Originally Electronics Engineer entered research in early 90s in satellite communications building simulations, models and algorithms for satellite communications systems. The mobile revolution meant that this became mobile device systems models in later years. Spent most of my time tree-walking models and graphs of systems using proprietary software (read, homegrown "hey! we made this!") which meant that no two models were comparable in any way. Along came systemc and things improved somewhat. We now have models in C++ and, as we all know, C++ was specifically designed to model communications and SoC :)

So, here I am, still building models and simulations and designing system architectures. But, as I am a huge believer in "The Model is All" philosophy I now use EMF/xcore/xml/Xtend (Xtend is really great for walking trees and graphs) and generate the code (isn't the code just another view of the model?).

I play chess abysmally, read books I don't really understand and look out my window upon the sea while drinking vast amounts of coffee and call it work.

Software Philosophy

1. Make it work
2. Make it work better
3. Do it in that order

If you don't have a model, what exactly are you building?