[r6rs-discuss] Rationale issues

From: AndrevanTonder <andre>
Date: Thu, 28 Jun 2007 17:01:49 -0400 (EDT)

On Thu, 28 Jun 2007, Thomas Lord wrote:

> Thus, in
> the draft, implementations are expected to "peek into" the internal
> structure of library names and do things like version number
> comparisons according to the editor's particular whim of what those
> comparisons entail. What is missing from that "pro" argument is
> any rationale that would persuade us such a feature ought to be
> standardized. The list of languages which do not specify built-in
> version numbers as part of how library names are resolved is quite
> impressive -- the draft proposes that Scheme should be quite
> exceptional in this regard.

Not to mention unmodular. Each time I check in an update
of a library in a large project an obtain a stable release,
am I to update the source of all 43 dependent libraries to
reflect this? And once I update them, how about the 43
libraries that depend on each of these?

There are good versioning tools that keep things consistent
on a higher level than the source of the stuff being kept
consistent, are free, and are in common use in the industry.
Anyone in his right mind who does projects large enough to
need version control will already be using one of these.

I cannot imagine that the editors would have chosen someone
ignorant or without experience of such tools to come up with
the current proposal. So maybe there is another reason for
the current proposal that Scheme systems duplicate this
functionality by having an extremely handicapped built-in
versioning tool, unstated for political reasons, perhaps
to do with the distribution method of one or more large
Scheme implementations?

Andre
Received on Thu Jun 28 2007 - 17:01:49 UTC