For _Z1=0, _B1=0, you get x=0. So something is a little weird -- I guess I need to go re-read the help page on solve to figure out what it really means with those _B's!

However, it does not seem easy to patch (by Maple users), since the implementation appears to be partly in Java (ie com.maplesoft.mathlib.worksheet.Worksheet), partly in the kernel (sscanf with %m format), so this would be a thorny one for people from the community to track down properly. Send an email to support@maplesoft.com, and report back what you hear!

I missed commenting on this for a while, but since this comment will put me at 700, it seemed fitting to use it to look at what has happened. Will continues to nibble towards the top spot, but his posting rate has slowed a lot. I just may get there first... Mariner and roman_pearce are still tied, and are both heading towards the top ranks. Not to be outdone, Axel Vogt and gkokovidis are also steadily climbing; they may actually have a bit higher rate than Mariner and roman_pearce, so may well catch up to them. acer got his gold leaf this week (congrats!). dcasimir, you've got some work to do. Scott03 and Thomas Madden continue to post steadily. In this time-frame, it is in the bronze leafs that we've seen the most movement. Dave L, Tim Van Dusen, and Alex Smith have all been quite active. And of course, we now welcome william budd onto the first page!

As far as I know, pdsolve is completely classical (ie no distributions). And yes, you should write to Frederic (Chyzak) about this problem in Mgfun.

As far as I know, pdsolve is completely classical (ie no distributions). And yes, you should write to Frederic (Chyzak) about this problem in Mgfun.

You can even get rather fancy: L1 := zip( `[]`, x1, y1 ); will do the same as what you show above. I believe that all contructors in Maple now have names too, so that temporary functions do not need to be created just for data-structure creation (for simple cases anyways).

You can even get rather fancy: L1 := zip( `[]`, x1, y1 ); will do the same as what you show above. I believe that all contructors in Maple now have names too, so that temporary functions do not need to be created just for data-structure creation (for simple cases anyways).

First, the underlying theory uses ideals of polynomial operators (and has nothing to do with Lie theory, BTW). And the equation diffeqxalpha does annihilate f, but is not contained in the ideal spanned by just diffeqx and diffeqalpha. Second, you are too trusting of pdsolve, ie that it finds the 'complete' solution. In fact, it is missing some ``non-standard'' component of the solution. For example h(x, alpha) + (_C3*alpha + _C4/alpha)*Dirac(x) is a solution of {diffeqx, diffeqalpha}, but not of diffeqxalpha. [Those solutions can be found via Fourier transforms].

First, the underlying theory uses ideals of polynomial operators (and has nothing to do with Lie theory, BTW). And the equation diffeqxalpha does annihilate f, but is not contained in the ideal spanned by just diffeqx and diffeqalpha. Second, you are too trusting of pdsolve, ie that it finds the 'complete' solution. In fact, it is missing some ``non-standard'' component of the solution. For example h(x, alpha) + (_C3*alpha + _C4/alpha)*Dirac(x) is a solution of {diffeqx, diffeqalpha}, but not of diffeqxalpha. [Those solutions can be found via Fourier transforms].

To post output from Maple (like above), use <pre> brackets (which uses fixed-width fonts), otherwise the result is pretty much unreadable (as I am sure you can see).

To post output from Maple (like above), use <pre> brackets (which uses fixed-width fonts), otherwise the result is pretty much unreadable (as I am sure you can see).

Last I knew the code, solve did not really consult the assume facility (ie is, coulditbe, assuming, etc). A shame, yes.

Last I knew the code, solve did not really consult the assume facility (ie is, coulditbe, assuming, etc). A shame, yes.

Maple does not implement a full algorithm, but has some fairly effective heuristics (in solve) for inequalities. As with all heurisitics of this kind, they are generally fast, often useful, but completely unpredictable (ie when they will work/fail).

Maple does not implement a full algorithm, but has some fairly effective heuristics (in solve) for inequalities. As with all heurisitics of this kind, they are generally fast, often useful, but completely unpredictable (ie when they will work/fail).