Newsgroups: gnu.misc.discuss,comp.lang.tcl,comp.lang.scheme
Path: cantaloupe.srv.cs.cmu.edu!nntp.club.cc.cmu.edu!godot.cc.duq.edu!news.duke.edu!MathWorks.Com!europa.eng.gtefsd.com!howland.reston.ans.net!EU.net!uknet!liv!news
From: bruce@liverpool.ac.uk
Subject: Re: Why you should not use Tcl
In-Reply-To: gsipe@pyratl.ga.pyramid.com's message of 27 Sep 1994 16:46:29 -0400
Message-ID: <BRUCE.94Sep28120500@iasc3.scm.liv.ac.uk>
Sender: news@liverpool.ac.uk (News System)
Nntp-Posting-Host: iasc3.scm.liv.ac.uk
Organization: IASC, University of Liverpool
References: <9409232314.AA29957@mole.gnu.ai.mit.edu> <364bq2$mjd@topaz.sensor.com>
	<367br7$275@pyratl.ga.pyramid.com>
	<PSMITH.94Sep26184324@lemming.wellfleet.com>
	<36a0b5$aui@pyratl.ga.pyramid.com>
Date: Wed, 28 Sep 1994 11:05:00 GMT
Lines: 42
Xref: glinda.oz.cs.cmu.edu gnu.misc.discuss:18459 comp.lang.tcl:19502 comp.lang.scheme:10049

>>>>> "George" == George M Sipe <gsipe@pyratl.ga.pyramid.com> writes:

> In article <PSMITH.94Sep26184324@lemming.wellfleet.com>, Paul Smith
> <psmith@wellfleet.com> wrote:

>> Furthermore, the GNU project has embraced *many* packages which
>> aren't licensed under the GPL or the GLPL; it is *not* a
>> requirement that everything GNU distributes be placed under the
>> GPL.

> What are some of the *many* packages that the FSF distributes which
> are not covered by the GPL or LGPL?  Such a long list of significant
> software would go a long way to convince me to reconsider this
> position.

X and TeX spring to mind.  They may not count as many, but that's
surely *lots* of code!

I don't think rms would have any problems with the legal bits of TCL
so long as he could be sure that they'd stay the same (which does seem
assured); his problems with it appear to be technical, and I agree
with him.

TCL doesn't seem to have been designed to allow compilation even down
to bytecode; it lacks some structuring (although there are extensions
to allow this).  I think a language that you want to apply to a large
number of applications needs to be a decent language, because for at
least some of them, you're going to want to write largish programs,
even if for the most part you're just writing little scripts.

It may well be that TCL as it is can be bytecoded (at least for the
most part, assuming well-behaved source code) by using various hacks,
and this would be valuable for the existing code.  I can, however,
understand why rms would want to persuade people to move away from
such a language as much as possible.

(I happen to think that relatively minor changes to TCL's syntax could
make it closer to scheme in terms of bytecoding, simply by changing
it's obsession with strings to one where the primary statement
"quoting" really meant list construction.  It would break some
existing code, sure, but I think the result would be as readable as
TCL and it could run faster.)
