 Video streamVideo streamVideo stream |
 |
 |
|
|
|
| Question indexQuestion indexQuestion index |
|
|
|
|
|
| Interview transcriptInterview transcriptInterview transcript |  Discuss this interviewDiscuss this interviewDiscuss this interview |
|---|
|
The last couple of years have been an interesting couple of years. As you are probably aware I was in Microsoft for quite sometime, just left this past June and at Microsoft I actually was in a variety of roles. I started off as a software developer of all things, quickly learned that software development was not exactly what I liked to do on a daily basis.
TSS: Not fond of slinging code everyday?
I just like meeting people I love talking to people and talking of technologies, isn't much like writing the code. TSS: Oh come on admit it you are just like us, you just like to talk.
You know I found my niche for a while and that was, for about 2-1/2 years at Microsoft I actually did technical evangelism. So during the dot com days going out to companies like EBay and MarketWatch and some of these big brand name companies and showing them ASP and COM and trying to get them to move to Microsoft's platform that was a lot of fun. That was a lot of fun because it was a really competitive time for Microsoft. That was when J2EE was starting to come out and that was a lot of fun and then of course most recently before I left working on ASP.NET, that was just so much fun. TSS: Doing a lot of presentations with Scott Guthrie
Yeah, doing the presentations. When I joined the Internet team there were I think about fifteen of us and it was just a small team and having an opportunity to design some of the 1.0 features and then also start building the community around that. I love building communities where people come together that share information and then being able to kind of have an early hand in that with regard to ASP.NET. You look at the ASP.NET community now it is just gigantic. Then being able to work on ASP.NET 1.1 ASP.NET 2.0. For me it was a dream come true. I started off doing development writing gosh on UNIX machines writing CGI PERL scripts, it was kind of my first entrance into the webs zone. TSS: And you can finally come clean about that, now that you no longer work for Microsoft, right?
That's right. I was first introduced to ASP when I saw the three-letter extension on the Microsoft web site in the earlier days and now I have been part of that technology and so it is so exciting, lot of fun. Now I have left Microsoft. I started my own company. Before I came to Microsoft I had my own consulting company and I left, I had a kind of a five-year goal where I wanted to go somewhere, learn the industry, learn the technology, meet some people and then leave and start my own company again so that is what I did. I actually gave notice to Microsoft back in March 2003.
TSS: Oh wow!
I gave them about 15 months' notice that I was planning on leaving.
TSS: That's generous of you.
Yeah, I actually wanted to leave with a really good relationship. So, I went off and started a company and the company is called Telligent Systems based out of Dallas, Texas although I'll throw in some buzz words here, we actually have a work web. We have people in Italy, in New Zealand, in California, in New Jersey it's all over the map. Our main office is in Dallas. We are actually opening up our office this summer though in Dallas. We are actually opening an official office.
We do a variety of things right now, we do anything that makes money that allows us to keep playing with technology is what we do right now. Our initial business was to start off kind of doing training. Lets do some training to kind of get our name out there and start working for companies and we did one training event, and the next phase of our business which is services, which is what we are doing a lot of now. Primarily helping a lot of well-known companies manage applications and develop applications. For example the ASP.NET web site is a site that our company now manages and runs for Microsoft as is Blogs.MSDN.com. Our long-term goal is obviously to become a software vendor, some of the people we have hired would be Scott Watermasysk who is the creator of .Text and Jason Alexander who is the creator of nGallery and one of the things that we are trying to do long term is something that we have already put out in the market that people are adopting is this platform we are calling Community Server and the idea behind Community Server is that they are open source platforms which have all the source codes to it where people can push code back into it, a forum solution, a blogging solution and a photo gallery solution all integrated together. TSS: Sort of in competitions to like DotNetNuke. No it is more complimentary to DotNetNuke. I look at DotNetNuke can share point portals as containers for the type of functionality that we build. So, DotNetNuke is a great container that allows you to easily plug in components. What our software does is we can run outside of DotNetNuke or SharePoint, so we also can plug in. So, that means if you want to do blogging by yourself you can do that, if you want to do blogging inside of DotNetNuke you could also do that. TSS: Okay interesting, interesting. Well good luck with that, we will get that out of the way. No thank you. TSS: Obviously you are best known to the .Net community as one of the two voices for ASP.Net the other one really being Scott Godfrey. Oh yeah. I think we did get there. I think the initial goals were just building a development platform that excited web developers and made it easier, made it better. Looking back now in a consulting role its fun, when I go and I see companies and I see how they write software and going and looking at ASP code again and looking at just what that looked like, not having really had to have met with it for several years, it just amazes me. It amazes that how successful people were with it. It was obviously very powerful, but you look at it now and you compare it with ASP.NET and it is just a night and day difference. When I look at ASP applications, look at ASP.NET and just look at making small changes or just structural changes to the way applications behave it is just a night and day difference and when you can look at that and make those kind of easy determinations you know you have hit a home run. It was not so much what was wrong with ASP, it was more just changing some of the characteristics of what ASP delivered to the end user. ASP was more of a scripting technology. One of the big areas was performance, it was how do we step back and make the application perform better, how can you do smarter things with regard to performance. One of the features that I was really passionate about and worked on a lot was caching. A lot of developers in ASP for example, the way they would do caching would be to use applications state. They would stick data in there and they'd have some kind of mechanisms that pull back out and validate it. Now you have what the caching API for.NET and all those problems just go away. You look at it today and you're like why didn't someone do that for ASP, some simple things like that just make more sense. The complied nature of the code, the structured nature of the code oh such a huge difference, the eventing model. The way I like to explain it to people even if you never did a VB6 form development. It is like VB form development for the web, of course you use other languages, I can officially say my preference is C#. TSS: Does it feel good to come out... I feel good to be able to finally come out of the closet and make that announcement. I love VB but C# is what I am really passionate about. The great thing about ASP.NET was when you think about how you write software, how you really write good quality software and how you want the componentize it up and make it into discrete units of functionality those were the things you just couldn't do with ASP. ASP was just this amalgamation of just script and HTML and COM and looking at ASP code today any developer would probably agree it is just hard to wrap you head around it. You open an ASP.NET application site, here is the load event, this is where things are starting, here is an event that fires when you click on a button and just the nature of being able to program to server controls, I'm big on server controls. In fact with our Community Server product, it is like 110 server controls. All server controls, no page code whatsoever and I think that is the way of the future even if you look at ASP.NET 2.0 I think that is dramatized even more, because you look at ASP NET 2.0 and one of the big points Microsoft makes, is always making about is the savings you get in productivity. Why that is because of the declarative programming model. You don't have to write code for lot of these simple tasks. You can do things declaratively through the structured and server control model, it just makes things easier. When I joined the team a lot of the base infrastructure, some of the base kind of ideas were already in place. Obviously, written by Scott Guthrie and it is funny you look at the way technologies have evolved. When you look at ASP and one of the things I have always stated and believed in is really if you look at JSP and compare that to ASP you will notice a lot of similarities, lot of classing is the same, a lot of behavior is the same. There are some different functionality and characteristics with the way the application performs. When I was doing a lot of the design work for ASP.NET, I definitely read books on competitive technologies I read books about Cold Fusion, I had used Cold Fusion previously, I had used a lot of the PERL libraries in the previous lifetime. You always draw upon past experiences, especially when you are building technology. You draw upon what you have used in the past to say this was what I used before. This is where it was good. This is where it was deficient and taking from those experiences and building on top of it. The entire team like Scott Guthrie's background was from IIS and ASP so a lot of our experience was just looking at ASP and saying hey these are deficiencies, we know these are problems. There may be some ideas that came in from other angle, so we very clearly stated within Microsoft we could not evaluate the technology. We could not download J2EE and install it on our machine and go look at it and deconstruct that. That was not really done. I mean I pickup up books on different J2EE technologies and wrote competitive analysis papers that were sometimes looked at, sometimes were not looked at, but a lot of it was pretty just entrepreneurial invention, that's what we did. I say it's just a matter of opinion. There is no one right solution for any problem. There are always going to be a multitude of solutions for any given problem in most cases especially when you are dealing with technology. I work with customers a lot now and it is interesting to see how the perspectives and the driving forces that change the way software is developed based on the requirements of the customer. Sometimes they want a highly architected grandiose system that's laid out perfectly with all the best practices followed with all of the coding patterns followed and then you have another guy who says gosh! I need to get this problem done. Here is the problem I don't want to spend a million dollars solving the problem, I want to solve it one way, the right way, the quickest way, get it done. I think when you look at ASP.NET, I think what ASP.NET gives you is a spectrum whereas most Java developers will come to you and say there is one way you do things, which is this MVC pattern that you have to use. I think what an ASP.NET developer can say is I have a variety of options. I can give you a very simple application I can whip together in 10 minutes that allows you to do data entry, you hit a submit button, there is page that handles an event, push it into a database, done. Do you need model view controller to do all that? You could use it, might be overkill in that case. Now I would also come back and say there are many ways in which I think the ASP.NET pattern of a page architecture with eventing and server controls can be of superior pattern because its simpler. Its cleaner, its easier to understand. When you look at the page metaphor of ASP.NET. I am a big fan of this and so we will buy... TSS: We will take your biases in stride. The concept of having a page and being able to have just the physical and contextual reference point to start from when you are handing off an application to someone, makes things a lot easier to understand versus here is a set of class libraries. There is an entry point based on this URL here is the code go find it. So it's really just a matter of perspective about how applications can be constructed, is MVC the right way, I think it's one way, I think the ASP.NET page control architecture definitely a peer or a parallel option as well. Yep. Did you find a lot of the arguments to be religious in nature or were people fairly open to the idea of being well maybe you're right, how much did you see out there? There's a good mix. There is definitely a lot of religious fanaticism that says this is the way and really I think what it boils down to is not so much religious fanaticism, it is a comfort zone. Developers tend to get into a comfort zone of what they know and when you challenge that there can tend to be some reactions like very defensively this is what I know, this is what I believe in and I am going to defend this until I die because this is the principle that I was shown for how to build a software. For example, n-tier Architecture, everyone always talked about n-tier Architecture, 20 years ago it was DNA Architecture whatever you may want to call it and there is a whole line of thinking that everything had to be physically architected separate from one another, so everything was just modular, you can swap things in and out, and that is what's great on paper. I come up to the board and I drive you a firewall and I will drive you a couple of web servers, I will tell you, you've got to scale those problems, you just add this web server things in here and oh you got some scalability in the middle tier, we just add some more servers in here and database, you know what that's not as easy as it sounds. There are all kinds of challenges with managing hardware, there are all kinds of performance issues and security issues. As soon as you are talking of process boundaries and hopping across processes, all kinds of new issues pop up and in fact one of things that I like to advocate a lot of developers especially when they are building high performance web sites where security is a big issue, maybe it is just more of an advertising site not like a e-commerce site, really advocating, Hey follow this architecture pattern and this is another religious issue. Follow the same architecture pattern, they don't always constrain yourself to physical limitations, say that a business logic component can exist and live within the same process as ASP.NET. You can go home and not feel like you have to take a shower to wash the dirt off yourself. It is okay, it really is. Again at the end of the day its not so much, going out talking to developers it is not so much a matter of there is always a right way, I think it matters saying there are always going to be new and different ways. That's what is fun about technology. You can always explore these different options. Well, I think there is a continuum, I think one thing is that Microsoft has done a phenomenally good job at is commoditizing development of software. A lot of developers especially when we look back to the ASP days they didn't initially have a traditional computer science background. They weren't trained in C++, they didn't learn C++, so they didn't learn PASCAL, they had no idea about object-oriented programming. Do you know honestly the way they started, they heard about the thing called the Internet, they learned that you could write some text in a text pad editor, you could save it and it would show up on a web page and that was cool, and that was motivation enough for these people to learn that hey I can write HTML, oh gosh now I can add some of the script stuff in there, now I can do stuff like talk to database. That is how lot of ASP development... TSS: Lot of art majors went into graphic design and so forth?
A lot of people are a bit surprised I come clean and tell people, I don't have a CS degree, I'm completely self-trained. I started programming on the Commodore 64 way back then getting the magazines and typing in the program listings from the back of that. You remember doing that, you spent all day like typing these lines in the code.
That's when $20 for the code disc just seem really worth it.
But I guess my point is that with Microsoft technology is that they don't try to raise a false bar, they don't say, that you have to be this type of person to play in this market and I think that's what nice about Microsoft technology. They say its open, you can start here and you can be at the beginning and you can be a high-end developer, you can be the Don Boxes of the world and still play in that area.
It is not necessarily a good thing. You don't want bad code, writing bad software is a poor reflection often times on the technology and not the developer. You asked a minute ago about the patterns and practices and I think that is one thing Microsoft production is doing a phenomenal job at now, it is starting to really push and invest a lot of time and energy in the patterns and practices, groups and they have been publishing some patterns for .NET some of which I think are phenomenal some of which are so-so, but I think at the end of the day were making the statement that they are putting effort and they see value there and they are putting time and energy there to educate the developer. You can still have people who don't start off as developers, write code, solve problems that down the road cause other problems because in the way they built their systems, but again I think as long as developers always, if you step back from your role, someone who understands technology and you can make technology work together and look at the problems you are trying to solve for people, its often times more important to solve their problems than initially do things 100% the right way. So for example, if I am being employed by a company to write software for them and they say they need x functionality, well I could go shoot the moon, make it localized, make it completely modular, make it super scalable, all they wanted was a way for their customers to submit a request for more information. So part of it is also really starting to, you don't have to be a classic developer like a CS trained developer to be able to go and do that kind of stuff.
It is a challenge, I'll be completely honest with you. I had a customer e-mail me today and say lets take the scope of what you define and scale it back because we don't care about localization, we don't care about extension, we are never going to replace SQL Server, so we don't need to abstract the data layers much. So you look at these kinds of problems and me as a technologist gosh I want to get in there and I do want to make it architected correctly and I do want to make it so that in three years if I need to service it that, I can do so in a meaningful way, but it is a trade off, I think when we start looking at consuming time and money, it takes a lot more time to really develop things the way they properly should be done and sometimes there are areas we have to look at and say, hey Mr. Customer or Ms. Customer, I'm sorry, I think that this is the functionality, I think you might need in three years. Here's the opportunity, its going to take x amount longer, what do you want to do and often times they are going to come back and say lets worry about that problem in three years versus now. I think that would have to be, there are actually two. My first was database cache invalidation. That was something that I came on to the team from day #1, that was the problem they kind of threw to me and said, what do you think about this, how do you solve this and being part of even in 1.0, thinking about how to solve that problem was a tremendous amount of fun and of course we wound up doing pulling the other teams SQL Server, ISA, Internet Security Acceleration Server, ASP.NET, IIS, and collaborating together to build a solution that spans those technology and this has been, I can't wait to see that technology come to market, because I think it will change the way developers write software. It will because when you can pull data out of your SQL Server database and hold on to the memory it will guarantee you that when the data in the database changes that you can flush that memory, that takes the performance characteristics of ASP.NET and just sets them far and above and beyond any other competitive technology can even hold a candle to, cause you can just do some amazing things. The other piece of technology that I was super passionate about that I worked on was a provider model design pattern. The idea behind the provider model design pattern was looking at ASP.NET and looking over the different APIs, it is APIs that developers become familiar with. They grow to know and use, look at the Response class, which has been around since ASP.NET 1.0, the Request class. These classes and these methods in them have behaviors that developers expect to use and expect behaviors behind and looking at a way we could move from that being a fixed set of behaviors where we would work on a known set of data an HTTP request and translating that to move to a somewhat less fixed set of data that we are working on, data in the database whether it be SQL or web service or somewhere else and the important thing about the provider design pattern was coming up with a mechanism that allowed Microsoft to publish APIs, to say here is a membership API, you can use the invalidate a user and create a new account and educate developers on that, write books about it and get a lot of information out there, but at the same time, talk to your architect, you're one of the people that really wishes to say I would have designed this differently, I want to change the way this behavior works and being able to say yeah you can. Here's how you do it, you unplug this functionality, you write your own class, you drop that back in and guess what you use these from your APIs, you can cast out and use your own what if it give you is it allows you to tap the developer market and say hey I need a developer, oh its using my provider unbeknownst to you, but you are not a part of me. That's so leveragable, you can talk to the business people who are making justifications for making software decisions and you give them that level of flexibility that gives them a whole new level of comfort where they just say, look I have options now, I can deploy this, if I don't like it I can unplug it and plug something else in and find a developer that knows how to use it.
You can use them ASP.NET 1.0.
How do you do that, there are a lot of people out there who would love to get hands on that now I have an article coming out on MSDN. Today's what October 27th, it should be coming out in the next couple of weeks, my MSDN column which is about five months overdue right now, Ken Short, he is going to throttle me but I have an article that is coming out that talks about how you can do database cache in validation with ASP.NET 1.1 and 1.0., using one of the same techniques we use in Whidbey and the provider model design pattern is something that really fell out of a lot of work that I was doing in with now the Community Server product the Forums, the work Scott did on .Text, the work that Jason did on nGallery. To be honest with you when I was at Microsoft, the forums that is now with the Community Server, that is what I prototyped membership in, role manager, personalization, provider design pattern, a lot of the functionality that went in the cache and the behaviors there skinning, theming a lot of that stuff. So you can use a lot of functionality today, it's not as clean and as elegant when you use this in ASP.NET 2.0 but you can use it. I seem to be asked that question a lot for some reason.
I think that is a struggle that will continue to exist for some time now, it is the short answer. The long answer is I think that there is a market that tried to start about two or three years ago, but I think it is now it is starting to really take root, which is the ASP market. The application service provider market where I think SalesForce.com is the example that most people are familiar with, the no software is kind of their advertisement. They are completely server driven. They are completely web based. You don't require any software installed, set up to do anything to get that application running. I think that's kind of one extreme. I think that looking at some of the software now that we are trying to get into the ISV market, and looking at some of the software we are building, there are absolutely cases where smart client makes more sense. In fact, one of the things I talked to Scott Guthrie about as I was leaving Microsoft is my belief is I think there is a tremendous amount of opportunity in the smart client market. If you look at what has happened with ASP.NET done, the level of growth and the adoption and you can say that the smart client market, or the smart client adoption in the community its definitely very different. So I think that there is actually a tremendous amount of opportunity, I think if Microsoft matures its technology, things like XAML and some of the new things that are coming out, I think there is going to be more and more opportunity for the smart client developers to do some very powerful things, because there is a lot of flexibility that exists on the client machines aren't being taken advantage of. People are buying these high-end, dual-processor one gig RAM machines and then using a web browses to run their application. When you look at that capability you think, gosh, how can I use that? How can I use that to make my experience for the end user that much better and so you're absolutely going to see more smart client applications. I think we are. I think what is important to kind of frame the discussion around is that the Internet was a really disruptive technology. It changed the way people work with software and delivered software to their customers. Before the Internet what was the traditional model? It was all client server based. There were a lot of VB6 applications, there was PowerBuilder, applications such as that, the Internet was disruptive. It forever changed the way people thought about writing software because of the concept of the web browser. Are we going round and round the wheel. No I don't think we are. I think the Internet was very disruptive, it changed how information can be delivered, because it connected systems better together more easily. I think that what we are seeing happening is a progression of using the Internet, using wide distribution network and the new capabilities that you are seeing now. I plug into the hotel here and get a hundred megabit line. How much longer is it going to take for downloading, running, really high performance smart client applications across that line. To do, to use the Internet, in many of the same ways a browser is done, to do more things with it. Yes and no. I think the way to look at this is what software would your grandmother, mother whomever else typically run on the computer. What are the typical things they would be doing, probably e-mail, probably they want to be able to surf the web, they are probably gonna have something like AOL or MSN or some other kind of service provider that is going to install some additional functionality and services onto the computer. When you start looking at software that is delivered that way, you start looking at Office, you start looking at some of the functionalities coming out there. Having .NET built in is something that is going to start happening when they install new software in the upgraded Office or they are always going to be the type of user that doesn't really use that functionality. They are always going to be limited to e-mail and browsing the web and I don't think browsing the web and Internet is going away anytime soon. When you look at business users, when you look at organizations who are actually using a computer not just for information purposes but are using them more as tool to accomplish tasks, distributing the software then makes total sense because you are looking at people that are probably using a high-end operating system. They probably have software requirements that get pushed down onto their desktops. I think the last number I saw was .NET was installed in like 120 million desktops or just some amazing number and then it becomes a little more real. Then you can start seeing, hey if there is installed base of a 120 million desktops, investing in smart client technology makes more sense. TSS: Last question. Okay.
Any change I could make. Addition, subtraction, modification... Well there was this little feature that a lot of people probably have forgotten about, that we had in ASP.NET 1.0 in some of the beta versions and it was called My Web and what it was is a little feature that allowed you to literally be viewing your ASPX files in File Explorer inside of Windows and double click an ASPX and it would just work and run so basically there is a built-in web server behind the scenes that recognized that extension would load it out would do the proper things. That's the feature that I think a lot of people who saw it missed the most. Talk to Kent Sharkey, he will go on and on all day long about the My Web feature, I forget what we actually wound up calling it, but that is a feature that I would really love to see. Other features I think ....
Well, I think that where it gets a little interesting because imagine that instead of having to run all of my web applications from a centralized server, imagine if I could send you a zip file, you can unpack the zip into a directory and double click on a file and things just worked. That is pretty powerful so it provided a new way to deliver information where I didn't necessarily had to have a central server acting as a web server. TSS: So that was great. Anything else you want to say to the listening audience? Yeah, I think the last thing I would say is just I love to hear from people. So my e-mail address is RHoward@TelligentSystems.com and I just love to hear from people, see what they think what kind of technology they are working with, things they are up to, I just love to hear and learn about how people are using technology especially what they think of ASP.NET. TSS: It is your baby for the rest of your life. I'll always love it... Deep passion for ASP.NET TSS: You never stop being a parent. Right Rob. That's right.
|
