[R6RS] Semantics

Michael Sperber sperber at informatik.uni-tuebingen.de
Thu Feb 16 04:41:29 EST 2006


In Tallinn, the four of us were there briefly discussed the issue of
the presentation of the formal semantics in the report.  There are two
obvious options:

1. Amend and extend the denotational semantics in R5RS.  (I could
   fairly easily copy and paste the semantics I wrote for our Scheme
   workshop paper, which includes specifications of dynamic
   environments and DYNAMIC-WIND.)

2. Ditch the denotational semantics and subcontract Robby Findler and
   Jacob Matthews to donate their small-step operational semantics
   from last year's Scheme workshop.  Here's the paper:

   http://people.cs.uchicago.edu/~robby/pubs/papers/scheme2005-mf.pdf

There are advantages to either approach.  Robby's and Jacob's
semantics is more complete (it covers unspecified evaluation order,
for instance), and is immediately executable, but I find the
denotational semantics easier to read and understand, and more useful
as a guide to implementing the language.

Preference, anyone?

-- 
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla



More information about the R6RS mailing list