.NET Editor-in-Chief Derek Ferguson sat down to chat with Microsoft Software Legend David Chappell at the recent Professional Developers Conference 2003. In this exclusive interview, Chappell talks about how he came to focus on Microsoft technologies, and why he enjoys teaching and speaking about new technologies.

-Interviewed by Derek Ferguson, .NETDJ Editor-in-Chief

.NETDJ: I'd like to start off by having you tell me a little bit about yourself, please.
DC: I am a speaker, writer, and consultant. I talk about technology - how to use it, when to use it, why to use it, and where to use it. My audience is developers, architects, and IT decision makers. For the past decade or so, I've been focused on Microsoft technologies.

.NETDJ: Why Microsoft technologies?
DC: I spent the first 10 years of my career as a Unix guy, until something happened that abruptly changed my path. I spoke at a USENIX conference - an event for hardcore Unix geeks - right around the time Windows NT was first released. Microsoft had sent a guy to do an evening session on NT, which was like sending Daniel into the lion's den. He changed my life! I talked to that guy after the session until midnight.

It seemed so clear to me that with NT, Microsoft was going to do to the server market what they had already done to the client market: commoditize the hardware. The technology was also so impressive - NT was a much more modern operating system.

So the truth is that now - almost 10 years later - my initial thoughts have been proven correct. Windows has commoditized the server market, and the Unix world has responded by moving more and more to Intel-based Linux systems. I still think I made the right choice in changing my focus.

.NETDJ: What was it that excited you so much about Windows NT?
DC: It was first of all the business aspects that made me think Windows would be dominant. But it was also the technology. Unlike Unix, NT wasn't derived from a system built in the late '60s - it was a brand new operating system.

.NETDJ: What is your primary claim to fame as a Software Legend?
DC: Initially, it was a book I wrote called Understanding ActiveX and OLE. Even after all these years, it's still the book I've written that people are most likely to remember. My latest book is Understanding .NET. It's the first book in Addison-Wesley's Independent Technology Guides series, for which I am series editor.

I cannot overemphasize how much I regret giving my earlier book that title, though. I should have called it Understanding COM. But ActiveX was a brand name that Microsoft created just before the book was published, so it seemed like a good idea at the time. Unfortunately, they largely killed that brand within a year.

.NETDJ: So, what do you think was so special about Understanding ActiveX and OLE?
DC: It was a book about COM that had no code. I wanted to write about the technology itself, not the differences between how it looked in VB versus C++. Also, I wanted to write a book that developers could not only understand themselves, but could also give to their bosses to explain the platforms they work on.

.NETDJ: Speaking of platforms - how did you get involved in .NET, given your roots in COM?
DC: I've always marveled at Microsoft's willingness to make radical changes - it takes guts - and so the impact of such a big change was one thing that interested me right off the bat.

Beyond that, I really liked the unification and relative simplicity offered by .NET. Compared to the world of C++ and Visual Basic, which was held together with the duct tape of COM, the unified CLR-based approach of .NET was very attractive. And the .NET world was obviously more modern than the technologies it replaced, another very good thing.

.NETDJ: How did you get started making the leap from COM to .NET then?
DC: I have done a number of speaking tours over the past few years, many of which Microsoft has sponsored. I had one right after the 2000 Professional Developers Conference (PDC), which was where Microsoft first announced .NET, so I was the first person to talk about .NET to a lot of the audiences that I encountered. That was a lot of fun.

I spoke to all sorts of people - developers, architects, IT decision makers. Everyone was extremely interested, but of course not everyone was enthusiastic.

.NETDJ: Why do you think some people were resistant?
DC: .NET was a big change to their existing investment, and it meant lots of retraining. Also, plenty of people were confused because the .NET message - especially initially - was presented in a complicated way. One example is that the core question of "What is .NET?" was very difficult to answer cogently at that point.

.NETDJ: Do you think that it is any easier to answer that question today?
DC: Oh yes! Today I think Microsoft has pulled back from their original very broad use of the .NET brand. Today .NET pretty clearly refers to the developer technologies associated with programming Windows: the .NET Framework and Visual Studio .NET.

I think the new simplicity is great. It's a much clearer and stronger use of the term.

.NETDJ: What made you choose the speaking and teaching lifestyle?
DC: The main reason I do this is because I really like it - it's a great life. It's fun to interact with such diverse groups of people all around the world. I get to speak with so many smart people about such cool technologies!

.NETDJ: And what do you currently perceive to be the absolute coolest new technology?
DC: Well, here at the PDC the thing that I have found to be the most interesting has been Indigo. But in a larger sense, I'm really interested at the moment in the whole notion of services and service-oriented architecture (SOA), of which Indigo is a big part.

.NETDJ: I've heard a lot of buzz about SOA lately, but wouldn't you agree that SOA can easily be overused?
DC: Sure. One of the issues with Web services is that using them is generally more expensive than using similar, binary formats for distributed computing like DCOM or CORBA. And it's not hard to design things that overuse services at the expense of good performance.

.NETDJ: Are there any examples out there today of organizations that you think are really using Web services properly?
DC: Two of the most visible and well known are Google and Amazon. I think these are very interesting uses of Web services, but what they're not an example of is the dominant use of Web services. Google and Amazon's Web services are on the Internet, whereas the majority of people using Web services today have them inside their firewalls. This has a lot to do with the lack of really effective security in Web services today.

.NETDJ: What's wrong with just using SSL?
DC: SSL isn't good enough. It's fine for point-to-point interactions, but it's not hard to get into scenarios that are more complex, cases for which SSL is not sufficient. This is why Indigo implements WS-Security and its related technologies.

.NETDJ: So as a developer, Indigo will give me WS-Security. Anything else?
DC: A lot more. It makes it much easier to develop service-oriented applications; it's a platform for building this kind of app. It also unifies the current diversity of Microsoft technologies for building distributed applications. Enterprise Services, ASP.NET Web Services, and .NET Remoting all get effectively replaced by Indigo.

.NETDJ: Very cool! Do you think your COM background has fueled your interest in Indigo?
DC: Before I started working in the Microsoft space I worked with Unix-based distributed computing for many years. Actually, I was primarily interested in COM because of Distributed COM. The truth is that technologies for building distributed applications are what I have been interested in for most of my career.

.NETDJ: Since you've raised the topic of Indigo, which is one of the new Longhorn technologies being unveiled here at the PDC, let me raise another: Avalon. Any interesting thoughts on it?
DC: To me, the most interesting aspect of Avalon is that it's part of the new Longhorn application model, which lets developers blur the distinction between ordinary Windows applications and Web-based applications. Thinking of an app as a set of page-like abstractions that can be downloaded on demand, whether or not the app is running inside a browser, is both powerful and simplifying: a great combination.

.NETDJ: How about WinFS - Longhorn's new storage system?
DC: I've been waiting for Microsoft to start bundling a DBMS with Windows, and WinFS looks like a long step in this direction. The notion of a schematized file system is potentially very useful, too. End users probably won't care much about dealing with the WinFS metadata in a direct way, but one big benefit that they should get from this is better integration among their applications. So, for example, because there are standard schemas that define things like contacts, contact information can potentially be shared among many applications.

.NETDJ: What will you do with all of the new stuff you have learned here at the PDC?
DC: I have a European tour scheduled for early 2004, where I'll be talking about Indigo and Longhorn. I'm also writing a book on Indigo that will take up a lot of my time.

.NETDJ: Tell us about that book?
DC: The goal for all the books I write is pretty much the same: I want to write the first book a developer reads on a topic and the only book that the developer's manager reads. That's why my books are never cookbook, screenshot-filled, how-to tutorials. IT managers don't care about that, and I don't think developers should start there, either. Developers should start with the big picture of the technology, and that's what I try to provide.

.NETDJ: Would you call your books "introductory reading" then?
DC: My books are used in upper-level courses at some universities, so they aren't introductory from the standpoint of introducing people to programming - you need to have some knowledge to make sense of them. They are intended to introduce people to a technology, though, such as the .NET Framework or Indigo.

I will also produce a second edition of my last book, Understanding .NET, that will be updated for Whidbey. I think it's kind of a shame that at this PDC Whidbey has tended to be swamped by Longhorn.

.NETDJ: What do you think are some of the outstanding features of Whidbey that are getting lost in the shuffle, so to speak?
DC: I think that the new language features in Whidbey, like generics, should definitely be getting more attention.

I also think that, in terms of the service-oriented world we're headed for, some of the forthcoming tools for Visual Studio that help with design and deployment of service-oriented applications are very interesting, such as the Whitehorse technology.

.NETDJ: Do you think there's a market for Longhorn books this early in the cycle?
DC: Yes. I think that the large number of people here at the PDC and the percentage of them that went to the Indigo sessions says a lot about the interest in these technologies. Even though developers can't use this today in their day-to-day-jobs, it's always useful to know what's coming. And besides, who doesn't want to learn about cool new technology?

.NETDJ: The only people I can think of would be those having to focus 100% on their current jobs in order not to lose them in this economy. This brings me to my final question: How do you think the economy has affected .NET adoption?
DC: Microsoft had the great misfortune to release .NET into the worst economic climate the IT industry has ever seen. This isn't fatal - it has just slowed down adoption. Regardless, the core point is that developers really care about new technology, and their managers care about the benefits that technology can bring. The truth is that new ideas and new applications of those ideas are fundamental to what we do. If you don't like change, get out of the software business!

About David Chappell
David Chappell is principal of Chappell & Associates. His books on enterprise software technologies are used at MIT, ETH Zurich, and other universities. He is series editor for Addison-Wesley's Independent Technology Guides, and his most recent book, Understanding .NET, was published in 2002. David's consulting clients have included HP, Microsoft, and Stanford University. You can check out his web site at www.davidchappell.com.