Wikibooks talk:Original research

What is the aim of this proposal?
Or, perhaps more clearly; what type of material is it supposed to keep out of Wikibooks?

As User:SB Johnny mentioned in Staff lounge, there is a problem with the lack of clarification of the phrase "No Original Research" since the Wikipedia definition would imply that many Wikibooks don't belong here.

I'm just wondering what exactly the aim of it is. We currently exclude "original works of fiction or literature" and "primary research in any field". From this proposal's current text it seems that it is the latter definition that is too strict. It is currently clarified as "proposing theories and solutions, original ideas, defining terms, coining words, et cetera." which is very similar to the opening text of this proposal. Could this text be modified?

I'm going to go out on a limb and throw the idea out there that we limit the "original research" clause to scientific research. Might all non-fiction, non-scientific works that don't belong here be covered by WB:NPOV? --Swift 17:16, 13 October 2006 (UTC)


 * Actually, the reason I'm bringing it up is quite the opposite concern: defending what is permitted in wikibooks. Not everything has to be about "restricting freedoms", so this is more along the lines of a "bill of rights" than a "censor's guide" (except in the hypothetical case where it might be used to tell a censor-oriented person that there are a lot of things here that shouldn't be censored).


 * There is a lot of interesting material on wikibooks which I think is perfectly appropriate, but would fall under wikipedia's definition of "OR". Examples include:
 * Most, if not all of the cookbook recipes (many of these are the results of the authors' culinary experiments)
 * Most, if not all of the "unofficial user's guide" books (the only one I can think of at the moment is AEM EMS Guidebook, which I've done some structural work on and copyediting, but honestly I have no idea what an AEM EMS actually is, but is obviously not "peer reviewed", otherwise the company would probably have it in the official manual).
 * My own happy project, A Wikimanual of Gardening, includes stuff about controlling particular weeds that don't appear in any peer-reviewed source, but I know they work because it's my job (in real life) to control them.
 * Software manuals such as C++ Programming would either have to be mere exegeses of "official" manuals or just plain old copyvios, rather than the "sharing from experience" sorts of books that they are.
 * Just about any book that doesn't have at least 5 footnotes per page.


 * SO, in a nutshell, this is meant to be a positive statement of what's good and acceptable, as opposed to a dictum saying what's not allowed. Does that sort of clear it up? -- SB_Johnny | talk 17:57, 13 October 2006 (UTC)


 * Thanks for the clarification. I, actually, already understood that it is not to weed out books that we have, but to plant them in the definition of the scope of the project. My suggestion was not to introduce any restrictive measures, but to ease the ones that we already have.
 * We do need some restrictions. One of the current ones is that of "original works". Rather than make exceptions for all the things we'd like included, could we just modify the current clause?
 * I like everything on your list of good original works projects and agree that they, at least partially, fall under original research. None of these are, however, scientific. I admit that I didn't give it terribly much thought, but the only kind of research that wouldn't be suited (and so, by my own oppinion) would be scientific. If we modify the current definition to make it less restrictive, we allow for the inclusion of current works that we agree are useful for the project &mdash; including every item on your list. --Swift 20:07, 13 October 2006 (UTC)
 * I totally do not get it that somehow "scientific" OR should be de-privileged. Why consign Scientific OR to the dungeon while privileging, say, cookbooks? Are scientists "evil"??? If working up cirriculum at wikiversity,is one not allowed to create novel textbooks at WB because it would constitute OR??? What about novel aggregations of secondary sources - wouldn't that technically be OR? Isn't really every new creation except for a pure Import, OR????? I feel like I need more coffee...Geofferybard (discuss • contribs) 21:31, 19 May 2011 (UTC)

Rewording
Given the discussion above, perhaps it would be prudent to reword some of this policy to show that some facts that are not referenced, but which are readily verifiable (killing weeds is a good example) should fit under this proposal. Also, there are only certain types of "original research" that we really do remove from the project historically, and we should probably mention that only certain types of them need to be worried about? I'm going to monkey around in the text of this proposal, and try to put some of these changes into effect. --Whiteknight (talk) (projects) 21:41, 23 October 2006 (UTC)
 * Actually, I take that back, I am not going to make any changes right now. However, I would like to list a few points that should probably be addressed in the final text of this proposal:
 * New conclusions that are drawn directly from pre-existing and verifiable data are probably acceptable. Just because nobody ever jumped to a particular conclusion doesn't mean that the conclusion is obvious and justified from the given information.
 * Just because nobody has ever stated a particular fact doesnt mean that it isn't true, that it can't be verified, or that it is "original".
 * Common knowledge isn't original research (but it should be demonstratable).
 * New ways of teaching old subjects do not constitute "original research". For instance, new styles of writing a book, so long as the information in the book are verifiable, do not constitute OR.
 * Research that is performed by a wikibookian outside of wikibooks, and is appropriately published and peer-reviewed externally, can be used here.
 * The last point is a vanity point for me, because I am doing some "original research" as part of my thesis work, and once I get my things published, I would like to be able to share those results here. --Whiteknight (talk) (projects) 21:50, 23 October 2006 (UTC)


 * That's a special case though... successful theses and dissertations would always, I assume, be considered verifiable sources since they have passed the muster of peer review. OTOH, your actual thesis should probably go on wikisource, with a copy of it here for further development. -- SB_Johnny | talk 15:50, 10 December 2006 (UTC)

revolution
It looks like an interesting revolution might be starting here, quite a change of pace for Wikibooks since there has always in the past been a stated strict claim of adherence to NOR and NPOV. Since it might become possible for Wikiversity to include some original research activities, there has recently been some discussion at Wikiversity of different types of original research (here and here).....possibly some of that discussion might also be relevant to Wikibooks. In many active fields of biology it is not possible for a textbook to avoid a form of original research that involves the textbook author being selective about which results to include and describe, which to emphasize, and, generally, what kind of strategic approach to take to a subject. Further, very few biology textbook authors follow a NPOV policy. Successful and influential biology textbooks tend to be iconoclastic and define a particular novel POV that attempts to selectively exclude mistaken views, particularly views that are still well-known and "popular" but not supported by newer and better research results. In my opinion, for any active field of biology, a textbook that is constructed according to NOR and NPOV will be an artificial construct that nobody who truly knows the field would want to write or recommend to a student. It would be great if Wikibooks is finally open to discussion of these kinds of issues. --JWSurf 05:00, 11 December 2006 (UTC)

Edits
I see this proposal kind of slowed down and then sped up again. I edited the lead. View my edits here. I intended mostly to tighten the prose and clarify a few things. --Iamunknown 01:26, 24 December 2006 (UTC)

Reasons for "No Original Research"
I'm not making any dispute with any of the above discussions that have taken place on this page, but I would like to mention two "unofficial" reasons why the original "no original research" policy was created on Wikipedia, and inherited on Wikibooks for many of the same reasons:


 * Becoming a Vanity Press - In spite of some problems in the area of philosophy/theology Wikibooks, this project has not really had too many problems with books becoming or being viewed as a sort of vanity press, as would be suspected from an initial overview of the main goals and aims of this project. Still, this is something that we need to be ever vigilant about keeping a perception of Wikibooks as an easy way to stroke your ego as a published author.  While it goes beyond the scope of this policy, having a strong (and genuine) editorial peer review of the content of each Wikibook is one way to help filter out the garbage.


 * Telling UFO and Paranormal Researchers to find another forum to publish their ideas. While I will say that there are some books that are borderline problems such as Making an Island and Colonising Mars, the no original research policy has been largely responsibile for keeping even these kinds of books under control.  I will note that for the Colonising Mars book in particular, I explained very early to the author/contributors of this book what the original research policies were and why some of their projects (like collaboratively writing a constitution for the government of Mars and social structures like Martian currency and banking systems) needed to be written somewhere else other than Wikibooks.  It amazes me to no end that people are still coming up with Perpetual motion machines and trying to justify them in the face of centuries of hard science to prove they can't exist.  Obviously a "no original research" policy is a more polite way of telling people to take a hike instead of being more blunt and telling them that they are stupid and scientifically ignorant.


 * A Free Web Hosting Service. I can't emphasis this enough.  Particularly because of how open Wikibooks can be for allowing you to add and create content, there are always people who try to use Wikibooks simply as a way to simply throw some web pages up without having to pay for them.  I got into a bit of a mess about a year ago when I tried to remove the Wikimania Presentations from here, citing the no original research policy as a reason to remove the content.  I will note in particular one of the major rationales being used to keep the presentations here on Wikibooks was that they were being cited in scholarly literature and referencing Wikibooks.  I countered that this was a problem precisely because it was original research.  In this particular case, one of the telling signs that it was original research was that the nature of the content would keep you from making substantial changes to the content except for the original author.  The Wikimania organizers openly admitted that the only real reason for using Wikibooks was that they needed a spot to host the content, and acknowledged that had they tried the same thing on Wikipedia it would have been deleted right from the beginning.  While not so much a common occurance on Wikibooks at the moment, many individuals who have had ideas that would likely become a seperate sister project (using Wikijunior and Wikiversity as prime examples) have used Wikibooks as an incubator to host these ideas.  Quite often, in addition to other issues related with being an incubator, original research has also been a frequent theme, or at least a less offensive way to tell people to find alternatives for things that simply don't have to deal with content creation of books at all, much less textbooks.

None of this is suggesting that these points ought to be mentioned specifically within this policy. But at the same time, these are all underlying issues that really are the reasons why the "no original research" provisions have been established in the first place. To pretend that there are higher motives involved is only a way to add dressing to the policy and not addressing these root causes and concerns. --Rob Horning 07:32, 6 March 2007 (UTC)


 * Many of these ideas that you mention here are part of the current text of WB:WIW, although several people have high-hopes that the unstable branch will be accepted soon (even though interest in that project has waned lately). There is a tendency to move away from the core ideas in many of the policy rewrites, not because anybody disagrees with them, but usually because a good succinct explanation beats a chant or a mantra any day. Saying something like "Wikibooks does not host Original Research" is far less helpful then saying "Wikibooks does not host new or original information, or information that is inaccurate or unverifiable".
 * With all of that said, this page exists to help flesh out the details on the unstable WB:WIW branch. I'm generally of the opinion that policy documents should be short and sweet, and the details should be handled by a series of supporting guideline documents. There is a school of people who believe that the number of policy/guideline pages should be minimized, thereby creating a small set of monolithic policy documents that are all self-contained. There is yet a third school of people who believe that there should be almost no official policy whatsoever, except for a "use good judgement" guideline, and an "assume good faith" one as well. My point with this digression is that specific issues like those you mentioned above are easy to integrate into this proposal, while they would likely be difficult to mention explicitly in the text of WB:WIW/Unstable. If you feel that these should be mentioned more explicitly then they are, you should try to add that text to this proposal. --Whiteknight (talk) (projects) 14:00, 6 March 2007 (UTC)
 * It seems that part of the problem is that the precedent set on wikipedia is irrelevant and counterproductive when naturally we tend to apply it to Wikibooks. TSimply proclaiming that Wikibooks does or does not encourage OR, without a rationale for doing so, constrains the project with no compelling reason to do so. There is originality and there is a research component in any action which is short of outright mechanistic plagiarism. .  Geofferybard (discuss • contribs) 06:19, 13 March 2011 (UTC)

Self Verifiable information
The conversation about this proposal has long since stalled. I have some difficulty with the final section of this proposal, particularly what it would mean about mathematics. The nature of a mathematical proof is one that proceeds from known facts and through a series of self-verifiable steps arrives at a new conclusion. Thus this would allow for pretty much any and all mathematics to be published here. Something a little less permissive is needed. Thenub314 (talk) 22:09, 8 July 2010 (UTC)


 * Or maybe clarification is needed? In the case of the math the intent would mean that steps taken to arrive at the new conclusions would need to be included and that anyone reading it could successfully repeat the steps to arrive at similar conclusions or the same results consistently. A person could not claim to have arrived at a new conclusion and simply say they were able to verify it for themselves without explaining how they were able to reach that new conclusion, what was done to reach that new conclusion, why they reached that new conclusion, etc. for other people to verify for themselves. Another way to look at is take experiments. A book could explain how readers can verify the presence of static electromagnetism for themselves by asking readers to do an experiment involving rubbing a balloon against their clothing and than stick the balloon in their hair to see their hair stick to the balloon. Also the first section defines what verifiable means in this proposed policy:

""Verifiability" in these cases means that a reader of Wikibooks must be able to reproduce the information for themselves safely and reliably."


 * Maybe that can be better worded?

"Verifiable in this policy means that readers must be able to safely and reliably confirm facts and conclusions through independent and consistent reproduction."


 * --dark lama  12:06, 9 July 2010 (UTC)


 * Well I see the difficulty between wanting to allow books we like versus excluding original research. I think something more than clarification is needed.  The difficulty of "a new conclusion" is defining what a new conclusion is can be as slippery as specifying what original research is itself.  There is a rather large body of new mathematics dedicated to producing old results by new methods.


 * I think the idea of experiments is maybe a fruitful one to think about. Verifiability should imply that I can check in at least some ternary source (or at a very minimum a peer reviewed primary source) that the experiment is safe and reliable, without conducting experiment.   As with the information I objected to that used to be in Drugs:Fact and Fiction that gave specific doses of drugs to try taking.  We should at least require as the notion of verifiability that it can be verified to be safe before attempting to follow the steps.  After is too late.  This applies equally well to the cookbook as a book about physics or chemistry experiments, as poorly written recipes and experiments can cause serious injury. Thenub314 (talk) 14:52, 9 July 2010 (UTC)


 * I think identifying an experiment as safe to do and realistic for most people to do so is the key point for this proposal. If an experiment is dangerous or unrealistic than it is not an a substitute for a source citation. An example of possibly safe experiment that is not realistic might involve testing the toxicity of mercury in which proper safely precautions are taken into account like using a hazmat suit. The experiment is unrealistic because most people are unlikely to have access to mercury nor a hazmat suit. Recipes are used in the proposal as an example of a safe and realistic experiment that most people can do. Most people are likely to be able to follow any precautionary steps and have access to the needed ingredients which makes recipes generally safe and realistic. Following any poorly written instructions can cause serious problems for anyone. How would one go about verifying that recipes are generally safe to follow? While I requiring at a minimal that the doing an experiment is verified as safe does appeal to me to some degree, I'm not sure that it will always be realistic. In the absence of being able to verify the safeness of doing an experiment the decision could be left up to the community on a case by case bases through RFD, should anyone have concerns. --dark lama  15:52, 9 July 2010 (UTC)


 * Well sure each case will be left up to the community as an RfD. The point I am trying to make is that is not possible to tell if a recipe is safe by reading it alone.  A page should only live up to a verifiable standard if it can be checked against external in other sources that the experiment is safe to do.  In the same way mathematics should only be allow if it can be checked against external sources.


 * Without external sources it is even difficult to tell if some things are computer generated, nonsense made up by person, or real and useful information. (Just check out snarxiv of some very clever computer generated nonsensical physics.) This is why I think external sources must be key to the whole idea. Thenub314 (talk) 16:25, 9 July 2010 (UTC)
 * What are you saying? Are you saying that this proposal shouldn't call safe and realistic experiments "verifiable", or are you saying that experiments shouldn't be allowed if there are no external sources? --dark lama  16:37, 9 July 2010 (UTC)
 * Experiments (and other things) shouldn't be allowed if there are no external sources. Most of the time of course this is just a clean up issue.  But yes, external references should part of our definition of reliable. Thenub314 (talk) 17:04, 9 July 2010 (UTC)
 * I disagree in that case. Part of the motivation for this proposal is to allow limited exceptions when there are no external sources. --dark lama  17:11, 9 July 2010 (UTC)

But in what case are there no external sources? And why should we allow books on these subjects? The objection given about the C++ Programming book given above is really about how closely a reference needs to be followed. Basically we shouldn't require every paragraph and thought to list which source it comes from, but rather we should allow for a bit more lax style of referencing. Basically we should allow books to give a bibliography where facts may be checked without requiring in advance that every fact have a foot note. This would take care of the "5 footnotes per page". I think we can judge things like cook recipes by the same standard. You don't need to follow exactly the recipe as given in a reference, we may allow a bit of leeway, as we would with C++. But there are endless amounts of cookbooks available and we can cite similar recipes in these books. The same goes for gardening, there are a multitude of books about weed control. Presumably we would benefit from require referencing, as the book would contain more effective methods. And if someone somehow comes up with an ultra effective method that on one has ever heard of, that is OR and shouldn't be here. Thenub314 (talk) 11:07, 10 July 2010 (UTC)
 * Advice shared by people with personal experience may have no reliable external sources/references. Methods can be heard of without there being any reliable external sources/references that mention it. Collective experience and knowledge can move faster than journals might keep up with. That doesn't mean the advice and methods are wrong or even original. Whether its call limited exceptions, relaxed requirements, or giving leeway, doesn't really mater, but as pointed out before without some clear differences with Wikipedia's definition and requirements, a lot of books that Wikibooks allows would need to go. Writing software manuals often involves an original way of teaching how to use software. Really any book that could be written on Wikibooks involves an original way of explaining and teaching something, otherwise it would involve reusing approaches, ideas, examples, etc. which are likely protected by copyright law. --dark lama  12:09, 10 July 2010 (UTC)


 * I concede that some advise may have no reliable external source, and I do see the need for some level of lenience, I just feel the current guidance goes much too far. I am not advocating we use wikipedia's definitions, but something a bit stronger then the current idea of self verifiability.  I am pretty happy with the sections on common knowledge and some other parts of the current document. (Not so much the intro for the same reason.)


 * I am concerned by the idea that we want to exempt things because "collective experience and knowledge move faster than journals". To me this sounds as if the knowledge has hope of being published in a journal they do catch up.  Knowledge that will eventually published in a journal should clearly fall under the heading of original research, and we should require that people simply wait until such knowledge appears in print before uploading it here.  Because to be honest, if it is outside the field of the other editors who end up looking at it, it would be indistinguishable from complete gibberish.


 * About the last part of your post I can only say that, from my experience publishing dead tree books, I strongly believe that reusing examples, methods of explaining, ideas, approaches etc. are not protected by copyright law. (If this were the case about 99% of mathematics texts would violate such a law.) Provided the source is referenced the extent to which the source is followed exceedingly closely. Thenub314 (talk) 16:08, 12 July 2010 (UTC)
 * IMO this discussion would fare better if there were specific cases to examine. Geofferybard (discuss • contribs) 21:37, 19 May 2011 (UTC)