First the bad news: Java is still equal parts problem and promise. Remember all those claims Sun Microsystems made about Java being secure? Virus-free? An easy transition for C++ programmers? Well, none of that has worked out, but one thing remains the same - performance is a significant limiting factor in building Java applications.
The corollary to this is that Sun still doesn't seem to know how to make money from Java. The big problem? Wintel's price/performance is improving and Java's isn't. And there's not much the company can do about this besides what it's already doing: letting Java-mania distract customers from the problem.
Security is going to become a bigger and bigger problem as time goes on. What would the world be without viruses, packet flooding, cellular cloning and so on? We'll never know - at least not until we start to build new technologies that are secure and ready to face a nasty world when they first arrive.
Novell isn't getting better. Rumours of NetWare's demise aren't true yet, but they may be self-fulfilling prophecies. It's third and long for Big Red. Maybe Oracle will buy them out.
On the bright side
Now the good news: Lotus Notes isn't dead. Many people seem to have given up on Notes, as if Microsoft Exchange were the answer to all questions known and unknown. I compare Notes and Exchange and almost see complementary applications. Sometimes, messaging is the key; that's where Exchange shines. Other times, replicated data makes an application go, which is why Notes still has a future. In addition, I expect to see Lotus bring out some interesting Internet-based applications.
Apple isn't dead. Buoyed by improved financials, better pricing and the emergence of something resembling a strategy, Apple's condition should continue to improve over the next year. Unless, of course, everything goes completely wrong, which can never be ruled out. But I'm feeling guardedly optimistic, to use the official medical term about Apple's condition these days.
Trifecta! Unix isn't dead, either. Windows NT is a nice operating system, but you'd be a fool to trust your mission-critical applications to it. That's not me talking; that's the collective wisdom of the user community. That doesn't mean NT won't improve, but between now and 2005 the world needs something comfortable and Unix will be it.
Forgetting my concerns about Java for a moment, the Java-based network computer is an interesting option for corporate computing. People tend to fall into three camps on this issue. One group immediately sees the value of network computer hardware and applications as a replacement for dumb terminals and dedicated PCs. The second group is those who "get it" once the network computer is explained in detail. And then there are the bitter-enders who say their users will give up their PCs when you pry their keyboards from their cold, dead fingers.
This latter group misses the point entirely. People who have a personal relationship with their computer aren't good candidates for the devices. But network computers could have a big future in operations where computers are assigned to functions or tasks rather than individuals.
David Coursey, an analyst and consultant, is editor of coursey.com., an online newsletter available at www.coursey.com