Re: future of PEAR (was [PEAR-DEV] [Call For Votes]PHP::Fork)
| From: | Daniel Khan | Date: | Tue, 11 Nov 2003 00:26:17 +0000 |
| Subject: | Re: future of PEAR (was [PEAR-DEV] [Call For Votes]PHP::Fork) | ||
| References: | 1 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-23408@lists.php.net to get a copy of this message | ||
Hello,
having followed the recent discussion I want to share some things I have in mind.
I am also not that sure if a PEAR 2 would be a good idea.
Yes - there are much things that could be better in the current infrastructure, concurrent versions are the most important IMHO.
I also often tend to throw whole projects into the trashbin after reviewing them after some time and finding out that things can be done in a better way.
BUT if we do a PEAR 2 we may end up with a semi mature PEAR 1 and a non mature PEAR 2 doing endless discussions on how to do PEAR 2 and loosing time for making any of the branches perfect.
Especially as someone has to do this work and time is limited.
Someone spoke about the Apache httpd project. Well the first steps towards Apache 2 were made in 1996 - so they did a bunch of planning while actively developing the 1.x branch.
ATM there is so much we could do on PEAR 1 and on how the community works.
1. Yes the package approval should be automated to get rid of these philosophic discussions about if a package makes sense or not.
There have to be clear rules and asking the community should be the last step if no decission could be made.
2. Let's reanimate PEAR QA. Let's write a todo list for pear QA. Let's do the things on this todo list ;)
3. Finally - let's create a todo list for the current PEAR (don''t know if this allready exists anywhere)
So people not following each and every discussion or even new people can add ideas and/or contribute work.
It would also shorten endless discussions about things that were allready discussed.
We should simply see all the current problems at one piece and if we then find out that the current infrastructure doesn't meet our needs we can discuss on how a PEAR 2 should look like.
Stopping in the middle to start new isn't the best way I think.
greetings
--
Daniel Khan