IBM is moving forward with what it calls "On-Demand Development and Morphogenic Systems." Despite the mysterious and unfortunate choice of names (what the heck is morphogenic?), there are glimmers of hope that the company will produce tools to make your life easier and your systems more stable.

The basic ideas IBM is pursuing are not new: These types of tools have all been built already. Most of them have failed because prototyping and code-generation tools generate bloated software, and until recently that was a problem. These days, though, CPU speeds have increased to the point that this concern is not so significant in some applications. Do I want my high-performance Web server to be running generated code? No. Do I think this is a good idea for the interface to that legacy system that's hit a few thousand times a day? Sure.

The other issue that has held this type of toolset back is not so easily overcome, however. It is resistance. You'll find it difficult to convince a staff that has been able to customize or fix code that a generation tool will not limit their ability to respond to problems and business needs. By lumping the tools in this kit together, IBM may in fact be invoking this same resistance.

Individually, there is much to be said for some of these tools. Automated generation of integration code for legacy systems is a great idea. As for having a tool to automate Java standards conformance and perform even minimal coding error checking, there are others out there, but I would try one by IBM if it were available.

Overall, when IBM starts touting Morphogenic Computing, make sure you ask for details. Find out what the company is offering you. IBM is taking a risk by hiding some excellent tools under a strange-sounding name; if you don't delve deeper, you risk missing some tools that could help you better serve your customers.

Post a comment or question on this story.