Transcript: Longhorn at the PDC

Robert Hess, Group Manager for Microsoft Corporation and Show Host.

ROBERT HESS: Welcome to another episode of the .NET Show. For those of you who were at the recent Professional Developers Conference in Los Angeles, you heard Microsoft talk about a variety of technologies that we're planning on coming out in the next couple years, as well as specifically Longhorn, the next version of Windows. For those of you who did not attend the PDC, we have a special treat for you. We were down there and we got a chance to talk with some Microsoft people about Longhorn, the technology they're developing, as well as some attendees to find out what they thought of Longhorn and how they think it might help them in their future programs and development efforts. But before we get into that let's check in with Erica and the news.

Hosted by Erica Wiechers, Program Manager.

ERICA WIECHERS: Hello, I'm Erica Wiechers. Welcome to the MSDN News Update.

Longhorn Unveiling at PDC 2003

On October 27 at the Microsoft Professional Developers Conference (PDC), Microsoft unveiled Windows "Longhorn," the code name for the next generation of the Windows platform. The PDC was an important first step in the journey to delivering the next version of the Windows platform and enabling the next wave of computing. Attendees were given pre-alpha bits to the Longhorn preview so they could begin exploring this new platform. The Longhorn preview is now also available on MSDN Subscriber Downloads.

New Longhorn Developer Center and SDK

To support the unveiling of Windows "Longhorn," MSDN launched a new Developer Center dedicated specifically to this new platform. Here developers can find everything they need to get started with Longhorn, including articles, samples and community features, such as forums and blogs. In addition, MSDN released the Longhorn SDK on its new Labs site. The SDK contains documentation, samples, build environments, and tools designed to help developers build applications and libraries that target Longhorn.

Visual Studio "Whidbey" and SQL Server "Yukon"

Also at the PDC, Microsoft gave attendees prerelease versions of Visual Studio, code-named "Whidbey," and SQL Server, code-named "Yukon," the next versions of Microsoft's development tools and database. These releases center on continued advancements in developer productivity, making Web Services easier and more powerful, and simplifying data access and management by taking advantage of the latest in Extensible Markup Language (XML) and managed code technologies. Microsoft expects beta versions of these products to be widely available to MSDN Universal subscribers in 2004.

Tablet PC SDK version 1.7

Microsoft also announced the availability of the new Tablet PC Software Development Kit (SDK) version 1.7. The SDK is designed to make it easier for developers to create rich, ink-enabled applications by providing new and improved tools such as support for Web- or HTML-based pages, real-time stylus and ink support, a context tagging tool to define data field types without getting deep into code and writing recognition enhancements. The Tablet PC SDK 1.7 will be available for download in the second quarter of 2004.

This has been the MSDN News Update. I'm Erica Wiechers.

Robert Hess meets with Pablo Fernicola, Quentin Clark, Brad Abrams and Chris Anderson to discuss what the key messages are that they hope to communicate to attendees of the PDC.

ROBERT HESS: Welcome back. Like I said, today were at the PDC talking about what Longhorn means to the attendees and Microsoft presenting it to the attendees. Joining us here we have Pablo Fernicola, Quentin Clark, Brad Abrams, Chris Anderson and Erica and myself, of course. Hopefully, we'll take and give you some ideas what the PDC was about and how we help people understand what Longhorn meant to them. Well, thanks for joining us, guys.

ROBERT HESS: Welcome back. Like I said, today were at the PDC talking about what Longhorn means to the attendees and Microsoft presenting it to the attendees. Joining us here we have Pablo Fernicola, Quentin Clark, Brad Abrams, Chris Anderson and Erica and myself, of course. Hopefully, we'll take and give you some ideas what the PDC was about and how we help people understand what Longhorn meant to them. Well, thanks for joining us, guys.

QUENTIN CLARK: Thanks for having us.

ROBERT HESS: So, some of you guys have actually had some sessions already, talking to you about Longhorn. I'd like to kind of get your ideas on what you think was important to you to get across from your sessions and what you think the audience has been taking back about what Longhorn is and what some of the technologies are that meant the most to them. Pablo, why don't we start with you? What do you think some of the aspects of Longhorn was that you wanted to communicate to people, and how well do you think those messages are getting across?

PABLO FERNICOLA: Well, in many ways the message is that the client's back. We have a lot of new technologies for people to develop applications on the client, we are taking care of a lot of key problems that people have today in relation to deployment, for example. One feedback from the attendees has been fairly consistent is that they're surprised at the magnitude of technology that we're presenting at the conference. In many ways people were expecting perhaps just incremental changes. We're showing a lot of new technology and people are excited about it. We're getting a lot of questions, a lot of people coming to the booth, a lot of people coming to sessions. We have sessions overflowing that have to be repeated. So we're getting a lot of good feedback.

ROBERT HESS: Now, you said the client's back; now, what do you exactly mean by that, the client's back?

PABLO FERNICOLA: So, for example, the technologies I'm working on relating to Avalon are all about developing applications from the client. So things related to a new set of controls, things related to graphics, things related to data binding as well as things related to just application deployment and how you build an application, how do you integrate an application with the user experience on the shelf. So basically covers the whole wide range of things that you do on the client.

ROBERT HESS: Basically then the smart client applications, you're actually writing binary code that's running on the client rather than relying on the browser and compatibility between different browsers and such. We're focusing on making sure that developers can write really rich and exciting applications that take full advantage of the client.

PABLO FERNICOLA: That's right, and actually one of the selling points is that it's all managed. Even the code that we write ourselves, a large percentage, probably 90% of our code is managed with most of our developers writing in C#. We have a few that write managed C++, where we interact with Direct3D. That's another key thing, another key message for this conference. Basically all of Windows on the client is going to be using Direct3D from the moment you boot to the moment you shut it down.

ROBERT HESS: The managed Direct3D interfaces?

PABLO FERNICOLA: No, we would basically talk directly to the lower levels of Direct3D. We do a lot of the management ourselves.

ROBERT HESS: Okay. Now, from a managed standpoint, I know Brad, managed code is really important to you.

BRAD ABRAMS: Absolutely.ROBERT HESS: So how does that apply to the different technologies we're talking about here at the PDC?

BRAD ABRAMS: Well, I really think Longhorn is kind of the fulfillment of what we've had with the promise of the .NET Framework and that we're now seeing really ubiquitous support for the .NET Framework across the entire stack within Longhorn. So that means developers that are already familiar with C#, Visual Basic .NET and the broad .NET Framework class library that we have, are prepared to move to Longhorn. They already have the skills necessary to take advantage of the cool stuff Pablo talked about, the cool stuff going on WinFS and Indigo.

ROBERT HESS: Now as Pablo mentioned, they're using the Direct3D interfaces to take and show things with Avalon and stuff like to get as high performance as possible, but they're not using the managed 3D interfaces. How does that message play out with, let's say, okay, on one side we're using managed and the other side we're not using managed. Should developers be concerned about that at all, or is that actually a good thing?

BRAD ABRAMS: Well, like with any new technology there is some staging of the technology where we want to be sane about that. And so there is certainly some parts of the system that are very low level, that are at the place where, device driver level, that kind of thing, where it's really not appropriate to write a managed code. But for all broadly used APIs, the first class way to access them is with managed APIs. So, for example, the Direct3D APIs that Pablo's talking about, you can access all those directly from C# or Visual Basic .NET or any of the managed languages.

ROBERT HESS: So basically then by the operating system taking very carefully using the unmanaged set of APIs, you're able to get as high performance as possible. That then allows the applications using managed APIs on top of that to really access as much powers and capabilities to the system as possible.

BRAD ABRAMS: Yep, exactly. We have both the power and the simplicity; it's really a great world.

ROBERT HESS: Okay. Now Chris, I know you've been doing an awful lot of work with what I'm trying to understand from a programmability standpoint how developers using these APIs to write great applications. You had a great demo during Jim's talk, the lap around Longhorn, that I know was really well accepted. What are some of the feelings you have about what developers should be taking away from the PDC and Longhorn?

CHRIS ANDERSON: Well, I think it goes back to echo what they've been saying, which is it's about managed code. It's about raising the abstraction layer on the platform. We have the new declarative programming language with XAML that'll allow you to write your application using XML and then we can compile that into an executable code. And so I think we're taking deep steps to try to make the simplicity of the platform be orders of magnitude better than it's been in the past. If you look at the last major platform that we had, which was the move to Win32, and you think about where the tools were and what the abstraction layers that we gave you and the languages that you had to work with, I think we've taken it to heart to have the developer be the primary focus of all of our processes internally. We have these huge reviews of all of our APIs that we're producing, and even the fact that we're getting the bits out at the PDC is really key because we want that early feedback from all the developers here. And so it's really looking forward to that.

ROBERT HESS: And Quentin, I know that WinFS is a real big thing, important to you, well WinFS is something that's maybe not quite as visual as is Avalon and some of the controls are. In my mind, I think WinFS is probably one of the coolest things about Longhorn, because it really finally allows the data that people work with to make an effort in their lives and access different applications like that. What are some thoughts you have?

QUENTIN CLARK: Yeah, Longhorn from where I sit working on the WinFS Team, we're really able to start talking about the kinds of innovations that developers are going to be able to do on the new platform. From our angle, that's where all of the excitement's been. There's been UI before, no offense.

BRAD ABRAMS: Hey, hey, not UI like this.

QUENTIN CLARK: Not UI like this. And there's been networking before, and there's even been things like file systems or databases, but WinFS fundamentally is opening up a set of new possibilities. One of the things I've been hearing a lot of over the past couple days here at the PDC has been, wow, will this mean that I can do this or that? Or what does it mean if I put this data here, what else can I do with it? So we're hearing people explore possibilities that we've never heard before when talking about storage inside the desktop.

ROBERT HESS: Okay. Now, how many of you have done breakout sessions already so far?

QUENTIN CLARK: Just you, right?


ROBERT HESS: So what did you talk about exactly, and what do you think the audience took away from it?

BRAD ABRAMS: Okay, I did the .NET Framework overview presentation, and there we talked about some of the cool stuff we're doing in the Whidbey release of the .NET Framework. And as you know, the Whidbey release ships in about a year to the V 2.0 release of the .NET Framework. In addition to that we talked about how the .NET Framework is a good lead-in to Longhorn and how your investments today in the .NET Framework kind of lead right in to Longhorn.

ROBERT HESS: Did you get a chance to talk to the audience after the talk at all?

BRAD ABRAMS: Yeah, I got a chance to talk to a few people, and the other thing that was really cool is I got a chance to read a couple of bloggers that right after the session they had already typed in detailed transcripts, detailed notes. It's great to see those messages that came through and that people are writing them down and you can access them. So I think some of the things that people really took away are the advancements we're making with ASP.NET, how we're really into that. The new support work that we're doing in generics.

ROBERT HESS: Now, did you discover that the audience maybe took away or needed something slightly different than what you came in thinking they might need? Did they kind of understand things a bit differently? Or did you learn anything, I guess is what I'm asking about?

BRAD ABRAMS: Did I learn anything? Well, I think one of the things that surprised me a little bit was the... I don't know actually.

CHRIS ANDERSON: Brad knew it all, you see. So he came into it...

ERICA WIECHERS: Can you give us an example of what kind of questions people were asking? Was there any question that just caught you off guard, or what's some really interesting things that people were concerned about?

BRAD ABRAMS: Well, one of the things that people were concerned about, but frankly, we did anticipate it a little bit, was people were just blown away by the demo that Chris and Hillel did around Avalon and the kind of UI support that you have there, and one of the things that we talked about, some people have already invested in Windows Forms and they're looking at that going forward, and they're kind of concerned about the transition there and if their investments today in WinForms will make sense going forward. So that's been a topic of a lot of discussions for a while. And one of the things that should be clear is that we have a great story for Windows Forms in Avalon, and if people want to start preparing for Avalon, the best way to do that today is to invest in Windows Forms. And Microsoft's gonna keep improving that over time. It's still the preferred way to build applications that have to run both on older OSs, Windows XP, Windows 95, as well as run great on Longhorn. And Avalon would be the best technology for you to write if you're able to run just on Longhorn and take full advantage of the experience there.

PABLO FERNICOLA: That was a message that we highlighted during the Avalon overview, that probably the best way to get it started right now is to go for Windows Forms, and then you can take your UI and add Avalon controls later on. Or if you do things through XAML, the markup language, you're also able to put your Windows Forms controls as part of that markup and they'll interop seamlessly, including things like data binding and security, we're able to have all those integrated. So the best way to get ready for Longhorn is, do your UI in WinForms, separate your UI from your data, and then you can hook up your different UI or augmented UI later on in Longhorn.

CHRIS ANDERSON: Well I think that, in typical fashion, we tend to get really focused on our own technologies. I think that's a great analysis of where we are in the UI space, but Longhorn has a lot more innovation than just the UI, and I think that's a key thing that, because Avalon's so flashy, people are kind of attracted to that look and they're wondering about it. But in reality all the innovations like WinFS and Indigo will be accessible via any managed code. So you can actually take your WinForms app and keep the UI in WinForms; you don't even have to de-invest in that if you are happy with where that's going. And you can leverage WinFS and you can use Indigo and you can do remote messaging, you can do all those things. And in fact, you can access a lot of this from your Win32 code. Using the advancements in the C++ language you'll be able to do seamless use of managed code from C++ code without any rewriting. And so you'll actually be able to use WinFS and Indigo from those places as well. And in fact, I think WinFS has native APIs in Win 32?

QUENTIN CLARK: Yeah. Yeah, the whole integration and support for existing Win32 APIs I think has been an important message we're trying to get across throughout the entire PDC. WinFS itself does support in fact the Win32 file I/O APIs like CreateFile and FindFirst FileIndex.

BRAD ABRAMS: Good that those keep working.

CHRIS ANDERSON: But directly in WinFS.

QUENTIN CLARK: Directly in Win FS.

CHRIS ANDERSON: Just call CreateFile and it works.

QUENTIN CLARK: Absolutely. You see a namespace that looks just like a Win32 namespace. If your Legacy application -- like just recently shipped, Office 2003 -- to managed code is a Legacy application kind of perspective. It calls in and can access it, thinks it's just the normal file system. And the benefits around having metadata and we're being able to relate things together and act on information are still there, even though those applications have still used Win32. So we're really trying to take a crawl, walk, run approach to letting people dive in and get the benefits out of Longhorn's surface area.


ERICA WIECHERS: So I want to know if each of you can answer this in your own way. What do you think is the best way developers can get ready for Longhorn? Pablo?

CHRIS ANDERSON: Managed code.

QUENTIN CLARK: Yeah, write managed code.

PABLO FERNICOLA: Managed code.


ERICA WIECHERS: Okay, you're unanimous. That's not fun, come on.

ROBERT HESS: I know one issue is that we're giving the attendees at the PDC some very early developer prerelease version of Longhorn that's...

CHRIS ANDERSON: Very, very early.

BRAD ABRAMS: Very, very early.

QUENTIN CLARK: Very, very, very early.

ROBERT HESS: Now, what's the real purpose of taking and giving such an early view of something that's gonna be out in, not very soon?


BRAD ABRAMS: Absolutely.

QUENTIN CLARK: It's all about learning from the people who are here, and the people that the bill will go out to as well, especially around betas and whatever else, to get their feedback to make sure that we're developing a platform that's gonna provide the benefits that we're hoping it does.

PABLO FERNICOLA: We really want to develop a central community for the developers out there. So one of the things we have is this Developer Center where people can go and read articles. We're going to be posting samples, we're going to be posting some executables over there. There's a set of newsgroups. So we want to have a lot of interaction, so people should feel free, they should go and participate. We want to hear questions, or if people want verifications of things. A DSDK is available online as well. So there is quite a bit of material and we want to post more material so people can sort of get a two-way interaction.

CHRIS ANDERSON: It's also interesting, because we're doing this so early, typically by the time we've released a beta we're pretty far along the development cycle. Here it's really early and the feedback can have a direct impact on the product. We'll actually be able to incorporate it into this version. It won't have to be next version.

ROBERT HESS: So if someone really felt it's important that we no longer had a white mouse cursor, but had a green one instead, we could actually get that feedback and include it in the system?

QUENTIN CLARK: We're here to listen.

CHRIS ANDERSON: Yeah, we're here to listen. We're here to listen.

BRAD ABRAMS: But seriously, I really think this is a historic moment that the industry finds itself in. It's kind of like if you roll back the clock when the Windows API was first being developed, who got a chance to give feedback into that? And we're still dealing with some of the problems. I've heard attendees here talking about some of the problems with the Windows API that we have to live with. So now we find ourselves at the very moment that we're giving birth to the WinFX API, and that's an API that's gonna have to carry us for another 10+ years. So now is the time when the whole industry should really get a chance to give us feedback on that. So my advice is if you're not gonna give us feedback, then you can't complain about it later. If I hear you complaining about the WinFX API and I didn't get feedback from you, then...

ROBERT HESS: So that means you're keeping a list of all the attendees of PDC, so if any of them complain, you kind of quickly look them up...

BRAD ABRAMS: Exactly. We have this log, yeah, yeah, we're storing it in WinFS. Yeah, the ones that can complain, yeah.

ROBERT HESS: So with the big changes going on for WinFX and its API from the past Win32 model, do you find attendees concerned about that, saying oh, suddenly now I've got to forget about Win32 and learn this brand new API system; why can't you just improve Win32 and let me use that properly?

QUENTIN CLARK: I think they're excited.


QUENTIN CLARK: I'm not hearing the, oh I don't want to; I'm hearing the, how do I best take advantage of, and I'm hearing more of that thought. I think they want to have an environment where they can be more productive, where their applications can run more robustly. They want to take advantage of these new parameters in the system and I'm not hearing any remorse over not being able to call -- well, you can still call CreateFile, but I'm not hearing any remorse about, (unclear) item thing.

BRAD ABRAMS: Right. Right.

PABLO FERNICOLA: I can tell you from the graphics side, for example, people are running into limitations themselves, so they're glad that we're removing those limitations. So if you look at GDI, for example, you don't really have good hardware acceleration compared to Direct3D. If you look at being able to have color pipelines that are more than 8 bits per channel, that's something that we couldn't do in GDI. Things like sub-pixel clear type, another thing that we couldn't do in GDI. Another direction we're moving to is having double-precision full coordinates for graphics, which is fairly important when you're doing the high end graphics on CAD, and that's something, again, you couldn't retrofit that in GDI; there was no way.

BRAD ABRAMS: One thing I have heard a couple attendees talk about is they're on VB 6.0 today, and they haven't been able to move over to VB.NET yet or they're still writing MFC apps and haven't been able to move to the .NET Framework, and we're really taking time, especially with the Whidbey release, and thinking about what are the pain points involved. Like for the Visual Basic developer, we've added Edit and Continue back to the Visual Basic language, and that was a huge blocker to a lot of our Visual Basic customers. So getting those people over to the Whidbey version of the .NET Framework really puts them on the right course to getting to WinFX when it releases.

ROBERT HESS: So moving forward, what do you kind of think are some of the major things that people are gonna focus on from an application development standpoint? Are we gonna be seeing new types of applications coming out for Longhorn, or are we seeing just better versions of the same thing we've already got?

QUENTIN CLARK: No, I think we're hoping for the former. Gord said it best. People ask us, what's the killer app that's gonna be built on Longhorn? We don't know, but it's in the back of someone here's brain. Something really interesting that's never been done before that the platform now opens up the possibilities. It's just like going from Win16 to Win32. No one knew what the killer app would be. No one picked at Outlook, for example, when we were building Win32, but there it is. I was actually talking to Will Kennedy the GM for Outlook, he's like, couldn't have built it on Win16, because the OS systems were just not there enough to make it possible. So we're hoping for the same kind of phenomenon.

BRAD ABRAMS: Yeah, that's one thing that's so exciting about this time. This is a unique moment in history where we're seeing this inflection point happen.

ROBERT HESS: Now, do you think that the market is ready for applications to finally come back into their own again after this big long time of having Web pages being the main things people focused on? Are you seeing kind of this tightness spring ready for the application style?

CHRIS ANDERSON: People have suffered enough with raw html pages. They're ready to get back to real applications that they can use, and yeah, and I think you see it. I was just completely shocked at the reaction we got from Amazon. Because they were firmly committed to the Web; that's the place they wanted to be. And when we showed them the stuff in Longhorn, they just like, wow, this can solve our problems. This can give us a better user experience. And I think users are getting tired of Web pages that don't refresh right, and not being able to run offline, and all of these things. So I definitely think it's the time.

PABLO FERNICOLA: One of the key reasons that people went to the Web in the first place is the ease of deployment. Deploying a Win32 client application is fairly hard, it's fairly costly. Deploying a Web page you just navigate to a URL and you're set. So we're going to bring that simplicity to the platform in Longhorn. You will be able to navigate to the URL, your application will get installed, you will be able to have it offline, and you will be able to take advantage of all the functionality that's on the client, something that you couldn't do for html.

ROBERT HESS: Now, that brings up a good point because when you browse to a Web page, technically I can understand that since a Web page really can't access your computer at all, it's basically perfectly safe. So navigating to it and having that stuff coming up is perfectly fine, whereas we're now talking about creating smart client applications that can, through WinFS, get real rich deep access to your operating system, the pixels on your screen, everything else like that, where a Click Once style deployment sounds a little on the scary side to me. What are we doing to actually pay attention to make sure these new smart client applications that are as deployable as Web pages really are secure and can't do anything bad?

PABLO FERNICOLA: We're doing quite a bit of work on something called the Security Execution Environment, and it's basically a security model that's built on top of the Framework Security Environment. And it's something that applies both for applications that are locally deployed as well as applications that you access through Click Once. And it will enable you to have full access of functionality and the performance for the machine, but still do that in a protected way. Security is something that for the past three years that we've been working on Avalon for example, from the beginning security's something that we've been looking at.

ROBERT HESS: Now, Quentin, since you've got WinFS, which is kind of the data model, which I think of for my mind, that's what I want to be secure. What specifically is WinFS maybe doing to make sure that these applications are secure?

QUENTIN CLARK: Well, we have two big advantages. The two advantages are: we get to bear the full weight of the CLR security model and the framework model on it, as well as NTFS. So we don't have to go out and invent an ACLing system or a code execution environment system; we're just taking advantage of the fact that we're marrying together the CLR relational engine and the file system to create the data model in the first place. So we've had to design that in from the very beginning. Every line of code has to consider how we're taking those benefits in, using these technologies to ensure that the environment is in fact secure.

ROBERT HESS: Chris, you have anything to add? Put you on the spot here.

CHRIS ANDERSON: It's all great. No, on the security I think it is definitely pervasive and everybody is thinking hard about it. All the recent press is that the timing of the, kind of the security crisis across the industry that we're all facing has come at, for us, actually a relatively good time. It's a time when we were already looking deep into our system and looking at how we're gonna innovate for lots of other reasons. And so as we've gone through and looked at the basic building blocks, the file system, the UI presentation system, the networking stack, we're revisiting those and we've taken security as a top-level item for all of those. It's priority one in all the pillars. So I think it's great.

ROBERT HESS: How tough is that ourselves, internally, to focus on security and understand what the security problems might be in some of the code we're writing, and what are we doing to resolve those?

CHRIS ANDERSON: It's very interesting because it moves from a model where features are your primary goal to where security is a primary goal. And I think it's going to be very interesting to see what customers react to it because there will be less features because we've spent time with security. Obviously we have a lot of features. If you look at the stuff we presented this week, we have a lot of features. But yeah, in the end there will be less things because we have to work on the security of it. But I think that's what we have to do. It's our obligation because of the size of our platform, the number of installed users, the level of those users. My mom doesn't want to worry about the security of her PC; it has to just work.

QUENTIN CLARK: I think a lot of us feel badly about some of the past years with the security things.


QUENTIN CLARK: And a lot of us have really taken it to heart. In fact, I don't know anybody right now that doesn't worry about it, that isn't concerned about it, doesn't think about it. The past years' events around security have really sort of shaken the tree a little bit and got us all very motivated around there.

BRAD ABRAMS: Yeah. It's definitely been a culture shift I think at Microsoft. It's kind of at a grass roots level even, the way people just interact. You're in code reviews with people, you're doing spec reviews, and the security topic comes up, no matter what you're doing.

CHRIS ANDERSON: I don't know if you've had Michael Howard on your show before, but you should definitely get him on. He has great stories of how hard it used to be for him to convince teams to fix security issues, because they would be things where they would look at it and they're like, well, the ramification isn't that bad, it's okay. Maybe it's a DOS issue or something like that. But now that we've hit this where it's really brought to the attention of what the ramifications are, that's never a concern anymore. All the teams, the immediate reactions are: we have to fix this. There's no way you can punt this bug; we must take it. And that's a huge culture shift. Yeah. Security's not an optional feature.

ROBERT HESS: Well, I think we're probably about out of time right now, but I'd like to get some brief input from each one of you of where you think some of the PDC attendees and the non PDC attendees should be thinking next to go to for resources on how to move forward writing their applications, getting ready for Longhorn, adding more security to them and stuff like that. Do you have any suggestions on resources and avenues for development?

CHRIS ANDERSON: The Longhorn Development Center, that's the definitely the key place to start. That's where we're gonna have links to all the content we find. It'll have links to the relative MSDN content as well as third party community stuff. I think that and the newsgroups are gonna be the key area to focus on.

BRAD ABRAMS: Yeah. And especially for people that weren't able to come to the PDC, I think you can get a sense for not only the technology that we've talked about, but also the atmosphere and the buzz by going to, I think it is, and there you can kinda see just what attendees think about what we're talking about and kind of hear it in their own words.

QUENTIN CLARK: Yeah, the blogs have been a significant, different phenomenon at this year's PDC. You're able to get this real-time feedback; you walk out of the session and them boom --

BRAD ABRAMS: It's there.

QUENTIN CLARK: You're like, okay.

CHRIS ANDERSON: It's not on an evaluation form; it's on the Internet.

BRAD ABRAMS: Yeah, publicly. Yeah, exactly.

PABLO FERNICOLA: It is important to stress that Longhorn is still quite a while away, so people should really spend their time right now focusing on Windows Forms, learning that technology and applying it to their applications that they're going to develop and deploy between now and the time that Longhorn ships.

ROBERT HESS: And managed code, right?

QUENTIN CLARK: And managed code.

BRAD ABRAMS: And managed code in general, right.

CHRIS ANDERSON: We already gave you a unanimous answer last time.

ROBERT HESS: Thank you guys for joining me. I'm glad to have this chance to talk with you, and hopefully our audience will understand a little bit more of what we were kind of talking about with the PDC, with Longhorn, the technologies, and managed code.



PABLO FERNICOLA: Thank you very much.

ROBERT HESS: So, hopefully that gives you some insights on what we've been talking about at the PDC, what some of the attendees have been hopefully learning about Longhorn, and advanced developers are doing. And now in our next segment we're gonna actually talk to some attendees to find out what their viewpoint is. But we'll be back right after this short break.

In this special "Public Service Announcement" from the Red vs. Blue team, we learn how the PDC could bring about world peace... yeah... right...

Robert Hess meets with Richard Hale Shaw, Simon Stewart, and Robb McLarty to hear an attendees perspective about the PDC.

ROBERT HESS: Welcome back. Now, earlier we talked with some people from Microsoft about why they came to the PDC, what they intended to actually communicate to some of the attendees here and what they thought was important about Longhorn and Longhorn development getting forward. What I've got with me now is some of the attendees of the PDC. I want to take and find out from them why they came to the PDC, what they found's been really important for them, what they learned, what they discovered, and what sort of things maybe they still have some questions about. What Longhorn might do and what the time between now and the Longhorn shipment they plan on doing for focusing on their development efforts. So what I'd like to do now is have each one of my guests introduce themselves, let you know who they are and what they're doing. Why don't we start with you, Richard?

RICHARD SHAW: Sure, thank you. My name is Richard Shaw. I have a training consulting firm in Ann Arbor, Michigan. We do .NET and UML and XML training across the U.S. and other parts of the world as well.

ROBERT HESS: And why did you come to the PDC?

RICHARD SHAW: The PDC is for us a chance to see, obviously, where Microsoft wants to go over the next couple years, might go, sometimes they go there and sometimes they don't go exactly there, but the idea is to see where they'd like to go. It's also a chance for us to reconnect with old friends and see a lot of people that you just don't always get to see on a regular basis. We were just out, a couple of colleagues of mine and I, having coffee out there and see Anders Hejlsberg going walking across the room and hailing him. All praise A. It's very exciting to see guys like that and be able to talk to them and ask them routine questions right there on the spot. A lot of fun.

ROBERT HESS: So essentially since your livelihood is based upon helping other people understand how to use Microsoft technologies or the development services in their own work, you need to make sure that you're understanding where we're going so you can try to be either one step ahead or at least in lock step so that your customers can get more benefit out of what you're providing them?

RICHARD SHAW: Would you like to write my marketing copy for me in the future like that? That's great, Robert. You're exactly right. And it's critical, I think anybody, even end user firms who are going to keep track of what Microsoft's doing, at least touch base with what Microsoft would like to be doing. And that's I think what the PDC is all about, to give you like a two year, here's where we'd like to be a couple years down the road picture.

ROBERT HESS: Okay, thanks.

SIMON STEWART: My name's Simon Stewart and I work with a company called ID Tech in South Africa. I'm one of the chief architects, and I think the most important thing for me, as far as coming here and seeing what goes on, is really to see what Microsoft wants for the future and also the products that they want to use to actually get them there. That's been the main things.

ROBERT HESS: And what sort of development stuff do you do work on right now?

SIMON STEWART: Primarily it's .NET, custom development in .NET, mainly in C#, also some C++. So, yeah, the same as Richard, seeing some of the new stuff and also hearing guys like Anders Hejlsberg talk about the new enhancements. And really seeing it from the creator of the language has been fantastic.

ROBERT HESS: Okay, thanks. Thank you. And you are?

ROBB MCLARTY: I'm Robb McLarty. I work for an investment firm in New York. I'm not a developer at all, but I somewhat understand the technology, and I'm here just to get a feeling for what Microsoft is doing, and also to understand what the implications might be from an economic perspective.

ROBERT HESS: You say you're not a developer, but still it was important enough to you to come to a developers' conference to understand these sorts of things. How does that actually play out? Not being a developer, why come to a developers' conference?

ROBB MCLARTY: Because this is the first glimpse of Longhorn, and you get so much more information, you get different information going to a business conference than you do going to a developers' conference. It's a different perspective, and it's a different way to get valuable information, by not only by talking to Microsoft people that are technology people, but also real developers, people that are using tools and putting the Microsoft implementations out into the world.

ROBERT HESS: I know that at the PDC here this would be the first time we actually have unveiled some of the aspects of Longhorn stuff like that to people. I believe Richard's probably seen some of this stuff because he's got some good contacts inside of Microsoft and being of a press kind of person that helps him out quite a bit, but the rest of you probably really didn't know other than what you've read in blogs and stuff like that. Could each one of you maybe kind of briefly say what your expectations were about Longhorn, and maybe some of the things that surprised you that you saw here at the conference?

SIMON STEWART: Okay. Yeah, I think what I was expecting was perhaps more focusing from the service side and what they were gonna do from things like IIS and more stability from the actual core engine itself. But the front end stuff's really good. Some of the new stuff they're doing and also the fact that they're opening it up. There's nothing worse than seeing some really good functionality but you can't actually use it. So, that's been good to see that focus from Microsoft.

ROBERT HESS: So from a development standpoint, most of your own needs are server-based, and your company is taking focus more on working on things on the server, and that's why you're wanting to focus more on the server aspects?

SIMON STEWART: Yeah, everyone needs a server if you're going to do enterprise applications. The front-end's nice, but if the back-end falls over, then it doesn't really count for anything.

ROBERT HESS: Yeah, yeah. Now, from a client side perspective, what you're seeing Longhorn providing, do you see anything there that meets well your needs on the server side of things?

SIMON STEWART: Yeah, I think what they've done with Avalon is fantastic, especially some of the data, the relationships, ibeing able to set up relationships from inside Windows. That could open up a whole new avenue of development. And also the navigation, the Carousel control, that looks really good. Yeah, I look forward to seeing it in a real world application.

ROBERT HESS: Okay. What about you, Robb? What are some of the things that kind of caught you by surprise when you came here? You said, wow, that's some kind of exciting. I wasn't really expecting it.

ROBB MCLARTY: WinFS was interesting. I was just having a conversation about that. Taking the tack that I do, where I look for how it might impact other businesses. I'm still trying to figure it out, but it sounds like WinFS might be consuming a lot more hard disk than regular Windows 2000 or anything that uses NTFS. So that has implications on the industry in certain ways.

ROBERT HESS: In other words, better understanding how WinFS is gonna actually be utilizing the disk resources it has is kind of important. Because clearly right now you've got the data file and then you've got the FAT table you're dealing with, and very little footprint on top of that perhaps adding to the metadata, where WinFS has got a lot more metadata that's coming into the picture that's gotta be stored someplace.


ROBERT HESS: And in order to make that metadata as rich and useful as possible for the search inquiries, that means more and more information needs to pull to the metadata, because you clearly don't want to be searching through the files every single time you do a query. And so I think that's where a lot of the performance comes in. All I can say is, fortunately the hard disks are getting cheaper and cheaper these days.


RICHARD SHAW: Robert, along those lines though there's an interesting point you're raising. If you think about side-by-side execution in .NET, which is now an old story in a lot of respects, right? I guess for a lot of customers it's still a new story or if they haven't discovered it yet, but for the early adopters it's an old story. The implication was that hard disk storage space -- this goes to Robb's point -- hard disk storage space is actually relatively cheap, and our advice to customers is, except in the embedded or handheld environments, if you can't afford hard disk space you're in the wrong business, because it is the cheapest commodity out there. I would rather have them chew up my hard disk space, and maybe I've got to readjust my future assumptions for that kind of space, in order to get that kind of rich information. We've already seen in managed code development the benefits of having rich metadata that permeates the entire system or just overwhelm any of the possible negatives you could have as fallout, so I would say, please let them chew up my disk space; I've got bigger problems to solve trying to ship product to customer, and metadata's gonna help me with that.

ROBERT HESS: So if you can get value and/or performance by requiring a larger hard drive, money well spent.

RICHARD SHAW: Yeah, that's exactly right.

ROBERT HESS: So what have you seen at the conference here maybe that caught you by surprise or has got you kind of excited about what Longhorn's providing?

RICHARD SHAW: I think overall, and long ago under NDA I heard people mention that Longhorn was kind of .NET becomes Windows or something like that. If you think about it, the managed environment explodes into the operating system and the implications of what you can do in a managed environment, which goes to your earlier point about, we ought to be writing managed code. I think we were talking about this earlier.


RICHARD SHAW: If the people aren't writing managed code by the time Longhorn comes out, they're just not gonna be able to get the benefits of it. If you start thinking about the implications of a managed code environment and then how that permeates the entire operating system, it's very rich. There's lots more opportunity than just saying, well, are we running Win 2K or Win XP with the .NET Framework installed? It goes way beyond that. Historically though, it's interesting to me because this is one more step on a long path for a lot of us who have been around the business for a while towards Cairo. So we're getting closer to the whole idea of what used to be called Cairo, code name from 10 years ago, a fully object-oriented operating system, but it's beyond that because you have this rich type information and far better services than any of us ever imagined back then that makes it easier to build lots of different types of applications.

ROBERT HESS: Now, talking about managed code, since you do some development service for people and you're dealing with trying to get people to understand managed code better, do you find that people are really understanding the benefits of managed code, or is that something that we still need to do a better job of educating people on?

RICHARD SHAW: It's a good question and the answers are yes and no. Yes, they are understanding managed code once they have made the commitment to move at least to the .NET Framework, although a lot of time that's motivated by the fact that somebody somewhere has made a decision in that organization for them, not necessarily the person we're talking to, has decided this is a good thing. Once we or somebody qualified shows them what you can do in the managed code environments, you won't ever have a developer saying, I want to go back. They'll be screaming if they're forced to go back to work in an unmanaged environment.

And so the tricky part -- and this is where all of us have to work harder at getting the message out there as what the benefits of managed code are and why we need to get as many developers as possible who are still working in unmanaged environments like Windows without .NET or Visual C++ prior to .NET, or VB 6.0 or any of that. They are spending so much time and effort doing workarounds for things that someone in a managed environment does with a wave of a hand. And it's really hard to express the major productivity gains somebody in those older environments would get if they moved to managed code; they're not gonna get them staying there. But it's kind of like being in a country where people don't have any feeling in their hands. You've got hands, you tell them, look at these great - look at all these things I can do with my hands, and they have no idea what you're talking about. It's tough.

ROBERT HESS: Yeah, that might be interesting for Simon and Robb maybe to respond to. How is managed code working its way through your companies, and do you have any feeling for what's going on there?

SIMON STEWART: Well, yeah. We've been using .NET for about three years now and I agree with Richard; we just couldn't go back. I think things are moving so fast now and it's gonna be so much real deep integration into the operating system going forward that if you don't make the jump now it's gonna be really, really difficult to do that. But yeah, there are so many benefits and it's been fantastic. We've got Perl programmers and Java programmers that have migrated to C# and they've really stuck with it and that's really their favorite language now. Yeah, we're sold.

ROBERT HESS: Okay. What about you Robb? Are you aware of any managed code aspects inside your company at all, or is that not something you're dealing with an awful lot?

ROBB MCLARTY: Not really. I know that we have developers programming in C# on the .NET Framework, but I've no clue as to anything really beyond that.

ROBERT HESS: Now, personally do you understand what managed code is and why it might be important?

ROBB MCLARTY: I understand the benefits in terms of garbage collection and improved security, but I don't have a great understanding.

ROBERT HESS: But from a non developer standpoint, understanding those two things, those are really kind of important.


ROBERT HESS: Because having garbage collection...

ERICA WIECHERS: It's probably more than I can say. That's good.

ROBERT HESS: I think that's great. I think as long as people are focusing on it. And quite often I'm surprised when I do developer events and I'm talking to people, whether it's developers or nondevelopers, and they say, managed code, I just haven't quite got into that. It shocks me. Maybe it's just because I'm living and breathing it so much, and you guys are too obviously, that when someone doesn't understand what managed code is it's kind of strange.

RICHARD SHAW: It's the message that I think Microsoft - if you talk about your original question about, have we been successful or not getting the message out about managed code, the answer is no in terms of Microsoft's marketing. But it's a tough message to get out there, so I'm gonna give Microsoft marketing a freebie that they can put out there right now, and the message they should give to their customer is, if you don't move to managed code, your competitors will.

ROBERT HESS: Yeah, yeah.

RICHARD SHAW: I think that's it right there.

ROBERT HESS: Yeah. Yeah, that's a good statement.

ERICA WIECHERS: So, along those lines, is there anything at the conference that you haven't seen enough of that you wish that you maybe had a little more guidance or information on?

SIMON STEWART: So far, no. I've seen a lot. Perhaps Indigo, just understanding a bit more about where that fits in, and also how we can leverage Office. But so far it's been great. Learned a lot, seen a lot of people, and that's really part of the big thing of coming here.

ROBERT HESS: What are some of the main sessions you've been to? You mentioned Indigo and stuff like that.

SIMON STEWART: Yeah, I've been to som Don Box sessions that have been great.

ROBERT HESS: Did he sing any songs there?

SIMON STEWART: No he didn't. He spoke about his mom a bit. And also some Anders Hejlsberg sessions. And that's been my main focus really. But also some of the productivity enhancements in the IDE have been pretty impressive. Also understanding the tradeoffs that they made, why they did certain things, why they didn't do certain things, and it's good to really hear it first hand.

ROBERT HESS: What about you, Richard? What are some of the sessions you've been to?

RICHARD SHAW: Actually I'm gonna use this opportunity to plug INETA, if I may.


RICHARD SHAW: Because I'm an INETA speaker, but more important, is that for a lot of people that don't know, INETA is the Independent Association of .NET user groups, and there is a speakers' bureau consisting of some unbelievable people -- you just go to the INETA Website and take a look -- who can come in and give talks, either at your organization or at your user group. And for fun, Carl Franklin, myself, George Bullock who's a Microsoft guy, Jeff Prosise, another INETA speaker, Shawn Wildermuth, also an INETA speaker. All of us got together last night and put together the INETA Speakers Band and we kicked - I can't say what we kicked, that wouldn't be any good on camera - but we kicked - we rocked.

ERICA WIECHERS: I heard about that.

RICHARD SHAW: It's a great, great band. I dare say you can even have the INETA Speaker Band come and play and give you great technical talks if you just ask INETA. So go to for that.

ROBERT HESS: So it sounds like you've spent most of your time just having fun then?

RICHARD SHAW: Well, some of this I've seen before. So yeah, I'm doing some social things too, but it's woven into the technical fabric, right?

ROBERT HESS: Yeah, yeah. Yeah, yeah. And Robb, what about you -- have you been to any sessions and stuff like that? And if so, which ones?

ROBB MCLARTY: I've enjoyed the general sessions probably the most, because a lot of the breakouts actually get into code. So I tune out a little bit there. I can follow, but I find that the general sessions focus more upon the big picture, which is more where I'd be also. That's where the greatest benefit has been for me.

ROBERT HESS: What are some of the key takeaways that you took from some of the general sessions specifically?

ROBB MCLARTY: It was interesting to hear this morning's talk with respect to Microsoft Research. Some interesting applications in terms of Tablets. What else? I enjoyed Bill Gates' talk, obviously, the one that kicked it all off. That was really great. Again, mostly just the bigger picture of Longhorn and what it could mean.

ROBERT HESS: Now, one of the reasons why we're giving people such an early glimpse of Longhorn like this -- because we're still several years away from shipping, which creates some strange marketing issues and problems like that, opening up the kimono so early like this -- is we want to make sure that Longhorn had the features and capabilities that all you guys are wanting, and each one of you kind of represent a different market segment to a certain extent. Can you kind of let us know what you think you would like to see in Longhorn? If you can just give me a brief description of what Longhorn would be, what would you say? Well, gee, I sure hope they're gonna solve this problem. Maybe, Richard, you can start out.

RICHARD SHAW: Wow, I don't know if I can give you a really accurate answer there other than I think staying on the path to finding as many ways to exploit the managed code environment in the context of the operating system, and maybe not even stopping with where they are. I think that would be probably the best general answer I could give, because there's so many different ways that this could be applied and I don't know if that's even been thought through completely. I like what I've seen so far.

There is always the concern that what you're seeing however, is still gonna change a lot between the time you're seeing it at a PDC and two years from now when it ships, because there'll be the approximate ship date that's probably been announced this week, and then there'll be the usual eight month time lag between the time that they say it's gonna ship and it really does appear. So many things can change. Sometimes I think the vision that's announced with so much energy at a place like this gets lost between now and then. And I would say stay true to that vision as much as they can. Try not to make too many compromises strictly for the sake of trying to ship something. Microsoft's best work has been done when they have taken the hits in the press, delayed something, and done the right job for the customer rather than saying, oh we gotta ship, we gotta ship and then they make a lot of shortcuts and leave things out or cut something short that's really not as good as it could be. And again, the .NET Framework actually is a good example of that. There the development teams were allowed to call the shots and say, we're not letting this out until we think it's where it should be. I think that made a big difference. I hope the operating system teams will do the same thing.

ROBERT HESS: Yeah, that's actually very well taken because it takes an awful lot of time to develop an operating system and Longhorn has an awful lot of stuff in it. On one hand, I'm really surprised at how well it worked today; my team has been running a lot of the demos and stuff on it. While there are some rough spots, it works. I mean, I can write code in C# and I've got Avalon things and XAML that I can do on it that's kinda cool, but I know there's still an awful lot further to go on this. What about you, Simon? What would you like to see in Longhorn?

SIMON STEWART: Well, I think some of the main things are definitely performance, and obviously also the stability on the platform. Security is also really important, and it's good to hear Microsoft talk about that, and some of the things they're doing and also the book that they gave us as well. But I think what's also really important for me is what they advertise that the operating system can do that I can actually do. Adding things to the sidebar, I want that to be really easy for me. I don't want to have to write the plumbing if Microsoft's done it. It would be really good if I can just plug into a DLL. Currently there's some workarounds you have to do and lots of third-party components, but I think definitely, if I can do what they can do, what they're showing at the conference, then I'm happy.

ROBERT HESS: Yeah, I know one of our goals is to try to make programming for Longhorn as simple as possible. The days of having to write tons and tons and tons of COM code is hopefully out the window. While you can do an awful lot of power with COM code it still took some work to get that sort of stuff done. I don't know whether you saw Eric Rudder's talk or not, but in there he was talking about how we did some code examples of doing a security-type application that was connected to the Internet, and here it is with 50,000 lines of code, and then we take the modification now we got it under 25,000 lines of code, and now we've got it down to three lines of code doing the exact same functionality. And so that's what we're trying to constantly do. I think .NET kind of shows what we really can do. We say, okay, we really don't need to make programmers write all this code, do we? They're doing it over and over and over again; why don't we just do it once, get it out of the way, make them make one call, set a couple properties and have it be done? So I know that's hopefully part of the goal of this. What about you, Robb? What would you like to see Longhorn do to solve some of the problems your company has or just maybe some cool, consumer kind of issues too?

ROBB MCLARTY: From a user perspective, I'd like to see faster and fewer reboots. I used to work in the telecom space; I actually used to design for voice switches which were very old machines but they worked like a charm. If one ever went down it would come up, another side would switch in immediately. And so I think that if I can say anything to Microsoft is, look towards other places in the technology realm such as telecom where they have had a higher bar for reliability and perhaps performance, and look to them for a lead. Because as a user it kills me to wait sometimes a minute to get my PC to boot up. That's not the case all the time, but I think that that would really improve the experience if it would just come up a lot faster. It's a simple thing but I think it's valuable.

ROBERT HESS: The problem is there's so much stuff that's going on during boot up, whether it's reconnecting to the network or verifying your authentication or stuff like that. It's almost mind boggling that it takes as short a time as it does, but still it's frustrating when you turn your machine on and it just takes so long to happen. I know that that's a big problem that a lot of people.

ERICA WIECHERS: Just be able to flick a switch.

ROBERT HESS: Yeah. Yeah. So I think we're almost out of time now but I'd like to kinda get some quick feedback from you maybe. How would you like to see Microsoft communicating information to you, like a PDC information, like the .NET Show and stuff like that? Help me better understand where we're planning on taking our platform and stuff like that. Are there ways we can improve where we communicate to you your specific need for the platform?

ROBB MCLARTY: I've been pretty happy so far. I've just recently gotten onto MSDN and watched some of your videos. So I'll continue to do that. I think that using blogs and feeds are really useful.

ROBERT HESS: So you get online and read blogs and stuff like that of what people are doing and stuff?

ROBB MCLARTY: Yeah. So I find it pretty helpful and I don't really have any points for recommendations.

SIMON STEWART: Yeah, I think blogs are also useful, but I think it would be good for Microsoft to maybe have more interviews with people and really discuss why they did certain things to really get that information out into the community. But yeah, I think definitely, just keep putting the developer first, and yeah, just push the information to them.

RICHARD SHAW: I would say along the same lines, any way that any of this can be, if it's not already scheduled to be part of MSDN Universal or those packages to put it in there. A developer with MSDN Universal is basically armed to the teeth to do anything they need to do on a Microsoft platform. They don't need to lower the price, but you always want to just find it in there; you don't want to have to go looking anywhere else. That ID should be able to get you any information that you could want to get that if it isn't in your current DVD set, it's on the site. As long as that stuff is packaged where MSDN subscribers can get to it easily, I think that's the key. I think it's worth asking developers to pony up that price and then packaging it and packing as much in there so that we can get to it very, very easily and quickly.

ROBERT HESS: Okay, okay. Well, thank you. I really appreciate the time you've spent here with us to talk about your thoughts about the PDC and Longhorn and stuff like that. There's still another half day going on of sessions, and the sessions tomorrow, so you can all attend the symposiums and the panel discussions going on. So, thanks for joining us.

ROBB MCLARTY: Thank you.

RICHARD SHAW: Thank you, Robert.

ROBERT HESS: So hopefully that gave you an idea of what Longhorn is, what the PDC is, and what the two different sides have kind of come up out of it. We had Microsoft people talking about the technologies and what they hope to communicate to people at the PDC, and then we heard from the attendees of what they hope to get from the PDC and what they plan on doing moving forward with their own development efforts. Thanks for joining us.

Erica Erica Wiechers, Program Manager, Microsoft Corporation interviews Don Box, Architect for XML Web Services.

ERICA WIECHERS: With me today is Don Box, an architect at Microsoft. Don, welcome.

DON BOX: Thanks for having me, Erica.

ERICA WIECHERS: Good to have you, Don. So why don't you start by telling us what exactly you do at Microsoft, what are you an architect for?

DON BOX: So, I'm an architect on the Indigo Team. I work on both the protocol stack and the implementation of that for doing messaging using XML on our platform.

ERICA WIECHERS: Okay. Can you get into some specifics about what you're working on? Any projects?

DON BOX: Sure. Right now, except for the PDC, which I'm obviously at right now, right up until now and of course when I get back next week, we've been working on the milestone, which is gonna be our Beta 1.0. And we've just been doing a lot of simplification, a lot of reduction in surface area making the product simpler to use for everybody.

ERICA WIECHERS: So, Don, how long have you been at Microsoft?

DON BOX: I've been at Microsoft for 20 months.

ERICA WIECHERS: Twenty months exactly.


ERICA WIECHERS: So, you're fairly new.

DON BOX: I am relatively new, yes.

ERICA WIECHERS: So what did you do before Microsoft?

DON BOX: Before Microsoft I traveled the globe with a bell telling people about COM. I would walk up and down the streets and declare, IUnknown, IUnknown, IUnknown.

ERICA WIECHERS: So tell me about your fascination with COM, where'd that start?

DON BOX: It started on August 8th, 1994. At one moment, COM just became pretty obvious to me. Before that I was always in a complete haze and knew nothing about it. And I kind of figured it out on August 8, 1994, and I just kind of fell in love, and so I've been doing that for a long time.

ERICA WIECHERS: Is COM still near and dear to your heart?

DON BOX: Yes, we're basically rebuilding COM one little brick by brick at a time.

ERICA WIECHERS: So you've been doing technology all your life basically?

DON BOX: Yeah. Yeah, I've been doing software since about 1984.

ERICA WIECHERS: So, what do you like to do outside of work?

DON BOX: I love working to get senior executives to do new things in their keynotes. This year I was able to get Jim Allchin to come out of his shell and confront the fact that he loves using VI. I plan for the next PDC, which may be 2005 or 2006, my goal is to get Steve Ballmer to get up and program in Edlin, which is the original line editor for DOS.

ERICA WIECHERS: Okay, beyond work though, what do you like to do outside of work?

DON BOX: Erica, that's not work, that's what I like to do. But I have other hobbies too. I have a band called Band on the Runtime, and we've been playing together for a couple years now, and we get a chance to play at most Microsoft events, and that's been pretty fun.

ERICA WIECHERS: Can you give us some examples of your song titles?

DON BOX: Sure, we have -- let's see, this year we added a bunch of new tunes. We did "One More Version" to the tune of "Like a Virgin. " We did "On Longhorn" to the tune of "On Broadway." Oh, and I sang -- I'm doing some recruiting, I'm trying to get Miguel de Icaza to join us at Microsoft, and I sang a love ballad to him, which was of course called "Miguel."

ERICA WIECHERS: Okay, do you think he'll appreciate it?

DON BOX: I know he appreciated it. He's told me privately many times that he was moved.

ERICA WIECHERS: So, can folks out there get your music out on CD?

DON BOX: We don't have a CD yet, but there were many A&R people, because we're in L.A., you know, what are you gonna do? There were a lot of scouts out there and we'll see what happens.

ERICA WIECHERS: And can you tell us what you see about your future at Microsoft? What are some things you'd like to work towards?

DON BOX: I really don't know. Right now I'm pretty heads down trying to get Indigo out the door with 250 of my closest friends. I really don't know what I'm gonna do next.

ERICA WIECHERS: And what is your favorite thing about working at Microsoft?

DON BOX: My favorite thing about working at Microsoft is the people I work with.

ERICA WIECHERS: Good crowd. Well Don, thanks so much.

DON BOX: My pleasure, Erica.

ROBERT HESS: And that brings us to the end of another episode of the .NET Show.

ERICA WIECHERS: And to continue our Longhorn coverage, next time we'll bring you an overview of Longhorn.

ROBERT HESS: And until then, we'll see you on the Web.

Top of Page Top of Page