StackOverflow modding itself out of existence
movingfulcrum.com
movingfulcrum.com
I hate to be contrarian, but I have not experienced much of the negatives people are complaining about here. StackOverflow has been a fantastic resource in my programming career.
I used to be pretty active on SO and Programmers.SE as a young buck, and back when SO was still pretty new and shiny. Nowadays it just feels like a race to the bottom as someone trying to answer a question. The winning strategy is to post a shitty, incomplete answer and get the upvotes and while that is happening, update to a better, clearer answer, oftentimes poaching content from the guy or gal who took the time to do a detailed, "good" answer. I get the notion of consolidating the answers into that one perfect answer, but working on a good answer and having it poached by the highest voted answer will leave a bad taste in most folks' mouths.
In the case of Programmers.SE, there was a shift in the site topic and, in my opinion, some rather overzealous moderators along with the handful of folks who just seem to sit on the site and close-vote any and everything that does not perfectly align. Always be wary of those that seek power over their fellow man...
StackOverflow and the surrounding sites are a good force overall, don't get me wrong, but as someone who tried to be active and use the site, it leaves a lot to be desired. I'm not interested in answering "How do I do X in jQuery" and I'm not interested in having someone plagiarize me.
That's true, if you care about the points, badges, etc.
Jeff Atwood has discussed this subject fairly extensively, and it does matter to a lot of people, which helps make SO more successful than it would otherwise be.
Could you build an incentive strategy and reputation system around collaborative answers?
While I don't argue the demonstrated knowledge and skills, having talked to job recruiters, SO is actually very far down their list to my surprise.
So if you are getting discussion, you are actually just experiencing the common case of getting your needs addressed before the terrible moderation kicks in and closes good content.
Also confusing from the point of view of the site operators - aren't they missing a lot of site interaction and pageviews and additional free content, etc. ?
That's not very smart. Sometimes, even a technical problem's context can be complicated enough to merit several different "good" answers, depending on other factors.
Maybe someone should formulate a law -- over time, any site's moderation starts to suck. There are factors involved with this, such as depletion of low-hanging fruit, popularity drawing more attention from spammers, and increasing workload of moderators.
It happens quite often that I look up something that I already have (or could write) correct working code for. It's when I notice that there's more than one way to write that code, and I want to do it the "right" way the first time, so the decision doesn't come back later and bite me, when I go out looking for what is considered "best practices" in such a situation.
Usually there's more than a few "correct" answers to what is considered the best practice, and the choice depends on the context of your problem. That is when the discussion is invaluable, without it I cannot determine which of the choices apply to my problem.
But of course it's your fair right to say "well okay, but SO is not intended for that".
I don't think that SO is a bad tool, but it may well be a bad tool for a certain subset of programmers seeking help. I would be interested in figuring out who that segment is and what they want!
The issue I see is that they have been so successful that now they are THE place where you'll find someone capable of helping you.
And then you have a question where you are sure that you could get great input from the expertise assembled on SO, but you aren't supposed to use it because the question has the wrong format. And there is no other (obvious) place to get the answer, because SO has all the mindshare, because they cover most questions.
A site that only caters to "questions not applicable to SO" would have a very weird scope, and a general site has to compete with SO for the SO-type questions, which is hard to do because SO does those very well.
Additionally, if the relevant information ever changes, no new answers can be submitted.
I'm not sure if that's the best way to do it, but it's an option.
Something tells me that's never going to happen, though.
I think if a question is closed, that would mean no corrections or updates would be possible.
You can't vote on closed questions or answers to closed questions.
A question can't just be closed. There needs to be a reason to close it. "This question is a duplicate of this other question" is frequently used. Often the questions are similar but different enough to need a new question. You can't ask a broad question to capture all the variation - that'll get closed as too broad. You can't focus in on your exact problem -that'll get closed as too narrow. The gap in the middle is full of dupes. It's pretty demoralising to have people picking over your question and closing it for (seemingly bogus) reasons or asking for pointless clarification.
Anyhow, SO isn't the same site it was 5 years ago. It's actually pretty darn toxic for new userbase and seems to cater to the mods themselves.
That's another frustrating gap in SO: the absolute ignorance of the notion of time and change. Both questions and answers often have a significant timeliness attribute (ex: what are the best online resources for learning <x> as of Jan 2015), and SO has no support for this. To be fair it's a hard problem, but I'm not even aware if it has even been discussed.
I'd like to think that the platform is now mature and tuned enough that they could start to introduce some major new functionality rather than just keeping the lights on.
Here is a question of mine for example (http://stackoverflow.com/questions/4701538/javascript-based-...).
Leaving aside the question if the question is indeed off-topic, I don't know why anyone will come 4 years later and question intelligence of person asking the question and person answering it. Are these mods suddenly wiser than participants 4 years ago or here to stroke their ago?
There used to be a question about popular programming blogs. To me it looked like a good and popular question, so I asked a similar one - which was then closed, because by then the consensus among the mods had moved on to discourage that kind of question.
Which brings us back to the original question: why isn't it welcome. Yes, I understand the very real requirement for SO to drift towards a forum, but I think they've drawn the line very far from the optimal place. Furthermore, the insane success of SO has largely killed off specialized forums where that type of very valuable conversation formerly took place. I wish SO would introduce a discussion forum site or "layer" (integrated into SO itself in some way).
Jeff's motivation for SO was something along the lines of making the internet a better place - in my opinion, many of the current mod's actions are contrary to that initiative. SO is still a massive net benefit, but it could be better with a little common sense.
"Update 2: Thanks to the Hacker News community, the post has now been reopened!"
One would hope it would prompt some discussion and soul searching, but I expect not.
Over the past year or so, we've altered the mechanics of close voting to encourage fast closure and discourage gradual accumulation of votes. If you're interested in the details, see: http://meta.stackexchange.com/questions/250223/age-close-vot...
A question about a *nix software on an Apple device? That could be superuser, or unix, or askdifferent, or maybe the ubuntu one whatever that's called.
I don't see why "closed" as such is problem. I could interpret it as "someone asked this question and it's been answer, we're moving on".
Sometimes I find an answer on a random web forum - if the forum still permits answers, there will 20 or 100 pages to look through and the usually the only good things will be right after the question.
The web at scale is just messy and cruel.
Closed as "out of scope"/"duplicate" isn't saying "this is finished, moving on", but "don't ask this kind of thing, go somewhere else, even if you probably would get a good answer to your related question here".
Voting is also possible on closed questions and answers.
That said, the question - "Of course the Unsafe class is undocumented, but how can I use it in a real world scenario" - is pretty much the definition of "too broad".
That said, I voted to undelete and reopen, which appears to have succeeded.
The rules should be bend when it is appropriate. I'd argue that this is one of those cases.
The true rule should be: Is this question useful to people? Will other experts be able to provide useful back?
The rule most in need of context-sensitive, question-specific interpretation is the closing of opinion based questions. The number of useful, highly-upvoted questions closed for this reason is the evidence that something needs to change.
Exactly, and it's my understanding that this is very explicitly not the rule - moderators do not care if something is useful, they care only if it "fits the guidelines" - actually guidelines is not the right word, laws would be more accurate.
I would disagree with it being called 'laws', because in the general sense laws can be disobeyed on occasion, and regularily are (extremely minor traffic laws, such as using the horn while stopped, anti-tailgating laws, etc. are routinely broken here in the UK -- I don't condone it, but it does happen quite a lot). In some cases police have the discretion to ignore offences of some extremely minor laws in favour of using that time instead to catch proper criminals. Hence to me at least, that does not convey the right mannerisms.
Perhaps 'sacrosanct law' would be a better alternative? :D
It might even introduce new terminology: "The chances that this question will stay open forever are approaching infinity..."
For every answer to this question, SO would be better served with a question asking "How does [Java internal or widely used library class] accomplish [thing that seems impossible in Java]" for which the answer is "It uses sun.misc.Unsafe in [this way]".
Yes, the information on this page is useful, but it's not stack overflowish.
Still, seems crazy to close it if more sensible versions of the question don't exist.
Either you're being sarcastic, or you have no familiarity with the history of science.
The rest of your answer shows another aspect of what's wrong with Stack Overflow: zealous mods who get an idea of the perfect question into their heads, and then who go out to purge anything that doesn't conform, regardless of whether others find it (and its response) useful or interesting.
This question is useful as a place for blogposts to link to, not as a primary answer to a question you might Google. So it's 'wrong' for stackoverflow in that sense. But no, I'm not agreeing that mods should have closed it, and I didn't say it should have been purged from SO. Luckily, it hasn't been, so we can stop worrying.
I completely disagree. If you are a new programmer who knows nothing about a certain topic/keyword/etc, googling "what would I use <x> for" is not only not being "up to no good", it's just plain common sense and curiosity. How else might one learn about entirely new things?
Maybe they would be better categorized into some group that is more appropriate, but then you lose the possibility of someone within that knowledge domain being able to answer the question appropriately.
Those questions run right alongside reading documentation for fun.
I suppose an overhaul to the infrastructure of question answering could help solve this problem - like questions have a maturity cycle / process in which they are developed into a more refined / edited sorts of comprehensive documentation, or even a better 'smarter' way of linking clusters of question together, or a way of moderating the authoritative direction of a question (rather than delete or replace, show a progression of modifications so people can learn what are model questions and what are less useful kinds).
Getting people to stop asking kinds of questions will never work, because they ask those questions precisely because they have not been trained in any sorts of formal internet etiquette. If you want to prevent the community from collapsing in on itself, I think the best solution is to modify the flow and connectivity of questions, so that they are intelligently inter-related, rather than relying on a sort of competitive singular universal form.
Also you give no reasons as to why your preferred way of asking questions is superior other than stating as a fact that "SO would be better served" by it.
That is an extremely good and useful question. It is called the bottom up approach. The fact that it might or might not be asked generally, has nothing to do with what we should do the times it is asked.
reminds me of accurate wikipedia articles closing or being removed because wikipedia was trying to be an encyclopedia, instead of the first source for information on anything and everything you want to know about
My detailed question was closed because I mentioned Webmin installed.
I no longer actively answer questions on StackOverflow, because it is too hard to find worthwhile questions between all the noise. If someone specifically asks me to answer something, I will do it, but trying to find interesting questions on StackOverflow has become impossible due to the high influx of very low quality material.
There was a period where I'd try to help moderation by closing an deleting questions, but at some point you just realize it's a loosing battle and give up.
I don't think many people are arguing against moderation of questions in general.
My point here is that, in my experience, the main problem of StackOverflow is too little or ineffective moderation, rather than the other way around.
Eventually I just deleted my account and left since my motivation to answer stuff over there was a bit more than "I'd like to help people". I also liked to find questions I can't answer and figure out how to solve them on my own. Those became buried under lazy questions to a point where I had to dig for hours to find one. Maybe android was/is a bit of an extreme case, but still I see the same pattern more or less when I occasionally browse SO after finding an answer to my own problems over there.
As for the OP, I think a lot of people don't get the SO format. Yes, that post contains a lot of very useful information. That's great, but: No it's still not a good fit for SO, SO is for specific questions with a clear answer and that post isn't that. It's overly broad in scope and has no right answer. One can argue about whether deleting really helps anybody (don't think so personally), but closing it seems appropriate considering how the site is supposed to work.
EDIT:
Also reading that question (thanks for the archive link!) that looks like the kind of question I would like to see more of actually. Sure it's somewhat subjective since it'll be experientially dependent but it certainly moves the ball forward in terms of helping someone understand a technology, language or language feature.
I suspect the link was provided mostly for the users that can still see it. There is an "undelete" link[1] for users with that privilege level, but it takes ten votes to restore the question, with no guarantee that it won't be blown away again right after its restoration.
There's nothing wrong with that idea, they just have a bug in the implementation. Somehow they've incentivized the closing of good questions that are frequently on-topic.
Here's an example of a plainly on-topic-yet-closed question that annoyed me recently: http://stackoverflow.com/questions/2847655/find-threads-runn...
StackOverflow is for programming questions. That question, as it was formulated, is not about programming. It belongs on either Unix&Linux or Super User instead.
Of course, it does [1] (as an aside, read the comments on the answer there, most of them are people bitching that they can't earn rep on community wiki posts). The first link in the answer there points to an old blog post about "The Future Of Community Wiki" [2], which I highly recommend to anyone interested in this controversy as food for thought.
I see a lot of parallels between this problem and the problem that Reddit is having right now. It all stems from the question of how much these sites should be considered "communities." To what extent is StackOverflow really a "place"? Do fun, off-topic or poorly-formatted questions really ruin the reputation and mission of StackOverflow if they are but a subset of pages that happen to be hosted side-by-side with other pages that include content that can be found with a search and that many people will find invaluable?
[1] http://meta.stackexchange.com/questions/11740/what-are-commu...
[2] https://blog.stackexchange.com/2011/08/the-future-of-communi...
Perhaps you can take issue with the fact that StackOverflow doesn't allow overly broad questions, but personally I find it helpful. When I find something on StackOverflow (or more commonly Cross Validated), it's almost always someone solving or pointing to a way to solve a problem very similar to what I encountered. Allowing extremely broad questions would diminish the likelihood that these results can be easily accessed.
https://web.archive.org/web/20150404191857/http://stackoverf...
Having said that, SO has been nothing short of amazing when you look at the vast amounts of information it provides. However, I never really had a good experience receiving responses on anything I've ever asked. I'd typically get a comment or two but answers are rare.
Funny that https://docs.google.com/document/d/1GDm_cAxYInmoHMor-AkStzWv... is top post too.
Appealing after adding additional info showing how the questions were different didn't improve anything as the "reviewers" got points by "Reviwing" the appeal-- and denying it, of course.
That was the last time I used Stack Overflow.
You cannot have a community generated content site with community moderation where bad behavior is incentivized and rewarded. (Something that was a serious problem on HN in the past.)
Explain for a newer HN user?
Jeff Atwood used to think very deeply about this type of thing, the nuances of human behavior. It seems to me he no longer has the time or interest, and no one has replaced him in that crucially important role.
Put the information up in a blog post, or anywhere else. It doesn't have to live on SO to be useful to everyone. SO is really terrified of turning into a reddit or a forum full of "What's the top ten coolest ...?" type questions.
What is a good place to pose such questions? I don't believe Ask HN is one, since time-on-frontpage is very short, and if thread does not get upvotes quickly, it goes into oblivion, and also, while too broad for SO, they're probably too specific for HN :)
Any recommendations?
Maybe there's a business case for a portal "TooBroadForStack.com" or sth like that ;)
Here are some other places. Each has their own problems. I'll leave it as an exercise to the reader to determine if they are better.
reddit, /g/, IRC, lobste.rs
Say you find a bug with Product X v1, create a workaround, and post it to your blog. In V2, the problem has been fixed, or exists in a slightly different incarnation with a slightly different workaround that someone figures out only after reading your content. Maybe they'll leave a comment on your blog about it, but you don't really have any incentive to update it. If your post was an SO question, they could post another answer or make an edit, and the page becomes the canonical source for information about the bug.
How many times has HN discussed the digg/reddit/HN decline in quality as the population grows?
In their minds (and I have to give them the benefit of the doubt given their awesome accomplishment!), keeping strictly on topic is one way to prevent the point of the site from drifting.
It may be that the SO people "want" is a better one. But that's the decision that Atwood & co. have made. To wish for a SO that allowed off topic responses is to ask for a different product.
Doesn't this ignore the power of having a common place to go for questions and answers, back and forth, etc.? I visit a couple of sites almost every day, and (I hope) help as well as be one of the people helped. But I'm not googling extensively for newly-posted things every day.
After reading through the help center info linked, it seems like this question would fit all the requirements in its current form. It's hard to look at past versions since the page doesn't exist on SO anymore, but how was the question originally stated?
http://meta.stackexchange.com/a/224922 has previous Stack Exchange data dumps, which include many of the deleted questions. With sufficient effort somebody could use to construct a partial archive. (Provided that they follow the requirements of the Creative Commons license: http://s.tk/attribution.)
I wonder if the people from archive.org are interested in long term archival of these data dumps.
At the very least, overall it's.. inconvenient for our tech community to regard information as volatile so often. Especially Q&A. Take threads on forums for example, that get closed simply because they are old. What's the word again? Necroposting?
Sometimes I find an old solution to something I'm currently struggling with. And that goes to show that age says nothing about quality. So it's a real downer to me when I see people complain to someone who responded to a three year old topic when at the same time it's clearly still relevant to someone.
When someone has a problem, searches for a solution, finds that same question asked 4 years ago on a forum and unanswered, later finds a solution, and decides to share that solution by reviving the 4 year old forum post, they get nothing but thumbs up from me.
Information simply does not expire. Wish people (moderators in particular) would stop acting like it does. In most cases, if someone replies on topic, it is clearly still relevant.
Sorry for the digression, it's just a pet peeve.
Anecdote is not data. StackOverflow is pretty well designed in regard to the sort of problem the author of the linked article encountered.
I'm still convinced that the answers are interesting and useful. At the moment it's locked because of "historical significance" which makes me little proud even.
[1] http://stackoverflow.com/questions/961942/what-is-the-worst-...
This is one of these rare cases. I'm glad that this question was re-opened and got protected - I think that was the right thing to do.
All that said, I'm sick and tired of all the winning I hear about SO lately, so many complaints, yet, everyone keeps using it.
I do not encourage a vicious behavior as sometimes seen in SO but I also don't like to watch people post again and again the same questions that have been answered multiple times and could be found in a simple google search (and which probably show on the right bar when they compose the question).
I'm also tired of people complaining about closing questions - dudes, if you don't want your question to get closed pretty please put some effort in it! do your research, post what you're tried, compose a minimal working example that reproduces the error, post a FULL stacktrace and show which line triggered it. But no, people don't like to "work hard" so they put all the burden on the people that are here to help them. This is NOT cool. So yes, fuck yes - I will keep voting to close poor questions, and even (potentially) good questions that the author didn't bother to give a fuck about making it clear, readable & reproducable (but please do not dump all your code and expect me to read it if only one small method is relevant) and no - this is almost always NOT a language/culture issue - regardless of what some people say.
Bottom line: they say "in rome act like romans" - well when you join a community you should follow the community rules. Take a few minutes to go through the help center, see if what you're asking is on-topic. Learn how to post a new (good) question. Hell, even ask someone you know to read what you've written and give you the feedback if it's clear and if it makes sense. You want help? so put the time and effort so that we won't waste our time & effort when we try to help you!
Have a nice weekend!
P.S. Those who know me, know that I like to help people regardless of what I just wrote (example): http://stackoverflow.com/questions/31358932/is-there-a-payme...
but I really am tired of hearing all the criticism. It bothers me especially when I recognize that it comes from intelligent people that I'm not sure if they're just "playing dumb" or being a "smartass".
~~~~~~~~~~~~~~~~~
Help us help you!
I'm not sure if your typos are "dumb" or ignoring "autocorrect". See how we can both be pompous dicks?
The worst part about SO is pretending like you are trying to help people learn, when, as other commenters have called out, the answers are spread across multiple replies. If a person asks how to do X, and your answer fails to do that, you are part of the problem.
Please don't do this in HN comments. It's true that upping the ante is common on internet forums, but as you probably know from lurking here, we try not to be that way.
All in all, I've moved onto just reading the documentation and figuring things out occassionaly asking a very narrow question on SO.
If anyone is thinking of disrupting SO's space now is a perfect time. Ironic that people are now searching for stackoverflow alternatives.
the largest barrier to this is user training-- informing and getting people to move over. I originally conceived of this barrier 5 minutes ago when reading about HappyFinder or some other fuzzy finding tool that made it to FP-- why do we have to have a complex name that redefines a word to fit the solution space? The most efficient would be 'AHelmAlternative'. Similarly the competition to 'StackOverflow' would be 'StackOverflowAlternative1'
sounds horrible, but uses encoding to efficiently replace extensive time/marketing to gain mindshare. Oh, Alternative1 didn't work, we don't like it so we're going to start our own, it'll be called Alternative2. 'Hey, we made it, we're going to rename in 6 months. You can start using the new domain name now.'