I concur with Mark.  This is a very challenging area and no change would be warranted without keeping the current accuracy guarantees, definitive speed improvement, extensive tests, and consideration of the impact on users of having float reprs change yet again.

This was a nice tracker item for improving our awareness of alternate algorithms, but it isn't a credible feature request without the qualities listed above.  

I'll mark this as closed/rejected.  Feel free to reopen if we get some solid evidence that python would be significantly improved enough to warrant the disruption for users and the investment of developer time.