On Nov. 3, 2010, Richard Fateman wrote:
>>
>> Here is a record of the new definitions
>> the calc. of the 'true value' and the
>> symbolic calculation's reduction
>> to floats as a check of correctness.
>>
>Have you read the articles on this topic that appeared in the SIGSAM
>Bulletin,
>previously referenced?
no. I clicked on all the links and they wanted dollars to
see text.
> There is also an MS thesis at UC Berkeley on
> branch cuts and computer algebra systems.
how easy is it to see this?
> Are you able to convince yourself that you have
> (a) a complete specification of the problem you are trying to solve
no.
> and
> (b) a solution?
no.
> It is hard to tell from a single example.
Agreed. If all one needs is one data
point, one can prove a lot of nonsense.
As a physicist, not a computer scientist or
programming science expert, I am hoping my "data point"
will stimulate those on the mailing list more qualified
than I to pursue these issues.
Ted
RJF