guile-devel
[Top][All Lists]
Advanced

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

Re: what to do


From: Thien-Thi Nguyen
Subject: Re: what to do
Date: Wed, 15 Aug 2001 21:58:56 -0700

   From: Tom Lord <address@hidden>
   Date: Wed, 15 Aug 2001 17:35:13 -0700 (PDT)

   I experimentally dropped in to this list with a bunch of contextless
   suggestions for features from Systas that I think help make it a
   nicer dialect of Scheme.  I also raised the issues of inadequate
   testing in both implementations, and a deep design flaw in the GC
   they have in common.  But the result seems to be more the usual
   random walk around the design space.

   Any clever suggestions?

well, you've suggested some sound engineering initiatives.  why not
spearhead those efforts given the current state (code/people/momentum)?
these initiatives -- i'm thinking of, specifically, more rigorous
performance testing, and generally, more structured design process --
would be nice to have in place regardless of the design, and may bolster
redesign requirements / efforts.

in contrast, i disrecommend the wipe-the-slate-clean approach because,
as much of a rush as that entails, the downside is to belittle the very
engineering practices which you seem to be proposing, which undermines
your credibility, no matter your ability.  proof by existence is fine,
but if you want to walk the rigorous path, your proofs must be of same
form.

personally, i'm looking forward to you sharing your experimental
methodology that resulted in your above conclusion, since, analogous to
code that is not rigorously designed, experiments w/o published
methodology are not so pleasing.  on the other hand, if your methodology
consists solely of going rhetorical and emitting "the usual random" spew
based on half a dozen responses to your two dozen queries, maybe i'm not
so interested after all.  my question then would be: what did you learn
about people in the last five years?

these suggestions are not clever in my eyes, but i express them
sincerely in the hope that i (selfishly) can learn from your involvement
w/ guile things that i cannot learn from other people.  [insert geezer
whine about not wanting to get into Yet Another Scheme Implementation.]

thi



reply via email to

[Prev in Thread] Current Thread [Next in Thread]