Re: The future of PHP

Thomas Deliduka wrote:

A little background... Skip to "THE JIST" if you wanna make this quick. I am on this webmaster's list where most of the people are fairly new webmasters. They're just getting the hang of things. I am probably one of the only advanced ones on there (not to toot my own horn really) and I'm one of two who use PHP (the other is just learning.) I tout PHP and MySQL as awesome, which they are, but I don't say they're the only technology out there. There is another guy, a kid, who is a major Java guy and he's BIG into XML and JSP. Some guy on this list asked what a good dynamic web solution is, I immediately chimed in and said PHP was great, free, and on most all linux hosts. Told him how powerful it was and what could be done with it. This kid chimes in and says, something to the affect of that if the guy wants to live in the past and not let his website go anywhere then he should go with PHP but JSP is the wave of the future and it's more powerful, and has the backing of the almighty Sun and the Open Source community (as if PHP doesn't).

I may be offbase a bit, but I think straight servlet style is more 'wave of the future' than JSP. JSP was/is more of a marketing rebuttal to ASP than a solid approach to web page, in my experience. Knowing Java, JSP to servlets to whatver shouldn't be a huge leap, but I'm always wary of Java people who get really *excited* about JSP specifically. Again, my limited experience, it's cumbersome and slow. http://www.zdnet.com/enterprise/stories/linux/0,12249,2646052,00.html Revisit that - the JSP stuff was by far the slowest of the bunch. Yes, a better JVM would help, but would a different JVM give a 200% speed increase (what's needed to put it side by side with PHP in terms of speed?).

THE JIST So, the jist is, what does PHP have to offer to the web in the future? I think it's still a viable option that will be around for at least another 6-7 years. This kid thinks it's shelf life is another 3. What do ya'll think?

Personally, I'd like to see the Java integration in PHP get better (easier to set up, perhaps different architecture so it can run faster) so people wanting interaction with Java and PHP can integrate the two easier. That might help put these issues to rest more. XML? I'm pretty sure there's more XML stuff available for Java than PHP - by XML stuff I mean libraries and routines predone that'll help speed development. More SOAP stuff is happening in Java than PHP right now, so they're 'ahead' in that respect. At the end of the day, a language needs to be able to interact with a web server (or run as a standalone process on port 80!) to accept incoming data, process it, and send it back in some specific format (HTML, XML, etc.). Java can do this. PHP can do this. CF, ASP, Perl, Python, Ruby, TCL and others can do this. To some extent I engage in the same kind of 'turf wars' of PHP v ASP, JSP, CF, etc. I've spent most of my time developing in PHP, so I have a vested interest to promote it as a viable answer to client problems. Someone who's done the same with JSP will no doubt react as this person does. One of PHP's strengths is also a weakness, imo. Because there's no standard way of doing certain things, many ways to approach the same problem get developed - I'm thinking code routines, etc. With Sun and MS, although you can often come up with multiple ways of doing things, these companies (ime) propose a 'standard' way of doing things - even if its slow/cumbersome/inefficient, that's the way people will learn how to do stuff. Tangential case-in-point: the new ASP.NET stuff - the "web forms" stuff - put in a value, put new tags in the form to specifiy what's required, what's not and other validation routines, and the server will automatically fill in the values if it needs to redisplay that form. It's just a damn templating system built-in - but those practices will be learned by a whole new set of ASP people in the coming months and years who will espouse it as the best/awesome/cool way of doing stuff. You couldn't really do good templating systems at all in ASP up until recently, so most people over there never knew what they were missing anyway! We've developed our own internal templating system/logic flow that allows for this type of stuff to be developed about as quickly as the ASP.NET version (possibly a bit faster once you get the hang of it). But because it's not a 'standard' in the community, it will never be as widely adopted as the ASP templating system they are promoting. <rant off> So - what do I think about shelf-life? At least 3 years in its present form. I'd like to see PHP evolve to more of a standalone system to act as its own webserver - or have closer integration with Apache ala mod_perl. If either of those happen, imo, it'll be much longer than 3 years. If someone could write a PHP -> Java bytecode compiler, PHP code could run on any JVM, any the whole point would be moot anyway. :) The rebuttal to 3 years, of course, is "who knows what the hell will be happening in technology 3 years from now?". How many people in 1997 saw the internet as it ended up in 2000? ------------------------------ Michael Kimsal http://www.tapinternet.com/php/ PHP Training Courses 734-480-9961