Fun to read. This seems to have much in common with type theoretical approaches to program correctness. The ultimate type system is one where a program has a type iff it implements a desired algorithm (expressed as a type). So in your terminology type theory defines "essentially the same" as "having the same type," or, more properly, provably equivalent types since these systems generally can't have minima.
You might look at http://www.nuprl.org --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Algorithm Geeks" group. To post to this group, send email to algogeeks@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/algogeeks -~----------~----~----~----~------~----~------~--~---
