Hi Andreas
Thorsten Scherler wrote:
On Mon, 2009-03-09 at 14:07 +0100, Andreas Hartmann wrote:
[email protected] schrieb:
up to now, there wasn't much feedback on the 2.2 discussions (apart from
Felix' comments). Is anyone else planning to actively participate in the
migration to Cocoon 2.2?
If not, I'd suggest to stop the migration and revert the SVN (use the
2.0.2 branch as trunk again).
...
Not sure, I had a look at the revisions and it would be a pity to lose
this work.
Can you give us a quick update of the status of the migration, any
problems or blockers and what needs to be done to get Lenya working
again? I didn't want to interfere with you while you were doing the
bulk of migration but if there is a need I can pick up some tasks on
the todo list.
Would it make sence to skip the migration to 2.2 and wait until Cocoon 3 is
available ?
IMO this makes sense. Cocoon 2.2 has some nice concepts, but if feels a
bit like a hybrid between Avalon and Spring, and not like the holy grail
yet. Maybe C3 will be more streamlined and pruned of legacy stuff.
I share this opinion. I see good concepts but it is not yet the best
practice approach. Further the development of 2.2 is a wee bit slower
since Grzegorz (who took the lead in the dev) stop working on 2.2.
Cocoon 3 is MAYBE the best choice, but IMO it makes sense to keep the
trunk based on 2.2. up for a while. It would be very interesting to see
how the performance has developed with using cocoon 2.2.
IMO Cocoon 3 is far from being stable and recommendable for lenya.
Cocoon 2.2 is THE big step for Lenya moving away from
Avalon to Spring and make use of the Maven build process. I assume
that the move from Cocoon 2.2 to Cocoon 3.0 won't be that much
work if our code is based already on the Spring Framework and Maven.
Alternatively we can develop useful modules and publications to demonstrate
the power of lenya to the public.
That would be very important in any case.
The power of cocoon 2.2 lies in the blocks and spring configuration. In
your other mail you asked whether it makes sense to drop modules in
favor of the block concept and IMO it does.
+1
salu2
- Felix
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
