Wikipedia talk:No original research/Archive 15

From Wikipedia, the free encyclopedia

Archive This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Contents

Nutshell text spilling onto Wikipedia policies box?

That's how it looks to me, and very ugly. But is it only me? Does it look OK to everyone else? Or has my machine dropped acid? qp10qp 03:43, 6 October 2006 (UTC)

Fixed This is happening to a number of david gerard recent nutshell restorations --Birgitte§β ʈ Talk 22:32, 6 October 2006 (UTC)

Is it original research if... (redux)

Is making a purely mathematical calculation to yield a result original research?

If a verifiable source, say a census bureau gives the population of a geographic entity, and also gives its area, is the mere quotient of the two (yielding the population density) original research if the census bureau does not explicitly provide that number? What can be done consistent with no original research: (a) forgo the information even in our info boxes; (b) search for some other reliable source which has decided to do the math for us (which may not be available, especially as the census bureau tends to be more "up to date" than most other secondary sources, so the calculations from reliable sources may not be made using the same numbers); (c) do the math ourselves <sarcasm>and cite the maker of our calculator the reliable source</sarcasm>. Carlossuarez46 17:48, 6 October 2006 (UTC)

On one hand I don't think you need a citation to state the 8/2=4. However, when aggregating statistics it is important to take into account if the statistics were designed to go together. For example, what is the range of error for the 2 values? Perhaps you can put a "~" before it and a footnote on it stating how it was calculated. HighInBC 20:12, 7 October 2006 (UTC)
Citing the two values and calculating the ratio seems unexceptionable. Is anyone actually making an issue of this? Robert A.West (Talk) 03:08, 10 October 2006 (UTC)

Wikipedia:Attribution

A proposal that WP:NOR and WP:V be combined, and WP:RS ditched. All views welcome. SlimVirgin (talk) 05:19, 11 October 2006 (UTC)

I like the current set up. While they all work together, they are all very different sets of rules. HighInBC 14:40, 11 October 2006 (UTC)
As I noted elsewhere, I think it's a good idea, which will streamline our fundamental policies and make the process more efficient. There is too much overlap and confusion currently among WP:V, WP:RS and WP:NOR. It will also provide a single focal point for discussing sourcing issues. Crum375 18:52, 11 October 2006 (UTC)

Problem on an article

Per the dispute resolution page, I would like some people with a long standing knowledge of this policy to come and post a couple of comments on the RFC discussion at Talk:Make Love, Not Warcraft#RFC. An example of the content which is under discussion is this version of the page [1] which editors are claiming is acceptable. Thanks, Localzuk(talk) 20:15, 11 October 2006 (UTC)

sources question

Hi, this is an attribution question that seems to apply to a combination of WP:V, WP:NOR and WP:RS, but since WP:NOR appears to be the most relevant, it may make sense to post it here. Here is the scenario: assume we have an editor with access to copies of affidavits from an official government agency (in this case, the Patent Office). Assume that we want to rely on these letters as a reference in an article to show that the witness in that affidavit made certain claims (which he affirms in the affidavit). We aren't sure, however, how to verify these documents or obtain them, presumably from the government directly. The editor with access to the original letters uploaded the scanned letters into WP as a PDF file of scanned images. The letters seem like affidavits; they have government seals and signatures and appear authentic. The question is: can the article (which includes some controversial claims) make reference to these letters and point to the Wikicommons scanned image as source? Would doing so violate any existing WP sourcing/attribution policy or guideline? How, if at all, might one go about verifying this document, and does this verification method reasonably fulfill Wikipedia requirements? Thank - Che Nuevara 23:24, 11 October 2006 (UTC)

The documents should be available to anyone from the government for the normal copying fee, or should be available for inspection to anyone who goes to the appropriate office. The citation for the document should make it clear how to obtain a copy of the document, if that isn't common knowledge. By the way, patents for at least the last few decades are available at the patent office's web site. I hold two patents; if you tell us what images you are talking about, I might recognize what kind of document they are, and just possibly, where to find them.--Gerry Ashton 23:42, 11 October 2006 (UTC)
In this case it isn't a patent, but rather an affidavit of testimony in a patent case. Does that make any difference? - Che Nuevara 00:01, 12 October 2006 (UTC)
Yes. If it is a recent U.S. patent,it's trivial to read it on the web. I have no idea how to get a copy of an affidavit of testimony in a patent case. --Gerry Ashton 00:08, 12 October 2006 (UTC)
The patent is here. The uploaded scanned documents are here. The question is assuming for the moment that we cannot easily get them from the Patent Office or elsewhere, is that uploaded file sufficient as a reliable and verifiable source by itself. Crum375 00:11, 12 October 2006 (UTC)
Inherent in the question, of course, is can we get it easily from the Patent Office? We haven't tried yet. - Che Nuevara 00:45, 12 October 2006 (UTC)
I don't think scans of documents that are stored in an unreliable repository, when the same documents are not available from any reliable source, are acceptable as sources for Wikipedia articles.
On another note, patent applications used to be confidential until the patent was granted. (Since 2000, applications are published after 18 months since the filing of the application, unless the inventor wants to forgo foreign patents.[2]) It is routine for the inventor to apply for a wide range of protection, and the patent examiner to pare down the protection so it is much narrower than the original request. I'm not sure, but it is possible that documents that relate to claims (that is, areas of protection) that were denied may be confidential forever. --Gerry Ashton 00:50, 12 October 2006 (UTC)
This claim was granted a number of years back, so theoretically it's public record now. Do you know how one might request the records (ie affidavits and such)? - Che Nuevara 01:48, 12 October 2006 (UTC)
I don't know how to obtain the information. My point about claims was this: an inventor makes claims, and provides a description to support the claims. If the claims are allowed, then no one else may make what is in the claims for 17 or 20 years, for old or new patents respectively. Usually the examiner will allow some claims and deny others. I expect that documents related to claims that made it into the final patent are available to the public, but I don't know about documents related to denied claims. --Gerry Ashton 02:16, 12 October 2006 (UTC)
Let me ask a couple of questions. Let's assume I am the inventor of gizmo X. I decide to write up an article about X. I want to include a statement in the article that I know was made in an affidavit submitted to the PTO. I have a copy of that affidavit, but I can't find it online anywhere. My first question is: can I just scan and upload the affidavit into WP as a PDF file and refer to it as my source? Second question: If not, can I just say that I have this source at my home, and if anyone else wants it, they can ask the PTO for it, or I can fax them or email them a copy, and in the meanwhile use the uploaded source as reference?
Thanks, Crum375 02:42, 12 October 2006 (UTC)

(unindent) I would say Crum375's approach would not be satisfactory, because that Wikipedian is trying to be a publisher, but (as far as I know) does not have a reputation as a reliable publisher. --Gerry Ashton 03:33, 12 October 2006 (UTC)

I understand what you are saying, but I am not trying to be the publisher - the 'publisher' would be the PTO. It would be like some rare book, of which I happen to own a copy, and I make a reference to some item in it. If some reader or editor questions the item, they can always try to find a copy through their local library - wouldn't WP accept this type of reference? And if yes, why can't I do the same with my 'rare' copy of the affidavit that people in theory may be able to get from the PTO? And to make life easier for everyone, I upload them my copy as a PDF image into WikiCommons. Crum375 03:49, 12 October 2006 (UTC)
Rare books are acceptable if they are held in a library that is open to the public. So maybe your not a publisher, maybe you're a library. But are you a reputable library?
Also, I consider scans held on non-reputable archives, including Wikipedia and Wikicommons, to be unreliable. There is no way to know if the document that was uploaded was authentic, nor is there a way to be sure of the identity of the person who uploaded it. And when the sources are related to a patent that, at first glance, appears to be an embarassment to the USPTO, the evidence needs to be exceptionally reliable. There is good reason to question the authenticity of evidence related to highly unlikely claims. --Gerry Ashton 04:16, 12 October 2006 (UTC)
But can't I just (effectively) tell any skeptical readers to either trust me, or go get these documents from the PTO on their own if they don't, and check for themselves? Crum375 12:44, 12 October 2006 (UTC)
I believe that's the issue, of course -- if it is possible to go get the docs from the PTO, then a reader can go verify the documents. If not, then obviously not. I was under the impression, however, (and this may just be ignorance on my part) that it's possible to request this sort of document. - Che Nuevara 16:18, 12 October 2006 (UTC)

I came here looking for an answer to my question on this. ([semi-]Long version: I requested a record from the Wisconsin Department of Military Affairs in re the Wisconsin State Defense Force. They sent me a letter back saying that the record does exist [that they do not currently have a plan to organise it]. I think that that information would be `useful' to have in the article) Short version: I sent a letter to the government. They sent me a letter back. Anyone can get what they sent back by simply asking them. Can I use this in Wikipedia as a source? (Which from the dicusion, I would say probably) How would I cite it? Benn Newman 01:16, 18 October 2006 (UTC)

Chess engine analysis

I should like a view on this. Nowadays, chess engines are routinely used to analyse games. If an editor adds chess engine analysis to an article I would argue that it is not OR because it is verifiable by another person who can simply run the same program. I should welcome views, please. BlueValour 15:50, 15 October 2006 (UTC)

I would disagree. Having used such engines to analyze many games, I am aware of all the many parameters involved, each of which can change the results, potentially. Yes, one could identify exactly each such parameter, but IMO for the reader who wants to verify or reproduce the results it becomes a non-trivial task. I think we are allowed to point people to published literature only; making them install and properly run specialized software packages (with the exact same version) to verify the article goes far beyond going to to your local librarian and asking for a copy of some book. There is a lot of skill involved in properly analyzing a game with an engine; sometimes you may decide to follow deeper on some branches because they 'look' promising or interesting - I think it's clearly WP:OR. Crum375 16:20, 15 October 2006 (UTC)
In this specific case, the engine, the depth, the full length of the primary variation, and the evaluation were all cited. I think it may be difficult to verify, but it's not original research since the source is cited; and in general, use of chess engines to analyze games helps the project immensely, since extremely strong evaluations of positions are difficult to come by for free -- unless you are using a strong engine like Rybka. Besides, anyone can purchase this engine from the website for the approximate monetary cost of a large paperback book, with considerably less time spent locating it. --Ryan Delaney talk 02:59, 16 October 2006 (UTC)
I say it's WP:OR, and obviously so. To be sure of the same result you not only have to purchase the same software and set it the same, but you also need to purchase the same computer hardware, because these chess engines run better on better machines. Then after all that, the editor decides which analysis to publish and which not to publish. Rocksong 03:14, 16 October 2006 (UTC)
That's not right. Rybka at the same depth will find the same evalutation, regardless of hardware. Better hardware would only mean she would reach that depth faster. --Ryan Delaney talk 13:31, 16 October 2006 (UTC)

To avoid OR does not require the source to be cheap or free, it means to have verification. Say a fractal image, the exact same source file can be used to produce the same image. Ease of confirmation does not seem to the be an issue, foriegn language sources are allowed. The only concern is if in fact this game engine can give the same results when ran by other people in a deterministic fashion. HighInBC 03:22, 16 October 2006 (UTC)

Provided the parameters are specified then running the same chess engine would give the same result. On that basis the analysis is verifiable. BlueValour 03:36, 16 October 2006 (UTC)
This still does not address my other objection, that the Wikipedia editor decides which computer analysis is relevant to run and include. Contrary to what Ryan Delaney says, good analysis is not hard to find for free, at least when it comes to significant games: there are a number of Grandmaster commentaries, free on the web, on the game in question (the final tie-break game in the recent FIDE World Chess Championship 2006). If these Grandmaster commentators do not think the computer analysis is relevant to include, why should we? I still say it's WP:OR. Rocksong 03:58, 16 October 2006 (UTC)
That is an editorial judgment that must be made in the case of incorporating information from any secondary source. Such disputes can be resolved on the talk page. --Ryan Delaney talk 13:33, 16 October 2006 (UTC)

Hmmm, many lab experiments could be duplicated if documented properly. That hardly means that scientists should skip scholarly journals and just add their findings to Wikipedia articles. I'm not seeing why Wikipedia should be a dumping ground for stuff no one else is interested in publishing. Wikipedia is not a blog. --W.marsh 04:13, 16 October 2006 (UTC)

This is a disanalogy. Running chess analysis with an engine is not an experiment, any more than asking Grandmaster Ian Rogers to analyze a position is an experiment. Incidentally, Rybka's opinion is more verifiable than that of Mr. Rogers, since any person can have his own personal copy of Rybka, but there is only one Ian Rogers. --Ryan Delaney talk 13:36, 16 October 2006 (UTC)
Using a complex engine, with a complex algorithm, with a complex set of parameters and criteria, to produce results is WP:OR. To ask people to reproduce the results, as W.marsh stated above, is like the scientist asking others to reproduce his/her results. We are only allowed to present previously published data, and can only ask the readers to get a copy of that data from the library (or other source) and read it to verify we quoted our source properly. Anything more would become original work. Also, the very fact that you stop at a specific search level on the engine (say 14ply) is also WP:OR, since you are effectively telling the reader to 'trust you' that by going deeper on some branch the analysis won't change. Crum375 13:55, 16 October 2006 (UTC)
The settings are all at default. Anyone can reproduce this with no special configuration. --Ryan Delaney talk 14:21, 16 October 2006 (UTC)
None of this goes to why Wikipedia should be publishing information no one else cares to. No matter how easy it is to duplicate, it's still original work (as in, this is the first place it's being published). --W.marsh 15:36, 16 October 2006 (UTC)
My thoughts on this

I think this is easier to understand if we simply think of a computer's opinion as functionally the same as that as any extremely strong chessplayer. For example, if I ask Garry Kasparov what his opinion on a position is, and then report on that, am I doing original research because Garry Kasparov is hard to get access to and so it would be hard to find him and ask him the same question again and again? In fact, it's even less reliable and less verifiable to ask Kasparov, since he might give a different opinion on a different day, and he's extremely difficult to get access to. However, where there is only one Kasparov, anyone can have his own Rybka. I might have made up everything Kasparov said, but if I make up what Rybka says, anyone can run it for himself to expose my fraud.

Really, I think that what I am doing here is interviewing an extremely strong chessplayer and then reporting on what he said. This is not original research any more than that would be. --Ryan Delaney talk 15:20, 16 October 2006 (UTC)

Interviewing? Reporting? That's not what we do here. Again, this is not a blog/newspaper, sorry to sound trite but we aren't going for original coverage, we're reporting what reliable sources have already published. Generating new information covering a topic is what reporters do, most Wikipedians are untrained high school and college students, generating original work like a blog would is exactly what our critics make fun of us over. By and large we aren't very good at it. We are good at writing articles based on reliable, published sources though, and that's what we should stick to. Plenty of stuff can be verified reliably, the question is whether anyone's bothered to publish it. Wikipedia is not a publisher of original thought... this is one of core concepts. --W.marsh 15:36, 16 October 2006 (UTC)

These are all valid points, and I agree with you in general, but what I'm trying to present here is an explanation for my belief that this generalization does not apply to this specific case. The original research policy is intended to stop people from inserting their own ideas and their own work into Wikipedia for the purpose of popularizing them, and that's good. But this is not such a case; I am not putting my own work or my own ideas into Wikipedia. I am using chess engines to supply commentary, which may be illuminating to readers who are not chess professionals into the extremely dense and complicated positions of world-championship level chess. This makes the articles better. In fact, chess engines are more reliable, verifiable, and accessible than the opinions of strong humans. --Ryan Delaney talk 18:07, 16 October 2006 (UTC)

You have still not responded to my question above: how do we know that the specific search depth that you used is sufficient (e.g. 14 ply)? Aren't you basically telling the reader 'trust me, this depth is enough'? As I am sure you know there are many examples where going deeper (at any given depth) on one specific branch can totally change the analysis, sometimes from win to loss. So if you are making this decision for the reader, you are in fact presenting your technical opinion that "this depth is sufficient", which could be correct or not, however once you present an opinion, you are editorializing and that's definitely WP:OR. Crum375 18:19, 16 October 2006 (UTC)
If the reader wants to find out what the engine would say at a different ply, they can do that themselves. However, I'm not against a general agreement that, for example, the engine should have searched to at least depth 20 (or some other number deemed appropriate) before we accept its analysis. --Ryan Delaney talk 18:31, 16 October 2006 (UTC)
I would definitely not go below 20, then, and 6 man EGTB endgame analysis. But I am still uneasy about the concept of editors running big computational jobs to analyze positions. I realize it is a mechanical type of work, but it's still significant work and is prone to errors. I would feel much better if the analysis were done off-wiki at some chess site (wiki-chess?) and then linked into from WP. Crum375 18:46, 16 October 2006 (UTC)
I understand/appreciate that the goal is improving articles, and maybe this is a situation to "ignore all rules" since it seems harmless and we shouldn't avoid improving an article of principle. However I do feel like this approaches original work though, since no one else is apparently publishing this stuff. That Wikipedia should be the first to publish something, however verifiable it may be, should always be scrutized. --W.marsh 18:16, 16 October 2006 (UTC)
Absolutely, and I appreciate your good faith in applying a strict scrutiny here. --Ryan Delaney talk 18:19, 16 October 2006 (UTC)
The purpose behind OR is to avoid physics-cranks and rewriting history. None of those issues apply in this case. We're talking about something that is fairly simply verifiable. But aside from that, it is also rather trivial. *All* commentary on chess that can't be solved (endgames can be solved completely because their are a finite number of moves remaining) will either have to have some sort of arbitrary analysis or no analysis at all. Adding chess engine results *is* useful and I don't see the downside of including it. So what if a chess engine is incorrect and there is a "better line". It is irrelevent. *All* lines are subject to this kind of change by mathematical necessity. In terms of articles themselves, adding chess engine results serves to illustrate the point, so long as the article is not worded as follows: "This chess line is great for white because xyz chess engine gives it a +2.3". But say that a book on chess or a chess player has published "this chess opening gives white a slight advantage" and we add that to an article. Well then it makes sense to illustrate the point by giving a chess engine result. The article is not asserting that the chess engine's result is correct, it is merely illustrating the point. The chess engine's results are just mathematics. Say I have a fictious "ramchessengine" that evaluates the same position at -3.7 because it assumes that White usually makes lousy moves and Black makes good ones. Does this matter at all? It doesn't argue against the original point that the chess line is considered to be good for white. So long as the chess engine's result is not the *only* assertion, but instead illustrative of an existing point, I have no problem with its usage in this case. It should be noted that if someone even understands the chess engine results, then they already know that such results are generally arbitrary by definition. -- RM 18:27, 16 October 2006 (UTC)

broken link

Wikipedia:No_original_research#Reputable_publications seems to go back to the top of the page. --Espoo 10:26, 16 October 2006 (UTC)

Ghost ramp

Can someone take a look at ghost ramp? Does this violate NOR (or another policy)? --NE2 00:49, 18 October 2006 (UTC)

On a superficial look it seems you are right - I see not a single source naming any of these as 'ghost'. By the definition provided, each would have to be shown as abandoned, per reliable source. As for notability, I guess it would not be an issue if some reference could be made to the overall phenomenon being notable, but I think that each individual case would have to be proven 'ghost' on its own, with its own source(s). The GoogleMap or equivalent sources are nearly worthless,IMO, as they don't prove the 'ghost' issue. Crum375 00:59, 18 October 2006 (UTC)
Note that there is a second (subsidiary) definition, "The term can also refer to a ramp which at one point handled traffic, but was abandoned for some reason (and never demolished)." Between this and the "built or partially built but abandoned," doesn't that cover just about anything that would show up in an overhead view? That being said, the sourcing for most of the discussion of individual items is fairly minimal. --Mr Wednesday 06:29, 18 October 2006 (UTC)
If you read on, you'll see: "The term does not refer to inactive or partially-built ramps which are intended to connect to a roadway which is actively planned or under construction; it only refers to ramps which have been abandoned for some reason." (emphasis added by me)
Which, of course, means that there is no way to tell from a visual inspection; you'd have to provide reliable sources per entry to prove that they meet that requirement. As of now, I see none. Crum375 12:10, 18 October 2006 (UTC)
Hence my observation, "sourcing for most of the discussion of individual items is fairly minimal." --Mr Wednesday 15:36, 18 October 2006 (UTC)
We may be talking about the same thing, I am not sure. What I am saying is that according to the definition given, you would need a reliable source per item -- just visual observation is clearly insufficient. None of the items that I looked at (and I may have missed some) have any source that I could see showing they meet the inclusion criteria. Hence the article is pure WP:OR at this point. If this is what you are saying, then we are in agreement. Crum375 17:55, 18 October 2006 (UTC)
I was agreeing that almost all of the items lack a reliable source. If I get a chance, I'll try to come up with something for the item or two that I added myself. --Mr Wednesday 20:03, 18 October 2006 (UTC)
Out of curiosity, do you see an acceptable source for any item in that list? Crum375 20:07, 18 October 2006 (UTC)
I was hedging with the "almost", I didn't want to take the effort to see if there were any. At this time, all of the Houston entries should be adequately sourced, and it looks like another user added a citation for one of the Florida entries. --Mr Wednesday 22:16, 18 October 2006 (UTC)
I picked one of the Houston ones at random, the one with reference to page 184 in this book, and I couldn't find where it says there is any leftover stub with no plans for connection. In fact, even the article text doesn't claim that. It seems to me, if this is an example of the 'good part' of the article, the bulk of the article is mostly likely somewhere between plain wrong and unsourced, unless proven otherwise. Crum375 22:36, 18 October 2006 (UTC)
Excuse me? The overhead view is proof that the stubs exist (and there are also pictures in the reference). The fact that the corridor is under study (per the reference) demonstrates to me that there are no plans for connection (it's merely one of the alternatives under consideration). What's your standard of proof here? I don't disagree with your assessment of the portions that have not been referenced, but I would respectfully request that you reconsider your evaluation of the three Houston items.
Note that I stripped one entry due to the impossibility of finding any reference at all (Kirby with Beltway 8) and another due to evidence that there are now plans to connect up the ghost stub (Crosby Freeway). --Mr Wednesday 00:57, 19 October 2006 (UTC)

(outdent) This article is essentially a list, and a part of the inclusion criteria (which I quoted previously) is: "The term does not refer to inactive or partially-built ramps which are intended to connect to a roadway which is actively planned or under construction; it only refers to ramps which have been abandoned for some reason." The critical words for me are "actively planned": if there are no plans being made to extend/complete the stub, it would be considered 'ghost' and hence includable. In the single Houston example I checked, it specifically said there are plans being studied for extension/completion, hence I don't see it as eligible, i.e. it is not 'abandoned' by the definition in the intro. What am I missing? Crum375 01:27, 19 October 2006 (UTC)

The study was to determine what to do with the corridor, generally. Rather than complete the original express lane plan (which I believe, based on personal observation of the right-of-way, to have been four lanes in each direction), the "most feasible alternative" was to create two toll lanes in each direction over the entire length of the corridor. The study appears to have terminated, but I find no indication that there are any active plans to implement it — TxDOT's "Houston projects" page shows only the feasibility study, nothing subsequent to it. See [3]. --Mr Wednesday 02:43, 19 October 2006 (UTC)

I'm also curious about the use of the term "ghost ramp". As far as I know, this term is only used by roadgeeks. --NE2 21:04, 18 October 2006 (UTC)

As you may be aware, the page was tagged for a while with a recommendation for renaming. I had nothing to do with the removal of that tag, so I can't comment on it. --Mr Wednesday 00:59, 19 October 2006 (UTC)

Is the result of a vote allowed to override this policy? --NE2 22:18, 21 October 2006 (UTC)

  • No, but it is clear that not everybody views it as OR, and the AfD discussion is not a vote, as stated many times by closing admins, and the editors' interpretation as to what constitutes OR is part of that discussion, if not the main point. If the consensus is that there is no OR in the article in question, then there would be no "overriding" of the policy, but, in fact, a reaffirmation of it. 147.70.242.40 23:10, 23 October 2006 (UTC)

Are most Wikipedia "Lists of" original research? (fork from “Ghost ramp”)

Wikipedia contains many "lists of", some trivial, and some more serious. Consider for example List of polydactyl people and List of countries. I am wondering whether most of the lists on Wikipedia constitute original research to some degree, and whether that original research isn't so inextricably linked to some of these lists that they don’t even belong on Wikipedia.

Most lists contain a set of members, and the presence of the member on the list is supposed to attest that the member does indeed have whatever property that list is supposed to identify. In forming lists this way, I see two ways that original research can slip in.

First, determining whether a member does or does not have the necessary property can be difficult. Should a member be on a list if reliable sources contradict each other as to whether they have the property? What is the citation standard for deciding that a member has a property?

Second, the very act of assembling such a list strikes me as consisting of original research, given that the editor(s) are making conscious decision which members should or shouldn’t be on the list.

Consider List of countries – on first glance it seems this should be an easy list to compile. But instead, the article contains a lengthy subsection explaining what criteria were used for putting members on the list. In effect, original research to create a Wikipedia-specific definition of “Country”. Someone, by listing Pridnestrovie for example, has done original research to decide that Wikipedia can recognize this entity as a country.

I suspect, if NOR is truly to be respected, the only “Lists of” that Wikipedia should contain are those where the list is compiled and maintained by a reliable source outside of Wikipedia, and we merely reproduce that list.

This comes to mind in light of the above discussion of “ghost ramp”. Although that article doesn’t explicitly call itself a “list of”, it is in effect one, and that seems to be where the NOR issue arises. - O^O

I think that to collate and present information using some selection criteria, you must define your terminology, and then determine if individual items meet the selection criteria. I don't consider it WP:OR to define inclusion criteria for a list (e.g. "all buildings taller than 1000'"). But each selected item must be shown to meet the defined criteria per reliable source ("building X is 1,234' tall[1][2]"). Crum375 18:48, 18 October 2006 (UTC)
Extending your example, for this List of buildings teller than 1000 feet, do broadcast towers make the list? What about oil drilling platforms? What about the CN Tower? Even for a subject as simple as you posted, there can be (and indeed are) genuine debate over what makes the list. If we end up writing our own definition of what should make the list, then we are in effect doing orginal research (to create the definition) instead of just accepting some common definition used outside of Wikipedia.
So, returning to List of countries, where is a non-wikipedia source that says we are using the right definition of country? And, if there isn't one, then isn't our definition (and thereby the list itself) original research? - O^O 20:27, 18 October 2006 (UTC)
Every piece of writing within and without Wikipedia explicitly or implicitly defines the scope of the writing; if authors were forbidden to do that, we would cease to be a literate society. --Gerry Ashton 20:40, 18 October 2006 (UTC)
And to respond more specifically to your question, your implication is correct: we do need to carefully define our terminology, especially so in list articles. For the 'building' example we would need to say: "for the purposes of this list, a 'building' is a free standing structure (no guy wires) and its height is measured to the top of any rooftop antenna or other solid protrusion" (for example). Yes, it can get complicated, but you have no choice if you are going to present precise, high quality information. None of that process involves WP:OR - it's just planning what you are going to present and how you are going to do it. The actual data you present has to be verifiable, but the criteria and format are yours to decide. Crum375 21:06, 18 October 2006 (UTC)

Policy changes by SlimVirgin

SlimVirgin (talk · contribs) has been making major changes to this policy without prior discussion here. Take a look at these extensive diffs since October 16th. SlimVirgin had zero edits to the talk page during that period, but made approximately 24 changes to the policy itself.

These changes need to be carefully examined by others. --John Nagle 07:27, 25 October 2006 (UTC)

Skimming over the diffs, they all appear quite reasonable. Do you have something particular that you object to? Crum375 12:38, 25 October 2006 (UTC)

Is juxtaposition of A and B equivalent to OR ?

I wish to invite discussion on the following point. The Wikipedia Original Research Policy states in part: "A and B, therefore C" is acceptable only if a reliable source has published this argument in relation to the topic of the article."

I have recently become involved in a discussion on the Stephen Barrett Talk Page [[4]] where I have been accused of doing Original Research. I have attempted to state two juxtaposed facts A and B, but have not stated a conclusion. The ABC version of the policy as presently stated is enforceable. A policy which said that editors could not state two juxtaposed facts A and B because readers would draw a conclusion from them would not me enforceable, since readers will inevitably draw conclusions from juxtaposed facts. Here are some examples:

When we read an encyclopaedia article about Hitler we inevitably think: "Hitler must have been mad" yet I shouldn't think the word "mad" is used to describe him in an encyclopaedia article.

To take another example. There exist charlatans who claim to have supernatural powers which they know they do not posseess. Nobody will state in his Wikipedia contribution outright: "Mr X is a liar and a charlatan." But the longer these people go through their careers the more facts accumulate about them. So anyone who reads a Wikipedia article about someone like this will read a collection of facts from which readers will almost inevitably infer that the subject is a charlatan.

So, should the OR policy only ban explicit inference as I say ? Or are we going to get into grey areas and matters of nice judgement decided on a case by case basis ? Robert2957 08:11, 25 October 2006 (UTC)

Robert, if your two juxtaposed facts A and B imply a conclusion, C, then it's as if you stated it. If that conclusion has not been made by a reliable source in relation to the topic, then you're likely engaged in original research. The material has to be presented as neutrally as possible, sticking closely to what reliable sources have specifically said about the topic, and not what can be deduced from what they have said. SlimVirgin (talk) 17:12, 25 October 2006 (UTC)
Slim is right. Put it another way, Robert: you are admitting that you want to arrange passages so that the reader will reach a conclusion. If that is a conclusion you want the reader to reach - and you have as much admitted this is the case - then this is both a perfect example of original research and I'd argue a clear violation of NPOV in that you are trying to use our article to convey your POV. Wikipedia is not a blog or a soap box; the internet has plenty of other venues where you can express your own views. None of us editors are supposed to do that, we have to be self-conscious to try not to do it and grateful when another editor catches us slipping up. Of course, if a verifiable and reliable source makes the conclusion you want the reader to understand, just say so: "According to x ...." and provide the citation. If you can't do this, that in and of itself is a good reason to think you are violating NOR. Slrubenstein | Talk 17:18, 25 October 2006 (UTC)

No, I wouldn't wish to arrange passages to lead people to conclusions. I have not admitted this. What I am trying to argue is that an editor should not be accused of OR if he merely states two juxtaposed facts from which readers happen to make an inference, or if he juxtaposes a fact of his own with another editor's fact. Close consideration of my example about charlatans should make this clear. Any sourced and verified fact should be acceptable. Conclusions for which no authority is quoted, of course, should not be allowed. The ABC rule is OK. An AB version is not. Robert2957 20:17, 25 October 2006 (UTC)

Perhaps it would help if you could give an example of the edits that caused the problem. SlimVirgin (talk) 20:19, 25 October 2006 (UTC)
I don't think it would. My question is about a principle of very great generality Robert2957 20:38, 25 October 2006 (UTC)
I think SlimVirgin's request is perfectly proper, especially considering that this is your specific issue that is being highly disputed by all the other involved editors.
It also involves issues other than OR, where you seem to wish to turn the article into a place to start debunking possible errors. This would set a precedent for using articles as debunking dumps for every single possible error one can find on a website (in this case a major 3000+ page website that has disclaimers and doesn't claim to be inerrant). That's not very encyclopedic, and certainly not the purpose of the article. I don't think it would be wise to even begin down that path. -- Fyslee 21:14, 25 October 2006 (UTC)


With all respect I am here raising an issue of very high generality which is completely independent of any one controversy about any one website. I am saying that an NOR policy of the ABC type is justiciable, and an NOR policy of the AB type isn't. This question not only can, but should, be pursued independently of any particular discussion. I don't want to bring a particular argument about a particular example into this. If I did so any number of other particular OR discussions could be brought into this discussion and those particular discussions should take place on their respective talk pages. Robert2957 21:20, 25 October 2006 (UTC)

Generalities are only as as useful (generally :) ) as the way in which they're applied to specific situations, most times. The article in question, SV, is Stephen Barrett; see related talk page for more details. — ripley\talk 21:46, 25 October 2006 (UTC)
I don't have time to look at the article/discussion in detail, so I will comment generally. If an article juxtaposes verifiable and significant facts A and B, which tend to produce conclusion C, but makes no mention of C, I wonder what is going on? Surely the sources are not idiots. Is C consensus? If so, it would be fair to mention it. If not, does a reliable source assert C? If not, this is beginning to look like covert OR and possibly an NPOV problem? Does the article omit verifiable and significant facts X and Y, which tend to produce conclusion not-C?
Of course, if the natural way to present the facts tends to suggest a conclusion, and all the facts are fairly presented, then I see little to object to. It is natural to list a performer's credits in chronological order. If this demonstrates a gap of five years following the birth of her child, there is a natural conclusion and there is no need to leap through hoops to avoid it. Robert A.West (Talk) 23:15, 25 October 2006 (UTC)

To answer Robert's question to me, if an editor justaposes a and b and readers draw conclusion c, then the article is presenting original research even if the editor who made the juxtaposition did not have this intention. If the editor did not have the intention, surely s/he would have no objection to the edit (juxtapostion) being changed so as not to violate the policy. This as abstract as I can get. Slrubenstein | Talk 08:47, 26 October 2006 (UTC)

To get an idea of precisely what he is trying to do, read this. He seems to think that Wikipedia is the place to "get certain facts" mentioned. I have encouraged him to make a blog. He doesn't seem to understand the purpose of Wikipedia. -- Fyslee 09:33, 26 October 2006 (UTC)
Also this diff [5]. — e. ripley\talk 13:17, 26 October 2006 (UTC)

Precisely what I am trying to do is to establish and ABC rather than an AB version on the NOR policy as the official policy of Wikipedia. Any previous controversy I may have been involved in is irrelevant to the discussion I am trying to initiate here. I do not think that Wikipedia is the place to "get certain facts" mentioned.I understand the purpose of Wikipedia. Robert2957 10:33, 26 October 2006 (UTC)

But, don't you see that, if the juxtaposition of A and B is a deliberate editorial choice, and the juxtaposition implies a conclusion, then the effect is the same as if C were explicit? Or, do you not agree that (as Slrubenstein pointed out) edits must be evaluated based on their effects, rather than on guesses about the intention of the editor? If the conclusion can't be sourced, the juxtaposition should be avoided, unless failure to juxtapose would look strange or otherwise harm the article. Robert A.West (Talk) 23:53, 27 October 2006 (UTC)

Serve/serves

Jesup, the singular is correct here. "... or any unpublished analysis or synthesis of published arguments, ideas, data, or theories that serves to advance a position."

To use serve is equivalent to saying: "Any man or woman who arrive on time get the job." SlimVirgin (talk) 17:06, 25 October 2006 (UTC)

Review of Policy

I have reason to believe that this policy needs some review to discss the incorporation of factual original research. An editor recently put a tag on a very small article I began to write stating that it needed a complete rerwite because it didn't state sources etc. and it looked like original research, which it was. I have no problem with people questioning, and have no problem with knowing when I'm wrong, when somthing needs to be changed or when something needs to be done differently. I do, however, have a problem with this policy for a few reasons:

  • The policy discriminates against people who don't have the means or the money to publish their own arguments, ideas, knowledge, etc. For example, if I had more money I could publish a book containing all of my ideas and cite it as a reference on wikipedia, although it would conform to wikipedia policy, I feel it wouldn't be reliable as my updated ideas as I'm always learning new things and discovering new stuff that would need to be updated. As I don't have the means or the finances to publish my arguments, ideas, etc, this policy states that my ideas, arguments etc, mean nothing and acording to this policy should not be included on wikipedia. I feel this is wrong.
  • What is the definition of an expert and what gives a higher educated person more credibility over another person? Most of the the credibility it deserved, but never always, and education isn't everything, we can't forget that people learn more in day to day life and what they see than what they do during formal education, yet there is no method to categorise this level of knowledge. So why ignore it? You can't run away from uneducated knowledge, and it is vastly more valuable than any other form of knowledge. Education can only get a person so far, this is true in industry, careers and personal development.
  • I do agree that the minority wreck it for the rest of us, there are alot of people out there who think they're opinions, ideas and arguments, etc, are the factual ones, reality, when they actually are far from it. The reality is that no one, no one, regaurdless of their ability to conform to policy or their level of education or level of expirience is unfalible. Everyone should be questioned, and everyone should be heard. I wrote an article on a small suburb/town called Warrandyte in which I used a hell of alot of original research, combined with "acceptable references" to write it, it would have taken a person who has never been to or knows nothing about Warrandyte, a very long time to uncover published references to write the article in the level of detail I did, and even then they may have cited references that themselves had never been to the town and seen it with they're own eyes, therefore rendering it less reliable than my original research. I can verify my original research through the use of individual knowledge of the town, photographs, my own sources such as local newspapers, local groups such as historical societies, things I have seen with my own two eyes (for example, "many people canoe in the river all times of the year", why on Earth would I have to cite a source for that when I go there all tiems of the year and see them with my own eyes? My point is that if anyone has a problem with my original research, I will listen to them, take their critique seriously and with an open mind, and see if there is any merit in their critique, if there is, the content can be changed, if there isn't then it won't have to be. Slapping people with WP:OR tags I feel is just lazy way to say to people you don't trust their level of knowledge so this article should not be trusted. If you see an article that you don't believe should be trusted because they've used original research, ask them where they got their research from. And tell them what you think needs improving, what you disagree with, etc. If they then say to you no bugger you I'm right your wrong then they don't have a leg to stand on. But if they say sure what needs improving, lets get it right then whats the problem?

I apologise for my inability to spell and write with correct grammar to a highly educated level, if my parents had the money perhaps that would be different. And I apologise for not making clearer sense and organising my opinion so it is easier to read, but I figure, if you have taken the time to read this then perhaps you might understand what I'm saying, if you didn't read it then your attention span is too short to be discussing wikipedia policy.

Thankyou for your time. Nick carson 06:05, 29 October 2006 (UTC)

Hi Nick, regarding your first point, if you paid to have a book published, we couldn't use it as a source anyway because it was self-published, except in very limited circumstances. So money is not the issue. SlimVirgin (talk) 06:13, 29 October 2006 (UTC)
I guess that raises the question, how do you proove to people that your intentions are not selfish? And why do people instantly assume that I am a selfish perosn and am only writing articles in an encycopedia for personal gain? Nick carson 06:20, 29 October 2006 (UTC)
Nick, the best thing is to use reliable sources whenever you write articles, and then no one can question your intentions. SlimVirgin (talk) 06:23, 29 October 2006 (UTC)
I agree, but what do I do if I can't find a reliable reference for information I know to be true? Do I then leave that info out because of the policy? Sorry if I seem to be antagonistic, some things in this policy just don't sit quite right for me. Nick carson 07:21, 29 October 2006 (UTC)
A rewrite is being undertaken at the moment which will hopefully deal with that issue. However, even under the rewrite, it's best not to add your personal knowledge over any issue that is likely to be challenged, and if another editor challenges an edit, you must supply a source for it — unless the person is being completely unreasonable, in which case ignore them; but I do mean "completely unreasonable." :-) SlimVirgin (talk) 07:24, 29 October 2006 (UTC)
Best I go and find out what constitutes a reliable resource then! Glad to hear that issue is being delt with, and thanks for your response to my questions and comments, is it just me or does it seem like critique and feedback are hard to find on Wikipedia? I get you :) Thanks again! Nick carson 07:37, 29 October 2006 (UTC)
It can be hard to get feedback on these issues, in part because we're asked questions like these a lot, and so sometimes people get a bit weary of answering them. However, reasonable questions like yours are always welcome. If you have problems getting answers in future, you might want to try the village pump. Cheers, SlimVirgin (talk) 07:47, 29 October 2006 (UTC)

Policy dispute regarding primary sources and OR (requested comment)

Cross posted to WP:RS, WP:OR and Wikipedia:WikiProject Medicine. There is a dispute going on at Talk:Depo Provera#Disadvantages and side effects WP:NOR violations. One user found primary sources, reliable journal articles, that said depo provera may do certain negative things. Another user is disputing this, citing original research. The argument, I guess, is that the 1st user is interpreting the source, and drawing conclusions not published anywhere else. The studies only dealt with rats (not humans), and there is nothing in the patient drug information (or any other way to verify the claims outside of the studies in question). The logic goes that making a connection that a study dealing with rats may effect the use of this drug in humans as a contraception is original research. Furthermore, it may be pushing a POV that this drug is unsafe by mentioning these studies (that are not verified outside of the individual study, and not mentioned in patient drug information). I feel like I am repeating myself, sorry. The counter argument is that a) citing primary sources is a good thing b) the claims are cited and verifiable and reliable, fulfilling almost every wikipedia criteria for inclusion. So I guess there are two issues. Is using the information in this manner original research? And is it giving undue weight to a minority view by citing obscure studies like this? Sorry if I am missing anything or misrepresenting a side. Please direct comments to Talk:Depo Provera#Disadvantages and side effects WP:NOR violations. Thanks!--Andrew c 03:32, 3 November 2006 (UTC)

Hi Andrew, first of all, the article is not about "Depo Provera use in humans"; it is about the drug in general, and so any information about the drug that is published by a reliable source may be used. Even if it is a study with a minority conclusion, it may still be used. However, it should not be used to draw conclusions that the study itself did not draw. It should simply be described, sticking closely to what the authors of the study said, and going no further. SlimVirgin (talk) 03:41, 3 November 2006 (UTC)

Thanks for the imput. When writing up my summary request for comment, I was thinking that would be a good solution. Don't drawn new conclusions from the study, or make it seem as if the results fit under a category they don't, but instead relay them in a manner that is representative of the study in question (this fixes the OR and RS concerns, but still, there are questions about notability and NPOV). --Andrew c 15:24, 3 November 2006 (UTC)


Um, SlimVirgin hasn't said that the one sentence study summaries in "Disadvantages" exist a category in which they shouldn't.

Also, please see your own comments regarding the herpes rat study, the "long quote" dispute in which the anon previously contested the inclusion of the herpes rat study (only in hostile edit summaries, not on talkpage) on the grounds that the source was "in the opinion of a native american women's health care activist," and would not stop making what could be considered possibly racist and sexist ad hominem attacks against the source until it was pointed out firmly to him that the source cited more than one pubmed ref/study in her biblio--i.e., the very herpes rat study citation being contested now as "OR" you agreed was appropriate in both category and one-sentence summary, and is being brought up again now...

What SV does do, I think, is make the excellent point that an article doesn't have to be a rehash/mirror of a drug product insert or a textbook, nor should it be...?

As far as "notability" goes--this hasn't been brought up before. What are you referring to? (And, as far as I know, "notability" is a concept which refers to BLP.?)

No NPOV argument has been made either--merely an ad hominem attack against me which violates WP:CIVIL and WP:NPA by the anon who uses the string of IPs (the same one who previously made the "native american women's health care activist" ad hominem attack against the source in the herpes rat study citation...)

I think it might be really helpful, Andrew, for you to read all the studies re Depo and STDS, to help elevate discussion/move it back to sources and facts at Depo talkpage--the first cite in disadvantages re Depo/STDs was not added by me, but linked to the others (Depo appears to suppress immunity in general, making users more susceptible to pretty much all of them...) Cindery 19:08, 3 November 2006 (UTC)

C'mon! O.R. is EVERYWHERE!

Take any recent (or not-so-recent) computer game or Japanese comic-book character. Or a pop song. Look them up in Wikipedia. Original research, almost all of them, with one fan adding a fact or asserted fact and the next fan adding to that one. Leave aside all this tallk about "published, refereed research." You will find it lacking in many if not most WP articles. I don't want to be flip, but Get Real. Actually LOOK at any one of the Random Articles in the column at the left; go ahead — choose one. Chances are it will be almost totally based on Uncited Sources. If you find one that's NOT, list it here. Sincerely, and with great good will, GeorgeLouis 06:03, 3 November 2006 (UTC)

This is quite true. It's also true that there are lots of grammar and spelling errors. But all of this only means that there is lots of work to be done, it's not a reason to give up, if that's your implication. By that logic, law enforcement should let all crime go unpunished because there's too much of it everywhere, and healthcare workers should let all disease go unchecked, etc. Crum375 15:48, 3 November 2006 (UTC)
I think the thing to remember is that this is an encyclopedia - not a fansite. If we allow original research to stay then the site will simply be useless for its intended purpose. Personally, I think an all out cull of original research, unverified and unsourced material would do the site a huge amount of good. Ok, in the short term it would mean that the site would be a little more empty, but then we aren't here for quantity - we are here for quality.-Localzuk(talk) 15:55, 3 November 2006 (UTC)
Agreed. Let's not use pokemon as a model for this site. Certainly not as a precedent for opening the door to OR. Most editors here have not even looked at a pokemon article. Chances are if more started looking at them there would be a massive culling of material. However, i can't bring myself to spend time on those article when there are so many more important, in my opinion, stubs and clean up jobs on more academic topics. Fancruft is fancruft, I'm sure people that read those articles on wikipedia will understand why they have so much OR. David D. (Talk) 16:44, 3 November 2006 (UTC)
This is not original research. Most of this information is clearly present in the original game, book, etc. It is not verifiable in secondary sources and it is a serious problem for creating a valid encyclopedia, but it is not the same thing as original research. —Centrx→talk • 17:42, 3 November 2006 (UTC)

I don't think that people should not be allowed to put in original research. I WAS planning to ask Jason Lewis about his potential role when I see him again, but now I suppose it is useless. (I met him at Allen Savage's). Just because information isn't published doesn't mean it shouldn't be in the article. —The preceding unsigned comment was added by Superstormfanatic (talk • contribs) 01:56, 15 November 2006 (UTC)

I'm pretty sure most people will disagree with you. David D. (Talk) 18:44, 15 November 2006 (UTC)
Original research means "using Wikipedia as a journal" -- ie. you're not allowed to use Wikipedia as your first publish. You don't publish things on Wikipedia. Wikipedia is for things published elsewhere. The stuff you brought up usually has been published/discussed elsewhere. "No original research" = "Wikipedia is not a journal". 70.101.147.74 03:10, 26 November 2006 (UTC)

Question about Census data and crime stats

Do census data and official crime stats constitute primary sources? Is it OK to say 'X is the 3rd largest city in country Y, according to the latest census results'. Or is it OK to say that 'city Y has the third highest crime rate according to national crime statistics.' Curtains99 09:11, 7 November 2006 (UTC)

Official government reports are considered reliable, because they have gone through layers of fact-checking. If the relevant report ranks cities by population, that qualifies as a non-controversial factual claim. The primary/secondary source distinction is mostly there to get people to read competent historians, rather than writing their own histories from letters and land grants. If the report does not rank, but you can trivially compute a ranking, the only objection I see is the possibility that you made a mistake, and the nuisance of checking up on it. Robert A.West (Talk) 09:53, 7 November 2006 (UTC)
Raw data would almost certainly qualify as primary sources. For example, you can buy microfilms for 1930 and earlier census sheets filled out by the census takers who went from door-to-door. At the same time, the census bureau compiles the data and analyses them for trends; the analysis may be a secondary source. --Gerry Ashton 19:11, 7 November 2006 (UTC)
You omit to mention that primary sources are fine for, "Descriptive claims the accuracy of which is easily verifiable by any reasonable adult without specialist knowledge." I think that census figures and rankings fall within the central intention of that proviso. Robert A.West (Talk) 19:51, 7 November 2006 (UTC)

Translation(s)

I am intending to translate this page into Urdu for Wikipedia ‎Urdu. In this regard I have one question.

Does this policy cover those situations where a ‎‎new translation is created for foreign words/terminologies for which the language doesn't already have any ‎words?

There are hundreds of scientific words and terminologies in English for which ‎there are absolutely no replacements/translations in Urdu. I have noticed some users at ‎‎Wikipedia Urdu are creating/inventing translations for such words and ‎terminologies. These translations have never been published before anywhere else. I ‎think it is a serious situation should be addressed at the highest possible level of ‎Wikipedia administration as, in my opinion, it is an open violation of No original research policy.

Szhaider 23:52, 8 November 2006 (UTC)

IMO this should be addressed on the UR-WP NOR Talk page and agreed to by consensus, like all policies. My own opinion is that a translation should ideally have a source for the translation (of a word or term), e.g. from a dictionary, or sample translation of scientific papers, if any. In the case of no available source of any kind for the translation, I would go by simple logic and common sense. An alternative would be to insist on English (or some other foreign) words when no source for translation exists. But it boils down to what the consensus of Urdu speaking editors agree to, on the page I mentioned. WP is not about ruling by decree - we are ruled by common sense and consensus. Crum375 00:02, 9 November 2006 (UTC)
I have tried to address this issue but because of one certain member's arrogant behaviour and total disregard for anybody else and the policies, most of the members have backed off (constant contributers are less than 10). I had only one success where he translated a trademark into Urdu and I persuaded him to delete the translation and use the tranliteration instead. Only two members were able to confront him on this recent issue. That is why I need some kind of force or backing by some clarification of policies to gather consenus to stop him from ruthlessly creating translations without any citations. Szhaider 00:16, 9 November 2006 (UTC)
It sounds like you are going about it the right way. If you have one editor who is not cooperating, there are ways to deal with that, through various conflict resolution methods, all the way to ArbCom. I am unfamiliar with the Urdu WP, but I am sure you have those mechanisms in place. I think you need to deal with the situation accordingly. It doesn't sound like a generic NOR issue - more like a specific (but common) content dispute. Crum375 00:21, 9 November 2006 (UTC)

Info on translation

I've just edited the policy to include info on translations of outside text and using translations by Wikipedians of outside text as sources. Specifically, I've said it's out-of-bounds. Edit is here. It's new, but I'm trying to be bold and I think it's a good idea. Even if this is eventually beaten down, there should be a section saying it's ok. This came up over a minor content dispute I'm in about adding an original translation to an article, and to my surprise, I was informed that it's not covered one way or another by this policy.

I really think it should be forboden, as it's no more verifiable than an original scientific experiment is. After all, when you analyze (even without performing) an experiment on your own, you're "translating" the data into a readable conclusion. So why should an editor be able to translate something on his/her own and include it as verifiable fact?

No translation of significant size is going to be identical, similar to how no opinion on global warming statistics, symbolic effect in The Scarlet Letter, or the signifance on the recent swing in power in the U.S. Congress will be identical. Therefore, Wikipedia should rely on verifiable, outside translations only. 66.231.130.70 01:25, 10 November 2006 (UTC)

Link to section is here. Also please don't blindly revert me if you disagree, I've edited a sentence elsewhere that is a useful clearing of of ambiguity. 66.231.130.70 01:28, 10 November 2006 (UTC)

My own opinion is that if there is no translation available by outside parties, then a WP editor can provide it. Others may disagree and/or tweak the translation by consensus. Even if an outside translation is available, that doesn't make it unique - if there are several versions (of something important and relevant) they should all be presented. I can also see a WP editor using a dictionary or equivalent source to add a note that "X can also be translated as Y". IOW, we shouldn't just apply a blanket ban on translation by editors. Of course the original foreign text must always be presented (or linked) as 'primary' evidence in these cases. Crum375 03:16, 10 November 2006 (UTC)
  • I concur wtih Crum. >Radiant< 13:31, 10 November 2006 (UTC)
I am not going in to the argument whether the policy page should or should not be changed. However, I will try to explain the disputes of translation by using a couple of examples.
Einstein's theory of relativity is called in Urdu Nazriae Izaafiat. Nazriae or Nazria is universally accepted and used word for a scientifc theory. All Urdu scholars have used the word Izaafiat for Relativity in scientific terms. One user (who is also an admin there and blocked at English Wikipeida) at Urdu WP is using Nisbiat for Relativity simply because he doesn't agree with the usage of Izaafiat. In the same way all Urdu-speaker and writers have practically accepted the word Computer as an adopted word. All computer related scientific books in Urdu use English terminologies of computer science except for a few. But this particular user at Urdu WP has translated all of computer related terminologies including the word computer which he calls Shumaarinda. This kind of words are nowhere to be found in Urdu dictionaries or any other books, and causing a lot of trouble for other editors who are trying to write scientific articles without unnecessary translations. In addition, his such articles are completely useless for those who are trying to read about any scientific terminology at Urdu WP. Szhaider 18:06, 10 November 2006 (UTC)
I think your presentation of the problem is clear. It seems to me that if there is a dispute regarding the translation of a word, then it should be resolved using the existing WP dispute resolution techniques. Ideally it should be solved on the Talk page, otherwise it can be escalated through mediation all the way to ArbCom (I am assuming there is an Urdu ArbCom). My own personal opinion is that the most common translation be used, as decided by consensus. As a footnote, other translations may be added, if they are supported by verfiable and reliable sources. I would also set up a special 'WP:Translation' policy in the UR-WP to clarify this. But the point is that one editor or admin should not bully the rest. The Urdu community must use the various Talk pages to reach consensus, and ultimately ArbCom if that fails. Crum375 21:08, 10 November 2006 (UTC)

Working through the backlog

I am trying to begin the painful work of working through the backlog of pages tagged as OR. Most of them seem either incorrectly tagged or tagged without any further comments on the talk page. If anyone would like to help out it would be great because I really do feelthat the OR tag should be temporary and not something just slapped on the article because someone disagreed with it or didn't know how to tag it correctly. For example, a lot of pages are tagged as OR when the correct more indicative tag would be to tag it as unreferenced or containing weasel words. The OR tag is a pretty serious one and should be used with caution and an explanation why it was used. MartinDK 15:02, 10 November 2006 (UTC)

Question Regarding Self Publishing

Could someone please define this for me? Is this material someone submitted for publication themself, or a company publishing something themself, etc? Thanks! Q Jenkins 15:57, 14 November 2006 (UTC)

It's when you publish something that you have written yourself e.g. your blog, or when you send a manuscript you've written to a vanity publisher, and you pay them to turn your manuscript into a book. SlimVirgin (talk) 15:59, 14 November 2006 (UTC)

Ok, so that excludes company self published material, like, an instruction manual or a FAQ? Also, I know this probably isn't the place, but how do you do the indent? Q Jenkins 16:03, 14 November 2006 (UTC)

To learn about formatting, just look at what others are doing in the 'edit' mode. In this case, indent is via a colon ':' at the beginning of the line. Regarding the self-published manual, if it's published by the author/inventor or a program or device, it would be acceptable as a source in an article about that program or device, per WP sourcing policy. Crum375 16:26, 14 November 2006 (UTC)
Some companies have a good system to publish high quality publications, and can be considered reputable publishers. Their publications would not be self-published, even if they don't give the name of the authors. This would apply to user manuals for Microsoft programs, for example. Of course, not every publication by a reputable publisher is a reliable publication; letters to the editor of the New York Times are not reliable; publicity and advertisements on the web site of a reliable publisher may not be reliable. --Gerry Ashton 19:43, 14 November 2006 (UTC)
I don't think that you have to be anywhere near MS for your manual to be an acceptable source. Even if you are the smallest software publisher on earth, a tiny garage operation (let's say), if your product has a WP article (i.e. its notability was properly established), the instruction manual could be used as a reference about the software. This is equivalent to our ability to use a self-published biography by an author as a source for his/her own article, as it relates to him/herself. If the claim is controversial, we would have to phrase it carefully. If it maligns someone or something, we may want to avoid it altogether, unless properly sourced elsewhere. So it's not the size of the corporation, it's the way we use the source that is important for self-publications. Crum375 20:10, 14 November 2006 (UTC)
My point is that when an organization uses a reliable editorial process to produce a publication, that publication is not self-published, it is published by a reliable publisher. The parts of the policy about self-published material simply don't apply. As for how big an organization needs to be to be a reputable publisher, it need not be nearly as large as Microsoft, but one person isn't enough. --Gerry Ashton 20:44, 14 November 2006 (UTC)
I think it depends on the situation. Yes, a large corporation would have more eyes going over its publications than a small one or an individual, but there could still be clear conflict of interest, as all those extra eyes are still paid by the corporation, hence not necessarily neutral. So it depends on the specific item that is being sourced from the publication, and its controversiality or lack thereof. In any case, size is not a determining factor. A couple of layers of very few neutral editors would trump a brigade of loyal company employees towing the corporate line. Crum375 20:57, 14 November 2006 (UTC)

Translating the Policy page

I am trying to translate this policy page in to Urdu for Urdu WP so that I, as an admin there, can begin to force it specially on those who are using WP for publishing their own theories and agendas. However, everyday when I begin to trnaslate a new paragarph I notice that previous translation is out-of-date because some one has already changed the policy page. Can an admin point me to the final and finest form of this page. As it is an official policy, every word should be carefully translated which is becoming hard with daily changes in the project page. Szhaider 05:07, 16 November 2006 (UTC)

I apologize that the policy appears unstable; it should be stable. The current version is fine. Any small changes that occur will not affect the meaning. Good luck with the translation! :-) SlimVirgin (talk)
Actually that version isn't stable, variations are under debate here and here. In particular, I don't think a literal translation of this or any policy page is required for an other-language Wikipedia, as long as the meaning comes across. Most policy pages on the English Wikipedia have grown verbose and convoluted over the years, and a starting project could do with something much simpler. (Radiant) 10:36, 16 November 2006 (UTC)
You are right about the complex language of this page. That's one of big hindrances in translation. I have to carefully choose Urdu words which are not literal meanings but have same contextual meanings and uses. Degree of violations at Urdu WP needs a comprehensive set of policies and I think policies at English WP are quite comprehensive, and can be practically implemented, at least, at Urdu WP. Szhaider 18:18, 16 November 2006 (UTC)

Synthesis of published material serving to advance a position

The NOR rule has been invoked to try to disallow inserting comments which are direct quotes from reliable sources. There is a big difference between advancing an argument about the Chicago Manual of Style and noting that "the Chicago Manual of Style defines plagarism "using a source's information, ideas, words, or structure without citing them"" NBeale 10:41, 16 November 2006 (UTC)

Specifically, it has been suggested that if fact A is in an article and fact B is added then this "serves to advance the argument (cleverly left implicit)" However this would be an argument against inserting any facts into an article since adding any fact can be seen as advancing an implicit argument. So with some trepidation I've tried to make this clear in the article. What do people think? NBeale 17:27, 18 November 2006 (UTC)

As noted in my edit summary, I think that by WP adding B it is editorializing, possibly violating NPOV and NOR. Let a reliable third party add the B and then quote or cite it. When we do it ourselves, we are no longer a neutral observer, which is our mandate as an encyclopedia, but a participant. Crum375 18:06, 18 November 2006 (UTC)
Hi Crumb375. It's clearly possible that adding B might violate NPOV (esp cumulatively). But I don't see how adding a fact can be OR. If so then the whole of Wikipedia is OR, which surely can't be right. (and if "X says B is relevant" becomes a fact then adding that would be OR on this reading). What do others think?NBeale 19:02, 18 November 2006 (UTC)
The mere juxtaposition of independent facts can easily convey a POV and effectively be OR. In a controversial WP:BLP case, where we need to be extra careful to maintain neutrality and use only high quality sources, by adding our own 'facts', well sourced on their own but independent of the specific entry, we can easily convey an editorial message which would be OR and POV. For example, say we have an article about a notable person who apparently misprepresented his age in some government document X some years back. If we add the statement, "In the state of Virginia, it is a felony to knowingly misrepresent one's age on form X [1][2][3]", we are in effect saying: "And we, WP, think this person is a felon, per our research and understanding of applicable Virginia law". OTOH, if we include a newspaper article saying the same, then we are just a neutral observer, as we should be. Crum375 19:20, 18 November 2006 (UTC)
Crum375 is right about the juxtaposition of facts. Material added to articles should have been mentioned by a reliable source in relation to the topic, if it serves to advance a position.
For example, suppose we have an article about a scandal involving a politician having an affair with an actress. Reliable newspapers are alleging the actress is or was a prostitute. Then let's suppose a Wikipedian discovers a little-known published biography of the actress from when she was first starting out. It notes that, while working as a waitress in Hollywood and going for auditions, she was often seen around town in her pink Ferrari. None of the newspapers writing about the prostitution scandal have noticed this biography, but the Wikipedian who found it adds the pink Ferrari detail to the article.
This would not be allowed, because it raises the question "How could she afford a Ferrari if she was working as a waitress?", and that advances the prostitute position. Even though it's well-sourced, we would have to wait for a published source to mention it in relation to the prostitution scandal before we could do the same. SlimVirgin (talk) 23:57, 18 November 2006 (UTC)
Hi Slim. I don't think this example quite works (far more likely to be a gift from an 'admirer'), but suppose it was "police suspected she was a prostitute, but could never prove anything." Then I think you could argue that it would be wrong to include this, but this would be under NPOV and BLP. Finding information from reliable sources and then including it cannot be OR because if it is then the whole of Wikipedia is either poorly sourced or OR. NBeale 06:44, 22 November 2006 (UTC)

OR vs. deficit of research

Hey everyone.

Just wanted people's opinion on an issue about WP's OR policy. I am currently preparing an article on Erdheim-Chester disease. Unfortunatly, due to the rareness of ECD, there is no widely accepted treatment. I have presented the treatments that have been tried, with varrying levels of success. There is also some case studies available that discribe various treatments and their levels of success for a particular individual. This could be considered OR, however, I feel that it would be useful to the reader to add, for example: "Two patients were reported to respond to prolonged therapy with vinblastine and mycophenolate mofetil (Jendro et al., 2004)." Is there a way to disclaim that treatements for ECD are still under research and there is no consensus amongst physicians? I'd prefer to add the information and disclaim it, than simply leave it out. Let me know. All the best! --JE.at.UWOU|T 17:04, 16 November 2006 (UTC)

I don't think that anecdotal information per se, when reported in an otherwise reliable and verifiable source, is original research. Of course the information has to be properly presented as anecdotal and/or of limited statistical significance, and the conclusions cannot be any more than what is in the source. Crum375 17:17, 16 November 2006 (UTC)
Just to clarify, although anecdotal evidence per se, when reliably and verifiably reported, is not automatically excluded, the usual strict requirements for scientific or medical sources, such as reputable peer reviewed journals, are still in place. And these strict sourcing requirements are even more stringent if the scientific or medical claims are 'exceptional'. Crum375 18:43, 16 November 2006 (UTC)

Newspaper accounts

Newspaper accounts are primary sources. An analysis of an event, on the other hand, is a a secondary source. From secondary sources (my highlight):

An example of a secondary source would be the biographyof a historical figure which constructed a coherent narrative out of avariety of primary source documents, such as letters, diaries,newspaper accounts, and official records. It would also likely utilizeadditional secondary sources (such as previously-written biographies)as well. Most, but not all, secondary sources utilize extensive citation.

I corrected acordingly ≈ jossi ≈ t • @ 00:10, 17 November 2006 (UTC)

An example:

A 1941 newspaper account of the Perl Harbor attack is a primary source. An comparison of the attack on Perl Harbor with the 9/11 attacks, published in a newspaper or magazine are a secondary sources. ≈ jossi ≈ t • @ 00:12, 17 November 2006 (UTC)


This policy should be deleted

The fact that we must paste from other websites makes Wikipedia look like a giant plagiarizing bulletin board rather then a Encyclopedia. The fact that some 12 year old who posts from an article based off some news blog is considered more reliable then a college professor is an insult to all definitions of knowledge.

There are plenty of well educated people making great contributions that get reverted simply because they are not plagiarized. Also, it seems as if there is a secret rule stating that any fact not found on Google must be a lie. Not to mention the sheer amount of information that is not on the web; but in Books, Labs, and in real life. This rule is entirely biased.

—The preceding unsigned comment was added by Roxanne Edits (talk • contribs) 00:46, 17 November 2006 (UTC)

Blogs are not acceptable as reliable sources but books are. See WP:RS for details. Curtains99 14:22, 17 November 2006 (UTC)
Also, it seems you are a little confused about what we require for information to be included. Information simply has to have been published by a reliable source to be able for us to summarise its content - this is not plagarism, it is summarising it. Can you not see the purpose of this rule?
I agree though that there is a over-obsession with google checking to see if something is a factual piece of information. However, we shouldn't need to check if a source is provided in full.
You should check out our citation policy, verifiability policy and our neutrality policy. Hopefully these will explain things a little better.-Localzuk(talk) 15:38, 17 November 2006 (UTC)
You seem to be unfamiliar with what an encyclopedia actually is. You also seem to be confused about what plagiarism actually is. --Mr Wednesday 22:37, 17 November 2006 (UTC)

Image:Vineland 55-60.jpg

Are my statements at User talk:Route 82#Re: New Jersey Route 60 correct - that a detailed map of a route that was only proposed in general terms is just as bad as a detailed description? --NE2 00:56, 17 November 2006 (UTC)

This edit that you removed is certainly original research if no source can be found for it. As for the map, it depends on who produced it. SlimVirgin (talk) 01:00, 17 November 2006 (UTC)
He made it; he has a whole "fanfic" about the highway that he was adding as an external link. --NE2 01:07, 17 November 2006 (UTC)

Newspapers

Jossi, a newspaper story is a secondary source, unless it's an old one, in which case it acts as a primary source about the period. Regarding recent stories, an eyewitness's statement about a traffic accident is a primary source. A newspaper's report about that eyewitness's statement, and about the accident in general, is a secondary one. SlimVirgin (talk) 01:51, 17 November 2006 (UTC)

Sorry, but I do not follow the logic. Could you explain why time makes a difference? My understanding that wires and newspaper reports of current events, if these do not contain any kind of commentary or analysis, are all primary sources, regardless of time horizons. ≈ jossi ≈ t • @ 02:27, 18 November 2006 (UTC)
Some examples:
  • Primary sources are records of events as they are first described, without any interpretation or commentary. [6]
  • [primary sources] may also include published pieces such as newspaper or magazine articles (as long as they are written soon after the fact and not as historical accounts) [...] Secondary source materials, then, interpret, assign values to, conjecture upon, and draw conclusions about the events reported in primary sources. These are usually in the form of published works such as journal articles or books, but may include radio or television documentaries, or conference proceedings. [7]
≈ jossi ≈ t • @ 02:34, 18 November 2006 (UTC)

Reliance on primary sources

The policy provides (apparently as guidance rather than a hard and fast rule) that "edits that rely on primary sources should only make descriptive claims that can be checked by anyone without specialist knowledge". This guidance is regularly, and quite correctly not adhered to. For example, a firsthand report of a sports match published in a newspaper is regularly used to evidence the score and what happened in the game, even though those without the specialist knowledge of that sport would not understand it. From a primary source saying that the result of a Gaelic football match was Team A 2-13, Team B 1-17, it is perfectly reasonable for those writing on Gaelic football to conclude Team B won - yet that's not clear at all if you know nothing about Gaelic football. From this scorecard it is reasonable for someone who knows about cricket to conclude that Jim Laker had an absolutely amazing match. But someone unfamiliar with the game may question why that is so if he only scored 3!

I suggest changing the text so that it better reflects actual (quite reasonable) practice. Maybe "edits that rely on primary sources should only make claims that can readily be deduced by anyone with specialist knowledge in the field"? jguk 17:43, 19 November 2006 (UTC)

Perhaps Jguk is right, that anything but a direct quotation could be impossible for a non-specialist to compare to a source in some fields. But if this change is made, then the rule is the same for primary, secondary, and tertiary sources, so just drop the whole sentence. --Gerry Ashton 17:50, 19 November 2006 (UTC)

Splitting hairs here? I do not think that anyone will dispute the score of a football match as reported by a newspaper on any grounds of NOR... ≈ jossi ≈ t • @ 18:04, 20 November 2006 (UTC)

I know in practice no-one's going to stop our articles using primary sources that require some knowledge to interpret them from being used. It would wipe out too much good work that's going on. Plus it's somewhat perverse to say, yes we have a primary source that is reliable and easily understood by those who are knowledgeable in that field, but as someone who knows nothing about the subject in hand won't understand it, we won't use it. It's just that the sentence "edits that rely on primary sources should only make descriptive claims that can be checked by anyone without specialist knowledge" does not reflect what current (and past) WP practice is. It should either change so that it does reflect practice (as I suggest), or be deleted in full (as Gerry Ashton suggests). jguk 18:53, 21 November 2006 (UTC)
Sorry, but I disagree. The formulation reflects current understanding about primary sources. ≈ jossi ≈ t • @ 19:09, 21 November 2006 (UTC)
You are saying that if we have a primary source that we have assessed as being reliable for a claim that we decline to use it if someone totally unfamiliar with the subject will not be able to understand it, and yet if we have a secondary source that we have assessed as being reliable for a claim we are quite happy to use it even though someone totally unfamiliar with the subject will not be able to understand it? I really don't see this distinction being made anywhere on WP (and indeed a large proportion of WP:Cricket's articles would disappear if this approach was enforced as people totally unfamiliar with the sport will not be able to understand scorecards, the laws of cricket as published, press releases from the International Cricket Council, press (ie eyewitness) reports of matches, etc. etc.) jguk 19:16, 21 November 2006 (UTC)
You are missing the point. Please re-read (my highlight): "should only make descriptive claims that can be checked". It is not about the ability to understand, rather, about the ability to check. ≈ jossi ≈ (talk) 00:22, 22 November 2006 (UTC)

YouTube art as primary source

Someone made a trailer--an original work of art--for a film about a guy named Brent Corrgian. A guy who appointed himself the YouTube police tagged if for copyright verification. When the author verfied authorship/licensing, the YouTube police guy said then it's OR and still can't be used. I say it's a primary source and fine as an external link; original art in which the subject of the article appears advances no position and is in no way OR. (YouTube-police-guy might have had a vanity argument, except that the link has now been posted by someone else--me--and I don't know the author. Or a commercial argument--except that the film isn't finished and has no studio distributor, ad budget, marketing plan or source of revenue--the trailer at this point is just a stand-alone original work of art on the indie-film level, which enhances the quality of the article as a relevant, valuable, interesting external link.) Please see: [8] Opinions? Cindery 21:35, 19 November 2006 (UTC)

NOR concerns aside, there is no argument that the link actually adds anything of value to the article. ---J.S (t|c) 02:34, 20 November 2006 (UTC)

According to the view-meter, it's been viewed more than 10,000x. That's 10,000 vs. one--I'd say you're outvoted by a rather large consensus about whether it has any value. But, this is the NOR board. You autocratically informed the author of the link that it was orginal research--after telling him or her that if they objected to your deletion of YouTube links, that they should post on your project talkpage so "experts" could review their objections. I think the NOR talkpage is a better place to review OR accusations. Fellow editors can discuss, weigh-in; not a single self-appointed "expert" who refused to even discuss the matter with the author. Cindery 03:01, 20 November 2006 (UTC)


If it is a primary source or not, is irrelevant. YouTube videos are not reliable sources for anything. ≈ jossi ≈ t • @ 18:03, 20 November 2006 (UTC)

Sure they can be. A youtube copy of the original Where's the Beef commercial would be a reliable primary source. It's easy to think of dozens of others. SchmuckyTheCat 19:24, 20 November 2006 (UTC)
No they cannot be as these are not verifiable to be "originals", unless these have been uploaded by the creators themselves. Today, anyone with free editing tools can add, remove, and modify video content too easily. Only these youTube uploads that have been uploaded by the copyright holders can be assured to be originals. Theses uploaded by fans, or amateurs that ripped these videos from DVDs, off-air TV, etc are not reliable to be used in Wikipedia. ≈ jossi ≈ t • @ 19:57, 20 November 2006 (UTC)

The Brent Corrigan video was uploaded by the original author, and he or she vouched that every element of it is his or her original work--hence the OR accusation. (As for YouTube in general as a reliable source, there is the issue of substantiality. For example, ripping a YouTube clip of 2 minutes of a 70 minute documentary is very arguably fair use. There are several legal elements to a copyright vio case. It is highly likely that in if a tiny segment of an oscure political documentary--or an old commerical--is ripped into a YouTube clip, that the copyright holder would be happy for the exposure, and hence deleting-on-source-bias would serve no one. In additon, "fair use" and sampling are very big issues right now in alternative media. "The elements can be manipulated by anyone" is a disturbingly oversimplified argument--there's nothing inherently wrong with the fact that the medium is manipulable by the masses, and Wikipedia would lose a lot of valuable links/enriching sources if a source bias persists. Rather than a blanket bias against YouTube as a medium, the links, when contested, should--like blogs and personal websites--be decided on their individual merits. If, for example, somone taped all the old Max Fleischer cartoons off of tv--that would be great. They aren't under copyright. If someone ripped 10 minutes of Alexander Nevsky and subbed in their own original kazoo composition as audio, that would be great--Nevsky isn't under copyright either: "Ten minute clip of Alexander Nevsky, with original music by <blank>," etc. For anyone who knows anything about film, YouTube is an amazing delivery medium--all kinds of old 8mm and 16mm film people have personally shot can be digitized and YouTubed merely for accessibility--like photos taken by Wikipedians, these are invaluable visual resources, which editors have to offer under GFDL.) But, to bring it back to OR-- I see a problem with insisting that the author vouch that it is not a copyright, and then when he or she does, accusing them of OR--so, if someone else posts the link, it's a potential copyvio because YouTube is the source, and if the original author posts it, it's OR--that's a catch-22, not a means to verify the reliability of a source... Cindery 22:14, 20 November 2006 (UTC)

Fair use has its problems as fair use is dependent of jurisdiction. Nevertheless, I was referring to the probability of spoofing and/or tampering with video content. So, editors need to exercise caution, in particular if the issue is contentious or controversial. As for old 16 mm videos, these would be primary sources and thus these should be also be used with caution, and only for non-controversial material (i.e. a movie of the Red Square in Moscow, for example), and as external links and not as sources. ≈ jossi ≈ t • @ 22:42, 20 November 2006 (UTC)

"Fair use has its problems" is better phrased/understood as "fair use has its complexities"--which is an argument in favor of caution and discussion, as opposed to deletion-on-source-bias (deleting because the source is YouTube). "Spoofing or tampering with video content" could also be better phrased. Altering content, such as ripping a clip, making a collage, or subbing audio, is not at all necessarily bad. "External links" is primarily what we've been discussing, but video can be a great cited source, especially for places and people that no longer exist. See: Barrington Hall--three floors of murals were wiped out/painted over in 1989. Only one 8mm film exists of the murals. As visual documentation that something existed, the film isn't just a good source, it's the only source--no photos capture the building in long takes, showing that it was a building, that one mural bled into another for a city-block. In the current ongoing purge of YouTube links, no one seems aware or has even mentioned that video can exist as photographs do--totally noncontroversial excellent sources. Original photographs are just as manipulable by authors as video--I can photoshop in whatever I want, etc., and yet we appreciate these photos, and are not purging them en masse because they are original works which could have been manipulated without establishing that they were manipulated in some way which undercuts their validity as sources. Cindery 23:47, 20 November 2006 (UTC)

Sure, but there are significant differences between photos uploaded to commons or to Wikipedia, and videos uploaded to youTube. The former go through a process of validation and verification and are uploaded under the GDFL. We control what is uploaded and what is not uploaded, what can stay in WP and what is removed and we do so via the WP:IFD process.
In YouTube we do not have such control. Uploaders can add, remove and modify content at will, and there is almost no possibility of validating if the uploader has rights or not. So, sure, we may use YouTube videos in external links, but mots certainly we cannot allow YouTube videos to be used a sources, as in most cases these will be in violation of WP:V, WP:NOR, or WP:COPYRIGHT. ≈ jossi ≈ t • @ 00:34, 21 November 2006 (UTC)

No, I think you're just making an erroneous conflation, based on assumptions, that the lack of control over the YouTube= "most cases will be violations." A lot of them could be, but the source itself does not equal violations, and a source bias is harmful because it implies to editors that the problem is the source, not the potential V/NOR/C policy issues. People shouldn't be discouraged from using such an incredibly useful medium; they should be encouraged to evaluate the use of YouTube as link or source based on policy. Mass deletions of YouTube links don't do much to raise awareness about policy and how to make sure useful links can meet policy, they just imply that there's something inherently wrong with YouTube, and there's not. Any copyright holder with an objection to use of any image used on YouTube has clear redress to both YouTube and Wiki--aggressive before-the-fact purging isn't necessary--the purge should be less aggressive/is erring to far on the deletionist side, and is giving people the impression that YouTube can't be used, not that YouTube should be used judiciously. In the Corrigan case, I see no OR issue--what I see is source bias. Cindery 01:15, 21 November 2006 (UTC)

You are not addressing the concerns raised, Cindery. I have no problems, if used with caution, to use YouTube in the external links of an article, if pertinent, and if anything, editors should err of the side of deletion rather than inclusion, unless there are no doubts about the origin and accuracy of the link. The main point, though, is that YouTube videos should never be used as a source to support a claim in an article, unless it is absoluelty certain that there are no copyvio issues and the material is highly relevant and not available from another, more reliable source. So, what is your position pertaining External Links vs sources for an article? ≈ jossi ≈ t • @ 01:59, 21 November 2006 (UTC)

That's funny--my feeling is genuinely the opposite; that you are not addressing concerns raised. Perhaps it clarifies matters to state that a template in is in wide circulation stating that "99% of YouTube links" should not be used (and the most conservative--and biased/unscientific-- estimate of the person who is conducting the purge is 90%...) "Editors" are not erring on the side of deletion--a small handful of people are deleting all the YouTube links without careful evaluation or discussion on talkpages of articles in --they have stopped even posting a warning template (with an exaggerated biased number of 99%).

I think I have made it clear that my position on YouTube links as sources is that they can be invaluable as visual documentation, as photos are. (But I see YouTube overall as potentially useful external links, as blogs and websites can be.)

YouTube videos should never be used as a source to support a claim in an article, unless it is absoluelty certain that there are no copyvio issues and the material is highly relevant and not available from another, more reliable source--there is no disagreement here (except with "more reliable source")--the disagreement pertains to how this is being established. Talkpages of articles are appropriate; mass deletions by people who do not participate in article talkpage discussions/don't know anything about the subject or the link are not. A less biased warning template should be used, and discussion allowed. If there's no discussion after 2 weeks or so, I don't see a problem with deleting the link; and I don't see a problem with deleting a napster-ish link like a whole recent popsong without warning. The problem is that the YouTube purge is over-deletionist and fairly hostile--it is not educating editors about how to use YouTube per policy, it's just giving them the idea that YouTube shouldn't be used, and that is both incorrect and harmful. Moreover, reasonable objections to deletions--such as, the owner has copyright and is willing to license link under GDFL!--are being ignored in favor of trying to delete all YouTube links period. "More reliable source" is just source bias. YouTube offers stable storage and easy accessibilty for film/video--it's not any less or more reliable as a storage/delivery medium than a website. Cindery 02:26, 21 November 2006 (UTC)

Sorry, but I don't follow. How can you upload content to YouTube and tag it as GDFL? There are zero checks and balances in YouTube, and there are no admins or editors that can evaluate the content and tag it appropriately. I am not surprised that there is an effort to remove links to YouTube material. If there is any material that needs to be included, let those that want it carry the burden to argue for its inclusion. (I have not seen a centralized discussion on YouTube as a source. Is there such a discussion?) ≈ jossi ≈ t • @ 02:44, 21 November 2006 (UTC)
YouTube cannot be used as source per WP policy. Crum375 02:51, 21 November 2006 (UTC)
OK, I now realize that WP:RS is just a guideline. Live and learn. Crum375 03:03, 21 November 2006 (UTC)

Exactly--just because you can't upload content to YouTube and tag it as GDFL doesn't mean you can't release it under GDFL (and one would assume that if the author uploaded their own YouTube work to Wiki, they were licensing it, since editing Wikipedia is done under the agreement "You agree to license your contributions under GDFL." ) Establishing that the Wiki publisher and the YouTube publisher are the same, and/or that the Wiki publisher has copyright/author permission/knowledge should be established via discussion, not mass deletions which imply the problem is YouTube, rather than policy. As far as I know, this is the beginning of centralized discussion of YouTube as a source (And sorry it has gone off-topic per OR--it should be discussed on all the policy boards, though, esp RS and C.) The problem with "burden to argue for inclusion" is that the mass-deleters are deleting even when there are valid objections, and directing complaints to their own talkpage, not policy boards or article talkpages, where they claim they are "experts," and tell people, for example, that if there's no copyright issue it's OR...A lot of valuable external links and sources could be lost that way. Cindery 02:58, 21 November 2006 (UTC)

Look, I think that YouTube is a great technology and an great social experiment. But that has nothing to do with the arguments presented. I still do not see how we can validate the content posted in YouTube. Same as we cannot validate content posted on a blog, personal webpage or online forum. No difference. As such, YouTube videos, and not reliable sources to support any type of claims in an article, and if used as external links, these should be evaluated on the basis of appropriateness, and lack of copyvios by involved editors. I am not concerned about "mass deletions". Any link that is good will be surely sooner or later added back if it fits the criteria. ≈ jossi ≈ t • @ 15:31, 21 November 2006 (UTC)

We can validate the content used in videos the same way photo content is validated--case by case, based on knowledge of the subject. Videos--whether they are linked to from a website or YouTube (and there is no difference except convenience and stability of storage)-- can be cited as sources to verify that someone or somthing existed; that something happened; that something changed or is gone but was once there. Stop-motion photography can illustrate how something happens, like cell division..."He read at the Bowery Poetry Club in 2006" can be footnoted with a video of him reading at the Bowery Poetry Club...

Any link that is good will be surely sooner or later added back if it fits the criteria.--you're ignoring the original point made--when the author of the Brent Corrigan video verified the GDFL license and added back the link, the same person deleted it and told him/her it was OR. He deleted it after I noticed what was going on, and didn't stop until I said I was crossposting to NOR. He is not objecting to the use of the video as a source--he's deleting an external link on the grounds that it is OR (after telling the author to go to his page, not a policy board, for "review" by "experts" if there were any objections to this. It's fairly obvious that the person who posted the Brent Corrigan link thinks the deletion, the "No YouTube 99% of the time" template, and the review process by "experts" is "official"/Wiki policy--and it's not. That certainly makes adding back links problemmatic. The YouTube purge is not being conducted according to policy. If there is an urgent need to mass-delete YouTube links, policy should address that.) Cindery 17:14, 21 November 2006 (UTC)

As I noted above, use of YouTube as a source is contradictory to Wikipedia's Reliable Sourcing guideline. The reason is that there is no editorial board that controls the validity of the YouTube contents and stands behind it. Anyone can edit and modify anything on that site any which way s/he wants, hence it is not considered reliable. Contrast that with a news media outlet, that strictly controls all its contents, and is legally liable for it. The non-reliability includes both false/misleading/bogus information, as well as potential copyright violations. If the relevant WP site is WP:BLP-related, IMO anyone can remove any YouTube link with no dicussion necessary (an edit summary would be nice), with no WP:3RR limit. Crum375 17:42, 21 November 2006 (UTC)
Using that logic, shouldn't we disallow any kind of fair use image uploading to Wikipedia? After all, we cannot know whether an image that was uploaded was manipulated in any way by the uploader. The only thing we could do is find the source and compare. Exactly the same is true for videos on YouTube, and I'm having trouble seeing the difference. Don't get me wrong, I see that this could be a problem, but this looks like a rather rare one to me. --Conti|✉ 18:14, 21 November 2006 (UTC)
I agree fully that any WP image uploaded by a user can be challenged at any time by anyone, and should never be relied upon as a source. We never know what modifications an editor has done, and an anonymous editor, or even a normal mortal non-anonymous one, cannot be a 'reliable source'. Crum375 18:36, 21 November 2006 (UTC)

Fox news certainly "edits and modifies" the news any way they see fit. :-) But, the point is that there is no difference between uploading a video to a personal website and uploading it to YouTube--the source is not the issue. Establishing that the source is valid/use of the source is valid has to do with policy, not YouTube as a source; the same policies apply to any use of a source as apply to use of YouTube as a source. (And personal publishers are just as liable as whole news outlets; lack of legal accountability is not relevant). Deleting YouTube links on sight from BLP articles because they are YouTube links is not a good idea--a query should be placed on talkpage, unless you know enough about the subject to know that the YouTube link is or contains false and defamatory information. There is, for example, a YouTube link of Joshua Clover reading at the Bowery Poetry Club in 2006. The author is happy it's there; the subject is happy it's there; readers are happy it's there. No one who was at the reading--including the subject and the author-- believes the link is anything other than a completely unmodified recording of the reading. In the absence of any complaint from the subject, the author, regular editors of the article, or any editor who happens by and thinks the video of the reading is somehow a BLP violation, removing it simply because it is a YouTube link would be idiotic, and fighting to reinstate the link would waste the time of the subject, the author, and the regular editors of the article. If there is some need to verify all YouTube links in advance, that should be clearly addressd in policy--for example, by requiring that YouTube authors post on talkpages of articles "I license this under GDFL" after they have posted a Youtube link, or someone else has posted their link. Everything that is put into an article is subjected to scrutiny under all policies. Using YouTube is not a de facto policy violation--so unless there's a policy violation problem with a YouTube link, YouTube links should not be deleted just because they are YouTube links. Cindery 18:17, 21 November 2006 (UTC)

In the absence of any challenges, a YouTube link may be used as an external link, but 'never as a source to substantiate a claim. ≈ jossi ≈ t • @ 18:23, 21 November 2006 (UTC)

That is flatly untrue, per RS--YouTube can be used as a source, because film and video can be used as source; it just usually isn't--like blogs and personal websites, YouTube is more oft used in external links, but there are exceptions--and they should be clearly delineated, with examples, in policy. I'm still waiting on any discussion regarding whether use of a (GDFL-verified) external link wholly owned by the author is OR. (clearly, it's not OR. Like blogs or websites relevant to the subject of an article, a YouTube external link about the subject is not OR). Cindery 18:48, 21 November 2006 (UTC)

This is not about WP:RS, but about WP:V that is policy. YouTube is not a verifiable source, neither is a reliable one for the arguments already made. GDFL is not applicable to YouTube as there is no feasible way to verify that the content uploaded is owned by the uploader. Policy is not designed to address exceptions, but the rule. Exceptions can be discussed on a case-by-case basis. ≈ jossi ≈ t • @ 19:07, 21 November 2006 (UTC)

No, none of that is true. YouTube can be used as a primary source, and V delineates exceptions. And GDFL is applicable to YouTube. The feasible way to verify GDFL is to query on the talkpage of articles about whether the YT/Wiki publisher were the same, and/or if the Wiki publisher has GDFL permission from the author and can attest to that. Assuming that the GDFL license is not in effect because the source is YT is in direct violation of C policy: "It is not the job of rank and file Wikipedians to police copyright violations"--thing to do is query on talkpage or make a report at Copyright problems.

Per Rs, this is all there is about YT:

YouTube

Some concerns have been raised about the use of YouTube as a source. YouTube is a website where the contributors are unknown, and in which material that may be useful to Wikipedia articles is almost always suspect of copyright violations. As such, linking to video content in YouTube should almost always be avoided as a source.

Per V, this would apply to self-published YT:


Self-published sources in articles about themselves


Self-published material, whether published online or as a book or pamphlet, may be used as sources of information about the author, so long as there is no reasonable doubt who wrote the material, and so long as it is:

  • relevant to the self-publisher's notability;
  • not contentious;
  • not unduly self-serving or self-aggrandizing;
  • about the subject only and not about third parties or events not directly related to the subject

Cindery 19:22, 21 November 2006 (UTC)

I will try again to explain my arguments in the context of your comments above:
  1. There is no feasible way to ascertain that the uploader is the owner of the material, thus GDFL is not applicable (you have to be the owner of the material to make it available as GDFL);
  2. Self-published sources: There is no feasible way to to ascertain the identity of the uploader, thus self-published sources guidelines are not applicable for videos uploaded to video sharing sites such as YouTube, Google video and others;
  3. Although It is not the job of rank and file Wikipedians to police copyright violations, also note that If some of the content of a page really is an infringement, then the infringing content should be removed.
≈ jossi ≈ t • @ 19:49, 21 November 2006 (UTC)

..the feasible way, per copyright policy, is to query on the talkpage of the article, not delete the links on source bias. If some of the work really is an infringement, such as a recent pop song, that would be separate from the problem of establishing who the publisher is/GDFL. If there's a question about copyvio, such as GDFL, copyright policy clearly states that a query should be placed on talkpage and/or a cr report made. I think V/NOR/C and RS should all be updated to address YouTube a little better; and that it should be made very clear on pages where the mass-deleting is being done, via template, that "No YouTube" is not a Wiki policy, but that use of YouTube is covered under extant policy. Cindery 20:01, 21 November 2006 (UTC)

We have some agreement here, Cindey, at last! If the issue is a copyvio challenge of a YouTube video in the external links section, of course it needs to be discussed in talk. But your other arguments, such as the GDFL and self-published sources, do not stand. ≈ jossi ≈ t • @ 20:12, 21 November 2006 (UTC)

While I am extremely confident and hopeful that we will reach agreement, I still think we're missing each other--the problem with the mass purge isn't at all the purge of clearly suspected copyvios, like whole pop songs (a lot of YT links are whole popsongs/music videos, and I see no problem with delete-first let the publisher justify it, and I think the YT purge is very helpful to Wiki in the thankless task of getting rid of those. It would be a waste of everyone's time for talkpage discussion to go down about those--edit summary more than sufficient.) The problem I see is specifically with the no-GDFL assumption, and no explanation about how to affirm GDFL license. They're mass-deleting with the edit summary: rm per EL; no licensing info--that's the technicality they're using. Talkpage template and discussion should happen for cases where the only suspected copyvio is a question about authorship/GDFL. I don't even see the need for mass deletions for that reason--better to wait for complaints. It makes a lot of busywork for everyone, and leaves editors confused--either they think YT is not allowed period, or they're left asking, well, how can I license it under GDFL? And I see a problem with the bias in the YT purge--for example, instead of explaining, oh, just verify GDFL on the talkpage, the project finds another excuse--like OR--to try to purge Wiki of YouTube. If there's a huge GDFL prob with YT, that should be explained in policy for editors: put a talkpage notice giving license, etc. Cindery 20:29, 21 November 2006 (UTC)

Can you explain how can anyone ascertain GDFL of content uploaded to YouTube or Google video? I do not see that as a possibility, Cindery. If such a possibility existed, then I could agree with you. ≈ jossi ≈ t • @ 20:40, 21 November 2006 (UTC)

GDFL can be ascertained by affirming that the YT publisher and the Wiki publisher are the same, or that the Wiki publisher has the permission of the YT author, if they are not the same. The problem is: because it is not possible to look at YT link on Wiki and determine that the author of YT and Wiki is the same, there can be questions. Since it is not known either way, it is not a clear copyright suspicion--it's a question which should be asked. In order to help editors answer this question, a clear explanation should be provided on the talkpages of articles where the question is asked, and in policy. Say, for example, that I want to publish a YT external link. I do so. It is purged with no talkpage discussion or explanation--just "EL-license" summary. I figure out what "no licensing information" means exactly and what to do about it, and post on talkpage that I am the author of both the YT link and the Wiki edit including it, satisfying GDFL license requirement. If it were already explained in policy "because GDFL license cannot be absolutely determined from YT, please post on talkpage re authorship when including a YT link," that would make everything much simpler. It is not C policy to delete questionable cr violations, but to inquire on talkpage. When the YT links are deleted without discussion or even a template, people do not get any info about what they should do to affirm GDFL. In the gap, some great links will be lost to confusion--in addition to C policy about talkpage posting re cr vios, that's the argument in favor of querying instead of deleting wothout notice or discussion (esp. in absence of policy info re YT/GDFL). A great deal of YT links are not under copyright and will never be under copyright--the vast majority of YT publishers are releasing their work into the public domain. As a self-publishing medium, the cr issues don't pertain much to the users, but to cases where the users are using material they don't hold copyright to under fair use. Publishing on both YT and Wiki is not a reasonable suspicion of copyvio on its face--unless the material has cr issues, it is extremely unlikely that GDFL is an issue. Nobody is putting anything on YT and hoping it won't be linked anywhere else--people publish on YT for public domain. Cindery 21:04, 21 November 2006 (UTC)

Personally? I can't see a clear reason here why having a bot tag every article containing a YouTube link with a "Copyright Ok?" notice of some kind isn't possible. In the same way that various other items are flagged, people can then follow up behind and review the link to make sure its kosher.
Consider, a tag being generated on the Talk page saying that the existing YT links need to be verified, then giving a certain period of time for that to be done, before the links are removed. Regularly watched pages will no doubt be checked sooner, but even for those dormant or less frequented pages, there will still be a record showing that the YT links were tagged, a check period was initiated and YT links were deleted because as a precaution -- but can be restored if someone does the leg work at another time.
What bothers me is the presumption that "99%" or even "75%" of existing YT links are copyright violations. While the percentage may be true for links in total, that percentage isn't necessarily true for a link in particular.
Summary umm... execution of links is, I think, against the spirit of WP.Jodyw1 21:21, 21 November 2006 (UTC)

Youtube can any type of source: primary or secondary. If someone publishes an original video on Youtube, it is primary source. If someone posts something from TV, it is a secondary source. --Ineffable3000 23:51, 21 November 2006 (UTC)

Youtube is exactly equivalent to a personal web site or a blog. The only difference is that there are lots of them. The material posted to a blog or personal web site is totally unverified - anyone can modify anything, regardless of external labels or appearances. There is 0 editorial oversight and 0 "chain of custody" or liability for bogus info, copyright violation or doctored materials. Hence Youtube is unacceptable as a source of any kind, primary, secondary or tertiary. It could be used in some very special cases as an external link to 'enhance' existing articles with an 'example' or 'extra material' if it otherwise meets WP:EL, but the video cannot be relied upon in the text as a source. Any challenged Youtube material can be removed, and if copyvio is suspected or BLP is involved, without regard to WP:3RR. Crum375 00:07, 22 November 2006 (UTC)
That is my understanding as well. ≈ jossi ≈ (talk) 00:20, 22 November 2006 (UTC)

Proposed amendment

I propose that we add a section that says:

Do not cite sources that you have written yourself. There are no exceptions. --Ineffable3000 23:54, 21 November 2006 (UTC)
There are many exceptions. If you wrote a scientific paper and it got published in a respectable peer reviewed journal, you may cite it. Similarly if you are a journalist and write for a mainstream newspaper you can cite it. Same for a book that you wrote and got published by a big respectable publisher, photos you took and got published by respectable media etc. etc. Crum375 00:13, 22 November 2006 (UTC)
See relevant guidelines of WP:AUTO and WP:COI. ≈ jossi ≈ (talk) 00:18, 22 November 2006 (UTC)
Certainly conflict of interest and neutrality and all other WP policies would have to be adhered to, but there is no blanket prohibition of sources that a given editor wrote or created. Crum375 00:29, 22 November 2006 (UTC)


Self-publishing credibility depends on the credibiliy of the source, and what the source is used for. For example, an acclaimed poet who videorecords another acclaimed poet and posts it on YT can be evaluated differently for source credibility than <insert less credible example>. The source in that case is not YT, but the acclaimed poet. If the self-published source is used in a BLP, it's a great source for what they subject has said, believes, etc--that's why a notable subject's blog is included often in external links.

The problem with current YT purge is the terms--"EL, sites that don't provide licensing information," which directs the editor to the EL guidelines, which currently read "YouTube should not be linked"--but no policy states that YT should not be linked. Using EL is a case of using a guideline to trump policy, and guidelines don't trump policy. Any YT link with a material suspected copyvio, like music video etc., should be deleted with edit summary: suspected copyright violation. Any YT links for which there is merely a GDFL question should be left alone, or a query posted. The EL guideline is currently inaccurate per policy, and directing people to it after deleting with that justification wrongly implies that YT can't be linked. Cindery 00:23, 22 November 2006 (UTC)

I thinl that we have exhausted the discussion. YT content should not be linked to unless there is a very specific need to do so. ≈ jossi ≈ (talk) 00:31, 22 November 2006 (UTC)

I do not think it is ok in any situation. If you wrote a notable article, someone else will write about it. --Ineffable3000 00:40, 22 November 2006 (UTC)

That is one viewpoint. The consensus has long been, however, that self-citation is viewed with suspicion but allowed. Youtube can be linked, but it's a rare work indeed on YouTube that is both notable/authoritative and not a copyvio (exceptions include things like videos intentionally published on YouTube for exposure or advertising purposes, etc.) Deco 00:48, 22 November 2006 (UTC)
Yes Deco, there are videos uploaded to YouTube under agreement with with news organizations, studios, and manufacturers. These may be OK to link to. ≈ jossi ≈ (talk) 00:57, 22 November 2006 (UTC)

I would say the discussion has only just begun where the EL guideline is being abused to contradict the C policy--it may even be a huge violation of WP:POINT. Cindery 02:22, 22 November 2006 (UTC)

You are pushing this to an extreme. WP:EL is a guideline and is applied at the discretion of editors. ≈ jossi ≈ (talk) 02:26, 22 November 2006 (UTC)

No, I'm pointing out that the purge has been pushed to an extreme--deleting under EL with EL link takes editors to the EL guideline, which is currently in conflict with/contradicts policy. Cindery 02:46, 22 November 2006 (UTC)

I don't see any conflict here. There are 2 critical policies: WP:V and WP:C. Including a source from YT would violate at least one, and possibly both, most of the time. There are almost no cases where both would be satisfied. This is the same as linking to any old blog or personal web site - 0 verifiability of the contents, regardless of what they seem to be, and very possibly copyvio in many cases. Therefore, it makes ample sense to delete all such links a priori. In the very few cases, if any, where it can be proven that a YT link is acceptable, we can deal with it on a case-by-case basis, like any blog or personal web site. This becomes even more critical in cases of WP:BLP, where the sources and links need to be of extra high quality. Crum375 03:05, 22 November 2006 (UTC)
That is my understanding as well. ≈ jossi ≈ (talk) 03:46, 22 November 2006 (UTC)

No, you're just wrong--the current wording of EL is that Youtube cannot be linked, and that is not policy--YT is not prohibited. Moreover, the most conservative estimate from the person doing the deleting is that at least 10% of the YT links are legit. (He thinks he's deleted about 2000, so that's 200 legit links--I think it's a significant underestimate because he's not carefilly evaluating them.) There should be a clear copyvio suspicion per content, and the deletions should be labelled as suspected copyvios--deleting them under EL is is not policy, and harms Wikipedia by misinforming editors--the problem is not YT per EL, and deleting YT under EL doesn't tell people how to use YT for the benefit of Wikipedia. Cindery 03:52, 22 November 2006 (UTC)

I think that everyone has been extremely patient with you on this. Repeating "you are wrong" does not make you to be right. The arguments presented are quite clear and well grounded, so I would suggest that you take this to the Wikipedia:Village pump (policy) so that other editors can comment and hopefully be better at clarifying this for you. ≈ jossi ≈ (talk) 06:24, 22 November 2006 (UTC)

No, following research at EL, I have discovered that the EL policy was changed about two weeks ago to exclude YT on the EL technicality, so I am addressing the change there (where a regular EL guideline discussion page editor noted in the first place Nov 3 that no change was necessary, because C covers C violations already). This is very recent--excluding YouTube on an EL technicality--and the same two people who made the change then immediately began deleting all YT links as EL violations. YouTube copyvios are already covered under C policy; excluding them under EL contradicts every other policy--because YT links are not prohibited by policy simply because they are YT links. I noticed also that you were quick to jump into the initial YouTube discussion at AN before it went to EL--you appear to have a strong anti-YouTube bias. Cindery 06:47, 22 November 2006 (UTC)

Cindery, I don't think either Jossi or I have any bias here. The issue is fairly simple: verifiability. Everything else follows from it. There is no way to prove verifiability on YT, or any similar personal web site or blog, as there is no editorial board controlling its content. Hence, any claims to copyright status become unreliable, as well as the content itself. Again, YT is no different than any personal web site and blog, which are clearly not verifiable or reliable sources. If we allow YT, we might as well allow everything and become one more web forum. I don't think there is any consensus here to subvert our main mission, which is to become a reliable encyclopedia. Crum375 13:21, 22 November 2006 (UTC)
One small point. It is not verifiability that is the issue, as anyone can verify that a source is on YouTube and states or shows what it shows. Verifiability is at heart, can someone else, independently confirm, that such a source shows or states what it shows or states. Verifiability has nothing at all to do with truth. Your real issue it seems to me, is, is the source reliable. That is, does it make an attempt, to reveal the "truth" as it knows it or as its underlying source states it. Again the "truth" (in quotes) is not the Truth in capitals. We always need to be aware of the distinction. We can certainly report here, that a YouTube by Y shows X, as long as we can confirm or have a valid belief that Y is a reliable source. YouTube is not a source, it is the underlying author who is the source. Wjhonson 17:32, 22 November 2006 (UTC)
One might argue that if a source is notorious for violating copyright laws, a reader who respects the rights of authors would make a modest effort to determine if the source was in compliance with copyright laws, and if it was not, would not view it. If a source is one that readers who respect authors would refuse to view, then it is not verifiable except by readers who don't care about authors. --Gerry Ashton 17:39, 22 November 2006 (UTC)

A source which is "notorious" for violating copyright laws can be dealt with under WP:C. As the EL guideline editor said on Nov 3: copyvio links are already forbidden, and it isn't a specific problem with Youtube but with all publicly contributed sites As far as "readers who respect authors" goes, Gerry doesn't seem to have seen the many valuable self-publishing examples unique to YouTube, which make it an amazing resource with potential to be more amazing (Geriatric 1927?) Propagating a source bias denigrates the legitimate and valuable uses of YouTube; it doesn't merely get rid of the lame music-vid copyvios. Daily Kos is a blog, and needn't be confused with thousands of lamer blogs merely because it is a blog--we didn't ban blogs, and we didn't ban YouTube. No self-publishing mediums have been banned, there are rules for their use and inclusion. The problem here is that a tiny handful of people, without much discussion, changed the EL guideline two weeks to contradict policy, and are mass deleting using the guideline, which is creating confusion and could result in the loss of excellent sources and links. And yes, jossi was one of the original handful--see his talkpage and the Admin Noticeboard discussion. Jossi thinks YouTube is a disaster in the making, which justifies a pre-emptive strike. Again, deleting the copyvio links is a great project. Banning YouTube on a technicality is not--there are too many valuable exceptions, and there is no reason to treat YouTube differently than any other self-publishing medium. As Wjohnson rightly pointed out, it is the underlying author who is the source, not YouTube. Conflating all those authors with the bad apples who upload music vids is wrong and does a disservice to the quality of Wikipedia. Cindery 22:48, 22 November 2006 (UTC)

A YT source is not verifiable simply because it is generally an individual with a web page who can change the content on that site at any instant at will to say anything, with no editorial supervision or corporate oversight or any other legal restriction. This is very different from a respected official site of some reputable news organization (for example) that stands behind its reports, or a respectable publishing house that has many layers of validation and fact checking before publishing information to the outside world. If the article is in state A at a given time, it can be in state B 1 millisecond later. If it says it's quoting or showing a statement by person A, it could easily be doctored - there is nobody verifying that the person is really A and those are the original undoctored words. In short, YT is just like a personal web page or blog, with 0 validation or verification - it does not meet WP:V - what one editor sees at moment X can be very different the next moment when another editor clicks on it to 'verify' it, plus the statement made on that page is totally unverified and quite possibly a forgery. Crum375 00:06, 23 November 2006 (UTC)

...now you're going way out on a hypothetical limb. By that argument: the internet could be hacked at anytime! Because of this vulnerability, the very possibility that anyone could hack into and change it, nothing is truly verifiable or reliable, and we may as well stop now. I mean, how do we really know when we link to the NYT that the Flying Spaghetti Monster didn't rerrange all the words with his noodly appendage after we linked it??? :-) It's an epistemological mystery that can never be solved by mere mortals...Meanwhile, the EL policy was just changed--it no longer states that YT is prohibited under EL.

ps: Have you read the YouTube article?--"reputable news organizations" license content to YT. And many self-publishers with perfectly fine reps publish on YT for public domain. The assumption that because people can publish whatever they want, therefore we cannot trust them is predicated on the assumption that freedom=unreliability. (I think there's an opposite argument to be made--that a lack of freedeom; being financially beholden to the advertising dollars of large corps-- can adversely affect reliability.) But, anyway, most YT links will be used/are used as external links, not sources. If the subject, the author, or any editor thinks that such a link is unreliable in any way, they can, do, and will speak up about it--as they do for blogs and websites. YT is no more nor less a reliable source; anything can be changed on a blog or website at any time. Cindery 02:04, 23 November 2006 (UTC)

Hey, hey... please... Material in YouTube that is uploaded under license by a news organization, a studio or a record label, is OK. But all other content uploaded by individuals is not verifiable, not reliable, and not to be used in Wikipedia to support any claims in any articles. The use of YT in external links is also very questionable, and should be assessed in a case by case basis. Now, if you disagree with that, then go back to EL and make proposals to change the wording. ≈ jossi ≈ (talk) 02:10, 23 November 2006 (UTC)

You seem to have missed the part above where I said EL was just changed back--the guideline now accurately reflects policy. Per RS and V, there are exceptions whereby YT can be used as source. Perhaps you should go to RS and V to argue that they should be changed--falsely stating on OR that YT is not to be used to support any claims in articles on NOR discussion page won't change V or RS. Cindery 03:41, 23 November 2006 (UTC)

YT is equivalent to a personal web site or blog. I fully agree with Cindery there. And WP generally forbids linking to such sites, except in some very limited cases. A web site belonging to a reputable organization has an oversight mechanism that verifies/validates the contents at all times, by definition. Such a site, e.g. cnn.com, may be linked into as source. OTOH, a personal site contains random unverified data and clearly does not meet WP:V - every click on it may yield another bogus or just totally unvalidated version, with no editorial control whatsoever. If WP were to rely on such personal sites, it would become one more online forum, not a reliable encyclopedia. Crum375 02:13, 23 November 2006 (UTC)

What you seem not to get is that the possibilty that the use of a source is unreliable doesn't make the source unreliable. A YouTube video used as photographs are can be a legitimate and extremely useful use of a source. A specific complaint about a specific video or photo would be necessary to argue that the photo or video was unreliable; that photography and film are user-manipulable sources is not a sufficient generality to exclude all uses of film and photography on principle. Cindery 03:39, 23 November 2006 (UTC)

I think you have it reversed. A personal web site or a blog is considered unreliable and unverifiable due to its total lack of editorial controls. Hence it is unusable as a WP link except in some very special cases. So the rule is that you can't use it, the very few and rare exceptions need to be argued and supported in a special way. So by default, all personal web pages and blogs, including YT which is equivalent, are unacceptable and stay out. Crum375 03:51, 23 November 2006 (UTC)

For video evidence that something happened or used to exist, I could post a digitized YouTube link of a film and you would have to argue on talkpage of article, provide evidence, and gain consensus that it was unreliable (on some other basis than because it was self-published, as most Wikipedia photographs are. Depending on who I am, I could be an extremely reliable or unreliable source--think back on the acclaimed poet videotaping the other acclaimed poet). If I re-published a public domain Krazy Kat cartoon I videotaped off my laptop from the Library of Congress site in a more easily accessible YouTube link, I could cite it as a footnote in a sentence making a descriptive claim about the specific cartoon. If you wanted to object, again, you would have to discuss on talkpage and gain consensus; provide evidence/make argument that the YT link was unreliable. Cindery 04:15, 23 November 2006 (UTC)

You are implying that evidence provided from a personal web site or blog is 'reliable until proven unreliable', whereas the WP rule is exactly the reverse. We assume that any evidence on a personal web site or blog, that has not undergone a formal 'publishing' process by a respectable editorial board is hopelessly tainted and can only be used in some very limited cases. This is not only YT or video. By your argument, we can take any personal web site, containing anything, and use it as source until someone proves it's wrong. WP does not work this way. We must start out with verifiable and reliable sources, and those are sources that have been vetted and published by respectable and reputable entities. If we followed your proposed sourcing rules, WP would be just another online forum, not an encyclopedia. Your argument flies in the face of the sum total of our sourcing and linking rules. Crum375 04:28, 23 November 2006 (UTC)
Cindery, I think you need to know when to stop. Please listen to the arguments made and contemplate the possibility that your understanding of policy may be a bit off. ≈ jossi ≈ (talk) 05:15, 23 November 2006 (UTC)

Re Crum75, Yes, it's reliable until proven unreliable, because we assume good faith that editors are adding information to improve the encyclopedia. If there is some problem with any addition, other editors revert edits and explain why. In a talkpage discussion regarding a YT link used as a source, if the link was a good source, you would lose if your argument was "even if it is a good source, it should be excluded because it is YT." Source reliability isn't established by policy or guideline--it's established through the editorial process, which relies upon policies and guidelines.

RE Jossi, I realize that you would like to denigrate my understanding of policy rather than admit that the EL guideline was changed two weeks ago --without adequate discussion or consensus-- to contradict policy, but it's too transparent to take seriously. I did notice that you immediately ran over to EL to try to reinsert a prohibition on YT, though! :-) Cindery 06:43, 23 November 2006 (UTC)

There is no such "prohibition". EL is a guideline and it evolves with time as new challenges surface. I invite you to bring the discussion at WP:EL. ≈ jossi ≈ (talk) 16:13, 23 November 2006 (UTC)

As you are aware, I have brought it up at EL. You've reverted changes by Barberio and Wjohnson which deleted YouTube, and are now up to 2RR--without bringing it up on talkpage yourself. Is there some reason you are reverting the other editors without discussion? Cindery 16:26, 23 November 2006 (UTC)

Synopsyes of novels

I have a question regarding a possible conflict between WP:NOR and WP:copyvio (and I apologize if this question has been raised and answered before - it's not in the current NOR article or talk page, from what I can see). If one is creating an article about a novel, such as my recent The Antipope article, how is it possible to insert a synopsis of the novel that is neither OR (if one writes it oneself), or copyvio (if taken from a "reputable" source)? Would this case be an exception to the NOR policy? Thanks in advance. Carre 20:09, 22 November 2006 (UTC)

In practice, most editors seem to think this policy doesn't apply to plot summaries. If there has only ever been one plot summary written, and this may be the case with some novels which only the publisher has ever gone to the effort of describing, and one wants to follow a strict reading of NOR, one may well be restrained to selectively quoting that one summary, with the result of our summary being very short. Of course, if you can locate a number of different plot summaries, this problem goes away. Jkelly 20:30, 22 November 2006 (UTC)
Why not simply cite the work of fiction? Spoo, a featured article, includes references to episodes. --NE2 20:53, 22 November 2006 (UTC)

Need input on page in this matter

Hi. I am not sure where else to post this. I am wondering if some people can give me their opinion as to whether I have crossed the line into original research on a page I recently made. The material in question is not currently showing on Saipan Sucks but can be seen at http://en.wikipedia.org/w/index.php?title=Saipan_Sucks&diff=90000413&oldid=90000009 Thanks! C.m.jones 22:55, 25 November 2006 (UTC)

You would be better asking that question on the article's talk page, or alternatively, ask for non-involved editors to take a look, by placing an request for comment. ≈ jossi ≈ (talk) 22:57, 25 November 2006 (UTC)
Okay, will do, thanks. C.m.jones 23:06, 25 November 2006 (UTC)
If anyone wishes to follow up, my entry at RfC is at Wikipedia:Requests_for_comment/Biographies. Thanks! C.m.jones 23:35, 25 November 2006 (UTC)

Since I have not gotten any comments I am requesting here again. Please help.

Is this OR:

At the webpage http://www.saipansucks.com/about.htm the author's name does not appear. The page is written anonymously.

But not really. Because with the same page opened in your browser, go to View > Page Source (in Mozilla) or View > Source (in IE).

On line 39 of the html source code you see the name William Betz as author.

Can I cite the html source and have that not be OR?

Also, if one does a whois with the domain saipansucks.com, the name William Betz is shown as the Administrative Contact.

Can I cite that and have that not be OR?

In both cases one must take an ACTION. And no published source mentions that by these actions one can find the page's author. I in a sense "created" the way, did the research for this.

Can I cite the whois lookup and have that not be OR?

C.m.jones 22:26, 27 November 2006 (UTC)

The whois lookup does not tell you the author. It just tells you the name given by the guy who put the website up. I presume you believe the author is the same person who has this website, and that is why you wish to make the claim. However, you really need a more reliable source than a whois lookup to be positive about the link to him. jguk 13:02, 28 November 2006 (UTC)
.C.m.jones is not using whois; C.m.jones is viewing the html source code of the page. HTML text can be commented out and not displayed by the browser in the same way that text in Wikipedia articles can be commented out; preceed the text with <!-- and follow it with -->. This same issue has come up with respect to digital signatures; if a person applies a digital signature to some text, is the person agreeing to the text as it appears in the program that is usually used to read it, or is he agreeing to hidden text as well? I don't think a difinitive answer has been reached. --Gerry Ashton 19:46, 28 November 2006 (UTC)
Splitting hairs. If whois tells you the site was put up by "John Brown" then you could say in the article that "WHOIS says the site was put up by John Brown". If the HTML code says <Author>John Brown</Author> then you could say "the HTML code says the Author is John Brown". And so on. It's getting pretty fine-grained in here! This is not original research. Research is not "original research". Original research involves you, creating new material, not finding someone that someone else has created. Finding material is source-based research. You are using sources, in this case, Whois, and "View Source". You are not creating what they say. Wjhonson 16:06, 29 November 2006 (UTC)
Stating a verifiable fact, such as "the HTML code says the Author is John Brown" should be no problem as we are not asserting that John Brown is the author, just saying that the HTML code states that. But we cannot say which John Brown is this, as there is no way to verify that. ≈ jossi ≈ (talk) 17:06, 29 November 2006 (UTC)
BTW, this "problem" is quite a known one... if you use Word to generate a web page, Word will copy all info from your Word properties into the HTML code. This has created some embarrassing situations for many people that thought that they will remain anonymous via a web page... Non-techie beware! ≈ jossi ≈ (talk)
There is also the possibility that an author might copy some HTML from a friend or another web site as a starting point, and never notice that the HTML has some now-incorrect author information in it. --Gerry Ashton 17:45, 29 November 2006 (UTC)

Change needed? Also, a question

Hi.

I've noticed these:

  • The prohibition against original research limits the possibility of an editor presenting his or her own point of view in an article. By reinforcing the importance of including verifiable research produced by others, this policy promotes the inclusion of multiple points of view in an article. Consequently, this policy reinforces our NPOV policy.
  • "If your viewpoint is held by an extremely small minority, then — whether it's true or not, whether you can prove it or not — it doesn't belong in Wikipedia, except perhaps in some ancilliary article. Wikipedia is not the place for original research."

For the first, one can include multiple points of view even if one is not verifiable, so this (WP:NOR) to me sounds more like an issue of verifiability and not neutrality unless we are saying that if it's unpublished it could therefore be simply a personal opinion. Saying it "promotes the inclusion of multiple points of view" doesn't seem to make much sense, rather it might be better to say "promotes the inclusion of only verifiable, attributable points of view" or something along these lines as it doesn't just promote the addition of any ol' POVs.

For the second, this sounds more like Notability, as in order to "belong" in Wikipedia all that is needed, really, according to WP:V is publication in citable, and ideally, reliable sources, and "original research" as defined in this very policy page (WP:NOR) means "unpublished material", not how many people hold the viewpoint. Then again, I could be wrong... 70.101.147.74 03:06, 26 November 2006 (UTC)

I agree that both these statements are suspect on this page. I'd add the note that NPOV is not the same as multiple points of view. Better to just remove them - the existing policies cover them just fine. Deco 04:21, 26 November 2006 (UTC)

A person as a secondary source

Part of the project page used to say that a journalist could be a secondary source. For the purposes of wikipedia, no person can be a source, whatsoever. A person is not verifiable, only their work is. So a person is not a source. The interview, the article, the book, is a source, but a person is never a source. I cannot imagine the horror of citation to "Mr John Brown at Rutgers" and then editors having to call him up to interview him to verify the article. That would be an absolute nightmare, not to mention eminently impossible to achieve. I strongly hope that others agree with my position, that only works are sources. Wjhonson 07:06, 26 November 2006 (UTC)

Good points. At best people could be considered primary sources, but our definition of primary sources includes transcripts of interviews. I think the key point is that for us a "source" is some sort of text. These texts are indeed produced by people, but it is the text that is the source, not the person him/herself. Slrubenstein | Talk 07:13, 26 November 2006 (UTC)
I think disagree with this. When we allow a blog to be used as a source, it's because we regard its author as reliable. While, strictly speaking, the blog is our source in the sense that it's the published material we refer to, it's the author of the blog who carries the reliability criterion, and therefore I'd say the person is our source. SlimVirgin (talk) 01:52, 2 December 2006 (UTC)
I think it is necessary in policies and guidelines to speak strictly. It is perfectly legitimate for authors in most other media besides Wikipedia to interview a person and report the results in an article. Because Wikipedia does not accept this widespread practice, our policies and guidelines should make this clear by never suggesting that a person can be a source for an article. --Gerry Ashton 02:06, 2 December 2006 (UTC)

A request for advice

As some of you may know, I've been creating a series of articles about the local administrative units of Ethiopia (they are called Regions, Zones & woredas or districts). Until yesterday, I've been having a fair amount of luck sticking to my plan of simply providing the data from reliable sources (the Ethiopian Central Statistical Agency for the 2005 population numbers, maps from UN-OCHA for geographical details) with the minimum amount of interpretation needed. (Let's not debate whether my work is in part or whole OR; if it is, then so is every contribution Rambot has made.) There's been a few places where my sources have contradicted each other, but I've been able to avoid the thicket of OR up until now.

Yesterday, I was exploring a new source -- the Disaster Prevention & Preparedness Agency of Ethiopia's website -- & discovered that it had a number maps that I could make good use of. However, many of these maps introduce a number of important complications into my work: not only do these maps document new subnational administrative units (the existence of some I suspected from other sources), but different boundaries for many of these units. From everything I can tell, this source is of equal reliability as the others (it's part of the Ethiopian government, just like the CSA, & the other materials are produced by the UN). In other words, I have a problem that needs some kind of explanation.

The best solution -- find a source that explains these discrepencies -- is not an option. I've done a few Google searches, & I failed to find any explanations: that one source was using incorrect information, or perhaps these were routine administrative reorganizations. (I'm not surprised: with the possible exception of the military, the Ethiopian government is understaffed & providing English translations of administrative changes is not that high on their lists.) I've done enough local library searches for information on Ethiopia to know that I don't have printed sources close to hand. (I might in maybe in 9 or 12 months, with help from a dedicated reference librarian -- books on the recent Ethiopian are uncommon.)

If this wasn't being created for Wikipedia, I'd provide the most reasonable explanation -- that the differences are most likely due to the expected administrative reorganizations reported at different times -- & leave it at that. On the other hand, if I were following the steps I described in Wikipedia: These are not Original Research (& I'll confess, I was the one who started that essay) of simply providing the information & leave this for the reader to sort out, I believe this would abandon the reader at the moment they need some kind of guidance -- even if it is bad advice.

Some might say this is a good case where I should ignore all of the rules, & just provide my reasonable explanation. However, in the last few months the "Ignore All Rules clause" has been abused (or alleged to have been abused) so often that I'm not sure that I could effectively invoke it here. While I'm following the spirit of "no original research" here (& I doubt anyone will argue that administrative reorganizations never happen), I am concerned that some enthusiastic Wikipedian will declare that I am violating the letter & engage in a protracted & disruptive edit war to remove all of my work. I'd rather work towards a solution first. -- llywrch 19:30, 27 November 2006 (UTC)

I don't see how what you are doing can be affected by WP:NOR, as it's quite apparently that you are not making up your own subdivisions of Ethiopia and promoting them. I do see issues around WP:V, and in particular whether the information you are adding is supported by a reliable source, given that you yourself are casting doubt on some of the sources as they are contradictory.
One route is not to report anything where there are contradictory sources, or you are not concerned about the reliability of the sources. The alternative is to report what the sources say, but to make it clear what your source is. If the source is a map from 1996, for example, say so, and then everyone reading it knows that what you write may be inaccurate for 1997 and later years.
From your comments, it appears that you assess the sources you have available as being reasonably reliable, but not perfect, and that the precise application of them to any particular date is unclear. I'd therefore recommend the latter route - use your sources, but make the inherent limitations of your sources clear to the reader. jguk 12:56, 28 November 2006 (UTC)

A case study (and improving Wikipedia to make a point)

I have chosen to use the time-honoured device of improving Wikipedia to make my point about these two sentences:

For that reason, anyone—without specialist knowledge—who reads the primary source should be able to verify that the Wikipedia passage agrees with the primary source. Any interpretation of primary source material requires a secondary source.

My point is that it is silly to have a requirement that primary sources have to be intelligible to someone who knows nothing about the subject. And also it is silly to try to prohibit interpreation of primary source material where there is no originality in that interpretation - ie an interpretation that no-one who understands the subject would genuinely disagree with.

I have therefore created the article English women's cricket team in Australia and New Zealand in 1934-35.

Apart from bit about the quatrain in the lead, everything else in the article has been derived from primary sources - either scorecards of cricket matches, or from photographs. And I have readily interpreted those primary sources. Every cricket fan (specialist), if they were so motivated, would be able to agree that what I have written follows immediately from the primary sources I have cited. Someone who knows nothing of the sport (non-specialist), will not.

My question is therefore: is it reasonable for this new article, English women's cricket team in Australia and New Zealand in 1934-35, to exist? jguk 17:57, 1 December 2006 (UTC)

Uh, while I have no doubt that you did it in good faith and it seems to be a worthwhile article, but if you had reason to doubt that the article would be acceptable, then you may have disrupted Wikipedia to make a point. Your user space would have been a better place to try it out first if you're just looking to make an argument. –  Anþony  talk  00:52, 2 December 2006 (UTC)
I believe my article is improving Wikipedia by covering an area where there is little information collated on the web where it can be easily accessed. I also don't believe that anyone would really seriously disagree with me having an article of this nature. My point, which is made by improving rather than disrupting (hence my joke at the start:) ), is that what is written in this policy does not accord with WP practice, and is not a desirable criterion to enforce. I'd welcome comments on the issue being illustrated. jguk 08:35, 2 December 2006 (UTC)
True, but "I thought I was making it better" is too easy of an excuse for anyone violating WP:POINT. In complete honesty, if this article has to go through AfD or even just be prod'd, you're wasting someone's time to delete it. It's not really that serious of course and I hope I don't sound like I'm chastizing you. Lord knows I've done some stupid things on Wikipedia. (Like confusing policy with a TV show talk page ... jeez.) Anyway, that doesn't detract from your argument, but like I said I really don't know enough about the details to comment. –  Anþony  talk  09:58, 2 December 2006 (UTC)
The example article is, IMO, replete with what appears to be good-quality original research. A simple example should suffice: Myrtle Maclagan is stated as having "dominated" the matches, although Betty Snowball scored more points. That conclusion strikes me as open to argument on its face. It may be a good evaluation, but it obviously requires knowledgable balancing of the relevant factors. That is exactly what Wikipedia should avoid doing. Surely someone with credentials has written about this fascinating incident, and could be cited. I agree that doing so is far less interesting than writing more prose, or arguing on policy pages, but if editing Wikipedia were easy, anyone could do it!
To take an example more familiar to me, there is no doubt whatsoever that Mike Schmidt was the greatest third baseman in the history of professional baseball. Every knowlegable fan with access to the relevant data will come to the same conclusion. Nevertheless, there are at least two sources for that conclusion in the article. I grant those sources are easier to find than the ones you need, but the low-hanging fruit have been largely picked, and new articles are likely to require more effort as time goes on. Robert A.West (Talk) 19:07, 2 December 2006 (UTC)

Robert, thank you for your comments. I think they help illustrate what I am saying.

Any cricket fan looking through the article or the scorecards would conclude that Myrtle Maclagan was the star player - her wickets need to be considered as well as her runs. It is something any specialist (here, cricket fan) would agree on, but that a non-specialist (ie non-cricket fan) cannot immediately see. (In fact the Morning Post quatrain, which is from a secondary source, further supports the conclusion.) By making that conclusion, I have added to the article, and hopefully made it more interesting to the reader, who may then be tempted to learn more about Myrtle Maclagan (and probably Betty Snowball too). It is a worthwhile thing to add. Given that it is an interpretation easily checked by the reader (and remember that cricket fans are the target audience), I think I should be able to add that interpretation to the article.

Yes, I am dealing with something on which relatively little has been written - unlike your baseball example, which I agreee appears to be comparable in every other respect. This makes what I write in WP more important (if you want to learn about the English women's cricket tour of 1934-35 where else can you go, whereas if you want to know more about third basemen, I guess you have hundreds of options other than WP open to you). It also means that it is reasonable for the article's author to draw out on WP as much information that is available to the author. Not so as to give voice to his own theories - I agree that that is against this policy - but so as to place the subject in proper context.

The wording of the policy makes it clear that I can interpret a reliable secondary source in an unoriginal way even if that secondary source cannot be understood by a non-specialist. It seems odd that I can't do the same with a reliable primary source. jguk 18:47, 3 December 2006 (UTC)

Where does the policy states that one can "interpret a reliable secondary source in an unoriginal way"? ≈ jossi ≈ (talk) 18:50, 3 December 2006 (UTC)
The only interpretation of primary sources we can use, are those made by reputable secondary sources. We do not interpret neither primary or secondary sources. That is what WP:NOR means. Summarizing sources is one thing, interpreting sources, be these primary or secondary is not the job of editors. ≈ jossi ≈ (talk) 18:52, 3 December 2006 (UTC)

All articles are interpretations of source information. That's what putting together an article that is not plagiarised involves: interpreting your sources and putting together a text that conveys information already in those sources in a way that will get your point across and interest the reader. When I refer to interpreting sources in an unoriginal way, that is the process I am referring to.

The wording of WP:NOR certainly does make a distinction between primary and secondary sources in this respect. It says:

For that reason, anyone—without specialist knowledge—who reads the primary source should be able to verify that the Wikipedia passage agrees with the primary source. Any interpretation of primary source material requires a secondary source.

This requirement is not reproduced for secondary sources. So it follows that we are allowed interpretations of secondary sources provided those interpretations are unoriginal - ie they do not introduce new ideas or concepts.

If, however, you mean that we should not make original interpretations full stop, but we can reproduce original interpretations of others (ie by reference to a secondary source that makes original interpretations of a primary source), then we really need WP:NOR to say that. jguk 19:09, 3 December 2006 (UTC)

It already says so. Consider the following passage from [WP:NOR#Reliable sources]], "The only way to show your work is not original research is to produce a reliable published source who writes about the same claims or advances the same argument as you." Consider the entire section WP:NOR#Synthesis of published material serving to advance a position. Robert A.West (Talk) 19:16, 3 December 2006 (UTC)
All articles are interpretations of source information.???? You may have missed what this policy is about... Articles are summaries of sources of information, not interpretations. Interpretations of sources violate this policy. ≈ jossi ≈ (talk) 19:28, 3 December 2006 (UTC)

Removing original research

Is there a consensus as to whether or not original research can be removed at any time? Taxico 08:08, 3 December 2006 (UTC)

Many people confuse "source-based research" with "original research". They are not the same thing. Source-based research is strongly encouraged. You may remove original research at any time. Your edit description should state it. Personally, I recommend moving it to the Talk page of the article, so you and other editors can discuss it. It might be that other editors will not concur that it is original research, and that procedure would seem the most non-confrontational way to handle the situation. Wjhonson 08:23, 3 December 2006 (UTC)
Very good suggestion. Taxico 08:28, 3 December 2006 (UTC)

First sentence

Please see discussion at Wikipedia_talk:Verifiability#The_first_sentence - it effects the first sentence of this page too. --Tango 00:07, 4 December 2006 (UTC)

Audio recording

Is an audio recording, such as a song, considered a primary source? Can it be used for information, or is that original research? A small discussion (and many different reverts) has been going on at Love (The Beatles album) (edit|talk|history|links|watch|logs) and Talk:Love (The Beatles album) (edit|article|history|links|watch|logs). —Gordon P. Hemsley→✉ 01:46, 8 December 2006 (UTC)

I took a glance at the most recent revert [9] in which 70.153.27.27 adds In My Life to the list of songs that "Strawberry Fields Forever" contains elements of. However, the footnote at the end of the sentence credits a BBC web site for the information, and that web site does not mention "In My Life", so 70.153.27.27's version of the article misrepresents the source. I leave it to others whether references to two songs is sufficient for Wikipedia readers to decide if one song contains elements of another, or whether an published opinion from a music critic is required.
By the way, there is no reason an audio recoding couldn't be a secondary source, depending on just what kind of material it contains, who the author is, and who published it. --Gerry Ashton 03:16, 8 December 2006 (UTC)
Using a source for information is "source-based research" not original research. Original research consists in a creative act. Listening to a song is not creative. Wjhonson 04:19, 8 December 2006 (UTC)
Audio recordings that do not specifically state where its samples come from can't be used as a source of cited analysis in that regard. Just64helpin 14:16, 8 December 2006 (UTC)
Further, non-trivial analysis cannot be done by wikipedians. But I would like to point out that there are relatively few audio recordings that specifically state where their samples came from. That's not the bar. The bar is, does some reliable source state it. The recording itself does not need to state it. Wjhonson 18:17, 8 December 2006 (UTC)