I was just about to append to my former post a little PS, but instead I will post it here: It is quite possible that in Maple there exist some simpler way of performing the calculations, but I have not been able at present to find any such. As a theoretical physicist, having studied general relativity, I have no problems in dealing with tensor- and differential calculus, but I do, unfortunately, have problems implementing these using Maple, primarily because I find it difficult to figure out the proper use and syntax of the appropiate packages.

I was just about to append to my former post a little PS, but instead I will post it here: It is quite possible that in Maple there exist some simpler way of performing the calculations, but I have not been able at present to find any such. As a theoretical physicist, having studied general relativity, I have no problems in dealing with tensor- and differential calculus, but I do, unfortunately, have problems implementing these using Maple, primarily because I find it difficult to figure out the proper use and syntax of the appropiate packages.

Thanks for letting me know the reason. What problems are you experiencing when trying to post? Problems with the pre-tag, or the maple-tag? Sometimes I do have difficulties with the latter.

Thanks for letting me know the reason. What problems are you experiencing when trying to post? Problems with the pre-tag, or the maple-tag? Sometimes I do have difficulties with the latter.

Nice to hear that you too find the output of ToInert spectacularly ugly. Unlike you, however, I do not find it easy to program with, but probably that is due to insufficient present knowledge on my behalf, which, of course, just brings me back to the desire for a proper introductionary text.

If you want your expression written in terms of tan, you can use the following:

> convert(sin(x)^3/cos(x)^3,tan);
                                  3
                            tan(x) 

It seems to me that the function MakeOpList in the link you provide does pretty much the same as my function toTree in the thread Expression trees. At least, it gives the exact same expression tree for the example provided. PS: I would like to use this opportunity to apologize again for my bad behaviour in the above-mentioned thread, which still makes me feel ashamed.

An introduction from you on issues like ToInert and FromInert would surely be widely appreciated. Searching through (using Find in Acrobat Reader) the Maple 11 Advanced Programming Guide, I am unable to find anything on ToInert or FromInert, or for that matter dismantle, disassemble, pointto, or addressof. Quite surprising, I think.

Depressingly, you are probably quite right. I guess, I demand to much of this world. To me, the most important part of your post is that on Mathematica: to be honest, I have thought about changing CAS, but if what you write is really true, then, of course, it would be quite stupid to do so.

Thanks for that historical illumination. What really surprises me is that it was only later on that people starting expecting things like matrices of matrices to work. To me that is foundational, compare But why should it be an error?. But I guess it is to late for that now!?

I wonder why I did not see that function while searching the help text. Regarding ToInert and FromInert: I have to admit that I am unable to penetrate their syntax and use. Is there any good introductionary text on them somewhere, explaining all those parameters?

Nice initiative of yours, but, as you write yourself, it probably is a long shot. Lately, after having upgraded to Maple 11 to no real avail, as previous frustrated threads of mine clearly show, I have got the feeling that MapleSoft is only pretending to be obliging. I cannot help having a feeling that MapleSoft (as a company, not the developers, I am sure) seem to care more about how their products look (because that seemingly sells), than how they work:

• What does it help to have all these nifty XML features, if it really just makes something as completely trivial as tabbing a nuisance, if not to say an impossibility (compare the thread I declare defeat: rolling back to Maple 9.5)?
• What does it help to market a product saying that it is the only CAS right now having a physics package if that package seems impenetrable even to a physicist (compare the thread Impenetrable Physics Environment?)?

Until now I have not been able to use Maple 11 for anything which I could not do using Maple 9.5. I keep asking myself what my money was spent on.

Hi Alex, Up to a sign (of the second sum) I reproduce your result if the squaring (inside the integral) of the difference of the sums is left out. I guess you have simply overlooked it, I myself did that first too. Including the squaring leads to a much more complicated problem because of the double sums. As of yet I have no nice, if any at all, solution to that problem.

Hi Alex, Up to a sign (of the second sum) I reproduce your result if the squaring (inside the integral) of the difference of the sums is left out. I guess you have simply overlooked it, I myself did that first too. Including the squaring leads to a much more complicated problem because of the double sums. As of yet I have no nice, if any at all, solution to that problem.

I have been thinking along similar lines, but I was unable to make it work. I suppose one of my mistakes was to try something like AntiSym(4), that is, using parentheses instead of square brackets, as you do. But even if I had thought of square brackets myself, I think that a line like

if type(procname, 'indexed') and type(op(1, procname), posint) then

would have been outside the scope of my present abilities. So, thanks! It seems, though, that you have left out the line

if row > col then row,col,sign := col,row,-1 end if;

which makes the whole endavour meaningful, for without it no antisymmetric indexing is done. Also, to me it seems that the `if`-clause in generators is unnecessary. Finally, I could not help smile (that's nice, of course) reading your comment on the missing flattening of the matrix, because that is exactly what I do not want, compare the thread Width and depth. So, in this case, Maple behaves as I would like it to.