This question is an exact duplicate of:
In the comments on this post, there is a discussion about how much research you should do before asking a question.
My standpoint is, I'm asking the question as soon as:
- I did some basic check whether this is trivially found on Google or already asked on StackOverflow.
- I am sure that the question will be valuable for others and is constructive (i.e. other people will have the same question/problem and any answers will help them).
- I feel comfortable and safe in my understanding of the subject as a whole.
- I have some background knowledge about the question.
- I am quite sure that the question is not a stupid one.
In any case, I would probably figure out the answer after spending enough time doing research. That is, of course, always the case for everybody - if you just spend enough time, you'll probably find the answer.
So, the question is, what is enough research.
From the comments, people suggest to me that it is bad to do as little research as I do before asking. I don't see why that is. The question will be valuable anyway (otherwise I would not ask it). And if I already figured it out myself (i.e. after doing own research), I will more likely not ask it anymore. The whole point is that there is maybe someone who can easily answer the question.
In most cases, after I spend some time researching it myself and after I asked the question, I continue to research it and to find out myself. Often, a few hours later or a few days later, I find out and if nobody else was able to answer in the meantime, I add the answer, so that it is helpful for other people.
To give some references, these are some questions I asked lately (which have generated the whole discussion because I got hit by the 6 questions/day limit):
- Autogenerate MSVC import library (LIB-file) from a DLL
- Crash when calling gd function
- Homebrew/apt-get/Portage or similar for Windows
- How to get MSVC compiler messages in English
- MSVC fails with compiler errors without compiling any sources
- MSVC and boost::lambda::bind error: T0: standard-argment not allowed
See also other questions I have asked in the past.
- Do you think they are valuable for the community?
- Under what reasoning would it have made sense to do more research before asking them?