Re: GNU Extension Language Plans

Peter da Silva (peter@nmti.com)
Sat, 22 Oct 1994 17:36:03 GMT

In article <38a3mk$lr8@csnews.cs.colorado.edu>,
Tom Christiansen <tchrist@mox.perl.com> wrote:
> Peter, be aware that the common man has neither the time for nor any
> appreciation of your spartan minimalism.

Spartan minimalism? Tcl is hardly spartan... it's just designed for a
specific job and does it very well. Perl is designed for a different
job and does THAT very well. I don't think it could do Tcl's even as
well as Tcl does Perl's.

[preaching to the choir omitted]

Hey, I'm responsible for some of the features in Tcl that *are* there, like
the way strings work. Karl and I worked out the semantics of Tcl arrays on
his whiteboard when he worked here. We did Tcl Extended, because the original
language was too minimal, and a lot of that has been picked up. For that
matter we picked some ideas up from Perl... some of them didn't make the
cut and still aren't in the core language (like the filescan stuff).

But there's always been this basic assumption: that you don't add a feature
just because it sounds good. You add it because you need it. If there's two
ways of doing something you use the one that avoids complicating the language.
The classic example in Perl is the postfix if statement. It doesn't add
any capability to the language, and it confuses new users. In an extension
language that's a bad thing... because most of the time most users are
new users, because they're not using the language to do a job, they're
using it to configure the tool that does the job.

> The problem is, you see, is that quite simply, you're designing the wrong
> languages for the wrong crowd.

Who, me? I'm not designing a language at all. Or redesigning one. I'm trying
to keep a bunch of people from inventing yet another camel when the specs
don't even call for a horse.

[a bunch of stuff that doesn't seem to have anything to do with me at all,
skipped]

> When it comes to lisp or tcl, while the extensive run-time nature of
> those languages make machine language generation (at least of certain
> constructs) difficult, compiling them into native perl (probably
> with a run-time emulator library) should in theory present no insurmountable
> hurdles.

Certainly with a runtime emulator library... especially when you're running
around loading stuff on an ongoing basis at runtime and using code fragments
as your communication channel between components. And since you have to keep
doing that, what's the point to putting Perl in the loop at all? It's not
technically infeasible, it's just not very useful. And that's why I think
it's unlikely.

-- 
Peter da Silva                                            `-_-'
Network Management Technology Incorporated                 'U`
1601 Industrial Blvd.     Sugar Land, TX  77478  USA
+1 713 274 5180                       "Hast Du heute schon Deinen Wolf umarmt?"