Re: PEAR: The next biggest thing??? Maybe...

On Sun, Jul 15, 2001 at 04:43:17AM -0700, Alan T. Miller wrote: > There seems to be an issue of whether application frameworks are appropriate > or not, I don't see why they should not be. PEAR after does stand for > "extension and add-on repository." Might not an add-on be construed as more > than a straightforward class? I really do not understand why, if someone > wants to develop an application framework that builds on the solid PEAR > base, what the harm would be to add a section, or area for them. I mean we > are here trying to create a common code base, but we do not want to include > anything that actually uses that code base? That does not make much sense. > It seems to me, that something that used the PEAR framework, would make a > great addition to the pear project, and in turn would attract a whole hell > of a lot of interest in PEAR, this is turn would probably create a lot more > interest in contributing to the PEAR core base code. I'll repeat my thoughts on the subject once more so that they don't get missed. I don't think application frameworks belong in PEAR any more than full PHP applications. I do think it's appropriate for individual _components_ of those frameworks to be added. For example, assuming an application framework had classes for authentication, online help, usage statistics, and image manipulation, we might be a PEAR hierarchy like this: pear/MyFramework/ Auth/ Help/ Stats/ Images/ In the case of Binary Cloud (which, again, I don't know well at all), we would be able to import just the Binary Cloud authentication class. I don't think it's unreasonable for a PHP application to rely on Binary Cloud's authentication interface any more than it would rely on PEAR's authentication later (assuming there was one). I hope I'm getting my point across here: packagable components, not large blocks of code. -- Jon Parise (jon@csh.rit.edu) . Rochester Inst. of Technology http://www.csh.rit.edu/~jon/ : Computer Science House Member