Re: 4.2

[sandra%defun@cs.utah.edu (Sandra J Loosemore)]

> Cough, cough, you'll recall that my minor remarks on the compilation
> chapter asked the question how much of this should apply to both the
> compiler and interpreter, and ....
>
> All I'm trying to do is see whether there is some concensus among this
> smaller group before doing anything about it in the larger group.

As I said before, I have no objection to the idea of reorganizing all
of chapter 4 to integrate the discussion of compilation into the
presentation of the evaluation model.  I also agree it would probably
be helpful to have the standard specify what a well-formed program is
and that having such a thing might make it possible to simplify parts
of the compiler discussion, but I'd have to see a more detailed
proposal before I could give an opinion on whether the specification
you have in mind is reasonable, or how it should affect the
presentation of the material on compilation.

I do think, though, that just gluing material on constraints on
conforming programs that don't have much to do with the process of
compilation into this section as it stands now confuses (rather than
clarifies) the presentation.  And that's what's happened in the draft
Kathy sent me to review.

-Sandra
-------