Why do they all answer their own private wannabe version of the question? How do we fix it?

Q: How do I make A work with B?

A: Why don't you use C instead of B?

Comment: Why don't you try to answer the question? If you have no idea, just shut up.

Maybe people shouldn't be allowed to answer questions too quickly. When they hit 'post' they should be drawn back to the question, the browser would hang and they would be forced to stare at the question for 30 secs again.

I think having stronger reputation loss for downvotes, as discussed elsewhere, might help.

I think too many people just shoot answers looking for easy reputation, backed by the strong unbalance between up- and downvotes.

  • I hate these too, but only when they obviously didn't read the question vs. read it and actually have a good reason for suggesting another option (and yes, there are some users who make a habit of doing this). I'm all about making downvotes send a stronger message, but... more than anything... Just down-vote diligently; some folks actually do pay attention to them.
    – Shog9
    Jul 31, 2010 at 18:03
  • 15
    How do I parse XML with regular expressions?
    – Welbog
    Jul 31, 2010 at 18:05
  • 11
    @Welbog: you use jQuery!
    – Shog9
    Jul 31, 2010 at 18:06
  • 3

5 Answers 5


Because C may conceivably be the better option. Many people asking questions on SO demonstrate what is known as the Dunning Kruger effect. The reverse is true as well, as many people answering questions often undervalue their skill. You must permit the possibility that you are trying to solve the wrong problem, and someone with more experience is likely to notice that quickly. Surely you have heard the expression:

"When the only tool you have is a hammer, everything starts to look like a nail."

I'm going to give you the best possible answer that I can, which sometimes may be "I don't think you're ever going to get the results you want with B, have you considered using C?"

Telling people to shut up is never a good idea, especially when they are simply trying to be helpful.


You are also free to ignore the "Just drop that and use JQuery" crowd at your discretion. Note, nothing against JQuery. It's just an example of how people might look at two points that could possibly work together, and instead try to introduce a whole other solution. Again, that solution may be better; it just doesn't immediately fix your urgent problem. You aren't the only person who will benefit from the answer; anyone else searching for your problem will likely find SO. The technically best answer should, naturally, rise to the top.

Still, getting upset over it isn't going to help anything. You have a down vote and the ability to not engage that user. If you find yourself upset while using SO, there is a very good chance you are taking it too personally.

Messing with the weight of down votes punishes a whole lot of people. A down vote to me is "Something is wrong with this answer, let me go see what it could be". Should I lose 30 rep for forgetting a semicolon in a code snippet in an otherwise great answer?

It's better to just reward great answers, which the system already does rather well. And, again, refraining from telling people to shut up :)

  • 5
    I understand this thinking, but I still don't like the practice. Tell people to use C in a comment if you want, but limit the answers to actually answering the question. I can't count the number of times I've searched for a problem on Google, found a link to a forum where someone asked that exact question, only to find the 20 "answers" people gave were all "you don't want to do that, in your particular example using X would be better" where X completely doesn't apply in my case Jul 31, 2010 at 18:03
  • 1
    Agreed. But that must definitely happen in the minority of cases. One would think people have a nice thought about the problem, narrow it down to the core and THEN ASK on SO. So, if I ask, I really want an answer to what I ask. Really I think most people just shoot answers looking for reputation. That's the REAL effect we're seeing, not Freddy Krueger.
    – Mau
    Jul 31, 2010 at 18:06
  • 3
    @Mau: I think you'll find you're in the minority. Most of the time people don't know enough about what they're asking to ask the right thing. It's not always true, but it's true most of the time.
    – Welbog
    Jul 31, 2010 at 18:10
  • @Matthew P. Welboggingue: Most of the time this is true, but I also had quite a few cases where I specifically asked not to mention C or D because I absolutely knew I wanted B and people still answered with C,D or E. This behaviour is really annoying especially since it draws away people looking at your answer because it already has been "answered".
    – RickyA
    Jan 4, 2013 at 11:42
  • I think each and every person on SO should come to the table with the understanding that there is someone more knowledgeable out there. If you ask a question, prepare to be edited or corrected. If you really don't know the answer to the question, someone else probably does so don't give the SO community garbage.
    – demongolem
    Aug 21, 2013 at 15:41

Most people asking questions on SO frankly do not know what they are doing. For example, you get a lot of C++ questions about how to use character pointers, malloc, strcpy etc. It does the people asking these questions no favours to answer them directly - the proper answer is "don't do that - use std::string".

  • 1
    Obvious +1 from me. We aren't here to answer peoples questions, per se, but solve their problem. Too often people ask how to solve a step instead of solve a problem; we just choose to solve the problem.
    – GManNickG
    Jul 31, 2010 at 18:57
  • +1; in the Perl world we call this an XY Problem (perlmonks.org/?node=XY%20Problem)
    – Ether
    Aug 1, 2010 at 2:18
  • This implies that no employer has ever mandated the use of malloc and strcpy over the use of std::string and new (in b4 someone takes this literally), and that there are never terrible codebases already in existence and in need of maintenance. Sometimes (read: often) it simply isn't possible to do things the Right way. A decent answer would ATFQ, but add a footnote explaining why it is Bad and Wrong to do things that way. This isn't an approach that requires a great deal of thought or imagination to adopt, just an actual desire to help.
    – user234461
    May 31, 2018 at 10:48

If people continuously answer with obvious solutions that don't work for you, try pointing out why they are not what you are looking for.
We can't read minds and thus figure out that your problem is a bit more convoluted than the question tells us.


Because it's hard to discern people who really need to use B instead of C from those who just don't know about C.

I remember a relatively recent question on iPhone development. The guy was using sockets to retrieve data over HTTP, and was asking for a way to interrupt the execution of a function he doesn't have control over after 60 seconds if nothing happens. People told him he should use NSURLRequest as it does HTTP, and has a timeout parameter. That's what he did and it worked for him.

You can't say every time someone wants to do something weird that they really need to do it this way. I believe it's your responsibility as a question asker to let other people know what else you tried and what didn't work, as usually people want an answer to a problem as a whole, not necessarily to a single step of a problem.


I agree that, in general, it is safe to assume a SO questioner doesn't know how to solve the problem they're offering a (partial) solution for, but it is also bothering when the questioner has clearly (IMHO) simplified the question to emphasise a point about which they want an answer and even then people offer alternatives :-(

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