Re: GNU Extension Language Plans

Tom Lord (lord@x1.cygnus.com)
Sat, 22 Oct 1994 03:06:33 GMT

Excessive exegesis is the greatest vice of internet discourse. I
would much rather use my posting minutes to work on the project than
to argue about the sources of its inspiration. However, I feel I have
a responsibility to clear up one point for the record, and to extend
an offer.

John Ousterhout quoted RMS and replied:

Sun recently announced a campaign to "make Tcl the
universal scripting language." This is a campaign to
convince all the developers who *don't* prefer Tcl that
they really have no choice. The idea is that each one of
us will believe that Sun will inevitably convince everyone
else to use Tcl, and each of us will feel compelled to
follow where we believe the rest are going.

Please understand that this "campaign" exists only in Stallman's mind.
[...]
If anyone has concrete evidence to back up Stallman's accusations,
please post it so we can all see it.
[...]
The only information I recall from Sun about Tcl is a couple of job
listings and one or two personal messages from me, which were intended
to keep the Tcl community informed about what my group is doing and
to solicit input.

Was the campaign only in RMS' mind? In one ``personal message'' to
the Tcl community, Dr. Ousterhout wrote:

I'm enclosing below my "official blurb" on what is happening in my new
group at Sun. [....]

Here's the blurb:

The Tcl/Tk project that I'm heading at Sun has the long-term goal
of making Tcl and Tk into a universal scripting language for the
Internet.

One of the job postings was written this way:

_________________________

SUN LABS Tcl/Tk Project
_________________________

Sun Micorsystems Laboratories, Inc. is embarking on a new project
directed by Dr. John Ousterhout. Our goal is to make Tcl/Tk the
universal scripting language.

By all means, Dr. Ousterhout is forever free to clarify his meaning or
even simply change the stated goals of his project -- but RMS' posts
are very much responsive to a reasonable interpretation of what
Dr. Ousterhout wrote in the past.

But the news isn't all contentious. Dr. Ousterhout explains the
positive emphasis of his campaign this way:

I also hope that Tcl and Tk will become a universal scripting
language for the Internet, but I hope to do this by making Tcl
and Tk so attractive that people *want* to use them, not by
somehow preventing people from using alternatives.

In that case, Dr. Ousterhout, you should find the GNU extension
language plans very exciting, and i hope we can find ways to cooperate
more directly.

For example, a program that supports the GNU extension language will
be programmable using Rush or a Rush-like language, provided only that
a suitable translator has been written. (Rush is, in fact, already
written as a translator to Scheme).

Rush is semanticly and syntactictly very close to Tcl; the differences
are not likely to be noticed by many programmers. The performance of
Rush is generally superior to Tcl. It is quite plausible to view Rush
(or a Rush-like language) as a direction in which Tcl can evolve
smoothly. Upward compatability is a delicate matter, but I am sure
that, working together, we could handle it with the greatest
gentleness towards existing users.

I am at your disposal to discuss the possibility of cooperation off
line. (Our offices are within minutes of each other).

-t