[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Issue: HASH-TABLE-STABILITY (version 1)

This proposal is so long that I got lost while reading it.  From the
examples, one would think it was proposing some rules about what users
can expect when they modify objects that are used as keys of hash
tables.  However, I couldn't find anything actually proposed about that.
Most of the proposal seems to be about what performance users of hash
tables should expect, but I didn't see anything specific enough that I
could write a Common Lisp program to test whether an implementation
conforms to the proposal.

I think this proposal needs to be shortened and rewritten.  I would
prefer to see it speak about the behavior a user can or cannot expect
 from a Common Lisp implementation, rather than in terms of internal
details of how Common Lisp might be implemented.  The essay on
implementation techniques could go in the discussion section, or could
be published separately, but I don't think it is suitable as a language

It might be a good idea to break this into two proposals, one on key
modification and a separate one on performance.  The reason I say that
is that I think standardizing performance is extremely difficult, and I
would hate to see the problems with that sink the other proposal.