On Wed, Aug 22, 2001 at 12:39:12PM +0100, Andy Wardley wrote:
> Nathan Torkington, High Priest of the Slippery Lighter, wrote:
> > Here are my reasons why the idea of using PHP on big projects scares
> > me:
>
> Forgive me for slipping into marketing buzzwords, but I think
> the most succint description of PHP's downfall is that it
> doesn't support an "n-tier" model. It doesn't support 3-tier
> and it doesn't even support 2-tier.
>
> [...]
Forgive me if I don't fall into the trap of "Why I don't like X
(where X != Perl)". :-)
The web is a *big* place. No one tool/paradigm/approach/technique
will solve all problems. So, while n-tier is preferable to a
certain class of problems (perhaps the vast majority of problems
to be solved on the web), there are legitimate (and frequently
non-technical) reasons to choose 1-tier approaches.
Perrin Harkins has a story on perl.com comparing the myriad templating
systems available on CPAN. It's a very delicate subject, and I
think he handled it very well: each has it's niche and which one
you use depends on a variety of constraints, including scalability
vs. performance vs. learnability vs. complexity.
http://www.perl.com/pub/a/2001/08/21/templating.html
I have no opinion for or against any aspect of PHP. However, it
is undeniable that PHP is doing something well - it is a simple,
robust, portable system for developing 1-tier web applications[*],
and there is only one of it. That makes it much simpler to focus on,
adopt, support, and write about.
As long as there are 27 different 1-tier web toolkits on CPAN,
they will mostly get lost in the noise while PHP usage increases.
(Ditto for Cold Fusion and ASP to some degree).
Z.
*: Like it or not, 1-tier applications are going to be written.
Deal with it as you like. Writing axkit/tt2 to support n-tier
apps is probably the most productive way to deal with it. :-)
