Re: when GC is permitted Tom Lord 10 Jan 2004 21:48 UTC


    > From: Michael Sperber <xxxxxx@informatik.uni-tuebingen.de>

    > >> From: Richard Kelsey <xxxxxx@s48.org>

    > >> I can understand wanting the first FFI SRFI being a safer, more
    > >> general one, perhaps based on JNI or Pika.  This SRFI isn't that
    > >> SRFI because that isn't the type of FFI that Mike and I needed.

    > Tom> In off-list conversations about SRFI-50, that paragraph has raised a
    > Tom> few hackles. [...]

    > You're misinterpreting what Richard wrote.  He's explaining the
    > *history* of the draft, not the rationale for the future, which
    > Richard, I, and you don't know yet.  In your response, you changed all
    > occurrences of "need" to the present tense (note the past tense above)
    > which, urmh, changes the semantics to something that raises a few
    > hackles with me.

I'm sorry I raised your hackles.  My intention was only to give direct
voice to concerns that are "out there" (and that I share) -- but I
hoped to do that in a bridge-building way, giving you both plenty of
opportunity to answer those concerns (which, in fact, you've begun to
do).  I also wanted to use the occaision to propogandize a bit for
what the SRFI process can accomplish when it is functioning at its
best -- especially in this context where we're starting off not with
some clear win (in a simpler domain :-) like Olin's list library but
rather, with a huge potential win that will ("IMO", of course) need to
undergo a lot of revision and work by the community.

As I said: my favorite axiom in trying to puzzle out the meaning of
you two is "the reasonable person principle."  I would ask that you
apply that same principle in reverse: One possibility might be that
I'm being gratuitously insulting to you but that would violate the
reasonable person principle (it would imply I am not a reasonable
person).  Perhaps some other possibility is therefore the more likely.
That an expression may be unpleasant or critical does not (or at least
should not) mean that it is disrespectful: it may be (in this case I
hope was), in fact, respectful.

As for changing the semantics in my paraphrases: I think that the
paragraph from Richard is at least ambiguous -- though I wonder if
there isn't a difference in perception because we come from such
divergent linguistic communities.  To my ears, the paragraph doesn't
(other than in the slim evidence of the past tense of "needed")
contain anything that suggests the possibility the authors will come
around on these issues.  On the other hand, the present tense of "This
SRFI is", the use of "SRFI" rather than "draft" in "This SRFI is", the
impersonal and indirect "I can understand wanting", all in the context
of earlier messages -- all of these strongly suggest, to my ears, that
the authors are not taking seriously the proposal to switch to
different calling conventions.

Let's get back to work then.

-t