Michels Addresses SCO Forum

Speaking recently at SCO Forum to a group of developers and resellers, Doug Michels, cofounder and chief technical officer of the Santa Cruz Operation, had some pointed words to say about the direction open systems was taking throughout the industry. As UniForum's cofounder and one of our association's visionary leaders, Michel's comments, excerpted below, will interest the UniForum family and those of us whose jobs and careers are tied to the future of open computing.

This industry is exciting and dynamic. It changes so fast that I know you're afraid to take a vacation because when you come back you know you're going to be obsolete.

The pace of innovation and change is enough to make you seasick. The changes never stop and this year is no different--but it is, sort of. This year's changes are more dramatic and profound than what we've seen in a long time.

Let me set the stage. When we started SCO, Unix was just becoming popular, and PCs were not even thought of yet as a movement. We picked Unix because of the movement in the industry. This was the time of the proprietary trap where customers had to use one company for everything--printer, cables, you name it. If they didn't, it wasn't going to work. They had to choose once and never again. Once they made that investment in the software, they were trapped.

The open systems and Unix movement were driven by people saying, "We're not going to take it anymore; we've got to have freedom of choice. We may love our vendor, but we don't like the feeling of never choosing again." The concept of not choosing doesn't seem fair during IT business negotiations. It's worse if the vendor changes strategies or goes away. So the open systems movement was driven by the need for standard interfaces across all platforms. It was an idea that disrupted the whole industry. We were the enemies of the incumbent vendors, like Ken Olsen at DEC, who was preaching about Unix as snake oil. (And now DEC is a leading advocate of open systems.) I'm here to declare victory!

The open systems/Unix wars against proprietary computing were a resounding success. We won everything we wanted. Users are no longer locked in. In fact, the biggest complaint I hear from users is, "Gee, I miss the old days when I just had to make one call." We broke the monopolies. Now everyone is sitting around saying "Well, that's nice. What do we do now?" I don't think we should give up the good things about open systems. After all, they are important and vital. People bet their careers and their companies on them. But those things are not enough anymore.

We have a new movement. The whole PC revolution has changed the desktop. When we started the open systems movement, it was all about dumb terminals, character cell devices, and big servers that could run lots of users. Now we have a new world that's got to do with networked PCs, GUIs, and a whole new set of de facto standards on the desktop. Two years ago I said that we at SCO had to be Windows-friendly. I got a lot of people mad at me that day. They called me names like "traitor", and said that I "didn't believe anymore." I did believe, but I also knew you can't fight reality.

The world is moving to a productivity desktop that has huge support and acceptance in the marketplace, and support from the ISV community. They're not as open as us true, open-systems zealots, but they're pretty open. They publish interfaces, they run on all sorts of hardware, and they do a good job of providing users with as much freedom as they really need on the desktop.

Customers of business systems still want more. They're really saying, "We got our freedom, but we aren't getting what we need to make our users happy." Users sit there with a powerful thing on their desktop running powerful office automation software, and MIS sits here with huge servers from all the big Unix players with gigs of great data, and it's still too hard. Users can't pull up data from two databases. Users are rebelling. They're not terrified by the high priesthood of MIS anymore.

We're starting to deal with users who grew up with a computer, who went to college with a laptop. But they don't know how to do business computing. They don't know how hard it really is in business computing to get integrity, reliability, backup, and recovery. What we have are users who have expectations that are at one level regarding ease-of-use, power, and flexibility and business systems at another level in terms of what they can actually deliver.

It's time to stop thinking about technology and start thinking about how to solve user problems. We must move away from the religion that says "if we can just get common APIs we'll have solved everything," and move toward the religion that says "give data to the users so they can solve problems."

We're at a point now where we must stop talking about open systems (not stop using them) and start talking about open solutions. We've got to use all this power of standard technology and standard interfaces and begin to put it together so people can solve business problems. It's a big challenge for all of us. We have a clear understanding now of the results we want to achieve, but we still have difficulty getting real solutions.

In the past all we cared about was open systems. All we had to do to be architecture experts was go to meetings at X/Open to find out what was going to happen to open systems. We then shared that information with our partners.

But as we move into the world of open solutions, it's not as easy anymore. We can't just go to X/Open to get an architecture that will tell us what the next thing in open systems is going to be. We must invent our own architecture that leverages what's happening in open systems and at the same time leverages what's happening with Windows, so that the combination allows you to put them together. Where are we going in the future? What are my predictions?

First of all, I think that solution complexity is going to get much higher. We're going to be asked to solve much harder problems than we were asked to solve a few years ago. This will change what we all do. Users can do the easy stuff themselves--accessing files, plugging in users, and going to the computer supermarkets. The real added value will be found in the harder problems of integrating corporate databases in wide-area networks, building gateways to the Internet, and combining information that flows between different businesses.

Second, the industry is maturing. We'll see consolidation of all kinds. We see fewer players today and stronger players emerging out of consolidation. The complexity of solutions now forces users to use fewer vendors. The more vendors, the harder it is to make things work.

Another thing to watch for is the 64-bit APIs with all the leading Unix vendors and Intel. The industry seems to be saying "we know this change is coming, so let's all get ahead of it and be ready." For the first time with one of these standardization efforts there is nobody against it. Everybody participated and wanted to be part of it. We don't need to argue about it. Let's just pick answers to problems with development and use it. The 64-bit Intel P6 solution will be a shift point in the industry.

We're entering a time where people are becoming much more pragmatic. Idealistic thoughts about computing are going away, and true believers are few and far between. These pragmatists say, "I just want something that works. This is just too hard." It's no longer acceptable to be totally involved with one religion. We must be driven by solutions and not just a single way of thinking.

My advice? It is really important that all of you understand how big these upcoming changes really are. You need to know how the transition from an open systems attitude to an open solutions attitude will affect customer business, and you need to start learning about things you don't know about. If you're a pure Unix person, you've got to learn about Windows. If you're a pure TCP/IP person, you've got to learn about IPX/SPX. There's not a lot of room to be a specialist. You've got to learn how to integrate solutions and deliver answers, not just more questions.