Wikipedia talk:No original research/Archive 45

Archive 40

←

Archive 43

Archive 44

Archive 45

Archive 46

Archive 47

→

Archive 50

recent deletion of sentence re using self-publication of discovery as source

Slim Virgin, Could you explain the reason for the deletion of this sentence? Thanks. --Bob K31416 (talk) 16:09, 15 June 2009 (UTC)

We say it needs to be published by a reliable source, and we above say what we mean by "reliable source," so adding anything else seems repetitive. SlimVirgin ^{talk|contribs} 16:11, 15 June 2009 (UTC)

Thanks for your response. In a previous edit summary you moved a sentence back to the previous paragraph saying that it was a "non-sequitor" in the last paragraph. So it appeared from your remark that there was a need to show how it wasn't a non-sequitor. Thus I put in the sentence, "But note the above regarding self-publication of your discovery." By now saying that it is repetitive, you seem to be contradicting your previous edit summary remark that it was a "non-sequitor". Could you clarify? Thanks. --Bob K31416 (talk) 18:42, 15 June 2009 (UTC)

Sorry, I don't follow. SlimVirgin ^{talk|contribs} 18:04, 16 June 2009 (UTC)

By using the term "non-sequitor", you seemed to say that the last paragraph of the section wasn't related to the moved sentence.[1] Is that what you meant? --Bob K31416 (talk) 21:46, 16 June 2009 (UTC)

That's right. It was a non-sequitur (it didn't follow from the preceding sentences), and it was repetitive. We already say material needs to be in an RS and we say what that means. There's no need to add explanations, such as "but note what we say about self-published," or "note what we say about newspapers being acceptable," etc. That was what I meant by repetitive. SlimVirgin ^{talk|contribs} 21:59, 16 June 2009 (UTC)

Here's the paragraph as it currently is.

If you are able to discover something new, Wikipedia is not the place to première such a discovery. Once your discovery has been presented in a reliable source, however, it may be referenced.

In addition to the above, there are two versions that we are discussing.

1) In the 1st version, the statement in question wasn't repetitive (since it was moved from the previous paragraph) but you felt it was a non sequitor.[2] The sentence in question seems to follow the train of thought of the paragraph. First there is the notion that editors' discoveries shouldn't be premiered in Wikipedia. Then it is mentioned that they can be published in an RS and then referenced. Then it is mentioned that self-publication is generally not considered an RS, in case the editor tries to get it into Wikipedia by that route. The 3rd sentence is no more of a non sequitor than the 2nd sentence.

2) In the 2nd version, you felt the problem was that the sentence was repetitive.[3] In this version, it seems that the second sentence is no less repetitive than the 3rd that you objected to. After all, referencing an RS is already mentioned in WP:NOR and many places in Wikipedia. If you still feel that the third sentence of this version is repetitive, perhaps you should delete the 2nd sentence too.

Please note that my preference is for the 1st version and I offered the 2nd version to accomodate you as a compromise.

--Bob K31416 (talk) 23:35, 16 June 2009 (UTC)

It's probably a moot point now, as someone else has also reverted. But for what it's worth, in the first version, the final sentence is a non-sequitur in that position, because no mention had been made in the previous sentence about self-publication, explicitly or implied. The same sentence in its original location was fine:

"If you are able to discover something new, Wikipedia is not the place to première such a discovery. Once your discovery has been presented in a reliable source, however, it may be referenced. But note that self-published material, whether on paper or online, is generally not regarded as reliable, with some exceptions. (See these sections of Verifiability.)"

In the second version, the final sentence is repetitive, because the sentence above about self-publication had been moved back to its original location, so this was just repeating it:

"If you are able to discover something new, Wikipedia is not the place to première such a discovery. Once your discovery has been presented in a reliable source, however, it may be referenced. But note the above regarding self-publication of your discovery."

Experts may indeed be able to self-publish their discoveries. We explain it all in the SPS section. No need to hint at it here, where we can't explain it properly without being unnecessarily repetitive, but where, if we don't explain it properly, it will be misleading. SlimVirgin ^{talk|contribs} 00:37, 17 June 2009 (UTC)

Sorry, but it seems like a repeat of your previous ideas, rather than a response to my message. Specifically, I made comparisons to the second sentence of the current version of the paragraph, and you didn't address those comparisons. --Bob K31416 (talk) 09:32, 17 June 2009 (UTC)

Just for the record, I prefer this section without the editorial note pointing to SPS. The point being made in the section under discussion is that Wikipedia should not be used as the first place of publication for new discoveries or ideas. Pointing to SPS makes that simple point more confusing. Blueboar (talk) 14:13, 17 June 2009 (UTC)

OR and dynamic numbers

I've run across a couple of cases of where for purposes of quantify a size to include a number that is dynamic such as views on YouTube, the number of downloadable songs that can be added to music video games that are introduced weekly, etc. My preference, however, is to only add these numbers when they are sourced through a different reliable source instead of directly using the source - so for youtube counts, articles that talk about the rise to fame of a well-notable video (eg Pork and Beans (song)) are necessary as I feel using the actual count is a bit of OR-ishness. Of course, the problem is when you add these, newer editors will tend to assume they should be freely updated by just actually looking at that number. Often what happens is that I will include a sourced number that says "there have been X views through DATE", with that being sourced to a reference to establish the sense of growth, but then editors will overwrote X and DATE with updated numbers, not recognizing that the reference no longer applies and that the statement no longer serves its original purpose.

Technically, reading OR, this would fall under "routine calculations" and suggest that it's appropriate, but at the same time, I'm concerned that without limiting this, editors will continue to push to update every dynamic number they find. Am I barking up the wrong tree here or could we include something to this effect? -MASEM (t) 13:47, 18 June 2009 (UTC)

I agree that this doesn't fall under OR. It's a routine calculation, and common sense. If there's a separate reason that it's bad for people to "update every dynamic number they find", I haven't understood it from your post. Axlrosen (talk) 17:24, 18 June 2009 (UTC)

Example article is Left 4 Dead 2, under the Boycott section. There's a group that wants to boycott this game, and it made news shortly after the game's release, so inclusion of the group's fast growth seems appropriate. In the days following, there's a handful of data points about the size (reports start at 10k, go up to about 21k) and then there were comments by Valve (the company that made the game) that basically neutralized any significant talk about this group. However, the group has continued to grow so the problem I've seen is people update the group's new size based on looking at the membership count where the group is hosted. That count itself, while true, does not reflect *why* the initial count was so important according to the news articles that reported. That's why I think there's something about OR here; including that number without saying why it is important feels like OR to me, particularly when other sources state why their reported number was that way. --MASEM (t) 00:14, 19 June 2009 (UTC)

The existing policies are already very clear on this issue. We are warned to use primary sources (e.g. raw numbers on a website) with great care, if at all. If the primary information is not self evident to an average person and requires any kind of interpretation or analysis, we may not use the primary information on its own, and require a reliable secondary source to analyze it for us. So if it's an issue of the downloads per specific time period which is important, as opposed to just total downloads, a reliable secondary source must explain that, and interpret the raw data, and we can't just copy down numbers from the website. As bottom line, primary data must be used sparingly and with great care, and if there is any question of interpretation, we must let a secondary source do it for us. Crum375 (talk) 02:55, 19 June 2009 (UTC)

Not having read the article, couldn't both numbers be useful? i.e. the membership was X on date Y (from a secondary source), and is now Z (from the web site). I'm a little wary of calling this OR, I think that would be more strict that necessary. Instead, it sounds like the primary problem is correctly using this information to make a coherent article - the kind of thing that affects every article and is separate from OR. Axlrosen (talk) 12:46, 19 June 2009 (UTC)

Primary information, such as numbers gleaned from a website, must be used with extreme care. If anything is controversial or not obvious to an average reader, we need reliable secondary sources to interpret it for us. Crum375 (talk) 19:33, 19 June 2009 (UTC)

"below"

Slim Virgin, You've introduced the word "below" into the RS section 3 times recently,[4] [5] [6] and you've done so without discussion after it was removed by myself and another editor. Here's the subject sentence.

Self-published material, whether on paper or online, is generally not regarded as reliable, but see the discussion of self-published sources (below) for exceptions.

It doesn't seem to make sense since the link doesn't point to "below" and nothing down there mentions self-published material. Perhaps you are trying to refer to the section Citing oneself which is a few sections below, but that section doesn't mention self-published material.

Could you explain what you are trying to do? Thanks. --Bob K31416 (talk) 14:56, 18 June 2009 (UTC)

I apologize, I made a mistake. I was thinking we were on WP:V, and I couldn't understand why no one wanted to say that the SPS section was "below." :-) I've fixed it now. Sorry for being dense. SlimVirgin ^{talk|contribs} 01:57, 21 June 2009 (UTC)

My I suggest something more along the lines of:

• Self-published material, whether on paper or online, is generally not regarded as reliable, except under limited circumstances (see our policy on Self-published sources). Reliably published material written by an editor may be included as per Citing onself (below).

Perhaps this reflects what everyone is trying to say? Blueboar (talk) 16:00, 18 June 2009 (UTC)

That's fine with me. SlimVirgin ^{talk|contribs} 01:54, 21 June 2009 (UTC)

Bengali romanization

A discussion involving original research issues has been instigated at Talk:Bengali script. Please weigh in if you care. — AjaxSmack 01:51, 21 June 2009 (UTC)

Is describing the results of a googlesearch original research?

Can someone state that a particular image of something is the most popular representation of that object and use as a reference the results of a search on google images that shows more hits for that partiuclar representation than for any others? Or state that a partiuclar spelling for a city is most popular and use as a reference the results of google search showing that this is the way the city is most often spelled? Would this constitute original research or would it be merely a descriptive statement that dos not involve research on the part of the editor? Is there a way to word a summary of the results without venturing into original research territory? Thanks...Faustian (talk) 04:13, 25 June 2009 (UTC)

Google can certainly be used as a source for talk page arguments. It is also patently unsuitable as a source in articles. But there is really no reason to use google as a source in an article; for example, one can write "a common spelling" instead of "the most common spelling" to avoid the issue. — Carl (CBM · talk) 04:20, 25 June 2009 (UTC)

In which case, how would one source "a common spelling" (or would it be unsourced?) For example, with respect to the spelling of Ukraine's capital Kiev/Kyiv [7], could the article include a statement about the fact that the former spelling is more common than the latter, based on the results of google?Faustian (talk) 05:39, 25 June 2009 (UTC)

The article on Kiev is a good example. There's really no need to say "Kiev" is the most commonly used; simply saying it is the traditional name conveys essentially the same information to the reader. On the other hand, it may be that we can agree to say "Kiev" is the most commonly used spelling without a specific source.

Instead of doing these things, the article has an attempt to source the "most common" claim. But the footnote there, "As of 2008, the Oxford English Dictionary includes 19 quotations with 'Kiev' and none with any other spelling. This spelling is also given by Britannica and Columbia Encyclopedia.", is a little desperate. Because the OED and Britannica have a reputation for correctness, I would say this is OK as a compromise, as long as everyone realizes it for what it is. But a google search (the original topic) would be even worse. — Carl (CBM · talk) 12:29, 25 June 2009 (UTC)

"Is describing the results of a googlesearch original research?" Yes, and bound to change at a whim as Google results are extremely fluid. If you want to go this route you'd need a reliable source describing the results, but then I don't know tht it'd be notable anyway. Things can be wrong for multiple reasons because multipl policies are always at work. DreamGuy (talk) 13:37, 25 June 2009 (UTC)

Faustin, a google search should never be referred to in an article, if that's what you mean. It's one research tool among many, sometimes helpful sometimes not, but it's useful only for editors' research purposes, never as a source in an article. SlimVirgin ^{talk|contribs} 15:35, 25 June 2009 (UTC)

Thanks!Faustian (talk) 18:47, 25 June 2009 (UTC)

Citing number of occurences in a Google search may not fail original research, but it does fail WP:V. Search engine results, unless some third party logged them and published them someplace, are ephemeral and somebody else searching even a few minutes later may get a different result. While search engine results are sometimes useful on talk page discussions as a rough idea on how to phrase things in an article, they shouldnt be cited from mainspace. Squidfryerchef (talk) 16:39, 29 June 2009 (UTC)

Video Media Proof

Is visual proof, such as an in-game or in-movie action or scene that be viewed clearly by anyone considered original research? What I'm asking is that if I were to say, "The sky is blue", would I need a source, even though anyone can go outside and see for themselves that the sky is blue? I ask this because of an OR claim on one of my edits. --70.184.239.162 (talk) 08:19, 28 June 2009 (UTC)

The short answer to your question is "maybe", which is why providing an actual link to the edit in question is going to help more than asking a general question. there is truth in the statement that 'common sense is not all that common.' What seems obvious to you may not seem that clear to someone else. For example, one person's blue sky is someone's else's gray sky. Some people see Christian overtones in the death and resurrection of Superman, whilst others see a marketing ploy to boost sales. Almost all of life is relative, which is why Wikipedia is composed of community effort; an agreed-upon reality is far more manageable than an absolutist one. - Arcayne (^{cast a spell}) 00:01, 29 June 2009 (UTC)

If someone challenges an edit, it's always better to provide a source, even when it's something trivially true. If the same person continues to request sources for edits that really don't need them, you could ask an admin for help. As for watching movies, that can be problematic, unless the scene is described in such a way that no one could disagree that it was purely descriptive, but even then you face the problem of, "Why pick out this scene?" For anything at all contentious, it's best to stick to secondary sources. SlimVirgin ^{talk|contribs} 00:11, 29 June 2009 (UTC)

The nearest-run thing you ever saw in your life

see Wikipedia talk:No original research/Archive 44#The nearest-run thing you ever saw in your life

I left this section alone in the hope that it would generate more comment, and now (of course)it has been archived (Sigh!). Of the four who commented three thought the older wording better than the current wording. So do we go back or do we modify the current wording to make it reflect what the old wording said? --PBS (talk) 08:13, 2 July 2009 (UTC)

Can you say what changes you'd like to make? SlimVirgin ^{talk|contribs} 09:10, 2 July 2009 (UTC)

Question for Laurent only

I am getting very confused about who is posting about what, so I'd like to ask this of Laurent, and could he alone please respond?

Laurent, do you object to the addition of your example here, and if so, can you say how (succinctly, please, if possible, because volume often confuses). It says:

---

"A simple example:

"The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world.

"Although no conclusion is drawn and both facts are true, the sentence implies that the UN has failed to maintain world peace. If no reliable source has combined the material in this way, it constitutes original research. It would be easy to imply the opposite using the same material, illustrating how, when no source is provided, facts can easily be manipulated:

"The UN's stated objective is to maintain international peace and security, and since its creation there have been only 160 wars throughout the world."

---

If you object, can you say what you would like to see it changed to, and why? SlimVirgin ^{talk|contribs} 04:32, 4 July 2009 (UTC)

I don't object to the addition of the example. I've initially put it in a different section but I know realize that it's indeed an example of WP:SYNTH. I think the way you've reworded it makes it very clear. Laurent (talk) 19:11, 4 July 2009 (UTC)

Thanks for the reply. SlimVirgin ^{talk|contribs} 19:13, 4 July 2009 (UTC)

OR examples

Is there any reason we don't have Wikipedia:Original research examples, in the same way that we have Wikipedia:Reliable source examples? Surely, there is room for more than one example on such a page? Viriditas (talk) 11:47, 4 July 2009 (UTC)

Another good idea. SlimVirgin ^{talk|contribs} 19:13, 4 July 2009 (UTC)

I'm curious if Bob K31416 would help me with it. Viriditas (talk) 13:16, 5 July 2009 (UTC)

Not sure what you would like from me personally. We could discuss this on my talk page. Regards, --Bob K31416 (talk) 09:01, 6 July 2009 (UTC)

Aren't you interested in introducing examples into the OR policy page? I was thinking we could work on them on a separate page, much like the reliable source examples. If you aren't, that's ok. Viriditas (talk) 09:39, 6 July 2009 (UTC)

"synthesis" - factually erroneous explanation

http://dictionary.reference.com/browse/synthesis : the combining of the constituent elements of separate material or abstract entities into a single or unified entity (opposed to analysis ).

Obviously (and I also saw it mentioned in an explanation of what Wikipedia is about), Wikipedia articles are meant to be a synthesis of existing knowledge. Therefore my sugestion, instead of :

"Carefully summarizing or rephrasing source material without changing its meaning is not synthesis—it is good editing."

Write: "Carefully summarizing or rephrasing source material without changing its meaning does not advance a position —it is good synthesis."

Harald88 (talk) 14:29, 5 July 2009 (UTC)

The term is used differently in the context of Wikipedia, but I agree that we have ongoing problems with defining terms. Even the term "original research" is problematic. Here, we are saying that synthesis means the act of combining multiple sources/arguments to advance a POV or reach a conclusion not found in the original sources or reflected in secondary sources. This is a very serious problem on Wikipedia, and the community as a whole has not been able to adequately deal with it. One way we try to address it is to evaluate the types of sources being used and to verify the material is found in reliable sources. Viriditas (talk) 10:28, 6 July 2009 (UTC)

Should not have an article wording, a potential problem

If no reliable third-party sources can be found on an article topic, Wikipedia should not have an article about it

There's a few potential problems with having this in the NOR policy. While I sympathize with the intent of discouraging new users from creating trivial articles, this isn't the place for it. This is the original research policy, not the general notability guideline, and this belongs at the guideline level not the policy level.

Someone could write an article using only first-party primary sources ( i.e. a corporation's website for an article about one of their products ) without introducing original research. That wouldn't be an acceptable article, but it would be unacceptable for reasons other than OR.

Also this wording is dangerously close to being misread as requiring third-party secondary sources. While the notability guidelines very strongly encourage secondary sources to show notability, there is a longstanding acceptance of a presumed notability for articles about towns and other geographic topics, and these articles often start off relying heavily on third-party primary and tertiary sources such as Census data and maps. Squidfryerchef (talk) 14:22, 21 June 2009 (UTC)

An article that relied entirely on primary sources would amount to OR. OR is not only unsourced material. It is material that no secondary source has seen fit to mention. An article based only on a company's website, assuming no other sources could be found, would fail both the notability and the OR requirement, as these are closely linked. SlimVirgin ^{talk|contribs} 01:15, 23 June 2009 (UTC)

Note that when I say "secondary" in this context, I mean tertiary too. SlimVirgin ^{talk|contribs} 01:17, 23 June 2009 (UTC)

That's not a correct assertion; an article based entirely on primary sources could fully pass WP:OR, if no synthetic claims were made. Many of our biographical articles are based primarily or entirely on primary sources. — Carl (CBM · talk) 04:22, 25 June 2009 (UTC)

I don't think an article based entirely on primary sources would pass OR. Can you give an example? I promise not to interfere with it, but I'd like to see what you're referring to. SlimVirgin ^{talk|contribs} 12:20, 2 July 2009 (UTC)

This is made more difficult by the tendency of some people to redefine primary sources as secondary, merely so they can say that an article uses secondary sources. But, for example, 2009_Wimbledon_Championships only cites primary sources, although I doubt there is any OR there. Brinkman v. Miami University is another example; it has some citation issues but not OR problems. However, you can see that your claim is wrong a different way. It implies that any completely unsourced article violates the OR policy, which would be absurd. — Carl (CBM · talk) 12:41, 2 July 2009 (UTC)

Viriditas' argument is also sound: just because primary sources were used, does not mean secondary sources do not exist, and the existence of appropriate secondary sources is enough to satisfy the OR policy. — Carl (CBM · talk) 12:54, 2 July 2009 (UTC)

But, the point here is that secondary and tertiary sources can easily be found supporting our biographical articles, not that primary sources could be used to write them. Editors also perform original research using primary sources when they cherry pick what they personally think is important for an argument. But, what happens when there are no secondary sources that verify the significance of the material they have chosen? The reason we rely on the secondary sources is so that we have an authoritative (depending on the quality of the source) measure of importance independent from the bias of Wikipedia editors. This is why we should rely on the best sources written by experts in their field. Right now, we have the following statement in the policy: 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. But this avoids directly addressing the problem of cherry picking. It is not just the interpretation and analysis that is OR; It is also the very act of picking and choosing from the primary sources without secondary sources to support a particular selection of material. Some editors will choose unimportant items from primary sources to make a point. For example, an editor might choose a series of quotes from archival tape transcripts to further an argument, and those quotes might not appear in the secondary sources. A secondary source might support one aspect of the overall line of reasoning, but the rest would be an interpretation of the primary source, usually made to further an argument that the editor feels is important. Where does the OR policy specifically address this problem? Obviously, it is implicit in the act of writing an interpretation, forming analyses, and making synthetic claims, but the act of selecting content from a primary source without a secondary source that supports such a selection needs to be explained in more detail. Viriditas (talk) 11:35, 2 July 2009 (UTC)

I was referring to the claim "An article that relied entirely on primary sources would amount to OR". That claim is simply false; your explanation is one reason why.

The issue of cherry picking is relevant to the idea of OR, but if editors pick primary sources in a way that gives appropriate weight to them then there is no problem. Picking and choosing from numerous sources in the process of creating an article is ordinarily considered part of the writing process.

The first paragraph of the policy page explains when this goes wrong: when the analysis presented on is out of line with the analysis already present in the literature. As long as the analysis we present agrees with the analysis in the literature, we have achieved our goal. Just as the editors of each page have to judge due weight for NPOV, they have to judge originality for NOR. There is no way to remove editorial judgment from the writing process. — Carl (CBM · talk) 12:20, 2 July 2009 (UTC)

The original research policy simply means that every statement in a Wikipedia article must have been published someplace else before; Wikipedia should not contain anything that's citable only to Wikipedia. I don't understand how an article based entirely on a company's press releases, with every statement cited, could contain original research. It would be a bad article for a variety of reasons, but not original research. Squidfryerchef (talk) 16:34, 29 June 2009 (UTC)

We would need a secondary source to tell us that the company was worth writing about. Viriditas is right about the danger of relying on primary sources, because people do cherry-pick what to mention and what to leave out— either deliberately or because they're not familiar with the material. The difficulty of writing policy to exclude this kind of reliance on primary sources is that we don't want editors to feel they have no leeway whatsoever when it comes to using primary-source material, so getting the wording right would be tricky. SlimVirgin ^{talk|contribs} 12:20, 2 July 2009 (UTC)

Knowing that the company is worth writing about is an issue for WP:N, not WP:OR. Please avoid bringing other policies into the discussion. People also cherry-pick because they are familiar with the material and are picking only the relevant things to include. There is nothing wrong with that. — Carl (CBM · talk) 12:45, 2 July 2009 (UTC)

There has always been an overlap between notability and OR, so it's not a question of bringing other policies in; indeed, that's why the policy says if no secondary sources exist, we shouldn't have an article. SlimVirgin ^{talk|contribs} 15:06, 2 July 2009 (UTC)

Nevertheless, Squidfrychef is correct that an article which cites only press releases but does so without making any analytic or synthetic claims will not violate the V or OR policies. Brinkman v. Miami University is another example of an article that only cites primary sources but meets the OR policy anyway because it makes only descriptive claims. — Carl (CBM · talk) 15:20, 2 July 2009 (UTC)

Also, this policy does not say what you just claimed. It says, "If no reliable third-party sources can be found on an article topic, Wikipedia should not have an article about it." The word "secondary" is not there. WP:N has the word in its nutshell but not the actual GNG text. — Carl (CBM · talk) 15:25, 2 July 2009 (UTC)

SV, any chance we can discuss the overlap between notability and NOR in the policy? Viriditas (talk) 06:00, 3 July 2009 (UTC)

Sure, go for it. SlimVirgin ^{talk|contribs} 08:07, 3 July 2009 (UTC)

I was thinking, a nice graphic showing the interrelationship between all the policies and guidelines would be more effective than just talking about. Does such a thing exist? Viriditas (talk) 08:30, 3 July 2009 (UTC)

Not that I know of, but it would be a great thing to have. SlimVirgin ^{talk|contribs} 08:50, 3 July 2009 (UTC)

Break 1

Death of Michael Jackson is another high-profile article written entirely from primary sources: newspaper stories about the event published contemporaneously with it. I don't view this as a bad thing; it's just part of Wikipedia that we do in fact permit many articles to be written from primary sources. — Carl (CBM · talk) 15:39, 2 July 2009 (UTC)

Those newspaper reports aren't primary sources for us. They will be in 100 years' time, but for us, now, they're secondary sources. SlimVirgin ^{talk|contribs} 08:07, 3 July 2009 (UTC)

This is the problem I was referring to earlier. Redefining newspaper stories as secondary sources does not actually improve the sourcing of the article. It appears to me that the only reason anyone pursues this redefinition is to avoid admitting that the articles are, in reality, based upon primary sources. — Carl (CBM · talk) 12:13, 3 July 2009 (UTC)

The designation changes depending on its temporal context. Newspaper articles become primary sources when they are historical in nature. Current newspaper articles are considered secondary sources for reporting about contemporary information derived from primary sources. It's very confusing. Viriditas (talk) 12:22, 3 July 2009 (UTC)

That is the viewpoint SlimVirgin is proposing. Others consider the newspaper stories to be primary sources permanently from the moment of publication. None of the "references" in the PSTS section says anything about newspaper articles changing from secondary to primary over time. All but one of them simply categorize newspaper articles as primary sources, the last categorizes them as secondary. — Carl (CBM · talk) 12:40, 3 July 2009 (UTC)

It's not so much her viewpoint. If you look into the matter, you'll find a lot of style guides supporting this statement. If you remember, Wikipedia:Evaluating sources was supposed to explain this in detail, but editors continued to remove it. When I last edited it, it said, "...the meaning of primary sources might change based on the historical context. Thus, secondary sources will often discuss the historical context of the primary source, and such information is often invaluable, and frequently there is a consensus as to the context within the academic field. Such views as to the context are almost always relevant to the discussion...If a secondary source was written by a historical contemporary of the primary source, that source may have significance beyond its role as a secondary source, and may also be primary, and often represents an important point of view as to the original meaning of the author and the relevant historical context." There was also a footnote which said, "Since a source is only a source in a specific historical context, the same source object can be both a primary or secondary source according to what it is used for."[8] This was removed by several editors, and I protested its deletion. And here we are again. Viriditas (talk) 13:09, 3 July 2009 (UTC)

That language was meant to explain that, for example, a history textbook that is a secondary source now might be considered a primary source in 1000 years. It did not explain why contemporary newspaper articles would ever be considered secondary sources, when they have none of the aspects used to define secondary sources. I followed the earlier discussion closely and have read many guides about primary/secondary sources in the course of that. Very few of them seem to accommodate the blanket reassignment of all contemporary newspaper articles as secondary sources. — Carl (CBM · talk) 13:20, 3 July 2009 (UTC)

If you could point me to one of those guides, I'll take a look. Many textbooks are considered tertiary sources since they are composed of secondary sources. Most newspaper articles report about topics using primary sources, so they are considered secondary sources. For example, look at the news citations used to report on Obama's speech, "A New Beginning". Most are secondary sources reporting about the speech, the reaction to the speech, highlighting key parts of the speech from transcripts, audio, and video recordings, etc. Viriditas (talk) 13:28, 3 July 2009 (UTC)

Some examples are [9], [10], [11], [12]. The sources used in the Obama article can just as well be classified as primary sources: newspaper articles written contemporaneously with the event they discuss. Now if someone writes about the speech in a book next year about presidential speeches, that will be an unambiguously secondary source for us to use. In the meantime, claiming that our article is based on secondary sources is misleading. — Carl (CBM · talk) 13:54, 3 July 2009 (UTC)

Those links confirm what I have said above. Take a closer look at them and you will notice that newspaper articles are classified both as primary and secondary sources. The newspaper cites in the Obama article are (for the most part, obviously there will be exceptions) considered secondary sources. If you read the links you provided closely, you will see that they confirm this. I admit it is confusing because there is a lot of ambiguity. For example, the Yale link you provide lists newspaper articles as primary sources that are "from the original time period (contemporary to events) and have not been interpreted or evaluated by others". But they are not talking about contemporary newspaper articles, but rather historical newspaper articles. This becomes clear further down when you see that "newspaper accounts" are listed as a primary source under the discipline of history. The BMCC library link lists "newspaper articles that interpret" as a secondary source under the discipline of humanities. Most of the newspaper sources in the Obama article are analyzing and interpreting the importance of the speech. For our purposes now, newspaper articles about the speech and the reaction to the speech, and the interpretation of the speech, and the significance of the speech, are considered secondary sources. On Wikipedia, the primary source is considered the speech itself. If a robot decides to write about Obama's speech in 2109, and chooses to use newspaper articles about the event from 2009, those newspaper articles are now considered primary. Viriditas (talk) 14:29, 3 July 2009 (UTC)

I am aware of the interpretation you are giving but I completely disagree with it. The broad area of study in which one writes about the speech is history, so the standards within the field of history are what we should be employing. The newspaper articles cited in the Obama article are simply reporting on the speech; they are not academic historians interpreting its importance. It would be impossible, in fact, to evaluate the importance of the speech without the benefit of hindsight. I have seen no guide that has a statement such as, "newspaper articles written at the same time as an event are primary sources, unless the event was very recent, in which case they are secondary". In fact I have only encountered that argument on WT:NPOV. However, I do see "A primary source is a source that was created during or immediately after the event or period it documents",[13] and "...a letter written by a soldier during the Vietnam War is primary material, as is an article written in the newspaper at the time of the Vietnam War." [14]. These are the sorts of examples you asked for, and they are not as vague as your latest post implies. — Carl (CBM · talk) 14:43, 3 July 2009 (UTC)

But Obama's speech is considered contemporary, not historical. The newspaper articles only become primary sources when they are connected to the event by future historians studying the past. For our purposes now, they are considered secondary sources and the speech itself is considered primary. On the other hand, historians writing about the Gettysburg Address will use newspaper articles about the speech from 1863. Those articles, which were considered secondary at the time of Abraham Lincoln's speech, are, in 2009, classified as primary. Viriditas (talk) 14:49, 3 July 2009 (UTC)

As far as I can see, the articles on the Gettysburg address articles were primary sources when written as well. None of the links I provided (at your request) explicitly treats contemporary newspaper articles differently than historical ones (which is not to say that contemporary events are not also in the scope of history). The links and quotes I showed you suggest that what matters is the close connection in time between the source and the event it is describing, not the distance in time between us and the event. — Carl (CBM · talk) 14:58, 3 July 2009 (UTC)

In other words, I am simply pointing out that if there has not been enough time elapsed after an event for secondary sources to be written, this does not mean we must treat the primary sources created along with the event as if they are secondary. It is true that we do not have as many sources for Obama's speech as for the Vietnam war, but that doesn't mean we have to pretend that the sources we do have for Obama's speech are something they are not. In articles about contemporary event we will necessarily be forced to use primary sources to write our articles. We can add secondary sources later. — Carl (CBM · talk) 15:10, 3 July 2009 (UTC)

A newspaper article from 1863 about a speech in 1863 is used as a primary source by historians because it is a now an artifact from 1863. This old article is no longer considered a secondary source because it gives us a first-hand account of a past era, not just the event of Lincoln's speech. And as you rightly observed above, these newspaper articles from 1863 have become important primary sources. But, a newspaper article about a speech in 2009 is considered a secondary source because we use it only to evaluate the primary source, the speech itself. We are not interested in the historical importance of a newspaper article written in 2009. We do not consider the reporters to be active participants in the speech or representative of a larger historical era, nor are we interested in their tone or perspective. Rather, these journalists and writers are simply reporting on the event and that's all we care about. We aren't interested in analyzing a BBC News article as a historical artifact, although one could do that, in which case it would be considered a primary source. But obviously, for our purposes, it isn't, so it's secondary. We don't consider the BBC News or The New York Times as historically important in this context. They are merely reporting about the speech, and offering us interpretations. And, we aren't using the 2009 news articles as original documents related to the event. For our purposes, now, the original documents are transcripts of the speech. Viriditas (talk) 15:19, 3 July 2009 (UTC)

Break 2

← Our articles about events in the recent past are indeed history articles, not somehow completely different than articles about the ancient past. The events they describe do take place in a larger historical context (2009), and we must keep in mind that no contemporary source can possibly have sufficient separation to give an objective analysis or interpretation. This is the fundamental reason for classifying sources as primary in the first place: they are the sources close enough to the event to capture its zeitgeist and be unaffected by later analysis or by changes in societal and academic opinion.

Independent of that, the argument you just gave also implies that I can use a newspaper article on the Gettysburg address as a "secondary source" as long as I don't focus on the newspaper article itself as an historical artifact (which is essentially never done on WP). I continue to point out that none of the links above says anything about contemporary newspaper articles being treated differently than older ones. I know that one can interpret things that way, but I have never seen it in print. I think it is a wikipediaism. — Carl (CBM · talk) 16:28, 3 July 2009 (UTC)

That's interesting that you think that. I've seen it in many books, particularly on Google Books, and was aware of it before I came to Wikipedia in 2004. I can provide links to the material, but I thought you already knew about it, which is why I didn't. Viriditas (talk) 11:32, 4 July 2009 (UTC)

Please do, I would be interested to see what they say about it. One thing that is clear to everyone is that there are many different variation on "primary" and "secondary" sources. The thing that I am pointing out here is that many readers (and sources) will simply treat newspaper stories as primary. The sort of source that you would want to find for me would say that an old descriptive newspaper story on a old event counts as a primary source but a new descriptive news story on a new event counts as a secondary source. Everyone agrees that new stories on events from long ago will be secondary sources. — Carl (CBM · talk) 12:53, 4 July 2009 (UTC)

This problem arises in one other way. In order to read this policy the way that some do, it is necessary to redefine both newspaper stories and academic journal articles as "secondary sources". This is unfortunate, opaque to newcomers, and confusing in general. It would be clearer if we avoided this redefinition, but there is some fetish towards calling things "secondary" that leads to wording in the policy that requires it. — Carl (CBM · talk) 12:17, 3 July 2009 (UTC)

I don't think we should avoid it; I think we should explain it. Viriditas (talk) 12:23, 3 July 2009 (UTC)

We could say that wikipedia articles should primarily be sourced to reliable books, magazines, journals, newspapers, and websites, although some references may be to other reliable published sources. — Carl (CBM · talk) 12:48, 3 July 2009 (UTC)

Break 3

Carl, do you have a source that says contemporary newspaper articles are primary sources, even if written by a journalist not involved in the event? I haven't seen anyone describe them that way. SlimVirgin ^{talk|contribs} 05:04, 4 July 2009 (UTC)

Yes, I provided several higher up, such as: "A primary source is a source that was created during or immediately after the event or period it documents",[15] and "...a letter written by a soldier during the Vietnam War is primary material, as is an article written in the newspaper at the time of the Vietnam War." [16]. It is not uncommon to see all newspaper articles classified as primary sources, with no exception for contemporary ones.

Moreover, as a thought experiment, consider two hypothetical articles:

• An article on the death of Abraham Lincoln, cited exclusively to newspaper articles from the month after his death
• An article on the death of Michael Jackson, cited exclusively to newspaper articles from the month after his death

Imagine these are both written in the normal WP way that does not analyze the sources themselves, but only cites them for information about the death. It is quite hard to see how the sources in these articles, which are used in exactly the same way, could be classified differently. — Carl (CBM · talk) 05:26, 4 July 2009 (UTC)

I think your source is just not right (I don't know what the Internet Public Library is), and is looking at the issue from a purely historical perspective. But historians aren't the only people to use primary/secondary sources. When a journalist works on a story, for example, the last thing they want to do is rely on other newspapers, yet according to your definition, other newspapers would be their primary sources. That definition also confuses "contemporary" and "contemporaneous."

Contemporary newspaper articles are secondary sources, unless written by someone directly involved in an event, in which case they are primary. SlimVirgin ^{talk|contribs} 09:21, 6 July 2009 (UTC)

The sources aren't used in the same way, which is the point. Newspaper articles about Lincoln's death are now considered historically significant and are used as a primary source by historians. Historians use these articles to form opinions about Lincoln's death. On the other hand, newspaper articles about Michael Jackson's death are now considered secondary sources. We would not treat these articles as primary sources since we are only using them for their opinions about Jackson's death. Primary sources in this case, would be Jackson's autopsy report, death certificate, interviews and speeches, music lyrics, etc. The definitions of what is a primary, secondary, and tertiary source depend not just on the field of research but on how we use them. Some of the articles about Jackson's death might actually be tertiary sources. We are not studying the words or arguments in the newspaper articles about Jackson's death; we are only interested in the opinion of the journalist reporting on the event, including the interpretation and analysis of the event. So, we are using the Jackson sources as secondary sources. The Lincoln sources, on the other hand, are used precisely for their content, and for the type of coverage provided by each source, and for comparing the different versions and reports of Lincoln's death. In this way, we are using the newspaper articles about Lincoln's death as primary sources. None of the articles about Jackson are being used in this way, but we could if that was our focus. So, if we were to write about the different news coverage of Jackson's death, and analyze the reports and compare the coverage, then in that case, those articles would be used as primary sources. Viriditas (talk) 11:32, 4 July 2009 (UTC)

In my thought experiment, the sources are used in the same way, to create wikipedia articles. In other words, if the sources in the article on the death of Michael Jackson qualify as "secondary", then an article about Abraham Lincoln's death written in exactly the same way from newspaper stories would also be written "from secondary sources". I am thinking of wikipedia articles and nothing else. As you say, we would not be studying the words or arguments in the newspaper articles about Lincoln's death; we would only be interested in the opinion of the journalist reporting on the event. If doing this for Jackson counts as using a secondary source then doing it for Linconln would as well. — Carl (CBM · talk) 12:53, 4 July 2009 (UTC)

The crucial difference and missing piece of the puzzle is this: In the example above, historians are writing (and publishing) opinions about the death of Lincoln using 19th century newspaper articles (once removed from the subject) as primary sources. We can't do that on Wikipedia. Wikipedia editors cite opinions about the death of Michael Jackson, in this case, journalists, reporters, writers, and authors writing about the event. Since we are citing the opinions and interpretations of writers twice removed from the subject, these sources are, for our purposes, classified as secondary. Viriditas (talk) 13:44, 4 July 2009 (UTC)

There are no historians in my example, just two wikipedia articles written by wikipedia editors. So I think you are arguing that newspaper articles about Lincoln's death published just after it are secondary sources for us as long as we only use them to cite things about Lincoln's death? I am not talking about historians here, just wikipedia editors. — Carl (CBM · talk) 23:37, 4 July 2009 (UTC)

Contemporaneous newspaper articles about Lincoln's death are primary sources for us (and for historians and anyone else now alive). SlimVirgin ^{talk|contribs} 09:24, 6 July 2009 (UTC)

Yes, I agree. But this means that contemporaneous newspaper articles about Michael Jackson's death are also primary sources for us. That's the point I'm bringing up here – the mere lack of passage of time does not transform otherwise primary sources into secondary ones. — Carl (CBM · talk) 12:33, 6 July 2009 (UTC)

Yes, it does. The passage of time changes secondary sources into primary ones. See below. SlimVirgin ^{talk|contribs} 14:36, 6 July 2009 (UTC)

I can address this directly. Please pick one Jackson source you consider primary. Viriditas (talk) 12:57, 6 July 2009 (UTC)

I don't doubt you can make an argument why some source is secondary. My point is that that same argument would apply to a parallel newspaper story on Abraham Lincoln's death. So here is an AP wire article [17]. When you argue it is a secondary source, please explain why your argument would not also imply that a newspaper article about Lincoln's death used by us in exactly the same way is a secondary source. — Carl (CBM · talk) 14:16, 6 July 2009 (UTC)

A newspaper article about Lincoln's death is a primary source because Lincoln's death and the newspaper article were (a) written a long time ago, and (b) were written at the same time. The newspaper report is contemporaneous with Lincoln's death. It therefore gives us a lot of insight into that period, and what people were saying and doing and thinking at that time. It is very close to Lincoln's death relative to us.

A newspaper article about Jackson's death is a secondary source for us, because we stand at the same level as the newspaper, if you like. It is contemporary, happening right now, and I can give myself as much information about what it's like to be alive now, as the newspaper can. The newspaper gives me no special insights that I can't get from a thousand other places. It is not very close to Jackson's death relative to us. A primary source about Jackson's death would be the autopsy report, for example, which is very much closer than a newspaper article.

Closeness is what counts when judging whether a source is a primary one. And closeness is always going to be relative. SlimVirgin ^{talk|contribs} 14:35, 6 July 2009 (UTC)

I keep pointing out that "happened long ago" is not a standard for sorting primary/secondary sources that I have never encountered in print, and you have not provided any sources that back it up. It appears to be something invented on wikipedia.

That leaves a story about Michael Jackson's death equally close to Jackson as a story on Lincoln's death is close to Lincoln. The fact that we have no better information than newspapers for Jackson's death, and the fact that we have no better perspective than the newspapers, doesn't magically transform those newspaper sources into secondary sources. It also does not mean that the newspaper stories on Jackson's death are not close to his death; that's an argument without much strength. — Carl (CBM · talk) 19:59, 6 July 2009 (UTC)

CBM, none of what we are talking about was invented for Wikipedia and is commonly found in research and style guides. We can evaluate a source as primary depending on a number of factors, including how we evaluate its proximity to the topic as well as how we use it. In the case of Jackson, most newspaper accounts aren't yet classified as historical documents. But let's take a look at what you offered. You gave me a link to an Associated Press article.[18] But, I don't see this source used in Death of Michael Jackson. When I said, "please pick one Jackson source you consider primary," I assumed you understood that the source should be picked from the article we are talking about. Usage is one aspect of evaluating a source. So you are asking if the article by Adams is considered a primary source? Well, first of all, the AP article in question was written by ten different writers. The reporter, Lisa J. Adams, relied on other people to write the report. These other people collected evidence and Adams used their information to file her report. Because of the mix of sources and interpretation and analysis ("Whatever led to Jackson's death, his passing left a deep impression on fans and fellow singers worldwide") it is difficult to give you an accurate answer, however, we do not see any evidence that Adams participated in or observed the event. Danto (2008, p. 82) discusses how we should evaluate a source like this:

...Bormann (1969, p. 173) has drawn up a useful distinction: "When a reporter writes of what he or she has personally observed, be it printed or broadcast, he or she becomes a primary source." But when a reporter draws on material from participants or "informed observers or sources close to the government and 'undisclosed' sources," then the report is a secondary source.[19]

Adams has used multiple writers from around the world to compile her article (Istra Pacheco, E. Eduardo Castillo and Jose Osorio in Mexico City; Frank Bajak in Bogota; Deborah Seward in Paris; Gregory Katz in London; Yuri Kageyama in Tokyo; Kim Yong-ho in Seoul; and Teresa Cerojano in Manila). In her article, Adams draws heavily on participants like security consultants, marketing managers, law professors, heads of state, and entertainers. Adams even surveys the media and reports on what they are saying about Jackson. We are not really looking at a primary newspaper account here, as you are arguing, but rather a collection of second-hand observations made by other people, including interpretation and analysis. Adams is literally interpreting the event through the use of other sources, and none of the information is particularly unique. There is no evidence Adams is directly reporting on the event. So, with the information we have at hand, this article is a secondary source, although it is only accurate to evaluate this source in the context of its usage. Since you have not provided evidence of a source being used, my comment cannot fully address the issue. Viriditas (talk) 22:45, 6 July 2009 (UTC)

That the terms primary and secondary are relative, and that they switch according to time and context, is one of the most basic issues in historiographical research. Nothing is being invented for Wikipedia. SlimVirgin ^{talk|contribs} 22:51, 6 July 2009 (UTC)

Break 4

← Viriditas: the source was used in this version of the article [20]; it was removed by SlimVirgin with the vague-if-not-inaccurate edit summary "+link", so you probably missed the removal [21]. Since I don't edit that article I cannot promise the sources I mention are not removed. Pick any AP wire story you like.

If that's today you're talking about, I made two edits and only one ended up in the history—a server quirk. Why did you pick on that source? SlimVirgin ^{talk|contribs} 00:45, 7 July 2009 (UTC)

I picked it yesterday as a random AP wire story. — Carl (CBM · talk) 02:17, 7 July 2009 (UTC)

I picked an AP wire story because these are the epitome of direct contemporaneous news reporting on an event. Like I said, the goal you have to meet is not arguing that this is a secondary source, since the sources vary widely one can argue essentially anything is a secondary source. Instead you have to argue that in a way that does not also make newspaper wire stories about the death of Lincoln (or, say, the death of JFK) into secondary sources. These will be, for all intents and purposes, equivalent to the wire stories about Jackson in they way they are written from reporters' notes as you describe. In particular the quote you gave above would imply equally well that articles on Lincoln are secondary sources.

Moreover, the death of Jackson is an historical article in the same way the death of Lincoln or JFK is an historic event. So we must analyze the sources in the same context of historiography. It's almost solipsism to claim that contemporary documents are not historical documents because they seem more familiar to us than documents from ancient Persia. — Carl (CBM · talk) 00:26, 7 July 2009 (UTC)

CBM, the AP wire story was used to support this statement: ""Vigils were held by members of the general public after the death."[22] The journalist was not directly involved in this report but was summarizing the work of others. Of course, it could be used as a primary source in an article about the death of Michael Jackson if the journalist was reporting directly from a vigil, because the reporter would take on the role of an eyewitness reporting original observations from the scene. Additionally, a reference can be used both as a primary and secondary source. For example, if we were comparing the media portrayal of Jackson vigils, the report by Adams would be considered a primary source. Viriditas (talk) 05:00, 7 July 2009 (UTC)

So if I were to find a newspaper article from the time of Lincoln's death and use it to source the fact that vigils were held for Lincoln, that would make the 100-year old newspaper story a secondary source in my wikipedia article? That seems like a tough sell. — Carl (CBM · talk) 05:12, 7 July 2009 (UTC)

That newspaper article is now a historical document associated with the event on a first-hand basis. In 2009, we are no longer using it as a secondary source. Viriditas (talk) 05:21, 7 July 2009 (UTC)

We can't have it both ways: if we are using both the Lincoln article and the Jackson article in identical ways, and they are essentially identical sources, we can't claim one of them is a primary use and one is a secondary use. — Carl (CBM · talk) 05:28, 7 July 2009 (UTC)

Ok, we have identified your disconnect. Actually, yes, we can claim one is primary and the other is secondary. Can you tell me why you don't think we can? Here, I'll let you prove your own point. Show me how you would use the historical document from Lincoln in your above example as a secondary source. Viriditas (talk) 05:32, 7 July 2009 (UTC)

I am using it exactly like the source in the Jackson article, to source the bare fact that there were vigils after Lincoln's death. The text in the two WP articles is identical. So if you claim that usage is secondary in the Jackson article, and the usage is identical in the Lincoln article, then the usage in the Lincoln article is also secondary. — Carl (CBM · talk) 05:45, 7 July 2009 (UTC)

CBM, please provide me a real example. Your hypothetical Lincoln source is considered primary, not secondary, and there is no direct relationship between the usage of a historical and a contemporary news account. An Associated Press summary of Jackson vigils is not a historical document. In 100 years, it will be. Viriditas (talk) 05:49, 7 July 2009 (UTC)

I keep asking you for sources for that claim. The Jackson article seems to me to be equally as much an historical document, just a much more recent one. Of course there is no real example for Lincoln, because people don't write WP articles about Lincoln cited to newspapers. See my refresh section below for a different direction that I think could make continuing this discussion more productive. — Carl (CBM · talk) 05:53, 7 July 2009 (UTC)

WP:EVALUATE used to discuss this with sources, one of which was Stebbins, Leslie Foster (2006), Student Guide to Research in the Digital Age, Libraries Unlimited, pp. 61-79, ISBN 1591580994.[23] Viriditas (talk) 06:00, 7 July 2009 (UTC)

The sentence that source is attached to has the usual proviso that a source can change from secondary to primary if the way it is used changes. This implies that if the way the source is used does not change, then the usage that starts out secondary will remain secondary indefinitely. I am looking for a ref for the claim that the same usage changes from secondary to primary only because of the age of the source, not because the way the source is used changes. — Carl (CBM · talk) 06:15, 7 July 2009 (UTC)

You got the source you requested ( see p.62). The implication you raise is not in play, and nobody ever claimed the the usage changes only because of the age of the source. They said (as the source does) that contemporary newspaper articles aren't usually considered primary sources. The evaluation of a source does not depend on only one criteria and whether a source is primary or secondary depends on many factors, such as the discipline, its distance from the event (firsthand, secondhand), and its age. Viriditas (talk) 06:38, 7 July 2009 (UTC)

← Thanks for the source. I see it is somewhat self-contradictory, because although the author does say contemporary newspaper stories are not generally primary, she says on the same page that newspaper articles published during the time of an event are primary sources. Looking through pp. 64-65, one interpretation that makes sense is that she only means that contemporary articles on past events are not primary sources. To which I think everyone agrees. Or perhaps she is referring to the practice of historians not to study contemporary events, a luxury we do not have on WP. — Carl (CBM · talk) 07:08, 7 July 2009 (UTC)

It's not self-contradictory. The context is of a historian studying past events and using firsthand accounts - newspaper articles - that are published at the time of the event. It's not a reference to contemporary secondary sources. I think we have covered this several times. If we are writing about contemporary events, chances are we are using secondary newspaper articles, not primary ones. Of course, it is possible to find primary news articles and to use them, carefully. I think the reason you don't see this as often is because the art of investigative reporting is rarely practiced these days. Viriditas (talk) 07:18, 7 July 2009 (UTC)

The newspaper articles about Lincoln's death were secondary sources back then too. And our secondary sources will be someone else's primary sources in 100 years' time. SlimVirgin ^{talk|contribs} 00:47, 7 July 2009 (UTC)

So if the current version of our article on Michael Jackson's death were to sit in a time capsule unchanged for 100 years, when it emerges it now is written from primary sources instead of secondary sources? Even though the sources and the way they are used have not changed in any way? That's a hard position to maintain. If the articles on Jackson's death would count as primary in 100 years if we use them in a certain way, they must be primary now if we use them in that same way. — Carl (CBM · talk) 02:17, 7 July 2009 (UTC)

I'm not sure where the disconnect is, but let's try to find it. For our purposes, the Adams' report is a secondary source about the public and media reaction to the death of Michael Jackson. It has been used in an article about Michael Jackson's death to support the claim that vigils were held after his death. Now, fast forward 200 years. Future historians might want to understand funerary practices of the 21st century. And they might start by studying the funerary practices of high-profile figures, such as entertainers. In this case, Adams' report becomes a useful primary source for providing data for these researchers. Viriditas (talk) 05:17, 7 July 2009 (UTC)

The statement that vigils were held is not an interpretation or analysis of public and media reaction. It's just an observation of current events. That's what wire stories are for: they are not deep analyses, just news reporting. Please remember I am not talking about historians, I am only talking about wikipedia editors.

So as I said above, it seems to me you are arguing that if I find a newspaper story about the vigils after Abraham's Lincoln's death, and use it in a WP article to source the fact that there were vigils, that 100-year old newspaper story is claimed to be a secondary source in my WP article? It would be an essentially identical source used in a way essentially identical to the way the story was used in the Jackson article. But that's crazy; we know that that sort of story is a primary source. So the one about Jackson is as well. — Carl (CBM · talk) 05:28, 7 July 2009 (UTC)

First of all, secondary sources are often composed of primary sources. The statement that vigils were held is not an eyewitness observation from the reporter, but rather a summary of collected reports after the event. There is actually no firsthand news reported in the AP article, unlike an actual primary news report from a reporter in the field. This is a secondary source about the vigils, and meets Bormann's definition as a secondary news story. Granted, the distinctions aren't that clear, but this is why it depends on how the source is used. Viriditas (talk) 05:36, 7 July 2009 (UTC)

It doesn't matter that the reporter is not an eyewitness participant. As I pointed out, Borman's definition also would make 100-year-old newspaper stories about Lincoln's death into secondary sources, unless the reporter actually saw the death. That would be very strange. But let me try to start over in the next section. — Carl (CBM · talk) 05:42, 7 July 2009 (UTC)

It most certainly matters that the reporter is connected to the event firsthand, as that is one criteria for a primary source. The AP article is reporting about the vigils using other sources; it is not an eyewitness account. The AP article is also interpreting and analyzing the event. You are claiming that the AP article is being used for its raw data, namely that "vigils occurred". That may be true. But when one looks at the source, we don't see any raw data, but rather a summary of secondhand reports that quotes other sources who have interpreted and analyzed the event. Viriditas (talk) 05:54, 7 July 2009 (UTC)

The same would be true if I cited a news story about Lincoln's death written by a reporter who didn't witness the death firsthand? In any case, I know what you are saying, and I would suspect you know what I am saying (you can see it in the sources I provided a while back), so let's stop this subthread. — Carl (CBM · talk) 05:57, 7 July 2009 (UTC)

refresh

Let me refresh the discussion, since I think I can state my point more clearly now. Imagine that I come across a Wikipedia article about George Washington that is written entirely from newspapers. Some industrious editors have found newspaper stories to cite Washington's birth, the important events of his life, and his death. Everything is well cited, but only to newspaper stories published contemporaneously with Washington's life and death.

How would a random editor evaluate that article at peer review? I'm sure she would say, "the article isn't awful, but it uses only primary sources. Why don't you use some secondary sources as well? Find a biography or an academic history book to cite." And she would be correct.

The issue I am pointing out is that articles like that do exist: many of our BLP articles are written entirely from primary sources such as newspaper stories, rather than from biographies and academic texts. These articles suffer from all the problems one imagines the George Washington article would suffer. The practice of renaming contemporary primary sources into "secondary sources" only serves to hide this problem, which is that for many articles we write there simply are no secondary sources available yet. — Carl (CBM · talk) 05:42, 7 July 2009 (UTC)

But, it's not a problem, it is standard practice. Look, instead of appealing to hypothetical examples, why not find a Wikipedia article you believe supports your point? That would be helpful. Viriditas (talk) 06:02, 7 July 2009 (UTC)

I pointed out the article on the death of Michael Jackson. There is the article on the recent Honduras coup. There are really so many examples that if you cannot find them, I have not explained my point clearly enough.

I agree it is standard practice to use only newspapers and similar for these articles, because there are no other sources available. The problem is not that practice, it's the renaming of these sources into "secondary sources" to mask the lack of better sources. An article on the 2009_Honduran_political_crisis written from newspaper articles (as ours is) is no better than an article on the Falklands War written entirely from newspapers would be. But the latter has a bibliography full of books. Where are the books in the article on the Honduras coup?

When we say that Wikipedia articles should be based on secondary sources, we are trying to say articles should be more like the one on the Falklands war. But even articles that are based on primary sources, such as the one on the Honduras coup, can still satisfy the NOR policy; that was what I was saying at the very top of this section. — Carl (CBM · talk) 06:11, 7 July 2009 (UTC)

Ok, here's an easy one: What featured article is based solely or for the most part only on primary sources? Viriditas (talk) 06:39, 7 July 2009 (UTC)

2009 Orange Bowl is on FAC and looks like it will be promoted. 2008_Monaco_Grand_Prix and Subtropical Storm Andrea (2007) were promoted recently. In each case I looked through the refs and it appears that the vast majority of them are primary sources, if not all.

But NOR is about all articles, not just FAs; the FA people usually have higher requirements than policy does for sourcing. The original claim that I responded to above is that any article based on primary sources would fail NOR, not that it would fail FAC. So focusing too much on FAs avoids the original claim. — Carl (CBM · talk) 06:58, 7 July 2009 (UTC)

The only article that even qualifies for discussion here is Subtropical Storm Andrea (2007), which relies heavily on primary sources. I'm glad you brought this up as it raises an important point that is avoided by our policies and guidelines, namely, that each discipline will use primary and secondary sources in a different way. The tropical cyclones WikiProject appears to rely on primary sources for their articles more than other projects. This is not a problem unless the topic is controversial, challenged or disputed. For the most part, the primary sources are used to support unambiguous hard data collected by official organizations. If you look carefully, the interpretive data is supported by secondary sources. Looking at the FAC, it looks like it passed without much discussion. This says more about the FAC process that it does about the use of references. I'm only seeing 19 kB of prose, and that's fairly small for a FA. Viriditas (talk) 07:12, 7 July 2009 (UTC)

You don't have to sell me on the shortcomings of FA. On the other hand, the statement I was originally responding to had no provision for variation between subjects: "An article that relied entirely on primary sources would amount to OR." — Carl (CBM · talk) 07:17, 7 July 2009 (UTC)

An article that relies entirely on primary sources would amount to OR. That's true. As you can see, there are plenty of secondary sources in Subtropical Storm Andrea (2007). News indexes show hundreds more. So, no problem here. The problem we are concerned about is, for example, someone getting a hold of Michael Jackson's police reports, autopsy reports, and other primary sources, and cherry picking elements out of those reports into a Wikipedia article and promoting a POV based on those individual elements with no secondary sources to support it. And if you think this is far-fetched think again. I can point to an article that a Wikipedia researcher rewrote with an emphasis on primary sources, and for all intents and purposes, it looks like they did this to push a POV. In other words, they cherry picked data from primary sources that is not given significant coverage in secondary sources. That is the kind of thing we are worried about. Viriditas (talk) 07:27, 7 July 2009 (UTC)

Please remember I do not accept your analysis that news stories magically count as secondary sources. From this stricter definition of primary sources, the three FAs I listed are virtually exclusively sourced to primary sources. I don't see any secondary sources at all in Subtropical_Storm_Andrea_(2007). If you don't accept this analysis, imagine that 20 years have passed, so that the newspaper articles have become primary sources in your criteria as well. — Carl (CBM · talk) 07:32, 7 July 2009 (UTC)

What counts as a primary and a secondary source is not magical; Some news stories are not secondary, and I've explained how news stories can be used as primary sources (Bormann, Stebbins, etc.). Viriditas (talk) 11:04, 7 July 2009 (UTC)

I have been arguing that most news stories are not secondary, so we seem to agree now. The argument I have to refuse is that all contemporary news stories are magically secondary only because not enough time has passed for them to become primary sources. — Carl (CBM · talk) 15:40, 7 July 2009 (UTC)

I didn't say most news stories are not secondary, I said some news stories are not secondary. Nobody said all contemporary news stories are secondary. Viriditas (talk) 21:16, 7 July 2009 (UTC)

(unindent) CBM wrote: "So if the current version of our article on Michael Jackson's death were to sit in a time capsule unchanged for 100 years, when it emerges it now is written from primary sources instead of secondary sources?"

Yes, that's correct, because the notion of primary and secondary is always relative to the observer, if you like. Think Einstein, not Newton. SlimVirgin ^{talk|contribs} 10:31, 7 July 2009 (UTC)

When deciding whether a source is primary or secondary, we ask ourselves, "How close is it?" By that we mean how close is it to the event, or person, or idea we're writing about. Let's suppose we are writing about the Queen's view of Diana, and our source is The Times of London. We say, "Oh, not very close. It's just a newspaper with no direct knowledge, a secondary source good for analysis and overview, that's all."

But the second question we must ask ourselves is, "How close is it from whose perspective?" From our perspective, The Times is not at all close to the Queen. But in a thousand years' time, The Times of July 2009 has become magically, thrillingly close to the Queen , closer than anyone living in a thousand years time could ever be. It existed in the same society, was in contact with people who actually knew the Queen. It's a miracle to have found it! It would be a cherished primary source.

When looking at the primary/secondary issue, always ask those two questions when studying X: (1) "How close is the source to X?" and "How close is it from whose perspective?" SlimVirgin ^{talk|contribs} 10:50, 7 July 2009 (UTC)

That's simply not correct. If an article uses source X as a secondary source, and the article is never changed, then the article continues to use source X as a secondary source forever. The idea behind source typing that that it is a static thing, depending ony on the source used and the method in which the source is used (and similar objective factors). It appears your argument implies that every document written more than 100 years ago is written entirely from primary sources, since the things the authors thought were secondary have become primary in the meantime.

You are right that an article from The Times will be regarded as a primary source in 1000 years. We must similarly treat it as a primary source right now. To claim that our present viewpoint is as privileged as you claim is nearly solipsism. — Carl (CBM · talk) 15:40, 7 July 2009 (UTC)

What you're arguing—that sourcing is a static issue—goes against everything that people are taught when studying history, for example. SlimVirgin ^{talk|contribs} 17:23, 7 July 2009 (UTC)

Sources do change their status over time... what was once a Secondary source can become a Primary source. For example, a Roman historian may have written a history using at Primary documents that existed when he wrote... this history was a Secondary source at that time. However, the documents he used might have been lost in the centuries since the history was written. To us today, the history is a Primary source. Now think to the future... a biography of George III is a Secondary source to us today, but at some point in the distant future the letters and records of George III's reign that were used to write this biography may be lost... the biography might become a Primary source for the information it conveys.

Sources also can change their status depending on how we use them. Gibbon's "Rise and Fall of the Roman Empire" is a Secondary source (allbeit an outdated Secondary source) for the history of the Roman Empire... Today, some two hunderd years after it was written, it can also be seen as a Primary source for an analysis of how 18th century historians saw and wrote about Roman History. It remains Secondary from a historical perspective, but it is Primary from a historiographical perspective. Blueboar (talk) 17:45, 7 July 2009 (UTC)

Proposed slight amendment to the "synthesis" guidelines

I propose that the synthesis guidelines be revised slightly to include the following:

In cases where a given fact is reasonably well established by authoritative sources, and an additional source supplies clearly erroneous information to the contrary, it is acceptable to acknowledge the error on the part of said source as such in an article. For example, it is well established that Michael Jackson was born on August 29, 1958. Therefore, if X magazine states that Michael Jackson was born on August 29, 1959, it is permissible for editors to note "X Magazine erroneously reported that Michael Jackson was born on August 29, 1959." A third source explicitly stating X magazine is wrong shall not be necessary. However, such a statement may only be included when there is consensus that the given fact is indeed reasonably well established by authoritative sources.

—Preceding unsigned comment added by Sylvain1972 (talk • contribs) 14:28, July 7, 2009

Oppose - There is no need to include this in the policy. If it is clear that a source contains information that is erronious, and that other sources are correct, then the best solution is to simply rely on the information contained in the other sources (ie simply ignore the single source that says Jacko was born in 1959). If it is not clear which source incorrect, then (per WP:NPOV) we should mention what each source says without making judgements as to which is correct and which is not. What we should never do is include a statement in the article along the lines of "X is incorrect" without a source that specifically states that X is incorrect. Blueboar (talk) 15:07, 7 July 2009 (UTC)

I believe that the policy is not very explicit on this point as it now stands. I also believe that there is no reason to tie editors hands here. In some cases it is better to simply ignore incorrect sources, and in some cases it is better to acknowledge them as such, and it would be good for editors to have recourse to both approaches.Sylvain1972 (talk) 15:17, 7 July 2009 (UTC)

There is a difference between acknowleging an error on the talk page (and editing the information in the article with this acknowledgement in mind), and overtly pointing the error out in the article. To overtly point out an error in an article we need a source that specifically mentions that the error exists. Blueboar (talk) 15:33, 7 July 2009 (UTC)

Strong oppose per WP:IFITAINTBROKE. Solves nothing and introduces new problems, plus it's just plain WP:CREEPy. DreamGuy (talk) 15:25, 7 July 2009 (UTC)

Oppose. I don't really see any reason to allow an entity to be singled out in such a way. What makes that one source more important than any of the others? To point out such a mistake would seem to be a violation of WP:UNDUEWEIGHT. — HelloAnnyong ^{(say whaaat?!)} 15:27, 7 July 2009 (UTC)

I also don't see the benefit in adding this. These issues boil down to common sense and are better dealt with on the page in question. SlimVirgin ^{talk|contribs} 15:51, 9 July 2009 (UTC)

Strong oppose per WP:IFITAINTBROKE and WP:UNDUEWEIGHT. There are enough guidelines to deal with this sort of problem already.--BruceGrubb (talk) 12:45, 10 July 2009 (UTC)

Question regard NOR

Suppose (good example) it's an article and in the article the size of a game world is needed for a context statement. Your average game company isn't going to calculate the size of the game world for you and say "oh yeah it's XXX square kilometres". Instead they may state the size of a given unit of area (or provide a way to find the size of a unit of area) and leave it up to the players to figure out the size if they so desire. in this example (which has actually occured on Daggerfall, I see no way of it ever being able to be acceptable to the strictness of NOR, short of pandering by a gaming site for the sake of Wikipedia. Stormscape 14:19, 10 July 2009 (UTC)

Could you give us an example of "the size of a game world is needed for a context statement"? I am not sure I understand what you are referring to. Blueboar (talk) 15:10, 10 July 2009 (UTC)

The entire second paragraph of Daggerfall#Game_world is number estimates given by players of each game, and is needed because Daggerfall has been quoted as being "one of the largest game worlds ever made" at some point on the ancient Bethesda Forums before they switched to Invision. The context of showing the two later title's game world shows just how big it is, and none of these numbers could be found without OR. Stormscape 22:45, 10 July 2009 (UTC)

There's something here that alludes to the size of Daggerfall. [24] You could write to the company and ask for the details, and also ask if they're available online anywhere, or in a leaflet, something you could use a source. SlimVirgin ^{talk|contribs} 22:59, 10 July 2009 (UTC)

SV, doesn't Wikipedia:These are not original research cover this? See Simple or direct deductions. Viriditas (talk) 23:24, 10 July 2009 (UTC)

Yes, I think it would, so long as the deductions were clear enough. SlimVirgin ^{talk|contribs} 23:32, 10 July 2009 (UTC)

Hmmm... calculating the size of a game world might fall under the "routine calculations" exemption... but comparing the result to the size of other game worlds, and connecting it to the statement that Daggerfall is "one of the largest gameworlds ever made" seems like basic OR to me. Blueboar (talk) 14:01, 12 July 2009 (UTC)

OR and the usage of historical logos

There has been a question raised at Wikipedia_talk:Non-free_content#How.2C_when.2C_and_why_for_historical_logo over whether the usage of historical logos to convey how a brand changes over time would be or be not original research if no sourced text explaining this was in the article itself. WhisperToMe (talk) 04:53, 11 July 2009 (UTC)

I'm not seeing anything related to OR in that discussion. Viriditas (talk) 01:58, 12 July 2009 (UTC)

There was a brief discussion... quickly settled. The discussion moved on to other issues. Blueboar (talk) 13:49, 12 July 2009 (UTC)

Court cases clarification needed

Over at the MLM page I sited the BERR vs Amway court case and an editor removed it saying "seems a primary source, at best; not usable under WP:BLP" while I did fina a secondary reference I am confused regarding how Wikipedia defines primary and secondary regarding court cases:

Primary sources are sources very close to an event. For example, an account of a traffic accident written by a witness is a primary source of information about the accident. Other examples include archeological artifacts; photographs; historical documents such as diaries, census results, video or transcripts of surveillance, public hearings, trials, or interviews;...

However

Secondary sources are at least one step removed from an event. They rely for their facts and opinions on primary sources, often to make analytic, synthetic, interpretive, explanatory, or evaluative claims.

By their very nature wouldn't court decisions like BERR vs Amway (Case No: 2651, 2652 and 2653 of 2007) or SAFFORD UNIFIED SCHOOL DIST. #1 v. REDDING (No. 08-479) 531 F. 3d 1071) be considered secondary rather than primary? If not why not?--BruceGrubb (talk) 06:12, 12 July 2009 (UTC)

This should be addressed on the appropriate noticeboards, not here. You have several different issues going on, mostly related to reliable sources and how to appropriately cite them in an article. You certainly did find a secondary source for the primary, but the secondary isn't really about the topic of Multi-level marketing, nor does it seem is the primary. Since you are adding material to a criticism section, it is especially important that the sources discuss the topic directly. From what I can tell, neither of these do that. Viriditas (talk) 09:08, 12 July 2009 (UTC)

Several things here

First, your statement about BERR vs Amway a case involving a Multi-level marketing company not having anything anything to do with the topic of Multi-level marketing is a big "Huh?!?" That is like saying State v. Scopes, Scopes v. State, 152 Tenn. 424, 278 S.W. 57 (Tenn. 1926) and Tammy Kitzmiller, et al. v. Dover Area School District, et al., Case No. 04cv2688 have nothing to do with the topics of either Evolution or Creationism. It just makes no sense.

Second, I know the Scheibeler quote is not stellar but it was the only version of it I could find that didn't have reputable publisher problems. It also appears "Merchants of Deception: An Insider's Look at the Worldwide, Systematic, Conspiracy of Lies That is Amway/Quixtar and their Motivational Organization" but that is a self published e-book and I didn't want to go through the hassle of showing Scheibeler want a expert in the field.

Finally, you never answered the question that brought me to this board in the first place. Why are court cases considered primary when by their very process they look like secondary? THAT is the question I am asking and THIS board is the most relevance place to answer THAT question. Kindly stick to the question being asked rather than go down irrelevant tangents.--BruceGrubb (talk) 12:02, 12 July 2009 (UTC)

As I said before, your question belongs on the noticeboards, not this page. This page is for discussing the policy. And, as I said before, neither the primary or secondary source you cited is directly related to the topic of MLM. Both sources are about separate and distinct topics. The primary source you used is about Amway, not MLM, and the secondary source you used is about Amway Global and its funding of the Republican party, not MLM. I hope that is clear enough for you. Viriditas (talk) 12:07, 12 July 2009 (UTC)

Why are court cases considered primary sources? Many reasons, although I suspect none will satisfy you. Essentially, for our purposes, trial transcripts are considered raw data. Articles about the cases would be classified as secondary. As editors, we don't have the ability to sort through a court case and pick and choose what we think is important. We use secondary sources to do that, and cite the primary as needed. Can you find me a single research guide that describes court cases as secondary sources? They are universally classified as primary sources because they are directly connected to the event. Viriditas (talk) 12:32, 12 July 2009 (UTC)

We have to differentiate what we are talking about when we talk about "court cases". Are we talking about a transcript of trial testimony... or the final decision of the Judge. A tiral transcript of testemony is indeed a primary source (raw data) and would only be reliable for attributed statments as to what the person testifying stated during the trial. Such a source should be treated as the opinion of the person testifying, and statements of opinion have very limited usage in a BLP. The final, published decision of the Judge, on the other hand, should be considered a secondary source. The Judge is a legal expert who analizes the law and reaches an expert conclusion. Legal decisions undergo a process similar to peer review (actually they can under go two forms of peer review... The appeals process, and commentary in journals such as the Law Review).

As to your specific situation... it would be OR to discuss BERR v. Amway unless a) that decision specifically discusses MLM, b) if a decision involving MLM specifically discusses the BERR v Amway, or c) if some other reliable source has discussed BERR v Amway in direct connection to MLM. Blueboar (talk) 13:47, 12 July 2009 (UTC)

Trial transcripts would be primary-source material. They're not taking a step back to analyse what happened in terms of a wider context, and they're notoriously easy to misuse. They also shouldn't be used in a BLP unless to augment secondary sources who have discussed that trial. SlimVirgin ^{talk|contribs} 21:41, 12 July 2009 (UTC)

Slim, I think we are all in agreement as to trial transcripts. I am not sure if you are familiar with how legal cases are cited... when someone cites a case (as in: BERR v. Amway) they are not citing the trial transcript... they are citing the court's final decision. Blueboar (talk) 02:40, 13 July 2009 (UTC)

Should we follow the Bluebook? :) Viriditas (talk) 06:20, 13 July 2009 (UTC)

When I talk about court case I am talking about the final decision of the Judge which is arrived at by relying on what Wikipedia considers primary sources (eyewitness testimony, police interviews, expert testimony, etc) to make an analytic, synthetic, interpretive, explanatory, or evaluative decision ie the very definition of secondary source. Viriditas seems to be saying that if I wanted to use "Wilson did not have sufficient suspicion to warrant extending the search to the point of making Savana pull out her underwear" out of Redding in a relevant article I would have to find a valid secondary source quoting it to do so. That is ridiculous.

As for the Bluebook remember that is for US court cases. As this case was in the UK wouldn't OSCOLA be the correct standard? If so the proper reference if I am reading that correctly could be BERR vs Amway [2008] EWHC (Ch)1054 with the appeal being BERR vs Amway [2009] EWHC Civ 32.--BruceGrubb (talk) 06:47, 13 July 2009 (UTC)

How can I seem to be saying something that I never said? Please explain. Viriditas (talk) 06:53, 13 July 2009 (UTC)

You said "Why are court cases considered primary sources? Many reasons, although I suspect none will satisfy you." and then went into your reasoning saying along the way "As editors, we don't have the ability to sort through a court case and pick and choose what we think is important. We use secondary sources to do that, and cite the primary as needed." As Blueboar points out when most people cite a case they are referring to a court's final decision not transcripts. In fact, for practical purposes the only way most people are even going know anything that would be in a transcript is a secondary source quotes it. We had a similar talk over at Talk:Jack T. Chick about what do you do when a reference used by a living person is blatantly wrong as demonstrated in Big Daddy. Note that Edwards v. Aguillard, 482 U.S. 578 (1987) has nothing directly to do with either Hovind or Jack Chick but it does relate a claim they both have made ie "It has never been against the law the teach the Bible or creation in public schools". Edwards v. Aguillard clearly shows this statement to be false and even Science and creationism: a view from the National Academy of Sciences by National Academy of Sciences on page 29 flash states that teaching "Creation Science" as scholarship is unconstitutional (ie illegal).--BruceGrubb (talk) 14:18, 13 July 2009 (UTC)

Bruce, trial transcripts are used all the time on Wikipedia, so your statement that "the only way most people are even going know anything that would be in a transcript is a secondary source quotes it" completely misses the point about OR and primary sources by several astronomical units. If you say I require a secondary source, and the court decision is a secondary source, then what is your point? I don't see one. But to address your latter example, unless a Jack-Chick related source directly discusses an error made by Chick, it would be OR to introduce it from sources unrelated to it as you describe. This topic is discussed in Intelligent design in politics in the section 2001 Arkansas House Bill 2548, and while the information does appear accurate, the sources are either 404 or do not appear to support the content. Without sources that directly discuss the topic, your example illustrates original research. Viriditas (talk) 22:10, 13 July 2009 (UTC)

Trial transcripts are used all the time on Wikipedia? Didn't know that. Do you have some examples of people quoting raw transcripts so I can go and fix those pages? As for 404 errors those are easy enough to fix--check if Internet archive has a copy of the web site (which in this case it does for the actual bill) and use that--end of problem. This does show a major problem with how web sties are referenced in Wikipedia-the use of raw urls without anything to tell you what the blazes it goes to. I had to clean up a bunch of those in the Chick Publications Inc. and the reason is simple--urls will eventually break and you need to know what the URL originally referenced so you can find it elsewhere (perhaps in even dead tree form)

In the case of the Big Daddy mess over at Chick Publications Inc. it is supported by secondary sources. Foley, Jim (August 31, 2001) "Fossil Hominids: Big Daddy?". talkorigins.org directly refers to the errors of Hovind, Spinney, Jacob (November, 2004). "A Close Look at Dr. Hovind's List of Young-Earth Arguments and Other Claims". Skeptic Report. http://www.skepticreport.com/creationism/hovind.htm states "In fact, there have been repeated rulings by the Supreme Court declaring it illegal to teach Creationism in public schools." (but fails to mention the cases involved), and the BioSence article mentions the cases stating in no uncertain terms that teaching creationism in a US public school is unconstitutional ie illegal. Please note that the sentence is so structure to avoid WP:SYN.--BruceGrubb (talk) 03:56, 14 July 2009 (UTC)

Why would you go and "fix" those pages? If you know how to use Google to search Wikipedia, you can easily find trial transcripts used in Wikipedia articles. And, the use of primary sources is perfectly acceptable when we have supporting secondary sources. I see that you changed the dead link in the Intelligent design in politics article, but the content is still not supported by references. Viriditas (talk) 09:33, 15 July 2009 (UTC)

Actually, Viriditas, it would be helpful to see some examples of trial transcripts being used in Wikipedia (whether they need to be "fixed" or not)... and since not all of us do know how to use Google to search Wikipedia, it would be helpful if you provided us with some links to articles (from your comment, I am assuming that you do know how to use Google to search Wikipedia, so this should not be a huge burden for you). Thanks. Blueboar (talk) 13:17, 15 July 2009 (UTC)

trial transcript site:en.wikipedia.org. Viriditas (talk) 13:58, 15 July 2009 (UTC)

That will return hundreds of articles. Mumia Abu-Jamal, a featured article, makes extensive use of trial transcripts. BTW, this search also revealed (via another FAC) that the WP:RS policy circa September 2006 used to claim the following:

In general, Wikipedia articles should not depend on primary sources but rather on reliable secondary sources who have made careful use of the primary-source material. Most primary-source material requires training to use correctly, especially on historical topics. Wikipedia articles may use primary sources only if they have been published by a reliable publisher e.g. trial transcripts published by a court stenographer, or historic documents that appear in edited collections.

Whatever happened to this? Viriditas (talk) 14:39, 15 July 2009 (UTC)

That language was changed because the consensus was that it was misleading... the point being made at the time centered on the need for primary sources to be published, but it did not take into account the issue of when it was appropriate to cite them in the first place. We now make it clear that while there are situations when it is ok to cite a primary document (such as a trial transcript) ... these situations are limited. And one situation where doing so is very limited is in a BLP article. In other words, in some situations you may cite a trial transcript... while in other situations you may not cite a trial transcript. Blueboar (talk) 15:14, 15 July 2009 (UTC)

Just a word of caution... I am not saying that this is the case in this situation, but many people equate "OR" with "Primary source" and "Not OR" with "Secondary source". This is not accurate. Primary sources can be used without violating NOR (although, as SV points out, their use is severely limited in BLPs)... and one can definitely engage in OR using Secondary sources. While a court decision should be seen as being a secondary source, it can still be used in a way that constitutes OR. Blueboar (talk) 15:24, 13 July 2009 (UTC)

Assata Shakur, a featured article, uses secondary sources to source material from trial transcripts that are apparently unavailable in their original form. There was a discussion about this on the FAC. This is an interesting instance where the trial transcripts would have been useful for fact-checking specific claims, but could not be accessed for whatever reason. Viriditas (talk) 14:18, 15 July 2009 (UTC)

New example in SYN is not synthesis

Here's the example that was recently put into WP:SYN with the sentence of explanation that follows it.

The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world.

Although no conclusion is drawn and both facts are true, the sentence implies that the UN has failed to maintain world peace.

The second part of the sentence alone, i.e. "since its creation there have been 160 wars throughout the world" can be said to imply that the UN has failed to maintain world peace. The first part of the sentence is not necessary for this implication. Therefore it isn't synthesis. --Bob K31416 (talk) 20:58, 12 July 2009 (UTC)

The conclusion that is obvious in any reader's mind is that the UN has failed to achieve its objective. That is not part of the second clause. There is no problem at all if a reliable source out there makes this entire argument. Problems arise when Wikipedians present suggestive statements of this sort that imply a novel argument, one that is not found in the literature. JN466 21:58, 12 July 2009 (UTC)

The reader can infer from only the second part, with the comment about 160 wars since the UN began, that the UN has failed to maintain world peace. Also, please note that we are talking about implication, so that it won't show up explicitly in any case, synthesis or otherwise. The 2nd part is POV and needs to be corrected for that reason. --Bob K31416 (talk) 02:43, 13 July 2009 (UTC)

Bob... your point about POV is irrelevant... when joined with the first statement this is a very good example of a synthesis, whether the second statement is POV or not. Blueboar (talk) 13:25, 13 July 2009 (UTC)

I only added the comment about POV so that it is understood that I think the sentence is still a problem, even though it isn't synthesis. The point I'm making isn't a question of POV, it's that the second statement alone is sufficient for the implication, so it's not synthesis, as I previously mentioned. --Bob K31416 (talk) 13:40, 13 July 2009 (UTC)

Of course it is... The reason why it is a synthesis is that the implied conclusion is predicated on both parts of the sentence. Change the first part and the reader is led to a different conclusion... suppose the UN was not created to "maintain world peace", but to "coordinate joint military action"... if that were the case, and this was stated in the first part, then the conclusion implied would be that the UN was successfully carrying out its mission. (Which, by the way, would also be an improper synthesis if no single source connected the two parts to reach that conclusion). Blueboar (talk) 13:46, 13 July 2009 (UTC)

A characteristic of implication is that it uses the knowledge that the reader already has. The second part can stand alone because the reader is already aware that the UN represents itself as seeking to maintain world peace. --Bob K31416 (talk) 18:10, 13 July 2009 (UTC)

That was a counter-factual to prove a point. I agree that if you remove the first part completely, you no longer have a synt situation... but that just reinforces my point... with it (or with some other first part as in my counter-factual), you do. Oh... one last point... you should never assume that your reader is aware of the same things you are aware of. Not everyone knows what the mission of the UN is. Blueboar (talk) 18:25, 13 July 2009 (UTC)

Removing the first part eliminates the synt situation? I didn't say that. On the contrary, I was saying that even with the first part there it isn't a synt situation because the second part alone makes the implication.

Re "you should never assume that your reader is aware of the same things you are aware of." - That's a good thing to remember, not only in this case, but always when editors here are trying to explain policy at WP:NOR. I considered this before posting my last message, and it seemed that the vast majority of the people who come to Wikipedia, would know that the UN claims to seek to maintain world peace. --Bob K31416 (talk) 20:49, 13 July 2009 (UTC)

For the reasons I stated before I do not think "this is a very good example of a synthesis" as the examples hang on the use of "but", and "only", without those there is not syntheses. For example "Since the creation of the UN, with its stated objective to maintain international peace and security, there have been 160 wars (ref 1 and 2)." combines the two sentences but does not advance a position, and I think that the two simple examples are likely to confuse people into not summarising sources, but quoting them or using disjointed sentences. -- PBS (talk) 17:08, 13 July 2009 (UTC)

PBS, I very much disagree that your re-write does not advance a position... I think the position that "the UN is not successful in its mission" is still clearly implied. We still have an impropper synthesis being drawn from two sources. Blueboar (talk) 17:21, 13 July 2009 (UTC)

How is it implied? Without knowing how many wars there were in the 60 years preceding the creation of the UN one can not know how successful it has been, (and by making that statement I am making a huge assumption that there is a correlation between the number of wars and the success or failure of the stated UN mission). --PBS (talk) 17:36, 13 July 2009 (UTC)

As almost a side note, I don't think 160 wars can be called "international peace and security," no matter how many wars were fought in the years prior to the formation of the UN. Dreadstar † 22:37, 13 July 2009 (UTC)

This is exactly why it is an impropper synt... even if we try to put a positive spin on this... the argument that is being made by placing these two parts together is that there is a correlation between the number of wars and the UN's mission (success or failure). To make this argument (even through implication), we need to cite a single source that makes the connection, not two sources that don't. Blueboar (talk) 18:12, 13 July 2009 (UTC)

You have already gone a step to far for me. Is "The UN's stated objective is to maintain international peace and security. Since its creation there have been 160 wars throughout the world." syn? What about two sentences in the same paragraph, or in the same section ... --PBS (talk) 18:25, 13 July 2009 (UTC)

(ec) We can't pin down every example of SYN here, Philip, nor would we want to. It gets to the point on some articles, particularly contentious articles, where SYN is used an excuse to stop people from joining sentences together, unless everything is quoted. The more we say about it in this policy, the greater the chances of it being misused in that way.

The SYN violations we see sometimes boil down to POV pushing, but are just as often the product of bad writing. People often don't know how to summarize source material without a SYN violation. As I said earlier on, we had classes when I was at school called "comprehension." They involved reading a text, then being asked to summarize what it said, without using its words or straying from its content. We did this from the age of six-ish to mid-teens, until it became second nature. Judging by the editing I see, not everyone is taught this. These are problems that go far beyond what this policy can cope with. It would be great if we could set up a page about reading comprehension and how to summarize accurately without copying word for word, but it would be a big job, and I wouldn't be sure how best to approach it. SlimVirgin ^{talk|contribs} 20:08, 13 July 2009 (UTC)

And to answer your question... Yes, Placing those two sentences next to each other like that is indeed syn... placing them in the same paragraph or section might be (or might not be) depending on the wording of the rest of the paragraph or section. The key is whether two sentences (each cited to a seperate source) combine to form an argument that is not made by any source... it does not matter whether that argument is stated or implied.

A synthesis is formed when two or more things combine to form a conclusion. If that combination and conclusion is not made by a single source (ie if it is a Wikipedian who is setting up the combination and conclusion) then that synthesis is a form of OR. (note: it is still a synthesis, but isn't OR, if a single source combines the parts and reaches the conclution.) Proximity agrivates the situation and makes it easier to detect (and conversely lack of proximity will ease the situation and make it less obvious), but it is not the central issue. The lack of sourcing for the connection is. Blueboar (talk) 20:22, 13 July 2009 (UTC)

Blueboar, I think SV and I am singing from the same song sheet, the only point where she and I seem to differ is if the simplified example is a good example to include in this policy given that editors have to understand the subtleties of but and only in the examples -- and be able to take those examples and not misuse them in content disputes -- (She thinks they are good examples and I don't), but I think that you and I are much further apart. I'm interested to hear what other editors think. PBS (talk) 20:54, 13 July 2009 (UTC)

OK... then chalk me up as "its a good example" (but there should be a third that is even easier)... I'll let it go for now. Blueboar (talk) 21:20, 13 July 2009 (UTC)

If an editor is having trouble with the "subtleties of but and only", then I imagine they'll have far larger and more basic issues with editing than understanding OR or Synthesis. As I stated earlier, I don't see any problem using a simple and clear example that utilizes a conjunction, as a matter of fact, I think the use of "but" in the example makes it even clearer and simpler. The current examples in NOR Polcy are fine. Dreadstar † 22:25, 13 July 2009 (UTC)

More examples

Here's a couple of examples of 1) implication by synthesis and 2) simple synthesis with a stated conclusion. They are based on an edit of an actual article.

1) Implied conclusion.

Intelligence officers of the attacking forces claimed to have telephoned the residents of buildings where military assets were suspected of being stored, to warn them to leave before it was bombed.^[23] During the war, 90% of the telephone system was down.^[24]

2) Stated conclusion.

Intelligence officers of the attacking forces claimed to have telephoned the residents of buildings where military assets were suspected of being stored, to warn them to leave before it was bombed.^[23] During the war, 90% of the telephone system was down.^[24] Thus the attacking forces were unable to warn 90% of the residents whose buildings were bombed.

--Bob K31416 (talk) 23:04, 13 July 2009 (UTC)

Good example. Here another, also from an actual article:

Organisation X, with which A. N. Other was associated, called for a pre-Christmas product boycott. They subsequently claimed on their website that it had a dramatic effect. [properly sourced]

In the run-up to Christmas that year, Retailer Z suffered disappointing sales,[source mentions sales drop for retailer Z, but does not mention Organisation X, nor the boycott] and its share price dropped.[share price source for Retailer Z does not mention Organisation X or the boycott either]

Implied conclusion:

Organisation X's boycott may have had something to do with the drop in sales and subsequent drop in share price of retailer Z. [not in any of the sources] --JN466 23:15, 13 July 2009 (UTC)

And here's an example that JN466 mentioned on another thread.

On February 19th, 2009, 3,000 followers of Maitreya Yogeshwara chanted in New York to help the world economy.(source1) The following week, the Dow Jones Index rose by 10%.(source 2)

--Bob K31416 (talk) 13:59, 14 July 2009 (UTC)

I agree, Bob's new examples are good. If this helps resolve the issue, I would not object to using them instead of SV's (and this from someone who thinks SV's example is fine). Blueboar (talk) 00:29, 14 July 2009 (UTC)

This is not a good example:

"Intelligence officers of the attacking forces claimed to have telephoned the residents of buildings where military assets were suspected of being stored, to warn them to leave before it was bombed.[23] During the war, 90% of the telephone system was down.[24]"

It's too wordy, and it isn't as obvious as Laurent's example. You could make it more obvious with "but," which implies their mission to warn people was thwarted. Without "but," it's much less clear. And "claimed" to have telephoned introduces confusion. If they only claimed to have telephoned, and perhaps really didn't, then it makes no difference that the telephone system was down. Finally, why invent an example, when Laurent added a real one? SlimVirgin ^{talk|contribs} 01:10, 14 July 2009 (UTC)

It's a good example to think about, but I agree it is not a good example for putting in the policy. I think what's in the policy right now works well. JN466 02:22, 14 July 2009 (UTC)

SlimVirgin,

1) Re "but" - Use of "but" is a point of contention with PBS.[25] It wasn't in the real edits that either my example or yours is based on.

2) Re "claimed" - could you clarify what you mean? Also, we may be able to remove "claimed".

3) Re "Finally, why invent an example, when Laurent added a real one?" - Invented? Here's a link to the real edit that it was based on. [26] And for reference, here is a link to the real edit that your example is based on. [27] Seems like both are based on a real one.

--Bob K31416 (talk) 09:46, 14 July 2009 (UTC)

Can I ask why there is a need to use a "real" edit as an example? Why does it matter whether the example matches something actually placed into an article? I would think that the goal is to use an example that clearly shows an impropper synthesis, regardless of whether it is "real" or not. No? Blueboar (talk) 14:56, 14 July 2009 (UTC)

I also think the "phone" example is too wordy and unclear. To be honest, by the time I got to the end of the two sentences, I forgot the initial statement about the calls and didn’t even come close to connecting the two on re-read…I was like…”yeah, so…90% of the phones were down, so what?" As Jayen466 states, it's a good "thought" example, but not good for including in the policy. Even if the word "claims" is removed, it's still too clunky to include in Policy, but perhaps these examples are a good start to an "OR examples page" as previously discussed. The current example in policy by Laurent and SV is much clearer, simpler and obvious right off the bat. I like the idea of having a real example from a real dispute or article on WP, this gives a view of a real case that editors can actually see in action here on Wikipedia. This makes the example relevant, and if they exist, it gives the ability to track down any relevant discussions or more detail on the reasons it was OR/Synth and what was done to correct it. Dreadstar † 16:09, 14 July 2009 (UTC)

Agreed... I suppose my question is why people seem to be quibbling about the fact that the current Laurent/SV example is based on and not actually the same as something put in an article? (they use "but" while the article uses "however", and SV combines it into one sentence instead of two). I don't see a need to have an example that exactly matches something from an article. If amending a "real life" synt makes for a clearer example, I have no problem with doing so. Blueboar (talk) 16:32, 14 July 2009 (UTC)

Oh, right, I see what you mean. I agree with you, I don't have any trouble with the minor changes to the "real-life" article content in order to make it a sounder, clearer and even broader Policy example. It doesn't need to be exact, just firmly based in reality (or even loosely-based...but never Lucy-based...unless it's the Comedy Policy...:). Dreadstar † 16:51, 14 July 2009 (UTC)

Well... if we want to make it LUCY-based we should tweek it to say:

Lucie's stated objective is to perform at Club Tropicana, but since I Love Lucy first aired there have been over 160 times when Ricky told her she couldn't do so.

Although no conclusion is drawn and both facts are true, the sentence implies that Lucy has failed to perform at the Tropicana.  :>) Blueboar (talk) 17:15, 14 July 2009 (UTC)

LOL! I love it! We'll definitely have to include this LUCY-based example in our OR Examples page as an example of fictional comedy OR... :) Dreadstar † 18:58, 14 July 2009 (UTC)

The problem with the phone system example is that it's too complex to be a replacement for our simple example. There are at least two possible conclusions in it (possibly more if we look closely):

(1) that because 90 percent of the phone system was down, the attacking forces were unable to warn the residents of the bombed buildings i.e. the attacking forces meant well, but were necessarily ineffectual because of the broken phone lines;

Or (2) the fact that 90 percent of the telephone system was down shows they almost certainly did not telephone the residents, though they claimed to have done so (the example does not say "the attacking forces claimed to have tried to telephone," it says, "the attacking forces claimed to have telephoned"). Perhaps when they made the claim, they didn't realize that 90 percent of the system was down when they were supposedly making their calls. Therefore we know they are liars.

SlimVirgin ^{talk|contribs} 17:19, 14 July 2009 (UTC)

Another possibility: "how lucky was it that all the targets were in the 10% of the phone system that actually worked...so they all got warned" Dreadstar † 18:58, 14 July 2009 (UTC)

That's the nature of implication for both the examples. That's why examples involving implication are intrinsically not simple. What they imply depends on the reader. Please note these possible conclusions for your example.

1. the UN made the situation worse
2. the UN was ineffective and neither helped nor hurt the situation
3. the UN helped the situation but not much
4. the UN helped the situation but couldn't remove all the little wars
5. the UN failed at maintaining world peace
6. it's impossible for wars to be eliminated
7. no conclusion

--Bob K31416 (talk) 18:15, 14 July 2009 (UTC)

I disagree Bob, the UN example provides a simple, clear and effective OR/Syth situation which is clearly meant to repudiate the UN's stated mandate, there's no way 160 wars can be considered "world peace" in the context of how the sentences are presented. However, the "phone" example is much less clear, more complex, and less effective because it is too wordy and has a nebulous connection between the two statements, which left me wondering what the example was really supposed to be telling us. Dreadstar † 18:53, 14 July 2009 (UTC)

(edit conflict)BTW, example 2 in my message at the top of this section has a stated conclusion, so there aren't the ambiguities and complexities that come from implication. --Bob K31416 (talk) 18:55, 14 July 2009 (UTC)

And that's a problem, Bob, it has to be explained, so it's not simple and doesn't really lead to the conclusion chosen. The current examples in the policy are far superior. Dreadstar † 19:00, 14 July 2009 (UTC)

For the implication example that was recently added to WP:SYN, the conclusion is selected there with the following phrase, "the sentence implies that the UN has failed to maintain world peace." This isn't explaining the conclusion, it's selecting a conclusion, which is misleading since there are other possible conclusions, as I mentioned above.

My point in mentioning example 2, which is in my message at the top of this section, is that it is an example that includes statement of the conclusion. There is only one conclusion, the stated one. So it could be an example in WP:SYN that is simple because it doesn't need to explain the conclusion, unlike examples that involve implication which have to explain/select a conclusion from several. --Bob K31416 (talk) 19:17, 14 July 2009 (UTC)

All SYN examples involve an implication of some kind. SlimVirgin ^{talk|contribs} 19:29, 14 July 2009 (UTC)

Didn't quite understand your point. You're invited to clarify if you like. Please note that in example 2 that I mentioned, the conclusion is stated. --Bob K31416 (talk) 19:34, 14 July 2009 (UTC)

All violations of SYN involve an implication, whether explicit or otherwise, that isn't contained in the sources used for the different parts of it. That's what it is to be a violation: that a position is advanced in some way, a suggestion made or hinted at. SlimVirgin ^{talk|contribs} 19:53, 14 July 2009 (UTC)

In the context of this discussion, implication meant a conclusion that is not explictly stated, which is what your example has. In the example 2 that I mentioned, the conclusion was explicitly stated. --Bob K31416 (talk) 22:09, 14 July 2009 (UTC)

I haven't added an example of my own. If you mean Laurent's examples, the conclusions are barely hidden. SlimVirgin ^{talk|contribs} 22:18, 14 July 2009 (UTC)

In addition to my above objections, example 2 has too explicit of a conclusion given - the easiest way to combat that conclusion is via WP:V, which makes the example even less effective for NOR policy. Neither example 1 nor example 2 in this section are satisfactory for inclusion in NOR policy. We're looking for a synthesis of sentences, not a blatantly unsupported conclusion - especially one that isn't clear unless it's explained to the reader. Dreadstar † 23:00, 14 July 2009 (UTC)

Re "example 2 has too explicit of a conclusion given - the easiest way to combat that conclusion is via WP:V, which makes the example even less effective for NOR policy." - Please note that if only WP:V is used, without WP:SYN, an editor may argue that the conclusion is in the combination of the two sources. WP:SYN is needed to make it clear that, although the conclusion comes from the combination of the two sources, the conclusion is prohibited because it is in neither source taken alone. That's the basic purpose of WP:SYN.

In any case, example 2 was meant for editors who are unfamiliar with synth, who are at the stage where they are trying to understand the basics, before going on to more subtle or complex examples. --Bob K31416 (talk) 02:56, 15 July 2009 (UTC)

Yes, I'm completely aware of the purposes of both examples 1 and 2, but I disagree that they are good enough for inclusion in Policy for the many reasons stated throughout this section. The current examples in policy are far superior to either example 1 or 2. We already have excellent examples and explanations that clearly describe and show what Synthesis is, how it works and why it's not allowed. I am welcoming and eager for any examples or improvements to what is already in place to make OR/Synth better and easier to understand; however, I do not see examples 1 or 2 in this section as improvements over what we already have. These proposed examples may be good for inclusion in an "OR Examples" page, but not in main policy. Dreadstar † 17:28, 15 July 2009 (UTC)

<outdent> A page with SYN examples attached to this policy page would be useful. The WP:ORIGINALSYN section of Phenylalanine's essay started off that way. SYN seems to be a concept that takes Wikipedians quite a long time to get their head around; witness people arguing again and again that "I didn't say 'but' or 'however', therefore it is not SYN." Some more examples on a subpage might help. JN466 18:16, 15 July 2009 (UTC)

SYN addition

Laurent1979 recently added a new example of synthesis to a different section. [28] As several people have suggested adding a simpler example to the synthesis section, I've taken Laurent's and added it here, as it's a very easy example to follow. Thanks for fixing it, Dreadstar. :) SlimVirgin ^{talk|contribs} 23:38, 26 June 2009 (UTC)

My pleasure..:) It's a great addition...! Dreadstar † 05:41, 27 June 2009 (UTC)

Laurent's contribution is a type of synthesis not yet discussed in WP:SYN. It involves an implicit conclusion, rather than synthesis with an explicit conclusion that is described in WP:SYNTH by A+B=C. The C is not stated in Laurent's idea. It is implied by juxtaposition. Perhaps this should have been considered for inclusion in WP:SYNTH as an idea in addition to what was there already, rather than a "simple example" of what was in WP:SYNTH, which it wasn't. Thus I reverted the edits that are trying to use it as a "simple example". It's an idea that wasn't expressed in WP:SYNTH before, rather than an example of the ideas that were already there. --Bob K31416 (talk) 13:42, 28 June 2009 (UTC)

There was some discussion of implicit conclusion here. I think the present idea of Laurent1979, with the juxtaposed aspect clarified by Crum375, is better. As Crum375 mentioned in the edit summary, juxtaposition is key. --Bob K31416 (talk) 15:58, 28 June 2009 (UTC)

I don't follow what you're saying, Bob. We can't have two sections that say essentially the same thing. Is it implicit v explicit you're think of as the distinguishing factor? If so, it makes no difference. A SYN violation is when an implicit or explicit conclusion is reached by synthesizing sourced material that wasn't explicitly reached by the source(s). (And a juxtaposition is just one form of synthesis.) SlimVirgin ^{talk|contribs} 21:50, 28 June 2009 (UTC)

By reverting another editor's revert of your new change, without getting consensus, it appears that you are edit warring. --Bob K31416 (talk) 22:51, 28 June 2009 (UTC)

I don't see any disagreement. All I did was move it to the right section. Dreadstar agreed. The editor who added it obviously wanted it there in some form. The editors who worked on his example must have wanted it too, or they'd have removed it. But there's clearly no need for two sections about SYN. SlimVirgin ^{talk|contribs} 22:55, 28 June 2009 (UTC)

Pardon me for stating the obvious, but why wasn't this talked about in discussion before being placed in a core policy article? I am not even looking at the content of Laurent's addition, but think its extraordinarily stupid to do something like that. It's akin to altering the design of the main space to 'make more sense' or whatever. Being bold is all fine and dandy, but NOR potentially affects thousands if not millions of articles. Changes in these polices should be glacially and exhaustively discussed before being altered. - Arcayne (^{cast a spell}) 23:35, 28 June 2009 (UTC)

And no, we don't need two sections that say the same thing. Let's discuss the specific parts that Laurent is seeking to add to the SYN policy. - Arcayne (^{cast a spell}) 23:41, 28 June 2009 (UTC)

You're right. Normally Laurent's addition would have been removed as too bold without discussion. But in this case, several people had recommended adding a simple example to SYN, and Laurent's was a good example, so it was kept. That is, the discussion had already taken place, in effect. We were just waiting for that good, simple example to be found. SlimVirgin ^{talk|contribs} 00:13, 29 June 2009 (UTC)

I think it doesn't matter whether the new example is in a new or synthesis section. However, the example and accompanying wording should be made as simple as possible. In current version the emphasis is put on an argument that synthesis is bad because it violates NPOV. I don't think that is a point we should strive to make here. It should be clear that synthesis is bad in it self, whether it violates NPOV or not. -- Vision Thing -- 08:25, 29 June 2009 (UTC)

I can't see where we say it's bad because it violates NPOV. SlimVirgin ^{talk|contribs} 08:44, 29 June 2009 (UTC)

From two examples that push an opposing POV (UN's work has been (un)helpful) someone can draw a conclusion that implicit synthesis is not a problem if a sentence is written in a neutral way. That can be avoided with different wording. -- Vision Thing -- 10:11, 29 June 2009 (UTC)

One example is written as if anti-UN and the other as if pro-UN, making the point that, with unsourced synthesis, we can easily produce whatever our own POV is. Therefore, we need to stick to sources. It is true that unsourced synthesis is only a problem when serving to advance a position. Without that, we just have regular synthesis, which is what all articles are. As soon as you put two sentences together, it's unsourced synthesis if you're not plagiarizing. So, yes, you're right that there's an inherent NPOV connection, for sure, but for the purposes of this policy, it's OR we're more concerned about: that any synthesis that serves to advance a position that isn't the source's position must be the editor's, and therefore OR. SlimVirgin ^{talk|contribs} 11:15, 29 June 2009 (UTC)

But wouldn't previous example ("The UN was created in 1945 with stated objective "to maintain international peace and security". Since 1945, there have been 160 wars throughout the world.") also be violation of NOR? If the source for second sentence doesn't talk about UN at all, wouldn't this also be an example of OR even though presentation is neutral? -- Vision Thing -- 18:49, 30 June 2009 (UTC)

Slim Virgin, You wrote, "All I did was move it to the right section." - That's not true. You modified it considerably when you moved it.[29]

Here's how it was before you changed it.

Juxtaposing facts in a way that advances a position

Do not juxtapose facts to advance a position unless the juxtaposition is supported by a reliable source.

Sometimes it seems neutral to simply state the facts and let the readers draw their own conclusions. However, you should not try to arrange these facts in a way that implies more than what is written.

For instance, consider the following sentence:

The UN was created in 1945 with stated objective "to maintain international peace and security". Since 1945, there have been 160 wars throughout the world.

Although no conclusion is drawn, and although both statements are individually verifiable, their juxtaposition implies that the UN has failed at maintaining world peace. If no reliable source combines these two facts in the way it is done here, it would be original research for us to do so.

Here's how it is after you changed it.

A simple example:

The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world.

Although no conclusion is drawn and both facts are true, the sentence implies that the UN has failed to maintain world peace. If no reliable source has combined the material in this way, it constitutes original research. It would be easy to imply the opposite using the same material, illustrating how, when no source is provided, facts can easily be manipulated:

The UN's stated objective is to maintain international peace and security, and since its creation there have been only 160 wars throughout the world.

If you had simply moved it to the SYN section, there wouldn't be this disagreement. I would have supported that move. I suggest that you fix that now, i.e. make a subsection of WP:SYN titled Juxtaposing facts in a way that advances a position, that contains the previous version that was indicated in the first show/hide above, and then we can work from there. This seems to be the situation that would have the most agreement, since that is the version that Laurent1979, Crum375, Tool789789 and Vision Thing worked on and I expressed support for. --Bob K31416 (talk) 17:37, 29 June 2009 (UTC)

I'd strongly disagree with adding the term "juxtaposition" to the NOR policy, never mind as a section title. While Bob's example is one that clearly advances an original thesis which may or may not be true according to the cited sources, I guarantee WP will have problems with editors only reading as far as "juxtaposed" and using it to remove content. We had similar problems with people ignoring the subtleties of the original Smith/Jones example. For the next couple of years after the example was added, every time a deletionist didn't like something, even something as basic as how an article was organized, they would trot out "original research" as the excuse to block it. We've had debates over "OR by juxtaposition" before but it thankfully never reached consensus. Making it explicit in the policy would only reignite those problems. Squidfryerchef (talk) 19:12, 29 June 2009 (UTC)

I think I found what you are referring to: Wikipedia_talk:No_original_research/Archive_39#Insidious_OR. Was this it? If so, I wasn't able to find where they discussed the problem of using the word "juxtapose" or variations thereof, as you mentioned above, although they used those words in their discussion. Maybe what they were discussing was the idea of implicit conclusion, which BTW is in both versions here.

Perhaps one of your points is that neither version should be in WP:NOR, since they both involve the idea of implicit conclusion which may be misused? The "misuse" aspect is something important to consider. --Bob K31416 (talk) 20:45, 29 June 2009 (UTC)

P.S. Re "While Bob's example..." I can't take credit for it since I didn't contribute a single word. --Bob K31416 (talk) 21:07, 29 June 2009 (UTC)

Apparently both of these versions require discussion and a version should only be put into the policy page after consensus is reached. Thus I reverted to the state before either was added. --Bob K31416 (talk) 22:07, 29 June 2009 (UTC)

I'm reverting to the version where it belonged in one section. We can discuss the contents, and remove the example if that's what people want, but there's no reason at all to have two sections on the same subject.

What do you see as the difference, in this context, between a synthesis of sentences and a juxtaposition? Note: I mean only in this context. I know what the words mean, so no need to explain the difference. But how is the difference functioning here? SlimVirgin ^{talk|contribs} 22:43, 29 June 2009 (UTC)

• This seems to be a dispute about whether it was added "in process" or "out of process" rather than "Is it a good idea?" Is there anyone who thinks it's not a good idea, and if so, why? Reverting because "It wasn't discussed first!!!" seems a bit pedantic. ⟳ausa کui ^× 23:05, 29 June 2009 (UTC)

This appears to be an attempt to introduce into a policy page an addition that is supported by only a minority. If you think otherwise, take a survey and get the opinions and positions of editors so that we can see if there is a consensus for SlimVirgin's version. In my opinion, SlimVirgin's version is a misrepresentation of a "simple" example of the basic principle of a violation of WP:NOR, which is A+B=C, where C is an explicit conclusion that is not sourced. Laurent's version recognizes that the idea of C as an implicit conclusion is a more subtle aspect.

From Sqidfryerchef's remarks, whether or not that editor intended, it appears that there may be problems with the implicit conclusion idea that is in both versions. I think I found the discussion that he was referring to, but we'll see which ones he meant if he cares to respond. In addition to Squidfryerchef's recollections, I seem to recall similar discussions where the implicit conclusion idea wasn't accepted when it was proposed previously, possibly by Phenylalanine. This would require some digging to find. Anyhow, there appears to be sufficient reason to discuss and reach consensus on this talk page before putting either version into the policy page. Seems fair and consistent with normal Wikipedia practice, rather than trying to edit war a change into policy. --Bob K31416 (talk) 23:53, 29 June 2009 (UTC)

Bob, could you answer the question, please? Repeating it: "What do you see as the difference, in this context, between a synthesis of sentences and a juxtaposition? Note: I mean only in this context. I know what the words mean, so no need to explain the difference. But how is the difference functioning here?"

As for the example, there's been a clear consensus for months that we needed to add a simpler example to the complex one. You yourself said so. To revert it now that Laurent has added it is very odd. SlimVirgin ^{talk|contribs} 00:00, 30 June 2009 (UTC)

I concur with the change as posted by SlimVirgin, it's a good, simple example and there's clear consensus for adding one. It doesn't need a separate section, and limiting it to "juxtaposition" is incorrect. Dreadstar † 00:39, 30 June 2009 (UTC)

You shouldn't have edit warred to get it in with your recent revert, but wait for a consensus on this talk page. --Bob K31416 (talk) 01:08, 30 June 2009 (UTC)

AFAICT, there is consensus for the addition, you seem to be the only one really disputing it. As for for edit warring, please see WP:KETTLE: [30][31]. Dreadstar † 01:30, 30 June 2009 (UTC)

Re "What do you see as the difference, in this context, between a synthesis of sentences and a juxtaposition?" - I think we are in agreement on this, so I'm not sure of the purpose of your question. Anyhow, synthesis of sentences is a more general concept that can occur wherever the sentences are, juxtaposed or not, and is usually followed by a conclusion, or the sentences may carry with them an implication. Juxtaposition is an effective way of making an implication without giving a conclusion. Is that what you were asking?

Re "there's been a clear consensus for months that we needed to add a simpler example to the complex one" - OK but like I said in my previous message, the one you're proposing doesn't fit the bill for the reasons I mentioned. Take a survey and see what others think so that we can see if your proposed addition has a consensus. Good luck. --Bob K31416 (talk) 00:44, 30 June 2009 (UTC)

I can't see what your objection is, in that case. You said that the word "juxtaposition" was important to this example. But it is just a subclass of synthesis, and I'm not sure the example added by Laurent is an example of juxtaposition. I don't see it as that, anyway. So could you explain your objection? SlimVirgin ^{talk|contribs} 00:55, 30 June 2009 (UTC)

I referred to my objection in a previous message, which so far you haven't addressed. Also, I fixed the bisecting you did of my last message. Are you sure you want to behave in these disruptive ways? Please note that I don't carry on discussions with disruptive editors. --Bob K31416 (talk) 01:08, 30 June 2009 (UTC)

Maybe I'm just fuzzy headed today, but I also can't really tell what your objections are Bob, can you please outline them for us? And please stop commenting on the other editors, no one is being disruptive here. Dreadstar † 01:12, 30 June 2009 (UTC)

Actually I feel that "original synthesis" should be limited to explicit conclusions, and that implied conclusions are really a subclass of original essay, not of logical synthesis. Basically there's three kinds of original research: There's simply publishing original information; there's original synthesis, which is publishing facts deduced from cited sources; and there's original essay, which is arguing an original position from cited sources. I think the implied o.r. by juxtaposition in the U.N. example is more akin to writing a POV essay than it is to a logical synthesis.

The reason I feel so strongly about keeping "soft" o.r. out of the synthesis section is because there needs to be a strong concept about what is and is not synthesis so that synthesis actually means something in a debate. To me an original synthesis is either an interpretation that adds some unique art or judgement beyond what is contained in the sources ( for instance a graph of stock prices which could be interpreted more than one way), or a deduced fact which may not be true under the same conditions as the cited facts ( just because X is considered to be a Y, and Y's generally have a Z, doesn't mean we should say X has a Z ). We run into trouble here because most real life examples are not "Socrates is a man", but "a tomato is a fruit" where we're dealing with fuzzy and not Boolean boundaries.

By the same token, simple arithmetic, translations, conversion between anglo and metric units, and some very simple deductions are not o.r. because there's only one way to add two numbers or convert feet to meters; no original facts are created. By limiting the syntheis section to explicit logical synthesis, synthesis questions can be settled by simple logical debate and explaining how the conclusion C might be incorrect even if the supporting facts A and B are rock-solid. The more subjective questions about implied facts belong in the more subjective "original essay" section.

To get back to Bob about juxtaposition, no, it wasn't that particular debate, which was a doozy, where I was actually taking the deletionist side and arguing that in an article on raw foodism there was something very strongly implied, and unlikely, about human evolution that wasn't in the cited sources. But I wasn't making an o.r. argument, I just said there was a problem with the wording.

What I was thinking of was a series of discussions with Jayjg ( I wasn't involved in all of them, and some were on article talk pages ), where he would argue that if an article is titled 'X', every source cited in the article must specifically mention 'X'. Which sounds fine, unless you're dealing with merged articles or articles that need simple background information. For example, an article about an experimental aircraft that uses some unique fuel which we don't have an article about, might spend a paragraph talking about the properties of the fuel and which sources might not mention the aircraft.

While suggesting that each source must mention the subject title is a useful "Occam's Razor" type rule of thumb for keeping the articles in focus, especially in articles about Israel and Zionism that are extremely prone to POV-forking, that's handled by the WP:RELEVANCE essay and shouldn't be added to the policy level as a fourth type of OR, by juxtaposition of citations.

For instance, I once had trouble citing an article about licorice candy flavoring in an article on jellybeans (the reason was that most non-gourmet "licorice" candies in the U.S. aren't flavored with licorice or any facsimile thereof, but are actually natural or artificial anise flavorings); it was considered "original research" because the article didn't specifically mention bean-shaped sweets(!) Something like that might be a tangent, but please don't call it original research if it doesn't actually introduce new facts. Squidfryerchef (talk) 14:32, 1 July 2009 (UTC)

Do you oppose Slim Virgin's recent addition to WP:SYN that was put in again by Dreadstar? --Bob K31416 (talk) 01:56, 2 July 2009 (UTC)

I agree with most of what Squidfryerchef wrote in his/her last posting to this section. "Socrates was a man". "A tomato is a fruit." is a juxtaposition one might conclude that Socrates is not a tomato, but I do not think that it is SYN unless someone adds to an article: "Socrates was not a tomato", unless they have a source that says so, (and even if no source is presented, if it is a conclusion that everyone would draw I think it should come under a simple logic provision similar to that of "Routine calculations", as it would stop people using this section to avoid introducing a NPOV to a subject for which the majority of sources are skewed in one direction. For example the there are probably enough sources claiming that the last Israeli invasion of Gaza was a genocide,[32], and if the subject heading is drawn up narrowly enough there may not be any direct refutation of that potion for there to be a balanced article. In Wikipedia we assume silence means consent, but in scientific disciplines silence does not mean acceptance (it is the number of times an article is positively cited by a scientific discipline which indicates acceptance) and there is a danger that the Wikipedia OR policy can be used to present scientific fringe views as majority views (hence breaching NPOV), either because the general public (most of us) do not have access to and do not know about specific expert publications, or because the view was not thought worth a reply by the majority of scholars in a particular field.

A couple of my lectures used to use "Time flies like an arrow, [but] Fruit flies like bananas" are we going to start to have a policy on syntactic ambiguity? If so we need to discuss it fully and not add it as an extension to this section. If we are going to go that way then we need to include WP:WTA which has started to stray into areas of rhetoric. I think that it may well be time for an essay on how rhetoric, grammatical tricks, and other types of presentation, (whoops a tricolon) can be misused to get over a point, (Eg it could be argued that the example given in this section is a breach of WP:WTA because using "claim" is to as strong as "stated" and introduces a POV. Equally putting smith first introduces a POV because it makes a statement that is then refuted, when the sentences could be constructed to put Jone's POV first.

Most people do not read the polices every day to see what has or has not changed since yesterday. version at the start of the year, version about a month ago. So in an argument over whether to use "Juxtaposing facts in a way that advances a position", or not, as the wording has only been there for the blink of an eye, as people have been reverting it out, I think it is up to Bob K31416, to show that is is the consensus view. I also think that the examples with the UN are not helpful, as they the word "only" is covered by NPOV.

One of the points about Wikipedia policies is that not only must they reflect the opinions of most Wikipedians but they need to reflect it in a way that is clear and easy to interpret so that time is not wasted in long debates over angels on pinheads. For example what precisely does "Juxtaposing" mean? If "The UN's stated objective is to maintain international peace and security.^[1] Since its creation there have been 160 wars throughout the world.^[2] is "Juxtaposing", what about "The UN's stated objective is to maintain international peace and security.^[1] The UN is now over 50 years old.^[2] Since the UN's creation there have been 160 wars throughout the world.^[3]" is that "Juxtaposing". Further the statement "Although no conclusion is drawn, and although both statements are individually verifiable, their juxtaposition implies that the UN has failed at maintaining world peace." is making an assumption, that just because one editor infers something all readers will infer the same thing (if they do then it probably comes under routine logic as with "routine calculations"). As no comparison is made for other periods in history, no inference can be drawn, and personally I do not infer the conclusion forced upon me by the passage in "Juxtaposing", because I am reading this above ground and not as a troglodyte in a nuclear fall out shelter.

I think the more complicated example is a much better example because it is not the juxtaposition, which is syn but the specific extrapolation in the sentences that start "If Jones did not consult the original sources ..." because those specific sentences are OR. So I think we should go back to the section as it was at the start of last month and drop both extensions to the section. --PBS (talk) 10:22, 2 July 2009 (UTC)

The issue of "juxtaposition" is a red herring here, because Laurent's examples aren't examples of juxtaposition. The key to SYN, and they are clearly examples of SYN, is that they constitute sourced material combined in such a way as to advance a position not advanced by the sources. SlimVirgin ^{talk|contribs} 10:26, 2 July 2009 (UTC)

In my previous long ramble, I never said it was just "juxtaposition". I don't think the examples help with explaining what SYN is, and it is straying into areas to do with summary style. If we go to far in this direction we push editors into creating articles that are nothing but a string of quotes and disjointed sentences, with problems over plagiarism. --PBS (talk) 10:53, 2 July 2009 (UTC)

That's true, but it's a question of knowing how to write, how to summarize. We used to be taught this in school; they called it "composition," I think. I don't know whether it's taught nowadays. But the point of SYN is that it only excludes material that advances a position not advanced by the sources. Regular summarizing is what we do all the time. SlimVirgin ^{talk|contribs} 12:22, 2 July 2009 (UTC)

SlimVirgin, It's clear that both versions are disputed. Only your version is in the article. Perhaps you will now agree that your version should be removed so that neither of these new additions are in WP:NOR until consensus is reached? Thus, I've changed WP:NOR to the state without either of these new additions. Please note that the burden of proof should lie with the editor that is proposing new additions to WP:NOR and consensus is needed. Thanks. --Bob K31416 (talk) 14:51, 2 July 2009 (UTC)

If the problem is the implicit conclusion, why don't we make it explicit? For example:

"The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world. Thus, it has failed at maintaining world peace."

Statement A and B are sourced, statement C is original research. Additionally, perhaps we could mention that it doesn't matter whether the conclusion is explicit or implicit (as it was in the original edit), as in both cases it is considered original research. I also think that we need a simpler example in WP:SYNTH and, in my opinion, the UN one is much easier to understand that the current example. Laurent (talk) 15:57, 2 July 2009 (UTC)

We're in danger of getting our knickers in a twist with talk of implicit, explicit, conclusions, implications, juxtapositions, syllogisms, consequences, inferences, entailments, deductions, interpretations, and what have you. Let's keep it simple. The point is that SYN is not allowed if it advances a position not advanced by the sources. We need an easy example, which everyone gets, and a more complex one of the kind people stumble over. Thanks to Laurent, we now have both. SlimVirgin ^{talk|contribs} 16:24, 2 July 2009 (UTC)

SlimVirgin, You seem to be persisting with the notion that your version is the same as Laurent's, which isn't true as I indicated previously. One of the significant problems with your version is that you are representing it as a simple version of synthesis, which from this discussion it should be evident to you that it isn't. --Bob K31416 (talk) 16:36, 2 July 2009 (UTC)

[edit conflict]P.S. We should consider Laurent's most recent suggestion, in the message previous to SlimVirgin's. Without the additional comment regarding "implicit", it would be a simple example of synthesis. --Bob K31416 (talk) 17:07, 2 July 2009 (UTC)

But it is a simple example of SYN. What is complex about it? SlimVirgin ^{talk|contribs} 17:00, 2 July 2009 (UTC)

Because the C in A+B=C is not obvious without the reader being told it is there. In Laurent's most recent example, it is obvious. --Bob K31416 (talk) 17:14, 2 July 2009 (UTC)

His first example wasn't obvious? It couldn't be more obvious. SlimVirgin ^{talk|contribs} 17:16, 2 July 2009 (UTC)

<outdent>SlimVirgin, Perhaps this excerpt from PBS's message, which you may have overlooked previously, might help you understand.

Although no conclusion is drawn, and although both statements are individually verifiable, their juxtaposition implies that the UN has failed at maintaining world peace." is making an assumption, that just because one editor infers something all readers will infer the same thing (if they do then it probably comes under routine logic as with "routine calculations"). As no comparison is made for other periods in history, no inference can be drawn, and personally I do not infer the conclusion forced upon me by the passage...

--Bob K31416 (talk) 17:27, 2 July 2009 (UTC)

There is no juxtaposition, so I don't understand that comment, and PBS said he didn't mention juxtaposition, so I'm further confused. I also don't know what "I do not infer the conclusion forced upon me" means. I think people here are perhaps talking about different issues. Bob, could you say, in your own words, how the example isn't an obvious example of SYN? SlimVirgin ^{talk|contribs} 19:34, 2 July 2009 (UTC)

arbitrary break

It isn't a simple example of syn because the conclusion isn't stated. Here's your example.

The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world.

This can be compared with Laurent's recent suggestion which is a simple example because it states the conclusion.

The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world. Thus, it has failed at maintaining world peace.

--Bob K31416 (talk) 22:18, 2 July 2009 (UTC)

So Laurent isn't objecting to the addition; he would just like to tweak the writing. I think it's better before with the hidden conclusion, which is really very obvious. If someone can't see the conclusion in that, they'd probably have difficulty doing any kind of editing. I suppose we could add in brackets what the hidden conclusion is and spell it out, but my preference would be to leave it as it is to avoid unnecessary wordiness. SlimVirgin ^{talk|contribs} 05:40, 3 July 2009 (UTC)

Please note that with your version there is more than one possible conclusion, for example:

1. the UN made the situation worse
2. the UN was ineffective and neither helped nor hurt the situation
3. the UN helped the situation but not much
4. the UN helped the situation but couldn't remove all the little wars
5. the UN failed at maintaining world peace
6. it's impossible for wars to be eliminated
7. no conclusion

Thus adding a conclusion in brackets won't work because you don't know what would be in the reader's mind. Also, just having to use brackets makes the example more complex. Laurent's example is simple because the conclusion is stated. --Bob K31416 (talk) 10:29, 3 July 2009 (UTC)

Based on the previous discussion, I've incorporated a simple example into WP:SYN. It is based on the most recent example of Laurent and it seems to be acceptable. But if it isn't, I'm sure I'll find that out. If someone does revert it, please discuss it here. Thanks. --Bob K31416 (talk) 03:09, 4 July 2009 (UTC)

I reverted because I couldn't see how your edit improved it, and also because you seem to have done more than just tweak the example. SlimVirgin ^{talk|contribs} 03:13, 4 July 2009 (UTC)

This is a response to your removal of my recent addition and the simultaneous installation of your addition again, for the 3rd time, which you didn't mention either here or in your edit summary. You did not respond to my previous message regarding the problems with your suggestion, but instead you installed your previous version again. I haven't reverted it, even though I've identified the problems with it in a number of my previous messages and I don't think your behavior was proper. You shouldn't edit war to get your addition into WP:SYN.

I recently put in a simple example (based on Laurent's suggested example), and you removed it. In your edit summary you wrote, "the writing isn't as good in places, and there seems to have been more changed that one example being added".

What improvement would you suggest? Here's what I put in.

Here is a simple example:

The UN's stated objective is to maintain international peace and security,^[23] but since its creation there have been 160 wars throughout the world.^[24] Thus, it has failed at maintaining world peace.

Both of the facts in the first sentence are from reliable sources. These facts were combined to form the conclusion in the second sentence. If this conclusion does not appear in a reliable source, it is a synthesis that is original research and can not be included in the Wikipedia.

And for reference, here is the example that Laurent suggested.

The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world. Thus, it has failed at maintaining world peace.

BTW, you really should remove your addition that you've put in for the 3rd time. It's been disputed above and it certainly doesn't have consensus.

--Bob K31416 (talk) 04:06, 4 July 2009 (UTC)

I don't see who is disputing it. People seem to be talking about different things. Will look at the rest of your post later and respond. SlimVirgin ^{talk|contribs} 04:24, 4 July 2009 (UTC)

Your addition certainly isn’t juxtaposition Bob, it is first and foremost a blatant violation of WP:V – which only serves to add confusion about the purpose of this Policy; beyond that it is purely OR synthesis, of which we already have a good example. We’re trying to find a new example, specifically a simple one that shows a synthesis of sentences….your addition is not that. And if you think there’s any objection to SlimVirgin’s addition besides your own singular objection, then I suggest you prove it. Dreadstar † 04:46, 4 July 2009 (UTC)

Informal survey

Please express whether you support or oppose SlimVirgin's addition.

The new example:

A simple example:

The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world.

Although no conclusion is drawn and both facts are true, the sentence implies that the UN has failed to maintain world peace. If no reliable source has combined the material in this way, it constitutes original research. It would be easy to imply the opposite using the same material, illustrating how, when no source is provided, facts can easily be manipulated:

The UN's stated objective is to maintain international peace and security, and since its creation there have been only 160 wars throughout the world.

Poll responses

• oppose - It misrepresents the example as one that is simple. It involves the notion of implication which is controversial and not simple, as described in previous discussions. --Bob K31416 (talk) 04:35, 4 July 2009 (UTC)

• Support SlimVirgin's addition, disagree with Bob's proposals. Dreadstar † 04:47, 4 July 2009 (UTC)

Please note that this is a survey on SlimVirgin's addition, which has just recently been put into the wiki. --Bob K31416 (talk) 04:57, 4 July 2009 (UTC) )

Yeah, I got that, Bob, but we're certainly free to express our opinions on your counter-proposal, and I'm thinking it's a good idea to include both. Dreadstar † 05:39, 4 July 2009 (UTC)

• Oppose The language If no reliable source has combined the material in this way, it constitutes original research is very problematic. Besides codifying the controversial idea of "orignal research by juxtaposition" in policy, editors most definitely will use it to disallow any combination of citations they don't like, whether it promotes a particular thesis or not. Squidfryerchef (talk) 15:44, 4 July 2009 (UTC)

Hi Squidfryerchef, one of the concerns that Slim's version addresses is exactly that: it does not codify OR by juxtaposition; instead, it is an example of OR by a synthesis of sentences, something that already violates WP:NOR policy, you cannot utilize a sentence from a source that is presented in such a way that in combination with another sentence is clearly meant to advance a position by combining two differently sourced statements. It's purely an example of what is already a violation of WP:NOR policy, it doesn't create anything new, nor does it narrow the focus down to juxtaposition nor does it allow editors to disallow any combination of citations they don't like, they can only do that if a combination of sentences advances a position as does Slim's example. Dreadstar † 18:37, 4 July 2009 (UTC)

Re Dreadstar's comment, "nor does it allow editors to disallow any combination of citations they don't like, they can only do that if a combination of sentences advances a position as does Slim's example." What is the safeguard against this? It seems that it is a matter of interpretation since the conclusion isn't clearly stated. In my previous message of 10:29, 3 July 2009 in the arbitrary break section above, I gave examples of how all sorts of conclusions can be drawn when a conclusion isn't stated. In the same way, someone can say that two facts imply a conclusion and delete one of the facts as a violation, according to SlimVirgin's new example in WP:SYN, by contriving an unstated conclusion. --Bob K31416 (talk) 12:27, 7 July 2009 (UTC)

Wikipedia Processes safeguard this, Bob. Simply stating a conclusion in the policy example is not going to clarify this, only consensus and following the dispute resolution process will find an ultimate resolution for any disputes over synthesis. As I've said before, a stated conclusion that's not supported by the sources would first and foremost violate WP:V and may not even necessarily be a synthesis. Dreadstar † 02:52, 10 July 2009 (UTC)

• Oppose because the second of the two recently introduced example sentences only works if the word only is included. As I said above this is starting to stray into questions of rhetoric, and the style used to summarise sources (with articles consisting of nothing but a strings of quotes and disjointed sentences). --PBS (talk) 13:59, 6 July 2009 (UTC)

Utilizing a conjunction such as "only" is a perfectly legitimate, clear and simple means to illustrate a synthesis of sentences. Dreadstar † 02:52, 10 July 2009 (UTC)

But it is not clear because the whole example hangs on that, and one has to be quite familiar with the sublities of SYN to spot it. I think that the more complicated example is better. --PBS (talk) 11:15, 12 July 2009 (UTC)

• Support - only minor objection I have is that "simple" should be removed. Second example is more complex but this example probably shouldn't be termed as simple. -- Vision Thing -- 11:02, 7 July 2009 (UTC)
• Polls are evil - (somebody had to say it) Blueboar (talk) 12:36, 7 July 2009 (UTC)

Please note that it had been previously asserted by SlimVirgin and Dreadstar that there wasn't any opposition to the new example. This poll has been very useful in correcting that misunderstanding. --Bob K31416 (talk) 12:42, 7 July 2009 (UTC)

Sorry Bob, but not only was there no "misunderstanding" in the first place, you'll note that I also mentioned Consensus, which is not a numbers game, but centers on reasoning. AFAICT, consensus is still not against the addition. There appear to be several misunderstandings in the poll itself. In any case, the rest of us have gone on to discuss the example, I suggest you do the same and quit making comments about other editors. Dreadstar † 02:52, 10 July 2009 (UTC)

Dreadstar, Here's what SlimVirgin wrote, "I don't see who is disputing it. People seem to be talking about different things. Will look at the rest of your post later and respond. " [33]

And here's what you wrote, just after SlimVirgin's message, "...And if you think there’s any objection to SlimVirgin’s addition besides your own singular objection, then I suggest you prove it. " [34]

Seems like the poll proved there are other editors objecting. Regards, --Bob K31416 (talk) 06:24, 10 July 2009 (UTC)

Bob, while technically true (your poll does indeed demonstrate that another editor besides yourself opposes SV's addition... in fact, it shows that a total of two other editors do so), please don't use your poll to draw any conclusions as to consensus. Only four other people have responded to your poll (not counting my "Polls are evil" comment). This is not enough to determine a consensus one way or the other. A poll of five people is hardly an accurate reflection of the views of the broader community. Blueboar (talk) 13:32, 10 July 2009 (UTC)

But it does not show that there is a clear consensus for the addition. So I am surprised to see that it has been added, when the subject is still under discussion. --PBS (talk) 11:15, 12 July 2009 (UTC)

The question still remains of why an example with opposition and no clear consensus was able to get into WP:NOR. --Bob K31416 (talk) 12:48, 17 July 2009 (UTC)

• Oppose per Squidfryerchef and PBS. The addition does indeed codify bad precedent for defining what OR is. If something is advancing an argument unsupported by sources, it is NPOV and leave it at that. older ≠ wiser 12:25, 12 July 2009 (UTC)

older/wiser... if something is advancing an argument unsupported by sources... that is the very definition of Original Resarch as stated in the very first paragraph of the Policy ("Wikipedia does not publish original research or original thought. This includes unpublished facts, arguments, speculation, and ideas; and any unpublished analysis or synthesis of published material that serves to advance a position"). No new codification. Blueboar (talk) 12:59, 17 July 2009 (UTC)

Is the burden of proof on new additions?

If a new addition to WP:NOR is removed, is the burden of proof on the new addition to show that it has consensus before it is restored? Comments are invited. Thanks. --Bob K31416 (talk) 09:11, 6 July 2009 (UTC)

Short answer is Yes... any change or removal requires consensus. But that leaves open the question as to what constitutes "proof" of consensus. In most policy content disputes, both sides will be able to "prove" that their view has conseneus. Blueboar (talk) 12:50, 7 July 2009 (UTC)

If a poll is taken, and the majority oppose the addition, or it is about evenly split, wouldn't the addition lack proof of consensus, and hence should not be restored if it is removed? --Bob K31416 (talk) 13:22, 7 July 2009 (UTC)

Yes and No... Poll's are evil, especially on policy pages because a Poll only reflects the opinions of a few editors (those who are currently watching the policy talk page and enguaging in the discussions)... policy pages work on a much larger consensus (that of the entire Wikipedia community ... something that is much more difficult to determine). I have seen changes made to policy pages with full consensus of the five or six editors actively involved in the talk page discussions at the time, only to have the change negated a week later when some editor who was not involved in the discussion comes along and discovers the change. Blueboar (talk) 13:59, 7 July 2009 (UTC)

Please note that this section is about not restoring new additions without consensus that have been removed, rather than putting new additions into policy. --Bob K31416 (talk) 14:07, 7 July 2009 (UTC)

Yes, I know. This is why I answered above with "Yes". It is clear from the limited Poll you ran above that there was mixed reaction to the recent addition among the few people who expressed an opinion, with some expressing approval and others expressing disapproval. The problem I am pointing out is that we can not determine whether this limited poll reflects a wider consensus. It may... it may not (I have no opinion on that).

Another problem is that multiple steps that have been taken without a clear demonstration of consensus... Was there consenus for the initial addition? Was there consensus for removing it? Was there consensus for returning it? Is there now consensus for undoing that return? My point is that a limited poll on the talk page will not determine the consensus for or against any of these steps... as it only relfects the opinions of a few people and not the wider community. May I suggest that we slow down a bit... go back to square one and give more people some time to look at the initial issue. Blueboar (talk) 14:56, 7 July 2009 (UTC)

Regarding the example that you brought up, you mentioned that there is uncertainty whether there was a consensus for the recent addition, and that there is uncertainty whether there is consensus for removing the recent addition. By giving the same status to the recent addition and its removal, you're clearly advocating that the burden of proof does not lie with a new addition. Is this how you feel all new additions should be treated, i.e., when there is uncertainty regarding consensus, the burden of proof does not lie with the new addition? --Bob K31416 (talk) 16:47, 7 July 2009 (UTC)

No, I take each addition seperately and don't make generalizations. Blueboar (talk) 17:12, 7 July 2009 (UTC)

So you feel that sometimes when there is a new addition, and it is uncertain if the new edition has consensus, it is sometimes justified to restore the new addition after it has been removed, even if consensus for the new addition is uncertain? --Bob K31416 (talk) 18:14, 7 July 2009 (UTC)

Yup... Sometimes yes, sometimes no... My real point here is that a poll rarely establishes whether there is consensus for any addition/removal or not. Polls simply reflect the opinions of those who happened to look on the talk page at a particular moment in time and respond to the poll. I am willing to disreguard poll results if I am confident that an addition reflects or does not reflect my understanding of the general consensus... dispite what the discussions on the talk page among a few editors and a snap poll may indicate. Of course, others are free to disagree with me. Writing Wikipedia policy is a very messy process. Blueboar (talk) 18:48, 7 July 2009 (UTC)

All that your last message has added to the discussion is how consensus polls can be uncertain, not how a new addition can be allowed if consensus is uncertain, which is what you are advocating. Well, thanks for the discussion. I think we've reached the end of it between you and me. --Bob K31416 (talk) 19:00, 7 July 2009 (UTC)

OK... but before I go, let me clarify ... I was purposely not addressing your original question (whether a new addition should be allowed if consensus is uncertain). Instead, I was trying to point out that I think the question itself is flawed, because consensus is rarely certain when it comes to policy pages... and hardly ever determined by a poll. Instead of conducting polls, I think people should make good solid arguments for and against the change/revert/return and apply their understanding of pre-existing consensus to those arguments. Numbers don't matter... well thought out arguments do matter. Thus, I am quite willing to allow a change which is well explained on the talk page, even if a snap poll contains a majority of poorly explained votes against it. I would also favor a revert that was well explained in the face of lots of poorly explained poll votes.

I have been away for two weeks and I am slowly catching up on the discussions that lead to your poll, and thus to this thread... I have not yet made up my mind as to the underlying issue (I need to read more). I think you are right, however, in ending this discussion. It does not really help us determine if the change should stay or not. Blueboar (talk) 21:02, 7 July 2009 (UTC)

I only ended my discussion with you because it seemed to become repetitive and appeared to have run its course. Surprisingly, you came up with some new ideas afterwards. Good!

Let's consider your subjective approach of determining consensus. Suppose you use your criteria and determine that there is no consensus for a new addition to a policy page. You revert the addition, and the editor reverts it back, saying that there is consensus. then you give excellent and correct arguments and the editor says NO, there is consensus for it and my addition stays. If there is no quantitative way to evaluate whether there is consensus, like a poll, what do you do? --Bob K31416 (talk) 21:24, 8 July 2009 (UTC)

LOL... welcome to the joys of policy writing. You have several options... you can edit war (not recommended, but it is an option)... or you can continue the discussion and try to convince people that your version is the best way to go through qualitative means (ie through quality arguments). Unless you think that the other version is seriously harming Wikipedia, relax and don't worry about which version is currently in the policy... continue the discussion. Another option is to leave the change, but tweek it so it is better. Blueboar (talk) 21:39, 8 July 2009 (UTC)

Thanks for the good humor! Actually, it's not a matter of me having a version. It's a matter of not wanting a version that is being forced into policy without first getting consensus. When a new addition is opposed and reverted, it seems that it should be required to have consensus before it is restored. In the specific case of the new addition to WP:SYN, there are three editors who wrote Oppose and gave reasons and were involved in extensive discussions related to the new addition. Prior to that poll, it appeared to me that those editors expressed their opposition to the new addition in their extensive comments, but the two editors that were restoring the new addition said there was no opposition, at least none other than me. So I took a poll, not to establish consensus, but to show that there was opposition to the new addition. What would you do if you were in a situation like we have now, where you felt there wasn't consensus for a new addition but it was being edit warred into policy anyhow? --Bob K31416 (talk) 22:46, 8 July 2009 (UTC)

I would continue to argue about it on the talk page, re-expressing repeatedly in as many different ways as I could why I think the addition is a mistake and should be removed. I would engage the person who feels that the addition has consensus, and try to convince them that it is not a good edit even if it does have consensus.

I think we are getting sidetracked and it is time to stop our discussion... lets go back to discussing what is the best wording for the policy. Blueboar (talk) 23:27, 8 July 2009 (UTC)

Thanks for your response. --Bob K31416 (talk) 23:35, 8 July 2009 (UTC)

One more question regarding your approach to determining consensus. In your opinion, does the recent addition of the part identified as "A simple example" in WP:SYN have consensus? Thanks. --Bob K31416 (talk) 05:25, 9 July 2009 (UTC)

An excellent question that moves us away from bickering about process and towards returning to a discussion about substance. I am going to answer in two parts... 1) For over a year, multiple editors (myself among them) have been calling for a simpler example in the SYNT section. So I would say that adding a simple example does indeed have a clear consensus. 2) Given this, the question narrows to whether the specific example that was chosen has consensus. That is less clear, but also (to my mind) less important. I don't think it harms the policy to have it, so I would lean towards keeping it in while further discussions take place.

Personally I like this example, as it addresses something that I have long seen as being unaddressed (ie that a synthesis with an unstated/implied conclusion is never the less a form of synthesis). Blueboar (talk) 12:45, 9 July 2009 (UTC)

Re your item 1, I don't agree that this is a simple example for the reasons I mentioned previously in the poll and before. Even one of the people in the poll who supports the addition believes that the phrase "simple example" should be removed.

Re your item 2 and the part "I would lean towards keeping it in while further discussions take place." - LOL, isn't that natural for anyone who supports an addition, and isn't it just as natural for those who are leaning the other way or are neutral, to want to keep it out until consensus is reached, to preserve the stability and status quo of a policy page? It seems like Wikipedia favors some stability, with principles like Bold, Revert, Discuss WP:BRD, rather than Bold, Revert, Revert back, Discuss. But I appreciate your opinion and I note your objectivity when you remarked that consensus for the new addition isn't clear, even though you are leaning in favor of the addition. --Bob K31416 (talk) 13:10, 9 July 2009 (UTC)

Sigh... that was an attempt to shift the discussion back to the merits of the addition rather than discussing the process by which it was added. I hoped that if I stated that I liked the addition and why I liked it you might engage me in discussing why you don't like it. I will make one more attempt... Forget who added, reverted, unreverted, etc. etc. Look to the future rather than the recent past... Pretend that the example has been in the policy for years... now... convince me (and others) that it should be removed. Blueboar (talk) 13:54, 9 July 2009 (UTC)

Re "I hoped that if I stated that I liked the addition and why I liked it you might engage me in discussing why you don't like it." - But I answered that! I directed you to my comments in the POLL. Also, the process is important so that policy is made by reason, rather than by force of edit warring. Please read those comments of mine in the poll, respond here, and I will respond back here. OK? --Bob K31416 (talk) 14:16, 9 July 2009 (UTC)

Back to discussing the example and not the process

I'll make it easy for you and copy my POLL comments to here.

It misrepresents the example as one that is simple. It involves the notion of implication which is controversial and not simple, as described in previous discussions."

Re Dreadstar's comment, "nor does it allow editors to disallow any combination of citations they don't like, they can only do that if a combination of sentences advances a position as does Slim's example." What is the safeguard against this? It seems that it is a matter of interpretation since the conclusion isn't clearly stated. In my previous message of 10:29, 3 July 2009 in the arbitrary break section above, I gave examples of how all sorts of conclusions can be drawn when a conclusion isn't stated. In the same way, someone can say that two facts imply a conclusion and delete one of the facts as a violation, according to SlimVirgin's new example in WP:SYN, by contriving an unstated conclusion.

Please note that it had been previously asserted by SlimVirgin and Dreadstar that there wasn't any opposition to the new example. This poll has been very useful in correcting that misunderstanding.

--Bob K31416 (talk) 14:34, 9 July 2009 (UTC)

Bob, thanks for repeating your comment. it does help.

I disagree with you, in that I think the example is indeed simple... but I can understand the argument that it isn't simple enough. I know that someone suggested adding a clearly stated conclusion (something along the lines of "This demonstrates that the UN does not promote world peace"... the final wording may change) so that the example avoids the "unstated/implied conclusion" issue entirely. Would this be an acceptable compromise?

That asked, I do think we will need to address the issue of unstated or clearly implied conclusions. But we can contine to work on that once the immediate issue is resolved. Blueboar (talk) 15:44, 9 July 2009 (UTC)

I would disagree with that, Blueboar, because it would involve adding a third example. The beauty of this one is that it's very simple and shows how easy it is to add hidden conclusions, so easy, in fact, that it could be done inadvertently. It's a very good example for that reason. SlimVirgin ^{talk|contribs} 15:49, 9 July 2009 (UTC)

Blueboar, I agree to your compromise. (I fixed the typo of US instead of UN.) --Bob K31416 (talk) 16:03, 9 July 2009 (UTC)

Hi SV, glad you're back in the conversation. Actually, I think having a third example is exactly what we need. We would start off with very basic example, to illustrate the most obvious form of synt (because there are editors out there, especially newbies, who don't understand even the basic form of synt). Then we would move on to warn against subtler forms of synt... such as implied conclusions (whether inadvertent or not)... and finally we would discuss very complex forms of synt, as illustrated by the old Smith and Jones example. I think this progression from basic, to slightly more complex, to very complex would be a great help to editors. Blueboar (talk) 16:05, 9 July 2009 (UTC)

This is the policy page and it needs to stay fairly tight. I wouldn't mind adding a third example if we needed it, but in my view we really don't. How about Viriditas's suggestion above that we have an NOR examples page? SlimVirgin ^{talk|contribs} 18:37, 9 July 2009 (UTC)

I am hesitant as to the wisdom of adding extra pages (examples, FAQs, etc) to policy pages... they tend not to get watched, and can end up contradicting what is said in the policy.

If the goal is tightness... perhaps we should simply get rid of all examples and just make sure the main text is clear. Not my favorite option, but worth discussing. Personally, I think we can have three examples and maintain tightness. Blueboar (talk) 18:54, 9 July 2009 (UTC)

I normally agree about not having additional pages, but in this case, I'd see it as a welcome addition. I'm not sure it would end up contradicting this page. The problem is not that people think something is OR when it's not. The problem is only ever the other way round, so there's a limit to how far an examples page would be likely to stray. SlimVirgin ^{talk|contribs} 19:01, 9 July 2009 (UTC)

Hmmm... I am not against the idea of an examples page... just hesitant. I certainly won't throw up any road blocks if others feel that having one would be helpful. If we do go that route, I would suggest that we discuss the issue of whether the examples page is intended to be "part of" the NOR policy or not. We don't have to discuss it immediately, but we should discuss it... because that question is going be raised at some point, and we might as well get the debate over with sooner rather than later. Blueboar (talk) 19:41, 9 July 2009 (UTC)

I'm not against an examples page either, and in this case it might be very helpful, because there are obviously many variations on the themes presented by NOR policy - as long as we don't make the examples page actual policy..  :) I do like your proposal Blueboar, but I think such a clearly stated conclusion is not necessarily synthesis, and since it would also blatantly violate WP:V, it might further confuse those trying to utilize or understand NOR policy. SlimVirgin's example is a subtle, yet clear and simple synthesis of sentences, and I think is a good addition. Dreadstar † 01:14, 10 July 2009 (UTC)

I agree that any examples page should not be part of policy, but just a helpful adjunct. Perhaps what we could do is, whenever we see a good example of OR, post it somewhere on this page, and then move the examples to a separate page if we get enough? That way, we could slowly create it over time with the best examples (best as in simplest, and most subtle). SlimVirgin ^{talk|contribs} 01:20, 10 July 2009 (UTC)

For reference, here is the example with Blueboar's suggested second sentence:

The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world. This demonstrates that the UN does not promote world peace.

--Bob K31416 (talk) 02:55, 10 July 2009 (UTC)

In addition to being a general example of synthesis, "This demonstrates that the UN does not promote world peace," is also a blatant and clear violation of WP:V. Potentially confusing and a tad bit redundant to what is already in NOR Policy. Dreadstar † 03:08, 10 July 2009 (UTC)

This is splitting hairs. The example is: "The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world." And we go on to explain, "Although no conclusion is drawn and both facts are true, the sentence implies that the UN has failed to maintain world peace." There is no need to add yet another sentence to the example to explain it, unless we also want to add "and world peace means people not fighting," and "demonstrates is another word for 'shows'." Anyone who can't get the example as written is going to have difficulty editing an encyclopedia. SlimVirgin ^{talk|contribs} 03:17, 10 July 2009 (UTC)

SV... that is your preferred example. What Bob and Dreadstar are splitting hairs about was my proposed compromise version (with the conclusionary sentence added). Dreadstar... can you come up with another example of a very basic synthesis, one that does not involve WP:V? I am certainly not wedded to using the UN example twice. Blueboar (talk) 13:52, 10 July 2009 (UTC)

I meant that the proposed compromise (I believe originally proposed by Bob, although I find all the posts hard to follow) is splitting hairs. The current example is very easy to understand, and it's the kind of OR that happens a lot. We can't possibly spell out the meaning of every sentence, and every part of every sentence. At some point, we have to assume we're writing for rational adults who understand English. SlimVirgin ^{talk|contribs} 15:23, 10 July 2009 (UTC)

In case it's not clear in all the above, I agree with SlimVirgin on all counts. I think the current examples in the policy are fine, and if we're going to continue adding more then we should create a separate examples page. Dreadstar † 15:31, 10 July 2009 (UTC)

And in case it is not clear... I also agree with SV as to the current example... My point is that I don't think her example is easy enough. Because it involves a second issue (ie clearly implied, unstated conclusions) it is not the best example to start with. A lot of editors (especially new ones) really don't understand the concept of synthesis (at all), and I think an example of very basic synthesis that demonstrates the concept at its most basic level (A + B thus C, where all elements are stated) is needed. Once we have given an example of that, then we can expand on the concept by discussing other, more complex forms of synthesis, such as current one. Blueboar (talk) 15:59, 10 July 2009 (UTC)

Do you have a suggestion for an easier example? Not the proposal about spelling out the hidden conclusion in the current one, but an entirely different one. SlimVirgin ^{talk|contribs} 23:01, 10 July 2009 (UTC)

I would have to think about it to come up with a really good example... but I could come up with a lot of poor examples ... Just off the top of my head: "Bob is a New York State Senator.^{cite A} The New York Times has said that all NY State Senators are certifiable idiots.^{cite B} Thus, Bob is a certifiable idiot." Of course we would need a short paragraph to explain why this is an example of impropper Synt. —Preceding unsigned comment added by Blueboar (talk • contribs) 13:25, 11 July 2009 (UTC)

It seems like the new example has digressed to more of an example of a violation of WP:NPOV, rather than a violation of WP:NOR. It shows how a single word can determine the POV. Here's the two parts of the example with the POV words in bold.

The UN's stated objective is to maintain international peace and security, but since its creation there have been 160 wars throughout the world.

The UN's stated objective is to maintain international peace and security, and since its creation there have been only 160 wars throughout the world.

Why is there this digression to an example that needlessly adds a violation of WP:NPOV?

Perhaps the added POV was put there to make the example seem more improper than it actually is. This suggests that any OR problem in the example is not so clear and an additonal problem was needed to give a clear example of prohibited editing by adding POV with the words but or only.

With that in mind, here's the example without the extra POV words.

The UN's stated objective is to maintain international peace and security, and since its creation there have been 160 wars throughout the world.

Is this a violation of WP:NOR according to the new example? An editor could argue that it is not because it doesn't contain strategically placed words like but or only. Was that the intention of the new example in WP:SYNTH, that the above is not a violation of WP:NOR? Thanks. --Bob K31416 (talk) 11:10, 11 July 2009 (UTC)

Bob... I don't see a problem with including an example that also has NPOV issues. Many (indeed I would argue most) SYNT violations also involve NPOV to some extent. It is one of those areas where NPOV and NOR overlap. The key is that the paragraph that follows the example explains why it is a SYNT issue (and thus a NOR issue). Blueboar (talk) 13:25, 11 July 2009 (UTC)

I’m not sure how much easier Synthesis can get than in the current example already in NOR Policy, if it gets too simple it stands a greater chance of becoming an obvious violation of other policies, such as WP:V. The very problem with synthesis is that it is subtle and isn’t always easy to identify. The "new" example isn't necessarily violating NPOV, the source's actual sentence could really use "but" or "only" in its original wording, but when combining - or "synthesizing" - the two sentences, it then becomes a synthesis of sentences which may have additional the effect of violating NPOV and V, however the primary and key violation is of NOR Synth. We don't want to be too blatant in violating other policies when giving examples that violate synthesis. Or are there actually easier, simpler synth examples that are clearly synth and not confusing with other policy violations? I haven't yet seen anything superior to what we already have in policy. Dreadstar † 16:41, 11 July 2009 (UTC)

Bob's first suggestion above (NY senators) is a syllogism, which arguably is allowed under SYN, one of the few kinds of SYN that's not problematic, so long as the syllogism is a valid one. This is the problem here. We risk introducing yet more complications. The current examples are clear and unproblematic. I'm fine with leaving the policy as it is now. SlimVirgin ^{talk|contribs} 17:46, 11 July 2009 (UTC)

SV... the NY Senators suggestion was mine, not Bob's. The reason why it is more than just a sylogism is that the various parts of the example stem from different sources, neither of which puts two and two together... in other words it is a Synt violation. However, I do agree that the NY Senators example is not the best example (I even said so at the time I posted it). Blueboar (talk) 19:41, 11 July 2009 (UTC)

I appreciate your attempt because we need a simple example of synth with a stated conclusion, for editors who are unfamiliar with this policy and are at the stage where they are trying to understand the basics, before going on to more subtle or complex examples. --Bob K31416 (talk) 12:55, 17 July 2009 (UTC)

arbitrary break - Back to discussing...

None of you responded to the question in the last paragraph of my message regarding the following sentence. Please respond. Thanks.

"The UN's stated objective is to maintain international peace and security, and since its creation there have been 160 wars throughout the world."

Is this a violation of WP:NOR according to the new example?"

--Bob K31416 (talk) 18:11, 11 July 2009 (UTC)

It's SYN unless your source connects these two facts. JN466 18:19, 11 July 2009 (UTC)

Is the unsourced connection that you are referring to, the connection of the the two facts with "and" in the same sentence? --Bob K31416 (talk) 18:45, 11 July 2009 (UTC)

Yes, because it is you "connecting the dots", and not a source. If you connect the dots and see an image, then it is OR. If a reliable source connects the dots in that way, it is a different matter. JN466 19:24, 11 July 2009 (UTC)

How about this?

The UN's stated objective is to maintain international peace and security. Since its creation there have been 160 wars throughout the world.

--Bob K31416 (talk) 19:58, 11 July 2009 (UTC)

Still the same. You're connecting the dots by placing the two sentences next to each other. If there is no source out there that connects those two dots, then our article shouldn't connect them either. You have to check if the best and most reputable sources on the topic connect the dots and follow their approach. JN466 21:01, 11 July 2009 (UTC)

Please note that there is more than one possible conclusion, for example:

1. the UN made the situation worse
2. the UN was ineffective and neither helped nor hurt the situation
3. the UN helped the situation but not much
4. the UN helped the situation but couldn't remove all the little wars
5. the UN failed at maintaining world peace
6. it's impossible for wars to be eliminated
7. no conclusion

This doesn't look like research of any kind. If anything, it's possible POV if the majority of readers might interpret it one way or the other, and it may need to be balanced with more info to make it more neutral.

Also note that these conclusions can be drawn from just the second sentence alone so that it really isn't synthesis anyhow. --Bob K31416 (talk) 03:02, 12 July 2009 (UTC)

The mere fact of putting two facts together suggests that they have a bearing on each other. If secondary sources do not put these two facts together, then neither should we: because putting two facts together which are unrelated in the literature is evidence of a (novel and as yet unpublished analysis) that has gone on in a Wikipedian's mind. If that analysis has not also gone on in the mind of reliably published authors, it has no business being in Wikipedia. JN466 11:59, 12 July 2009 (UTC)

Please note that the implication is contained solely in the second sentence, "Since its creation there have been 160 wars throughout the world." This is not synthesis, it is purely POV using one sourced fact. The first sentence is not necessary for an implication. This is not a case of synthesis where A and B imply C, where A is the first sentence and B is the second sentence. It is a case of POV, where B alone implies C. --Bob K31416 (talk) 12:20, 12 July 2009 (UTC)

If you want to present the two facts together, check that there are reliable sources that also present them together. That's all. Simple. Don't put two facts together that are not put together in reliable sources. Also simple. That is all this is about. JN466 21:52, 12 July 2009 (UTC)

I think that we are agreed that the sentence is a problem. We seem to differ in the appropriate reason for correcting it, SYN violation or NPOV violation. For now, we might as well limit our discussion to the section New example in SYN is not synthesis, so we don't have to repeat ourselves. We can always pick up the discussion here later. OK? (To others following this, please note that the message below about Yogeshwara is an older message.) --Bob K31416 (talk) 03:08, 13 July 2009 (UTC)

Let me give you another example. The above example suffers from being too sensible; I am sure there are sources out there that have made similar statements about the UN and wars. So look at this:

On February 19th, 2009, 3,000 followers of Maitreya Yogeshwara chanted in New York to help the world economy.(source1) The following week, the Dow Jones Index rose by 10%.(source 2, not mentioning Maitreya Yogeshwara)

Do you see the problem? No "and", no "but", just two sentences, but still WP:SYN, unless a reliable source connects these two events. (And by the way, it's also WP:SYN if the Dow Jones Index fell the week after the followers chanted. ;) Unless a reliable source connects those two facts, our article shouldn't connect them either.) JN466 21:07, 11 July 2009 (UTC)

Hi guys, I haven't been following the discussion, but I wrote an essay a while back that might be helpful: WP:NORDR. This might also be helpful:[35] and [36]. Anyways, good luck. --Phenylalanine (talk) 19:19, 11 July 2009 (UTC)

Hi! : ) If you have the time, it might be helpful if you mention how the ideas here compare to what you did. --Bob K31416 (talk) 20:04, 11 July 2009 (UTC)

User:Jayen466 I disagree with you whole position, on this issue. --PBS (talk) 13:48, 13 July 2009 (UTC)

Philip, would you do us the courtesy of explaining why you disagree? Blueboar (talk) 14:03, 13 July 2009 (UTC)

I can but it will repeat what I said earlier in on this page (See above "I agree with most of what Squidfryerchef ..." and "In my previous long ramble..."). --PBS (talk) 16:50, 13 July 2009 (UTC)

OK... I lost track of who said what where. Your pointer to where you said it is enough. Thanks. Blueboar (talk) 17:23, 13 July 2009 (UTC)

Help with novice editor

Resolved

Novice editor Ring Cinema (talk · contribs) does not believe that you can reference films as primary sources to write plot summaries in a purely descriptive manner on the films' Wikipedia articles. Several of us seem unable to explain the relationship of primary sources and plot summaries to him. He thinks that Philosophical Investigations is an example of using a primary source. Can anyone who frequents this talk page take a moment to explain it to him at User talk:Ring Cinema#Guidelines? —Erik (talk • contrib) 19:57, 13 July 2009 (UTC)

Well... that article goes well beyond a basic plot summary written in a descritive manner... so there are OR issues with it. Blueboar (talk) 20:31, 13 July 2009 (UTC)

Feedback for "Historical and scientific accuracies" write-up

Films are mainly works of fiction, and filmmakers sometimes use history or science as the basis of their films. They incorporate these topics in their films in a way that suits their storytelling and filmmaking abilities. Their approaches to incorporating these topics or others' reactions to their approaches can be interwoven in the film article's article body in sections such as the "Production" section and the "Reception" section, respectively. If ample coverage from secondary sources exist about a film's historical or scientific accuracy, editors can pursue a sub-topic sharing such coverage in a section titled "Historical accuracy" or "Scientific accuracy" ("accuracy" being applied as neutral terminology).

Wikipedia's "No original research" policy says about synthesizing, "Do not combine material from multiple sources to reach a conclusion not explicitly stated by any of the sources." Readers and editors should take for granted that there are many ways films conform to, and deviate from, history or science. Analysis should be introduced by reliable published secondary sources that compare the film with history or with science. Avoid listing miscellaneous information about accuracies or inaccuracies whose relevance are not backed by secondary sources. If analysis is limited, links should be provided to historical or scientific articles so readers can read about topics based in reality after reading about the work of fiction that uses these topics with dramatic license.

I'm an editor from WikiProject Films, and we have guidelines to shape film articles. I have been updating the guidelines slowly but surely, and my most recent goal is to add a "Historical and scientific accuracies" section. Above is my write-up for the section, and I was curious to know if frequenters of this page agree with my guidelines. My main question: Is the argument of synthesis properly applied? Editors tend to like challenging the historical or scientific accuracy of a film with their own comparisons between the work of fiction and academic materials (or their own knowledge). For a pretty sloppy example, see Historical deviations in Gladiator (2000 film) with such comparisons (particularly unnecessary since Martin M. Winkler's Gladiator: Film and History is perfect to implement). Can others share their thoughts on my write-up or the argument against these kinds of comparisons? —Erik (talk • contrib) 02:22, 17 July 2009 (UTC)

This looks excellent Erik. About the only thing that I could see being added would be some explanation of how long lists of petty inaccuracies can be seen as a form of WP:TRIVIA, which wikipedia strives to avoid. Blueboar (talk) 02:51, 17 July 2009 (UTC)

Good suggestion! It will discourage simplistic comparisons. Do you think it's fair to say that a reason is that it lacks immediate relevance to the topic when not supported by a secondary source? —Erik (talk • contrib) 00:25, 18 July 2009 (UTC)

No it's simply that a lot of "inaccuracies" are simply trivial. Blueboar (talk) 02:30, 18 July 2009 (UTC)

Informal survey – proposed example for WP:SYN

Members of attacking forces attempted to telephone the residents of buildings, to warn them to leave before they were bombed.^[23] During the war, 90% of the telephone system was down.^[24] Thus the attacking forces were unable to warn 90% of the residents whose buildings were bombed.

Please consider the above example for inclusion in WP:SYN. It is for editors who are unfamiliar with this policy and are at the stage where they are trying to understand the basics, before going on to more subtle or complex examples. Thank you. --Bob K31416 (talk) 20:23, 15 July 2009 (UTC)

• Oppose I'm not sure it works as a simple example because the conclusion seems too arbitrary (one could argue that perhaps all the people who were bombed were in the area where 10% of the telephones were working, so everybody ended up being warned). So I would tend to tag the conclusion with a "fact" template before even considering WP:SYNTH. In other words, I think the example draws the reader's attention on the potentially incorrect conclusion rather than on the juxtaposition of sentences. Laurent (talk) 20:36, 15 July 2009 (UTC)

• Oppose for the reasons given by Laurent. It's not straightforward enough to be a simple example. SlimVirgin ^{talk|contribs} 20:39, 15 July 2009 (UTC)

• Oppose, per Laurent, and for all the reasons outlined in the section above Dreadstar † 20:42, 15 July 2009 (UTC)
• Polls are evil - and polls of four or five people are worthless. Blueboar (talk) 00:03, 16 July 2009 (UTC)

• I'm not keen on the polls either. SlimVirgin ^{talk|contribs} 00:16, 16 July 2009 (UTC)

• Poll? What poll? Is there a poll? We don't need no stinkin' polls!  :) Dreadstar † 01:04, 16 July 2009 (UTC)

• Looks like someone we know and love participated in a poll here, and may have even started it by writing Oppose. : ) --Bob K31416 (talk) 02:10, 16 July 2009 (UTC)

OOPS... sorry... I should have said: Informal Surveys are serious sins... (I never said I don't participate in polls by the way... just that they were evil and worthless in small numbers... Hell, I've even proposed the occasional poll... but they are still evil... EEEVILLL I tell you!) :>)

Seriously... it was fairly clear how each of us felt before this survey was proposed. What we need are a whole bunch of outside opinions. Instead of polling the same few people, may I suggest an RFC or a post at the Village Pump, something to get more people take a look at the issue? Blueboar (talk) 02:49, 16 July 2009 (UTC)

• It is a good and valid example of synthesis, but a bad example to write in the policy. JN466 21:30, 16 July 2009 (UTC)

Comments

• Re Laurent's remarks - Too arbitrary? I disagree. However the potential for inaccuracy is a typical problem with OR. That's why statements like that have to be sourced. Please note that a source could just as well have stated the same thing as fact. Also note that tagging with a {{fact}} template, which results in , is what one could do for suspected OR in general, since a source isn't given for an OR statement. --Bob K31416 (talk) 21:18, 15 July 2009 (UTC)

• Here's a nice example of synthesis from the article Let's trim our hair in accordance with the socialist lifestylediff:

Let's trim our hair in accordance with the socialist lifestyle was part of a North Korean government propaganda campaign promulgating grooming and dress standards. It was broadcast on the state-run Korean Central Television in the capital of Pyongyang. The television program claimed that hair length can affect human intelligence, in part because of the deprivation to the rest of the body of nutrients required for hair to grow. This is at odds with the mainstream medical understanding that hair growth is unaffected by the amount of previously grown hair that remains attached to the body.

After about 10 minutes of google searching, I was unable to find any specific rebuttal of the campaign message. It appears that media outlets take it for granted that readers know that the claim is false. Hope this is helpful. Cheers, Phenylalanine (talk) 21:15, 16 July 2009 (UTC)

It's a good one, thanks. I've added it to the new examples page. SlimVirgin ^{talk|contribs} 18:05, 17 July 2009 (UTC)

Examples page

I've gone ahead and created an OR examples page: Wikipedia:No original research/Examples, input and examples are welcome! Dreadstar † 17:57, 17 July 2009 (UTC)

It's looking nice but I wonder if the intro sentence "The violation of SYN is in bold" is appropriate. I think it makes it looks more simple than it is, and it may mislead readers into thinking that if they remove the sentence in bold, there's no more a SYN violation. However, whether the conclusion is explicit or implicit is irrelevant. What matters is that no sources combine the sentences in the way it is done there. Perhaps, rather than bolding the sentences, we could add a short description to each example and explain why it's a synthesis? Laurent (talk) 18:21, 17 July 2009 (UTC)

It's just a starter, so please feel free to work on it! Why don't we all collaborate on the examples talk page: WT:NOREX. Dreadstar † 18:36, 17 July 2009 (UTC)

Great job guys! Could we make the link more prominent, maybe add it in the Synthesis section? --Phenylalanine (talk) 19:19, 25 July 2009 (UTC)

I have added in the two examples currently given in the text of the policy. The creation of this sub-page raises a new possibility... we could cut the current examples from the policy text entirely, and instead state:

• "For some examples of synthetic statements, and explanation of why they are original research, see: Wikipedia:No original research/Examples."

I am of two minds on this... what do others think? Blueboar (talk) 19:47, 25 July 2009 (UTC)

I don't mind moving the link to the Synthesis section, but I'd like to keep a couple of main examples of synth right in the policy. Unlike subpages, main Policy pages are watched closely so there's less chance of having poor or misleading examples added. I also think it's good to have a couple of examples right in the policy, helping give direct substance to the concepts outlined therein - subpages can help expand on those, but aren't necessary to the core understanding of the policy. Dreadstar † 21:02, 25 July 2009 (UTC)

A bold proposal

In an attempt to turn a divisive RfC into something productive I have created a new page. I believe people who watch this page have valuable perspectives and I hope you will look at this new page, and do what you can to help make it work: Wikipedia: Areas for Reform Thanks, Slrubenstein | Talk 15:22, 22 July 2009 (UTC)

"NOR" policy not compliant with usual scientific standards in the field of law

Legal topics often relate to controversial issues. Some even argue that the essence of law is its contentious nature. The usual standard for legal scientific publications - of course - requires the facts to be thoroughly documented but it does not prohibit the author from giving an opinion, as long as it is clearly marked as such. While opinions in this context should be based on arguments, by nature they are not facts, and open to disagreement. But in the field of law, opinions are important. A reader may agree or disagree with the opinion given by the author - and in either case, it is helpful.

While in natural sciences there is usually no major disagreement - at least on topics that are included in an encyclopedia - some central aspects of law are controversial. An encyclopedia can not ignore this ambiguity. To mention just one example: human rights are often conttradictory, and open to multiple interpretations that are all equally current. IMHO a proper article on (a specific asdpect of) human rights could help the reader form an opinion by giving its own opinion - after a thorough discussion of the facts, of course.

Perhaps the essence of legal writing is that they are not only used to convey facts, but also help to form opinions. Rbakels (talk) 23:38, 24 July 2009 (UTC)

I don't see much to disagree with in your post, but could you explain how it relates to our policy on original research. We do aspire to write about disagreements with a neutral viewpoint so that readers can make up their own mind. (By the way, it is not correct to say there are no major disagreements in the natural sciences relevant to our articles, but that's not relevant to this thread.) — Carl (CBM · talk) 23:42, 24 July 2009 (UTC)

We have to be careful with terminology here... in Law there is a difference between an opinion and an Opinion. The first (small "o") is mearly the thoughts and ruminations of any old person. Joe Blow might have an opinion on a legal issue, but his opinion should not be included in Wikipedia as he is probably not an expert in the Law. An expert (such as a noted legal scholar) can have an opinion on a legal issue... if this is published in a reliable source such as a peer reviewed journal, then we can include it in Wikipedia. Then there are Judge's Opinions (capital "O"), such as are issued by the Supreme Court. These are very different and (at least in the case of a majority Opinion) can have the force of setting legal precident and in effect becoming Law. What I am getting at is that we need to note who is expressing the opinion, and under what circumstance (ie is it actually an Opinion). At least this is how it works in the Anglo/US system of jurisprudence. Things may be different under other systems. Blueboar (talk) 20:04, 25 July 2009 (UTC)

Relaxation of NOR to allow for GPS coordinates on articles

Recently, there was another unproductive row about geo-coordinates.

Would it be possible for a relaxation of NOR to allow people who have bothered to go and survey the location of places themselves with GPS units, to input that data into Wikipedia?

Allowing users to survey such locations themselves starts to remove the need for co-ordinates to be obtained from non-free sources. The data gathered could also be used to help improve other 'free' maps, which for various reasons cannot be done currently.

Sfan00 IMG (talk) 18:35, 27 July 2009 (UTC)

Stan, can you explain more about how this is done, and what safeguards there are against errors? Also, what is the problem with obtaining coordinates from reliable sources? SlimVirgin ^{talk|contribs} 20:54, 27 July 2009 (UTC)

GPS coordinates are obtained by going to a location with a GPS device, and clicking a button on the device to mark the location. The GPS device obtains the location from U.S. government satellites. GPS coordinates can be accurate within 10 ft (~1 meter). --Aude (talk) 00:26, 28 July 2009 (UTC)

FWIW, I've noticed that user-provided locations in Google Earth are often misplaced.   Will Beback  talk  21:09, 27 July 2009 (UTC)

Asking people to get their own GPS locations for places, is in essence not much different from what OpenStreetMap

is doing, although on a lesser scale. Someone goes out to a place, gets a co-ordinate, records it and then puts it on article.

The main issue with current 'reliable sources' is mainly that some people have issues in using co-ordinates from Google and other 'non-free' mapping providers, to the possible extent that such geocoordinate information is considered practically un-useable in any jurisdiction which implements 'database rights', the use of 'non-free' sources also in some peoples view taints the data set again making it unusable, despite Wikipedia nominally being under a CC style licence.

There is NOT a problem with obtaining 'co-ordinates' from reliable 'free' sources, such as US Fedral Govt mapping, or Open Street Map.

However, Open Street Map is based on a collaborative editing model, so I fail to see how if OSM can manage not to have glaring errors, a similar approach on Wikimedia projects could not also work.

Sfan00 IMG (talk) 22:47, 27 July 2009 (UTC)

If you can identify the site on GoogleEarth (for example), there is no reason not to copy the coordinates into WP, since this is verifiable by any reader. Crum375 (talk) 22:50, 27 July 2009 (UTC)

Some items are easy to identify, and some aren't. It's not hard to pick out the Disnleyland, though folks could differ on exactly where to put the crosshairs. The entrance? The center? Likewise for a city. (Do we always use city hall as the location?) But many buildings and natural features may be harder to identify. Getting back to the issue, if I type in a placename name and Google Earth zooms to a spot, that's information sourceable to their database. On the other hand, if I use my own judgment to decide where a notable location is located then that'd be original research, just like with any topic. It seems like if there was a dispute there'd be no way of solving it other than finding a real source.   Will Beback  talk  23:31, 27 July 2009 (UTC)

I agree, in general, esp. that if there is a bona-fide dispute, a real source would be helpful. But if the site is clearly visible and identifiable on GoogleEarth and friends, as it very often is, this is clearly verifiable in the WP sense. GE has amazing resolution, so "the entrance", "the center", even "the bus stop", should be identifiable, assuming the general site has good resolution (not all sites do). Disneyland has quite good resolution on GE, and my guess is that you can get accuracy of a meter or better for any identifiable fixed structure or object. The bottom line is verifiability: if the GE (or GoogleMap) identification is clear enough that a reasonably intelligent reader can verify it, then the the coordinates are not OR. In my own experience, the published coordinates for many locations are not as accurate as the ones you can obtain with GE. Crum375 (talk) 23:48, 27 July 2009 (UTC)

People arguing about whether Disneyland is at the gates or at the center of the park is the reason we don't allow this? Seriously? That sounds like one of the lamest edit wars I've heard of, and it hasn't even happened yet. This ought to be simple, really. If someone's coordinates are dead wrong, correct them; if they're basically right but you're pedantic enough to think that the crosshairs should be at the gate instead of the statue of Mickey Mouse, let it go. This problem has nothing to do with what the OR policy is supposed to prevent, but it has been interpreted so overbroadly that we get into these bizarre situations. ⟳ausa کui× 00:01, 28 July 2009 (UTC)

Indeed. The OR policy is not meant to cover things like this. — Carl (CBM · talk) 00:10, 28 July 2009 (UTC)

Obviously, this doens't concern Disneyland. Could User:Sfan00 IMG give details of the actual dispute so we're not dealing with hypotheticals?   Will Beback  talk  00:15, 28 July 2009 (UTC)

Google Maps is a non-free, copyrighted source, with their data coming from third-party sources (e.g. Tele Atlas). When we systematically collect coordinates, we are copying their data (well beyond fair use) and can be considered a copyright infringement. Because of the copyright issues, OpenStreetMap will not import Wikipedia coordinates and placemarks into OSM. We really need to get away from relying on non-free sources for coordinates. On the otherhand, it would be okay for people to get coordinates from OpenStreetMap or public domain sources if possible. That's not always possible, so getting coordinates via GPS is a very good option, in the same way we allow people to go take photographs for Wikipedia. --Aude (talk) 00:21, 28 July 2009 (UTC)

Google Maps is no more non-free than The New York Times, and both are perfectly valid sources. If we take large chunks of data we'd be violating their copyright, but to get the coordinates of a specific site, all we need to do is to verify its location on the map and grab the coordinates. This is perfectly acceptable because it's verifiable, and we are not even using their database when we verify (and/or tweak) the coordinates ourselves. Using a GPS to get coordinates in my opinion is a clear violation of WP:OR, because the reader has no way to verify the information, short of visiting the site. The reader could verify the coordinates on an online map, like GoogleEarth or GoogleMap, but then we are better off starting with the map, which is directly verifiable. Taking photos is different, in that we make a special exclusion in WP:OR to allow editors to shoot their own pictures, and present them. We make no such exclusion for editors to use an electronic instrument and read data from it, possibly transposing digits, or even confusing different sites. Crum375 (talk) 00:42, 28 July 2009 (UTC)

Regardless whether the user gets the coords from their own GPS, from Google maps, or from any other source, the coords are still verifiable, and so there is no issue with WP:OR or WP:V. I think the analogy to photographs above is apt: we believe we are able to verify that a user-taken photograph really is of the person claimed, and geographical coordinates are easier to verify than that. The OR policy is meant to prevent people publishing novel theories on WP; applying it to GPS coordinate is, in almost every case, overreach. — Carl (CBM · talk) 00:48, 28 July 2009 (UTC)

How do you know the 'reliable source' maps are accurate? I once read a comment that someone made about a certain area where the

'official' maps were notoriously unreliable, such that people used an alternate produce by a local concern. Sfan00 IMG (talk) 00:47, 28 July 2009 (UTC)

If photos have an exception, why shouldn't reasonably sourced 'data' also have an exception, mis transcribing or site confusion can occur even when using 'reliable' maps (although probably less so) Sfan00 IMG (talk) 00:47, 28 July 2009 (UTC)

Sfan00 IMG - is this all hypothetical or is there an actual dispute ?   Will Beback  talk  01:00, 28 July 2009 (UTC)

Photos have an exception because we have a dearth of freely licensed images, and because images enhance the aesthetic value and readability of an article. Also, photos are self verifying to some extent, in that they at least show an object that relates to the subject matter. If we allow editors to take electronic instruments and start collecting their own measurements, we might as well remove WP:OR altogether, because we'll be creating reams of primary digital data, with no editorial oversight except other WP editors. WP is supposed to summarize and present material collected or created by others, not to create original content, photos/artwork excluded. Crum375 (talk) 00:54, 28 July 2009 (UTC)

You just said above that these coords are verifiable in other ways. If they are, there is no OR issue. If they are not, we have the same issue as with photos. — Carl (CBM · talk) 00:57, 28 July 2009 (UTC)

Yes, if we start from an online map, identify a location, and carefully copy-paste its coordinates, that's equivalent to finding a verifiable published source for a textual fact. The reader can click on the coordinates and verify on the map that the identifiable object is where it should be. But if we allow editors to go out into the field with electronic instruments and take measurements, there is no immediate way to verify it, because the object may or may not be identifiable on the online map. If it is, we might as well start from the map, and be done with this discussion, and if it isn't, then that would be non-verifiable and OR. Crum375 (talk) 01:03, 28 July 2009 (UTC)

If there is no way to verify the coords with an online map, but the location is reasonably accessible to people with GPS units, then I see the situation as completely parallel to photos of living people. — Carl (CBM · talk) 01:12, 28 July 2009 (UTC)

I don't agree that coordinates are like images. The idea is that we want everything in Wikipedia to be based on published sources... but... If you use a published photo, you have to use the entire thing... and doing this can violate copyright laws. We thus make an exception to the NOR rules to allow for user created photos. Taking a coordinate from a published source (Google Earth, a map, or what ever) is more like quoting one sentence out of a book. With coordinates you are mearly taking one piece of information from the source, and that does not violate copyright laws. So there is no need to make an exception. Blueboar (talk) 01:13, 28 July 2009 (UTC)

I agree with Blueboar. We make a special exclusion for images because we have so few free ones. In the case of GPS, we can glean those coordinates off online maps, and have the reader verify them directly from the map. There is no reason to make an exclusion, as Blueboar says. Crum375 (talk) 01:22, 28 July 2009 (UTC)

You were just arguing that we cannot get them off online maps (if we can, they are verifiable anyway, so OR doesn't apply). Have you reversed your position there?

Re Blueboar: the OR policy doesn't say that everything must be based on published sources, only that the information presented must also be presented by published sources. — Carl (CBM · talk) 01:26, 28 July 2009 (UTC)

No... but WP:V and WP:RS do. Our polices and guidelines do not work independently of each other. Blueboar (talk) 11:59, 28 July 2009 (UTC)

We can get coordinates from offline maps if those maps are reliably published, although in that case we should also verify them on online maps, as a sanity check. Once we post the coords on WP, they are a couple of clicks away from a variety of online maps, so it makes sense to verity that they all (or at least a couple) seem correctly localized. As far as the WP:OR and WP:V policies, they do say that anything which could be challenged should be verifiable by a published source, so it makes sense to start with the verifiable source, not to get our own originally researched material and let our readers (or other editors) do the verification on published sources for us. Crum375 (talk) 02:30, 28 July 2009 (UTC)

If something is verifiable, then it does not matter what we start with. Verifiable does not mean "referenced", it means "able to be referenced". If (as you are claiming) these coordinates are able to be referenced, then it makes no difference how an editor obtains the coordinates to put into the article. The only time it would matter that coordinates were measured by some individual editor is when the coords could not be verified elsewhere. — Carl (CBM · talk) 13:17, 28 July 2009 (UTC)

Is this even an OR issue?

Or is it a copyright issue? This is starting to seem like its on the wrong talk page. ⟳ausa کui× 03:09, 28 July 2009 (UTC)

It would be copyright if a non-free database were copied to WP. I don't think anyone is suggesting that here, so that's not the issue. The issue we are addressing is whether it is Original Research for Wikipedians to go into the field with GPS receivers to measure the coordinates of a location, and then post their results on WP article space. Crum375 (talk) 03:19, 28 July 2009 (UTC)

The top posts above are exactly about non-free databases being copied. On the other hand, if coordinates are already published, then there is no way that posting them here can violate the original research policy. — Carl (CBM · talk) 03:24, 28 July 2009 (UTC)

I don't think anyone here condones copying a non-free database into WP. Creating fresh and independent content avoids the copyright issue. Crum375 (talk) 03:47, 28 July 2009 (UTC)

But, "creating fresh and independent content" is practially the definition of Original Research... since published coordinates exist and can be used, there is no valid reason to make an exception (unlike images, where the copyrite issue is a valid reason to make an exception). Blueboar (talk) 11:59, 28 July 2009 (UTC)

If the data is already published, then editors who measure the coordinates themselves are not creating new data. — Carl (CBM · talk) 13:17, 28 July 2009 (UTC)

But the only reason anyone is even wanting to go into the field with their own GPS devices is, apparently, because someone thinks its a copyvio to use Google Maps data on Wikipedia. Right? ⟳ausa کui× 13:24, 28 July 2009 (UTC)

But if using Google Earth to find co-oridnates is not creating new data, then any co-ordinate obtained from a Geo-data source like that should thus be sourced, because it's verifiable and to show it isn't OR? ( And because they are sourceable the usual rules about over use of a specfic source and so on come into play..)

On the other hand, if as some people have suggested the copyright issue is a red herring, then the co-ordinates 'traced' are possibly 'new data' created by the Wikipedia contributors concerned and by some logic expressed here that's OR, it being an armchair survey rather than a real-world one. Sfan00 IMG (talk) 13:29, 28 July 2009 (UTC)

Sfan00 IMG has tried before to get the use of coordinates derived from unfree maps banned on copyright grounds. --NE2 12:04, 28 July 2009 (UTC)

I assume the idea was shot down? The copyright issue is a red herring, only raised because it related to user created images. It is clearly not Original Research or a CopyVio to use and cite coordinates from a published map, whether online or paper. There is no need to allow user compiled coordinates, because we can use published coordinates. Blueboar (talk) 12:41, 28 July 2009 (UTC)

Exactly. We are allowed to cite information from the NYT or get coordinates from a non-free map — neither is WP:OR or copyvio. We are not allowed to grab entire long paragraphs from the NYT, or copy chunks of data from a non-free database, as both would infringe on their owners' copyrights (in many countries). Crum375 (talk) 13:09, 28 July 2009 (UTC)

Disscusion point : But unless it's explicitly stated as text, is getting Coordinates from a source like Google Earth's arieals of itself OR? Granted you can cite Google Earth, but Google Earth is only really allowing you to do an 'arm-chair' survey as opposed to a real world one. I will note here that I've had WikiMapia co-ordinates (that I'd reinstated) removed as being 'unreliable', seemingly BECAUSE they were user surveyed from Google Earth, precisely what some coordinates in Wikipedia are... But maybe this isn't the best forum for a discussion on source reliability. Sfan00 IMG (talk) 13:20, 28 July 2009 (UTC)