woods at weird.com (Greg A. Woods) is widely believed to have written:

> In Smalltalk syntax one of the problems with symbols like you suggest,
> or any new keywords that look like words, is that they're much more
> easily confused with message selectors or variable names, etc.
But, but, they _are_ message selectors!

> Perhaps a better choice than '@' would be an entirely new glyph, though
> at least there's precedence for using '@' (i.e. in ObjectOriented APL).
> This shouldn't be a difficult problem for Squeak, though even Squeak
> tries to make it easy to transliterate Smalltalk code to ASCII, and
> since Squeak doesn't use '@', it seems like as good a choice for these
> new "operators" as any other special character would be.
Squeak uses '@' - all over the place. As in a at b being a Point with x=a
and y=b.
> 
> I suspect anyone who works long enough with F-script will learn the new
> operators much as an experienced C programmer learns the unique idioms
> of C, or indeed even as a Smalltalk programmer learns its idioms.
Mixing operators with message sending is a recipe for confusion. :-(
> 
> > Ah, yes. u@(v"(1+lv,_)). Of course ;-)
> 
> I wonder what that'd do in TECO!  ;-)
Uppercase everything on disc, write it to disk sector 'v', replacing any
instances of "( followed by more than a single 1 with lv, doing it all
in memory treated as a single string. Maybe.

tim
-- 
Tim Rowledge, tim at sumeru.stanford.edu, http://sumeru.stanford.edu/tim
Useful random insult:- Always sharpening his sleeping skills.