detail treatment" for the purpose of criteria 1, as it would be considered a trivial mention for the purpose of GNG (which expressly requires more than fourteen words). James500 (talk) 06:01, 9 November 2018 (UTC)

No, normally the only information they provide on the book is a summary of the book's contents, without independent fact-checking, etc., making them only as usable for that content as the book itself. Technically, for almost anything else on the encyclopedia, it should be in the form "X, citing Y, says Z", but "Reviewer X, citing the book itself, says Z about the book" is ridiculous. Again I will refer to Elliott's review of Man'yo Luster, which clearly got all its information from (Elliot's misreading of?) the book itself, and is completely unreliable for everything but Elliott's opinion of the book itself (which amounts to "I like it you should read it"). Hijiri 88 (聖やや) 07:21, 9 November 2018 (UTC)

I don't think this is accurate, in a broad sense. No one will dispute that there are bad reviews, even if the source is typically considered reliable. Still, a good review will contain helpful information like publisher, release date, page count, and maybe a comment or two about various editions available. All of that can be used to build a publication section. Good reviews may also be used to source themes, style, allegories, and references to other works that would be considered OR for an editor to claim on his own. Identifying and excluding bad reviews like your example is the responsibility of knowledgeable editors. Argento Surfer (talk) 13:49, 9 November 2018 (UTC)

Yes, but most of that information would be available from the book itself; indeed, normally simple newspaper book reviews will just take that information straight from the book itself without fact-checking, since that's not what the readers are interested in. Hijiri 88 (聖やや) 22:41, 9 November 2018 (UTC)

You can describe the themes, style, and allegories from the book itself without it being OR? I'm skeptical. Argento Surfer (talk) 13:40, 13 November 2018 (UTC)

@Hijiri88: One review by one reviewer in one periodical does not say anything about other reviews by other reviewers in other periodicals. One example does not a pattern make. As far as I am aware, that sort of thing is the exception rather than the rule. There are certainly a number of problems with that example: (1) The reviewer is not contemporary with the book reviewed. You can't expect a newspaper journalist to be an early medieval historian. A review of that book published in a medieval history journal would be a better example than one published in an ordinary newspaper. A review of a contemporary book would have been an even better example. Criteria 1 is primarily meant for contemporary fiction. (2) The review is not written in the same language as Man'yōshū. Man'yōshū is written in Japanese, whereas the review is written in English. (3) Man'yōshū is obviously notable and would satisfy WP:NBOOK even if it had zero reviews. Book reviews did not exist in the eighth century, so they are not required for works of that era. So the review in the Japan Times has not generated a false positive. So all things considered, I don't think that particular review is a good example. Further, since footnote 2 to criteria 1 expressly excludes unreliable reviews, there is not much point in proving that unreliable reviews exist, because unreliable reviews don't count towards criteria 1 anyway. James500 (talk) 21:22, 9 November 2018 (UTC)

You're getting confused. The subject of the review is the book Man'yo Luster, which is (largely) in English, and whether it meets our notability guidelines (it is not the same thing as the famous anthology Man'yōshū) is up for debate. I wouldn't want a review that contains so many obvious errors to be used to demonstrate the notability of the book. It attributes authorship of the "original" Japanese book to the non-existent "Susumu Nakahashi", whose real-world counterpart Susumu Nakanishi wrote the book(s) from which the modern Japanese kana representations of the poems were taken, while crediting Levy (the actual author of the book under review) as the "translator". These kind of errors make me very skeptical about the "new translations" the review attributes to the book, since Levy has published various MYS translations, beginning with a complete translation of about a quarter of the whole work, over 30 years or so, a fact of which the reviewer seemed unaware. (As a comparison, I noticed some slight changes to the formatting in the "Japanese edition" of One Hundred Poets, One Poem Each -- I don't think anyone would reasonably arguing that removing a comma makes it a "new translation", but it seems entirely possible that that kind of change is all that was made.) Hijiri 88 (聖やや) 22:41, 9 November 2018 (UTC)

BTW: I'd suggest you read up on WP:CIVIL and adjust your tone accordingly. SMcC is a well-respected member of the community (he couldn't be as active in the drama-filled parts of the project, like MOS, as he is and yet have an almost-clear block log otherwise), and dismissing everything he says by repeatedly (six times and counting) referring to it as "nonsense" is inappropriate. Admittedly, some of that is from eight months ago, before you "apologise[d unreservedly"] for similar comments elsewhere, but the continuity of dismissive language between March and November is actually more concerning given that an apology and an apparently sincere statement of intention to do better came in the interim. Hijiri 88 (聖やや) 23:58, 8 November 2018 (UTC)

If I understand User:SMcCandlish correctly, his concern would be addressed by changing "This includes published works in all forms, such as..." to "This can include published works in all forms, such as...". Worded this way, an editor would be less like to misinterpret the criteria as saying that every review is significant.

That aside, this seems like a molehill. Is there a large volume of AfDs that are being closed as keep because a book received two reviews and no other coverage anywhere? Argento Surfer (talk) 16:22, 8 November 2018 (UTC)

While I still don't think any change is necessary, Win Bigly is my example of a book that meets the guideline enough to survive AfD but probably shouldn't be a stand-alone article. power~enwiki (π, ν) 17:06, 8 November 2018 (UTC)

But it hasn't actually survived AfD. Argento Surfer (talk) 17:21, 8 November 2018 (UTC)

SMcCandlish, could you confirm that what Argento Surfer suggests above would address the issue you raised above (one of them, anyway)? If so, is there any opposition to such a change? If we are agreed that a review is not automatically non-trivial in every instance, this shouldn't be controversial, I imagine? — Rhododendrites ^talk \\ 04:10, 9 November 2018 (UTC)

@Rhododendrites and Argento Surfer: Sorry, I didn't see this until just now due to all the noise injected by a single editor. Anyway, that would definitely help, and it's better than the weird "Published works includes ... published works" edit that James500 reports making, below, and which I hope has been reverted. And, yes, there are in fact AfDs closing as keep on the basis of "has two reviews" or I would not have come here and raised the issue. I have no idea how often this happens as a statistical matter (I spend as little time at AfD as possible), but we have at least one extreme inclusionist editor here making "Keep because reviews" votes in AfD at every opportunity, without regard to the nature or depth of the review, reputation of the publisher, or actual encyclopedic notability of the work. The ambiguous wording of this guideline line-item makes it easy for this argument to be prima facie accepted without any analysis, almost as a kind of thought-terminating cliché. It's not like I just wake up in the morning and think up irrelevant copyedits to propose for my own entertainment.  :-) In closing, while I support Argento Surfer's specific revision idea, I'm not certain it will go far enough. We may need an actual short sentence on reviews and that they are not all equal. The substantive parts of this thread already provide everything we need for that.  — SMcCandlish ☏ ¢ 😼  12:57, 11 November 2018 (UTC)

Are reviews the only aspect you're concerned with, not documentaries/newspaper articles/best seller lists? Perhaps the reviews could be moved to a separate bullet point and given a higher number than two? Argento Surfer (talk) 13:46, 13 November 2018 (UTC)

I went ahead and WP:BOLDly made the proposed change several hours ago: [11]. James500 (talk) 04:16, 9 November 2018 (UTC)

Gah. Can't believe I took this as face value. No, you did not, unless you misplaced this text. The proposed change by Argento Surfer was to change "includes" to "can include." Not to change "This includes published works of all kinds" to "'Published works' includes published works of all kinds". I've undone that bizarre business and added the above. — Rhododendrites ^talk \\ 16:05, 11 November 2018 (UTC)

I want to amend my comment above. In rereading the edits now, I can see that changing "this" to "'published works'" could be understood as clarifying, in the sense of referring back to a specific part of the preceding sentence, to remove the possibility that "this" includes "non-trivial." Still, I think the phrasing suggested above is cleaner. — Rhododendrites ^talk \\ 01:19, 12 November 2018 (UTC)

• Agree with McCandlish here. For academic works, "two reviews" is ridiculously low as a notability threshold. I've worked in academia myself, and in my field pretty much every decently published academic monograph, down to mere Ph.D. theses, was getting two or more independent reviews. I've myself published works that would meet this criterion with flying colours, and I've also written (independent, in-depth, properly published) reviews for plenty of others. But I'd never dream of the idea that either my own work or those others I reviewed would warrant a Wikipedia article. Fut.Perf. ☼ 10:45, 12 November 2018 (UTC)
• Comment I think it's very helpful that James500 prefaces so many of his posts with warnings such as, "This is utter nonsense from start to finish". Saves a lot of time and unnecessary reading. EEng 17:39, 13 November 2018 (UTC)
• I think the current standard is appropriate. The reviews should be from reliable sources. I would argue, were I to participate in such a discussion, that Publisher's Weekly reviews should not be used for notability unless they are starred reviews (also would suggest it is RS for things like sales numbers as reported by the publishers). Other trade publications like Library Journal are RS for reviews. Certainly indepth reviews of the kind found in many major newspapers are helpful in establishing notability. I say all this as a reminder that whether a given source is RS and helpful in establishing notability is different from whether a SNG is set at the right "presumptive" level. As a new page patroller I feel we are far more overrun with mediocre or poor articles in several areas covered by SNGs -NBOOK is not one of them. It would therefore not be the place I would look to tighten our presumptive notability standards becuase it seems to be a solution in the need of a problem - the limited examples shown in this discussion not yet being convincing. Best, Barkeep49 (talk) 01:37, 15 November 2018 (UTC)
• Oppose as reviews in a reliable source indicate notability but they should be full reviews to qualify not capsule reviews and the source is also qualitive. For example a full review in a national publication such as the New York Times or major regional publication such as the Chicago Tribune would be more indicative of notability than a book review of a local author's work in a small town newspaper. The publishers weekly and library journal are acceptable as they are independent but I have doubts about reviews in publications that mix paid and unpaid reviews such as Krircus Reviews for example. In summary, reviews can be significant coverage but not all reviews, regards Atlantic306 (talk) 22:18, 21 November 2018 (UTC)

@Atlantic306: So ... you support changing the guideline to be more discriminating? The actual wording of your comment implies you agree with SMcC, myself and others, but the first word of your comment doesn't gel with that -- am I missing something? Hijiri 88 (聖やや) 08:40, 6 December 2018 (UTC)

• Oppose changing other than clarifying how "in detail" (per WP:GNG) could apply to book reviews. Not sure why is it claimed that film articles do not follow anything like this criteria; there is similar criteria at WP:NFILM that supports using film reviews in general but excludes "capsule reviews" for not being in detail. Furthermore, WP:GNG clearly defines "independent", and it has nothing to do with "the same narrow field". If that is a problem in some areas, the fact it says "two or more" provides latitude to seek a third and fourth review since WP:GNG also says, "There is no fixed number of sources required since sources vary in quality and depth of coverage..." It's not two reviews, full stop. If the reviews are insufficient in some way, that can be challenged. Erik (talk | contrib) ^{(ping me)} 18:20, 2 December 2018 (UTC)
• Oppose Still seeing no significant evidence that this is a problem that needs fixing. Andrew D. (talk) 14:06, 4 December 2018 (UTC)

More needless obstruction from the guy who wrote this before ignoring the follow-up questions. I wish I could say I was surprised. Hijiri 88 (聖やや) 08:40, 6 December 2018 (UTC)

• That's not relevant evidence because Mottainai is not an article about a book and it's not a problem either. Andrew D. (talk) 19:32, 6 December 2018 (UTC)

I was commenting on your general tendency to engage in obstructive behaviour when your "opponents" are trying to accomplish something, regardless of whether it would benefit the project, and on your refusal to constructively respond to criticism of the same; your changing the subject actually supports me on the latter point. Hijiri 88 (聖やや) 01:22, 7 December 2018 (UTC)

• Support I wasn't aware we were !voting on anything in particular at this point, but if we are I might as well state re-affirm that I basically agree with SMcC's assessment of the situation, as I did the last time he and I found ourselves talking about reviews as sources. Hijiri 88 (聖やや) 08:40, 6 December 2018 (UTC)

While we don't agree about the idea, I think we are in definite agreement that we weren't at a voting stage for this. It feels like we'd need a proper RfC to make this change. Best, Barkeep49 (talk) 19:11, 6 December 2018 (UTC)

• Oppose I find the present guidelines workable and sufficient. All these objections to Kirkus Reviews and Publishers Weekly reviews do not meet my experience in finding reviews from those publications. They review books worthy of a review, worthy of reading. They do not review every book that is published. Library Journal usually has their reviews written by a named person at a specific library. When life is good, there are reviews in newspapers, and as newspapers get stronger, those may appear more often. So I am agreeing with the people who say that there is no issue here, nothing to change, the current guidelines are workable and working. --Prairieplant (talk) 12:16, 4 February 2019 (UTC)

  On what planet are newspapers "getting stronger"? EEng 13:01, 4 February 2019 (UTC)−

• For those who don't see the issue, imagine a book whose only coverage is in Publishers Weekly and Booklist (i.e., trade publications designed to help library acquisition staff), each with single-paragraph reviews (<250 words) that are mostly synopsis, book jacket stuff. Would these two sources be "non-trivial" coverage for the purposes of this guideline? And likewise, would these two sources be sufficient for the guideline (and the GNG) as "significant coverage" for independent notability warranting a separate article? How could that be the standard while WP:NFILM excludes capsule reviews? czar 08:34, 21 December 2019 (UTC)
• Agree I have seen many instances in which this criteria was gamed to auto-confer notability to a book. We should either remove it or clarify the meaning of "non-trivial" to indicate that the outlet reviewing the work is not a trivial outlet, not that the review itself is substantial. We have had numerous instances where an RPG rulebook has been successfully defended from deletion on no other basis than it had a two paragraph review on a couple of random game blogs (and no other mention - even a mere mention - in any other source known to man). This is in direct contravention of the spirit of Notability. Chetsford (talk) 17:21, 21 March 2020 (UTC)
• Oppose - the "spirit of Notability" as far as I can tell is not that the topic must be important, but that it must be independently documented. Whether a documented subject should have its own article, or is "important enough" to be discussed on WP, is to be decided on the basis of WP:NOT and WP:DUE, not on the basis of Notability. I see no problem, therefore, with the potential result that every print publication that had received two independent reviews in RS - even capsule reviews, and even in specialized sources - should be deemed to meet NBOOK and the GNG. If consistently understood, this principle saves time at AfD and allows questions of "importance" to be sorted out in merger and content discussions, as they should be.

To respond directly to SMcCandlish, the paradigmatic case of "trivial" coverage is a passing mention, and no capsule review - no matter how brief - could constitute a passing mention, because by definition it is primarily concerned with the publication reviewed. Since any RS review does in fact provide documented information about the publication reviewed, there is no GNG-compliant reason why it would not contribute to Notability. And per WP:NOTPAPER, there is no real rationale for gatekeeping the articles WP carries on print publications, as long as they are written to comply with NOT. Newimpartial (talk) 19:05, 8 April 2020 (UTC)

• Oppose. If there's a problem here, it's the vast number of book articles that are based primarily on the book itself, or worse that copy-and-paste the table of contents directly into the article. Articles on books should be based primarily on what independent sources say about the book, and notability should be about the existence of adequate sources on which to base an article (usually, reliably published reviews), not on our own subjective judgements of what makes books significant. If people are counting blog posts as reviews, they shouldn't be; that's an issue for WP:RS. And if you want to tighten up rules, the ones I think more in need of that are the ones that allow plot summaries to be completely unsourced, or that allow wholesale copying and pasting of tables of contents. —David Eppstein (talk) 20:01, 8 April 2020 (UTC)
• Oppose The above criterion is a rebuttable presumption. Say someone makes an article on a book with only 2 short reviews that backs it up, and no other apparent sourcing. If you see this and make the effort to find that there is no other significant sourcing for it such that it is impossible to improve the article, then you have a means to challenge the presumption of notability and take that article to AFD explaining where you looked to show why there are no other sources. This puts the onus on those wanting to keep to identify further sourcing; they cannot just fall back on this criteria at that point. So it may be easily game-able but it is also not a guarantee to allow an article to be kept. -Masem (t) 20:21, 8 April 2020 (UTC)
• Support. I do not usually edit much in this area but was shocked to see how expansive this part of WP:NBOOK actually is. It would encompass pretty much any non-fiction book printed by any University Press. I have just written two reviews for respected peer-review journals myself and can tell you that the grounds for inclusion usually have far more to do with the publisher choosing to send a review copy to the journal - although independent in the literal sense, there is a clear inter-dependence. I also do not see how a flurry of post-publication reviews can meet WP:SUSTAINED. —Brigade Piron (talk) 09:17, 19 November 2020 (UTC)

  Sustained is a criteria that explicitly only applies to events. I agree with all of your facts but not the conclusion. Why not have articles on any university published book? Best, Barkeep49 (talk) 10:43, 19 November 2020 (UTC)

• Oppose - I didn't think people were still !voting on this, which clearly isn't an RfC in any case, and we have had multiple apparent !votes from the same editor, etc. But I might as well record my essential reaction to the proposal, namely that the current version of NBOOK is less gameable than either the proposed change or the GNG itself, so the change would not be an improvement. Newimpartial (talk) 12:44, 19 November 2020 (UTC)

Anti-Zionism on Campus is an article about a book that appears to have notability primarily of the two book reviews rule. I nominated it for deletion before knowing about this rule, but it was kept which clearly is correct in light of this rule (I hope I'm not breaking the wp:canvas rule here because the afd voting is over). ImTheIP (talk) 09:09, 19 August 2020 (UTC)

• Oppose for now. I know that this is old but I want to put my two cents in. This would have to be made extremely more clear before it could pass. I will say that nowadays books tend to need more than 2 reviews to pass notability guidelines unless those two reviews are in very, very good sources such as an academic/scholarly journal, an outlet like the New York Times, or written by people seen as very authoritative sources in their given field or genre. For example, I wouldn't create an article on a book if the only sourcing that exists were trade reviews in PW and Kirkus, as I don't view that as strong enough coverage. Now a publication like the Horn Book Guide would be far stronger because while it is brief, the HBG is pretty selective about what works they actually cover.

A vaguely worded policy would also pose an issue as to what would be considered in-depth. Not everyone holds the same opinion on this, so it's theoretically possible for someone to argue that a review in the NYT by a notable critic wouldn't be in-depth because the critic reviewed three books in the same article, giving each book about the same amount of text they'd have in say, a PW review. There are also reviews in scholarly and academic journals that can be similarly brief, even if the next review in the journal is 1-3 pages long. This change didn't really cover how we would define something as in-depth, which as shown can be hard to concretely define to where it can be easily used as a standard.

This also could have a domino effect. If we do something that could potentially consider reviews - at least some of them - to not grant notability, then this could in turn have an impact on other notability guidelines that use reviews as a sign of notability. It could result in thousands of articles getting deleted despite otherwise being seen as notable, such as articles on films, games, anything that could conceivably receive a review.

I also have to say that very, very few books gain coverage in even the trade reviews. I'd wager that about 5% of books actually gain enough coverage to justify an article and only slightly more actually even get weak coverage from trades. I'm not saying that this should make all trade reviews usable, just saying that the guideline is far less inclusive than one would think. As someone who regularly creates book and author articles, it's far more common that I don't find enough coverage than I find enough to establish notability. I don't even start compiling sources on my sandbox unless I see enough in the first 1-2 Google search pages to make me think that the person or book could be notable - and I've had to clear out more than a few times because that promise never came to fruition.

I think a better alternative would be to single out which review outlets shouldn't be used and to create a list similar to WP:VG/RS that we can point people to. I don't believe that WP:BOOKS has ever had a list like that, or at least one that's easily found. An argument could be made that we should up the number to 3, as that would actually cut the number of book articles that could be created dramatically, but I think it would be better to have a good discussion about which sources should be used and which shouldn't. 04:35, 20 November 2020 (UTC)

I think that's a splendid idea ReaderofthePack. Best, Barkeep49 (talk) 04:37, 20 November 2020 (UTC)