A long time ago, in what now seems like a galaxy far, far away, a software guru by the name of Gerald Weinberg wrote a book called The Psychology of Computer Programming. Several of my colleagues and I stumbled upon the book by accident when it first appeared in 1971, and we were all thunderstruck -- no one had ever suggested that software development might be considered as a human activity.

A few days ago, Apple Computer announced quarterly earnings of 68 cents a share, far exceeding the predictions of Wall Street analysts. This was the fourth consecutive profitable quarter for the company that even diehard loyalists like me had just about given up on. I hate to jinx the company's future fortunes, but I think the time may have come to celebrate Apple's phoenix-like rise from the ashes.

Companies thinking about upcoming changes in their corporate architectures are going to want to pay special attention to the evolving role of application servers. High-end application servers will increasingly serve as the heart of most new distributed Internet applications.

When I got into the software field, back in the Dark Ages, I couldn't believe that someone would actually pay me an above-average salary to have fun. I suppose there are other industries where people have fun -- baseball players seem to enjoy themselves, from what I can tell -- but it was hard for me to imagine that there was anything more intriguing, more challenging, more satisfying, and more exhilirating than developing and debugging a large complex software system.

I had lunch a couple weeks ago with George Schussel, the founder and CEO of Digital Consulting, Inc., one of the country's largest IT conference organizers. We compared notes on interesting trends and developments in the field, but he stunned me with the news that DCI's "Database World" conference has shut down. It was the company's flagship conference for over 15 years, and it enjoyed a monopoly for a year after the demise of its only competitor, the massive "DB Expo" conference.

Traditional software developers often dismiss Y2000 as a simple problem, arguing that it involves nothing more than a large number of intellectually trivial code-fixing problems. Y2000 projects don't require any systems design or requirements-definition activity, right? Yes, there is some testing to be done, the software developers acknowledge, but even that should be simple because all of the test data is concerned with the same date-related issue.

Following are the typical justifications used to support outsourcing. Some have been well-covered in the media, while others are less public and more controversial. In next week's Alignment E-Mail Advisor, we'll look at the case AGAINST outsourcing.

A respected colleague who specializes in CASE tools and object- oriented methodologies wrote to me a few days ago, suggesting that a comparison between large Y2000 projects and large "normal" projects was like comparing apples and oranges. "All in all," he suggested, "Y2000 projects lack many of the complexity drivers which condemn a large percentage of new developments to failure. This should mean that they are generally much more likely to be completed in time."