Senior VP cites strategy for making platform work smoothly with Java.
As Microsoft Corp., of Redmond, Wash., rolls out Visual Studio .Net, a product its touting as the best available for developing and delivering Web services, Eric Rudder, Microsofts senior vice president for developer and platform evangelism, sat down with eWeek Senior Editor Peter Galli to discuss .Net and the road ahead for the companys Web services vision.
eWeek: What are the arguments that Microsoft uses to convince developers and customers to go with a .Net-based platform rather than a Java-based one when considering their Web services needs?
Rudder: XML is native to our platform, which was designed from the ground up to be a Web services platform. There are no extra classes bolted onto the side, no extra configuration files need to be created on the side and no extra build steps are needed. Customers are smart enough to compare for themselves which platform is easier to build Web services solutions on, which is easier to run solutions on, which is easier to operate and which offers better price/ performance. I think on the characteristics that people find attractive in any platform, .Net compares favorably.
eWeek: What are your plans to ensure that Java works with .Net, as many potential users want the guarantee that the language they have chosen is 100 percent supported on the platform?
Rudder: We have a threefold strategy with Java. One is interoperability, so where you have a Java platform, the investment in that platform will be protected. Secondly, were making sure that we have a great Java language implementation for .Net, which is Visual J#, and were committed to shipping that product so developers who want to code in the Java language can do so on the .Net platform. The next and probable beta [version] of J# will likely be in early March. The third is for people who have a source base of Java and who want to come over to the .Net platform but are more comfortable in C#. We have tools that let people convert code bases over to C#.
eWeek: Microsoft is apparently telling its Web services partners that the rollout of the first wave of .Net My Services may be delayed and that you are rethinking the business model around this. Some of the changes you are considering deal with concerns that have been voiced around security, privacy, lock-in and Microsofts perceived role as the gatekeeper of the Web services world. Is this correct?
Rudder: One of the reasons we shipped a beta of .Net My Services at our Professional Developers Conference in Los Angeles late last year was to get feedback. We gave out some 7,000 copies of the beta then and have gotten back a lot of feedback, which has been consistent. The feedback was around how we partition data, how we operate it, how we make sure customers can work together and bring some trust. A lot of it was about concerns. We take that very seriously and are looking at how best to incorporate some of that into the first version, and I think what were working toward now is how to balance schedule with the needs of our customers. Were figuring out what things we have to get right in Version 1. One of the messages that came across clearly is that customers want to own their own data and that My Services can be run behind the firewall. If they chose to use My Inbox, that doesnt mean Microsoft has access to all their mail or anything like that. So we will continue to work on federated versions of the system that give individuals and corporations control over their data and will continue to take these considerations and feedback into account. We will adapt the system in such a way that customers will be happy with it.
eWeek: Going forward, what are your goals for your Web services vision?
Rudder: Our teams are working on making Web services even easier to use going forward. While [Simple Object Access Protocol] is a great base-line implementation, we are continuing to evolve the frameworks to make it even easier to program Web services. Part of this is developing new technology like scale-in, scale-out, clustering and high-end availability. So we continue to push the envelope in how to make Web services the core of an enterprise architecture.
eWeek: Are you planning to target the business/enterprise market with a separate set of Web services and initiatives?
Rudder: We dont see the architecture splitting into a consumer and enterprise branch. They have different needs, and thats a challenge to us, as is how the Web services specs evolve to meet those needs by providing flexibility within profiles. Take security, where in an enterprise there could be one way to provide credentials that could be different for the mass consumer market. However, that doesnt mean you necessarily need two security architectures. I think you can factor your architecture and make it composable in such a way that we dont need to diverge.
eWeek: Is Microsoft considering porting, or allowing someone else to port, the .Net platform to Linux?
Rudder: As you know, we are working with Corel [Corp.] to port .Net to the FreeBSD platform. We will ship the first milestone of that work fairly shortly, and people can then take a look at it. We will examine commercial interest and react accordingly. We have also opened the .Net specifications up, and youre seeing things like the Mono Project being implemented on top of Linux. Our primary strategy is one of interoperability, not portability, and theres a question about whether portability really works. Weve talked about portable specs before, but it turns out that to get better solutions working, there are real reasons to extend the platform. Well see what happens and whether we let the platform specialize and make sure it interoperates or whether portability is worth following.
eWeek: What is Microsofts current position regarding the Liberty Alliance?
Rudder: We would like to join the alliance, but we need to make sure that the intellectual property clauses in the charter work for software companies, as many of the founders are not software companies.