(j3.2006) Liaison to IFIP WG 2.5

[Toon Moene]

Van Snyder wrote:

> IFIP WG 2.5 are very interested in interval arithmetic.

In 1992 I followed one of the initial courses in Numerical Meteorology, 
given by the European Centre of Medium Range Weather Forecasts.

Because I had received Arnold Neumaier's " Interval methods for systems 
of equations" as a gift when leaving SARA (http://www.sara.nl), I asked 
(the then director of ECMWF) Dr. Dave Burridge about his opinion on 
interval arithmetic.

His answer: We have better (meteorological based) limits on the extremes 
of our forecasts.

This has been formalized in the mean time by ensemble forecasts 
(http://ecmwf.int/research/predictability/).

It is my professional opinion that - unless we want to restrict Fortran 
to purely mathematical computation - that (almost) every physics (or 
other science) subject that uses mathematics as its language of choice, 
has better ways of abstracting its uncertainty than interval arithmetic.

Kind regards,

-- 
Toon Moene - e-mail: toon at moene.indiv.nluug.nl - phone: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
At home: http://moene.indiv.nluug.nl/~toon/
GNU Fortran's path to Fortran 2003: http://gcc.gnu.org/wiki/Fortran2003