[alexAME]

There should be a real difference between good questions and questions which could be solved by having a look at the docs.

People are just too lazy, SO became a replacement for many references, docs and trivial things.

At the current state, it's not possible to judge a developer by the SO points, we have to check the profile, browse through a few given answers and eventually create an opinion based on the quality of the answers.

It's simply because they might have answered a question called "How do I fill an array with 4 elements". Thousand of people are lazy and voted that question up and also the answer.

What we have now, is a guy who got a ton of points by asking a stupid question and a guy who answered a trivial stupid question and got 120 up votes on that.

Why should we change that?

SO is a great huge site, developers can show what they have to offer, you can set up resume and you can search for jobs over SO.

SO just establish itself as an important tool for recruiter and companies looking for top notch developers but the points are misleading, someone who has answered lots of trivial questions isn't necessarily a good developer because of those trivial answers to trivial questions.

On the other hand, we got many people with below the 3k mark, who just answered really hard questions with 3-4 up votes.

What I'm afraid of is that someone with more points gets preferred in a decision between two developers from SO for a job position.

[rprospero]

I see and understand your point about reading quantity over quality. On the other hand, I also get paranoid about reflexively telling people to rtfm.

As my anecdote, I was trying to learn a new framework last month and team into what should have been a trivial problem. I spent days scavenging through the docs and finding nothing. Finally, I gave up and went to stackoverflow, where some other poor soul had asked my question before and been given -10 for their troubles.

The question had several comments about hours the site has deteriorated with questions from people too lazy to read the docs. Thankfully, there was one answer that explained the solution (the framework authors had overloaded a non-intuitive operator to handle this functionality) and a link to the homepage of the documentation with an admonition to read that before posting to stackoverflow.

In the month since, I still haven't encountered a single reference to that behaviour in the docs, despite referring to then quite regularly.

[brudgers]

I've listened to pretty much all the back catalog of Stackoverflow podcasts. From the early days the big idea was to surface information. The answer to a question may be in the documentation, but MSDN, for example, is a really big site and the dependency graph to understand any particular page can be enormous. Reading the documentation is a long term solution, but it is not always a useful one in the near and medium term.

On the other hand, there's always been a tension over how simple a question is appropriate. Spolsky and Atwood fell on different ends of the spectrum...See Spolsky's "How do I move a turtle in LOGO?" [1].

[1] http://stackoverflow.com/questions/1003841/how-do-i-move-the...

[hasenj]

> People are just too lazy, SO became a replacement for many references, docs and trivial things.

That was its original purpose. It wasn't meant to be a place for answers to detailed questions. It was designed with the aim of being the first Google result for things like "How do I collapse border tables?". Trivial questions where the answer is buried deep inside technical documentation.

[fratlas]

This. So much easier than trawling through terribly structured docs. I find the CLI version even more useful for this purpose.