Q&A: Kevin Klustner

Connecting the HP 3000 to the Enterprise

Walker, Richer and Quinn (WRQ) practically defines multivendor connectivity in the midrange system market, but it wasn't always that well linked. WRQ began by knowing the HP 3000 terminal better than anyone. The company started on the path to today's $115 million in revenues by selling one of the earliest HP 3000 terminal emulators, PC2622, in 1981. A few years later that product became Reflection, and by now it's well into Unix, Digital, IBM and other markets, riding a range of connectivity tools such as TCP/IP stacks, intranet access tools and mobile connectivity packages. The HP 3000 business, however, lies at the heart of the company's interests.

The HP 3000 is also fundamental to WRQ vice president of sales and marketing Kevin Klustner. Klustner joined the company six years ago when WRQ was less than half its current size, coming directly from eight years of HP marketing duties, including Worldwide Customer Support and HP's headquarters in Geneva. Klustner started as the HP product manager for WRQ, which in 1990 translated into knowing what the HP 3000 customer needed. As someone who cut his teeth reading HP Desk messages and helping to steer the Reflection line into a multiple-host offering, Klustner represents both WRQ's heritage as well as its vision of the future. We asked him to give us his view of how WRQ sees the HP 3000 connecting in enterprises, and how the lift of Internet networking ideas can influence the 3000's future.

Would you say it's essential to use nodes smarter than terminals to fully network today's HP 3000 users?

It's essential if you want to network them to more than a 3000. You can network PCs, and it's much harder to do that with a terminal. For a lot of customers, the PC is overkill for what it's being used for. But it's one of the democratic phenomenons that once end users get the power of the PC, dumb terminals just aren't going to hack it. I don't think we're ever going to go back to a cheaper alternative.

HP 3000 terminal emulation is no simple matter, but it continues to draw new entrants on a regular basis. What's WRQ's advantage to sites who might be considering one of these newcomers because they know them from another market?

That's exactly why Attachmate and White Pine are doing the HP stuff. Attachmate is responding to a lot of their IBM-oriented customers that have HP 3000s, and those customers want to buy connectivity from a single vendor. The difference for us is that we were born and grew up in the HP 3000 world, and we know HP terminals and the HP 3000 better than anybody else outside of HP itself.

But is there really a big difference for most customers in how much you know about 3000 emulation?

They don't care about terminal emulation so long as it works for them. As soon as some quirky escape sequence comes down, well... More importantly, connecting via network to an HP 3000 is no simple trick.

You've got competitors selling 3000 connectivity at a lower price point. How do you cost-justify a Reflection purchase to a customer who's shopping on price issues?

Our competition has been smart at carving out the low end of the market, and they've had a fair amount of success. Our target customers depend on mission critical applications running real well. We take great pains to test the software, and we make sure we make the investment in technical support to back us up if there is a problem. We still provide all that stuff for free. There's value there that's necessary in the marketplace and that our customers see in WRQ, and it's worth something to them. We can't provide that level of service or support our product quality if we give the product away.

Microsoft is trying to push the market toward bundled connectivity software. How does WRQ plan to counter that philosophy?

I'd be lying to you if I told you I didn't worry about Microsoft doing that. But I'd also say that Microsoft has bundled a terminal emulator that included HP terminal emulation since Windows 3.0 in 1990, and it just wasn't very good. The challenge for a company like WRQ is to continue to provide value over and above what Microsoft offers. This is what we do for a living, and we feel we do it better than anyone else.

Would you say that marketing can make products survive in competitions like that?

There's some validity to that, although I think you're going to see a backlash from all the marketing hype on this Internet stuff. It's already starting to happen.

Do you think the Internet is the "killer application" that can revitalize the HP 3000?

HP is putting enough effort into it so people who don't want to make the investment in adding Unix servers can use their 3000 to do these things. It's possible that a company might use a 3000 as an Internet server or a Web site, but I don't think it's a huge mainstream business. HP's got to get this proprietary networking access to the 3000 figured out before that can happen. It sounds like with the 5.5 release of MPE/iX you may not need NS/VT to connect to the HP 3000. If HP does a good job of that, the 3000 may be as transparent as a Unix server.

But won't that shave off some of your revenues from selling NS/VT connectivity?

It's the best thing for our 3000 customers. It will have a negative impact on our business. But DOS has gone away too, and we've managed to survive.

How do you rate the networking standards support of the HP 3000 compared to the HP 9000 or Windows NT?

I'm not an expert on this stuff, but my sense is they've done a pretty good job considering the proprietary nature of the 3000 itself. HP has had to rely on doing a lot of this itself, as well as encouraging its existing 3000 partners to develop network applications for the 3000. Given where the 3000 is and the size of the market, they've done reasonably well. I give HP a lot of credit for being aggressive in the Posix environment, which brings them into the mainstream networking environment without sacrificing the performance and investment in applications.

Companies have been proposing "Network Computers" for the last year now. Do you think such a device has a part to play in WRQ's offerings? What kind of impact could such an "NC" have on the HP 3000 customer?

From what I can figure out with my pea brain, it sounds like a diskless workstation. This stuff never dies. It would be supremely ironic if a terminal puts WRQ out of business, considering we started with terminal emulation.

In this case, the 3000 environment isn't any different from any others. I think there may be a role for these network terminals, just like there's a role for regular terminals. But the horse is out of the barn as far as computing on the desktop. The PC is a very powerful resource that end users have come to really like. It will be hell to pull that back in and give them a reduced set of functionality of a network computer.

We may make Reflection run as an applet inside a browser, but nobody has figured out a funding mechanism to incent people to do a lot of work on this stuff. There aren't a lot of applications that can be accessed in that kind of environment, because I don't think any of us have figured out the right economic model to get paid for it.

Speaking of business models, WRQ continues to provide free support that has recently won awards. How can you justify the expense of operating free support when others can't?

It gets back to where we provide value. Our customers feel like they're buying a long-term partnership with us. That includes the best pre-sales support and the best post-sales support once they buy the product. Providing the best quality tech support that we can, free of charge, is part of our strategy of providing a solution for IS managers. If they can't sleep at night, at least they're not awake thinking about our stuff. It's a competitive advantage for us.

HP is mulling over the idea of putting MPE/iX on the Merced RISC chip. How important do you think this is to the installed base of HP 3000 customers?

It's important psychologically for the customers. We all want to make sure that when HP is thinking about the future, they're thinking about the 3000. I doubt there's a huge performance problem with the current architecture.

Will there ever be a completely dominant operating environment for business computers, or is diversity a more likely outcome? In other words, does it matter if the HP 3000 isn't a leading choice?

Microsoft would love everybody to consolidate onto NT long-term, but I don't see that happening. The 3000 and the AS/400 and the mainframes will continue to play large and important niche roles. The HP 3000 has never been the number one or number two choice in minicomputers. The strength of the HP 3000 is that it's focused on a couple of niches, and it's really delivered over time a high quality environment to develop the applications people still rely on. That's still going to be viable as long as people feel the environment has a future. From a technology standpoint, it's very hard to run into a 3000 customer that doesn't like the system.

Copyright 1996, The 3000 NewsWire. All rights reserved.