Wikipedia talk:Reference desk/Guidelines

This is the talk page for discussing improvements to the Wikipedia:Reference desk/Guidelines page. For other discussions about the Reference Desks, please go to Wikipedia talk:Reference desk.
Please sign and date your posts using four tildes (~~~~). New discussion topics go at the bottom of the page; click here to start a new topic. New to Wikipedia? Welcome! Frequently asked questions. Old discussion is available at this revision.	Talk page guidelines Be polite Assume good faith No personal attacks Don't bite the newcomers!

Reaching consensus

I am guessing that someone decided to put the template on the main Reference Desk pages in order to draw contributors' attention to the proposed guidelines and elicit responses. I propose that contributors who don't want to respond to each of the points raised on this discussion page give a basic response to the guidelines and whether they oppose them, support them, or support them with small exceptions or caveats. Marco polo 18:58, 5 April 2007 (UTC)

I will start. Although I realize that I have not always adhered to the proposed guidelines, I generally agree with and support them. I am also open to small changes if there is a consensus for small changes. Marco polo 18:58, 5 April 2007 (UTC)

I oppose them because they seem too ridged and long-winded. People come to the Reference Desk for answers to simple questions, not to read a whole page of guidelines and polices. Tweak the existing rules. Think outside the box 19:40, 5 April 2007 (UTC)

Agreeing with this chap. I came to RD to get away from a lot of the policy, bureaucracy, and red tape BS that infests most of Wikipedia. Can't we just keep things simple here without expecting people to commit to memory yet another long-winded guideline document? The short bulleted list that adorns the top of each RD page is totally sufficient, in this writer's opinion. Why do we even feel the need to create a handbook for RD usage? It's just rule creep. -- mattb @ 2007-04-06T06:27Z

You're right the short bulleted list that adorns the top of each RD should be totally sufficient. The reason that this guideline has evolved is that some users prefer to have a more explicit description. This guideline is not meant to replace the short bullets. It is for those who need more rules and lines in the sand than others. Take it or leave it, but some will appreciate it. David D. (Talk) 06:59, 6 April 2007 (UTC)

Responding to mattb's question "Why do we even feel the need to create a handbook for RD usage?" : Matt, how would feel if someone deleted your recent RD post here without providing any explanation ? And when you notice it has been deleted and restore it, it is deleted again and you got a terse message on your talk page telling you it had been deleted because it was unsourced opinion and your restoration was disruptive vandalism ? And if you are unhappy with this and restore it again because, after all, you were only trying to help the questioner, then you get a 24 hour block for edit warring ? The RD guidlines originated as a reaction against incidents just like that. Gandalf61 12:38, 6 April 2007 (UTC)

Well, if someone did that to me I'd call them a jerkface, ignore them, and go about my business. Most people who sit around here know what's appropriate and courteous without having to have rules spelled out in lurid detail. We already have an annoying mass of Wikipedia policies that, in one way or another, apply to WP:RD. Again, I see this as rule creep. However, if enough people find it useful, by all means go for it. I just don't personally find this endeavor to be productive, probably wouldn't ever bother to read the page in its entirety, and definitely wouldn't bother to keep track of the day-to-day disputes over two words of phrasing that these lovely guideline pages tend to attract. -- mattb @ 2007-04-06T14:34Z

When the idea came up for this, I also argued it was instruction creep and unnecessary. We have had contributors doing inappropriate things and claiming they can't know what's appropriate without explicit rules, but I think the worst offenders have been banned or have otherwise moved on. The obvious problem is, the people who need this kind of rule won't really benefit from them- they'll just ruleslawyer and continue on their disruptive way. However I can't say that these guidelines particularly hurt anything, except as you say, it's one more page to maintain and argue over. If people followed the standards we already have, this guideline would be unnecessary. Friday (talk) 14:41, 6 April 2007 (UTC)

Contradiction?

It looks like this paragraph below is saying that the Reference Desk does things that are not its purpose:

"Within our capabilities, we do try to help answer scientific, semantic or historical questions, help explain concepts, and point the questioner to possible sources of solutions or further information."

Now, look at the alleged purposes of the desk:

"The primary purpose of the reference desk is to help the growth and refinement of Wikipedia by encouraging article contributions, additions and improvements.The secondary purpose of the desk is to help the readers by facilitating access to the information contained in Wikipedia."

And there only two purposes listed, so... where does that in the first paragraph fit? A.Z. 19:26, 5 April 2007 (UTC)

The first paragraph doesn't say where these sources of information are. Ideally, they're Wikipedia articles. Friday (talk) 20:43, 5 April 2007 (UTC)

If Wikipedia contained all answers, there would be no need for a Reference Desk nor for us trying to improve Wikipedia nor for any other source of knowledge. I suggest we take out the words "contained in Wikipedia". A.Z. 21:46, 5 April 2007 (UTC)

We are here as a Reference Desk for Wikipedia. way. Many other parts of the guidelines already indicate (or imply) that we will point people to external sites to supplement (and to complement) the information in Wikipedia. Though we are available to everyone, we should not state our secondary purpose to be as broad as "to facilitate access to sources of information" (by your suggestion, "contained in Wikipedia" has been removed). This would imply that we are a universal reference desk, regardless of whether or not the problem relates to Wikipedia in any way. No no. That would certainly be a hindrance to the first purpose. − Twas Now ( talk • contribs • e-mail ) 21:59, 5 April 2007 (UTC)

I would never think of the Reference Desks of Wikipedia as being a "guide to Wikipedia". I have been asking and answering a few questions here for a year now and I never thought of it this way. I thought the Reference Desk provided reference and that's it, no matter whether it had to do with Wikipedia or not. Now, if you admit that many other parts of the guidelines imply that we will point people to external sites, then I think you agree there is an explicit contradiction between the two guidelines there. A.Z. 01:36, 6 April 2007 (UTC)

Changes by Twas Now

I have just made several changes throughout this entire document. Many of these changes clarify the guidelines, either by rectifying ambiguous phrasing (as in the very last bullet), or by removing idioms (which the guidelines themselves discourage!), or by using more precise—and more accurate—language (e.g. proper use of 'and' and 'or'; replacing 'seriousness' with 'sensitivity' and 'diligence'). I have also added the important point that questions should be presented concisely and clearly, rather than long-windedly or vaguely (to the best of their abilities, of course, since we are not all native speakers of English).

I must also acknowledge that User:Friday has a couple of edits in the midst of that diff link I provided above. These were additions to "Removing others' posts" ([1]), and to "What the reference desk is not" ([2]), the latter of which I subsequently consolidated into a very similar guideline. − Twas Now ( talk • contribs • e-mail ) 21:07, 5 April 2007 (UTC)

Forum

So "the reference desk is not a chatroom or forum". I don't know what this means and I wouldn't know how to follow this guideline. Wiktionary says a forum is a place for discussion. Discussion is defined as a "talking with other people". So "the reference desk is not a place for talking with other people!". Yet, people talk to each other on the reference desks.

I know it may seem to some people that I'm just joking or trying to disrupt the discussion. I think I understand more or less what some people mean by "this is not a forum", but for me this sentence does not mean the same thing and in some senses the reference desk is a forum. I am just saying: change the whole phrasing and make it clear what you mean. Using less words does not mean being clearer: many times, using few words makes everything obscure and not understandable. Even the people who support a view of the reference desk different than mine should try to make the phrasing clearer so everyone including myself can know exactly what it means. So, when people violate this policy, people won't argue that "according to Wiktionary, everything you ever wrote on the reference desk is content of a forum".

And sorry for the English mistakes. A.Z. 21:58, 5 April 2007 (UTC)

When we say "forum", we mean a message board—generally a place for idle chat, a place where you are free to express your opinions, talk about whatever, make random posts. Check out the article Internet forum. If we did not have a "no forum" rule, the Ref Desk might turn into this:

A: I have heard a Girl Talk remix of Grizzly Bear's song "Knife". What is the second song that is mixed with "Knife";?

B: It is "Wamp Wamp (What It Do)" by Clipse.

C: Which is off their album Hell Hath No Fury.

Anon 1: That song sucks, you suck!

Anon 2: No, that song is awesome, shut up Anon 1!

Person D: R0X0RZZRZ!!!!!11one

Person C: You are all stupid.

Anonymous 1: No I'm not, your face is stupid.

This cannot help Wikipedia, unless we want to throw into the article "the song has been described as 'R0X0RZZRZ!!!!!11one', but also as a song that 'sucks'. However, the people who said that were both called stupid". In the context of Wikipedia, there would be a lot of unnecessary, non-contributory material being posted to the Ref Desk, and it would be a hassle to sort out the good stuff from the bad. − Twas Now ( talk • contribs • e-mail ) 22:24, 5 April 2007 (UTC)

I did not say you shouldn't have a forum rule, I just said it is not well-explained. I agree that it will be bad if the Reference Desk turns into this thing you wrote, but I don't think it's totally impossible to prevent it from happening without an explicit written rule saying so.

I still think the rule is not well-explained. A.Z. 01:30, 6 April 2007 (UTC)

personal opinion vs. social opinion

I feel that some questions with seek an opinion regarding a large social generalization should be permitted. This should be distinguished from asking a question about PERSONAL opinion. In practice this often already happens and I think it is not harmful as long as it does not turn into a debate. In fact I think it is one of the things that make the reference desks so useful. For instance, let’s say I’m a prospective foreign exchange student who will be going to Germany soon. Not wanting to stand out as an American in Germany, I ask the reference desk what is generally considered normal street clothes in Germany. This sort of question about a fairly well excepted social standard seems like it should be distinguished from “asking for personal opinion.” Any thoughts? S.dedalus 06:39, 6 April 2007 (UTC)

I don't see a reason to worry much about that type of question. It seems less likely to degenerate into an ugly, useless debate than a lot of the questions we see on controversial topics. It's not a reference desk question, per se, but I don't see that it's very harmful either. Friday (talk) 19:01, 6 April 2007 (UTC)

I had written a response to this earlier today, but I somehow lost it. My thought was that "what kind of clothes should I wear in Germany?" is not a question that one would ask a real live reference librarian in your local public library. If one did, then the librarian would probably, at best, redirect you to an appropriate resource. Bring the analogy back here, then, the response would be something like "check the discussion forums at Expedialocitibitz.com". Perhaps, however, the real live ref librarian had just been to Germany, and so can directly answer the question. In that case, the interaction becomes a discussion, or a "chat" if you will. The discussion can continue, but not actually at the reference desk, since that is not the purpose. --LarryMac 19:17, 6 April 2007 (UTC)

I still feel that the Reference desk guidelines should make a clear distinction between these types of questions. After all there is a very great difference between asking “How can abortion not be considered murder.” and asking “What is generally considered to be one of the best representation of film noir? In the first case the discussion is likely to degenerate into an ugly flame war. In the second case the answer is likely to be something like “Well according to the American Film Institute's 100 Years, 100 Movies [3] blah, blah, blah, etc.” I think we have an advantage in some ways over a traditional public library reference desk in that here people can ask questions of several thousand people who may be able to provide unique information that a “traditional” library might not have access to. S.dedalus 21:54, 6 April 2007 (UTC)

In the case of “How can abortion not be considered murder ?”, a strictly factual answer can be given: "Those who don't consider it to be murder don't consider the embryo/fetus to be a live human at that point, based on the lack of a heartbeat, undeveloped brain, etc.". Now this could develop into an argument, but doesn't have to, so the question itself should be allowed. StuRat 15:11, 12 April 2007 (UTC)

Guidelines for dealing with suicidal people?

I think that there should be some guidelines for how best to deal with suicidal individuals. We often get people on the reference desks who feel that there is no one to turn to, and so they ask for help from total strangers on Wikipedia. In such cases responding incorrectly could easily push already desperate people over the edge. Perhaps the reference desk guidelines should include a list of some online resources that editors can recommend to suicidal individuals. A thought on this? S.dedalus 23:04, 6 April 2007 (UTC)

I think the volunteers here are capable of helping suicidal people and the reference desk happens to be a great place to turn to when you feel suicidal. Please DO NOT EVER send those people away: instead, invite them to stay, sit around, create an account and chat and also help on the desk! A.Z. 00:32, 7 April 2007 (UTC)

Rather than inviting them to chat – with people who may be well-meaning but definitely aren't trained or experienced in dealing with suicidal individuals – I think that having a list of resources to which we can guide them would be very helpful. We have to acknowledge that some problems are just beyond the scope and abilities of the Ref Desk, and that it would be unethical and even dangerous for us to try to solve them.

This issue occaisonally comes up elsewhere on Wikipedia, as well. Perhaps we ought to have a centralized list of resources and tools, if one doesn't already exist. TenOfAllTrades(talk) 00:41, 7 April 2007 (UTC)

I'd say revert on sight. 99+%Many of them are trolls, and even if there's the occasional poor soul who isn't, there's little we can do for them. We're ill-equipped and it's outside our scope. However, a list of resources such as Ten suggested would certainly not be a bad idea either. I suggest it be put on the user talk page tho, (even for an IP address) and the ref desk content be reverted. The last thing we want is a bunch of people making stupid jokes about it, on the off chance that it's not just a troll we're dealing with. Friday (talk) 00:45, 7 April 2007 (UTC)

This is a really stupid idea. People do I wish people did not start to make stupid jokes. A.Z. 00:49, 7 April 2007 (UTC)

Unfortunately they do, A.Z. I remember two instances, one last August or September, one more recent one. Both times frosty advice on how to actually perform the act of suicide as well as cynical comments (and stupid jokes) were added. Both original posts were made by registered users, and not for one second, did I assume they were trolls, nevertheless, they were treated in an incredibly unhelpful and immature way. In one case I tried contacting the poster per e-mail, but never received a reply, though the user continued to edit a few days later. Perhaps s/he was embarassed, and, if I'm not mistaken, the post was eventually overseen (meaning removed in a way that even makes it impossible to view the edit history). I'm at a loss what to recommend here, and would welcome more thoughts on this topic. ---Sluzzelin talk 01:54, 7 April 2007 (UTC)

Sluzzelin, in cases where editors are making jokes about the suicide message I think the only remedy is for other editors to be very diligent about reverting such replies. Perhaps, such replies could also be treated as especially saver cases of vandalism.S.dedalus 02:07, 7 April 2007 (UTC)

Jokes about suicidal people that come here asking for help should be reverted immediately. A.Z. 02:17, 7 April 2007 (UTC)

Ah, but they were, and at least one person saw this as an infringement of our "freedom of speech" and wished to reinstate them, and a discussion regarding "unilateral" censorship and policies ensued (not kidding). This is an extreme example, I'm glad we both agree on how it is best dealt with. I wish everyone agreed. ---Sluzzelin talk 02:24, 7 April 2007 (UTC)

Friday, I must respectfully disagree with your assertion that most suicide messages should be reverted as trolling. I remember reading about one of the few surviving Golden Gate Bridge “jumpers.” He said that he was standing there reconsidering whether to jump or not, when a tourist asked him to take her photo. He thought, there really is no one who cares, so he jumped. I think reverting a ”cry for help” will almost certainly result in an obituary. It is far better to needlessly respond to a few trolls than accidently fuel a suicide attempt. As for putting replies on the person’s user talk page, I seriously doubt that people who are not experienced with Wikipedia would think to check their talk page. The “centralized list of resources and tools” sounds like a good idea to me. Perhaps under Wikipedia: Dealing with suicidal individuals. I would suggest [4] would be a good resource to start with. Is this a workable idea, and are there any other editors who would like to help write this? S.dedalus 01:52, 7 April 2007 (UTC)

But with your example you've just underlined the weakness of this whole idea. With suicidal people, there is no formula correlating the degree of goodwill of the interlocutor equalling the likelihood of the jumper not jumping. The guy on the bridge chose to take that woman's request as an indication of the state of the world; not surprising, considering his frame of mind. But the woman could just as easily have been thinking that by getting him down off the ledge to get her camera, she could grab him or call for help. It's easy to look at that situation and say, 'Oh, she obviously said the wrong thing.' But if he'd gotten down to take her camera and she'd grabbed him or done whatever, it would have been the right thing. It was only judged the wrong thing to say because he jumped. But HE jumped. HE chose to interpret her words the way he did, and HE is the cause of his own attempted suicide. She could have been completely well-meaning but simply done the wrong thing. Or maybe there was no right thing. I have a friend who was talked down from suicide by someone who said, when my friend asked if she should do it, 'You can do absolutely whatever you want. It's YOUR choice.' Her response was, 'I choose to stay for a while.' Her interpretation of all the well-meaning help she'd been getting from others had just made her feel powerless and desperate. That's my point; when someone says, 'I am planning to kill myself, any ideas?' We can never know if saying, 'Sure, cyanide.' will make the person go out and buy cyanide, or whether saying, 'Sure, counselling.' will make the person go out and buy cyanide. Anchoress 01:01, 14 April 2007 (UTC)

The phone number in your link is only for the United States. A.Z. 02:44, 7 April 2007 (UTC)

That’s true , however I would like to include a number of recourses in the proposed guideline page. Do you have any suggestions for international resources? S.dedalus 03:35, 7 April 2007 (UTC)

A person asked a question about chat-based online counseling on the miscellaneous reference desk on March 29, but I can't find it in the archives. A.Z. 03:57, 7 April 2007 (UTC)

Is this it? S.dedalus 04:13, 7 April 2007 (UTC)

This is the useful link from that online counseling discussion. S.dedalus 04:33, 7 April 2007 (UTC)

The handy thing is- we don't have to guess which ones are trolling. Regardless of which are which, replying usefully on user talk and discouraging further comment on the ref desk is a good thing, don't you think? My main concern is for the ones that aren't trolling- we don't want to make their bad situation worse. Altho, come to think of it.. even if the original post is trolling, it's still not good to encourage a bunch of suicide-related humor. A real person in a bad situation could read that, too. Friday (talk) 02:09, 7 April 2007 (UTC)

I think there should be comments and advice on the reference desk. I like that idea. At least until someone creates a little Wikimedia project to help all the suicidal people who come here. A.Z. 02:11, 7 April 2007 (UTC)

The problem is, Friday, that I don’t think many IP address users even know they have a talk page. Perhaps we could post replies on the persons talk page, but then put a note on the reference desk saying that we have responded to their question on their personal discussion page. We could then include a link to that page. S.dedalus 04:19, 7 April 2007 (UTC)

Maybe we should develop a standard response for suicidal people. Anyone who claims to be so (whether they're perceived to be a troll or not) could then be directed to that page, and any other comment (whether humorous, helpful or indifferent) would be banned. What would go in that response is of course a matter of discussion. How the ban on other comment could be enforced is also a challenge. As many have noted, some of these people really are suicidal, and the worst possible result would be for Wikipedia to unwittingly (or wittingly) assist a person to kill themselves. It's one thing to be a reference desk for all comers, but sometimes we do ourselves a disservice by taking that to its extreme, by simply telling them what they want to know, eg. a non-messy way of killing themselves. In my view, that is simply not on. JackofOz 04:31, 7 April 2007 (UTC)

Using my bad English, that I wish so much was better right now, let me just say that, to some people, Wikipedia is more than an encyclopedia. To some people, Wikipedia also means hope that the truth will reach a lot of people and that everyone in the world will know what are logical fallacies and cognitive biases and will avoid them, and everyone will find out that gay people have contributed to mankind the same way straight people have. There are people here who have been bullied and harrassed in the real world and, in the place that at the start was just an encyclopedia, ended up founding good people that would help them feel good with themselves. Respectful, intelligent people that care about other people and that, as themselves, want the truth to reach everyone. I know this is an encyclopedia. I know it is not supposed to be for helping suicidal people. I do understand now Friday's worries and I didn't before. I really understand him now because people have made jokes about me and making those jokes can really lead to an obituary. The fact is that... I don't know to where on the Internet you could send all those MANY suicidal people who come here for one reason or another. Who come here to write either on the reference desk or in the articles. Who explicitly say they are suicidal or not. People are just not nice to each other in the real world and it seems that the community of Wikipedia somehow became a little more civilized than the average... This WILL attract a lot of people searching for help. Some of them will press the "save page" button, some of them will say "how stupid to turn to an online wiki encyclopedia" and jump off a bridge. Just please realize how important it is what you are deciding right now. You are actually deciding what will become of some people's lives. Even if they were few people, it would matter a lot. But there are A LOT of people who come here for that, and the number is only likely to increase. A.Z. 07:21, 7 April 2007 (UTC)

I have a concern that adding something about this to the guidelines may have a WP:BEANS-like effect on trolls.  --Lambiam^Talk 08:07, 7 April 2007 (UTC)

Yeah, I'm not only talking about the guidelines of the reference desk here. I actually think Wikimedia should start to draft a policy for this sort of thing. A.Z. 08:11, 7 April 2007 (UTC)

Thank you A.Z. I totally agree. Is it time to bring this up on WP:VPP? S.dedalus 19:06, 7 April 2007 (UTC)

Yes, by all means, if no one has brought it up there before! I am not a native speaker of English, so I wouldn't be the right person to bring it up there, since it's such an important matter and I wouldn't be able to explain everything that has to be said well enough. You seem to care about the suicidal people who come here and to understand the importance of such a policy. However, I was more thinking of Wikimedia, since suicidal people may go and cry for help on the other projects as well. A.Z. 20:50, 7 April 2007 (UTC)

Out of Wikipedia's original scope, but how about setting up a 'Special Help Desk' for individuals who are indeed desperate, and even suicidal? This can be manned by registered members, needing a password to reply (to stop the immature comments and to better know who is talking to whom). These registered members could be equipped with the knowledge of the best places to go for 'visitors', and could use them in accordance with the specifics of the problem that each visitor has, should preliminary talking fail. I was chairman for an NPO that specialized in family counselling in Japan, so I would certainly do it, if asked.ScouseMouse - スカウサーUK! 00:48, 8 April 2007 (UTC)

Unfortunately I doubt that what you propose would go over very well with the Wikipedia powers that be. It would skirt dangerously close to being considered a counseling service and in the US it is probably illegal for an organization of people with no medical background to set up a service like that. At the very least it would probably leave Wikipedia open to a torrent of lawsuits. S.dedalus 01:36, 8 April 2007 (UTC)

While it is an interesting idea, I note that such a project – operating within the confines of the Wikipedia project – isn't technically feasible at the moment. Currently we can limit editing of pages to admins only (page protection), or to registered users only (semiprotection); I don't think we can implement the sort of fine-grained control described.

I suspect that we're probably best to confine ourselves to the response that a real-life reference desk would offer to an individual contemplating suicide. We should provide a list of resources and qualified professionals who can help the person in need. Leaving aside the issues of civil and criminal liability that S.dedalus (quite reasonably) mentions, I'm just not comfortable with an encyclopedia project (remember, that's what we're mandated to do) assuming this sort of role. TenOfAllTrades(talk) 03:42, 8 April 2007 (UTC)

As per Wikipedia:How to create policy I am creating a policy draft of the proposal under the tentative title Wikipedia:Responding to suicidal individuals. If you are so inclined, please help me draft this policy as I have no experience writing this kind of Wikipedia page. When the page looks okay we should probably post the proposal on Wikipedia:Village pump (policy) and Wikipedia:Requests_for_comment/Policies. From the ideas suggested in this discussion I’ve started to draft the policy around the following points: how and how NOT to respond to a suicidal individual’s post, the stipulation that editors should not attempt to provide medical advice (e.g. "you’re suffering from depression"), a list of public online or telephone recourses where suicidal individuals can seek help, and the requirement that editors should try to respond to suicidal messages on the user’s personal talk page, but also leave a message on the reference desk linking the person to their talk page. Anything I missed? S.dedalus 06:19, 8 April 2007 (UTC)

Two of the worst things we could do to a person who claims to be suicidal are to delete their post or give them a stock reply. Being ignored or getting the equivalent of a "form letter" could put them literally over the edge. I believe standard advice to those who encounter suicidal individuals is to keep them talking, then try to find something meaningful in their lives they can focus on. StuRat 01:15, 10 April 2007 (UTC)

I agree about not ignoring them. I'm somewhat concerned about us acting as personal counsellors, however. Just as we're not in the business of providing legal or medical advice, I don't think we should be in the business of talking a suicidal person through their crisis. There are thousands of places that are equipped to do precisely that, and we should tell the person about them. If they're serious about wanting help with suicidal thoughts, they'll be interested in finding someone who's skilled in that area. Wikipedia is not such a place. Some previous examples of the way we've dealt with suicidal people here were models of what not to do, with bickering amongst users becoming the main focus of attention. The person ended up becoming incidental to the discussion, which was tantamount to ignoring them. We need to assume they're for real, and focus on them, not each other. But does that mean we should assume that Wikipedia is the one and only place they've gone to for help, and assume total responsibility for their lives? Sometimes the best advice we can give is to direct them to where the best advice can be found, not try to make it up as we go along. That direction would be one of the things I'd definitely include in the "form letter" (an expression I would never have used in this context). If they really want help, they'll read it, and not just latch on to the Reference Desk and make it solely responsible for the answers to all their problems. JackofOz 01:39, 10 April 2007 (UTC)

I agree with Jack's response. As for Wikipedia not being the one and only place where the best advice can be found, I recently read a related discussion somewhere else on Wikipedia, and one editor, justifiably in my opinion, pointed out that suicide threats here were a form of emotional blackmailing. Maybe Friday was thinking of something similar too. I'm certainly against ridicule or (even potentially) pushing over the edge, but our responsibility should end after pointing users to sources with the professional expertise necessary to help people in these desperate and dark moments of their lives. ---Sluzzelin talk 03:13, 10 April 2007 (UTC)

I agree of course that Wikipedia cannot provide a counseling service. I mentioned this above in my reply to ScouseMouse. All I am proposing in this policy is that Wikipedia respond to any suicidal individual the same way any good citizen would respond to a person on a ledge. In other wards treat them in a way that will not increase their likelihood of jumping, and then point them to a source of help. Unfortunately Wikipedia has often failed miserably at this in the past. S.dedalus 06:04, 10 April 2007 (UTC)

Of course we should treat them in a way that will not increase their likelihood of jumping. But the analogy does not quite hold water. Wikipedia is, in some respects, a many-headed hydra, speaking with just as many voices. Who, in real life, would want a hydra with them on the ledge while they contemplate jumping? They need one person. The Ref Desk cannot hope to speak with one voice, and should not try to. We are inherently many. That is why we have failed at this in the past: the lesson, imho, is that we should never have even tried to help, except for your suggestion that we point them to a source of help, which might be achieved inter alia by developing a resource that becomes a standard response. JackofOz 06:46, 10 April 2007 (UTC)

"treat them in a way that will not increase their likelihood of jumping" is subjective. Suicide threats should be reported to authorities and should not be responded to by any means. In an ideal scenario, I'd like to see them be reverted without discussion. It is not our job to be psychologists, and we may do something unforeseen that actually encourages them to commit suicide. God knows some idiot kid will stumble upon the suicide threat and start blabbering about U SHURLDNT KIL URSELF LOLZ. "I believe standard advice to those who encounter suicidal individuals is to keep them talking, then try to find something meaningful in their lives they can focus on." - StuRat, what you've heard of "standard advice" doesn't cut it in life-or-death situations that often have vastly different causes: general depression, mental illness, chronic pain. The only "help" offered should be a list of resources (phone numbers, websites, etc) and no attempt at trying to, as with any other health issue on the refdesk, diagnose or treat anything. I think JackOfOz is on to something here - standardized template with resources. I'd prefer that it be reverted, then added to the user or IP's talk page so that we don't have newbies coming in and trying to offer counseling that may be harmful. -Wooty Woot? contribs 03:54, 16 April 2007 (UTC)

Wooty, this discussion is now continuing over at Wikipedia talk:Responding to suicidal individuals. It's moved on somewhat since last week. Please join us over there. Cheers JackofOz 13:02, 16 April 2007 (UTC)

---

Please continue the discussion at Wikipedia talk:Responding to suicidal individuals.  --Lambiam^Talk 07:19, 16 April 2007 (UTC)

Illegal Practices?

Should we enforce some sort of ban on questions about perpetrating illegal acts (e.g. murder, rape, etc.)? I was going to add this directly to the guidelines, but I'm not sure about other people. bibliomaniac15 01:04, 7 April 2007 (UTC)

To me, it's a tossup whether we want to mention this or not. It might be asking for trouble. It's certainly the sort of thing that would make a question more likely to be seen as trolling, though. Friday (talk) 01:25, 7 April 2007 (UTC)

Perhaps we could create a template for use in this situation. Something to the effect of “Thank you for visiting Wikipedia! However, we cannot provide you with advice on the best way to rape and murder your Aunt Sally. Sorry.” S.dedalus 01:58, 7 April 2007 (UTC)

@Bibliomaniac15 unfortunately, instituting (let alone actively enforcing) such a ban could entail some undesirable consequences. For example, suppose the ban is indeed instituted, enforced with 100% accuracy and zero 'false-positive' removal of content. That would arguably be inconsistent with Wikipedia:General_disclaimers, because this track record of banning content could be argued as a form of "advice" about what is and is not unlawful (yes I know, lame argument but it has been made before). Someone could claim they detrimentally relied on the "ban system" to protect them from bad stuff, and they could complain they were harmed if one bannable question actually got through. dr.ef.tymac 21:32, 7 April 2007 (UTC)

I suppose we could just put a disclaimer instead of enforce a ban. bibliomaniac15 01:30, 8 April 2007 (UTC)

I wouldn't object to a ban on advice on how to commit an illegal act, but only if those activities are illegal everywhere. Prostitution and marijuana are legal in some places, for example. StuRat 01:02, 10 April 2007 (UTC)

Netspeak

I would like to propose that the use of Chatroom language be discouraged and kept to a minimum, especially when posting queries. Reading queries that begin with 'OMG', or 'WTF', and having 'LOL' or 'LMAO' and stuff like that makes me not want to answer them, and I am sure I am not alone in this. I may not be the Brain of Britain, but including language like this reduces the chances of a poster getting an answer from a serious professional. ScouseMouse - スカウサーUK! 23:29, 7 April 2007 (UTC)

I haven't noticed this as a serious problem. Is there a specific desk on which you see this?  --Lambiam^Talk 00:47, 8 April 2007 (UTC)

I don't think of it as a serious problem, but it just seems to me that the impression is that we are all just having a laugh. Many times we are, of course, as we are all contributing to something that we like, but Chatspeak makes me feel like I'm dealing with a teenager who isn't really serious about what he/she's asking. Sometimes this is fine (that is why I say 'should be kept to a minimum' and not 'eradicated'), but sometimes.... I have only noticed it on two of the desks I usually answer on: Language and Misc.ScouseMouse - スカウサーUK! 00:56, 8 April 2007 (UTC)

Givnan, how do you propose that chatroom language be discouraged when we can't even get folks to type four tildes? ~ hydnjo talk  18:53, 8 April 2007 (UTC)

Good question. As we've all seen from experience, getting reasonable editors to agree that a certain thing should or should not be done is the easy part. Getting unreasonable editors to actually do or not do what's been agreed on is the hard part. Friday (talk) 19:34, 8 April 2007 (UTC)

Givnan, I recall the first times that I went on-line and thinking that it was cool to say "i" and "OMG!" and "lol" and other "chatspeak". My own experience suggests that the example of others was extremely influential in realizing that that dumbed down language wasn't "cool" at all. If "chatspeak" is a problem for you then just skip over to another inquiry, don't let it bother you. Let me illustrate in another way, do you really think that giving the middle finger salute to a rude driver will decrease the number of rude drivers? ~ hydnjo talk  20:37, 8 April 2007 (UTC)

We had specifically talked about the use of the term "OP" before, and agreed not to use it, as many OPs won't know what an "OP" is. I rarely use "LOL" myself, instead preferring a smiley. The one place where I still use "LOL" is where that is my only response, as "Ha ha" just doesn't seem to work, and a smiley all alone is confusing. StuRat 00:57, 10 April 2007 (UTC)

Do your own homework

This came to my attention recently when one user responded to a question by drawing attention to this point and so not addressing the quesiton, and then shortly after another user 'assumed good faith' and answered the question anyway. I understand the reason this point may be included at the top of the reference desk pages but it's often hard to judge if a question is a homework question, and, more importantly, a well answered question, whether it be a homework question or not, is surely more beneficial than no answer at all. I also think this is slightly at odds with 'assume good faith' part of answering the question. Surely anyone with the cheek, for want of a better word, to blatantly ask what seems to be a homework question isn't going to be put off by this notice? So my question is, is this piece of advice really useful or necessary in the 'how to ask a question section'? Johnnykimble 00:00, 10 April 2007 (UTC)

Some questions are in a gray area (they might be homework and might not be), but others are entirely clear. Obvious homework: "Please list 5 arguments in support of X and 3 arguments opposed, with supporting evidence for each claim." Not so obvious: "What are the orbital characteristics of Pluto ?". StuRat 00:50, 10 April 2007 (UTC)

well its kinda hard to find out which are homework questions and which arnt but the most obvious signs of one are when someone ask somethink like; what is the tempreture of liquid nitrogen and how cold is it,in ferinhite, celcius, kelvin. Obvious homework question for chemistry as this is what they focus on. however there are some people that just want to know this for the heck of it. but i see most of the times that people who post homework questions sound more or less like a homework question. Maverick423 21:07, 10 April 2007 (UTC)

I don't see this in contention with AGF at all. The purpose of the reference desk is not to do peoples' homework for them, and if there is reason enough to believe that a question is just that, I see no problem in pointing out the rule. Sure it won't stop people from asking, but it gives us something to point to. -- mattb @ 2007-04-10T23:35Z

so your saying if it sounds like a homework question then we should treat it as such and point out the rule but at the same time assume good faith and help them out anyways? or just point out the rule and ignore them? and ya your right no matter what we do we will never stop people from asking for help with homework. Maverick423 13:52, 11 April 2007 (UTC)

There is nothing wrong (unless the homework task is to find pointers) with giving some pointers directing the questioners to the approach to be taken. For example, if the question is to solve 2x² + 7x + 3 for x, we can point them to the article Quadratic equation. If they come with an answer, we can say whether it looks right, and if they are stuck, we might give a hint to help them take the barrier. What we must not do, is do the homework for them, so that they can just copy and paste the answers. Theoretically it is possible that a clever questioner succeeds in disguising a homework assignment so well that it is no longer recognizable as such. But apparently most are not clever or shrewd enough for that; almost all questions that could be homework are also obviously homework questions. A real giveaway, for example, is those that take the form of a statement followed by "Discuss." I think we should try to avoid suggesting that the questioner is a shirker unless the question is obviously homework (but it is not bad, in my opinion, to ask: "This is not homework, is it?"), but in cases of doubt many honest questioners will be happy enough with pointers in the right direction, and if not they can always come back with questions for further elucidation.  --Lambiam^Talk 16:40, 11 April 2007 (UTC)

Lambian is exactly right. If we want to assume good faith, or not bite the newbies, the thing to do is just not scold them for asking an obvious homework question. But assuming good faith does not have to mean being a sucker and doing people's homework for them! (As happens, it seems to me, with distressing frequency.)

If someone asks what the temperature of liquid nitrogen is, and it's an obvious homework question, we can provide a link to Nitrogen, which just happens to be a 100% appropriate answer even if it wasn't a homework question. Likewise, if someone asks about the orbital characteristics of Pluto, and even if we're not sure whether it's a homework question or not, we can simply point them at the Pluto article, which is again a 100% appropriate answer whether or not it was a homework question.

Now, if someone is really adamant about not giving a lazy slacker any help whatsoever, they might feel that even giving them a ink is more than they deserve. In that case, you can either (a) decline to answer, or (b) console yourself that (like it or not) knowing where to look for information is a skill that not all students have, so providing them with a link is a legitimate piece of help, since they still have to find the exact information on that page, which is still potentially a piece of work for them. (Or, (c), console yourself with the fact that someone who is so lazy as to be cutting and pasting their homework questions onto a Wikipedia Reference Desk page, and expecting to be spoon-fed a complete answer, may well be too lazy to even follow the proffered link.) —Steve Summit (talk) 15:01, 14 April 2007 (UTC)

"Too long"

In response to User:Radiant! tagging this page with {{toolong}}, I think we should summarize the medical disclaimer. It should be a short paragraph, simply emphasizing that we do not give medical advice (diagnosis, prognosis, or suggested treatment), and then directing the reader to our "official" medical disclaimer at Wikipedia:Medical disclaimer.

Since that page is somewhat short, we might also want to transfer some of information from this page (Wikipedia:Reference desk/guidelines#Medical advice guidelines) to the medical advice page (Wikipedia:Medical disclaimer). I would do that myself, but the page is protected. − Twas Now ( talk • contribs • e-mail ) 20:00, 11 April 2007 (UTC)

I moved the medical advice section to a subpage, and removed the entire "Inappropriate responses" section. I thought it best that that section should be removed for the following reasons:

1. It was to detailed, the guideline should not attempt to list what "may" or "may not" be removed, or what is or is not appropriate. It's a guideline, and should merely describe in general terms the desired content on the desk.
2. We're unlikely to get any consensus on a detailed list of what could be subject to removal, or on the removal process itself. If we are to eventually put a guideline page in place, then let's focus on what we can get consensus on.
3. Inappropriate content is removed from Wikipedia every day, and we already have guidelines and policies in place which help us decide what content is appropriate and how to remove what is not.

—eric 17:40, 12 April 2007 (UTC)

Good decision. Who likes a CREEP? − Twas Now ( talk • contribs • e-mail ) 09:24, 15 April 2007 (UTC)

I'm really very likable in person.—eric 20:55, 15 April 2007 (UTC)

Removing others' posts

The current version starts with: "Incorrect information or advice that may be dangerous if acted upon should be immediately removed and discussed on the talk page; it should not be restored unless the outcome of the discussion is that the information was correct (or at least not dangerous) after all."

As it stands, the guideline seems to give carte blanche for anyone who disagrees with anything anyone else says to simply remove it, on the basis of it being "incorrect information". This section is essentially about dangerous information, and it should focus on that subject alone. I suggest the highlighted words be removed from the opening sentence. JackofOz 03:30, 12 April 2007 (UTC)

The wording is ambiguous, but I assumed the intention was: Information or advice that is incorrect and may be dangerous if acted upon.  --Lambiam^Talk 13:59, 12 April 2007 (UTC)

Allegedly "dangerous" responses

I oppose this entire item. After all, many things are somewhat dangerous, like driving a car. Should we remove all advice related to cars on the ground that "cars are dangerous, and we don't want to encourage dangerous activities". You might think I'm taking something relatively safe and making it sound dangerous to make a point, but let me point out that info I provided with sources saying that the elemental liquid form of mercury wasn't very dangerous (other forms of mercury are) was removed as being "dangerous advice" while far more people die in car accidents each year than of mercury poisoning. In fact, I don't believe anyone has ever died from exposure to elemental mercury, but only from other forms. So, my point is that this policy will be applied very unevenly to remove comments on things which are incorrectly perceived to be dangerous (like elemental mercury) while leaving in comments about things which are incorrectly perceived to be safe (like cars). Another example is that, in the US, nuclear power is perceived to be dangerous, while coal burning power plants are perceived to be safe, even though coal burning plants release far more pollution, and even more radiation, into the environment (the radiation comes from uranium in the coal). So, should anyone from the US feel free to remove advice on moving to nuclear power on the grounds that this is "dangerous advice" ? StuRat 14:56, 12 April 2007 (UTC)

The assumption is that participants at the desk will bear in mind all the relevant guidelines for conduct on Wikipedia (and indeed, in real life), particularly WP:DICK, WP:POINT, Wikipedia:Use common sense, and WP:CIVIL. I presume that the subsequent discussion on WT:RD would reveal if people were abusing the 'remove dangerous advice' provision. I suspect that it is meant mostly to target bad medical advice. ("I have a toothache, what should I do?" "My grandmother told me to soak some solder in vinegar overnight, and then drink the liquid." "Apply a dab of turpentine directly to the affected area.")

Though I agree with StuRat to the extent that he raises a point I – and others – brought up a long time ago: attempting to create a policy which exhaustively and finally addresses every possible case will always fail—there's always something we didn't think of, and wikilawyers will (if allowed) always abuse or misinterpret written guidelines in the most twisted and perverse way possible.

And StuRat, please let go of the mercury dispute. First, you misstate your position—you stated flatly that handling liquid mercury "isn't dangerous"; you didn't qualify your answer on the Desk as you did here (by saying "wasn't very dangerous"). Second, elemental mercury has caused a number of deaths, as can easily be determined through a quick Google search: keywords: "elemental mercury exposure" death. (See for example case studies [5], [6], etc.) In a poorly ventilated space, mercury vapour can easily accumulate to dangerous levels. More often, of course, it causes neurological impairment or organ damage; it's a very slow toxin, and most people seek medical attention when they start to show symptoms. TenOfAllTrades(talk) 19:47, 12 April 2007 (UTC)

Neither of those examples you gave involved death. In the first case they had a large quantity (21 pounds) of mercury which they threw at each other and tried to light on fire. Had they tried to set aluminum, plastics, or many other materials on fire, and inhaled the fumes, they would also have been taken ill. I stand by my assertion that far more people die from car accidents than from elemental mercury poisoning, and that treating elemental mercury as if it is far more dangerous than cars is therefore not justified. Allowing people to censor evidence of the safety of things which are perceived, incorrectly, to be highly dangerous cuts off the discussion and does not allow it to reach a logical conclusion, leaving people to forever wallow in their own superstitions. If you don't like the mercury example, I also provided the nuclear power example, and we can use that as a basis for discussion, instead. StuRat 03:18, 14 April 2007 (UTC)

My mistake; I linked to the wrong case study. Here are some deaths caused by elemental mercury: [7]. In any case, you well know that a comparison to cars is specious. Many more people have access to or interact with automobiles than with mercury. The average person receives training (formal and informal) from an early age on how to behave around and in motorized vehicles. People who drive are formally trained and tested, their conduct is extensively regulated by government, and their behaviour is monitored by police. For the vast majority of the population, no formal or informal training in handling elemental mercury is ever provided (witness the bloody stupid things some people have done in the linked case studies). Flatly telling people who lack appropriate training and background that handling elemental mercury "isn't dangerous" is irresponsible. Is it less dangerous than organomercury compounds? Yep. Is it less dangerous than an automobile? To someone who isn't trained to safely handle an autombile, certainly. Is it harmless, particularly in the hands of someone without proper training? Nope.

Leaving that aside, however, I'm not sure what sort of discussion you're looking for on this; I thought that my position was clear, and if you would like clarification of something specific feel free to ask. The nuclear power example you bring up is easily answered—removing the discussion would be absurd, and someone who tried to cite this clause to do it would be censured. Repeated abuses of that sort would be met with blocks for disruptive behaviour. Wikilawyering tends to wear very thin very quickly, and individuals who feel they can rely on perverse readings of policy to support obnoxious behaviour are generally rapidly disabused of such notions. TenOfAllTrades(talk) 04:10, 14 April 2007 (UTC)

Cars are still highly dangerous to people who have gone through all the training the government offers. Somewhere around 43,000 people die from car accidents each year in the US, which is perhaps 1 in every 7,000 people who ride in a car each year in the US. Over a lifetime, the death rate from cars might be around 1 in 100. I doubt if anywhere near 1 in 100 people who handle mercury die from it. Here is a good site which compares risks of various items (although not elemental mercury): [8]. They actually put the lifetime risk of death from a car at 1 in 70, and try to show how people have the same type of bias against certain types of risks, like chemical and nuclear, while far more serious risks are accepted as "safe". StuRat 05:20, 14 April 2007 (UTC)

Many more people die from slipping in the bathroom than from trying to imitate Spiderman by jumping off the Empire State Building. Nevertheless, it would be irresponsible to suggest to kids that statistically speaking the latter is relatively safe.  --Lambiam^Talk 13:13, 14 April 2007 (UTC)

But you failed to look at the ratio of people who tried to those who died. If 100% of people who jump off the Empire State building die as a result and only 0.001% of people who use the bathroom die, then we can safely say that jumping off the Empire State building is more dangerous. I've provided a link showing that 1 in 70 people die from car accidents, which makes it a major risk. I don't have the figures, but, as I said, I suspect that far less than 1 in 70 people who handle elemental mercury die as a result, making it considerably safer than cars. StuRat 15:47, 14 April 2007 (UTC)

The link you provided is not a reliable source; it says "the risk expression may look like ...". I'm convinced that 1 in 70 is much too high even for the U.S., and definitely for the world's population. In 2005, there were V = 5361 traffic deaths in Germany (see de:Verkehrsunfall#Unfallstatistik Deutschland). Assuming a more-or-less stationary population of N = 80 million, life expectancy of L = 75 years and chance of dying from a traffic accident, this gives for Germany a chance of VL/N, or about 1 in 200, of dying from a traffic accident. In countries that, unlike Germany, have a speed limit, this is in most cases probably even less. More importantly, you forget that many people get in their cars every day and make several trips. If as many people handled elemental mercury every day several times, and the stats would still show negligible amounts of risk, yes, then you were right. But we don't have such statistics, and I don't believe for one second that that is what they would show if they existed. Much fewer than 1 in 70 people die of those who once take a car ride and then never again. You are making the very mistake that you are trying to pin on me.  --Lambiam^Talk 19:24, 14 April 2007 (UTC)

The death rate from car accidents is probably much higher in third world countries, however, where roads and vehicles are in bad shape, there is little enforcement of traffic regulations, etc. And it is entirely reasonable to compare lifetime risk of death from handling mercury with lifetime risk of death from driving cars. For example, if one was to choose between a career as a truck driver and a chemist working with mercury, such statistics would be important in making the decision. StuRat 04:21, 16 April 2007 (UTC)

Did you have anything to say about the topic of this thread? TenOfAllTrades(talk) 06:10, 14 April 2007 (UTC)

I've had quite a bit to say about it, and I've now said it. In short, allowing unilateral deletion of allegedly "dangerous" advice is a bad idea, because people's perceptions of relative risk is extraordinarily poor, as you've demonstrated in this thread and as was covered in the link I provided. StuRat 15:58, 14 April 2007 (UTC)

You also gave some rather questionable safety advice under Wikipedia:Reference_desk/Science#Dry_ice_safety, such as saying that it's safe to plunge your bare hands into liquid nitrogen, but I won't go and remove it unilaterally, or call you stupid for having done so. I would hope I could expect the same courtesy from you. StuRat 19:34, 15 April 2007 (UTC)

Actually, I said that it "isn't all that bad" for "modest lengths of time". I didn't make the unqualified statement that it was "safe". I also noted the mechanism that provides (some) protection in the circumstance, and indicated clearly that contact with most other materials at liquid nitrogen temperatures could cause rapid injury. Please be precise when you comment on my remarks. TenOfAllTrades(talk) 20:51, 15 April 2007 (UTC)

Someone lacking expertise who tried that could easily injure themselves, however. Thus, it was "dangerous" for you to say that. StuRat 04:14, 16 April 2007 (UTC)

Somebody lacking expertise could hurt themselves with liquid nitrogen or dry ice period. Should we tell them that if they do not know what they are doing to stay far away from either, and not say anything else? [Mαc Δαvιs] ❖ 19:59, 16 April 2007 (UTC)

If this policy becomes official, then that's just what we would be forced to do. This is one example of why I oppose this policy. StuRat 01:52, 17 April 2007 (UTC)

Which policy are you talking about? Is there a proposed policy that prohibits providing correct and verifiable information about the risks of certain behaviours? Where is it being discussed?  --Lambiam^Talk 14:39, 17 April 2007 (UTC)

As near as I can tell, the proposal is "advice that may be dangerous if acted upon (in the opinion of the deletor) should be immediately removed". Under this definition, advising people that plunging your hands into liquid nitrogen "isn't all that bad for modest lengths of time" seems to qualify for immediate, unilateral deletion. StuRat 19:41, 17 April 2007 (UTC)

Debates and controversy

I agree with most of the what the reference desk is not-section, but I don't think that last entry about controversy and debate should be included specifically. It might send the wrong message and make people think they cannot ask questions about it. It falls square in the "not a soapbox-rule" we already have established. - Mgm|^(talk) 10:23, 12 April 2007 (UTC)

Hmm, I liked that bit. To me it made it clear that the answer we should give is not our own opinions, but rather to pointers to information about the debates on these topics. But, yeah, it's a bit redundant. Merging might be good but IMO we take care to retain the full meaning. Friday (talk) 14:08, 12 April 2007 (UTC)

The answer we give may not be our personal opinion on the subject of the question, but certainly whatever reference we cite will be our choice based on our own personal opinion of which references are credible or not. If you assume that the people who ask questions here don't know how to separate the good from the bad when it comes to our explicit personal opinions, then you can't possibly defend that they have the ability to decide on their own which references were good and which ones were bad. A.Z. 02:51, 16 April 2007 (UTC)

This is a good point, POV is inevitable, even when using reliable sources. This reminds me of a US news station which was accused of having a liberal bias for broadcasting names, pics, and brief bios of every soldier killed in Iraq. They said "we aren't biased, everything we say is true and verifiable". But then it was pointed out that they choose to list soldiers killed in Iraq, but didn't list civilians killed by Saddam, or people killed in any other way, like accidents or disease. They chose to focus on that one aspect exclusively, and in that choice lies the POV. StuRat 02:27, 24 April 2007 (UTC)

Re: Use external links from answers as sources to improve our articles

Any chance we could get a plug in for Wikipedia:WikiProject Reference Desk Article Collaboration somewhere in this section? I don't wish to spam it though, so I'll leave it up to someone else to decide. Rockpocket 04:25, 16 April 2007 (UTC)

WP:RDAC took an early break and could use some refreshments. I like the suggestion, but don't wish to spam the guidelines either, plus it's too early to write coherently. ---Sluzzelin talk 08:08, 17 April 2007 (UTC)

Yes, I was thinking a bit of a revamp might be in order. Rockpocket 09:15, 17 April 2007 (UTC)

Misleading sarcasm or irony

The recent fracas between myself and another contributor over what (in my view) was a 'joke gone bad' highlights the problems with understanding when something said in an online communication is intended as a joke or not. Whilst I agree that an outright ban on humor would be a bad idea, I believe we do need to have somewhat stronger guidelines than we currently do.

The problem comes about when a respondant answers a question ironically or with sarcasm and the result is harmless when interpreted as a joke - but disasterously bad advice when not construed as a joke. This kind of thing is dangerous because:

1. Our readership may not have a strong command of the english language and be therefore unable to realise that this is a joke.
2. Even with a strong command of the language, if you are unfamiliar with the subject matter, it might easily be that a totally outrageous suggestion intended ironically becomes understood as real advice.

I don't want to rake over yesterday's events again - so let me give you a 'real world' example where I made this very mistake. Many years ago, back when computers ran MS-DOS, I was working with a not-very-computer-literate co-worker. He phoned me up one day saying that he was running out of space on his hard drive and did I know what he could do about it? I jokingly said: "Well, you could always type "DEL *.*" (which deletes all of the files on the hard drive - including the operating system). I went on to try to explain where he might find files to safely remove - but as I was doing so, he said "No - that didn't work - now my computer has locked up"...he'd gone ahead and actually typed what I suggested.

This kind of thing can happen so easily - and is yet more likely to occur in an online conversation where one's tone of voice is missing.

So - I'd very much like to see some kind of concrete rule that says "Thou shalt not give bad advice - even in jest."

Aside from that, the guidelines seem perfectly fair. Thanks! SteveBaker 18:36, 16 April 2007 (UTC)

Were you sent by Friday? Or did you join the dark side of your own choice. By the way, i agree with your sentiments. David D. (Talk) 18:49, 16 April 2007 (UTC)

I try to include a smiley after anything that could possibly be misinterpreted. Or, we could just do on the spot executions for anyone who makes a joke. :-) StuRat 19:17, 16 April 2007 (UTC)

So you think up a joke then try and find some way, however tenuous, of making it relevent to the question? Bang bang. You're dead. Rockpocket 01:07, 17 April 2007 (UTC)

Oh boy, so i'm a goner for the above :-) David D. (Talk) 19:21, 16 April 2007 (UTC)

For jokes, no. Support on the spot executions for the inclusion of any "smileys" in a response.—eric 19:28, 16 April 2007 (UTC)

But they are there specifically to ensure that everybody knows the statement is not to be taken seriously, thus they are critical. StuRat 19:30, 16 April 2007 (UTC)

I'm convinced—nicely demonstrated.—eric 19:46, 16 April 2007 (UTC)

But smiley faces are not universally understood. My mother is 70 years old and occasionally consults Wikipedia - she has no idea what :-) means...much less :) or =P it's very easy to miss such things if you aren't hip to the Internet culture. I'm not saying don't make jokes - I'm saying that when a joke could be misinterpreted as untrue information - don't do it. SteveBaker 01:47, 17 April 2007 (UTC)

I suppose we could add "just kidding" to make it absolutely clear. StuRat 01:58, 17 April 2007 (UTC)

(edit conflict) I agree with the thrust of that, Steve. I'm as guilty (if that's the right word) as anyone of adding humorous asides to threads. In principle, I not only can't see the harm in this, but also think the occasional bit of humour helps to maintain the friendly tone we have here. As long as it's not done gratuitously or insensitively. Very often, jokes are appreciated by the questioner and other responders, although sometimes they fly swiftly over their heads (or appear to, based on lack of comeback) - which is as good as falling flat (to mix a metaphor or two). What I don't usually support are in-jokes, where the questioner is effectively excluded. I say "usually" because sometimes the temptation is just too great to resist. Such is the lot of a punster. JackofOz 02:06, 17 April 2007 (UTC)

Yes, anyone who makes in-jokes should be attacked by a flock of ravenous seagulls. StuRat 02:30, 17 April 2007 (UTC)

• We say: "Having a bit of humour in your answer is almost always a good thing but...". Really? I can think of many, many examples of questions where humour is pretty innapropriate (for example, in questions about death and diseases). As it is phrased it reads like we are suggesting one should attempt to add humour to improve an answer. I suggest we change "is almost always" to "can be" to reflect this. Rockpocket 18:33, 18 April 2007 (UTC)

We definitely don’t want to come across as not having a sense of humor here. But we don’t want people to accidently take us seriously. Couldn’t we just have a template to use on non-serious posts? Maybe something similar to the “This page is intended to be humorous and does not reflect Wikipedia policy. . .” one? 71.112.23.141 22:03, 19 April 2007 (UTC)

Consensus ?

Under whose authority was it decided that a consensus had been reached on these guidelines ? I see no consensus discussion, or even announcement, made here on the talk page. I've only today found out that somebody did this. StuRat 23:23, 21 April 2007 (UTC) -+

Stu, please take the time to read Wikipedia:Consensus as, based on your recent comments, i think you don't really have an understanding of this. Nobody has any "authority" to declare that a consensus has been reached, it is normally "reached as a natural product of the editing process." Also, nothing is ever "finalized" (as you asserted in an edit summary), consensus can change and bold, revert, discuss is the right way to try and make that happen. But please, a little moderation eh? You've not shown much interest in this page during the months it has been developing, to show up now and try and make major changes by edit warring is not going to work.—eric 23:42, 21 April 2007 (UTC)

Well, I don't agree that there is a consensus, so I'm going to remove the template that says it's a guideline reached by consensus. And I've shown plenty of interest, count the number of edits I have on this page, it's dozens. StuRat 00:39, 22 April 2007 (UTC)

I just counted my name 72 times on this page. Those aren't all edits, but I'm sure most of them are. I'm also the one who started the whole process. StuRat 00:53, 22 April 2007 (UTC)

How many of those are reverts to your preferred version ;P Rockpocket 01:19, 22 April 2007 (UTC)

I think you misunderstand. My name is on this talk page that many times, not in the history. StuRat 09:01, 22 April 2007 (UTC)

Why is the guideline tag being removed? Where's the problem? Friday (talk) 16:09, 22 April 2007 (UTC)

Because EricR, unilaterally, decided to declare the discussion period over and a guideline to now be in force: [9], with no edit summary as to why he thinks a consensus has been reached, despite there having been daily changes to the guidelines and several ongoing discussions on this talk page. No discussion or even announcement of this change in status was made, and the links asking people to discuss the proposed guidelines were also quietly removed from the Ref Desks. I can't think of any more inappropriate way to go from a proposal to a guideline than that. StuRat 17:18, 23 April 2007 (UTC)

Please take the time to read my comment above, that i "decided to declare the discussion period over and a guideline to now be in force" is quite contrary to my attempted explanation. I think you are making things here altogether too personal, and wonder you didn't comment when this document was moved to {{guideline}} status back in January or when i thought there was {{No consensus}} back in March? Erroneously ascribing opinions and motivations to me is not at all advancing your cause and certainly setting back the overall process. If you have such a problem with my edits in particular, as you have implied on a number of occasions, then please, take your concerns to RFC so we don't further drag down this discussion. Otherwise, make at least some attempt to read and understand my comments.—eric 18:37, 23 April 2007 (UTC)

I don't see any explanation above, just that you apparently think anyone can just change it into a guideline from a proposal whenever they want. Look at what Ten did below, as far as actually asking people if it's ready to become a guideline, and then actually waiting for replies, and hopefully listening to what people say. StuRat 02:07, 24 April 2007 (UTC)

Yes, absolutely, now you're catching on. Anyone can (if they think it improves Wikipedia) just change it into a guideline from a proposal whenever they want (as long as they are not edit warring.) It's called the "wiki process" and is the method which will ultimately determine whether this document is a guideline or a failed proposal. See the bold, revert, discuss page i linked above for a more complete explanation.—eric 02:51, 24 April 2007 (UTC)

Wrong, you shouldn't "be bold" with major changes like that, you should first verify that a consensus exists. StuRat 03:06, 24 April 2007 (UTC)

I've read a bit more now. StuRat, your edit summaries make it clear that you don't have a clue what you're doing here. NPOV requires that we express "all" points of view? Ridiculous. Go take it up on that talk page if you really must object to NPOV, but keep in mind you'll be meeting all sorts of new editors that way, and many of those editors may not have much patience for your eccentric interpretations of policy. Until StuRat can present a reasonable statement of what he's objecting to, and his statement is one that comes from an understanding of other core policies rather than a rejection of them, I strongly suggest he stop reverting. Friday (talk) 16:36, 22 April 2007 (UTC)

Saying "you don't have a clue what you're doing here" is a blatant personal attack, please stop immediately. As for what NPOV means, it clearly doesn't mean, even in an article, that each editor must only make NPOV contributions. For example, in articles with "advantages" and "disadvantages" sections (or "support" and "criticisms"), an editor is allowed to only contribute to one of the sections without being reverted for POV. On the other hand, if they went and deleted the other section, that would be an NPOV violation. Similarly, on the Ref Desk, each editor is allowed to present a single POV, so long as they don't delete the POV of others. Thus, the total response will hopefully be NPOV. Only by deleting one POV and allowing another to remain do we violate NPOV. StuRat 04:26, 23 April 2007 (UTC)

We are in danger of confusing representing a POV in a NPOV manner ("white extremists think blacks are inferior") and simply representing a POV ("blacks are inferior"). Stating only the former is perfectly acceptable, even though it is simply representing only one POV. Stating the latter is not even if contrasted with the opposing view. A statement saying "Blacks are inferior" countered by "Blacks are not inferior" is not NPOV, even though we are "expressing "all" points of view". This is what WP:NPOV is about in this context: representing positions in a neutral manner. It doesn't imply any editor has to represent all positions equally, simply that the position they do represent should be stated neutrally. Rockpocket 17:39, 23 April 2007 (UTC)

Racism goes well beyond POV. Multiple POVs are valuable, but racist statements are not. StuRat 02:19, 24 April 2007 (UTC)

What the desk is not

I like the guideline suggesting people don't use the Desks as forums for their own opinions, but I wonder if we could also suggest that answerers address the question itself. Too often, my queries are answered with, in effect: 'Well, I have no idea what the answer to your question is, but I hate to say 'I don't know', so I'll tell you this instead' - and give some piece of irrelevent info instead. I admit, very occasionally the info given is kinda interesting, so the guideline would have to be carefully worded... Adambrowne666 23:55, 21 April 2007 (UTC)

Thats a good point. However, often the querent will as a very non-specific and general question, meaning you have to interpret what they actually wish to learn. Different people will interpret such questions in different ways, leading to tangential answers. Rockpocket 00:25, 22 April 2007 (UTC)

For example, in response to "What's the best car", we might get some commenting on value, some on mileage, some on looks, some on popularity, some on safety, some on profitability. Also, some may mention specific cars ("the car that won the Indy 500 in 2003"), some might mention models, some may mention platforms, some may mention brands, and some may mention manufacturerers. StuRat 00:43, 22 April 2007 (UTC)

There's a bit in there under Guidelines for responding to questions: "...responses to posts should always attempt to answer the question..."—eric 00:34, 22 April 2007 (UTC)

Yes, I agree with all your points, but I still think a guideline could allude to this, as long as it was very carefully phrased so as to include the provisos you guys have mentioned. I am thinking of something a bit more specific than the line you point out, Eric - I'd love to be able to weed out that kind of contribution from the desks by quoting something explicit from the Guidelines. Adambrowne666 23:07, 22 April 2007 (UTC)

Serial commas

Some of the lists in this guideline use serial commas, some do not, and this inconsistency presents an appearance of amateurishness. Although I am (as you can see) a fan of such commas, my main concern here is consistency. Matchups 12:32, 22 April 2007 (UTC)

I have no objection to you making the comma usage consistent. Have at it ! StuRat 04:00, 23 April 2007 (UTC)

So, what is it then...

...a guideline, or a failed proposal?

To my reading, this document is a damn fine proposed guideline, and it's time to make it official. There seems to be some conflict over whether or not there is agreement on that point, and I'd like to gauge the level of approval from the community at large.

Bearing in mind that guidelines necessarily have some flexibility to them, and that there is still room for evolution of this document in the future, and that it's not necessary to deal with every possible conceivable eventuality in these guidelines, could people summarize briefly below any deal-breaking concerns that they have about this document?

People who are comfortable with the guideline as written are welcome to indicate their approval; if you're okay with it but would like to suggest tweaks, that's great too.

If there are any 'deal-breakers' – issues that you absolutely cannot support this guideline without – express them in the second section.

I'd urge involved parties not to declare the discussion closed until there's been a chance to hash out any disagreements, and in any case to allow at least seven days from the start of this thread. Thanks for your suggestions and your participation, TenOfAllTrades(talk) 02:29, 23 April 2007 (UTC)

Hello again. I'd also like to suggest that people not make any substantive changes to the guideline for the next few days, just to avoid confusion about which version people are commenting on and (dis)agreeing with. Feel free to make changes that are cosmetic or improve readability, flow, and style, but take care not to alter the sense or meaning of the proposed guideline at this time. Obviously I can't compel people not to make major edits/additions, but I hope everyone can see why I would ask this.

Finally, I encourage everyone to bear in mind that this document is a guideline. As such, its application will be flexible, and people will be expected to use their common sense and acknowledge that there is going to be some 'wiggle room' in its interpretation. (A guideline is meant to be a framework, not a straitjacket.) I also don't expect this document to necessarily address every possible situation that might arise, and anticipate future development. TenOfAllTrades(talk) 15:51, 23 April 2007 (UTC)

Looks good. (Minor tweaks are bulleted.)

• TenOfAllTrades(talk) 02:29, 23 April 2007 (UTC)
• David D. (Talk) 03:20, 23 April 2007 (UTC)
• Yes, this document is for the most the same as the prototype, Ten's RD thoughts, and despite some reordering and rewording what's in the guideline now has remained substantially unchanged for months now. More importantly tho, the desks themselves (in my opinion) have since improved. In general there seems to be more of an effort to keep responses on-topic and to cut out all the soapboxing. I think the guideline is a pretty good description of what most editors want to see on the reference desk.—eric 03:59, 23 April 2007 (UTC)
• Sluzzelin talk 05:36, 23 April 2007 (UTC) (Thanks to everyone for co-operating and working hard to achieve this. My main hopes projected into these guidelines are twofold: First of all, that they provide useful information for newcomers, people who wish to ask questions, and all other editors too. And furthermore, that we will see less drama and disruption at the desks and their talk page. See also my caveat below.)
• I'm pretty happy with things as they are. Rockpocket 05:56, 23 April 2007 (UTC)
• Looks reasonable to me. >Radiant< 09:32, 23 April 2007 (UTC)
• I'm sure there's room to quibble over details. But generally this is nothing new- it's just an explicit explanation of how to apply other policies and guidelines to the ref desk. Friday (talk) 14:06, 23 April 2007 (UTC)
• I think that generally it looks good. There is always room to quibble over minor details, but we don't really want to get caught up over the color of the bike shed, do we? ♠PMC♠ 06:42, 2 May 2007 (UTC)

Can't agree to this without...

• ...a clause added to the permitted humor section that specifically states that misleading or potentially misleading humor is disallowed. (See earlier discussion here) With that addition, I would support this proposal. SteveBaker 02:41, 23 April 2007 (UTC)

• Taking away all contradictions, like the one I have exposed above in the section "Contradiction"; take out the sentence "the desk attempts to provide services comparable to a library reference desk", as this reference desk doesn't have the limitations that a library reference desk has and can provide a lot of services that those can't; delete the part that says "our standards on verifiability, neutral point of view, or no original research apply to the Reference Desk as they do to the rest of the project"; deleting the sentence "first and foremost, we're here to answer questions", as I feel like people will be watching me to see whether I am doing disruptive stuff and will delete anything that is not either a question or an answer or something that they like; creating guidelines saying that it is not OK to delete other people's posts. A.Z. 03:05, 23 April 2007 (UTC)

• Agree with AZ. Also, we need to state that you shouldn't just delete anything you disagree with, or think is incorrect. If you disagree, state why, and provide any evidence, but leave the original response. This has been a problem recently. Here is the language I would add: [10]. Here are the other changes I feel are needed: [11], [12], [13], [14], [15], [16]. Also, we need to state that argument from authority and ad hominem attacks are not allowed. StuRat 03:57, 23 April 2007 (UTC)

• Can you give us some examples of it being a problem recently? David D. (Talk) 04:04, 23 April 2007 (UTC)

• Sure: [17]. StuRat 04:12, 23 April 2007 (UTC)

• How frequently does this sort of deletion occur? David D. (Talk) 04:19, 23 April 2007 (UTC)

• I am sure StuRat could give you a bunch of examples, but I must say that people are just stopping to write things that they know that Friday and Eric will disagree with because they are sick and tired to see their edits being reverted. I for one have not added anything these past weeks to the reference desks. I will do so when I am sure that people will not delete my posts. The reference desks are not a welcoming place anymore. If you take a look at the history of the talk page, you will see a lot of people defending posts of the kind that they are trying to forbid here (things that do not fit the deletionists obscure requirements), people who have stopped editing the reference desk. So, yes, it is a major problem. And the deletionism problem is not only affecting the people who answer questions, but also the people who ask questions. They just don't like when some bold guy says "hey, take it somewhere else, this is an encyclopedia!" and no one else answers anything because they don't know that the deletions are unilateral actions unsponsored by the community and they fear they'll be doing something "wrong" in answering the question. A.Z. 04:23, 23 April 2007 (UTC)

• Well said. As for how often it happens, I'm not likely to notice secret deletions (done without notification of the author(s) or even a post on the Ref Desk talk page), so I have no idea how many there are. And having even one regular editor who thinks it's appropriate to delete answers he judges to be incorrect is enough to present a serious problem. StuRat 04:32, 23 April 2007 (UTC)

• Any answer that follows the guidelines will not get deleted. For the record, i do not support the deletion of wrong answers. However, I do believe that there is precedence for deletion being a tool for the triage of the ref desk when these guidelines are not followed. I would not agree to a deletions are forbidden clause in the guidelines. I note that is not what StuRat is proposing, literally. It is the future interpretation of that clause that i worry about. David D. (Talk) 04:35, 23 April 2007 (UTC)

• I don't see the problem. If we add a "don't delete wrong answers" rule, how could that possibly be used to prevent deletions of, say, obscene racial slurs ? StuRat 04:44, 23 April 2007 (UTC)

• If the slur was part of an answer then could I delete it? But I am happy with the philosophy of don't delete wrong answers if it is not used to prevent all deletions, i.e. of racial slurs. i say put it in. David D. (Talk) 04:49, 23 April 2007 (UTC)

• I'd love to put it back in, but Rockpocket threatened to 3RR block me if I do. Would you mind putting it in: [18] ? StuRat 07:50, 23 April 2007 (UTC)

• The interpretation of any guidelines is a whole new ball game, but we have to tackle it. I think that somewhere the point needs to be made in the guidelines that Ref Desk answers are far more analogous to conversations or talk pages than to polished articles. The Ref Desk does not purport to present one and only one comprehensive and accurate answer to a question at any one time, as our articles strive to do. Ref Desk answers almost always contain different points of view from different editors, and often there is no one "right" answer. This is more likely to be the case at Humanities, Languages and Miscellaneous. Over at Mathematics, Computing and Science, it's more likely that a single correct answer to the question will be arrived at, either immediately or by discussion - but even there, different users may present different valid approaches to a problem, with the same outcome, and the questioner is free to choose whichever approach suits them best. Once questions are finally done with, they are rarely revisited, another very significant difference from articles, which are constantly being changed. These differences are why, imo, a full record of the Ref Desk discussions needs to be maintained, and deletions of posts should occur only in cases where the only discernible purpose is wilful harm - what I call "outright vandalism", or grossly gratuitously offensive or libellous posts (I know A.Z. probably does not agree even with those exceptions, but it's my view). JackofOz 05:13, 23 April 2007 (UTC)

• I agree completely. Well said. StuRat 07:08, 23 April 2007 (UTC)

• i don't want to revive the argument from authority debate, but i will note that StuRat and I have a very different interpretation of what constitutes an argument from authority. David D. (Talk) 04:38, 23 April 2007 (UTC)

• I'd rather state in the guidelines: "Arguments from authority and argumentum ad hominem are logical fallacies". Just that. I sort of read the thread and don't think StuRat and David D. disagree as to what constitutes an argument from authority! A.Z. 04:51, 23 April 2007 (UTC)

• I agree with the guidelines, with the following minor misgivings:

• It is not always possible to adhere to"Our standards on verifiability, neutral point of view, or no original research apply to the Reference Desk as they do to the rest of the project". There have been numerous instances where a responder has come up with an answer in the form of a personal anecdote, which by definition is OR. Many such answers are much appreciated by questioners.
• Also, "If it is impossible to answer a question without some calculation, please make this clear in your answer with a phrase such "My calculation is as follows …"." suggests that OR is welcome in such cases.
• To easily fix this, I'd be happy if we could say "Our standards on verifiability, neutral point of view, or no original research generally apply to the Reference Desk as they do to the rest of the project ...." -- JackofOz 03:59, 23 April 2007 (UTC)

• I agree that this is a problem. Here is the text I proposed to fix this problem: [19]. StuRat 04:37, 23 April 2007 (UTC)

Jack, what do you think of the nutshell version: "Keep the core policies and guidelines in mind when responding..."?—eric 06:05, 23 April 2007 (UTC)

I'm happy with both Stu's and your suggestions on this, eric. JackofOz 06:14, 23 April 2007 (UTC)

Well, i like the current wording and think we need mention V, OR, etc. specifically, but when combined w/ "ignore are rules" as Sluzzelin says below i think the nutshell version is the summary we end up with. Think about the policies and then provide the best response you can for the reader.—eric 06:55, 23 April 2007 (UTC)

I agree with eric. I think we should be trying to adhere to V, OR, RS ect. That should be the benchmark. Of course, depending on the question, that might not be achievable. But guidelines are just that - a guide. Rockpocket 16:58, 23 April 2007 (UTC)

I disagree, we shouldn't put policies into a guideline which we know are often impossible to follow on the Ref Desk. And, if we do, we should at least state that we know it's not always possible, and that, in those cases, the Wikipedia policies don't apply. StuRat 01:54, 24 April 2007 (UTC)

• As with everything here, WP:IAR, depending on circumstance and circumspection. ---Sluzzelin talk 05:39, 23 April 2007 (UTC)

diversity of roles

RW reference librarians have never totally agreed about the relative role of direct information-providing and instruction. There's a somewhat different attitude at public libraries than at academic, where the librarians see themselves as primarily instructors in library use. WP of course is closer to a public library, but most public librarians too often blend in some instruction about how to deal with future questions--at least indirectly by explaining where you're finding it--and I think that the practice here actually is to do this, not just give the answer and the source. My own experience is in higher education exclusively, but we also answered questions from anyone, and I made a distinction between questions that the inquirer would never have to address again, and those where here would be a point in teaching.

Librarians also work in the public, just as WP editors. Except for removing any personal identification provided, I think absolutely nothing should be deleted. It can be marked strikeout, but not deleted. Professionals are responsible for their mistakes, and are expected to admit them and correct them, and those trying to work like them should take such responsibility also, just as they do in other areas of WP.

(This is a comment on the interpretation of the guidelines--I think the text as it stands is acceptable) DGG 03:27, 23 April 2007 (UTC)

I agree that only strikeout should be used, and I would restrict that to your own text, as well. StuRat 04:03, 23 April 2007 (UTC)

"NPOV doesn't apply here.."

Responding to A.Z.'s justification for this edit, see m:Foundation issues:

The Wikimedia projects as a community have certain foundation issues that are essentially beyond debate. People who strongly disagree with them sometimes end up leaving the project.

number one on the list—NPOV.—eric 06:15, 23 April 2007 (UTC)

You have misrepresented what A.Z. and I have said in a straw man argument. We support NPOV, but not the misinterpretation being used here that each response must represent all points of view, or just the most widely accepted view. Only an entire article, or in the case of the Ref Desk, an entire question and it's responses, should aim to be NPOV. Here is the relevant text from the policy:

The neutral point of view is a means of dealing with conflicting views. The policy requires that, where there are or have been conflicting views, these should be presented fairly... Readers are left to form their own opinions.

StuRat 06:55, 23 April 2007 (UTC)

Thought it's probably a good idea to try, no one has stated that each and every response individually "must represent all points of view". I rather thought y'all were trying to remove the "significant views published by reliable sources" language as you did in your quote above w/ the ellipsis. If so, you should try on the NPOV talk page.—eric 07:24, 23 April 2007 (UTC)

What is the point in even listing NPOV here, since it doesn't apply to an individual reply ? That is what these guidelines are for, after all, the content of each individual reply. StuRat 07:44, 23 April 2007 (UTC)

How would it be to our advantage to encourage replied that aren't neutral? We do want individual replies to be neutral. If someone were hanging out at the ref desk, giving out biases replies disguised as factual, this would not be a good thing. Friday (talk) 14:38, 23 April 2007 (UTC)

It's a misuse of the NPOV policy to claim it says each Ref Desk contribution must be NPOV, it doesn't say that at all, nor is that even possible. If somebody asks what causes psychotic behavior, for example, it's entirely reasonable for one person to reply with psychological reasons and another with biological (brain structure/chemistry) reasons. Neither answer should be deleted for only containing one POV. StuRat 14:47, 23 April 2007 (UTC)

The best answer gives the whole picture, rather than just part of it. I think the only way an answer is in danger of deletion for npov reasons is if it's seen as soapboxing rather than an attempt to provide a neutral answer. I think the "keep in mind the pillars" bit addresses this adequately, but as always we depend on reasonable editors interpreting this reasonably. Friday (talk) 16:29, 23 April 2007 (UTC)

No, your version of the nutshell implies that every response must be NPOV, so only including the biological perspective in the example above would not be allowed. This is completely wrong and it should be explicitly made clear that each response most definitely does not need to take all possible POVs into account and attempt to come up with a single NPOV response. My version made that quite clear. StuRat 17:32, 23 April 2007 (UTC)

If you're still on about "including all possible POVs", you don't understand npov. If you're going to object to npov, this page isn't the right place for it. Please make some effort to interpret our guidelines here in the context of Wikipedia's other guidelines and policies. Friday (talk) 19:03, 23 April 2007 (UTC)

If you're going to misinterpret NPOV policy to mean that each edit must be NPOV, when it really says each article must be NPOV, you should go and complain on the WP:NPOV page and try to get the policy changed there, not here. Specifically, you need to change the word "article" to "edit" every place it occurs in the policy, which has some 40 occurrences StuRat 01:48, 24 April 2007 (UTC)

Without getting into whether anyone is misinterpreting anything, I support StuRat's basic argument here. To illustrate: There was a recent debate (which for the life of me I can't find anymore) about why some editors write "edit conflict" before certain posts. The point was made in that debate that there's a good deal of ego involved in contributing to the Ref Desk, and that's not a bad thing unless taken to extremes. The reasons why users contribute to the Ref Desk will vary, but I suggest that one of the reasons is that they derive some personal satisfaction from helping others. That personal satisfaction is essential for a healthy human environment, and it comes - at least in part - from having their viewpoint recognised and validated by others, if not necessarily accepted as containing the complete answer to the question asked. To that extent, the majority of posts are inherently POV. Usually, a collection of such POV statements will comprise the complete answer (or as complete as we go in any one case). Many posts are acknowledged as guesses, but they're made in good faith and in a genuine attempt to shed some light on the answer. Such guesses would be unacceptable for an article, but they can of great value to a questioner who comes to the Ref Desk for help. I'm not suggesting we abandon the NPOV principle, but at the same time we have to recognise the reality of how the Ref Desk works, and has to work. It's not like an article. The Wikipedia pillars etc apply to the Ref Desk, but with some substantial qualifications and exceptions. JackofOz 01:57, 24 April 2007 (UTC)

Its one thing to accept that there will inevitably be some movement from the most strict interpretation of the pillars - I don't think anyone disagrees with that. Its something altogether different to establish guidelines that explicitly legitimizes those in general terms. We should all be aiming for WP:NOR, WP:RS and WP:NPOV, because when it is possible to provide that sort of answer, those are the best answers we can give. Once we have a guideline that states "No attempt should be made to suppress the views of others." - I just know that it would be quoted ad infinitum to justify a preference for stating opinion over fact ("thats my view and the guidelines says that shouldn't be suppressed"). All of a sudden we find ourselves with a chatroom.

Lets have guidelines that guide us towards best practice, but give us the required flexibility. We all know that exceptions will exist, and we use common sense when those occur, but there are no need to ringfence them with explicit qualifications. Rockpocket 02:22, 24 April 2007 (UTC)

Your use of the term "best practices" is incorrect. Answering a math question by doing some math is not an "inferior answer" just because it uses OR. Answering a question on what causes psychotic behavior by giving the biological POV is not an inferior answer. Stating that a Cadillac is a GM vehicle in response to a question is not an inferior answer just because the statement lacks authoritative sources. The goals of WP:NOR, WP:RS and WP:NPOV only apply to some questions, not all, and perhaps not even most. StuRat 02:45, 24 April 2007 (UTC)

I know that deletion of others' posts is like waving a red rag at a bull, and causes no end of controversy. Which is why it should occur only in extreme cases (and why I have never done it). The dilemma is this: If we say nothing about suppressing others' opinions, it will be read by some as meaning it's acceptable to delete others' posts if they disagree with them - which it's not. If we say something about it, it might be perverted in the way you suggest, Rockpocket. It's a risk either way. Deletions occur infrequently in the scale of things, even if the heat generated by them is out of proportion to their frequency.

On balance, I'd rather have a guideline that promises a full record of discussions, even if they contain what some might call trivial, irrelevant or unuseful material, and even if they occasionally become a chat session. The wording needs to be carefully constructed, granted. The alternative is to implicitly condone someone deciding unilaterally to delete another person's post without the most compelling of reasons. Mere difference of opinion about the subject under discussion should never be considered a compelling reason to delete. It's such differences of opinion that bring life to debates and make the Ref Desk a healthy place for everyone, not least the questioner, who is often forgotten in all our behind-the-scenes talk. We're here primarily to be of service to them, and any improvements to the encyclopedia that might result are the icing on the cake, not the cake itself. The post that was deleted might have been exactly what the questioner wanted to know, but thanks to the actions of one high-handed user, they'll never get to see it. That's not being of service to our primary clients. -- JackofOz 03:30, 24 April 2007 (UTC)

I agree completely. I am wholeheartedly opposed to half-assed deletions and will give no quarter to the perpetrators. I support any language that will prevent this. StuRat 06:02, 24 April 2007 (UTC)

I also agree with JackofOz. Specifically, I agree that an individual RD response will often be POV, and this cannot be avoided; unilateral deletions on the RDs should only be done for the most compelling reasons; and "because I disagree with the answer" or "because I have a source that disagrees with the answer" are not sufficient reasons to delete an RD post. Gandalf61 08:14, 24 April 2007 (UTC)

I am also oppose the deletion of constructive and informative answers by editors in good standing in any format (why I have never done it, either, in spite of having been given the unhelpful "deletionist" label by certain other editors). Such answers may include OR (such as in the case of maths or computing questions) or POV (such as in the case of questions that explicitly ask for opinion) or personal anecdotes. We already have this covered in the guidelines: "If it is impossible to answer a question without some calculation, please make this clear in your answer". So if the questions calls for OR, POV or opinion - all of which are inherent in making calculations - and that is made clear in the answer, then there is no basis for that being deleted. However, since the Ref Desk is not a discussion forum or chat forum, there is no need to have explicit guidelines warning against "suppressing others' opinions", because the primary goal not for the expression of opinion, its for answering questions informatively. Therefore, I think the exceptions to the pillars are already adequetely covered and I'm not adverse to a guideline about removal of content, but it should be couched in terms of not deleting information useful to the OP, not in terms of enforcing freedom of expression. Rockpocket 17:25, 24 April 2007 (UTC)

That seems a promising direction. Our concern is utility. TenOfAllTrades(talk) 17:56, 24 April 2007 (UTC)

The problem with your answer, Rockpocket, is that whether answers are "constructive and informative" or contain "information useful to the OP" are themselves matters of opinion. When we leave it up to everyone to decide which other answers are "constructive and informative" and delete them if they don't think they contain "information useful to the OP", we get exactly what we want to avoid, people deleting answers they think are wrong or not useful. We don't want to allow that at all, as only the OP should make those decisions, so we need stronger wording than what you seem to propose. StuRat 20:29, 24 April 2007 (UTC)

That is a possibility, StuRat, but the alternative is equally open to abuse - the explicit protection of freedom of expression could be used to justify any comment on the Desk whatsoever, no matter how off topic or irrelevent. We certainly don't want to allow that at all. You seem to be suggesting that every answer that isn't strictly sourced or phrased in strict NPOV language would be deleted by these guidelines. Have a look at the Desks at the moment. There are many, many opinions stated that are not being deleted. At the end of the day, the guidelines are a matter of interpretation. They should be stated in terms of benefit for the querents, and we must have good faith that respondants will interpret them for the benefit of the querents. Even if there was a guideline supporting freedom of expression on the Ref Desk, I would still delete a purposely offensive comment aimed at the OP if I considered it serious enough, per WP:IAR and WP:CIV. In other words, there is no "freedom of expression" on Wikipedia if in doing so one violates our core policies. However, to address your specific concern about the removal of "wrong" answers, I would be happy to endorse a guideline prohibited discouraging the removal of good faith attempts at providing information useful to the OP. Rockpocket 20:57, 24 April 2007 (UTC)

(resetting long indent) I have a comment but I see Rockpocket already slightly tweaked his statement. Discouraging the removal of good faith attempts at providing useful info, I'm fine with. I wouldn't want to see the guideline say we prohibit it though- we've had some ridiculously stupid answers that have been defended as good faith attempts to provide useful information. A bull in a china shop may have good intentions; this doesn't make it less disruptive. Friday (talk) 21:05, 24 April 2007 (UTC)

Yes, after posting that occurred to me also, hence the tweak. Rockpocket 21:08, 24 April 2007 (UTC)

Wrong answer

Everyone should attempt to give correct answers, but a few wrong answers are inevitable. If you believe your own answer is wrong, please strike it out and add a comment as to why you no longer think the first answer was correct. If you think somebody else's answer is wrong, add a comment saying so, why you think it is wrong, and providing evidence, if you have any. However, do not remove the incorrect answer, as it may actually be you who is wrong.

We should discuss how we deal with incorrect answers. Please see what I wrote above and comment on it here. StuRat 22:54, 21 April 2007 (UTC)

Friday has now removed the last line: "However, do not remove the incorrect answer, as it may actually be you who is wrong" and replaced it with "The best way to avoid disagreement is to provide answers from a reliable source.": [20].

I disagree both with the removal and with the new text. The new text doesn't seem helpful at all, as many reliable sources can be found which disagree with one another. I believe there is a substantial majority which agrees that answers perceived to be incorrect should not be removed. If I read the discussions right, this includes myself, A.Z., JackofOz, Gandalf61, DGG, Rockpuppet and David D. with only EricR and Friday opposed. I can't quite read TenOfAllTrades's opinions on this.

StuRat 06:43, 25 April 2007 (UTC)

Since the intent of edits to this document seem to be often misunderstood, i provided a "b version" for comparison. I added it before seeing this talk page section, hoping for more discussion, not a vote between versions. Please, edit away on the "b version".—eric 21:03, 24 April 2007 (UTC)

I made a minor tweak to the 'b version', by removing the requirement that a request for references be placed only on the editor's talk page: [21]. Putting such a request there may not be ideal for a number of reasons. In many cases, there are distinct advantages to putting a request on the Desk itself. First, it takes advantage of our high-traffic Ref Desk. Someone (or several someones) can do some legwork and rapidly firm up support for an uncited claim. Second, you don't know how long it will take before another editor even sees a message—if someone's asleep, it could be hours; if they've gone away for the weekend it will be days. If the posting was from a dynamic IP, then they might never see the request for references. Finally, in cases where a dubious answer has been given, I don't see anything wrong with asking politely, "Hey, I'm not sure if the above is correct. Can anyone provide a citation for this statement?" It (politely!) reminds the original poster to take everything they read on the internet with a grain of salt, and it encourages other editors to provide clarification or sources. If someone posts a response without providing any sources, it's not unreasonable to express doubt, even without sources. I don't want to see a situation where only the first response is privileged, and a higher standard of evidence is required to express doubt. (I think we've gone over this concept in one of the archives.)

Otherwise, things look reasonable, though I agree that jumping straight to a vote probably isn't a good way to go. TenOfAllTrades(talk) 22:28, 24 April 2007 (UTC)

Oops, meant as a suggestion to prevent argument on the desk not as a requirement, if the original editor is a reasonable sort then yes, better to add it in as a response. I actually like the higher burden for challenges (tho not quite as StuRat has presented it) you already know there is at least one other editor who disagrees. Simply questioning a response doesn't seem like adding much value to the desk; whereas looking over the articles and doing some research does. Maybe it's all too fine a point, but that's what i was thinking w/ "due diligence", and maybe that text needs some tweaking also.—eric 23:34, 24 April 2007 (UTC)

So, let me explicitly ask, do you support or oppose the sentence:

"However, do not remove the incorrect answer, as it may actually be you who is wrong." StuRat 06:43, 25 April 2007 (UTC)

Support (possibly with tweaks)

• Support. StuRat 20:52, 24 April 2007 (UTC)
• Support. JackofOz 06:06, 25 April 2007 (UTC)

Oppose

• I don't think we want to explicitly say to remove or not remove an answer. Sometimes it may be appropriate to remove a very wrong answer, but we certainly don't want people removing other editors' posts lightly. I believe in cases where an answer has been seen as very misleading or dangerously incorrect, there's sometimes been consensus for removing them. Hopefully this happens rarely, if at all. Friday (talk) 20:56, 24 April 2007 (UTC)

• In that case, we could say "don't delete answers you believe to be incorrect unless you first gain a consensus to do so on the talk page". StuRat 21:04, 24 April 2007 (UTC)

• Well, in cases of very bad answers, it's often better to remove it sooner than later. If an editor uses good judgment and does this only when he believes consensus will support him, I think this is good enough, whether or not he explicitly gets consensus ahead of time. Friday (talk) 21:07, 24 April 2007 (UTC)

• I strongly disagree. If we provide proof that an answer is wrong, that's good enough until we have time to gain a consensus that the answer should be removed. Leaving it up to each editor to delete things based on their own judgment (including their own judgment as to whether others will agree) leads to bad deletions and edit wars. StuRat 21:25, 24 April 2007 (UTC)

• That's a problem we already have. Editors already need to be able to reasonably guess what will or won't be controversial. This is how you know whether to just boldly make a change, or take it up on the talk page first. Friday (talk) 21:30, 24 April 2007 (UTC)

• I spoke too soon, perhaps. I like Eric's "b version" better than the "a version". It really comes back to verifiability. Friday (talk) 20:59, 24 April 2007 (UTC)

• I oppose the deletion of answers just because another user disagrees with them, even if the original answer is demonstrably wrong. Discussion of opposing opinions is what the Ref Desk is all about. This is not like getting one answer to your question from a librarian on a library ref desk. There are many of us here, and opinions differ. Finding out why a genuinely held view is wrong is a learning experience - and you can bet that if one person holds such a view, many others do too. Deletions should occur only in the most compelling of cases, such as gratuitous insensitivity perceived to be designed to cause gross offence to some group; or libellous statements that could bring Wikipedia into disrepute or legal difficulties. JackofOz 06:15, 25 April 2007 (UTC)

As for "discussion of opposing opinions is what a Ref Desk is all about"- no,no, a thousand times no. That's exactly the point we're making with "The ref desk is not a forum". It's not for you and me to debate our opinions. This leads to all manner of trouble and doesn't further the goals of the project in any way so we want to avoid this. We can point people to articles about controversial issues but rehashing the controversy on the ref desk is not productive. The only debate we want at Wikipedia are the unavoidable debates about what's best for the project. Debates about the world at large have no place here. Friday (talk) 14:46, 25 April 2007 (UTC)

I disagree, we very much do want to see all the different POVs on a topic, like the biological and psychological POVs on the causes of psychosis, for example. StuRat 15:11, 25 April 2007 (UTC)

I think the main issue here is to be objective and have a balanced response. This might entail adding other sources (let's forget the RS argument for now) but should not entail one fighting for ones own POV. In other words soapboxing should be discouraged. David D. (Talk) 15:15, 25 April 2007 (UTC)

I believe in a balanced total collection of responses, but firmly believe we should allow a biologist to only give the biological POV, if they wish, and let the psychologist only give the psychological POV, if they wish. If they feel confident enough to stray from their own area, that's fine, but they shouldn't be forced to do so. StuRat 05:56, 26 April 2007 (UTC)

@ Friday - I'm not talking about debating the world at large, but the question under discussion on the Ref Desk. The undeniable reality is that we're not robots but humans. Humans talk about things. If someone asked "Does 1 plus 1 equal 2?", it would be fine if someone answered with "Yes" and nobody else said anything at all. But the very fact that someone would ask such a basic question suggests that they are searching for a deeper understanding of why it is so. It's not inappropriate to proffer some thoughts on the basic concepts involved. And it's not inappropriate for other editors to pipe in with related thoughts. Even with such a question, difficult concepts can arise, and opinions start to vary. You might well argue that any further discussion would be going well beyond the bounds of what we're supposed to be here for, and going well beyond the question, into irrelevant and complex areas of mathematics and logic that a person who asked such a question might be unlikely to comprehend. The question then becomes: where does the line get drawn, and more importantly, by whom? I'm certainly not advocating that all Ref Desk questions should be followed by a long discussion - but equally, I'm never going to be the one to say "Guys, this debate has gone on long enough". When I said "discussion of opposing opinions is what the Ref Desk [not "a" ref desk, btw] is all about", I meant that as a description of the way our Ref Desk actually operates (and imo operates very well in the vast majority of cases) - but I never intended that to be written into the guidelines. JackofOz 01:57, 27 April 2007 (UTC)

Many questions are not true vs false. I would (for example) never give a definitive answer on the causes of psychosis. Even if a brief answer, I would mention that the great majority but not quite all psychologists agree. it's only fair that the person asking a question gets some idea of the degree of certitude. If you know there is anything controversial, you have to say so. Example: Q. "are UFOs real?", A." nobody really knows. Most scientists think not, but here's our summary page UFology and you'll see all the different things people have to say about it." -- I may think the odds of them being real are about 0.1% and you may thing 50%, but we should give the same answer. Even if you thought 99%, you can say "I personally think they are, but I know most scientists say otherwise. Here's our page..." DGG 02:25, 27 April 2007 (UTC)

Discussion

In my opinion what is at issue here is the removal of potentially dangerous and not cited material, and responses for which a source is requested but the editor refuses to provide one. Framing these issues as the removal of "wrong" answers really doesn't move us any nearer a resolution.—eric 08:09, 25 April 2007 (UTC)

I am proposing changing it to: "However, do not remove the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources". This makes it clear that this can't be the sole reason for removal, but something can be removed for other reasons, like being dangerous. However, answers with uncited sources are allowed, so not having sources isn't a justification for removal, either. StuRat 14:56, 25 April 2007 (UTC)

I'd rather see us say what to do than what not to do, but this may be small beer. Friday (talk) 14:58, 25 April 2007 (UTC)

"The threshold for inclusion in Wikipedia is verifiability, not truth." ?

• I oppose adding this to these guidelines, as the threshold for inclusion on the Ref Desk is definitely truth. If I state that 1234 - 3 = 1231 in response to a math question, that should be included, because it is true, whether I can find any reliable sources which state this or not. Conversely, if I find a verifiable reliable source that states it is some other answer, I should not include that. However, in neither case should an answer be deleted. StuRat 21:17, 24 April 2007 (UTC)

• Pillar issue. Take it up on Wikipedia talk:Verifiability if you must. Friday (talk) 21:23, 24 April 2007 (UTC)

• There you go again, back to the same old misdirection. The issue isn't whether the policy itself is bad, it is whether it applies to the Ref Desk, or just to articles. StuRat 21:28, 24 April 2007 (UTC)

Essentially saying the ref desk is not a part of Wikipedia. The community takes a very negative view of Wikiprojects which attempt to form their own little domain away from the encyclopedia proper. As far as questions surrounding 'truth' and 'verifiability' go, there's loads of discussion on the WP:V and WP:ATT talk pages.—eric 21:31, 24 April 2007 (UTC)

It says no such thing, that's just more misdirection. Many policies on Wikipedia are written for articles only. That doesn't mean that non-article pages aren't part of Wikipedia. StuRat 21:36, 24 April 2007 (UTC)

For what it's worth, the rest of the guideline already makes it pretty clear that we don't just throw core policies out the window because it's the reference desk. Verifiability is already explicitly mentioned as applying to the ref desk. Friday (talk) 21:34, 24 April 2007 (UTC)

Not to the extent of tossing out true things which aren't verifiable and including verifiable lies, which is what this says we should do. StuRat 21:38, 24 April 2007 (UTC)

Newbies raise that exact objection to verifiability all the time. Look through the history for hundreds of standard responses to this objection. Or just go read the actual policy- it explains what is meant in more than just a couple sentences. Friday (talk) 21:43, 24 April 2007 (UTC)

It is true that policies were written with articles in mind. All of Wikipedia exists with articles in mind- the articles are the important stuff. The rest is overhead. However notice that the five pillars are described as defining Wikipedia's character rather than defining the character of the articles. As has already been agreed on and documented on many pages, policies are important in places such as talk pages as well as in articles. We don't need to start completely from scratch in describing the character of the ref desk- they're part of Wikipedia. Friday (talk) 21:51, 24 April 2007 (UTC)

Perhaps for article talk pages, but certainly not all Wikipedia policies apply to all talk pages. It would be absurd to suggest you can't put anything on your own talk page unless you have a verifiable reliable source for it, for example. StuRat 22:09, 24 April 2007 (UTC)

Exactly, It would not be encouraging "tossing out true things" and "including verifiable lies" any more than the articles do that. Besides, we have already esablished that there are cases where deviation for core policies may be required on the Ref Desk. In the rare case that a reliable source is incorrect, there is nothing wrong with replying "reliable source X says blah blah blah, however I have reason to believe that to be factually incorrect because, by my calculation, blah blah blah." I don't believe anyone would find justification in our guidelines for deleting that sort of statement, because it is at the very least providing useful information for the OP (the source: verifiable if not true) and perhaps also providing some context by calculation, so the OP can make their own decision on how reliable it is.

I really like eric's version B and would support it as is. However, I also think there is room for a sentence discouraging the deletion of others' answers, just because one considers them to be wrong. Rockpocket 22:00, 24 April 2007 (UTC)

I put in a sentence discouraging removal. It would need to be tweaked I think. I'm still not convinced we need to address this issue, but I can't see how it would hurt anything either. Friday (talk) 22:23, 24 April 2007 (UTC)

You put in "We generally discourage editors from removing answers they think are wrong - outright removal of a response should be done only in exceptional cases". This is definitely better than nothing, but still seems to say that there are cases where an editor can remove an answer solely because (they believe) it's incorrect. I don't believe that the majority here think that's appropriate, but that evidence showing why the answer is wrong should be added, instead. StuRat 06:37, 25 April 2007 (UTC)

I agree with StuRat here. I don't believe there is any justification from removing an answer solely because it is incorrect. That said, there may be good reasons for removing an incorrect answer (and that it is incorrect could be a contributing factor e.g. a suggestion a young OP drink bleach because its harmless). Can you think of a way of tweaking the phrase to address this? Rockpocket 06:52, 25 April 2007 (UTC)

How about if we tweak the line I had: "However, do not remove the incorrect answer, solely because it is wrong, even if you can prove so" ? This makes it clear that a wrong answer can be removed for other reasons (like being dangerous). StuRat 14:33, 25 April 2007 (UTC)

I don't like the "even if you can prove it". We don't want to encourage original research. I think the version that's there now is decent. Friday (talk) 14:35, 25 April 2007 (UTC)

I was referring to "proof" using reliable sources. However, I said "proving it" should not be used as a justification for removal, which hardly encourages OR. However, if you want to make it even more clear we could go with: "However, do not remove the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources". StuRat 14:47, 25 April 2007 (UTC)

Are there any guidelines about what constitutes a reliable source? I ask this because we often see edit wars in articles, where one editor cites a source that he or she believes is reliable, but other editors are of a different opinion. It's impossible to disprove a lot of so-called "unreliable sources", but possible to disprove others. On articles, the majority viewpoint more often than not wins the day and the "unreliable" source becomes no longer part of the article's references. In other cases, differing sources are presented in an effort to be NPOV, and readers can make up their own minds. However, on the Ref Desk, it should be possible to have virtually any viewpoints presented - whether they're backed up by citations or not - without deletion. Arguments against the veracity of another's opinion is what works - not deletion of it. JackofOz 05:59, 25 April 2007 (UTC)

Well, there is WP:RS, of course. Personally I don't see quite how reliable a source is to be an issue on the Ref Desk, as long as it is appropriately attributed. Consider a library Ref Desk staffer: when asked for information regarding a scientific subject, it is not their job to critically assess the whether the info in Science is more reliable than Nature. Thats said, they would be expected to distinguish between The Beano and Encyclopaedia Brittanica. If we provide a non-personal source - any source - and then provide a general indication of the type of source (a newspaper, a website, a scholarly journal), then we are providing useful information for the OP. What is not generally useful, is a multitude of contrary personal opinions with no indication of reliability whatsoever. Rockpocket 06:15, 25 April 2007 (UTC)

Also Wikipedia:Attribution/FAQ, tho not policy, is a good read.—eric 07:00, 25 April 2007 (UTC)

Thanks for those links, people. I agree that reliability per se should not be an issue on the Ref Desk. However, it comes into play when one editor disputes another user's answer and cites what they consider a reliable reference to support their case. I have concerns with this, from WP:RS (my highlights) - "Reliable sources are credible published materials with a reliable publication process; their authors are generally regarded as trustworthy, or are authoritative in relation to the subject at hand." What's credible to one editor may not be so to others. And what's authoritative to one editor may not be so to others. Ultimately, in many cases it comes down to individual opinion. JackofOz 13:20, 25 April 2007 (UTC)

This is why we remove ourselves one level and don't assert that a given source is right or wrong. There's a classic example mentioned frequently- if the New York Times reports that the Sun does not rise tomorrow, we can report that they said it. We would also show where other sources claimed that the Times was out of their minds. We don't try to be in the business of making our own determination of the accuracy of what a source says, for at least a couple reasons: 1) we're not experts, so we use sources instead of being the source. Editors can claim to be an expert but they're often lying. 2) The minute we try to be in the business of critiquing the sources, we'll get a bunch of quacks in here giving their own complicated and utterly stupid theories about why everything reported by Nature is wrong. We don't want to waste our time debunking such nonsense. Friday (talk) 14:13, 25 April 2007 (UTC)

Actually, according to the quote provided by Jack, we can't remove ourselves from evaluating each source, as we each are given the responsibility to determine whether a given source qualifies as credible and authoritative. If it isn't, it doesn't qualify as a reliable source, and shouldn't be referenced. I wonder if Wikipedia itself qualifies as an RS, since many doubt if it's credible and authoritative. StuRat 14:25, 25 April 2007 (UTC)

Yes, we can't avoid using judgment about the sources- but we take care how we do this. If Nature has a reputation for being a good source, that's all the further we look into it. We're not in the business of picking apart one of their articles. If another source does this, then it's fair game but not until then. Wikipedia itself is not considered a reliable source for our purposes. But the sources we use should be reliable. If you want to understand core concepts, the best way is to go read the core policies- much of what we're discussing here has been asked and answered hundreds of times before. Friday (talk) 14:30, 25 April 2007 (UTC)

But what if we have no idea what their reputation is ? We also frequently must evaluate a source provided by an OP or other responder. For example, survey results can be evaluated based on the numbers of people in the sample, the margin of error, whether the questions are unbiased, whether the sample group matches the target population in age, gender, race, income level, etc. StuRat 14:41, 25 April 2007 (UTC)

Yes, there is room for endless nitpicking. But we don't want to encourage this here, it's outside our scope. Let other sources do the nitpicking for us. They're allegedly qualified, we are not. Friday (talk) 14:49, 25 April 2007 (UTC)

So, if an OP or responder references a survey which states "80% of the American people support the President on the war in Iraq", we should accept that as fact, even though the sample size was 5 people, if we can't find any other source which examines this "survey" ? StuRat 15:03, 25 April 2007 (UTC)

Of course not. Please, go read what I wrote and the policies people are referring to. All of this is already well understood. We don't have to accept as fact what the sources say. We just say that they said it and avoid asserting that it's true or false. This is the basics of neutral point of view. We'd save all kinds of time if we came into the discussion as editors already well versed in the core concepts of Wikipedia. Is it too much to ask that editors make an effort to come up to speed on these issues before debating the ref desk guideline? Friday (talk) 15:15, 25 April 2007 (UTC)

Why shouldn't we point out an obvious flaw in a survey, like having a sample size of only 5 ? StuRat 15:38, 25 April 2007 (UTC)

Friday, I have to tell you that the interpretations of policies and guidelines, here and anywhere you care to name, are permanently under discussion and re-evaluation. Nothing is set in stone. What is the yardstick for "coming up to speed"? Forgive me if I misunderstand your position, but it sounds like you're saying that Friday's interpretation of what a particular policy means at any given point in time is the only correct one, and everyone else must agree with Friday's understanding, or risk his apparent irritation. JackofOz 06:28, 26 April 2007 (UTC)

(resetting the every-growing indent) My interpretation is certainly not the only useful one. And yes, the rules are not meant to be set in stone, they're meant to be interpreted by reasonable editors. My frustration with StuRat is that after many months of his raising the same objections over and over, he still gives no indication that he has any grasp of core Wikipedia standards. When editors tell him "here's how we handle these situations at Wikipedia", he rejects this, and wants to start from scratch, completely reinventing the policy wheel. We would make much more useful progress if we'd come into a ref desk guideline discussion already understanding the important Wikipedia guidelines. To me it looks like he goes out of his way to prevent useful discussion, by misrepresenting the views of others and trying to sort editors into warring factions. This is either intentional on his part, or he has astoundingly bad communication skills. Maybe I've let my irritation get the best of me in this case, who knows. But I do remember other editors having the opinion at times that StuRat is intentionally being difficult to work with. A pattern of behavior like that is not acceptable conduct for an editor. Friday (talk) 14:01, 26 April 2007 (UTC)

I would say almost exactly the same things about you, that by constantly misrepresenting a question (whether a particular Wikipedia policy applies to the Ref Desk in the same exact way as to articles) as a misunderstanding of the policy itself, or a challenge to that policy, you are trying to quash any discussion. A pattern of behavior like that is not acceptable conduct for an editor, much less an Admin. If this isn't the time and place to discuss which Wikipedia policies apply to the Ref Desk and how they should be applied, then what is ? As for poor communication skills, I've repeated the same thing many times now, but you never seem to understand. You need to work on your listening skills. StuRat 17:34, 26 April 2007 (UTC)

@ Friday - "they're meant to be interpreted by reasonable editors" - I'd be happier if you said "they're meant to be interpreted in a reasonable manner". Suggesting that Stu, or anyone, is not "a reasonable editor" is, imo, personalising the issue too much.

@ Both of you - your claims and counter-claims of misrepresentation fail to assume good faith on the other's part. It seems you're both bringing in stuff from your history with each other, and allowing it colour the discussion of this topic. I believe it's best to enter into every discussion with a clean slate and with "beginner's mind" in relation to other editors. Have your private opinions of them by all means, but respond to their posts as if you'd never had anything to do with them before. Easy to say, harder to achieve - but you might be surprised how effective it can be in having useful and positive debates that proceed more quickly than might otherwise be the case. JackofOz 02:56, 27 April 2007 (UTC)

That's good advice, and I will try, Jack. I'm certain I can avoid starting an argument, but must admit it is far harder for me to walk away from one started by another. StuRat 06:15, 27 April 2007 (UTC)

Why we should not be supporting the concept of "freedom of expression"

This is the sort of utter nonsense [22] that a guideline supporting freedom of expression would legitimise. Plain wrong in so many ways, and considering the potential damage it could do if taken at face value by the OP, I can think of no good reason whatsoever one could argue against its removal except for "freedom of expression". Rockpocket 00:21, 25 April 2007 (UTC)

So you're so damn afraid of the bad things that could happen that you don't even bother to take the time to look the other way and see the wonderful things that inclusionism could produce. By the way, this diff that you are linking to is utterly legitimate and definitely pertinent: it was completely on-topic. In fact, I really don't know why something like that would be considered plain wrong! I thought at first that you would have linked to something rather difficult to justify, like someone posting a photograph of another editor naked or something (a case that would truly be hard to deal with), but that edit right there is just plain normal. The guy is just saying his opinion on the matter, as all other editors are. I for one didn't find it particularly interesting or instructive, but I won't ever argue that just because something isn't interesting for me it should be deleted. A.Z. 03:33, 25 April 2007 (UTC)

As long as posts agree with WP:CIVIL/WP:NPA, there shouldn't be a problem --h2g2bob 07:49, 25 April 2007 (UTC)

Stuff like that is one reason many of us are hesitant to encourage giving our own opinions here. If we're talking about which brand of toaster we like, it's less likely to upset anyone. But when you get into a sensitive topic, and you get rants like this. I too have seen a couple of people repeatedly assert that all opinions are welcome here and any opinion is as good as any other. This view is not compatible with Wikipedia. If we were running a forum, I might almost agree. But, we're a reference work- some random editor's opinion on who should be maimed or killed and in which specific way has no encyclopedic value. You want a forum, the Internet is over there, and it's full of them. Go find one. Nobody's trying to tell anyone what they can't do- we're only saying that Wikipedia has a scope, and certain things are outside of it. Friday (talk) 14:20, 25 April 2007 (UTC)

Where are we on wrong answers?

I'm reading (skipping through) the discussuion above and there has been some flux in the guidelines during the discussion. Are we ALL gravitating to the b version that is currently on the guideline page? David D. (Talk) 15:05, 25 April 2007 (UTC)

I don't think so, I know I'm not, and I haven't seen anyone else say they are. StuRat 15:14, 25 April 2007 (UTC)

OK so i need to read all of above, thanks for the reply. And sorry for being lazy and asking for a litmus response. David D. (Talk) 15:16, 25 April 2007 (UTC)

That's OK, by why don't we just ask if they prefer version a or b ? StuRat 15:27, 25 April 2007 (UTC)

a) Everyone should attempt to give correct answers, but a few wrong answers are inevitable. If you believe your own answer is wrong, please strike it out and add a comment as to why you no longer think the first answer was correct. If you think somebody else's answer is wrong, add a comment saying so, why you think it is wrong, and provide evidence, if you have any. However, do not remove the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

b) The threshold for inclusion in Wikipedia is verifiability, not truth. While we should keep the verifiability policy in mind while answering, a strict application is not always efficient or useful for some questions. If you believe a response should provide a reference but does not, exercise due diligence before challenging the edit. Check Wikipedia's articles for conflicting or supporting information. Search out your own sources which could confirm or deny the assertion. Use common sense when challenging a response on the desk, present your own sources as a response to the question or politely ask for references. If one of your own responses is challenged don't take it personally—it is after all the Reference Desk—provide a source or reconsider the response. If you cannot find anything to support you edit then indicate such on the desk itself: strike out, remove, or add additional clarification to your response.

I prefer version a

• I prefer A. We need to establish which Wikipedia policies apply to the Ref Desk and how they apply. With respect to deletions, the Ref Desk behaves more like a talk page than an article. StuRat 15:27, 25 April 2007 (UTC)

I prefer version b

• I like B better. We don't need to reinvent core policies here. Friday (talk) 15:21, 25 April 2007 (UTC)

Discussion

Please stop trying to turn this into a vote. It's not helpful. Friday (talk) 15:28, 25 April 2007 (UTC)

How is it not helpful to know which version the majority prefers ? StuRat 15:32, 25 April 2007 (UTC)

Why not work on a solution that's acceptable to all instead? Is there a way to mix the two versions? I've worked on "your" version, are you willing to work on version b? Friday (talk) 15:37, 25 April 2007 (UTC)

If you remove the first sentence and last two sentences, the rest of version B is fine in addition to version A, but not alone, as it fails to say you shouldn't delete things just because you think they're wrong. StuRat 17:04, 25 April 2007 (UTC)

The first sentence came directly from a core policy. Friday (talk) 17:09, 25 April 2007 (UTC)

I don't have a problem with the first sentence in B. It is a statement of established Wikipedia policy and the following sentence makes it clear that the Ref Desk should adhere to that in general character (as the core policy suggests), but not always strictly. As for the last two sentences, whats not to like - seems like good advice to me? I'd like to see a merge, basically version B with two sentences from A (in italics)

The threshold for inclusion in Wikipedia is verifiability, not truth. While we should keep the verifiability policy in mind while answering, a strict application is not always efficient or useful for some questions. Everyone should attempt to give factually correct answers, but a few wrong answers are inevitable. If you believe a response should provide a reference but does not, exercise due diligence before challenging the edit. Check Wikipedia's articles for conflicting or supporting information. Search out your own sources which could confirm or deny the assertion. Use common sense when challenging a response on the desk, present your own sources as a response to the question or politely ask for references. However, do not remove the incorrect answer, solely because it is wrong. If one of your own responses is challenged don't take it personally—it is after all the Reference Desk—provide a source or reconsider the response. If you cannot find anything to support your edit then indicate such on the desk itself: strike out, remove, or add additional clarification to your response.

Rockpocket 17:52, 25 April 2007 (UTC)

The compromise version seems good to me. Friday (talk) 17:55, 25 April 2007 (UTC)

I wouldn't argue with that. (Yay compromise!) TenOfAllTrades(talk) 18:02, 25 April 2007 (UTC)

Let me list my specific reasons for objecting to those lines. The first line states that we aren't interested in the truth. This might be the case on Wikipedia articles, but we are most definitely interested in the truth here. Many true, but unverifiable, statements (like math solutions) are commonplace on the Ref Desk presently. The last two lines (one and a half, actually) seem to imply that any answer which lacks sources must be withdrawn, if challenged. This should not be implied. Going back to the math example, I shouldn't be asked to strike out, remove, or explain a math solution, just because I can't find a reliable source which supports it. StuRat 05:45, 26 April 2007 (UTC)

You are misrepresenting what it says: It does not say "we aren't interested in the truth". We should aim to verify our material whenever possible. When it is not possible (like in your example), "efficient or useful" then its not required. Would you be happier if opening sentence was more explicit? How about

"The threshold for inclusion in Wikipedia articles is verifiability, not truth. While we too should keep the verifiability policy in mind while answering, a strict application is not always efficient or useful for some questions."

As for the last sentence, I don't read that as a demand, rather as good advice. The only clarification one would expect in your example would be a "By my calculation..." We should always be making sure we inform the OP of what type of information we are providing (sourced, opinion, calculation, anecdote). If that information is already there in the original response, there would be no need to "strike out, remove, or add additional clarification." To address your concern, how about:

"If you cannot find anything to support your edit then consider indicating such on the desk itself: strike out, remove, or add additional clarification to your response.

Rockpocket 06:12, 26 April 2007 (UTC)

Why say "the standard for inclusion isn't truth", when this is definitely an important goal of the Ref Desk, to give truthful answers ? To say it and then try to unsay it subsequently makes no sense. Just don't say it in the first place. And I reject the notion that a responder should be told to do anything with their answer if subsequently asked to provide sources. They can, if they wish, but why should we imply that they then have some responsibility to "strike out, remove, or add additional clarification" ? If their response remains unsourced, so be it, the OP can then decide which answer to believe. What I'm particularly worried about is having user X, who dislikes user Y, challenge all of user Y'a answers as a means to "get even". The more work we ask Y to do following a challenge, and the less work we ask X to do before issuing a challenge, the more effective this "vengeance challenge" will become. StuRat 17:10, 26 April 2007 (UTC)

Ok, were are tying to be constructive here. Not obstructive. There will always be examples of situations where guidelines could be abused, but we have to draft them with the OP in mind, not the prtection of the person answering.

"Truth" is subjective. Your truth and my truth might be very different. Independent verifiability is not subjective (although the reliability of any verification is) - this is why it is the "standard for inclusion" for encyclopaedic content. No matter how much you might dislike this - it is an absolute and fundamental core policy that simply is not going to change. So where exactly is the problem? If it is simply a concern that you think we are discouraging "truth"? If so, we could aways remove that:

The threshold for inclusion in Wikipedia is verifiability. While we should keep the verifiability policy in mind while answering, a strict application is not always efficient or useful for some questions.

I don't see that being a big deal. If your issue is that you don't believe we should keep the "verifiability policy in mind while answering", then we may as well stop this discussion now, because we clearly have polar opposite views on what the Ref Desk is for and are not going to agree. As for the last sentence: It does not "tell" anyone to do anything, it now asks you "consider" how you can assist the OP in deciding what information is accurate (which is surely what our goal is). Again, if your answer is preceded by "In my opinion...", "In my calculation..." or "I don't have a source, but my understanding is..." no sources, clarifications or strikeouts are required. If it is not, and you don't chose to strike or source, going back and adding that clarification is all that is required, and that would be extremely helpful to the OP in determining the value of the information. How can this be abused for "vengeance challenges" because if the original response informs the OP of which type of answer it is (as it should), there is nothing to challenge. We could always add a sentence indicating this:

"If you cannot find anything to support your edit then consider indicating such on the desk itself: strike out, remove, or add additional clarification to your response to better assist the querent in determining the reliability of your response.

What do you think? Rockpocket 17:46, 26 April 2007 (UTC)

I think it's a bit silly to ask people to say they don't have a source. Doesn't the lack of a listed source already show that they don't have a source ? I think the Ref Desks are quite long enough already without requiring that we all state the obvious. StuRat 18:59, 26 April 2007 (UTC)

Again, that was an example, Stu. There is nothing in the guideline that says you have to say that. The point is that it is helpful to the OP to give some indication of where your answer is coming from. Say I ask, "how do rats and mice get on with one another?" and someone makes a statement in answering this question (e.g. "Rats and mice co-habit well"). Now its not clear from the answer if it is a personal opinion, a gut feeling, a calculation from personal experience, a recounted fact (even if you don't know the exact source of the fact) or a sourced fact (though you have not provided the source). If the responder was to preface their response with which of these types of answers it is, then then it would be much more valuble to the querent. Without such a preface, it is entirely reasonable to "challenge" an answer you consider to be questionable, because without some type of clarification, it could be highly misleading to the OP. Once that clarification is made there is nothing left to query: either it is now sourced or it is clear that a source is not required. Problem solved. I fail to see how this is anything but a good thing.

You didn't respond to my question about the crux of the "truth" problem, and I have another question for you, what exactly is the problem with the last sentence - are you against encouraging responders to state what type of information they are providing? If so, why? If not, then what is the issue? If we can get to the bottom of this, we should be verry close to getting consensus, so please continue to work constructively to improve, rather than simply pointing out flaws. Rockpocket 19:30, 26 April 2007 (UTC)

I am opposed to including verifiable info that you know is incorrect (like using Bush's WMD claims as "proof" that they existed). I also don't think we should tell editors they should consider striking out or removing their answer if they can't find a source. They should only strike out their answer if they believe it to be incorrect, regardless of whether they have a source. My recommendation for improvement is that we omit those lines. StuRat 19:53, 26 April 2007 (UTC)

Response based on standard application of core Wikipedia concepts (verifiability and neutral point of view): Bush's claims do not verify an assertion that WMD's really existed in some place at some time. It only verifies that Bush said what he said. Perhaps this is a subtle point, but it's vitally important to understanding what we do here at Wikipedia. Neutrality requires that we don't assert the truth or falseness of claims this way- we simply describe who said what, according to who. This is an example of a time when I think a proper understanding of existing Wikipedia guidelines solves our problems for us. Friday (talk) 20:42, 26 April 2007 (UTC)

If someone asked if WMD have been found in Iraq, and you responded with "Bush said there are" and references to prove that, this would be most unhelpful to answering the Q. A far better response would be "No, no substantial quantities of WMD have been found in Iraq, to this date", whether or not you had a source for that statement. StuRat 21:17, 26 April 2007 (UTC)

Sorry to butt in, guys, but I completely fail to see why you debate paragraphs a) and b) as alternatives to choose one of. They have separate subjects – a is about truth, b is about verifiability – both of which are worth giving guidance on. So I humbly suggest that a solution to the disagreement is to have two paragraphs, one about truth and one about verifiability. Here is a humble attempt:

c) Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is wrong, please strike it out and add a comment as to why you no longer think the first answer was correct. If you think somebody else's answer is wrong, add a comment saying why you think it is wrong, and provide evidence if possible. However, do not remove the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

Verifiability is a core policy of Wikipedia, and reference desk answers should be sourced whenever possible. There are some questions, however, such as those involving straightforward mathematics calculations, where a strict requirement of verifiability would not be efficient or useful. If you believe a response should provide a reference but does not, feel free to add comment to that effect. If somebody asks for a reference for one of your own responses, please try to provide one or indicate that you cannot.

(The above is an edited version of my first post a few minutes ago: pure requests for references cannot be sourced, only challenges to truth can.) --mglg_(talk) 18:25, 26 April 2007 (UTC)

That's a distinct improvement, mglg, but I feel the part you suggested then removed is needed, too, so here's my version:

c) Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is wrong, please strike it out and add a comment as to why you no longer think the first answer was correct. If you think somebody else's answer is wrong, add a comment saying why you think it is wrong, and provide evidence if possible. However, do not remove the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

Verifiability is a core policy of Wikipedia, and reference desk answers should be sourced whenever possible. There are some questions, however, where a strict requirement of verifiability would not be efficient or useful. If you believe a response should provide a reference but does not, feel free to add comment to that effect, and, if you also challenge the answer given, if possible source your challenge with the same care that you are requesting of the original responder. If somebody asks for a reference for one of your own responses, please try to provide one or indicate that you cannot.

I also took out the math example, as that might imply that only math answers are unverifiable. StuRat 18:52, 26 April 2007 (UTC)

My main point was to separate truth from verifiability. Thus I do feel that discussion of challenges to the truth of an answer, as opposed to its lacking references, logically belongs in the truth paragraph, not in the verifiability paragraph. Also, I dislike the idea of discouraging challenges by redundantly referring twice to providing evidence or sources for them, once in each paragraph. A false answer is much worse than no answer, and if somebody asks for a reference just because their b.s. alarm goes off, that is a good thing. For both these reasons I argue that the discussion of providing evidence and sources for challenges should be done only once, in the truth paragraph. How about this version, where I flip the paragraph order so that we can logically refer to providing sourced evidence in the truth paragraph:

Verifiability is a core policy of Wikipedia, and reference desk answers should be sourced whenever possible. There are some questions, however, where a strict requirement of verifiability would not be efficient or useful. If you believe a response should provide a reference but does not, feel free to add comment to that effect. If somebody asks for a reference for one of your own responses, please try to provide one or indicate that you cannot.

Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is wrong, please strike it out and add a comment as to why you no longer think the first answer was correct. If you think somebody else's answer is wrong, add a comment saying why you think it is wrong, and provide sourced evidence if possible. However, do not remove the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

--mglg_(talk) 20:14, 26 April 2007 (UTC)

That looks pretty good, I like it. One minor nit-pick, you changed "provide evidence" to "provide sourced evidence". Not all evidence is sourced, like when showing that a previous response contained a math error. StuRat 21:24, 26 April 2007 (UTC)

Reconsideration

In response to the concerns of some editors, I have tried to promote consensus by merging and modifying the proposals A and B. I think, taken as a whole, is addresses the idea of how we should deal with differences in opinion over answers. I don't think the issue of "truth" - and the subjectivity that involves - should be dealt with seperately. I now offer this for reconsideration:

The threshold for inclusion in Wikipedia is verifiability. While we should keep the verifiability policy in mind while answering, a strict application is not always efficient or useful for some questions. Everyone should attempt to give factually correct answers, but a few wrong answers are inevitable. If you believe a response to be incorrect, or should provide a reference but does not, exercise due diligence before challenging the edit. Check Wikipedia's articles for conflicting or supporting information. Search out your own sources which could confirm or deny the assertion. Use common sense when challenging a response on the desk, present your own sources as a response to the question or politely ask for references. However, do not remove the incorrect answer, solely because it is wrong.

If one of your own responses is challenged as being factually incorrect don't take it personally—it is after all the Reference Desk—provide a source or reconsider the response. If you cannot find anything to verify your edit then consider indicating such on the desk itself: strike out, remove, or add additional clarification to better assist the querent in determining the reliability of your response.

Rockpocket 20:24, 26 April 2007 (UTC)

Do we really want to have four sentences discouraging people from pointing out incorrect answers? Why should the guideline put more emphasis on due diligence for challenges that for initial answers? Is that conducive to providing correct information? --mglg_(talk) 21:03, 26 April 2007 (UTC)

All very good points. I think the very same due diligence should be encouraged before providing any answer (i.e. to encourage some effort being put into providing the correct answer rather than providing an answer). I think the point of this is to discourage unconstructive arguments, debates and discussion. Its very easy to just say "your wrong", but that doesn't help the querent anymore than the wrong answer does. I think everyone should be providing reliable attribution (or, in the absence of that, an explanantion for the source of the info), but if you plan to call out some one else for being wrong, its particularly important you can verify why you believe so. Rockpocket 21:20, 26 April 2007 (UTC)

I don't care for this version at all. I've already discussed most of the reasons. This is going backwards from the far better mglg version. However, I have no objection to inserting the "please use due diligence before challenging any edit" text into mglg's version. StuRat 21:31, 26 April 2007 (UTC)

Here it is with that line added:

Verifiability is a core policy of Wikipedia, and reference desk answers should be sourced, whenever possible. There are some questions, however, where a strict requirement of verifiability would not be efficient or useful. If you believe a response should provide a reference, but does not, feel free to add a comment to that effect. However, please use due diligence before challenging any edit. If somebody asks for a reference for one of your own responses, please try to provide one or indicate that you cannot.

Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is wrong, please strike it out and add a comment as to why you no longer think it's correct. If you think somebody else's answer is wrong, add a comment saying why you think it is wrong, and provide evidence, if possible. However, do not remove the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

I'd like to see "due diligence" as a link to another page that explains what this means in the Ref Desk context (using much of RP's material), not the current generic Wikipedia article. StuRat 21:44, 26 April 2007 (UTC)

I'm happy to endorse mglg's version, however i'm somewhat perplexed at your support for it considering one of your reasons for disliking the other version was

"I think it's a bit silly to ask people to say they don't have a source. Doesn't the lack of a listed source already show that they don't have a source?"

Tet mglg's, which you like, asks for the same thing:

...If somebody asks for a reference for one of your own responses, please try to provide one or indicate that you cannot.

Still, I'm pleased that I was able to talk you around on this particular issue and will support this version ;) Rockpocket 22:04, 26 April 2007 (UTC)

There's a difference, you seemed to be asking them to state this up front, in which case it serves no purpose, while mglg's version only asks that you respond to a request for a source. At that point, saying that "I looked but couldn't find a source" or "I don't have time to look" are both new info which could be of value to the OP or other responders. StuRat 22:14, 26 April 2007 (UTC)

Actually, I was suggesting it as an example of clarification in response to a challenge if it wasn't noted upfront (because if it was noted upfront there would be little point requesting a reference). I certainly don't think it is silly to note the reason for lacking sources upfront though. You seem to recognise the info "could be of value to the OP or other responders", so why wait to be challenged? Why not start an an unsourced answer with "I looked but couldn't find a source, however I think..." or "I don't have time to look for a source, but my understanding is..."? It can't hurt and it can be of great help to the OP in judging the reliability of different answers. Anyway; no harm, no foul. Lets move on an see if others are willing to support this version. Rockpocket 22:32, 26 April 2007 (UTC)

Presumably the majority of people don't provide sources because they don't think any are needed. That certainly is my reason. If you want to challenge my assertion that no substantial WMDs have been found in Iraq, I suppose I could find sources, but why should I ? You prove your case, that they were found. This gets to the "due diligence" part. If the challenger did provide some evidence of their POV, then I would actually make an effort to do so, too. In any case, am I correct in thinking that you agree with the version directly above, with the "due diligence" line added ? StuRat 22:53, 26 April 2007 (UTC)

Why wouldn't a source be needed? I know you are a smart guy with a remarkable breadth of knowledge. I know that because you are usually reliable in the accuracy of your answers on subjects I know a lot about. Sure, you sometimes miss overlook the subtleties and occasionally generalise, but overall your answers are very good for a non-expert. I assume that this level of accuracy is consistant across subjects I don't know anything about, and thus give your answers some credibility in the absence of a source.

So you know you are right, and I might think you are right, but how is some random IP querent on their first visit to the Ref Desk supposed to know that? They are presumably asking about something they don't know the answer to, so for all they know your unsourced answer could be utter nonsense or could be 100% correct. So whats the goal here: the OP getting the best info or every responder getting a chance to have their say? If the goal is to help and inform OPs, the point of providing sources should not be thought about in terms of a requirement for fending off a challenge, but about providing the best info for the OP to make their own decision. Sources aren't always needed, but they almost always help.

That said, I do support the version including the due diligence part, but if we are including that, I would also like to see similar advice that all responders do due diligence before offering an answer - that is the best way to ensure our answers are as correct as possible. Rockpocket 23:37, 26 April 2007 (UTC)

Sorry for interjecting when you two are actually approaching agreement, but I hate the "due diligence" phrase – it is abstract and unclear now that it is not followed by (the previous too lenghty) explanation. Why not say what we mean here rather than in a linked page? I also agree with Rockpocket that at least equal caution about providing sources should be directed to original responders as to challangers. How about this:

Verifiability is a core policy of Wikipedia, and reference desk answers should be sourced, whenever possible. Even if you know the answer, try to provide supporting references to Wikipedia or elsewhere. There are some questions, however, where a strict requirement of verifiability would not be efficient or helpful. If you don't have a quotable reference, it is useful to indicate why not, or where you got the information, to help the reader judge the reliability of your response. If you believe a response should provide a reference, but does not, feel free to add a polite comment to that effect. If somebody asks for a reference for one of your own responses, please try to provide one or indicate that you cannot.

Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is wrong, please strike it out and add a comment as to why you no longer think it's correct. If you think somebody else's answer is wrong, add a comment saying why you think it is wrong, and provide evidence, if possible. Make a serious effort to locate supporting sources, as you would in an original answer. Do not remove the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

--mglg_(talk) 00:27, 27 April 2007 (UTC)

Very nice work, mglg. I endorse this version. Rockpocket 00:33, 27 April 2007 (UTC)

Having just woken up and come in at the tail end of this long discussion:

• I also despise the stock phrase "due diligence" and hope it remains out of our guidelines (particularly as the once-proposed link tells us virtually nothing about what it means in our situation, or how to apply it).
• It seems to be implicit, but it might be useful to state explicitly that the only person who is permitted to strike out a post (or any part of a post) is its author. (This is a corollary of the general principle I adopt, that of not interfering with another person's edits, except where eg. a leading space leads to it being difficult or impossible to read; to add a question title; or in a really extreme case that might warrant the post being entirely removed). If a wrong statement has been disproved, it's still the author's choice whether to strike it out, or leave it and allow the flow of the discussion to remain for the benefit of the OP.
• I don't know how strict we should be about an editor always trying to find a suitable reference to support their answer, particularly if it's the initial response to the question. Some OPs do ask for sources to help them with their own research, but the majority of OPs come here expecting us to be the source of the answer. If we can provide a good answer, and nobody else disagrees with it, it shouldn't generally be a requirement to provide an external source for it. Linking to Wikipedia articles is always a good idea, but again, this shouldn't have to be mandatory.
• Otherwise, I support the text above. JackofOz 01:19, 27 April 2007 (UTC)

• Jack, I think you misunderstood the due diligence link. While it currently links to the Wikipedia article on the legal concept, I was proposing that we change the link to point to a guideline subpage which would define what we mean in the Ref Desk context. Much of it would be what RockPocket said. That said, I'm also fine with us explicitly describing it without the link, and omitting that hated term, if you wish. I agree that this last version puts too much emphasis on providing sources. Most of our readers don't care about sources, as you've said. Also, any blatantly wrong answers will be quickly commented upon, with the correct answers provided, so the risk of an OP going away with the wrong answer is low. My main reason for not including sources is simply time. I can answer 100 questions (with 99 of them correct) in the time it might take to answer 10 with full sources. Are the 89 unanswered questions worth the 1 incorrect answer which I might spot if I looked up full sources ? I also rarely know where I got my info, I just know it. I find it hard to believe that others recall the exact source of every bit of info in their mind. StuRat 03:24, 27 April 2007 (UTC)

So, how about this:

Verifiability is a core policy of Wikipedia, and reference desk answers should be sourced, whenever possible. There are some questions, however, where a strict requirement of verifiability would not be efficient or useful. If you believe a response should provide a reference, but does not, feel free to request one. However, please have a good reason before challenging any edit. If somebody asks for a reference for one of your own responses, and you can't provide one, it is useful to indicate why not, or where you got the information, to help the reader judge the reliability of your response.

Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is wrong, please strike it out and add a comment as to why you no longer think it's correct. If you think somebody else's answer is wrong, add a comment saying why you think it is wrong, and provide evidence, if possible. However, do not remove the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

Any better ? StuRat 03:34, 27 April 2007 (UTC)

Yes, definitely better, Stu. Thanks for the redirect about due diligence - I'll suspend further comment until such a sub-page is developed. I'd also ask that these further minor changes be considered:

Verifiability is a core policy of Wikipedia, and reference desk answers should generally be sourced, whenever possible. There are some questions, however, where a strict requirement of verifiability would not be efficient or useful. If you believe If a response does not should provide a reference, but does not, feel free to request one. However, please have a good reason before challenging any edit. If somebody asks for a reference for one of your own responses, and you can't provide one, it is useful to indicate why not, or where you got the information, to help the reader judge the reliability of your response.

Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is wrong, you may wish to please strike it out and add a comment as to why you no longer think it's correct. If you think somebody else's answer is wrong, add a comment saying why you think it is wrong, and provide evidence, if possible. However, do not remove or strike out another editor's the incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

I'm not sure I'm entirely happy with "solely because it is wrong, even if you can prove so with reliable sources". This suggests there could be other good reasons to strike out or delete another's posts. I believe there are almost no good reasons to remove another's post, and no good reasons at all to strike out another's post. But this is a minor disquiet in the scheme of things, and if we wait for perfection we'll be waiting forever. JackofOz 04:10, 27 April 2007 (UTC)

version m

I agree completely, even including the reservation. Here is a clean copy of your version, with some extremely minor tweaks (like contractions):

(Moved to bottom of this page to get more comments.) StuRat 21:40, 28 April 2007 (UTC)

Generally yes, with one concern regarding the wording: If a response doesn't provide a reference, feel free to request one. However, please have a good reason before challenging any edit. " I'm not sure quite what that means; what is a "good reason" for challenging? As far as I'm concerned there is "good reason" to ask for a reference to back up any unreferenced statement of fact that could be reasonably considered questionable (that good reason being I would like to establish the material is correct and verifiable, not just the bullshit of some cocky knowitall). I presume this statement is meant to discourage procedural requests for every single, non-controversial statement (e.g. "Some dogs are black", "Do you have a source for that?") If that is the reason for it being there, I would prefer it be reworded along the lines of

If a response doesn't provide a reference, feel free to request one. However, do so for clarification or verification, do not ask just to illustrate a point.

Rockpocket 18:53, 27 April 2007 (UTC)

How would that prevent people from asking for a source for every response ? They would just say "I wanted clarification or verification". I want them to either have a specific need for a source (like to add to a Wikipedia article or a paper they are writing) or have a real doubt, preferably with evidence, that the response is correct. StuRat 21:51, 28 April 2007 (UTC)

While I agree with the sentiment and understand the concern here, we don't need to reinvent the wheel—someone who goes through every sentence on the board and asks "Do you have a source for that?" is already disrupting Wikipedia to make a point. Going through and asking for a source for every statement made by a particular editor is likely a violation of our guidelines on harassment. In all cases, we expect people to abide by WP:DICK. In other words, we don't need to put all our eggs in this basket; Wikpedia already has policies to deal with editors behaving obnoxiously. As Rockpocket says, we shouldn't discourage people from asking for sources for statements of fact. TenOfAllTrades(talk) 19:11, 27 April 2007 (UTC)

(ec)Agree in principle, w/ certain qualms as to the language:

• Verifiability is a core policy of Wikipedia...; "...verifiability not truth" is, i think more useful, we've already asked contributors to "keep in mind" the core policies. Please see the attribution FAQ for what is, in my opinion, an excellent explanation of what the phrase means to most editors. Something along the lines of: "Ref desk contributors care very much about the truth, but are mindful of their own limitations...". The first part of the policy quote "The threshold for inclusion...", taken literally is not really consonant with the rest of our guideline and should probably be explained.
• ...answers should generally be sourced, whenever possible; Quoted material should always be sourced. Material concerning living persons should almost always be sourced, especially if negative. Responses that could be controversial should usually be attributed or sourced. Generally the information we provide should derive from reliable publish sources, but a citation need not always be provided up front. And we should always consider our responses in light of the verifiability policy before adding them to the desk. Is the above in accord with "answers should generally be sourced, whenever possible"?
• If you think somebody else's answer is wrong...; i dislike language such as: "answer is wrong", "incorrect answer", and "prove so with reliable sources" and would rather see it framed in a way compatible with "verifiability not truth". Basically encourage some research before answering, a little (more) work before challenging, and willingness to reconsider responses and alter or remove them if necessary.
• ...do not remove...; My opinion is that the wiki process is beyond the scope of any guideline document, and we can't forbid any good faith edits. That said, i'll of course conform to any consensus, written into this document or not, that advises on the matter. For the purpose of the guideline, i'll just post a question and see if there is a consensus answer: what happens when and editor continually ignores this guideline, when they add information which conflicts with Wikipedia's articles and all available sources, and they refuse to provide any sources of their own or alter their response on the desk?

—eric 19:15, 27 April 2007 (UTC)

I would think that such a consistently disruptive editor would get invited to not participate anymore. But I don't think the guidelines need to address unusual cases like this. For that reason I also share eric's reservations about prohibiting removal. Friday (talk) 19:21, 27 April 2007 (UTC)

Sorry, guys, but I think there's a rather strong consensus against removing Ref Desk responses just because they are wrong. Just post a correction and let the wrong answer stay. StuRat 02:18, 28 April 2007 (UTC)

If anyone was talking about removing responses for being wrong, your comment might be relevant. Friday (talk) 06:38, 28 April 2007 (UTC)

I thought you and EricR were objecting to the line "However, do not remove or strike out another editor's incorrect answer, solely because it is wrong, even if you can prove so with reliable sources". If not, what are you objecting to ? StuRat 13:53, 28 April 2007 (UTC)

Now now, boys. I think an editor who did what eric mentions would soon get tired and bored of having their statements swiftly rebutted with authoritative references, and go off to make trouble elsewhere (or maybe, just maybe, they might realise what they're doing, achieve enlightenment, and later become the best editor the Ref Desk has ever had). I don't think it's a crime to have an opinion that nobody else shares, but when all of one's opinions are in that category, we're talking about either a genius or a madman/woman. If they persistently refused to source their lonely opinions, then this would be evidence that they're being disruptive for its own sake, which would justify them being blocked from contributing altogether. Until that happens, just removing their posts would be contrary to the principles of tolerance and preparedness to argue one's case. JackofOz 06:53, 28 April 2007 (UTC)

The reference desk is not for arguing one's case. Wikipedia is not a forum. Friday (talk) 13:34, 28 April 2007 (UTC)

A certain amount of "arguing" is a good thing:

Responder 1: "I believe the answer is A."

Responder 2: "That doesn't sound right, I believe the answer is B."

Responder 1: "My answer is supported by X, Y, and Z."

Responder 2: "OK, you may be right."

As long as the argument remains logical, not personal ("you obviously don't know what you're doing here"), and is reasonably short, it's in the interest of the OP to have any questions of accuracy debated and resolved. StuRat 13:53, 28 April 2007 (UTC)

I don't want to argue here, but I feel I should clarify that I consider StuRat's example to be less an argument and more a clarification ;) Rockpocket 19:16, 28 April 2007 (UTC)

If X, Y, and Z represent reliable sources then we could do with a whole lot more of this type of "argument".—eric 04:10, 29 April 2007 (UTC)

It could be that, or could be "you've misplaced the decimal point on your third calculation", etc. StuRat 09:37, 29 April 2007 (UTC)

What the reference desk is not...

I think that there should be a clause saying that the reference desk is not a help desk; that is, it does not answer questions given about the editing of Wikipedia. Questions on the help desk are often given to the reference desk, so how about the guideline saying that questions on it are just for technichal knowledge? -- Casmith_789 (talk) 13:22, 27 April 2007 (UTC)

User:Friday added this:

• The reference desk is not for help using Wikipedia. Try the Wikipedia:Help desk.

That's fine by me. StuRat 15:45, 27 April 2007 (UTC)

Apparently harmless. A.Z. 21:45, 27 April 2007 (UTC)

Verifiability and wrong answers

(Moved from above to get more feedback.)

Verifiability is a core policy of Wikipedia, and reference desk answers should generally be sourced, whenever possible. There are some questions, however, where a strict requirement of verifiability would not be efficient or useful. If a response doesn't provide a reference, feel free to request one. However, please have a good reason before challenging any edit. If somebody asks for a reference for one of your own responses, and you can't provide one, it's useful to indicate why not, or where you got the information, to help the reader judge the reliability of your response.

Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is incorrect, you may wish to strike it out and add a comment as to why you no longer think it's correct. If you think somebody else's answer is wrong, add a comment saying why you think so, and provide evidence, if possible. However, do not remove or strike out another editor's incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

Does everyone agree with this version ? (If not, please list specific changes you would make). StuRat 06:06, 27 April 2007 (UTC)

I do. JackofOz 06:15, 27 April 2007 (UTC)

I do. StuRat 21:41, 28 April 2007 (UTC)

Too long. We don't need to tell editors what to do if they change their own minds, for example. Also it diverges too much from standard project expectation. I'll try making specific changes on the guideline page. Friday (talk) 07:05, 29 April 2007 (UTC)

Yes, I see that your attempt to shorten it involved removing any suggestion that it's improper to remove a response you believe to be incorrect. Needless to say, I put it back in. StuRat 08:08, 29 April 2007 (UTC)

Discussion

Was there something inappropriate about the feedback you received above on this version, or did you just not like it so much?—eric 04:08, 29 April 2007 (UTC)

When it got into discussion of what we were and were not discussing, it seemed to have gotten sufficiently off-track that we needed to get it back on track, with a new section to contain the endless discussion of discussion. StuRat 08:13, 29 April 2007 (UTC)

I refer you to the comments above which you have failed to address. Specifically, I believe we need to clarify what a "good reason" is for challenging. I propose a change:

If a response doesn't provide a reference, feel free to request one. However, do so for clarification or verification, do not ask just to illustrate a point.

Rockpocket 04:59, 29 April 2007 (UTC)

I did respond, here it is again:

How would that prevent people from asking for a source for every response ? They would just say "I wanted clarification or verification". I want them to either have a specific need for a source (like to add to a Wikipedia article or a paper they are writing) or have a real doubt, preferably with evidence, that the response is correct. User:StuRat 21:51, 28 April 2007 (UTC)

We don't make guidelines based on "What if someone is being unreasonable in this specific way?" kinds of scenarios. We have to assume guidelines are interpreted reasonably by editors who are trying to collaborate, not antagonize. Don't be a dick is already assumed. Friday (talk) 08:50, 29 April 2007 (UTC)

We make guidelines like that all the time, such as "no personal attacks". StuRat 09:43, 29 April 2007 (UTC)

If you want to modify Friday's version, though, go right ahead. StuRat 08:13, 29 April 2007 (UTC)

I took a stab at shortening, merging, and reworking the content about wrong answers into a section about verifiability. Some of the content under "wrong answers" was now redundant, so I trimmed it a bit. I still don't really agree with the sentence that remains, though- it seems like an oddly specific instruction that might not always be good advice. Friday (talk) 09:11, 29 April 2007 (UTC)

I see you decimated the paragraph, as follows:

Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is incorrect, you may wish to strike it out and add a comment as to why. If you think somebody else's answer is wrong, add a comment saying why you think so, and provide evidence, if possible. However, do not remove or strike out another editor's incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

Aren't you even going to give any reasons for removing this material, which most have agreed to on this talk page ? Here are my reasons for keeping each line:

Responders should make every effort to give correct answers, but inevitably a few answers will be wrong.

(Editors should be told it's OK if they aren't always perfect. Many newbies are nervous enough without the pressure to always be right.)

If you believe your own answer is incorrect, you may wish to strike it out and add a comment as to why.

(We want to suggest the proper way to deal with this situation. Clio, for one, left her incorrect comments in, apparently thinking no edit was allowed.)

If you think somebody else's answer is wrong, add a comment saying why you think so, and provide evidence, if possible.

(This is specifically to avoid having people challenge answers when they have absolutely no basis for that challenge.) StuRat 09:33, 29 April 2007 (UTC)

The explanation for removing the material is immediately above your comment. As I said, it's now redundant with the section I added. Friday (talk) 19:11, 29 April 2007 (UTC)

It is not redundant. If you believe it to be so, show me which lines say the same things as each line I have included above. For example, where do you instruct people to strike out their own incorrect answers ? StuRat 20:09, 29 April 2007 (UTC)

This is the reference desk guideline, not the introduction to Wikipedia. We don't need to instruct people on various ways to edit their own content. The dispute here is about editing other people's comments. Friday (talk) 20:13, 29 April 2007 (UTC)

So now you've retracted your argument that it was redundant and decided it was not needed, instead ? If we don't want to tell people how to edit their own posts, then we should remove all the lines encouraging people to include reliable sources, too. StuRat 20:18, 29 April 2007 (UTC)

I don't see how there's a retraction involved. I've given a couple reasons why I don't like the remaining sentence and explained it a couple times using different words. The lines about sources are really about verifiability- this is why I reworked the "wrong answers" section into the "factual disputes and verifiability" section. I thought it was more tightly focused on things we should be focused on. I figured the "use common sense" sentence could replace specific instructions on when to remove or not remove. I've attempted again to edit the guideline to address my objection- maybe the results are satisfactory? Friday (talk) 20:36, 29 April 2007 (UTC)

No, you've just managed to weaken it further so now it says removing wrong answers is "not recommended" (but apparently allowed). The majority on this talk page clearly does not want answers to be removed solely because they are wrong, so I've now restored the Wrong Answers section which you've weakened with every edit. It also is not "different words", you claimed that the text you removed was redundant (repeated elsewhere), when it is not. StuRat 05:33, 30 April 2007 (UTC)

I put back the versions a and b thing since there's apparently still no agreement. The "factual disputes and verifiability" was meant by me to replace the "wrong answers" section, but for now both sections are there. I put in the "wrong answers" version I attempted as option c. Friday (talk) 06:34, 30 April 2007 (UTC)

I moved the old "verifiability" paragraph ("b") to the verifiability section (that was a no-brainer). I'm fine with your new verifiability para as an alternative to the old V para, but I'm not fine with you deleting or weakening the para on wrong answers. StuRat 14:27, 30 April 2007 (UTC)

when someone writes "(edit conflict)"

Hi. I've seen this a lot, people often write "(edit conflict)" when they answer a question. Is it used when someone dissagrees, or when they actually experience an edit conflict, or something else? If it's for when dissagreeing, why would it be called edit conflict, and if it's an actual edit conflict, how come I haven't seen "edit conflict" written on other talkpages? Is this in the guidelines, but if not, should it be mentioned? Please mention this as I see it a lot. It would be helpful if I knew what it meant. Thanks. – AstroHurricane001^{(Talk+Contribs+Ubx)(+sign here+How's my editing?)} 23:10, 28 April 2007 (UTC)

It just means the editor tried to "save" their post, only to discover someone else had edited the same page in the meantime. In order not to have to change their post in light of what the other person just posted, they sometimes write "edit conflict" to indicate this new post was written before they saw the one that just preceded it (if that makes sense). It's usually a response to the post 2 threads ago, or earlier, not the one that's just been posted - so it neither agrees nor disagrees with it. It does appear on other talk pages, but not as frequently on any one talk page as it does on the Ref Desk, and on some talk pages not at all. Maybe you've only been reading the latter ones - stick around, you'll encounter some more in due course. It's not in the guidelines afaik, but since it's entirely informal, I don't see any need for it to be. JackofOz 04:25, 29 April 2007 (UTC)

Agreed. Edit conflicts are far more common at the Ref Desk than most other pages due to the high volume of changes on Ref Desk pages. StuRat 08:06, 29 April 2007 (UTC)

Call it a guideline yet?

Several things have improved since this question was last asked, including things specifically mentioned as objections. Are we close enough to call it a guideline? Friday (talk) 07:25, 29 April 2007 (UTC)

No, there are many issues yet to discuss. StuRat 08:03, 29 April 2007 (UTC)

There always will be. We call it a guideline when it's workable and generally agreed on, not when it's complete. Friday (talk) 08:51, 29 April 2007 (UTC)

I don't believe it is generally agreed upon. StuRat 09:03, 29 April 2007 (UTC)

Other than a couple spots at the end where we need to choose between a couple of possible versions, are we in decent shape yet? As before, specific suggestions or objections ("I can't support this unless..") are way more useful than just saying "No, it's not ready yet". Friday (talk) 14:30, 30 April 2007 (UTC)

No, I have several more objections, which I've opted to put on hold until the current issues are resolved. StuRat 14:38, 30 April 2007 (UTC)

Is there some advantage to waiting? If you say what your objections are, maybe some smart person can figure out how to resolve them. Ned Wilbury 15:45, 30 April 2007 (UTC)

The advantages is to keep focus on one issue (well, two) at a time until resolved. If we bring up everything at once each discussion gets buried in the rest and we never reach a consensus on anything. StuRat 15:49, 30 April 2007 (UTC)

Perhaps more editing of the guideline itself (versus the talk page) would help? Even after the guideline is labelled a guideline, there'll be people objecting to various part of it at once. We can't really prevent this so why not start doing it now? Ned Wilbury 15:54, 30 April 2007 (UTC)

Direct edits just caused edit wars, so we should discuss changes first. StuRat 15:59, 30 April 2007 (UTC)

The entire rest of Wikipedia runs by people making changes as they see fit. But, if you wish, let's discuss changes. What are your other objections? Ned Wilbury 16:01, 30 April 2007 (UTC)

Please read the above. I don't wish to cause more confusion by bringing up new issues until the current issues are resolved. StuRat 18:04, 30 April 2007 (UTC)

The thing is, StuRat, that a number of different parts of the guideline are interrelated–as is becoming quite clear with the current discussion about handling incorrect answers. If you keep a queue of objections to yourself and only let them out on to the talk page one at a time, I'm afraid we're going to be here for years and never formally accept a guideline. Or, we will have a guideline, but it will be one that lacks your endorsement. TenOfAllTrades(talk) 18:27, 30 April 2007 (UTC)

I agree with TenOfAllTrades. One thing that might help is: we could leave out the parts we don't agree on yet, call it a guideline, and then add in the removed parts as we reach agreement on them. Ned Wilbury 18:36, 30 April 2007 (UTC)

That would mean leaving out huge chunks of it. StuRat 18:38, 30 April 2007 (UTC)

Are we really so far off? Sounded to me like people were thinking we were close to being able to call it a guideline. In the bit at Wikipedia_talk:Reference_desk/guidelines#Can.27t_agree_to_this_without... I thought the remaining objections were few, and I think they're mostly resolved. What am I missing? Ned Wilbury 18:41, 30 April 2007 (UTC)

I, for one, listed some 9 objections, few of which have even been discussed, yet. StuRat 19:23, 30 April 2007 (UTC)

Well, which of those objections are of the "can't support this unless" variety? We all need to be willing to compromise or there's no way we'll get anywhere. Hopefully by this point the disagreements are mostly over minor things. Also keep in mind, this can get labelled "guideline" without unanimous consent. If you want your input considered, you'll have to do some of that work yourself. Keeping your objections secret doesn't help. Ned Wilbury 20:06, 30 April 2007 (UTC)

They aren't "secret", I've listed them before and will do so again, once the current issues are resolved. I would like to, at the very least, discuss all 9 issues before I will lend my support to any guidelines. StuRat 23:01, 30 April 2007 (UTC)

Could you list them right here, for the sake of clarity? Numbered list (using #) or fourth-level headings (====, so we can comment on them without any getting 'lost') would be fine. We'd like to have an idea of what you see as genuine deal-breakers here, as well as what might fit into the existing guideline, what redundancies or overlaps may exist, and what might well be beyond the scope of this document. We've been pushing words around on this page for months, and I suspect that most of the editors here really don't want to spend a couple of months apiece on each new issue. If you refuse to support the guideline before all nine of your issues are discussed, well—discuss. TenOfAllTrades(talk) 23:16, 30 April 2007 (UTC)

StuRat's suggestions

Fine, here's a few, to start (they could be combined into one, if we wish):

1) No argument from authority. Your response will be judged based on whether it is logical and has supporting evidence, not solely because somebody with a PhD in the field (you or others) made the argument.

2) No ad hominem attacks. A person's response should not be denigrated because of their age, gender, country of origin, English language skills, or lack of a degree in the area of concern. The only valid ways to question a response are with logic and supporting evidence.

3) The Ref Desk standard for verifiability is that it is preferable if a source be provided, but not necessary.

4) The Ref Desk standard for neutral point of view is that all POVs should be respected, although, as in the rest of Wikipedia, it isn't necessary that each author include every POV in their contribution.

5) Answering questions by referring to articles or even reliable original sources is consistent with these key content policies, but not required.

6) The Ref Desk standard for no original research is considerably relaxed, as many Ref Desk questions require OR, such as math questions.

7) We shouldn't state or imply that Wikipedia policies apply to the Ref Desk as they do to articles. Some apply exactly the same, like WP:NPA, some apply differently, like WP:NPOV (where we want the totality of all responses to be NPOV, but don't require that every response be NPOV), and some don't apply at all, like WP:NOR (doing math probs is "original research").

8) The "wrong answers" topic being discussed below. StuRat 03:27, 1 May 2007 (UTC)

1) Verifiability is already covered. 2) Courtesy is already covered. If you think there's more to say on these topics, why not add to the sections that already address these topics? Friday (talk) 02:15, 1 May 2007 (UTC)

Those sections certainly don't cover those topics now. When I've tried adding such info in the past, you've reverted me. StuRat 02:32, 1 May 2007 (UTC)

Part of what you're saying in #1 sounds directly at odds with the notion of verifiability, so that may be why. There's no way some random internet person's opinion gets the same weight as a citation from a reliable source. Friday (talk) 02:52, 1 May 2007 (UTC)

No, a reliable source would be "supporting evidence". StuRat 02:59, 1 May 2007 (UTC)

As for 3-6, much of this is already addressed, altho your standards appear to be a bit different. #4 is just weird, tho- I don't think we need anything along thoses lines at all. It makes it sound like we want to encourage free speech, where everyone gives their own personal opinions, which is totally irrelevant to our purposes. I did soften the language a bit in response to your concerns, though- people shouldn't think that the ref desk has literally the same expectations as an article, altho they should expect our core policies still apply. Friday (talk) 02:37, 1 May 2007 (UTC)

There's nothing weird about 4, a psychiatrist gives the psychiatric POV, a biologist gives the biological POV, etc. They are free to each give other POVs, if they wish, but this certainly isn't required, as the old NPOV text seemed to imply. StuRat 03:06, 1 May 2007 (UTC)

(After edit conflict) Okay, that gives us some idea of where you're coming from. I hope the last three issues are on their way? Here are my thoughts on those points. I really don't think we're that far from agreement; a lot of those suggestions seem to be pretty much part of the proposed guideline, though the phrasing may be different.

1. I suppose we could write a guideline which says, "Never supply an answer which employs any sort of logical fallacy", but it would be damnably difficult to police. I don't see anything wrong with an answer that indicates an editor's area of expertise or credentials, particularly if that answer relies to any extent on that person's judgement or experience. I wouldn't want to see an editor trying to bludgeon another one with his degree, but that's a straight civility and courtesy issue (already covered, see #2).

In reply to your comments about #1, I think we should let the reader weigh the evidence presented as they see fit—including, where applicable, any credentials presented by the parties. If the person asking the question ultimately says, "I'm going to trust Rockpocket here, because he has the cutest username", that's their call to make. If someone makes a specious argument (an argument from inappropriate authority, presenting a false dilemma, positing a strawman, etc.) by all means politely call them on it.

The whole point of allowing – where necessary – answers from personal experience and professional expertise (dare I say oiginal research?) is to tap the broad pool of knowledge here at the Desk. I don't think it's unreasonable to indicate one's relevant experience or qualifications when answering such a question so that the questioner might have an additional basis on which to evaluate your remarks.

Mind, if an editor is using his professional status to browbeat others or otherwise be obnoxious, then it becomes a user conduct issue covered under WP:CIV. TenOfAllTrades(talk) 03:37, 1 May 2007 (UTC)

I don't believe that does cover it, see the examples below. StuRat 03:54, 1 May 2007 (UTC)

I'm in broad support of Ten's view here. If the question was about the efficacy of certain medical treatments for infants, and the first few answers were not what the questioner wanted to know (eg. flippant, guesses, irrelevant, ...), I'd have no problem with the next editor saying "I'm a paediatrician and I can tell you there's been a huge debate about this. For what it's worth, my opinion is ....", or something like that. I think the questioner would be much more likely to take note of such an answer and distinguish it from the unuseful ones. Also, the Ref Desk attracts people with relevant knowledge, skills, experience etc as volunteers, because of the pleasure they derive in sharing their knowledge with others. There's broad agreement that providing published sources is recommended, but not generally necessary; so, if a source isn't provided, no harm is done by the volunteer stating where their answer comes from, viz. their personal involvement in that field. This is very different from saying to another volunteer "I'm a PhD and you're not, and therefore I know better than you". JackofOz 01:00, 2 May 2007 (UTC)

2. I'm inclined to agree with Friday that #2 is already in the guideline; attacks on other editors, for whatever reason, are banned by sections 2.1 (Maintain the highest standards of courtesy) and 2.2 (Don't poke fun). Furthermore, it's already part of the larger corpus of Wikipedia policies, as WP:CIV and WP:NPA.

No, 1 and 2 aren't covered. If somebody says "my evidence is that I have a PhD in the field and you don't", that certainly wouldn't qualify as a WP:NPA violation. Either would "you will understand once you've lived as long as I have". However, we still don't want to allow this type of discourse. StuRat 03:20, 1 May 2007 (UTC)

I still don't see how #2 isn't covered by WP:CIV and WP:NPA. Could you elaborate? TenOfAllTrades(talk) 03:37, 1 May 2007 (UTC)

I just gave two specific examples. Do you think those quotes would violate either policy ? StuRat 03:47, 1 May 2007 (UTC)

As I read them, both of those quotes fall under your point #1 (potentially fallacious arguments from authority) not #2 (attacks on the person). Again, it's extraordinarily difficult to prohibit all fallacious argument, and the best response is probably to politely point out the flaws in such reasoning as necessary. Otherwise, we're in the boat of "You just made a fallacious argument from authority, so I removed your comment and am now trying to get you banned"; I just don't think we want to go there. I'm quite reluctant to try to codify every single possible case of WP:DICK into this document, and I fear that specific provisions would tend to be abused ("You just mentioned your qualifications to answer this question, that's not allowed!"). TenOfAllTrades(talk) 05:07, 1 May 2007 (UTC)

3, 5, and 6. These seem to be incorporated into the body of text already.

Where ? StuRat 03:15, 1 May 2007 (UTC)

4. I don't know quite where you're coming from with this. I would agree that all editors should be respected, but some POVs are just plain loony, or out-and-out-wrong (Examples: Q: What causes gravity? A: Leprechauns. Q: What's the capital of the United States? A: Brasilia. Q: What's 2 + 2? A: 5.) By all means treat editors civilly (see #2), but there's no need to grant equal weight or credence to every point of view.

4) That's not what I said at all. I said each answer does not need to be NPOV, not that wacky answers are encouraged. StuRat 03:12, 1 May 2007 (UTC)

Based on your response to Friday above, I think I have a better understanding of what you're saying. I think you're arguing that no response should be required to necessarily address the entire breadth and depth of a topic, and that editors are invited to offer their expertise to answer particular facets of a question, even if they are unable to answer a question in its entirety...?

I don't think there's anything unreasonable about that...have I read your statement correctly? On the other hand, I don't think there's anything in the guideline as it stands now which requires an answer to address all aspects of a question. I've tried to make this explicit with this edit. TenOfAllTrades(talk) 03:50, 1 May 2007 (UTC)

That helps, but I've seen comments deleted with edit summaries like "NPOV vio", and this should not be a justification for a removal, as each person is allowed to only provide their own POV. StuRat 04:02, 1 May 2007 (UTC)

I haven't seen that. Can you provide some examples? On it's surface, it sounds like something that shouldn't be happening, though I can see exceptions in cases where a lot of ranting is involved. I could also see cases where an editor is attempting to use the Desks to push (aggressively) a particular fringe opinion to the extent of being disruptive. TenOfAllTrades(talk) 05:07, 1 May 2007 (UTC)

So, what think you? TenOfAllTrades(talk) 03:07, 1 May 2007 (UTC)

7. Okay, I see that Friday already softened the language a bit, to indicate that the application of Wikipedia policies is necessarily flexible (here and everywhere on Wikipedia, really). Anyone editing here or anywhere should of course bear in mind Wikipedia:Ignore all rules and Wikipedia:Use common sense (those two policies definitely apply everywhere), but I don't think that a specific mention of them would be helpful—old fogeys already know them, and newbies are apt to be confused. TenOfAllTrades(talk) 05:17, 1 May 2007 (UTC)

And 8 is being discussed below, so what's number 9? TenOfAllTrades(talk) 05:17, 1 May 2007 (UTC)

Preferred version ?

We now have two versions of each of these sections:

Factual disputes and verifiability

a) While the reference desk is not an article, verifiability and other core policies should still be kept in mind. However, there may be some questions where a strict requirement of verifiability would not be efficient or useful. Use common sense. If you have reason to doubt someone else's answer, ask for a source, or better yet try finding your own source that gives a better answer. If someone asks for a source, don't take it personally. Disputes may be best avoided by providing a reference for your answer in the first place.

b) The threshold for inclusion in Wikipedia is verifiability, not truth. While we should keep the verifiability policy in mind while answering, a strict application is not always efficient or useful for some questions. If you believe a response should provide a reference but does not, exercise due diligence before challenging the edit. Check Wikipedia's articles for conflicting or supporting information. Search out your own sources which could confirm or deny the assertion. Use common sense when challenging a response on the desk, present your own sources as a response to the question or politely ask for references. If one of your own responses is challenged don't take it personally—it is after all the Reference Desk—provide a source or reconsider the response. If you cannot find anything to support you edit then indicate such on the desk itself: strike out, remove, or add additional clarification to your response.

Wrong answers

a) Responders should make every effort to give correct answers, but inevitably a few answers will be wrong. If you believe your own answer is incorrect, you may wish to strike it out and add a comment as to why. If you think somebody else's answer is wrong, add a comment saying why you think so, and provide evidence, if possible. However, do not remove or strike out another editor's incorrect answer, solely because it is wrong, even if you can prove so with reliable sources.

b) Even if you believe someone's answer is wrong, remember to be polite about it. Editors may see it as very rude if you remove or strike out another editor's comments, so this is not recommended.

Which do you prefer ?

• I prefer version A of both. StuRat 14:38, 30 April 2007 (UTC)

• Factual disputes and verifiability: I don't have any deal-breakers about either of them, but on balance I prefer version A.
• Wrong answers: I definitely prefer version A. JackofOz 01:05, 1 May 2007 (UTC)

I still think it's worth merging whatever we say about wrong answers into the previous section- "Factual disputes and verifiability" is a better heading anyway IMO. I put a sentence in saying "Don't just remove an answer because you think it's wrong." Is there more that needs to be said about it, and if so, can it be worked into the existing section? Ned Wilbury 01:17, 1 May 2007 (UTC)

Yes, we also want to say that you shouldn't strike out another editor's answer, and we want to say that this is true not only if you think it's wrong, but even if you have proof, such as from a reliable source. We also want to say that a wrong answer can be removed if there is another reason for removing it, such as it containing racial slurs. Based on all this, we should probably say something like, I don't know, maybe: "Do not remove or strike out another editor's incorrect answer, solely because it's wrong, even if you can prove so with reliable sources." I wonder why nobody thought of this before ? StuRat 01:44, 1 May 2007 (UTC)

Discussion

Either of the first two is fine. Don't need a "wrong answers" section in my opinion. If we want to talk specifically about wrong answers, this should be a topic under the previous section. Ned Wilbury 14:56, 30 April 2007 (UTC)

What do you mean by "first two" ? The two section A's ? StuRat 15:54, 30 April 2007 (UTC)

Sorry, I meant either a or b under "verifiability and factual disputes" is fine. Ned Wilbury 16:08, 30 April 2007 (UTC)

"verifiability and factual disputes": B. There is no need for either wrong answers section. The key points could be merged into the first sections as has already been proposed. Rockpocket 17:28, 30 April 2007 (UTC)

If I were writing these guidelines myself, I'd keep what I consider the best parts of both versions (of both topics). But since we're being asked to choose one over another, I've voted as I have above. I can certainly live with whatever consensus emerges (subject, of course, to ongoing debate and development once the guidelines have been promulgated). JackofOz 01:15, 1 May 2007 (UTC)

There's nothing saying we have to choose one or the other- that's just one idea. I like combining the best of both- go ahead and try it, if that's your preference. Ned Wilbury 01:18, 1 May 2007 (UTC)

Yes, Jack, please go ahead and combine them together and let's take a look. StuRat 01:30, 1 May 2007 (UTC)

We are going round in circles here: that has already been proposed, but was then dismissed by StuRat and no-one else expressed an opinion. What, exactly, was wrong with this attempt at merging them?

The threshold for inclusion in Wikipedia is verifiability. While we should keep the verifiability policy in mind while answering, a strict application is not always efficient or useful for some questions. Everyone should attempt to give factually correct answers, but a few wrong answers are inevitable. If you believe a response to be incorrect, or should provide a reference but does not, exercise due diligence before challenging the edit. Check Wikipedia's articles for conflicting or supporting information. Search out your own sources which could confirm or deny the assertion. Use common sense when challenging a response on the desk, present your own sources as a response to the question or politely ask for references. However, do not remove the incorrect answer, solely because it is wrong.

If one of your own responses is challenged as being factually incorrect don't take it personally—it is after all the Reference Desk—provide a source or reconsider the response. If you cannot find anything to verify your edit then consider indicating such on the desk itself: strike out, remove, or add additional clarification to better assist the querent in determining the reliability of your response.

Rockpocket 01:43, 1 May 2007 (UTC)

We were actually starting to make some progress until Ned came in and scrapped it all and had us start over from scratch. Your version is better than Ned's, but "due diligence" was not a popular term (I proposed a subpage to explain it). StuRat 01:55, 1 May 2007 (UTC)

Really, I can't see anything about it that would cause me not to support it as a very good inaugural guideline. It will undoubtedly change over time, but are there any serious objections to this? JackofOz 01:58, 1 May 2007 (UTC)

To answer Rockpocket's question: One of the facts of Wiki-life is that we sometimes take 2 weeks to cover the same ground that we'd cover in 2 hours if we were all sitting around a table face-to-face. In 2 weeks, we all have a bazillion other things on our minds, and then there's the outside world on top of that. So, we forget a lot of the detail of what was said days ago, and/or haven't got the mental energy to go back and see who said what when. JackofOz 01:58, 1 May 2007 (UTC)

We should all meet for coffee and discuss the guidelines one day. (doesn't have to be for coffee, could be tea as well) A.Z. 04:08, 1 May 2007 (UTC)

Great idea. We should organise a Meetup specifically for Ref Desk Regulars. I nominate Brazil for the first meeting. A.Z., how big is your place? We'd only need 20 or so beds.  :) JackofOz 10:15, 1 May 2007 (UTC)

My humble opinion: if a man (so I can use the pronoun) writes on the reference desk something that you think is wrong or unverifiable, just say it on the thread and disprove him. Maybe one specific post by him could have been wrong, unverifiable and not helpful to the reference desk. Then again, we should be entitled to write on the reference desk things that are wrong and unverifiable, as long as everyone else is entitled to disprove us. (I actually wrote that first on the talk page of my essay, to which there's a link on my user page) A.Z. 04:17, 1 May 2007 (UTC)

Yes, that's a good thing to do--but I assume we all know not to word it quite that way. (smile) "Additional information that yo might like to consider" is one way to word it. If it is going to be debatable, there is probably a subject talk page debating it on WP, and I'd suggest the inquirer took a look there as well. DGG 06:48, 1 May 2007 (UTC)

I think we should explicitly say: "It's OK to post wrong and unverifiable things, although not intentionally", or else the deletionists (innacurate word, but useful, anyway), the deletionists will just come up with a lot of excuses to delete stuff and the ridiculous lame edit wars and discussions on the talk page will never end. If the deletionists win this battle and the guidelines end up saying that it is OK to delete answers that you dislike even though you can't disprove them, then there should be a place to redirect all threads whose posts have been deleted: Wikiversity could be such a place. A.Z. 07:35, 1 May 2007 (UTC)

There already is such a place, on Wikipedia - it can be found by going through the history page. JackofOz 10:15, 1 May 2007 (UTC)

Who are these fabled deletionists, anyway? They sound scary. I sure wouldn't want to meet one. It's always so difficult being in a war with an enemy that you can't see; and it's so sad that these invisible creatures are out there on the Desk right now, mangling innocent answers that don't yet have a guideline to protect them. Er....

Can we tone down the rhetoric a notch, please? Where did anyone, anywhere, suggest a guideline along the lines of "it is OK to delete answers that you dislike", except as part of a straw man argument? Thinking of this as a war – complete with winners, losers, fascists deletionists, communists inclusionists, battles to be won or lost, and the fate of the Desk hanging in the balance seems a bit melodramatic. Honestly, this guideline has been in about the same shape for months. It essentially describes the way the Desk has operated for even longer; people are already following it because it's just an elaboration on how some specific aspects of WP:DICK, WP:CIV, WP:IAR, and WP:SENSE happen to apply (and are applied). There's no ravening horde of deletionist wikilawyers waiting in the hills, eyeing the tasty loopholes that will let them delete the Desk. There just isn't.

On the subject of Wikiversity, if you feel that a particular discussion is better suited to that venue, feel free to make the suggestion; they're a sister Wikimedia project, and I wish them all the success in the world. TenOfAllTrades(talk) 13:48, 1 May 2007 (UTC)

Again: the guidelines have not changed much recently just because StuRat lost a couple of edit wars. He said that a few months ago this was the consensus: [23].

I'm going to tone down the rhetoric a notch. Thanks for the advice. A.Z. 19:54, 1 May 2007 (UTC)

You might as well link to the local copy, A.Z.: User:StuRat/rules. The last edit to that version was way back in mid-December; I agree that there was never consensus support for that version. It has been based on a very short period of discussion along with some straw polls that some people found to be poorly structured. A number of stakeholders did not even participate, feeling that the discussion was being cut short. Consequently, it was abandoned more than four months ago.

Shortly after, I decided that developing policy wasn't going to work until we had some handle on the desired goals and outcomes. I slapped together User:TenOfAllTrades/RD thoughts, which was just my personal understanding of what we were trying to accomplish. I'm flattered to note that that document was later pressed into service to act as a framework on which to hang a new policy; the bulk of that work took place in late December and into January. Since then, the document has been pretty stable; between the end of February and now, there have been a few changes: diff. The bulk of them relate to formatting and word choice, rather than affecting the meaning of the guideline. There are some expansions to give examples, and some of the wording is tightened up. The biggest two changes are that the section about medical advice has been shuffled off to a subpage, and the 'Factual disputes and verifiability' section has appeared (which we're discussing now).

We really are getting into a mature, stable guideline, and I think that the relatively few changes should be necessary from here to bring everyone (or nearly everyone) on board.

There's one more point I'd mention to you, A.Z.—nobody 'wins' or 'loses' an edit war. An edit war is what happens when a) consensus has not been achieved, and b) parties are failing to engage in discussion. TenOfAllTrades(talk) 21:07, 1 May 2007 (UTC)

Archived

Okay, this page was getting to be unmanageably long, so I've archived everything up to around the beginning of April. We could probably do with an even more aggressive pruning, but I didn't want to step on any toes.

If there's an important thread to which anyone would like to refer, see Wikipedia talk:Reference desk/guidelines/Archive 1, or copy the thread back here. TenOfAllTrades(talk) 03:22, 1 May 2007 (UTC)

Request for clarification

Disclaimer: I was invited to join in. I haven't studied any of the discussion here since April 23. Please don't see this as disrespect or laziness (well, maybe that), but I chose this approach deliberately. After comparing the current guidelines with the ones I had read before signing the agreement eight days ago, I noticed, to my delight, that most of the sections received only minor changes and a few saw additional improvement. I guess it comes down to the last paragraph, the one on factuality and verifiability, eh? It sounded strange, after having read the rest. Consider the first two sentences.

"The threshold for inclusion in Wikipedia is verifiability, not truth. While we should keep the verifiability policy in mind while answering, a strict application is not always efficient or useful for some questions."

I stumbled and had to re-read them. The emphasis shifts from one sentence to the next. Is there an explanation? ---Sluzzelin talk 16:06, 1 May 2007 (UTC)

I think it's the result of a merger between a couple (or several) variants of that passage that were spread over a few different locations. There's still some massaging going on (look up this talk page); if you can think of a good way to smooth out the lumps, it would be welcomed. TenOfAllTrades(talk) 16:38, 1 May 2007 (UTC)

Thank you, Ten. The paragraph's intention isn't clear to me. There are different possible ways of smoothing it, but what's the first sentence's message? Is it supposed to provide context for the next sentence, which makes it relative. (i.e.: "Though verifibiality, not truth, is generally Wikipedia's threshold for inclusion, a strict application is not always efficient or useful for addressing some questions at the reference desk." Is it supposed to provide context for the subsequent part. ("Wikipedia's threshold for inclusion is verifiability, not truth, so please don't say someone else's answer is false, and don't remove it if you think it is. Instead, present other references and point out the conflicting views etc.") I guess I don't understand what the last paragraph is trying to tell the reader. ---Sluzzelin talk 09:46, 2 May 2007 (UTC)

The page just didn't change because StuRat lost the edit wars. I don't like edit wars too much, so I didn't participate. However, there is definitely no consensus about the guidelines. A.Z. 19:47, 1 May 2007 (UTC)

Editors have a choice. You can complain, or you can participate. If you elect not to participate, you might find nobody pays much attention to your complaining. The current guideline draft seems to me to have a pretty decent consensus among those who have been participating. As always, specific suggestions for improvement are much more valuable than nonspecific complaints or assertions that there's no consensus. Friday (talk) 22:14, 1 May 2007 (UTC)

What I meant by "I didn't want to participate" was just that I preferred to discuss the guidelines here on the talk page and help reaching consensus rather than engaging edit wars.

There's no consensus because a lot of editors think that the guidelines as they currently stand don't reflect what the reference desk is all about, and are likely to allow the destruction of the potential that it has. This is mainly due to the fact that the guidelines as they stand allow for the deletion of content. There is no reason to delete wrong nor unverifiable opinions, since everyone is allowed to disprove those!

Really good editors who come here to answer questions just go away the minute they realize that people are going to arbitrarily delete their responses. A.Z. 22:26, 1 May 2007 (UTC)

Please do read the guideline. There's even a sentence in there specifically saying not to remove content just because you disagree with it. This whole "people will arbitrarily delete whatever they don't like" was a straw man months ago, and it's a straw man today. Friday (talk) 22:32, 1 May 2007 (UTC)

Friday, I totally agree with what is written. I just think that each user will interpret it in a very particular way. Some ways will be even "eccentric". Hipocrite said: "disruptive answers and questions will continue to be removed from the desk". To me, this is like saying that you will delete something because you disagree with the content. My main point here is: saying what your opinion is (with or without sources and arguments to support you) does not violate NPOV and verifiability. A.Z. 22:43, 1 May 2007 (UTC)

You are using a very different definition of the word "disruptive" than I am. Hipocrite - «Talk» 22:48, 1 May 2007 (UTC)

Maybe we're just quibbling over words. When I delete vandalism, I would not say it's because I "disagree" with it. I would say it's because vandalism disrupts Wikipedia. Perhaps the key is in what people mean by "disruption" at Wikipedia. The page WP:POINT may clarify. Friday (talk) 22:49, 1 May 2007 (UTC)

How does this page help to clarify things? And... were you both wrong about Froth after all, when you said that he was disruptive? Or today you would stop reverting his posts, accordingly to the guidelines? A.Z. 22:54, 1 May 2007 (UTC)

Concern

This guideline appears to be overly long. Is a codified set of rules more necessary than "WP:ENC, WP:DICK (WP:POINT), WP:5P?" When this process started months ago it was transparently an attempt to protect obvious nonsense from being removed from the reference desks. I certainly hope all of the editors here realize that the strong community consensus that pooflinging is not protected expression anywhere is not going to be changed by a very long guideline that, in paragraph three of section four subpart 5 sentence 2 says "pooflinging is ok."

To be more specific, disruptive answers and questions will continue to be removed from the desk regardless of whatever some individuals believe this policy says. Hipocrite - «Talk» 22:21, 1 May 2007 (UTC)

We don't agree as to how the policies apply to the reference desk, Hipocrite. You can explain to us your positions and let the other editors judge for themselves and show everyone whether they're fallacious or credible. A.Z. 22:27, 1 May 2007 (UTC)

I agree that the guideline is not needed. However, as long as what it says is reasonable, I don't see that it particularly hurts anything either. Reasonable, experienced editors don't have much need for long rules, it's true, but if some folks claim having the guideline helps them understand what the ref desk is about, why not let them have it? Friday (talk) 22:29, 1 May 2007 (UTC)

I understand and agree with that sentiment, and generally follow the principles of WP:IAR for myself. Nevertheless, I don't mind helping editors who need a bit more structure to – within reason – assemble a slightly more detailed set of guidelines. I'm not trying to impose any particular changes on the Ref Desk or rebuild it in my own image; I'm just trying to help build guidelines that reasonably accurately reflect accepted practices on the Desk and on Wikipedia in general.

Mostly this means keeping a handle on the scope and aims of the policy to avoid too much WP:CREEPiness, and trying to avoid writing things in or out that will encourage ruleslawyering in the future. I'm also trying to avoid 'gotchas'—things added, omitted, or misstated that might lead a novice who is unfamiliar with Wikipedia culture and practices into bad habits or behaviour.

I expect that most experienced editors will intuitively understand and naturally follow the guidelines that we lay out, without ever needing to read this guideline or even knowing that it exists. TenOfAllTrades(talk) 22:45, 1 May 2007 (UTC)

I strongly concur with ToaT. The 5P will always supercede these guidelines. I too will continue to remove disruptive answers and questions if they interefere with the project, and should those be continually reverted on a WP:POINT of principle, then there is policy to deal with that too.

Although I see no harm in having a more detailed guide for what editors should expect, and not expect, from the desk. It would be misleading and confusing to have guidelines that would appear to allow open season for chat on the desk, only for other editors to invoke 5P and remove it. Therefore if we do have these guidelines, it is imperative that they are consistent with the wider culture of Wikipedia. This is why A.Z.'s attempts at re-inventing the Desk are pointless. There about about 10 individuals involved in forming these guidelines, no matter what we decide here, we have no mandate to overrule the core characteristics of the project. If he somehow managed to claim consensus among us that "anything goes" is our new policy, the project would correct that soon enough. Rockpocket 23:36, 1 May 2007 (UTC)

Except that I'm not trying to re-invent the desk, as a lot of editors of the desk agree with me about what the desk is; this understanding of what the reference desk is does not overrule any core characteristics of the project (arbitrarily deleting stuff does that); and the closest to "anything goes" that I can find here is eric's view that anything is allowed when it comes to deleting stuff. A.Z. 00:17, 2 May 2007 (UTC)

(I was invited to rejoin the discussion.) I have come to doubt the wisdom of trying to generate guidelines at all. Even among the handful of editors who work actively on the proposed guidelines, there is an obvious lack of consensus not only about guideline formulation but fundamentally about how the reference desk should be run. Any official guidelines that reflect a compromise would almost certainly be used by some faction as a blunt instrument to defend dubious behavior on the desk. The existence of the guidelines would then have a negative influence on the desk. My feeling is that the attempted guidelines should be abandoned. --mglg_(talk) 00:08, 2 May 2007 (UTC)

Being honest

I'm going to be honest here. I would like that there were no guidelines at all, I would like it if we didn't need them. But people misinterpret the policies and use them as an excuse to delete posts from the desks: that's a problem. I am willing to take this to the next level now. I don't know who is it who will decide, I don't know whether it is the village pump the right place or which place it is. Maybe the talk pages of the policies. The only reason why I didn't want this to get out of this page is that I was afraid that people elsewhere would have a hard time understanding why the policies apply to the reference desks sort of counter-intuitively: this only happens because people are used to apply them to articles, and the reference desks are quite different from articles.

Now, it just looks as if it's never going to end. Just look at Rockpocket's post above: he clearly sees things a lot differently than I do. I would like to do as suggested above and gather us all in one place so we could understand one another, but this seems unlikely to happen. Why don't you pay attention to what I'm saying? Why don't you realize that things that are opinion, wrong and unuseful in your opinion do not violate unverifiability nor NPOV? They could violate notability, is that what you are talking about? What belongs or doesn't belong to the articles and the reference desk is a matter highly subjective, so you should allow the subjectivity to be explicit when no harm is done!

Anyway, where should we take this to let some other people solve this for us once and for all? A.Z. 00:34, 2 May 2007 (UTC)

I agree with you to the extent that the reference desks are quite different from articles, in some significant ways. I would support the main Wikipedia policies and guidelines explicitly acknowledging that, so that we can have some freedom to construct some Ref Desk guidelines that reflect these significant differences, and not always be hamstrung by arguments to the effect that "the Wikipedia policies say X, the Ref Desk is part of Wikipedia, and therefore we must always abide by X".

But I disagree that taking these discussions anywhere else would be helpful. We are the ones who are most affected by whatever outcome this process produces. Anyone else on Wikipedia can see these discussions and is welcome to come along here and join in. JackofOz 01:11, 2 May 2007 (UTC)

You are welcome to take this wherever you wish, but I don't think anyone, anywhere, is going to have much interest in debating it with you. The general response of Non-Ref Deskers to disputes about the Desk is: "Its not building the encyclopaedia so delete the whole thing." Now imagine what they are going to say with a proposal that it should be immune from our core policies. Good luck with that.

The fact is that there will be absolutely zero enthusiasm from anyone outside the handful of people here to promote anything that deviates from the 5P. Wikipedia's policies will continue to govern the Desk as they always have. The options are we can either draft specific guidelines within that framework to assist those that are not familiar with our policies, or we can carry on without specific guidelines. However, If you are serious about getting wider consensus to grant the Desk explicit exceptions from WP:5P then I suggest you draft a proposal and invite comments. I'll be happy to give you mine. Rockpocket 01:22, 2 May 2007 (UTC)

There's no need to change anything with the policies. They apply just fine! The articles are not signed, so a person cannot write "abortion should be legal". The person can only write "The president of Brazil thinks that abortion should be legal." On the reference desk, we use four tildes to sign our posts and everyone reading it understands that our posts are our personal opinions. The fact that the posts are signed and the acknowledgement that the sentence "abortion should be legal", when written on the reference desk, actually means "User:John thinks that abortion should be legal" makes answers with personal opinions NPOV and as verifiable as it is that a person is being truthful about their own opinion!

Also, they say "It's not building the encyclopedia, so delete the whole thing." What do you think? Is it building the encyclopedia?

It's just that I am having trouble figuring out why you don't say "delete the whole thing" as they do. A.Z. 01:36, 2 May 2007 (UTC)

A.Z., I'm a bit concerned that your interpretation of NPOV differs somewhat from the way it's usually understood on Wikipedia, but in any case, the sort of debate that you'd be encouraging is really outside the scope of the Desk. (It might be welcomed in some other forum, however.)

Our response to the question "Should abortion be legal?" should be along the lines of "There are many schools of thought on this issue. See our articles on Ethical aspects of abortion, Religion and abortion, and Legal protection of access to abortion for discussions of these perspectives and considerations. Our article on abortion also links to a number of other useful resources. For country-specific information, see List of articles about abortion by country." The Ref Desk isn't the place to rehash the entire debate about abortion, with every participant offering their own opinion ("It should be legal!" "No it shouldn't!" "Yes it should!" "Why do you hate women!" "Why do you hate babies!"). TenOfAllTrades(talk) 03:28, 2 May 2007 (UTC)

What do you think is counter-intuitive? This guideline is just a common-sense explanation of how existing Wikipedia standards apply to the reference desk. Friday (talk) 02:37, 2 May 2007 (UTC)

I think that the notion that the collection of answers to a question will not look exactly like a Wikipedia article even though they follow the same rules is counter-intuitive: in the collection of answers, the biases will be more "explicit" and there will be things perceived to be off-topic, which doesn't happen with the articles. A.Z. 04:55, 2 May 2007 (UTC)

Where does this continued fear of "arbitrary deletions" come from? Has this ever happened? Are a small minority of editors trying to stir up fear of some unlikely scenario? If someone were to remove a post, they'd be expected to have a darn good reason for it. Where is the problem? Friday (talk) 02:28, 2 May 2007 (UTC)

Please don't make huge changes without discussion

There seems to be a general feeling that we're in decent shape. Major changes like this one should not be made out of the blue at this stage. I've reverted back to how it was before. Friday (talk) 03:12, 2 May 2007 (UTC)

But, you disagree with the changes or not?

You said to me earlier: "As always, specific suggestions for improvement are much more valuable than nonspecific complaints or assertions that there's no consensus." Well, you just said "there seems to be a general feeling...", which is a very nonspecific complaint about the changes I made. A.Z. 03:13, 2 May 2007 (UTC)

I doubt Friday was compaining about the words per se. I think he was complaining about a fairly large edit to a document that we're coming close to achieving agreement on. By all means make suggestions, A.Z., but when they're that major, discuss them here first, rather than just changing the policy document without any discussion. JackofOz 03:20, 2 May 2007 (UTC)

I'm all in favor of people being bold- that's how we get things done. But, don't be reckless. Specifically, saying that stating "your opinions on any subject" is in line with core policies is a complete non-starter. Friday (talk) 03:23, 2 May 2007 (UTC)

People who are capitalists can go to the article on capitalism and write "capitalists believe this, according to this source". On the reference desk, you can do the same. A.Z. 03:30, 2 May 2007 (UTC)

Apples and oranges. Encouraging sourced statements is vastly different from encouraging editors to give their personal opinions. Friday (talk) 03:48, 2 May 2007 (UTC)

Please, explain why, Friday. A.Z. 03:50, 2 May 2007 (UTC)

This must be the language barrier at work. I can't think of a way to be more clear. I can tell you that sourced opinions are verifiable but I can't imagine how reading that page will help if the above statement doesn't make sense to you. Friday (talk) 03:55, 2 May 2007 (UTC)

A personal opinion may be or not be a sourced statement, but someone saying "black people are inferior" is different from someone saying "I think that black people are inferior", as Rockpocket said. A.Z. 03:57, 2 May 2007 (UTC)

Errr... any chance you could rephrase that so it doesn't potentially read like I'm a racist? Rockpocket 06:51, 2 May 2007 (UTC)

Applying the policies

We are in danger of confusing representing a POV in a NPOV manner ("white extremists think blacks are inferior") and simply representing a POV ("blacks are inferior"). Stating only the former is perfectly acceptable, even though it is simply representing only one POV. Stating the latter is not even if contrasted with the opposing view. A statement saying "Blacks are inferior" countered by "Blacks are not inferior" is not NPOV, even though we are "expressing "all" points of view". This is what WP:NPOV is about in this context: representing positions in a neutral manner. It doesn't imply any editor has to represent all positions equally, simply that the position they do represent should be stated neutrally. Rockpocket 17:39, 23 April 2007 (UTC)

What about saying "I think that blacks are inferior"? It is NPOV. A.Z. 03:30, 2 May 2007 (UTC)

Any statement starting with "I think" is by definition POV. JackofOz 03:37, 2 May 2007 (UTC)

What about "capitalists think" and "Jack thinks" and "Hitler used to think"?

And "I think that 1+1=2" ?

What about a neurologist or a psychologist making experiments with people and asking them "OK, now we're gonna do a thinking exercise and I'll evaluate the results. I'll say a word and you tell me what do you think". The person will say "I think..." and that sentence will be the object of the scientific study.

I mean, I'm just trying to show why it isn't POV... A.Z. 03:54, 2 May 2007 (UTC)

Do not feed the hand that bites you

Is there some rule in the notion of building consensus that we should indefinitely keep taking account of the arguments of editors who insist on argumentation in a style that is virtually indistinguishable from trolldom? Perhaps we should not feed them either.  --Lambiam^Talk 11:11, 2 May 2007 (UTC)

1. I see what you did there. Hipocrite - «Talk» 12:27, 2 May 2007 (UTC)

I'm willing to ignore unhelpful talk page contributions. This is difficult for me because I tend to be optimistic and assume that people can be reasoned with. But it may be time to cut our losses in some cases. Friday (talk) 14:15, 2 May 2007 (UTC)

I see anticipaton of disruption at the desks (edit wars) and indignation on the talk page (endless) as one reason why a few editors choose to not ignore the argumentative style. I have only the highest respect for this approach, but it does focus a lot of attention on one single "conflict" not too many editors seem interested in dealing with anymore (the conflict is specified in the next section below). Yet I know no alternative, since the edit wars in particular seem to be unignorable. ---Sluzzelin talk 14:50, 2 May 2007 (UTC)

Straw poll?

Can we have a straw poll on the following issue: Should inappropriate content sometimes be removed? A possible list of answers:

1. Yes, inappropriate content should always be removed.
2. It depends, but some inappropriate content ought definitely to be removed on sight.
3. Perhaps, but only after the consensus has been reached on the talk page that the controversial content is to be deleted.
4. No, content must never be deleted.
5. Other (don't agree with the question or answers; don't understand the question; am taking the Fifth; ...)

 --Lambiam^Talk 11:31, 2 May 2007 (UTC)

If removing the content improves the encyclopedia, remove it. Hipocrite - «Talk» 12:28, 2 May 2007 (UTC)

The concept of inappropriate content is ill-defined. If I say "1 + 1 = 7", nobody would agree this is correct, but everyone is free to dispute it and prove it wrong. It's clearly inaccurate, but is it inappropriate? And if so, should it be removed? I say no. If I say "all Buddhists eat pigs' entrails", then few if any would object to it being removed on sight because of its offensive nature. It is clearly inappropriate. So, it depends. I vote for 2. JackofOz 12:49, 2 May 2007 (UTC)

At the moment my question is not whether inappropriate content should be removed sometimes and if so when, but whether we can have this straw poll (possibly amended as the result of discussion here by rewording questions, including others, or merging some). I'll wait before starting any polling (if at all) for at least 48 hours, and not until any meaningful discussion peters out.  --Lambiam^Talk 13:03, 2 May 2007 (UTC)

Quite, Jack. The real answer is 2, based on policy and practice. Please note that I would discourage any attempt to create an exhaustive list of content that would qualify. Creating a such a list will lead to an endless amount of bickering; the existence of such a list will discourage editors from employing common sense; and the use of such a list will encourage wikilawyering.

Worth considering as well is that removal is not the only route to dealing with edits that aren't appropriate to the Ref Desk—a pattern of unhelpful editing is not just a Ref Desk problem, but a user conduct issue that can be dealt with through other processes (mediation, warnings, probation, blocks, page bans, and – if all else fails – site bans.) Please everyone try to remember that this document doesn't need to address every single conceivable circumstance to be a workable guideline. TenOfAllTrades(talk) 13:21, 2 May 2007 (UTC)

Yeah, 2 seems to be about right. There are plenty of harmless but off-topic questions and answers which nobody seems too bothered to remove. It's the potentially harmful stuff we should worry about. Friday (talk) 13:29, 2 May 2007 (UTC)

A straw poll might be helpful at this point. It might show that no one holds positions #1 and #4, and it might boil down to how positions #2 and #3 are split in the comnunity of those who cast their vote. ---Sluzzelin talk 14:00, 2 May 2007 (UTC)

Option 2 doesn't actually exclude option 3, and I think the choice to employ one or the other (or some unlisted option) depends strongly on circumstances. This might be another reason to discourage this particular straw poll—it may inadvertently create the impression that a) there is more disagreement than actually exists, and b) that there are two warring camps working on these guidelines (a popular myth that we're trying to stamp out). TenOfAllTrades(talk) 14:38, 2 May 2007 (UTC)

A very reasonable sentiment, but #2 and #3 do seem to be mutually exclusive they way they're phrased now. ("some content definitely to be removed on sight" vs "but only after consensus has been reached at the talk page"). Yeah, I'd prefer to vote for something like "some content may be removed on sight, in the interest of damage control, but persuading the poster to remove his/her own comment is the preferable course of action". ---Sluzzelin talk 14:59, 2 May 2007 (UTC)

#2 is the only option per WP:BLP and WP:BAN. Extreme examples of these types of inappropriate content should be removed immediately. Rockpocket 17:25, 2 May 2007 (UTC)

What about something like this? Personally, I think it should go, but I'm afraid of being assigned to a faction. FWIW, this is exactly the type of thing a certain banned user used to post; when I tried a unilateral deletion of something similar way back when, I was reverted and his action was supported by others, leading to a hiatus in my attempts to work the RDs. --LarryMac 18:17, 2 May 2007 (UTC)

I think we've come a long way. The remaining editors who fanatically object to any removal of off-topic content are now generally kept in check by the rest of us. I hope reasonable editors feel comfortable applying common sense to the reference desks. Friday (talk) 18:22, 2 May 2007 (UTC)

Friday, all I will say is that the desks where I make most of my contributions are now infinitely better, and more serious, than when I first joined Wikipedia last October. Most of the puerile schoolboy tendency have now gone, along with the 'oh-so-clever' dialogues. Clio the Muse 18:36, 2 May 2007 (UTC)

It appears to me that differences of opinion on the issue of removal of content deemed inappropriate have led to warring behaviour in the past, and are likely to continue to do so unless we can reach some form of consensus. My purpose in having this poll is to see how far we are from consensus, and what roads might lead to consensus on the issue. If the consensus is #2 (something that has neither been established nor disproved yet), we can try to work out some compromise on what constitutes delible content – perhaps covering more than some would like, and less than others would like, but something we all can live with. If there is no consensus for #2, then it might be time to consider asking for wider input from the Wikipedia community.  --Lambiam^Talk 18:45, 2 May 2007 (UTC)

I think the point has been made that artificially restricting our choices – by saying all matters must be handled using either method #2 or method #3 – isn't going to work. It isn't how things are done now, and it doesn't acknowledge that the process chosen must depend on the nature of the content removed and the history of the editor(s) involved. Presenting this as an 'either/or' situation is likely to push people here further apart, not result in a useful consensus. (Please see also my comments above on how trying to create a list of forbidden comments is likely to be unsuccessful.) TenOfAllTrades(talk) 18:57, 2 May 2007 (UTC)

Choices #2 and #3 are logically exclusive of each other. If a user is in favour of allowing some content to be removed on sight, they should choose #2, and definitely not #3, which excludes that option. If a user feels that content should never be removed on sight, they should not pick #2, but instead #3 or #4. You cannot have guidelines according to which both some content may be removed on sight, and no content may ever be removed on sight. Such guidelines would be internally inconsistent, contradicting themselves. So the distinction is not at all an artificial restriction of our choices; the restriction is already imposed by the natural requirement of consistency.  --Lambiam^Talk 20:01, 2 May 2007 (UTC)

Cearly the answer is #2 - the other options are straw men. The question that this begs is what exactly is "inappropriate content". Is a possibly wrong answer inappropriate ? Is a tasteless joke inappropriate ? Is a joke of any kind inappropriate ? Is a remark about seagulls or bagels inappropriate ? Is a response that start "In my opinion ..." or "In my experience ..." inappropriate ? Is a response without links or sources inappropriate ? Agreeing on #2 proves nothing. This is where we came in. Gandalf61 21:35, 2 May 2007 (UTC)

"Fairly and without bias"

Editors should attempt to represent fairly and without bias significant views published by reliable sources.

At the risk of stating the blindingly obvious, that is total horse doodle. No single answer can or should be expected to represent all significant views on a subject fairly and without bias. My responses, for example, will inevitably be biased towards mainstream science rather than pseudo-secience and left-wing rather than right-wing political views, simply because those are my opinions and beliefs. There are differing significant views published by reliable sources, but I won't be presenting them in my own answers - I will leave that to other editors. Like I said - obvious really. Just common sense. Gandalf61 21:28, 2 May 2007 (UTC)

If you don't believe in WP:5P, perhaps you should find a different project? I don't find it difficult to provide answers that represent fairly and without bias significant views published by reliable sources what-so-ever. If you are unable to distance your personal opinions from your answers to specific questions, perhaps this guideline is a hint you shouldn't be answering those type of questions? Hipocrite - «Talk» 21:32, 2 May 2007 (UTC)

It doesn't say that any single answer "should be expected to represent all significant views on a subject." Simply that the verifiable view you choose present should be represented fairly and without bias. I understand that to mean that you represent the source accurately, and don't misquote it or use it out of context. Rockpocket 21:34, 2 May 2007 (UTC)