JBoss.org Podcasts
Gavin King
Seam Project Lead
Mark Newton interviews Gavin King about Seam Framework, Web Beans (JSR-299) and a variety of other topics.
Recorded Thursday, 10th May 2007
Published Tuesday, 18th of September, 2007
YouTube: Part 1, Part 2 | MPEG-4 (Video) | MP3 (Audio Only)
Podcast Transcript
| Mark | I'm Mark Newton, I'm a member of the JBoss.org team. I'm here at JavaOne 2007 with Gavin King. Hi. |
|---|---|
| Gavin | Hey |
| Mark | We're going to talk a bit about Seam, what the future holds, and some of the new information that's around on the new book.
Gavin, you said in your blog at the end of March how Seam has had similar download numbers to Hibernate. What do you attribute such good adoption rates to? |
| Gavin |
I think that people today are really out there looking for a whole
solution to the problem of application development. I think that
they're kind of sick of the days of going out and looking for ten
different frameworks to solve each of their narrow technical problems
and then trying to figure out how to wire them together, and how to
use them together. They want to take it up to the next level and
have something that solves the whole problem of building an application. I think the first steps in that direction were actually taken in some scripting language environments like Ruby, but I think there's a lot of people who are super keen on getting that same kind of productivity in Java. There's some great advantages to working in a type-safe language that supports all the wonderful tooling that Java has. The JVM is really a great execution environment. A wonderful execution environment, in fact. And the enterprise services which are available in Java EE are unmatched anywhere else. I think that people want the kind of pre-built environment and they want it in Java. There is relatively few projects which are trying to go there. There's a number of projects trying to go there. Clearly Seam is one of them. |
| Mark | Excellent. At the moment, there's a lot of noise around Seam being used with web application development, but can you also use Seam for different kinds of Java applications other than web applications? |
| Gavin |
So, today it's all about web applications, currently, but we've
been flagging for months that it's more than that. The fundamentals
of Seam, the contextual component model, state management, the
integration with things like Drools,
jBPM, the fundamentals of the
programming model are applicable beyond a web application. We are
currently working on integration with JBoss ESB,
JBoss WS. We've been
promising to have that our for months, and it's taken longer that
we'd have liked to get there, but it'll get there eventually.
We have heard from people who are currently integration Seam with GWT, with Adobe Flex. We're very keen to see the results of that work. At some stage, we'll also turn to something like Swing, Eclipse RCP, or some environment like that. |
| Mark | Excellent. With the new split with the Red Hat business model now so we have Seam as a framework and it's an integral part of the JBoss Enterprise Application Platform, how's it fit in with that? What benefits does it provide? |
| Gavin |
One thing I know Norman's been working on very hard of
late is getting Seam working on the latest version of
JBoss AS that's going to be rolled into the Enterprise
Application Platform for the first release. This will be
the first time that JBoss has done a release of a platform
that is "as tested". In the past, if you wanted to use
jBPM and Seam and Hibernate on the application server you
had to go out and download those individually and it was
kind of a bit for you to take the risk that they were all
going to actually work together. In practice, things tended
to work pretty well together since things were designed with
each other in mind.
What we're doing with the app platform is taking the stack of technologies through an extra level of QA that we never had before. From my point of view, as lead of the development team, my job is to keep pushing community releases, pushing out the latest features, to a user-base looking for the latest and greatest. For people with Seam applications in production today, I'm not imagining this group of people, quite a number of people have come up to us over the last couple of days and talked about their applications they've had in production for 6 months or a year, and so for these people, these people need a QA'd environment for deployment and so the app platform really provides that. It frees us, to a certain extent, from being so concerned with issues like backwards compatibility across previous versions of app servers, previous versions of Hibernate. We can know that the productization team is going to be taking care of all of that. |
| Mark | Sure, so there's multiple integrations points that'll be taken care of. So we have this new book out by Michael Yuan, on JBoss Seam, and I think we've run out of copies both in the booth at JavaOne, for various reasons.. |
| Gavin | Yeah, so the word is that Michael Yuan's Seam book was the most stolen book at JavaOne. We know of at least 8 stolen copies. But also the bookshop sold out. |
| Mark | Which is good news. |
| Gavin | Which is fantastic. It was one of the top 10 selling books. That's really great to see that happening, and congratulations to Michael. That's great news. And there's another book out on A!Press, and I spoken to a couple of people who are working on books for Manning and I forget the other one. But there's a raft of books coming out which is really excellent. That means when people come to me and say "hey, well, how do I get started with Seam?", I can point to this book. "Go look at that book." |
| Mark | Excellent. I love books. It's a great way for me to learn about it.
Now you're also heavily involved in the Web Beans JSR spec. What can you tell us about that? |
| Gavin |
The spec's taken a few months to get off the ground, but it's definitely off the ground now. We've had quite a few con-calls, and good participation from Sun, and Google, and Oracle, and especially some of the individual members are making great contributions. People like Jacob Hookom, Richard Kennard.
What's most interesting that's happened is that Bob Lee and I have discovered a great deal in common, and I've been spending a lot of time looking at Guice, and Bob's been spending some time learning about Seam. We've each discovered things in each other's model that is very interesting. So what we've been working on, is how to build a model that has the contextual, state-management of Seam, and also the nice things about the Guice model, of its type-safe approaching to wiring components together, which Seam doesn't have. We're emerging with a model which looks different to both of them, but hopefully has the good features of both. And that's pretty exciting and interesting. It's always fun at the beginning of these, when you go out there and you build a system from the ground up, and you make all kind of decisions along the way. And sometimes, some decision you made, you don't necessarily see the full extent of its impact at the time you make it. So it's kind of nice to be able to go back and start questioning all of your assumptions again, and revisit everything with a group of people who don't necessarily share your prejudices, and see "was that really the right way to go?" or is there some other approach there. It really helps you understand the problem a lot better. It's a lot of fun. |
| Mark | Great, I can't wait to see what comes of it. |
| Gavin | It'll be a while, yet. |
| Mark | Are there any other interesting technologies you've seen at the conference this year? |
| Gavin | It's clear there's a lot of excitement around the JSF ecosystem. The top-selling book at the conference was, I think, a JSF book. Things like the technology we acquired from Exadel, RichFaces, and things like ICEFaces are really hitting their stride now and really becoming mature at the same time as Seam is becoming mature. The JSF ecosystem is really one of the most exciting things around at the moment. It was amazing when I got up and did my talk today, and showed people how amazingly easy it is to build rich AJAX apps in JSF, I think some people were like "Oh? Is that all there is? That looks way easier than I thought."
Once you're outside the echo-chamber of javablogs.com, where everybody sits around and tells each other that JSF sucks, and they tell each other back-and-forth so much that everyone knows JSF sucks, except that none of them has actually used it. So it's interesting to see people's faces. JSF's doing great. The Groovy and Grails community is really looking very healthy. Grails books are selling well. Groovy books are selling well. Grails is cool stuff. What Emmanual's been working on with Seam for the last few weeks is learning to write Seam components in Groovy. We mention that people and they're like "all, cool!" It'll be interesting to write Seam applications in Groovy. That's definitely a very strong community building around there. |
| Mark | What about JavaFX? Have you had a look at that? |
| Gavin | I can't say I've had enough of a look at it to make a comment. |
| Mark | Fair enough. To wrap up, I understand you're living in San Francisco these days. What attracted you here? |
| Gavin | What attracted me here was a woman. But I've always loved San Francisco. This is a great city. There's so much to do around, riding various kinds of two-wheeled forms of transportation is a lot of fun. It's a great place to live, actually. |
| Mark | Thanks for you time. |
| Gavin | Thank you, very much. |