This is the mail archive of the guile@sources.redhat.com mailing list for the Guile project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Translation for extension is a bad idea



> From: Han-Wen Nienhuys <hanwen@cs.uu.nl>
> 
> * there is consensus that Scheme is scary due to prefix notation. This
>   suggest that there should be some infix dialect of Scheme.
> 
>   (would Dylan be an appropriate language for this, IIRC it was a
>   derivative of LISP?)
> 

Dylan is in fact a derivative of Lisp with infix syntax.  There was also a
prefix form at one time, but it was abandoned due to marketing pressures.
Dylan would be a pretty good target for guile translators, since it also
supports a CLOS-like object system.

To sum up, based on comments that have been made in this thread, I'd like
to see translators serving these purposes:

1) A "standard" infix-ish syntax for guile a la CTAX;

2) Syntaxes that are more-or-less the same as python, tcl, dylan etc. but
   with more schemey semantics.  In other words, not
   python-translated-to-guile but guile-with-python-surface-syntax.  This
   is about 1% of the amount of work that would be required to match the
   semantics of the target language fully.  The drawback is that users will
   expect the semantics to be identical when they aren't -- they'd be
   superior :-)  Hopefully this wouldn't bite them too often.

There was also the mention of a different issue, that infix math
expressions are cumbersome in pure scheme.  An "expr" macro was proposed as
a solution for this.  I totally support this, but it's orthogonal to the
desire for translators (in fact, it would probably be useful for anyone
using guile for numerical programming).

Mike


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]