Wikipedia talk:Attribution/Poll

This poll and its talk page are a frequent source of heated debate. Please try to keep a cool head when commenting here.

This poll was nominated for deletion on March 30, 2007. The result of the discussion was Speedy keep.

Please join the debate at Wikipedia_talk:Attribution/Community discussion

Pre-poll Archives: Archive 1 • Archive 2 • Archive 3 •

Archives from after poll start: Archive 4

WP:RS?

I'm on Wikibreak and have probably missed something, but the "advert" on my watchlist states this poll includes merging of WP:RS, while the blurb at the head of the project page mentions just WP:V and WP:NOR, before weasel-wording re RS. Seems inconsistent to me; important, as one of the two key reasons for my oppose was not wishing to merge WP:RS into a policy, as it's currently a guideline. I'd want consensus on RS becoming policy on its own before considering a merge. --Dweller 12:22, 5 April 2007 (UTC)

Whatever you read was badly worded, Dweller. Aspects of RS that were already policy (which had been taken from V or NOR) were merged into ATT. The rest was abandoned or may be worked into the FAQ, if suitable. The only thing in RS that I know of that wasn't policy that ended up in ATT was "exceptional claims require exceptional evidence." It's a pity about the poor wording because I saw a few comments from people saying they'd opposed because they didn't want RS to become policy. SlimVirgin ^(talk) 12:25, 5 April 2007 (UTC)

Hmmm. It's the standard wording at the top of everyone's watchlist (unless they've dismissed it). --Dweller 12:56, 5 April 2007 (UTC)

The wording of "Professional self-published sources" was also taken from RS, was it not? In fact, I remember the moment SteveBlock made the edit. It was one of the light bulb moments because we'd been arguing about the exceptions for so long. The wording from RS is a longer version of what was already in V, so it was already policy. Marskell 13:12, 5 April 2007 (UTC)

I agree with SlimVirgin... the original intent was simply to merge V and NOR... Once again, I think people are confusing the concept of RS with the guideline page WP:RS. The problem is that both V and NOR contained sections on the concept of RS (pointing the reader to the guideline WP:RS for further info). Shortly after ATT went live, The short cut to WP:RS was redirected to the section in WP:ATT that discussed RS. Looking back, that was probably a mistake. People payed attention to the links and not the content... they assumed that WP:RS had being merged into ATT and promoted to policy, when in fact all ATT did was restate what had already been part of V and NOR.

Perception is everything in this debate... WP:RS was never meant to be included in this debate... but it got tossed in by happenstance. If you look back at the debates on the wording of this poll, you will see that I too percieved WP:RS as being part of the merger. It wasn't until I looked into it further that I saw what actually was intended.

The "Oppose - don't merge RS" comments highlight something important... One thing that we will have to do over the next few weeks is to analyze the comments in this poll. If you look at it as more than just a head count... there are comments made in the "Oppose" section that actually express some degree of support of ATT, and there are soome interesting critical comments in the "Support" section. Blueboar 13:21, 5 April 2007 (UTC)

The original intent of ATT was to merge V and NOR and delete RS so that descriptions of reliable sources are policy and policy only. There would cease to be a reliable source guideline—in this sense ATT is a merger of RS. I don't think it was a pity to ask about it in the slightest. It needed to be asked about and should be asked about again if there is no consensus: why is this fundamental concept treated in a guideline? Why does V (if V is to remain) rely on a guideline for it's expanded definitions? Marskell 13:30, 5 April 2007 (UTC)

Because the question of what sources are reliable, for what points, is essentially a matter of judgment; it probably can't be summed up, absolutely correctly, in a readable length. Guidelines can be vague and approximate, because {{guideline}} admits the existence of exceptions; policies can't. Septentrionalis PMAnderson 16:10, 5 April 2007 (UTC)

Can we be careful with our terminology here... there is a huge difference between the concept of Reliable sources (RS) and the guideline page Wikipedia:Reliable sources (WP:RS) ... I don't want people to mix them up. Now, I agree that some of the creators of the ATT page had the intent to delete the guideline page: WP:RS... but that intent was somewhat debated. I think almost all the contributors to ATT saw that WP:RS is flawed, but there was less consensus on what to do about it.

I would agree with the view that WP:RS is flawed... As I have said before, it is somewhat unique - sort of betwixt and between. On one hand it is a guideline that was treated by the community as if it were a policy (even Jimbo refers to it as if it were policy in the comments that got us started on this poll). On the other, it discusses a concept that is clearly part of Policy, which people can ignore because "it's only a Guideline". It is inherantly unstable and should be deleted or re-concieved in some way.

I have suggested elsewhere that we need to do a complete re-think of the page WP:RS ... leaving the any statements that could be construed as "rules" to either WP:ATT or a combination of WP:NOR and WP:V (and let us not forget that RS is part of WP:NPOV, in either case). I feal that we need a new RS guideline that should focus on giving guidance as to how to determine if something is reliable... not giving quasi-rules on whether something is reliable or not. And to make a clean break, I would re-name it to something like Determining Reliability. I am sure others may have different ideas. But I think a lot of us agree that the problems with WP:RS, and the RS issue in general does need to be discussed further.

But all of this is besides the point... the fact is that a lot of people percieve ATT as "promoting" WP:RS to policy status. I think we would agree that THAT was never the intent. Blueboar 14:16, 5 April 2007 (UTC)

I am confused about SlimVirgin's and Marskell's assertions above that some "aspects" or portions of WP:RS are already policy, and that RS is a "fundamental concept" of Wikipedia. It seems to me like there may have been some instruction creep here over the last year or so, since I don't recall RS being introduced as policy before. (Admittedly I do not keep track of the intricacies of day-to-day policy modifications, and my own mental map of Wikipedia policy is based on year-or-two-old versions of WP:NOR and WP:CITE, so things may have changed without my noticing, but I think it is a problem if the policy is really shifting in meaning, rather than in the details of its wording, over time.) Can someone summarize what parts of WP:RS are already purported to be consensus-backed policy? -- Rbellin|Talk 16:11, 5 April 2007 (UTC)

Look at the opening sentences of WP:V: "Verifiable" in this context means that any reader should be able to check that material added to Wikipedia has already been published by a reliable source. Editors should provide a reliable source for quotations and for any material that is challenged or is likely to be challenged, or it may be removed. Then scroll down the page to the section on Sources...

Look at the second paragraph of WP:NOR: Wikipedia is not the place for original research. Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not presenting original research is to cite reliable sources that provide information directly related to the topic of the article, and to adhere to what those sources say. Then scroll down and look at the section entiteld Sources.

It was from these two "Sources" sections that the majority of the RS section of ATT was composed. Blueboar 20:55, 5 April 2007 (UTC)

I'd just like to point out that Wikipedia:Neutral Point of View also discusses reliability/reputability. NPOV says to "to find the best and most reputable sources you can" and "work for balance, that is: divide space describing the opposing viewpoints according to reputability of the sources." So, NPOV sees reliability/reputability as a scale, where the degree of reputability is one way to determine due weight, or degree of inclusion. Wikipedia:Attribution says "The threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true." So, ATT sees reliability and inclusion as binary. There was already contradiction between NPOV and V before this, but the new wording, as opposed to "verifiability not truth", with no mention of reliability, seems to increase this contradiction.

The tension between those who think of reliability as binary, and those who think of it as a scale, is the reason Wikipedia:Reliable sources has been a self-contradictory mess, in my opinion. Keeping one side of the debate in ATT and another in NPOV, without a separate page to try to reach consensus, does not seem like the solution to me. Besides, RS contained a lot of helpful advice of too much detail and insufficient consensus to be full policy, in my opinion.

— Armed Blowfish (mail) 21:41, 5 April 2007 (UTC)

Blowfish... I am not understanding how you turn the "threshold" quote above into "ATT sees reliability and inclusion as binary." The "whether it is attributable, not whether it is true" line is not disussing the reliability of the attributable source, just that you have to have one. To me if confirms NPOV... as it means that statements of opinion are allowable. Even if people do not think the opinion represents "the truth", you can not exclude it since it is attributable to a reliable source. I suppose you could say that ATT does have one binary feature... if there is a reliable source you can add it... if not you can't. But that is not in conflict with NPOV. NPOV also says things have to be backed with reliable sources. The only difference is that NPOV tells us a bit more about what is and is not considered reliable. Blueboar 00:34, 6 April 2007 (UTC)

"Whether it is attributable to a reliable source" seems to imply two options: either it is attributable to reliable source or it isn't. The word "whether" seems to make it a true/false condition, at least the way I read it. If you took out the reliable source part and made it clear that this was only one threshhold, it might work. (E.g. "The first test for suitability for inclusion in Wikipedia is whether the material is attributable to a source, not whether it is true. After that, see WP:UNDUE.")

I am not objecting to the "not whether it is true" part, just the inclusion of reliability in a true/false condition. That said, add an "According to X" and it becomes true (philosophical arguments about whether or not X exists aside).

— Armed Blowfish (mail) 01:01, 6 April 2007 (UTC)

It seems to me that having policy that says that all material needs to be attributable to a reliable source, but only having a guideline to what constitutes a reliable source is exactly the right thing to do. Sometimes editors are just going to have to reach a consensus value judgement about whether a source is reliable or not for a particular purpose. This is particularly true about online sources as the line between published and self published is going to continue to blur over time. To take a real world example, if the Alfred Russel Wallace Page website expresses an opinion about one of Wallace's works is it a reliable source? It is true that the website is largely the work of a single person (Charles H. Smith) and is essentially self edited without peer review, but anyone who does much research on Wallace knows that it is an important site with a strong positive reputation. Sooner or later you reach a point where policy can't eliminate the need for good judgement and that is what guidlines are for. Rusty Cashman 17:25, 6 April 2007 (UTC)

Just an observation.

The poll states at the beginning that ATT has been edited for months by more than 300 editors, which is by itself a significant number. However, in comparison with the number of active accounts[1] which is more than 4 million, it seems that only 1 in about 10,000 editors was interested/aware/involved in this article. In these few days (not months!) of poll, since a note had been put on 'watchlist' pages, almost 800 editors have expressed their opinions, which tells that ATT policy making was far from transparent, and not well advertised at all. I am sure if a note was put on more visible pages, like 'edit' pages, the response would be even greater. Lakinekaki 21:42, 5 April 2007 (UTC)

Numbers are fun... but they can be twisted to say almost anything... You could look at the same numbers and ratios as saying that ONLY 800 editors have exptessed opinions. Assuming the trend holds by tomorrow we should have around 900 opinions. That is only a little better than 3 in about 10,000 editors who are interested/aware/involved in this article - even after banging them over the head with a watchlist notification. I would bet that our results include most of the 300 who worked on the article (and I am sure there are some of them who voted "oppose" dispite their involvement.) Face it... most editors don't really care about editing policies. They just want to work on their articles.

As for the supposedly low number of contributers to ATT... I think this is somewhat true of all our policy and guideline pages... if you look at the number of editors who regularly contribute to NOR or V for example (discounting those who pop in just to ask a question) it is far smaller than the number of editors who worked on ATT. As with articles, our policy pages attract only those who really care about their subjects. It takes something as profound as a querry by Jimbo to move policy issues into the public eye. Most of our policies and guidelines are created by a only a few people... and almost all were declared to have consensus based on far fewer than 300 opinions.

Could this have been advertized better?... of course. But let's be fair to the creators of ATT... if there was a problem with how it was promoted, that problem is system wide and not limited to just this policy. All this policy debate did was highlight the need for a re-thinking on how all of our policies and guidelines get created/changed/etc. Blueboar 00:04, 6 April 2007 (UTC)

I Agree! Lakinekaki 00:46, 6 April 2007 (UTC)

Extend?

Resolved

– Re-opened until previously-declared closing time.

I probably should have said something sooner, but is there any need to close the poll so soon? People still seem to be adding their views, why not let this continue? — Armed Blowfish (mail) 02:06, 6 April 2007 (UTC)

The watchlist message said that the poll would not close for another day... —bbatsell ¿? ✍ 02:07, 6 April 2007 (UTC)

I agree. We have advertised the closing as Apr. 7, 01:00 UTC, there are still votes and comments coming in, and I see no need to close early. Crum375 02:09, 6 April 2007 (UTC)

Yes, the poll should stay open until the time and date announced on the WatchList page! --Rednblu 02:11, 6 April 2007 (UTC)

(double edit conflict) It has apparently been closed and protected. [2] [3] — Armed Blowfish (mail) 02:12, 6 April 2007 (UTC)

And re-opened, to continue until the "advertised" (not my term) closing time. — SMcCandlish [talk] [contrib] ツ 13:53, 6 April 2007 (UTC)

What does "No consensus" mean?

Given the discussion, the comments, and the number of people that have participated, closing the poll with a "no consensus" and a call to "move on and relax" does not address the core issues raised. We cannot dismiss 400 votes (pro or con) with a wave of the hand. ≈ jossi ≈ (talk) 02:16, 6 April 2007 (UTC)

We could consider compromises. I know you don't like the hierarchical suggestion, where everything would be kept live but not all as policy, although I do not fully understand why. We could talk about it. Or someone could magically think up another compromise. I don't know.... — Armed Blowfish (mail) 02:22, 6 April 2007 (UTC)

I have unprotected and corrected the closing date as per the announcement in the watchlist. As per the closing, it should be closed without comments and allow Jimbo, that requested the poll to make a statement about moving forward, given the results. ≈ jossi ≈ (talk) 02:23, 6 April 2007 (UTC)

Thanks! : ) Armed Blowfish (mail) 02:25, 6 April 2007 (UTC)

I fully agree with Jossi. The poll should continue to run to its highly advertised conclusion on Apr. 07, 01:00 UTC, as votes and comments are still coming in. I also agree that at closing time, the page should be protected, and Jimbo should then decide how to interpret the results. Crum375 02:31, 6 April 2007 (UTC)

Thanks. I suggest that we should devise among ourselves a proposed next step. Jimbo's job is to make sure that we solve our problems in a reasonable fashion. It is not his job to invent a direction we should go. --Rednblu 02:32, 6 April 2007 (UTC)

The way I see it is very simple. Jimbo asked for this discussion and for the poll. As such, we need to close the poll without a closing statement on April 7 01:00 UTC and send a message to Jimbo about the fact that the poll has closed, inviting him to check the poll page and suggest a way forward. ≈ jossi ≈ (talk) 02:35, 6 April 2007 (UTC)

Alternatively, Jimbo intervened because he saw that we were not solving our problems. We should be facing and resolving our problems here, I suggest. --Rednblu 02:40, 6 April 2007 (UTC)

(ed conf)Not really. Jimbo's intervention was related to the fact that he was discussing an edit and was surprised to find WP:NOR superseded by WP:ATT. ≈ jossi ≈ (talk) 02:45, 6 April 2007 (UTC)

We have already cut down a thousand virtual forests with our rhetoric. I see no reason to expect that another thousand will get this resolved. I agree with Jossi that we should hand the poll results to Jimbo to let him decide. Crum375 02:43, 6 April 2007 (UTC)

I hope Jimbo does not pull a Solomon on us and say "Bring me a sword. Cut the living child in two and give half to one and half to the other." :) --Rednblu 02:50, 6 April 2007 (UTC)

(edit conflict) Jossi's procedure sounds good, except for closing on 7 April. Why not extend until 14 April? If we close this while there is still a stream of votes coming in, they will probably spill over into the talk page. — Armed Blowfish (mail) 02:52, 6 April 2007 (UTC)

... because we advertised it as closing on April 7. ≈ jossi ≈ (talk) 02:59, 6 April 2007 (UTC)

I'm not sure if we'd learn anything more, but it wouldn't really bother me if the poll went on longer. The whole idea was to get a greater amount of community input. It wouldn't hurt anything, and I find it very ironic that the users who started the poll early, on a whim, are so insistent on closing it on a specific date. It doesn't matter if we said it was going to close April 7th, if we want to extend it we can. -- Ned Scott 03:02, 6 April 2007 (UTC)

The reason there is a date is that it's a cutoff date. Anyone that cares enough will vote, and if you don't cap the date it's just an endless discussion/debate. No concensus means that by no means have we all reached a mutual agreement about the topic at hand. As such, no changes are made at this time. Further discussion is encouraged, as for me I still stand firm on my view. Fr0 03:09, 6 April 2007 (UTC)

People who care, yet miss the cutoff date, might very well vote on the talk page. I've seen it happen before. WP:STRAW suggests not even having a set closing time. — Armed Blowfish (mail) 03:25, 6 April 2007 (UTC)

Once again... the perception of regular editors is very important in this debate. Just as we could not close this poll early without someone yelling foul... we can not hold it open without the same reaction happening. Because we stated that there would be a closing time on this, if we hold it open after that time the perception of the average editor will be that it was held open in order to favor one result or another (it does not really matter which). As for people posting after the fact votes on the talk page... I see nothing wrong with that. Remember, the reality (as opposed to the perception) is that this isn't about exact numbers... it's about consensus, which is a mushy concept. We can already tell that there isn't a huge consensus in favor of the merge (what is trickier is that there isn't a huge consensus for un-doing the merge either). So a few late votes would not really affect anything, unless they had some brilliant idea. Since Jimbo is probably going to check this talk page anyway (he should, as there are some good preliminary discussions here on what the next step might be), if someone suddenly shows up with a brilliant idea or comment, we want them to share it. But there is no reason why they have to post it in the poll page. Blueboar 14:55, 6 April 2007 (UTC)

Well said, Blueboar. ≈ jossi ≈ (talk) 15:13, 6 April 2007 (UTC)

If you really think that's a risk, I guess it is for the best. — Armed Blowfish (mail) 15:35, 6 April 2007 (UTC)

Jossi asked:

Given the discussion, the comments, and the number of people that have participated, closing the poll with a "no consensus" and a call to "move on and relax" does not address the core issues raised. We cannot dismiss 400 votes (pro or con) with a wave of the hand. ≈ jossi ≈ (talk) 02:16, 6 April 2007 (UTC)

and while that thread, immediately above, grew in various directions, no one actually addressed that question.

I would like to.

No consensus means simply that it isn't over/decided yet, and we go on as we were before, business as usual, until the issue arises again. I think that some people are being confused by the scale here. This is no different, at its core, from a discussion/poll on an article's talk page in which there are only a dozen editors who give a darn enough to bother to weigh in, even after 2 months. If the discernable opinions are 50/50 or 40/60 or whatever, and defensibly so (vs. "I like it", etc.), then you have no consensus. Things go back to the status quo before the issue arose, and further discussion ensues (else the issue just goes away), typically with people building their "case" in the background, shifting things, twidding things, addressing concerns, resolving objections, and making compromises, so that when it comes up again, hopefully it won't be a big split but a clear yes or no.

This really isn't anything new. If the article one is working on survives AfD narrowly with a closing admin's determination of "no consensus", then one takes that as a sign that a lot of work needs to be done, or the next time you can bet that "no consensus" (much less "keep") is a remarkably unlikely outcome. This poll is a reprieve. Use it wisely, and maybe we'll all be singing WP:ATT's praises. Use this time to continue to attack, ignore or ridicule those with criticism for or concerns about ATT and/or the underlying processes, and you might as well just stop now and go do something else, like write a new article.

— SMcCandlish [talk] [contrib] ツ 10:12, 6 April 2007 (UTC)

~~Actually~~, someone foolishly asked the entire community, most of whom don't care; and now they feel compelled to randomly make up an opinion very much on the fly, on a relatively short time scale. That's not to say these people aren't intelligent or anything, it's just they may not have had 5 months to mull over possible compromises, so this poll wasn't very fair to them or their opinions. :-P --Kim Bruning 12:25, 6 April 2007 (UTC)

Agreed. The "Actually..." phrasing seems to indicate that we might disagree on something here; I'm skeptical that is really the case. — SMcCandlish [talk] [contrib] ツ 13:24, 6 April 2007 (UTC)

Fair enough. Summarising: I think we simply asked a silly question and got the silly answer we should have expected. (that, and we just wasted a lot of peoples time). :-/ --Kim Bruning 13:28, 6 April 2007 (UTC) (Possibly the actually was in there because I was thinking of Jossi's text too).

What kind of statement is that? "We" did not ask any question, did we?; This was a poll requested by Jimbo. If you have any issues with that, you can ask him. ≈ jossi ≈ (talk) 14:15, 6 April 2007 (UTC)

Or saved a lot of it (people's time I mean). The debates here have been remarkably heated yet remarkably circular. If I failed to short-circuit that cycle, then @#$% it, I'll have to try again some other day. (Oh, and I just noted your... I'd call it a parenthetical, but it's not. Your "post-sigular" to make up a new word, and all is clear in-context now.) — SMcCandlish [talk] [contrib] ツ 13:45, 6 April 2007 (UTC)

PS: I also just now clued to your much wider meaning of wasted people's time. Duh. — SMcCandlish [talk] [contrib] ツ 14:01, 6 April 2007 (UTC)

Nevertheless, we are left with some good material for future discussion. It's brought out: some big problems people have with the Attribution proposal in its current form, some problems people have with any merger at all, some outstanding issues with current policy and a few misconceptions about both current and proposed policy. Identifying issues is of course the first step to resolving them. --bainer (talk) 13:53, 6 April 2007 (UTC)

Indeed. — SMcCandlish [talk] [contrib] ツ 14:02, 6 April 2007 (UTC)

Yeah, it's been an eye-opener in many ways. Metamagician3000 14:05, 6 April 2007 (UTC)

This is not an AfD. It is different. The process was different, the issue is different, and the closing should be different. ≈ jossi ≈ (talk) 14:12, 6 April 2007 (UTC)

True. We can't dismiss either side. As for compromises, I know you don't like the ones that have been thought of so far, and we don't know how the community feels about them since we didn't specifically ask. But where does that leave us? — Armed Blowfish (mail) 14:19, 6 April 2007 (UTC)

Re "good material" and "identifying issues," I'm thinking that we should workshop the three pages individually and sequentially, drawing on the responses in the poll. I suggest starting with RS. There are a lot of issues and misconceptions surrounding it. Marskell 14:38, 6 April 2007 (UTC)

I don't know. People above are mentioning "no consensus" and back to status quo before the poll. But ... what is the status quo? WP:ATT was policy for more than a month before the poll, was used in ArbCom cases and widely cited. So my point is: Jimbo asked for a discussion and a poll, and we have done that. As such, Jimbo needs to play a central role in the aftermath of this. So, we close the poll April 7, 01:00 UTC, protect the page, and wait for Jimbo to comment. After all, it was his intervention that triggered all of this. ≈ jossi ≈ (talk) 14:42, 6 April 2007 (UTC)

That all sounds good, jossi, except I would suggest extending the poll. Not that it's really that big a deal, but I don't see the point of having votes on the talk page while we are still waiting for Jimbo anyway. Perhaps we could extend it "until Jimbo gets around to closing this"? — Armed Blowfish (mail) 14:47, 6 April 2007 (UTC)

See Blueboar's comment in the previous section. ≈ jossi ≈ (talk) 15:14, 6 April 2007 (UTC)

Suggestion for how to close the poll: We can all work together to edit a document which summarizes the various positions about the merge, exercising our encyclopedic skills to (as much as possible) write it in a NPOV way. Working on helping to express the various points of view other than one's own can help to promote understanding and possibly even willingness to compromise. One idea would be to do it in two steps: first produce a document with short sentences such as "<number> users favour keeping all the pages as policy" following by copies of the actual votes. Afterwards, on a separate page, produce a document in paragraph form attempting to describe all sides of the issue. The question of what to do next would still remain, but I think it would be valuable to produce such a document. It could be helpful to Jimbo: "Here's our summary, and you can also read the individual poll votes if you like." We're not just doing this for Jimbo, though. I really think the process of producing such a document, and the document itself, wuold be helpful in somehow figuring out how to proceed. --Coppertwig 15:53, 6 April 2007 (UTC)

Jossi wrote "what is the status quo? WP:ATT was policy for more than a month before the poll, was used in ArbCom cases and widely cited." WP:ATT was declared policy on the grounds that it represented a consensus to merge WP:V, WP:NOR, and WP:RS. The current poll surely raises the question whether such a consensus ever existed. It seems to me that the appropriate course is to assume that WP:V and WP:NOR are policy, WP:RS is a guideline, and WP:ATT is a work-in-progress. Bucketsofg 18:46, 6 April 2007 (UTC)

I respectfully disagree. The poll shows a divided community as it pertains to this issue. ≈ jossi ≈ (talk) 18:54, 6 April 2007 (UTC)

This issue being what, specifically? Bucketsofg 19:27, 6 April 2007 (UTC)

I agree with User:Bucketsofg above. The true status quo is that ATT is a work in progress, and the older policy/guideline pages are current until the s consensus to change or merge them. DES ^(talk) 23:19, 6 April 2007 (UTC)

Jimbo can jump in any time he wants. We should not wait for Jimbo. We should proceed in a normal fashion. This isn't a remake of Life of Brian. Once everyone who wishes to present an analysis has had such a chance (ie give people a few days) then normal wiki editing of policy pages and their talk pages is fully in order. This is a wiki, let's act like it. WAS 4.250 18:51, 6 April 2007 (UTC)

I think inviting Jimbo to comment, and giving him a couple days to do so, is reasonable. We might also want to invite opinions from the rest of Wikipedia on what the poll has told us, but I would rather not finalize a closing statement or take other action without waiting for Jimbo to chime in. TheronJ 21:16, 6 April 2007 (UTC)

Agree... ≈ jossi ≈ (talk) 21:28, 6 April 2007 (UTC)

Agree. Metamagician3000 22:44, 6 April 2007 (UTC)

I concur as well. — SMcCandlish [talk] [contrib] ツ 00:37, 7 April 2007 (UTC)

It would be a mistake for Jimbo to comment at this time, other than to say "Carry on" -- because we have not done our job yet of coming up with a solution for this problem. --Rednblu 01:19, 7 April 2007 (UTC)

Well and concisely said, Rednblu. I agree.

I also agree with User:Bucketsofg and DES. I haven't seen evidence that WP:ATT ever went through a reasonable process of becoming policy, or that there was ever consensus that it was policy. The current poll demonstrates that lack of consensus. The status quo, in my opinion, is that WP:NOR and WP:V are policy, while WP:ATT is a work in progress.

I think it's OK to wait a few days for a reply from Jimbo before making major decisions, but I also think it's important not to waste time but to continue discussing and trying to reach consensus. Developing a joint poll-closing document as I suggest above can proceed either before asking Jimbo for input, or while Jimbo is thinking. Probably better to do it before. If we can manage to reach consensus, or at least get closer to it, it would be better to present that to Jimbo. Think how you'd feel in his shoes. --Coppertwig 01:26, 7 April 2007 (UTC)

I also concur with this. Lest that sound self-contradictory: Yes, status quo is ATT = work-in-progress, and much discussion is required. But also, yes, Jimbo started this inquiry, and will be required at some point to resolve the mess and confusion resulting from his declaration and actions, and perhaps to also resolve or declare resolved the ultimate issues after aforementioned continued discussion. — SMcCandlish [talk] [contrib] ツ 01:35, 7 April 2007 (UTC)

There seems to be a rather strange logic in some of the previous discussion of this issue, namely "Jimbo called for the poll, therefore Jimbo should decide what it means". Certainly Jimbo's comments will be interesting, but he clearly identified himself with the oppose side before the vote even took place. He may well rise to the occasion, but he is far from an ideal arbitrator. No consensus does not mean "let Jimbo decide", it means "read what people (on all sides) said and then move forward more informed". Geometry guy 09:46, 7 April 2007 (UTC)

!! Yes. --Rednblu 18:44, 7 April 2007 (UTC)

A positive outcome

Although Attribution ~~as a combined policy~~ as a replacement for V and NOR does seem to ~~be shelved~~ have not achieved a consensus for now, I think the work that went into it was still of great value. To my eyes, ATT clarified the existing policies in a number of ways, and I hope these clarifications can brought over into the once-again-separate V and NOR pages.--Father Goose 10:53, 7 April 2007 (UTC)

ATT is not shelved any more than V or NOR or RS are shelved. The only thing that is shelved is confident talk that any one side has a claim to consensus. WAS 4.250 11:33, 7 April 2007 (UTC)

I agree that "shelved" is not the operative word here. Nonetheless, we've now discovered that only slightly more than half of the community can describe itself as being "in broad support of WP:ATT". We can hardly describe it as policy given the lack of consensus supporting it. It is best regarded, I suggest, as a work in progress. Bucketsofg 12:31, 7 April 2007 (UTC)

Poll closed

I have protected the poll and header pages, as the poll is now closed. Crum375 01:05, 7 April 2007 (UTC)

Post-poll comments

If you missed the poll, feel free to comment here. Your comments will not affect the vote outcome, but as the point of this poll was to gather and gauge opinions, not produce a vote count, your comments will be considered along with those in the poll. As on the poll itself, please do not post argumentative or supportive replies to comments in this section. For overall discussion of the issues, please see Wikipedia_talk:Attribution/Community discussion.

Please add this to the notes section

An important point to make in the notes section is WP:!VOTE. It is extremely important that newcomers and readers be reminded that when viewing the results, that poll numbers aren't always the basis here on Wikipedia. Please be bold and make this change. Kopf1988 04:04, 7 April 2007 (UTC)

OK

So what happens next? JulesH 17:57, 7 April 2007 (UTC)

Find someone who disagrees with you, and discuss with them. Then find some more. :-) (you now have a nice list of people you can discuss with :-) ).

You can try hardest first (find someone with the totally opposite opinion), or you could team up with folks with ideas closest to yours to refine your arguments, or you could try for an easy sell towards someone who almost seems to agree with you, or... well, take your pick. :)

--Kim Bruning 21:25, 7 April 2007 (UTC)

I object to the classification of my vote and the votes of the 101 other editors who voted in the "Neutral/qualified/compromise/other" section as "neutral".

The title of the section was "Neutral/qualified/compromise/other", not "Neutral". Not all of the 102 people who voted in that section are "neutral". The wording of the count in the header should be changed accordingly. Also, the other two sections were not "support" and "oppose", but "broad support" and "broad oppose". — Armed Blowfish (mail) 00:58, 8 April 2007 (UTC)

agreed, I think those votes should be moved accordingly, just like it would in a RFA poll. --RiseRobotRise 01:06, 8 April 2007 (UTC)

Same here. My opinion was "This is terrible as is, but it may be possible to do it well". Counting that as being neutral towards "the merger" completely misrepresents it. The question asked was not "What do you think of this merger?", but "What do you think of the idea in general?" -Amarkov moo! 01:20, 8 April 2007 (UTC)

* What do you think would work to resolve our little policy problem here? --Rednblu 01:18, 8 April 2007 (UTC)

- Either the policy has to be rewritten so that nothing is changed, or people have to stop insisting that nothing was changed. I can't really evaluate it until one of those two is done. -Amarkov moo! 01:20, 8 April 2007 (UTC)

While I did not vote in the same section as some of the commenters above, I do concurr that the wording in the header does not clearly summarize the intent of the respondents. The vote count summary should precisely match the headings of the sections where the votes were entered so no possible bias is introduced by any change in the words. If it does not match (as currently is the case), then the summary is not accurately reflecting what the editors responding were seeing when they entered their votes and comments. That could affect the perception of the results of the poll for someone who only reads the summary, or even someone who reads the summary first and then reviews the votes.

Here is a suggestion for improved wording to avoid biasing the summary of the responses in any way:

424 poll responses were entered in this section: In broad support of WP:ATT
354 poll responses were entered in this section: In broad opposition to WP:ATT
102 poll responses were entered in this section: Neutral/qualified/compromise/other

I request replies for consensus on this (or for alternate wording suggestions). Once we have consensus on this, if we can attain that, I ask that someone with the authorization and technical tools and to implement the consensed change please make it so. Parzival418 02:29, 8 April 2007 (UTC)

I've changed the header to better reflect the headings in the poll. --bainer (talk) 03:45, 8 April 2007 (UTC)

Thank you for the quick action! Parzival418 04:00, 8 April 2007 (UTC)

Attempt to summarize

I've gone through each of the votes for and against and have attempted to summarize them. I think that capturing the issues is more valuable than counting votes, per se. The key points for were much easier to capture than those against, which were more numerous and nuanced than the former. The latter could probably be further condensed, but I'm getting "editor's fatigue", so I think I'll just post them as they are at this point. They are not listed in any particular priority, just that general order in which they fell while I was trying to consolidate similar thoughts/issues/concerns. I've striven to be comprehensive, but I'm sure I must have missed something or two.

My intent in producing this synopsized listing is the hope that it will serve as a basis for identifying the more fundamental issues and coming up with ideas about where we might all go with this toward finding a more consensual outcome. My gut feel from reading all this is that a more agreeable product can evolve from WP:ATT, but it's going to require some further compromise on both sides. Please add your comments, observations, suggestions, etc. beginning in the Comments subsection that follows at the end of these two lists. Cheers, Askari Mark (Talk) 02:59, 8 April 2007 (UTC)

In broad support:

'Attribution' (ATT) does not change policy, but consolidates and clarifies similar and mutually relevant material; this is specifically the case for merging 'Verifiability' (V) and 'No Original Research' (NOR), but there remains disagreement regarding 'Reliable Sources' (RS). Essentially, most see V + NOR = 'Attribution' (which is "at the core of what we do"), and "Attribution to Reliable Sources" = Verifiability.
Most do not see that the two or three documents being merged in ATT are "different, overlapping, but distinct aspects of the same thing" or else they feel those differences to be semantic and trivial, especially with regard to their practical application. The policies may be conceptually distinct, but they are practically equivalent.
A consolidated document greatly eases the reading, learning, referencing, maintaining, updating, deconflicting, and understanding of these fundamental aspects of Wikipedia – something generally absent and impractical with the status quo – for both new and experienced users. Fundamentally, simplification = better understanding.
Streamlining how this set of policies is defined and described will make it easier for editors to reference them and for new editors to understand the guidance they provide. While these related policies have been on separate pages, they have often become inconsistent. Those who maintain policy pages – which are a constant target of single-issue edits – have a terrible time trying to keep them in sync, and a merged policy page would greatly ease their burden.
ATT provides far superior explanations and elucidations of these "fundamentals" to those currently present in the existing separate articles individually.
Condensation of Wikipedia’s policies is essential to prevent "instruction creep", especially as the project grows ever larger and has to create additional mandatory/core policies like 'Biographies of living persons' (BLP) to deal with new challenges it faces. In the beginning NPOV was our only really dominant policy; NOR and V developed as independent attempts to address some of the same problems. 'Original Research' (OR) is and always has been a special case of lack of verifiability/attributability. ATT gets us to coordinate our policies in a way we really should have done long ago.
The fewer policy pages there are, and the more detail in those policy pages, the less opportunity there is for wikilawyering and holding "the letter above the spirit of the law".
Consensus was already achieved over several months of work on ATT.
Fundamentally revisiting and re-thinking our core policies every three or four years is necessary to keep Wikipedia a healthy, learning organization.

In broad opposition:

Most agree with Jimbo's statements that each policy expresses a significantly different idea. They don't agree that the combined policy is either clearer or more streamlined, but rather assert that it obfuscates important aspects of our policies and weakens each of them. Merging them would only serve to confuse or dilute them, rather than clarify and strengthen them. There may be deficiencies in the existing documents, but this is not the way to fix them. The merge into ATT is exactly opposite of what should be happening.
This merger is not at all what many supportive poll respondents perceive it to be. It is not a 2-into-1 or (with RS) 3-into-1 merger; in reality, is it a 3-into-2 merger or even 2-into-2 (since ATTFAQ would exist whether it incorporated RS material or not).
It is not true that with ATT "we aren't changing things, we are only clarifying things". ATT actually makes major changes to our policies, both in breadth and in spirit. For example, the "unpublished synthesis of published material" (WP:SYN) section is new policy not previously addressed by RS. Moreover, although it is claimed that ATT only merges "existing policies", the actual process followed was one in which existing policies were molded to ATT before work on them was somewhat abandoned.
The "synthesis clause" as a subset of WP:OR should not be lumped together with fundamental 'Verifiability'. First we need to cite our references (WP:V and WP:RS), then we can worry about scenarios where there is OR (even with the references).
It's important to keep policy pages separate for the sake of clarity. 'Verifiability', 'Reliable Sources', and 'No Original Research' are very independent components of high-quality Wikipedia pages. Each one is a stand-alone principle. Having these policies merged would create a policy that conflates too much. Something is lost in most mergers, and the result of merging these as currently enshrined in ATT will inevitably dilute the understanding and even-handed application of all three as they are demoted to being just single points among many. Moreover, over time the merged document will evolve and that means there will be an unavoidable progressive mixing and recombination of its contents, further diluting and confusing these key principles. Such dilution of our policies makes it harder to defend Wikipedia against editors who do not have high standards for citing reliable sources.
WP:V and NOR are perceived as being very different concepts with V having little, if anything, to do with NOR. Along with RS, they are seen as quite independent (although some see them as interrelated) components of high-quality Wikipedia pages, and each one is a stand-alone principle. They see the development of ATT as reinforcing the widespread misconception that they are merely two sides of the same thing. Moreover, a lot of 'Original Research' (OR) which is found in Wikipedia articles is of the "synthesis" type, and is perfectly verifiable; it is the putting them together which constitutes the original research. Eliminating separate V and NOR pages will make it even harder for newbies and others to understand and appreciate their distinct and unique contributions to the whole. What is needed is more focused work on improving these separate documents (particularly RS).
Conceptually, the merger misses the mark by blurring important distinctions and placing undue weight on peripheral issues, notwithstanding adherents' assertions that it reflects the existing policies. In particular, the resulting product ('Attribution') suggests that the fundamental test for inclusion at Wikipedia is one of form rather than substance. "Attribution" was originally designed to ensure a representation of the variety of views that exist on any given subject. While this remains a laudable goal (especially on controversial topics), placing it up front and center creates the impression that the existence of a source for information is more important than its quality.
While 'Truth' is generally (but not universally) recognized to be an unachievable standard, they perceive the purpose of these policies as being to try to get as close as possible to an objective presentation of the facts – i.e., to ensure that Wikipedia is a high-quality, reliable information resource. That central aim is obscured and rendered far more difficult (if not made impossible) by the current proposal; worse still, "attributability" trumps "accuracy", which greatly weakens Wikipedia's ability to resist POV-pushers.
The chosen name, 'Attribution', actually seems to promote WP:OR at the expense of WP:V, although the text is neutral. Furthermore, Wikipedia:Citing sources, which is also directly related, isn't even mentioned in the poll.
ATT also encompasses the promotion of RS from a (very good) guideline to a policy. However, there is disagreement over whether this elevation is a good idea. Those that support it claim that doing so would strengthen the reliability of sources — as well as strengthen verifiability (although some say ATT strengthens RS but actually weakens V) — but these usually deprecate merging NOR. They see these three key content policies as operating in different ways, and feel that the best way to reinforce this is to have separate pages.
Those opposing the elevation of RS to a policy are leery of the impact. Elevating WP:RS to policy globally skews the whole whole rationale for and approach to citing sources.
Some don't mind if WP:V and WP:RS are merged, as they could easily cover the same material; but most feel that WP:NOR must be kept separate; 'No Original Research' is a focus that defines an encyclopedia and Wikipedia itself, in that it is the publication of a compilation of work from other people, and that needs to be stressed in its own article. Verifiability and reliable sources are important for stylistic reasons, in that for people to write good articles, they need to reference correctly and use the correct type of sources required for a good encyclopedic article. They are different things.
In practical terms, the policy pages' most important role is as a guide to settle content disputes, which requires tools of precision. Policy components need to be formulated in bite-sized concepts in order to be of value to users.
WP:ATT is too cumbersome to read. It's unrealistic to expect a newcomer to really read in one sitting all of WP:OR+WP:V+WP:RS...or an even longer WP:ATT. The 'Simplified Ruleset' is a much better place to send them to start. The longer the article, the sooner "reader fatigue" sets in. A large page is harder to follow and digest, and therefore less likely to be observed. Moreover, it is much more confusing to newbies to be referred to ATT, with its overwhelming abundance of information, than to specific separate pages where specific rules may be more easily learned. It's easier for newbies to understand V, NOR and RS one at a time rather than trying to take it in all at once in a huge gulp. Some, though, see a possibility that ATT could serve as a useful guide for providing an understanding how all three of these principles interplay. However, while these objectors feel it's a good idea to have the ideas of each all in one place, they believe it's a poor idea to do so for the purposes of making new policy.
The 'Reliable Sources' part must be hashed out more. The wording in the old RS guideline is much stronger. All kinds of controversial or even false things can be "attributed" to generally 'reliable sources'. In controversial topics (not just scientific ones, either), RS is essential to prevent nonsense, or anonymous IPs posting so-called "reliable info" from their own websites. Indeed, "attributable but not necessarily attributed" can be a means of avoiding providing sources for material, while verifiability supports that notion that Wiki's credibility depends upon the ability of our readers to verify that our information comes from reliable sources. "Attributable but not attributed" opens the door to "experts" to challenge the need to cite information. ATT falls into the "expert" problem by the subtle switch from verifiable to "attributable to someone, somewhere, sometime", but unless you're an expert on the topic, you don't have the right to ask exactly where it's attributed. Verification and NOR are not synonyms. 'Truth' – at least in the form of 'accuracy' – matters, not just 'attribution'. "Verifiability, not truth" is a much clearer, more accurate and more elegant expression than this clumsy "attributability to a reliable published source, not whether it is true" monster. Otherwise, you are really suggesting that an editor who knowingly includes cited, but false, information, is making 'helpful' contributions in writing this encyclopedia.
OR, RS, and V answer three very different questions: OR says "Did you find this or do it yourself?" RS says "If a group with an agenda on a topic says something outrageous, is it appropriate to add?" and V says "Did someone make this up or is it real?" ATT seems to be "It's OK if it came from someplace, we don't care where." The original three policies do not apply equally in all cases (a source can pass V and OR and still not be RS). Therefore, they should be separate. ATT undermines the encyclopedic process.
The core policy of WP:ATT that "[t]he threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true" is profoundly misguided. The threshold for inclusion in Wikipedia should be accuracy resting on facts "attributable to a reliable published source." Facts should always and in all circumstances be attributable, but provided that articles conform to NPOV, arguments not only need not be externally attributable, but requiring them to be so gives POV editors a sword to wield against attempts to correct plainly wrong information. In short, NOR suggests that WP would rather an article be attributable and wrong than accurate. That is an open invitation to POV.
Wikipedia has developed a culture which is a vital component to its success – more vital than the exact wording of its policies. The terms V, NOR, RS, NPOV, AGF, etc., are part of the language of that culture. We have thousands of pages of discussion where those terms are used. Merge the terms and over time those discussions, where various boundaries have been hammered out and consensus formed, become meaningless to newcomers. Absent a compelling problem with the existing structure that can't be solved any other way, the importance of retaining our traditional policy structure weighs against a merge.
Merging policies into a single place for "simplicity" is not necessarily a good thing in of itself. If we wanted, we could merge all of Wikipedia's policy into one concise page, but it would be ridiculous and thus it's the same thing here.
ATT does not have, and never had, consensus – either prior to its being elevated to "policy" or now. It is the product of a small "cabal" developed out of due process for achieving "consensus" on fundamental aspects of what Wikipedia is all about. Opposition seems to have been ignored, and some very bizarre arguments were used at times for not listening to editors.

Comments

Excellent work. This seems like a very good summary of most of the key arguments. It is interesting to see how the actual arguments put forward differ from both the pro and con position papers. --bainer (talk) 03:40, 8 April 2007 (UTC)

A very good summary indeed. -- Ned Scott 03:45, 8 April 2007 (UTC)

A thing of beauty, even. Clear even for newbies. Jfarber 03:47, 8 April 2007 (UTC)

What is interesting is that the summary above, condenses 424 comments in broad support into 9 points, and 354 into 20. This could mean that there is strong consensus about the raison d'etre of ATT amongst those that supported it, and that there is a lack of clarity about these reasons by those editors that opposed it. Food for thought. ≈ jossi ≈ (talk) 04:16, 8 April 2007 (UTC)

Well, I suppose it could mean that, though that seems to me less likely than that in this kind of process, commentators sometimes want to make new points, even if they're minor. One way or the other, the diversity of opinion suggests that it will be hard to form a concensus around WP:ATT. —The preceding unsigned comment was added by Bucketsofg (talk • contribs).

It could also be a confirmation that people who like ATT were the people who worked on it over the last few months (and have thus had plenty of opportunity to clarify their thoughts on it), and that more people who don't like it have only recently become aware of it. --bainer (talk) 05:23, 8 April 2007 (UTC)

The oppose section included gems such as "The core policy of WP:ATT that "[t]he threshold for inclusion in Wikipedia is whether material is attributable to a reliable published source, not whether it is true" is profoundly misguided, shows a total lack of understanding of our core policies, regardless of ATT. Given this, a continuous and vigorous debate about the importance of these core policies is necessary to continue. ≈ jossi ≈ (talk) 04:19, 8 April 2007 (UTC)

Another misguided notion is the one that asserts that RS was a"very good" guideline. Well, that was not the case, until RS was redesigned based on the work done at ATT. ≈ jossi ≈ (talk) 04:21, 8 April 2007 (UTC)

That's an interesting point. My impression was that WP:RS (at my last consultation of it) was better than the ATT/FAQ. But it is surely the case that improvements can still be made. Bucketsofg 05:08, 8 April 2007 (UTC)