Comment on PEP-0238

[Greg Ewing]

Edward Jason Riedy wrote:
> 
> For many geometric and scientific codes using floating-point,
> this is exactly what you want.  You bump the precision up
> once, or you multiply it after trying to get a good-enough
> answer.

Yes, I can see that this is a different sort of 
precision problem. Even so, I wouldn't like the idea
that some piece of code I have no control over could
lower the precision and mess up the results from
*my* code. So I still don't think it should be a
piece of global state, not even per-thread. Some
sort of scope control is needed.

-- 
Greg Ewing, Computer Science Dept, University of Canterbury,	  
Christchurch, New Zealand
To get my email address, please visit my web page:	  
http://www.cosc.canterbury.ac.nz/~greg