Gentlemen, a call to arms!
Matt S Trout
dbix-class at trout.me.uk
Tue Oct 17 08:46:26 BST 2006
Simon Wilcox wrote:
> On Mon, 16 Oct 2006, Jonathan Rockway wrote:
>>> These are all very well but when they break you're stuffed.
>> Yup. The advantage of Catalyst is that you can get rid of the CRUD
>> framework and write things properly. With other "frameworks", you're
>> stuck with what they give you, and you end up bending your app around
>> the framework. Catalyst stays out of the way and lets you handle things
>> (like a ... catalyst). That's not quite as "fun" as watching rails or
>> jifty make your forms for you, but it's what most people need in the end.
> My problem is, or was, that these frameworks were touted as "get going
> quick" solutions. Once installed and not working it was nigh on impossible
> to work out where the breakage was occuring.
> My experience, admittedly a long 10 months ago, was that these CRUD
> frameworks were a distration and a dead end. I should revisit them now
> that the whole landscape has matured a bit.
I've always found CRUD frameworks to be such; if you think Catalyst is one
then you've entirely misunderstood the nature of the project - one of its main
points is it doesn't ship with a DB library at all, there are users using it
for pure fs backend stuff, LDAP, SVN etc. (and at least two search engines
that I know of).
I certainly don't tout it as a "get going quick" solution so much as "here are
a bunch of the wheels you always need for a web-app nicely backaged and ready
to do things".
>> Catalyst is a great platform for building other frameworks; that's why
>> we have Enzyme, InstantCRUD, Reaction, etc. (I am even brewing up a new
>> framework in the back of my mind, but I'm waiting to see how the others
>> pan out before I actually write it.)
> I don't want to write frameworks, I want to write web applications and
> right here is the key, the single biggest difference between the rails
> camp and the catalyst camp. Rails is an emergent framework being evolved
> through the development of real world applications. Catalyst, although
> there are obviously sites being built with it, has the feel of a project
> that exists as a solution in search of a problem.
I think the original development of Catalyst was very much that; these days
however, it's absolutely being evolved through the development of real-world
applications - all of the core team members have Catalyst sites that they work
with on a day-to-day basis (in my case, given we do a lot of consultancy and
contract dev it's often not the same site, but the point still stands :).
I ran into the project a while back looking for something entirely different,
realised what it was and fell in love. That I'm now involved in development as
a direct result of using it successfully in the wild and that the vast
majority of our active contributors are the same should say something.
> I'm no rails fan-boy, much of it is crufty and solution specific, much
> like php actually and a by-product of the way it's being developed, but
> they are being successful with it.
> I don't think Catalyst wants to be successful in that way but that's a
> very perlish attitude and we should probably commend it for that :-)
Not particularly, we're trying to build a good substrate atop which that sort
of thing and many others can be built; most of the rails and PHP defectors
I've seen have moved to Catalyst not because it's a better one of what they're
already using but because they've outgrown the capabilities of what they're
currently using's One True Way and want to move to a situation where they can
use Whatever Way Works For Their Requirements - and *that* is a perlish
Matt S Trout Offering custom development, consultancy and support
Technical Director contracts for Catalyst, DBIx::Class and BAST. Contact
Shadowcat Systems Ltd. mst (at) shadowcatsystems.co.uk for more information
+ Help us build a better perl ORM: http://dbix-class.shadowcatsystems.co.uk/ +
More information about the london.pm