How long will the Seam project be active before people focus is placed on WebBeans?

[Daniel Hinojosa]

Is Seam's life short?

[Jason Porter]

Seam will continue to be developed after Web Beans. Seam 3 will use the Web Beans RI internally, but Seam, I'm sure, will continue to have a very long life after the RI for Web Beans is complete.

[Dan Allen]

This is a great question and I'm glad you asked it. It's important to understand that Web Beans is not a replacement for Seam. The best way to think about Web Beans is as a kernel, the future kernel of Seam.

Currently, Seam bootstraps its own "container" where it manages references to component instances (objects) in various scopes. I quote container because underneath the covers, it is simply using the Servlet API scopes, plus the jBPM business process scope to store the objects. It then wires instances together at runtime using a method interceptor that reflects on fields annotated with @In and injects component instances by name.

Fast forward a year or two and we get to Seam 3, where likely the Web Beans kernel will be the dominant core. Web Beans uses typesafe injection built on a feature known as "a binding annotation". It's a revolutionary way to approach dependency injection. While powerful, it is going to take a while for people to understand. Thus, the current Seam container will be formfit to it to support both styles of programming.

In short, the answer is "no", Seam isn't going anywhere. Add to that the fact that it is wrapped up as a product of JBoss with long-term support and you can be sure you can develop Seam 2 applications for 5 years or more. Personally, while I like the Web Beans concept (it's really brilliant) there is no way I am jumping ship on the Seam 2 core given how productive I am with it already.

All the existing integrations of Seam and the ones to come, especially the Facelets-style markup, you can enjoy even in a Web Beans world.

[Dan Allen]

Btw, read here for more info on Web Beans.

Web Beans overview

[Jignesh Patel]

I am confused here, Whatever I read for Web Beans(offcourse not too detailed) I found Web Beans is more sophistacated version of Web Beans and it is driven by the same people who involved in creation of Seam.

[Hendy Setyo Mulyo]

Originally posted by Dan Allen:
This is a great question and I'm glad you asked it. It's important to understand that Web Beans is not a replacement for Seam. The best way to think about Web Beans is as a kernel, the future kernel of Seam.

Hi Dan,

You mentioned that Web Beans is a kernel, or would be the future kernel of Seam. So, is there any possibility that there will be another Application Framework, just like Seam, which will be using this "Web Beans kernel" in the code? If there is, then it would be very interesting to have Seam competitor in the future .

[Dan Allen]

I am confused here, Whatever I read for Web Beans(offcourse not too detailed) I found Web Beans is more sophistacated version of Web Beans and it is driven by the same people who involved in creation of Seam.

It is true that Web Beans was proposed by Gavin King, the creator/visionary of Seam and is now being implemented by the Seam developers (likely to be the reference implementation). It's mechanism of dependency injection is more sophisticated than what Seam has, but also more difficult to understand for beginners. But of course, Seam goes well beyond dependency injection.

[Dan Allen]

There will most certainly be other implementations of Web Beans, and they may provide the core of something completely different than Seam.

[Valerie Griffin]

Whether or not Seam is the hottest WebBeans framework three years from now, it's a great way to be a relatively early adopter of the WebBeans technology (which I believe will be a significant development framework), and to start getting experience now.

No, it's not perfect, but nothing ever is. By being part of the early community, you also have more chance to influence its development.

[Dan Allen]

Hey Valerie, glad to see you over here during the promo.

I agree that paying attention to Web Beans is great advice. I read through a very recent version of the spec and frankly I think that Gavin et al have discovered an incredibly innovative way to use Java in the midst of many people claiming the language is too old to learn new tricks.

I am very suprised by the commitment to type-safety, because it has been one of those gray areas where the Java community has gone back and forth. It makes sense that if you are going to have a type-safe language, you might as well take advantage of the guarantees it provides. This is where Web Beans shines.

If you are interested, try out the Web Beans implementation the Seam developers are putting together. I'm sure they will welcome feedback (but don't give too much right away since they are still filling in large gaps).