32

Compared to Stack Overflow, the number of questions which are down-voted, put on hold, or closed is astounding. I think it's a sign that the criteria (and possibly usefulness) of this site is poorly defined.

I originally came to this site thinking it's good place to engage in design discussions and ask general questions about platforms, libraries, and development. What I found is that many of the participants approach questions and answers with a vindictive attitude (as manifested by routinely down voting decent answers and reasonable questions.)

After avoiding this site for about year, I thought I should come back and ask a design question. Frankly, I'm very put off by the apparent negative attitude of the community here. I was even more chagrined to find one of my questions migrated here from Stack Overflow (where it was answered in a totally appropriate and useful way) and then closed.

  • Are there any specific questions (I assume on the current front page) that you believe shouldn't be down voted or closed? It can sometimes be challenging to get people to read what the scope of the site and general SE is. – user40980 Oct 25 '13 at 18:35
  • 2
    See also Why are almost half the questions on the front page marked “closed”? - a bit old, but still quite applicable. – user40980 Oct 25 '13 at 18:53
  • 6
    @MichaelT: honestly I think the criteria is applied inconsistently. I don't have the time to sort through all the questions to compare, but if I look at my own (still open) questions, I see several that should be closed by todays standards. For example this one. There should be a site where broad development questions can be asked. I thought it was here, but it's really not. – Sam Goldberg Oct 25 '13 at 19:05
  • 2
    comparison to SO isn't quite fair, as explained in more details here. In brief, their unmanageable (yet) CV queue behaves much differently (yet) from that at Programmers – gnat Oct 25 '13 at 19:30
  • 4
  • 15
    I find it frustrating to see legitimate conceptual/idea based questions being flagged/put on hold because they're "opinion-based". Programmers is a forum for concepts and ideas -- how can that /not/ breed opinion-based answers? There is also virtue in opinions being stated so long as they are backed up with reference material. – brandonscript Oct 28 '13 at 19:52
  • 6
    @r3mus: That was also always my impression of the value of this site versus SO. Meaning, a place where developer can have high level conversation about design, concepts, and also ask for open ended advice on those things. If a question gives no concept, or appears to have no effort behind. – Sam Goldberg Oct 28 '13 at 21:14
  • 6
    @SamGoldberg yeah... so what do we do about it? I feel as if the moderators have taken over and decided it's not going to be that way. Sort of wish you could vote mods up/down. – brandonscript Oct 28 '13 at 21:36
  • 2
    @r3mus mods close very few of the questions. Most questions are closed by 5 people with 3k or more rep. The thing to do about it is to also get 3k rep and participate in /review yourself. – user40980 Oct 29 '13 at 12:42
  • 2
    @SamGoldberg conversations can be had, and often are had, in chat. It is often difficult to carry on a conversation about design in comments or in questions and answers. They are set up to be a more "this is the question" to "this is the answer" format. Write or wrong, the open ended advice just doesn't work well to generate meaningful long term content for the site. A key point is that the SE network isn't a forum and specifically was designed as a rejection of the forum style for getting help on a particular problem. – user40980 Oct 29 '13 at 14:08
  • 10
    Absoluetly correct. The Karma of this location is awful, there are way too many janitors around "protecting" the site from growing. They are choking the place to death and are proud of it. – JensG Nov 1 '13 at 12:03
  • 3
    @JensG I don't think there are too many janitors really... just a handful. Here's a Data.SE query that shows the top close voters on Programmers for this year. It should be noted that the query doesn't include deleted posts, which get auto-deleted 9 days after they're closed if they meet a certain criteria – Rachel Nov 1 '13 at 13:14
  • 2
    @MichaelT - Quote "Most questions are closed by 5 people with 3k or more rep" - Does that not raise the chances of questions getting closed more arbitrarily (as opposed to being closed by a moderator)? – talonx Nov 1 '13 at 13:23
  • 1
    @talonx Stack Exchange isn't a democracy. /couldn't resist plugging my other site. – yannis Nov 1 '13 at 14:19
  • 1
    @talonx It requires enough people to say that something is closed (or remain open). If three people say that it should remain open it is removed from the queue and is less likely to get 5 close votes. However, all of this requires people to review. Our judgement isn't perfect (thats why it requires 5). It consistently disappoints me that more people don't do reviews and act as a check on others. It also takes 5 to reopen a question, another difficult vote to muster when people aren't doing reviews. – user40980 Nov 1 '13 at 14:24
12

It is often challenging to have people ask appropriate questions. The front page sometimes shows this. There are a fairly active group of people voting and P.SE doesn't have as much traffic as SO - thus there is more attention on each question and answer that gets posted - both up votes and down votes.

One of the important aspects of votes is that questions that have a negative score and 0 or negative score answers (and no accepted answer) gets automatically deleted by a process after a period of time. Thus, to try to keep these questions out of the accumulating pool of questions (that are having an improving quality), down votes are necessary.

Looking at the current 15 newest questions:

Design questions are certainly welcome - look at the ones that are up voted to see many examples of such questions (one such recent question - What are the safety benefits of a type system? - incidentally migrated from SO) . But, we're getting a fair number of questions from people who are asking questions that just aren't appropriate or were cross posted from StackOverflow, or trying to get around a Stack Overflow ban.

Yes, it is a constant battle to try to maintain the direction and scope of the site (what is currently an upward trajectory for quality and activity). Closing and down voting the questions that don't fit is part of this ongoing maintenance.

If you don't know it will fit or how it will be received, try asking us in chat.

  • 2
    I doubt anyone is asking questions here that don't work on SO. SO is much more permissive (and gets many more responses) for most specific implementation questions. Probably it's more confusion on the posters's parts. I think asking here for recommendations (libraries, books, IDEs) is very reasonable, e.g. Agile Books for Sales People. The way I found this site originally is someone migrated to here (a year ago), a question I asked about IDEs for C++. Today that question would be closed as too broad. I think the niche this site filled in StackExchange universe is rapidly closing. – Sam Goldberg Oct 25 '13 at 19:00
  • 4
    @SamGoldberg Broad recommendation questions are off topic on all Stack Exchange sites, not just Programmers. – yannis Oct 25 '13 at 19:04
  • 1
    @SamGoldberg We try to avoid migrating questions that will get closed on Stack Overflow for either SSCCE or minimal understanding. At one time it was not uncommon for us to regularly reject SO migrations to P.SE for quality. There is a growing number of close votes that are "this question should be on SO but wouldn't meet the minimal question quality standards" In such cases, they get bounced back to us - all around a poor experience for both us and the person who is asking the question. – user40980 Oct 25 '13 at 19:06
  • 1
    @YannisRizos: I may be misunderstanding the defintion of broad, but a quick search on SO, turns this up, The Definitive C++ Book Guide and List. And I know I've seen others like that "question". I think these judgements are highly subjective. – Sam Goldberg Oct 25 '13 at 19:10
  • 3
    @SamGoldberg "This question has historical significance, but is not a good example of an appropriate question. Read and learn from this post, but please do not use it as evidence that you can ask similar questions." :| – yannis Oct 25 '13 at 19:11
  • @YannisRizos: that was one example. There are plenty of others similar. – Sam Goldberg Oct 25 '13 at 19:15
  • 2
    @SamGoldberg if you find them, I am sure that people would appreciate you flagging them for closure with the appropriate reason. There are likely many that deserve attention of the more focused close reasons. – user40980 Oct 25 '13 at 19:17
  • 2
    @YannisRizos: how is How to make Sprint planning Fun less broad than asking for recommmendations for Agile Books for Sales people? Both are potentially asking for a list of recommendations. The question and responses for How to Make Sprint Planning Fun are quite verbose, and probably violates the "too broad" guideline, yet is not closed. What makes that less broad? – Sam Goldberg Oct 25 '13 at 19:20
  • 2
    @MichaelT: actually, I'm looking to close fewer questions, not more. Let there be a free flow of information. If you want to differentiate this site, do it by subject (development process, meta issues, architecture) not exclusivity (question not asked well enough, too broad, etc). I used to try contribute here regularly, but got frustrated with the constant down voting (for no reason) of answers that work and thought went into. It is very easy for people to skim over questions that don't interest them. I don't think the clutter issue is as big as it's made out to be. – Sam Goldberg Oct 25 '13 at 19:23
  • 1
    @SamGoldberg I don't know, the topic of the question is outside my expertise (and interests). If you have more examples (of questions incorrectly closed, or incorrectly left open) please add them to your question. I'm sure you have a point, and I'm sure the community will be more than eager to help (and re-open any question that shouldn't have been closed), if only you take the time to de-rantify your Meta question and point us to some actual examples. – yannis Oct 25 '13 at 19:24
  • 1
    @SamGoldberg: I agree with you that "Agile books for sales people?"(programmers.stackexchange.com/questions/215562/…) contains a good question, so I have rephrased it to fit better in the SE format. – Bart van Ingen Schenau Oct 26 '13 at 13:03
  • I'd be careful saying that a question is not a good fit because it's "a design review". That particular question is too broad, but if it's you can clearly show your design and are asking specific questions about specific problems/concerns with it, it's probably on-topic. Generic critiques or very abstract questions probably aren't good fits, though. – Thomas Owens Oct 28 '13 at 12:44
  • @ThomasOwens: Design questions encompass both conceptually large and abstract questions (which may tend toward theoretical or philosophical), and also down in the weeds specific design questions for an existing set of code. Are you saying that that only very narrow design questions can be asked here? If so, is there a different StackExchange site where those broader questions can be asked? – Sam Goldberg Oct 28 '13 at 15:02
  • 1
    @ThomasOwens good point. That specific question was a "here's my design, whats good and bad about it" - specifically I have drafted a proposal for my final year project and would appreciate any feedback. The feedback can be anything constructive either specific to this proposal, the area that I will be working and researching in or my ideas. which I believe falls under the "general critique" category. – user40980 Oct 28 '13 at 15:05
  • 3
    @SamGoldberg it used the be the case that broader design questions were standard here but the general direction of the community has been away from them. At one time I know some of the previously active members here were talking about setting up a more discussion-oriented community for those types of question, but I haven't heard much about that since, so I don't know whether the project faded. I hope not because programmers.SO as it stands is unfortunately too narrow to be useful in most cases, mostly as a consequence of the strictures of the format. – glenatron Oct 29 '13 at 10:33
6

Another thing to consider is that there are asked a lot more questions on Stack Overflow. And the crew in Stack Overflow is mostly organised in smaller groups around interest areas, you have your mini-communities, which govern certain "tags".

10 most recent questions on Stack Overflow have less than 100 views combined, the most recent question at the moment on programmers.stackexchange has more than that.

  • 1
    Exactly. So long story short, community moderation is working well here and our moderators are doing a great job. – dcaswell Oct 28 '13 at 14:56
  • I'm not really sure whether this is good "success criteria" or not. (Maybe it is.) I will say that in pretty much any question I research, large or small, I am far more likely to find a link on StackOverflow than anywhere else. I can't think of any topic I've ever done a search for, which has led me here. To me, that would indicate very little historical value to the questions. But if the purpose of this site is to solve very narrow practical problems, which may only be of interest to one developer at a given moment in time, then perhaps that is all this site aspires to. – Sam Goldberg Oct 28 '13 at 15:06
  • 2
    But if the purpose of this site is to solve very narrow practical problems, which may only be of interest to one developer at a given moment in time -- No, that's Stack Overflow's purpose. You have some good observations here (especially the one about number of views), but Google Page Rank is not necessarily definitive proof of community value. Stack Overflow gets tons of page rank for a number of reasons, yet I've still had Google search results point to Programmers questions. – Robert Harvey Oct 29 '13 at 16:46
  • @SamGoldberg - SO is massively larger. Of course search engines are more likely to find matching text there. – psr May 15 '14 at 22:18
5

The biggest problem of this site is, in my opinion, that the scope of this site is subjective questions, but the line between subjective and opinion-based is very thin, and the latter has very little tolerance here. From what I see, even the questions which fall under the guidelines described in Good Subjective, Bad Subjective still get closed a lot. It is almost as if the site fights itself.

Right now I'm looking at top questions on the website.

And there are tons of questions like these. I'd venture to say that nobody even knows what the site is about anymore. Sure, in theory we know which subjective questions are good and which are bad, but in reality questions just seem to be randomly closed because no one can draw an exact line between good and bad, and questions get closed "just in case". Even I can't always guess which question will get closed, though I've been reading this site for a long time, so I have no idea how newcomers are supposed to do that.

We really need to lighten up on what we consider opinion-based. Frankly, I would say the value of the site as it exists now is rather questionable. And we are definitely going downhill. Just rerun the query on upvotes and downvotes from that question, compare that to other sites, and you'll see what I mean.

  • 4
    I couldn't agree more. Having a site which is about sharing development ideas is valuable. This site seems to be positioning itself as a StackOverflow2.0, just with a more snobby outlook. In my opinion, good developer questions don't always have a narrow scope, and often opinions or recommendations are the best response. – Sam Goldberg May 13 '14 at 16:39
  • I noticed a downvote on my answer, but no comment. Of course, I may be wrong about something, but then I expect that somebody explains to me: OK, you are not right about this and this. I know I'm saying unpleasant things, but that's the way the site exists right now, and it needs to be changed. – Malcolm May 14 '14 at 15:26
  • @Malcolm Early on in P.SE's history, there were a lot of popular questions ("what is your favorite comic?", "what do chair do you sit in?", "What should I name my cat in a programming theme", "What is your most controversial opinion?"). Many of these got LOTS of up votes... and they were awful questions for a stack exchange site. The vote graph reflects that change from popular questions to ones that are more useful. – user40980 May 14 '14 at 15:47
  • 1
    @Malcolm: I sympathize -- this site is good a down-voting without comments on how to improve. This is part of what I found to be the judgmental and unconstructive dynamic of this site. – Sam Goldberg May 14 '14 at 17:44
  • 1
    @MichaelT: comparing the types of questions we are talking about to questions not at all programming related, is a wild exaggeration in the discussion we are having. No one is going to argue that Cat and Comic questions are out of scope. What is really the point of your comment? – Sam Goldberg May 14 '14 at 17:45
  • 1
    @MichaelT You are correct that there used to be useless popular questions, but as Sam Goldberg pointed out, we are discussing a completely different problem here. No one misses the questions about the best name for a programmer's snake. The days of those questions are long gone, but the upvotes have been steadily declining irregardless (and not just the upvotes). – Malcolm May 14 '14 at 21:21
  • 3
    Amen. SO is fine for boolean-style "How do you do X?" questions, but P.SE was always intended to be for questions dealing with software that cannot be specifically answered. Software isn't something that has rote textbook answers and solutions, there are opinions for a reason. – Wayne Molina May 16 '14 at 0:12
  • @WayneM the problem is, the layout of Stack Exchange doesn't work well with that type of question. It is focused heavily on providing a solution to the 'problem -> solution' type format. Going into DIY.SE and asking "How do I build a house" or Cooking.SE and asking "How do I bake a cake" or Biking.SE and asking "How do I fix a bike?" - these aren't good questions for the format. They lead to discussions, short + poor answers, and similar things that Stack Exchange was intended to avoid. – user40980 May 16 '14 at 14:30
  • 1
    @MichaelT You are talking about generality of the questions, and Wayne M meant subjectivity. If you make your questions more specific, they are perfectly answerable (How to bake a chicken?, How to bake french fries?). – Malcolm May 16 '14 at 21:48
  • 2
    @Wayne M: I agree with you. And I respectfully disagree with those who say that a StackExchange site can't accommodate this style of question. It can and it has. Just look at the early days of P.SE or really any Meta site for crissakes! – Jim G. May 21 '14 at 23:52
  • @user40980 yeah, sure. Nobody's saying that asking a bad question should result in upvotes. Question are: 'where's the line?' and 'why are good questions not upvoted well-enough?' The simplest few word answer is "haters community". Malcolm is right in saying unpleasant things. – Igor Soloydenko Oct 6 '17 at 23:17
-1

Site scope and question quality expectations seem to be set, stabilized and well understood by majority of regulars long time ago, as discussed eg here: Are you still confused about what Programmers is for?

Per my observations, most troublesome questions are either asked by newcomers or are old ones bumped by answers from newcomers which makes me feel that somehow, site "image" for those newcomers is blurred.

Based on several months of observations on questions that are exposed / promoted to SE network users via collider I firmly believe that to large extent, misguiding newcomers about Programmers is caused by the way how questions are selected for exposure at collider.

The way how I see collider damage to Programmers is described in more details in this answer: What can we do to help users understand our site better?

I think this is likely related to issues with answers quality in hot questions (discussed here in more details):

  • "Hot garbage waves" in the answers once or twice a week
  • poisonous effect these mis-answers have on questions, making interesting and well presented problems look the same as non-constructive popularity contests
  • regular ways to deal with this kind of issues just don't work...

An "outsider" reading low quality answers in hot questions, could naturally think,

hey here I can chit-chat about how git is fantastic and get my portion of cheap upvotes, c00l

Taking into account that these are highly visible posts, with thousands of views, it is pretty possible for them to be a steady source of misguided contributors.

...the way how users can be confused about what Programmers are about is probably best explained in a comment made by Ben Brocka:

Often hot questions have lots of bad or meh answers that are bad enough to drag down quality of the content and general discourse (encouraging similarly bad new answers)...


Side note. For the sake of completeness, I don't think that collider damage is contained within Programmers (and Workplace for that matter).

As discussed eg here, it looks like damage is spread over whole SE network, even though indirectly:

...really, say guys at SO are protected from direct impact. Meaning when they see meh, they can flag no-code and get crap deleted. But thing is, it's SO users who look at collider and who visit "sticky" questions and who pick the crappy attitude and they get back to SO and get posting meh answers there
and, well, there are hundreds and thousands of them. And meh attitude gets spread over multiple questions and SO mods only can delete minor part of crap that has been influenced through collider

they start thinking it's the norm. That's what low quality answers in hot questions teach readers. That's what "educated" readers spread further, to their answers to other questions. That lowers overall quality of answers in multiple other questions, that is the site-wide damage
This makes it look like good questions are those having many meh answers, the effect that is amplified by these questions being highly visible to collider audience - hundreds and thousands of SE users. Misguided users spread acquired attitude further into other questions and answers, posting stuff that follows what they saw at the "cool" ("hot") questions.

nobody is really protected, collider spreads it across all the SE network

  • 1
    If you ask me the site itself is misguided in what it should be; SO was for questions that had specific answers, Programmers was intended to have more opinion-based questions since most things in software development don't have straight answers... – Wayne Molina May 16 '14 at 0:14

You must log in to answer this question.

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