Tuesday, May 27, 2003

Okay, now to business, or rather the future of business.

I've been pondering the areas of research that need to be done in computer sciences. Over the last couple of weeks I have come up with a couple that are intriguing.

1. 'the id of the computer' - this is an idea - double entendre intended - that goes to the heart and soul of a computer. How is a computer identified and how does it identify itself? How do we reliably and securely recognize a computer, and vice versa, how does a computer reliably and securely recognize itself? It seems that the core of this is memory, the organization, and the association of memories....

2. 'visualizing software development' - this goes to the thought that if you can see something being developed, you have a better chance of determining if it is going well, adjust as needed, and you can have an entire team working hand in hand on the same problem, rather than discrete unconnected pieces. My first thought on this is to use CVS as a source for visualization data. Using the frequency, quantity, and quality of changes in a region of CVS could be used to give visual hints about the activity that is going on in a software project. This relies on an assumption that the organization of the source is somehow reflectvive of the overall project. The key would be to use some 3d tree viewer that could color code the sections that were being modified. You could see higher frequency by the shading -- with, for example, red being a very hot area of development.

No comments: