[racket] Blog post about Racket

Neil Toronto writes:

 > My dissertation is on programming languages for Bayesian analysis, which 
 > can handle arbitrary, possibly recursively defined models and arbitrary 
 > probabilistic conditions.

I have only skimmed it until now, but it looks like a very important
step in a direction I like. Beyond the application to Bayesian
analysis, I like the idea of computation in "Cantor's paradise",
which looks like an interesting compromise between today's
dominant trial-and-error approach to correctness and the optimal
but extremely laborious approach of formal proofs.

 > My broader research agenda is to do exactly what Konrad is hoping for.

Sounds good :-)

Konrad.