Acknowledgements
This work was partially supported by the National Science
Foundation, Grant  DMI 97-14142.
This presentation has been influenced by numerous
discussions on the mailing list
             reliable_computing@interval.louisiana.edu