11
$\begingroup$

I understand the need to make questions clear when asking them and agree telling the OP to make the question clearer, if it is not, is a good thing. But can't this be done without locking the ability to post potential answers?

Say for example a question seemed unclear to the mod, but was understood by someone who potentially knew the answer: all that is being achieved by putting the question on hold is stopping the answer from being submitted. And if the question seemed unclear to the mod, and was unclear to someone who did know the answer, then no answer would be submitted anyway, and there would be no need to put the question on hold.

Furthermore as an extension to the first scenario where a potential answer is blocked if the OP did not edit his question or forgot to, then there would be no way of answering the question forever for the potential benefit of all others who might face the same problem, which in my opinion is a gross failure of the philosophy of this website and what it tries to do in helping people through knowledge sharing.

So my question is, what advantage does putting unclear questions on hold actually achieve over just asking the OP to make the question clearer?

$\endgroup$
  • 15
    $\begingroup$ As a general proposition, the worst possible action is to attempt to answer an unclear question: unless the answer is very well crafted, future readers will have no way to know what interpretation of the question it addresses. That's not "knowledge sharing": that's simple confusion. Closing does not affect anybody's ability to post comments to ask for clarification. $\endgroup$ – whuber Sep 25 '17 at 13:10
  • 2
    $\begingroup$ Agreement with many of the excellent points made on this thread. However, the OP raises a point which deserves underscoring and that's the standards to which queries are to be held. Are they high and, therefore, more or less readily intelligible by the many quantitative experts on this blog or can they be permitted to be couched in more vernacular, nonexpert terms? An example is this query where the OP marked it as "answered" only to have the query put on hold 1 day later...stats.stackexchange.com/questions/305006/… $\endgroup$ – Mike Hunter Sep 28 '17 at 16:41
  • $\begingroup$ There appears to be a consensus among the senior participants of this blog that there are two standards to which CV threads are held: 1), creation of a high-quality, online statistical resource and, 2), ensuring intelligibility to future readers. These are laudable objectives. My only observation is that these standards are quite inconsistently enforced for if they were, then the relatively high proportion of posted queries receiving no comments, edits or responses of any kind would be much lower (last I knew that proportion was running about 1/3rd of all posted queries). $\endgroup$ – Mike Hunter Sep 29 '17 at 10:58
  • 2
    $\begingroup$ @DJohnson Your logic is puzzling. I see no connection between the rate of unanswered questions and consistency of moderation. You seem to be suggesting that this community ought to summarily remove any question that has not elicited any response within some short period of time. Although that would be feasible, I'm sure it would create an uproar amid allegations that we were trying to hide our failures, disguise our inadequacies, favor only those topics we currently understand, etc. If you seriously think we should go in that direction, then please start a Meta thread about it! $\endgroup$ – whuber Sep 29 '17 at 14:46
  • 1
    $\begingroup$ Untouched questions (ones without answers, comments or votes) are eventually removed automatically in any case. Better for a question to be placed on hold and improved than ignored in any case. $\endgroup$ – Glen_b -Reinstate Monica Sep 29 '17 at 23:46
  • 1
    $\begingroup$ One thing worth keeping in mind is that while five users with sufficient reputation (or one moderator) can put a question on hold, five users with sufficient reputation (or one moderator) can also reverse it. It's a temporary state which ultimately helps the OP get to a question that can attract a suitable answer. $\endgroup$ – Glen_b -Reinstate Monica Sep 29 '17 at 23:50
  • $\begingroup$ @whuber Apologies for assuming too much. Let me try explaining it one more time, if that doesn't work, it's not worth further effort. First, you did not object to the two "standards," objectives or policies (call them what u will) that I proposed: confirmation that they were not in error. A corollary to these unwritten "policies" is the consistency with which they are enacted or observed. That fully 1/3rd of CV queries go unedited, uncommented on and/or ignored suggests the haphazard, arbitrary inconsistency wrt observing these unwritten "policies." This seems key to the OPs main point. $\endgroup$ – Mike Hunter Oct 3 '17 at 13:00
24
$\begingroup$

This is misleading in so far as it implies that closing is always and only done by a moderator. On the contrary, closing as unclear is often done by non-moderators without any moderator intervening. Then a consensus of five high-reputation users is needed.

The point otherwise is well taken: "appears unclear to me" does not mean "is unclear to everyone": it could be sufficient that one competent person understands it and can answer. And judgments are fallible.

I would argue, however, that present principles and practices are about right. If anything, we need to work harder at identifying unclear questions (more generally, questions that aren't good enough) and either getting them improved or closing them down.

First, deciding that a question is unclear is often much easier than understanding what the question is. Without trying to be exhaustive, here are some common examples.

  • The question boils down to "What should I do with my dataset?" but that and the underlying problem are too vaguely described to allow good advice.

  • The question is too brief and cryptic to be clear.

  • Presentation is just too messy to attract all but the most noble readers.

Second, the gentle model -- just leave it there; someone may understand -- is not nearly as kind as it seems.

  • It is in the original poster's best interests to get prompt and decisive feedback on the question. They can then try to improve it. If they don't, and/or if they get a negative impression of CV, that's unfortunate in some senses, but keep reading. If they can, a much better question is much more likely to get a good answer.

  • It is in nobody else's interests that CV is cluttered with questions that no one understands -- or wants to answer. Indeed, this is true already! The fallacy here is that CV is a helpline in which ideally every questioner and every question deserves an answer. Not so; the ideal is to build up a site with an archive of good, distinct, well-answered questions. Every question is a candidate for that status, but many questions fail.

Third, there are enough checks and balances to correct over-zealous voting to close. Needing 5 non-moderator votes has already been mentioned.

  • The SE model does tend to mean that people with vote to close privilege have done enough around the site to have read lots of questions and answered many. That doesn't rule out puzzling votes to close, but no system is perfect.

  • Answering a question is the best way to show that someone understands it -- so long as they do understand it! Guessing wildly and guessing wrong won't help.

  • Do your bit by editing a question, or proposing edits, that make it clearer. Sometimes just a little care and attention can tip the balance. It could be something as simple as a better title.

  • Comment in threads where you think voting to close is wrong.

  • A vote to close brings a thread to the attention of high-reputation users who are happy to disagree with each other if some see merit in a question. In this sense a vote to close as unclear from a non-moderator is no more than "this seems unclear to me; do others agree?".

$\endgroup$
  • $\begingroup$ Edited to mention an excellent point brought out by @Scortchi's answer. $\endgroup$ – Nick Cox Sep 25 '17 at 13:37
  • $\begingroup$ and @whuber's comment. $\endgroup$ – Nick Cox Sep 25 '17 at 13:37
20
$\begingroup$

A small point not covered in the excellent answers so far: a premature (though well intentioned) answer to a currently unclear question can do more harm than good. When a question bears more than one reading; makes false assumptions or relies on implicit, unexamined premises; lacks crucial context—answering it can create much confusion, even mislead the OP & other readers. Better to first contribute to its clarification.

$\endgroup$
17
$\begingroup$

I understand the need to make questions clear when asking them and agree telling the OP to make the question clearer, if it is not, is a good thing. But can't this be done without locking the ability to post potential answers?

If the question is unclear, then it is rather unlikely to get a high-quality answer and we don't want this site to be a repository of low-quality questions and answers.

On another hand, I agree with you that sometimes we, as a community, lock questions too quickly. I myself have answered a few questions that simultaneously got put on hold as "unclear". I agree that the mechanism is not perfect. However, notice that every question can be re-opened, or even un-deleted: you just need enough community votes, or a moderator for that. So if you feel that some question should be re-opened, then you can always (a) put this in comments, if there is an active discussion under the question, (b) discuss it on chat (mods often visit chat), (c) discuss this on Meta, (d) flag the question for mods' attention asking to re-open it.

Say for example question seemed unclear to the mod, but was understood by someone who potentially knew the answer, all that is being achieved by putting the question on hold is stopping the answer from being submitted. And if the question seemed unclear to the mod, and was unclear to someone who did know the answer, then no answer would be submitted anyway, and there would be no need to put the question on hold.

First of all, it is not only the mods who can put questions on hold. I'd argue that in most cases questions go on hold by gathering community votes. So in most cases you need at least three people to agree that some question is unclear. In most cases, if it is unclear for three people, then it is unclear.

Furthermore as an extension to the first scenario where a potential answer is blocked if the OP did not edit his question or forgot to, then there would be no way of answering the question forever for the potential benefit of all others who might face the same problem, which in my opinion is a gross failure of the philosophy of this website and what it tries to do in helping people through knowledge sharing.

If OPs didn't bother to edit their question that were unclear, i.e. do not care about their questions, should we have to care about that? The community does not benefit by collecting answers for unclear questions. Moreover, if an OP does not care about their question but you feel that behind it there is an interesting question to be answered, you can always ask the question, and potentially answer it yourself. Last time I did this I got +156 votes for the question; moreover as a result the OP has edited the original question and it got answered.

So my question is, what advantage does putting unclear questions on hold actually achieve over just asking the OP to make the question clearer?

Making the question clearer is the advantage. Moreover, it is a mechanism that guards the site from being flooded by low-quality questions, which would lead the site to start looking like the "SPAM" folder in a mailbox and become useless.

$\endgroup$
  • 2
    $\begingroup$ +1. Much overlap with mine, which I take as all to the good. $\endgroup$ – Nick Cox Sep 25 '17 at 8:31
  • 3
    $\begingroup$ posted first while I was drafting mine.) $\endgroup$ – Nick Cox Sep 25 '17 at 8:54
  • 1
    $\begingroup$ Main problem is that there are a lot of questions which were just out of the knowledge base of moderators and marked as unclear/off-topic, while at the same time a lot of junk gets promoted because mods decided to have some fun. For example I had a question (already deleted) about ethics in statistical science which was immediately closed as off topic, while what is your best statistics cartoon gets hundreds of answers. $\endgroup$ – Cagdas Ozgenc Sep 28 '17 at 10:59
  • 5
    $\begingroup$ @CagdasOzgenc please consider the facts that (a) nobody is monitoring the site 24/7 and looking through all the questions that are posted, the review is selective (if you see something vote to close or flag it), (b) there is lots of old currently-off-topic threads that were not off-topic in the present standards (from time to time we close, protect, or lock the threads). Moreover, I see no problem with keeping one or two threads with jokes and cartoons, we don't have to be that serious. $\endgroup$ – Tim Sep 28 '17 at 11:23
  • 1
    $\begingroup$ Aside from whether we need to be completely "serious", I think there is something to be said for grandfathering some old threads from when standards were different, rather than pursuing at all costs a hobgoblin of consistency $\endgroup$ – Silverfish Sep 28 '17 at 22:23
  • 1
    $\begingroup$ @CagdasOzgenc Your first claim is not substantiated and not really testable, which is all the more crucial because it contains an element of criticism of identifiable individuals. I know of no dataset on moderators' expertise that lets anyone judge impartially if and when they are playing away from home. You should flag speculation as what it is. $\endgroup$ – Nick Cox Sep 28 '17 at 23:02
  • 1
    $\begingroup$ @ all above: I agree 'we' (i.e. random commenters) cannot judge the extent of the pool of moderator's knowledge. Why would we even speculate when it is obvious the mods do their utmost best to serve the community, but cannot always be right, let alone omniscient (although from my perspective some do seem so...). But all that aside, I do agree with @CagdasOzgenc, based on the (vague) idea of 'not knowing what you do not know'. A question about the ethics behind statistics seriously interests me, and I think a community debate about it being on or off-topic here on CV could send sparks flying! $\endgroup$ – IWS Oct 2 '17 at 11:51
  • 1
    $\begingroup$ cont'd So maybe @CagdasOzgenc would want to honor to start a meta thread about it? $\endgroup$ – IWS Oct 2 '17 at 11:52
  • 1
    $\begingroup$ @IWS It seems worth mentioning that a great many moderation decisions can be made without any subject matter knowledge whatsoever: badly formulated, vague, ambiguous, or unreadable questions or answers will need improvement, period. For moderation purposes, the main benefit of knowing anything about the subject lies in being able to find duplicate or closely related threads. Indeed, there is a standard boilerplate reason for moderators to decline flags on posts that goes something like "flags shouldn't be used to indicate wrong answers": it's not the mod's job to assess correctness. $\endgroup$ – whuber Oct 9 '17 at 16:15
  • 1
    $\begingroup$ (Contd) Having said that, I ought to add that a broad knowledge of the subject matter and its terminology can be a bit of a curse, because it means one can readily see multiple interpretations of many questions. Where most readers might make only one interpretation--whichever one conforms with their experience--a broadly experienced reader may see many possible meanings. I would therefore suggest that anyone who believes a question has been unfairly closed first think very hard about the possibility they might have overlooked plausible alternative interpretations. $\endgroup$ – whuber Oct 9 '17 at 16:21
  • $\begingroup$ @whuber, I agree. The point I was trying to make was supporting CagdasOzgenc. A few comments above he introduced ethical considerations into the mix of possible interpretations of Q&As. I believe a thread about whether or not the community wanted guides on how to deal with such differences of opinion/world view would be valuable. Lo and behold: stats.meta.stackexchange.com/questions/4970/…. That said, I do not think the moderators put question 'on hold' too often or unjustly, but would support keeping this/a discussion going anytime ;) $\endgroup$ – IWS Oct 10 '17 at 7:24

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .