Re: [PHP4BETA] Re: [PHP-DEV] Re: [PHP4BETA] PHP 4.0.0 Releasepackaged

On Mon, 22 May 2000, Manuel Lemos wrote:
> Right, it doesn't have to be the actual programmer to document the
> contributed material.  I think the way it could work is that new features
> are only accepted when somebody enters the relevant documentation. That
> could be written by someone else, but until it is done the feature would
> not ship.

That's a bit naive. Here's a scenario, because I know you like them:

1. Programmer A write a super cool PHP extension X.
2. Programmer A passes off the extension to documenter B with some notes
   and words of support.
3. Programmer A goes on vacation.
4. Documenter B spends a day writing docs, but he is not even half done.
   Documenter B either gets bored, has other stuff to do, or - goes on
   vacation.
5. Due to unfortunate leaks people are now aware of extension X and are
   screaming bloody murder because they want to use extension X _now_.

My take on this is that it's fine to put in new features or extensions.
Sure, programmers should make a proper effort to get them documented - I
did that for most of my functions. But most of the functions I add, for
example, start getting used right away. If I waited until I have time to
document all of them, it would detract from the PHP.

-Andrei

"Claiming Java is easier than C++ is like
 saying that K2 is shorter than Everest." 
 -- Larry O'Brien (editor, Software Development)