On Tue, 5 Dec 2000, Stas Bekman wrote:
> But, you all know that php pretty much takes over. Why? For two reasons:
> 1) initial corporate pushing (press/ads)
> 2) once well known, the word of the mouth does the rest.
oh, there's also the part about php being so much easier to
setup and to program ;)
if you really feel the need to compete with php in the
lowest tier web app space, you need to make simplicity your
#1 goal. php is awesome entry level technology, and i almost
always recommend it over perl to people who only have the
desire to do casual programming for personal sites and small
projects. and that's a significant percentage of the people
i know doing web programming.
i'd love to see us evolve mod_perl with simplicity and ease
of use in mind. but precisely because php exists, i think
there's a more important goal for us: creating
infrastructure that can get perl accepted by the pointy
heads in the engineering shops and corporate it depts.
competing with java, rather than php.
java's got 2 things perl doesn't: a huge, well funded
marketing machine, and a large set of standard apis *that
were designed to work together* which have almost universal
programmer acceptance. i know it's heresy, but while
tmtowtdi is a cute slogan, it doesn't market us well outside
the circle of perl programmers.
how many of you have tried to hire engineers with
significant mod_perl experience? it's almost impossible,
even in san francisco which is supposed to be the wellspring
of geekery. 5 out of 10 perl programmers i interview have
not used mod_perl; 4 of the rest have only ever used
registry; and the last guy has an equal shot at having
experience with embperl, mason or his own homegrown
presentation layer. there is almost no talent pool with a
consistent skill set. makes it really difficult as a manager
to want to continue with a mod_perl-based application when
senior engineers who love java are more or less a dime a
dozen.
we need to create standards for ourselves and fully embrace
them. DBI is a great example. the explosion of mod_perl
presentation technologies is a great counter example.
we need to achieve significant vendor support, in the
development tools area, the enterprise software area, the
content management area, etc. how many of you use rational
rose for analysis & design? how many of you would use it for
code maintenance as well if it could round trip engineer
perl?
a premise of the perl philosophy is that it's fun to write
code. yes, it is. but it's also fun to make money. and
writing innovation applications that people pay for is much
different than writing (or often re-inventing)
infrastructure components. how many of you are out there
trying to re-invent cisco products? i bet zero. you buy
them, unwrap them and turn them on. infrastructure software
should be the same way. only a few people will ever make
money on threads and databases, but many many more will make
money on email messages, calendar events, news articles, and
item sales.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
