Christian Dickmann wrote:
We don't have to prove, that your new package breaks BC,
you have to prove, that it does not!
Not to encourage further argument, but I feel that this is an unreasonable statement. Since there are no unit tests to verify, nobody can prove anything is BC or not. This makes patch-work, with or without a proper patch, and analysis of new contributions much more difficult, and more prone to error. In turn, this creates a barrier to entry for PEAR contributions, which is IMO a bad thing.
I would say that more importantly than we've held coding standards even, we should be holding two things: unit tests and documentation. I really hate it when I find a cool PEAR package that sounds like it does *exactly* what I need, then find out I have to read the entire source to figure out how to use it. And unfortunately at that point, I don't take any consolation in the fact that the code is at least "readable" according to PEAR standards. ;)
Lux
--
PEAR Development Mailing List (http://pear.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
--
John Luxford
President and Chief Developer
______________________________
SIMIAN systems
Driving Web Content Management
______________________________
web :
http://www.simian.ca/
email : lux@simian.ca
phone : 204.452.8537