.NETDJ: What can you share with us about the current state of developing and licensing the .NET Compact Framework?
DT: It's currently in beta, and people who are interested can download it from our device developer site: http://msdn.microsoft.com/vstudio/device/default.asp. There are currently more than 10,000 developers in our beta program giving us feedback and building cool applications.

.NETDJ: What do you think will be the "killer app" for mobile .NET computing?
DT: We've seen lots of creative development going on using location services and integration with service providers using XML Web services. I've also seen solutions that have provided immediate solutions to some of today's problems. For example, one early adopter upgraded a manual paper-ordering process using Pocket PCs, the .NET Compact Framework, and XML Web services. Now they punch in orders electronically, and the orders get processed in minutes rather than hours. This has resulted in big savings in operational costs. But really, that's just scratching the surface of what's possible. Ultimately, it's developers who will come up with the next big killer app, not us. My team's job is to give developers the tools to write the killer app.

.NETDJ: What will be the biggest challenge in making .NET succeed, and how will Microsoft overcome it?
DT: So far, I'd actually say that we've been incredibly successful ­ we have hundreds of case studies reaching into the Fortune 100, millions of developers using the .NET Framework and Visual Studio .NET, hundreds of books, and an incredibly active developer community. Getting that kind of momentum was our first challenge. Our next challenge is getting people other than developers to understand what .NET is and why it's important. Developers know what it means to them ­ an elegant framework and toolset for writing applications and XML Web services ­ but we've been vague in our use of .NET for enterprise customers and end users. Going forward, you'll see us introducing clear guidelines ­ focused on end users and enterprises ­ about what .NET is and why it's good.

.NETDJ: What can you tell us about new features in the next release of the .NET Framework?
DT: You can find out a lot at http://msdn.microsoft.com/vstudio/productinfo/roadmap.asp. As you'll see there, this next release is focused on dependability. For us, the higher order bit is making the APIs as stable, secure, and reliable as they can be. If I had to call out one especially cool feature, it's the integration between IIS 6 and ASP.NET in Windows .NET Server, most notably between IIS 6 and ASP.NET. Of course, the rollout of the .NET Compact Framework will also be a major "feature" as well.

.NETDJ: Do you think Microsoft deserves any of the media hype about alleged poor security ­ or is it all just a question of averages? Or is it something else altogether?
DT: Security is a shared responsibility ­ Microsoft, our customers, our broader community, and even the press need to participate actively. At the core of it, Microsoft must deliver a secure stack and a great security model that make it easy for developers to design and write rock-solid applications. When we set out to build the .NET Framework over four years ago, that was our top priority. We were able to deliver on this commitment with features like evidence-based security, role-based security, and new concepts in authentication and authorization, and isolated storage. Delivering a great implementation is just a small part, however. Microsoft is making a huge investment to ensure that we're evangelizing best practices for security coding to developers. We've introduced Trustworthy Computing sessions to all our developer conferences, we're rolling out books and prescriptive architectural guidance documents, updating our Web sites, and generally making a huge investment in education and evangelism around security.

.NETDJ: How tightly integrated will .NET be with future versions of Windows? Is this at all a concern, in case anyone would ever want it unbundled for some reason?
DT: Microsoft's strategic direction for developers is toward managed code ­ the kind of code developers write on the .NET Framework. By default, every product we ship going forward will have a managed-code component. At the heart of this is Windows: we are the Windows company. We believe in an integrated platform that provides programmability and application services as a core part of the OS, and our customers love that the plumbing is there and just works for them ­ no assembly required. Including the .NET Framework as a core feature of Windows .NET Server and all other future versions of Windows is a great example of how we're producing this integrated platform.

.NETDJ: What is your favorite .NET language and why?
DT: Ah, that's always a hard one to answer, since the answer depends so much on what you want to develop! Five or so years ago, when I was doing lots of hard-core development, I probably would have picked C# or C++ for their power and flexibility. However, these days I'm not doing as much development myself, so I usually fire up VB.NET when I need to do some coding, as it's the easiest way to get off the ground and write cool applications quickly.

.NETDJ: Given the ease and power of ASP.NET, why isn't everyone using it to build their Web sites today?
DT: They are! Our research shows that developers are incredibly excited about ASP.NET and are building ASP.NET applications at an amazing rate. There have been over 100,000 downloads of our ASP.NET Web Matrix Project development tool (www.asp.net), and our ASP.NET discussion groups get posts from thousands of unique developers each day. I haven't met a developer yet that hasn't raved about the improvements over traditional ASP when they've used ASP.NET. When Web developers give it a shot, they fall in love with it.

.NETDJ: As Microsoft allows .NET to be integrated with an increasingly diverse set of technologies, how will you avoid lending additional support to competing products (for example, Apache versus IIS)?
DT: For us, it's not an issue of avoiding; it's about partnering with other vendors to ensure that the customer gets the best possible experience. Microsoft has always worked closely with our competitors ­ companies like Oracle and IBM are some of our best customers and we help them build great products on Windows. We will talk with anyone who wants to make investments in .NET, and the reality of the times is that that means having to work with companies who also support competing products.

.NETDJ: As Microsoft continues to add functionality to .NET, how will you avoid stepping on the toes of your third-party component vendors?
DT: We have partners integrating tools, languages, and controls directly within the Visual Studio .NET IDE. We have enterprise customers integrating their in-house tools and online communities within the IDE using the Visual Studio .NET Integration SDK. And we also have component vendors building their components on the .NET Framework, which extends their offerings to .NET developers, enabling them to be more efficient. This gives .NET developers over 100,000 components at their fingertips to seamlessly plug into their applications. That's a clear indication that we won't be stepping on their toes ­ we actually work closely with them to broaden and extend the capabilities of .NET development. Ultimately, it enables developers to have a broader and more universal development experience.

.NETDJ: Which is more important to Microsoft: winning the OS market or the Web services market?
DT: XML Web services is a set of technologies to integrate applications. Without an implementation of XML Web services ­ that is, some software that speaks these common protocols ­ there's nothing to talk about. Our implementation ­ the .NET Framework ­ is a feature of Windows. So, by definition, implementing your XML Web service with .NET is a Windows win.

.NETDJ: Are you at all surprised by Sun's slowness and resistance to Web services ­ given their obvious benefits, even to Sun?
DT: Not really. I don't think Sun recognized the power of application layer integration, and they definitely viewed XML Web services as a threat to Java, but we're happy they finally came around. More industry heavyweights supporting XML Web services equals more ubiquitous technology, which ultimately equals a drastically improved development experience.

.NETDJ: Given that VS.NET lets you create Web services in minutes, why isn't everyone building Web services today?
DT: You might be surprised. IBM has said that 80% of their new consulting work is around XML Web services. We're seeing similar patterns in our own work with customers. So our customers are building Web services today, and we have lots of case studies to demonstrate that. Today, our customers are mostly integrating systems behind their firewalls, making legacy code and systems work side by side with new projects. We expect this integration to take place outside the firewall in a B2B scenario in the near future, and in a B2C scenario in the next 12-18 months.

.NETDJ: Since Web services are, by definition, platform independent ­ why choose Microsoft?
DT: Time to market, cost, performance, and standards compliance. We have the best platform and tools, making it easy to build applications that expose and consume XML Web services. We offer integral support for the entire stack of XML Web services standards ­ XML, SOAP, WSDL, and UDDI ­ and a powerful, familiar programming environment that most developers already know. We offer development in over 20 popular programming languages. We've got high-performance access to SQL Server and Oracle databases. The list goes on and on. But overall, we said we'd agree on standards and compete on implementation. We've poured lots of blood, sweat, and tears into making our implementation rock.

.NETDJ: What is Microsoft's position on Mono and other non-FreeBSD attempts to port .NET to other platforms?
DT: Of course, we're partial to our own port of the CLI to FreeBSD ­ the Shared Source CLI. We created and delivered over a million lines of code that developers can download for noncommercial uses free of charge on FreeBSD and Windows XP. Regarding Mono specifically, we're really happy to see that Miguel [de Icaza] is also finding the platform technologies to be compelling and useful. They are a concrete example of people building on the ECMA standards process.

.NETDJ: Does Microsoft feel that Web services must remain rooted in XML in order to succeed? What about other, non-XML based approaches to Web services that are currently gaining momentum?
DT: Web services are about connecting. IBM, Microsoft, and other industry leaders have pushed XML as the base technology to build on because of its ubiquity, flexibility, and power. The openness and ubiquity of XML are key to the connectivity vision of Web services, and all the major software vendors have invested a lot of time and money in trying to add XML to their application server offerings. Other approaches ­ binary protocols, or protocols that perform simple operations ­ have been tried and have failed to deliver on global-scale, multiplatform integration. XML is not only a viable approach, it's the only approach based on current technology that can deliver.

.NETDJ: So, what would be your response to critics who say that XML is far too bandwidth and processing intensive to form a foundation for interoperability technologies such as Web services, which should ideally run well on very limited devices and platforms?
DT: That's a fairly common criticism of XML, but look at how other technologies that emphasize interoperability and scale over absolute performance have fared over the years. A great example is TCP/IP. In the late 1980s and early 1990s, there was a lot of debate in the industry about whether TCP/IP would ever be ubiquitous, since it was much "fatter" than other network transport protocols like NetBEUI and XNS. Some people claimed that TCP/IP would never run well on PCs since it was so heavyweight relative to the other protocols. The fact is that it needed that extra heft in order to achieve its primary goals, and the enormous forces caused by Moore's Law and the Internet simply overwhelmed any marginal performance advantage that the other transports had. I expect that the same thing will happen with XML and Web services: the interoperability will prove so compelling that any downsides are swept away as the industry works hard to apply the power of Web services to a broad array of devices.

David Treadwell is the general manager for Microsoft's .NET Developer Platform team, part of the Developer Division. The team is responsible for the .NET Framework and Common Language Runtime (CLR), the core runtime platform of .NET; the .NET Compact Framework, which brings the platform to small devices; and the Microsoft Shared Source Common Language Infrastructure (CLI), which makes the platform, including underlying source, available on several operating system platforms. David played an important role in the establishment of Microsoft's .NET strategy for developers. David holds a BSE in electrical engineering from Princeton University.