> So far, from everything I've seen I think sympy + whatever > we already do in SAGE is the best way forward for > pure symbolic manipulation in SAGE. > > I'll include the latest version of sympy (version 0.5.3) in > sage-2.8.4.1 so people can take a look at it. See > > http://code.google.com/p/sympy/wiki/Changes > > for what's been happening with sympy lately. > > In the long run (which for me means "1 year") it seems to me > that the only way to have excellent symbolic > manipulation in SAGE is that it's going to have > to get written "from scratch" since Maxima (and > several other systems) are just too archaic. > The sympy developers are hard at work at exactly > such a project, and I hope people with an interest > in calculus/symbolic computation will try out what > they're doing, subscribe to their mailing list, > post bugs, etc.
We need to implement the _sage_ methods in SymPy as agreed, we just didn't yet find time to do it. Also we are still changing internals if we find that we can speed things up, or clean things up. The end user shouldn't notice much changes though because we rarely change tests, nevertheless until we reach the version 1.0, we can change the interface if we find it badly designed. As to what kind of features one can expect in SymPy to work well: Calculus and stuff needed in physics and applied physics, because this is the motivation of me, Pearu, Kirill and others. The other half of developers are mostly doing computer science or mathematics. I am myself not interesed in other things, but of course we gladly accept anything that is pure python and can be integrated in sympy. My impression is that the main developers in SAGE are mostly interested in mathematical stuff like number theory, elliptic curves and stuff, but I don't need this at all. I want to do calculations in general relativity, one can do some easy calculations already, like the Schwarzschild solution [1]. Then I want to do some calculations of cross-sections in quantum field theory, this really stresses the manipulation facilities a lot (unfortunately, for anything interesting in QFT, SymPy will be too slow, that's why ginac was started, because Maple was just too slow and buggy). And soon I and Kirill want to use SymPy for calculating matrix elements in electronic structure calculations (there are some boring integrals and spherical harmonics calculations, so SymPy should be fast enough). People are used to using Mathematica/Maple for those kinds of things. I want to change that. I think SAGE will change that. So my priority now is to integrate SymPy with SAGE more, to see how maxima is standing next to sympy. I am curious if it's going to be faster/slower and how much. Ondrej [1] http://sympy.blogspot.com/2007/08/midterm-evaluation-is-over-and-noone.html --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
