re: extension to STk

From: Tom Hicks <hickst_at_lx2.tucson.saic.com>
Date: Thu, 16 Feb 1995 13:31:41 +0700

Subject: re: extensions for STk

>Date: Thu, 16 Feb 95 13:54 EST
>From: dsf_at_gti.net (David Fox)
>To: stk_at_kaolin.unice.fr
>Subject: Re: Building Extensions for STk

>These theoretical discussions are all very well, but not if our
>conclusion is that scheme is a lost cause (and I know I'm quoting a
>little out of context.) I think the original message is worth
>re-emphasizing - if you are creating an interface to some set of
>system services, take a look around and see if there is a popular
>implementation already out there whose interface you can mimic.

As the person who sent out the extensions that kicked off this whole
thread, I feel I have to speak up. I agree that it's probably a good
idea to conform to what has been done by others. I apologize for my
limited knowledge of what's available in other Schemes. I don't even
know what the "popular implementations" of Scheme are. I tried to
use names and argument structures which were familiar to C/Unix
programmers. STk appears to do the same thing with its extensions
('flush' for example). But please remember that, as I said, I
created seek/tell for use on a project within our company.
  Our project appreciates all the work that has gone into STk and
when I saw a request for seek/tell functions, I thought that I might
save the requestor (and perhaps others) some effort. It seems to me
that an extension is, by definition, something which is not
standardized as a core part of the language. Use them or ignore them
as you see fit.
        -tom hicks
        hickst_at_aries.tucson.saic.com
Received on Thu Feb 16 1995 - 21:29:37 CET