Meet the new Web, same as the old Web.
COMPUTERUSER has been spotlighting Web technology in its March issue since 1996. As one of the first “experts” of the Web–according to then Editor-in-Chief Steve Deyo–I wrote the first two cover stories about Web technology in 1996 and 1997. At the time, COMPUTERUSER only owned our Minneapolis edition, while it licensed content to a dozen or so independent publishers in 15 other markets. We served licensees’ various needs and diverging ideas about how to publish the magazine (some called it a newspaper) as best we could. We have since purchased all but two licensees, which helps promote consistency from market to market.
Just as our 1997 Web technology cover was published, I was hired as COMPUTERUSER’s managing editor. I remember our Seattle licensee complaining about the hiring largely based on the two covers I had written as a contributing editor. He said that there was no new information in the second cover, which somehow showed that I was not the “expert” Deyo claimed I was. When Deyo left a month later leaving me in charge, my relationship with the licensee became very interesting. Unfortunately, we do not have electronic archives back that far. But I have a pretty good idea what I wrote in those covers. The first one was long and comprehensive. The second one was short and elegant. But the Seattle licensee was right, to some extent. The second cover went over largely the same territory as the first, updating many of the points I had made the year before.
The Seattle licensee was wrong about me and my research, however. I was a top student in one of the few graduate programs in the country that focused on the new Web medium when I wrote those covers. But the year I transitioned into this job, the Web experienced a temporary lull in new developments. HTML and XML standards efforts were in their early stages and companies such as Microsoft and Netscape were elbowing each other in standards meetings for its key features. The standards battle slowed the pace of Web development for about a year. After the battle was over and the World Wide Web Consortium (W3C) released key standards, the Web exploded into a flurry of developments. Our 1998 cover, written by Contributing Editor Nelson King, was much more interesting than either of mine had been because there was that much more to say.
They say history repeats itself, and, whoever they are, they are usually right. This month’s cover story by Editor at Large Nelson King on Web services goes over much the same territory as the last Web services cover he wrote in June 2002. Again, it is not due to Nelson’s lack of credentials or effort; he is one of the foremost experts in the field on both Web technology and e-business. No, there has been a lull in Web services developments since last June. And, once again, standards are the culprit, at least in part.
While the economy doesn’t help, the biggest hindrance to Web services is the widespread notion that companies must choose between one of two Web services forks in a path through the wilderness: Microsoft or Java. Whichever way they go, their path will dictate a lot of other choices, such as development platforms, server architectures, database applications, etc. Choose the wrong path and it will be much harder and more expensive to bushwack to the other path down the trail.
With these big issues on their minds, it is no surprise that CIOs are squeamish about going down the primrose Web services lane. Even if they wanted to integrate their enterprises with Web services, they can’t get the money to do so with so much uncertainty in the market. If they knew ahead of time that both paths lead to the same destination, the choice would be simpler. Improved and expanded standards efforts could give them the confidence to move forward, knowing that the end of the trail is the same no matter which fork they take.
And the end should be the same. The whole idea of the Web from the day Tim Berners-Lee thought it all up consisted of a cross-platform technology, enabling users to exchange data and collaborate with anyone else on the planet no matter what systems are in use. Berners-Lee now runs the W3C, which has been very busy since the mid-’90s, churning out enough standards to make your head swim. In the case of Web services, we’re talking about integrating or sharing applications between organizations. The Web services working group is no exception. But we are facing a time similar to 1996, when standards development slowed and there was a realization that much more work needed to be done.
Java only adds to the confusion. The standard is held not by a neutral group like the W3C, but by Sun, which is trying to walk the fine line between opening it up to everyone (including its sworn enemy in Redmond) and retaining the rights to the technology so as to profit from it. So far it has leaned toward the latter route, much to the consternation of Microsoft. Rather than accept this and adopt the Sun Java standards (J2EE), Microsoft has developed competing languages and development environments, such as J++ and C#, that work like Java but don’t work with Java. (Microsoft used to call the suite of technologies .Net, but the new term is .Net Connected. We’ll just refer to it as Microsoft’s approach.) In theory, the openness of the other Web services standards, such as XML and SOAP, should allow C# applications to work with J2EE applications. In practice, that is far from certain.
“Can’t we all just get along?” is a question Neil Charney from Microsoft, one of our Q&A subjects in this issue, hears often in dealings with customers. In his view, it is a matter of two lines of code or a couple of clicks of the mouse to turn any Visual Studio application into a Web service that should work with any other Web service-enabled application. For him, it’s a myth that Microsoft’s more proprietary approach makes it harder to build Web services with non-Microsoft partners. That is not the way others in the industry see it, including Bob Suter from IBM, the other Q&A subject in this issue. Suter says in the ’90s, the Web didn’t wane just because its standards were immature. It was stagnant as standards were developed. He thinks Web services work will be slowed–but not stopped–by the lack of standards in this case as well.
Until the standards are cleared up and there is proof that both forks in the road lead to the same destination, we may have to rest content with an elegant restatement of the major issues related to Web services. But if 1997-98 is any guide, we could see an explosion of new developments when the uncertainty lifts in 2003. Stay tuned.