I just noticed that Kevin O'Bryant tagged a few closed questions with the tag "closed". May I ask what the rationale is? It seems redundant.

I believe Mark is right. I think it's a good move, but in my opinion a question that is closed should be removed from the front page.

I don't think we should have a [closed] tag. I also don't think we should have a special mechanism for ignoring closed questions.

If a question is closed because it's totally off-topic or is somehow a clear-cut "bad" question, then it should just slip off into oblivion. Retagging it [closed] only makes it more visible. The oldest question on the home page is currently 13 hours old. Of the nine questions that have been tagged [closed], none of them would be on the front page right now if they had not been retagged. I realize that this argument doesn't apply quite as well if you use the questions page rather than the home page, but I think it still holds water.

If a question is closed, but is somewhat controversial, or is subsequently revised, then it should be highly visible so that people can vote to reopen it if appropriate.

I propose removing the [closed] tag from the ecosystem (via a moderator merging it into [tag-removed], as usual), but I'll hold off on doing that for a while in case somebody has an objection.

We might expect some questions to be repaired and reopened, and there is no reason to remove the tags from them.

Yes, we should remove all tags on closed questions. This should be automatic upon closure.

I disagree. I just don't see what good that would do. As Scott said, it is often expected that the question will be revised. Even if you expect a question to be closed permanently, why would we want to remove tags which are appropriate to the question? As François suggested, the whole point of keeping a question around if it is closed as a duplicate is to make it easier for people to find the original. Moreover, removing all the tags from a question would make people take closure even more personally than they already do, which would be awful.

10k+ rep users can vote to delete closed questions, so even if we do get a tsunami of terrible closed questions, we don't have to worry too much about them getting in the way.

Hi all! What I was aiming for was a way to call up all of the closed questions. I've gotten email from some people indicating that they believe there's some bias into what gets closed, and how quickly. I wanted to look through all closed questions, rather than falling victim to anecdotes. I stopped tagging when I realized that it was pushing these (almost uniformly) awful things onto the front page.

Is there another way to accomplish what I was trying? For instance, how many closed questions have 10+ upvotes? For that matter, how many closed questions are there?

Kevin, you should do your analyses on the public dumps of the database. There's links to it somewhere around here. Then your experiments won't affect the main site.

I didn't know about the public dumps before now. BTW, I'm not expecting answers to the questions in my post a few minutes ago; I was just giving examples of the sort of question I'd like to be able to answer. The dumps should enable me to do so.

Are the highscoring questions not the soft questions that had been around for a while but were closed after, at least in some cases, the OP had asked for them to be closed? At least there was a consensus to close them, if I remember correctly. If so they skew the data a bit don't they? I mean they were not closed because they were bad but because they were ok in the beginning but had run their course.

@Gretar: you're right. Certainly, the closed questions with votes 27, 21, and 14 should be removed from any data analysis - those were my questions. Two were closed by consensus and the third (at 27) is the infamous "Singing in the rain" question.

The high-vote closed questions were generally closed because the new answers coming in were of declining quality or were repeats of older answers, and it didn't seem like anything more could be said on the subject.

@obryant: I think your suggestion of a "retired" tag is a good one. I don't think there is any discussion in the public dumps regarding the closure of such questions, but previously there has been some discussion in meta about this. Basically, in addition to what Qiaochu said, the consensus was to close some of these older "stale" soft questions so as to reserve more real estate for fresher and more research-oriented questions.

I see that some of the closures are "closed as no longer relevant", while some are "closed as off-topic". Perhaps the "closed as xxx" phrase could be tweaked to ease hurt feelings and to give encouragement to rewrite when appropriate. Then again, that seems to be redundant, as the comments always have that info already. Never mind.

I'm still in favor of a retired tag and a closed tag (it wasn't my idea, but it would be nice to be able to screen the closed questions out of the RSS feed), but only mildly.

@obryant: I believe this came up previously as well. If I recall correctly, currently the SE software does not give moderators/admins the ability to rephrase the reasons for closure, unfortunately.

Are there any remaining objections to merging [closed] into [tag-removed]?

Excellent idea!

Because [tag-removed] has the same meaning as [closed], but we've been using it for 6 months.

When I cast the final vote to close on a question that is complete garbage, I remove the tags and replace them with [tag-removed]. If it seems like it can be reopened with a bit of work, I leave the tag.

I think that this is a good solution.

@Mark: it was when I said

Retagging it [closed] only makes it more visible. The oldest question on the home page is currently 13 hours old. Of the nine questions that have been tagged [closed], none of them would be on the front page right now if they had not been retagged.

I guess retagging with the [closed] tag doesn't make things any worse for people who have [closed] on their ignored list and who have set the preference that ignored tags are hidden completely, but it's really annoying for everyone else. In particular, for "lurkers" who are just casually reading MO (we know there are a lot of them), it cultivates the impression that MO is more and more heavily populated by garbage. I also think there are good arguments for not setting the preference that questions with ignored tags are hidden completely.

If there's another mechanism to keep them from appearing on my screen at all when I look at MO, I'd be interested to know what it is.

Here's one solution, but it requires you to push a button. Create a new bookmark in your browser's bookmark toolbar. For the "location", give it

javascript:$('h3>a:contains("[closed]")').parents(".question-summary").hide();void(0);

Now whenever you're looking at a list of questions on MO, you can get all the closed questions to disappear by clicking that bookmark.

Because [tag-removed] has the same meaning as [closed]

I disagree. You should never add the [tag-removed] tag unless there are no appropriate tags for the question. The meaning of [tag-removed] is that there was some tag that I (or another moderator) decided shouldn't exist. There are lots of questions with the [tag-removed] tag which haven't been closed. If you see one near the top of the home page, please remove the [tag-removed] tag from it.

@Harry: sometimes there are good reasons for retagging a question solely with [tag-removed], but they are few and far between. In the situation where you want to do it, you can probably be pretty confident that the question will be closed and you'll basically never see it again. In that case, the easiest, least disruptive, and most diplomatic thing you can do is leave the tags alone. Remember that people take it pretty personally when one of their questions gets closed. Retagging it with something like [tag-removed] or [spam] sends the message that you actively think the question is worthless. This will be incredibly annoying to the question owner, regardless of how inane or worthless the question was.

If you want to retag something with [tag-removed], consider the tradeoffs. What do you gain by doing it? Is it worth bumping the question and annoying the owner? The most reasonable situation where it makes sense is if the question keeps being bumped, but that tag doesn't match the question. In that case, you can help the situation by leaving a comment like "I've removed the [graph-theory] tag because this question is about the graph of a function, not about graph theory." Try to make it clear that you're not retagging the question as a way to emphasize how bad it is (but don't use those words).

I think we should try to be civil --- both outwardly and inwardly.

@Harry: It seems like you want to punish people for asking bad questions, presumably in the hope that either this will discourage them from asking more bad questions, or it will make an example of them for future users of the site. Such a policy may lower the number of bad questions, but I feel that it may alienate mathematicians who would otherwise contribute positively. I would consider this a suboptimal outcome.

I would prefer a scenario in which the people asking bad questions came away from the question-closing interaction with the impressions that 1) we mathematicians are generally nice people, but 2) their questions are definitely not well-suited for MathOverflow. The act of removing tags unnecessarily detracts from the first impression, and doesn't really add to the second. While our first priority in exercising vote-to-close powers is to maintain a high standard of content, I don't think it requires much additional effort to foster good relations with neighbors.

@Harry : I think "contempt" has no place in scholarly discourse. Even in regards to people who ask bad questions.

Harry, "I have contempt for people" pretty much summarizes your whole attitude. I am a real mathematician, actively participate in MO, at least for now, and you've demonstrated contempt for me as well. From what I've seen in meta and, to a lesser extent, on MO itself, your attitude is at best unhelpful and at worst, harmful. Kevin Lin gave you very good advice:

I think we should try to be civil --- both outwardly and inwardly.

Now, I've seen many questions that, in my opinion, clearly fall below the threshold of MO as defined by FAQ. Some of them get closed after getting downvotes, some of them generate helpful comments and even answers, occasionally even gaining upper single digit reputation. I also see very tactful efforts (sometimes, exceedingly so) to get the best out of a question. At the same time, I see partisan efforts to close perfectly reasonable questions on ideological grounds. That works like a double whammy: people wonder why instead of uniformly enforcing the clear rules about closing too elementary (routine homework type) and purposefully vague ("tell me about X") questions, the powers that be are spending their energy arguing over nebulous issues "Is this type of question good for the future of MO?"

I don't know what the best answer might be for dealing with this situation, that's something that needs to be carefully discussed and implemented on a systemic level. But retagging questions and especially making disparaging comments about users, whether here or on MO, isn't going to make the problem go away. It does, however, tend to discourage some good people from participation due to the surrounding drama and incivility, even if they wouldn't say so directly.

You have made plenty of disparaging comments to specific users; you have definitely called more than one user on meta a liar. Let's not go down this road again; the point is that many people have a certain perception of you regardless of how justified you think that perception is.

Fair enough. I apologize for bringing up the issue at an inappropriate time.

Returning to the issue at hand, let me see if I can summarize the current direction of the discussion. There is some disagreement about how freely the tag [tag-removed] should be used. Several users believe it could be seen as insulting and should only be used sparingly, whereas you wish to use it to actively discourage bad questions. Yes?

I think Scott, Anton, and the others made good points. We should always be ready to give users and questions the benefit of the doubt, and [tag-removed] is an unnecessarily aggressive move in many situations.

My apologies again! I misread Anton's stance on the subject. I was somehow under the impression that he was talking about borderline questions.

I now see that I disagree with Anton and some of the others. To my mind, the point of removing tags is so that people who want to browse questions by tag (e.g. because they only feel qualified to answer questions in a particular subject) can do so without having to see irrelevant or low-level questions, and the point of [tag-removed] is as a placeholder because the system requires that every question is tagged.

At the time that I used [tag-removed], I didn't see it as an aggressive action (at least, not any more aggressive than closing the question). I may have to revise this point of view, but I still think that in the interest of making it easier for mathematicians to see and answer research-level mathematics questions, low-level questions should not have tags. I guess I agree with you after all, more or less!