Re: [MIT-Scheme-devel] improving integration of VALUES and CALL-WITH-VALUES

[Taylor R Campbell]

Any preference now about which transformation to commit, between the
conservative and aggressive options?

Closer inspection of LIAR's subproblem ordering suggests that in
programs that the transformations are applicable to, LIAR wouldn't do
anything interesting anyway, since it wouldn't recognize the programs
as cases in which it can find any better ordering of subproblems than
a random one (i.e. right-to-left).  So I'm inclined to commit the more
aggressive one, partially just because it's a little easier to
understand the code, to which I can add the more detailed comment.