This new user's first question got closed as 'not a real question', way too abruptly in my opinion. As a first question from a new user it was pretty ok, considering what else is out there. He did post code - admittedly too much - but if you read his code you can in fact see where he was having difficulty, even though he did not clearly state the code's intent. All he needed to do was trim it down and expand on precisely what the desired behavior was. (Bear in mind not using OO makes that more difficult - maybe he doesn't know OO.)

I posted an answer, then only 2 hrs later, before he could even respond, it got closed. I am too lowly to vote to reopen.

I think this is unreasonably abrupt and harsh to newbies. Do we have a consistent standard of how much time and warning we give a newbie before closing as 'not a real question'? Or do we just guillotine it? I would at least have given him 48hrs - it's different to if he was told to clarify it and given time and simply didn't respond. If there is any consensus, can we add that to the FAQ? I am happy to help Python newbies who show some effort, and I resent what might well have been a useful interaction getting cut off abruptly.

(AFAIK there is no automatic warning 'Please edit this within to improve the question, or else it will be closed - several users have voted to close it as not a real question')

(For the 'big dump of code' scenario, do we need a new closing reason 'question does not isolate the specific problem and solution' - which is distinct from the generic catchall 'unclear'?)

When should we downvote, when should we close, when should we leave a comment asking them to clarify, what is a reasonable time expectation for each of those, and how much extra allowance do we make for new users?

As far as user-friendliness of reopening goes, virtually no new users know this, nowhere in the FAQ does it state: What can I do if my question got closed? If the reason was clarity, you can edit your question and it will then be queued for voting on reopening it.

Related discussions:


People have a bad habit of thinking closure is permanent; you called it the "guillotine" in this case. The process is theoretically extremely simple:

  1. If a post is broken, it's closed, as soon as humanly possible
  2. If somebody is being nice, they leave a comment explaining what's wrong with the post
  3. The OP fixes the post, which pushes it back on the active list
  4. People see the fixed post and vote to reopen it

Unfortunately #4 is harder in practice than it should be, but people have a habit of stopping at #1 -- at least get to #3

  • Related-ish: meta.stackexchange.com/questions/70061 – Daniel DiPaolo Jul 19 '11 at 22:49
  • 2
    It IS a gullotine because the guy is unlikely to respond now. I cannot vote to reopen, and I will not get the privilege to vote to reopen for months, and it only attracted 1 other user's attention. Also, where is it spelled out to the new user that they can have it reconsidered if they edit it for clarity? What % of the time does it actually happen that closed questions get successfully reopened? I suspect <10%. Do you not think a warning and grace period are better than closure, at least for newbies? – smci Jul 19 '11 at 22:51
  • 1
    The status merely says 'closed', not 'closed until clarification added'... very user-unfriendly. Few new users will be aware they can get it reopened. – smci Jul 19 '11 at 22:53
  • 5
    @smci In addition to relying on the community to reopen the post, you can flag it for a moderator to review if it's ready to be reopened. An unedited bad post deserves to stay closed. – Adam Lear Jul 19 '11 at 22:54
  • 1
    @Anna: he's not likely to edit it now, so I would be wasting my time in flagging the closing. That's my point. The guy was not given a chance to fix it. Did you actually read his question and are you saying it was a bad post? I think it's okay and doesn't deserve to have been closed abruptly. – smci Jul 19 '11 at 22:58
  • 2
    @smci he still does have a chance to fix it, you can edit closed posts – Daniel DiPaolo Jul 19 '11 at 23:00
  • 2
    I doubt he knows that (it was his first question on the entire site), and I doubt that >10% of newbies with questions closed for NARQ do get them successfully reopened. Anyone (mods?) got any statistics on that? – smci Jul 19 '11 at 23:01
  • 4
    @smci It is pretty bad. There's no explanation of the actual problems he's encountering and just a wall of code. The post has a comment explaining what needs to be done to improve it and makes an explicit reference to editing the post. Seems fair enough to me. – Adam Lear Jul 19 '11 at 23:10
  • 1
    @Anna If a user flags a (fixed) question for a moderator to review and that request is rejected, does the user's flag weight go down? If so, it would seem to discourage people from flagging a question to be reopened given that there is an inherent bias against approving such a request. – Chris Frederick Jul 19 '11 at 23:33
  • 1
    @Chris I personally (as a moderator) would not dismiss a "Please review this question" flag as invalid. There's no way a request like that can be. But, I agree that it might somewhat depend on the moderator. Honestly, though, focusing on flag weight is counter-productive. How many questions do you (general you, not you specifically) flag for reopening every day? It's a pretty rare event. – Adam Lear Jul 19 '11 at 23:44
  • 1
    It almost never reaches #3 with a new person. Well, other people might fix it but usually not the OP. Way to be optimistic. :) – Jeff Mercado Jul 19 '11 at 23:55
  • 1
    @Jeff et al: Right. I'd say that alone proves new users aren't aware of this. Surely a case for more user-friendly warnings and statuses? – smci Jul 20 '11 at 0:03
  • 2
    @Michael: btw, on pretty much every other coding Q&A site, closing is permanent. You know that, right? Newbies will not know SO does this. I called out that the FAQ does not state it clearly. – smci Jul 20 '11 at 0:17
  • 3
    @smci: But in all cases, there is usually one or more people commenting on how the question could be improved (add more details, clarify some points, etc.). That is (IMO) much better than what an automated response could provide. And even then, more than half the time, the OP never responds or doesn't add any clarification that moves forward. I don't think there's much more that could be done about that. But you do have a point about letting it be known that it can be reopened. – Jeff Mercado Jul 20 '11 at 0:44
  • 1
    Really, who cares whether the know this or not? It's not like we're trying to encourage users who ask this type of question. – Cody Gray Jul 20 '11 at 1:46

We deal with a very large number of flags on Stack Overflow every day, sometimes the number climbs over one thousand. I'm the one who closed the question, and I'll explain my reasoning and then elaborate a bit more.

It wasn't just me that closed the question

While I am the only (and super) vote involved, I was reacting to multiple flags from high reputation users of Stack Overflow, who had (likely) ran out of close votes themselves. I agreed that the question was quite difficult to understand, I also agreed that the community tends to react negatively to questions that basically boil down to "please run my code and tell me how to fix it".

Closing is not a permanent state

If we want to close a question with prejudice, we also lock the question to prevent even high reputation users from modifying, un-deleting or re-opening something that simply does not belong on the site. Users are still free to disagree with us, however a discussion must happen (here) before the decision is reversed.

Simply closing the question stops the flags and rude comments and gives the original author time to improve their question with feedback from the community. If a question is sufficiently improved, it can and should be re-opened by the community, or reviewed and opened by a moderator.

The text provided under the closure notice is not at all difficult to comprehend:

It's difficult to tell what is being asked here. This question is ambiguous, vague, incomplete, overly broad, or rhetorical and cannot be reasonably answered in its current form.

If the user is not able to understand this, there's a very good chance that the user is also unable to sufficiently improve their question. This doesn't mean the question is going to go off to the meadow and die, but it will need some help from experienced users in order to be edited into shape. Still, during this time, the question needs to remain closed, or we're sending a clear message that we'll accept content that contradicts our quality guidelines.

Many poorly asked questions are never improved by the author

There are some users that quickly catch on to how Stack Overflow works and improve the quality of their writing as well as their interaction with the community. Unfortunately, most of the time (as of late), this simply is not the case.

I have, on may occasions left comments strongly suggesting improvements, only to visit flags the following week to realize that the advice was ignored or perhaps never even read. None of us have the bandwidth to star ever single thing we mark as needing improvement. To prevent disruption to the community and keep our quality bar as high as possible, our only choice is to close or improve poor quality questions on sight. If a question is really bad, beyond the point of being salvaged, we may elect to just delete it so it doesn't block a better attempt at a similar question.

I did consider improving the question

I clicked the edit button, started to edit and gave up. I could not accurately tell what the OP really wanted, beyond someone to run their code, point out the bugs and indicate how those bugs could be fixed.

With 200+ other flags in the queue, my best option was to cast my close vote and move on to other more urgent flags.

Given the current popularity of Stack Overflow, we have to realize that not everyone can be helped into our community. The trick is identifying those who can, then helping them. If the author of the question had made even an attempt at improving it to the point that the actual question could be discerned, I would have happily spent more time helping them make additional edits and re-opened the question.

Even after help from someone else in the community, the question as it stands now is still not suitable to be re-opened.

  • Tim, all of that is true, but what is your reaction to my observation that newbies do not know they can edit so that it may get reopened, and thus the process (and close notice, and FAQ) are user-unfriendly? Should we not add something to both the close notice and FAQ? – smci Jul 20 '11 at 6:06
  • 1
    @smci The vast majority of users see and use the edit link, and there's quite a bit of prodding about how the system works when new users ask a question. I think, at this point, the official stance is that we've basically done all we can do to help the type of user that would likely become a frequent and valuable contributor. That being said, anyone in the community is welcome to leave a comment and spend as much time as they'd like helping the OP. – Tim Post Jul 20 '11 at 6:16
  • That's not the specific issue here. What % of first-time question askers upon getting it closed, do successfully get it reopened? I bet you it's <10%. Tell me the number please. Anything that improves that number is IMO worth implementing. The FAQ is derelict on this issue and needs to be improved. – smci Jul 20 '11 at 6:19
  • 1
    @smci - I don't have a percentage, just a bird's eye view. So, I fired up incognito to go through the paces (as a new user would, to see what sort of help they get when asking) only to realize that Stack Overflow is no longer accepting questions from my IP when I'm not logged in. :/. What I'm pretty sure of is that there are a lot of measures in place to keep people from creating poor quality questions, and I know users see some extra informative links when a post is deleted. I'm just not sure if they see it when a question is closed. – Tim Post Jul 20 '11 at 6:50
  • @smci - A good way to proceed from here, since you feel strongly about it would be to open a new meta post to suggest specific changes to the FAQ. I would honestly like to help every single user that arrives, but unfortunately I simply can't. Stack Overflow is just way too large. – Tim Post Jul 20 '11 at 6:54
  • FYI all, here's an example of a worse question which has been left open for 5 months. – smci Jul 20 '11 at 9:44
  • @smci - when dealing with a system run by human beings, there will always be a 'worse' example. If you really feel passionate about this, open a new post to suggest edits to the FAQ. – Tim Post Jul 20 '11 at 10:29
  • 1
    This is a faq-worthy answer. I'd highly suggest you ask a question such as "I'm a new user and do not understand why my question is closed, or what to do about it!", add this as the answer and then faquify it. Would be a great resource to point noobs to for just this situation. – user1228 Jul 20 '11 at 12:44

Do we really want to cater to the newbies who can't be bothered to spend 30 seconds reading the guidelines on what questions are appropriate, then give up immediately when their poorly written question is closed?

It's hard enough for the higher-rep users to keep up with the influx of well-intentioned but unintelligible questions, or dupes (there were like 5 questions today in SQL about the exact same thing, custom sequences, and 5 or 6 the day before and the day before...), or off-topic questions, etc.

I'm not sure how much effort it would be for the dev team to add warnings for close votes, but I suspect that would just lead to comments like WTF guys why are you closing my question I just want you to plz send teh codez!

I've noticed a trend in the past few weeks of questions from newish users along the lines of:

Hey guys, here's my problem. I need a function that will do <insert operations here>. It needs to be really efficient and it can't use <insert exceptions here>.

<If this is a SQL question, put in a terrible table design you insist you can't change>.

I need it really fast!! Thanks!!!

We get I'm sure dozens of new users every day who ask thoughtful, relatively well written questions and contribute to the knowledge base here. How much energy do we need to expend to retain the vocal minority who put forth little effort to begin with?

  • None of those complaints are relevant to this guy's case. He didn't ask other people to write the fn for him, he had already written his own (buggy) code and posted it. I eyeballed it and responded. Your 'send teh codez' comment is plain irrelevant, and disrespectful to him. As to reading and following guidelines, yes he did. The guidelines specifically allow that 'If other users ask you for more information in the comments, edit your question.' But allowing the guy between 2am-4am to respond is not reasonable. 48hrs seems reasonable. – smci Jul 20 '11 at 0:10
  • 1
    The send teh codez wasn't addressing your specific example but others I have noticed recently. Your example he didn't even have a question, so I doubt he read the guidelines, which refer to "What kind of questions can I ask". He posted a statement with a code block. You are within your rights to ask him to clarify, but others are perfectly right to close it as "Not a real question" since questions normally have question marks and other indications of an interrogative. – JNK Jul 20 '11 at 0:17
  • 1
    @smci - Also, this is slightly tangential, but you personally don't have the responsibility of closing bad questions so it's easy to suggest things like this. It's different when you're the one that has to either bookmark it to come back in two days (two days = hundreds of questions per tag) or just let it slide and remain a bad question that's cluttering up the site. – JNK Jul 20 '11 at 0:20
  • 1
    He posted his (buggy) implementation code for his moveFrom(), moveTo(), processMove() fns, and wrote 'I can't seem to implement a movement function for the pieces. If anyone see any errors in my code, or can offer help, I would appreciate.' Objecting that that doesn't contain a question mark is IMO petty. Are we agreed that the one-liner 'Where is the error in moveFrom(), moveTo(), processMove() fns?' would have prevented this being closed? – smci Jul 20 '11 at 0:53
  • 1
    I would allow more grace period to those who post code (as in this case), people showing a good-faith attempt, and first-time users. This example is not the usual crud that demands instant closure. – smci Jul 20 '11 at 0:55
  • If someone were to write: 'Here is my code for fns a,b,c, please point out the bug.' but without a question mark, you honestly think that is grounds for closing? – smci Jul 20 '11 at 0:57
  • 1
    1 - Maybe. I wouldn't have voted to close on that particular question since it's not my area of expertise. You call it petty, which is fine, but you have to have a line somewhere. 2 - That's a fine opinion, others obviously disagree. Get some more rep and you can vote to reopen those questions. 3 - not necessarily, but it's a red flag for me sometimes. – JNK Jul 20 '11 at 1:06
  • I'll meet you halfway and say if he had specifically identified the offending fns, by name (which he didn't): "The implementation of a,b,c is buggy, please critique", I would consider that IMO a 'question', and allow it so stay open pending clarification. Either way, the FAQ on this is not user-friendly and needs additions. – smci Jul 20 '11 at 1:13

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