| QUESTION & ANSWER |
BEA's Carges on open source and runtime SOA |
 |
By Michael Meehan, News Writer
06 Oct 2005 | SearchWebServices.com |
|
|
Before becoming chief technology officer at BEA Systems Inc., Mark Carges headed up BEA's Enterprise Framework Division -- including integration and portal products -- and was one of the original architects for the Tuxedo transaction system.
Part 1 | Part 2You've made a push around SOA runtime capabilities, do you feel that sluggish runtime execution has proven a barrier to entry for some would-be SOA users?
Mark Carges: That goes back to BEA's pure value proposition. What people expect from us and why they give us money typically comes down to the fact that we give them infrastructure and allow them to build solutions that work and scale and that they can put in production for the next 15 years. Those operational qualities are why people come to BEA. You can go get a J2EE app server or Web services stack for free in open source. The value we provide is how well we weld that to the operational robustness. That's what we focus on.
What does runtime SOA still need?
Carges: To help make it a reality for the business would be a tie-in between all of the information flowing in the service network between the producers and consumers and a direct tie-in back to the business. How much money is currently flowing in my service network? Not how many messages, how much money. How many orders are currently flowing in my service network? If I have a loan origination service that's being used across many different portals, I want to be able to look at the enterprise level at any particular time and see how customers and dollars and outstanding loans are currently in flight right now. So true business activity, not just how many bytes and bits and messages and so on. If I can see what business value is flowing through my service network, that would be tremendous. It's not just being able to do the composite apps, it's being able to tie those back to those business metrics.
What value do you believe you can add on top of open source frameworks?
Carges: The key value-add, take Spring as an example. Can a customer get support? Can they call up and get support anytime of the day or night when a mission critical app is down when they've built an app in this blended model? They need that support because if it goes down, the production ops people look at it and say, 'Oh, it failed in here, what is this Spring thing? Who supports that?' Everyone says, 'Not me, not me, it's this small company over here.' No one wants to hear that.
The second thing is the tooling support. If you look at how you build an application in Spring, there's the code I write, the plain old Java objects, there's all the meta data and XML files. The reason these things are so productive is a lot of the plumbing information is not in the code anymore, it's out in the meta data. Well, you could easily get into a problem where you've got all of these little meta data files. I need tooling to keep all of that together. Since Workshop is based on Eclipse, now we can provide those tools right in the Eclipse environment.
Then, when I am in production, I'd like to actually see a deeper level of integration so when I go into my WebLogic console I'd like to sort of see, 'Here are my EJBs, here are my SpringBeans, here are my other components.' I'd like these to show up as not just 'these things take up a bunch of memory, I don't know what it is, but it's over here.' We've done integration in the management console, so you get that full roundtrip of development, operation and support. You do all three of those and you're providing real value to a customer, and they're leveraging these frameworks that they're getting a lot of productivity benefits from.
In a way you're letting the open source community act as your R&D shop then?
Carges:
The open source community is doing a tremendous amount of innovation and some of those innovations are really catching on at the Web tier, at the business object tier and at the data tier. For those that are really catching on with our customer base, we want to provide the development support, the runtime management support and then the production support.
Does open source need vendors as much as vendors need open source?
Carges: The customer doesn't know what the future is of these frameworks. The people who created them, what if three years from now they move onto something else? Look at BEA with Tuxedo. Tuxedo is 21 to, 22 years old now. BEA is still putting out new versions and supporting old versions. Just like IBM and CICS, they know we'll be there for a long time. With the open source model, you don't have that same comfort level. Remember, they're not tied to this thing. They created it, but they can all move. Customers are worried about it with Linux, of course, because you've got major corporations saying, 'Don't worry, we're behind this.' There's a whole spectrum between that and what happens if interest wanes. That's happened with a number of open source projects. It's happening right now with Struts and JSF. People are moving to new things. Well, what if you put all these Struts-based apps into production, what happens? That's starting to catch up with people and that's where the blended model is important because you can see what corporations are betting on and who's standing behind it.
|
|
|
|
