This was in my last letter I sent (sorry about the muti-threading, I
just got on a kick and can't seem to stop), I think you're looking at a
separate project through incubator which would, eventually, move into
MyFaces once it's ready.
The nice thing about incubator is that it's built around community
building which is important for starting projects. The community will
be smaller, so developing consensus will be easier. And when you
finally hit Apache main-stream, you'll have your community already there
to hit the ground running. There is extra buerocracy involved, but I
think it would vastly benefit a project of this size. I know that
Trinidad benefited greatly from its incubator process.
I'm still somewhat new to Apache myself (only working with it for about
three years), so others may have more informed opinions.. :)
Scott
Michael Concini wrote:
Whether it is done by updating one of the existing projects to 2.0
first and then enhancing it with the additional functions or by
starting from scratch, I think that it is a good idea to get one solid
component set for JSF 2.0 that would consolidate the requirements that
are currently addressed by all three.
On a somewhat related question, at some point in the future,
potentially with any 2.0 component set(s), would it be best to move
the component sets into a separate project instead of as subprojects
of the MyFaces core? It might make sense to maintain one project for
the spec driven development done in the core and one for the widget
libraries.
I'd also like to chime in my agreement with Curtiss' that it would be
a good idea to start looking at creating a new branch soon for JSF 2.0
core development now that the draft spec is nearing completion. I
would certainly be interesting in devoting my time to the project as
well.
Mike
Scott O'Bryan wrote:
I would still echo sentiments that it would be most helpful to start
from an existing project. There are so many issues and requirements
that the existing renderkits have had years to work out, I think it
would be a much better starting point. Encouraging people to move
off of their existing renderkit is the only way to get developer
support for OpenSource projects and the only way to do that is to
address all the requirements from all three projects.
Scott
Curtiss Howard wrote:
Jesse Alexander (KSFH 323) wrote:
I am wondering whether the event of JSF 2.0 would not be a good
moment to create a new component set.
I'd like to chime in here with my +1.
I imagine maintaining three separate-but-similar component sets is
quite a bit of work and, from what I can tell with the JSF 2.0,
there will be enough major differences in the programming model to
make backwards compatibility quite a chore (as Jesse noted).
Merging the three sets would be difficult, so then why not try for a
new set that takes all the best parts and is built "from the ground
up" to leverage the JSF 2.0 programming model and concepts? It
would not preclude Tomahawk, Tobago et al from moving to JSF 2.0 if
that is the choice, but at the same time it would provide a fresh,
unified JSF 2.0 component set that isn't hamstrung by backwards
compatibility concerns and could move in its own direction if need be.
On a related note, what is the status of MyFaces and JSF 2.0? Is
there any interest yet in creating a branch and "skeletoning" an
implementation as the draft is released/updated? This is an area
I'm interested in pursuing and would offer some effort if the
community is interested.
Curtiss Howard
