Wikipedia:Reference desk/Refdesk reform RFC

Background information

 * In a library, a reference desk is a "public service counter where professional librarians provide library users with direction to library materials, advice on library collections and services, and expertise on multiple kinds of information from multiple sources."
 * The Wikipedia reference desk claims to work the same way but in reality it is an open forum where any and all subjects are discussed and even debated, where opinions are sometimes presented as fact, with comments by  whoever happens by and feels like commenting.
 * Users not already familiar with the reference desks are strongly encouraged to spend some time reading the current refdesk threads so that they can see how it currently works before commenting.

Why is this a problem?
The result is that replies at the various reference desks vary widely in how helpful they actually are to the user posing the question. They might get directed to source material that can help them understand the topic, or they might get a bunch of wrong answers and wild guesses that contradict one another. There is also an issue with persons who, apparently amused by these tendencies, persistently post ridiculous questions, as well as a few outright trolls  and serial sockpuppeteers. And lastly, there is a persistent issue wherein ref desk regulars clash over what is and is not a request for medical or legal advice.

In short, it appears the reference desk system suffers from several serious problems in that policies which apply everywhere else on Wikipedia are not observed and much time is spent in prolonged back-and-forth between users who sometimes don't seem to realize that the purpose of the reference desk is to help people find information, not to spark debate or even discussion of the topic.

Potential reforms

 * The first six options were presented when this process was opened on June 3, 2013 at 17:17 UTC. Further proposals may be added as this process is underway, they should be timestamped when added so that it is clear they came later and may not have been seen by early participants.

Option one:Make the ref desk work like an actual reference desk
The reference desk will cease providing direct answers of any kind. Discussions will be kept brief and stricly limited in scope. It will only provide direction to appropriate resources such as pertinent Wikipedia articles the sources used by them, and other reliable sources. The reference desk will be kept in compliance with the principle that Wikipedia is not a forum or a social network. Comments which are general discussion of topics or speculation will be removed. A standardized question process such as is used at real reference desks may be developed to make it possible for almost any Wikipedian to competently assist those posing questions.

Endorsements of option one

 * 1) This would be my preferred option. After seeing how the refdesk regulars are reacting to any and all attempts at reform option two is now my first choice, but I still endorse this idea should that not happen. Beeblebrox (talk) 17:22, 3 June 2013 (UTC)
 * 2) Sounds like a good idea to me. NinjaRobotPirate (talk) 19:27, 4 June 2013 (UTC)
 * 3) Use Wikipedia's epistemological model - NPOV, NOR, RS - at the reference desk. Anthonyhcole (talk · contribs · email)  13:10, 5 June 2013 (UTC)

Discussion of option one
While I appreciate that there are users who put a lot of time and effort into giving good answers to questions at the refdesk, there is also a lot of nonsense in the form of outrageous questions and uninformed speculative answers. The refdesk is supposed to be a place to get help finding information about legitimate questions, not a place to discuss nonsense like how long it would take to walk from the earth to the sun. (not making that up it was a real refdesk thread that got a lot of replies) Beeblebrox (talk) 17:25, 3 June 2013 (UTC)
 * It's a nice idea, and in a less disorganized world I'd support it, but who will be in charge of keeping the New and Improved Ref Desk on track? Will we appoint clerks (if so, out of what ranks will we pull them?)? Will the clerks have any authority, or will they just hope people abide by their judgments? Will regular users be expected to police themselves? What happens when there's disagreement about what's "legitimate" or "appropriate"? I suspect that the only way people want to use the ref desk is how they're using it now - the askers for asking general, unanswerable questions, or for homework help; the answerers to show off things they think they know rather than referring to reliable works, and to occasionally make jokes or pick fights with each other. If that's the case, it's not going to do much good to try to police the board unless we use heavy-duty clerking, and I don't think that's a useful way to expend our limited manpower resources. A fluffernutter is a sandwich! (talk) 17:40, 3 June 2013 (UTC)
 * We'll police each other, just as we do on articles. Anthonyhcole (talk · contribs · email)  13:22, 5 June 2013 (UTC)


 * I really believe it's bad for any of us to be the arbiters of whether a question is "legitimate" or "nonsense". The person asking how long it would take to walk the distance from earth to sun is quite possibly a child who knows that the sun is a long way off, who has been told that it's such-and-such millions of miles - but who has trouble getting his/her head around what all of those zeroes at the end of the number really mean.  Journalists routinely simplify these kinds of things for the general public.  They tell us that "A DVD can hold the equivalent of three libraries of congress" or "The USS Nimitz has a flight deck the size of ten football fields" or "This bacterium is a hundred times smaller than the diameter of a human hair" (I made those numbers up BTW).  Those kinds of thing really do help people to visualize very large or very small numbers.  Sure, they may not help you - as a matter of fact, they don't help me either - because I'm some kind of Asperger-syndrome-ridden geek who really wishes that the stupid journalist would have said 7Gbytes, 1,200m, and 2 microns (or whatever).  But there is absolutely no doubt that some people would have a better feel for just how ungodly far away the sun is if they knew how long it would take to walk there.


 * So, IMHO, it's none of our damned business why the person wants to know this - it is enough to know that they do.


 * We're required to WP:AGF - and in this case, that means the when someone takes the trouble to come to our service and type in the question, we must assume that they really do want an answer. Sure, we get the occasional troll - but WP:AGF says that you don't start off assuming that until you have a substantial body of evidence.  Only then can you say that they probably don't care about the answer and that we shouldn't answer it - and we don't need new rules for that: WP:DISRUPT has us covered. SteveBaker (talk) 20:28, 5 June 2013 (UTC)


 * I agree in principle with everything you've said here (and with your excellent analysis elsewhere on this page of how many "bad" questions the Reference Desks actually get), but I do think we have to be careful here. If we don't want to be continually defending the RD against charges that it's a trollfest that should be abandoned, I think we might want to think about ways of sort of "dialing back" the assumption of good faith a bit.  Assuming Good Faith is a wonderful concept, but if it's taken too far, it can seem (especially to the RD critics) to result in a situation in which we're compelled to respond to every outlandish question, and forbidden from treating them as suspicious.  And sometimes we come pretty close to destroying ourselves -- I'm sort of reminded of the way that an autoimmune response causes an organism's immune mechanisms to turn against itself.  (And that's exactly the way I feel when I find the RD or some other part of Wikipedia powerless to defend itself against a skilled troll.) —Steve Summit (talk) 21:59, 5 June 2013 (UTC)


 * This isn't a matter of assuming good faith or assuming bad faith. Questions can have major flaws and questions can have minor flaws. We should not be concerned with the person asking the question. Rather we should be concerned with the question. Is it presented in a form that can be constructively responded to? If it is a reasonably well formulated question, we can just respond to it. But if we have concerns we should politely present our concerns to the person originally presenting the question. A failure to respond to our expressed concerns about the original question after a reasonable amount of time should be seen as a good reason to abandon the question. If a person was asking a question and checking back to see if their query was responded to, they should also be able to provide the additional input that we may be asking for. Bus stop (talk) 22:17, 5 June 2013 (UTC)

Comment: Bear in mind that we who respond to questions on the ref desks are not paid to do it; so we won't do it unless it is interesting to us. Simply providing links and refs would not be interesting to me, and I suspect to most others who participate. I frequently complain about frivolous questions, but this is not a viable solution. Looie496 (talk) 19:27, 3 June 2013 (UTC)
 * That's sort of what I was trying to get at in my comment, actually, even though you and I are arguing (I think) two different conclusions. The reference desk is only interesting to many of its denizens when it doesn't function as an actual reference desk. Unless/until its users want to use it in a different way than the way they're using it now, attempts to change the way it functions will almost certainly be unsuccessful (either because people will drift away from it, or because they'll resist the changes). A fluffernutter is a sandwich! (talk) 19:33, 3 June 2013 (UTC)

Comment. If I'm reading this aright, it would forbid supplying readers with identifications of half-remembered movies, unfamiliar (to them) music, or of plants, animals, and the like. It's been my experience that (correct) answers to these sorts of questions garner the most effusive thanks from querents, so that they are also among the most satisfying to answer. I don't see how such questions are outside the scope of a ref desk, and I'd therefore oppose this option. Deor (talk) 19:46, 3 June 2013 (UTC)


 * Keep in mind the ref desk began as an offshoot of the help desk. If it's to be restricted to linkable articles, then they might as well close its current setup and just roll it back into the help desk. ←Baseball Bugs What's up, Doc? carrots→ 20:11, 3 June 2013 (UTC)
 * Yes, but then how would you fill your time if you had no one to annoy here? Perhaps you could start helping at the Help desk--- perhaps you could start now, and leave us alone like many people have been asking--86.181.25.138 (talk) 18:22, 30 June 2013 (UTC)


 * I think the vast majority of Refdesk users will oppose this option, but supposing that by some "no consensus" magic it becomes enacted over our objections, I will register a request that should this somehow be enacted, that we have time to write bots and not be hindered by deletion or other mechanisms while we get all the archived content transferred over to Wikiversity, which admittedly has considerable overlap with its purpose. Wnt (talk) 21:48, 3 June 2013 (UTC)


 * Support comment. A real ref desk usually cannot provide direct, correct answers to "what's that movie" or "what's this bird" -- but we CAN! The key difference between us and a traditional ref desk is the number of staff on hand. We are modeled after a traditional desk, but we should also leverage our advantages. Of course refs improve answers, and even when obvious consent is not reached on these questions, they also tend to not spawn spite-filled debate and disruption. SemanticMantis (talk) 23:40, 4 June 2013 (UTC)

Comment: A lot of mathematics questions are better answered by providing a short proof/calculation than by citing a reference, especially since references of this type may be hard to find or will require purchase; this, probably, is even more likely if the question is less advanced, meaning "references only" hampers answering the questions of those less able to do it themselves. This same line of reasoning applies to a lot of computer questions too. I don't think this is a good option.Phoenixia1177 (talk) 04:21, 4 June 2013 (UTC)
 * Yes; math is a special case, as it (in principle) need not rely on any outside authority. Well-presented proofs do show up on the math desk, and they are to be commended, not deprecated! SemanticMantis (talk) 23:40, 4 June 2013 (UTC)

Comment: Sometimes seemingly ridiculous questions are asked honestly and truly expect an honest answer. For example, we recently had an OP asking the most wierd and seemingly ridiculously obscure questions about everything from how 1940's era French railroad signal boxes work to whether it would be possible to bounce a rather specific german hand-grenade off of two walls. It eventually emerged that this person was writing a novel about the French resistance during WWII and what emerged was a long series of very interesting, reference-filled questions that I believe was a shining example of what we can do here. I have personally become an expert on German stick-grenades as a result and have improved several articles about them as a direct result of the research I undertook to answer these questions.

If someone asks how long it would take to walk from the earth to the sun, I won't answer "3,600 years", I'll say that "Walking says that an average pedestrian walks at between 4.5 and 4.75 kph and that Earth says that the orbital distance of the earth around the sun is between 147,000,000 and 152,000,000 kilometers and that dividing one by the other will produce the time in hours as between 30 and 34 million hours - or roughly 3650 years." Someone else will doubtless add the obvious in that there is no surface to walk on and even if there was, without gravity you can't walk, that the intervening space is a pretty decent vacuum, that you'd be cooked by the sun's rays before you got there and that the radiation of deep space would kill you before you got very far out of our atmosphere, that walking away from the earth would get easier and easier as you moved out of the gravity well - and soon after, the sun would pull you with enough force that you'd fall the rest of the way very quickly. Then deeper discussions of the nature of orbits and how that previous point is over-simplistic - and so forth.

The point is that the OP (who was probably a child) may have started off with a simple curiosity - but with any luck we've managed to provoke him/her into thinking more deeply about the issues involved, how to find the information and do the calculation for him/herself and so on. This is an extremely important educational function and our very best work at the ref desk is when we can engage a young mind and push them to move from curiosity to using the encyclopedia to answer questions like this. Refusing to answer them or deleting them would be to miss a hugely valuable educational opportunity for a budding new mind. I'd be happier to get rid of all the other kinds of question if only we could keep this kind!

Where we run into problems is if we do all this and the followup question is "How long would it take to walk from Mars to the Sun?". Now we know that we probably have either a very stupid person or a troll on the line. My followup response would be "Look at the Mars article to find the orbital distance - then divide by the average persons' walking speed, just like we did in your previous question.".

Ideally, the thread would end there - but (regrettably) there will probably be a bunch of OP-sniping from some of the more annoying RefDesk folks. If we got a third question of a similar nature, then I'd advocate simply not answering it on the "Don't Feed The Trolls" principle. SteveBaker (talk) 13:51, 4 June 2013 (UTC)

Comment: I wonder if anyone has actually worked on a Library Reference Desk? I have. My experience there closely parallels the experience we seem to have at the WP Refdesk, in that we get people who know a little of what they're looking for, some who don't know how to ask for what they want, or in a way we can respond to, or sometimes we had people trying to catch us out, oh and then there were the "mystery shoppers" who were paid by the College to check we were doing our jobs and not drinking tea and chatting, not to mention the old faithfuls who turned into friends... You get the drift. Sometimes I think that those here who criticise the way the WP Refdesk works would benefit from a spell on a College/University Library Refdesk, and their opinion of what we should be doing would change. (Not to mention a course in customer service.) We need to be able to retain the latitude to actually *help* people, rather than just become a link factory. --TammyMoet (talk) 19:10, 4 June 2013 (UTC)
 * I also work on a library reference desk, and agree with much of what TammyMoet says. In a "real" reference interaction, we have the benefit of engaging in a process with the questioner, helping them figure out their information needs, discussing strategies already used and strategies to consider, evaluating information found, reformualting searches, &c.  Typically (not always, but often) that doesn't happen here, as the OP will ask their question and then not comment further.  That's supplanted somewhat by the discussion among those of us answering the question.  When the discussions remain civil and at least reasonably on-topic, they tend to illuminate new perspectives that any one of us individually might not have considered.  Which I think is kind of the point of this whole enterprise.  --some jerk on the Internet    (talk)  20:50, 4 June 2013 (UTC)
 * Ditto me. This isn't so far off of reality, particularly in the library where I work (public; primarily elementary aged kids and low income).  They ask as much for home ask simple curiosity, no matter how odd. Mingmingla (talk) 22:48, 6 June 2013 (UTC)

Comment@Anthonyhcole : The ref desk does strive to follow NPOV, NOR and RS - but RS a tough standard. Take the kinda silly example above: "How long would it take to walk the distance from the Earth to the Sun?" - we can (and should, and generally do) link to a Wikipedia article to show where we get a typical human walking speed from and the distance from Earth to Sun rather than pointing to some obscure, hidden-behind-a-paywall astronomical journal full of incomprehensible orbital dynamics math and some book on human gait patterns that costs $120 and which none of our readers will ever read. Generally, we use Wikipedia articles in preference to true reliable sources because we want people to use the desk as a way into the encyclopedia. We regard it as the responsibility of those articles to have the reliable sources. However, we're never in any reasonable world going to be able to come up with the time it would take to walk that distance - so we're really forced to violate WP:SYNTH and synthesize an answer from the various facts we've dug up. This pattern is a very common one - and probably applies to 90% of the answers we give. The Ref Desk does need some leeway from strict adherence to all of those Wikipedia guidelines - although I believe that we strive to follow the spirit of them. I'll admit though that we very often fail - and some of our contributors are really bad at providing references as to where they got their data. SteveBaker (talk) 14:13, 5 June 2013 (UTC)
 * @SteveBaker. Presently, with some noble exceptions, the RD is just a blatherfest of "I think..." Any site can host that, and many, many do. If the responders there worked within our epistemological model it would exalt the RD to truly exceptional value. --Anthonyhcole (talk · contribs · email) 04:23, 6 June 2013 (UTC)
 * I speak from my own viewpoint, and to me using language such as "I think" invites contributions from those with other viewpoints. After all, the pace of knowledge moves on so quickly these days - not to mention proofs available for opposing viewpoints - that the words "I think" are probably the only valid words any of us can use on a Reference Desk! People who come to us often don't want links or references: they want an interpretation of the available research that puts it in language they can relate to. There is a case in point recently on the Science Desk, where someone relatively new asked a question regarding the sun never rising in the north in the northern hemisphere. The links and references provided made little or no sense to the questioner: it was the explanations given by the rest of us that helped the information click. Please don't assume that all the questioners are of a sufficent level of intelligence (or competence in English for that matter) to read and understand the links we give them. Following Anthonyhcole's protocol would prevent us actually being useful. --TammyMoet (talk) 19:13, 6 June 2013 (UTC)

As Phoenixia1177 points out above, this won't work for math and physics, at least not for questions that are above primary Kindergarten level, because to explain the answer to a non-trivial question, you often need to discuss the topic in detail. Count Iblis (talk) 18:25, 5 June 2013 (UTC)

Comment: Consider someone who asks "how do I stop cheese from getting moldy in my refrigerator? Neither Cheese or Mold has a good answer to that question, but I happen to know that if you place a small piece of paper towel soaked in vinegar in a plastic bag with some cheddar cheese, the cheese does not get moldy. I also happen to know that stuffing a large onion deep down a gopher hole and adding water makes it rot and that the gas from this drives them away. Do we really want to exclude answers like these? --Guy Macon (talk) 16:22, 6 June 2013 (UTC)

Option two: Get rid of it altogether
Close the reference desk entirely. Wikipedia is a research tool and all one need do to find information is type a query into the search box. The reference desk is not really part of the encyclopedia and should be done away with. The tea house and the help desk exist to help users find what they need on Wikipedia, the refdesk is a redundant distraction from those resources.

Endorsements of option two

 * 1) I think this is the only option that's actually workable. All proposals to restrict usage of or access to the ref desk to try to jump-start it into being functional, while nice in theory, run into the same problem: who's going to enforce them? The last thing we need is one more area that needs clerking, one more area for people to disagree over how to administer it, one more area for patrollers to disagree over what constitutes "in scope" or "social" or "blacklisted", one more area for people to fight about who gets to count themselves as more important or expert than who else. The reference desk doesn't really function as a reference desk in any recognizable sense of the word; it's not worth attempting to reorganize something that isn't a workable concept to begin with. A fluffernutter is a sandwich! (talk) 17:33, 3 June 2013 (UTC)
 * 2) Agreed. Like so many of our sideshows it ultimately tends to get in the way of genuinely improving the encyclopedia.  I don't think there's any way to ignore the cold hard fact that if all the time spent on the refdesk was instead used expanding, copyediting, and verifying articles, we'd be a far better encyclopedia at this point. Andrew Lenahan -  St ar bli nd  23:29, 3 June 2013 (UTC)
 * 3) I cannot think of a single thing RefDesk does that couldn't be done elsewhere. It is, IMO, superfluous and an eternal timesink. :) · Salvidrim!  ·  &#9993;  23:34, 3 June 2013 (UTC)
 * 4) Support: As much as I hate to put extra load onto the Teahouse and Help Desk (and probably village pumps), this seems like the most sensible option based on the problems as stated. Technical 13 (talk) 00:58, 4 June 2013 (UTC)
 * 5) Strong support - I thought the purpose of Wikipedia was to make an encyclopedia, not a library. If it is kept, only experts on the topic should be answering questions. Reaper Eternal (talk) 12:41, 4 June 2013 (UTC)
 * 6) Support It's a distraction from our goal, contrary to WP:CREEP and WP:NOTFORUM. Readers with questions should be directed by the help desk to the most appropriate article and its talk page. Warden (talk) 10:25, 5 June 2013 (UTC)
 * 7) I am honestly surprised to be adding my endorsement here, but after seeing the flat-out denial that there is aven a problem from so many of the regulars there I have despaired of effecting any real change. I'm appalled to see so many of them with blinders on, unwilling to consider even the possibility that the refdesk has serious issues. I'm afraid I now consider this the most viable option. Beeblebrox (talk) 23:13, 5 June 2013 (UTC)
 * 8) Support. Yeah, kill it, WP:NOTFORUM. I enjoy it very much and it's about the only area of Wikipedia that I still use these days, but I agree that it's not what the project was intended to do. Tevildo (talk) 12:49, 8 June 2013 (UTC)
 * 9) Support Per User:Technical 13. User:Anthonyhcole and User:SteveBaker seem to describe ongoing problems Wikipedia doesn't need. I didn't know we even had a reference desk until now. Chris Troutman  ( talk ) 16:04, 17 June 2013 (UTC)

Discussion of option two

 * The failed premise of several of the previous questioners is that the people who man the ref desk would switch to editing articles if the ref desk was shut down. For me, and I think for almost every other Ref Desk denizen, that is absolutely not the case:
 * Searching the web (and the encyclopedia) to find an answer to a tricky question is like solving a crossword puzzle, it relaxes me and fills in an odd 5 minutes I may have to spare during the day. Editing articles doesn't do that for me.  Without the ref desk, I'd spend those odd moments doing something completely unrelated to Wikipedia.
 * Researching for writing or improving an article is a very different kind of activity. When I do it, it is generally because I have a specific interest in the subject matter and I'm making a long term commitment to work on it.  That's a very different thing from dipping into the ref desk a couple of times a day.
 * It takes a different kind of person to be good at working on articles...many of our Ref Desk denizens don't work in article space at all.
 * The ref desk DOES result in article improvement. It happens all the time.  In the course of writing an answer, it'll become evident that there is a gap in our coverage and a new article may spring up. That's how Draper point got written - when I answered a question about red-hot and white-hot iron.  Then I found that I needed to improve John William Draper - and so forth.  Similar things happened when I needed to look at how a Waterhouse stop works - again, there was a gap in our encyclopedia, so I created a new article, found that the John Waterhouse article needed considerable improvement, did some Wikignoming about Mechanics' Institutes and so forth.  One ref desk question can easily result in the improvement of half a dozen articles.  Obviously that doesn't happen all the time - but it's frequent enough to be worth-while.
 * So my point is that the ref desk is the seed from which many new articles are created. It provokes editors into looking at obscure articles that we'd never consider working on otherwise - and it emphatically DOES NOT detract from focussed editing in article space.
 * SteveBaker (talk) 14:10, 4 June 2013 (UTC)
 * To mirror what Steve has said and add to it; Wikipedia does not only exist for its writers, or even primarily. Wikipedia exists for its readers, and in that vein, the Reference Desk serves a valuable tool in directing readers to get information about topics they are having a hard time locating on their own.  -- Jayron  32  14:16, 4 June 2013 (UTC)


 * With regard to SteveBaker's point 4, I don't write many new articles at all, but I wrote George Henry Horn in response to a ref-desk question about a bug he named, and I wrote Kadleroshilik Pingo in (extremely belated) response to a redoubtable admin's question regarding the world's highest pingo. Deor (talk) 14:20, 4 June 2013 (UTC)
 * Yep, exactly. It's highly unlikely that anyone would wake up one morning and think "Hmmm - I wonder if we have an article about a Waterhouse stop?" - then go and write it.  Those kinds of gaps in our coverage only become evident when someone searches for an answer and doesn't find it.  If we're very lucky then the person doing the searching is a Wikipedia editor and goes off and writes a new article - but the probability of that happening is very small.  If a Wikipedia reader tries to find an answer and doesn't get one, he'll most likely give up.  The reference desk catches a considerable number of such people - and directs editors like Deor and myself to go and write something.  There are at least hundreds - and possibly thousands of articles that would never have been written if a ref desk question had not been asked about the subject. WP:AfC kinda covers that same territory - but having worked there for a while, I can tell you that it's an incredibly ineffective tool for the job. SteveBaker (talk) 15:11, 4 June 2013 (UTC)
 * I'm reminded of a book I read recently - "The Professor and the Madman" - it's about how the Oxford English Dictionary was created (it's a WONDERFUL book BTW - I highly recommend it). It talks at length about the problem of finding words that need to be defined.  Writing their definitions turns out to be the easy part.  Finding the most obscure words in the first place is tough.  So they'd get a huge number of people to read quirky, old or forgotten books and send them lists of words that they encountered that seemed strange or unfamiliar.  We have a similar problem.  How do we know that there is a hole in our coverage?  There was a time when we could compare ourselves to other encyclopedias and say "Huh!  Britannica has an article about such-and-such, but we don't - so we'd better write one." - but we've by far surpassed every other information source on the planet and finding holes like Draper point, Waterhouse stop and George Henry Horn is exceedingly difficult.  Sure, we could do things like ask whether there are any Japanese railway stations that we don't yet have article for - and our Wikiprojects do that very well - but the Ref Desk is unique in that people come to us with actual questions that they want answered.  We then look in the encyclopedia for an answer, and generally, we find one.  But quite often we don't - and that's a golden opportunity to create or expand on an article - or improve the searchability with a redirect or whatever.  We're not only finding holes - but we're finding holes that our readers are actually falling into.  I can't think of a better way to do that. SteveBaker (talk) 15:23, 4 June 2013 (UTC)


 * I've seen a lot of mathematics and physics article talk pages that have people asking questions about the topic, many times they end up getting referred to the reference desk. I think it's better to have a place connected directly to Wikipedia to refer people than it is to tell them they can't ask questions there. I think to a lot of people it's natural to want to direct questions about a topic to their source on that topic, the refdesk is the closest thing we have that can fulfill this. If the point of writing articles is to inform people, it makes sense to have some part of the site that can deal with the questions that arise from that process.
 * As to anyone claiming we are, "not a library", we are also not a print encyclopedia. The point of an encyclopedia is to inform its readers, if other encyclopedias could communicate with their readers, I don't doubt that would want to include a feature that could answer questions those readers might have.
 * Referencing point 1, if there was a stronger dialogue, and greater awareness, between article space and the refdesk, answering questions posed by people visiting articles would help in clarifying difficult parts of articles. I don't think article talk pages currently serve this purpose since it's hard for those that don't understand a topic to articulate their difficulties, and it's hard for writers to discern the specific issue from what is, and can be, said. The process of identifying what isn't clear to someone isn't something that can be easily done without trying to address the misunderstandings and questions of the reader, that's something that talk pages cannot do.
 * I don't think the reference desk seriously injures the credibility of the project. Most of the criticisms I've heard is that anyone can edit, the we have a million articles on arbitrary things (like television episodes or pokemon), that you don't need to even register an account, etc. I've never heard anyone say that they don't trust wiki because someone got a wrong answer on the refdesk; I doubt that anyone seriously misunderstands that asking questions is a whole different thing than reading articles and that mistakes are more likely in the former.
 * Another major criticism I've heard, pertaining to theoretical subjects, is that our articles are obtuse. It can be impossible to get the gist of a science topic that is mentioned in a scifi novel if when you look it up, you get a no-thrills formal treatment and need to follow 30 links to other articles to penetrate it. Given how difficult theoretical subjects can be, this is understandable if we at all intend to cover the subject. However, if our goal is to actually provide a service to readers, lacking a place for someone to just ask about the topic in their context is a sure way to fail at that.
 * Final point, promise. Finding the reference desk has lead me, personally, to hundreds of articles I otherwise would never have read. A lot of answers have links to articles I would never have thought to look up, or was unaware of, finding these has enriched my experience as a reader of Wikipedia. I don't doubt that it's done the same for others; so the refdesk can also be a useful tool to get a wider a audience for articles.Phoenixia1177 (talk) 04:04, 5 June 2013 (UTC)
 * I agree. Specifically:
 * Forwarded questions from article talk pages: I agree. Remember that the rule with article talk pages is that they are reserved for discussion of the article itself, not for discussion of the subject matter of the article.  If some reader wishes to have a point clarified, they aren't strictly allowed to post their question to the talk page.  The best they can do is say something like "I think the article should answer the following question..." and hope that one of the editors decides that this is a good thing to do and updates the article accordingly.  That takes months to years for most articles!  But even if the OP does (incorrectly) post a direct question to the talk page, our article-space contributors are not strictly allowed to answer these kinds of question there at all (although they sometimes do).  Also, many articles are not patrolled anywhere near regularly enough.  I've had "conversations" with article authors where I've asked a question - and gotten a response over a year later - by which time, I've forgotten why I needed to know!
 * Not a library. Yeah - we're here to inform readers.  To spread knowledge.  To educate.  In it's heyday, Encyclopedia Britannica had a mail-in research service almost identical to the ref desk...so yes, Encyclopedias do sometimes have this kind of service.
 * Stronger dialogue with article-space talk: There is an additional point here.  Sometimes a perfectly valid question can only be answered by reading multiple articles...some of which may be at a greater technical level than our OP's ability.  Again, the example of "How long would it take to walk the distance from the earth to the sun?" can't be answered using only the information at Walking or Earth - and to even find the distance from the earth to the sun requires understanding what a Perihelion is.  Very often (in the "Can you identify this plant?" type of questions), the OP has already failed to find the correct article.  This is where the ref desk excels.
 * Injuring credibility: I agree.  Our quality has been measured against other online question-answering pages and we score higher than any of the others.
 * Obtuse articles: This is really an argument for improving our articles...but certainly our readers are often extremely unsophisticated.  Many questions can be answered simply by copy-pasting the question in to Google and taking the first search result.  They need help finding an answer - and they need education in explaining it.  We do that exceedingly well.
 * Leading to other articles: Yes, it not only leads readers to new articles - but also leads editors to gaps in our coverage, missing dabs and redirects, articles that are abandoned and need cleanup...you name it.
 * SteveBaker (talk) 14:34, 5 June 2013 (UTC)
 * The proposal to close down the Ref Desk seems founded on a number of false premises. "The tea house and the help desk exist to help users find what they need on Wikipedia."  No, as far as I know they exist to help editors find how to contribute to Wikipedia.  Further down, among the arguments in support, reference is made to WP:CREEP and WP:NOTFORUM.  I have never looked at WP:CREEP before, but it relates to a steady increase in rules and policies, so not relevant here.  And WP:NOTFORUM says article space should not be allowed to become a discussion forum, explicitly recognizing that talk pages and other space may contain some debate.  The existence of the ref desk may actually help in keeping such material off the article pages.  Sussexonian (talk) 19:09, 10 June 2013 (UTC)

Option three:Limit scope
The type of questions allowed at the reference desk will be severely limited in order to prevent trolling, homework questions, and other types of inappropriate or problematic topics, but the method of replying will remain basically the same. Questions not meeting the criteria (to be determined at a follow up RFC) will be speedily closed.

Endorsements of option three

 * Support We took what I saw as a bad step last year simplifying the guidelines at the top of the page. Guidelines as to questions that can be posted should be clear and strictly enforced. Recent questions which have seemed fine to some have been "what does a cat do when it smells petrol" and "which of my four beers of the day tastes best".  Questions need to be strict requests for references answerable and answered with links to articles or outside sources or of encyclopedic value, such as "please identify this animal/novel/building/mechanism.  Questions inviting debate and speculation need to be swiftly closed and no reopened because an editor finds the subject "interesting". μηδείς (talk) 17:50, 3 June 2013 (UTC)

Discussion of option three
This misidentifies the issue. The problem is not bad questions per se, it is the frequent occurrence of people who flood the desks with one bad question after another. Individual bad questions can be ignored or handled gently -- it's series of bad questions that challenge the system. Looie496 (talk) 19:30, 3 June 2013 (UTC)
 * Except for extremely obvious trolling, a question that looks fishy could be boxed up, rather than deleted, and should be discussed on the talk page rather than edit-warring within the ref desk topic page. Concensus would drive whether to (1) delete; (2) leave boxed up; (3) un-box it. ←Baseball Bugs What's up, Doc? carrots→ 20:20, 3 June 2013 (UTC)
 * Even that is easily handled. When you get exasperated with someone who does this (and I'm reminded of "Planet Colors Guy" who annoyed the heck out of me!) you can simply stop answering their questions.  If people want to go to the trouble to discover that cats are particularly sensitive to some constituents of gasoline - or that there is indeed some science behind the fact that one's appreciation for flavor diminishes the more times it is tasted - then that's OK.  It doesn't consume unreasonable amounts of bandwidth or impinge on the efforts of other editors.  What does need to stop is the off-comments and general sniping at the OP.  This is not a matter of major ref.desk reform.  It should be noted that there is emphatic and widespread criticism of User:Medeis for doing exactly what is endorsed by that user, above.  I can't see this becoming an acceptable way to run the ref.desks going forward. SteveBaker (talk) 20:25, 3 June 2013 (UTC)
 * SteveBaker needs to link to what he is objecting to rather than just asserting it exists. Simply "not answering" a flood of nonsense is not acceptable when the flood still exists and others are happy to wallow in it. μηδείς (talk) 20:53, 3 June 2013 (UTC)
 * If the same OP bombards us with useless questions, that can easily be handled under existing WP:DISRUPT guidelines. We don't need to change the ref desk to make that happen. SteveBaker (talk) 14:17, 4 June 2013 (UTC)


 * If an OP asks a question that can't be reasonably answered without more info, then the OP needs to be told so. That doesn't mean editors should say, "You're an idiot" to the OP. But to operate like a "real" ref desk, the OP has to give some. ←Baseball Bugs What's up, Doc? carrots→ 20:30, 3 June 2013 (UTC)
 * The trouble is (and this has happened several times, recently) is that one editor thinks that there is no reasonable answer to a question, and boxes it up - then someone else comes along and finds a decent answer for it. Lack of knowledge of an answer is not an authoritative way to decide to close a question.  So the simplest and cleanest thing to do with an unanswerable question is to not answer it.  Actually, what I prefer to do is to wait a day or so to see if a decent answer emerges - and if not, suggest to the OP that they rephrase it or provide more information.  Most often, they don't - but sometimes they do - and a decent answer can subsequently be given. SteveBaker (talk) 20:58, 3 June 2013 (UTC)
 * By far the most problematic trolls on the Wikipedia Refdesk are the people who come in looking to enforce policies. I'll gladly trade any one of those for any ten people who ask questions about their homework, believe me! Wnt (talk) 22:10, 3 June 2013 (UTC)


 * "Decent" answer? As in entertaining or interesting answer?  The purose of the ref desk is not to find interesting answers for requests for opinion or debate.  When we get questions like "which of my four beers of the afternoon will be the best" be dedicated one-purpose accunts who do nothing other than contribute nonsense  AND ' then we get the usual four or six regulars jumping in to make stuff up then the ref desk has a problem that needs outside attention. (Notice said discussion was hatted after the complaints of at least five editors, then reopened because one editor said he an another were not part of that consensus.
 * If I buy four tins of beer on a sunny day and drink them, which will be the nicest?
 * If this is not proof positive that there is really no question closeable as a request for opinion, speculation or debate, what could be? μηδείς (talk) 00:39, 4 June 2013 (UTC)
 * Maybe the "not a forum" rule should be abolished. ←Baseball Bugs What's up, Doc? carrots→ 00:57, 4 June 2013 (UTC)
 * Indeed! Why should Wikipedia have it so that it's against policy to tell someone about some interesting science peripherally related to some question, but perfectly OK to spend months keelhauling editors and going on about how useless all their work is?  (True, we have a correction for that - you can go after the person griping for "incivility" and "battleground behavior".  If I'd written this as a comic scene I'd have thought I went too far!) Wnt (talk) 01:38, 4 June 2013 (UTC)
 * Medeis has a rather idiosyncratic standard as to what questions are, or are not, appropriate for the reference desk. Irregardless of whether or not people who respond are able to provide references to help the particular OP provide answers for their questions, Medeis seems to feel the need to close or at leats vehemently oppose questions they have some personal objection to; though I (for the life of me) cannot figure out their system or standard for deciding what is, or is not, an appropriate question.  The above cited question is a perfect example.  While some idle chit-chat went on, several respondents were able to provide citations and references to reliable sources, and yet Medeis inexplicable maintained their objection to the question.  Evidence that Medeis has an idiosyncratic standard, bordering on inexplicable is clear based on the frequent and widespread objections to many of their closures of questions and attempts to divert, circumvent, or otherwise steer off topic questions that Medeis doesn't agree with or could provide answers for themselves.  If Medeis's standards for what is, and is not, an allowable question were widely supported (or even widely understood, which I posit they are not even that), then one would see at least some support for their actions on a consistent basis.  I've not seen any evidence of even a modicum of support, or even understanding, of the kinds of questions Medeis will "allow" compared to the kinds of questions Medeis feels the need to try to shut down.  -- Jayron  32  05:50, 4 June 2013 (UTC)
 * I followed Medeis' link to the "four beers" question expecting to find what Medeis describes. The question itself relates to happiness research and the economics of diminishing returns, about which much scholarly research has been published, enough to give a clear answer (the first beer) but with some caveats about other things that can affect the enjoyment of beer. Jayron32 did some searching and found useful pointers to relevant WP articles and reliable sources. Other editors denigrated the question and the answers or brought in issues that seem irrelevant to the consumption of four beers in an afternoon. So what I saw was the refdesk working well with an unusual, but definitely not ridiculous, trolling or fatuous question. Why are the questioner's other contributions at all relevant? It's still a valid question, not "nonsense". I've probably drawn the opposite conclusion about the refdesk from what Medeis would want me to. MartinPoulter (talk) 12:58, 4 June 2013 (UTC)
 * That's not at all an accurate description of what happened, Martin. The single purpose user Horatio Snickers asked a blatantly subjective and utterly random question soliciting speculation and opinion.  He was immediately responded o by two editors, not myself, saying his question was inappropriate.  He then said wat he really meant was "Is there any scientific research into..." his subjective reaction to four beers.  That's patent nonsense on the face of it, of course he is not the subject of any research.  And, of course, worst, it means that any invalid request for opinion or speculation could be made valid by prefixing it with "Is there any scientific research on..."  This would meanthere are no criteria whatsoever.  Is that what is supported here? μηδείς (talk) 18:25, 4 June 2013 (UTC)
 * I don't think there is merit in demanding that people write their questions in some particular way to pass our guidelines (reminds me of the annoying stupidity in the Jeopardy! quiz show where you have to answer in the form of a question). It annoys me that when someone asks "I keep sneezing - do I have a cold?" - it's clearly medical advice and we can't answer - but when someone asks "Is sneezing a symptom of having a cold?" then we can.  I don't know how we fix that one - but it's irrelevant to me whether he prefixed his question with "Is there any scientific research on..." or not.  In either case, we should respond with research-backed fact if it's at all possible.  If you think it's a junk question - then don't reply.  No harm done.  SteveBaker (talk) 19:19, 4 June 2013 (UTC)
 * I can read the question for myself and it is not "utterly random" nor "soliciting speculation and opinion". So you're definitely not giving an accurate description of what happened. Are you actually denying that there has been scientific research into subjective experiences of taste, pleasure or satisfaction? You seem to take the question as nonsense because it's in some sense about the questioner: would you have the same response to "If I jump out of a plane in mid-air, will I fall down or up?" This whole episode reflects well on Jayron, who actually gave useful information in response to the question, and not so well on contributors who attacked the questioner, or who want to mark this sort of question as somehow beyond the pale. MartinPoulter (talk) 13:00, 5 June 2013 (UTC)
 * EXCUSE ME Firstly I would like to thank Medeis for showing their usual etiquette in not notifying me that I am being discussed in this forum. I find their behaviour towards me thoroughly unreasonable, but this is not really a part of this discussion. However, I would object to the fact I am described here as a "single purpose account", let alone that my questions are somehow making trouble. I have been trying to become more involved in Wikipedia and have met some users who have reacted with utter hostility, which I would suggest goes against any of the aims of this project. Personally, the reference desk provided a good introduction to this encyclopedia and encouraged me to get more involved in editing it, despite the hostile reaction of some editors to what I (and given the discussion above, several other) editors would consider reasonable questions. I personally would ask for the reference desk to remain, but this is not really my purpose here - I came here in response to this message, as I am increasingly becoming more interested and keen to contribute to Wikipedia, and would be interested to hear further about the functioning of the reference desk, only to find that I am subject to a character assassination where none of those involved considered it may be courteous to drop me a note informing me of this situation. The fact that Medeis chose to link to my contributions with an external link (and thus further hinder me from learning that I was being discussed) is typical of their behaviour that I have experienced since I first came to this project. Horatio Snickers (talk) 18:00, 3 July 2013 (UTC)
 * "What does a cat do when it smells petrol?" sounds on the same level as "Can I train a rat to sniff out bombs?", and yet we have Gambian pouched rat. The beer question was definitely lighthearted in the way it was asked (and it's intent probably), however context and circumstance do determine how refreshing something is, so it's not unreasonable to suppose there is an answer, or at least information related to the question. I don't think we should remove questions for seeming silly since that's relative, instead we should only answer those questions that we can give a detailed and informed answer to; if the only responses you get are none or serious ones, the impetus to be disruptive diminishes.Phoenixia1177 (talk) 04:12, 4 June 2013 (UTC)
 * That was a pretty dumb question - but the problem here is that one editor may look at such a question and think "That's pretty dumb" and toss it out, when another may happen to know of some interesting research and be able to come up with a really good answer - referenced and detailed. This happens all the time.  I'm constantly amazed (and educated) by some of the answers we can come up with to seemingly unanswerable questions.
 * The best response to a question that you think is dumb or unanswerable is to simply ignore it. Don't answer it.  If nobody can answer it, then it scrolls off the top of the page after a day or two.  If the question was from a troll or an idiot - then they are not entertained and they'll soon stop pestering us.  If somebody else can answer it - then why not let them?  The concern is not the dumb questions - it's the poor responses to them.  We need to encourage respondants to ignore things they can't answer rather than jumping in with whatever comes to mind - or to make some flippant or insulting response.  This is something the ref desk has been able to self-police, but I'd endorse a firmer guideline along these lines and some teeth with which to enforce it. SteveBaker (talk) 14:25, 4 June 2013 (UTC)
 * Agree strongly with both SteveBaker and Phoenixia1177, and others who have expressed similar views. The questions are, by and large, not the problem, unless you are offended in principle by the prospect of answering someone in good faith when they are trolling (and if a trollish question generates a relevant and well-sourced answer that contributes to the knowledge base, then who cares what the OP's motives were?  Not me.)  Unanswered questions vanish unmourned in a couple of days, and even a flood of them (what's a flood?  Six?  Sixteen?) will eventually dry up if the answers are either brief and on-topic or nothing at all.  No, the difficulty lies with those who can't or won't provide a well-sourced neutral answer, but who just can't resist saying something, so we get a personal opinion, or a guess, or a joke, or whatever.  Cue a pile-on response, and bingo! Another load of irrelevant rubbish cluttering up the desks and providing fodder for those who would love to see Option 2 instigated because they don't believe the Refdesk's benefits outweigh its drawbacks.  It's not always the same people by any means: newbie responders often go for these questions because they are so much easier to answer than something requiring time and research.  Medeis favours hatting such questions, if not removing them altogether, to prevent problem answers being given, and I have some sympathy with her stance, which is the logical extension of removing unequivocal medical/legal requests, although her definition of what constitutes an unanswerable question has frequently not coincided with that of others, myself included.  But for me, the solution lies perhaps 10% in managing the questions, and 90% in managing the behaviour of those who answer them. (re-edited to add signature)   Ka renjc 18:54, 4 June 2013 (UTC)
 * Agree with SteveBaker, Phoenixia and Karenjc. At the very least we can link to the relevant WP article: if you don't feel like answering a question, then don't. It's not compulsory. Going back to my days in the library, if we didn't take our users seriously but instead turned on them accusing them of all sorts of nasty stuff, we'd be fired. We need to preserve our ability to help people whose questions are inchoate or poorly formed, and maybe this means tolerating some stuff that we find offensive or irrelevant. Sometimes I think we all need to remember that ours is only one point of view in an ocean of human experience, and a little humility wouldn't go amiss. --TammyMoet (talk) 19:16, 4 June 2013 (UTC)
 * Agree with SteveBaker, Phoenixia, Karenjc, and TammyMoet. If we're going to remove all but the questions we deem to be worth of our consideration, what's the point of having the space to ask questions?  --some jerk on the Internet    (talk)  21:02, 4 June 2013 (UTC)
 * Nine out of ten questions are not problematic. Actually enforcing the standards, especially in case of trolls whose sole repeated behavior is to make repeated subjective or provocative questions is not going to shut down the desks.  It might actually improve them. μηδείς (talk) 17:57, 5 June 2013 (UTC)
 * I would agree that nine out of ten questions are utterly uncontroversial. The other 10% are a mixture of very peculiar questions, unanswerable questions, non-questions, questions that violate our guidelines...and trolls.  But I think it's wrong to label all of these as "problems". "Problematic" is a better word..."iffy", "dubious".  Most of them seem to be asked honestly - in light of WP:AGF.  It only seems like we're being drowned in horrible questions.
 * Let's take some actual data. I looked back at the 60 questions asked since June 1st on the Misc and Science desks (which I happen to frequent the most) and see what goes wrong:
 * On the misc desk, I can find only two questions out of 18 that probably shouldn't be there.
 * "Laura Ingall's Wilder Bio" was a "thank you" note - well intentioned I'm sure - but hardly a question.
 * "Page hacked" was a complaint about incorrect data put into an article - evidently due to vandalism, which one of our regulars swiftly corrected - even though it's not our function to do that.
 * Sure, that's fractionally more than 1 in 10 - but neither of them were malicious or nasty or really objectionable in any way. Both were simply people not honestly not knowing where to come to say those things - and the Misc desk does tend to attract people who don't know where to ask.  Deleting or chastising the people who posted them would be WP:AGF and the best we can do (which we did) is to politely point out their errors and move on.  There is really no harm done here - and so long as our regulars are polite (they were) and helpful (that too) - then it's a net win for Wikipedia and our readership.
 * On the science desk, there were 32 questions, and really about four that are "iffy".
 * Two ("Gabentine" and "Malignant lump") were debatably medical advice. The first one, I didn't think fell into that category, the second I did (and you and others complained that I was overreacting).  In the case of inappropriate medical advice, we tend to react first - and debate later because of the legal and ethical risks of not doing so.  This tends to inflate problems and to blow them out of proportion.  But if we can't agree on whether these questions violate our terms of use - then how could we expect our questioners to know?  These are tricky, borderline cases.  Moving the line between acceptable and unacceptable won't change this - any line you draw in the sand suffers from the problem of questions that straddle that line.  Any rule we came up with would have them - and we just have to work through the issues.
 * "Physics heat project topic" appeared to be a request to do someone's homework - but the question was so badly phrased that we couldn't parse it - let alone provide an answer - so it was kinda moot.
 * "Biology" - a badly titled question which asks "What happens when a cat smells petrol?"...certainly a very odd question! Tough to answer in any meaningful way - yet we did come up with a couple of salient points.  What's problematic here is that this user account is used only for asking questions at the ref desks - at a rate of about one every couple of days.  They are almost always very weird, off-the-wall kinds of questions, badly worded, hard to answer - generally uncomfortable.  I'm not sure whether we have a troll here - or merely some strangely obsessed, but genuine, person who has a lot of strange thoughts pop into their heads.  WP:AGF suggests that we should assume the latter - but it's getting harder by the day and we may need to take this person gently to one side and suggest limiting themselves a bit.  We have had to ban this kind of person in the past...but it's only with great heartache because we recognize that the kind of personality that exhibits this kind of weirdly obsessive behavior is likely to be a younger person somewhere on the autism spectrum who doesn't understand when societal norms are being breached.  A gentle nudge in the right direction generally sorts these problems out.  Real, actual trolls are much, much rarer.
 * So my conclusion is that out of 60 questions, only one came close to being something we might want to take action over. These very genuine - but weirdly obsessive - questioners do pop up once in a while, but not at anything like 10% of our questions.  2% maybe.  The rest are borderline or simply weird or confused posters...no new rules will change that.
 * SteveBaker (talk) 19:33, 5 June 2013 (UTC)


 * The heading here is Limit scope. We need to limit the scope of a small number of regulars to 'hat' or 'box' questions or to engage in debate, on the actual page, on the acceptability of the question.  One problem is that most hatting is done by one individual; the person who has given Support to this option above.   A hat/box actually draws attention to the action, inviting the visitor to look under the hat to see what's there  (it would be better in these rare cases to place the text within the border and small type without the content being hidden: is this possible?).
 * It really isn't possible to define the acceptable types of questions in order to 'limit scope'. OK we would probably prefer that the cat/petrol, four beers, My Little Pony or ashkenazi/sephardi etc questions didn't get asked, but then we might also prefer there were not 3 million articles about individual songs, tv episodes, porn stars, pokémon or non-notable bands/software/professors.  At least the refdesk question will be gone in 10 days or so.  Honestly these type of questions do no great harm to the project and a gentle suggestion (as has been given to the My Little Pony questioner) combined with a disinclination by the regulars to answer the question should suffice.  Sussexonian (talk) 21:22, 5 June 2013 (UTC)


 * The "problem" can be stated in more than one way, just as neither the chicken nor the egg comes first. Those responding to problematically formulated questions who fail to rectify the problem are complicit in perpetuating the problem. Intelligent dialogue with the person posing the question should always be expected. Any time a question is unclear, illogical, or in any way problematic, we who are responding should feel it is our duty to engage with the person in a dialogue to set the question up in a form that meets our standards as to how the Reference desks should be used. I am opposed to "limiting the scope" of what the Reference desks are for. That is a "knee-jerk" reaction that doesn't address the underlying and most general problem. Bus stop (talk) 21:42, 5 June 2013 (UTC)


 * I favor "limiting scope" with an instruction to the answerers: if you think a question does not belong, don't answer it. Just move on to the next. We have too many self-appointed moderators already, and having power over what others can or can not write is too addicting to allow. --Guy Macon (talk) 16:32, 6 June 2013 (UTC)


 * We who are participating in this thread are "the regulars". I don't think too many people who have only asked but never "answered" questions are participating in this thread. In a sense we all are "self-appointed moderators", even if only by dint of participation in this thread. I understand that someone can simply not answer a question. But the situation is generally not so stark. There are "questions" asked that are not quite answerable, or that we hesitate to answer for numerous reasons. None of us are exempt from asking poorly thought-out or poorly-articulated questions. It would seem a waste of human resources to not engage in conversation with the person asking such a problematic question in order to further clarify the aim of the question. Much of this page is preoccupied with fixing "the system", by which I mean the mechanics of how the Reference desks function. I am suggesting that we actually work on our own psychology, or the way we relate to people posing questions. I am suggesting that we have to be more interactive, and that we should expect more interaction from people posing questions. Too often a section consists of one half-baked question with no further input from the person posing the question, and below it a whole slew of responses that may or may not be on-target. In a nutshell, I'm saying that we "regulars" are complicit in the problem. I don't think it's a big problem. You could say this is just my "pet peeve" with the Reference desks. Bus stop (talk) 11:33, 7 June 2013 (UTC)

Option four:Status quo
The reference desk is fine the way it is, or at least not so flawed that any radical changes or needed.

Endorsements of option four

 * 1) Support It isn't broken. No need to fix it.  Questions that actually ask for sources get referred to sources.  Questions that ask for opinions get opinions.  Yes, the opinions are often contradictory, but that has a benefit... it informs the reader that there can be more than one viewpoint on an issue, and possibly more than one "correct" answer to their question.  I don't have a problem with this. Blueboar (talk) 18:11, 3 June 2013 (UTC)
 * 2) Support - The reference desk has been shown by independent studies to be more effective than 'real' library reference desks - and vastly superior to other online question-answering services such as ask.com. It's not without issues - but the community has historically done a good job of self-policing the worse excesses.  "Fixing" it will likely result in an exodus of the experts who've gathered there to provide.  It's not broken.  SteveBaker (talk) 18:58, 3 June 2013 (UTC)
 * 3) Support - The ref desks are certainly not perfect, but we musn't let the perfect be the enemy of the good. Looie496 (talk) 19:34, 3 June 2013 (UTC)
 * 4) Support. I frequently find aspects of the ref desks exasperating—particularly responders who seem compelled to chip in their comments on everything, whether or not they have any useful contribution to make—but I haven't been able to come up with any means of solving its problems that wouldn't result in its being even less effective. Deor (talk) 19:56, 3 June 2013 (UTC)
 * 5) Support. It's not seriously broken. The major source of contention are the petit-police who want to enforce their peculiar and usually narrow view of what is appropriate. Giving them more rules to enforce will not make the problem better. Moreover, the reference desk effectively has a secondary function as a safety valve for the rest of the project,  giving us a place where people can be sent to get informed who would otherwise clog up article talk pages. For that secondary purpose, having a broad scope works better. --Stephan Schulz (talk) 20:19, 3 June 2013 (UTC)
 * 6) Support. To me the reference desk is a bit of an anomaly, but a harmless one. The worst thing that the reference desk might be doing consuming some editor-time, but this is a volunteer project anyway and if we shut down the Ref desk that time would probably be spent on WikiAnswers (or similar) rather than on Wikipedia. If it has problems, it's quite possible it's actually containing these from the main body of the encyclopedia (as per Stephan). Lastly, it's probably one of our better user-feedback channels. --LukeSurlt c 20:44, 3 June 2013 (UTC)
 * 7) Support There's some occasional shenanigans, but no more so than the rest of Wikipedia. -- Jayron  32  20:44, 3 June 2013 (UTC)
 * 8) Support I only really have experience of the maths reference desk, but it seems to work quite well. The questions are broad ranging from school level homework to post grad stuff. Most get some sort of answer. Sometimes its used for questions about wikipedia articles and these can lead to improvement in the articles. Yes discussions can sometimes get off the point, but thats what you expect if you put two mathematicians in a room.--Salix (talk): 20:53, 3 June 2013 (UTC)
 * 9) Strong support. The Refdesk is one part of Wikipedia that is still mostly about sharing knowledge, not powergaming and holding play trials and voting each other off the island.  The reason why off topic conversations occur and are permitted is that everything starts with a question that can be about anything, and so it only makes sense to allow other people to ask questions about anything vaguely related during a discussion rather than starting a new one. Wnt (talk) 21:38, 3 June 2013 (UTC)
 * 10) Support I don't really see the problem. The reference desk provides a needed service to our readers. Questions that purely consist of invitations to speculate or debate are usually closed. The existence of the occasional sockpuppet or troll isn't a good reason to shut it down. Editors disagree on the proper application of rules almost everywhere in the project, and it isn't surprising that this sometimes happens on the reference desk. Hut 8.5 21:55, 3 June 2013 (UTC)
 * 11) Strong Support, per all of the above. I believe that the refdesk is working just fine the way it is now; I've been reading it for a long time and have learned so much along the way. Even to the point that I was inspired to start answering questions myself, and it's been a lot of fun. The refdesk is a truly great feature of Wikipedia; one I wish I had known about back when I first joined. --Yellow1996 (talk) 23:54, 3 June 2013 (UTC)
 * 12) Strong support. Ironholds (talk) 01:31, 4 June 2013 (UTC)
 * 13) Support The current nature of allowing debate (and additionally, WP:OR) allows querants to gain a broad, deep answer and the medical and legal disclaimers are adequate.  Additionally wikiversity has a help desk where querants can answer questions.  With these mechanisms in place, I see no urgent need for the reference desk to require reform.Curb Chain (talk) 02:24, 4 June 2013 (UTC)
 * 14) Support I don't think changes in policy will improve the refdesk, if there is improvement needed, then I think it's going to be generated by the community at the refdesk, or not at all. When I go to the opera, I don't scream and cheer when my favorite violinist plays a solo; when I go to a hockey game, I do cheer and scream when my favorite player has the puck. The difference is not because of policies at the opera versus a hockey game, but because of what is implied acceptable and tolerable. As for more strict rule enforcement, given the anonymity of the internet and the ease of changing online identities, I think a certain section of people would see this as an invitation to be even more disruptive.Phoenixia1177 (talk) 04:46, 4 June 2013 (UTC)
 * 15) Support - Honestly, I never thought that the Reference desk was suffering from problems. It's working the way it is now. It was the reference desk that got me bask to editing, and the knowledge I've learned there has allowed me to understand a lot of things better. I prefer it to sites like Yahoo! Answers which are useful sometimes, but other times doesn't give answers as detailed as the reference desk. Probably the only thing that needs fixing is that we should continue to strictly enforce the policy/guideline of not giving any medical/legal/professional advice, although it would be a good idea to inform users who request such advice of websites that serve that purpose. Narutolovehinata5 tccsdnew 12:19, 4 June 2013 (UTC)
 * 16) Strong support; it's useful, it removes these discussions from talk-pages, it works. Yes, there are trolls, unintelligent people, those who don't care about policy, and so on, it's no worse than the rest of WP. And inappropriate discussions are usually closed fairly quickly. -- Ypnypn (talk) 15:25, 4 June 2013 (UTC)
 * 17) Support I am often utterly amazed at the erudition of some of the contributors. The answers frequently help me and I suppose others are helped too. I sometimes think some of the answers should be incorporated in relevant articles but I don't see how any of the other original proposals would help achieve this. Thincat (talk) 16:41, 4 June 2013 (UTC)
 * 18) Support No, it isn't perfect, but I don't think it's broken.  The variety and depth of knowledge of some of the contributors, and their willingness to share it for free, is a fantastic resource and entirely within the spirit of the project. We should definitely use the material we generate to help improve articles even more than we already do (per Thincat). -  Ka renjc 19:14, 4 June 2013 (UTC)
 * 19) Strong support It ain't broke folks. --TammyMoet (talk) 19:18, 4 June 2013 (UTC)
 * 20) Support I really can't see how any of the proposed alternatives would be considered improvements. They range from thoughtful to absurd, but overall, this has become a circus. --BDD (talk) 20:17, 4 June 2013 (UTC)
 * 21) Support Policy and especially police patrols will not make the Refdesk better.  A little collegiality, an honest desire to be helpful, and the exercise of some common sense and good judgement will.   --some jerk on the Internet    (talk)  20:54, 4 June 2013 (UTC)
 * 22) Support I may have time to post a lengthier analysis later, but for now, a simple support here will suffice.  The Reference Desks do have problems, of course, but they're not radically different from the problems which the rest of Wikipedia (or the rest of the world!) has, and I do not believe they require radical solutions at this time.  Moreover, none of the other options on this page really address the real problems or would provide effective relief. —Steve Summit (talk) 13:58, 5 June 2013 (UTC)
 * 23) Support The main thing we need to do is spend less time harassing and arguing, and more time helping OPs. That's already in the rules, so we just need to encourage a better community within our existing framework. SemanticMantis (talk) 16:19, 5 June 2013 (UTC)
 * 24) Support. The ref desk is beneficial, and the problems it does have are usually not severe.  Also, I agree strongly with Steve Baker's comments in the discussion section under options 2.  Dragons flight (talk) 19:05, 5 June 2013 (UTC)
 * 25) Support. I have found the ref desks useful on quite a number of occasions. If sometimes answers develop into an educated and informed discussion, then all the more interesting, provided this is kept within reasonable bounds of remaining on-topic, which normally seems to be the case. Whatever is done, short of destroying all ability for people to post questions, there will always be trolls. As far as I can see, the regulars at the desks do their best to deal with these using experience and common sense, and I'm not sure that any other system would be dramatically better. 86.128.1.46 (talk) 00:21, 6 June 2013 (UTC)
 * 26) Support Please assume good faith, at least. The discussions going on there are within the scope of, and is focused on, answering questions. And they also provide citations and sources for the necessary ones, in most cases. The existence of ref desks is in no way against our goal in constructing an encyclopedia. A user is expected to search for the answer in related Wikipedia articles first and ask on ref desks if the former didn't help. Hence, it also helps in finding the missing information in the encyclopedia, and several articles have been created and improved after the ref desk questions. ···V ani s che nu「m/Talk」 16:16, 6 June 2013 (UTC)
 * 27) Support The only purpose for making radical changes would be to drive away people who like it the current way it is. RJFJR (talk) 16:28, 6 June 2013 (UTC)
 * 28) Support. There are quite a few problems with the way the Ref Desk functions, and there are things we can do to improve it. However, actually making improvements will require a lot of constructive efforts, it cannot be done by outsiders not familiar with the Ref Desk on the basis of the perception that on the Ref Desk we don't strictly adhere to Wiki-Ideology. One has to take a careful look at what works and what doesn't, evaluated on the basis of the quality of answers and the satisfaction of the people who post questions. If I do that myself, I come to the conclusion outlined in option eleven below, but this is not going to get much traction right now. Count Iblis (talk) 00:13, 7 June 2013 (UTC)
 * 29) Support Sure there are idiotic trolling questions, as well as lazy students wanting someone do do their homework, and sure some answers are merely opinions by self-proclaimed experts. Sometimes the contributors engage in bashing of one country or another, or attempt to mount the soapbox. But it is far more common to see sensible questions from someone who lacks the skills to find the answer via Google search, Google book search, Google news archives, search of Wikipedia, or from an actual book. The Ref Desk is the place people are sent when they try to ask questions on the talk pages of articles about the subject of the article. When a question comes to the Ref Desk, it sometimes means that our article on the subject area of the question needs improvement, or even that a new article is needed. This has led me to seek out references not just from the internet but from a research library, and to make improvements to the relevant article, or to create a new article. It is common to see links to Wikipedia articles as well as other online data sources or to offline books. There is a corps of educated and experienced contributors who spend many hours working to answer questions, who have been unjustly demeaned by some of the exaggerated criticisms here. Edison (talk) 03:58, 7 June 2013 (UTC)
 * 30) Support RD is working fine, keep it as it is! --Tito Dutta  (talk • contributions • email) 00:15, 8 June 2013 (UTC)
 * 31) Support. There is nothing wrong with the Reference Desk as it is.  RNealK (talk) 05:53, 8 June 2013 (UTC)
 * 32) Support. Some questions receive problematic answers. I don't think any of the proposed reforms could fix this, except option two. I find that the usefulness of the reference desk outweighs the problematic sides, hence my support. --NorwegianBluetalk 09:15, 8 June 2013 (UTC)
 * 33) Support: I'm not in disagreement that the reference desks sometimes get tremendously off-topic and argumentative, but it's about the only non-mainspace area of Wikipedia that's actually entertaining, informative, useful, curious, and busy with purpose.  It also, I'm convinced, does no harm.  I think a prominent template (as in Option twelve below) could safeguard this a little more but isn't particularly necessary.  Julia\talk  18:56, 8 June 2013 (UTC)
 * 34) Support for reasons similar to those just stated.  The claimed problem is not serious emough to warrant any of the actions proposed on this page.  I also detect a suggestion from the advocates of drastic change/closure that they consider things have got worse.  I don't see that either.  Sussexonian (talk) 19:13, 10 June 2013 (UTC)
 * 35) Support The refdesk is still an important and handy wiki-resource (often better than other similar Q&A sites). Personally I don't see any compelling reason for scrapping or other radical actions. Brandmeistertalk   23:31, 10 June 2013 (UTC)
 * 36) Support as imperfect but valuable for the project (e.g. in concert with proposal #14).  -- Scray (talk) 20:52, 12 June 2013 (UTC)
 * 37) Strong Support When I first came to this RFC, I was for Option 1, but after reading this entire page, I'm strongly supporting keeping things the same. The reference desk certainly has its issues, but as others have pointed out it is a good resource, does provide value to the encyclopedia and our readers, and generally does an alright job at what it purposes to do.  Certainly it has issues, but these can be fixed by a manner of gradual policy and culture changes rather than large scale reform.  I strongly oppose option 2. Zell Faze (talk) 00:09, 22 June 2013 (UTC)
 * 38) Support. It hits the occasional pothole on the road to enlightenment, but keeps chugging along. Objectionable questions can be ignored. Clarityfiend (talk) 00:52, 22 June 2013 (UTC)
 * 39) Support. Amongst the problems I have seen a lot of work here. I think Julia W characterized the current situation well. TheGrappler (talk) 20:03, 24 June 2013 (UTC)
 * 40) Support. I completely understand how such an RFC could be prompted. It's a fact that these desks are (and always have been, though to varying degrees) plagued by edits and exchanges that don't harmonize with all sorts of stuff listed in What Wikipedia is not (not a forum, not a soapbox, not a place for gossip, ..., not even a manual:-), but I think the benefits outweigh whatever costs our inane clutter might generate. While there is noise, there is also a good amount of signal in a majority of cases. I still think the best thing that can be done on an individual level is to give relevant and referenced information while ignoring debates, polemics, bad jokes, etc. Not surprisingly, some of the most active anti-troll vigilantes are the ones who are actually rewarding the trolls and pseudo-trolls, and unintentional trolls ... Support " [...] or at least not so flawed that any radical changes or needed." Ignore the crap and focus on the question; the desks will be fine. ---Sluzzelin  talk  03:18, 26 June 2013 (UTC)
 * 41) Support - Rarely post here but sometimes read. I fail to see a big problem. Just perhaps hat troll discussion s faster. Garion96 (talk) 23:11, 30 June 2013 (UTC)
 * 42) Support - RefDesks are an overall benefit to project as they attract readers and encourage research; inappropriate questions and answers are only a small minority of posts; occassional problems handled effectively by RefDesk regulars. Gandalf61 (talk) 12:19, 3 July 2013 (UTC)

Discussion of option four
Seeing this reaction has pushed me into supporting option two. We can't fix the "cultural problem" at the refdesks if so many of its regulars refuse to even consider the possibility that it needs to be reformed. Sad, really, but I'd rather not have it at all than have it continue in a dysfunctional manner that has become so routine there that nobody even realizes it any more. Beeblebrox (talk) 23:19, 5 June 2013 (UTC)


 * Beeblebrox, can I ask you a couple of favors?
 * Please don't turn this into an us-vs-them thing. It doesn't help the discussion (and I don't see anyone else doing it).
 * Please be careful about what you see as self-evident assertions. Your original proposal contains a certain amount of this, and now your "so many of its regulars refuse to even consider the possibility that it needs to be reformed" is, I'm afraid, worthy of a  tag.  I see plenty of people ("regulars" included) agreeing that the Reference Desks are not perfect.  But the way this RfC was constructed, the only way to say "the Reference Desks could stand to apply their existing guidelines and uphold their original goals better" is to support the option labeled "status quo".  But that doesn't necessarily mean that the supporter believes that there are no problems, or that they're acceptable.
 * Thanks. —Steve Summit (talk) 01:00, 6 June 2013 (UTC)


 * The reason I haven't supported this option (yet) is precisely because I am considering the possibility that it could perhaps do with changes. These changes needn't be phrased in what is allowed and what isn't, we could also think of less regulatory ways of encouraging desired behaviour and discouraging undesired behaviour. I really want to think about things and make some suggestions I haven't seen here yet, but it will take time, uninterrupted focused time, a rare good right now. ---Sluzzelin talk  01:11, 6 June 2013 (UTC)


 * Beeblebrox, most of the suggestions (including option 1) aren't so much reforms as they are ways to kill the current Reference Desk with the possibility of maybe replacing it with something quite different. That's too radical, which is how I ended up in Section 4.  Part of the problem may be that you are expecting the Reference Desk to focus on providing references, when in fact the responses at the desk tend to focus on trying to provide an answer.  Limited references, in the form of wikilinks and/or URLs, are often included, but for the most part responses at the Ref Desk tend to center around providing answers more than they do on steering people towards places for additional research.  Some library information desks also provide a similar answer service, but such services are rarer for libraries than simply trying to provide lists of relevant references.  By contrast, most online services (e.g. Wikianswers, Yahoo Answers, etc.) also tend to focus on attempting to provide an answer in much the same way that the Ref Desk here does.  The fact that the Wikipedia Reference Desk is de facto more of an answer service than a reference service, doesn't inherently make it bad.  Most of the time the service it does provide is decent (sometimes excellent and sometimes very crappy, but okay on average); however, it may mean that one's expectations need to be recalibrated about what to expect.  Dragons flight (talk) 02:10, 6 June 2013 (UTC)
 * The thing is we aren't Yahoo or Ask.com or any of those other sites. We're supposed to be something different, a website where a reference is always provided to back up anything that there may be any doubt of. I don't believe we should have an "ask a self appointed expert" service here. If we can't bring the refdesk up to the same standards of the rest of every other part of the project due to inertia in favor of the current way it does things I'd rather we just didn't do it at all. It's emberassing to me to have something like this hosted here. We have enough problems with reliability without it. I know its a hobby for some users who do virtually nothing else, and that they may stop editing if we got rid of it. That's unfortunate and I would rather retain those users and redirect them to a more worthwhile endeavor here, but if their only interest is in answering random questions, well quite frankly that just isn't what Wikipedia is supposed to be.
 * That being said, I don't wish to make this into an us-vs-them thing, and there are some refdesk regulars who are supporting other positions or making other proposals, but a good number of them have wound up in the "everything's fine" camp.
 * If the refdesk were forked off somewhere where it was not pretending to be part of an encyclopedia I'd be all for it, but I think a free-for-all that doesn't require sourcing to back up an answer is as out of touch with the Wikipedia ethos as the "experts only" proposal below. We don't do that anywhere else. Even the drama boards require diffs as evidence for controverisal statements. So, from the perspective of compliance with the five pillars, the core policies and philosophy of this project, the refdesk is more off base even than ANI. That's not good. Beeblebrox (talk) 02:34, 6 June 2013 (UTC)
 * Thanks for making your position clear. —Steve Summit (talk) 03:13, 6 June 2013 (UTC)

The delineation - in mission OR application - between what the refdesk does vs. teahouse vs. helpdesk vs. village pump (edited to add) isn't always clear to me, and I've been around a very long time. I can't imagine the confusion a new or occasional editor (or user) has about the situation. I can see the original mission value, but there doesn't seem to be a choice for 're-invigorate and re-organize to mission'. EBY (talk) 05:38, 6 June 2013 (UTC)


 * The phrase "...refuse to even consider the possibility that it needs to be reformed" bothers me. Where is there room for the option of not "reforming" by changing the rules but instead by using gentle persuasion to encourage participants to change how they answer? --Guy Macon (talk) 16:38, 6 June 2013 (UTC)


 * Beeblebrox misrepresents the supporters of the Ref desk by saying "so many of its regulars refuse to even consider the possibility that it needs to be reformed." He set up the form of his RFC, where his option 4 says "The reference desk is fine the way it is, or at least not so flawed that any radical changes or needed." To the contrary, many have stated that there are recurring issues to be dealt with, and which have been dealt with by hatting inappropriate threads, or by the disciplinary methods available in Wikipedia to restrict trolling editors, or by just asking someone to strike or revise inappropriate postings. Read over the Ref Desk talk page archives and you will see a self-correcting process, just as the encyclopedia in general has and uses corrective means. Edison (talk) 04:10, 7 June 2013 (UTC)
 * WP:NOTPAPER. With Jefferson, "He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me." Even if the reference desk does not conform to any particular editor's view of what Wikipedia should be, is this uneasiness significant enough to remove a feature that has been around since, roughly, time immemorial, and that, with some exceptions, seems to work well? I fail to see the real damage. We also have about a billion articles without proper references, many of them quite good, but many, at best, well-meaning but misguided  - does than mean we should shut the project down? Of course not. I think in either case the collateral damage far outweighs the possible benefit. --Stephan Schulz (talk) 12:55, 8 June 2013 (UTC)

Option five:Experts only
In order to prevent speculative or contradictory answers to queries, only users who are verifiable experts on a subject will provide answers that contain anything other than pointers to reference material. The volunteer response team team will verify all claims of expertise via OTRS while still preserving anonymity for those who desire it. Any user will still be permitted to provide simple pointers to appropriate Wikipedia articles or other resources.

Endorsements of option five

 * Strong support second to complete removal. If people are going to be engaging in original research or synthesis of sources, at least let's make them be experts who are likely to actually know what they are talking about. Reaper Eternal (talk) 12:44, 4 June 2013 (UTC)

Discussion of option five

 * This is strongly contradictory of standard Wikipedia practices. SteveBaker (talk) 19:00, 3 June 2013 (UTC)
 * I agree with SteveBaker. This is currently the only option I feel I need to denounce. Pseudonymous Rex (talk) 19:20, 3 June 2013 (UTC)
 * Agree with the above. This isn't a viable option; we don't stratify editors into ones whose knowledge matters and those whose knowledge doesn't. That way lies madness. A fluffernutter is a sandwich! (talk) 19:35, 3 June 2013 (UTC)
 * This would kill the ref desk, so it's really equivalent to Option 2, "Kill the ref desk". ←Baseball Bugs What's up, Doc? carrots→ 20:08, 3 June 2013 (UTC)


 * What determines an expert? and do we want a Citzendum style model? ShakespeareFan00 (talk) 23:08, 3 June 2013 (UTC)


 * The problem is this bans all but a self-approved elite (I speak fluent Spanish, but have no certification in it) while allowing single-purpose trolls to continue spamming the desk with questions. I would be extremely surprised to learn that good answerable questions within the guidelines are getting evul responses by regular contributors. You almost never find a problem answer to a non-problem question. μηδείς (talk) 01:45, 4 June 2013 (UTC)
 * I agree with SteveBaker, I think this is the opposite of what Wikipedia is, and is supposed to be, about. If I can edit any article I so chose without even registering an account, why should I need to provide real life certification to answer questions? I think a policy like this will kill the refdesk; and even if I did have real life credentials, I wouldn't participate with any site with such a policy on principle.Phoenixia1177 (talk) 04:28, 4 June 2013 (UTC)
 * What's an expert? Who decides who's an expert? In what fields? How do you prove you're an expert? Who appoints the WP Thought Police? Oh no, this is about the worst of all options. Strongly oppose. --TammyMoet (talk) 19:20, 4 June 2013 (UTC)
 * I'm not defending this idea, but the proposal does in fact explain that the volunteer response team would verify claims via OTRS, the same way they verify that users who claim to be a well known living person are really that person. I would assume an advanced degree or being a published author on a particular subject would be a reasonable claim of expertise. So, just saying, it could be done if the community actually wanted it done. Beeblebrox (talk) 19:21, 5 June 2013 (UTC)


 * Defined authorities with central mechanisms to endorse them do not scale well and are contrary to the spirit of Wikipedia. Option five sounds to me like what Citizendium's Reference Desk would be. —Steve Summit (talk) 22:13, 5 June 2013 (UTC)
 * Pretty much, except these experts would only be considered experts for purposes of responding to refdesk questions. They would not have any special authority over content as they do at Citizendium. The other difference is that people have actually heard of and use Wikipedia. Although I did write this proposal I don't actually support it and expected it to get the reaction it is getting, I just thought it was important to consider a broad range of ideas, even some like this that seem antithetical to the WP ethos. Beeblebrox (talk) 23:27, 5 June 2013 (UTC)


 * Absolutely not. That is not the Wikipedia way.  It would be the thin edge of the wedge, and then you could kiss Wikipedia goodbye.  --   Jack of Oz   [Talk]  06:20, 6 June 2013 (UTC)
 * And there's the question of practicality. Supposing you could get consensus on exactly what type of qualification conveyed sufficient gravitas to give the green light to answer a particular specialist question (good luck with that), and ignoring all the people who had jumped in to answer in the interim and may have already managed to provide useful references without the benefit of an advanced degree in the subject, by the time you had got your self-proclaimed expert rubber-stamped by OTRS the question would have been archived and the OP would be long gone, or quite possibly dead. - Ka renjc 20:52, 6 June 2013 (UTC)

Option six:Blacklist
Users who persistently post incorrect or speculative answers, frivolous questions, overly argumentative posts, or ther problematic content will be placed on a blacklist of users who are topic banned from all refdesk editing for a minimum of three months. After three months they may appeal the ban a maximum of once every two months. Blacklisted users who violate the terms will be subject to sanction in the form of escalating blocks, up to and including indefinite blocks. Individual admins or other users may not unilaterally add users the the list. Users will be added to the blacklist only after a consensus is established on the talk page, said discussion to remain open for minimum of one week.

Discussion of option six

 * This is a reasonable response to problems - but no "reform" is necessary. We already have the means (in terms of Ref Desk guidelines and general Wikipedia guidelines) to deal with abusive/disruptive editors.  Indeed one such person has received an indef ban in just the last few days.  Nothing new is required here. SteveBaker (talk) 19:02, 3 June 2013 (UTC)
 * They could start by banning editors who (1) insist on giving medical advice; and/or (2) attack other editors without making any attempt to answer an OP's question. ←Baseball Bugs What's up, Doc? carrots→ 20:14, 3 June 2013 (UTC)
 * Sure, in cases where someone persistently violates the ref.desk guidelines (eg by giving medical advice) or who is abusive or condescending to the OP's - then we could ban them. But this RFC is about reform.  No reform is needed to do this.  We can (and sometimes do) exactly that under existing guidelines against disruptive editing, vandalism, WP:AGF violations and so forth.  We don't need special changes to the Ref Desk for that...merely keeping to the rules and societal norms that we already have - and which are common throughout Wikipedia - and applying the usual administrative measures where needed  - is plenty enough here. SteveBaker (talk) 21:03, 3 June 2013 (UTC)
 * I would caution that there are questions that are clearly asking medical or legal advice, and there are questions where some editors want to expand the definition of medical and legal advice to cover questions that really are not medical and legal advice. And of course the opposite can happen as well. --Guy Macon (talk) 16:47, 6 June 2013 (UTC)


 * This can already be done on an ad hoc basis, and doesn't need to be explicitly put into policy. -- Jayron  32  01:02, 4 June 2013 (UTC)

If you expand the grounds for banning to "not sticking to the rules", then you'll end like physicsforums where the rules are gamed by the people who are the de-facto moderators. I am banned on that forum for "not sticking to the rules", and I'm pretty sure I and many others will end up violating the "rules" here too. Count Iblis (talk) 18:34, 5 June 2013 (UTC)
 * Who has been indef banned in the last few days? μηδείς (talk) 01:36, 4 June 2013 (UTC)
 * Sorry - I mis-spoke. Wickwack was indefinitely topic-banned (But since he claims to only edit the Ref Desk, the result is essentially the same). SteveBaker (talk) 13:24, 4 June 2013 (UTC)
 * Point of order; the discussion in question has not yet been closed and evaluated by an uninvolved party; such closure is usually required to enact the results of the discussion. You're likely right that the topic ban will be enacted as soon as someone gets around to doing so, but formally, the discussion has not ended and thus the ban has not yet been implemented.  -- Jayron  32  13:27, 4 June 2013 (UTC)
 * I apologize - I'm not familiar with the details of these administrative processes. However, my point here remains. There is ample means to deal with disruptive editing on the Ref Desk - and it is occasionally used.  I'd support a modest increase in the way we control adherence to guidelines - but this RFC is about sweeping reform and not small tweaks to the way we operate.  I don't see a sweeping reform being enacted as a result of this option. SteveBaker (talk) 14:29, 4 June 2013 (UTC)
 * No, no apology necessary. You're entirely correct in the sense that the Ref Desk can, and does, deal with disruptive elements by the same mechanics as the rest of Wikipedia.  We've done so in the past (Light Current, Timothyhere, Bowei Huang, etc.) where such users have been dealt with via Wikipedia's standard procedures.  We'll continue to do so, and I, like you, have no idea why the Ref Desk needs its own explicit rules and systems.  It doesn't; it's a part of Wikipedia and should be dealt with like any other part of Wikipedia in this regard.  -- Jayron  32  14:37, 4 June 2013 (UTC)
 * Well, we do need some explicit rules and systems. The "No Medical Advice" thing, for example, is a rule we live by which is not a general Wikipedia guideline.  Our community-endorsed "Kainaw's Criterion" by which we judge it is also Ref-Desk-unique.  But, yes - we can (and do) call on admins and apply Wikipedia standards where we can.  We also employ standards such as WP:DISRUPT when our own guidelines or community norms are repeatedly breached - even when no specific Wikipedia rule exists. SteveBaker (talk) 15:35, 4 June 2013 (UTC)
 * Actually, not so much. Medical disclaimer has existed in roughly its current form, with some tweaks, since 2003.  Ten years is a pretty long time, and there's no mention now, or in the past, of the Reference desk in that policy.  It is written from an encyclopedia-wide perspective.  The Reference Desk was created in 2004, so the Medical policy has existed at least a year before the Ref Desk did.  -- Jayron  32  15:39, 4 June 2013 (UTC)


 * This idea does have an inherint risk in that it could turn the refdesk into a "walled garden" controlled by the regular contributors there, but it already kind of is that in my opinion. Rules that we already have such as WP:NOTFORUM are completely ignored on the refdesks. Most of Wikipedia is unaware of that as they don't use the refdesks, and so it has gone on for so long that it is considered routine there. Beeblebrox (talk) 23:31, 5 June 2013 (UTC)

Option seven: Absolute strictness on medical and legal questions
Whilst the worst that most ref desk issues could result in is some on-wiki disruption, mishandling of the medical/legal issue could result in serious real-world consequences. As such this proposal is that the medical and legal question ban be tightened, such that anything that is even remotely tied to these areas is shut down quickly. No "I'm just curious" or "this is for a college project". Even if those qualifiers are genuine, the discussion could be read by another person who may base important decisions on the answers. LukeSurlt c 21:05, 3 June 2013 (UTC)

Endorsements of option seven

 * Strengthening a preventative measure against the one potentially major problem the ref desk could cause --LukeSurlt c 21:05, 3 June 2013 (UTC)

Discussion of option seven

 * Why on Earth would you need this? What possible reason is there to ban people from talking about medicine or biology?  After listening to the rants about how unethical we are for answering questions to the best of our ability, or for merely wanting to, I can think of no higher accolade to a person's morality around here than to be regarded as completely, inherently, and unapologetically unethical.  I also procedurally object to the attempt to "settle" this by one out of seven options to an RfC while ignoring the volumes and volumes of past discussion that people here have actually made in regard to these restrictions, the essays about them, etc. Wnt (talk) 21:52, 3 June 2013 (UTC)
 * Actually, YOU are the reason it's needed. It doesn't seem like it rises to the level of a numbered "option", though, but merely an ongoing specific issue. ←Baseball Bugs What's up, Doc? carrots→ 22:32, 3 June 2013 (UTC)
 * You need the policy because I expressed my disagreement? That is scarcely a way to develop consensus policy. Wnt (talk) 22:37, 3 June 2013 (UTC)
 * No, YOU need the policy enforced ON YOU, because YOU keep trying to violate it. ←Baseball Bugs What's up, Doc? carrots→ 22:40, 3 June 2013 (UTC)
 * Last week people were trying to ban you from the Refdesk as one of the "trolls". The truth is, the real "troll" here is the existence of power policy - the pretense that we can have an encyclopedia anyone can edit but still ban people. Wnt (talk) 22:49, 3 June 2013 (UTC)
 * And you would love my being banned, as it would be one less voice confronting your continual attempts to give medical advice in defiance of the rules. ←Baseball Bugs What's up, Doc? carrots→ 22:53, 3 June 2013 (UTC)
 * Then I must have forgot to vote. Now true, if I were pure of heart I should have spoken in your defense, but we are not perfect.  I'm not sure if this process is the price for that - I don't know if their attempts to get you banned are what got Beeblebrox in on this - but it is possible.  Because there is nothing to come of all that nonsense.  We need merely agree to stop caring about policy and power and focus solely on the love of knowledge for its own sake, and "wasps and snakes will not bite us, hawks and tigers will not claw us." Wnt (talk) 23:07, 3 June 2013 (UTC)
 * If I get banned, no one is harmed, including me. If you're allowed to give some of your quack medical advice, and someone dies, that is BIG trouble for both you and Wikipedia. You have NO knowledge about medicine that is of any value here to any user, and that will remain true no matter how much you try to argue the other way and argue for defying the rules. ←Baseball Bugs What's up, Doc? carrots→ 23:35, 3 June 2013 (UTC)
 * Wnt knows that his is the leftmost edge of opinion on this policy, I don't think a long discussion is needed unless he wants to say there's a user whoe policy opinion is even more unstrict. μηδείς (talk) 00:31, 4 June 2013 (UTC)
 * You and bugs know that you are on the rightmost edge of opinion on this policy. Does that also mean that no discussion is needed? --Guy Macon (talk) 16:56, 6 June 2013 (UTC)
 * On the medical questions, we simply need to enforce Kainaw's criterion more firmly — people who don't provide enough information need to be told to come back before any answers are given, people who are clearly asking for advice need to be shut down immediately, and people who provide enough information and aren't requesting advice need to be helped as much as they would be if they were asking on any other topic. No opinion on what to do with legal questions, since as Kainaw notes, legal advice is a different process.  Nyttend (talk) 03:47, 4 June 2013 (UTC)


 * I'd like to be abundantly clear on this point before any more conspiracy theories are posted here: No one user or incident caused me to open this. In fact I don't know anything about the banning discussions referred to above. I have been thinking about opening this rfc for the better part of a year, but the time never seemed right. Some of you refdesk regulars may recall that about a year to year and a half ago I was an active refdesk contributor for a while, mostly on the humanities and miscellanous desks. While I saw a lot of good answers given, I also saw a lot of nonsense. I quit contributing there but I kept looking in on it from time to time, and every time it was exactly the same, some helpful on-point answers, some useless speculation, some arguments over what is and is not medical advice, and some utterly ridiculous questions asked for reasons that have nothing to do with learning. That's it. I have not spoken to anyone in any way about this until I posted it today.


 * So please, let us all stay on point and discuss the refdesk an not each other. WP:RFC/U is thataway if you need it. Beeblebrox (talk) 05:19, 4 June 2013 (UTC)


 * And on that point, getting back to the actual subject: After reading the above proposal I can't agree, but I would offer a middle road between the two extremes. A standard disclaimer could be developed that would be posted to any question in either of these areas. In fact, two disclaimers would be better as we could specialize them. Anyone who posts a general question as opposed to asking for advice for one specific situation gets the disclaimer added directly below their question, between it and the answers. The basic idea would be that the disclaimer point to General disclaimer, which apples everywhere on Wikipedia and states that Wikipedia is not to be considered in any way a form of professional advice. The disclaimer would be make it clear that everything said is to be considered opinions of amateurs . Although I am sure there are some bona fide experts who post there, On the internet nobody knows you're a dog.


 * In this way we can provide generalized statements and opinions while still being clear that it is solely the opinion of whoever happens to comment. Suggested language and formatting of first line of both disclaimers: ] If you have a real  problem you are looking for answers in the wrong place.  We could make it red and have stop signs on it and it could even have obnoxious flashing animation if we want. Beeblebrox (talk) 05:46, 4 June 2013 (UTC)
 * Lo and behold, we already have just such a template. I think if we added this after every single question asking for medical advice nobody could claim we weren't being explicit about the value of the comments:


 * pretty hard to ignore, don't you think? Beeblebrox (talk) 18:05, 5 June 2013 (UTC)
 * The problem is that you're demanding a "bright line" solution where no bright line exists.


 * Any rule you write (and I happen to think that the one we have - as interpreted under Kainaw's criterion - is just fine) will have corner cases. When someone comes to the ref desk and asks for a blatantly obvious diagnosis, we do jump right on it and nuke it to oblivion with minimal if any debate.  The difficulty is that a good 50% of the medical-related questions are right on the fuzzy borderline.  Sure, you can write a new rule with stiffer penalties or bigger, redder banners - but the problem remains that a piece of open format English prose can always be found that is sufficiently fuzzy to straddle your bright line and cause the kinds of debates that we have.  The problem isn't whether we hat the question or delete it or delete the answers or put a big red scare banner there or indef ban the questioner and all those who answered condemning them to eternal enslavement at http://ask.com.  None of those things solves the real problem - which is how to get the community to agree on those fuzzy-boundary questions.


 * The best that can be done (IMHO) is to have a clear policy on how an average editor is supposed to deal with a question that he or she believes is an unacceptable medical/legal question. "Take the following four steps...do this, then do this, then discuss here, then do that or do not depending on consensus.".


 * But what makes that difficult is time. If a genuine violation of the policy has occurred ("I have chest pains, what should I do?"..."Take an antacid and don't worry about it!"...or..."You're obviously having a heart attack, call 911 right now!!")...then there can be actual bodily harm and possible legal consequences.  So we must allow 'first responder' editors to act instantly to remove bad answers and warn bad questioners - without needing consensus to do it.


 * But then, we'll get perfectly acceptable questions removed that shouldn't be removed because some editors are over-enthusiastic about nuking anything even slightly out of their comfort zone.


 * In the end, we clearly need to go to the Talk: page and discuss it - but we know that once a question has scrolled up the list by a dozen or so entries, very few people will continue to respond to it and our OP will have lost interest or been offended and gone away unhappy. So having a full-scale Wikipedia debate with consensus and Support/Oppose !voting and closing discussions and all of that good stuff is far, far too slow and unwieldy.  We need a small group of experienced and trusted "second tier" ref-deskers to be able to deliver a swift opinion, reversing or affirming the actions of that lone "first responder"...ideally within just a few hours of that first response - and without a full-blown consensus debate.  So long as what they decide generally predicts the will of the consensus, we've lost nothing.


 * If those second tier folk don't generally follow the will of the consensus then further debate can continue after that - but it's all water under the bridge by then and the only possible value is to tweak the policy or change the group of "second tier" debaters for next time.


 * SteveBaker (talk) 20:08, 5 June 2013 (UTC)
 * Well then let's skip the idea of removing or hatting and do it like this : After a user posts a question of a medical nature, the first thing that should happen is that that big ugly notice be added to the thread, where it will should between the question and any subsequent replies. There's no need to fight over what is and is not an appropiate question then, we just tell them "you're in the wrong place for medical advice and you shouldn't take these answers as being well informed or correct" and then the query can be answered in the usual manner. If we do this  nobody can ever claim it was not abundantly clear that the answers were not to be trusted or considered professional advice, and they are advised to contact actual medical professionals if they have a real problem. Seems like a win/win for both sides of this issue. Beeblebrox (talk) 23:09, 5 June 2013 (UTC)


 * We should always post the "big ugly notice". I agree with that much. Although I think it should be toned down a little bit. It is too obtrusive. Bus stop (talk) 23:15, 5 June 2013 (UTC)
 * That could be done easily enough. I actually found this because i was going to try and draft a new template, when I typed in the name I was going to give it I found it was already used by this. I had planned to make it fairly obnoxious so nobody could say they had not seen it, but not quite this eye-poppingly obnoxious. Beeblebrox (talk) 23:34, 5 June 2013 (UTC)
 * I would be willing to put up with such a ridiculously oversized WP:DISCLAIMER, despite the lack of need for one, to be able to answer people properly. However, I should emphasize that the medical lobby here is not going to accept that solution.  Medical ethics is solely about profit, and anything that potentially reduces the number of people seeing a doctor, or reduces the health penalty for not seeing a doctor, will never be acceptable to them. Wnt (talk) 03:20, 6 June 2013 (UTC)

Get rid of this stupid "no medical advice policy" asap. We are here to serve the people who ask questions, and when their questions are not answered because it can be argued to be about some medical issue, they don't like that. This policy thus harms our image, it doesn't solve any real problems. Patients who ask questions on internet forums are helped by getting suggestions that they can talk about with their own doctors. The net effect of giving medical advice being positive, our policy thus actually harms patients. Then by the very "do no harm principle" which is often cited by the proponents of the current polcy, we should actually overturn the "no medical advice" policy. Count Iblis (talk) 12:25, 6 June 2013 (UTC)
 * The policy is one of liability. I have no doubt that you or Wnt or whoever could offer good advice, but that's irrelevant.  The real world is such that if we offer bad advice and someone suffers from it, we and the project could be legally liable.  Does it help a patient foolish enough to ask here? No.  The policy is about covering our asses.  It's also one that is taught in real library schools and real libraries.  See here for a real library with the same policy. Whether or not Wnt's Big Medicine is responsible is irrelevant, even if it were true.  We wouldn't be any less liable. Mingmingla (talk) 23:09, 6 June 2013 (UTC)


 * I know the current Ref Desk policy is similar to policies I've encountered at real libraries. I've had the impression that there were specific laws about practicing medicine / law without a license, and that such laws were a significant part of what motivated these policies.  Specifically, that one can get in trouble even if you are right, in addition to the possibility of liability for being wrong.  However, it appears that Quora, Yahoo Answers, and others allow the handling of medical questions.  If that is so, then what gives?  Do they have a different understanding of the liability and legal issues involved?  Can you provide references that help explore the liability considerations in more detail?  At first blush, it certainly makes sense that we should worry about potential liability (and also bad PR) from giving harmful advice, but then I still don't understand why other sites seem willing to do it.  Dragons flight (talk) 00:34, 7 June 2013 (UTC)


 * As we all know, dangerous advice can only happen in comments about the law and medicine. No way can advice about, say, engineering, be harmful. Now, as I was saying, feed your Uranium hexafluoride into your high-speed isotope separation centrifuge... --Guy Macon (talk) 01:13, 7 June 2013 (UTC)
 * I didn't say it was right or logical. Danger doesn't really enter into it, though. In many jurisdictions you can't practice medicine or law without a license and if you do, it's a crime. Misrepresenting yourself as either profession can net you legal charges and possible jail time.  Not all professions have that same restriction, and as a result stupid/dangerous advice can be legally given without liability.  Maybe Wnt's onto something with Big Medicine, but again, it doesn't make us (or those other sites, or physical libraries) less liable.  Mingmingla (talk) 01:21, 7 June 2013 (UTC)
 * To be more clear, even WebMD claims not to offer advice. For legal reasons, I'm guessing.  If they don't, we certainly won't. Mingmingla (talk) 01:26, 7 June 2013 (UTC)
 * And here, from their "Symptom Checker": "This tool does not provide medical advice It is intended for informational purposes only. It is not a substitute for professional medical advice, diagnosis or treatment. Never ignore professional medical advice in seeking treatment because of something you have read on the WebMD Site. If you think you may have a medical emergency, immediately call your doctor or dial 911." Mingmingla (talk) 01:30, 7 June 2013 (UTC)


 * Those WebMD examples are helpful, if no other reason the illustrate the silliness that goes one when it comes to certain kinds of advice. If all that matters is telling people we don't give advice and then giving them "information", then arguably we (or even your local library) could offer quite a bit more "information" than we presently do.  On the other hand, if giving medical advice over the internet is really forbidden then providing tools to check individuals symptoms doesn't seem logical.  Frankly, I don't really understand how WebMD or Quora or even the local library has decided which information they feel is okay to offer and which isn't.  Many of these institutions seem to come to different conclusions and then they all try to add an extra layer of protection by saying "we never give advice", etc.  Dragons flight (talk) 01:51, 7 June 2013 (UTC)
 * To my point: we don't offer it for liablity reasons, an issue that even explict medical sites have to deal with. The best way to avoid the situations that requires such a disclaimer in the first place? Don't offer the service at all. Plus it just stupid and reckless to offer any medical advice to any person sight unseen, no matter how confident you are that they give you their symptoms truthfully and honestly.  You don't know them, you don't know their history.  Hell, you can't even see them.  To suggest we should give advice because it "helps" them is arrogant, dangerous behaviour.  You could be a great healer in real life, but to take a single paragraph some random person of indeterminate gender with any myriad health conditions they might not even know about and use that to try and figure out there problem... I don't know what more I can say.
 * As to the question about what one can offer or not, I might remind what I said above: practicing medicine without a licence is illegal in many jurisdictions. Any information that a library or individual website gives that could be interpreted as offering a suggested course of action or a potential diagnosis could create an unfortunately legal entaglement.  Those legal entanglements determine what information is offered.  It's a crappy world we live in when litigation decides whether or not to help someone, but there it is.  What want and what the world really is like are sadly two different things. Mingmingla (talk) 02:30, 7 June 2013 (UTC)
 * The anti-medicine/biology faction has for years played this game of making all sorts of legal assertions about how it is prohibited for us to answer the questions - then retreating to claims that it is merely "unethical" when these are refuted, which they have been many times before in many discussions just like this one. If it is legal for Wikipedia to have an article about multiple sclerosis (though I'm sure many of you would love to change that) then it is legal for us to tell someone what the symptoms of multiple sclerosis are when they ask.  If it is legal for Wikipedia to describe a list of conditions that cause dizziness in an article, it is legal for us to tell a poster whatever we can think of that causes dizziness.  It's not "practicing medicine without a license" if we're not representing ourselves as physicians.  Even if one of our contributors were attacked someday, it would either be (a) someone showing exceptional irresponsibility far beyond what people do on chat sites all over the Internet every day, or (b) someone singled out in an act of legal terrorism by which we shall not be cowed, and to which we may well choose to respond. Wnt (talk) 03:44, 7 June 2013 (UTC)
 * I am in basic agreement with Wnt on this even though, delicate flower that I am, I wouldn't quite phrase it the same way he would. What I do not understand is statements like "it just stupid and reckless to offer any medical advice to any person sight unseen". Why is it OK for me, sight unseen, to tell you how to wire up an outlet in your bathroom (it should be protected with a GFCI, and not all metal water pipes make a good connection to earth ground, in case anyone was wondering), build a black-powder cannon, or answer other engineering-related questions?


 * In fact, according the the current rules I can tell you where the best place to steal plutonium is and how to build an atomic bomb once you have the plutonium. However, if partway though building your bomb your hair starts falling out and you start puking, I cannot tell you that it might be radiation sickness and advise you to avoid plutonium -- that would be medical advice.


 * Why is engineering advice like that OK but it is not OK to tell someone that "eat less, move more" is good advice for most people "spend lots of money on homeopathic cures" is not? --Guy Macon (talk) 08:41, 7 June 2013 (UTC)


 * Just so people know, the concerns about "practicing without a license" and "liability" are only part of the story.
 * In general, practicing law or medicine without a license involves (a) misrepresenting yourself as a properly-licensed professional and (b) taking money, neither of which we would or could do. (But that doesn't mean giving medical or legal advice would be okay.)
 * Wikipedia has some decent disclaimers. (But that still doesn't mean giving medical or legal advice would be okay.)
 * User TenOfAllTrades has written an excellent essay describing the real reasons we really don't want to be giving medical or legal advice.
 * And, in any case, whether we should or shouldn't be offering medical advice is
 * one of those age-old, never-ending debates which option eight is talking about and which
 * this page is certainly not the place to reopen or debate further.
 * —Steve Summit (talk) 12:45, 7 June 2013 (UTC)
 * Everything in the TenOfAllTrades essay applies equally well to engineering advice, yet I can answer a question about wiring up a swimming pool pump. Why restrict only one and not the other if the best arguments you can come up with apply equally well to both? --Guy Macon (talk) 14:49, 7 June 2013 (UTC)
 * Because, in general, over the past ten years since the medical disclaimer and no medical advice policy has existed, more people have found the arguments against the giving of medical and legal advice compelling than those that do not. No matter how hard Wnt and Iblis and a few loudly vocal others object to it, the bulk of people do not feel comfortable with allowing them to give out medical and/or legal advice; but are not similarly uncomfortable with the giving of the wiring of swimming pool pumps.  That's really what it comes down to.  IF we want to overturn the current policy, THEN we need to start an explicit RFC which has a simple, confined scope that says simply "Do we want to keep or overturn the current prohibition against Wikipedia volunteers from giving out medical and/or legal advice".  We hold that RFC, and see where consensus goes.  We can argue back and forth on why we think it's a good or bad policy, but that gets us nowhere at all, regardless of how long the wall of text is that we make that either argues for or against the policy.  You are not going to get it overturned by creating your own wall of text, no matter how sound you feel your own logic and reasoning is when you make it.  Create that RFC, or follow the current policy.  That's the options.  -- Jayron  32  14:57, 7 June 2013 (UTC)
 * Yes, I agree that we should start such an RFC about medical advice in due course. This RFC is not the right place to go into too much details. Count Iblis (talk) 15:31, 7 June 2013 (UTC)
 * If this RFC has the power to abolish the Refdesk completely, to limit its scope, to move it, to impose worse restrictions, why is it that the one power it would totally lack is the power to put a stop to the worst trolling on the Refdesk, namely these inane claims that we can't give out medical information when we're not pretending to understand an individual case? That is simple interpretation of the existing guidelines, as has been consensus, on paper, with things like "Kainaw's criterion"; it has simply been marginalized by intense lobbying from those opposed to our involvement in biomedical areas of expertise.  I should emphasize that Wiki Med Inc. is an organization which actually seeks to obtain money from outside funders, and some of their people can be seen prominently in these discussions, so when I say "lobbying" I may even be speaking in the literal sense. Wnt (talk) 16:23, 7 June 2013 (UTC)
 * (Beating the dead horse)I've got no problem giving medical information. Check my history, you'll find I've defended that many times.  I see a distinction you don't between that and advice.  Despite appearances on this particular page, I'm closer to your side than you might think (though obviously I don't completely agree with you.)  We absolutely should offer medical and legal information, and there certainly are people who push it way to far in trying to close any question regarding medicine and medical conditions. All I know is that I apply the same standards to medical advice here as I do on the real reference desk I work at, for the reasons I outlined above.  But it's true: this isn't the place to keep this going.  A new RFC would be better for that. Mingmingla (talk) 16:44, 7 June 2013 (UTC)


 * Note that this RFC failed to achieve anything (from Beeblebrox's perspective), because Beeblebrox didn't do his homework before starting this RFC. So, I think we can do better with a new RFC specifically about the medical advice issue. We should carefully consider what will happen in practice if we implement the various options, given the way the Ref Desk works in practice, given the way other websites that do allow medical advice work in practice, given the way real people interact with their doctors, their friends in real life and and on the internet in practice. We then move away from discussing hypothetical situations that never happen in practice that are only raised to defend certain positions to real issues. Count Iblis (talk) 16:50, 7 June 2013 (UTC)

Option eight: Special board for disruptive questions
We have similar problems on some mathematical article talk pages where some old chestnuts get repeatedly brought up with endless pointless discussion. To solve this problem we have developed /Arguments pages. For example Talk:0.999.../Arguments, Talk:Monty Hall problem/Arguments, Talk:Gödel's incompleteness theorems/Arguments, Talk:Electron/Arguments. Those of an argumentative bent can use those pages to their hearts content, normal editors can happily ignore the page. Occasionally questions on the main talk page get moved to arguments.

A similar technique could possibly be used with the reference desk, a special board Reference_desk/Perennial discussions perhaps, with the trolling questions moved there.

Endorsements of option eight

 * 1) ✄ (talk) 17:20, 17 June 2013 (UTC)

Discussion of option eight

 * Can you clairfy what the scope of Arguments pages are? On some topics, an academic argument about the topic might be notable enough that it should be in the article itself. This is true for cutting edge science topic, and likely in some humanities topics as well.  A good 'academic' debate on an arguments page could be something that helps improve the article.ShakespeareFan00 (talk) 23:06, 3 June 2013 (UTC)
 * For 0.999... there is a well established mathematical result that 0.999...=1. This seems counter-intuitive to some and plenty of people challenge the result with arguments which have serious flaws. On the Monty Hall page there is a similar counter intuitive result. Here the arguments can be more subtle and have led to an academic paper. The other two don't tend to get used much. The pages serve to keep the talk page clear for discussions about the actual content of the article. I suppose there is a similarity with Perennial proposals of the village pump.
 * What could work is a page for long running discussions. If say a thread has run for more than a week it could be moved to that board. It does not kill the discussion but removes the distraction.--Salix (talk): 05:44, 4 June 2013 (UTC)
 * We do already have Reference desk/FAQ and Reference desk/Language/FAQs - but I doubt that many people check them before posting (and I'm not sure we even have a prominent link to them) - and I'm pretty sure that many of the questions listed there are not in fact frequently asked!  But just as our OP's don't check the archives - they aren't going to check a FAQ either.  We know that that a significant fraction of our questions can be answered by copy/pasting their question into Google and picking the first search result.  Our users are not sophisticated at using search engines...and expecting them to search a FAQ or our archives before posting is a wildly optimistic stance!  Those resources are mainly useful for ref desk respondents who can provide a quick link to a previous answer. SteveBaker (talk) 15:48, 4 June 2013 (UTC)

Though I think that an arguments page could be fun (for some of us...) I don't think it would really solve any problems. If the arguments pages were broadly advertised (highly visable to visitors,) they would be a troll attractor. If they aren't, then nobody but the regulars would find them and they would be barely used. --Yellow1996 (talk) 17:38, 19 June 2013 (UTC)

Option nine: Split to a "Research Forum" at Wikiversity
Sometimes in getting an answer to a refdesk question, contributors have used multiple sources to produce what in larger terms might amount to original research, or at the very least a form of synthesis. In articles both of these are generally discouraged.

It would therefore be better to host the 'explanatory' side of the reference desk in its current form to a wiki whose policy supports the sort of academic synthesis and original research that occurs on the Reference Desk but which is generally discouraged in articles.

Wikiversity would seem to be an ideal place for a Research Hub so the proposal is that the Reference Desk at Wikipedia reverts to dealing with finding actual resources, Example " Which Journals discuss current theories on meson acceleration?" whereas the "Research Hub" on Wikiversity would handle the more abstract or explanatory questions. Example: " What are the current theories of meson acceleration, commonly accepted?"

Endorsements of option nine

 * ShakespeareFan00 (talk) 23:21, 3 June 2013 (UTC)
 * Weak support - I don't support being a part of Wikiversity - but having a "WikiRefDesk" would work for me. Being a separate 'branch' of the Wiki family would have advantages for the Reference desks.  For one, we could vary our rules more appropriately than Wikipedia guidelines allow.  We would actually become MORE visible on the front page than we currently are.  There would also be the possibility of changing the way the underlying software works so that we could more flexibly archive questions only when no further answers are forthcoming, etc.  I'm not sure though - this would be a gigantic step with massive difficulties in terms of getting the change enacted.  I can't see it happening. SteveBaker (talk) 14:35, 4 June 2013 (UTC)

Discussion of option nine

 * Obvious problem #3577; nobody uses Wikiversity. Ironholds (talk) 03:24, 4 June 2013 (UTC)

Option ten: Expand and support the Refdesk's mission to establish a database of questions
We've all seen the ads on TV. When kids have a question, they ask their closest friend and companion, their Android phone, "Google, why is..." -- and they get their answer. Of course, they also get their voiceprint registered with the company and likely, despite heroic legal resistance, shared widely under any one of a hundred national security laws, and their location tracked by GPS, and their place in a company's incipient monopoly over the very idea of searching, asking a question, translating, etc. But imagine for a moment that there was a competitor after all, a free culture alternative, someplace else that had a database of questions and answers that people could get an answer from. Can you think of one?

We have a huge archive of answered questions. I've written up a module to index it with a little bit more information than even before (WP:Reference desk/Archives/Lua) though there is more that can be done there. Still, with enough volunteers, or better programmers, we could categorize, and then subcategorize, all of a decade of existing questions, and eventually group similar topics quite closely, provide curated, concise answers based on the information we have collected. We could create a database of hundreds of thousands of questions and answers ready to be used by any party interested in delivering answers to questions. Maybe some open source programmers can even witch up a script to search through them based on a voice request, you never know.

What I propose is that we don't treat answering questions like we're doing a favor, or wasting our time - we treat it like we're building a useful encyclopedic resource, a database of questions. We stop worrying about whether someone is asking too many, or whether it's ethical to help kids with their homework. We thrust aside with prejudice any notion that it is wrong to answer questions about medical or legal topics that state the same things we should rightly include in the encyclopedia, provided we are not (falsely) claiming to know an individual's case or what is "best" for someone. In short, we stop thinking of the Refdesk as a service to users in the short term and start thinking of the accumulated discussions as a permanent knowledge resource we are building for the world to use. Wnt (talk) 02:30, 4 June 2013 (UTC)

Endorsements of option ten
Wnt (talk) 04:41, 4 June 2013 (UTC)

Partial support. I admire your contributions and ideas on database/indexing, and have some sympathy on the boogeyman of "medical advice. But the two are separate, and only the first has any traction at the moment. SemanticMantis (talk) 23:56, 4 June 2013 (UTC)

Support. Somewhat tepidly, because I can't imagine how it would look. But one of the constraints of RefDesk is that previously asked/answered questions aren't immediately available as a resource. There is an amazing amount of value somewhat hidden away. EBY (talk) 15:02, 5 June 2013 (UTC)

Option eleven: Make the Ref desks more prominent by allowing Googlebots to crawl the pages
The problems with the Ref Desks are caused by the Ref Desks being isolated in cyberspace. A random person doing a google search isn't directed to the Ref Desk, rather they will be directed to Yahoo Answers, Stackexchange or other sites. This means that people with questions and people with expertise who can answer these questions will not come here, they won't know that we even exist. We are thus condemned to remain a small community that answers questions from a small group of posters.

By expanding the community of experts, the quality of answers will go up and you get less of a pile up effect where quite a few regulars will give replies addressing problems with the previous replies. What will happen is that there will always be a great answer posted, so then you don't want to post an imperfect answer. So, you are only going to answer if you can provde for a perfect answer to a question that hasn't already been answered. This dynamics is at present completely absent on the Ref Desks.

Count Iblis (talk) 18:59, 5 June 2013 (UTC)

Endorsements of option eleven

 * Support There's little point doing this if the Q&A isn't searchable. Warden (talk) 11:30, 6 June 2013 (UTC)

Discussion of option eleven

 * Strong oppose!! - this would be an open invitation to spammers and evildoers of all stripes to spam our ref desks into oblivion. Bad, bad idea! SteveBaker (talk) 20:13, 5 June 2013 (UTC)
 * The proposal doesn't seem to take into account that ref desk entries are archived after a few days. So anyone searching e.g. cat smell would only find the desk in that brief time window.  Plus, I don't see how the proposal is predicted to attract experts in quantity, it would surely attract questions in much greater quantity.  Sussexonian (talk) 20:40, 5 June 2013 (UTC)


 * How about if the archives, and only the archives, were made crawlable (the archives that don't show on the page, that is), and the archives were protected so that "spammers and evildoers" could not get at them? That might take care of Steve's objection. Looie496 (talk) 23:27, 5 June 2013 (UTC)
 * I'm confused about what this is proposing. The reference desks are not noindexed, in our robots.txt or anyway made non crawable to bots beyond that which applies to most of wikipedia (certain misbehaving bots may be banned) that I'm aware of. There was a successful proposal to noindex unpatrolled pages Requests for comment/NOINDEX, I don't think this was supposed to include anything outside the article namespace so shouldn't have affected the archives and in any case, it's been abandoned for now because of implementation problems anyway AFAIK Wikipedia talk:Page Curation/Archive 6. There have been other proposals to noindex all wikipedia namespace pages by default NOINDEX of noticeboards, Search engine indexing but these all failed AFAIK and it was a common view that there needed to be an optout if it was implemented and the reference desk was one area that should be opted out.
 * Definitely I've found the reference desk and archives when looking for specific questions plenty of times with an external search engine and a few simple tests shows it still works. It does seem to be quite difficult to actually find the reference desk without a very specific search, I usually include 'reference desk' or similar in my search. But it's easy to prove you don't have to do this, e.g. try 'effects of marijuana as a drink very nasty indonesian' on Google or 'effects of marijuana as a drink very nasty indonesian film darah' or 'finding a relationship between elliptic integrals and the factorial/gamma function' in Bing or just do a phrase search.
 * I don't know why we tend to be so low rated/difficult to find, my guess it's a combination of a lack of external links to the RD, the lack of keywords and a large number of other factors we can never know because the secret sauce of search engines is usually kept quite secret. Ultimately, I don't know what we can do about it, and it's not clear to me what this proposing is suggestion we do about it. The proposal, or at least the commentators sound to me like they believe we aren't indexed or shown in search engine at all at the moment. It's true all external links in wikipedia are 'nofollowed' including the ones on the reference desk, I guess it's possible this harms us in some way but this definitely is never going to change.
 * Nil Einne (talk) 19:04, 10 June 2013 (UTC)

Option twelve:pseudo-forking
The reference desk wil make it clear that it is not part of the encyclopedia and that the standards that apply everywhere else on this project to insure reliablity do not apply in any way at the refdesk. It will continue to be hosted on en.Wikipedia but will be trreated as a semi-separate entity. A notice on each refdesk page and an edit notice will be displayed to emphasize this at all times to all those posing questions so that they understand that the refdesk isn't really the Wikipedia they know that has striven so hard to be reliable. 02:43, 6 June 2013 (UTC)

Endorsements of option twelve

 * 1) As proposer. This is the closest I can think of to a "have your cake and eat it too" solution. I'd honestly rather not "kill" the refdesk but I don't think we should act like it is part of the actual encyclopedia when it ignores so many of the core policies. Beeblebrox (talk) 02:43, 6 June 2013 (UTC)

Support. I've given my reasons in the discussion below. --TammyMoet (talk) 19:34, 6 June 2013 (UTC)

Discussion of option twelve
I knocked together a draft of what that warning might look like:

Please be aware of the following I'm sure someone with more template coding experience may be able to make it a bit more sexy, but you get the idea. Beeblebrox (talk) 02:54, 6 June 2013 (UTC)
 * The Wikipedia Reference Desk is not part of the encyclopedia. Rules that apply in articles such as always providing reliable sources do not apply to answers given here.
 * None of the users giving answers can be assumed to be actual experts on any topic, and no advice should be taken as representing a professional opinion on the subject.
 * The rule that Wikipedia is not a forum or chat room does not really apply here either. You may find that instead of one simple answer you get a wide range of answers, some of which may be speculative or contradictory.


 * I edited the formating of your template. My primary goal was to remove the usage of which was adding an entry to the table of contents.  Dragons flight (talk) 04:34, 6 June 2013 (UTC)


 * Do you have similar templates for User:Jimbo Wales, WP:AN, WP:RfA? I'd like to see that one - WARNING: None of the admins here can be assumed to have any actual administrative or HR experience of any kind, nor should any scarlet letters they demand to maintain on your talk page after they block you should be taken as representing a factually based conclusion, nor does WMF assume any responsibility for it... Oh my, I think that disclaimer would go on for pages. Wnt (talk) 14:11, 6 June 2013 (UTC)
 * Admins at AN aren't pretending to be doing anything other than administrating this website. The refdesk is pretending to be an actual reference desk, which I'm pretty sure we all agree it is not. Beeblebrox (talk) 14:51, 6 June 2013 (UTC)
 * Really? They don't, say, set up permanent lifetime walls of shame where they brand teenage editors identified by their real names for things like "pedophile-related disruption", in remarkable contrast to homilies about BLP?  "Administering the site" seems to have a very loose definition, just like "reference desk".  But I see comments above from people who say that the Refdesk works exactly like a library reference desk.  Can you name one non-profit organization or government whose "administration" consists of encouraging people to totally trash each others' projects and humiliate one another? Wnt (talk) 15:06, 6 June 2013 (UTC)


 * I like this one. It is merely stating what actually goes on here in the manner of the ISO9000 series of Quality Assurance - "say what you do and then do what you say". It would mean that those who think the Reference Desk should only provide links would have to amend their thoughts on the RD, but then it is usually easier to get the minority to change their views than the majority to change their behaviour! It also means we can provide references which are not considered reliable but which are often the only possible answers. In other words, we can help more people than we do now in a friendlier way than we do now. For me, that's a good outcome. --TammyMoet (talk) 19:32, 6 June 2013 (UTC)
 * Despite TammyMoet liking it (and I agree with Tammy on this RFC otherwise), I think this proposal is the clearest example possible of trying to make a WP:POINT. (Just look at the statement The standards that apply everywhere else on this project to insure reliablity do not apply in any way at the refdesk.) I would be happy to see a statement added to the text at the top of the Ref Desks which says "This page is not part of the encyclopedia" as we find on user pages and elsewhere.  The Reference desks are not part of article space, in the same way that policies, talk pages, teahouse, xFDs etc etc are not.   The standards which aim for reliability do apply here, even though they are not always achieved any more than on article pages.  We do not expect article editors to be experts in the field, nor refdesk helpers.  You may find that instead of one simple answer you get a wide range of answers, some of which may be speculative or contradictory.  That's just silly.  Sussexonian (talk) 19:47, 10 June 2013 (UTC)

Option thirteen: Rename the Reference Desk
Short version: Let's acknowledge reality and call ourselves an answer service rather than a reference desk.

The Wikipedia Reference Desk, like the services at sites such as Wikianswers and Yahoo Answers, tends to focus on trying to answer questions. Where a library reference service might provide users with a list of resources they can consult to hopefully find an answer, we are generally more oriented toward providing an answer directly and hopefully explaining it in enough detail that it is understood. In the process of answering questions, responses are generally expected to include wikilinks, urls, or other supporting documentation to corroborate the response, but that is often only a small part of what is said. In practice, the Wikipedia Reference Desk is primarily about finding answers and only secondarily about providing users with avenues for further research.

If the Reference Desk is going to continue to operate in more or less the present mode, then I would suggest that it would be helpful to acknowledge this reality by changing the name. My suggestions would be Wikipedia:Answer Service or Wikipedia:Answer Desk or maybe just Wikipedia:Answers. There could be other possible names that might be even better. A name change could help visitors and participants understand that we are more like other online answer services and less like a traditional reference desk. Obviously, people who think the Wikipedia:Reference Desk shouldn't be an answer service are likely to disagree with this proposal, but I am making this proposal on the assumption that large changes to present mode of operation of the Reference Desk seem unlikely to gain consensus.

This change would be primarily cosmetic, simply serving to better acknowledge a reality that has long been true. The documentation should continue to encourage / require the use of references, at least in the form of wikilinks and preferably more.

Endorsements of option thirteen

 * 1) This is my suggestion for a minor improvement. Dragons flight (talk) 03:21, 6 June 2013 (UTC)
 * 2) This dovetails perfectly with option twelve and if we aren't going to be able to change the status quo then both changes should be implemented. The refdesk will be allowed to continue as it does now, but it will be made clear that it is neither part of the encyclopedia nor an actual reference desk. Beeblebrox (talk) 03:37, 6 June 2013 (UTC)

Discussion of option thirteen
The problem I see with this is that the best answers generally do come with sources. I think the Refdesk does a much better job than Yahoo Answers, and that is one of the reasons why. I particularly dislike the notion of calling it a "service" for the reasons I lay out in option 10. I'd consider something like WP:Question Database, but that is too awkward, though perhaps a better phrase can be devised; for now I don't see a reason to toy with the status quo. Wnt (talk) 14:06, 6 June 2013 (UTC)

Calling it an "Answer service" implies that all questions will get answers. It implies some form of quality of service which I'm not prepared to sign up to - even more than the term "Reference desk" does: outside the confines of academia, this term means "a point at which reference may be made" rather than "a service which provides references". --TammyMoet (talk) 19:36, 6 June 2013 (UTC)

In all this discussion no-one seems to have mentioned WikiProject Resource Exchange/Resource Request so I shall mention it if only to give it a plug. It has far too low a profile. That forum is rather strictly a means of requesting references and it is focussed on obtaining sources of direct relevance to developing articles. Thincat (talk) 09:06, 7 June 2013 (UTC)


 * Definitely it has a related purpose - I remember adding a link to it somewhere on the Reference desk main page - but it serves the narrow purpose of interlibrary loan - one important part of the desk. Also, I think there is the fear that (like borrowing CDs and movie DVDs at the local library) if it became popular enough, somebody would be along in short order to ban it, which makes people reluctant to put a lot of effort into promoting it.  (Heck, as it is maybe some of the same people here will go after them next) Wnt (talk) 04:02, 9 June 2013 (UTC)

Note: This (or a very similar) proposal is currently being discussed separately at Wikipedia_talk:Reference_desk. Since the RFC here might be closed soon, I suggest commenting on the name change there. ---Sluzzelin talk  23:24, 3 July 2013 (UTC)

Option fourteen: Encourage the culture of using research at the desks for improving the encyclopedia
Different people use different points of motivation and inspiration for doing whatever they do, including for working on an encyclopedia. There are a number of editors who regularly use questions at the desks as a springboard for improving and creating articles in mainspace. Some editors have mentioned this above (see "Discussion of option two"). For one small sample, a few articles had been (and very occasionally still are) added to the Category:WikiProject Reference Desk Article Collaboration.

Whatever works to increase this kind of service to the encyclopedia is welcome. Little things (like perhaps moving "Use external links from answers as sources to improve our articles" further up, on the list of guidelines) to bigger things, like ideas of how to get others to help — while the idea of an article-creation-"project" never quite took off the way it was enthusiastically hoped for, what does often work is mentioning, in the threaded question while it is still hot and interesting, that one is struggling with how to weave a certain piece of information into an article. Bringing the more arcane questions to the attention of editors at our various projects might get better results in the thread as well as in mainspace. And so forth.

Not a very concrete or exciting option, I'm afraid, but it's worth thinking about how we could increase the desks' actual value to the encyclopedia.

Endorsements of option fourteen

 * 1) (As proposer). Added before supporting or commenting on other options. ---Sluzzelin talk  01:36, 11 June 2013 (UTC)
 * 2) Certainly. There are situations where a WikiProject is a better place to go for help, but there are also plenty of situations where the Reference desks can do a good job.  I've used them for article-development questions myself a number of times. Looie496 (talk) 03:31, 11 June 2013 (UTC)
 * 3) Support, since the end result, in the best case, is that people sought out reliable sources to provide answers to important, interesting, or useful and encyclopedic issues in cases where the information was missing, or poorly presented in the relevant articles. The net result is an improved encyclopedia. I suppose the downside would be is someone canvassed to get support for his side in some factual dispute (Tesla was Serbian. No he was Croatian!) and it just mustered more cannon-fodder to some edit war. Edison (talk) 17:27, 12 June 2013 (UTC)
 * 4) Support. In this vein, I recently created Eagle effect as a result of researching a question on RD/S. -- Scray (talk) 20:50, 12 June 2013 (UTC)
 * 5) Support: it is possible to see new information added to articles, or articles clarified, as a result of ref desk discussions. Rarer to see a new article created. A "good" question - one that requires someone who knows more about the subject to go away and do some research outside Wikipedia, and bring that new information to the ref desk - has the potential to spawn a new article or improve an old one. I wonder what the best way is to encourage "content-spawning" though - I don't think it helps the way that questions "disappear" into the archives. Perhaps particularly good question discussions should be linked to in the talk pages of relevant articles so future editors (or especially curious readers) can fish them out, or there could be some kind of link to "articles for creation"? TheGrappler (talk) 20:07, 24 June 2013 (UTC)

General discussion
Please use this section to discuss any issues beyond the individual proposals


 * So, just to get the ball rolling here, I think most of us that are familiar with the reference desk can agree it doesn't always work anything like a real library reference desk. I've never seen librarians gather in a small group and just start making up answers to questions. That isn't what happens every time, but it happens enough to be a problem. I've also never seen librarians put up with patrons who ask one ridiculous question after another when they are clearly just asking for the sake of it and not because they actually have some reason to care what the answer is. I would imagine a real librarian would tell such a person to go read a book and figure it out for themselves after being asked three or four pointless questions. Something needs to be done about these issues, and I have tried to come up with a broad spectrum of possible solutions, including some that are fairly radical. As you can see there is also space for any new ideas to be added. Beeblebrox (talk) 17:35, 3 June 2013 (UTC)
 * Speaking as someone who works a real reference desk in a real library, we are annoyed occasionally by people who ask odd an incessant questions. We don't kick them out after 4 dumb questions: we ask them to step aside when another person comes up, maybe, but we don't just kick them out, and we do allow them to come back.  That's because it's what we are paid to do.  The problem is that we don't know for sure why they ask the questions.  They could be genuinely curious but simply not very smart or able to formulate anything that makes more sense.  On these ref desks, however, if we don't want to answer, it's even easier than in a real library: we just ignore them and leave them unanswered. We don't get to choose what people ask, but we do get to choose to leave them alone. Mingmingla (talk) 21:45, 3 June 2013 (UTC)
 * Plus 1 the preceding comment! SteveBaker (talk) 14:36, 4 June 2013 (UTC)
 * Sure, for 4 dumb questions -- but what about 40, with no prospect of stopping? (Yes, that has happened.) Looie496 (talk) 02:36, 6 June 2013 (UTC)


 * Well, that begs the question of whether our reference desks should work the way a library's reference desk does... or whether it should instead work some other way. Wikipedia is not a library, after all... it's an encyclopedia.  Indeed, it may well be the only encyclopedia to actually have a "reference desk" where people can ask questions.  That means we are breaking new ground.  We can follow the model set by libraries... if we wish to... but we can follow a different model if we want to.  We can even invent our own model and take the concept of what a "reference desk" is in a completely new direction.  Our reference desks can be whatever we want them to be.
 * I opted for option 4 above (the Status quo) ... because I don't really see a problem with the way we handle our reference desks at the moment. We get all sorts of questions... and I think we do a fairly good job of answering them.  When a question asks for sources, we appropriately give sources... when it asks for facts and figures, we give facts and figures (and when the facts are debated it becomes quickly apparent - we disagree and call the facts into question.  Thus highlighting that there might be more than one "correct" answer to the question)... and when it asks for opinions, we give opinions... usually quite good ones (granted, we could do a better job of supporting our opinions with facts and sources). Blueboar (talk) 18:42, 3 June 2013 (UTC)
 * Actually, Encyclopedia Britannica did once have such a service. I forget what they called it - but many years ago when I bought a print-edition Britannica ($3,000!!!), it came with a bunch of little "postage stamps" that you could affix to a question and post it to their headquarters where the Britannica team would research an answer for you.  I used this service many times - and the responses were pretty similar to the Wikipedia Ref Desk - except that since this was in an era before the Internet was huge, references tended to be in the form of photocopied articles and such.  The questions I asked (and the answers I got) were very similar in quality and tone to the Wikipedia Ref Desk. SteveBaker (talk) 19:06, 3 June 2013 (UTC)


 * The benefits of having the ref.desk go far beyond simply answering questions. It is extremely common to find gaps and errors in Wikipedia's articles following these questions - and I, personally, have started dozens of articles that have grown into useful material following questions to the ref.desk.  I'm also a fairly frequent user of the desks - mostly in areas like language where I'm no expert myself.  I have never failed to learn something from the responses I've gotten.  It's a valuable resource and it's worthy of Wikipedia to supply the trivial amount of resources it requires to maintain it. SteveBaker (talk) 19:12, 3 June 2013 (UTC)
 * This is a nuclear solution without any obvious problem to solve in the first place. If you're really offended some guy asked about how cats react to petrol fumes, you could have chopped out his question and been done with it.  But even the most preposterously pointless question isn't really that bad - it's hard to answer, after all, but doubtless there's some interesting biology there.  There are (or at least, were) natural oil seeps, wells naturally tainted with petroleum; there probably have been places where animals for millions of years have decided whether or not to use a watering hole.  I bet there's a fair chance we've inherited some P450 enzymes evolved in such a place, if only we had a way to tell.   As long as people deluge us with creative ideas, this isn't really an attack we have to resent.  It's some sort of intellectual pillow fight, really quite harmless. Wnt (talk) 22:05, 3 June 2013 (UTC)
 * It's not about being "offended" by a question - it's about wanting to help a user by answering a question that can't be answered without further input from the user. Any ref desk worth anything would reasonably expect a user to clarify a vaguely-worded question. ←Baseball Bugs What's up, Doc? carrots→ 22:44, 3 June 2013 (UTC)
 * Is this an official RfC? I don't see it listed at Requests for comment/All. —Steve Summit (talk) 00:21, 4 June 2013 (UTC)
 * Ah. Answering my own question, I see it's at Requests for comment/Wikipedia proposals. —Steve Summit (talk) 00:30, 4 June 2013 (UTC)


 * I voted option three, but this looks like a snow-close for option four. μηδείς (talk) 01:46, 4 June 2013 (UTC)

It's really too bad that Beeblebrox has dropped this surprise RfC without making any prior attempt to engage with any part of the community. I don't think he could have more effectively sabotaged his attempt to improve the Ref Desk if he had tried. My thoughts on this are at Wikipedia talk:Reference desk/Refdesk reform RFC TenOfAllTrades(talk) 02:47, 4 June 2013 (UTC)
 * Maybe you could link to that for us? μηδείς (talk) 03:22, 4 June 2013 (UTC)


 * The key thing lacking at the Reference desks is dialogue. I think the Reference desks are great, but we have some problems. Whether inadvertently or deliberately, questions are posed that are problematic. Clarification is sometimes needed. We should be engaging the person posing the question in dialogue. This should be expected on our part. We have a burden to intelligently engage the questioner in dialogue. And they have an obligation to respond. I find it frustrating to find a dozen regular Reference desk people responding to a "question" that has not had requisite clarification supplied. This is frustrating because people are guessing what question was asked. The consequence is a widening of the area of discussion. And this becomes truly forum-like. An expectation of give-and-take dialogue is lacking at the Reference desks. I am not recommending conversation for extraneous purposes. I am recommending conversation for the purpose of homing in on the question and weeding out trolling and such. We don't have to tell anyone they are a troll. This can become apparent in the course intelligent dialogue, in which questions asking for clarification are posed to the questioner. If the questioner refuses to engage in dialogue we should apply the maxim of "don't feed the trolls". That could mean hatting the question but I would prefer just ceasing to respond. Bus stop (talk) 03:29, 4 June 2013 (UTC)
 * Well, as I said in "option ten", I think a major purpose of the Refdesk - even the main purpose - is to create a database of questions and answers. For that purpose it doesn't really matter if you're answering what the questioner had in mind or not, whether he's a troll or not; you're just using him as a sort of Monte Carlo style random number seed to sample part of the intellectual landscape.  What we need is not a true knowledge of his exact mental state and interests; we are simply harvesting his creativity for use in our researches, whether he is aware of that or not. Wnt (talk) 03:42, 4 June 2013 (UTC)
 * While it seems attractive to be building a database of answers, I doubt very much whether many of our OP's would take the time to search it before posting. If they did - then our existing "search the archives" function has that covered.  It's more likely that (as at present) someone asks a question and the Ref Desk experts recall a previous answer and refer them to it.  Hence any such database would become nothing more than another research tool for ref desk denizens. SteveBaker (talk) 14:55, 4 June 2013 (UTC)
 * I'd like to posit that some people actually do know what the questioner is talking about and can provide a good answer to their question. This position is based on the fact that, while we all speak English, we come from all over the world and bring with us our culture and experience. So while you yourself may not understand a particular question - or maybe it pushes a hot button for you - bear in mind that not everyone will share your reaction to that question, and maybe curb the tendency to jump in with all feet shouting. Our diversity of knowledge and experience is our strength - if only we are allowed to use it! --TammyMoet (talk) 19:28, 4 June 2013 (UTC)


 * My main problem with this RFC is that it seems to be a general "I don't like the ref desks" bitch session. It doesn't lay out a specific problem or set of problems that needs to be resolved.  If the problem is "too many dumb questions" then the option to abolish the ref desk is ridiculous over-kill, ditto if the problem is "too many medical/legal questions are being answered".   We need to debate this in bite-sized chunks.  Either:
 * We should list the current set of problems - attempt to address them individually - and only if it seems that they are insolvable should we start to address "nuclear options" such as abolition.
 * We should decide whether we want the ref desk to exist at all...and if we do, we should address any problems it has.
 * Attempting to do both at once is creating a messy RFC that's hard to resolve with a simple Support/Oppose response. In many cases, I can see people are trying to say "If we could resolve X and Y then we should keep the ref desks - but if we can't it should go"...and it's just not possible to express that position clearly with the catch-all structure of this RFC. SteveBaker (talk) 14:55, 4 June 2013 (UTC)

Right now, we have 9 options - of which only two are supported by more than two people. So in the end, we have 5 !votes to abolish the ref desk and 19 !votes to keep it without change - and nothing close to consensus for any other course of activity. There is considerable rebuttal of the remarks made by the abolitionists with no contradictory responses to those rebuttals. We have no rebuttal whatever of the points made by the 19 status-quo-ists. Clearly debate isn't going to overturn this response...and without overwhelming abolitionist consensus we aren't going to change anything anyway. To overturn anything, the five abolitionists would have to convince most of the 19 status-quo-ists and the handful of "we-can-fix-it-ists" - and they have so far offered not one line of rebuttal to do that. There isn't any realistic prospect of a sweeping consensus to change anything. I think it's over. SteveBaker (talk) 19:43, 4 June 2013 (UTC)
 * Don't be ridiculous Steve, this has only been open one day and so far we have mostly heard from refdesk regulars. This is listed on WP:CENT and over the next month the broader community will express its opinions as well, and who knows what the final result may be, but it is waaaayyyyyy to early to call for a close or declare the debate over. Beeblebrox (talk) 20:04, 4 June 2013 (UTC)

Just to throw this out there, this sort of behavior on the refdesk is what puts me over the line to wanting it deleted. I was horrified then that the denizens of the refdesk apparently felt their behavior in the thread (cf here) that ANI references was acceptable, and I continue to be horrified by it now, on re-reading about it. Questions like that one should absolutely not be responded to, and certainly not in the manner the reference desk responded to it, and as long as the culture of the refdesk is such that that is the response to a possible minor asking about a sexual assault, I find it to be more liability than help to the project. A fluffernutter is a sandwich! (talk) 20:32, 5 June 2013 (UTC)


 * Clearly that was a pretty horrifying situation. It's bad that the question was asked so obviously and in such vivid detail (I'd have gone with "There are some embarrassing photos of me posted at such-and-such places, how do I get them removed?") - and it's utterly unacceptable that DriveBy answered in that way.  However, this is Wikipedia and worse things happen in article space every single day.  Kids "out" their gay friends in the article about their schools, people post inappropriate photos to Commons and Wikipedia and it takes a few hours for someone to notice and remove them...many MANY things like this happen.


 * I've been working the ref desk off and on for many years - and this is by far the worst incident I can recall (and I happened to be on WikiBreak at the time). But to dispense with a service that has helps around 10,000 people a year with useful answers to often difficult questions - solely on the basis of the coincidence of a poorly thought out question and a disasterously insensitive answer is really taking things a bit far.  The ref desk DID self-correct.  If the ref desk didn't exist, then maybe a desperate kid would have posted it at Talk:Penis or Talk:Facebook and gotten at least as poor a response...and without the experience of ref-deskers to jump in and fix it in a timely manner. SteveBaker (talk) 21:19, 5 June 2013 (UTC)
 * Sorry, let me clarify, Steve. DriveBy's behavior there was, of course, absolutely unacceptable. But what horrified me more was everyone else's behavior, where they threw out random answers and offered their personal advice, and it didn't seem to occur to anyone other than one person who was reverted that dealing with a legal/sexual assault question with those kinds of casual, inexpert, non-"Get this off a public encylopedia page and call the police if you think you've been assaulted" answers was a horribly wrong choice. Everyone was participating in good faith, but no one seemed to have had the judgment to go "Hey, this is not a question it is safe or appropriate for us to be answering, nor is it particularly wise for the asker to pose it here." The refdesk didn't so much self-correct as someone had to go to ANI to get it corrected while the refdesk continued digging a really unfortunate hole. A fluffernutter is a sandwich! (talk) 13:36, 6 June 2013 (UTC)
 * You do realize, Fluffernutter, that DriveByWire, who made the inappropriate request, was the long-blocked troll Light Current? It seems odd to place the blame for such a mess at the feet of "ref-desk regulars" when the objectionable part of the discussion was by a troll who we've been battling for years.  His inappropriate response was removed, ultimately.  -- Jayron  32  04:04, 9 June 2013 (UTC)
 * There were two issues in that thread, and you seem to be thinking my issue is with the first rather than the second. First, we had the long-term troll under the guise of DriveByWire. He made a horrible comment, and though there were some false starts, he and his comment were more or less dealt with appropriately. That the refdesk is sometimes trolled is not the problem here - the problem is the second layer of things that happened in that thread - the part where utterly non-trollish, well-intentioned, trying-to-be-helpful users also participated in that thread, giving responses that were well-meaning but also completely inappropriate. My problem is not with DriveByWire's trolling - trolls happen - but with how everyone else's judgment appears to have failed in that thread as well, in regard to each person's handling of the question (note that I say their handling of the question, not their handling of DBW or his comment), and my overarching point is that as long as we have a reference desk that gives (well-meaning but) incredibly inappropriate advice to people who ask those sorts of questions, I'm not convinced it's a service we should be offering. A fluffernutter is a sandwich! (talk) 14:01, 10 June 2013 (UTC)
 * I think that it is revealing that a Refdesk retractor thinks that one inappropriate comment is more important than tens of thousands of correct answers. It appears it is true that when we say that proponents of censorship make it their sole priority, that is not an approximation.  There is literally not one nanogram of weight reserved for consideration of the good people can do by participating in this project, and no plausible scenario under which, given an opportunity, they would not set the server building on fire and walk out singing hymns to their glory. Wnt (talk) 03:24, 6 June 2013 (UTC)
 * You may find people take your comments more seriously if you leave out the overblown hyperbole. Beeblebrox (talk) 03:38, 6 June 2013 (UTC)
 * To quote Jack Nicholson, "Is there any other kind?". --   Jack of Oz   [Talk]  06:29, 6 June 2013 (UTC)
 * Alright, to track down this silliness to the source, I found the original thread, still archived at Reference_desk/Archives/Miscellaneous/2012_July_11. Note that not only was the answerer  DriveByWire permanently blocked ; so is the person who asked the question, both based on the presumption that they are "sock puppets".  How the admins decided that the guy asking the question was a troll, I have no idea - nothing links to his user or talk page, so there can't have been much of an investigation!
 * Now to be clear, not one of those responses is within Refdesk guidelines, because everyone insisted on giving some sort of advice, and we're not supposed to be giving any. However, I'm not as concerned by the actions of somebody who says "you ought to change your email address" as with those of someone who blocks someone in such a desperate situation based on speculation that he is up to no good, or prohibits people from trying to come up with ideas to help.  I compare the action of DriveByWire, who mocks the kid under the assumption that he is a troll, and the admins, who block the kid because they think he is a troll, and of the two, I prefer LightCurrent to the Wikipedia admins.  At least he was trying to be funny!  In any case, if the two are indeed but one troll, the fact that some people here are discussing shutting down the entire Refdesk system because of his actions is the ultimate accolade that any troll could hope for, and it is time for all of us to take a moment to kneel in homage. Wnt (talk) 16:02, 10 June 2013 (UTC)
 * DriveByWire was a fairly obvious sock Cuddlyable3, the behavioural evidence was fairly strong. The link to LightCurrent was a mistake by someone who was fairly unfamiliar with the RD. An investigation was started here Sockpuppet investigations/Light current/Archive but no attempt was made to link it to C3. I suspect any evidence about C3 was too old for a linkage by the CUs anyway.
 * The other editor raised in that CU was Bred Ivy (who wasn't involved in the fuss over that specific question IIRC). I don't think I personally ever worked out who that was (Jayron32 and perhaps some others may know), I don't personally think it was C3. The non linkage is one thing but it also appeared they were going to a great efforts to imply they were someone else, while at the same timet somewhat incriminating themselves of sockpuppetry, and some of this seemed to beunrelated to the case or even before it. So in the end, it was perhaps a moot point as their comments strongly suggested an inappropriate use of multiple accounts. (While they mentioned a desire to avoid linkage to their RL, it appeared to be a case of 'I don't want to people linking my harshly worded comments and possible personal attacks on others to my real life' not 'I don't want people knowing I'm involved in editing these or have these views'.)
 * As for the Daniel account which started the question, I'm not sure what evidence there was for that. But although I may have occasionally used humour when dealing with trolls (more commonly with non-trolling problematic editors), it's generally considered best to WP:DENY (i.e. block, delete, ignore) for a reason which you yourself seem to have alluded to, they want the attention!
 * BTW, if you look at the blocklog, the Daniel account was blocked by someone else who didn't participate in the CU request and before the other accounts so I don't believe there's any suggestion Daniel was C3 (or "Bred Ivy" for that matter). The block also came a while after the question or for that matter the ANI thread and SPI so I'm not sure the block was just because of the question.
 * Edit: Actually the logs of the person blocking [//en.wikipedia.org/w/index.php?title=Special:Log&dir=prev&offset=20120626141740&limit=20&type=&user=Jpgordon] seems to confirm it was mostly unrelated. I presume a CU was done as although the edits to the RD and in general often show a similar style (and one has a nick that's similar to some of the questions), they didn't all edit the RD. I'm still not sure who it is, some of the edits suggest it could actually be LC, but then again one of the accounts was created in 2008 and although never editing the RD I would have expected it to be found in one of the many CUs done on LC. Either way I don't care but it seems clear the blocks were all appropriate.
 * Anyway I don't think any of this is going to make a big difference to fluffernutter, it seemed clear from their early comments and this was re-affirmed by their later comments that their concern isn't going to depend on the trolling even by the editor who asked the question. While it may seem unfortunate that the big deal is over something which was trolling anyway, that isn't going to negate what they see as a problem. Ultimately it's possible that a non-troll will ask these sort of questions. I don't personally share their view on how poorly it was handled but it's not something I can be bothered discussing hence why no replies until now.
 * Nil Einne (talk) 20:33, 10 June 2013 (UTC)


 * Something I find somewhat concerning is the use of the RD to seek opinions on content disputes, instead of the mechanisms intended for the purpose (WP:3O, RfC, &c.)—in effect leakage from article Talk pages or a form of forum-shopping. Where it gets to the point that the conversation becomes sprinkled with WP:XX shortcuts it must get awfully confusing to readers who don’t frequent the venues where such discussion normally occurs. If there’s to be a warnings-&-disclaimers banner, perhaps it should include some language discouraging such questions, or emphasizing that the RD is not an alternate WP:DRN.—Odysseus 1 4 7  9  03:27, 10 June 2013 (UTC)
 * We do occasionally get people seeking opinions on content disputes etc. We generally direct them to appropriate places. Occasionally we may offer opinions or help although that's more common with the helpdesk sort of stuff or cases where it's not a content dispute but simply someone confused about how to edit and use wikipedia. Nearly always where it matters, people make it clear their responses are intended generally and shouldn't be used to back their side in a content dispute. Sometimes there may be a bit of to and fro, and people may occasionally visit the relevant pages of discussion (i.e. talk pages) themselves. But while they may have been effectively forum shopped, I don't think I've ever seen a case where what was said at the RD came in to it (or perhaps only once, I have a vague memory of something). In any event, I don't know that you'll get much support for expanding the header. It used to be a lot longer and while not mentioning this, people felt it was too long and should be shortened so it was more likely to be actually read. And these sort of cases aren't that common. At most, perhaps a template or closure system (which is very rare on the RD). Edit: Just realised you were talking about the proposed warning banner. In that case, perhaps there would be support but I'm not particularly sure since the banner would already seem to imply there's no point coming here for content disputes. Nil Einne (talk) 20:49, 10 June 2013 (UTC)
 * I haven't seen this abuse - the RD can be useful in content disputes, namely, in coming up with data to support a position (or both); we don't hold "!votes" in the question discussions. Wnt (talk) 03:40, 11 June 2013 (UTC)

At one point The RefDesk Guidelines insist that responses to questions fall into one of three categories, Sourced Answers, Requests for clarification, or clarifications. A couple of paragraphs later, it allows that Answers may contain humor. Sadly, many people have misinterpreted this rule (or interpreted other people's behavior) as allowing them to post replies that are wholly jokes, comments or conversation.

Right now there's a culture on the RefDesks that certain parts of the guidelines should be enforced with a zeal that goes beyond the written guidelines, but other parts, like the part I just mentioned, are mere suggestions not to be taken seriously.

Rather than "reforming" the reference desk guidelines, I think there would be a dramatic positive change in the RefDesk if the existing guidelines where more uniformly enforced. APL (talk) 19:40, 19 June 2013 (UTC)