In response to John Coggeshall's Apache 2.0, My Two Cents.

I have been a longtime user of both the Apache platform and the PHP scripting language. I follow the apache dev lists extensively and chip in from time to time on a number of things that I feel concern me.

Right now, I see a few advantages in Apache 2.0 which have provided the business impetus for me to make the change despite having only the prefork MPM availible for use with PHP.

1) LDAP Support. Yes, i suppose you could invest the time to implement the support neccessary to support basic authentication and so forth based on LDAP data sources but as someone stated earlier int he thread... why bother? Oh wait....

"HEY LOOK -- YOU CAN SAVE SOME MAJOR HEADACHES USING EXISTING IDENTITY DATA FROM LDAP DATASOURCES TO POWER YOUR APACHE BASIC AUTH MECHANISMS!"

Active directory and a number of generic email-related identity management systems come to mind for support... Specialized support and implementation? Yes. Small number of users affected? Probably Yes. Advantage of 2 over 1.3? Yes.

2) While arguably not really an enhancement to the average Apache/PHP/MySQL Developer out there, the availibility of the mod_python platform is a definite business driver in commercial production environments due to the ease of RAD with Python scripting.

3) There was recently some noise on the apache dev list about discontinuing support for Apache 1.3. While I strongly resounded against implementing an arbitrary End-Of-Life date, there were indeed many who did not share my position and felt that removing or de-emphasizing the level of continuing development to the 1.3 codebase would strengthen 2.0 and especially 2.1 efforts. Lets face it, as any product manager knows, having to do backport implementation for existing product codebases for each new security bug, et al, is a real pain in the rear. I dont think the 1.3 source tree should be arbitrarily made inaccessible or something stupid like that but at the same time, realize that whether by policy or no, there continues to be diminishing contribution to that branch in favor of adding widgets to the more robust hosting platform of Apache 2.

4) The Apache API changes have resulted in a greater level of feature accessibility and process transparency. This, in turn, has led to some great innovations (and applications) in modules for security, proxying, etc. Proxying changes in particular are significant because these changes have provided a business driver for some organizations to move to Apache 2 for use of advanced hosting modules which involve forwarded CMS content or architectures in which content is actually requested and provided by secured locations (secured-network-segment apps, et al).

I would also like to point out that the non-thread-safe nature of many of the PHP-associated libraries should really not be seen as a shortcoming of apache but rather a weakness in PHP that we (as a developer community) should really try to fix or help fix. This may be either fixing it by making direct contributions to up-stream projects or by providing community pressure to up-stream projects to get these bugs fiixed. Hold on...

"HEY LOOK -- If the PHP community can specifically identify non-thread-safe libraries and so forth that PHP interoperates with and uses, then apply presure to make the libraries thread-safe, PHP on the Apache 2 platform will provide a much more robust hosting platform for commercial organizations."

I think we can all agree that this would further both the PHP AND Apache project goals of building the community of users. I dare not make any claims about speed increases as that is largely dependant on configurations and loaded modules and any subsystems (E.G. LDAP or MySQL Query Times, etc).

In summary, I personally believe that the argument "Well, using apache 2 doesnt provide any compelling feature increases specifically for my PHP scripts so I wont help enhance my product for the new platform" is bull. Its like saying "well i dont care that my users see other requirements that are met by the new platform but since *I* dont get anything cool, im not working with it".

The debate of whether or not 2.0 is "better" than 1.3 is a moot point. The debate of whether or not 2.0 provides anything new or "better" to PHP is a moot point. The simple fact of the matter is that the PHP community has a large base of users out there, many of whom do and will benefit from new features, enhancements, and modules brought to the 2.x platform of apache and I think it very irresponsible of an otherwise excellent project to essentially abandon them.

Put your weight behind getting PHP and associated libraries thread-safe. Your community is the community at large... not just those who would see specific gains to strictly PHP applications.