[Bryar-devel] future direction of this thing

[Justin D. <ju...@de...>]

Cool, there might be some movement on this now;)

Bryar 2.6 leaves a lot to be desired.  For instance, actually
working out-of-the-box would be a big improvement I think (I was
using the CGI frontend - calendar bug bit me).

When I was looking for a blog package my criteria were:
-in perl
-using TT or at least have a way to implement the view in TT
-elegant implementation
-easy to install, cutstomize, and extend

Bryar looked like the only one that fit the bill.  Unfortunately it
took quite a few hours to get it working satisfactorily.  Even now
its still not as nice as I had wanted.  There is really no reason
that Bryar can't be all I was looking for and much more.

It looks like Feuilleton has some nice ideas (Plugin API, ACLs, SPAM
proofing, etc...).  It also looks like Simon has already taken a
hack at this.

http://feuilleton.simon-cozens.org/
http://cvs.simon-cozens.org/viewcvs.cgi/feuilleton/

I wonder what we can glean from his attempt...

The problem with Feuilleton is that it increases the complexity and
dependancies in order to achieve its goals.  I'm not a fan of
requiring even more modules than Bryar does now.  Actually, I think
that Bryar should be made to have the ability to work with a very
minimalist set of modules.  This way it will be possible to run in
restrictive environments as well as have a "minimalist" method of
use for those that simply want it.  Complexity could then be scaled
to the desired/required amount as a choice.

Having a minimalist ability will allow Bryar to better compete with
something like Bloxsom.  And the ability to scale things up as you
need them would allow it to compete with things like MT.

In short I'd like to see Bryar do the following:
-clean up the current API
-add a robust plugin API
-allow for a minimalist install

Any thoughts?

I'd be willing to help out in any way I can.

Justin DeVuyst