1

I was pretty happy with Stack Overflow when it started, however I have seen the noise to signal ratio growing over time. At this point the only responses I get to my questions are from people who are obviously trying to get points rather than answer my question. I am sad to say that I no longer turn to Stack Overflow first. Furthermore I will likely not be turning here at all going forward. I am confident that you have a very popular site, but I no longer consider it a useful one.

Responses to comments below. Of the 15 questions I have asked I have received 3 solid answers and answered one myself. So I am running about 1 for 3.

It is not just answers to my personal questions, but answers in general. I find that I must make my way through many useless answers to other questions that I have found here that interest me.

Yes there are few Git experts here, unfortunately there are few Git experts period. I have asked the questions asked here on the Git forums as well. Aside from having my head bit off I have received little positive information. I appear to be the only person who finds Git not ready for prime time.

Examples. First, a good example: Can a Windows Forms control have a Design Time-only property?

Here are several that I am disappointed by:

I guess what I am seeing as I go through these, is that I choose not to spend the time that it takes to get an answer here. So perhaps this is not issues with stack overflow, so much as my issues with stack overflow. The first couple of questions I asked resulted in answers within a couple of days with little effort on my part. The rest have resulted in no answer, or answers that I had already considered. Maybe stack overflow is just not that useful to me. Still the "see community wiki" answers and others in a similar vein are an obvious attempt by some to gain reputation points without doing the work. I do not have an answer for this issue. I understand the role that reputation plays here and agree that it is an adequate solution. I have changed my mind in that I will not avoid stack overflow in the future, but it will move from being one of my first stops.

Improve this question
16
  • 14
    15 questions, 10 with an accepted answer, 1 with no answer (yet), and you answered 18 questions. I'd say things worked well for you! Commented Feb 1, 2011 at 14:16
  • 2
    "At this point the only responses I get to my questions are from people who are obviously trying to get points rather than answer my question." - Those two are not mutually exclusive. I can't say I agree that there is a decline in the signal to noise ratio. Commented Feb 1, 2011 at 14:31
  • 3
    Welcome to Meta, Pat! Sorry it's not under better conditions. If you were happy when you started, and you've only got 15 total questions, how many of them have been bad? Six? Seven? I'm sure that's frustrating, but it's not a huge sample size. You could just be hitting a patch of bad luck. Overall, people seem to be pretty happy with the site, so... stick with it a bit longer? Posting costs nothing -- well, two minutes of typing -- so you might as well try SO in addition to Google or whatever else. Now that all that's said... you haven't asked a question here, so I'm voting to close this post. Commented Feb 1, 2011 at 15:26
  • 3
    The last few questions to which you haven't received a (good?) answer were about git. Looks like there aren't enough experts on git on stackoverflow to answer your questions. Commented Feb 1, 2011 at 17:22
  • 9
    can you provide specific examples? Commented Feb 1, 2011 at 18:50
  • 4
    I actually replied to one of your questions and you gave no comment what-so-ever that my answer was crappy - I'm actually quite embarrassed and offended to see this post on meta: stackoverflow.com/questions/783790/… Commented Feb 3, 2011 at 15:57
  • So why didn't you accept Matt Cook's correct answer about the web sockets? Commented Feb 3, 2011 at 15:59
  • @Arjan - It requires a proxy server, so it is essentially a javascript interface to a telnet client on a separate server. It's a good approach to the general problem, but could hardly be called a javascript telnet library. Commented Feb 3, 2011 at 16:06
  • Ah, I see, @Pollyanna. Then I agree with chibacity and others: commenting to the answers might have been very nice (as the answer did point the OP to web sockets, and the proxy server itself is written in JavaScript too?). Commented Feb 3, 2011 at 16:14
  • @Arjan - The proxy server cannot be written in javascript, because javascript does not support raw sockets. Websockets have additional overhead that cannot be removed. Commented Feb 3, 2011 at 17:17
  • Ah, thanks again, @Pollyanna, now I am learning, rather than the OP! ;-) Ah, I now see that the proxy server is an extension to JavaScript. (I think; not my cup of tea.) Commented Feb 3, 2011 at 17:22
  • 1
    @Pat: you apparently also need to learn how to format your posts. I found several hard to read until I properly formatted them. Commented Feb 3, 2011 at 19:33
  • Clearly you are correct and I am not up to the challenge of using Stackoverflow. Thanks for clearing that up for me :-) Commented Feb 4, 2011 at 14:01
  • Signing off. Not sure what else to say, but have to use 30 characters. Commented Feb 4, 2011 at 14:51
  • 6
    Not sure what to say either, seeing as you have not seen it necessary to really respond to any of the feedback you've been given here. If you accept inferior answers in order to "be a good neighbour", how do you expect that question to yield any good new answers? Maybe you really aren't up to the challenge of using Stack Overflow. Commented Feb 4, 2011 at 14:51

6 Answers 6

Reset to default
15

I have to say I disagree. Sure, you get the odd people who don't give very useful answers but I've had considered responses to all of the questions that I've asked, and I find that people really wanting to get points are what usually causes them to provide a thoughtful and good answer (or at least try.)

Can you give some examples of questions that you consider you've just had "noise" to as a response instead of a decent, well thought out answer? I'd be interested to see (that's a genuine request, not sarcastic in the slightest - online text is awful for portraying things sometimes!)

Improve this answer
8

Have you bumped the questions to see if they get a better response with a different set of eyes? Have you considered adding bounties to these questions to get more attention?

I notice that of the examples you posted, 50% have accepted answers. Of the ones where you haven't marked an answer, you've only asked follow up questions to one of them. The other two have upvoted answers but you don't do any follow up to tell them why their answers won't fill your need and ask what other information they have.

I understand your frustration, however it honestly appears that you are expecting magic answers to pop out of stackoverflow with no additional effort beyond posting the original question.

They all received some response, but you haven't taken the time to cultivate your question by bumping them, responding to answerers that aren't helpful, and modifying the question to make it more clear why the given answers are not meeting your needs.

To more directly address the three questions where you haven't accepted an answer:

  • There may not be a tool that does an SDK check as you'd like. Sometimes there is no answer for a question. That doesn't mean there's a problem with the signal to noise ratio. I'm very curious to see your response to the one answer that was upvoted, though - it seems useful to me, but without you commenting on it no one can tell why it's inadequate.
  • Your assertion is incorrect - websockets do NOT implement a plain socket connection, they require special server side support to understand their pseudo-HTTP behavior. I've posted an answer to this question.
  • Your git problem is unique, and you have not provided enough information for someone to replicate the problem. I've posted a few suggestions for troubleshooting techniques, but unless it's a common problem (and it's not, in this case) then you won't find someone who knows the answer off the top of their head.

But in general your questions might receive a better response if you communicate with those that do provide answers, update your question as you discover new information, and bump it occasionally or add a bounty.

Improve this answer
3
  • 1
    I have accepted the answers in large part as an attempt to be a "good neighbor". Clearly as was pointed out in the comments above I am not qualified to use Stackoverflow. Commented Feb 4, 2011 at 14:02
  • pollyanna, how do you legally bump a thread? I had assumed that bumping threads by posting an answer to your own question was against the rules. Commented Feb 5, 2011 at 6:58
  • 1
    @Michael You should bump the question by editing it. While you should re-read the question and try to make substantial changes so that it will be easier to answer, if you honestly believe the question is perfect then you can make simple edits to force a bump. Commented Feb 5, 2011 at 14:39
6

I actually replied to:

I am looking for a tool like LibCheck to evaluate different versions of the same SDK

I have to say that my jaw just hit the floor when I see you moaning about it on Meta!

If you have a problem with answers then comment on them. You have made no indication what-so-ever in your question that you are unhappy with my answer.

I am almost speechless.

Improve this answer
1
  • 1
    I did not intend to point out individual examples, I appoligize if I have offended. I appreciate your time and effort in answering the question. Commented Feb 4, 2011 at 14:05
3

Yes, I think there is a problem with the signal-to-noise ratio, but in the opposite direction of what you're stating. There are a lot of low-quality, lazy questions entering the system as Stack Overflow becomes a more popular Google destination for people searching for easy solutions. This concern is addressed in this question here, and a filter was introduced for those who repeatedly ask bad questions, but the problem still exists.

In my observations here, the number of people capable of providing quality answers has only increased over the last two years. For example, I was struggling with this issue for days, finally decided to ask a question about it, and received the exact answer I needed within six hours. This was a localized, highly technical query that I doubt I would have gotten an answer for a year or two ago (the user who answered it has only been a member for six months).

Poor answers tend to be voted down, so people get discouraged from placing them. Non-answers are now being removed at a higher rate due to the new /review page. Spam also seems to be under control.

Low-quality questions seem to be the last area that could be cleaned up further. Yes, being inundated with these may cause some experts to be driven away from the site (like has happened in many mailing lists and forums before this), but it seems to me that this is not happening yet. People still love to answer interesting, well-written questions.

Improve this answer
2

If an answer is added to your question that is a wild-guess or does not meet your requirements then leave a comment and/or down vote it. If you don't see activity at first, just be patient (not every user is alway online at once).

If you have a question that you think someone will know the answer to but you think it just needs more visibility then add a bounty to it.

Improve this answer
0

+1 to Pat O for this comment ("Clearly as was pointed out in the comments above I am not qualified to use Stackoverflow"), not really the original complaint.

Regarding the original complaint, it's unfortunate that the community couldn't help you, but not everyone is an expert at everything. I would try to answer your questions if I felt I could give remotely decent answers, but I am unable to do so. No one is more saddened by that than myself. For those that you feel are merely attempting to pad their reputation, you can always downvote them... though I would suggest you first leave them a comment asking them to clarify how their response answers the question.

This comment, however, it is spot-on. StackExchange is a challenge for new users. (See my previous meta post.) The owners/moderators might not like me saying this, but the only reasons this site is popular are (1) high visibility in google results, and (2) some of the people here are extremely knowledgeable. This site is not popular because it is easy to navigate or easy to use. Yes, there are a lot of great features here, but there are a few that are poorly done or not done at all. I think those that run this site overlook the fact that when new users come here for help (via google results, mainly, would be my guess), they are already concentrated on something else and are not predisposed to drop what they're doing to spend hours learning what, in some respects, is a hard-to-use system. It's basically the same as when you've hunted down a program that you need in order to accomplish some main objective, and then you don't stop to read the 30-page EULA (unless, I suppose, you're Richard Stallman.)

Improve this answer

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.