That you don't know what community means is kind of sad, but it certainly explains the decisions that drove me (and others here) away from Stack Overflow.

Speaking as a Wikipedia admin, you're replicating a mistake that I think has harmed Wikipedia immensely without mirroring one of the key safety valves.

Wikipedia has a problem that's well-known in the community: active participants are burning out, editor activity is stagnant, and the existing social environment is very off-putting to newbies. Petty enforcement of petty rules, an obsession with particular notions of quality, and a lot of behavior that seems to the uninitiated like high-handed dickishness.

But one of Wikipedia's saving graces is this core policy: "If a rule prevents you from improving or maintaining Wikipedia, ignore it." It reminds people that rules are the tail, not the dog.

With Wikipedia, strict rules aren't what leads to a valuable artifact. it's that a lot of people care about creating the world's best reference work together. They understand very clearly that it's the community that produces the encyclopedia. The rules are just there to record community consensus, and to help newbies think things through.

Is the quality of the site being kept up, though?

Questions certainly feel "worse" than they did 2 years ago when I first started following SO.

Way more "homework" or "newbie south-east Asian programmer asking for THE CODEZ" type questions than there ever was back in the day.

I'm not really sure how you fix that, although slightly raising the barrier to entry (perhaps require they at least pick a username?) might help reduce the "drive bys" who ask a poor question, and never accept an answer.

All true, and far more real a risk. That's why we now (as of about 6 months ago) require registration to ask questions, for example.

SO is slowly driving away the more senior participants who haven't been drawn in by the gamification of the site. Every few months it seems like the questions on the site are more and more of the "I could have googled this but it was faster to ask it on SO." variety.

In short, SO is becoming a Mechanical Turk for searching for answers to software questions, and the reason is because Google is becoming less and less useful (as it tries to be more and more helpful) for searching for answers to software questions.

I don't know how that fits into the long-term goals of SO.

I think SO will be happy if they will be able to become a database for most common programming problems.

That is where the crowd is, mediocre programmers are plenty and they won't be asking expert questions either. They need hand holding at every step. And if you could answer all they question they have, you could serve a huge crowd.

On other hand, asking expert questions may make you an awesome site. But the target audience is to little.

Because number of people needing answers to expert questions themselves are scanty.

(blog author)

+1 aha this!

This is really the root cause perhaps?

This is why Joel and the chap whose own code is a horror are here talking quantity in reply to a blog post about quality perhaps?

> Like Wikipedia, strict rules are what lead to a valuable artifact.

What? Wikipedia's strict rules (really, a hideous mish mash of conflicting obscure scattered policy and guideline and suggestion and essay and convention) detract from the purpose of getting information about a subject from a variety of trusted sources into an article about that subject.

See, for example, the amount of time that goes into deciding whether to allow someone to have a username consisting of a bunch of digits or not. (Such a user name is "confusing". But what is it confused with? It's not confusing by itself. The policy is to prevent people using names that imply power. and so on for several megabytes, several times a year.) And then, when the community has taken several months (at a short time estimate) to agree on a policy they realise that the software has hard-coded limits which are different to the policy anyway. So it all starts again.

SO's mods are doing a good, but difficult job.

Wikipedia's admins, New Page Patrollers, Anti-Vandal Patrol, Recent Change Patrol, twinkle user, rollback user, are doing an inconsistent job.

I'll agree that I'm pretty anti the current WP experience.

"[StackOverflow's rules] are intended to create a body of precise questions and answers ..."

Suitable for informing a junior code monkey working to someone else's plan. Yes, we all understand this. Yes, we agree it is useful.

But junior code monkeys grow and learn. On their way to becoming a senior software engineer, they will ask a question like "how do I solve such and such problem with a Ruby hash" and the answer is "you probably don't want to--take the stakeholder to lunch and do some requirements analysis, then come back and we'll help you solve the systems architecture problem that you will have just discovered".

This answer will be expunged by the moderators for being subjective and argumentative. Yes, of course it is. The most valuable questions about software creation have to do with purpose and architecture. Software creators are people who need mentoring, not robots looking for answers from autistic oracles.

I have no objection to putting only objective questions on the standard view. It's expedient for people who need to learn some detail right now. It will get more eyeballs, more Google juice, and more monetization. But there is no need to delete subjective questions when they can be trivially tagged and filtered off into another view.

Actually, that answer would very likely be appreciated as long as you phrased it politely. I've very often told people on Stack Overflow that they were asking the wrong question, told them how to start off on the right path, and gotten the checkmark for it.

(I actually will sometimes answer the original question if it's not too involved or impossible, and include a note along the lines of "See this? This is why it sucks for what you want to do.")