PL seminar - David S. Wise, Pure/Impure LISP and Compiling

• To: plseminar@cs.indiana.edu
• Subject: PL seminar - David S. Wise, Pure/Impure LISP and Compiling
• From: Y. Annie Liu <liu@cs.indiana.edu>
• Date: Fri, 15 Nov 1996 15:27:13 -0500 (EST)
• Cc: cs.students.grad@moose, cs.colloq@moose

	    Programming Languages Seminar
	        Friday, Nov. 22, 1996
	         10:05am-11am, LH101

What compiler could solve Pippenger's problem?

David S. Wise

Nick Pippenger has posed a remarkably simple
problem that establishes a complexity difference
between Pure LISP and Impure LISP.
Several restrictions surround the definitions of those languages,
including matters of real-time, operational semantics.
Compiling, in its broader sense, is forbidden.

Setting aside operational semantics, the question arises
of how to solve his problem by writing a high-level functional
program that could be compiled into efficient target code.
What would it take to compile to his impure Lisp as target?

This seminar is intended to be a conversation, rather than merely a simplex talk.
Take a look at  /nfs/seafox/u/dswise/pippenger5.ps.Z  beforehand.
This is a revison of his POPL97 paper that is accepted for ToPLaS
and is posted with his permission for this seminar.  (Please do not distirbute.)

(Recommended:
    zcat < /nfs/seafox/u/dswise/pippenger5.ps.Z | psbook -q|2up|lpr -PpsXtumble  
where psX is your printer's ID.)

--
Looks like David is going to offer us an opportunity for a lot of
discussions even on polished work!

• Prev: PL seminar - Jon Riecke, Bell Labs, Security in Lambda Calculus
• Next: PL seminar - Erik Hilsdale, MetaML and Staged Computation
• Index(es):
  • Main
  • Thread