I was in the beta of SO. I almost never interact with it anymore.
Asking a question on SO is a last resort to me, and I get a horrid sinking feeling in my gut when I feel forced to do so. The people[1] who are still active on it seem to be people who thrive on pedantry and whose goal is to find any potential flaw in your question and feel smart for pointing it out.
You begin to realise no one is actually reading your question in good faith, so you start getting defensive: filling your questions with disclaimers about how your example code is just an example[2], how you know there are other ways you could do it but you're constrained toward this direction for various reasons[3], and so on and so forth, until you feel like you spend more time defensively shoring up your question from attacks than actually constructing the question in the first place[4]
I still read SO, but as someone who was around before it existed I don't really feel like the quality of answers is any higher than the random forum posts of yore, it's just that they're all under the same URL now, and the same user interface.
Which I suppose is something.
[1] Not all people™, but definitely the general feeling tends this direction
[2] classic situation: you simplify your code to Foo and Bar levels to show the problem cleanly, so people chastise you for having a complex data structure / worrying about performance / whatever for such simple code
[3] e.g., "How do I achieve X" gets turned into people saying "Why would you want to achieve X, that's stupid"
[4] This is not the same as researching the issue and trying as many things as you can think of, which is definitely helpful in any context of question asking
It's interesting to compare Stack Overflow to Quora, which was similarly great a few years ago and is now almost worthless, but in a different way. Stack Overflow suffers from militant moderators who close and delete reasonable submissions and answers due to draconian rules. Quora, meanwhile, has been taken over by spammers and idiots, and has lost any sense of trustworthiness. Just today I visited a discussion on Quora about WordPress plugins [1]. The top answer is an advertisement, the second gives an answer but offers no justification (and is also an advertisement), the third is probably an advertisement, and the fourth is again an answer without any justification. Repeat ad absurdum.
It's weird that both sites' communities have made it difficult for old users to take part, but in completely different ways.
Seconding the impression of SO. More than once I saw a question where it was kinda obvious that it was a beginner who was facing a very specific issue. Let's say it's a question that's not very difficult to solve for a seasoned programmer because even if you don't know it right off the batch, your experience can guide you pretty quickly to a solution. So, I spend the five to ten minutes to come up with a solution that works reasonably well -- in parts because I'd like to help out someone but in parts also because it's informative for me as well to learn something I didn't know before -- but just as I'm typing up my findings, the question gets closed. And it's not possible to answer closed questions (the rationale for which does not reveal itself to me immediately).
I mean, I get that SO wants to be a programming resource (as in "archive") where people with a problem should find a solution - not by asking but through googling. And so they want question/answer pairs that have a sort of general value, not an individual answer to just one person.
But then again: why? What's the big deal? Someone has a very specific question, and maybe nobody in the universe will ever have the same question again, but I'm willing to help that person out -- why shouldn't I be allowed to do it? Are they really worried about too much noise on the site? Please, come on.
In the situation I sketched above, I will still walk away having learned something new, but the person who posed the original question is left with a very negative user experience AND is none the wiser regarding their specific problem. At the same time, I was never allowed to help that person which I wanted to do not for the potential credit points but rather for altruistic reasons. Way to go.
I've been on Quora since 2011 when I wrote my first answer. I would easily spend hours back then reading incredible insights from all kinds of people. Sadly these days it's exactly as you say, a community filled with spammers and low-quality answers from popular personalities crowding out focused responses from people who are experience in that topic.
The product itself has also become 1000x worse to use and I think it's one of the biggest examples of a Silicon Valley company that remains successful despite the constantly bad product management.
Quora suffers from an extreme cult of personality where answers are primarily upvoted these days based on the popularity of the writer and rarely on the quality of the answer.
I generally agree with all your points, but I also remember the forum posts and I find that SO has at least one big advantage over those: it's really quick to pick out a good answer compared to a long thread. Just the fact that a) answers are clearly demarcated vs comments and b) the author can select a "correct" answer makes them way more skimmable than, say, a phpBB thread where every post looks the same.
I definitely see that side of those, though I think they both have pros and cons.
Especially in the JS space[1], there are lots of historic answers upvoted on SO that are now wrong and you have to scroll around (or find a slightly rephrasing of the same question that occurred more recently) to find a good answer. Unfortunately the longer a question is around both the more answers it has and the more likely the answers are to be wrong. On forums / mailing lists the question would just be reasked. On SO it is supposedly a wiki so new versions of the same question gets closed.
[1] Because it has evolved a lot, Rust has similar issues, as I'm sure lots of languages do
Not in my experience on the workplace so - in some cases the answer the OP needs is "NO do not do that", but the highest rated answer is long one that really does give the answer that is needed.
Sometimes you have to work out the underlying Q is and answer that even if the OP doesn't want to hear it.
I have over 6k rep, but I've come to hate it too - often within minutes of asking a question, some over-zealous numpty who plainly hasn't bothered to read the question, will flag it as a duplicate - despite it being no such thing.
I swear these people must sit F5-spamming their review queue, just itching to flag every question possible. They really make it an unpleasant place nowadays.
When I find questions with a poorly placed close vote, I try to immediately leave a comment why I disagree in the hopes that others will see it before too many votes are cast.
I often have good success finding a solution on SO although I have the feeling that most useful answers are several years old and not much new is happening. I never posted a question before but recently I have posted a few I had been struggling with for a while and I was a little surprised how unpleasant the responses were. They were all along the lines of "Why would you want to do such a thing?" or "If you don't know what you are doing, hire an expert" or stuff like that. After this experience I definitely won't ask anything there anymore.
> [4] This is not the same as researching the issue and trying as many things as you can think of, which is definitely helpful in any context of question asking
It would help showing that in your question. Often questions lack the necessary information to discern whether the asker has actually covered the basics
> [...] how you know there are other ways you could do it but you're constrained toward this direction for various reasons
Answerers are also programmers that like to apply best practices whenever possible. If you want to do something unusual which might smell a little without explanation your colleagues, code reviewers or similar would probably also ask "is this really necessary?". So you either have to write defensively from the start or deal with the patronizing-but-wellmeaning-comment-downvote-edit-reopen ordeal.
It's unfortunate, but there is an information asymmetry. The answerers can only see what you write, while you have all the background information specific to your case.
Agreed, but imho not so much a SO problem as how developers tend to enjoy treat eachother, or so I’ve witnessed. It often seems the most joy that can be had out of interacting with one another is leaving the interaction with a feeling of superiority.
>The people[1] who are still active on it seem to be people who thrive on pedantry and whose goal is to find any potential flaw in your question and feel smart for pointing it out. //
This is how the system is designed, IMO. It's the encouraged behaviour.
>I still read SO, but as someone who was around before it existed I don't really feel like the quality of answers is any higher than the random forum posts of yore, it's just that they're all under the same URL now, and the same user interface.
Hmm, we really remember things differently. I remember:
Correct on all accounts. SO is radically superior to what existed before. Faster, more comprehensive, tightly maintained, highly accurate.
I routinely cringe at reading some of the comments on SO, the general smugness on there. However there's no comparison to the past options. It's better in every way. Forums of old were overflowing with plenty of smug behavior as well, SO didn't spawn that.
Another problem on some of the more generalist advice ones workplace for example is you get people who tend to assume that the world is identical to the USA!
This does not help when offering basic advice on employment out side the USA you will see I highly rated posts go on about "right to work" - this has Zero if not negative value and could actively cause harm.
Judging by the links from Workplace.SE that show up in the Hot Network Questions list, the place seems to be full of handwringers. How are the posts you mentioned actively harmful?
Really? I'm a junior software engineer. I used a lot of SO before college/freshman college etc but then I realized last few years I never use SO. I really cannot think of problem that I couldn't solve reading the manual, man-page, info-page, project wiki, archlinux wiki etc... I'm not shit talking SO, I definitely read it rarely, but it doesn't seem like an irreplacable thing for me.
The people[1] who are still active on it seem to be people who thrive on pedantry and whose goal is to find any potential flaw in your question and feel smart for pointing it out.
How dare they feel smart huff puff
Because it really sucks to spend half an hour on an answer and then op goes "actually I'm doing this" and the spec shifts completely rendering your effort wasted.
A few times of that, and "clarify your question so it's very clear what you're asking" is a lot cheaper.
> Because it really sucks to spend half an hour on an answer and then op goes "actually I'm doing this" and the spec shifts completely rendering your effort wasted.
This is exactly how I feel about asking a question! Except "Actually I'm doing this" is "why don't you do that" or "why are you doing this" etc.
> Because it really sucks to spend half an hour on an answer
My feeling is that they aren't answering the question though, not the core one, there finding a thing that you're wrong about and talking about that instead.
It's like taking the above paragraph and refuting it by saying "*they're". Yes there is an (intentional) spelling mistake in there, but is that really the point of that it's trying to say?
Meh. Stack Overflow is suffering from the same "problem" as Wikipedia: the quality & quantity of existing content is now so high that it's becoming really hard to contribute new content. However, while this is indeed a real problem for "askers" trying to get heard in all the noise, and there's diminishing returns for the "answerer" who has less fun questions to tackle, it's also completely irrelevant to the 99% of us whose questions have already been asked & answered, and those answers can be pulled up in seconds by your favorite search engine.
Also, in case the blog author's name (Jon Skeet) didn't ring a bell, he's Stack Overflow's #1 contributor. If you've ever searched for anything related to C#, you've probably seen his answers, which have racked up over a million reputation points: https://stackoverflow.com/users/22656/jon-skeet
I was going to say the same exact thing, that it has the same problems as wikipedia. But then I was going to say "hard to contribute ... because control-freak insiders who've managed to take political control of what should have been an Internet-wide mass-community don't want you to."
Sounds like another case of geeks versus wonks. In Tucker’s “A Political Theory of Geeks and Wonks,” he delineates the latter with
Political wonks are fascinated by process. They love the game. They get as much satisfaction from observing as changing. They want to be players above all else. Ideals bore them. History is mere data. Intellectuals seem irrelevant. What matters to the wonk are the hard realities of the ongoing political struggle. They defer to title and rank. They thrive on meetings, small victories, administrative details, and gossip about these matters. Knowing who is who and what is what is the very pith of life.
Geeks, on the other hand:
They are no less fascinated by detail but are drawn to ideals. Observation alone bores them. They are drawn to the prospect of change. They don't want to be players as such; they question the very rules of the game and want to change them. They are happy to make a difference in the ideological infrastructure, whether big or small. They tend to work alone and totally disregard caste distinctions.
As to your frustration with Wikipedia’s political elite
The wonks are the ones who consolidate, stabilize, and entrench the status quo; the geeks are the ones who prepare revolutionary change. The wonks freeze it into place and make it work more efficiently; the geeks imagine and work toward a future that no one thought was possible. The wonks rule out drastic and extreme measures as imprudent and reckless; they geeks think these paths are the only ones worth pursing, and have confidence that the unknown future will somehow work itself out. The wonks try to bring the king around to their point of view; the geeks kill the king.
Tucker describes the conflict between the two camps.
The geeks and wonks can work together but there will always be a natural tensions between the two. The wonks think the geeks are hopeless, powerless, reckless outsiders whose heads are full of useless and unrealistic fantasies. The geeks think that the wonks are part of the system and, therefore, more than likely corrupted by it, and increasingly so.
Broadening the view, the struggle to control history is a battle between the wonks and the geeks …
According to his view, Julius Caesar represented the wonks and Brutus the geeks in Ancient Rome. In the America’s revolutionary generation, they were Alexander Hamilton and Thomas Jefferson.
Not sure what is political. I consider Wikipedia and SO to be perfect democracies of information. People upvote answers they consider correct. Those who contribute good answers frequently have demonstrated their skills and knowledge. If another talented contributor provides intelligent answers, we would happily upvote his or her answers as well.
This is really insightful. In some ways every community goes through the "wild west" -> settlers -> cities evolution. Happened with the internet at large, happens with most technologies that achieve mass adoption, and it appears to have happened with Wikipedia and Stackoverflow. (I read an interesting article about technology pioneers, settlers and other archetypes, but I couldn't find it.)
Once most of the pioneering work is done, and the low hanging articles are written, a community needs to decide how to proceed. Who are you going to maximize value for? In the early days it makes sense to focus on what is the rarest resource (in SO's case, probably the answerers) but as a community matures, the focus can shift. Now it is frankly the anonymous viewer that it makes sense to optimize for, both from a monetary standpoint (jobs display being a primary monetization strategy) and a social good standpoint (far more anonymous viewers than either askers or answerers).
> this is indeed a real problem for "askers" trying to get heard in all the noise, and there's diminishing returns for the "answerer" who has less fun questions to tackle
To address this issue, I put together a "deep-cuts" algorithm that uses the API to tries to find overlooked higher quality questions that have yet to be answered.
It's not perfect. But if you're looking to engage with Stack Overflow, I've found it much more useful than the rat-race to answer (or close) new questions.
>suffering from the same "problem" as Wikipedia: the quality & quantity of existing content is now so high that it's becoming really hard to contribute new content. However, while this is indeed a real problem for "askers" trying to get heard in all the noise, and there's diminishing returns for the "answerer" who has less fun questions to tackle,
I agree with your observation but Jon Skeet's essay seemed to be focusing on something else... the culture of hostility (or perceived hostility). The key words I noticed he used (multiple times) were "hostility", "jerks", "feelings".
On the other hand, the keywords I would associate with your observation would be... "boredom", "novelty of a new website", "low hanging fruit to quickly build easy karma", etc.
To discuss Jon's specific angle, let's use an example of a [tag] that's fairly new technology such as "[vue.js]". Since Vue is 2014, it doesn't have the "completeness of answers" of a very old tag like C# from 2008.
Maybe his question deserved all those downvotes. I don't know enough about Vue to guess if the downvotes were all caused by a pasted screenshot instead of pure text.
To Jon's "culture" essay, is the minus-4 the optimal way to communicate it's a "bad" question? Is it a bad question because of its format (the screenshot) or was it bad because "props" is a trivial LMGTFY (Let Me Google That For You Question)?
To generalize, if there's "[new technology]" that opens a vast untaken area for new contributions of answers, is the culture created by Stackoverflow's mechanisms (the downvotes, the karma system, the moderators' powers, the "closings" vs "moving", etc) ... the best incentive system to minimize hostility?
Are Jon's well-intentioned "covenants" realistic? Or are they naive to human tendencies when any website is open to mass participation?
The downvote button is marked "shows no research effort".
Where's the "I looked at this manual and that tutorial and this debugger and tried these things"? Nowhere, which makes it feel like "I put no effort in stackoverflow do it for me".
The downvote button is an awful way to do what it purports to do. In practise it's percieved as more of an "oh fuck off" button.
It's a single mouse click by a single person (or maybe a bot), but it carries so much weight.
That's one thing Jon didn't touch on, the downvote. It's the easiest thing in the world for the giver, but to the receiver it feels like a kick in the gut. It definitely contributes to the feeling of hostility.
Personally I try to use my downvotes only as a last resort, for factually wrong answers or questions where clarification attempts went nowhere.
I do hope that Stack Overflow will become usable again. In my opinion, the problem is mostly the answerers.
I went all the way from a simple coder to management, and now I'm doing both, running my second startup. I talk to people of different walks of life, and I always do my homework.
But on Stack Overflow, it's like I'm back to the early years of my career, a stupid newbie whose questions get randomly downvoted. I delete it, tweak the wording a little bit, and now it's suddenly accepted. I say something like, "a solution or a 3rd party library", an idiot downvotes me saying that "requests to recommend 3rd party libraries are offtopic". I change it to "a solution (3rd party library is OK)", and now it's suddenly fine.
It's like the mafia of high-karma answerers decided to adopt Les Grossman as their role model.
As a side effect, the good answerers are gone, and the really tricky questions are left unanswered, so I often answer them myself. Although, the way Stack Overflow is, my desire to contribute to the community is diminishing.
It's nice that in a discussion that is in part about being nice you call someone that downvoted you an idiot.
I usually don't downvote answer that ask for 3rd party libraries, just vote to close. Your example is a bit borderline and it's something I wouldn't close, but I can see how someone might think it warrants to be closed. Consider that the rule about external tools is meant to avoid a deluge of answers about libraries. They may be useful, but none of them is objectively right and can cause religious wars with people downvoting answers about libraries they don't like and massively upvoting their favorite library.
If we can avoid this by being strict, I think it's a good thing. But hey, I'm just one of hundreds of thousands users, so it's not like my opinion is somehow more valuable than yours.
> They may be useful, but none of them is objectively right and can cause religious wars with people downvoting answers about libraries they don't like and massively upvoting their favorite library.
Like many, I find the answers to these "religious" (in fact, they are not, because done properly the answers are supported by arguments) questions to be the most useful content.
In many ways, I can answer the most objective questions myself from the documentation or source code. But what I cannot do this with is distillation of expert knowledge that requires time, and the more distilled it is, the more subjective it is. Subjective doesn't mean unsupportable by facts (experience).
> It's nice that in a discussion that is in part about being nice you call someone that downvoted you an idiot.
I call them an idiot because they were nitpicking on a question that was asking for help well within the already Procrustean bed of regulations. I stand by my assessment: it's hardly a reason to refuse help citing bureaucratic regulations if you're a public servant, but that's somewhat expected. If, however, it's a fellow forum contributor, this is both unexpected and serves no purpose.
Not sure who's right, of course, but as you see, the number of people complaining is greater by day. Note that many of them are familiar with the Internet since 1990s and are netizens of good standing; still, they manage to somehow "violate" these rules.
As soon as there is a real competition to the "soup Nazi's shop", many people will jump ship without hesitation.
It's also interesting that other Stack Exchange communities are not that strict.
> I usually don't downvote answer that ask for 3rd party libraries, just vote to close
why ? why do you do this ? a lot of time there is an exact piece of code that does what OP asks for, what does it change if the piece of code is in a SO answer or in a github project ?
IMO it doesn't scale indefinitely. There's a site culture to be maintained, and as the site gets larger the core moderator culture becomes a smaller and more isolated part of the site.
You basically asked how to architecture a security aspect of your software. And you asked for other people's experiences to boot, instead of a canonical solution. Additionally the preface about the GDPR is a distracting, non-technical detail, instead you should have described which security properties you want, your threat model is etc. with the GDPR as a footnote in case someone wants to know your underlying use-case, not the other way around.
That's very much considered offtopic on SO. Maybe other sites on the SE network such as Information Security or Software Engineering might have taken it. But I don't use either of those, so I don't know if that would actually fit into their policies.
I think your question being closed is legitimate, it _is_ too broad. SO is at its core a Q&A site, and it's trying to be a _reliable_ and _responsible_ Q&A site.
Broad questions like that don't have a single good answer that fits every scenario, they have lots of different answers, each of which is the best fit for a different scenario.
The problem with a site like SO allowing questions like that is that you _would_ get an answer, that answer would be accepted, and that question would appear really high in search results. When the next person comes along needing an answer to the same question, but in a different scenario... they're gonna take the one that was given there, even though it isn't the best fit.
Scale this up to SO's scale across a few years and all of a sudden SO isn't a trusted resource any more; it's more like experts exchange used to be, where it was full of answers which _can_ be valid, but are basically useless to everyone, and taking that advice can actually be actively damaging.
I get why your question is on hold. But I find it unfair. Some of the best questions and answers have been highly opinionated. And I have often been led to on hold questions by google.
An opinionated SO site might be nice. In the old days we had newsgroups for this. But that audience has disappeared when everything moved to the web.
The commenters on that post are very narrow-minded and should not be downvoting and making pointless comments. That said, I think for SO you should word the question so that it can have a specifically correct answer. The farther you stray from this requirement, the more likely the question will get closed as "too broad". If you have more broad design questions try SoftwareEngineering Exchange.
It's like the mafia of high-karma answerers decided to adopt Les Grossman as their role model.
It’s like the high karma answerers are exactly the subset of the site that you want seeing and answering your question because the site supports the idea that they are good at doing that, but they’re always reflected as some kind of tyrants appointed by an evil overlord.
Broken site design somewhere that leads to that effect.
> It’s like the high karma answerers are exactly the subset of the site that you want seeing and answering your question because the site supports the idea that they are good at doing that, but they’re always reflected as some kind of tyrants appointed by an evil overlord.
Are you 100% sure?
Because from my experience, the most useful subset is between 100 and 2,000, those with experience but not arrogance. The top karma hoarders more often than not provide citations from official documentation and sabotage questions.
I wish they were appointed by an evil overlord because the overlord is better than a swarm of annoying mosquitoes.
In my opinion, the problem is mostly the answerers
And .. I downvoted you. I hope it will sting. People giving up free time to add value to your thing and you blame them for the site problems? Then the site design is really failing, if the problem is all the people being engaged with it.
Whereas on stackoverflow when I downvote it’s for “this question shows no research effort” which is the downvote button tooltip, but a single downvote is felt as a personal attack on character.
Stackoverflow is for programmers - people whose character traits are tainted to be nitpicky about small details - yet the bad experience angry posts call answers traffic cops and Martinets - exactly the kind of people who could help get your code past the compiler or browser or api or whatever.
“Neutral” responses feel like negative ones (see: “what have you tried?”).
As an answerer more than an asker, if I help someone understand then I feel good, if it seems someone asks from a sense of “help me” I’m endeared to the question and if they ask from a sense of “solve my business problem” I feel more abused (personally or site-being-abused) and resentful.
No amount of Jon Skeet blog posting about the classic usenet popup “this reply will be sent to thousands of people who will have to spend time and money on it, are you sure?” seems to work on the modern web.
If it were me reworking it, I’d get rid of the low quality question idea, let answered be the judge of that and google sort the wheat from chaff.
Then get rid of downvotes and close votes and “neutral” comments.
Leave only room for feedback buttons that in some way communicate “I tried to answer your question but couldn’t, Your question might be more answerable if ..”
Forget being a repository of high quality questions and answers in the same way software has forgotten about big design up front and perfect quality goals and turned into “put an MVP on GitHub and let the issue tracker be the driving force”.
The goal of “stop duplicates, everything that can be questioned and answered has been questioned and answered” forgets that there are 150,000 new humans every day, and thousands more existing people learning to code - a goal of turning everyone away with “that’s already finished nothing for you to do here” is .. daft.
Questions and answers are, if nothing else, practise. A good way to learn is to try explaining it to someone else - SO churn has that room.
But not on stack overflow, where following the site plan for downvotes and closevotes makes everyone feel shitty.
I even doubt it is “high quality questions” that are supported but a combination of “this question feels like the asker is like me” combined with “and they’re asking about a thing I like and feel positively about rather than an edge case of the tool I feel bad about and wish the asker would politely overlook it like the rest of “us” all do”.
You've repeatedly crossed into incivility in this thread, degrading the discussion noticeably in several places.
Please (re-)read the site guidelines and follow them if you want to keep commenting here. That means, among other things, posting with scrupulous respect for others and eschewing jabs and swipes.
I can't see whether I was downvoted but I see that the net result is very positive. But I'll take your word on that.
Perhaps your view is not shared as widely?
> People giving up free time to add value to your thing and you blame them for the site problems?
Do read my original post. My problem is people who don't actually contribute but nitpick on how the questions are worded. How does kicking me out or ignoring my question at best help me resolve my issue?
I would prefer them to keep the free time they dedicated to harassing me to themselves.
I do find your suggestions on design sensible though.
Because the difference is night and day. No pedantry (no pedantry on a math site! Impossible!), no assumption that askers are throwing their homework on them. In fact, my experience with mathematicians have had me question whether I choose the right career or if I should have become a mathematician. They all seem damn chill in comparison to most of us developers.
Anyway, the idea behind that site appears to be to help people learn math. Not to build the world's most polished question repository. I very much prefer the former approach.
I wonder if this has to do with the people who are entering a field. I suspect that people who post on a math forum really want to do math, have an interest in it and don't want to just get a job that pays the bills. In software you have more and more people who got into programming because it's a good job but don't really care for it.
This may diminish the quality of forums because a lot of people just want somebody to solve their problem so they can move on without doing any work. I see that also in a lot of new hires who don't seem to care about actually learning programming but just want to do the bare minimum.
I think there's also a level of cultural gatekeeping in play as well. Programming (and many of its sub-focuses) are considered a fashionable career path, and there are many people who want the reputation that being a programmer/hacker/etc conveys more than they actually want to do the work to learn.
The Security Stack Exchange is full of a similar sort of people. Almost all of the new questions at any given time are either tech support "I think I have a virus!!" or low-effort handholding questions "how do I use kali linux to hack server". I think folks who spend a lot of time on these sorts of websites have adjusted to be more jaded towards questions that look similar. I've learned to ignore it and be thankful that (in general), questions that I ask on stack exchange tend to receive helpful answers.
I think the difference is that math is cut and dry.
Stackoverflow has questions about some ridiculous configuration in languages and libraries where the asker couldn't even bother to pinpoint the issue but rather dumped all possible variables into the question.
Seems simple: math doesn't have to deal with that.
helping people learn:
achievable with very high reward
Creating a catlogue of perfect questions and answers: impossible and of limited resource as every answer only caters for a specific set of back ground knowledge
And what is funny is that with option 1 done right who needs option 2? why scrabble around in the dark when i could be connected with someone who actually wants to help me.
i think that software dev is saturated with people who like to learn lots of things by themselves (excellent work, this is a form of hero), however this is not a good representation of most people. if we want IT to be more inclusive we need systems that can handle people who dont want to read several textbooks before getting some code working.
I find the experience the same. I had to fight and defend my question multiple times and added all sorts of irrelevant details, and in the end, the question was still closed. I eventually found out the answer on my own.
I second the vast community differences across the StackExchange network. I like the mechanics one, the photography one, and the gardening one. They've been super helpful chill folks like you say. I've spent a little bit of time in the Bicycling one, and there may be some snobbery there but that's based on limited experience but the first impressions were a little tainted. The music one was a really warm receiving group.
I think within SO there's sub-communities too just by tech stack. Some of the deep esoteric stuff might attract some snobbery.
> A while ago I started writing a similar post, but it got longer and longer without coming to any conclusion. I’m writing this one with a timebox of one hour, and then I’ll post whatever I’ve got. (I may then reformat it later.)
While this is meta it pretty much sums up the situation regarding my feelings regarding Stack Overflow.
I was an early user of Stack Overflow, not quite beta but almost. At one point I even received a snail mail letter from Jeff Atwood, thanking me for my contributions. I read most of the early discussions about policy and, being a veteran of various internet forums at that time, it seemed to me the most sensible, most reasonable approach you could take to running a question-answer site. I went to CodeKen in London in 2011 where I met Jon and a lot of other excellent folks that made Stack Overflow happen as a community back then.
Today, I rarely consult Stack Overflow, and when I sometimes find a useful answer via DuckDuckGo it's almost certainly closed. I even joked that closing is a sign of quality at Stack Overflow. My last contribution is probably years ago.
The strange thing is that there is no point I "left", nothing I could point at and say, that's when Stack Overflow took the wrong turn. I think we just drifted apart. And for the conclusion I guess for me it's mostly: "The way to hell is paved with good intentions".
Is this a sign of strength or weakness though? Like Wikipedia, any given topic on SO will approach “completeness”. I use stackoverflow at least as much now as in its prime - but these days just like a Wikipedia for code. I don’t consider it a flaw.
I’m sure there are topics I don’t use (the latest ML framework or the js framework du jour) where the community is still growing the database. That I use it as a mostly completed reference work just means my field/tech is mature in terms of language/frameworks
It's really hard to consider an answer to a moving target to be "complete". If the question was answered for Rails 3 but you're using Rails 5 and have the same exact question, chances are that answer isn't going to work for you. Now you have to chance that if you post the question again, is someone even paying attention? Is someone going to close your question as a duplicate? Is someone going to give you the wrong answer?
Unless you're working with a dead technology, by the time the information is "complete", it's outdated. And I would consider much of the information on SO to be just that: outdated.
But people still ask those questions and the questions apparently get indexed. In my experience the result is that when coming from a search engine, even for non-fit topics, it is quite common that closed Stackoverflow posts occurr several times on the first result page.
It could be that I search for strange things, but at least for me the combination of closed but still indexed posts are not helpful, especially when the question didn't get any useful answers at all before being closed.
I don't mind closing topics, but I would prefer that closed questions with no even remotely useful answers wasn't indexed at all. Now, that might not be entirely under your control, but the issue remains, or so it seems to me.
Thank you for the letter and the stickers, I have one of them on my bicycle to this day.
> We hadn’t quite figured out that opinions and discussion questions were not a good fit in the early days.
This is something I had to learn as well, I guess one of my most upvoted questions falls well in that territory [1].
I just think that Stack Overflow was best when you were its primary moderator and it's arch enemy was Experts Exchange [2]. That battle is won, but the war goes on. Nowadays when the first search result is Quora I have very much the same feeling I had regarding ExpertsExchange back in the day and I can't help thinking that there is lost opportunity for Stack Overflow and Stack Exchange.
The sibling comment is good enough and gets the point across. But I want to elaborate because I've ended up in basically the same place as the GP.
I know you've written about SO not being a good place for opinion-based questions before, and perhaps in theory you are correct, but I (at least) am not satisfied with the implementation, and tend to link that at some level to a flaw in the theory.
What I mean: some of SO's most [in]famous questions are opinion questions where the askers are requesting advice from "expert" answerers to help them navigate a new topic. They're not just infamous because they may be controversial and "hard to moderate". They're famous primarily because the way in which the question was asked drove many users (hundreds even thousands of upvotes with millions of views) to the site and helped a lot of people through their journey working on a new technology or in a new problem domain.
SO said "nahwp, we don't want questions without clear cut answers". Not necessarily SO's fault, but nonetheless their problem: the community took that as, "err on the side of annoying pedantic asshole and close, downvote, and otherwise ostracize any user asking any question that might not be a good fit for the site under the new regime".
In my experience, it feels like people are trying to find a way justify being annoying instead of find a way to provide meaningful advice. I've asked questions that, if you took 30 seconds to think about instead of a quick search for a marginally related title (which I already did in the 4 hours prior to asking my question) and immediate vote to close as duplicate, or an almost bot-like flag as "too broad" based on word choice heuristics, are not bad questions, and often after hours of flagging and arguing with the original users who are stubborn because they don't want to be wrong (especially publicly) I can make my case. But by that point it's too late and too many times I've had to do that to fend off users trying to do good for the site over helping me with a problem. And I'm not even asking opinion questions...
By making one small decision, you created two problems.
1. You actively choose to diminish the value to practical real users of the site by setting a long term vision of becoming some bastion of objective knowledge. We have those; they're encyclopedias.
2. You unintentionally validated the pedantic/authoritarian desire for and abuse of power that has deterred many of your original users.
Honestly most people can handle snarky comments and opinionated edged answers. Everyone (users of SO) has a tiny (and sometimes large) "programmer ego" inside of them. On SO you could responsibly exercise it. Not anymore. What I can't handle is nonsensical authoritarian and meaningless abuse of questions because they don't immediately and explicitly adhere to some SO agenda. The former can be mildly annoying at times but tolerable and hell even fun on occasion, the latter is frustrating and infuriating and something I cannot tolerate.
The underlying problem to me is the fundamental flaws in crowd-sourcing and blacklisting as models to produce and manage content.
Websites like StackOverflow, Yelp, Wikipedia start off by being fantastically useful by using crowd-sourcing to quickly generate a large volume of content that covers topics too wide for a small group of humans by themselves.
But then crowd-sourcing becomes an albatross because an unpaid army of moderators have almost no oversight. They inconsistently apply (or don’t apply) standards, and blacklisting vitriol is a Sisyphean task.
Valid questions get blitzed and jerks proliferate because it’s seemingly impossible to police them all in real time. I saw a stat that said only 7% of SO users ever ask a second question - that means 93% probably had a bad experience.
While the problems are structural IMO they can do a few things. Downvoting should be temporarily disabled until Joel et al figure out a better model to surface important/good questions and answers; I personally now visit the forums of whatever framework or language I’m working in to seek advice because the downvoting on SO seems so inconsistently applied. Insults I can ignore but the DV makes it hard to have your issue seen.
Perhaps a multi-tier system where questions aren’t posted “at the top” for answering directly and moderated down, but instead a first step where the question is treated and discussed without possibility to answer. The community can suggest edits, and upvote it, but not close it. If it’s a duplicate the asker can get some time to explain why it isn’t, or retract it if it is.
If it gets a thumbs up, it qualifies for the regular Q/A where people can answer - and expect a certain quality (and don’t risk spending 20 min writing an answer to a perfectly good question that was closed while the answer was written).
No open platform, as far as I can tell, has solved the problems of vitriol, but the down-voting is absurd (as attested by some of the experiences described in this comment thread) and can absolutely by addressed by changes like the one you propose or some other system.
a first step where the question is treated and discussed without possibility to answer.
How would you enforce this - would you try to enforce it? If there's any kind of comment box, you'll get people posting answers in there, if it's amenable to a short answer.
> I saw a stat that said only 7% of SO users ever ask a second question - that means 93% probably had a bad experience.
That doesn't follow. The one-question-askers may be using throwaway accounts, or be confused about what SO is for; they might be in completely the wrong place, and have a low likelihood of ever needing to ask another SO question. They might also just not run into questions that are only answerable by asking the open internet that often.
> The one-question-askers may be using throwaway accounts
Why would you use a throw-away account? There something personal about a compiler error?
> they might be in completely the wrong place
I assumed mods moved questions to the right place or directed users to correct board.
> have a low likelihood of ever needing to ask another SO question
I can understand not asking 50 questions. But never asking a second question? Your entire programming career you only have one technical problem you need help with?
> They might also just not run into questions that are only answerable by asking the open internet that often
Yes, a lot of the technical problems I encounter at work are too specific to our implementation for SO.
No retention rate will be 100%, whatever the product. And there are plenty of legit reasons why you may not need to ask a lot of questions.
But 7%? They can't all be "noobs" asking dumb questions since plenty of HN devs here complain of similar problems: arbitrary downvoting, questions being marked as duplicate when they're not, unaccountable mods, etc.
To me the fact that 93% of SO askers don't bother with the site after their first experience signals a deeper problem with how SO is structured. As one of other comments in this thread argues, their setup was designed for its initial phase of growth but it's not ideal for the "mature" phase its in now.
1) it want to be a repository of canonical knowledge, a giant FAQ and wiki of everything. The problem is, that it does not fit the way the site actually works: people will write imperfect questions, precisely because they don't know the answer, and ask the opinions of their peers about things. But they are not domain experts writing documentation so they often get down-voted or their questions closed.
2) There is a karma system in which more karma gets you more power. Inevitably such system attracts people in search of fame or power, and this kind of people get off by telling others that they are smarter than them, or that they know this little site rule better and consequently censor their question. As regular users become fed up with the state of things and leave the site, the position of high-karma users become more entrenched. Then, those users obtain total control over the meta board to veto all fix proposals, make up even more draconian rules, and entrench their position even more.
Problem 1) causes inadequate meta rules, and 2) zealous enforcement of these rules. Some combination of 1) and 2) prevent the site owners from changing the rules.
If one operates outside of the mainstream, then SO still works like in the "good old days". Mainstream and the main volume of action is about the relatively basic questions about the popular programming languages and platforms.
But if one asks or answers questions about, for example, more advanced features of git, more advanced algorithms, numerical mathematics, GIS, anything that is not interesting to "the masses", then you still get genuine helpful answers, but you might have to wait hours or days, and there will not be a large number of answers.
Readit News
Asking a question on SO is a last resort to me, and I get a horrid sinking feeling in my gut when I feel forced to do so. The people[1] who are still active on it seem to be people who thrive on pedantry and whose goal is to find any potential flaw in your question and feel smart for pointing it out.
You begin to realise no one is actually reading your question in good faith, so you start getting defensive: filling your questions with disclaimers about how your example code is just an example[2], how you know there are other ways you could do it but you're constrained toward this direction for various reasons[3], and so on and so forth, until you feel like you spend more time defensively shoring up your question from attacks than actually constructing the question in the first place[4]
I still read SO, but as someone who was around before it existed I don't really feel like the quality of answers is any higher than the random forum posts of yore, it's just that they're all under the same URL now, and the same user interface.
Which I suppose is something.
[1] Not all people™, but definitely the general feeling tends this direction
[2] classic situation: you simplify your code to Foo and Bar levels to show the problem cleanly, so people chastise you for having a complex data structure / worrying about performance / whatever for such simple code
[3] e.g., "How do I achieve X" gets turned into people saying "Why would you want to achieve X, that's stupid"
[4] This is not the same as researching the issue and trying as many things as you can think of, which is definitely helpful in any context of question asking
It's weird that both sites' communities have made it difficult for old users to take part, but in completely different ways.
[1] https://www.quora.com/Which-is-the-best-WordPress-plugin-to-...
I mean, I get that SO wants to be a programming resource (as in "archive") where people with a problem should find a solution - not by asking but through googling. And so they want question/answer pairs that have a sort of general value, not an individual answer to just one person.
But then again: why? What's the big deal? Someone has a very specific question, and maybe nobody in the universe will ever have the same question again, but I'm willing to help that person out -- why shouldn't I be allowed to do it? Are they really worried about too much noise on the site? Please, come on.
In the situation I sketched above, I will still walk away having learned something new, but the person who posed the original question is left with a very negative user experience AND is none the wiser regarding their specific problem. At the same time, I was never allowed to help that person which I wanted to do not for the potential credit points but rather for altruistic reasons. Way to go.
The product itself has also become 1000x worse to use and I think it's one of the biggest examples of a Silicon Valley company that remains successful despite the constantly bad product management.
Anything that is spam worthy, like marketing or self help, is pretty much junk.
But other topics on there are fascinating. I read a few weeks ago an account of someone's grandpa who was an SS officer in Nazi Germany.
If you have an interest in a topic that isn't typically spammed by 'gurus', check it out on Quora.
Especially in the JS space[1], there are lots of historic answers upvoted on SO that are now wrong and you have to scroll around (or find a slightly rephrasing of the same question that occurred more recently) to find a good answer. Unfortunately the longer a question is around both the more answers it has and the more likely the answers are to be wrong. On forums / mailing lists the question would just be reasked. On SO it is supposedly a wiki so new versions of the same question gets closed.
[1] Because it has evolved a lot, Rust has similar issues, as I'm sure lots of languages do
Sometimes you have to work out the underlying Q is and answer that even if the OP doesn't want to hear it.
About that, there seems to be a recent trend that answer quality is so low that the only actual useful information is on the comments.
I swear these people must sit F5-spamming their review queue, just itching to flag every question possible. They really make it an unpleasant place nowadays.
Hey, you have to earn those sweet internet points somehow... Gamification has its downsides like this.
It would help showing that in your question. Often questions lack the necessary information to discern whether the asker has actually covered the basics
> [...] how you know there are other ways you could do it but you're constrained toward this direction for various reasons
Answerers are also programmers that like to apply best practices whenever possible. If you want to do something unusual which might smell a little without explanation your colleagues, code reviewers or similar would probably also ask "is this really necessary?". So you either have to write defensively from the start or deal with the patronizing-but-wellmeaning-comment-downvote-edit-reopen ordeal.
It's unfortunate, but there is an information asymmetry. The answerers can only see what you write, while you have all the background information specific to your case.
And you don't feel the same thing when you comment on HN?
I honestly think every point you made about SO can also be said of HN, reddit and almost any other place where people congregate online.
This is how the system is designed, IMO. It's the encouraged behaviour.
Hmm, we really remember things differently. I remember:
- Answers locked behind paywalls (ExpertsExchange)
- Searching pages of forum threads.
- Answers spread across multiple posts
- "Nvm, I solved it" https://xkcd.com/979/
In my experience, finding answers to programming questions is in a much, much better state of affairs now than it was in 2007.
I routinely cringe at reading some of the comments on SO, the general smugness on there. However there's no comparison to the past options. It's better in every way. Forums of old were overflowing with plenty of smug behavior as well, SO didn't spawn that.
This does not help when offering basic advice on employment out side the USA you will see I highly rated posts go on about "right to work" - this has Zero if not negative value and could actively cause harm.
StackOverflow has saved the shit out of so many developers. What would a junior engineer do without an SO thread?
How dare they feel smart huff puff
Because it really sucks to spend half an hour on an answer and then op goes "actually I'm doing this" and the spec shifts completely rendering your effort wasted.
A few times of that, and "clarify your question so it's very clear what you're asking" is a lot cheaper.
To quote myself in a comment on a now-closed question just yesterday:
> Your desired output would be invalid json anyway, what are you actually trying to do?
You really can't answer something reliably when the asker is posting things that don't make sense.
This is exactly how I feel about asking a question! Except "Actually I'm doing this" is "why don't you do that" or "why are you doing this" etc.
> Because it really sucks to spend half an hour on an answer
My feeling is that they aren't answering the question though, not the core one, there finding a thing that you're wrong about and talking about that instead.
It's like taking the above paragraph and refuting it by saying "*they're". Yes there is an (intentional) spelling mistake in there, but is that really the point of that it's trying to say?
Also, in case the blog author's name (Jon Skeet) didn't ring a bell, he's Stack Overflow's #1 contributor. If you've ever searched for anything related to C#, you've probably seen his answers, which have racked up over a million reputation points: https://stackoverflow.com/users/22656/jon-skeet
Political wonks are fascinated by process. They love the game. They get as much satisfaction from observing as changing. They want to be players above all else. Ideals bore them. History is mere data. Intellectuals seem irrelevant. What matters to the wonk are the hard realities of the ongoing political struggle. They defer to title and rank. They thrive on meetings, small victories, administrative details, and gossip about these matters. Knowing who is who and what is what is the very pith of life.
Geeks, on the other hand:
They are no less fascinated by detail but are drawn to ideals. Observation alone bores them. They are drawn to the prospect of change. They don't want to be players as such; they question the very rules of the game and want to change them. They are happy to make a difference in the ideological infrastructure, whether big or small. They tend to work alone and totally disregard caste distinctions.
As to your frustration with Wikipedia’s political elite
The wonks are the ones who consolidate, stabilize, and entrench the status quo; the geeks are the ones who prepare revolutionary change. The wonks freeze it into place and make it work more efficiently; the geeks imagine and work toward a future that no one thought was possible. The wonks rule out drastic and extreme measures as imprudent and reckless; they geeks think these paths are the only ones worth pursing, and have confidence that the unknown future will somehow work itself out. The wonks try to bring the king around to their point of view; the geeks kill the king.
Tucker describes the conflict between the two camps.
The geeks and wonks can work together but there will always be a natural tensions between the two. The wonks think the geeks are hopeless, powerless, reckless outsiders whose heads are full of useless and unrealistic fantasies. The geeks think that the wonks are part of the system and, therefore, more than likely corrupted by it, and increasingly so.
Broadening the view, the struggle to control history is a battle between the wonks and the geeks …
According to his view, Julius Caesar represented the wonks and Brutus the geeks in Ancient Rome. In the America’s revolutionary generation, they were Alexander Hamilton and Thomas Jefferson.
Once most of the pioneering work is done, and the low hanging articles are written, a community needs to decide how to proceed. Who are you going to maximize value for? In the early days it makes sense to focus on what is the rarest resource (in SO's case, probably the answerers) but as a community matures, the focus can shift. Now it is frankly the anonymous viewer that it makes sense to optimize for, both from a monetary standpoint (jobs display being a primary monetization strategy) and a social good standpoint (far more anonymous viewers than either askers or answerers).
http://blog.gardeviance.org/2015/03/on-pioneers-settlers-tow...
To address this issue, I put together a "deep-cuts" algorithm that uses the API to tries to find overlooked higher quality questions that have yet to be answered.
The code is available here:
https://github.com/klenwell/stack-overflow-deep-cuts/blob/ma...
It's not perfect. But if you're looking to engage with Stack Overflow, I've found it much more useful than the rat-race to answer (or close) new questions.
And some pairs do better/worse than others: http://minimaxir.com/img/stack-overflow-tags/so_tag_adjacenc...
I agree with your observation but Jon Skeet's essay seemed to be focusing on something else... the culture of hostility (or perceived hostility). The key words I noticed he used (multiple times) were "hostility", "jerks", "feelings".
On the other hand, the keywords I would associate with your observation would be... "boredom", "novelty of a new website", "low hanging fruit to quickly build easy karma", etc.
To discuss Jon's specific angle, let's use an example of a [tag] that's fairly new technology such as "[vue.js]". Since Vue is 2014, it doesn't have the "completeness of answers" of a very old tag like C# from 2008.
Here's a Vue question sitting downvoted to negative 4: https://stackoverflow.com/questions/49345098/how-to-get-prop...
Maybe his question deserved all those downvotes. I don't know enough about Vue to guess if the downvotes were all caused by a pasted screenshot instead of pure text.
To Jon's "culture" essay, is the minus-4 the optimal way to communicate it's a "bad" question? Is it a bad question because of its format (the screenshot) or was it bad because "props" is a trivial LMGTFY (Let Me Google That For You Question)?
Here's another Vue question at negative 1: https://stackoverflow.com/questions/49321665/how-to-add-a-ge...
To generalize, if there's "[new technology]" that opens a vast untaken area for new contributions of answers, is the culture created by Stackoverflow's mechanisms (the downvotes, the karma system, the moderators' powers, the "closings" vs "moving", etc) ... the best incentive system to minimize hostility?
Are Jon's well-intentioned "covenants" realistic? Or are they naive to human tendencies when any website is open to mass participation?
The downvote button is marked "shows no research effort".
Where's the "I looked at this manual and that tutorial and this debugger and tried these things"? Nowhere, which makes it feel like "I put no effort in stackoverflow do it for me".
The downvote button is an awful way to do what it purports to do. In practise it's percieved as more of an "oh fuck off" button.
It's a single mouse click by a single person (or maybe a bot), but it carries so much weight.
Personally I try to use my downvotes only as a last resort, for factually wrong answers or questions where clarification attempts went nowhere.
The guy is a legend. If there was ever an MVP in the C#/.NET space, it's this guy. His books are great too.
he has (had?) some competition from eric lippert. i used to love reading his blog when i did c# work.
I went all the way from a simple coder to management, and now I'm doing both, running my second startup. I talk to people of different walks of life, and I always do my homework.
But on Stack Overflow, it's like I'm back to the early years of my career, a stupid newbie whose questions get randomly downvoted. I delete it, tweak the wording a little bit, and now it's suddenly accepted. I say something like, "a solution or a 3rd party library", an idiot downvotes me saying that "requests to recommend 3rd party libraries are offtopic". I change it to "a solution (3rd party library is OK)", and now it's suddenly fine.
It's like the mafia of high-karma answerers decided to adopt Les Grossman as their role model.
As a side effect, the good answerers are gone, and the really tricky questions are left unanswered, so I often answer them myself. Although, the way Stack Overflow is, my desire to contribute to the community is diminishing.
I usually don't downvote answer that ask for 3rd party libraries, just vote to close. Your example is a bit borderline and it's something I wouldn't close, but I can see how someone might think it warrants to be closed. Consider that the rule about external tools is meant to avoid a deluge of answers about libraries. They may be useful, but none of them is objectively right and can cause religious wars with people downvoting answers about libraries they don't like and massively upvoting their favorite library.
If we can avoid this by being strict, I think it's a good thing. But hey, I'm just one of hundreds of thousands users, so it's not like my opinion is somehow more valuable than yours.
Like many, I find the answers to these "religious" (in fact, they are not, because done properly the answers are supported by arguments) questions to be the most useful content.
In many ways, I can answer the most objective questions myself from the documentation or source code. But what I cannot do this with is distillation of expert knowledge that requires time, and the more distilled it is, the more subjective it is. Subjective doesn't mean unsupportable by facts (experience).
I call them an idiot because they were nitpicking on a question that was asking for help well within the already Procrustean bed of regulations. I stand by my assessment: it's hardly a reason to refuse help citing bureaucratic regulations if you're a public servant, but that's somewhat expected. If, however, it's a fellow forum contributor, this is both unexpected and serves no purpose.
Not sure who's right, of course, but as you see, the number of people complaining is greater by day. Note that many of them are familiar with the Internet since 1990s and are netizens of good standing; still, they manage to somehow "violate" these rules.
As soon as there is a real competition to the "soup Nazi's shop", many people will jump ship without hesitation.
It's also interesting that other Stack Exchange communities are not that strict.
why ? why do you do this ? a lot of time there is an exact piece of code that does what OP asks for, what does it change if the piece of code is in a SO answer or in a github project ?
But you admit to trying to suppress his ability to express that opinion and see no problem in doing so?
IMO it doesn't scale indefinitely. There's a site culture to be maintained, and as the site gets larger the core moderator culture becomes a smaller and more isolated part of the site.
That's very much considered offtopic on SO. Maybe other sites on the SE network such as Information Security or Software Engineering might have taken it. But I don't use either of those, so I don't know if that would actually fit into their policies.
Broad questions like that don't have a single good answer that fits every scenario, they have lots of different answers, each of which is the best fit for a different scenario.
The problem with a site like SO allowing questions like that is that you _would_ get an answer, that answer would be accepted, and that question would appear really high in search results. When the next person comes along needing an answer to the same question, but in a different scenario... they're gonna take the one that was given there, even though it isn't the best fit.
Scale this up to SO's scale across a few years and all of a sudden SO isn't a trusted resource any more; it's more like experts exchange used to be, where it was full of answers which _can_ be valid, but are basically useless to everyone, and taking that advice can actually be actively damaging.
I get why your question is on hold. But I find it unfair. Some of the best questions and answers have been highly opinionated. And I have often been led to on hold questions by google.
An opinionated SO site might be nice. In the old days we had newsgroups for this. But that audience has disappeared when everything moved to the web.
It’s like the high karma answerers are exactly the subset of the site that you want seeing and answering your question because the site supports the idea that they are good at doing that, but they’re always reflected as some kind of tyrants appointed by an evil overlord.
Broken site design somewhere that leads to that effect.
Are you 100% sure?
Because from my experience, the most useful subset is between 100 and 2,000, those with experience but not arrogance. The top karma hoarders more often than not provide citations from official documentation and sabotage questions.
I wish they were appointed by an evil overlord because the overlord is better than a swarm of annoying mosquitoes.
And .. I downvoted you. I hope it will sting. People giving up free time to add value to your thing and you blame them for the site problems? Then the site design is really failing, if the problem is all the people being engaged with it.
Whereas on stackoverflow when I downvote it’s for “this question shows no research effort” which is the downvote button tooltip, but a single downvote is felt as a personal attack on character.
Stackoverflow is for programmers - people whose character traits are tainted to be nitpicky about small details - yet the bad experience angry posts call answers traffic cops and Martinets - exactly the kind of people who could help get your code past the compiler or browser or api or whatever.
“Neutral” responses feel like negative ones (see: “what have you tried?”).
As an answerer more than an asker, if I help someone understand then I feel good, if it seems someone asks from a sense of “help me” I’m endeared to the question and if they ask from a sense of “solve my business problem” I feel more abused (personally or site-being-abused) and resentful.
No amount of Jon Skeet blog posting about the classic usenet popup “this reply will be sent to thousands of people who will have to spend time and money on it, are you sure?” seems to work on the modern web.
If it were me reworking it, I’d get rid of the low quality question idea, let answered be the judge of that and google sort the wheat from chaff.
Then get rid of downvotes and close votes and “neutral” comments.
Leave only room for feedback buttons that in some way communicate “I tried to answer your question but couldn’t, Your question might be more answerable if ..”
Forget being a repository of high quality questions and answers in the same way software has forgotten about big design up front and perfect quality goals and turned into “put an MVP on GitHub and let the issue tracker be the driving force”.
The goal of “stop duplicates, everything that can be questioned and answered has been questioned and answered” forgets that there are 150,000 new humans every day, and thousands more existing people learning to code - a goal of turning everyone away with “that’s already finished nothing for you to do here” is .. daft.
Questions and answers are, if nothing else, practise. A good way to learn is to try explaining it to someone else - SO churn has that room.
But not on stack overflow, where following the site plan for downvotes and closevotes makes everyone feel shitty.
I even doubt it is “high quality questions” that are supported but a combination of “this question feels like the asker is like me” combined with “and they’re asking about a thing I like and feel positively about rather than an edge case of the tool I feel bad about and wish the asker would politely overlook it like the rest of “us” all do”.
Please (re-)read the site guidelines and follow them if you want to keep commenting here. That means, among other things, posting with scrupulous respect for others and eschewing jabs and swipes.
https://news.ycombinator.com/newsguidelines.html
I can't see whether I was downvoted but I see that the net result is very positive. But I'll take your word on that.
Perhaps your view is not shared as widely?
> People giving up free time to add value to your thing and you blame them for the site problems?
Do read my original post. My problem is people who don't actually contribute but nitpick on how the questions are worded. How does kicking me out or ignoring my question at best help me resolve my issue?
I would prefer them to keep the free time they dedicated to harassing me to themselves.
I do find your suggestions on design sensible though.
https://math.stackexchange.com/
Because the difference is night and day. No pedantry (no pedantry on a math site! Impossible!), no assumption that askers are throwing their homework on them. In fact, my experience with mathematicians have had me question whether I choose the right career or if I should have become a mathematician. They all seem damn chill in comparison to most of us developers.
Anyway, the idea behind that site appears to be to help people learn math. Not to build the world's most polished question repository. I very much prefer the former approach.
This may diminish the quality of forums because a lot of people just want somebody to solve their problem so they can move on without doing any work. I see that also in a lot of new hires who don't seem to care about actually learning programming but just want to do the bare minimum.
The Security Stack Exchange is full of a similar sort of people. Almost all of the new questions at any given time are either tech support "I think I have a virus!!" or low-effort handholding questions "how do I use kali linux to hack server". I think folks who spend a lot of time on these sorts of websites have adjusted to be more jaded towards questions that look similar. I've learned to ignore it and be thankful that (in general), questions that I ask on stack exchange tend to receive helpful answers.
Stackoverflow has questions about some ridiculous configuration in languages and libraries where the asker couldn't even bother to pinpoint the issue but rather dumped all possible variables into the question.
Seems simple: math doesn't have to deal with that.
helping people learn: achievable with very high reward
Creating a catlogue of perfect questions and answers: impossible and of limited resource as every answer only caters for a specific set of back ground knowledge
And what is funny is that with option 1 done right who needs option 2? why scrabble around in the dark when i could be connected with someone who actually wants to help me.
i think that software dev is saturated with people who like to learn lots of things by themselves (excellent work, this is a form of hero), however this is not a good representation of most people. if we want IT to be more inclusive we need systems that can handle people who dont want to read several textbooks before getting some code working.
I think within SO there's sub-communities too just by tech stack. Some of the deep esoteric stuff might attract some snobbery.
Dead Comment
Dead Comment
While this is meta it pretty much sums up the situation regarding my feelings regarding Stack Overflow.
I was an early user of Stack Overflow, not quite beta but almost. At one point I even received a snail mail letter from Jeff Atwood, thanking me for my contributions. I read most of the early discussions about policy and, being a veteran of various internet forums at that time, it seemed to me the most sensible, most reasonable approach you could take to running a question-answer site. I went to CodeKen in London in 2011 where I met Jon and a lot of other excellent folks that made Stack Overflow happen as a community back then.
Today, I rarely consult Stack Overflow, and when I sometimes find a useful answer via DuckDuckGo it's almost certainly closed. I even joked that closing is a sign of quality at Stack Overflow. My last contribution is probably years ago.
The strange thing is that there is no point I "left", nothing I could point at and say, that's when Stack Overflow took the wrong turn. I think we just drifted apart. And for the conclusion I guess for me it's mostly: "The way to hell is paved with good intentions".
I’m sure there are topics I don’t use (the latest ML framework or the js framework du jour) where the community is still growing the database. That I use it as a mostly completed reference work just means my field/tech is mature in terms of language/frameworks
Unless you're working with a dead technology, by the time the information is "complete", it's outdated. And I would consider much of the information on SO to be just that: outdated.
But people still ask those questions and the questions apparently get indexed. In my experience the result is that when coming from a search engine, even for non-fit topics, it is quite common that closed Stackoverflow posts occurr several times on the first result page.
It could be that I search for strange things, but at least for me the combination of closed but still indexed posts are not helpful, especially when the question didn't get any useful answers at all before being closed.
I don't mind closing topics, but I would prefer that closed questions with no even remotely useful answers wasn't indexed at all. Now, that might not be entirely under your control, but the issue remains, or so it seems to me.
> We hadn’t quite figured out that opinions and discussion questions were not a good fit in the early days.
This is something I had to learn as well, I guess one of my most upvoted questions falls well in that territory [1]. I just think that Stack Overflow was best when you were its primary moderator and it's arch enemy was Experts Exchange [2]. That battle is won, but the war goes on. Nowadays when the first search result is Quora I have very much the same feeling I had regarding ExpertsExchange back in the day and I can't help thinking that there is lost opportunity for Stack Overflow and Stack Exchange.
[1] https://stackoverflow.com/questions/961942/what-is-the-worst...
[2] https://blog.codinghorror.com/whos-your-arch-enemy/
I know you've written about SO not being a good place for opinion-based questions before, and perhaps in theory you are correct, but I (at least) am not satisfied with the implementation, and tend to link that at some level to a flaw in the theory.
What I mean: some of SO's most [in]famous questions are opinion questions where the askers are requesting advice from "expert" answerers to help them navigate a new topic. They're not just infamous because they may be controversial and "hard to moderate". They're famous primarily because the way in which the question was asked drove many users (hundreds even thousands of upvotes with millions of views) to the site and helped a lot of people through their journey working on a new technology or in a new problem domain.
SO said "nahwp, we don't want questions without clear cut answers". Not necessarily SO's fault, but nonetheless their problem: the community took that as, "err on the side of annoying pedantic asshole and close, downvote, and otherwise ostracize any user asking any question that might not be a good fit for the site under the new regime".
In my experience, it feels like people are trying to find a way justify being annoying instead of find a way to provide meaningful advice. I've asked questions that, if you took 30 seconds to think about instead of a quick search for a marginally related title (which I already did in the 4 hours prior to asking my question) and immediate vote to close as duplicate, or an almost bot-like flag as "too broad" based on word choice heuristics, are not bad questions, and often after hours of flagging and arguing with the original users who are stubborn because they don't want to be wrong (especially publicly) I can make my case. But by that point it's too late and too many times I've had to do that to fend off users trying to do good for the site over helping me with a problem. And I'm not even asking opinion questions...
By making one small decision, you created two problems.
1. You actively choose to diminish the value to practical real users of the site by setting a long term vision of becoming some bastion of objective knowledge. We have those; they're encyclopedias.
2. You unintentionally validated the pedantic/authoritarian desire for and abuse of power that has deterred many of your original users.
Honestly most people can handle snarky comments and opinionated edged answers. Everyone (users of SO) has a tiny (and sometimes large) "programmer ego" inside of them. On SO you could responsibly exercise it. Not anymore. What I can't handle is nonsensical authoritarian and meaningless abuse of questions because they don't immediately and explicitly adhere to some SO agenda. The former can be mildly annoying at times but tolerable and hell even fun on occasion, the latter is frustrating and infuriating and something I cannot tolerate.
Sorry that really turned into a rant, didn't it?
Websites like StackOverflow, Yelp, Wikipedia start off by being fantastically useful by using crowd-sourcing to quickly generate a large volume of content that covers topics too wide for a small group of humans by themselves.
But then crowd-sourcing becomes an albatross because an unpaid army of moderators have almost no oversight. They inconsistently apply (or don’t apply) standards, and blacklisting vitriol is a Sisyphean task.
Valid questions get blitzed and jerks proliferate because it’s seemingly impossible to police them all in real time. I saw a stat that said only 7% of SO users ever ask a second question - that means 93% probably had a bad experience.
While the problems are structural IMO they can do a few things. Downvoting should be temporarily disabled until Joel et al figure out a better model to surface important/good questions and answers; I personally now visit the forums of whatever framework or language I’m working in to seek advice because the downvoting on SO seems so inconsistently applied. Insults I can ignore but the DV makes it hard to have your issue seen.
If it gets a thumbs up, it qualifies for the regular Q/A where people can answer - and expect a certain quality (and don’t risk spending 20 min writing an answer to a perfectly good question that was closed while the answer was written).
No open platform, as far as I can tell, has solved the problems of vitriol, but the down-voting is absurd (as attested by some of the experiences described in this comment thread) and can absolutely by addressed by changes like the one you propose or some other system.
How would you enforce this - would you try to enforce it? If there's any kind of comment box, you'll get people posting answers in there, if it's amenable to a short answer.
That doesn't follow. The one-question-askers may be using throwaway accounts, or be confused about what SO is for; they might be in completely the wrong place, and have a low likelihood of ever needing to ask another SO question. They might also just not run into questions that are only answerable by asking the open internet that often.
Why would you use a throw-away account? There something personal about a compiler error?
> they might be in completely the wrong place
I assumed mods moved questions to the right place or directed users to correct board.
> have a low likelihood of ever needing to ask another SO question
I can understand not asking 50 questions. But never asking a second question? Your entire programming career you only have one technical problem you need help with?
> They might also just not run into questions that are only answerable by asking the open internet that often
Yes, a lot of the technical problems I encounter at work are too specific to our implementation for SO.
No retention rate will be 100%, whatever the product. And there are plenty of legit reasons why you may not need to ask a lot of questions.
But 7%? They can't all be "noobs" asking dumb questions since plenty of HN devs here complain of similar problems: arbitrary downvoting, questions being marked as duplicate when they're not, unaccountable mods, etc.
To me the fact that 93% of SO askers don't bother with the site after their first experience signals a deeper problem with how SO is structured. As one of other comments in this thread argues, their setup was designed for its initial phase of growth but it's not ideal for the "mature" phase its in now.
1) it want to be a repository of canonical knowledge, a giant FAQ and wiki of everything. The problem is, that it does not fit the way the site actually works: people will write imperfect questions, precisely because they don't know the answer, and ask the opinions of their peers about things. But they are not domain experts writing documentation so they often get down-voted or their questions closed.
2) There is a karma system in which more karma gets you more power. Inevitably such system attracts people in search of fame or power, and this kind of people get off by telling others that they are smarter than them, or that they know this little site rule better and consequently censor their question. As regular users become fed up with the state of things and leave the site, the position of high-karma users become more entrenched. Then, those users obtain total control over the meta board to veto all fix proposals, make up even more draconian rules, and entrench their position even more.
Problem 1) causes inadequate meta rules, and 2) zealous enforcement of these rules. Some combination of 1) and 2) prevent the site owners from changing the rules.
But if one asks or answers questions about, for example, more advanced features of git, more advanced algorithms, numerical mathematics, GIS, anything that is not interesting to "the masses", then you still get genuine helpful answers, but you might have to wait hours or days, and there will not be a large number of answers.
Although still relevant as I ran into the same problem as the OP but as nobody else answered I had to come up with an answer myself.