|
|
Posted by Chris Shiflett on 06/25/05 18:46
Paul Waring wrote:
> I've used both PEAR and CPAN for a few years now and I've noticed that
> CPAN tends to win hands down in terms of documentation and updates. That
> might just be down to the particular packages I've happened to use but
> given a choice I know which one I'd rather use.
Yeah, you're basing that on which ones you've used. The interesting
thing about CPAN is that it has far more crap than PEAR. This seems to
work out well, because the best packages trickle up in terms of
reputation. For example, most Perl developers use Test::More to
implement their tests, but there is a lot of stuff in CPAN that does the
same thing (and outputs a TAP-compliant protocol), and many of them
existed before Test::More.
PEAR is much more guarded, and it has a higher quality to quantity
ratio. This has some advantages. Of course, it also has disadvantages -
there will always be complaints about PEAR being political (independent
of whether it actually is) by those whose packages don't get accepted.
Another problem is stagnant or poor packages that solve an important
problem. With CPAN, I can just write a better solution, and if it is
actually better, everyone starts using that. With PEAR, I need to try to
work with the original author, which might involve enough effort just to
make contact that I give up, and the better solution is never developed.
It's a risk.
Anyway, take what I say with a grain of salt - I have contributed no
CPAN or PEAR packages (yet). :-)
Chris
--
Chris Shiflett
Brain Bulb, The PHP Consultancy
http://brainbulb.com/
Navigation:
[Reply to this message]
|