1:47 p.m.
On Sun, Mar 20, 2011 at 12:18, Denis Forveille <denis.forveille(a)gmail.com>wrote:
Dan,
I will give a look at Arquilian and find what is needed
Turns out that we have a Websphere 7 container already working to some
degree. See: http://community.jboss.org/message/559002
We are also in the process of splitting out the containers into their own
repositories to make them simpler to build.
, and yes I would like to see Seam3 run as smooth as possible on WAS
v8.0.
The intersting thing is that WAS v8 has a different implementation stack of
implementation of standards than the traditional ones used by Seam 3
(OpenJPA, CDI impl based on OpenWebBean, JSF 2.0 based on MyFaces etc.).
I'll try as much as possible to have the
That shouldn't matter. The whole idea of Seam 3 is that it's portable (as I
describe here: http://in.relation.to/Bloggers/IsSeam3GoingToBePortableOrWhat).
Of course, we will leverage vendor extensions as extra features, but that
doesn't get in the way of using a module. In fact, in the JCR module, we
have developers from two vendors working on it, which is an ideal case to
me.
Before being comfortable with Seam3, the new things I have to master (like
many other Seam 2 users I guess) include DI, CDI, JEE 6, Seam 3 etc in
addition to maven and git. The step is quite high (or should I say the
stair..lol) for an "old school" user and contributor to Seam 2, familiar
with svn and ant. It can take a long time to assimilate all of that before
being comfortable using and contributing to Seam 3 and play with Arquilian.
IMHO this may be an obstacle for Seam 3 adoption (but not the fault of Seam
3 but of the new way to develop JEE apps )
It's exactly the revolutionary approach Seam 2 took by boldly embracing Java
5, EE 5 and the many integrated technologies. We take these large leaps
forward so that we can embrace the inevitable change.
But we can also leverage a lot of the existing knowledge because the
programming model retains the same contextual, state-oriented philosophy.
The forthcoming migration guide will make that pretty clear.
But that's not to say we are bound to that philosophy either. Several of the
Seam modules are exploring completely new approaches, and that's great.
My first priority is to have the seam-booking app starting and running on
WebSphere v8.0 beta and educate my self on the technology.
Great. We'd definitely like to have seam-booking validated on WebSphere.
Then we can feel even more confident in our portability.
And if I can express some opinion on the rush to Seam3 final.I think that
many people that will look at Seam 3 will look first at the documentation,.
then will try the booking sample app and finally will check the wiki and he
forums
There will be only one Seam 3 Final release in the life os Seam 3.. Many
people are working very hard to reach this goal and we can not miss this
opportunity to show how Seam is a great framework to the world. It is better
IMHO to have the more shiny Seam v3.0 distribution as possible than to rush
it and let people wait for a v3.1 even if it is obviously not possible to
get everything perfect for v3.0.
For this, IMHO it is very very important for people to easelyt understand
Seam 3 and that the booking sample app just build and run "as-is" without
any tweaks and that the code is very "clean" and demonstrate the power of
Seam 3 (and CDI JSF 2 etc). Even if it is stated that the booking app is
target to GF 3+ and JBoss App 6+, it should be "as neutral" as possible for
people using other ASS (Oracle, WebSphere) to be able to taste it too...
About the WiKi, IMHO it is due for a minimal "cleaning" before Seam v3.0
It should be clear of what is related to Seam 2 and what is related to Seam
3. Seam 2 is still there for a long time as many people will not be able to
quickly switch to a JEE6 App server in production. I suggest to start from
bottom to top with the link on the left bar of the WiKI and check page per
page. Also on the left bar I would group the pages differently :
- Seam 3
- Documentation (Include a brief description of the role each module
here. Add the missing ones)
- Download
- Tooling
- Development
- Contribute
- Report bugs (new page, maybe a list here with all the links to the
JIRA for each modules, and a link to the SEAM3 JIRA for "global" bugs and
maybe a way to have on 1 screen all the bugs relative to a release of Seam 3
whatever the module. I don't know if it is possible with JIRA)
- <anything related to dev,. maybe architecture/design or so docs>
- Module Handbboks
- Seam 3 Modules
- Solder
- Catch Module
- <other modules>
Also the left menu "Community/contribute and JIRA" should be merge under
the Seam 2 and under the Seam 3 sections with the pertinent information for
each version
I hope the release of Seam 3 will be a great success. The goal of this mail
was to give my opinion and to contribute to this goal. I hope nobody will
take it as an offence to they great and hard work on Seam 3.
Denis, you've given us many things to think about, focus on and consider. I
applaud community members that come forward with ideas to make Seam a better
stack. We are listening.
With that said, we've decided that releasing Seam 3.0.0.Final now is
critical for it to continue to succeed. As I mentioned in my blog entry,
it's not an end, it's a beginning. This is a way to say that Seam 3 is a
real project, that we've demonstrated that the modular strategy can be
pulled together into a stack and that developers can begin prototyping
applications with it. Obviously, there is plenty of room for enhancements,
polishing and new features. We'll listen to the community and from that
feedback, determine how best to proceed with releases (whether to focus on
3.0.x releases or quickly advance to 3.1).
Hope to see ya in #seam-dev.
-Dan
--
Dan Allen
Principal Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://www.google.com/profiles/dan.j.allen#about
http://mojavelinux.com
http://mojavelinux.com/seaminaction