The Changing Face of Developers

I've included a couple of photos from the Gartner Application Development Summit held in Phoenix at the J.W. Marriott on September 27-29. The first is me talking to Gartner analyst Theresa Lanowitz, a research director for the testing market and testing advisory services within software development organizations.

The second is Kayla White, the PR manager I work with at Compuware. Insofar as I get ink in the trade press (yes, including FTP's publications), the credit goes entirely to her. While I've worked on both the trade press and the vendor side of the software industry, Kayla's efforts and results have once again reminded me of the strangely symbiotic relationship between vendors and the press that she manages so well. But that's a topic for another day.

For those of you who aren't familiar with Gartner Group (www.gartner.com) and its ilk (Meta Group, Forrester Research, IDC, and a host of smaller ones), they play the role of referee between vendors and end-user enterprises. Enterprises hire them to advise on technology strategies, while vendors hire them to...well, advise on technology strategies. If it seems like a conflict of interest, most of these research companies handle it reasonably well. Compuware spends a lot of money on Gartner services, but I mentioned to Theresa that I was well aware that she doesn't shill for us. Nor does she take the role of user advocate, but rather waters down the hype of technologies and products into something useful in individual situations.

I sat in on a session about Java and .NET interoperability presented by Mark Driver, who is a research vice president at Gartner and is often quoted in the trade press. Driver proposed that Java developers today are largely highly technical and code-centric. These are the developers who understand the platform to an extremely detailed level, and spend their own time studying and trying out new technologies. He refers to this group as Type A developers.

But he sees that changing over the next several years. The highly technical developers focused on the comprehensive details of Java would gradually be supplemented by an increasing number of developers whose approach to software is that of a job rather than a passion. Highly technical developers will make up only about 25 percent of the community by 2008, he predicted.

This is significant, in that these "Type B" developers will be looking for different things from the platform and from development tools. In particular, they (or we, if you fall into that category) want the ability to apply platform technologies in less complex ways, while also being more productive at delivering applications. In particular, Driver mentioned modeling and patterns as approaches that would be popular with Type B developers.

My take is that over the next several years we will see significant changes in how we build applications. And it's an understandable evolution that should be embraced by the veterans among us, rather than feared. I date myself when I say that my first development experiences consisted of command-line compilers, linkers, loaders, and debuggers, often from different vendors with no integration between them. I think we would all agree that the modern IDE is significantly better than that.

Just as the "tool chain" of the past has thankfully given way to the modern IDE, we can expect that the IDE will increasingly include the ability to build applications using Java technologies at levels of abstraction above writing code. If the Java community wants to increase its numbers while enabling those new members to build complex distributed applications, this direction is not only possible but inevitable.

Posted by Peter Varhol on 10/03/2004