Deep comparison of dicts - cmp() versus ==?

[Roy Smith]

In article <550bbfc1$0$13010$c3e8da3$5496439d at news.astraweb.com>,
 Steven D'Aprano <steve+comp.lang.python at pearwood.info> wrote:

> I cannot remember the details, and I don't have my copy of the Apple 
> Standard Numerics manual here to look it up

Amongst the details you don't remember is the correct name :-)  It was 
Standard Apple Numerics Environment (SANE).  I think I still have my 
copy of the manual somewhere.

> If you find that hard to believe, it's because you're spoiled by the 
> astonishing success of IEEE-754 floating point arithmetic.

There's two things we're spoiled by.  IEEE-754 is, of course, one of 
them.  But, more than that, are good math libraries built on top of it.

One of the truly unappreciated things to come out of the early BSD Unix 
releases was a C math library written by people who understood (and 
worried about) things like accurate FP calculations, rounding, corner 
cases, and all that gunk.  Sure, I could sit down with a C compiler and 
the Wikipedia Gamma function article and code up gamma().  It would work 
well enough to get a decent grade as a homework assignment for 
Introduction To Programming 101.  But, fundamentally, it would be a 
piece of crap because I know just enough about numerical programming to 
know that I should leave it to the experts.