X

Microsoft sees future beyond the PC

Bill Veghte, an 11-year Microsoft veteran who now runs the embedded/appliance platform group, says the company is making huge investments in devices other than the traditional beige box.

7 min read
For Bill Veghte, Microsoft's future lies in devices that skirt the traditional definition of a "beige box."

Veghte, an 11-year Microsoft veteran, has for the past year worked as vice president of the embedded/appliance platform group, the unit responsible for the Windows CE and Windows Embedded operating systems.

In a recent interview, Veghte outlined an alliance that will give chipmakers access to the next version of Windows CE, so they can design chips to power a range of embedded devices from industrial automation equipment to cell phones to handheld computers.

He also discussed the company's plans for this week's Embedded Developers Conference in Las Vegas and Microsoft's competitive position against OS maker Wind River Systems and the Linux operating system.

News.com: Given that Microsoft is a very large company with a lot of different things going on, where does the embedded group fit in?
Veghte: If you think about our strategy and our investments, Microsoft is making huge investments beyond the PC?And those investments are characterized by end-market devices such as the Pocket PC or the Xbox or Stinger reference design, but also in investments at the platform level with Windows CE and Windows Embedded.

One of the things that I think is interesting is many of our competitors ignore or belittle the PC. But we think the PC has a critical and important role simply because so much data is created there and so much data resides there for the user?Now the caveat to this--and (CEO) Steve (Ballmer) will be talking about this Tuesday--is that?we think it's critical to have smart devices.

Historically, industry pundits have debated the merits of thin vs. thick, PCs vs. non-PC devices. And the .Net and Microsoft bet is on smart clients prevailing over dumb clients?Smart means software intelligence at the edge of the network, as opposed to a centralized server model, which assumes dumb terminals on the edge of the network.

Where are some of the areas where Microsoft has seen the biggest support so far in terms of industries or types of devices?
I actually think one of the fun parts about this is that as soon as you start focusing on the rich applications and services, you see all sorts of interesting devices coming out. There's an Experience Music Project device, in which (the Seattle music museum) is using CE as a platform for their carry-around device as museum-goers go from exhibit to exhibit and interact with the exhibits via wireless connection.

Or Vail Ski Resorts, which is using a Symbol device to scan lift tickets and be able to provide a financial transaction on the spot if a the person doesn't have a valid or up-to-date lift ticket. Or Bally's Gaming, which is an industry-leading gaming and entertainment provider. They're using Windows NT for embedded to run a wide variety of games.

If you think about our strategy and our investments, Microsoft is making huge investments beyond the PC. What exactly is on tap for the conference?
It's something that we are bringing back after one year of hiatus and actually expanding it. Previously, the Embedded System Conference focused mainly on CE developers. This year, we're bringing together people who write applications and build hardware for all types of embedded devices--from Windows CE to Windows Embedded to server appliances.

It's an opportunity for Microsoft to give all our embedded developers a detailed roadmap, not only of the technologies we have and are planning on delivering, but of our vision for smart devices and the opportunity that .Net represents for them.

What other news is going to come out of the conference?
On Tuesday, we will announce the Windows Embedded Strategic Silicon Alliance. That specifically is a variety of silicon partners that are partnering with us in the software platform to optimize Windows CE for their architecture.

So that includes industry like ARM, Alchemy, Cirrus Logic, Hitachi America, Intel, MIPS Technologies, NEC, National Semiconductor, Texas Instruments and Toshiba. Members of this alliance will work in close partnership with us in joint development efforts, which include access to our next generation CE source code to ensure that embedded developers will be able to take advantage of the latest silicon technologies with optimized operating systems code.

Microsoft Windows CE goes into a huge variety of devices from the Stinger smart phone platform and Pocket PC to Internet appliances and industrial automation. What are some of the challenges and why do you feel that one basic OS kernel is right for all these different markets?
It's actually interesting...I don't think that one operating system fits all? We are not in the business of providing a platform for 8- and 16- bit devices. When you take a step back and you look at our strategy, we are definitely taking a two-pronged strategy. We think that Windows CE is great for these mobile or these lightweight connected client devices.

And then you're seeing us start to take NT 4 Embedded and then with the delivery of Whistler Embedded, drive that into these rich, high-end connected client devices as well as into server appliances. So we definitely think?there are multiple platforms necessary to service these smart devices.

Are there devices today that either use an 8- or 16-bit operating system or don't use any real operating system that tomorrow are going to need Windows CE?
I actually think there are a whole host of devices that are using 16-bit processors today?where a user is going to demand a richer experience. And as a result, they're going to move up to 32-bit processors; they're going to move up and demand rich applications and services.

The cell phone has had phenomenal growth both in numbers and in usage...The communication modules that you associate with the cell phone were initially around voice. Now increasingly, they're about voice, pagers or notification. Also, people are starting to look at it in the context of e-mail as well. That would be a great example of a device that demands a broader, richer set of experiences, and we're committed to providing solutions for those smart phones.

If you look at Linux as a platform, obviously Linux is still pretty adolescent in the tool set that they provide. Why is the embedded market financially important to Microsoft? If you look at our platforms business, our different Windows businesses, the business I manage and run is one of the fastest-growing platform businesses Microsoft has. But I think the impact goes beyond the financials because it really is about providing a platform that transcends devices. We think that the PC, in conjunction with the smart devices that I allude to, is an incredible opportunity.

When you look at the competitive landscape, obviously you have different competitors, some of which are charging, some of which are free, some of which are open source, some of which are not. How do you assess that competitive landscape?
Well, if you look at the traditional embedded marketplace, you would talk about Wind River. And the interesting thing about Wind River is that in the old embedded marketplace, it was fundamentally about (making a device work). It was not about applications... I think Wind River has done a solid job. But they have an incredible challenge because they haven't been able to build a platform for applications and services.

If you look at Linux as a platform, obviously Linux is still pretty adolescent in the tool set that they provide. If you look at the size of something like the Linux browser, it's very big and very bulky.

There's a tremendous amount of tinkering going on right now as people say, "Hey, I've got a Linux kernel booting on a device." But that's very different from creating a rich set of applications and services and experiences on those devices. So for Linux, it's about: Can the tool set mature? Can business models be strong. Can there be a strong business model behind it? And can that deliver a great applications and services platform?

The other question that people often (mention) is: "Well, Linux is free." And of course the run- time license is one portion of the overall cost of bringing a device to market. For example, there are Linux companies out there that say the kernel and the memory manager for free, but (add): "We're going to charge you for the device management module. We're going to charge you for the browser module." And by the time you add it all up, it's actually more expensive (than Microsoft operating systems).

So we think that there is no such thing as a free lunch. When you evaluate the total cost of the platform in delivering rich applications and experiences, we're pretty bullish about the opportunities that we provide developers and (manufacturers).

The group that you head up is a year old. Where would you say you are today compared with where you were a year ago?
I feel we've made a lot of progress. I measure it on the basis of people that want to participate in our partner programs. And if you look at (that), we were up over 125 percent since September, which is absolutely stunning to me?My two core products--Windows CE and Windows NT embedded--(are) showing revenue growth of 300 percent year over year.

You talked about some pretty impressive revenue growth in the past year. Is that something you think can continue at the same rate?
I think we will continue to have strong revenue growth year over year in both these products next year, next fiscal year, particularly around Windows Embedded because next year we'll deliver Whistler Embedded, so that product line will be refreshed.

But is tripling revenue something that can realistically be expected to continue?
I think if you look at our revenue growth vs. Wind River, for example--do I think we can continue to beat the market or meet them? Absolutely. I'm not in the business of forecasting our financials, but I do believe we can continue to grow much faster than the industry.

Are your units--Windows Embedded and Windows CE--are all of those profitable for Microsoft?
We are investing very heavily in this space. But?people have been?using Windows 2000 in an embedded context as well?And so if you look at my overall (balance sheet) that factors all those, yeah we're a profitable business for Microsoft.