Since I didn't make it to CICM this year, but still (with some help) follow
what goes on at the workshop, I've decided to post some reflection on it to
the OM mailing list instead, in the hope that it reaches the participants.
Readers beware; there may be ranting.
First and foremost, I don't think Michael should chair future discussions on
extending OpenMath; I've heard him do so at least twice by now, and the
results haven't been good.
It's not that I disagree with him -- on the contrary, I quite agree with his
position on the matter -- but it is clear that he is too keen on being the
advocate to do much chairing, and as a result the discussion degenerates.
The view I had on Monday didn't show much, but I could see a number of times
where Jan Willem raised his hand without anyone up front seeming to notice,
as a small group there seemed happy to keep the discussion between themselves.
(It also didn't help that what these people /were/ saying sounded distinctly
like 80% repeat of a discussion originally held last decade, where it
possibly rather was on the topic of strict versus pragmatic Content MathML.
If such reruns is what the oldtimers want to spend their time on then fine,
but please limit it to your own time, rather than wasting the precious 6
hours of OM workshop that we have once a year or two!)
That we should produce errata to the OM2 standard is a no-brainer! To not
issue errata, when there are obvious typos in the specification, is to be dead.
We should _also_ work on extensions, because there are kinds of mathematics
that the current standard doesn't do a good job of supporting. (Anyone who
doubts that claim is hereby invited to detail what in
http://ceur-ws.org/Vol-1010/paper-21.pdf they consider to be wrong. Note in
particular that appeals to lambdas or helper sets is tantamount to claiming
that MSO logic is pointless, something that the formal languages literature
very much disagrees with.) And the mathematics doesn't give a damn that
CMML3 has been blessed by the ISO -- blessings have no weight in any logical
system I've ever heard about.
What forms extensions to OM should take is another matter, but one that will
be easier to address once the possible extensions are made more concrete.
Having as engineering goal while developing an extension that it should be
implementable via a Language Extension Dictionary is fine by me, but it
should not be considered necessary, even if I suspect that such an
implementation will usually be possible.
Lars Hellström
PS: Yes (in case there was ever any doubt), I volunteer to work on the above
issues.
PPS: There is more to say, but this makes a nice part 1. Part 2 will be on
other topics.
_______________________________________________
Om mailing list
[email protected]
http://openmath.org/mailman/listinfo/om
