.NETDJ: How did you come to work for Microsoft on the ASP.NET team?
RH: I was originally on what was then known as the Developer Relations group at Microsoft. I was a technical evangelist. I was part of a team of folks that Microsoft would send to the top 100 "media metrics," which was a ranking tool that ranked the top Web sites and destinations during the dot-com boom.

My responsibility was to go and talk to companies and get them to adopt Microsoft technologies. We talked to Amazon and a whole slew of others that are no longer around. We went after the top sites not running our technologies.

.NETDJ: I was going to ask this question later, but it seems like it might be appropriate now: what is your favorite Web site running ASP.NET - the site you think is the best showcase for ASP.NET technology?
RH: The http://blogs.msdn.com site is a great showcase. Also, any of the community sites - ASPAlliance (http://aspalliance.com) or DotNetJunkies (www.dotnetjunkies.com), for example - are really great samples because they are always trying to do cutting-edge stuff. As far as consumer sites, every customer has at least one unique thing they are doing that other sites aren't doing.

We're starting to see more and more applications being developed where developers are writing Windows Forms applications and then hosting Web Forms inside of them. Now that's an interesting trend!

.NETDJ: It definitely is, but I digress - what happened when you were trying to convince Amazon and the other dot-com companies to move to ASP?
RH: .NET was in the works at that time - Project 42, as it was known. I was the first XSP evangelist - ASP.NET was originally known as XSP. I was to go and tell them about ASP.NET before it ever saw the light of day. I was going out and talking to folks giving them presentations - under NDA - all leading up to the launch at the PDC in Orlando.

Scott Guthrie approached me in 1999 or 2000 about trading in my evangelism role and becoming a program manager on the team. As an evangelist, I was also a programmer and I was concerned that becoming a PM would bring an end to my coding. Guthrie convinced me that that was very much not the case!

I joined the team shortly after we launched - when there were only 18 of us. Now there are hundreds of people on the team, which includes ASP.NET, IIS, etc.

.NETDJ: What is a typical day in the life of a Microsoft Program Manager?
RH: A year ago, when I was still in Redmond, a typical day would begin by coming in early (around 7 AM) and until about 10 AM, I would just be reading: reading e-mail, reading documents, reading specs, or reading bugs and checking them against our specifications. This would be followed by just a little application development.

 
Rob Howard - The ASP.NET evangelist

After this - from 10 AM to about 3 or 4 PM - it was meetings: design meetings, informational meetings, technical presentations, etc. In Program Management, there are design meetings almost daily for arguing about features.

.NETDJ: Was the decision to implement Web services via ASMX pages the result of some of those arguments?
RH: I think we were looking at how to build a simple and approachable technology for wrapping up SOAP messages. What we looked at and what we liked was the model of ASP.NET and we wanted to take some of that simplicity and apply it to that model. In many ways, processing a Web services request is no different than processing a Web page request. I think that is evidenced by the programming model - the model is the same.

.NETDJ: How did you organize the teams at Microsoft that first implemented the Web services functionality in .NET?
RH: We had PMs that worked in different areas - I was responsible for some of the Server areas that worked on the Web services stuff. There was another team led by Keith Ballinger that was responsible for the client side of Web services.

.NETDJ: I imagine the size and resources that those teams got was far greater than the resources that open source projects like Mono are using now to copy the technology!
RH: I haven't actually looked at Mono, but I have talked to a lot of Mono developers. I think it's great because our whole value statement about .NET is definitely validated by Mono. I think the challenge that those groups have, though, is the ability to keep up with our development cycles. I think it is going to be very challenging for them to keep feature parity with our commercial ASP.NET implementation.

.NETDJ: Definitely. As far as I am aware, for example, I don't believe they have any significant mobile functionality yet. You recently absorbed the Mobile Internet Toolkit technology into the base ASP.NET 2.0 functionality, didn't you?
RH: The Mobile Internet Toolkit team had developed a technology for building mobile applications using ASP.NET. We knew that our endgame was to get to the point where server controls are such that you put it on a page and it doesn't matter what kind of device is serving as the client. So, the absorption was pretty natural, really.

We actually merged a couple of years ago. We took them, their coders, their managers, etc. and began learning what they had done and took that technology and infused it into all of our controls. We did a lot of work to change our own infrastructure to add adapters and make it easier for folks that came from that team to plug in and to add functionality for ASP.NET pages.

When we do a merger, the team that we merge with doesn't just continue working on what they were doing before. Some of the mobility people got responsibility in other areas and vice versa. We also had some of our people start working on mobile stuff. When you look at it, team mergers are really about knowledge acquisition.

The goal is for everyone to think about ASP.NET as a mobile-friendly platform.

.NETDJ: But overall, developer productivity is the true killer feature of ASP.NET 2.0, no?
RH: It is definitely one of the biggest advantages - it allows developers to write an application faster and do their job better than people on other platforms. It's very easy for developers to sit down and do projects and rapidly build applications. We often hear stories of companies that switch to ASP.NET from other technologies, rewrite their stuff in a few months, and wind up needing fewer machines.

"Simple and easy to use" is what we say instead of just "productive." This means that our technology should be fun, in addition to allowing a lot of work to get done. I think this was driven primarily by Scott Guthrie. There was always a drive from the engineering side as to "what is the best way to solve the problem." But, most of the time, the best way is very modular and componentized and not necessarily easy to use.

One of the things that really impressed me about Scott was his focus on ease of use, not just on having the most abstract and modularized platform. I think that when you start approaching things from that point of view, you start addressing developers who don't care about the option to go in and extend stuff and modularize it. You don't have to be a rocket scientist as long as you understand how the system works.

.NETDJ: Were there pain points with ASP.NET administration that required "rocket scientist" skills, which is why you have revisited this topic in ASP.NET 2.0?
RH: There weren't really any pain points - we're just trying to go after a new audience. When we sat down, it wasn't necessarily to make the people who were having problems happy. It was to go after the administrators and managers of the servers - the people who are responsible for making sure that the servers running ASP.NET are up.

Our approach and philosophy has been that we want the administrators and managers to be just as thrilled to run our technology as the developers are to write applications on it. We tried to step back and say, "Let's go talk to administrators and managers and find out what expectations they have about the experience of running ASP.NET applications." It meant writing all sorts of stuff about instrumentation, events, administration interfaces, etc. Our goal is for administration people to be thrilled to hear that they have more ASP.NET stuff coming.

.NETDJ: So, what exactly have you done to improve the administration side of ASP.NET?
RH: I think there are two really big things. The first would be the amount of instrumentation - being able to monitor so many aspects above and beyond everything today - all kinds of WMI functionality.

The other big item is what we've done for configuration. Right now, ASP.NET has an XML-based configuration system which developers get, but from an admin point of view, they don't necessarily want to open up raw XML for editing. They want to write scripts and batch jobs without having to hand-code those files. In response, we've built out a full suite of configuration classes so that someone running an ASP.NET 2.0 application can run a script and have an application completely provisioned. This includes IIS 6, which has all kinds of rich scripting capabilities.

The thing to remember is that we are now the Web Platform and Tools team. Most people think of us as the ASP.NET team, but we actually cover all of IIS, ASP.NET, and the VS.NET Web Developer Experience. With the ability of having all of this under one umbrella, we can take advantage of better integration with IIS and VS.NET. That was a little more difficult before.

.NETDJ: How have you decreased the level of difficulty in role management and membership functionality development in ASP.NET 2.0?
RH: One of our biggest goals for this release is that we wanted to make things simpler and more approachable. When you look at any kind of significant Web application, it's going to require membership and role management features. We discovered this for ourselves as we started building ASP Starter Kits for our customers. We were writing role management code over and over ourselves.

We had also talked about doing role management and membership functionality as a part of version 1.0 of ASP.NET, but we just ran out of runway.

.NETDJ: A very specific question from a .NET Developer's Journal reader now: how are Master Pages better than User Controls?
RH: What's nice about Master Pages is that, rather than the way User Controls work - where you'd have to put them into every single page - Master Pages work in the exact opposite way. If you wanted to add something for multiple pages, you could do it just in the Master Page. It allows you to have one file you can touch and have it impact your entire site. If you want to read more about Master Pages, check out my latest book, ASP.NET v. 2.0 - The Beta Version.

.NETDJ: I've seen all of these ASP.NET 2.0 features in previous Tech Preview builds of Visual Studio 2005 - why should I install the Public Beta?
RH: Stability! It is definitely a more stable build and stuff will work that didn't work before. There were features in Visual Studio 2005 that were missing from the PDC build. We tried to tell everyone that it wasn't done. When you look at the newer builds, you're going to see a very high-quality product with all the features that were in 2003, plus a whole lot more!

.NETDJ: Do you think ASP.NET adoption has happened ahead of, behind, or right with the general .NET adoption curve?
RH: I think - I don't know the latest statistics - the information I've gotten in the past is that ASP.NET is one of the great "pull forwards" for .NET adoption. When we talk to companies about it, ASP.NET is often one of the first words out of their mouths. As more and more people look at ASP.NET, they are going to see that it is of increasing value to them. If you look at what we are doing with Web services, I just think it is a great way to open up people's eyes.

.NETDJ: So, with all of this excitement about using ASP.NET 2.0 to really drive .NET adoption - why are you leaving Microsoft now?
RH: Five years ago, when I started at Microsoft, I came in with the vision that I wanted to start a consulting operation. I knew I'd have to network and establish a reputation and I wanted to learn from one of the best software companies in the world. I wanted to learn why Microsoft has been so successful. Now, five years are up. That's really the honest answer. I've really enjoyed my time at Microsoft and leaving right now is just incredibly difficult because I love the technology and the work that I do. But, I would rather go out when things are going well. I'm very goal-driven and my goal has always been to start this company.

.NETDJ: Is the improved economy a factor in your decision?
RH: The economy is one factor. The fact that we are coming up on the 2.0 release of ASP.NET is another factor. The typical adoption trend is that Microsoft's 2.0 releases are when companies begin serious adoption. This represents an enormous opportunity for me to get out there as one of the technology's key developers and start my own company.

.NETDJ: How will your new company distinguish itself in the market?
RH: I think the unique value we bring to the market is our familiarity with the technology. We have a unique perspective on Microsoft and its technologies that other companies can't bring. Just from an organizational standpoint, Microsoft is like 5,000 small businesses under one umbrella. It is difficult for potential Microsoft partners to know who to deal with on what issues. The relationship we have with Microsoft is pivotal.

.NETDJ: So, what actual service will you use your relationship with Microsoft to provide better than your competitors?
RH: Our focus will be on application development - working as a virtualized team with companies that need expertise in .NET. Hopefully, we're going to grow our business by going after a largely untapped market, which is rich client development. As we move closer and closer to technologies like Longhorn, this will become even more essential for organizations.

.NETDJ: Whoa - it's kind of surprising to hear a major ASP.NET player talking about driving Windows Forms technology? Were you rivals with those folks at Microsoft?!
RH: Actually, there's a lot of camaraderie and joking between the two teams. The Windows Forms team used to call us the "browser-based technologies" team.

I'm of the philosophy that the most beneficial thing for a developer to do is to take advantage of whatever kind of processing functionality they have on the client. They can do things on the client that they just can't do on the server. As .NET evolves over time - as it becomes more distributed - there are a lot of things that you can do with Windows Forms that you may not be able to do with Web Forms.

I believe ASP.NET is a door opener for Windows Forms, though. Companies can deploy ASP.NET applications and be exposed to .NET technologies. The way I see it, there are a lot more Windows Forms applications today because people have been getting exposure to ASP.NET.

.NETDJ: Agreed. So, in closing, where do you see ASP.NET 10 years from now?
RH: I see ASP.NET as being the primary platform inside Microsoft for any kind of Internet-based appliance that needs to be written. It is, in fact, very modular and - as you watch the Internet and the growth of this wide-area network over the next several year - ASP.NET will be the infrastructure that can support the next generation of Web services and whatever their role will be. Also, when you start looking at networked appliances, those appliances will need some kind of processing stack and ASP.NET will be able to do that.