Re: [stack] syntax preference question

For the sake of completeness, Slava offered this definition for the  
quadratic formula:

   : monic ( c b a -- c' b' ) tuck [ / ] 2bi@ ;
   : discriminant ( c b -- b d ) tuck sq 4 / swap - sqrt ;
   : critical ( b d -- -b/2 d ) [ -2 / ] dip ;
   : +- ( x y -- x+y x-y ) [ + ] [ - ] 2bi ;
   : quadratic ( c b a -- alpha beta ) monic discriminant critical +- ;

This is quite a bit better than the version given here:
http://www.latrobe.edu.au/philosophy/phimvt/joy/jp-quadratic.html

Using the rather terse syntax in the last email, along with 't' for  
'tuck' and 'x' for 'swap', Slava's version could hence be written as  
such:

   t 2@[/] t sq 4 / x - sqrt `[-2 /] 2[+,-]

If the cleave combinators were type-aware (which is almost certainly a  
bad idea), we could do away with the arity notations. Further hell  
could result from making ',' infix and using brackets to disambiguate  
only when necessary:

   t @/ t sq 4 / x - sqrt `[-2 /] +,-

Alright, enough perversity for now.

- John