[om] critique of the notion of phrasebook

[Richard Fateman]

Andrew's construction of a mathematician's
workbench with copy/paste is useful, but there is
no need to select the representation.  An object
being copied could represented in all ways known, and the
receiver could select the one it understands, if any.

This means that to communicate between Maple and
Mathematica, one would have to recognize what
parallel constructions exist, and which do not.
For example, pattern matching is quite different.
Mathematica has the arguments to ArcTan2 reversed,
etc.

The idea that one can translate from N to one and then
from one to N, to make N encodings/decodings instead of
(approximately) N^2, is operational when one
is conveying extremely simple entirely conventional
data.  It is hardly more  (and maybe less) than the
common tendency to allow   a*b+c as a string as input
to maple mathematica macsyma reduce TeX..  But TeX
allows almost anything.

So from this perspective OpenMath contributes rather
little to the discussion.

Cheers.

RJF
--
om at openmath.org  -  general discussion on OpenMath
Post public announcements to om-announce at openmath.org
Automatic list maintenance software at majordomo at openmath.org
Mail om-owner at openmath.org for assistance with any problems