Unfortunately, StackOverflow is difficult-to-avoid nowadays… | Hacker News Reader

Unfortunately, StackOverflow is difficult-to-avoid nowadays…

marc.info

41 pointscnst21 comments

Looks like the case of snobbery outbreak right there. I mean, you can have issues with depth and precision and other stuff on SO answers - in any huge project, there are good things and there are bad things, and in fact there are tons of bad things you can pick out if the project is large enough. But ignoring the fact that lots of developers successfully use it to exchange information and enrich their knowledge and claiming it's just "some random site" is just ignoring the reality. And claiming "no one should be promoting SO in open source community at all" because it's imperfect just begs the question - ok, what should one be promoting instead? What is the perfect site which is as good an information source as SO and yet has none of its downsides? If this is not specified - in the same place - it just sounds as empty snobbery. SO is a practical tool, it's imperfect, it has obvious and known flaws - but dismissing it wholesale out of hand is only appropriate if one has better tool that solves the same problem, and better. So far I haven't seen any suggestions for that.

When a site is as big as SO you can never fit all the people all of the time.

1. It's community run

2. they make their money through legitimate means of job ads which fits perfectly

3. the ads they do have are to community voted places eg Sheldon Brown from cycling SE

4. It's about as transparent as you can get

5. The people with the most points almost always really know their shit

6. It's one of the few places on the Internet you can make yourself marketable by being helpful. I can prove my knowledge to potential companies through my answers

Or let's just all go back to Experts Exchange.

It all depends on how you use it. If something I make produces an error, or unexpected output, the first thing I do after the usual "stop and think" phase is google it. More often than not, that search will give me at least half a page of Stackoverflow pages - many of them completely irrelevant to my specific problem of course - but looking at the code can jog something in my mind that I wouldn't have thought of myself. It helps to look at somewhat related code. At least in my opinion.

I've lost count of how many times I've googled a question, had the first hit be SO, and the top answer to the question be "Have you ever heard of Google, n00b"? (Thread locked).

I've lost track of the number of times that I've gone to a question that actually deals with what I'm looking for just to find that it's locked and marked duplicate of a completely different question. Complete pain in the ass.

This is the worst and I see it often enough to want to respond to your gripe with support!

Although I've encountered this as well, in most cases I got a useful answer for my problem.

I often ddg problems related to Ubuntu. I had many encounters where someone had the same problem than me on askubuntu, but the answers where unhelpful, sometimes only the question was posted with no answer at all. Or the original poster stated he solved his problem, without posting the solution. In most cases, on SO I found relief.

Yeah, the moderation there needs improvement. When ever I use it I almost have to spend as much time preempting invalid attempts to close my question as I do actually asking it.

Then there's also the conceit that SO has hit upon a definition of what a good technical question is (or, more specifically, what isn't a valuable technical question), and they stick to it despite any evidence to the contrary. I've seen questions that fail their over-restrictive criteria that have far better answers than many that do.

I've lost count of how many times I've googled a question, had the first hit be SO, and the top answer concisely explains how to solve my exact problem, often with several other alternatives in the other answers and a nuanced discussion about the tradeoffs between them.

Stack Overflow is like democracy. It sucks sometimes, but it's still better than everything else we've come up with so far.

Quite a few times I’ve googled something only to hit SO answer as the first match where the accepted answer is my own... Oh, right...

I'm not sure what you search for but that has never happened to me in about 7 years of searching.

You can always appeal to get a question re-opened. You can see who voted to close it and then send them a message.

The worst place for things getting closed is Skeptics SE, but that's a hotbed of controversy and the moderators on there deserve danger money.

For me, it's alway Jon Skeet or Stephen Cleary with a well-written, correct answer that highlights some gotcha that is incredibly non-obvious in the official documentation. Where these guys find the time, I don't know...

Unless it's something to do with Skype or Lync. Then it's abandon all hope, ye who enter here.

I was curious, and I did a search for "deleted:1 user:me" to see if any of my posts had been deleted, and six had, almost all of them I had deleted myself. For context I have posted around 200 posts on Stack Overflow.

Being able to see deleted posts is indeed an eye-opener, and I also wish more people had this ability, but for a different reason. It makes you aware of the amount of very low quality posts that get posted to Stack Overflow, not to mention spam.

Yes, I see candidates for deletion in review queue, and most of it is either garbage or "I have this complex problem, please solve it for me after I give you the most vague description of it". Like the one I just saw recently where somebody has a huge pile of legacy Perl code and has trouble figuring it out (and that's all, no specific Perl question). I can feel their pain, but what can be done here except giving the poor soul a hug? So this kind of questions gets closed. And probably the person asking it is very frustrated, but really, what can one do here?

What about hiring an actual Perl programmer?

I've been a Perl programmer once, and I assure you given a pile of legacy Perl code, written in a good old style with good old "everything accesses everything", no comments and clever use of tricky Perl constructs, most people would be inclined to yell for help. It's not specifically Perl's fault, but a constellation of Perl being at the top of its popularity when thinking about who will support it 20 years later didn't exactly enter the mindset of most programmers.

Main issues for me: 1)too eager closing of questions for whatever reason 2)number of answers for given question growing over time and best answers not getting the best position.

I concur. Main issue I have faced is people down voting questions even when the question is perfectly valid and reasonably well explained(imo). It sometimes looks to me that when a question is about 'under the cover' stuff and not superficial 'how do i get this done/i did this but got this output', it would be down voted.

I had this happen when I was obtaining my degree a couple of years ago. I forget the issue exactly, but I had this homework problem that I was able to solve using Python docs but what was happening underneath wasn't really "clicking" in my brain. So I posted the problem I solved, my solution, along with my actual question about the underlying issue. I got a lot of "did you check the documentation?" and "is this for HW?"...sadly my question was closed and I never got an answer.

What I find hilarious is when I post a very detailed technical android answer and then someone comes along and edits it to be incorrect. I wish there was a way to make your posts only viewable or deletable. The people who go around editing do a terrible job...

That's a sweeping geneealization. More often than noy, those nifty, pristine posts with corrext syntax and grammar are the work of editors who clean them up.

Reply on news.ycombinator.com