Dear Senator!

Even if you are only pulling my leg I surely do appreciate the fresh approach!

I've been thinking on how to explain this better and without the jargon. Let me try this again...

One the goals of our project is compatibility checking. I was explaining how this will work to a knowledgeable bicycle school guy some months ago, and he pretty much immediately suggested that the "ultimate test" will be if two complete virtual bikes are built in our software, and then the frames are swapped.

Ouch!?

Go to the parts department guy at your LBS with the question: "What will be the issues will I have if I attempt to swap this frame for that one on this bike." If he can finish that in a few hours, then tell him you are also interested in another frame....

Our on the other hand we have already tested "firing" 6,000 focused little bite size "mini programs" written in our invented for bicycles language - it takes about 10 seconds. We see some exciting use for this technology.

However, these little programs (also called Agents, or Boids) require that the data be in unambiguous form. Also in order to keep these little programs small and simple we have to write "helper classes" to assist the programs - so interference fits is one helper, threads is another, and we think there might be about 20 of these for bicycles.

Maybe you can tell now some of the fun we are having trying to do something right that has never been done before. We need to start somewhere.

If we do this, your bike will be one of the best documented bikes on this planet.


Thanks,

Michael