In-Depth

A look forward, a look back

• By Tony Baer
• November 1, 2004

Because Longhorn remains a vague concept to most Microsoft customers, a clue to its eventual success might be found by looking at how the installed base fared with .NET, released barely two years ago.

The verdicts are mixed. “It’s been slow, but then again, our clients tend to be slow adopters,” observes Jeff Bocarsly, VP and division manager for functional testing services at Real-Time Technology Solutions, a New York-based software testing consulting firm. For instance, a cursory glance at message boards and online forums, such as DevX, reveals that developers are still asking basic questions about how to transition from classic Visual Basic programming to the newer .NET-based languages.

“Most evidence is anecdotal,” concedes Daryl Plummer, group vice president and research general manager for software infrastructure at Gartner, who characterizes most .NET adoption as mere lip service. Although almost all the clients he has spoken with claim they are building with .NET, he estimates that perhaps only 20% of the code is actually compliant.

“We don’t have hard numbers, but anecdotally, managed code is in,” counters Ted Neward, editor-in-chief of the TheServerSide.NET, an online forum focusing on .NET technology.

For instance, Corzen, a New York-based firm that specializes in syndicating employment data for clients in media, advertising and financial services fields, adopted .NET technology when it started two years ago. CTO and co-founder Steve Forte cites features such as ASP.NET, which automates the generation of Web pages, for enabling his team to roll out the company’s product in three months, barely a quarter of the time it took for a similar job at a previous company. Consequently, says Forte, Corzen was able to get to market and generate revenue sooner, thereby avoiding the need to raise venture capital. “That proved a huge difference in our business model. We otherwise would have been in debt or had a different business structure,” he says.

However, for Jay Glynn, development manager for a large insurance company, the going has been slower. Although his team has at least five senior developers relatively well versed in .NET, they are only getting around to their first major .NET project this year. In rewriting an insurance agent sales system, .NET’s object-oriented features should make the new system far more flexible, capable of responding to changes in state regulations or market demands without wholesale application rewrites. Glynn attributes the project’s timing to a slow economy, which discouraged major new projects. However, with agent hardware closing in on five years, the system refresh cycle provided the opportunity to justify a rewrite.

So how has the transition to .NET gone for Glynn and his colleagues? He notes that, although learning the syntax of the new .NET languages was fairly simple, internalizing the new object-oriented disciplines wasn’t. “Since day one, I’m still having ‘Aha!’ moments,” Glynn concedes.

Please see the following related stories: “Longhorn debuts, but few pay attention” by Tony Baer “The basics of Longhorn ” by Tony Baer