In a detailed interview Gupta outlined his company's plans for the new medium for delivering software services.
Where do you think customers will initially deploy Web services?
What we're finding is that for most IT organisations, they really intend to use the power of the Web services technology to integrate internal systems and the systems of their business partners. There's tremendous interest in IT organisations for that. What makes Web services exciting and interesting is the simplicity of the approach.
What will Computer Associates specifically be doing in this area?
We're doing a variety of things to enable enterprise-class, mission critical-class Web services. We are actually working to change all of our products so they can manage Web services.
So, for example, [CA's] UniCenter not only manages the regular infrastructure - which is everything from networking systems to database and application servers and apps and so on - but also manages the Web services component to ensure availability and performance.
Similarly, our portfolio of security tools would secure Web services and access to Web services in terms of who is authenticated and authorised. We have a high-performance, highly scalable directory for keeping track of who the users are, what the Web services are, who can use it and who can do what with which Web service.
There are some very interesting security nuances around Web services that have to be dealt with. We believe that our access control, Web access control, directory and certificate technology can leverage that. Another area is our portal technology, which is being enhanced so that it can automatically take any Web service and come up with a default visualisation for it.
In terms of the competition between Java 2, Enterprise Edition (J2EE) and .Net environments, which one do you think will ultimately dominate?
My game is to make money, and the way I make money is that I don't really care which one wins, as long as I help both those camps. In my opinion, there will be a heterogeneous Web services environment within an enterprise. People aren't going to change their mail service from Outlook to a J2EE-based thing. People aren't going to change some of their J2EE applications to .Net. That's the reality of the world. We are working very closely with Microsoft, and we're working very closely with Sun Microsystems.
Do you think that Web services will enable more software competition because it will be easier to integrate best-of-breed software applications?
The question really becomes: How fast can SAP get there with its Web services implementation? I think that the challenge that the little vendors will have is that even though you can use Web services to integrate, the standards are still up for grabs. Just because you have XML doesn't necessarily mean that there is a standard that everybody agrees on and how a new system should really interact with an order-entry system. There is an opportunity, but I am not sure that the opportunity will be something that others will be able to make the most of.
Computer Associates has made a push into the area of business process integration. What impact will Web services have in this space?
If you look at the Web services orchestration noise and hype, it is all around business process, specifically integrating a Web-flowing business process across multiple Web services. You have to let people use Web services as an integrated application.
Within our portal, for example, we're expanding our portal solution to provide support for Web services. It will also have support for things like some degree of Web services orchestration and workflow. I'm sure the app server vendors will probably provide something in that area as well. We see the co-ordination, orchestration and the management of the business process across multiple Web services becoming one of the pieces within our global server technology.
What will be needed to marry Web services with transaction-based processing requirements?
Right now, each vendor is actually trying to come up with their own orchestration stuff: IBM is doing stuff, Microsoft's doing stuff, BEA is doing stuff, CA is doing stuff and Oracle is doing stuff.
In terms of co-ordinating multiple Web services into a single transaction, I think that's really workflow-type technology. With Web services orchestration, you want to know exactly what's going on in my transactions and when they're actually happening and not happening, and where the bottlenecks are. We see delivering that through UniCenter.
But if you want to have transactional security rather than individual Web service security, that's a very different approach. Web services orchestration, I think, becomes an integration and workflow management challenge on one hand and then management of security challenge on the other hand. I think the jury is out on exactly what happens there. But Web services orchestration is really the next step.
How far along is Computer Associates with XML?
Our portal server today is completely XML-based. That's why we are excited about what we can do in the area of Web services with it. UniCenter 3.0 has XML interfaces to a variety of UniCenter services. I think we have actually gone a substantial way already, and actually talked about an appliance using XML and our products that support XML today.
What impact will Microsoft's myServices offering, otherwise known as Hailstorm, have on the enterprise?
It has more of an impact for the consumer side than on the business desktop. The challenge that Microsoft faces in terms of making Hailstorm technologies the sort of key thing for enterprises is that their back-end services are limited to Windows servers. I don't expect Microsoft to really come out with a Web services implementation that's more scalable on Solaris than it is on Windows servers. Because of that, I think that it'll be interesting to see how far along they get in terms of getting Hailstorm and Hailstorm technologies to sort of be adopted in the enterprise.
Beyond Web services, what else is Computer Associates excited about?
Wireless technology as a whole is extremely exciting. The whole notion of having mobile users with reliable high-speed wireless connections really changes the game. There is a tremendous amount of opportunity in the wireless Internet and the wireless Web. We can do software updates to the wireless infrastructure for devices. We do asset management of devices for wireless infrastructure. We manage the wireless infrastructure with UniCenter. We know the location of people based on the access points. We can track a device as it goes across wireless access point. We provide security in a wireless infrastructure, with both encryption technologies and things like access control authentication. We're working very closely with most of the leading vendors in the wireless space. I think that it will be awesome in terms of an opportunity for our underlying technologies.
How significant is peer-to-peer networking going to be?
With UniCenter 3.0, we did a bunch of peer-to-peer work, and we're doing more with the upcoming releases of UniCenter. One of the things you need peer-to-peer for is scalability. With UniCenter 3.0, we have a solution for managing networks that are scalable to hundreds of thousands, if not millions, of devices. It is very much a carrier-class network management solution, and I just see that continuing to grow.
Is there anything else on your immediate radar screen?
The whole storage space is continuing to evolve in a very interesting way. The 11 September incident brought tremendous focus on disaster recovery. As a human being, you hate to leverage tragedy, but at the same time the reality of business out there is that it took something like that for a lot of businesses to wake up when it comes to disaster recovery.
One of the biggest challenges is that the amount of data has grown tremendously, and the window for back-up and recovery has grown shorter. We're seeing huge interest in the extremely high-performance, highly scalable back-up and recovery solutions.