Forever loyal to the thin-client way of computing, Wyse Technology continues to extol the virtues of lowering IT costs by minimising complicated client technologies. Although the market has failed to live up to the hype of years past, recent developments are encouraging Wyse about the prospects of going forward. IDG's Dan Briody sat down with Wyse CEO Doug Chance to discuss the lay of the thin-client land.
IDG: After the hype of two years ago, has the thin-client market settled into a niche, or is it still defining itself?
Chance: I think that the thin-client market is at a very important period now. It's at a period that we might call the post-hype phase, where we're seeing a consolidation focus on two platforms, in contrast to the confusion of a year ago where there were all kinds of things that were called thin clients - Net PCs, network computers, Windows-based terminals, Internet terminals, etc.
We believe that the Net PC has really gone nowhere, that the network computer is a big question mark. The implementation of [Oracle CEO] Larry Ellison's vision never came about; it was a fat client, not a thin client. But we may see a revival there yet.
But for now, the two platforms that we see are the Windows-based terminal that's driven by the Microsoft involvement, and what we call the network terminal, which is a replacement for the dumb terminal but has all the capabilities of a built-in browser, as well as multi-Windowing access. So from that perspective, it's a period of convergence.
Are you seeing any replacement of PCs happening?
The thin client isn't directly competing with the PC. In fact, as PC prices have dropped in the last year, that's given people a chance to evaluate: do they need a PC or do they need a thin client?
However, I would say that it's replacing PCs being used as terminals. And there are a lot of 486-type machines out there.
Given Microsoft's proclivity toward fat-client software, do you feel that they are the right partner to be with in the thin-client business?
Wyse is a peripheral company, so we basically make peripherals for all the different computer platforms that are out there, and we haven't found the need to partner with one side or the other.
We've had two platforms virtually from the beginning. We don't sell the Java-based network computer anymore. The operating system was too fat. We also licensed the NCI Oracle application stack, and that was too fat. But the [Winterm 5000] that we recently introduced is in many ways a reincarnation of our first product.
It uses the Linux OS kernel, runs the Netscape Navigator browser, and supports all of the different terminal emulations in the world.
What do you see as IBM's role in this industry at the moment?
At the original thin-client events that Ellison put together, IBM took a leadership role. But it basically articulated a vision that had two parts to it.
One, that it is a systems integrator company and it will do anything anybody needs or anything anyone wants. And two, it is a strong believer in Java.
IBM sees Java as a way to rationalise all those different legacy operating systems that it has. So it's been a strong proponent of thin clients, and I would say that its product is very similar to our network terminal.
It has avoided the Windows-based terminal from a strategic perspective. It is anti-Microsoft. It acquired Lotus and it strategically doesn't want to acknowledge Microsoft in the large enterprise.
How have low-cost PCs affected your business?
Unfortunately, Ellison and [Sun President and CEO] Scott McNealy had the right idea for the wrong reasons. Their first-round implementations were not thin clients.
I think that the thin-client hype did some good for the PC business, in that it made everybody aware that what users wanted was lower-cost products. There was an inequality because the computer companies, led by Intel and Microsoft, believed that bigger and more powerful was better. The early studies that were done on thin-client computing showed that the primary reason to implement was lower total cost of ownership. There was a study that was done two months ago, that took into account the low-cost PC, and it came back and validated the same conclusions.
And there are two reasons. One is that if you looked at the total cost of ownership, the cost of the product was only something like 13 per cent of the total cost. So if you reduced the 13 per cent cost by 20 per cent, it was still peanuts compared to the total picture.
The second is that the low-cost PCs that everybody reads about are Windows 95- or Windows 98-based products; they're not Windows NT-based products. And in the commercial enterprise environment, these people are implementing NT. And the software cost really looks a lot different.