Daniel, Isn't that the general idea that Einstein used in his relativity formulation?
In any case, this thread is kind of split in two. Part of it is a discussion about Sussman using an interpreted language in doing physics and another part is about a specific physics problem.
Addessing the former, I thought I would show what I have been doing in the past few years with Prolog, and specifically the SWIPL distro. The semantic web infrastructure I lean on has incorporated the Notebook paradigm elegantly.
Add the SWISH package and it appears in the menu:
Query the knowledgebase using an RDF triple relation, i.e. rdf(Subject,Predicate,Object).
In this wildcard case, you see query results from the NASA JPL SWEET ontology.
It handles declarative representations of charts. Format the data structure and the notebook rule will attempt to render it,
The same for directed graphs.