chaynes@sunfish.cs.indiana.edu (Chris Haynes) (01/09/91)
Archive-name: languages/scheme/r3.99rs/1990-12-12
Archive: altdorf.ai.mit.edu:/archive/scheme-reports/r3.99rs.tar.Z [18.43.0.246]
Original-posting-by: chaynes@sunfish.cs.indiana.edu (Chris Haynes)
Original-subject: Re: Scheme standard approved
Reposted-by: emv@ox.com (Edward Vielmetti)
In article <TED.90Dec10215756@kythera.nmsu.edu> ted@nmsu.edu (Ted Dunning) writes:
In accordance with IEEE policy, drafts of the standard will no
longer be distributed by MIT and Indiana University.
does this mean that the draft will no longer be available via ftp?
That's what it means. IEEE holds the copywrite on the standard (and
drafts of the standard). They encourage wide (cheap) distribution of
drafts in order to obtain the widest possible comment during the
process of standard development; hence the prior ftp distribution of
drafts. However, in order to assure that all copies of standards are
official copies, and to make money, they (and all other standards
organizations I know of) limit distribution of standards to copies
they publish, for which they charge what appear to many to be
exorbitant prices.
Their concern that copies be official is legitimate: a document
purporting to be a standard that had been doctored could cause a lot
of damage. Their need to make money is also legitimate. Money made
on the sale of standards helps subsidize a formidable administrative
system intended to insure, among other things, that standards are
developed in a fair and open manner, that they are distributed in an
uncorrupted form, and that proper liaison is maintained with other
standards organizations so that we don't end up unintentionally with
duplicate or conflicting standards. You may feel that standard making
should be paid for in other ways so they can be distributed as widely
as possible. Or you may feel that the administrative machinery should
be streamlined so it isn't so expensive. You may have a point, and
are welcome to take up the matter with the IEEE Standards Board. My
slightly informed impression is that the IEEE runs a tighter ship than
some other major standards organizations.
Typically the relatively high price of standards is not a sore point.
In most cases they are purchased by a small number of companies
implementing products said to conform to the standard, and the cost of
the standard is inconsequential compared with the cost of a corporate
development effort. The Scheme community is exceptional in that there
are a great many amateur Scheme implementation efforts.
There is also a need for programmers to know what standard Scheme is
so they can write portable programs. For other languages this problem
has typically been solved by the appearance of books that spell out
the standard features of the language carefully enough for most
programmers (but not for professional implementors). I hope such
books appear soon for Scheme. There may already be plans for revising
existing books to satisfy this need.
if so, isn't this a giant leap backwards for the scheme community?
Not really. Before starting the Scheme standard, we obtained from the
IEEE an assurance that the RnRS documents would remain in the public
domain, even though the standard is based closely on a draft of R4RS.
This draft of R4RS is all that most amateur Scheme developers will
need, and it is still available from MIT via ftp. Hopefully the final
version of R4RS will carefully specify differences between R4RS Scheme
and Standard Scheme.
Sorry for the length of this note, but I expect resentment of the
standard's limited availability and high cost is widespread. Though
both are regretable, I hope this note has made them more understandable.
Chris Haynes