Microsoft Singularity

I just read my colleague Mary Jo Foley's post Want a peek at a non-Windows operating system from Microsoft? and thought I'd add a few additional thoughts on Microsoft's research project called "Singularity."

I was a bit amused by Microsoft's comments that Linux, UNIX, Windows and the like are all based upon 40 years of development and the futher implication that somehow that was bad. Gosh, it seems to me that 40 years of refinement and enhancement was something good not bad. It means that many people have examined the software and have worked to make it better and better. Software is not like bread, it doesn't get stale just because it has a track record that stretches back 40 years.

Organizations typically to follow the Golden Rules of IT when working with technology. This tends to preclude moving things just for the sake of moving things. So, this research project from Microsoft, if it ever hopes to be a commercial success, must make it easily possible to bring today's applications over without an enormous amount of pain.

While I applaud the effort to look at today's problems and think of new ways to address them, if the new approaches cause major disruptions, require retraining all of the developers, or render years of application developmenet obsolete won't be all that interesting.

I guess we're all going to have to sit back and watch this. It just may evolve into something interesting. It may, on the other hand, not lead to the promised land.