X

Promoting the promise of Longhorn

Microsoft's new developer division vice president, S. "Soma" Somasegar, plots a course for making Microsoft's Windows and .Net software the cutting-edge favorite of developers.

Martin LaMonica Former Staff writer, CNET News
Martin LaMonica is a senior writer covering green tech and cutting-edge technologies. He joined CNET in 2002 to cover enterprise IT and Web development and was previously executive editor of IT publication InfoWorld.
Martin LaMonica
6 min read
The success of Microsoft's Windows operating system is inextricably tied to its strong developer division. Since its founding, Microsoft has supplied programmers with the tools to build applications for Windows, the company's crown jewels.

S. "Soma" Somasegar has lived on both sides of the Windows developer fence. After working 15 years at the group responsible for building Windows, he became vice president of Microsoft's developer division earlier this year.

There's a close technical kinship between Windows and Microsoft's .Net development platform--particularly as the company continues to heap more and more software "plumbing," such as Web services connectivity software, into Windows. And developers are instrumental to getting advanced features such as better search capabilities into commercial applications.

Somasegar's job is to keep those programmers within the Microsoft fold. His strategy is simple: give developers the tools to write code securely and faster than Java and open-source alternatives. Last month, Microsoft cast an even wider net in its developer outreach program by announcing its Express initiative, a plan to lure amateur programmers.

Somasegar spoke to CNET News.com about the company's developers, the impact of open source and how Microsoft plans to entice the development community to devote its energies to the company's next big bet: Longhorn, the next version of Windows.

Q: Should we expect Microsoft to do more code sharing for developers?
A: We are learning as we go along here. If you had talked to me two years ago and asked, "Are you going to share an early build of your product?" I would have probably laughed and said there is no way I would know how to do that.

We have learned a ton. As we go along the journey, so to speak, let's see what works for our customers, what makes them more effective, and let's do it in a way that makes sense for us and for them.

How do you feel about the technology race between Java vendors and Microsoft's .Net? Is it becoming clear where people will use one platform over the other?
Sitting where I am, I feel really, really good about the momentum and adoption behind .Net. More than 60 percent of the Fortune 100 companies have .Net applications developed and deployed running in production environment.

And on the development tools side?
Over the last 10 years or so, we have always had a leadership position on what I call the design tools or the IDE (integrated development environment). In some sense, I feel good that Sun Microsystems has realized the importance of developer tools and is investing in that. That is good for Sun.

Tools have traditionally focused on developer productivity. But I've spoken to some customers who think that the management of finished systems is at least as important. Is that something you're trying to address?
Do I expect all developers to rewrite their applications completely on Day 1? Absolutely not.
The lines between a tool to help you develop rich, complicated applications and a set of tools to help you manage that are getting blurred. If you look at it organizationally, there is a division at Microsoft that worries about a set of tools to build to help customers manage their environments, manage their applications that that are deployed. And then we have a division that is focused on tools that help people build applications.

What is the relationship between the developer division and the Windows division? How do the groups fit together?
If you look at Longhorn and think about WinFX (programming interfaces for Longhorn) being the new managed programming experience and environment that we are providing in the Longhorn time frame, that is the collection of technologies that come from different parts of the company organizationally. But all of these groups are really contributing to one product: Longhorn. There is a great relationship that cuts across the organization boundaries, because we are focusing on one product and one set of programming experiences that we want to deliver to our developers and customers.

With so much new technology, I would think that taking existing applications to Longhorn and learning its innards would be a big jump.
Longhorn is in some sense a whole new programming environment, a managed environment. It's much richer in terms of the kinds of experiences that you can create.

We do think that application developers are going to take advantage of this functionality. Do I expect all developers to rewrite their applications completely Day 1? Absolutely not. There are going to be some developers who want to do that. There are going to be some who just want to get used to the platform and then start doing things over time. There will be other developers who will want to do a few things they wanted and then build on top of that.

Microsoft has said that this Express line is not going to be a real moneymaker for the company. So why do it?
Traditionally, we have been focused on targeting the professional developers--the people who are proficient in programming. If you look at the beginning programmer or the hobbyist--people who aren't necessarily programmers by profession but want to write a program or use something to help them in whatever they are doing--the cost of entry is too high, both in terms of the programming model and the tools that go with it.

About a year or two ago, we had this notion of making it easier to develop Web sites. And we came out with a downloadable set of tools and programs called the Web Matrix. That was really the genesis of our idea for Express, because the amount of customer response and excitement and momentum that we saw for that was just huge. That experiment blew our minds.

We realized that there is a huge developer space out there. I call them developers, but they aren't necessarily what I would call the professional developers. They are more like the hobbyists, the beginning programmers.

Would a systems administrator or business analyst possibly want to try out these Express tools?
The thing that I personally like about open source is the feeling of transparency that we have with customers.
Here is a way I think about it: The Express tools are a way to get started. They're for anybody in the world--it could be a student, it could be a 14-year-old kid who wants to get into programming, it could be a different kind of a professional who wants to get the fundamentals of programming. We think that there is at least an 18 million developer base of what we call the enthusiast, hobbyist, academic developer for whom we want to make it super-easy to get started. We hear about students learning to program for LAMP (Linux Apache MySQL PHP). How much has the rise of open-source tools affected your decision to introduce Express?
I wouldn't necessarily say that was behind the thinking of Express. The thing that I personally like about open source is the feeling of transparency that we have with customers.

Historically, we haven't been very transparent with our development process. So if you look at the development life cycle of Whidbey (the code name for Visual Studio 2005) we have come out with what we call community technology previews. To me, that is a huge step forward in terms of how we want to engage with our customers and really build a feedback loop with them.

When I have an early build of the products, I want to be able to share it with my customers and really treat my community of customers as an extension to my development team through the life cycle of the product.