You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by le...@hush.com on 2012/03/05 22:19:55 UTC
[math] Tensor algebra implementation
>>
>This is not *my* idea, it has been around for some time now (long
>before I joined the project). The aim of this thread is merely to
>try
>and find the best way to proceed and capitalize on previous
>discussions. I would love to hear your thoughts (and those of
>others)
>on that proposal (the wiki thing).
Sorry about that , I joined the mail list so recently that i am not particullary aware of what it has been discussed before. I guess i should get more familiar with markmail.
>
>>
>> I kinda think your idea could be expanded in a more broader way
>> by also adding support for Cartesian Tensor Algebra. I think a
>feature like that would be welcomed.
>>
>My topic is solid mechanics, so I for one would of course love to
>see
>some tensor algebra in CM at some point [1]. I have some personal
>classes for handling tensors with Voigt symmetry (most common in
>mechanics), which might at some point find their place in CM.
>
>As for higher-order or more general tensors, that would be an open
>field for you to think about, if you're interested in this area.
>If I
>were you, I would have a look to the Maxima Computer algebra System
>[2], which has a package for high-order tensors (with co- *and*
>contra- variant components!), maybe you could draw some inspiration
>from the api (I know it's not OO design, nor numerically oriented,
>but
>still...).
Thanks for your suggestion ! I Downloaded the sources and i will check them soon. If I am correct , Maxima is implemented in Lisp (or CLisp) but I guess it could be feasible , once they had been analyzed and studied , to translate those algorithms into Java.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [math] Tensor algebra implementation
Posted by Sébastien Brisard <se...@m4x.org>.
Hi Luc,
>
> Maxima is GPL licensed. We have absolutely not rights to translate it
> and release the translated code in an Apache licensed library. See
> <http://www.apache.org/legal/resolved.html#category-x>.
>
> We have to either implement the code from scratch using only reference
> paper or translate code that is published under a compatible license.
>
right. But we can still draw inspiration from the public API and the
functionalities that are offered, or can't we?
Sébastien
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [math] Tensor algebra implementation
Posted by Luc Maisonobe <Lu...@free.fr>.
Le 05/03/2012 22:19, leandro.pezzente@hush.com a écrit :
>
>
>>>
>> This is not *my* idea, it has been around for some time now (long
>> before I joined the project). The aim of this thread is merely to
>> try and find the best way to proceed and capitalize on previous
>> discussions. I would love to hear your thoughts (and those of
>> others) on that proposal (the wiki thing).
>
> Sorry about that , I joined the mail list so recently that i am not
> particullary aware of what it has been discussed before. I guess i
> should get more familiar with markmail.
>
>>
>>>
>>> I kinda think your idea could be expanded in a more broader way
>>> by also adding support for Cartesian Tensor Algebra. I think a
>> feature like that would be welcomed.
>>>
>> My topic is solid mechanics, so I for one would of course love to
>> see some tensor algebra in CM at some point [1]. I have some
>> personal classes for handling tensors with Voigt symmetry (most
>> common in mechanics), which might at some point find their place in
>> CM.
>>
>> As for higher-order or more general tensors, that would be an open
>> field for you to think about, if you're interested in this area. If
>> I were you, I would have a look to the Maxima Computer algebra
>> System [2], which has a package for high-order tensors (with co-
>> *and* contra- variant components!), maybe you could draw some
>> inspiration from the api (I know it's not OO design, nor
>> numerically oriented, but still...).
>
> Thanks for your suggestion ! I Downloaded the sources and i will
> check them soon. If I am correct , Maxima is implemented in Lisp (or
> CLisp) but I guess it could be feasible , once they had been analyzed
> and studied , to translate those algorithms into Java.
Maxima is GPL licensed. We have absolutely not rights to translate it
and release the translated code in an Apache licensed library. See
<http://www.apache.org/legal/resolved.html#category-x>.
We have to either implement the code from scratch using only reference
paper or translate code that is published under a compatible license.
Luc
>
>
> ---------------------------------------------------------------------
>
>
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [math] Tensor algebra implementation
Posted by Sébastien Brisard <se...@m4x.org>.
>
> Thanks for your suggestion ! I Downloaded the sources and i will check them soon. If I am correct , Maxima is implemented in Lisp (or CLisp) but I guess it could be feasible , once they had been analyzed and studied , to translate those algorithms into Java.
>
That's great Leandro. I was merely referring to the API, not really
the implementation of the different algorithms. However, if you feel
like diving into Lisp source code, good luck! ;)
Also, I seem to remember that you were willing to propose new features
regarding wavelets. I myself do not use these beasts, but I would tend
to think that such a feature would probably find a wider audience
among the CM users than tensor algebra. Maybe you could then
prioritize, as developing a whole new set of features will eat up a
lot of your time?
What do others think about wavelets vs. tensor algebra?
Best regards,
Sébastien
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org