Wikipedia talk:Identifying and using self-published works

Re-org

I just did a pretty big re-org of this article, but changed no meaning (I think). The old version had many redundancies, and the points jumped around quite a bit. I tried to put all the relevant info in the right places. MichaelBluejay (talk) 03:34, 20 January 2014 (UTC)

I looked through your changes, and I think you improved the page. Thanks, WhatamIdoing (talk) 05:25, 20 January 2014 (UTC)

Publisher as employer

About this passage: If the author works for a company, and the publisher is the employer, and the author's job is to produce the work (e.g., sales materials or a corporate website), then the author and publisher are the same. Isn't this also true of news organisations? If the author is a staff writer for The New York Times, the NYT is their employer and also their publisher, and it's the author's job to produce the articles for them. There's probably a good way of phrasing this that highlights the difference between this scenario and sales materials etc., but I can't think of it right at the moment. — Mr. Stradivarius ^{♪ talk ♪} 02:56, 25 September 2014 (UTC)

Dictionaries and similar sources seem to get around this by merely announcing the "traditional publishers" such as newspapers aren't self-published. WhatamIdoing (talk) 01:57, 12 December 2019 (UTC)

experts in the relevant field

Self-published sources may be considered reliable when produced by an established expert on the topic of the article whose work in the relevant field has previously been published by reliable third-party publications.[1] Take care when using such sources: if the information in question is really worth reporting, someone else will probably have done so.

This seems rather narrow. What about authors whose work has not been published by reliable third-party publications, but whose self-published work has been cited by reliable third-party publications? It seems rather absurd to say a source is inappropriate when the most respected authorities are citing them. I'm specifically thinking here of Leigh Rayment's website, which is used by Hansard, despite only being published on Rayment's personal website. john k (talk) 16:35, 17 December 2015 (UTC)

do you mean an MP cites the website & Hansard reports the MP's remarks? In general MP's are not "respected authorities" of the sort we look for. Rjensen (talk) 18:44, 1 October 2019 (UTC)

I suspect that john k meant that Hansard uses the website to make sure that its own publications are using the correct titles.

The answer to the question, though, is that WP:V imposes this on us. This page is only reflecting the central policy page. WhatamIdoing (talk) 02:00, 12 December 2019 (UTC)

Government source

If a government body publishes a report written by that body, and puts a copy of that report on its website, is that a self-publishing source? Despite the tendency of some governmental bodies toward propaganda, that seems like a very reliable source, especially if the source is mentioned in the text ("The FBI reports that... [ref FBI website]").

Is it different if the government in question is not universally recognized, like Taiwan? MathEconMajor (talk) 16:44, 19 October 2016 (UTC)

@MathEconMajor: I think you misunderstand. Although this essay doesn't specify it, the intent is to identify media publishing for which there is no fact-check or editorial review. It's not discussing a lack independent or impartial sources. The assumption is that government entities aren't publishing what is essentially the opinion of one civil servant or political apointee but the official release from that entity. It's not independent or impartial but it's not self-published. We're concerned with cranks putting out screed that gets used as a source here. Chris Troutman (talk) 17:43, 19 October 2016 (UTC)

Referring to a self-published improvement of a published work?

There is a published book that was self-republished by the author. Should I refer to the official publisher or to the Amazon account of the author in the reference URL? Amazon version contains some improvements (such as fixed typos) but is self-published.

The main trouble is that if I put a reference to the better self-published version, it could be removed as self-published. I need to note somehow that the self-published version is authoritative because it is a version of an officially published work. How? Should I add a note like "republished from ..."? How to add this note to the {{cite book}} tag? Should I mention in the note that the work was improved? What exactly to say?

(a copy of Wikipedia:Village_pump_(policy)#Referring_to_a_self-published_improvement_of_a_published_work?) --VictorPorton (talk) 13:08, 1 October 2019 (UTC)

How about two cites in one footnote: See XYZ1 (2003) page 99 reprinted as XYZ2 (2015) page 99 Rjensen (talk) 18:41, 1 October 2019 (UTC)

SPS

This may be of interest [[1]].Slatersteven (talk) 09:35, 3 November 2019 (UTC)

Lack of reviewer

Currently we say:

"Self-published material is characterized by the lack of reviewers who are independent of the author (those without a conflict of interest) validating the reliability of content"

Is that how it is characterized, or simply one of the concerns? I ask because earlier it states that a self published source is easily identified by being written and published by the same person (which is how I'd generally see it). That seems independent as to whether or not the author seeks some review of their work. Many self published authors will share their manuscript or other material with others before publication, but doing so doesn't seem to make it non-self published. Is there a better way to word this? - Bilby (talk) 22:03, 8 November 2019 (UTC)

Bilby, I think that this situation is characteristic of the kind of self-published works that the English Wikipedia doesn't want used – not of "self-published works", but of that subset of "self-published works that are likely to be unreliable". WhatamIdoing (talk) 02:02, 12 December 2019 (UTC)

All good. I might look at a slight reword to make that clear. - Bilby (talk) 04:27, 30 December 2019 (UTC)

Tertiary sources

User:Abductive, you have twice made changes to the description of a tehrtiary source that do not seem obviously correct. A timeline can be a work of analysis, and the items in it may be selected to prove a point. That would make it a secondary source. A ranked list of websites also means that it may be a work of analysis and selection. The main characteristic of a tertiary source is that it compiles information from other sources, but "The information is not analyzed or interpreted."

The classic example of a tertiary source is a phone book. The example needs to be clean. A "top 10 websites" list might be a tertiary source, but it might not be. That's why I prefer the example of "a simple list of tourist attractions": a "simple" list of businesses is a clean example. WhatamIdoing (talk) 04:07, 30 December 2019 (UTC)

• Wikipedia's own article says textbooks and encyclopedias are tertiary sources, and says "Almanacs, travel guides, field guides, and timelines are also examples of tertiary sources". There is no mention of a phone book, and I doubt you can find a reliable source that makes that claim. So, a blog saying a "simple list of tourist attractions" is not tertiary enough to use as a good example, since it it is not a travel guide (which assumes a certain level of curation). If we are trying to give good guidance for Wikipedia users, it is best not to have any ambiguity. "A simple list" is too diary-like (or could be interpreted that way), so that is why I thought that a local historian (many people have one in their family or group of friends) putting together a self-published timeline of local history would be a plausible example. A self-published travel guide would be acceptable to me, but less plausible. Abductive (reasoning) 05:05, 30 December 2019 (UTC)

Look at https://www.sciencedirect.com/topics/social-sciences/tertiary-source and http://ocean.otr.usm.edu/~w146169/2cReference.htm Both of them are talking about "Reference Sources", which they say can be either secondary or tertiary. Both of them list telephone directories as examples.

The second defines "Tertiary sources" as "Guides or pointers to material or information" and "Finding aids". I think that's what a telephone directory is. The timeline by the amateur historian fits their definition of secondary source better: "Material which has been modified, selected, or rearranged for a particular purpose or audience".

The addition of "self-published" to these examples is a serious problem with respect to the purpose of this page. One of my major goals in writing this page was to help editors understand that primary/secondary/tertiary are completely independent of who published the content. If a given bit of content is tertiary when you post it on your personal website, then it's still tertiary when McGraw–Hill prints ten thousand copies of it. It may not be reliable if you self-publish it (that's the point in the WP:SECONDARYNOTGOOD section), but its status as primary/secondary/tertiary does not depend upon whether it's self-published or traditionally published. WhatamIdoing (talk) 00:22, 2 January 2020 (UTC)

My take is clearer. Yours requires mental gymnastics and equating a phone book to an encyclopedia. Abductive (reasoning) 09:28, 2 January 2020 (UTC)

Perhaps we should use "an encyclopedia article" as the example, then. That's a clean example. WhatamIdoing (talk) 21:20, 2 January 2020 (UTC)

Is it self-published? Abductive (reasoning) 05:33, 4 January 2020 (UTC)

Well, ours is, although we're not using blog software. WhatamIdoing (talk) 03:50, 23 February 2020 (UTC)

Requested move 7 March 2020

The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this discussion.

The result of the move request was: not moved. Clearly no consensus to move this. Number 57 13:47, 12 April 2020 (UTC)

---

Wikipedia:Identifying and using self-published works → Wikipedia:Self-published works – The title is more concise. Interstellarity (talk) 23:53, 7 March 2020 (UTC) —Relisting. Steel1943 (talk) 20:27, 24 March 2020 (UTC)

• We do tend to include an "instruction" in the title such as Wikipedia:No personal attacks and Wikipedia:No original research but Wikipedia:Reliable sources was changed. Crouch, Swale (talk) 18:05, 10 March 2020 (UTC)
• Interstellarity has proposed multiple page moves along these lines. For some I don't care, but in this case, I prefer the longer title. I am concerned that the shorter title will give editors the idea that it's some sort of list of self-published works, along the lines of Wikipedia:Reliable sources/Perennial sources. Also, the shorter title does not do a good job of indicating that there is advice for how to use self-published works. This page is more than a definition of a term. WhatamIdoing (talk) 05:26, 11 March 2020 (UTC)

---

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Definition is misleading

I propose a change to the WP:FIRSTSENTENCE, because the current version if taken literally is inaccurate and at odds with the results of innumerable discussions at boards and talk pages. Such things as lack of editorial oversight are commonly understood to be part of it. Added this at the top, because the earlier version apparently was not clear; Mathglot (talk) 21:36, 7 July 2020 (UTC)

For years, the definition of SPS has been stated or implied here to be that where the author and publisher are the same. In 2014 in this edit of 03:31, January 20, this was formally added to the WP:LEADSENTENCE:

Self-published works are those in which the author and publisher are the same.

but this only formalized what had already been stated in other words, or in bullet points, in the succeeding paragraphs prior to that date. However, this definition, in my opinion, is either inaccurate, or meaningless for the purposes of verifiability and this encyclopedia. If A.G. Sulzberger decides to write an Op-ed piece in the New York Times, that does not make the New York Times an SPS. The implications of calling the Times an SPS are obvious.

Dozens of discussions at the encyclopedia appear to act contrary to this definition, or at least, ignore it. In reading various discussions at article Talk pages and the RSN archives, my impression is that we have a sense about SPS that is more about lack of editorial control and fact-checking, often run by a lone proprietor or small group, with or without the occasional invited guest; i.e., blogs and similar. I could list many discussions where this PoV is taken and seemingly accepted by all, and it was a surprise to me to find this definition here, so clearly out of sync with those discussions. In my opinion, the definition here needs to be replaced. Mathglot (talk) 19:26, 7 July 2020 (UTC)

It seems to me Mathglot has skipped some steps in his exposition, which prevents me from following it. Using a real example, Jonathan H. Adler submits an op-ed piece, "This is the Real John Roberts" to The New York Times, which publishes it on their website today, July 7, 2020. Johnathan H. Adler is the author. The New York Times is the publisher. We don't know the details of the interaction between these two entities; maybe there was none of the revising and negotiating one would expect between a reporter and an editor, but there certainly was editorial oversight in the form of the staff members responsible for the op-ed section receiving a large number of submissions, only the best of which are published. Jc3s5h (talk) 20:22, 7 July 2020 (UTC)

Agreed. The point isn't that NYT is an SPS, it's that it is not by common agreement of editors here about what an SPS really is. However, by the definition in the lead sentence, our common agreement about that is wrong. Something has to change. Mathglot (talk) 20:29, 7 July 2020 (UTC)

Note: related discussion is ongoing at WT:Verifiability. Mathglot (talk) 20:33, 7 July 2020 (UTC)

I have concern about some recent versions of this explanatory supplement because there is no single definition of "author" and "publisher" among those publications which take the trouble to cite sources. Cases could be contrived where, by some definitions, the author and the publisher are the same but the piece isn't truly self published. But I don't think there are any reasonable definition of "author" and "publisher" that would make Johnathan H. Adler and The New York Times the same. Jc3s5h (talk) 20:41, 7 July 2020 (UTC)

Once again, agreed. I think I haven't been clear. This is a proposal to change the WP:FIRSTSENTENCE of this page, to reflect what appears to me to be consensus over a wide range of discussions that an SPS is not what is described here, but rather, involves such things as lone proprietorship, lack of editorial backup, lack of fact-checking, and so on. But that is a significant change to this policy page. I have no quibble with your comments about the Adler piece, or what it implies. Mathglot (talk) 20:53, 7 July 2020 (UTC)

@Jc3s5h: Sulzberger and Adler aren't interchangeable as authors, since Sulzberger is publisher for the NYT and Adler isn't. I agree that it's sometimes hard to identify author and publisher. I gave a different example in a related discussion (here: [2]), where I don't know whether the publisher for Just Security articles is Just Security vs. the Reiss Center on Law and Security vs. New York University School of Law vs. some person or entity whose name I don't know. -- FactOrOpinion (talk) 21:29, 7 July 2020 (UTC)

To take Jc3s5h's NYTimes example, let's use Forbes Contributors as a counter example: where one can apply the same labels between the contributor and Forbes.com as "author" and "publisher" we absolutely do know that Forbes.com does no oversight of what is published, and for all purposes we consider Forbes Contributors a SPS (or at least unreliable). Same model would apply to a work like Medium.com. Its why the definition should be based on how much "editorial or publishing oversight" there is between the editor's completion of the piece and its publication. If the editor presses a button and its published with no checks, that's self-published, in essence. --Masem (t) 20:59, 7 July 2020 (UTC)

I would agree with Masem provided editor is replaced with author. Within Wikipedia we adopt the jargon that the person who writes or refines an article is an editor, but outside Wikipedia, a person who writes an article is an author. Editors sometimes are hired by the author and are an agent of the author; other times they're hired by the publisher and are an agent of the publisher. Jc3s5h (talk) 22:23, 7 July 2020 (UTC)

Good analysis. --David Tornheim (talk) 04:49, 8 July 2020 (UTC)

The problem with the "oversight" standard, aside from it being a standard that is never used anywhere in the real world, is that (a) we don't actually know how much oversight is involved in most publications, and (b) it substantially narrows the low-quality sources constrained by SPS. If "editorial oversight" – which can be obtained for money – is all an author needs to get exempted, then we'll get fringe theories promoted via self-published books and websites. The forewords to books written by would-be Wikipedia exploiters will sprout paragraphs about how grateful the author is for all the editorial oversight provided by professional editors (and lawyers), and we'll have unbalanced garbage in articles. SPS is meant to stop that, not to let people sidestep it by a claim that the editors they hired to fix up their book means that they're in the same league as someone published by an independent, traditional publishing house. WhatamIdoing (talk) 23:42, 8 July 2020 (UTC)

I think "editorial oversight" means the overseer can and will prevent the work from being published by the publication that the overseer is affiliated with, unless the work is of sufficient quality. An editor or vanity press hired by the author can offer suggestions, but the author can reject the suggestion and proceed with the publication. That isn't any kind of oversight, editorial or otherwise. Jc3s5h (talk) 23:49, 8 July 2020 (UTC)

In which case, we're functionally back to "the author and the publisher are [or aren't] the same". WhatamIdoing (talk) 23:55, 8 July 2020 (UTC)

• Thanks to Mathglot for bringing this up. I agree the first sentence surprised me too. It certainly makes sense for a website or blog when a single person owns the site and publishes whatever they want, whenever they want, with no one else's oversight (writer = editor = publisher = owner). I agree with Masem's statement.

The big problem with the WP:LEDE's first sentence is the precise use of the word "publisher" is likely unfamiliar to the average reader, who finds the word more ambiguous. Although the meaning of "publisher" is explored in detail and defined in WP:USINGSPS, I do not believe the average reader coming to WP:USINGSPS would be aware of it, and would be confused, as I was. Before coming to this discussion, if someone had suggested that Facebook, blogspot.com, webhosting sites, iMDB, or Huffington Post (re: contributors) act as "publishers" of things people post, I probably would have said that sounds about right. Yet it does seem in the industry, the answer is "No. Those are web-hosts, not publishers."

It reminds me of a strict math definition such as vector space "is a set V on which two operations + and · are defined, ...[such that...."] The lay person coming to that definition without a clear a priori understanding of the meaning of precise definitions of "set", "operation", etc. is unlikely to appreciate the rigor of this precise definition. --David Tornheim (talk) 04:49, 8 July 2020 (UTC)

I'm glad this discussion is happening here. Also, let's note that the current definition appears in multiple parts of the article, not just the lede. Anyway, Mathglot, what do you suggest as a preferred definition? Since a precise definition is hard to pin down, I'm reminded of I know it when I see it. -MichaelBluejay (talk) 10:40, 8 July 2020 (UTC)

In a related discussion on the RS/N, Kyohyi pointed out footnote 10 [3] on WP:V, and I’m wondering if “Self-published material is characterized by the lack of independent reviewers (those without a conflict of interest) validating the reliability of content” moves this discussion forward re: an alternative definition.

Whatever the conclusion of this discussion here, it has implications for the text of Wikipedia:Verifiability#Self-published_sources. That text needs to be revised regardless, as the footnote reference for footnote 10 — “if the information in question is suitable for inclusion, someone else will probably have published it in independent reliable sources” — isn’t substantiated by footnote 10 and likely isn’t true; see the paragraph here [[4]] contesting an earlier version of that WP quote, "if the information in question is really worth reporting, someone else will probably have done so.”

Also, as Jc3s5h noted above, we may need to help readers understand what’s meant by “editor,” as the WP-specific meaning of “editor” is distinct from its standard meaning, so it may be important to clarify which meaning is intended. (If people aren’t familiar with the concepts of Register_(sociolinguistics) and Code-switching, they’re relevant. Wikipedia has a linguistic register.) We may also need to help readers understand what’s meant by other terms (e.g., “publisher,” as noted by David Tornheim above). -- FactOrOpinion (talk) 12:35, 8 July 2020 (UTC)

What will probably be helpful is that once we establish what we want the "definition" to be is then to list out the examples of where the line is clearly drawn - eg one's social media page and Forbes Contributors are an SPS due to the lack of an editorial presence before the author self-publishing, but (at least based on this discussion) sites like SCOTUSBlog are not SPS due to the presence of an editorial board. There's still a grey area (Quackwatch) but that will help clarify the definition. --Masem (t) 17:10, 8 July 2020 (UTC)

Mathglot, if Sulzberger puts something in his paper, then why shouldn't Wikipedia treat that like a self-published source? WhatamIdoing (talk) 21:10, 8 July 2020 (UTC)

(Nitpicky side note: The publisher of a US newspaper writes an editorial, not an op-ed. Op-eds are these things that you let people with the "opposite" view point write for you. But don't let that distract you from explaining why Wikipedia should treat an opinion piece written by a person who literally cannot be refused its publication the same as an actual news story, written by a journalist, whose publication can be vetoed by both the editorial staff and the publisher.) WhatamIdoing (talk) 21:15, 8 July 2020 (UTC)

In looking at pages about SPS, or discussions at RSN or elsewhere, I mostly see allusions to the work, so that SPS is discussed as an attribute of where it appears, not who wrote it. Questions such as, "Is Just Security a SPS", or, "Is Infowars a SPS". Or even theoretically, "Is The New York Times a self-published source?" I guess if you think of SPS as pertaining individually to each article/editorial that appears, or attaches to its author rather than the work, then you could ask the question you did about Sulzberger, and then—what? "New York Times is an SPS on its editorial page"? I'm just not sure how this advances the discussion, so I don't know how else to respond. Maybe someone else will have another angle on it. Mathglot (talk) 21:53, 8 July 2020 (UTC)

SPS status can only be determined by looking at who wrote it, and who made it available to the public. When different parts are written by different people (the NYT and Quackwatch examples) or made available to the public by different people (a university website might operate this way: the publicity-related departments control the general pages, but the professors get to write whatever they want in designated areas), then you would have to evaluate each part separately. WhatamIdoing (talk) 23:32, 8 July 2020 (UTC)

Let's make some progress on this. We need to (1) Hammer down a good definition of SPS, and (2) Include it both here and at WP:V.

Are we going with SPS = (author = publisher) or something else?

Maybe the problem is that (author=publisher) sometimes works but doesn't always work? Is the solution to add a modifier (like "generally"), or to define exceptions, or both? -MichaelBluejay (talk) 17:04, 10 July 2020 (UTC)

Just a heads up that Michaelbluejay has created a "Let's define self-published sources" section on the WP:V talk page (here: [5]) to see if we can resolve the definition on both pages. I've tried to incorporate the issues that various people raised and have proposed "Material is self-published if the person(s) who controlled the material’s creation = the person(s) who controlled whether the material was made available to the public." If anyone has another proposal and/or wants to tweak my proposal, please join in there. -- FactOrOpinion (talk) 20:58, 23 July 2020 (UTC)

Changes to the lead

Changes to the lead

Self-published works are those in which the author and publisher are the same. There are many bestsellers that are or were self-published works, such as The Joy of Cooking and some all-time bestsellers like Fifty Shades of Grey. Therefore, self-published works are acceptable in Wikipedia so long as certain conditions are met.

Self-published works are those in which the author and publisher are the same. Anyone can self-publish information regardless of whether s/he is truly knowledgeable about the topic in question. Therefore, self-published works should be examined carefully for acceptability.

This is a pretty significant change in tone – from self-publishing being relatively common, to self-publishing being suspicious. Is that sense of suspicion what we want, for a page that's talking about how to use and identify them? WhatamIdoing (talk) 21:24, 8 July 2020 (UTC)

Thanks for posting this. Just to be clear, WhatamIdoing is not proposing a hypothetical above to provoke discussion; these are actual changes made in the last few days, by Michaelbluejay (talk · contribs). I echo WaId's question. Mathglot (talk) 22:06, 8 July 2020 (UTC)

• On the other hand, Michael’s change does bring this supplement closer to the cautionary tone found at WP:V and WP:RS. And I think it is good that editors are told to be CAUTIOUS about self published sources.

May I suggest that the lead here start by using the same language that is used in the actual policy/guideline pages, and then expand on that. Blueboar (talk) 22:29, 8 July 2020 (UTC)

I'm not satisfied with either option. I think we could do better. I don't think that the "anyone can publish, so assume none of them actually know anything" is a good approach, but I don't think that the commercial success of a work of fiction is really relevant (and I probably wrote that unfortunately constructed sentence). I think we want something in the middle. WhatamIdoing (talk) 23:19, 8 July 2020 (UTC)

I think the statement "SPS should be examined carefully for acceptability" is pretty reasonable and non-controversial. And it's certainly a different tone from WP:SPS which says bluntly that SPS are "largely not acceptable", which seems far too dismissive. Here again, I'm glad this discussion is happening, because whether we go with the new or old lede, both are at odds with WP:SPS, and we need to bring both articles into alignment. And yes, I removed the reference to 50 Shades of Gray because it seemed to be a poor example of a SPS, because I can't imagine what statement it could be used to support (besides the work's own existence, which is true even for non-best-sellers). -MichaelBluejay (talk) 14:52, 9 July 2020 (UTC)

Preference for non-SPS

This has been removed:

Self-published sources are largely not acceptable on Wikipedia, though there are exceptions. And even though a self-published source might be acceptable, a non-self-published source is usually preferred, if available. Examples of acceptable sourcing of self-published works:

I think that this page should state that, very broadly speaking, non-self-published sources are usually preferred for most purposes. What do you think? WhatamIdoing (talk) 21:36, 8 July 2020 (UTC)

This has been incorrectly transformed into an "always" statement in the new ===Unacceptable=== section: "A non-self-published source that verifies the same information is available. These are always preferable." Non-self-published sources are not always preferable. Sometimes the original is the authoritative and therefore best possible source. I do not think that this new catchall ===Unacceptable use of self-published works=== section is an improvement to this page. WhatamIdoing (talk) 21:41, 8 July 2020 (UTC)

I Agree... Not an improvement. When it comes to sourcing, we should avoid “always” and “never” statements. A LOT depends on the specific article, and the specific statement you are verifying within that article. There are situations where an original SPS is the most reliable source we could cite. Blueboar (talk) 22:01, 8 July 2020 (UTC)

Chiming in with agreement. Guidelines have lots of should's, where appropriate. Always and never should (ahem!) be limited to Wikipedia's licensing requirements, and legal obligations; and rarely if ever appear in policy or guidelines. Mathglot (talk) 22:09, 8 July 2020 (UTC)

I agree. I think that the context has to be kept central: a given source might or might not be acceptable (or preferable); that judgment always occurs in relation to the claim it's a reference for. -- FactOrOpinion (talk) 00:02, 9 July 2020 (UTC)

I was the one who added the wording "These are always preferable", thinking that I was correctly clarifying the intent of something I read elsewhere in the article. I certainly don't have any objection to changing the wording to "These are usually preferable", and moving it from the Unacceptable heading, which I just did. -MichaelBluejay (talk) 15:03, 9 July 2020 (UTC)

Thought experiment: delete this policy supplement page

First: I am not proposing we delete this page. But I am seeing a lot of things on this page, and other pages regarding SPS, that makes me think that it might be just a proxy for an aspect of reliable sourcing. One way to see if this is so, is to ask ourselves, "What would be lacking if we eliminated this page?" Or, conversely, "Is this page saying anything specific about one class of sources, i.e., self-published sources, that does not apply to all sources?"

Putting it another way, if we have a consensus (via RSN, or long practice, say) that so-and-so is reliable, ("Joy of Cooking", or whatever) does it matter that it's self-published? Conversely, if the Daily Mail is regarded as unreliable, does it matter that it is not self-published?

All sorts of things on this page seem to be address SPS, but really are about RS. In going through the section, #The problem with self-published sources, point by point, the assertions seem to be something that could apply equally to reliable sources. Examples:

• The UCB quotation, in explaining why SPSes are a problem, says that they may have "motives to get you to buy something or believe a point of view." – also applies to most books and newspapers. Perhaps not to scholarly articles; but even some of those.
• "While many self-published sources happen to be unreliable, the mere fact that it is self-published does not prove this." – Seems to undermine the whole point of having this page. If "self-published" is sometimes good, just like "editorially published" is, let's say, "often good", then what are we saying here?
• "Self-published sources can be reliable..." And books can be unreliable...
• "Sometimes, a self-published source is even the best possible source..." And sometimes a book is the worst.

The whole second part of this section (starting, "According to our content guideline...") is just summarizing the RS guideline; nothing new here.

Isn't this really all about proper use of reliable sources, irrespective of publishing status? Is "SPS" a kind of taint, that tends to lead us to prejudging, the way many editors view comments made by an anonymous editor with more suspicion because they are part of a class? Does the existence of this page foster a kind of publisherism prejudice about individual sources based solely on their publishing status class membership irrespective of their innate qualities, or is it a legitimate descriptor on its own for an individual source, that helps us decide something about reliability? If so, what? If this SPS page didn't exist, what instructions on the evaluation of reliable sources would be missing, that don't more properly belong in the RS guideline in the first place?

This Gedankenexperiment was spurred by the discussion of the Lead definition elsewhere on the page. I'm not trying to muddy the waters here. On the contrary: by asking an existential question about this page, I'm trying to pierce to the heart of the matter. Thanks, Mathglot (talk) 23:12, 8 July 2020 (UTC)

I wrote this page to address a persistent belief that a corporate website could be considered non-self-published so long as the corporation employed enough lawyers. IMO the core of the page is the "identifying" part, not the "using" part – the part in which someone reads it and thinks, "Oh, that's why they think this BLP isn't notable. It's because everything ever published about him is on Twitter, YouTube, and his employer's website. Silly me for thinking that Twitter, YouTube and the employer were 'publishing' all that stuff." WhatamIdoing (talk) 23:27, 8 July 2020 (UTC)

I think this supplement is useful, as it provides a definition (even though we might change it), and provides examples of acceptable and unacceptable use, since all this is missing from WP:SPS. I also like the bit in the supplement that says that SPS should not be a bit of jargon thrown around to automatically dismiss a source, since I see that a lot. If we delete this supplement, then WP:SPS needs those three things: a definition of SPS, examples of acceptable/unacceptable use, and a statement that SPS doesn't mean a source is automatically invalid. -MichaelBluejay (talk) 15:08, 9 July 2020 (UTC)

A reason to pay special attention to self-published sources, and perhaps keep this page, is that frequently the editor who adds information has access to the source, and all the other editors who follow the page do not. WP:Assume good faith possibly calls on us to suppose that the editor read the work and made a reasonable judgement that it's a reliable source. But if it's a self-published source, there is a much stronger chance that's it's a poor work, compared to a source from a traditional publisher. It's often easy to tell a work is self-published just by reading the citation, while obtaining access to the work is a lot harder.

So the special attention to SPSs lets us push the burden of showing it's reliable back on the editor who want's to use it.

A reason to delete this page is that it's hard enough to get people to read guidelines; getting enough people to read this page for the page to earn its keep is unlikely. Jc3s5h (talk) 15:22, 9 July 2020 (UTC)

User:Jc3s5h, said:

But if it's a self-published source, there is a much stronger chance that's it's a poor work, compared to a source from a traditional publisher.

This is exactly the issue I'm trying to point to, and asking 1) if it's fair, and 2) if it's useful. Saying that something has "a much stronger chance" of being <insert negative judgment here>, just because they belong to a class of items, where others in that class have evaluated poorly earlier, is a prejudgment. Are we besmirching a source for being SPS, because other SPSes are "bad" before we've examined whether this source is reliable? Does this help us reach a proper evaluation for this source? Mathglot (talk) 11:06, 11 July 2020 (UTC)

Besmirching SPS is neither fair nor useful. It leads to things like editors reverting changes, saying "SPS not allowed", which I've seen a lot. A source should be rejected only if it's a poor source, not because it's SPS. We don't have to think twice about quoting a science journal, but we *should* look twice at an SPS before using it as a source. And for that purpose, guidance on differentiating good from bad SPS is exactly what editors need. So again, keep this page, unless we're gonna add that differentiating explanation to WP:V and any other related policy. -MichaelBluejay (talk) 13:31, 11 July 2020 (UTC)

WP:BLP prohibits the use of SPS for claims about third parties, so in that case, under current policies, we do have to reject them for being SPS. WhatamIdoing (talk) 17:20, 11 July 2020 (UTC)

In that case, we don't reject the sources for being SPS, we reject them for for being SPS about third parties. That's a crucial distinction. The fact that we don't use SPS in certain circumstances means that we wind up painting SPS with a broad negativity brush. It's better to list when SPS acceptable and when they're not acceptable. -MichaelBluejay (talk) 17:06, 12 July 2020 (UTC)

Notability

User:Rjensen has changed the notability section from "never" to "seldom". WP:SPIP says Even non-promotional self-published sources, like technical manuals that accompany a product, are still not evidence of notability as they are not a measure of the attention a subject has received, which suggests that "never" is the correct standard. However, I wonder whether WP:N might be 'wrong' there – that the real problem with the non-promotional self-published technical manual is its lack of independence from the subject.

What do you all think? If, say, I set up my own website, where I post information about former businesses (or buildings or schools or whatever) in my small town, is that evidence of notability for them? What if it's not me doing this, but instead the website is written and published by the town's history club, or the city government itself, or a local history museum? (All of these are still self-published websites.) WhatamIdoing (talk) 18:19, 9 July 2020 (UTC)

Two reasons for my edit: (1) Wiki guidelines should not use "never" -- that term is restricted to policies set by Wikipedia Foundation. Here we have not even a guideline ("This page is not one of Wikipedia's policies or guidelines") (2) In this case for example a user manual can provide evidence of widespread usage by many different users that passes the notability criterion. Rjensen (talk) 18:25, 9 July 2020 (UTC)

I disagree with your (1) in principle, and I offer Wikipedia:External links as an example of a mere guideline that has used the word never for almost as long as it has existed.

I disagree with your analysis in (2), because the manufacturer's own publication never shows evidence of attention from the world at large, which is the point of notability. Also, what exactly would you find in a manual that is evidence of widespread usage, but wasn't self-promotional? "More than a billion sold" is self-promotional, and the typical non-self-promotional parts would provide no information about widespread usage. WhatamIdoing (talk) 19:13, 9 July 2020 (UTC)

Wikipedia:External links says never only for policy issues (copyright) set out by Wiki Foundation. Its general advice is the same as mine: it clearly states normally to be avoided Rjensen (talk) 22:42, 9 July 2020 (UTC)

I agree with changing "never" to "seldom." I also think it could be useful for us to try to generate a list of different categories of self-published works, so that we can explore whether a given claim is valid across the different categories. For example, now that I'm puzzling over what constitutes self-published work, my sense is that government publications are self-published works (government employees write them, they're often edited by other government employees, and the government generally publishes them without external review). But many government publications are significant, and sometimes they have considerable expert input, which can support a claim of notability. Or, consider a petition signed by a few million people; that kind of accumulation of individual support (where I think of each name/signature on the petition as being a self-published contribution to the petition, though I'm not sure that everyone would agree with that) could be one piece of evidence used in deciding that some topic is notable. -- FactOrOpinion (talk) 19:55, 9 July 2020 (UTC)

I think there are instances where what is technically a self-published source would be reliable enough to establish something is notable. For example if a 9-0 decision of the US Supreme Court stated X was an event that changed the world, that would establish X as notable. But undoubtedly X would have been described by numerous works that were not self-published, so there would be no need to depend on the Supreme Court decision for notability purposes. Jc3s5h (talk) 19:24, 9 July 2020 (UTC)

Jc3s5h, what do you think of Schneier on Security (a blog) as an example of a more obviously self-published source? If a security problem gets described there, then the blog post is probably evidence of notability. I'd probably count it as contributing towards notability. WhatamIdoing (talk) 17:13, 10 July 2020 (UTC)

I read that blog from time to time. I consider it a reliable self-published source, and for me, it would count toward notability. But my general sense is that the things Schneier writes about have substantial coverage in other sources too, so we probably wouldn't need the blog to make the case that something is notable. Jc3s5h (talk) 17:18, 10 July 2020 (UTC)

I looked up the origin of that SPS-related sentence. It started off as part of a footnote and used to include a phrase about attention from the world at large. I think it was really meant to be about Wikipedia:Independent sources instead. We should probably move this discussion to Wikipedia talk:Notability. WhatamIdoing (talk) 17:23, 11 July 2020 (UTC)

Notes and questions re: several edits I introduced

I've introduced several edits to the project page, and I did my best to explain what I did in the edit summaries. Some of my edits should be uncontroversial, such as updating the links for a couple of the quotes (and I hope that I was correct in changing the access date to today, since I had to add new URLs), whereas people may disagree with other edits, though I tried not to do anything drastic. Apologies in advance if I changed anything that should have been discussed first; I'm still in the "learning" stage as an editor, and although I try to be careful, I sometimes screw up. If I've done that, please let me know.

Re: the quote citations, the link for quote #1 still works, but there was no archived copy. I tried doing a Wayback save for that webpage, but the Wayback copy didn’t capture the podcast audio, which is where the quote comes from (i.e., listening to the podcast is the only way to confirm that the quote is accurate). If someone else knows more than me about archiving, it seems wise to archive the audio. I updated the links for quote #2 and #3, but haven’t checked yet whether those quotes are cited elsewhere on WP (e.g., on WP:V); if they are, the links should be updated on those other pages. Note that the Princeton citation is copyrighted in several different years (presumably as the work was updated), and I used the most recent copyright year since the quote still appears in the most recent version and that's all I could find an archived copy of.

I think it would be helpful to add something re: distinguishing between “third-party source” (which WP mostly treats as an “independent source,” though sometimes makes distinctions between them) and “tertiary source,” linking to additional info. Wikipedia:Party and person is the most relevant page, and it could use some work; other relevant pages include Wikipedia:Independent sources and Tertiary source.

Re: “Examples of self-published sources: Almost all websites except for those published by traditional publishers (such as news media organizations), ...” if there are examples of websites that aren't SPS and also aren't published by traditional publishers, I think it would be good to note that this is possible and add it to "Examples of non-self-published sources." I'd raised a question on the RS/N re: whether 3 specific online legal forums are or aren't SPS, as in my view, each has a degree of editorial oversight that isn't typical for online forums/blogs, so for at least some of their articles, it seems that author =/= publisher. That discussion hasn't concluded (and ultimately has to be guided by the decisions here and on related policy pages about whether independent editorial oversight is sufficient to make author different from publisher, as well as debate there about whether the editorial oversight of those forums is really independent), but in case it's helpful to have possible examples to push discussion here, here's a link: [6] -- FactOrOpinion (talk) 01:43, 12 July 2020 (UTC)

P.S. re: the claim "Sometimes, a self-published source is even the best possible source...":

I added two more examples for that claim, one of which was "A self-published source by an expert may include a significant opinion that hasn’t yet appeared in a non-self-published source." That addition was prompted by the article "On Wikipedia, Lawfare, Blogs, and Sources" ([7]), which notes in part that a blog by experts may "contain[] a large amount of commentary and thought from its major contributors that has never been published elsewhere," and is also mentioned on the WP controversies page ([8]). -- FactOrOpinion (talk) 14:30, 12 July 2020 (UTC)

Are government publications often SPSs and should this be addressed on the main page?

I think that government materials are often self-published, analogous to the statement on the page that "If the author works for a company, and the publisher is the employer, and the author's job is to produce the work (e.g., sales materials or a corporate website), then the author and publisher are the same." Given how much the government publishes and how often government texts are treated as authoritative, I think it would be useful to add some explicit text about this.

Some things that might be discussed: The government produces a diverse array of publications (primary - like court docs and congressional hearing transcripts, secondary - such as analyses of census data, and tertiary - such as the Congressional Research Service overviews). WP:BLPSPS requires users to determine whether a publication is first person (such as a court transcript) vs. third person (most government publications) and to refrain from using SPS for third-party BLP claims. Some government publications aren’t self-published (e.g., proceedings of government-sponsored conferences, where the authors/presenters don’t work for the government; PubMed info with the abstracts from articles published by peer-reviewed journals - [9]), and it may be that for some government publications, there’s an independent review process in the middle (e.g., reviewed and/or edited by other government employees, but where the government isn’t telling those editors whether to approve the publication or not, and where they may be acting more like newspaper editors). We could reinforce in this text that WP editors shouldn’t treat a given publisher as a uniform entity, but judge in relation to specific publications and claims. @Mathglot: highlighted this earlier ([10]): "In looking at pages about SPS, or discussions at RSN or elsewhere, I mostly see allusions to the work, so that SPS is discussed as an attribute of where it appears, not who wrote it. Questions such as, 'Is Just Security a SPS', or, 'Is Infowars a SPS'. Or even theoretically, 'Is The New York Times a self-published source?'" -- FactOrOpinion (talk) 18:11, 12 July 2020 (UTC)

Another consideration is that works created by US government employees in the course of their duties are not subject to copyright. This applies even if the work is published in a journal that ordinarily claims copyright on the articles it publishes. In this situation, the article may be published in a journal behind a paywall, and a free copy is hosted at the government agency where the author works/worked. In this case, the government agency is more like a printer than a publisher. Jc3s5h (talk) 19:07, 12 July 2020 (UTC)

Add info re: where to get help assessing whether a source is/isn't SPS?

Editors may sometimes be uncertain about whether a given source is/isn't SPS. I think it would be helpful to add a line somewhere on this page making explicit that if an editor needs help assessing whether a given source is a SPS, then the editor can seek help at the BLP/N if the material is about a living person and the question is whether the BLPSPS rule applies (or maybe I'm wrong about directing them to the BLP/N, and we should instead say to add a question to the relevant Talk page and then post a note to RfC/B for additional comments). Similarly, should we say that an editor can seek help at the RS/N if the question is a more general one, including questions about whether the creator of some material is expert in relation to a claim? As a newish editor, it wasn't clear to me where to ask for help determining whether a source is/isn't SPS. The descriptions on those pages -- BLP/N, RfC/B, RS/N -- don't mention anything about SPS as one of the things to ask for help with on that page. -- FactOrOpinion (talk) 16:20, 23 July 2020 (UTC)

Business, charitable, and personal websites

This line is problematic. The Washington Post is a business, and "charitable" could use elaboration lest it be interpreted to include all non-profits. — Rhododendrites ^talk \\ 16:52, 27 July 2020 (UTC)

Typo fix

I removed one "non".

• Previous: A non-self-published source that verifies the same information is usually preferred to a non-self-published one. If it is not clear which source is better, they can both be cited.
• New: A non-self-published source that verifies the same information is usually preferred to a self-published one. If it is not clear which source is better, they can both be cited.

I hope I got that right, because it sure didn't make any sense the way it was written. That language was introduced with this edit. Normal Op (talk) 08:31, 16 August 2020 (UTC)

Acceptable use of self-published works

I've been in an editing war with User:Michaelbluejay for 11+ years and strongly suspect he is trying to change the rules to allow him to use his own SPS. Lore E. Mariano (talk) 03:27, 17 August 2020 (UTC)

There's been a protracted discussion on the WP:V talk page, and I don't see what you suggest. -- FactOrOpinion (talk) 13:01, 17 August 2020 (UTC)

I agree with Alanscottwalker who comments: "...[N]o such change should be made to this central policy, without a formal 30 day widely advertised WP:RfC, per WP:PROPOSAL. Alanscottwalker (talk) 15:14, 17 August 2020 (UTC)" and "...[T]he approach to read and write SPS out of context is objectionable and bad policy writing. And per WP:PROPOSAL, the only sensible and correct sequence is to have the proposal written down to be commented upon at the beginning of the final RfC, not stop at just a sprawling, lengthy and largely impenetrable discussion among a few commentators. Alanscottwalker (talk) 16:42, 17 August 2020 (UTC)" Lore E. Mariano (talk) 21:51, 17 August 2020 (UTC)

To editor LoreMariano: You should post your agreement there, not here. It's also irrelevant to your initial comment and your suspicion not being born out in the discussion on that page. -- FactOrOpinion (talk) 22:05, 17 August 2020 (UTC)

Using self-published sources | For claims by self-published authors about themselves

Forgive me for making the change without bringing it up for discussion first. Changed: "the material is not unduly self-serving and ..." to: "the material is neither unduly self-serving nor ..." -- Johnnie Bob (talk) 23:24, 25 August 2020 (UTC)

Thank you, I like it. -MichaelBluejay (talk) 05:41, 27 August 2020 (UTC)

Suggested example of acceptability

There are many cases in which SPS (self-published sources) are, or should be, acceptable. I suggest adding this example to the list of acceptable uses of SPS:

• The very existence of the source supports the statement. For example, for the statement, "Members of his own party criticized his actions," self-published blogs by party members which contained such critical posts would be acceptable as a source. Similarly, for the statement "The organization purchased full-page advertisements in major newspapers advocating gun control," the advertisement(s) in question could be cited as sources, even though advertisements are self-published. (Note, this acceptability does not extend to supporting claims made in the advertisement, only the existence of the claims, though the claims might still be acceptable based on other items in this Acceptable Use list.)

Thoughts? -MichaelBluejay (talk) 16:55, 26 August 2020 (UTC)

What does this add, that WP:ABOUTSELF leaves out? Mathglot (talk) 07:27, 27 August 2020 (UTC)

Good question. My answer is that I don't see WP:ABOUTSELF as specifically covering the case I described. First, the "self" in ABOUTSELF seems to refer to the author, not to the work. Second, WP:ABOUTSELF seems to cover claims by the author about him/herself, which is quite different from using the existence of a work to support a statement. As for why this is necessary, I've had editors revert my edits (citing SPS), where the SPS should have been acceptable on existence grounds. So, existence should be an acceptable use of SPS. -MichaelBluejay (talk) 19:23, 27 August 2020 (UTC)

I'm not sure what you're talking about, but agree that WP:ABOUTSELF seems clear enough. Books/works don't "say" anything; an author says something. So citing an author's book is the same as citing the author. (But even then, I'm not sure this is relevant because your hypothetical situation is not clear in the first place.) Normal Op (talk) 19:54, 27 August 2020 (UTC)

Okay, how about this:

• CLAIM IN THE WP ARTICLE: "He attended Princeton from 1985-1989."
• SOURCE: The subject's personal blog: "I attended Princeton from 1985-1989."
• TYPE: The claim is supported by a statement in the source.
• POLICY: Covered by WP:ABOUTSELF

Contrast that with:

• CLAIM IN THE WP ARTICLE: "Members of his own party criticized his actions."
• SOURCES: Personal blogs of members of the subject's party, e.g. "Sen. Pinky's actions are bad."
• TYPE: The claim is supported by THE VERY EXISTENCE of the source. The source doesn't say, "I and other members of the party criticize Sen. Pinky's actions."
• POLICY: Not covered by WP:ABOUTSELF. The subject is not the author of the sources.

and

• CLAIM IN THE WP ARTICLE: "The organization purchased full-page advertisements in major newspapers advocating gun control."
• SOURCE: Self-published site which reproduced all the various advertisements. (Note, the advertisements themselves are self-published, and the website where they were republished is also self-published.)
• TYPE: The claim is supported by THE VERY EXISTENCE of the source. The ads don't say, "Such and such organization ran this ad in major newspapers."
• POLICY: Not covered by WP:ABOUTSELF.

-MichaelBluejay (talk) 12:15, 28 August 2020 (UTC)

Number 1 is probably uncontroversial; if contested then make sure there's a secondary source. Number 2 is WP:SYNTH, and since it's based off a self-published source, I wouldn't use it. Number 3: I wouldn't use either; there's no proof the ads were run anywhere even if a graphic artist created them (you can see them) and some non-RS stated they ran them. Normal Op (talk) 17:03, 28 August 2020 (UTC)

If by "Number 1" you mean the first existence-based example ("Members of his own party..."), then thank you, and this shows why WP:USINGSPS should explain that such sources are acceptable. Also, here's another example: The claim in an article is, "In his later years, he published a volume of poems." As it so happens, we can't find a source that specifically states "He published a volume of poems." However, the volume of poems exists. The volume of poems should be used as a source, since its very existence supports the statement. That's what I'm getting at. -MichaelBluejay (talk) 23:40, 28 August 2020 (UTC)

Don't read words I didn't write. WP:ABOUTSELF is the ruling policy, and if someone contests the information then you must provide a secondary source. Perhaps the part in between my words, the part left unsaid, is that if the information isn't controversial then no one will care and the edit might remain. But the moment some editor cares, and that edit gets removed... then that's your cue/clue that the edit was controversial. But under no circumstances should a non-reliable or questionable source be used as information about a second party. Normal Op (talk) 06:46, 29 August 2020 (UTC)

Michaelbluejay, my understanding is that the second of the three examples above would be excluded as WP:BLPSPS. I agree that the first of the three examples is fine under WP:ABOUTSELF and also agree with Normal Op that for the third example, an image solely of the ad doesn't guarantee that the ad was actually published. On the other hand, if the SPS showed an image of an ad with a newspaper's header / date / page number (as I think would normally appear with a full-page ad), then that would be acceptable (and likely provides enough info that once could go and find the ad in the newspaper's own microfilm archives, if one had access to those). Re: your newest example, I'm inclined to think that it's covered by WP:ABOUTSELF (e.g., the volume is dated and appears with the person's name as author, even if the poems themselves aren't about the person). -- FactOrOpinion (talk) 14:39, 29 August 2020 (UTC)

Publisher as employer revisited

I see this was brought up some years ago, bringing it up again. This "If the author works for a company, and the publisher is the employer, and the author's job is to produce the work (e.g., sales materials or a corporate website), then the author and publisher are the same." is confusing, because it can be used to apply to any content-producing agency. Could the original intent have been "If the author works for a company, and the publisher is the employer, and the author's job is to produce the work about that company (e.g., sales materials or a corporate website), then the author and publisher are the same."? Schazjmd (talk) 00:11, 14 October 2020 (UTC)

@Schazjmd, if we write that, it will be interpreted in a more limited fashion than it should be. Consider "But the work isn't about 'that company'; it's about the product/CEO/competitor".

Also, think about how you would write this when the employer is a political party or a charity whose primary purpose is to raise awareness about a social problem. Everything you write and post on "your" (including your employer's) website is self-published. There is no separate entity involved in the publication.

By contrast, if a wire service writes an article, and a newspaper publishes it, then we definitely have multiple separate entities involved (even if self-publishing didn't usually exempt "traditional publishers" such as news media anyway). WhatamIdoing (talk) 20:37, 2 November 2020 (UTC)

Is my book a reliable source?

Hi there. Sorry if this is not the right place to post this, but I was wanting to sort something out. I wrote a book about the manga artists Clamp, which I crowdfunded via Kickstarter and had published by a company local to me (Also listed on Amazon). I don't know if my book counts as a reliable source as it might fall under the banner of self-published works, and even then I don't know if I myself am allowed to use my own book as a source, rather than somneone else referencing my book. What exactly is the situation? Can I or someone else use my book as a source on Wikipedia? Thank you. ISD (talk) 15:13, 10 November 2022 (UTC)

@ISD, your book is self-published. Self-published works can be reliable. It's even technically "legal" for an author to WP:SELFCITE what they wrote (a little bit), although it's much much much better if you suggest it on a talk page and let other editors decide whether they think it's good for a particular point. (I assume that the goal would be to cite, e.g., something about a character described in your book. Self-published sources shouldn't be cited for information about any living person [except the author].) WhatamIdoing (talk) 21:43, 6 March 2023 (UTC)

@WhatamIdoing Thanks for the information. ISD (talk) 06:26, 7 March 2023 (UTC)