RE: [PEAR-DEV] The Nature of PEAR's Greatness
| From: | Richard York | Date: | Fri, 09 Apr 2004 22:49:55 +0000 |
| Subject: | RE: [PEAR-DEV] The Nature of PEAR's Greatness | ||
| References: | 1 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-27350@lists.php.net to get a copy of this message | ||
Tell the list, what do you think is great about PEAR?
>* The installer is great.
Let me reiterate that I think the installer is an awesome thing,
just a little hard to catch onto. If you don't know it exists
in the first place, its horrible, if you do then its a good thing.
In fact very convienient. In fact now that I know there's a remote
installer its my new best friend.
>* Coding standards rock!
And a moral obligation. I spend a great deal of time fielding begginer
questions in various forums, so I can appreciate the coding standards,
after having to deal with sloppy code all of the time.
I just like the one true brace convention, which isn't a knock on the
current convention, just personal taste, I find the PEAR convention
hard to read. One isn't really any better than the other, but I should
be afforded that much flexibility since it doesn't really hurt anything.
I _can_ appreciate the no tab requirement, another issue I can see both
sides on.
>* Naming conventions rock!
Rock on.
>* I like the idea of a common error handling scheme.
Its not necessarily a bad thing, but one that's difficult to upgrade. If
everyone relies on the same error package then we're stuck with that
error package for a good long time. Which was a point brought up when
error_stack was in the proposal phase. As others have mentioned this
won't be a problem with PHP5 and I'm looking forward to seeing error_stack
as well.
>* The categorization of classes, while it might use some work, is very
>helpful.
God help us without organization.
>* The structure of classes (tests/, docs/, etc) is very useful
Its OK.
I don't think putting examples in a docs directory is very intuitive.
A package should exist in the same directory. Just call me an advocate
for the hopelessly stupid (certainly myself included from time to time).
>* The peer review and commentary on code has been helpful to me in
>particular (not necessarily the proposal process per se, but the
>ongoing back-and-forth of good practice)
Peer review helps us all to see with new eyes, this is a very positive
thing.
I hope that no one takes what I have to say as negative, opinions are
as assholes... as the saying goes.
There is a great deal of very useful code in PEAR, and a hell of a lot of
people donating their personal time and or money to see it grow and become
better. So PEAR _is_ a wonderful thing. So it needs a little work.. I
think it has a pretty good group of folks to see it get there.
Regards,
Richard York