Re: Macros and little languages

[Matthias Felleisen <address@hidden>]

Apologies for my flippant tone. But, Lisp is a survivor. As PG has pointed
out in 'Revenge of the Nerds', which started this discussion, the McCarthy
clan came up with the 10 basic things back in 1958, and what we have today
is more or less what he thought of. 

Scheme is a survivor in the same way. We are now in the fourth
generation of Scheme implementations (Guy's Scheme, followed by
Scheme84/T/MIT Scheme, followed by Chez/Gambit/MacScheme, followed by
PLT Scheme/BigLoo/Twobit+Lareceny, etc). Let's see whether there will 
be 4 generations of Python implementations. 

The good ideas may not spread, but they survive. It doesn't make me
happy that the good Lisp/Scheme ideas don't spread, but when I see 
someone like PG or Orbitz succeed with the stuff, hey, I know that 
they are alive. And I personally work on spreading the ideas, but only 
for the sake of producing better programmers (programmers who hate the
C-style stuff and know better, but who are victims of pointed-hair bosses). 

-- Matthias


At Mon, 27 May 2002 11:04:15 -0700, Zooko wrote:
> 
>  Matthias Felleisen <matthias@ccs.neu.edu>:
> >
> > At Mon, 27 May 2002 09:26:39 -0700, Paul Prescod wrote:
> > > Re: infix versus prefix. I have more faith that you could convince the
> > > world to use esperanto than prefix notation. But by all means, have
> > > another go at it. Maybe it just needs one more try!
> > 
> > Well, perhaps we really just want to tease you. After all, if you don't
> > use prefix and parenthesized languages, we have an advantage. -- Matthias
> 
> I'm not sure of the tone of this conversation, but this quip just made me 
> realize something.
> 
> If we exist in a big pool of people who are all competing with one another, 
> then what you've said (Matthias) is true.  If we exist in a big pool of 
> people 
> who are all cooperating with one another, then the opposite is true -- by 
> using a syntax amenable to one hundred times as many people, I gain one 
> hundred times as many tool authors, clients, collaborators, etc.
> 

> Of course in reality we exist in a more complex situation were both kinds of 
> interaction are important.
> 
> But this has made me realize that there is an opposite side to Paul Graham's 
> "How I Used Lisp To Outdo My Non-Lisp Competitors" story -- it also implies that 
> he wasn't taking advantage of other people's code as much as his erstwhile 
> competitors could have.
> 
> FWIW the current Python culture (as well as, I'm sure, the current Lisp 
> culture) is very much on the side of cooperation viz competition with regard 
> to source code and object code.  The python-list is famously friendly, and 
> Python has mature traditions of interoperating with code from the C/C++ and 
> Java worlds.
> 
> (BTW, I'm a bit of an economic cynic and I expect this culture to prove 
> non-scalable.  ;-))
> 
> Regards,
> 
> Zooko
> 
> -------
> Secure Distributed Systems Consulting -- http://zooko.com/
> -------