I'm not trying to quell debate and discussion. I just wanted to bring up the reason we created the package in the first place and encourage its consideration in future design decisions.
Jason moorepants.info +01 530-601-9791 On Fri, Oct 23, 2015 at 4:47 PM, Justin <[email protected]> wrote: > > > On Friday, October 23, 2015 at 7:27:45 PM UTC-4, Nathan Goldbaum wrote: >> >> >> >> On Fri, Oct 23, 2015 at 6:25 PM, Justin <[email protected]> wrote: >> >>> I have a terrible way of wording things.. more of a discussion between >>> myself and the author where he mentioned that I should ask the community >>> what they think. >>> >>> On Friday, October 23, 2015 at 6:43:51 PM UTC-4, Jason Moore wrote: >>>> >>>> Justin, >>>> >>>> I don't think there is a debate. We have a very nice vector >>>> representation in the physics package, but it is based on mutable types and >>>> isn't very general. We created the sympy.vector package to make a more >>>> general vector object that was based on immutable types with the idea that >>>> the physics vector could eventually be deprecated. Our new implementation >>>> may not be general enough for the mathematicians' taste and we are willing >>>> to improve it so that it is, but we would still want it to eventually allow >>>> us to deprecate sympy.physics.vector. The addition of vectors from >>>> different coordinate systems is essential to this plan. So whatever you >>>> want to do to improve the package will have my support but I hope that you >>>> will keep this intended use case in mind when you think about bigger design >>>> changes. >>>> >>> >>> I come from a physics background and can't see when or why this would be >>> useful so my opinion is certainly biased. As to the generality of the >>> package there are no constrains on doing this and, bias and all, this tells >>> me there ought to be some. I am new to contributing so I will keep my head >>> down and add functionality as you mentioned. I am not trying to step on >>> toes here... >>> >> >> A vector (e.g. the mathematical object, not necessarily its >> representation) should be independent of the coordinate system, no? So >> long as there are well-defined translations between the coordinate systems, >> it should certainly be possible to do arithmetic operations on two vectors >> whose representations are written down in different coordinate systems. >> > > The way I see it is that well are defining two different (or the same) > coordinate systems with different (or the same) basis and allowing > arithmetic operations between these two different (or the same) coordinate > system. The thing is we don't know until we make some definitions of the > coordinate system whether the vector is well defined. For example: > > Say C1 is defined and the origin is set at (0,0,0). C2 is defined where > it's y-axis aligns along C1's x-axis and origin is set at (0,0,0). Any > scaling could be set on C2 wrt C1. Is the operation > C1.x * C1.i + C2.y * C2.j a well-defined vector? > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sympy. > To view this discussion on the web visit > https://groups.google.com/d/msgid/sympy/e1c47552-167f-4cb1-aa33-19bb896af677%40googlegroups.com > <https://groups.google.com/d/msgid/sympy/e1c47552-167f-4cb1-aa33-19bb896af677%40googlegroups.com?utm_medium=email&utm_source=footer> > . > > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAP7f1AgFKCvB3BnyZejf-kCvGmoPE3%3DYqRMUF78C3mNwuMKffw%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
