Much of the future success of Adobe hinges on the work done by its Platform Business Unit, which is headed by Kevin Lynch, the company's chief software architect.

But, if the pressure's getting to him, it's not showing. Plus, he believes, Adobe's got the Web at large working for him.

Adobe makes the bulk of its money from packaged software in its Creative Suite, which includes Photoshop, Illustrator and other creative design tools.

As it looks ahead, Adobe is trying to diversify into online services for consumers and businesses, and it would like to keep its audience of Web developers and designers loyal and not lose them to Microsoft, which is increasingly competing with Adobe.

That's where Adobe's platform group comes in. It designs the plumbing that will allow Adobe product groups to offer online services and other companies to write cutting-edge applications.

For Web developers, it has made more sophisticated tooling with Flex. More significant is the Adobe Integrated Runtime (AIR), cross-platform software that enables Web applications to run on a desktop.

During the company's MAX 2007 conference, Lynch, who came to Adobe through its acquisition of Macromedia in 2005, spoke about Adobe's strategy and its big bet on the Web.

Q: The big announcement from the first day was that you bought the company that made Buzzword, a Web word processor. Why are you getting into that business that Microsoft Office is in?
A: We thought that [Buzzword] was just a great example of potential of the runtimes that we are working on and also a great application in its own right. So it's not so much getting into the Office market space as it's just seeing a great Web application that has a lot of potential and shows what the technology can do. And I think it's a good example of the direction that we think that application development is really having now.

But Buzzword is also part of your whole services push. Give us an idea of where you're going with services and how you intend to make money on them.
Well, services is a new area for Adobe. We've been doing some services for a while, like Connect, which enables you to collaborate online. What we're working on now, however, is a number of new services for designers and developers. So the Share beta is a way to store documents that you are working on with other people. And that's a gigabyte of space â€" it's a free service. We're also providing APIs [application programming interfaces] so you can build your own rich internet applications around that service. We think that's going to be kind of a foundation service for us and a lot of other services and tools that work with that. Buzzword of course will be one of the first applications that will be hooking up to Share so you can work on documents on Share or you can work on them locally.

And the second one is one to enable rich collaborations. So screen sharing, whiteboarding, seeing someone else's video inside your application... The last one is in many ways the first one, which is Scene7, the dynamic imaging service.

So these are all base-level services â€" imaging, voice, collaboration, documents â€" and that's because we are just getting started in this space. But it has a lot of potential for other services as well.

The other piece of news is the number of applications that are coming out on AIR, the Adobe Integrated Runtime. These are Web applications that run on the desktop. Does AIR make the operating system less important?
I think what we are seeing is a big shift to the Web for application development and AIR is basically enabling a big trend for the Web to come full circle back to the desktop again. We're basically betting on this move to the Web. We're seeing all kinds of application development go that way already.

What we're now seeing is the desire for those applications to become more sophisticated, richer applications that do take advantage of local processing power, make use of local resources like your files, being able to notify you. These are the things we used to be able to do with desktop applications that you can't do right now with most applications on the Web. But we think that you should be able to make applications that have those capabilities and build them with Web technologies. There are hundreds of applications running on AIR [which is still in beta].

Will you use AIR in all your product lines?
Just like the industry is moving to Web software, we're certainly looking at doing that too. We have some of the most sophisticated pieces of software around â€" things like Photoshop a it's not like there's going to be an overnight shift from the way we currently build software. But you'll start seeing new products built for the Web that draw from that heritage.

So for example with Premiere â€" it's a great video-editing tool built on the classic way of a software package that you get in a box. We are now hosting Premiere on the Web so you can go to YouTube and Photobucket... That's branded Premiere Express. The Premiere team is building it, they're using web technologies [like scripting languages] to build it.

Same with Photoshop. Our chief executive Bruce Chizen mentioned that we're working on Photoshop Express, [which is] very similar to Premiere. So you'll see a lot of that happening from us.

You'll see AIR, for example, start to be used in some of our software in different aspects. Maybe you like using Premiere Express â€" you don't need all of Premiere, yet you want to put it on your desktop. That could be a bridge... That's not something we're currently doing, but you can imagine us doing it. There will be a spectrum.

What's the business model behind these hosted applications?
Well, with Premiere Express, it's an ad-share model. The site that it's hosted on, it's advertising-supported and we share in that revenue for people who are spending time using the Express editor. So ad-supported, subscription-based â€" all in new ways that people are monetising software.

How do developers make use of these web services?
We're going to make a bunch of APIs to these services available to integrate things like voice into your application. With Share, for example, there is a document repository, file repository that we're hosting. Everybody's got a free gigabyte of space, and there are some web APIs that anyone can use to access and make use of a data store. So you can build a file-browsing user interface or you can build your own image editor and use Share to hold the documents behind it. Developers can do whatever they like to make use of that â€" there's just APIs available.

On the business side of that, there is an amount of free storage and, beyond that, we will provide premium services that could have revenue associated with them either through ads or subscriptions and models like that.

You are doing a lot more in tools and, of course, Microsoft is very strong in tools and they're doing more with web development as well. How do you compete with that big machine over at Redmond?
We've actually had a long relationship with Microsoft. So there are areas where we co-operate in their areas; where we compete we are one of the biggest software vendors on Windows in the world.

The areas that we do compete, in web tooling for example, we've been competing for a long time. In the early days of web authoring, I worked on the Dreamweaver project at the time and we were basically last to market. FrontPage was already out; there were already a dozen tools already out there. None of them had really had gotten predominance or really popular with pro web developers and there was an opening for us to do that. We got in and designed a tool that really resonated with a community. We listened well and fed that feedback in. That managed to get an incredible level of popularity really fast. What we saw there was great success with Dreamweaver with pro web developers that beat FrontPage and other tools.

So even though Microsoft works on software, they don't always win in these markets. From what I've seen in these situations, you just have to be close to the customer, not be distracted by what the competition is doing in terms of trying to catch up with you or whatever, and stay on the leading edge and think about the next thing that is happening. And let the competition chase where you've been.

For us in the rich-internet-application space, that's brought us more into developer tools. Things like Flex Builder â€" that's a new tool for us and it's doing very well. We've made the Flex Framework open-source and free. A lot of the frameworks for the web now, of course, are free and open-source. So to play, that's kind of the table stakes. So we've anted up.

Thermo [a planned tool to let designers write their own web applications] got a lot "oohs" and "aahs" from attendees. But people have been trying to make development tools for people who aren't programmers for a long time, probably as long as code has been around. What are the limitations there?
I know! But we're not doing a general solution for anyone to build an application. That's one of the key things â€" we had to be specific about who the audience is. It will remain a holy grail for anyone in the world to decide" I'm going to write a rich internet application today" and to do that.

So it's just for designers?
Exactly. It's for people who are using tools like Illustrator or Photoshop and have a background in interface design and want to create a great experience for someone. But they are primarily a designer. Right now they work in partnership with developers and the way workflow works today, you create a comp â€" a composition â€" in Photoshop, a picture of what you'd like, and you hand it over to a developer to try to make it look like that. We've made some of the workflow work more seamlessly already.

But what Thermo does is, it makes it so the designer can not only draw what the application looks like, but they can also add the interactivity for how it works.

The magic of what we're showing with Thermo right now is that you can select elements that are just pictures on the drawing and you can say this actually represents a list box, or this represents a text edit field and we put the logic to convert the picture into a work component.

To fully complete an application you need to connect data â€" connecting back into web services, loading XML. Thermo doesn't go that far.

When you talk to people at Google, which is very influential in web development, they say they avoid Flash whenever they can and prefer using lowest common denominator technologies â€" JavaScript, Ajax. Is that a concern that developers are pushing the envelope in Ajax and staying away from Flash and maybe AIR?
No, it's not a concern. First of all, AIR supports Ajax and Flash and Flex and PDF. So if you're using any of those technologies, AIR is designed to run that code. If you're purely an Ajax developer, AIR is fully supportive of that and you can create first-class applications with that. I talk about betting on the web and that means both HTML and Ajax and Flash and PDF â€" all the stuff that people are really using on the web today. To the extent that people are building applications [for the web], I think... it's the right direction and we're going to work to enable that.

From the development point of view, using JavaScript in the browser is not that simple. You really need to be a sophisticated developer to do that.

Do you see more potential in the enterprise for these kind of applications?
AIR's value proposition is very similar to enterprises as it is to end users. You can use those same technologies but deploy them as desktop applications, and one of the things that's important in enterprises is that you can update those applications from your server where it came from. Just like the web, you can get that new application with auto update, so it solves the deployment problem. But you also get to have desktop integration, which enterprises want to have. We kind of lost that [with the web].

Flash is the dominant video format on the web. Now Microsoft has Silverlight and they are signing on partners. How do you want to stay ahead?
Right now, Flash is far in the lead so it will be difficult for any technology to get the kind of adoption that Flash has right now on the web. We were able to start in the early days of the web and there's a lot of use of it that has caused its adoption.

Now we're in a situation where we can add more functionality into Flash and, in a year, we get 90 percent of the world to update. No other client technology is in that situation right now, not Windows or Internet Explorer or any other technology. So it's going to be tough.

But we're not resting on our laurels at all. The Flash team is moving very fast.

Where do you expect AIR to go? In a couple of years, will most applications be written with AIR?
I hope so! Potentially. Because most applications are being written for the web and we're bringing those applications to the desktop with AIR, there's nothing really else out there right now [for that]. I think that AIR is positioned early to really be a leader on that, kind of like Flash was positioned early as the interactive multimedia leader.