@ecterrab 

Thank you! I think this is really an improvement - could be refined at some later point in time - but already provides much nicer and clearer output of differentials. I hope many users will appreciate the change.

@ecterrab 

I do like your suggestion of d(x) much more than the partial-d for this purpose. I personally think that setting it straight instead of italic is a very good idea, not only for mimicking textbooks but also as a clear indicator that this is a special operator that cannot simply be typed as d(x).

The bold version d(x) I think is kind of reminiscent of a vector valued function, especially if one uses the bold notation for 3-D non-projected vectors in the settings of Physics - but one could get used to that. I personally would prefer non-bold and maybe a grey (or any other) coloring of the d instead that can already be found in unevaluated integrals as an additional indicator that this is an operator. But maybe the grey color there is meant to signify a delayed evaluation, then this would be inconsistent and not a good choice but in this case any other color (and if it was black) could do the job. I personally like the specific coloring of operators, e.g. of the quantum operators in Physics a lot.

But again, I am only a user and can only suggest what would kind of look nice - you are the experts who know what formatting would be consistent with the existing basic Maple output and also the output of the special packages. And of course, since this is also a matter of taste, you should have the final say regarding such design changes. I think that a straight d is in any case a major improvement, no matter the coloring - and even if it has to be bold ;-)

@ecterrab

I agree, having one powerful, very general differential operator is a good thing and different displays might confuse users.

I personally however like it very much that the diff command now supports the display of a total derivative; it almost drove me crazy before this was implemented although I have no idea whether the majority of Maple users actually cares about such minor issues that are only related to the display of things.

If the display d(x) might confuse users to also type it this way, maybe the beautifully displayed dx that can be found in inert or evaluation-delayed integrals would be a good alternative? The inert %Fundiff for instance produces perfect, textbook-style output which, at least for me, makes Maple so much preferable over Mathematica. Not to mention the extremely beautiful implementation of the bra-ket formalism of QM that is unparalleled in this software!

Of course, functionality, internal consistency and logics should trump mere layout issues and being only a user I am not in the position to evaluate the former. And I also cannot estimate how many of the users would actually appreciate an even closer resemblance with textbook equations as long as they have the great functionality of the Physics package at hand.

@ecterrab 

Hello

Thank you for your comment on the input error and also concerning the series expansion. You are right of course, the expansion should be around r_ = 0, not around r__0_, which should give the familiar result

Thank you also for considering implementing such a kind of expansion in Maple.

Best wishes,

Peter

@ecterrab

I am using Maple for several years now starting with checking homework assignments for university and the like. I very much liked Wang's 'Physics with Maple' published by Wiley because without such instructions it is hard to use Maple appropriately for computational tasks in Theoretical Physics. The online help system is not so much of use here.

I immediately became a fan of the Physics package when I noticed how easy it's there to derive w.r.t. a function or even the derivative of a function and how cumbersome this was before without this package e.g. to set up Lagrange's equations of motion. Above all, the Vectors subpackage, the handling of arbitrary metrics, co- and contravariant tensors, Christoffel symbols and the like make this package just the perfect working environment for doing calculations in physics.

I hardly come along loading the most recent updates of this package and trying the newest improvements and fixes. There are so many of them since last summer in such short time intervalls that one has the feeling this software gets better and more feasible every week.

You should definitely write a book some day on Theoretical Physics using the Maple Physics Package since the latest development of this package outmatches older books like Wang's or Reineker's 'Theoretische Physik mit Maple' by far who simply didn't have such a tool to beautifully set up their equations with the physics package. I'd definitely be one of the first people to buy an exemplar in the book store!

To ScotGould,

I can see your point that you want your students to properly use units without making mistakes by clearly distinguishing between algebraic expressions and units. I think if you had raised this issue a hundred or maybe even two hundred years ago this would still have been a matter of controversy and taste and also I would probably have agreed with you. Nowadays there is a strict convention concerning this matter and I think this was also the very first equation I saw on the blackboard when I entered university (you should also find this in nearly every book on Experimental Physics and I also don't think engineers use units differently):

Q = {Q} [Q]

where {Q} is the numerical value of some physical quantity Q and [Q] is its unit. This is the only legal way of using the square brackets though some people are a bit  sloppy and use it also for the physical dimension, i.e. [force] = M L/T^2 where, strictly speaking,

dim(F) = ML/T^2 and [F] = gm cm/s^2 would be appropriate.

To write r = 3.0 [m] is (at least to my mind) wrong and violating very basic physics conventions, no matter whether it's nice from the viewpoint of avoiding mistakes or not. You could never submit a paper writing formulas in that style in the field of physics: Astrophysics, Theoretical Physics or the like, though I have to admit I don't know what's it like in the Life Sciences and I wouldn't care too much if they have their own conventions.

Maybe Maple-developers decided to use this bracket convention for the very reason you are claiming. I have to admit that I am not very fond of that decision but at least I am happy there is a workaround to fix this issue for your own worksheets - many thanks again for your solutions.

@Carl Love and Joe Riel: Thank you very much, `print/Unit` will do it! I wonder why such formatting rules are not standard in Maple, after all the typesetting is done very nicely anywhere else, if one just thinks of e.g. differentials dx with upright d and slanted x and the like. I'm happy there is a workaround for my specific issue.

Thank you so much for that detailed explanation! To be honest, I did not consider that dot X^i has no tensorial character, this was a very valuable hint.

As for the expansion of the cross-product, I tried the expand and Simplify commands but using an older version of the Physics library. I'm very glad that this works now.

Thanks again!

Thank you so much for that detailed explanation! To be honest, I did not consider that dot X^i has no tensorial character, this was a very valuable hint.

As for the expansion of the cross-product, I tried the expand and Simplify commands but using an older version of the Physics library. I'm very glad that this works now.

Thanks again!

@ecterrab Thanks a lot! I forgot about the mapping 0 -> 4. The possibility to also use 0 when refering to matrix elements, which is somehow more consistent, is very nice.

@ecterrab Thanks a lot! I forgot about the mapping 0 -> 4. The possibility to also use 0 when refering to matrix elements, which is somehow more consistent, is very nice.

@ecterrab Sorry, I have been too busy to think about any suggestions. But I certainly will do so.

In the meantime I encountered another issue of the physics package. When I set up the signature of the space time I can choose between the diag([1,1,1,1]) and diag([-1,-1,-1,1]). Maybe this is a convention used in the Landau books to regard x^3 as time-like component, but any German book on Theoretical Phyics I've been working with so far has the convention x = (x^0, x^1, x^2, x^3) = (ct, x, y, z). Also in QFT the so-called "signature -2" is common as I've read. I know there is a simple workaround for this issue but it would be very nice to choose a signature (+,-,-,-) also. Is there, or will there be such a possibility in a future release?

Thanks a lot,

Peter

@ecterrab Sorry, I have been too busy to think about any suggestions. But I certainly will do so.

In the meantime I encountered another issue of the physics package. When I set up the signature of the space time I can choose between the diag([1,1,1,1]) and diag([-1,-1,-1,1]). Maybe this is a convention used in the Landau books to regard x^3 as time-like component, but any German book on Theoretical Phyics I've been working with so far has the convention x = (x^0, x^1, x^2, x^3) = (ct, x, y, z). Also in QFT the so-called "signature -2" is common as I've read. I know there is a simple workaround for this issue but it would be very nice to choose a signature (+,-,-,-) also. Is there, or will there be such a possibility in a future release?

Thanks a lot,

Peter

Thanks a lot for the suggested workaround and for considering to fix it in the next release!

Btw: I am using the Physics Package in all my courses of Theoretical Physics to check my calculations and really love it for its functionality and pretty-print output that resembles so much what you write when doing calculations with a pencil on the paper. Yet I think it has quite a steep learning curve (e.g. working with Christoffel symbols) in spite of the Physics-Example-Worksheet and Maple Help Page and I'm sure I am not using it to its full. Will there probably be some additional documentation on how to use it for physical applications in some future release?

Thanks a lot for the suggested workaround and for considering to fix it in the next release!

Btw: I am using the Physics Package in all my courses of Theoretical Physics to check my calculations and really love it for its functionality and pretty-print output that resembles so much what you write when doing calculations with a pencil on the paper. Yet I think it has quite a steep learning curve (e.g. working with Christoffel symbols) in spite of the Physics-Example-Worksheet and Maple Help Page and I'm sure I am not using it to its full. Will there probably be some additional documentation on how to use it for physical applications in some future release?