> I've heard of using human choices for comparisons, when fewer decisions could provide a notable speedup.

Memoization is the usual solution when expensive computations are being repeated.  That technique preserves the work that was done and it avoids adding unnecessary complexity to the consumer code.

> Do you think it's worth posting to python-ideas to see 
> what people's use cases are?

No, that would be a waste of time and it wouldn't change the validity of Tim's insights.