Seam2, Seam 3 or JavaEE6

[kem]

Hi,

I have already developed different project with seam2 and jboss in the last 18 months. I have to start a 3rd similar project this month. But I am really lost between seam2, 3 and javaEE6: do I have to continue with seam 2-jboss or switch to JavaEE6-glassfish or wait for seam 3...
Also, javaEE6 share a lot of features with seam and is very well supported in NetBeans and other IDEs. So what seam 3, Forge... could bring NOW that we already don't have with JavaEE6 and a good IDE?
Any feedback, suggestion... is very welcome.
Thanks

khalil

[lvdberg]

Hi,

it all depends on the way you want to handle risks in your project. Seam-3 is still in the lab and not really ripe for production.At the same time, to make Seam-3 work , you need additional technology such as for example the JSF-2 comnponents (Richfaces/Primefaces or something like that) which also just left the lab. This means that you will be faced with bugs which have to be solved by the community or by yourself.

If you have a very tight deadline and don't need additional shiny functionality and don't want to cope with a lot of problems, stich to the Seam-2 environment.

Leo   

[kem]

Thanks Leo,

what about javaEE6 without seam2 ?
Does it make sense to use seam2 while many of its features are now part of javaEE6?

Regards,
khalil

[lvdberg]

Hi,

as mentioned in other threads, we like to play safe for projects with tight schedules and a high risk of failure. We consider Seam 2 proven, stable technology and prefer it in combination with AS6.

We are already looking at specific parts of Seam3, but it still misses an additional evolutionary step for us.

We expect to use Seam3/JBoss6 for the first time autumn this year.

Leo

[kem]

Hi Leo,

have you managed to switch to seam3/jboss6.

Do you still think it's better to continue with seam2 for small project?

khalil

[lvdberg]

Hi Khalil,

We are in the middle of the learning phase. Always complex because you have your regular projects and you need to switch to something new at the same time. Possibly we will have to migrate a very big project to Seam 3 somewhere at the end of the year. A big problem we have there is that the existing software contains jBPM and we really have the feeling that it will be dropped in favour of Drools. But we will surely succeed in migrating everything.

Your question about a small project is a bit difficult to answer, because it depends on the requirements. It all depends on the quality/availibility you need to deliver. Weld/Seam3 components is still at its start phase and you will surely encounter bugs, but you really don't want to find them in e deployed application. On the other hand, if you and your customer have the time , go for it, because it is really a huge step forward.

Leo