news.txt from the Maple Bugs Encyclopaedia
Vladimir Bondarenko
feb 1, 2006:
...when, now and then, I recollect dialogues with Keith
Geddes and Gaston Gonnet, I cannot help thinking over
and over again about the symbolic computations evolution...
nov 1980...
an important date in the CAS history: the Maple project
was conceived...
almost 25 years ago, jan 1982, Maple 1.0 was released;
the celebration is coming...
what's funny, it looks like almost e a c h Maple
function has bugs!...
ok, pre-announcing unparalleled Maple bugs results never
available before us...
keep calculating Maple 10.02 bug list...
new unusual bugs spring up...
the beta 0.2 of the 1st world's man+machine based Maple
review is coming...
meanwhile, why don't consider enjoying the beta 0.1 Review
at the Maple Crisis Review menu item to compare it
with the 0.2 ?...
read there about Maple 9.5.2 clay-solid numerics (aka
ROCK-SOLID NUMERICS according to Maplesoft) and much more...
new Encyclopaedia update is coming...
welcome to www.cybertester.com updated soon
............................................................
http://maple.bug-list.org/maple-crisis.php
............................................................
da...@axiom-developer.org
About 4700 of the 5500 items listed are integration routines
according to the encyclopedia counts.
As near as I can tell almost every integral has a
possible pole in the path of integration. And those
that don't seem to fit various corner cases of the
Risch algorithm which, as far as I'm aware, is not
fully implemented anywhere.
Can you suggest a general algorithm for handling these
badly behaved integrals? You've collected sufficient examples
to create various classes of integrals. It would be more useful
to do an analysis of the failing integrals and find algorithms
for handling each class. If I had a student that had collected
this kind of data I'd suggest that it would be more worthwhile
to analyze rather than continue collecting. Even a taxomony
of the integrals would be a worthwhile result.
The field moves forward because new algorithms are invented
and published.
Tim Daly
Axiom Lead Developer