Wikipedia talk:Handling primary, secondary and tertiary sources (proposed guideline)

• User talk:Yaris678/PSTS/Archive 1
• User talk:Yaris678/PSTS/Archive 2

Yaris's summary of the arguments so far

N.B. This is a summary, by one of the parties, of discussions which can be found at User talk:Yaris678/PSTS/Archive 2.

Yaris678 (talk) 23:57, 27 January 2010 (UTC)

OK. I think I have sorted the issue with the tag and I am prepared to accept SlimVirgin’s edits on scientific papers. Therefore, I think the main issue is whether or not we decide to have a separate guideline on primary, secondary and tertiary sources. I will try to summarise the arguments put so far. I will also add any obvious counter-arguments that haven’t been added yet. Obviously, I could be biased in this. I will try to be as fair as possible but all comments and suggestions welcome.

I will give each argument a letter and a number. The letter is either p for pro separate guideline or a for anti separate guideline. Counter arguments have their number appended to the argument they are countering. I hope it is obvious when you see it.

p1 - It allows the PSTS section of WP:NOR to concentrate on the original research aspect of PSTS. This fits with WP:Policies and Guidelines#Content, which says “Content should be within the scope of its policy”. The current section also gives definitions and talks about WP:CIRCULAR, which dilutes the message.

p2 - It allows us to write slightly more in the guideline and clarify a few things.

p3 - The current situation has lead to many (inexperienced) editors miss-reading the policy to mean, for example "Primary sources are not allowed." This problem could be reduced by having a separate guideline for the reasons given in p1 and p2.

p4 - Part of the reason why PSTS is important is that a primary source can not establish a topic's notability - but WP:NOTE is a guideline and not policy. Referring to a guideline from a policy could give the impression that the guideline has been upgraded to a policy.

p5 - WP:Reliable sources is a guideline that deals with a different aspect of source categorisation. It is arguably more important than WP:PSTS, so why is WP:PSTS at the policy level?

p5.a1 - WP:RS should never have been made into a guideline anyway. See a4.

a1 - This suggestion is the result of inexperienced editors thinking the current system doesn’t work. In reality it does.

a1.p1 - The current situation probably does work well for experienced editors. It has been observed to not work well for some inexperienced editors. If we clarify things it will make things easier for the inexperienced without making them any harder for the experienced.

a1.p2 - The implication that none of the people putting forward the suggestion has much experience is not true. Info from toolserver.org:

Username: Blueboar

User groups: rollbacker

First edit: Dec 07, 2005 18:41:06

Unique articles edited: 1,664

Average edits per page: 11.94

Total edits (including deleted): 19,875

Deleted edits: 586

Live edits: 19,289

a2 - The main reason for even talking about primary, secondary and tertiary sources is to do with original research. Even WP:NOTE is an application of WP:NOR - i.e. using your own judgement on the notability of a topic, without it being mentioned in secondary sources, is OR.

a2.p1 - "No original research based on primary sources" is an application of "No original research". However, that does not dictate that WP:PSTS should be part of the WP:NOR page. For example, the guideline WP:Do not include the full text of lengthy primary sources is an application of the policy WP:What Wikipedia is not; the guideline WP:Conflict of interest is an application of the policy WP:Neutral point of view.

a3 - This is a proposal to remove a key part of WP:NOR and reduce it to guideline status, which means people can happily ignore it.

a3.p1 - People do not ignore WP:RS.

a3.p2 - Under the proposal, WP:NOR would still talk about the original research aspects of primary, secondary and tertiary sources. It just wouldn’t have its message diluted, as per p1.

a4 - This would be a policy fork that would just confuse the situation, in the same way as WP:RS, a policy fork from WP:V.

a4.p1 - People seem to be able to use WP:RS and WP:V without any problems.

a4.p2 - While some policy forks (as you call them) should be avoided, there are times when it is a good thing. Wikipedia:Policies and guidelines says "Policies describe standards that all users should normally adhere to, and guidelines are meant to contain best practices for doing so." This would be impossible without policy forks.

Yaris678 (talk) 10:10, 15 January 2010 (UTC)

Oppose policy fork

I strongly oppose this for two reasons. First and most importantly, the issue of primary sources is a key part of the NOR policy, and has been in it for years. Secondly, I've been concerned to see in Yaris's editing the promotion of exactly the kind of non-notable primary source material that the NOR policy seeks to avoid. SlimVirgin ^{TALK contribs} 22:59, 27 January 2010 (UTC)

Also, I want to make clear that we would need a wiki-wide consensus to remove, downgrade, or fork such a key part of NOR, given the issues that arose over simply trying to summarize it in ATT. SlimVirgin ^{TALK contribs} 23:16, 27 January 2010 (UTC)

Hi SlimVirgin,

Nice to see that my upgrading the page to a proposed guideline has encouraged you to contribute again. However, it would be nice if you could engage with the points I made above rather than making this about me. There is no point in an argumentum ad hominem, given that I am not the only one putting forward this suggestion.

I agree that wide consensus is required.

Yaris678 (talk) 23:26, 27 January 2010 (UTC)

On the specific issue of a policy fork, see a4, above. Yaris678 (talk) 23:34, 27 January 2010 (UTC)

On the specific issue of downgrading a part of WP:NOR, see a3, above. Yaris678 (talk) 23:45, 27 January 2010 (UTC)

I support this proposal because it discusses more than just the "PSTS and OR" issue. There are PSTS issues that have nothing to do with OR (just as there are OR issues than have nothing to do with PSTS). These do need to be addressed, but they are out of place when adressed in the context of the NOR policy.

I completely agree that we need a wiki-wide discussion and consensus on this. Blueboar (talk) 23:38, 27 January 2010 (UTC)

See the confusion the existence of RS causes e.g. the recent list of suicides discussion. No more policy/guideline forks on issues as important as sourcing! SlimVirgin ^{TALK contribs} 00:22, 28 January 2010 (UTC)

I assume you are referring to this comment. The correct thing to do in that situation is refer the confused person to WP:V. As you said yourself here, that problem could be avoided by giving WP:RS a different name. My own suggestion would be to rephrase the hat note (although it does already mention WP:V).

I don’t think that is going to apply in this case anyway. Can you imagine the argument?

Editor 1: Editor 2, I removed your statement because it is based on an original analysis of primary sources. That comes under WP:NOR.

Editor 2: But WP:Primary, secondary and tertiary sources is just a guideline. I think we should ignore it in this case.

Editor 1: That’s irrelevant. Please read WP:NOR.

Yaris678 (talk) 00:57, 28 January 2010 (UTC)

I have just noticed that about 1 hour before I posted the above comment you had rephrased the nutshell and lead for WP:RS, so my suggestion to rephrase was a bit redundant. My apologies. Yaris678 (talk) 10:19, 28 January 2010 (UTC)

Strong redefinition of WP policies and guidelines

Much of this is strong redefinition of WP policies and guidelines, guaranteed to introduce Wiki-Drama no end. Promotion of "primary" sources is unwise, IMO, and the redefinition and placing of general encyclopedias with academic textbooks is also unwise. "If I were King of the Forest" ... I would also then separate secondary sources as to sources of fact and sources of opinion, and try to downplay the current excessive reliance of "editorial opinions" found in almost every controversial topic. WP has many articles now fully bloated with them. And, of course, I would stipulate than any section with a "grade index" over 12 should be ruthlessly edited to make it readable. Did you know some pages have indexes over 15? Sorry -- but the problem does not really lie with the current system - the problem lies in the abuse of the system which this does not prevent. Collect (talk) 11:32, 29 January 2010 (UTC)

The stuff about readability is interesting, but I don't see what it has to to with this page. How about WP:MOS?

Opinion pieces should mostly be treated as primary sources for an opinion, rather than secondary sources for the facts discussed. However, I can see there is a grey area where what seems like "analysis of the facts" to one person is "pure opinion" to the next person. Does anyone have a suggestion of how to deal with that issue? My first thought is that if it is an opinion, it is more likely to be contradicted by someone else's opnion, even if the two are based on the same facts. This is discussed in the lead of the article on opinion.

On your main point, can you clarify what you mean? Which bits do you think promote primary sources? The bit on quoting them? The statement that both encyclopedias and text books are tertiary comes directly from the current WP:PSTS so there is no redefinition going on there.

Yaris678 (talk) 13:10, 29 January 2010 (UTC)

Opinions are actually fairly simple to deal with... by clearly attributing the opinion to the opinion holder. A lot of POV debates can be settled by the addition of a simple: "According to..."

The hard part is determining whether mentioning the opinion is proper in the first place... we have to ask whether mentioning it gives undue weight to the opinion or its holder (per WP:UNDUE). Blueboar (talk) 14:48, 29 January 2010 (UTC)

You are right of course Blueboar. WP:NPOV is the relevent policy on opinions. I was trying to think of it in terms of PSTS, hence the "Opinion pieces should mostly be treated as primary sources for an opinion, rather than secondary sources for the facts discussed." However, it's probably best not to deal with it here and leave it to WP:NPOV. Good thinking. Yaris678 (talk) 15:16, 29 January 2010 (UTC)

Archiving of old talk

This is all getting a bit meta now, but I see no harm in archiving old talk. What is your problem SlimVirgin? The archives are accessible to everyone, as is the page history. Yaris678 (talk) 00:36, 28 January 2010 (UTC)

I've re-archived since I have received no reply. Yaris678 (talk) 01:31, 28 January 2010 (UTC)

Quoting primary sources

The guideline proposed the following:

For a direct quotation of a primary source, it is preferable to cite the primary source itself, rather than a secondary or tertiary source which also makes the quotation. However, if you can not locate the primary source say where you found the material. For large sources, page numbers or similar information should be given to allow the quoted section to be found. Do not include the full text of lengthy primary sources in Wikipedia; but if such a source is out of copyright, it may be appropriate to include it in Wikisource.

This seems to be fairly bad advice. Primary sources may have many editions, and many variations. Secondary sources may quote different versions of a primary source than an editor happens to own. This is an open invitation for WP:NOR. Jayjg ^(talk) 02:16, 29 January 2010 (UTC)

Which bit is bad advice? Are you referring to the first sentence? Are you suggesting that we should generally limit ourselves to passages which are quoted elsewhere? Yaris678 (talk) 08:27, 29 January 2010 (UTC)

Please note that WP:NOR explicitly says that we may cite primary sources. Quoting a source (of any type) is not OR. Misusing a source to form a synthesis, or going beyond what the source says and adding your own analysis, interpretation or conclusions is OR. The point behind this statement is that, in many cases, secondary sources may mis-quote the original (either intentionally or unintentionally), or may quote the original out of context (I have to deal with this all the time in editing articles on Freemasonry). We can not always rely on a secondary source to accurately quote the original... so when we wish to quote the original, it is best to locate and quote directly from the original itself.

Yes, there may be different editions or versions of a source (this is true for secondary sources as well)... which is one reason why we include the full publication data (including which edition we looked at) when we cite our sources. Blueboar (talk) 14:36, 29 January 2010 (UTC)

Yes, Blueboar, primary sources may be used, under certain circumstances, but they should be avoided except for the most basic, non-contentious facts. To begin with, absent a secondary source, primary sources should not be used at all for quotes. Secondary sources are what establish notability, context, meaning, etc. for primary sources, particularly things like quotes. Next, if a secondary source does bring a quote, and it disagrees with the primary source, we have no idea why that is the case. Perhaps the secondary source misquoted the primary; perhaps it used a different edition or translation; perhaps the primary source has been modified (in the case of an on-line source). It's not up to the Wikipedia editor to decide which source is better, and particularly not up to Wikipedia to decide primary sources are always more accurate. Rather, the quote from a secondary source should be used, and, if necessary, a primary source can be brought in a footnote for comparison. Jayjg ^(talk) 18:56, 31 January 2010 (UTC)

Yaris678, it's bad advice to tell people they should prefer primary sources for quotations over secondary. As I've said above, the secondary source may be using a different edition, version, translation, etc. than the primary. In addition, secondary sources establish notability and context for quotes. Jayjg ^(talk) 18:56, 31 January 2010 (UTC)

It says "For a direct quotation of a primary source..." It doesn't say "Please quote primary sources as much as possible", which seems to be how you are reading it. Yaris678 (talk) 19:26, 31 January 2010 (UTC)

It says primary sources are preferred to secondary for quotes. That's what it says and means. And it's both bad advice, and contrary to policy. Jayjg ^(talk) 19:49, 31 January 2010 (UTC)

I can not disagree more... when qouting it is always best to do so from the original as opposed to a pass through (which might contain an error). This is good advice and good editing, and does not contradict any policy (in fact, the language was directly taken from policy. See WP:NOR). Blueboar (talk) 20:02, 31 January 2010 (UTC)

Feel free to disagree, but since you've failed to respond to any of the reasons listed above, my point stands uncontradicted. Preferring primary sources for quotes is bad advice and bad editing, and in fact violates WP:NOR and WP:NPOV. Jayjg ^(talk) 20:11, 31 January 2010 (UTC)

Blueboar, I agree with what you are saying, but before you disagree with jayjg, it might be a good idea to confirm jayjg's understanding of what the proposed guideline says.

Jayjg, The proposed guideline says it is preferable to cite a primary source if you are quoting that primary source. It doesn't say it is preferable to quote primary sources.

Yaris678 (talk) 20:15, 31 January 2010 (UTC)

If that's what it said, then it would be saying nothing whatsoever. "If you quote a primary source, then quote the primary source." Yes, and if you quote a secondary source, you should quote the secondary source. And if you quote a tertiary source, you should quote the tertiary source. Jayjg ^(talk) 00:58, 2 February 2010 (UTC)

Neither NOR nor NPOV discuss this issue at all (and yes, I have checked). The closest we come to a policy statement on sources and quotes is from WP:RS which states:

• The accuracy of quoted material is paramount; the accuracy of quotations from living persons is especially sensitive. Quotations should be cited to the original source if possible; when secondary sources are used, those that cite the original source should be preferred over those that don't. Partisan secondary sources should be viewed with suspicion if they lack neutral corroboration. (italics mine for emphasis)

So when quoting from a primary source, we should indeed cite to that primary source. That said... I think what Jay may be concerned about is "cherry picking" quotes from primary sources to push a POV or OR conclusion... and I agree with him that doing this is improper. However, what makes it impropper is the act of cherry picking and formation (or even implication) of a POV or OR conclusion, not where you found the quotes you use to support the argument. To give an example... I might wish to quote the US Constitution in an article on "Gun laws in North America". If so, I should quote from the original and cite it rather than rely on a pass through. However, I might use that quote in the article in a way that pushes a POV or constitutes Original Research. That use would be improper... it would be a misuse of the quote (and of the source). Blueboar (talk) 04:15, 1 February 2010 (UTC)

Neither policy references this issue directly; policies rarely do reference every specific issue, they're quite brief. However, both forbid it, for the reasons outlined above. Jayjg ^(talk) 00:58, 2 February 2010 (UTC)

Why not add a new category above "References" called "Original documents" and allow listing of cites for such things as PDF documents which do not otherwise violate WP policy (i.e. arrest records with DOB, address, SSN, and other material not properly in an article)? The Gun control article thus could have the Bill of Rights appended, and usable without any possibility of it being used in the wrong context. Maybe? Collect (talk) 12:10, 1 February 2010 (UTC)

An option, certainly (for example, we can already include a link to documents on Wikisource)... but sometimes including a direct quote in the actual article text is simply good writing. We need to give editors flexibility, within certain limitations, to write their articles as they see fit. If they abuse that privilage we have mechanisms to deal with the abuse. Blueboar (talk) 15:57, 1 February 2010 (UTC)

Would it not then be harder for editors to cite out of context if the full context is a click away? Amazingly enough, it appears that some editors do pick and choose from documents ... Collect (talk) 18:57, 1 February 2010 (UTC)

Yes it would be harder... and in the case where the original is on-line we would include just such a link in the citation. Unfortunately, not everything is on line... and your idea of creating a PDF repository of texts to link to runs up against copyrite problems... to place quotes in context we often need to read larger chunks of the original text than the law allows us to copy. This is why we cite the quote to the original in the first place... so that if it isn't on line, someone can see where the quote came from, find the original, and check the quotation against it. Blueboar (talk) 23:00, 1 February 2010 (UTC)

Only if WP were the host for the PDFs -- I posit using links to where the sources are, not copying them to WP. The reader would have to use the link to where the actual source is. No copyright issues <g>. Collect (talk) 23:21, 1 February 2010 (UTC)

Well, your posit only works if the source is legally on line. We could not create a database to link to as that database would violate copyrite laws and it is still a copyrite violation for wikipedia to knowingly link to copywrited material on some other site.... and your idea would not work for dead tree sources as we would not have anything to link to. No, it is a good idea but not possible.

Besides... if you quote sources appropriately there is no need for all of this, as there is problem. To get us back to the point... the problem you seem to be conserned about is when editors cherry-pick sources, quote out of context, misquote, etc. That is a problem with the edit, not with the source itself. Blueboar (talk) 00:01, 2 February 2010 (UTC)

No, the problem is with a proposed guideline that tells editors to quote primary sources, when policy recommends using secondary sources, in order to avoid WP:NOR and WP:UNDUE. Jayjg ^(talk) 00:58, 2 February 2010 (UTC)

Jay... please point me to where any policy (but especially WP:NOR and WP:UNDUE) says not to quote primary sources. Blueboar (talk) 04:36, 2 February 2010 (UTC)

Not sure what you mean; your sentence doesn't match what I said. Jayjg ^(talk) 04:42, 2 February 2010 (UTC)

Collect,

Your idea is effectively what happens at the moment anyway. At least, if there is a version of the source available on line people often link to it in their citation. We don’t want to say that people can only use sources available on line but I suppose there is room to encourage people to do so.

Jayjg,

Part of the reason we are proposing this guideline is that people often misread the current policy as "primary sources are not allowed". This is certainly not what it says in any policy and the guideline is intended to clarify that issue. The guideline as we have written it certainly does not tell editors to quote primary sources, it just says that if they are going to quote them there is a preferred way of doing it. If you think this is not clear, perhaps you could suggest a better way to phrase it.

Yaris678 (talk) 09:28, 2 February 2010 (UTC)

But Yaris, that's quite obviously not what the proposed guideline says. Rather, it says that quotes from primary sources are preferred to quotes from secondary sources. Jayjg ^(talk) 02:38, 3 February 2010 (UTC)

It does not say quotes from primary sources are preferred to quotes from secondary sources. It's saying if you quote a primary source, take the quote directly from the primary source and cite the primary source. I'm assuming you know what the word cite means. Yaris678 (talk) 15:49, 3 February 2010 (UTC)

If that's what it's saying, then why isn't that the wording? I'll change it to that, and that should fix any confusion. Jayjg ^(talk) 02:16, 4 February 2010 (UTC)

That wording is fine with me. Yaris678 (talk) 08:52, 4 February 2010 (UTC)

Compromise?

I have boldly edited the passage to:

• Quoting from a primary source can be problematic. It is easy to quote primary sources inapproprately, in ways that constitute original research or give undue weight to a particular viewpoint. For this reason, great care should be taken when doing so. However, there are times when quoting from a primary source is appropriate and done appropriately. For a direct quotation of a primary source, it is preferable to cite the orignial primary source itself, rather than a secondary or tertiary source which quotes from the original (as the secondary or tertiary source may have misquoted the original or taken the quote out of context). However, if you can not locate the original, say where you found the material. For large sources, page numbers or similar information should be given to allow the quoted section to be found.

I hope this change will resolve Jay's problems with the section. My edit essentially repeats the warnings contained in WP:OR and WP:NPOV about how primary sources can be used inappropriately, but acknowledges that we are allowed to do so, if we do so approriately... and then points out that when you are quoting something (appropriately) it is preferred that you quote from the original. I think my edit correctly separates the issue of "use" (ie whether we can/should include a quote... which is what UNDUE and NOR are discussing) from the issue of "citation" (ie what we can/should cite if we quote... which falls under WP:V and WP:RS). This is fully in line with what is stated on other Policy pages. Blueboar (talk) 17:03, 2 February 2010 (UTC)

Blueboar, I like your additions. I have split the paragraph - as you said, the issues of whether and how to quote are separate. Yaris678 (talk) 18:19, 2 February 2010 (UTC)

I have tweeked a bit. Jay, does this ease your concerns about this section? Blueboar (talk) 18:22, 2 February 2010 (UTC)

Um, the first two sentences are fine. The rest, however, is the same bad advice that contradicts existing guidlines. I'll propose a sensible compromise below. Jayjg ^(talk) 02:38, 3 February 2010 (UTC)

Compromise2?

I've boldly edited the passage as follows:

When quoting, it's best to quote secondary or tertiary sources. They provide context for quotes, and indicate that they are notable, something primary sources cannot do. It is easy to quote primary sources inappropriately, in ways that constitute original research or give undue weight to a particular viewpoint. In cases where a secondary or tertiary source and primary source disagree, it's best to quote the secondary source in the article, and note in a footnote that a specific primary source disagrees. One should not assume the primary source is correct, since the secondary source may be using a different edition or translation of the primary.

When using a quote from a secondary or tertiary source, say where you found the material. Do not attribute it directly to the primary source. For large sources, page numbers or similar information should be given to allow the quoted section to be found.

I've incorporated the valuable parts of the previous wording, and tried to stick as closely to it as possible, but ensured that the advice actually conforms with policy and guideline. Comments? Jayjg ^(talk) 02:47, 3 February 2010 (UTC)

Jay, we still seem to be talking past each other and about different things. I am talking about what to cite in situations when it is appropriate to quote primary sources. Your version diverts the discussion to what to do when you quote secondary sources ... somehow we are disconnecting.

so that I can better understand where the disconnect occurs, do you disagree with any of the following?:

1. Sometimes it is appropriate quote a primary source.
2. When it is appropriate to quote a primary source, we are allowed to quote the primary source.
3. If you do quote the original, that is what you should cite.
4. It is important to quote sources accurately.
5. The most reliable and accurate source for any quote is the source itself.
6. Sometimes secondary sources quote from primary sources.
7. Sometimes secondary sources can (either intentionally or unintentionally) misquote the primary source or quote the primary out of context.
8. We do not want to include such misquotes or quotes taken out of context
9. The best way to ensure that something that is quoted by a secondary source isn't a misquote or a quote taken out of context is to check it against the original.
10. As long as we are checking the original, we should cite that (see #5... we should always try to use the most reliable source we can)

To me, this is a logic chain. one that clearly indicates that, in situations where it is appropriate to quote a primary source, the best practice is to quote and cite to that original primary source. Is there flaw in my reasoning? Blueboar (talk) 03:52, 3 February 2010 (UTC)

In response:

1. It's generally more appropriate to quote secondary sources.
2. A trivial conclusion that should not be highlighted in a guideline, particularly given how rarely the first clause is true.
3. Not relevant to this issue.
4. It all depends on the source, what it being quoted, etc.
5. Not relevant to this issue.
6. And sometimes primary sources change, or have different editions, or different versions, or different translations.
7. Favoring quotes from primary sources over secondary will not solve this problem, and will create other problems.
8. So long as you have an exact copy of the original used by the secondary source.
9. One generally shouldn't be checking primary sources, for the reasons given above.

Now, please let me know if you disagree with any of the following:

1. Wikipedia prefers secondary sources over primary, primarily because of the dangers of WP:NOR and WP:UNDUE when using primary sources.
2. Primary sources should be used rarely, and with great care on Wikipedia, because of these issues.
3. Primary sources often have different editions, versions, translations.
4. When providing quotes, secondary sources may be quoting a different edition, version, or translation than a Wikipedia editor is familiar with or has access to.
5. If a secondary and primary source give different versions of a quotation, one cannot always assume that the secondary source is incorrect and the primary source correct.

To me, this is also a logic chain, one that clearly indicates that one should not advise editors to prefer primary sources for quotations over secondary. Jayjg ^(talk) 02:15, 4 February 2010 (UTC)

Before I answer... do I still need to? Given your most recent edit you seem to now be satisfied with the language of the section ... so am I. Perhaps we stumbled upon a wording that resolved our disperate concerns. Blueboar (talk) 04:03, 4 February 2010 (UTC)

Well, at this point, the advice is still bad, but it's not quite as bad as before, and more redundant than anything else. Jayjg ^(talk) 05:07, 4 February 2010 (UTC)

Redundancy

Excellent. I'm glad we seem to understand what each other is talking about now.

I disagree that advice on how to quote primary sources is redundant. Jay, are you suggesting that it will happen so rarely as to be irrelevant? For an example of a good quote from a primary source, see yesterday's featured article on Marshalsea. It has an eye witness statement from the nineteenth century that definitely adds to the article.

Yaris678 (talk) 09:39, 4 February 2010 (UTC)

To which quote are you referring? Jayjg ^(talk) 00:55, 5 February 2010 (UTC)

I don't think this is what Yaris is referring to... but I noticed the quotes from Dickens, especially the one near the end of the article (last I checked "Little Dorrit" isn't considered a secondary source) Blueboar (talk) 01:41, 5 February 2010 (UTC)

Yaris, to which quote were you referring? Jayjg ^(talk) 00:49, 7 February 2010 (UTC)

I was actually referring to "Three lovely girls, the daughters of a prisoner, by visiting their father in prison, became acquainted with a villain, who, in conjunction with another fiend, accomplished the ruin of two out of three of these previously innocent females. In this case their mother attempted suicide, on becoming acquainted with their disgrace!"—Anonymous eyewitness. Yaris678 (talk) 02:50, 7 February 2010 (UTC)

That quote comes from a secondary source, specifically: Philpotts, Trey (1991). "The Real Marshalsea", The Dickensian, 87, no. 3 (autumn 1991), pp. 133–45. Jayjg ^(talk) 03:06, 7 February 2010 (UTC)

I would say that is a quote from a primary source (“Expose”) where the editor has applied WP:SAYWHEREYOUGOTIT. Perhaps this difference of perspective is related to our earlier misunderstanding. In my view, the fact that the primary source has already been quoted in “The Real Marshalsea” is one way of showing that using the quote is not being given undue weight (although it is an imperfect way since you can still quote from a secondary source selectively). I can see this makes it complicated.

For a more clear-cut example, see the article on The Slave Community. For that article, the book The Slave Community is a primary source but there are many quotes from it throughout the (featured) article. These quotes serve to put Blassingame's arguments in his own language. They also help to show that they are his arguments - not the arguments of the article - and hence serve the aim of NPOV. Yaris678 (talk) 11:38, 7 February 2010 (UTC)

I have just come across another example. The article on Ketuanan Melayu has a quote from a government-produced textbook. Textbooks are often considered secondary or tertiary sources. However, when the subject is Malay supremacy and how the idea is propagated in Malaysia, the quote is from a primary source. Yaris678 (talk) 22:08, 7 February 2010 (UTC)

Which quote do you mean? Jayjg ^(talk) 03:03, 15 February 2010 (UTC)

(remove indent) There are many quotes of The Slave Community in that article. The quote I am thinking of in the article on Ketuanan Melayu is:

A government-approved secondary school history textbook published in 2004 by Dewan Bahasa dan Pustaka, the government-owned publishing company, defined ketuanan Melayu as:

Semangat cinta akan apa saja yang berkaitan dengan bangsa Melayu seperti hak politik, bahasa, kebudayaan, warisan, adat istiadat dan tanah air. Semenanjung Tanah Melayu dianggap sebagai tanah pusaka orang Melayu.

Its English translation is as follows:

A passion for all that is related to the Malay race, such as political rights, language, culture, tradition and the homeland. The Malay peninsula is regarded as the Malays' traditional land.

Yaris678 (talk) 08:29, 15 February 2010 (UTC)

That seems like a pretty inappropriate quotation to use in the article. Why is a primary source being used for what is effectively analysis? Jayjg ^(talk) 03:37, 18 February 2010 (UTC)

I think you are putting the cart before the horse there. A secondary source analyses or interprets a primary source. This doesn’t mean that a primary source can not contain analysis, or that if it does it should not be used in Wikipedia. The important point is that this quote puts the government line on the subject of the article succinctly, and illustrates how it is taught to children. These are important things for this article. If, rather than giving the quote, the article had paraphrased it into a statement of fact, that would bring up problems with NPOV and would be an inappropriate use of a primary source. Doing it the way they have seems very sensible. Indeed, Ketuanan Melayu is a featured article. Yaris678 (talk) 12:18, 18 February 2010 (UTC)

On what basis did you conclude that "this quote puts the government line on the subject of the article succinctly, and illustrates how it is taught to children"? That, of course, is exactly the conclusion intended by the person who inserted that quote, but how do you know this quote is representative in any way of how the government defines ketuanan Melayu or teaches it to children? Perhaps this quote is an anomaly! The only way of actually knowing that "this quote puts the government line on the subject of the article succinctly, and illustrates how it is taught to children" is if reliable secondary sources indicate it does. This is an example of exactly the wrong way to use primary sources, and why such use should be avoided! Jayjg ^(talk) 00:57, 19 February 2010 (UTC)

Yes. That is the standard line against primary sources. Of course, you can only take it so far. Quoting a secondary source can also be done in a non-representative manner. Do you need to find a source that says the secondary source, or the bit you are quoting, has any relevance? Or that it gives a fair summary? At some point, the editors of an article need to use editorial judgement and reach a consensus. That applies to both primary and secondary sources. The only caveats are that you have to be particularly careful with primary sources and that you should not step over the line from editorial judgement into analysis or interpretation. Yaris678 (talk) 10:22, 19 February 2010 (UTC)

In case it wasn't obvious from my above statement, on the specific issue of the article on Ketuanan Melayu, I think using the primary source as they have shows good editorial judgment. It is not an example of the wrong way to use primary sources. Are you suggesting the quote should be removed? Or perhaps the article should be downgraded from featured status? Yaris678 (talk) 20:19, 20 February 2010 (UTC)

And that, in fact, is exactly why this proposal is a bad one. The example above is exactly the wrong way to use quotations; taking a quotation of unclear importance, from a primary source whose official status is also unclear (not everything in a school textbook is government policy), and using it to promote the view that the quote is an example of an official government position. And this proposal, and its authors, want to promote more of these kinds of policy violations, using this proposal. Jayjg ^(talk) 01:52, 21 February 2010 (UTC)

(indent)

Well. I think we have successfully got to the nub of the argument. Even if it took weeks of talking past each other.

Firstly, I'd like to point out, again, that Ketuanan Melayu is a featured article. This suggests that you are arguing against the way that Wikipedia works in practice.

Secondly, nothing in the wording of the current policy section WP:PSTS says that this quote should not be used. It says “primary sources are permitted if used carefully. All interpretive claims, analyses, or synthetic claims about primary sources must be referenced to a secondary source, rather than original analysis of the primary-source material by Wikipedia editors.” No such claims about the primary source are made.

Thirdly, I recognise that what we have been talking about could be called an “implicit claim”. i.e. By including this quote from a primary source, the editors are implying that it is representative of something. This sort of issue is dealt with by WP:NPOV (and especially WP:UNDUEWEIGHT) rather than WP:NOR. I have noticed a tendency to describe any kind of POV pushing as original research. This may help deter individual POV pushers but it is damaging in the long run as it marginalises WP:NPOV.

Fourthly, including this quote is entirely consistent with WP:NPOV.

Fifthly, I can anticipate that you will say "I know you say it is consistent with WP:NPOV, but can you find a source that says it is at all representative?" The response is that such thinking is based on WP:NOR. The important point is not whether I say anything, it is what the editors of Ketuanan Melayu agree through consensus. Specifically whether they think this quote is helpful and representative of anything or whether including it gives undue weight to the opinion of three text book writers in Malaysia. That is how NPOV works.

Yaris678 (talk) 08:43, 21 February 2010 (UTC)

1. Ketuanan Melayu was promoted in early September 2006, 3½ years ago. FAC evaluations vary in quality, the process does not ensure perfect compliance with policy, and in any event FAC standards have changed significantly since then.
2. Your own description of the quote made it clear it was being used for an "interpretive claim", even if implicit.
3. WP:NOR is not WP:NPOV, and the use of the quote violates WP:NOR.
4. Your assertion, is just that, an assertion, and a false one at that, since the quote also fails WP:UNDUE.
5. The policies work together, including WP:NPR and WP:NPOV. A small number of editors of a page cannot "agree through consensus" to ignore policy.

This discussion again highlights why this proposed guideline is inappropriate; it promotes violation of policy, and is written by editors who do not accept or recognize that their policy violations are not allowed. One should not modify or write guidelines with the intent of loosening up policy to allow specific kinds of violations that certain editors want to get away with; yet, it is regularly attempted, and this is just one example. Jayjg ^(talk) 15:38, 28 February 2010 (UTC)

No one is trying to loosen anything. In response to your numbered points.

1. I never said featured status guarantees compliance.
2. I think you might have been reading too much into what I said. Reading the article shows no interpretive claim is made for the quote.
3. Which part of WP:NOR does it violate?
4. I discussed WP:UNDUE in my previous post. How do you think that is violated?
5. I don’t think anyone has agreed to ignore any policy. I guess that is related to my points above.

Yaris678 (talk) 16:14, 28 February 2010 (UTC)

Hi Jayjg,

I notice that you haven’t responded to my points above. I don’t know if this means you accept them or you have had enough of discussing it. Either way, we were focusing on a pretty narrow point of a specific article. On the more general point of whether quoting primary sources is ever appropriate, I think the most obvious case is when the subject of the article is a work of literature. Such articles often quote passages – or the entire piece if it is a short poem, for example. Indeed, I think this is the sort of thing that the guideline Wikipedia:Do not include the full text of lengthy primary sources is talking about. An example of this sort of article is The Slave Community.

Yaris678 (talk) 10:49, 5 March 2010 (UTC)

Little Dorrit

OK... but the quotes from Dickens's Little Dorrit are from a primary source (and I defintiely would keep these quotes) 04:14, 7 February 2010 (UTC)

Well, it's a novel, and the quotes are used for color, not for citing facts, as far as I can tell. Is that the kind of use you meant with this policy? Jayjg ^(talk) 05:10, 7 February 2010 (UTC)

It is certainly one of the uses we mean. There are lots of reasons to quote primary sources. Some are appropriate and acceptable... others are not. Blueboar (talk) 15:15, 8 February 2010 (UTC)

See the examples I gave above of primary sources being used in the articles on The Slave Community and Ketuanan Melayu. Yaris678 (talk) 16:36, 8 February 2010 (UTC)

To which quote in Ketuanan Melayu were you referring? Jayjg ^(talk) 23:16, 15 February 2010 (UTC)

See my post of 08:29, 15 February 2010. Yaris678 (talk) 08:10, 16 February 2010 (UTC)

What's the point?

This isn't a policy. It doesn't set any criteria for editing. "Deciding whether primary, secondary or tertiary sources are more suitable on any given occasion is a matter of common sense and good editorial judgment, and should be discussed on article talk pages." Great advice, so why do we need the rest of this page? Fences&Windows 21:22, 29 January 2010 (UTC)

The "common sense and good editorial judgment" statement is taken directly from the current WP:PSTS in WP:NOR. You are right that things that are a matter of common sense are normally best left out of policies. This is just one reason why we are proposing that some of the issue of primary, secondary or tertiary sources be removed from a policy and put into a guideline. Yaris678 (talk) 02:29, 30 January 2010 (UTC)

To me, the point is that multiple policies and guidelines restrict the use of Primary sources in some way (or, alternatively, require the use of Secondary sources). The appropriate and inappropriate use of Primary sources is discussed at WP:NOR, WP:BLP, WP:RS, WP:Notability (just to name a few). I think it would be helpful to have a page that gathers together all of these policy statements, so that editors can read about the appropriate and inappropriate use of Primary, Secondary and Tertiary sources in one place (instead of having to jump about to multiple policies and guidelines.)

From my perspective, the point is not to remove the relevant policy and guideline statements from those policies (although I could see where some of them might be edited slightly... summarized a bit with a pointer to this page for broader explanation.) Blueboar (talk) 15:24, 31 January 2010 (UTC)

Would it be worth having a discussion about what would go into the PSTS section of WP:NOR? Perhaps some combination of the lead of this page and its section on no original research. The main thing is that it doesn't need to give definitions and it doesn't need to mention circular sourcing. I suppose the approach that would change the least of the policy page would be to leave the start of that section as it is (possibly including the common sense statement, which is why I just crossed out something I said earlier) and then have two sub-sections "Policy on primary sources" and "Policy on secondary sources". Yaris678 (talk) 18:26, 31 January 2010 (UTC)

Is the concept of labeling levels of sources still valid?

Someone has to ask. The labeling by WP does not conform to outside usage very well at all. The principles really should be broken down to:

1. Is the matter one of ascertainable fact? Such facts are, by definition, non-contentious. "A U.S. quarter is one ince in diameter. " Almost any source is good enough to be sure on these.

2. Is the matter one of contentious fact?

a."John Doe was arrested for DUI" is in this category. If it is for a BLP, then the source should be a fully reliable source, not tainted in any way, and not part of anything remotely considered an "opinion piece."

b."AGW has been proven". Requires reliable sources which are not initial studies of one aspect of the issue, precis from newspapers based on such studies, or government documents which do not make the simple statement for which they are used as a cite, but which are ideally subjected to rigorous fact-checking, either through a peer-review process for scientific studies, or through an editorial process which verifies facts before publication. "Governmental reports" which are one step removed from the hands of peer-review are now especially reliable for such statements of fact ("The Stimulus saved or created 800,000 jobs" is such an example.)

3.a. Opinions of notable people must be cited to either a reliable source as in 2b, or to material under their own control (thus self-published opinions should be usable as opinions of the person who publishes them).

b. Opinions of non-notable people can only be cited in aggregate, and only where the relliable source makes the connection as to the aggregate.

4. Gossip is not usable no matter the source.

Seem radical enough? Collect (talk) 19:43, 31 January 2010 (UTC)

contentious in nature (I suspect this is pretty clear).

I don't think we could gain a consensus for doing away with source typing completely. It is very ingrained. Blueboar (talk) 20:06, 31 January 2010 (UTC)

Ingrained or not - can such a radical redefinition of why the types of sources are, or are not, allowed, make sense? Can it be melded in? Right now the dispuutes over whether a government map is RS (primary) or a Rand-McNally one is (secondary? primary? what about fake towns on it?) etc. and the like eat up hundreds of hours for editors. Whether a specialized encyclopedia is tertiary or secondary, and can the Btritannica be used as a source on an arcane topic, and whether the OED provides definitions suitable for a technical article etc. all recur on RS/N. If we can draw a logical line instead of a difficult to understand system, ought we not try?

The problem is that it has been tried... several times... it meets with strong knee jerk reaction and shouts of "this is changing Long Standing^TM policy". I have given up on that line of reasoning. You are welcome to try again if you wish to. Blueboar (talk) 20:24, 31 January 2010 (UTC)

Collect, I really like your idea but I don't think it can replace the notion of primary, secondary and tertiary sources... At least, not in the short term. I suggest you write it up as an essay. Maybe when people get used to the concept it will get upgraded to guideline status and only after that would I start arguing about whether it supersedes the concept of primary, secondary and tertiary sources. Yaris678 (talk) 20:32, 31 January 2010 (UTC)

Done as WP:Commonsense on sources. Emendations are welcome. Collect (talk) 20:52, 31 January 2010 (UTC)

one thing that strikes me offhand is that sources aren't there to provide 'facts', sources are here to provide verification of points that editors include in articles. If an editor wants to add a statement that John Doe was arrested for DUI (and that statement is includable per other policies), then any source is acceptable for verifying that statement (whether it's a police blotter, and newspaper article, or a direct quote from John Doe), so long as it can be assumed that that source has reliable information about the topic. If an editor wants to add a statement that 'AGW has been proven' things becomes more complex, because part of assessing whether the source has reliable information involves assessing whether the source is in a position to accurately describe the viewpoint of scientists writ large (though for that particular statement we could probably exclude it on wording - no scientific theory is ever 'proven').

all I'm getting at is that the issue isn't whether the 'facts' are right; the issue is whether the statement can be verified in a source that is qualified (and in a position) to make an accurate and neutral statement. I'll go take a look at your essay a bit later, maybe piddle with it a bit, if that's ok. --Ludwigs2 01:11, 2 February 2010 (UTC)

I fear you misapprehend why the word "fact" was used -- it was to differentiate "fact" from "opinion" which is the same way "fact" is currently used in policies and guidelines. I also fear that the current standards are misused by some who wish WP to be the Enquirer (User:Collect/BLP shows such reasoning by an editor) and this really ought to stop - especially noting the BLP discussions now going on. I also worry that the current "published in a newspaper" standard is too lenient considering that many newspapers operate on the principle that "paper never refused ink" nowadays. WMF has made it clear, moreover, that there are substantial legal reasons for disallowing police blotters and the like. I am not about to argue with Mike Godwin <g>. Collect (talk) 11:41, 2 February 2010 (UTC)

Would a new help file be useful?

A while back I wrote up a page Wikipedia:How to use primary sources (biological sciences), with the thought that with some discussion it might be worked up into a Wikipedia help file. Do you think this would be a good idea?

I have the feeling that much of WP:Primary, secondary and tertiary sources is meant more to help editors than to create new rules. Perhaps it might best be presented as a new help file, with any actual change in policy left for discussion on existing policy pages. Mike Serfas (talk) 02:09, 2 February 2010 (UTC)

As a help file -- it makes substantial changes to the curent system. Collect (talk) 11:42, 2 February 2010 (UTC)

The idea here is definitely NOT to create new rules. The proposed policy/guideline is meant to be a restatement and consolidation of the various existing policy and guideline statements that involve source types... so that editors can find them in one place rather than having to hop around to multiple policies and guidelines. It is also meant to allow some of those policy and guideline statements to be rewritten or trimmed (for example, NOR could point the reader to this page for details and definitions as to what constitutes primary, secondary and tertiary sources ... which would allow it to better focus its discussion on to what constitutes OR when we use these source types) Blueboar (talk) 15:27, 3 February 2010 (UTC)

Mike, I don't think your understanding that guidelines provide new rules is correct. WP:PG says "Policies describe standards that (within the limits of common sense) all users should normally follow, and guidelines are meant to contain best practices for doing so." Yaris678 (talk) 09:00, 4 February 2010 (UTC)

Sorry, I didn't mean to suggest that, but the conversation seemed to suggest that policy changes might be intended. Mike Serfas (talk) 04:51, 5 February 2010 (UTC)

Some people are arguing that we are proposing a material change in policy. This view is generally formed from a misunderstanding of what current policies and guidelines say. This page merely draws some of it together and adds a little extra explanation in some cases. Yaris678 (talk) 06:32, 5 February 2010 (UTC)

Actually, I could see this proposal resulting in changes to policy (even though that is not the intent)... one of the by-products of creating a page that combines into one location all the disperate policy statements relating to the use of primary and secondary sources is that the result highlights where some of these disperate statements are inconsistent or conflict with each other. Even if this proposal never becomes a policy/guideline, that highlighting has been a useful excersise. Blueboar (talk) 15:53, 5 February 2010 (UTC)

Primary sources require expert interpretation

I've been away from this discussion for a while and just read through the current draft; I find that it misses one of the crucial reasons that the use of primary sources has long been deprecated. Although primary sources are close to the events they describe, their descriptions are not transparent but require careful expert evaluation, which is one of the things which Wikipedia editors are prohibited from doing. It is not just that primary sources invite synthesis, it is that the evaluation of primary sources is itself synthesis and is therefore prohibited. Primary sources can only be used when accompanied by secondary (or tertiary) sources that provide that expert evaluation.

A passage expressing that point is clearly lacking in this draft and would most appropriately be placed in the section on Original research. --SteveMcCluskey (talk) 04:01, 4 February 2010 (UTC)

I mostly agree. Except for your statement that: "Primary sources can only be used when accompanied by secondary (or tertiary) sources that provide that expert evaluation." ... This is not quite true... a primary source can be used for a simple discriptive statement as to what is contained in the primary source itself (for example, when giving a simple plot summary for a book or movie.) Or have I misunderstood your point. Blueboar (talk) 04:05, 4 February 2010 (UTC)

OK on that limited case of the plot of a fictional book or movie, you're right. If, however, we moved on to discuss what the novel Jane Eyre says about nineteenth century morality, religion, social class, or gender relations we're clearly getting into original research and have to cite how experts interpret that primary source. (I had just pulled that novel out of the hat, and find that the article discussing it has exactly the kind of interpretative problem I am concerned with). --SteveMcCluskey (talk) 14:00, 4 February 2010 (UTC)

The problems you identify with article on Jane Eyre are examples of interpretive claims. These need to be referenced to a source that makes the same claim, as clearly stated in the first paragraph of the proposed guideline. Yaris678 (talk) 14:18, 4 February 2010 (UTC)

Agreed. For an interpretive claim you do indeed need to cite a source that makes that same interpretive claim... and before we get all hung up on specifying that this source must be secondary... I will just point out that for the article on Jane Eyre, the only primary source is the novel itself. The point being that any source that interprets Jane Eyre would, automatically, be a secondary source... no need to specify. Whether any given source is a reliable secondary source is a different matter. Blueboar (talk) 17:01, 4 February 2010 (UTC)

There are many kinds of primary sources and not all require expert interpretation. For example, there are may business databases which list the address, gross sales, numbers of employees, etc. It doesn't require any interpretation to say that "Bob Jones is the CEO of Amalgamated Widgets" based on such a primary source. We should be careful of making generalizations about sources unless we're careful to consider the full range.   Will Beback  talk  18:44, 4 February 2010 (UTC)

The example you give is one of the few cases where a primary source does not require interpretation, but interpreting the significance of the gross sales, number of employees, etc. does require interpretation. I'm concerned about opening the floodgates to wikilawyers who would claim that their interpretations are only stating what's in the primary source. --SteveMcCluskey (talk) 19:43, 4 February 2010 (UTC)

The wording of the guideline should reflect that not all primary sources require interpretation. It's possible to include information without making OR decisions about its greater significance. Telling readers that WalMart has 100,000 employees (or whatever) informs the reader without leading them to any conclusion. However I grant that primary sources are full of pitfalls, such as making sure (in this example) that the same company is being referred to, or whether the definition of employee includes independent consultants, etc. But that's a slightly different problem than the one you've raised here.   Will Beback  talk 

We want to prevent throwing the baby out with the bath-water. Using a primary source in appropirate ways is fine. Using it in ways that constitutes OR is not. WP:NOR already makes it clear that any interpretation, analysis or conclusion requires a secondary source that directly contains that same interpretation, analysis or conclusion. Blueboar (talk) 20:18, 4 February 2010 (UTC)

Undue weight

I am thinking of adding a section on WP:UNDUE. Often the problem with a primary source is that it has a very narrow perspective. Using primary sources can give undue weight to the views and experiences of the author of the primary source. Secondary sources often have a wider perspective, but this is not guaranteed - the secondary source may be analysing or interpreting a single primary source or it may be written from a distinct perspective of its own. Anyone got any thoughts on this? Yaris678 (talk) 11:26, 5 March 2010 (UTC)

I have now added a section on undue weight. I think this is a BIG improvement. Undue weight can be a big problem with primary sources.

Any comments?

Yaris678 (talk) 23:49, 14 March 2010 (UTC)

Failed Proposal?

SlimVirgin, You seem to want to describe this proposed guideline as failed. I didn’t understand you last edit summary, but the previous one talked about a lack of consensus. As far as I can see, we haven’t actually started the process of measuring consensus yet, i.e. an RfC. The template at the top of all proposals says "The proposal may still be in development, under discussion, or in the process of gathering consensus for adoption." I would say we are still in development and discussion. Specifically, there is a discussion on quoting primary sources (although I think that has ended now so I may remove the underdiscussion tag from that section) and I recently suggested adding a section on undue weight. Once these issues are dealt with, it may be appropriate to go for an RfC.

I understand that you think this policy is a bad idea, but I don’t think anything is gained by rushing.

Yaris678 (talk) 08:28, 11 March 2010 (UTC)

Yaris's summary of the arguments, version 2

N.B. This is a summary, by one of the parties, of discussions on this talk page. Please see the above posts and User talk:Yaris678/PSTS/Archive 2 for a fuller picture.

Yaris678 (talk) 10:11, 26 March 2010 (UTC)

Right everything seems to have gone quiet, so I thought I would take this opportunity to summarise the discussion so far. I will also add any obvious counter-arguments that haven’t been added yet. Obviously, I could be biased in this. I will try to be as fair as possible but all comments and suggestions welcome.

As before, I will give each argument a letter and a number. The letter is either p for pro separate guideline or a for anti separate guideline. Counter arguments have their number appended to the argument they are countering. I hope it is obvious when you see it.

p1 - It allows the PSTS section of WP:NOR to concentrate on the original research aspect of PSTS. This fits with WP:Policies and Guidelines#Content, which says “Content should be within the scope of its policy”. The current section also gives definitions and talks about WP:CIRCULAR, which dilutes the message.

p2 - It allows us to write slightly more in the guideline and clarify a few things. For example, WP:UNDUEWEIGHT is one of the reasons that primary sources can be a problem, but that doesn’t get a mention in the current section of WP:NOR. Another example - People are always asking "are newspapers primary or secondary?" and this is addressed by the proposed guideline.

p3 - The current situation has lead to many (inexperienced) editors miss-reading the policy to mean, for example "Primary sources are not allowed." This problem could be reduced by having a separate guideline for the reasons given in p1 and p2.

p4 - Part of the reason why PSTS is important is that a primary source can not establish a topic's notability - but WP:NOTE is a guideline and not policy. Referring to a guideline from a policy could give the impression that the guideline has been upgraded to a policy.

p5 - WP: Identifying reliable sources is a guideline that deals with a different aspect of source categorisation. It is arguably more important than WP:PSTS, so why is WP:PSTS at the policy level?

p5.a1 - WP:IRS should never have been made into a guideline anyway. See a4.

a1 - This suggestion is the result of inexperienced editors thinking the current system doesn’t work. In reality it does.

a1.p1 - The current situation probably does work well for experienced editors. It has been observed to not work well for some inexperienced editors. If we clarify things it will make things easier for the inexperienced without making them any harder for the experienced.

a1.p2 - The implication that none of the people putting forward the suggestion has much experience is not true. Info from toolserver.org:

Username: Blueboar

User groups: rollbacker

First edit: Dec 07, 2005 18:41:06

Unique articles edited: 1,739

Average edits per page: 12.26

Total edits (including deleted): 21,323

Deleted edits: 563

Live edits: 20,760

a2 - The main reason for even talking about primary, secondary and tertiary sources is to do with original research. Even WP:NOTE is an application of WP:NOR - i.e. using your own judgement on the notability of a topic, without it being mentioned in secondary sources, is OR.

a2.p1 - "No original research based on primary sources" is an application of "No original research". However, that does not dictate that WP:PSTS should be part of the WP:NOR page. For example, the guideline WP:Do not include the full text of lengthy primary sources is an application of the policy WP:What Wikipedia is not; the guideline WP:Conflict of interest is an application of the policy WP:Neutral point of view.

a3 - This is a proposal to remove a key part of WP:NOR and reduce it to guideline status, which means people can happily ignore it.

a3.p1 - People do not ignore WP:IRS.

a3.p2 - Under the proposal, WP:NOR would still talk about the original research aspects of primary, secondary and tertiary sources. It just wouldn’t have its message diluted, as per p1.

a4 - This would be a policy fork that would just confuse the situation, in the same way as WP:IRS, a policy fork from WP:V.

a4.p1 - People seem to be able to use WP:IRS and WP:V without any problems.

a4.p2 - While some policy forks (as you call them) should be avoided, there are times when it is a good thing. Wikipedia:Policies and guidelines says "Policies describe standards that all users should normally adhere to, and guidelines are meant to contain best practices for doing so." This would be impossible without policy forks.

a5 - Some of Yaris's edits promote exactly the kind of non-notable primary source material that the NOR policy seeks to avoid.

a5.p1 - There is no point in an argumentum ad hominem, given that Yaris is not the only one putting forward this suggestion.

a6 - Much of this is strong redefinition of WP policies and guidelines, guaranteed to introduce Wiki-Drama no end.

a6.p1 - On the contrary, it is mostly a restatement of existing policies and guidelines. There is a change of emphasis, but this is designed to improve people's understanding.

a7 - The guideline shouldn’t encourage people to quote from primary sources.

a7.p1 - It doesn’t say quoting of primary sources is a good thing per se. It does say that if you do quote from a primary source it is helpful to cite the relevant passage in the primary source itself, rather than referring to a quotation of it in a secondary source. This prevents us repeating any errors in quotation and allows the reader to find the quote in its original context.

a7.p1.a1 - In that case, the advice is redundant because quoting from a primary source is almost never appropriate.

a7.p1.a1.p1 - It is often appropriate to include, say, a quote from a book in and article about that book.

(There then followed a long argument about some specific examples, which didn’t really get anywhere).

a7.p2 - This is perhaps the only new guidance given by the proposed guideline. It is not critical to the guideline as a whole and could be dropped if there is a consensus against it, without harming the proposal as a whole.

a8 - This isn't a policy. It doesn't set any criteria for editing. "Deciding whether primary, secondary or tertiary sources are more suitable on any given occasion is a matter of common sense and good editorial judgment, and should be discussed on article talk pages." Great advice, so why do we need the rest of this page?

a8.p1 - It isn’t a policy, it is a proposed guideline. The text on common sense is taken directly from the current section of WP:NOR.

a9 - I find that this proposal misses one of the crucial reasons that the use of primary sources has long been deprecated. Although primary sources are close to the events they describe, their descriptions are not transparent but require careful expert evaluation, which is one of the things which Wikipedia editors are prohibited from doing. For example the article on Jane Eyre interprets the novel to say things about 19th-century morality and so on, which it shouldn’t do.

a9.p1 - The problems identified with article on Jane Eyre are examples of interpretive claims. These need to be referenced to a source that makes the same claim, as clearly stated in the first paragraph of the proposed guideline. However, if you think that more should be said on the subject then creating this guideline is a good thing as it creates space to do that.

Yaris678 (talk) 10:11, 26 March 2010 (UTC)

Accept guideline on primary, secondary and tertiary sources

I would like people’s opinion on this proposed guideline. Should it be accepted as a guideline? If not, are there any bits of it that should be mentioned elsewhere? e.g. The linking of primary sources and undue weight. Alternatively, is it acceptable if certain bits are removed? Yaris678 (talk) 12:28, 10 April 2010 (UTC)

• Support accepting the proposal in full (as initiator of RfC). See #Yaris's summary of the arguments, version 2 for reasons why. Yaris678 (talk) 23:20, 10 April 2010 (UTC)
• Wasn't this already a standalone guideline at one time before it was merged into WP:NOR (at WP:PSTS) not very long ago? Gigs (talk) 18:56, 13 April 2010 (UTC)

I don't think so; the redirect from WP:PSTS to Wikipedia:No original research#PSTS dates back to 01:05, 23 June 2007. --SteveMcCluskey (talk) 19:11, 13 April 2010 (UTC)

• Support accepting the proposal. We have to explain this stuff over and over at WP:RSN, it would be nice to be able to reference this as a guideline. Dlabtot (talk) 00:06, 28 April 2010 (UTC)
• Weak oppose - I'm not sure Undue Weight section and to a lesser extent Quoting Primary sources works well with fiction, especially when there is the problem a vocal minority have with plot synopsizes, including info on characters and other elements, which have been done through primary sources including, on occasion, quoting passages from the works. These sections would seem to back the that very small, but vocal, minority when consensus is the opposite. This may be partly due to the fact when writing it, no one was considering this impact, but reading it from such a perspective it seems to go against the general consensus for what is established with regard to fiction articles.陣内_Jinnai 17:24, 28 April 2010 (UTC)
• Strong oppose, per reasons articulated at length above. Jayjg ^(talk) 06:40, 27 May 2010 (UTC)

Problems identified

OK. The RfC has ended and we seem to have one weak oppose. So let's deal with that before taking this any further.

Jinnai, can you clarify what problems you see with the sections on undue weight and quoting primary sources? The existing section on PSTS says "A primary source can be used only to make descriptive statements that can be verified by any educated person without specialist knowledge." Simple synopses would seem to fit into that category. Are you saying that some people believe that they don't? Is the problem that a lot of people don’t think descriptive statements are acceptable? Is the problem that some people have a tendency to slip interpretive claims into their synopses? To my mind, the main caveats to "descriptive statements are acceptable" are that you need to ensure that you are not giving undue weight and that the topic itself is notable

Yaris678 (talk) 13:01, 12 May 2010 (UTC)

Alright I'll go about this point-by-point.

• Using primary sources can give undue weight to [...] the experiences of the author of the primary source. Secondary sources often have a wider perspective since they will analyse many primary sources or seek to put a single primary source into a wider context.
  • Authors for literature are generally not considered the primary source, but a non-independant secondary source. This would essentially mean that the author's views on their book, which in literature are taken to be very important, especially in creation of the narrative, as secondary to outside observers. This is not saying outside opinion doesn't matter, but that because of the unique way in which authors of literature are not considered the primary source-the work itself is-this makes it seem as though their opinion does not matter when compared to more independent secondary sources. For non-critical commentary/real-life impact questions, they are generally considered among the foremost experts on their works and any bias they may have does not matter because they created it. To an extent this includes analysis, although such cannot be used to show notability.
• Providing an additional citation of a secondary source that quotes the primary can help to show that an article is not giving the quote undue weight and can also help the reader to find the quote if the primary source is difficult to get hold of.
  • That sentence is the one that is likely the problem. Very rarely will anyone quote a particular passage or phrase unless it is stand out like "All your base are belong to us". I believe that this would be used to quash any quotes that wouldn't be able to find a secondary source quoting that passage.陣内_Jinnai 19:24, 12 May 2010 (UTC)

Hi Jinnai,

I am still struggling to understand what you are getting at. Please don’t take that as an insult. I think it is mostly because the people who previously argued against this proposed guideline were under the impression that it would unleash a hoard of primary sources on an unexpected encyclopedia, whereas your concerns are very different.

Why do you think an author’s opinion on his/her work being a secondary source will mean that some people will be put off from using it if they follow this proposed guideline? To me the important thing is the WP:NPOV policy. Are you just saying that we need to say in the proposed guideline that an author’s opinion on his/her work is a secondary source? That seems like a reasonable addition, although we might want to add a caveat on independence as you did in your post.

The proposed guideline doesn’t say that citing a quote to a secondary source is required. It just says that it is one way to show that you are not giving it undue weight. This all depends on context of course. For many quotes, no one would ever suggest that the quote is giving undue weight.

Yaris678 (talk) 12:35, 14 May 2010 (UTC)

Yes, for authors of literature, I think you need to make it clear that they are secondary sources, but sources that cannot be used to establish notability per WP:GNG or the SNG.

For the latter, I realize this. However, in fiction, there is a very active and vocal minority who use policy and guidelines to enforce the removal of all plot material if its not commented on by a secondary source. I really don't want to hand them ammunition when consensus is clearly against this.陣内_Jinnai 00:22, 15 May 2010 (UTC)

I have added something on fiction in the section on context. I'm not sure how to proceed with your other point. We could easily get this proposed guideline bogged down in an argument between the two sides you describe. The current wording seems quite neutral to me. Have you got a suggestion? Yaris678 (talk) 09:26, 15 May 2010 (UTC)

Easiest way would be to asentance for fiction and link it to WP:Manual of Style (writing about fiction)#Primary Information which describes in detail what information can/cannot be used thus passing the buck to that guideline while stating unequivocally that fiction shouldn't be treated differently.陣内_Jinnai 18:21, 16 May 2010 (UTC)

That's a really useful bit of the Manual of Style! I have provided a link to it from the "See also" section. Yaris678 (talk) 07:51, 17 May 2010 (UTC)

As was explained above, there are strong objections to this proposed guideline, which have never been resolved. Jayjg ^(talk) 06:40, 27 May 2010 (UTC)

I had a feeling this would happen. That is why I didn't act on the results of the RfC straight away.

There certainly were objections raised previously and I agree they should be resolved. That is why I took the trouble to list all the arguments and counter arguments made. Perhaps it would help if you added anything to the list I made above that you think is relevant. e.g. Have I missed out or misstated any anti PSTSPROP arguments? Are any of the pro PSTSPROP arguments invalid for any reason?

Yaris678 (talk) 12:37, 27 May 2010 (UTC)

Which statements there summarize my objections? Jayjg ^(talk) 23:07, 1 June 2010 (UTC)

I've read the above sections, and I'm having a little trouble understanding what precisely your objections are, other than 'PRIMARY SOURCES = BAD'.

No doubt I've misinterpreted and your actual objections are more substantive and valid. Perhaps you could summarize your objections? Dlabtot (talk) 23:28, 1 June 2010 (UTC)

Jayyjg, I believe our discussion is covered by "a7 - The guideline shouldn’t encourage people to quote from primary sources." and the arguments and counter arguments I list below it. But if you think there is anything I have missed then please bring it up. Yaris678 (talk) 10:53, 2 June 2010 (UTC)

The guideline still promotes using primary sources for quotes over secondary. This is a fatal flaw. You've offered to remove that bad advice, but not acted on it, and, in the past, reverted any attempts to do so. I'm not going to try editing this proposal myself, since it is so thoroughly owned. Jayjg ^(talk) 22:10, 13 June 2010 (UTC)

Not going to try editing it? Beyond marking it as failed, presumably. I have reverted that edit. You may consider that to be me assuming ownership but I would argue that it is only fair - we haven’t actually finished the discussion about what to do with the proposed guideline. I also hope you realise why I created the heading that you removed, it made the difference in timing of the comments clear and the two discussions were on different subjects. But have that one your way if you insist.

When you say I offered to remove the bad advice, are you referring to "a7.p2 - This is perhaps the only new guidance given by the proposed guideline. It is not critical to the guideline as a whole and could be dropped if there is a consensus against it, without harming the proposal as a whole."? I stand by that statement, but this is the first time that anyone has responded to it. Does that seem like a sensible way to resolve the dispute?

Yaris678 (talk) 23:16, 13 June 2010 (UTC)

Concluding the RfC

OK... so it looks like the RfC had:

• Two people voicing support for accepting as a guideline.
• One person voicing week opposition. The reasons for the opposition was discussed and (I think) addressed by some changes to the proposed guideline.
• One person voicing strong opposition after the 30-day RfC period. The reasons for the opposition were discussed but the opposer stopped contributing.

This being the case, I think the best thing to do is to accept the guideline on primary, secondary and tertiary sources.

No one expressed an opinion on changing the contents of the section on the subject of primary, secondary and tertiary sources in WP:NOR. In addition, any such changes should really be discussed on WT:NOR. Therefore, I think the best thing is to leave that as it is for the moment. The guideline will need a shortcut and I suggest WP:PSTSGUIDE or WP:PSTSG. Yaris678 (talk) 21:48, 13 June 2010 (UTC)

Absolute nonsense. Two supports and two opposes is not enough to turn something into a guideline. Strong opposition has been clearly expressed throughout this proposal, by editors you've mentioned and editors you have not. Get some actual consensus for this first; you know, a couple of dozen people supporting, and almost no-one objecting. Meanwhile, as this proposal has obviously not gained any wide consensus (or even a narrow one), I've marked it as failed. Jayjg ^(talk) 22:03, 13 June 2010 (UTC)

I agree with Jayjg. this sort of isolated and almost private discussion is no consensus on anything. I;ve just learned of it, and added a comment, which summarizes my comments on earlier versions. This is such a wide-ranging matter that the true guideline is what we do on wikipedia, and I dod not think it capable of summary. DGG ( talk ) 00:39, 14 June 2010 (UTC)

I concur as well. I am one who voiced support but such things do require a wider consensus. Dlabtot (talk) 00:46, 14 June 2010 (UTC)

I played a minor part in early discussions and dug out some sources, but agree that this (like earlier attempts) has failed to gain consensus and is something of a walled garden. --SteveMcCluskey (talk) 00:56, 14 June 2010 (UTC)

So what do you reckon we should do now? Wait to see if more people turn up? Another RfC? Attract people's attention some other way? Jayjg would obviously like to declare the proposal failed.

Related fact - An hour ago I discovered that there had been a previous attempt to create a guideline on this subject and left a message with each user who had edited it or its talk page (including DGG). Yaris678 (talk) 01:02, 14 June 2010 (UTC)

N.B. RfC is the method suggested for getting consensus at WP:PROPOSAL. I'm a bit disappointed more people didn't respond. Yaris678 (talk) 01:08, 14 June 2010 (UTC)

Had I participated for the RfC, I would have been pretty disappointed if 4 people indicated enough consensus to implement what seems to be described here as a significant policy change. BigK HeX (talk) 07:19, 7 July 2010 (UTC)

Well, yes, we are continuing the discussion... but I don't think it is much of a policy change. No one has really said how this guideline, as currently written, contradicts any policies. Yaris678 (talk) 07:55, 7 July 2010 (UTC)

Primary publications in scientific usage, once more

I'm returning to this issue if only to report that I've finally found the full text of the report by Ellsworth B. Cook, "Proposed Definition of A Primary Publication" that we had discussed earlier. It was originally published in the Council of Biology Editors Newsletter, November 1968, pp. 1-2 and was reprinted in Herbert Stegemann and Barbara Gastel, "Council Classics", Science Editor, March – April 2009, Vol 32, No 2, pp. 57-8.

As I had suspected, the central concern of the original report was with the question of scientific priority; which publication was the first to disclose new data and ideas. To quote from the report:

In addition to their functions in communication and documentation, primary publications establish for investigators the priority of scientific observations and ideas. Reference and retrieval-type publications, consisting of citation indexes, abstract services, and other means for providing summary and broadened distribution or awareness have come to be called secondary publications. They have important functions in the communications network but they do not establish priority: they record, index, or retrieve the priorities already established in primary publications.

After reading this discussion, it doesn't seem that this sense of a primary publication.which has become common in scientific usage, is in any way related to the question of primary sources in Wikipedia. The issue of priority in discoveries is not our concern. --SteveMcCluskey (talk) 20:08, 25 May 2010 (UTC)

Steve,

Thanks for finding that reference. It’s a really good one to have.

When I started writing the draft guideline I was of a similar opinion to that which you are now. After all, a primary publication could contain considerable analysis and interpretation. However, I can see that it is not that clear cut. Take a look at the section on undue weight. You can see that primary publications will concentrate on one set of observations. You will often want to find a secondary publication for the same reason as you will often want to find a secondary source.

If there is only one paper on something, even by a renowned expert, it is often best to describe the contents of the source, as you would any primary source, rather than taking the contents as fact. For example, you could say “Astronomer X has recently presented evidence that black holes have property Y [Ref to paper by astronomer X]”.

Yaris678 (talk) 12:58, 27 May 2010 (UTC)

I have just changed the proposed guideline in line with what I said above. i.e. it is a grey area. Yaris678 (talk) 21:22, 9 June 2010 (UTC)

The reference given there is a specialized use of no direct relevance to Wikipedia. The true primary sources of scientific research in an experimental science like biology are the direct on-the-spot records experiments, not the published reports of the experiments: laboratory notebooks, field observations, and the like. The direct records of these experiments are not normally publishable as a scientific paper--a paper selects from them and interprets them, but as this paper is the closest thing to the data that is published, , it is called a primary publication. Normally, scientists in examining other's work do not look behind it to the actual records, but sometimes they do: the print-outs of instruments readings are used when needed in investigations of fraud, and lab notebooks are routinely used for priority discussions for patents. These records are normally preserved forever, but they are not part of library collections or available to the public.

It does not hold in any other field. In a theoretical science, such as theoretical physics or mathematics, there are no observations to record. The primary records in a sense are the scientists notes of his attempted calculations. when finished, they are revised and published, and that is the primary publication.

In the humanities, things are quite different: the primary sources are the texts of the works being studied. The work of the scholar is an interpretation of this, and is technically a secondary publication, but they are the actual primary source for the interpretation.

History is also special, In history, the records of the events are the primary sources, which may be published or unpublished. In particular, newspapers, which we traditionally consider in WP as a secondary source, are invariable classified in historical bibliography as a primary sources just as documents are, because only rarely is it possible to get behind them closer to the events. There is a sense in which reporters' notebooks if available are nearer a true primary source, but these too are actually an interpretation of what is told to the reporter or what the reporter sees, rather than the events themselves.

Therefore I think it wrong and misleading to refer to the academic classification of sources as relevant to wikipedia. DGG ( talk ) 00:36, 14 June 2010 (UTC)

DGG I think we agree on the irrelevance of the concept of "primary publication" to the Wikipedia concept of "primary source," which as I understand it's origins, grew out of the historians' definition. Your comment on laboratory notes being the primary source in experimental science is right on target, as is your distinction between experimental and theoretical science. --SteveMcCluskey (talk) 00:53, 14 June 2010 (UTC)

I also agree. The proposed guideline mentions the concept of primary publications partly to show that they are not the same thing as primary sources. Yaris678 (talk) 01:16, 14 June 2010 (UTC)

I agree that the Cook reference is not useful for Wikipedia, but I want to make sure that nobody thinks that we are crafting a novel definition of primary and secondary sources for use in Wikipedia that differs from the standard academic definition. We should stick with the well-trodden academic definitions, derived from the field of historiography. These definitions apply to all fields of study, including science and technology, because in essence, when you are analyzing sources, you are looking at the historical development of those sources over time--how later sources build upon earlier sources. If there is anything in this proposed guideline that is inconsistent with the primary source and secondary source articles, then we are merely adding confusion to a topic that is already difficult for most Wikipedia editors to understand. COGDEN 06:24, 14 June 2010 (UTC)

The definitions are taken from Wikipedia:No original research#Primary, secondary and tertiary sources. These definitions are close to those given in the respective encylcopedia articles. However, a problem, which I noted a while ago, is that these articles are all of a start class – or unrated. Yaris678 (talk) 14:19, 14 June 2010 (UTC)

Yes, that's a problem. But I think the articles will be good enough to rely upon, even if they are poorly written, as long as they cite the most relevant, consensus-level academic (preferably secondary) sources. I think that if there is any uncertainty about the difference between primary and secondary sources, it ought to first be hashed out in the articles, and then incorporated here only after consensus is reached in the articles. That said, I don't see anything in this guideline that is really inconsistent right now, but any proposal such as adding Cook's definition probably should be discussed in the articles first. I don't think the guideline should be leading in the effort to define primary and secondary sources. It should be following. COGDEN 18:42, 14 June 2010 (UTC)

Edits

I've made some copyedits and minor content tweaks; please review: [1] --JN466 02:54, 14 June 2010 (UTC)

"Quoting primary sources" rewrite

As there were concerns about the section on quoting primary sources, I have rewritten the section as follows:

Quoting primary sources

Quoting from a primary source is often problematic. It is easy to quote primary sources selectively, out of context, or in other ways that are inappropriate and constitute original research or give undue weight to a particular viewpoint. In contentious cases, providing a citation of a high-quality secondary source that quotes the primary source in the same way is essential to show that an article is not giving the selected quote undue weight. If no such secondary sources can be found, this is an indication that the article contains original research and undue weight problems that need to be addressed.

When citing a secondary source quoting a primary source, it is good practice is to verify that the secondary source is quoting the primary source correctly, and to add an ancillary citation to the primary source itself.

For large sources, allow the quoted section to be found by providing page numbers or similar information. Do not include the full text of lengthy primary sources in Wikipedia; but if such a source is out of copyright, it may be appropriate to include it in Wikisource.

Thoughts? --JN466 15:49, 14 June 2010 (UTC)

I think that text more accurately reflects current practice than the previous text. Therefore it is a good thing. My only slight quibble was that it dropped the reference to WP:SAYWHEREYOUGOTIT. I have stuck in some new text in that makes sense in the new context. Yaris678 (talk) 16:57, 14 June 2010 (UTC)

Thanks, that's a good addition. Overall, I think this would be a useful page to have as a guideline. --JN466 03:11, 15 June 2010 (UTC)

Jayen466, I think your wording has improved the section. Jayjg ^(talk) 03:15, 20 June 2010 (UTC)

I still say that can cause some issues with fictional articles that may use a particular quote of text and I think such should be left more to indivisual basis. In addition, it can cause problems for famous historic documents when quoting long passages as often smaller snippets are usually quoted rather than lengthy sections.陣内_Jinnai 05:40, 20 June 2010 (UTC)

We do say, "in contentious cases". If an editor cannot convince others on the talk page that a lengthy quote is WP:DUE, then I think it's fair that the onus should be on them to provide a precedent. --JN466 12:17, 20 June 2010 (UTC)

Thanks. --JN466 12:17, 20 June 2010 (UTC)

Jinnai, is your concern with the sentence "If no such secondary sources can be found, this is an indication that the article contains original research and undue weight problems that need to be addressed."? Although it just says "indication" are you worried that people would read this as "proof" and go around removing uncententious quotes of primary sources? Also, with the large primary sources, is your problem actually with the guideline Wikipedia:Do not include the full text of lengthy primary sources, which has been around for some time? Or is your concern that you might be prevented from quoting the most important or relevent part of the source for fear that you will be accused of cherry-picking from the source? Yaris678 (talk) 17:57, 20 June 2010 (UTC)

My fear would be more of the latter. Long quotes have generally needed more justification as the articles aren't suppose to be quoted text-if you want that go read the source-, but often important areas may be parapharased or summarized because they aren't very quotable text either because of length issues or because the passage ins't catchy and thus quotable.陣内_Jinnai 18:08, 20 June 2010 (UTC)

Title

I've changed the title to a "how to," in case this is in danger of becoming a guideline. I hope it doesn't become one, because there's a problem of it clashing with Wikipedia:No original research#Primary, secondary and tertiary sources. This proposal started life as a fork of that.

Therefore, in order to emphasize that this is a "how to" page, and not a "whether to" one, I've moved it to Wikipedia:How to distinguish primary and secondary sources, as that seems to be its focus. Or it could be "Distinguishing primary and secondary sources," or "Handling primary and secondary sources." That would keep it in line with Wikipedia:Citing sources and Wikipedia:Identifying reliable sources, which are written that way for the same reason—as "how to" pages, to make clear that they're not in competition with the parent sourcing policy, Wikipedia:Verifiability. SlimVirgin ^{talk|contribs} 09:46, 15 June 2010 (UTC)

First reaction: I'm not sure about the rationale, and am a bit suspicious because I think SV sticks to a mistaken view of primary/secondary source distinction, but I think this was a good rename. The title now more-so begs an interesting question: "Why would you want to distinguish primary and secondary sources". This question is an old one, and is answered, but perhaps not explicitly enough. --SmokeyJoe (talk) 10:10, 15 June 2010 (UTC)

"Handling primary and secondary sources" would seems to make more sense since the proposed guideline also points the reader to important policies and guidelines - NOR, UNDUE etc. – which should be considered when using them.

That said, I don't see what is wrong with "Primary, secondary and tertiary sources". This is a "how to" page... but what would be a "whether to" page? Whether to distinguish between primary and secondary sources? As SmokeyJoe says, the important question is "Why distinguish between primary and secondary sources?" The answer is to avoid original research, to avoid giving undue weight etc., as explained in the guideline. Yaris678 (talk) 16:05, 15 June 2010 (UTC)

The problem is that you set it up six months ago as an explicit policy fork, wanting the title to reflect the primary/secondary section in NOR that you don't like, and wanting WP:PSTS to redirect here. It's getting almost no support, but whenever anyone tries to add the failed proposal tag, you revert. :) Therefore, the best thing to do if you want to develop it is make clear from the title and the focus that it's not any kind of a policy fork, but instead is a page expanding on some ideas about primary sources—a "how to distinguish," "how to handle," "how to recognize", or "how to use" page. SlimVirgin ^{talk|contribs} 16:16, 15 June 2010 (UTC)

Yes... perhaps seeking to get the shortcut WP:PSTS to redirect here was too much of a leap. A better shortcut would be something like WP:PSTSG or WP:PSTSGUIDE. I am very keen that this should be a guide on how to distinguish and which policies and guidelines to follow rather than "policy by proxy". I don't think renaming it helps though. Perhaps it would be better to have a hat note that said something like "This is a guideline on how to distinguish between different types of source and considerations when using them. The policy on primary, secondary and tertiary sources is part of the policy on original research. Yaris678 (talk) 16:52, 15 June 2010 (UTC)

Yaris, please stop these games. [2] I've moved this away from the title with the word "guideline" in it, because it's not a guideline. SlimVirgin ^{talk|contribs} 19:21, 15 June 2010 (UTC)

Accusing people of playing games isn't going to get us anywhere. Can I suggest we agree on a name for the proposed guideline?

"Handling primary and secondary sources" is better than the last name you gave it but it does perhaps sound a bit too much like "Use of primary and secondary sources", to the neglect of "Categorising as primary or secondary". There is also the fact that it doesn't mention tertiary sources.

Another consideration is that any link to the proposed guideline under its old name is currently linking to the wrong place.

Yaris678 (talk) 19:35, 15 June 2010 (UTC)

It's hard to AGF when you set up a policy fork, copy the title and shortcuts from the relevant section of NOR, practically talk to yourself for six months, but won't allow it to be tagged as failed. Then canvass people you think will support you, and move the page to a title with the word "guideline" in it. That's not playing things straight, but if you don't play it straight, it has no chance of becoming a guideline, because people will see through it. Your choice. SlimVirgin ^{talk|contribs} 19:41, 15 June 2010 (UTC)

I don't think it is helpful to simply continue to repeat these accusations ad nauseum. In what ways, specifically, do you think this differs from our established policies? Dlabtot (talk) 20:00, 15 June 2010 (UTC)

I fail to see why this isn't simply an essay, in which Yaris678 explains his views about PSTS. If it's more than his personal views, and involves proposed changes to PSTS policy, then it should be discussed in WT:NOR, which covers WP:PSTS. Anything else just adds confusion. Crum375 (talk) 19:53, 15 June 2010 (UTC)

Slim, maybe if I explain things from my perspective it will help you to believe it was in good faith. Firstly, although I suggested that the shortcut WP:PSTS redirect here, I never implemented that suggestion and, like everything, it was always something subject to consensus. Secondly, I reverted the tagging of the proposal as failed because it didn't seem like the process of developing the proposal had finished. Thirdly, I did not canvas people on the basis of whether or not I thought they would support the proposal – I left a message with each person who was involved with the related proposal Wikipedia:Primary Secondary and Tertiary Sources, whether they had edited the page itself or just the talk page. This would doubtless include people who had all sorts of opinions on that proposal, never mind this one. Fourthly, the page title with “guideline” in it was my attempt at creating a compromise title. I had tried to discuss the title with you but got no response, so I thought I may as well change it to something vaguely neutral while I was waiting.

Crum, if we can't agree on a guideline then making the proposal into an essay would seem reasonable. I thought we might be getting somewhere since some good compromises seemed to have been found on a number of things but we might return to the essay idea if there is still a problem gaining consensus.

Anyway, if we can return to the issue that this section of the talk page was supposed to be about... what are we going to call this proposed guideline (or essay, if that's what it ends up being)? I think “Wikipedia:Primary, secondary and tertiary sources” captures the subject quite effectively. It is also what the proposal has been called for a while so I think there needs to be a very good reason to go for something different. That there is a section in WP:NOR by the same name doesn't seem like a good enough reason to me. The concept of independent sources is very important to Wikipedia and yet there is an essay by that title.

Yaris678 (talk) 23:30, 15 June 2010 (UTC)

Since there is already a policy section which specifically addresses PSTS, called WP:PSTS, it seems to me that it only adds confusion to call an essay explaining that policy by the same (or similar) name. I can see calling it "How to..." or "Identifying..." but it has to be very clear that this is an adjunct to the main policy section, not a confusing overlap. And having read the current version, although it has some good ideas, I think it also introduces controversial issues, which will only add confusion if this page is given any kind of official status (e.g. guideline). In fact, this is one main reason why I would oppose giving it guideline status: even if it were perfectly synchronized with the controlling policy today, it would start drifting away and conflicting with it tomorrow. And eliminating these contradictions is a maintenance hog and a source of chronic problems, as we have with WP:RS and WP:SOURCES. The main drawback is discussion venues: we end up discussing the same issues on different talk pages, with different people, often reaching different results. Clearly centralization is what's needed for coherent policies, not balkanization. Crum375 (talk) 23:54, 15 June 2010 (UTC)

Well if we were going to summarise what the page is about in the title, I suppose it would be Wikipedia:Categorising and using primary, secondary and tertiary sources. That is a bit of a mouthful so I would prefer something like "Guide to..." or "Summary of..." but I can anticipate someone might oppose that as well.

How about using a disambiguating term in parentheses, such as Wikipedia:Primary, secondary and tertiary sources (proposed guideline)? This could obviously be changed to Wikipedia:Primary, secondary and tertiary sources (essay), or (guideline) or (failed guideline) or whatever when the status of the page changes.

Another, perhaps archaic sounding, option is to call it Wikipedia:On primary, secondary and tertiary sources

Yaris678 (talk) 07:46, 16 June 2010 (UTC)

• How about calling it Wikipedia:Handling primary, secondary and tertiary sources?
• I am not averse to having something like this as a guideline, because PSTS has historically been difficult for people to get their heads round. Having this as a guideline would potentially be helpful enough to newer users to justify the maintenance overhead involved.
• If any of the present content in the draft is at odds with PSTS or any other policy, then we should sort this out; please detail any concerns. --JN466 12:20, 16 June 2010 (UTC)

Yaris, I'm not trying to be difficult, but if I'm coming across that way I apologize, but it's just that this is turning into something of a saga. You created this six months ago because you don't like WP:PSTS, and wanted it to be removed from NOR. That raises the red flag of a policy/guideline fork—trying to demote PSTS by the back door—so you need to go the extra mile to show that that's not what you're doing. Since then, almost no one has helped to edit the page or taken part in the discussion. But a guideline can't be developed in a vacuum.

If you want to develop additional advice about how to identify a primary from a secondary source—advice that complements the policy—I'd have no problem with that, but there can't be anything in the title, or on the page, that conflicts with NOR or that seems to duplicate it. We don't want the same NOR advice repeated elsewhere as a guideline, but not as policy, because that will confuse people. And it can't be a personal essay that's had little or no input from the editors familiar with the sourcing policies. SlimVirgin ^{talk|contribs} 16:00, 16 June 2010 (UTC)

In what ways, specifically, do you think this conflicts with our established policies? Dlabtot (talk) 21:44, 16 June 2010 (UTC)

No one seems to want to answer Dlabtot's question so shall we return to the subject that we were supposed to be discussing?

I still see no problem with calling this page Wikipedia:Primary, secondary and tertiary sources. However, thinking about alternatives, perhaps the best option is Wikipedia:Overview of primary, secondary and tertiary sources.

There is also the issue of what the page Wikipedia:Primary, secondary and tertiary sources should be. It is currently set to be a redirect to the WP:PSTS section of the original research policy. Perhaps it should be a disambiguation page. It can link to that policy, this page and (perhaps) the previous attempt to write a guideline on the issue.

Yaris678 (talk) 14:08, 19 June 2010 (UTC)

• Has this proposal ever been listed at WP:VP/P and WP:CENT? If not, then listing it there might get you broader community input. --JN466 12:05, 20 June 2010 (UTC)
• As I think I said once before, I think we could use a "Wikipedia:Interpreting WP:PSTS" guideline/essay. Perhaps similar to Wikipedia:What "Ignore all rules" means. --SmokeyJoe (talk) 12:26, 20 June 2010 (UTC)

I think "Wikipedia:Interpreting WP:PSTS" is part of what this page is about.

OK. Here is an idea. Let's call the page a name which does not define its scope too rigorously, such as Wikipedia:Primary, secondary and tertiary sources (proposed guideline) and then start a new conversation about what the scope of the page should be. If we can agree on that and it implies a more specific name then we will move the page to that name. Who knows, it might even inform what we put on the page! :-)

Yaris678 (talk) 19:49, 25 June 2010 (UTC)

Requested move

The following discussion is an archived discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

The result of the move request was: Not moved, no consensus billinghurst sDrewth 03:22, 9 July 2010 (UTC)

---

Wikipedia:Handling primary, secondary and tertiary sources (proposed guideline) → Wikipedia:Primary, secondary and tertiary sources — There is a disagreement about the name of the page. The page should be reverted to the name it had before the disagreement and a new consensus sought. This requires administrator assistance as the old name of the page has been changed into a redirect to Wikipedia:No original research#Primary, secondary and tertiary sources. Yaris678 (talk) 12:27, 1 July 2010 (UTC)

N.B. The logic I am following here is effectively the Wikipedia:BOLD, revert, discuss cycle, but admin assistance is required to perform the revert. Yaris678 (talk) 09:48, 2 July 2010 (UTC)

• Oppose Unless this becomes a guideline, the proper target of the phrase is the page that actually reflects (or at least should reflect) consensus. Septentrionalis PMAnderson 14:58, 1 July 2010 (UTC)

No consensus was gained before moving this page from Wikipedia:Primary, secondary and tertiary sources. No consensus was gained before changing Wikipedia:Primary, secondary and tertiary sources to be a redirect. There is no consensus on the current name for this page or any name held after Wikipedia:Primary, secondary and tertiary sources. The sensible thing is to return to the name that had consensus for many months and then discuss things and see if we can come up with a better name. Yaris678 (talk) 16:31, 1 July 2010 (UTC)

• Oppose This is still a discussion and there is no consensus that it is ready to become a guideline. The current title accurately reflects the current state of things and changing it would be premature. --SteveMcCluskey (talk) 16:50, 1 July 2010 (UTC)
• Comment What's in a name? That which we call a rose by any other name would smell as sweet. Dlabtot (talk) 23:56, 1 July 2010 (UTC)
• Oppose As I noted elsewhere, this could become confusing to new editors, who might mistake a proposal for the real thing, while there is an official WP:PSTS policy covering the same topic. Crum375 (talk) 00:09, 2 July 2010 (UTC)

Not really the same topic. There is a policy, but there is not a guideline as to how to implement it. Dlabtot (talk) 00:14, 2 July 2010 (UTC)

• Oppose. Yaris created this page in December 2009 as an attempted policy fork of the PSTS section of NOR, which he opposes. That's why he chose the same name as it. If he wants to create a separate page as an FAQ-type adjunct to the policy, that's fine, but then it needs a name that distinguishes it from the policy. But it seems to me that this has failed as a proposal anyway, because it's been going on for six months and Yaris is practically the only contributor. SlimVirgin ^{talk|contribs} 23:08, 3 July 2010 (UTC)

Alternative names

Can someone think of a better name? The current name is ridiculously long. The word "handling" seems overly restrictive. Wikipedia:Identifying and using primary, secondary and tertiary sources? Yaris678 (talk) 05:38, 2 July 2010 (UTC)

The above discussion is preserved as an archive of a requested move. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

Requested move 2

Wikipedia:Handling primary, secondary and tertiary sources (proposed guideline) → Wikipedia:Identifying and using primary, secondary and tertiary sources — Most people are against reverting to the original name and I think "Identifying and using" addresses the concerns of all parties. Yaris678 (talk) 09:29, 4 July 2010 (UTC)

N.B. Apologies for the mess I created before. I should have requested a move on this talk page... but instead I moved the talk page! That has been cleaned up now thanks to a very helpful administrator. Yaris678 (talk) 09:34, 4 July 2010 (UTC)

What's the problem?

I am interested in this subject, having been following it for years. However, I do not find it easy to engage in this discussion. I think the problem is that the problem being addressed is undefined. Are people not "Handling primary, secondary and tertiary sources" properly? What is the take home message (aka nutshell) of the page? --SmokeyJoe (talk) 09:47, 4 July 2010 (UTC)

I guess there are two take-home messages. “You can use all three types of source but there are some things you have to be particularly careful about with primary sources.” and “Categorising as one of the three types is dependent on the context”.

Yaris678 (talk) 10:18, 4 July 2010 (UTC)

OK. Are either of these two things in dispute? --SmokeyJoe (talk) 11:40, 4 July 2010 (UTC)

I don't think so. I think it is more the details where people are disagreeing. That said. There have been some good compromises on some of the details recently. Yaris678 (talk) 12:52, 4 July 2010 (UTC)

Failed proposal? Version 2

I notice that SlimVirgin has marked this proposed guideline as failed. That seems like the obvious conclusion to me. Although we seemed to be getting some good agreements on some things, we then had a completely dysfunctional conversation about the title of the proposed guideline. If we can't even decide on that then there isn't much hope. Yaris678 (talk) 11:17, 19 August 2010 (UTC)

Is a documentary film interviewing scientists considered a secondary source ?

Is a documentary film interviewing scientists considered a secondary source for WP referencing purposes  ?--— ⦿⨦⨀Tumadoireacht ^Talk/_Stalk 15:30, 14 October 2014 (UTC)