Re: Fw: [PHP-DEV] Function caching...

From: Date: Sat, 09 Dec 2000 19:39:35 +0000
Subject: Re: Fw: [PHP-DEV] Function caching...
References: 1  Groups: php.dev 
Request: Send a blank email to php-dev+get-40686@lists.php.net to get a copy of this message
Jon Tai wrote: > I'm sure the day will come when I'll need a > "full-fledged-load-balancing-clustering-whatever-app-server", and I'm very > relieved to hear that someone (lots of someones) are working on writing one. > However, I'd doubt that when it's all done, it will be called "PHP". Again: For a lot of people I know of, this day has been three months ago. They are not writing single threaded preforked application servers for fun. They do this because they have to stick to a crude hack since PHP does not offer a better scaling and more elegant solution as a native feature. And sure the resulting thing will need to run unmodified or only minimally modified PHP scripts or else it would be of little use to migrating users. Thus it should be called PHP. Ideally this would only be some build option to PHP, and for the low end users nothing will change at all except for a larger source download (the binary will be unchanged, and thus have the same size). Only if you enabled the --enable-application-server option and selected --with-cgi-stub, you'd get an application server binary and a CGI stub program in order to ship parameters and results. If you'd selected --with-apxs-stub, you'd get a mod_php_stub.so instead for apache, running the communication stub program as an Apache module. Apache 2.0 is only a partial solution for multiple reasons (like: not available, no release date, not even (obvious) documentation on http://www.apache.de and you commit yourself to a single webserver - an advantage of PHP always has been relative webserver independence). Anyway, this discussion leads nowhere. Zeev pretty much made clear that Zend is not interested into enterprise solutions, but into a preferably objectless web-BASIC for writing guestbooks. I agree with him insofar that simple things should be simple. But in my opinion complex things still should be possible. They should not be banned from a language or an architecture just because they are nontrivial. Kristian -- Kristian Köhntopp, NetUSE AG Siemenswall, D-24107 Kiel Tel: +49 431 386 436 00, Fax: +49 431 386 435 99 Using PHP3? See our web development library at http://phplib.netuse.de/

« previous php.dev (#40686) next »