Re: Forking PEAR

Stephan Schmidt wrote:

versions. Surely Greg could have been implementing channels a lot faster, if he did not care about keeping the installer compatible with pre-channel versions.

This is true. I could do a whole lotta things faster if I didn't care about quality or whether users could actually use it. I want users to type $ pear upgrade PEAR and have a working installer without any need to know how it works under the hood. I have broken BC within the installer itself in a few places (VERY few) because to my knowledge nobody extends it, but everybody uses it. Yes, I know this is a no-no, but in the case of the PEAR package, anyone who has extended it could not have done this just for a custom job for a client, and extending the installer is a huge task currently. After 1.4.0, this will not be true if I have anything to say about it, but there will be a transition time. This is not for Stephan, but all of you pining about me not being able to work fast enough because of messages I post to pear-dev or other theories, I would like to offer a reality check: I may appear to be a machine sometimes, but I really am human. Sometimes I just need a break from something I'm working on. I will listen to music, play chess, watch TV, code, or (gasp) spend time with my girlfriend. I'm never going to work monomaniacally on the installer because among other things, after a short while, I start to make really stupid coding decisions that take 15 hours to debug, rather than sleeping on an idea and coding a great solution in 5 minutes. So don't worry - nothing is preventing me from finishing but the fact that there isn't anyone other than me working on channels code and my own desire to keep my sanity intact :). Greg