OSDev.org

I have seen and understood for a long time (since I am around OS development forums, since 2004) that these forums are full of questions asked and answers, without taking into account the OS Wiki, which I haven't seen anywhere else; still, let's say it's rather incomplete compared to what one can find with a search engine, Wikipedia, eMule, etc.

And that's the point I'm going after, I see that asking something here that probably has been answered before anywhere else and is known by some, may not get a "Google It!" answer, yet the answers are very generic and won't show a definitive practical solution. That doesn't seem to happen in other ways of searching either in any case.

So what should the discussion in these forums should be used for, if it is very clear that most cases looking at Wikipedia or other websites for a very "simple" specific doubt would work better, just like that "Google It" suggests? A concrete example is looking for the Windows NT load process to have a better understanding of system loading and trying to use some of that knowledge.

If the answers are more easily solved in this way, then what should be asked or discussed to get actually helpful information in the forums, or maybe the point is that useful snippets arise from time to time even when they belong to other tasks of OS development?

I think this is a very interesting and good question.

I would say that part of the answer is that software design can be done an infinite number of ways. There is no right way. So anyone asking a software design question is 100% guaranteed to get a generic answer. The one time when that actually is helpful is when you don't have any idea how to code something like a "pipe", or whatever. You ask: what are some typical ways of coding up a 'pipe' system resource? You get some generic answers. You pick what sounds good and code it.

Hardware questions, on the other hand, have an exact and definite answer -- if someone in the forum happens to know it. I would say that these are the smartest questions to ask on the forums.

Then there are the "opinion" questions: "What is your favorite disassembler?" kind of stuff. You can google, and get 982,347 "hits" -- but you get no clue at all as to which ones are better than which other ones, unless you ask.

There are the "sage advice" questions: "What are all the possible functions that I might want to include in my VFS API?" stuff. Everyone knows the first obvious ones. But the most important ones are the ones you don't think of, at first.

And, last and least, there are the "help me find my bug" questions. They are no fun, but when you are at the end of your rope, what else can you do?

So, I would say that the answer lies in knowing what kind of question you are asking, and not doing a sloppy job of asking that question. If you ask clear questions, then the resulting discussions will be clear and helpful.

I consider the forums to be a location for human opinion and for review of your complex stuff. It can also be a showpiece, but I rather don't use it like that.