I really like your description of how models enable reasoning about the problem. Ya, abstractions vs models. I'll drop the "mental" qualifier, thanks. > get l

I really like your description of how models enable reasoning about the problem. Ya, abstractions vs models. I'll drop the "mental" qualifier, thanks. > get lost in... Totally. Too many times, the implementation comes to be treated as the model, completely obfuscating the client's actual problem.