[racket] shared and immutable structures

On 2/1/11 3:55 PM, Robby Findler wrote:
> The other David seemed to suggest he had an app where immutable cyclic
> data was important. Maybe it would be useful to hear a few more
> details to see if Matthew's reasoning holds up there?

That the cyclic data is immutable is not terribly important; it just 
seemed like the natural thing for the problem at hand (I'm trying to 
represent recursive environment structure in an evaluator).

David