begin  quoting David Brown as of Sun, Mar 16, 2008 at 09:46:55PM -0700:
> On Sun, Mar 16, 2008 at 08:39:53PM -0700, Darren New wrote:
> 
> >Granted, it doesn't have to be bad. But in my experience, any library that 
> >gives you *only* auto-extracted documentation has suckalicious 
> >documentation. Anyone doing serious development for useful libraries does 
> >the documentation first, and something winds up without a good place to 
> >put it in the code.
> 
> I don't think it is possible to write meaningful documentation for
> non-trivial and non-redundant code before the code has been written.  We
> like to maintain an illusion that things can work this way, but they don't.

Oh, I dunno. The less trivial and redundant the code, the easier it
is to write documentation for.

Trivial code is hard to write meaningful documentation for.

I also wouldn't worry about writing the documentation for more than
one a method or class.  Writing out the documentation for the whole
project before you write any code? I agree -- not useful.

> You can document the library this way the second time you write it,
> perhaps.

Well, if your first solution was a spike solution to see if you *could*
do it, yah. :)

-- 
Can we do it? I dunno. Let's try it and find out.
Stewart Stremler

-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg

Reply via email to