PEAR & code repositories (was: php and MS SQL Server....)

I think the way PEAR has been categorized as a place where generic, stable, well-tested code can migrate over time is good -- accessing PHP code that has some level of "editorial approval", as opposed to random code that you've found on a web site, can be a bonus. I pretty much trust anything I find in CPAN as reliable code (maybe I shouldn't :) and I think it's a win for PHP to have something like that as well. However, to keep up that high level of trust in and reliance on PEAR code, things that get committed to PEAR need to be discussed on the mailing list, reviewed, etc. to keep the quality up. That said, there's still IMHO a good reason to have code repositories (like PX) where other kinds of code can live. Code of different levels of maturity, appealing to different audience, code by someone who doesn't (yet) feel comfortable dealing with CVS, etc. Plus, I was thinking of having the PEAR code be web-accessible (read-only, obviously) through the PX anyway. :) -dave