Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 158 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

Let's consider a question I asked on the Stack Exchange site Travel as an example - Choosing a tour of Europe. I knew asking which tour to choose would be too subjective, so I tried asking about what to consider, but it was still marked as too general. I don't want to focus too much on this specific example, as I've seen this happen on many other Stack Exchange sites, but I think that Stack Exchange should handle these kinds of questions better.

If you read the answer I received before it was closed, you'll notice that it is possible to provide a good answer to this question. Rudy lists several points to consider including noting that tours with older people might slow the pace down too much and tours with children might result in an overly noisy bus. I think that it is a shame that we disallow questions like this that "teach a man to fish".

Here are some other kinds of questions that are very general, but for which it would be nice to have something to refer people to:

  • What should I consider when deciding which programming language to write my website in?
  • What should I consider when deciding which Python web framework should I use?
  • What should I consider when deciding which whether I go native or write a HTML 5 application for mobile phones?

I don't think it is reasonable to assume that these people are asking these questions because they were too lazy to use Google. These are inherently difficult questions, that aren't currently addressed well by either forums or Stack Exchange. What I am proposing is that we need a location (such as a wiki) to explain what the major tradeoffs are when making such a decision so that the person can decide for themselves, or ask a more specific question. For example, the programming language question, might have information on what language major websites use (and growth), major gotchas for major languages and an outline of framework availability. The scope of these question means that it would be better to have multiple people collaborating to answer the question using something like a wiki post.

Possible solutions

  1. Allow these questions on Stack Exchange - the main disadvantage is that this are "easy" questions to ask that can gain quite a large amount of reputation. Unfortunately, questions can't be marked as community wiki anymore
  2. Unofficial Wiki - official support would be much better, but this is a workable short term solution
  3. Blog posts - unfortunately, blog posts tend to be written once, difficult to collaborate on and rarely updated
  4. Questions on Meta - This is my preferred solution in the short term. I would advise filing these questions under a [topic-summary] tag. This would not be a free-for-all mechanism for getting any subjective or overly broad question into Stack Exchange. Rather, it would be reserved for providing general guides to topics that teach people how to figure answers out on their own or figure out which questions to ask. I know this stretches the definition of meta, but "what should I understand before I ask a question on this topic" is kind of metaish.
  5. Integrated wiki - The Wiki posts could show up the the "Questions with similar titles" and "Similar questions" section. The wiki posts could have tags and people should be able to upvote a Wiki page if they think that it is valuable. People could request the creation of a Wiki page and these requests could be closed or deleted like normal questions. Needless to say, the development effort would be rather significant.


I think that this would solve many of the biggest issues facing Stack Exchange.

  1. User acquisition - by answering these questions we could draw a lot of users in and by redirecting users to these answers, closing questions would be much less off-putting
  2. Too many non-reusable questions - if well written, these resources could answer many questions before they need to be asked
  3. Questions overly subjective - these resources could help people form more specific questions
  4. Reduced tensions between inclusionists and exclusionists - this would be a compromise between keeping Stack Exchange "pure" and loosening the restrictions


The original example used was "Which programming language should I use to build a website?"

share|improve this question
I think you and I have different definitions of what it means to "teach a man to fish". For example: the answer you got on Travel is not teaching you how to figure it out for yourself or "how to fish", it was just a random handout. It is a fish. – Wesley Murch Mar 18 '12 at 20:55

I know this is just a hypothetical question, but it's the one you gave :)

I'm building a website, what's the best programming language should I use?

This has already been discussed to death in dozens or hundreds of forums. The asker is here, asking the question, because they've not searched for it themselves, they've not done sufficient reading about the pros and cons of the dozens of languages that are applicable, and, worst sin of all, they've given no criteria by which one could judge "best". It might be "lets me have something running in a day", it might be "scales to millions of machines with nearly no effort", it might be the inherent safety of the language and interpreter, it might be the ease of acquiring maintenance programmers in a decade.

Part of being a high-quality programmer is knowing that even simple-looking problems need to juggle dozens or hundreds of criteria just like the handful I've given here, and picking a reasonable compromise. That takes experience -- which could be handily transferred via such open-ended questions as you've proposed -- but it does mean that one person's "best" won't come anywhere close to someone else's "best", even though they might ask the same question.

What that really calls for is a comparison table of options. This is something Wikipedia does very well. Those little red and green cells don't quite tell the whole story but I believe the information density in those tables far outpaces what we can reasonably provide here.

If a questioner comes up with a more specific question about implementation languages, I'm willing to entertain it -- perhaps more so than the norm here -- but it'd better be something more specific.

share|improve this answer
"The asker is here, asking the question, because they've not searched for it themselves" - that is not a reasonable assumption. When it comes to decisions like this, there are just some many variables that the beginner has no idea of where to begin. I know that they have given no criteria of what is best - the guide would help them understand the various tradeoffs so that they could write a better questions – Casebash Mar 18 '12 at 3:29
Wikipedia is not suitable for this. While Wikipedia has a semi-relevant page for this topic it wouldn't allow pages, for example, on improving performance in Java programs. Besides, just looking at that table wouldn't really help anyone wanting to pick a language to build a website in. The table would only be useful to people who already know quite a bit about various programming languages and so probably wouldn't need to ask this question – Casebash Mar 18 '12 at 3:35

"The asker is here, asking the question, because they've not searched for it themselves" - that is not a reasonable assumption. When it comes to decisions like this, there are just some many variables that the beginner has no idea of where to begin.

Then this beginner is not ready for Q&A yet.

It is important to understand the difference between Q&A and getting help. Q&A is about a specific, practical question. This question stands on its own, without any external context, without any knowledge of the experience level of the person asking it. The question has a specific, practical answer based more or less on objective facts and inferences made from them.

Some questions are a bit more subjective than others, but in general, most things that we here call "questions" have a single correct answer. The more possible valid answers it has, the less likely that it is an actual question.

Q&A is like communication via snail mail with a complete stranger. You write a letter, put it in a box, and a few days later, you receive a reply based solely on the contents of that letter.

Getting help is not about asking a question and getting an answer. It's about a process, where by the person seeking help engages in a dialog with one or more persons who attempt to provide help.

Getting help is what happens when you're on the phone with customer support. There's a lot of back and forth, where you provide a bit more info, they suggest a few things, you try them and immediately report back, etc.

Stack Exchange is not for engaging in dialog. It is not a place where you get "help". It's a place where you can drop a real question and get a real answer. Stack Exchange is, quite simply, not for a person who is a complete beginner at anything. This is a person who needs help, not Q&A. This is the wrong site for such a person.

And that's fine. No site needs to be everything for everyone.

share|improve this answer
"No site needs to be everything for everyone" - true, but I'm not suggesting this site be everything. Rather, I am suggesting a logical extension based on having used many different StackExchanges. This site could be so much more - I think the difference would be on the same level as StackExchange is compared to normal forums for Q&A – Casebash Mar 18 '12 at 8:33
@Casebash: But it's not a logical extension. SE doesn't want to be more than Q&A; SE is for Q&A, period. SE is what it is because it is less than normal forums, because it makes dialog and discussion (ie: "help") more or less impossible. – Nicol Bolas Mar 18 '12 at 8:55
@Nicole: I'm not suggesting free for all discussion be allowed - only a subset of questions that are valuable enough to deserve to exist despite not quite perfectly fitting (yet) – Casebash Mar 18 '12 at 8:57
@Casebash: My point is that those questions aren't valuable. They aren't good questions, even if they might produce good information. I'm not saying that information shouldn't exist. But it shouldn't exist here. You're trying to broaden a definition of "good question" to include subjective stuff that we deliberately banished. – Nicol Bolas Mar 18 '12 at 9:16
The way I remember SE back in the early days is as a site for professionals to share knowledge, which is fine. Once questions get too vague, we're essentially getting into consultancy, which is something people should be paying for. – Phil Lello Mar 18 '12 at 21:15

You must log in to answer this question.

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