> The key is the get better information out there and
> centralization at a wiki level is only one way to accomplish this

and

> possibly not the best if we agree that the only user we care about

is the

> GoogleBot. No one goes to a website to get information, they just

ask

> Google. Increasingly people are realizing that their only real

customer is

> Google and this applies even more so to open source projects.

If GoogleBot is your only customer, then it doesn't matter how you
structure the information, just don't put text in images and don't ask
for login to reveal info. For this, a static website, wiki, ml, forum,
online book, etc. is just as good or as bad.

So my problem is exactly the opposite: I'm tired to google around when
I should just go to a single web page and find it all there.

I'll quickly copy-paste a paragraph from my first message, to re-
argument this point:

<<So for example, when I want to know about "global temporary tables",
I go to the section with the same name and I get: documentation,
supported versions, changelog, implementation notes, user comments,
examples, support questions and answers, related bugs, architectural
debate, cross-links to "related" sections etc. In the current state of
affairs, I'd call that "research".>>