Why is tcl broken?

[Fernando Mato Mira]

At 08:54 AM 6/10/99 -0500, Gordon McMillan wrote:
>Fernando Mato Mira writes:
>> 
>>   I'm trying to collect a list of all the _current_ issues making
>>   tcl
>> a bad language choice.
>
>Why? Convincing your boss to use Python over TCL for your next 
>project is one thing, but in general this is a zero-sum (or worse) 
>game.
>
>The TCLers I know value its simplicity and accept its limitations. 
>Telling them they're wrong will only piss them off.

A `TCLer'  here is going to give a talk presenting the virtues of
tcl to the rest of the organization. As a hard-core lisper, it is obvious
to me that the core philosophical basis of the `New Jersey school' (i.e. the
whole idea of scripting languages) is broken, and obviously the syntax
is tasteless to me. But I'm looking for irrefutable signs of bad language
design that result in low maintenability. For example, dynamic scoping,
ambiguities, etc. Things like the `string issue' are also interesting if they
can be exposed in a way that the answer is not "So what?".

I'm not putting in question the virtues of Tk, as you can access that
through STk, for example. But if there are bad things there, I'd like
to know, too. (eg: is there a lot of tcl code making a Tk interface run?
If so, does speed ever become an issue? I assume that Tk works well
enough that you'd never have to look at the tcl code, and live happily
inside your reflective Scheme world)

Thanks,

Fernando D. Mato Mira                    
Real-Time SW Eng & Networking            
Advanced Systems Engineering Division
CSEM                             
Jaquet-Droz 1                   email: matomira AT acm DOT org
CH-2007 Neuchatel                 tel:       +41 (32) 720-5157
Switzerland                       FAX:       +41 (32) 720-5720

www.csem.ch www.vrai.com
<http://ligwww.epfl.ch/matomira.html>ligwww.epfl.ch/matomira.html