perl-embperl mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Gavin Carr <>
Subject Re: Some philosophical questions
Date Wed, 27 Jul 2005 03:22:07 GMT
On Tue, Jul 26, 2005 at 08:07:36AM -0700, Neil Gunton wrote:
> I think that people looking for the simpler, "beginner" solution turn to 
> PHP, whereas the heavier, "enterprise" users tend to go to Java. So maybe 
> mod_perl/Embperl is stuck somewhere in between in a kind of no mans land, 
> perceived neither as beginner nor enterprise level. That's a shame, since 
> we all know there really isn't anything that Java can do that 
> mod_perl/Embperl can't. But I don't want to be in the position of sour old 
> programmer crying into his beer about how he "coulda been a contender" but 
> backed the wrong horse. I'm increasingly getting nervous about basing my 
> programming experience on something that seems to not have the mindshare in 
> up and coming young programmers.
> Advocacy for tools like mod_perl and Embperl may be a way to go, but it 
> can't really replace "buzz" in the developer community. I think we lost a 
> lot of that "buzz" when we (as a developer community) took a massive time 
> out to develop the incompatible Versions 2 of Apache and mod_perl. People 
> kept hearing that it wasn't ready yet, and so they used something else. But 
> that's not the main reason, I think. PHP just gained mindshare because it 
> was so simple and cheerful. It's a lesson to geeks who think all that 
> matters is technical superiority. If we spent more time making our projects 
> easy to use and accessible to new users, rather than adding obscure, 
> complex new features that hardly anybody uses, then maybe our projects 
> would be more widely adopted.

Another issue with perl is the relative modularity of perl-based solutions,
which is an aspect of perl's TMTOWTDI mantra. This is both a strength and a
weakness. With PHP, while you can bolt on a number of different templating
engines, it's pretty capable for web stuff out of the box, and lots of
people/projects just use it like that. Perl isn't - you have to add in your
templating-system-of-choice (or CGI), sessions support, database-layer-of-
choice, etc. This also means that the 'buzz' around PHP is more localised, 
where the perl 'buzz' is fragmented across multiple projects - Embperl, 
HTML::Mason, Template::Toolkit, HTML::Template, etc. Choice is good, of 
course, but it does fragment the community and the evangelism.

I also think that Embperl has lost a fair amount of momentum with the 
length of time that Embperl2 has been in beta. This isn't intended as a
criticism of Gerald - I think it's mostly just some unfortunate timing
effects around Embperl2 being almost ready to do just as mod_perl2 got
seriously unstable. But whatever the reason, it seems like there's a
a lot more activity these days around the HTML::Mason and 
Template::Toolkit projects. You might also want to have a look at 
Catalyst ( sometime - there's a huge amount
of buzz around that these days as perl's answer to Ruby on Rails.

My 2c.


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message