Every day we get opinionated questions asked here. People seem to think that this is the place to ask them. New users point to the old open questions that have fallen through the cracks of time as reason that their question should be open too. Old Stack Overflow users continue to suggest reposting discussions on NotProgrammingRelated Programmers.SE. Consider this exchange on a Stack Overflow question:

This is not an appropriate question here because it is opinion-based. You might get help at Programmers. — Ed Cottrell (deleted)
@EdCottrell programmers has the same rules about opinion-based questions as SO. Please take a look at the help centre there before sending people our way. cheers. – MetaFight
@MetaFight my apologies. I see lots of these kinds of questions over there and didn't realize they weren't acceptable. – Ed Cottrell

We have been told in the past that when our site is full of crappy questions, our site sucks. Yes, the answers may be great - but the questions aren't. As said, fixing that is a painful process. It also means that we need to change the perception of what Programmers.SE is about. This means closing questions that are too broad, or primary opinion, or career advice from days of old.

Popularity of old questions does not mean that it is something that should be enshrined in a museum for all time. Well, that's what a historical lock is for.

This question exists because it has historical significance, but it is not considered a good, on-topic question for this site, so please do not use it as evidence that you can ask similar questions here. This question and its answers are frozen and cannot be changed.

(I will also point out that the new users who cite historical locked questions apparently don't read the text)

Programmers.SE one of the highest ratio of locked questions to questions of major sites (note also that the great contentious cleanup of SF hasn't been completely resolved). Digging into data.SE (feel free to refine the query - comparing it to search here, I do see discrepancies, however it is consistently applied to other sites):

Site | Questions  | Locked
P.SE |     37,017 |  744 (2.01%)
SO   | 10,258,635 | 2702 (0.02%)
Math |    496,668 |  638 (0.13%)
SU   |    284,455 |  764 (0.26%)
SF   |    205,689 | 4265 (2.07%)
AU   |    202,900 | 1233 (0.60%)
TeX  |     97,119 |  180 (0.18%)
U&L  |     74,315 |  220 (0.29%)

We are the most preservationist of the sites that are out there. We even tried moving some of our popular questions to the blog (it didn't work out well).

If you feel that it is necessary to reopen those closed questions of old popularity so that someone can add another answer and continue the discussion, then go right ahead. The site was down that road once before and we were told that this was something that wasn't sustainable.

In the meantime, the community that is here has defined our scope.

Programmers.SE is not Quora, or Reddit or any other site. We are a site to give practical answers to questions of software design and architecture. It says so right in our tour:

Ask about:
✔︎ software requirements
✔︎ software architecture and design
✔︎ algorithms and data structures
✔︎ development methodologies and processes
✔︎ software engineering management
✔︎ quality assurance and testing
✔︎ software licensing

Don't ask about...
𐄂 general workplace issues, career advice, job hunting, salary, or compensation
𐄂 implementation issues and coding tools
𐄂 what you should learn next
𐄂 what projects to do or books to read
𐄂 where to find libraries, tools, resources, or other product/service recommendations
𐄂 personal lifestyle or non-programming activities
𐄂 questions that are primarily opinion-based
𐄂 questions with too many possible answers or that would require an extremely long answer

As a Stack Exchange employee, reopening questions that are in direct conflict with what the site scope is about sends a confusing message to our users - both the ones asking questions and the ones closing them.


Yes, there are many wonderful blog posts in our old, off topic, too broad, and opinion answers. They may be worthwhile to try to preserve - as blog posts. It would be helpful to have people who are good editors to help us migrate these posts to a different medium and make posting blog posts something that is easier to do.


Referring to the specific question of Why aren't young programmers interested in mainframes? when I started the close vote process on it wasn't its popularity - it was the tag that is littered with bad questions and itself being on the verge of a meta-tag. When looking through a tag, I tend to start at the oldest as those are the ones that have the most answers and the least fixable. This often means seeing some of the old popular ones before the newer less popular ones. However, the phrasing of:

Closing questions with great content years after the fact simply because they became popular seems like an inherently spiteful and self destructive way to make a point. You can certainly control your own scope, but when groups of users go out explicitly to hunt down the most-popular questions specifically because they are popular doesn't make the site any better.

is incorrect about my motives and needlessly antagonistic. It also suggests that you have a different vision of what types of questions should be asked on Programmers.SE. If this is the case, then please state it clearly. We've gone through this in the past and there is even a blog post that I am sure you are familiar with. In it, it states:

Thus, questions that are not answerable -- discussions, debates, opinions -- should be closed as subjective.

The question asked in the post Why aren't young programmers interested in mainframes? falls exactly into that category:

Why is this? What makes mainframes unattractive to young programmers?

It is not answerable. It is subjective and requires extended discussion. It would be something to ask on a discussion board. The explicit question presented in the post is a poll of personal experiences. It is entirely composed of opinions and personal anecdotes. It is not within the scope of the site today. With its 26 answers, it speaks exactly to the first part of the blog post:

Most forums and chat rooms have a scale problem. As in, they don't. The more people that join the discussion, the more noise each of those connections bring. So the forums get progressively noisier and noisier, and suddenly one day … you stop learning.

Because we believe so deeply in learning, we are willing to go to great lengths to suppress the discussion, debate, and opinions that -- while plenty entertaining -- cause most forums to inevitably break down.

Note: as of this writing, your blog entry has a broken link, the correct link is provided above

These old, popular, and off topic / too broad / opinion polls are ones that new users often find when they first hit the site and seek to contribute. This is unfortunate because they are often not adding anything that isn't there in the previous twenty some-odd questions, lack good writing skills, and are unfamiliar with the scope of the site. All of this leads to a rather negative experience for new users (doubly so if they ask another question like it).

Allowing these old questions to remain open is in direct conflict with providing new users to the site a good experience - answering a good question or asking a good question.

Thus, questions that are not answerable -- discussions, debates, opinions -- should be closed as subjective.

By and large this is true, but do you ever stop long enough to consider if there is actually some pretty good content behind a long-standing question — content that didn't become a never-ending list of answers or break down into a needless, irresolute debate devoid of merit or reasonable explanation?

Not every question is one fact, one answer. The rare question I re-open is not full of never-ending nonsense continuing to attract low-quality answers for years to come. These are largely resolved threads with good content right there at the top.

When that rare exception occurs, and these mega-popular posts attract some great answers with a lot of voting which clearly drives the best answers to the top, it should be seen as something that adds to the site, and guarded jealously. It certainly brings in a lot of views, which is why I received these alarms in the first place — highly-upvoted, highly-popular content being closed suddenly, sometimes years after the fact. Popularity doesn't always equate to high-quality, but if it is good content, your priorities should be to see that it is curated properly… not closed and discarded.

The idea that there are hordes of people just waiting flood the site for want of a single precedent was one of those things we worried about before we had anything to back it up. But if you haven't noticed, Stack Exchange has moved on from a lot of those old adages foretelling of widespread doom in ways that simply never materialized in actual use. The "too localized" problem; the dreaded "shopping question"; not constructive; slap community wiki on posts with undeserved rep; don't ask us to recommend solutions… and every question must have exactly one absolutely factual answer.

If you want to hold on to those old ideals, at least use the historical lock — not with a condescending flag "to make Robert Cartaino feel better about the fate of this question with 50K views" — but because you understand what it is not to throw out the baby with the bathwater.

A lot of sites manage to strike this balance, enforcing a vast multitude of objective, answerable questions while still curating those rare threads where broader participation has created something widely useful to the site. Look at Ask Different; look at Mathematics; these sites are about as well-run as they come. Yet more and more of them manage to curate those rare gems which motivate a community to put in the effort to keep this stuff relevant. And yet they manage to hold that line to keep it from getting out of control.

  • 13
    But why should it remain open? Curation of existing content for questions that are off topic, primarily opinion, or too broad is most easily done when the question is closed. Mathematics has their own host of problems and hint we don't want to have answers like theirs. And every time we've had a big list type question (that the Apple.SE community has done well), its turned into holy wars and excessive comment threads here. For some reason, those people who read those questions are not as disciplined in answering and it puts a greater amount of community curation on us. – user40980 Oct 5 '15 at 2:13
  • 2
    I would welcome a SE community manager looking at our new and active questions and see if there are things that we can improve upon, or if there are issues with how we moderate and engage the community that does moderation here. Please, just look at the last two hours or so of questions (right now) and see if those are clear, on topic questions, or unclear, off topic, or too broad questions that we are getting. – user40980 Oct 5 '15 at 2:16
  • 5
    Or better yet, for the next week, just spend a minute a day on each new post that site gets. It should take under an hour. And while I do realize you are likewise a busy person, it is important to get a feel of the entirety of the site's activity rather than just the "they closed this old post." Read our meta questions and develop a more complete view of the site, the questions and answers it gets, and the community moderation that is being done. Otherwise, all you see is the alarmist "oh no, they're going to delete everything because they closed this old question" type picture. – user40980 Oct 5 '15 at 2:26
  • 2
    50K views and concerns of SE mgmt are solid reasons to request historical lock. Our moderators decline flags that fail to provide compelling reasons for their involvement – gnat Oct 5 '15 at 5:08
  • 2
    I respect your opinions on this Robert and I do agree that the sky probably won't fall when we allow some of these more subjective questions to exist in CW state. I will say though that I have enough respect for the diamond that if the community decided to close something, I almost never decide to reopen it without performing a number of actions to try to improve upon the question or answers. The last time this question was edited was 2011, and there are 26 answers, most of them pretty lame. You reopened to give it another shot, I respect that and won't interfere, cont... – maple_shaft Oct 5 '15 at 10:41
  • 8
    cont... however if I felt so strongly against something the community did that I would outright reverse with no other changes or improvements, I would typically work in Meta to try and justify my actions after the fact. I always give the community the chance to convince me I might be wrong. If you feel that Programmers is not heading down the right track and isn't in line with the StackExchange strategy anymore then please provide guidance and leadership. Trying to set precedents on a handful of questions I think is not a very effective way to achieve a course correction. – maple_shaft Oct 5 '15 at 10:45
  • 5
    I've applied a historical lock on the question, because it is a low quality question. I think that this community has been doing a really good job of improving or closing the low quality questions to indicate to others that the question (and others like it) do not fit well here. I think that your premise that closed questions are "discarded" is totally wrong. We are very careful to not delete questions that add value. But closing and/or locking is perfectly fine and is an indicator as to what our standards are. That appears to be the choice of this community, and I'm supportive of that. (cont) – Thomas Owens Oct 5 '15 at 11:46
  • 17
    Personally, I'd appreciate it if people who weren't active community members didn't use their powers without first raising the issue. Ping one of the Programmers mods in chat, drop by The Whiteboard, or post something here on Meta about your concerns. Don't just take action on posts because you disagree with it - ask why things are happening and understand this community. I don't mind Oded and Anna Lear taking actions, because they understand and participate in this community. But unilateral actions from staff that doesn't (cont) – Thomas Owens Oct 5 '15 at 11:50
  • 11
    participate in the community just raise issues like this. It isn't the first time that unilateral actions have led to a post on Meta, and it isn't necessary. Please: don't act on posts here or community actions unless you are a participant. If you have questions, ping a mod in chat, visit the site chat, or post on Meta. – Thomas Owens Oct 5 '15 at 11:52
  • 5
    For what it's worth, a common reaction here for "why is this closed?!" is that users point to other, open questions from the 2011/2012 timeframe as reference to why their question should not be closed. Questions which are similarly broad/opiniony/poll-like/etc. We ultimately can't know how many users find programmers.SE through old and popular content and assume it is appropriate for similar content now. All I know is I have been part of multiple comment discussions with users about exactly that. – Elysian Fields Oct 5 '15 at 13:29
  • 4
    Robert - How is the community expected to "hold that line to keep it from getting out of control" when routine, easily fulfilled requests that would help with that community moderation are ignored? And yes, that particular request has been asked for multiple times and is no different than what has been granted for multiple other sites within SE. – GlenH7 Oct 5 '15 at 13:59
  • 3
    So a question about the decline of mainframes is completely... unanswerable. I'm not going to tout that as the pinnacle of what this site should become, but my hopes in reaching out to communities and reopening such questions on occasion is to say it's okay, relax. It's largely a symbolic gesture to let y'all know it is okay to let things get a bit subjective on occasion. I didn't mean to cause such a stir. Other sites seem to benefit from what we've learned over the years, so at least consider the possibility that some stifling ideals are things you are only inflicting on yourselves. – Robert Cartaino Oct 5 '15 at 15:12
  • 10
    When someone complains about the closure of an off-topic but highly popular question on Stack Overflow, the question that I always ask them is "Why does this question need reopening? What purpose is served? What can you meaningfully contribute to a question that already has 56 answers posted to it?" Closure doesn't mean deletion, and I think many community participants confuse the two. All closure does it prevent new answers from being posted, and signal to the community that we really don't want these kinds of questions here. – Robert Harvey Oct 5 '15 at 15:43
  • 8
    @RobertCartaino - I'm sensing a measure of talking past each other here, which is counter-productive for all. I don't get the feeling that you're hearing the pain points behind Programmers. Likewise, if SE is mollifying its previous stances, then I would like to hear more about that in a constructive manner. If your only visibility into the site is high-view questions being closed, then you're not seeing a lot of the garbage we contend with on a regular basis. Many of the active users would welcome a constructive conversation (perhaps in Software Engineering Chat?) examining both sides of these claims. – GlenH7 Oct 5 '15 at 15:52
  • 9
    @RobertCartaino One thing that I see you are claiming is that there is no indication of a clean up action going on. There is. In essence cleaning up old tags and question from before progs.SE changed the on-topic rules. Handling these questions is either re-tagging if they are still on-topic or first closing and later maybe deleting or locking depending on the popularity/historical significance. Keeping the bad questions open is not an option. – ratchet freak Oct 5 '15 at 16:10

Something came up in the comments on Robert's answer that I'd like to address...

But why should it remain open? Curation of existing content for questions that are off topic, primarily opinion, or too broad is most easily done when the question is closed. -- MichaelT

Closure doesn't mean deletion, and I think many community participants confuse the two. All closure does it prevent new answers from being posted, and signal to the community that we really don't want these kinds of questions here. -- Robert Harvey

Similar sentiments were expressed in other comments and in chat. They're not entirely wrong, but I think they largely ignore the practical realities of how things work - both in terms of the software that drives these sites, and in terms of how this site operates using that software.

Closing as limbo

This - the notion that closing is a state wherein questions can remain until salvaged - has merit; indeed, the system itself reflects this usage by marking questions as "on hold" for the first few days after they're closed, a signal that they can and should be edited and reopened if they can be.

That said, most are not. Have a look at the closing statistics for the past 90 days here - in particular, the stats for questions closed as "Too Broad":

Close reason Closed % of total Edited % of closed Reopened % of closed Edited and Reopened % of edited
------------ ------ ---------- ------ ----------- -------- ----------- ------------------- -----------
too broad    603    19.03 %    56     9.29 %      6       1.00 %       3                   5.36 %

Only 1% of questions closed thusly are reopened; less than 10% are even edited, and of those barely 5% are reopened. In real numbers, we're talking about 3 questions where this ideal is actually realized; if you were wagering on the outcome of a question, this would not be where you'd want to put your money.

Closing as a permanent state which prevents only the addition of answers

Again, this isn't entirely wrong... The system does indeed block the addition of new answers to closed questions, and some closed questions do remain visible long after being closed... But ignores the other big change in status that closing implies, namely that the question is now eligible for deletion: privileged users may vote to delete it, and in some cases the system itself will remove such questions.

Here are some statistics on closed question outcomes for the past 90 days:

closed deleted reopened 
------ ------- -------- 
3169   2039    24       

Over 99% of closed questions stay closed. Over 64% are deleted relatively quickly. Now here are the same stats for questions closed in a 90-day period ending 1 year ago:

closed deleted reopened 
------ ------- -------- 
2837   2125    41      

The reopened % jumps to... less than 2%, while the deletion % rockets to over 74%.

This isn't necessarily a bad thing, mind you: if most questions that are closed aren't worth saving, then you'd expect the reopen % to be low and hope the deletion % is high. But... You'd also be naive to believe that closing is not an immediate precursor to deletion; clearly it is just that for the majority of questions.

Why does this matter?

Remember the premise of this question was that certain questions shouldn't be reopened. Whether or not that's actually the case, it's bad news if the folks who are closing questions do so misinformed of what that actually means. If you're voting to close a question, you should be comfortable with that question being deleted - statistically, that is what will happen. If you'd be disappointed in that outcome, then you should probably choose a different plan of action.

  • 1
    If the material is a good blog post candidate, while at the same time confuses new (and apparently old users at other sites unfamiliar with P.SE) users about the scope and results in a significantly negative first time experience with posting to one of these "old open polls that is no longer on topic" or posting a similar question... how should these two interests be balanced? – user40980 Oct 5 '15 at 19:43
  • 1
    What percentage of the off topic questions that were closed were edited by the asker? Is it easy to get a list of those questions? I think when people have questions which are actually on topic here and get edited by the asker those nearly always get discussions about being reopened. But if an off topic question is never edited by the asker... is it actually bad if it stays closed? – Elysian Fields Oct 5 '15 at 19:43
  • 8
    I agree with everything you've written here, except that I come to a different conclusion: Locking questions should be much easier, SO-wide. Let's see if I can find an MSE question on the subject, ah yes, some <s>troll</s> <s>troublemaker</s> user wrote this one: We should have a question status similar to locking that non mods can set on posts This way, we can actually preserve questions like the one above in a form that makes it very clear they're not acceptable, and cannot be easily reopened OR deleted. – durron597 Oct 5 '15 at 19:45
  • 1
    I would also strongly recommend looking at, for example, the last 19 hours of questions (including the deleted ones) - especially those that have been closed and consider what percentage of those are salvageable without significant work and input by the OP to change the direction of the question? Just looking at close, reopen, and delete stats could get washed out by the questions that we are regularly getting that aren't able to be fixed. – user40980 Oct 5 '15 at 19:45
  • 1
    My first and favorite answer to these questions is always "Edit", @MichaelT. The older a question gets, the less it becomes the responsibility of the author and the more important it becomes to the community to maintain it in keeping with current standards - that is, assuming there's something worth maintaining. If that's simply impossible, then there are various locks that can be used as a sort of archive... But first, if at all possible, Edit. – Shog9 Oct 5 '15 at 19:46
  • 7
    The statistics you present are compelling but do not take into account the context of those questions. Without dredging up all of the residual NPR issues (and yes, they are still an issue), it's a bit naive to assume that all questions are alike on this site. And it's a bit unfair to lump a 50k+ view question in with the majority of the low-quality items that rightfully get deleted off. We close out the poll / too-broad questions because they are constantly cited as examples justifying new polls. – GlenH7 Oct 5 '15 at 19:46
  • You can get a recent list from SEDE, @enderland; past a few days old, it hardly matters anymore. – Shog9 Oct 5 '15 at 19:47
  • 4
    @Shog9 this community does edit questions -- when they can be made on topic. This meta discussion though shows that most questions that were closed just... don't have potential here since they are completely off topic or don't have sufficient information to actually answer. Sure, the community could arbitrarily edit them but what is needed in the cases where more information could make it reopenable is only available from the OP - most never return... – Elysian Fields Oct 5 '15 at 19:48
  • Unless someone's proposing to treat 50K+ view questions differently, does it really matter @GlenH7? Perhaps these numbers are an argument in favor of such treatment, but that's a separate discussion. – Shog9 Oct 5 '15 at 19:48
  • 1
    I'd suggest it does matter, yes. On a question like that, we'd need 10 community votes to delete. With as small as the site is, that's very unlikely to happen. So to draw a parallel between VTC on a low quality question that ought to be deleted and a high view / high vote, off-topic question is a bit disingenuous. A VTC on that high vote question is not equivalent to a VTD on the question. – GlenH7 Oct 5 '15 at 19:51
  • 3
    As to editing, I would encourage you to make sure that you check the badge activity. Copy Editor (awarded 9 times). Archeologist (awarded 6 times). Refiner (awarded 3 times). And the users by edits tab. I do edit. Some still can't be fixed. – user40980 Oct 5 '15 at 19:52
  • I'm not judging you, @enderland. Just making sure you're making informed decisions. – Shog9 Oct 5 '15 at 19:52
  • 1
    Delete votes do not expire, @GlenH7 - unless of course the question is reopened. So unless you're claiming that this site will never have 10 10K users willing to vote, 10 is not a number that really stands in the way. – Shog9 Oct 5 '15 at 19:53
  • 7
    The point I'm trying to make is that context does matter when it comes to the question. A VTC on a high view / high vote question is not the same as a VTC on a low quality question. Regarding 10 10k+ voters casting a VTD - if that's the case, then yes, the question ought to have been deleted. That's an example of the community itself determining what is appropriate for the community. – GlenH7 Oct 5 '15 at 19:58
  • 2
    I'm not sure where you got the idea that I was arguing for more locking, @gnat. If the consensus is to delete everything posted before 2012, then do it - but be honest about it, don't pretend you're carefully archiving stuff that's really intended for the furnace. Assume that every closed question will, sooner or later, be deleted and discuss the fate of each question with this in mind - you'll make the right decisions. Above all, be honest. – Shog9 Oct 5 '15 at 22:42

I have recommended that this question be locked as "primarily opinion-based". In retrospect, it should be locked as duplicate. Do I need to provide an example?

I see no reason why there should be a double standard on this site for moderators and users with high reputation (ie. for questions that fit suit personal opinions). If someone were to ask "Why are all my posts being closed" it would be marked immediately as duplicate.

Look at this post Programmers tag line is misleading. Can we rephrase it? and see that it is down-voted 6 times whereas this post is up-voted eight times and closed as not-constructive. That is democracy for you. Where do you assert to yourselves such certainty of your own opinions? I think there is much merit to his post and less to yours. It should be one way or the other.

The post is also too long to read in its entirety and not clearly related to its title line. The post “Why aren't young programmers interested in mainframes?” is not the main subject of this question, just an ad for the author's opninion.

I don't feel that I can offer an opinion here, as it might be down-voted, thus affecting my reputation. But see what it is like to have your own post booted and condemned.

I agree with @Robert Cartaino, but look at the comments under his question? ex. "But why!!!???", with boldface and attitude - it is all opinion based.

Edit: I apologize for any real or apparent hatred against SO. I am not serious that this post should be closed, far from it. I'm just not clear why this post is open and all the others like it are closed. The opinions that you do not like should ALSO have their place on the SO meta.

  • Meta has no reputation. I can assure you that if someone asked "Why are all my posts being closed" I would not act to close it as a duplicate. The title of this post was changed, so no, it doesn't match the body anymore, and no the specific question was not the main issue but rather the apparent misconception by a SE staff member about the scope of the site and what the community considers subjective and not. And this is meta - meta is different. – user40980 Oct 14 '15 at 21:29
  • Yes, but look at all the other Meta posts that have been closed. I listed one example from the "related posts", and to me it is only expressing a different opinion on the same subject. I don't see where they differ, except that this is a staff member and that the opinion that of the general SO staff. Obviously this post should remain open. – user1122069 Oct 14 '15 at 21:38
  • 1
    Many of those were closed in different times than the current one. There was much gnashing of teeth when the site changed from Not Programming Related.SE to Programmers.SE. In that period, the same arguments were brought up again and again and again and some people got tired of saying "this is how it is" and went to closing questions instead of drawing out the drama again in yet another question. As an aside, another good place to discuss things is Software Engineering Chat (see The Whiteboard) where we might be able to more easily clear up any other confusion about how meta works and its culture. – user40980 Oct 14 '15 at 21:42
  • Fair enough, that post is 3 years old, but I don't get the feeling that times have changed. The author himself is asking for more restrictions, even refers to these old posts as the problem, "Allowing these old questions to remain open is in direct conflict with providing new users to the site a good experience - answering a good question or asking a good question." – user1122069 Oct 14 '15 at 21:46
  • 1
    That post makes a good suggestion, so if you think that times have changed, I suggest that it be reopened. It should be programming.SO and not programmers because a programmer is a person and so a programmer is pre-qualified to write anything, judging from the tag. – user1122069 Oct 14 '15 at 21:50
  • 2
    It has been stated quite firmly from Stack Exchange management (Anna works for Stack Exchange) that the site name will not change. This position hasn't changed. As this is going further and further from the topic of the question posted, I would encourage you to take this to Software Engineering Chat instead. – user40980 Oct 14 '15 at 21:55
  • Post about my post instead. I have no interest in talking about that subject. Main point - this question does not differ in principle from other questions that have been closed, at least not according to any applicable rules. – user1122069 Oct 14 '15 at 22:21

You must log in to answer this question.