Incongruent hash

Preston Briggs preston at tera.tera.com
Tue Feb 10 19:56:23 UTC 1998


>Um ... but aren't points (at least in 2-space) equivalent to complex
>numbers?  So shouldn't they sort the same way?  (So, of course, now we
>have to decide how complex numbers sort :-)

Right.  But most people don't compare complex numbers.
E.g., Fortran will only let you compare them for equality/inequality.
And since comparing floating-point numbers for equality is always
a little suspect, ...

I think <, >, et al. were a hack for points.
Makes some of the code for checking containment of points
in rectangles simpler.  Not much use otherwise.

Many things besides apples and oranges can't be meaningfully compared.

Preston





More information about the Squeak-dev mailing list