Yossi Kreinin on Tcl

[Colin Macleod]

See http://www.yosefk.com/blog/i-cant-believe-im-praising-tcl.html -
an interesting perspective from the author of the heretical C++
Frequently Questioned Answers (http://yosefk.com/c++fqa/index.html).
Colin.

[Bryan Oakley]

I like this quote half way through the article:

"Tcl is optimized for usage"

[Colin Macleod]

It's also kicked off a fairly animated discussion at
http://reddit.com/r/programming/info/6gmgr/comments/

[Bryan Oakley]

Colin Macleod wrote:
> It's also kicked off a fairly animated discussion at
> http://reddit.com/r/programming/info/6gmgr/comments/

I really liked this comment:

"...[tcl] really is like "extreme kung fu". At first you see this crazy
old man and laugh at his silly ways. Then you see how much of a badass
he really is."
--rook2pawn, http://reddit.com/r/programming/info/6gmgr/comments/

I nominate that for quote of the week, maybe of the month.

[EL]

Colin Macleod schrieb:

> It's also kicked off a fairly animated discussion at
> http://reddit.com/r/programming/info/6gmgr/comments/

This one comment from schlenk is very interresting... about the L
language that is a C like language, compiled to Tcl bytecode:

http://wiki.tcl.tk/17082

Amazing... does that mean it would be possible to implement other
languages on top of the Tcl bytecode? Maybe a Scheme or Lisp like
language, that can take advantage of all the Tcl/Tk goodies?

Not that I'd plan to do that - but it would be /really/ cool :-)!

--
Eckhard

[Bryan Oakley]

Of course it's possible. You have to realize, though, that several very,
very good programmers have been at this more than two years now and
don't yet have a production ready language (AFAIK). I think it's a
formidable challenge to do the job right.

[Neil Madden]

I'm not sure how suitable the Tcl bytecode engine is for Scheme
(tail-call optimisation, closures, continuations, etc) -- although
Miguel is certainly pushing things in interesting directions!

-- Neil

[Donal K. Fellows]

Neil Madden wrote:
> I'm not sure how suitable the Tcl bytecode engine is for Scheme
> (tail-call optimisation, closures, continuations, etc) -- although
> Miguel is certainly pushing things in interesting directions!

Tail-call optimization should be possible for the #1 case (recursive
calls to the current function) without bytecode engine changes as it is
a compiler-level thing, but the others you list do indeed require
significant work. Some of that work may actually be being done. :-)

Donal.