It has been over a year since Intel announced it would invest $US52 million in mobile connectivity vendor Xircom. While the obvious connection between Intel's networking group and Xircom's PC Card business has so far proved to be a successful combination, the two vendors also seem to share a common formula for success - a culture of paranoia that permeates the executive suite. For Dirk Gates, chairman, CEO and president of Xircom, that means worrying about how the growing popularity of personal digital assistants (PDAs) may impact the company. Sumner Lemon sat down with Gates recently to discuss trends in the mobile computing industry and the Xircom connectionLemon: How much attention are you paying to the evolution of new, smaller computing form factors, such as the PDA?
Gates: Four years ago there were those pocket LAN adapters - a completely different form factor - and we've followed the form factor to PC Cards because that is where the market is today.
I think these small devices will increasingly demand communications capability. Windows CE devices can now be PC Cards, but you're also going to see compact Flash form factors - and you're going to see different types of form factors.
I believe that over the next few years, in this Windows CE space, they will eventually consolidate around something that will be the next generation, smaller form factor communications interface.
And they're going to have a different set of requirements than notebooks - even lower power or a higher emphasis on wireless connectivity because you are not going to be standing in one place when you're using these things. The market is somewhat young at this point from a standards point of view and also from an installed base point of view.
Are there any specific products out there that you find particularly exciting?
I have a graveyard of PDAs and these kinds of devices, and for me there is something that I call the "one battery test". I buy one of these machines and if I replace the batteries, then I am on to something.
But, my Newton didn't make it. My Sharp and Casios of the past didn't make it and my first generation Win CE device didn't make it past the first battery. I haven't tried a Win CE 2.0 device but I might try that.
The one thing that I have used - although the batteries haven't run out yet - and find interesting and compelling is the Rolodex Electronics Rex. For me, this is interesting because it finally becomes an unconscious carry. I carry my business cards in it and I carry it wherever I go. So far, it's interesting.
But somewhere in here, between a Win CE device and this Rex device, I believe there will be a generation of devices that become an unconscious carry.
What the Rex is lacking is a communications capability to do simple messaging. The PalmPilot and Windows CE devices are still a little too big for me.
With Intel's investment in Xircom last year, how close are you to their research and development for mobile computing products?
I think we are reasonably well involved in a couple of areas. One, it's nice to have Intel as an investor because we get to walk the halls of Intel. We're there frequently enough that we understand what is going on. Our primary interface is with the networking group but we also have access to the processor group to understand their road map and where they're going. And that's always nice to know.
At the brainstorming level, we do spend a lot of time with the networking folks - trying to figure out where the next generation is going and how we get there. And how we get mobile computers there as fast as desktops will get there.
I think that's one of the driving motivations inside of Intel - to eliminate the mobile-desktop gap. With the MMX processors they've eliminated the gap. It's opened up slightly with the Pentium II but it's not that far behind. It's not like the 386 where there is a year or an 18-month gap between the processor on the desktop and the processor in a mobile.
The networking group at Intel is all about bringing fat pipes to the PC to drive consumption of higher-powered processors. They were doing that completely focused on desktops and now they want to bring those same fat pipes into notebooks and force people to buy higher-powered notebooks in the future.
How do you see the role of Windows unfolding in the PDA market?
We have been platform and OS-agnostic, if you will. For the most part, virtually everything we do is for the Intel platform because that's the way notebooks work.
In terms of network operating systems, there is a wide variety of those and we support virtually all of them either directly or indirectly through drivers that we've developed or helped others to develop. From an operating system point of view, it's the same thing.
Whether it's Unix or a Unix derivative or any version of DOS, there are all flavours there. It seems to be that in the notebook market of today, the world seems to be gelling around a Wintel solution.
Increasingly, you're going to see some versions of Windows, 95 and NT, running on an Intel-based platform. So I don't believe there is going to be much of a digression.
What is interesting is that this palmtop PDA market is in its very early stages right now. You could almost draw an analogy to the PC market pre-IBM where you had CP/M and all these various people with their own versions.
The market is trying to figure out where to go and we haven't seen someone step in and set a standard. Bill Gates is trying. He's trying with Windows CE, but Bill usually gets it right around revision 3.0. We're at revision 2.0 now, so we'll see where that goes.
One thing that he has going for him is his persistence. Microsoft never gives up. Right now it's an early enough market so we're going to have to see where that one goes.
With Windows CE, Microsoft has been trying to cram a large desktop OS into a smaller and more mobile package, apparently without much success. Is this a workable strategy?
I think that is in fact one of the failings of Windows CE. Instead of Windows CE devices getting smaller and leaner and figuring out the minimum subset of functionality required, I've watched Windows CE devices evolve up towards the notebook side as opposed to slimming down toward PalmPilot or less.
I believe that Microsoft, because of its OEMs for Windows CE, is being dragged in the direction of a light sub-notebook - something that is going to have a colour screen and a full ASCII keyboard. It's going to be a light version of something that can do e-mail. But, I don't know if that's where the market wants to go.
The PalmPilot has been a very successful device and the Rex is showing a lot of promise. But, people still want something lighter. I see Windows CE drifting in the wrong direction - it's getting bigger. PalmPilot can do things with hundreds of KBs of memory that Windows CE is now requiring 8MB to do.
Do you think that somebody coming out of left field, without millions of lines of existing OS code to worry about, is going to be better prepared to dominate this sector of the market?
Time and again in this industry, you have seen that whenever you get a major technology shift, that's when somebody new can pop up. That's when we saw IBM was upended in the shift from mainframe to mini by the HPs and Digitals of the world. Major shift. Those folks were upended again by IBM coming into the PC side. That's how Microsoft got its start.
I think everyone in high tech has seen this happen so often that we're all running paranoid, but even that paranoia can't always save you.
From Xircom's point of view, what scares you the most?
This new palmtop, PDA, Windows CE area. I see that in a gestation period. I see a lot of different ideas being tried and I think something is going to gel - something is going to catch on there.
I don't think we're late yet. We are looking and we're doing a lot of investigation. We can't afford at this point to put a whole lot of research and development resource against it because we'd be building interfaces for a huge number of devices if we wanted to address the market today.
But, I think as the thing narrows down and we start to get a couple of winning platforms and some winning solutions, then we start going after those in a big way.
It's just not clear to me where the whole thing goes