![]() | 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. |
Archive 15 | ← | Archive 18 | Archive 19 | Archive 20 | Archive 21 | Archive 22 | → | Archive 25 |
From all these discussions, I can't help feeling that there's some confusion, or a lack of clarity, as to what type of source (primary/secondary/tertiary) a given source is. As I understand the definitions, a primary source is a photograph, video, eyewitness statement (in the context of, say history), or a TV show itself, a script, a novel, etc. A secondary source is anything that isn't primary. A tertiary source is a special case of secondary sources which do not use any primary sources themselves.
Thus, if someone edited wikipedia with information they had gained first-hand, that would be unsourced, and the article would itself be a primary source. If they directly quote an eyewitness or a text, then that's sourced from a primary source, and the article is a secondary source. If the only directly used sources are secondary (including tertiary sources), then the article is a tertiary source (often seen as the ideal for wikipedia).
The principle here is original research - the composition of a primary source, or the synthesis of primary sources into a secondary (non-tertiary) source are research. If this is done for the article, then that is original research. Thus, if a person watches a TV show, reads a book, watches a film, or any similar activity and then writes a plot summary, that's original research. If they find one or more plot summaries elsewhere and reference them, then it's not original research, it's synthesis or composition of secondary sources.
Journal articles aren't a primary source, generally speaking - raw experimental data is a primary source. Primary sources are typically (but not always) devoid of interpretation. Interpretation is added through research in the synthesis of the secondary source.
WP:NOR is saying that we don't do interpretation, we do synthesis compilation.
Does that make sense to people? SamBC( talk) 23:07, 14 August 2007 (UTC)
Further to all this, I've just started a new "proposal" at Wikipedia:Classification of sources. I don't know for sure that it'll be useful, it's certainly not in the right tone yet, and if it is useful I don't know if it should stay a separate page. However, that's useful for working on it. If you think we should have a clear internal definition or illustration, please join me and help to build it up and put it in the right tone. It's mostly copy-paste from my initial comment above. SamBC( talk) 02:44, 15 August 2007 (UTC)
In case anyone's interested, there's a Village pump discussion on the pri/sec/tert sources issue. Dreadstar † 21:23, 15 August 2007 (UTC)
"Original research that creates primary sources is not allowed. However, research that consists of collecting and organizing material from existing primary or secondary sources within the provisions of this policy is, of course, encouraged: this is "source-based research," and it is fundamental to writing an encyclopedia.
Although most articles should rely predominantly on secondary sources, there are rare occasions when they may rely on primary sources (for example, legal cases). An article or section of an article that relies on a primary source should (1) only make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge, and (2) make no analytic, synthetic, interpretive, explanatory, or evaluative claims. Contributors drawing on primary sources should be careful to comply with both conditions."
Just some thoughts. Your thoughts? Vassyana 23:21, 11 July 2007 (UTC)
"
Original research that creates primary sources is not allowed.Research that consists of collecting and organizing material from existingprimary or secondarysources within the provisions ofthis policythis and other content policies is, of course, encouraged: this is "source-based research," and it is fundamental to writing an encyclopedia.
Although most articles should rely predominantly on secondary sources, there are rare occasions when they may rely on primary sources
(for example, legal cases). An article or section of an article that relies on a primary source should (1) only make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge, and (2) make no analytic, synthetic, interpretive, explanatory, or evaluative claims. Contributors drawing on primary sources should be careful to comply with both conditions."
There. ≈ jossi ≈ (talk) 00:07, 12 July 2007 (UTC)
I agree with the edits in the guideline. And I understand the preference for reliable, secondary sources, since these may be the result of more fact-checking than the primary source could or did. However, when secondary sources are not (freely) available, I have no problem with using a primary source provided the use thereof is clearly indicated in the article.
—
Xiutwel
(talk)
08:08, 12 July 2007 (UTC)
Due to the supporting comments and lack of objection, I have changed the page to reflect Jossi's proposal. Vassyana 13:10, 18 July 2007 (UTC)
It occurs to me that there's a pretty good, relatively objective criterion that we could use to cover when primary sources are appropriate, and it ought to cover pretty much every case:
Primary source are acceptable when they require and receive no interpretation in the article, but are merely a source of an objective fact.
Thoughts? SamBC( talk) 02:16, 15 August 2007 (UTC)
"Primary sources may be used in articles in the context they are employed in secondary and tertiary reliable sources. Primary sources that are considered accurate by reliable sources may be used for citation of "pure fact", including such raw data as U.S. census statistics."
(outdent) A couple words and an order swap might do the trick!
"Primary sources that are considered accurate by reliable sources may be used for citation of "pure fact", including such raw data as census statistics. Primary sources may also be used in articles in context as they appear in secondary and tertiary reliable sources."
Thoughts? Vassyana 05:17, 15 August 2007 (UTC)
I have given this issue a lot of thought for some time and I will direct to something I wrote previously about it. Pulling out my thoughts on question of when these sources are appropriate: sources which should only be be used with extreme care, making sure they are only cited to document that such a source makes such a claim (direct quotes of the source, careful paraphrase of what the author of the source claims, facts and figures in infoboxes).-- BirgitteSB 14:28, 15 August 2007 (UTC)
The problem I have with census data is that it can be cherry picked in ways that might even appear to be benign. TableManners 03:22, 16 August 2007 (UTC)
This line referring to OR contains an unnecessary assumption of bad faith:
I changed "therefore believed to be" to "could be" because we don't have to assume it's OR, the simple fact that it could be OR is enough. But the change was reverted. WP:AGF should be followed, not excepted in this case. Dhaluza 09:33, 15 August 2007 (UTC)
Yes, yes. So make some examples. If i say 'today it's hot' what sources do i must give to you? The C° meaured by my termometer? But it's OR, right? And maybe someone don 't find 35° 'hot' enough to be called so.
Behind my house there is an old church i don't know how wide it is and then i decide to measure it. Oh, my God, there is still an OR, and perhaps also bad faith, to add insult to injury.
We cannot copy other sources, because it' Copyviol, we cannot even resume them because it's OR (no syntesis, right?). Do you have the clue? Where is the good sense? Tell me how one can be authorized by your policies to do something that is not simply copy texts older than 90 years. Feeling as potential criminals every time that we post something that is not a he said so. That's the meaning of OR, IMHO.
I assert you cannot create an encyclopedia based entirely on quoting other sources, unless very often those other sources are other encyclopedias. Agree.-- Stefanomencarelli 15:58, 16 August 2007 (UTC)
{{ editprotected}} The edit war was successful at removing a long standing sentence:
“ | Wikipedia articles should rely on reliable, published secondary sources. | ” |
Can we put this back in and then reprotect? TableManners 03:13, 16 August 2007 (UTC)
No original research can be added to a wiki. So the only choices is to grab a researched document and copy inside the wiki but it's illegal (copyright) or you just can publish material that was created by someelse and changed it bypassing the copyright protection (cheating) or using published material that the original author allow to publish in the web.
Even more, there are a lot of "not so important material", that never was researched by a authority or a compentent entity, this kind of material cannot be showed in Wikipedia or even worst can be used by wikipedia only by a biased point of view, for example if we talk about Microsoft, there are little room for any outside to Microsoft to talk about it but Microsoft have almost all the rights and authority to talk about himself.
Anyways empiric talking, the "no original research" rule in wikipedia is not widely applied, common sense overule any no practical law. No original research is used currently to bash someelse, asking for any tiny details about a specific text, it's not anymore (and never was) a way to keep clean wikipedia. -- Magallanes 18:58, 16 August 2007 (UTC)
While you continue to discuss without problems about the santity of NOR and its stricth interepretations, in Wiki there are stuff like this: Domination of the skies, superior seamanship, and prudent, timely maneuvers helped to nullify the overwhelming odds. In the highest tradition of naval service, the finest qualities of the American sailor became commonplace during the heroic fight. Devotion to duty, daring courage, uncommon bravery, and an indomitable spirit were part and parcel of this victory.
So don't cause me LOL in so indecent manner. This piece is what could be called shameless agiography and pubblicity for US Navy, and there are thousands of this stuff. While i try to reduce every optimistic overclaiming made by italian aviators with the cross controls with several sources (it's unuseful to post that x has downed 12 enemy aircraft when are available sources that considering both sides, dismissed all this claimings without any problem), there are thousands of pages like this, written directly by J. Wayne. Obviousely you are not worried about, your goal is forbid brain activity. I am, instead. Babbling about NOR when gratuitus agiographies are so well spread is a thing that shows how these discussions and policies don't guarantee nothing. Sorry,but it so.-- Stefanomencarelli 13:59, 20 August 2007 (UTC)
I just noticed that the article has been protected due to edit warring, so I thought I'd take a look as to why. I think there's some merit to the "compromised" rewording of the sections in question.
For most articles common sense reasonably dictates that secondary sources are going to be the ideal sources to base an article off of. That gets thrown out the window, however, when we start talking about works of fiction. In that case you are usually writing an article about the primary source itself, in which case the bulk of your information will come from the primary source. A good chunk of it should also come from secondary sources, true, (otherwise it probably fails WP:N) but in that case I would not say that the article relies on secondary sources so much as it uses secondary sources.
Now obviously this situation only applies to a certain subset of articles, far from a majority, but also far from being "rare" as the current revision of the policy states. (Anybody care to try and count the number of articles on films, just for a warmup?) The basic idea behind the sentences in question is sound, but it needs to be slightly reworded to reflect the fact that some articles by nature rely on primary sources, not secondary sources. -- Y|yukichigai ( ramble argue check) 12:20, 18 August 2007 (UTC)
The NOR policy should reflect the dynamics of the subject matter at hand. An article about a breeding edge issue such will require different sources than an article about the French Revolution. Arebenti 13:19, 18 August 2007 (UTC)
Hi All, Requesting feedback on this: The National Institutes of Health posts a yearly Excel spreadsheet of medical grant awards. About 56,000 entries, $20 billion. Does this fall under WP:NOR: "Figure 2 lists entries from the NIH's 2006 Grant Database containing the words Giardia." On the one hand, it involves some kind of research activity by the author. On the other hand anyone can verify the statement by downloading the Excel file and repeating the search. Thx, Gastro guy 22:49, 18 August 2007 (UTC)
I am interested in putting up entries in WP on the Cairo metro system. No revolutionary theories, just the names of the stations, their general location in Cairo, prominent landmarks nearby, etc. This kind of content is common for other cities such as London, Paris, Moscow, etc. Most metro station pages I have read cite no sources anyway. Since no sources are cited, what is this referred to, if not original research? 41.196.184.19 10:40, 19 August 2007 (UTC)
I just spent somewhere around an hour reading all of the above (and several other pages), and I'm more confused now than I was before I started reading everything. So, without any further ado, here's a particular case in point, hopefully which this group can help solve.
I've been primarily interested in several articles where I was personally invloved in the subject matter, Axe Murder Incident and Joint Security Area (JSA), though I've also made numerous edits to a slew of other articles as well. So, I am an eyewitness to many of the events/articles that I have working on. This makes me a primary source, correct? As a primary source on these articles, "part" of what I edit may fall under original research, which I understand. I try to comply with this rule as much as possible by finding other references that I can cite which support my edits.
However there are several things which I either have problems with, or completely disagree with. First, let me start with a quote from this article that I vehemently disagree with, followed by (what I consider) a very good example. The quote is "The threshold for inclusion in Wikipedia is verifiability, not truth." So, in essence any lies (falsehoods) which have been published or otherwise made publicly available are perfectly legitimate Wikipedia inclusions since they can be verified from the published source, but somebody who was an eyewitness who hasn't published a "document" of some sort can't rebut the falsehood, therefore perpetuating the falsehood into eternity. Now for the example.
Regarding the Axe Murder Incident, the US Army released a statement the following day that one of the reasons for a slow reponse was that the Quick Reaction Force (QRF) was over a mile away from the DMZ at the time of the incident. This is blatantly false to anybody that has ever been there simply due to the fact that at NO TIME is the QRF ever 1 mile away from the DMZ. The farthest the QRF ever gets from the DMZ is approximately 100 meters, which is where the Advance Camp is located, anf this is usually just for a very short period of time. On this particular day and time however, the QRF was actually sitting at the entrance to the JSA which is right smack dab in the middle of the DMZ. Here is part of an email I received from the guy who was actually working the checkpoint at the entrance to the JSA that day.
"Thirdly, to answer the question about who recovered the USA soldiers bodies? It was 3rd Platoon. The information below, "In actuallity, the QRF was sitting in their trucks at CP#2, at the entrance to the JSA, at the beginning of the fight with the 3rd Plt. Ldr awaiting authorization from Capt. Bonifas (who was already dead) for authorization to go in, right smack dab in the middle of the DMZ," is a very accurate statement. Moreover, it took the actual JSA CDR, on this day, to order the 3rd Platoon Leader to go in to recover the bodies.
Lastly, Mike, as you may recall, the above quoted statement truly supports what I have been telling you, M.S. (from the History Channel) and others all along. And, that is, that I was assigned to UNC CP#2 at the time of the August 18, 1976, Axe-Murder Incident. When I initially saw the QRF platoon approaching my guardpost, I was somewhat happy that help was coming. Contrary to what was stated by the 3rd Platoon Leader during the December 28, 2004, airing of the documentary, it was he in the lead vehicle (and not an E-7 Platoon SGT) that I vigorously tried to wave into the JSA -- but my request fell on death ears. Instead, the foregoing platoon leader, subsequently, got out of his vehicle, went into my checkpoint, and then called the JSA CDR, whom, upon his arrival to UNC Check Point #2, ordered him to go in. You can imagine how much time had elapsed! I was saddened by his actions.
I specifically remember this aspect of the incident because due to my assignment at CP2, the LTC ordered me to assure that the soldiers congregating around to view CPT Bonifas' body did not, in fact, see him. I did, and I have been living with the residuals of it ever since. That's why I am telling this story, as it actually occurred."
BTW, my name is Bill, and the reference to Mike above is to another buddy of mine who was also included on the email. I also changed a person's name to just their initials in case they don't want their name known. So, since the US Army issued a statement almost immediately that was wrong in so many ways, it is allowed to stand since the other eyewitnesses and/or myself haven't published anything to contradict it, and the lie is allowed to florish forever? And seriously, if I was to issue to press relase staing the actual events that day, would any news organization really run it or care? No, because it would contradict the "official" version as published by the US Army.
I am very careful about my edits, trying to remain as nuetral as possible and report the actual events as accurately as possible, however this came to a head when somebody wanted to change the word "Murder" to "Killing", claiming that the term murder is POV and trying to diminish the actual events of that day Talk:Axe Murder Incident. I do have some references to a web page I have where I placed some of my recollections from Korea, and also from another buddy who was there with me (Mike, from the quoted part above). Both of our references are listed as "eyewitness account...". Also regarding the term "Murder" for the above immediate problem, a Google search for +"Axe Murder" +korea returns 1100 hits, while the same search that replaces "killing" for "murder" only returns 209 hits. Both sets of hits return some entries for other events, but more inaccurate results occured with the word "killing". So, the term "murder" is also more associated with the events of this day than the term "killing" is as well, adding prevalent thought and opinion to the term.
In conclusion, the history of the world would be in sad, sad shape if it was solely relegated to whichever side expended the most effort in promoting their version of events. I do see where this leads to edit wars, etc., but in many cases, I strongly feel that they need to be decided on a case-by-case basis, where a "group" of educated, non-partisan parties, preferably with both an interest and some background on the subject matter, can decide what should be "allowed" or not. Otherwise, as others have said above (and paraphrased by me), all Wikipedia is, is just a regurgitation of what's already available, so where's the added "benefit" of even using Wikipedia, if all you're are going to see is what's already available everywhere else? Has anybody ever tried (and had success) writing Encylopedia Britannica telling them a cited "fact" is incorrect, offering proof, and actually gotten them to change it? I haven't, and I seriously doubt if anybody else has had any luck in that regard either. That's what makes Wikipedia unique and adds value to many of the articles.
Thanks for any and all opinions on this, regardless of whether there's agreement or disagreement from my POV. wbfergus 15:57, 22 August 2007 (UTC)
The article describes a "battle" between Ken Jennings and Brad Rutter, two US game-show contestants. They have appeared on various game shows, sometimes head-to-head and sometimes not. The article attempts to compare their achievements and declare a "winner" of each "round". I believe that editors have invented the structure of this "battle" themselves, although each specific fact in the article is referenced to a source. Is this OR? -- Cinematical 16:27, 22 August 2007 (UTC)
Cogden, could you make sure you have clear consensus for any changes before making them, please? This is a policy page and it has to be stable.
The change you made didn't make sense to me. You say all sources should be handled the same way. You then say "the source should not be used to support or imply any analytic, synthetic, interpretive, explanatory, or evaluative claims, unless those claims are supported by some other reliable source."
That means that a secondary source cannot be used to make any analytic (etc) claim, unless supported by some other source. So you're now saying that multiple sources must be available for each edit, which is a major change.
It's also not true that all sources should be handled the same way. Primary and secondary sources are, as a matter of fact, handled very differently. The policy has to reflect that. SlimVirgin (talk) (contribs) 18:59, 22 August 2007 (UTC)
I'm not certain that changing the wording to say "based on how they are used" for the Primary, Secondary and Tertiary source use is a purely stylistic change. I think it needs to be discussed and a consensus reached before making this change. Dreadstar † 20:03, 22 August 2007 (UTC)
I did a brief review of who defines primary and secondary sources based on use, and I found some definitions that I think are particularly germane to this policy page, because their "use-based" distinction has a stronger connection to the idea of "original research". Here are a few definitions of primary sources:
Here are a few definitions of secondary sources:
Thus, whether something is a primary or secondary source really depends upon what aspect of the work is being cited. If I'm citing some original research (such as a new scientific theory or a new interpretation of history), it's a primary source, even if the original research was derived from evaluating and analyzing other primary sources (such as raw scientific data or diary entries). In other words, all original research begins its life as a primary source. If someone else adds an additional new idea, they too can be cited as a primary source as the originator of that new idea. Secondary sources don't add any new theories or interpretations, they just comment on primary sources.
According to this framework, sources can be both primary and secondary sources. A biography is a secondary source to the extent it describes, critiques, or analyses primary source material in a non-novel way, but a primary source to the extent it introduces novel theories, analyses, or interpretations about the subject's life. This is not the only way out there to distinguish primary and secondary sources, but given the focus here on the meaning and use of "original research", does adopting this type of framework make sense? COGDEN 21:53, 22 August 2007 (UTC)
This is somewhat related to the above thread, but with a slightly different focus. I would like the cited examples of primary and secondary sources to include something about the application of these terms to the scientific literature. To me, primary sources in this context include journal articles reporting on novel research conducted by the authors of the article. Examples of a secondary source, in a scientific/medical context, would be a review article synthesizing available primary sources, a textbook chapter, or a statement from a large/respected organization such as the WHO/CDC/NIH/etc. The basis for this? Editors may claim journal articles as "secondary sources" and selectively cite specific articles, out of context, to advance their point. Editors should not be in the business of deciding which, of the thousands of peer-reviewed journal articles published every week, are the most scientifically significant ones. Instead, such determinations should be made by experts in the field, as indicated in the sort of secondary sources I've described above. This dovetails with WP:WEIGHT. What I'm suggesting would be, in fact, a very minor change (proposed changes are in italics):
I don't want to open another can of worms about SPOV and so forth; these stipulations would exist to clear up an issue of what is primary vs. secondary within the scientific literature, and would not in any way prioritize scientific sources over non-scientific ones. MastCell Talk 22:35, 22 August 2007 (UTC)
Could you repeat your objections for my sake? No one is saying that an important journal article not covered in a review or the mainstream media needs to be removed. However, such articles are generally handled as primary sources de facto (see WP:MEDRS for example). They can certainly be cited, but when it comes to determining WP:WEIGHT, secondary sources like those I listed are essential. The idea that peer-reviewed == secondary source is one I strongly object to, and one that does not mirror current practice on Wikipedia. There are thousands of peer-reviewed articles published every week; calling them all "secondary sources" puts individual Wikipedians in the business of deciding which of these thousands are most relevant. That determination should be made not by editors here, but by reliable secondary sources such as reviews, textbooks, scientific bodies, etc. MastCell Talk 20:49, 23 August 2007 (UTC)
I'm probably violating something here, but I wanted others to readily see this. Why (or how) did the "Policy" page get unprotected? I though it was being protected until everybody could reach concensus and stop the edit wars on it, and it definately appears that there is no concensus yet. Does someone more vested in this care to revert it back to the protected version? wbfergus 17:26, 31 August 2007 (UTC)
I suggest that if anyone wants to continue a particular thread from the archive they should be welcome to bring that particular thread back in to activity with their new comment. We need not activate the entire recent history, nor refuse to discuss anything that is of particular interest to someone here.-- BirgitteSB 19:44, 31 August 2007 (UTC)
This page is now something like 70 kb meaning half of it should be archived - yet we have made no progress! For starts, can someone refactor the discussion in the sections below this one, moving what belongs into the sections above, and anything else off the page (perhaps to anyother policy or guideline talk page?) Slrubenstein | Talk 09:40, 1 September 2007 (UTC)
I believe the sources section should have two purposes:
I believe that when discussing sources:
I can't really figure out where the discussion is going right now, or why. I can't see why we can't just start with "category 1," discuss whether these are appropriate sources, these are inappropriate sources, these are appropriate sources in certain contexts, or the category should be divided into two or more parts, and then move on to "category 2." I'd suggest starting with tertiary sources, where we can probably reach the widest consensus, and then move "backwards." Jacob Haller 22:26, 31 August 2007 (UTC)
This page is now too long and must be archived, and we are no closer to any resolution. Please put all talk that is simply critical of NOR in another page. Nothing is stoping you from creating a page to propose replacing NOR with a different policy. But stop bitching about the policy here It is not the place. Be constructive. Discuss how to make the NOR policy stronger, or discuss on another page an alternative to NOR. You are free to do either one but they belong on separate talk pages.
My comment was meant to address many people which is why I did not use any names. In any event, talk that questions the value of an NOR policy itself can now go here: Wikipedia talk:Proposal to replace No Original Research Slrubenstein | Talk 12:51, 1 September 2007 (UTC)
My impression of the current policy (not based on any factual checking of the edits) is that it grew over time to include the terms and definitions that now appear contentious. Those same issues really don't belong on the policy page though, as they "really" have nothing to do with the NOR policy, their presence seems to have been added for clarity, but have now become a point of much discussion. So, if the policy was "rewritten" so that it only pertained to NOR, and then had very brief explanations of the other "terms" with links to those pages/guidelines, the policy would be easier to understand for other newbies like myself, and any problems people had with "Sources" or anything else, could be discussed and handled there, where they should be, instead of here. NOR has has nothing to do with defining sources or synthesis, though those are areas where OR usually appears. Keep them separate.
What I created in my Sandbox is an attempt (probably poorly done) to show how some minor changes could help improve the policy and minimize future disruptions of the policy, without actually changing the policy. In previous discussions (now archived) it was asked several times "...then how would you propose to reword ...", and people rarely did. I think Vassyana was the only one who took any effort, but that was solely towards the "Sources" problem. My attempt was to create something to minimize future disruptions of the policy, that others can see and edit without an edit war on the policy page (which looks really, really bad). Others who have problems with the current policy are invited to make their proposed changes there as well. Additionally, if anybody else agrees that removing the "contentious" sections from the policy and having those discussed and maintained separately would benefit the overall policy, I would invite those people to help with rewriting those and appropriately linking them. I don't think this can all be done (and agreed upon) very fast, but I see it as a better use of people's time than just constant "discussing" here. After all, more than a week has gone by with the policy locked, and no progress has been made.
I have two variations. One (using primary, secondary and tertiary source info) is at Sample NOR Policy. wbfergus 10:31, 2 September 2007 (UTC)
The other, using Vassyana's sources proposal is at Sample NOR Policy #2. wbfergus 14:27, 3 September 2007 (UTC)
I've been looking, and the discussion about primary/secondary has been going on for over a year in this talk page. One can look at a long effort like this and see either of two things:
(1) a diligent effort to reach consensus on something
(2) a diligent effort to escape consensus on something
Those who see (1) probably feel that to remove the something now as destructive of a long period of good, hard work. Those who see (2) probably feel that to remove the something now ends a discussion that should have ended far earlier.
At one point in the history I see complaints about a secondary discussion page having been created. I'm not one who asserted "stealth" but the discussion at that time sure resembles an objection to stealth. Sadly, I didn't keep track of when that was, but it was less than a year ago (as I have only looked a little more than one year back in the history.) When I see the date 1 September 2006 on primary/secondary discussion (that doesn't look all that different) I can only shake my head in wonder. -- Minasbeede 14:30, 2 September 2007 (UTC)
I think we should just advise that editors examine existing claims, and note claims which the sources given do not support and vice-versa. Ideally we should have ways to mark citations as fv, the source supports some but not all of the indicated claims; better sources required, the source can be interpreted to support the claim; better sources required, the source was not immediately available; I plan to check it later, and the source was not available; can other editors please check it?. We could even say that some kinds of sources require greater scrutiny than other kinds. Jacob Haller 21:00, 8 September 2007 (UTC)
Which policy is available for situations in which is given a source simply ignored by other contributors? And i not talk about strange and unproof things. I talk about guys that simply not rated at all the sources, even if Wiki should not have the purpuse to prune the sources, expecially if those sources simply not are liked from some guys not exacty NPOV. I have even posted links to sites from the info are extrapolated, but those guys are simply rollbacked accusing me to not have post any source at all. These are situations, sorry to say but that's in which no policy can do enough. Simply helps some very POV guys to act pruning. But still i cannot understand how they can say 'unsourced' when i post sources.--
Stefanomencarelli
18:39, 9 September 2007 (UTC)
If so could we add them to the list of examples of primary sources to make this clear ? Rod57 09:43, 12 September 2007 (UTC)
It is impossible to classify scientific articles as either primary or secondary sources. Take this 2004 paper as an example A trypanothione-dependent glyoxalase I with a prokaryotic ancestry in Leishmania major.
In the fist paragraph of the Results section I quote some dissociation constants I measured. If a Wikipedia article cited these results, it would be using the paper as a "primary source" - a source that presents novel, previously-unpublished data.
In the first paragraph of the Discussion on the other hand, I discuss previously-published data on Methylglyoxal synthase in the light of my results and bring these data together and interpret them. Here, if a Wikipedia article cited the paper on the possible problems caused by methylglyoxal synthase, it would be using the paper as a secondary source. Scientific papers are neither primary nor secondary sources, they contain elements of both. Tim Vickers 17:16, 12 September 2007 (UTC)
It would seem from the title of Tim Vicker's article that he is presenting data that is of evolutionary significance (for a single species.) He's saying (I think) the evidence indicates something about the evolution of a particular species, about how it should be classified. Is that conclusion primary or secondary? It is Tim Vicker's OR, published. His conclusion: primary or secondary?
I should point out that if/when others cite Tim Vicker's paper used here as an example they far more likely will be citing it for its conclusion than for its raw data or its review of other work. The conclusion is the part that constitutes the major part of the advancement of knowledge that the article presents. -- Minasbeede 20:05, 12 September 2007 (UTC)
One of the biggest concerns of the referees of a paper is whether or not it is novel, whether or not it is the primary source for a discrete (and novel) idea. If it just rehashes old material the paper ought to be rejected and usually is. The referees don't act to restrain publication to fall within what is already known/believed, they act to require that whatever is published extends what is known/believed. (Review articles and the like excepted.)
The referees may not know whether the author's conclusions are correct. If they see obvious incorrectness they will alert the author in their referee statements and exhort the author to correct errors. The editor may refuse to allow publication until the author does so. If they're not certain they may say that in their comments but recommend publication. (I think there's a story about Physics Review Letters giving priority to all the papers that the referees didn't understand. The referees did not want to impede the spread of new, good ideas and since they didn't understand they couldn't judge "good," so by default they said "publish.")
I agree that probably not all aspects of the issue were considered when the formulation was considered. We're doing some considering now. -- Minasbeede 20:41, 12 September 2007 (UTC)
Many peer reviewed articles on theoretical subjects should be considered primary sources. E.g., the article in which supersymmetry was first proposed, string theory, special relativity, General relativity, inflation theory, conformal symmetry, etc. etc. etc. Count Iblis 20:51, 12 September 2007 (UTC)
I don't want to mess up the section below (leave it clean for better responses) but this is a response to it. I've done a recent personal review myself and have recognized that my reason to be paying attention to WP:NOR had nothing to do with source typing: I'm troubled by the synthesis section (which I'll not discuss now: there are issues enough.) So I came into the source typing discussion more because I saw it and recognized I had thoughts and an opinion than because I had really strong feelings about it. (Or if I have strong feelings I don't have any agenda behind those feelings.) It looks to me a lot like source typing is inherently too cumbersome to save. The bottom line in source typing is "primary bad" or "primary suspicious" or "bogeymen lurk in primary sources" or something of the sort. The problem is that yeah, sometimes the problem is there - but sometimes it isn't. A policy applies to all. There are cases where I wouldn't hesitate to use 17th century sources. That's "use," not "misuse." No matter how the source typing section is written I'll pay very scant attention to it. There are overriding principles that existed before any source typing was done and those seem to me to be fully adequate. While there are 17th century sources I'd willingly use there are sources from this month I'd never even consider. I recognize that the issue is framed primarily as one of source type but there's the undercurrent of "that-primary-stuff-wasn't-checked-out" and "that-primary-stuff-is-from-long-ago-and-may-not-correspond-to-what-modern-scholars-say." The problem is that these are being made a universal (BEWARE ALL PRIMARY SOURCES) and that's not needed. On top of that is the poor choice of words and on top of that is the confusion that still exists over the Tim Vickers sample article, as an example. There's at least three aspects of the article that have been identified: the experimental data, the historical review (if that's what Tim Vickers called it) and the conclusions. If the article persists and is cited it will mostly be the conclusions that are the reason for the citation. Those conclusions are of the bugaboo "primary" source variety. I think the conclusions are not outrageous and almost certainly will survive every test of time (not having read the article I'm not basing this on what it says but on a feeling.) It's primary, it will (probably) remain always citeable. I think it is plain wrong to put all such articles under any sort of cloud. I don't think the cloud is justified. Yeah, yeah, the article was refereed but if it were a chapter in a book it wouldn't be: books, which on the one hand Wikipedia anoints as "good" sources, often are also primary sources and would fall under the source-typing cloud. Or there'd be a clause explaining why books, though primary sources, aren't really supposed to be watched (or, for all I know, the policy might end up saying "books in particular have to be treated with caution.")
Besides which an editor could be quite aware of many sources that agree with each other and select the primary source because it's the best selection for what the editor is doing. Or he could know that the source he chose is the only source available. He's supposed to use verifiable sources. As yet he's not required to exhaustively explain why he chose any particular primary source - but that could be coming, for all I know. Just like primary soures, the editor who chooses to use one is under a cloud.
Besides which it's rather absurd to make any general statement about source types ahead of their use. If 99.9% of some type of source is bad (whatever "bad" means) then probably that source type will be used infrequently. I can't see how anyone can feel that it's correct to a priori cast aspersions on an editors choice of source, including that 0.1% of the time when it's "good."
I can't help tighten the wording: my heart isn't in it. I'd rather chuck the whole thing. I can live with its being a guideline. -- Minasbeede 02:28, 13 September 2007 (UTC)
I propose the creation of a template for new editors that explains to them why their additions were reverted if they are original research. Unless there is one that already exists that I am not aware of? Sefringle Talk 18:15, 16 September 2007 (UTC)
I suggest that an example is agreed on to test in actual practice what we are talking about. Perhaps the article Gillian McKeith would serve. The use of Ben Goldacre's words from his blog versus the use of his words from a newspaper is one issue. The use of sources that refute the "poo lady"'s claims that do not use her name is another issue. As BLP's represent our strictest use standards, and the data and issues are small in number, and there is no ongoing major conflict at that article, perhaps it is a good choice. WAS 4.250 01:08, 17 September 2007 (UTC)
Let me be clear: I think this is a generally commendable policy, but one that can often be used in bad faith by "wikilawyers" or by people who just like to waste other people's time (like if someone says, "There is no citation here for the claim that these two words rhyme", and while that example is hypothetical, it's not exaggerated.)
I think two major things need to be added to the policy page:
- Jmabel | Talk 19:02, 16 September 2007 (UTC)
This is an alternative to Blueboar's proposal. I would have no objection to changing the word "sources" to "materials." it also may be possible to incorporate elements of Blurboar's proposal and mine. Slrubenstein | Talk 20:05, 16 September 2007 (UTC)
First, let me know Blueoar's version makes the same distinctions. Second, I tried to respond to your (and others) concerns in my rewording. Slrubenstein | Talk 20:37, 16 September 2007 (UTC)
I think Vassyana is right which is why in my version I wrote "However, some encyclopedias and other tertiary sources, such as Annual Reviews, have signed articles, and often articles that explicitly promote the author's own views. In this sense, a tertiary source can also be viewed and treated as a secondary source." perhaps Vassyana thinks this needs to be stated more strongly or clearly, or would suggest a specific change to my proposal? If it is in line with the sentence I just quoted, I am pretty confident I will hav no objection. If it diverges from what I wrote, I am certainly open to suggestions and will try to accommodate or dialogue until we agree. If you think the caveate at the end is simply insufficient, offhand I see two suggestions: (1) beaking it off into a new paragraph exclusively for when tertiary sources are encouraged i.e. one paragra[h that is proscriptive, and another that is prescriptive. OR, (2) divide "tertiaty sources" into two kinds, anonymous and signed. Other ideas? Slrubenstein | Talk 12:07, 17 September 2007 (UTC)
I think we are rapidly moving towards consensus. I have no objection to changing "sources" to "materials" in my proposal, which I think would please many other editors here. I have no objection to reworking the section on tertiary materials, in line with Vassyana's concerns, and I have suggested two possible solutions and invited other suggestions.
There has been little criticism of the section on the history of the proposal/its origins that I proposed; but I hope people will at least go over it for style and clarity.
I extended protection another week just so we can focus on these proposals and keep moving towards consensus. Hopefully all current conflicts will be resolved before the week is out. As we move forward any admin can move material into the policy even in its protected state. My intention is that the continued protection give us space to finish the forward movement to consensus on these issues.
By the way, I know Llywrch, Vassyana and others have been working on an essay on what is not original research. I suggest this: that after we achieve agreement on the two proposals (one on originas of the policy, one on materials) we move that essay to this page and consider working on it as a proposed new section of the policy next. Slrubenstein | Talk 12:21, 17 September 2007 (UTC)
OK, here is the Spenny re-write, which paraphrases the thinking, but adds in a couple of points. I've bolded the key phrase which is the root of what all this source jiggery-pokery is about. It is this key concept that we are battling with, but it is not explicitly stated. Spenny 23:59, 17 September 2007 (UTC)
Within source material we will find a mixture of facts and interpretations. Some sources will be basic observations that offer little in the way of analysis; other sources will offer analysis of information and draw conclusions. When adapting material for use in Wikipedia, it is important to only introduce facts and observations that already exist.
A Wikipedia article or section of an article can employ statements of facts only if the material is used (1) only to make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge, and (2) never to make analytic, synthetic, interpretive, explanatory, or evaluative claims. Contributors drawing on statements of facts should be careful to comply with both conditions.
Typically, statements of fact will tend to be found in materials such as historical documents; personal diaries, census results, video or transcripts of surveillance, public hearings, trials, or interviews; raw tabulations of surveys or questionnaires; written or recorded notes of laboratory and field experiments or observations; and artistic and fictional works (such as poems, scripts, screenplays, novels, motion pictures, videos, and television programs).
Wikipedia articles can include analytic, synthetic, interpretive, explanatory, or evaluative claims (1) only if such claims come from a reliable, verifiable source and (2) the point of view is clearly identified and accurately represented. In general, it is this type of source that is of most use to Wikipedia as an encyclopaedic article is a summary analysis of a topic and using raw facts to build an article will often require too much original work that could be challenged.
Further, it is most likely that works which are a comprehensive analysis of the topic will be most suited to providing a comprehensive demonstration of a particular viewpoint. Cherry picking individual concepts from different sources may suggest editing to support a point of view, and this can be especially so in the case of using passing references. (This is really NPOV, and I haven't worked this up well)
Useful analysis is typically found in academic journals, Government Inquiries, and media sources such as serious documentary programs and a few respected newspapers. {Yes, I know this bit is weak, but its past my bedtime).
Wikipedia strives to be a superb encyclopedia in its own right. Since Wikipedia does not want to be derivative, materials found in encyclopedias or similar reviews are often of limited value for Wikipedia research. Annual Reviews and Encyclopedia Brittanica articles often provide extensive bibliographies that are valuable tools for identifying important materials, and therefore of great use to Wikipedia editors. Nevertheless, these materials do not necessarily have the same content policies as Wikipedia and for this reason should not be viewed as authoritative. However, some encyclopedias and other material, such as Annual Reviews, have signed articles, and often articles that explicitly promote the author's own views which can be viewed as source material in their own right.
Ok... I think we are at the stage where we agree on the basic principals (1) we want to talk about material and not sources (2) the issue we want to adress is the misuse of primary materials - eitors taking those materials out of context and making analytical, interpretive, or conclusionary statements from them. (3) All but the most basic analytical, interpretive or conclusionary statements included in our articles should be backed by reliable sources (defined as secondary material) that contains the same analysis, interpretation or conclusion.
Now to the hard part... drafting language. As a starting point, I am going to actually propose the language I drafted above (in my "what if we just swapped words" thread). We probably need to edit it further (for example: dropping the tertiary material section?). But I think it may get us off to a good start. Anyone disagree? Blueboar 12:33, 16 September 2007 (UTC)
Well, I don't have a problem with the alternative... but I suspect that others may. For one thing, it does not solve the problem that not everyone agrees with our definition of what a "primary source" is. This version certainly defines it in a much clearer way, but it is a definition that is unique to Wikipedia. Personally, I have never had a problem with that... I understand that in Science the term means one thing, in History it means another... while on Wikipedia it means yet something else. When I apply the NOR policy on Wikipedia, I use Wikipedia's definition, just as I would use the Science definition if I were writing a scientific paper. But, based on the comments that keep being raised here... I am not sure if others can make that distinction. Was there a reason you wanted to keep the word "source"? Blueboar 23:38, 16 September 2007 (UTC)
I've had a look at the proposal above and also Vassyana's drafts. Sorry to be a dog with a bone, but I am still struggling to see what the differentiation of sources brings to the party. The list of primary sources still has the issue that as a generalisation it doesn't work. Simple example, TV documentary could well be a sound secondary source. The list does not explain what makes it a source for fact as opposed to a source for opinion and so this simply raises the query - is Pepys' Diary a source of fact or opinion? It also strikes me that the policy statements made aren't really dependent on this source typing. I guess the other observation is that the differentiation on fact and opinion doesn't really hold water with distinguishing with source typing by example, though it is clear that the distinction between fact and opinion is the issue that is being explored by the policy wording (which I think stands up pretty well on its own). Spenny 16:56, 17 September 2007 (UTC)
Two points - and if both of you agree, I would ask you to suggest appropriate changes. First, any list is of illustrative examples. By this I mean two things: first, it is not inclusive; second, it is not exclusive by which I mean these things are not always examples of (primary, secondary, etc.) (sources, materials, etc.) This shouldn't be too surprising: an apple pie could be (1) an example of my mom's cooking (2) an example of things you can do with fruit (3) an example of a desert (4) an example of symbols for America etc. This leads to my second point: whether something is a primary or secondary whatever depends on how it is used, the context. perhaps this gets closer to Ian's attmpt to get at the concept; i am suggesting that the concept involves how something is used and in what context. Does this make sense to you guys? If so, what language could more effectively communicate this? Slrubenstein | Talk 17:57, 17 September 2007 (UTC)
Some observations:
COGDEN 19:57, 18 September 2007 (UTC)
It's easy. Whenever there's a dispute between a more restrictive and a more flexible version of the policy, just find an example of something that would be allowed under one version but not the other, and claim that anyone who supports allowing that is against the policy and should therefore be excluded from the discussion. It's not necessary to actually succeed in excluding anyone; it's enough to put a chill on the making of certain types of assertions. -- Coppertwig 16:43, 17 September 2007 (UTC)
Taking what Ian wrote and tweeking.... mostly adding 1) that facts need to be cited to reliable sources as well as opinions... and 2) stressing that analysis etc should be the same as in the source. I could live with this approach. Blueboar 01:15, 18 September 2007 (UTC)
Within source material we will find a mixture of facts and interpretations. Some sources will be basic observations that offer little in the way of analysis; other sources will offer analysis of information and draw conclusions. When adapting material for use in Wikipedia, it is important to only introduce facts and observations that already exist.
A Wikipedia article or section of an article can employ statements of facts only if the material (1) comes from a reliable, verifiable source, (2) is used only to make descriptive claims and (3) never to make analytic, synthetic, interpretive, explanatory, or evaluative claims. Contributors drawing on statements of facts should be careful to comply with all three conditions.
Typically, statements of fact will tend to be found in materials such as historical documents; personal diaries, census results, video or transcripts of surveillance, public hearings, trials, or interviews; raw tabulations of surveys or questionnaires; written or recorded notes of laboratory and field experiments or observations; and artistic and fictional works (such as poems, scripts, screenplays, novels, motion pictures, videos, and television programs).
Wikipedia articles can include analytic, synthetic, interpretive, explanatory, or evaluative claims (1) only if such claims come from a reliable, verifiable source, (2) the point of view is clearly identified and accurately represented, and (3) the analysis, syntethis, interpretation, explanation or evaluation matches that in the source. In general, it is this type of source that is of most use to Wikipedia as an encyclopaedic article is a summary analysis of a topic and using raw facts to build an article will often require too much original work that could be challenged.
Further, it is most likely that works which are a comprehensive analysis of the topic will be most suited to providing a comprehensive demonstration of a particular viewpoint. Cherry picking individual concepts from different sources may suggest editing to support a point of view, and this can be especially so in the case of using passing references.
Useful analysis is typically found in academic journals, Government Inquiries, and media sources such as serious documentary programs and a few respected newspapers.
Wikipedia strives to be a superb encyclopedia in its own right. Since Wikipedia does not want to be derivative, materials found in encyclopedias or similar reviews are often of limited value for Wikipedia research. Annual Reviews and Encyclopedia Brittanica articles often provide extensive bibliographies that are valuable tools for identifying important materials, and therefore of great use to Wikipedia editors. Nevertheless, these materials do not necessarily have the same content policies as Wikipedia and for this reason should not be viewed as authoritative. However, some encyclopedias and other material, such as Annual Reviews, have signed articles, and often articles that explicitly promote the author's own views which can be viewed as source material in their own right.
I have been following this debate only intermittently, so some of the issues I mention below may have already been discussed. If so please simply point me to the correct section. (for easy labelling I'll continue to use PS, SS and TS terminology in my comments below)
Abecedare 02:47, 18 September 2007 (UTC)
I like this because it is a fresh approach, but I don't think it reflects Wikipedia practice, maybe in part because the language is not precise or comprehensive enough. "Statements of facts" are frequently used to support analytical, synthetic, etc. claims, particularly when the statements of fact are found in the same source as the analytical claims. Usually, an editor will cite the fact, cite the author's interpretation of the fact from the same source. The fact is used by the author to make the analytical claims, not the editor, it's not original research. Such a use of "statements of facts" would barred by this draft.
Also, since this changes the primary/secondary dichotomy to a fact/interpretation dichotomy, we have the additional problem of defining what is a "fact" and what is an "interpretation". This seems a bit too metaphysical for a pillar Wikipedia policy. For example, if Author X says "Joe was drunk", is that a fact, or an interpretation? And does it matter? Both facts and interpretations can be cited by Wikipedia editors, so long as neither the facts nor the interpretations originate with the editor.
As a corollary to the above, how can we really say that analytical sources are "of most use to Wikipedia"? Sources with pure, raw facts are useful too, maybe even more useful, since without the raw facts, what is there to interpret? COGDEN 20:27, 18 September 2007 (UTC)
The more I think about all of this, the more I think that we were approaching this back-asswards. The point of this policy is that we have to answer the question: "Oh yeah? Sez who?" with... "This reliable source over here sez so." If the answer is "I say so" or "the facts say so", or some thing like that, you are venturing into OR. The type of source being cited does not realy matter, the type of statement that it is being cited for does. For a statement of fact, we need to cite a reliable source that states that it is a fact. For a statement of opinion, we need to cite a reliable source that states that opinion. And for a statement of analysis, interpretation, synthesis, conclusion, etc., we need to cite a reliable source that contains the same analysis, interpretation, synthesis, conclusion, etc. Ian's approach is good... it does essentially approach the issue based on what the source is being used for, instead of what the source is. But it is a relatively long winded way of saying it. Can we be more concise? Blueboar 13:47, 18 September 2007 (UTC)
<undent> Good point that it's what the source is being used for rather than what the source is, it ties in with thoughts I've been trying to pull together. Primary material can be raw data, but of course it can also be an original synthesis. Thus something that's a secondary source becomes primary material when it's the subject of the article or of the part of an article. An example which includes no raw research is
Icons of Evolution. This book is a secondary analysis of primary sources, in a synthesis which is notable only as primary material about the ideas it promotes. The book has been thoroughly analysed by various secondary sources which form the basis of the article.
Perhaps another way of looking at the issue is that editors have to be careful to summarise accurately both the primary material providing the facts that are the subject of the article, and secondary analysis, description or summaries of those facts. Particular care has to be taken to avoid inadvertently summarising primary material in a way that presents our own opinion or understanding about the facts, so it's advantageous to use secondary material which shows attributable analysis.
In my view the primary/secondary distinction is useful as a way of assessing material, and while some examples are useful for explanation, we don't want a prescriptive list of what fits in each category. ..
dave souza,
talk
20:21, 18 September 2007 (UTC)
I'm becoming rather concerned with the view, largely espoused by Cogden, but seemingly being repeated by others now, that sources that produce analysis based on primary source/materials/whatever are, in themselves, primary sources for that analysis. That is not how the terms primary/secondary sources are used at all. Presenting new analysis, synthesis, commentary, etc etc etc is exactly what secondary sources do, and they are still secondary sources. Don't forget that these terms originate in history/historiography. SamBC( talk) 20:35, 18 September 2007 (UTC)
OK... on the theory of puting my money where my mouth is... here is a very rough draft of a short, concise... use oriented paragraph:
Citing the Right Source Materials
Within Wikipedia articles we will find statements of fact and statements of interpretation, analysis or conclusion. It is important to cite the right sources to back those different types of statements. Statements of fact should be cited to reliable sources that clearly demonstrate that fact. Statements of interpretation, analysis or conclusion should be cited to reliable sources that contain the same interpretation, analysis or conclusion. Editors should take care not to confuse a statement of fact with one of interpretation analysis or conclusion. To say that "The Constitution of the US guarantees the right to own a sub-machine gun" sounds like a statement of fact, but is in fact a statement of interpretation (the interpretation being that The 2nd Amendment applies to sub-machine guns). Thus we should not cite the Constitution itself for this statement, instead we should cite an article by a constitutional scholar that contains this interpretation.
The example may not be the best... but it was worth a try. In any case, is this more in the right direction ... or am I off base again? :>) Blueboar 20:23, 18 September 2007 (UTC)
For a better example, how about:
(If that's too U.S.-centric, feel free to substitute something else.) I think this has an advantage in indicating both the type of extrapolation that is allowed and the type that is not. -- Jmabel | Talk 23:18, 18 September 2007 (UTC)
Removed for clarity Spenny 17:24, 20 September 2007 (UTC)
Ok... I think this is the most recent version of the proposal, incorporating those comments that seem to have consensus. I would ask that people not make major edits to it until they have been fully discussed and have reached some degree of consensus. If you think something should be added/changed/deleted from this... post it as a new suggested draft with a new date/time heading. The idea is to keep the different versions clear so people can compare them and comment on them. Blueboar 16:05, 20 September 2007 (UTC)
Citing the Right Source Materials
Within Wikipedia articles we will find statements of fact and statements of interpretation, analysis or conclusion. It is important to cite the right sources to back those different types of statements. Statements of fact should be cited to reliable sources that clearly demonstrate that fact. Statements of interpretation, analysis or conclusion should be cited to reliable sources that contain the same interpretation, analysis or conclusion.
Editors cannot include their own interpretations of previously-published facts, unless that interpretation is either 1) an obvious and non-controversial consequence of the facts or 2) can be attributed to a reliable source. Nor can editors expand on an author's interpretations of fact, unless that expanded interpretation is also found in a reliable source.
For example, that Franklin D. Roosevelt survived polio and that he became president of the United States are both verifiable facts that may be cited in a Wikipedia article. The dates of each event are easily verified, and it would be perfectly admissible to say that he "survived polio, and later became president of the United States" without finding a source that specifically uses the word "later", or otherwise explicitly compares the dates. However, any statement about the effect of the disease on his career would require separate citation, since it is not an obvious or non-controversial consequence of these easily verified facts.
I like this... the FDR example is a great help, although I know that there will still be people who wikilawyer this exception to death. Blueboar 23:35, 18 September 2007 (UTC)
I could support this. It clearly illustrates the point. Vassyana 02:08, 19 September 2007 (UTC)
I could support this. It clearly illustrates the point. WAS 4.250 08:26, 19 September 2007 (UTC)
Good. Very good. However, is "editors cannot" the best description? We know that editors can, until someone reverts them. Would "editors cannot," "editors should not," "editors must not," or something else work best? Jacob Haller 04:51, 19 September 2007 (UTC)
See, here's why sourcing is so important, and why examples in policy are dangerous: FDR never had polio. See Franklin D. Roosevelt's paralytic illness. – Quadell ( talk) ( random) 12:07, 19 September 2007 (UTC)
I agree with the draft, with a couple of tweaks: We can call it a "paralytic illness" instead of polio, and there's a minor grammar issue. I think the example is a good one. I know examples are difficult in policy pages, but I think one would be very helpful here. Model examples can work effectively to elucidate policy (I'm thinking about the American Restatements of the Law as a non-Wikipedia example). COGDEN 20:19, 19 September 2007 (UTC)
As discussed above, it seems to me that the draft would work well with a subsection concisely explaining the usefulness of primary/secondary/tertiary sources. I've based this proposal (as shown at #Primary, secondary, and tertiary sources above) on the existing policy with some changes to bring it into line with the draft, and the examples etc. commented out. Slrubenstein's proposal may form a better basis for this subsection, but it seemed to me to be useful to put a proposal up for discussion about the principle. ... dave souza, talk 12:24, 20 September 2007 (UTC).. add link 12:38, 20 September 2007 (UTC)
We are moving fact and furious folks... and I am a bit confused as to which draft we are now discussing... would someone please post a "proposed" version that incorporates the latest thinking? (perhaps we should entitle them with: Proposed draft as of date/time and periodically update?) Blueboar 13:58, 20 September 2007 (UTC)
I'm opposed to any changes to the long-standing formulation of this key policy. The proposed changes open the door to a long and slippery slope, and for that reason alone I'd objection to their addition, much less their specifics. I appreciate the effort here, but once we start making exceptions, it will never end. FeloniousMonk 15:30, 20 September 2007 (UTC)
![]() | 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. |
Archive 15 | ← | Archive 18 | Archive 19 | Archive 20 | Archive 21 | Archive 22 | → | Archive 25 |
From all these discussions, I can't help feeling that there's some confusion, or a lack of clarity, as to what type of source (primary/secondary/tertiary) a given source is. As I understand the definitions, a primary source is a photograph, video, eyewitness statement (in the context of, say history), or a TV show itself, a script, a novel, etc. A secondary source is anything that isn't primary. A tertiary source is a special case of secondary sources which do not use any primary sources themselves.
Thus, if someone edited wikipedia with information they had gained first-hand, that would be unsourced, and the article would itself be a primary source. If they directly quote an eyewitness or a text, then that's sourced from a primary source, and the article is a secondary source. If the only directly used sources are secondary (including tertiary sources), then the article is a tertiary source (often seen as the ideal for wikipedia).
The principle here is original research - the composition of a primary source, or the synthesis of primary sources into a secondary (non-tertiary) source are research. If this is done for the article, then that is original research. Thus, if a person watches a TV show, reads a book, watches a film, or any similar activity and then writes a plot summary, that's original research. If they find one or more plot summaries elsewhere and reference them, then it's not original research, it's synthesis or composition of secondary sources.
Journal articles aren't a primary source, generally speaking - raw experimental data is a primary source. Primary sources are typically (but not always) devoid of interpretation. Interpretation is added through research in the synthesis of the secondary source.
WP:NOR is saying that we don't do interpretation, we do synthesis compilation.
Does that make sense to people? SamBC( talk) 23:07, 14 August 2007 (UTC)
Further to all this, I've just started a new "proposal" at Wikipedia:Classification of sources. I don't know for sure that it'll be useful, it's certainly not in the right tone yet, and if it is useful I don't know if it should stay a separate page. However, that's useful for working on it. If you think we should have a clear internal definition or illustration, please join me and help to build it up and put it in the right tone. It's mostly copy-paste from my initial comment above. SamBC( talk) 02:44, 15 August 2007 (UTC)
In case anyone's interested, there's a Village pump discussion on the pri/sec/tert sources issue. Dreadstar † 21:23, 15 August 2007 (UTC)
"Original research that creates primary sources is not allowed. However, research that consists of collecting and organizing material from existing primary or secondary sources within the provisions of this policy is, of course, encouraged: this is "source-based research," and it is fundamental to writing an encyclopedia.
Although most articles should rely predominantly on secondary sources, there are rare occasions when they may rely on primary sources (for example, legal cases). An article or section of an article that relies on a primary source should (1) only make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge, and (2) make no analytic, synthetic, interpretive, explanatory, or evaluative claims. Contributors drawing on primary sources should be careful to comply with both conditions."
Just some thoughts. Your thoughts? Vassyana 23:21, 11 July 2007 (UTC)
"
Original research that creates primary sources is not allowed.Research that consists of collecting and organizing material from existingprimary or secondarysources within the provisions ofthis policythis and other content policies is, of course, encouraged: this is "source-based research," and it is fundamental to writing an encyclopedia.
Although most articles should rely predominantly on secondary sources, there are rare occasions when they may rely on primary sources
(for example, legal cases). An article or section of an article that relies on a primary source should (1) only make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge, and (2) make no analytic, synthetic, interpretive, explanatory, or evaluative claims. Contributors drawing on primary sources should be careful to comply with both conditions."
There. ≈ jossi ≈ (talk) 00:07, 12 July 2007 (UTC)
I agree with the edits in the guideline. And I understand the preference for reliable, secondary sources, since these may be the result of more fact-checking than the primary source could or did. However, when secondary sources are not (freely) available, I have no problem with using a primary source provided the use thereof is clearly indicated in the article.
—
Xiutwel
(talk)
08:08, 12 July 2007 (UTC)
Due to the supporting comments and lack of objection, I have changed the page to reflect Jossi's proposal. Vassyana 13:10, 18 July 2007 (UTC)
It occurs to me that there's a pretty good, relatively objective criterion that we could use to cover when primary sources are appropriate, and it ought to cover pretty much every case:
Primary source are acceptable when they require and receive no interpretation in the article, but are merely a source of an objective fact.
Thoughts? SamBC( talk) 02:16, 15 August 2007 (UTC)
"Primary sources may be used in articles in the context they are employed in secondary and tertiary reliable sources. Primary sources that are considered accurate by reliable sources may be used for citation of "pure fact", including such raw data as U.S. census statistics."
(outdent) A couple words and an order swap might do the trick!
"Primary sources that are considered accurate by reliable sources may be used for citation of "pure fact", including such raw data as census statistics. Primary sources may also be used in articles in context as they appear in secondary and tertiary reliable sources."
Thoughts? Vassyana 05:17, 15 August 2007 (UTC)
I have given this issue a lot of thought for some time and I will direct to something I wrote previously about it. Pulling out my thoughts on question of when these sources are appropriate: sources which should only be be used with extreme care, making sure they are only cited to document that such a source makes such a claim (direct quotes of the source, careful paraphrase of what the author of the source claims, facts and figures in infoboxes).-- BirgitteSB 14:28, 15 August 2007 (UTC)
The problem I have with census data is that it can be cherry picked in ways that might even appear to be benign. TableManners 03:22, 16 August 2007 (UTC)
This line referring to OR contains an unnecessary assumption of bad faith:
I changed "therefore believed to be" to "could be" because we don't have to assume it's OR, the simple fact that it could be OR is enough. But the change was reverted. WP:AGF should be followed, not excepted in this case. Dhaluza 09:33, 15 August 2007 (UTC)
Yes, yes. So make some examples. If i say 'today it's hot' what sources do i must give to you? The C° meaured by my termometer? But it's OR, right? And maybe someone don 't find 35° 'hot' enough to be called so.
Behind my house there is an old church i don't know how wide it is and then i decide to measure it. Oh, my God, there is still an OR, and perhaps also bad faith, to add insult to injury.
We cannot copy other sources, because it' Copyviol, we cannot even resume them because it's OR (no syntesis, right?). Do you have the clue? Where is the good sense? Tell me how one can be authorized by your policies to do something that is not simply copy texts older than 90 years. Feeling as potential criminals every time that we post something that is not a he said so. That's the meaning of OR, IMHO.
I assert you cannot create an encyclopedia based entirely on quoting other sources, unless very often those other sources are other encyclopedias. Agree.-- Stefanomencarelli 15:58, 16 August 2007 (UTC)
{{ editprotected}} The edit war was successful at removing a long standing sentence:
“ | Wikipedia articles should rely on reliable, published secondary sources. | ” |
Can we put this back in and then reprotect? TableManners 03:13, 16 August 2007 (UTC)
No original research can be added to a wiki. So the only choices is to grab a researched document and copy inside the wiki but it's illegal (copyright) or you just can publish material that was created by someelse and changed it bypassing the copyright protection (cheating) or using published material that the original author allow to publish in the web.
Even more, there are a lot of "not so important material", that never was researched by a authority or a compentent entity, this kind of material cannot be showed in Wikipedia or even worst can be used by wikipedia only by a biased point of view, for example if we talk about Microsoft, there are little room for any outside to Microsoft to talk about it but Microsoft have almost all the rights and authority to talk about himself.
Anyways empiric talking, the "no original research" rule in wikipedia is not widely applied, common sense overule any no practical law. No original research is used currently to bash someelse, asking for any tiny details about a specific text, it's not anymore (and never was) a way to keep clean wikipedia. -- Magallanes 18:58, 16 August 2007 (UTC)
While you continue to discuss without problems about the santity of NOR and its stricth interepretations, in Wiki there are stuff like this: Domination of the skies, superior seamanship, and prudent, timely maneuvers helped to nullify the overwhelming odds. In the highest tradition of naval service, the finest qualities of the American sailor became commonplace during the heroic fight. Devotion to duty, daring courage, uncommon bravery, and an indomitable spirit were part and parcel of this victory.
So don't cause me LOL in so indecent manner. This piece is what could be called shameless agiography and pubblicity for US Navy, and there are thousands of this stuff. While i try to reduce every optimistic overclaiming made by italian aviators with the cross controls with several sources (it's unuseful to post that x has downed 12 enemy aircraft when are available sources that considering both sides, dismissed all this claimings without any problem), there are thousands of pages like this, written directly by J. Wayne. Obviousely you are not worried about, your goal is forbid brain activity. I am, instead. Babbling about NOR when gratuitus agiographies are so well spread is a thing that shows how these discussions and policies don't guarantee nothing. Sorry,but it so.-- Stefanomencarelli 13:59, 20 August 2007 (UTC)
I just noticed that the article has been protected due to edit warring, so I thought I'd take a look as to why. I think there's some merit to the "compromised" rewording of the sections in question.
For most articles common sense reasonably dictates that secondary sources are going to be the ideal sources to base an article off of. That gets thrown out the window, however, when we start talking about works of fiction. In that case you are usually writing an article about the primary source itself, in which case the bulk of your information will come from the primary source. A good chunk of it should also come from secondary sources, true, (otherwise it probably fails WP:N) but in that case I would not say that the article relies on secondary sources so much as it uses secondary sources.
Now obviously this situation only applies to a certain subset of articles, far from a majority, but also far from being "rare" as the current revision of the policy states. (Anybody care to try and count the number of articles on films, just for a warmup?) The basic idea behind the sentences in question is sound, but it needs to be slightly reworded to reflect the fact that some articles by nature rely on primary sources, not secondary sources. -- Y|yukichigai ( ramble argue check) 12:20, 18 August 2007 (UTC)
The NOR policy should reflect the dynamics of the subject matter at hand. An article about a breeding edge issue such will require different sources than an article about the French Revolution. Arebenti 13:19, 18 August 2007 (UTC)
Hi All, Requesting feedback on this: The National Institutes of Health posts a yearly Excel spreadsheet of medical grant awards. About 56,000 entries, $20 billion. Does this fall under WP:NOR: "Figure 2 lists entries from the NIH's 2006 Grant Database containing the words Giardia." On the one hand, it involves some kind of research activity by the author. On the other hand anyone can verify the statement by downloading the Excel file and repeating the search. Thx, Gastro guy 22:49, 18 August 2007 (UTC)
I am interested in putting up entries in WP on the Cairo metro system. No revolutionary theories, just the names of the stations, their general location in Cairo, prominent landmarks nearby, etc. This kind of content is common for other cities such as London, Paris, Moscow, etc. Most metro station pages I have read cite no sources anyway. Since no sources are cited, what is this referred to, if not original research? 41.196.184.19 10:40, 19 August 2007 (UTC)
I just spent somewhere around an hour reading all of the above (and several other pages), and I'm more confused now than I was before I started reading everything. So, without any further ado, here's a particular case in point, hopefully which this group can help solve.
I've been primarily interested in several articles where I was personally invloved in the subject matter, Axe Murder Incident and Joint Security Area (JSA), though I've also made numerous edits to a slew of other articles as well. So, I am an eyewitness to many of the events/articles that I have working on. This makes me a primary source, correct? As a primary source on these articles, "part" of what I edit may fall under original research, which I understand. I try to comply with this rule as much as possible by finding other references that I can cite which support my edits.
However there are several things which I either have problems with, or completely disagree with. First, let me start with a quote from this article that I vehemently disagree with, followed by (what I consider) a very good example. The quote is "The threshold for inclusion in Wikipedia is verifiability, not truth." So, in essence any lies (falsehoods) which have been published or otherwise made publicly available are perfectly legitimate Wikipedia inclusions since they can be verified from the published source, but somebody who was an eyewitness who hasn't published a "document" of some sort can't rebut the falsehood, therefore perpetuating the falsehood into eternity. Now for the example.
Regarding the Axe Murder Incident, the US Army released a statement the following day that one of the reasons for a slow reponse was that the Quick Reaction Force (QRF) was over a mile away from the DMZ at the time of the incident. This is blatantly false to anybody that has ever been there simply due to the fact that at NO TIME is the QRF ever 1 mile away from the DMZ. The farthest the QRF ever gets from the DMZ is approximately 100 meters, which is where the Advance Camp is located, anf this is usually just for a very short period of time. On this particular day and time however, the QRF was actually sitting at the entrance to the JSA which is right smack dab in the middle of the DMZ. Here is part of an email I received from the guy who was actually working the checkpoint at the entrance to the JSA that day.
"Thirdly, to answer the question about who recovered the USA soldiers bodies? It was 3rd Platoon. The information below, "In actuallity, the QRF was sitting in their trucks at CP#2, at the entrance to the JSA, at the beginning of the fight with the 3rd Plt. Ldr awaiting authorization from Capt. Bonifas (who was already dead) for authorization to go in, right smack dab in the middle of the DMZ," is a very accurate statement. Moreover, it took the actual JSA CDR, on this day, to order the 3rd Platoon Leader to go in to recover the bodies.
Lastly, Mike, as you may recall, the above quoted statement truly supports what I have been telling you, M.S. (from the History Channel) and others all along. And, that is, that I was assigned to UNC CP#2 at the time of the August 18, 1976, Axe-Murder Incident. When I initially saw the QRF platoon approaching my guardpost, I was somewhat happy that help was coming. Contrary to what was stated by the 3rd Platoon Leader during the December 28, 2004, airing of the documentary, it was he in the lead vehicle (and not an E-7 Platoon SGT) that I vigorously tried to wave into the JSA -- but my request fell on death ears. Instead, the foregoing platoon leader, subsequently, got out of his vehicle, went into my checkpoint, and then called the JSA CDR, whom, upon his arrival to UNC Check Point #2, ordered him to go in. You can imagine how much time had elapsed! I was saddened by his actions.
I specifically remember this aspect of the incident because due to my assignment at CP2, the LTC ordered me to assure that the soldiers congregating around to view CPT Bonifas' body did not, in fact, see him. I did, and I have been living with the residuals of it ever since. That's why I am telling this story, as it actually occurred."
BTW, my name is Bill, and the reference to Mike above is to another buddy of mine who was also included on the email. I also changed a person's name to just their initials in case they don't want their name known. So, since the US Army issued a statement almost immediately that was wrong in so many ways, it is allowed to stand since the other eyewitnesses and/or myself haven't published anything to contradict it, and the lie is allowed to florish forever? And seriously, if I was to issue to press relase staing the actual events that day, would any news organization really run it or care? No, because it would contradict the "official" version as published by the US Army.
I am very careful about my edits, trying to remain as nuetral as possible and report the actual events as accurately as possible, however this came to a head when somebody wanted to change the word "Murder" to "Killing", claiming that the term murder is POV and trying to diminish the actual events of that day Talk:Axe Murder Incident. I do have some references to a web page I have where I placed some of my recollections from Korea, and also from another buddy who was there with me (Mike, from the quoted part above). Both of our references are listed as "eyewitness account...". Also regarding the term "Murder" for the above immediate problem, a Google search for +"Axe Murder" +korea returns 1100 hits, while the same search that replaces "killing" for "murder" only returns 209 hits. Both sets of hits return some entries for other events, but more inaccurate results occured with the word "killing". So, the term "murder" is also more associated with the events of this day than the term "killing" is as well, adding prevalent thought and opinion to the term.
In conclusion, the history of the world would be in sad, sad shape if it was solely relegated to whichever side expended the most effort in promoting their version of events. I do see where this leads to edit wars, etc., but in many cases, I strongly feel that they need to be decided on a case-by-case basis, where a "group" of educated, non-partisan parties, preferably with both an interest and some background on the subject matter, can decide what should be "allowed" or not. Otherwise, as others have said above (and paraphrased by me), all Wikipedia is, is just a regurgitation of what's already available, so where's the added "benefit" of even using Wikipedia, if all you're are going to see is what's already available everywhere else? Has anybody ever tried (and had success) writing Encylopedia Britannica telling them a cited "fact" is incorrect, offering proof, and actually gotten them to change it? I haven't, and I seriously doubt if anybody else has had any luck in that regard either. That's what makes Wikipedia unique and adds value to many of the articles.
Thanks for any and all opinions on this, regardless of whether there's agreement or disagreement from my POV. wbfergus 15:57, 22 August 2007 (UTC)
The article describes a "battle" between Ken Jennings and Brad Rutter, two US game-show contestants. They have appeared on various game shows, sometimes head-to-head and sometimes not. The article attempts to compare their achievements and declare a "winner" of each "round". I believe that editors have invented the structure of this "battle" themselves, although each specific fact in the article is referenced to a source. Is this OR? -- Cinematical 16:27, 22 August 2007 (UTC)
Cogden, could you make sure you have clear consensus for any changes before making them, please? This is a policy page and it has to be stable.
The change you made didn't make sense to me. You say all sources should be handled the same way. You then say "the source should not be used to support or imply any analytic, synthetic, interpretive, explanatory, or evaluative claims, unless those claims are supported by some other reliable source."
That means that a secondary source cannot be used to make any analytic (etc) claim, unless supported by some other source. So you're now saying that multiple sources must be available for each edit, which is a major change.
It's also not true that all sources should be handled the same way. Primary and secondary sources are, as a matter of fact, handled very differently. The policy has to reflect that. SlimVirgin (talk) (contribs) 18:59, 22 August 2007 (UTC)
I'm not certain that changing the wording to say "based on how they are used" for the Primary, Secondary and Tertiary source use is a purely stylistic change. I think it needs to be discussed and a consensus reached before making this change. Dreadstar † 20:03, 22 August 2007 (UTC)
I did a brief review of who defines primary and secondary sources based on use, and I found some definitions that I think are particularly germane to this policy page, because their "use-based" distinction has a stronger connection to the idea of "original research". Here are a few definitions of primary sources:
Here are a few definitions of secondary sources:
Thus, whether something is a primary or secondary source really depends upon what aspect of the work is being cited. If I'm citing some original research (such as a new scientific theory or a new interpretation of history), it's a primary source, even if the original research was derived from evaluating and analyzing other primary sources (such as raw scientific data or diary entries). In other words, all original research begins its life as a primary source. If someone else adds an additional new idea, they too can be cited as a primary source as the originator of that new idea. Secondary sources don't add any new theories or interpretations, they just comment on primary sources.
According to this framework, sources can be both primary and secondary sources. A biography is a secondary source to the extent it describes, critiques, or analyses primary source material in a non-novel way, but a primary source to the extent it introduces novel theories, analyses, or interpretations about the subject's life. This is not the only way out there to distinguish primary and secondary sources, but given the focus here on the meaning and use of "original research", does adopting this type of framework make sense? COGDEN 21:53, 22 August 2007 (UTC)
This is somewhat related to the above thread, but with a slightly different focus. I would like the cited examples of primary and secondary sources to include something about the application of these terms to the scientific literature. To me, primary sources in this context include journal articles reporting on novel research conducted by the authors of the article. Examples of a secondary source, in a scientific/medical context, would be a review article synthesizing available primary sources, a textbook chapter, or a statement from a large/respected organization such as the WHO/CDC/NIH/etc. The basis for this? Editors may claim journal articles as "secondary sources" and selectively cite specific articles, out of context, to advance their point. Editors should not be in the business of deciding which, of the thousands of peer-reviewed journal articles published every week, are the most scientifically significant ones. Instead, such determinations should be made by experts in the field, as indicated in the sort of secondary sources I've described above. This dovetails with WP:WEIGHT. What I'm suggesting would be, in fact, a very minor change (proposed changes are in italics):
I don't want to open another can of worms about SPOV and so forth; these stipulations would exist to clear up an issue of what is primary vs. secondary within the scientific literature, and would not in any way prioritize scientific sources over non-scientific ones. MastCell Talk 22:35, 22 August 2007 (UTC)
Could you repeat your objections for my sake? No one is saying that an important journal article not covered in a review or the mainstream media needs to be removed. However, such articles are generally handled as primary sources de facto (see WP:MEDRS for example). They can certainly be cited, but when it comes to determining WP:WEIGHT, secondary sources like those I listed are essential. The idea that peer-reviewed == secondary source is one I strongly object to, and one that does not mirror current practice on Wikipedia. There are thousands of peer-reviewed articles published every week; calling them all "secondary sources" puts individual Wikipedians in the business of deciding which of these thousands are most relevant. That determination should be made not by editors here, but by reliable secondary sources such as reviews, textbooks, scientific bodies, etc. MastCell Talk 20:49, 23 August 2007 (UTC)
I'm probably violating something here, but I wanted others to readily see this. Why (or how) did the "Policy" page get unprotected? I though it was being protected until everybody could reach concensus and stop the edit wars on it, and it definately appears that there is no concensus yet. Does someone more vested in this care to revert it back to the protected version? wbfergus 17:26, 31 August 2007 (UTC)
I suggest that if anyone wants to continue a particular thread from the archive they should be welcome to bring that particular thread back in to activity with their new comment. We need not activate the entire recent history, nor refuse to discuss anything that is of particular interest to someone here.-- BirgitteSB 19:44, 31 August 2007 (UTC)
This page is now something like 70 kb meaning half of it should be archived - yet we have made no progress! For starts, can someone refactor the discussion in the sections below this one, moving what belongs into the sections above, and anything else off the page (perhaps to anyother policy or guideline talk page?) Slrubenstein | Talk 09:40, 1 September 2007 (UTC)
I believe the sources section should have two purposes:
I believe that when discussing sources:
I can't really figure out where the discussion is going right now, or why. I can't see why we can't just start with "category 1," discuss whether these are appropriate sources, these are inappropriate sources, these are appropriate sources in certain contexts, or the category should be divided into two or more parts, and then move on to "category 2." I'd suggest starting with tertiary sources, where we can probably reach the widest consensus, and then move "backwards." Jacob Haller 22:26, 31 August 2007 (UTC)
This page is now too long and must be archived, and we are no closer to any resolution. Please put all talk that is simply critical of NOR in another page. Nothing is stoping you from creating a page to propose replacing NOR with a different policy. But stop bitching about the policy here It is not the place. Be constructive. Discuss how to make the NOR policy stronger, or discuss on another page an alternative to NOR. You are free to do either one but they belong on separate talk pages.
My comment was meant to address many people which is why I did not use any names. In any event, talk that questions the value of an NOR policy itself can now go here: Wikipedia talk:Proposal to replace No Original Research Slrubenstein | Talk 12:51, 1 September 2007 (UTC)
My impression of the current policy (not based on any factual checking of the edits) is that it grew over time to include the terms and definitions that now appear contentious. Those same issues really don't belong on the policy page though, as they "really" have nothing to do with the NOR policy, their presence seems to have been added for clarity, but have now become a point of much discussion. So, if the policy was "rewritten" so that it only pertained to NOR, and then had very brief explanations of the other "terms" with links to those pages/guidelines, the policy would be easier to understand for other newbies like myself, and any problems people had with "Sources" or anything else, could be discussed and handled there, where they should be, instead of here. NOR has has nothing to do with defining sources or synthesis, though those are areas where OR usually appears. Keep them separate.
What I created in my Sandbox is an attempt (probably poorly done) to show how some minor changes could help improve the policy and minimize future disruptions of the policy, without actually changing the policy. In previous discussions (now archived) it was asked several times "...then how would you propose to reword ...", and people rarely did. I think Vassyana was the only one who took any effort, but that was solely towards the "Sources" problem. My attempt was to create something to minimize future disruptions of the policy, that others can see and edit without an edit war on the policy page (which looks really, really bad). Others who have problems with the current policy are invited to make their proposed changes there as well. Additionally, if anybody else agrees that removing the "contentious" sections from the policy and having those discussed and maintained separately would benefit the overall policy, I would invite those people to help with rewriting those and appropriately linking them. I don't think this can all be done (and agreed upon) very fast, but I see it as a better use of people's time than just constant "discussing" here. After all, more than a week has gone by with the policy locked, and no progress has been made.
I have two variations. One (using primary, secondary and tertiary source info) is at Sample NOR Policy. wbfergus 10:31, 2 September 2007 (UTC)
The other, using Vassyana's sources proposal is at Sample NOR Policy #2. wbfergus 14:27, 3 September 2007 (UTC)
I've been looking, and the discussion about primary/secondary has been going on for over a year in this talk page. One can look at a long effort like this and see either of two things:
(1) a diligent effort to reach consensus on something
(2) a diligent effort to escape consensus on something
Those who see (1) probably feel that to remove the something now as destructive of a long period of good, hard work. Those who see (2) probably feel that to remove the something now ends a discussion that should have ended far earlier.
At one point in the history I see complaints about a secondary discussion page having been created. I'm not one who asserted "stealth" but the discussion at that time sure resembles an objection to stealth. Sadly, I didn't keep track of when that was, but it was less than a year ago (as I have only looked a little more than one year back in the history.) When I see the date 1 September 2006 on primary/secondary discussion (that doesn't look all that different) I can only shake my head in wonder. -- Minasbeede 14:30, 2 September 2007 (UTC)
I think we should just advise that editors examine existing claims, and note claims which the sources given do not support and vice-versa. Ideally we should have ways to mark citations as fv, the source supports some but not all of the indicated claims; better sources required, the source can be interpreted to support the claim; better sources required, the source was not immediately available; I plan to check it later, and the source was not available; can other editors please check it?. We could even say that some kinds of sources require greater scrutiny than other kinds. Jacob Haller 21:00, 8 September 2007 (UTC)
Which policy is available for situations in which is given a source simply ignored by other contributors? And i not talk about strange and unproof things. I talk about guys that simply not rated at all the sources, even if Wiki should not have the purpuse to prune the sources, expecially if those sources simply not are liked from some guys not exacty NPOV. I have even posted links to sites from the info are extrapolated, but those guys are simply rollbacked accusing me to not have post any source at all. These are situations, sorry to say but that's in which no policy can do enough. Simply helps some very POV guys to act pruning. But still i cannot understand how they can say 'unsourced' when i post sources.--
Stefanomencarelli
18:39, 9 September 2007 (UTC)
If so could we add them to the list of examples of primary sources to make this clear ? Rod57 09:43, 12 September 2007 (UTC)
It is impossible to classify scientific articles as either primary or secondary sources. Take this 2004 paper as an example A trypanothione-dependent glyoxalase I with a prokaryotic ancestry in Leishmania major.
In the fist paragraph of the Results section I quote some dissociation constants I measured. If a Wikipedia article cited these results, it would be using the paper as a "primary source" - a source that presents novel, previously-unpublished data.
In the first paragraph of the Discussion on the other hand, I discuss previously-published data on Methylglyoxal synthase in the light of my results and bring these data together and interpret them. Here, if a Wikipedia article cited the paper on the possible problems caused by methylglyoxal synthase, it would be using the paper as a secondary source. Scientific papers are neither primary nor secondary sources, they contain elements of both. Tim Vickers 17:16, 12 September 2007 (UTC)
It would seem from the title of Tim Vicker's article that he is presenting data that is of evolutionary significance (for a single species.) He's saying (I think) the evidence indicates something about the evolution of a particular species, about how it should be classified. Is that conclusion primary or secondary? It is Tim Vicker's OR, published. His conclusion: primary or secondary?
I should point out that if/when others cite Tim Vicker's paper used here as an example they far more likely will be citing it for its conclusion than for its raw data or its review of other work. The conclusion is the part that constitutes the major part of the advancement of knowledge that the article presents. -- Minasbeede 20:05, 12 September 2007 (UTC)
One of the biggest concerns of the referees of a paper is whether or not it is novel, whether or not it is the primary source for a discrete (and novel) idea. If it just rehashes old material the paper ought to be rejected and usually is. The referees don't act to restrain publication to fall within what is already known/believed, they act to require that whatever is published extends what is known/believed. (Review articles and the like excepted.)
The referees may not know whether the author's conclusions are correct. If they see obvious incorrectness they will alert the author in their referee statements and exhort the author to correct errors. The editor may refuse to allow publication until the author does so. If they're not certain they may say that in their comments but recommend publication. (I think there's a story about Physics Review Letters giving priority to all the papers that the referees didn't understand. The referees did not want to impede the spread of new, good ideas and since they didn't understand they couldn't judge "good," so by default they said "publish.")
I agree that probably not all aspects of the issue were considered when the formulation was considered. We're doing some considering now. -- Minasbeede 20:41, 12 September 2007 (UTC)
Many peer reviewed articles on theoretical subjects should be considered primary sources. E.g., the article in which supersymmetry was first proposed, string theory, special relativity, General relativity, inflation theory, conformal symmetry, etc. etc. etc. Count Iblis 20:51, 12 September 2007 (UTC)
I don't want to mess up the section below (leave it clean for better responses) but this is a response to it. I've done a recent personal review myself and have recognized that my reason to be paying attention to WP:NOR had nothing to do with source typing: I'm troubled by the synthesis section (which I'll not discuss now: there are issues enough.) So I came into the source typing discussion more because I saw it and recognized I had thoughts and an opinion than because I had really strong feelings about it. (Or if I have strong feelings I don't have any agenda behind those feelings.) It looks to me a lot like source typing is inherently too cumbersome to save. The bottom line in source typing is "primary bad" or "primary suspicious" or "bogeymen lurk in primary sources" or something of the sort. The problem is that yeah, sometimes the problem is there - but sometimes it isn't. A policy applies to all. There are cases where I wouldn't hesitate to use 17th century sources. That's "use," not "misuse." No matter how the source typing section is written I'll pay very scant attention to it. There are overriding principles that existed before any source typing was done and those seem to me to be fully adequate. While there are 17th century sources I'd willingly use there are sources from this month I'd never even consider. I recognize that the issue is framed primarily as one of source type but there's the undercurrent of "that-primary-stuff-wasn't-checked-out" and "that-primary-stuff-is-from-long-ago-and-may-not-correspond-to-what-modern-scholars-say." The problem is that these are being made a universal (BEWARE ALL PRIMARY SOURCES) and that's not needed. On top of that is the poor choice of words and on top of that is the confusion that still exists over the Tim Vickers sample article, as an example. There's at least three aspects of the article that have been identified: the experimental data, the historical review (if that's what Tim Vickers called it) and the conclusions. If the article persists and is cited it will mostly be the conclusions that are the reason for the citation. Those conclusions are of the bugaboo "primary" source variety. I think the conclusions are not outrageous and almost certainly will survive every test of time (not having read the article I'm not basing this on what it says but on a feeling.) It's primary, it will (probably) remain always citeable. I think it is plain wrong to put all such articles under any sort of cloud. I don't think the cloud is justified. Yeah, yeah, the article was refereed but if it were a chapter in a book it wouldn't be: books, which on the one hand Wikipedia anoints as "good" sources, often are also primary sources and would fall under the source-typing cloud. Or there'd be a clause explaining why books, though primary sources, aren't really supposed to be watched (or, for all I know, the policy might end up saying "books in particular have to be treated with caution.")
Besides which an editor could be quite aware of many sources that agree with each other and select the primary source because it's the best selection for what the editor is doing. Or he could know that the source he chose is the only source available. He's supposed to use verifiable sources. As yet he's not required to exhaustively explain why he chose any particular primary source - but that could be coming, for all I know. Just like primary soures, the editor who chooses to use one is under a cloud.
Besides which it's rather absurd to make any general statement about source types ahead of their use. If 99.9% of some type of source is bad (whatever "bad" means) then probably that source type will be used infrequently. I can't see how anyone can feel that it's correct to a priori cast aspersions on an editors choice of source, including that 0.1% of the time when it's "good."
I can't help tighten the wording: my heart isn't in it. I'd rather chuck the whole thing. I can live with its being a guideline. -- Minasbeede 02:28, 13 September 2007 (UTC)
I propose the creation of a template for new editors that explains to them why their additions were reverted if they are original research. Unless there is one that already exists that I am not aware of? Sefringle Talk 18:15, 16 September 2007 (UTC)
I suggest that an example is agreed on to test in actual practice what we are talking about. Perhaps the article Gillian McKeith would serve. The use of Ben Goldacre's words from his blog versus the use of his words from a newspaper is one issue. The use of sources that refute the "poo lady"'s claims that do not use her name is another issue. As BLP's represent our strictest use standards, and the data and issues are small in number, and there is no ongoing major conflict at that article, perhaps it is a good choice. WAS 4.250 01:08, 17 September 2007 (UTC)
Let me be clear: I think this is a generally commendable policy, but one that can often be used in bad faith by "wikilawyers" or by people who just like to waste other people's time (like if someone says, "There is no citation here for the claim that these two words rhyme", and while that example is hypothetical, it's not exaggerated.)
I think two major things need to be added to the policy page:
- Jmabel | Talk 19:02, 16 September 2007 (UTC)
This is an alternative to Blueboar's proposal. I would have no objection to changing the word "sources" to "materials." it also may be possible to incorporate elements of Blurboar's proposal and mine. Slrubenstein | Talk 20:05, 16 September 2007 (UTC)
First, let me know Blueoar's version makes the same distinctions. Second, I tried to respond to your (and others) concerns in my rewording. Slrubenstein | Talk 20:37, 16 September 2007 (UTC)
I think Vassyana is right which is why in my version I wrote "However, some encyclopedias and other tertiary sources, such as Annual Reviews, have signed articles, and often articles that explicitly promote the author's own views. In this sense, a tertiary source can also be viewed and treated as a secondary source." perhaps Vassyana thinks this needs to be stated more strongly or clearly, or would suggest a specific change to my proposal? If it is in line with the sentence I just quoted, I am pretty confident I will hav no objection. If it diverges from what I wrote, I am certainly open to suggestions and will try to accommodate or dialogue until we agree. If you think the caveate at the end is simply insufficient, offhand I see two suggestions: (1) beaking it off into a new paragraph exclusively for when tertiary sources are encouraged i.e. one paragra[h that is proscriptive, and another that is prescriptive. OR, (2) divide "tertiaty sources" into two kinds, anonymous and signed. Other ideas? Slrubenstein | Talk 12:07, 17 September 2007 (UTC)
I think we are rapidly moving towards consensus. I have no objection to changing "sources" to "materials" in my proposal, which I think would please many other editors here. I have no objection to reworking the section on tertiary materials, in line with Vassyana's concerns, and I have suggested two possible solutions and invited other suggestions.
There has been little criticism of the section on the history of the proposal/its origins that I proposed; but I hope people will at least go over it for style and clarity.
I extended protection another week just so we can focus on these proposals and keep moving towards consensus. Hopefully all current conflicts will be resolved before the week is out. As we move forward any admin can move material into the policy even in its protected state. My intention is that the continued protection give us space to finish the forward movement to consensus on these issues.
By the way, I know Llywrch, Vassyana and others have been working on an essay on what is not original research. I suggest this: that after we achieve agreement on the two proposals (one on originas of the policy, one on materials) we move that essay to this page and consider working on it as a proposed new section of the policy next. Slrubenstein | Talk 12:21, 17 September 2007 (UTC)
OK, here is the Spenny re-write, which paraphrases the thinking, but adds in a couple of points. I've bolded the key phrase which is the root of what all this source jiggery-pokery is about. It is this key concept that we are battling with, but it is not explicitly stated. Spenny 23:59, 17 September 2007 (UTC)
Within source material we will find a mixture of facts and interpretations. Some sources will be basic observations that offer little in the way of analysis; other sources will offer analysis of information and draw conclusions. When adapting material for use in Wikipedia, it is important to only introduce facts and observations that already exist.
A Wikipedia article or section of an article can employ statements of facts only if the material is used (1) only to make descriptive claims, the accuracy of which is easily verifiable by any reasonable, educated person without specialist knowledge, and (2) never to make analytic, synthetic, interpretive, explanatory, or evaluative claims. Contributors drawing on statements of facts should be careful to comply with both conditions.
Typically, statements of fact will tend to be found in materials such as historical documents; personal diaries, census results, video or transcripts of surveillance, public hearings, trials, or interviews; raw tabulations of surveys or questionnaires; written or recorded notes of laboratory and field experiments or observations; and artistic and fictional works (such as poems, scripts, screenplays, novels, motion pictures, videos, and television programs).
Wikipedia articles can include analytic, synthetic, interpretive, explanatory, or evaluative claims (1) only if such claims come from a reliable, verifiable source and (2) the point of view is clearly identified and accurately represented. In general, it is this type of source that is of most use to Wikipedia as an encyclopaedic article is a summary analysis of a topic and using raw facts to build an article will often require too much original work that could be challenged.
Further, it is most likely that works which are a comprehensive analysis of the topic will be most suited to providing a comprehensive demonstration of a particular viewpoint. Cherry picking individual concepts from different sources may suggest editing to support a point of view, and this can be especially so in the case of using passing references. (This is really NPOV, and I haven't worked this up well)
Useful analysis is typically found in academic journals, Government Inquiries, and media sources such as serious documentary programs and a few respected newspapers. {Yes, I know this bit is weak, but its past my bedtime).
Wikipedia strives to be a superb encyclopedia in its own right. Since Wikipedia does not want to be derivative, materials found in encyclopedias or similar reviews are often of limited value for Wikipedia research. Annual Reviews and Encyclopedia Brittanica articles often provide extensive bibliographies that are valuable tools for identifying important materials, and therefore of great use to Wikipedia editors. Nevertheless, these materials do not necessarily have the same content policies as Wikipedia and for this reason should not be viewed as authoritative. However, some encyclopedias and other material, such as Annual Reviews, have signed articles, and often articles that explicitly promote the author's own views which can be viewed as source material in their own right.
Ok... I think we are at the stage where we agree on the basic principals (1) we want to talk about material and not sources (2) the issue we want to adress is the misuse of primary materials - eitors taking those materials out of context and making analytical, interpretive, or conclusionary statements from them. (3) All but the most basic analytical, interpretive or conclusionary statements included in our articles should be backed by reliable sources (defined as secondary material) that contains the same analysis, interpretation or conclusion.
Now to the hard part... drafting language. As a starting point, I am going to actually propose the language I drafted above (in my "what if we just swapped words" thread). We probably need to edit it further (for example: dropping the tertiary material section?). But I think it may get us off to a good start. Anyone disagree? Blueboar 12:33, 16 September 2007 (UTC)
Well, I don't have a problem with the alternative... but I suspect that others may. For one thing, it does not solve the problem that not everyone agrees with our definition of what a "primary source" is. This version certainly defines it in a much clearer way, but it is a definition that is unique to Wikipedia. Personally, I have never had a problem with that... I understand that in Science the term means one thing, in History it means another... while on Wikipedia it means yet something else. When I apply the NOR policy on Wikipedia, I use Wikipedia's definition, just as I would use the Science definition if I were writing a scientific paper. But, based on the comments that keep being raised here... I am not sure if others can make that distinction. Was there a reason you wanted to keep the word "source"? Blueboar 23:38, 16 September 2007 (UTC)
I've had a look at the proposal above and also Vassyana's drafts. Sorry to be a dog with a bone, but I am still struggling to see what the differentiation of sources brings to the party. The list of primary sources still has the issue that as a generalisation it doesn't work. Simple example, TV documentary could well be a sound secondary source. The list does not explain what makes it a source for fact as opposed to a source for opinion and so this simply raises the query - is Pepys' Diary a source of fact or opinion? It also strikes me that the policy statements made aren't really dependent on this source typing. I guess the other observation is that the differentiation on fact and opinion doesn't really hold water with distinguishing with source typing by example, though it is clear that the distinction between fact and opinion is the issue that is being explored by the policy wording (which I think stands up pretty well on its own). Spenny 16:56, 17 September 2007 (UTC)
Two points - and if both of you agree, I would ask you to suggest appropriate changes. First, any list is of illustrative examples. By this I mean two things: first, it is not inclusive; second, it is not exclusive by which I mean these things are not always examples of (primary, secondary, etc.) (sources, materials, etc.) This shouldn't be too surprising: an apple pie could be (1) an example of my mom's cooking (2) an example of things you can do with fruit (3) an example of a desert (4) an example of symbols for America etc. This leads to my second point: whether something is a primary or secondary whatever depends on how it is used, the context. perhaps this gets closer to Ian's attmpt to get at the concept; i am suggesting that the concept involves how something is used and in what context. Does this make sense to you guys? If so, what language could more effectively communicate this? Slrubenstein | Talk 17:57, 17 September 2007 (UTC)
Some observations:
COGDEN 19:57, 18 September 2007 (UTC)
It's easy. Whenever there's a dispute between a more restrictive and a more flexible version of the policy, just find an example of something that would be allowed under one version but not the other, and claim that anyone who supports allowing that is against the policy and should therefore be excluded from the discussion. It's not necessary to actually succeed in excluding anyone; it's enough to put a chill on the making of certain types of assertions. -- Coppertwig 16:43, 17 September 2007 (UTC)
Taking what Ian wrote and tweeking.... mostly adding 1) that facts need to be cited to reliable sources as well as opinions... and 2) stressing that analysis etc should be the same as in the source. I could live with this approach. Blueboar 01:15, 18 September 2007 (UTC)
Within source material we will find a mixture of facts and interpretations. Some sources will be basic observations that offer little in the way of analysis; other sources will offer analysis of information and draw conclusions. When adapting material for use in Wikipedia, it is important to only introduce facts and observations that already exist.
A Wikipedia article or section of an article can employ statements of facts only if the material (1) comes from a reliable, verifiable source, (2) is used only to make descriptive claims and (3) never to make analytic, synthetic, interpretive, explanatory, or evaluative claims. Contributors drawing on statements of facts should be careful to comply with all three conditions.
Typically, statements of fact will tend to be found in materials such as historical documents; personal diaries, census results, video or transcripts of surveillance, public hearings, trials, or interviews; raw tabulations of surveys or questionnaires; written or recorded notes of laboratory and field experiments or observations; and artistic and fictional works (such as poems, scripts, screenplays, novels, motion pictures, videos, and television programs).
Wikipedia articles can include analytic, synthetic, interpretive, explanatory, or evaluative claims (1) only if such claims come from a reliable, verifiable source, (2) the point of view is clearly identified and accurately represented, and (3) the analysis, syntethis, interpretation, explanation or evaluation matches that in the source. In general, it is this type of source that is of most use to Wikipedia as an encyclopaedic article is a summary analysis of a topic and using raw facts to build an article will often require too much original work that could be challenged.
Further, it is most likely that works which are a comprehensive analysis of the topic will be most suited to providing a comprehensive demonstration of a particular viewpoint. Cherry picking individual concepts from different sources may suggest editing to support a point of view, and this can be especially so in the case of using passing references.
Useful analysis is typically found in academic journals, Government Inquiries, and media sources such as serious documentary programs and a few respected newspapers.
Wikipedia strives to be a superb encyclopedia in its own right. Since Wikipedia does not want to be derivative, materials found in encyclopedias or similar reviews are often of limited value for Wikipedia research. Annual Reviews and Encyclopedia Brittanica articles often provide extensive bibliographies that are valuable tools for identifying important materials, and therefore of great use to Wikipedia editors. Nevertheless, these materials do not necessarily have the same content policies as Wikipedia and for this reason should not be viewed as authoritative. However, some encyclopedias and other material, such as Annual Reviews, have signed articles, and often articles that explicitly promote the author's own views which can be viewed as source material in their own right.
I have been following this debate only intermittently, so some of the issues I mention below may have already been discussed. If so please simply point me to the correct section. (for easy labelling I'll continue to use PS, SS and TS terminology in my comments below)
Abecedare 02:47, 18 September 2007 (UTC)
I like this because it is a fresh approach, but I don't think it reflects Wikipedia practice, maybe in part because the language is not precise or comprehensive enough. "Statements of facts" are frequently used to support analytical, synthetic, etc. claims, particularly when the statements of fact are found in the same source as the analytical claims. Usually, an editor will cite the fact, cite the author's interpretation of the fact from the same source. The fact is used by the author to make the analytical claims, not the editor, it's not original research. Such a use of "statements of facts" would barred by this draft.
Also, since this changes the primary/secondary dichotomy to a fact/interpretation dichotomy, we have the additional problem of defining what is a "fact" and what is an "interpretation". This seems a bit too metaphysical for a pillar Wikipedia policy. For example, if Author X says "Joe was drunk", is that a fact, or an interpretation? And does it matter? Both facts and interpretations can be cited by Wikipedia editors, so long as neither the facts nor the interpretations originate with the editor.
As a corollary to the above, how can we really say that analytical sources are "of most use to Wikipedia"? Sources with pure, raw facts are useful too, maybe even more useful, since without the raw facts, what is there to interpret? COGDEN 20:27, 18 September 2007 (UTC)
The more I think about all of this, the more I think that we were approaching this back-asswards. The point of this policy is that we have to answer the question: "Oh yeah? Sez who?" with... "This reliable source over here sez so." If the answer is "I say so" or "the facts say so", or some thing like that, you are venturing into OR. The type of source being cited does not realy matter, the type of statement that it is being cited for does. For a statement of fact, we need to cite a reliable source that states that it is a fact. For a statement of opinion, we need to cite a reliable source that states that opinion. And for a statement of analysis, interpretation, synthesis, conclusion, etc., we need to cite a reliable source that contains the same analysis, interpretation, synthesis, conclusion, etc. Ian's approach is good... it does essentially approach the issue based on what the source is being used for, instead of what the source is. But it is a relatively long winded way of saying it. Can we be more concise? Blueboar 13:47, 18 September 2007 (UTC)
<undent> Good point that it's what the source is being used for rather than what the source is, it ties in with thoughts I've been trying to pull together. Primary material can be raw data, but of course it can also be an original synthesis. Thus something that's a secondary source becomes primary material when it's the subject of the article or of the part of an article. An example which includes no raw research is
Icons of Evolution. This book is a secondary analysis of primary sources, in a synthesis which is notable only as primary material about the ideas it promotes. The book has been thoroughly analysed by various secondary sources which form the basis of the article.
Perhaps another way of looking at the issue is that editors have to be careful to summarise accurately both the primary material providing the facts that are the subject of the article, and secondary analysis, description or summaries of those facts. Particular care has to be taken to avoid inadvertently summarising primary material in a way that presents our own opinion or understanding about the facts, so it's advantageous to use secondary material which shows attributable analysis.
In my view the primary/secondary distinction is useful as a way of assessing material, and while some examples are useful for explanation, we don't want a prescriptive list of what fits in each category. ..
dave souza,
talk
20:21, 18 September 2007 (UTC)
I'm becoming rather concerned with the view, largely espoused by Cogden, but seemingly being repeated by others now, that sources that produce analysis based on primary source/materials/whatever are, in themselves, primary sources for that analysis. That is not how the terms primary/secondary sources are used at all. Presenting new analysis, synthesis, commentary, etc etc etc is exactly what secondary sources do, and they are still secondary sources. Don't forget that these terms originate in history/historiography. SamBC( talk) 20:35, 18 September 2007 (UTC)
OK... on the theory of puting my money where my mouth is... here is a very rough draft of a short, concise... use oriented paragraph:
Citing the Right Source Materials
Within Wikipedia articles we will find statements of fact and statements of interpretation, analysis or conclusion. It is important to cite the right sources to back those different types of statements. Statements of fact should be cited to reliable sources that clearly demonstrate that fact. Statements of interpretation, analysis or conclusion should be cited to reliable sources that contain the same interpretation, analysis or conclusion. Editors should take care not to confuse a statement of fact with one of interpretation analysis or conclusion. To say that "The Constitution of the US guarantees the right to own a sub-machine gun" sounds like a statement of fact, but is in fact a statement of interpretation (the interpretation being that The 2nd Amendment applies to sub-machine guns). Thus we should not cite the Constitution itself for this statement, instead we should cite an article by a constitutional scholar that contains this interpretation.
The example may not be the best... but it was worth a try. In any case, is this more in the right direction ... or am I off base again? :>) Blueboar 20:23, 18 September 2007 (UTC)
For a better example, how about:
(If that's too U.S.-centric, feel free to substitute something else.) I think this has an advantage in indicating both the type of extrapolation that is allowed and the type that is not. -- Jmabel | Talk 23:18, 18 September 2007 (UTC)
Removed for clarity Spenny 17:24, 20 September 2007 (UTC)
Ok... I think this is the most recent version of the proposal, incorporating those comments that seem to have consensus. I would ask that people not make major edits to it until they have been fully discussed and have reached some degree of consensus. If you think something should be added/changed/deleted from this... post it as a new suggested draft with a new date/time heading. The idea is to keep the different versions clear so people can compare them and comment on them. Blueboar 16:05, 20 September 2007 (UTC)
Citing the Right Source Materials
Within Wikipedia articles we will find statements of fact and statements of interpretation, analysis or conclusion. It is important to cite the right sources to back those different types of statements. Statements of fact should be cited to reliable sources that clearly demonstrate that fact. Statements of interpretation, analysis or conclusion should be cited to reliable sources that contain the same interpretation, analysis or conclusion.
Editors cannot include their own interpretations of previously-published facts, unless that interpretation is either 1) an obvious and non-controversial consequence of the facts or 2) can be attributed to a reliable source. Nor can editors expand on an author's interpretations of fact, unless that expanded interpretation is also found in a reliable source.
For example, that Franklin D. Roosevelt survived polio and that he became president of the United States are both verifiable facts that may be cited in a Wikipedia article. The dates of each event are easily verified, and it would be perfectly admissible to say that he "survived polio, and later became president of the United States" without finding a source that specifically uses the word "later", or otherwise explicitly compares the dates. However, any statement about the effect of the disease on his career would require separate citation, since it is not an obvious or non-controversial consequence of these easily verified facts.
I like this... the FDR example is a great help, although I know that there will still be people who wikilawyer this exception to death. Blueboar 23:35, 18 September 2007 (UTC)
I could support this. It clearly illustrates the point. Vassyana 02:08, 19 September 2007 (UTC)
I could support this. It clearly illustrates the point. WAS 4.250 08:26, 19 September 2007 (UTC)
Good. Very good. However, is "editors cannot" the best description? We know that editors can, until someone reverts them. Would "editors cannot," "editors should not," "editors must not," or something else work best? Jacob Haller 04:51, 19 September 2007 (UTC)
See, here's why sourcing is so important, and why examples in policy are dangerous: FDR never had polio. See Franklin D. Roosevelt's paralytic illness. – Quadell ( talk) ( random) 12:07, 19 September 2007 (UTC)
I agree with the draft, with a couple of tweaks: We can call it a "paralytic illness" instead of polio, and there's a minor grammar issue. I think the example is a good one. I know examples are difficult in policy pages, but I think one would be very helpful here. Model examples can work effectively to elucidate policy (I'm thinking about the American Restatements of the Law as a non-Wikipedia example). COGDEN 20:19, 19 September 2007 (UTC)
As discussed above, it seems to me that the draft would work well with a subsection concisely explaining the usefulness of primary/secondary/tertiary sources. I've based this proposal (as shown at #Primary, secondary, and tertiary sources above) on the existing policy with some changes to bring it into line with the draft, and the examples etc. commented out. Slrubenstein's proposal may form a better basis for this subsection, but it seemed to me to be useful to put a proposal up for discussion about the principle. ... dave souza, talk 12:24, 20 September 2007 (UTC).. add link 12:38, 20 September 2007 (UTC)
We are moving fact and furious folks... and I am a bit confused as to which draft we are now discussing... would someone please post a "proposed" version that incorporates the latest thinking? (perhaps we should entitle them with: Proposed draft as of date/time and periodically update?) Blueboar 13:58, 20 September 2007 (UTC)
I'm opposed to any changes to the long-standing formulation of this key policy. The proposed changes open the door to a long and slippery slope, and for that reason alone I'd objection to their addition, much less their specifics. I appreciate the effort here, but once we start making exceptions, it will never end. FeloniousMonk 15:30, 20 September 2007 (UTC)