Roundtable Transcript, Part 2 (Continued)

Predictions for the Future
[37:22] Phipps: OK, so we're getting to the point now where I'm going to ask you for your predictions for what will be true next year, which is what you've all been waiting for, I know. What I'm looking for here is to see what your prediction for the year is on any Java-related topic. You can predict what's going to succeed or fail in some aspect of the Java language. You can predict on the impact of an open source community of some kind. But preferably you won't be promoting a named product or a corporation of your choice. So predictions for the year, but would anyone like to volunteer? I would normally pick Dave, but he looks kind of worried. [Laughter]

[38:08] Elloy: Two for me. First of all, SOA. The orchestration and the manageability of what's deployed, and therefore the incorporation of SOAs, I think, is going to be the biggest challenge that we face in 2005. That's point A. Point B is we're all standing on the beach and there's two great big tidal waves rapidly approaching as one J2EE and the other one .NET written on it where J2EE equals Eclipse. And like it or lump it, it's going to happen. So how do we embrace that as tools vendors so as to not be caught off guard? And that's probably more of a 24-month as opposed to a 12-month.

[38:45] Phipps: So in terms of prediction, what do you think is actually going to happen?

[38:48] Elloy: I think that what will happen is the more widespread rapid adoption of the .NET framework in corporate America as well as corporate Japan and corporate Europe as a client-side, rich vehicle for deploying enterprise applications. Microsoft has not been a big proponent, has not been a big player in the enterprise to date. From what I can see that is changing rapidly. I'll tell you in 12 months.

[39:15] Stover: They've learned some valuable lessons from the Java community.

[39:17] Elloy: They absolutely have.

[39:20] Phipps: It's good to be a good prototyping community, isn't it? So, who's next? Steve.

[39:26] Stover: I'm obviously a big fan of standards. We talk a lot about Web services and their adoption, but the reality is we're still in early adopter phase. And a lot of companies have tried some things, and it's I think with the addition of the Web services specs in J2EE 1.4 over the next year, you're going to see the adoption of people understanding how best to utilize that technology. I mean, granted again, it's a start, but we have a ways to go incorporating additional standards for Web services into the J2EE platform. But I think we'll see a significant amount of activity just because that we now have that as part of our J2EE specs.

[40:10] Phipps: OK. Michael.

[40:11] Bechauf: Frameworks are going to be a very thorny road only because all the vendors coming out with all sort of different frameworks, they always say are the best, but on the other side of it, when there are more frameworks, the developers will come back and say these are not acceptable; all these very different-looking frameworks will not be acceptable, which will have a very large impact also on the tools. If the frameworks are different, which means if the runtime environment is very different, the tools will be different. It's a very confusing world, so I think the developers will push back and will basically very clearly articulate a need for standardization of the framework space, but on the other side it's going to be very thorny.

[40:52] Phipps: So are you predicting that this time next year we will see standardization in progress on frameworks?

[40:57] Bechauf: Yes. It needs to be. This time it's really not going to come from the vendors, but it's going to come from the developers.

[41:06] Phipps: Who's next? Mike?

[41:08] Milinkovich: I have two. First is that the persistence going into EJB 3.0 will be in retrospect one of the smartest things the J2EE guys ever did. And, two, that the rich-client platform—and sorry for plugging Eclipse—is going to spur a renaissance for client-side development in Java.

[41:35] Phipps: And it just isn't Eclipse actually; there are other approaches out there. One of the general session comments this morning was that the desktop is back. And you know I've noticed the JDNC project is going on as well as an LGPL project up on Java.net, which has got an XML scriptable GUI creation that looks incredibly simple, and it looks like any of us could build a tool that would allow our customers to become productive over a weekend. So I see that renaissance happening. So I'm sorry, I shouldn't really be supporting a prediction there.