Hello,
On Aug 24, 9:28 pm, "Ondrej Certik" <[EMAIL PROTECTED]> wrote: > > This was definitely very true a year ago. Did you get any sense that > > this had changed when you were at the scipy conference? Last year > > when I was at that conference, there was a lot of talk about improving > > matplotlib, but I don't know whether anything came of it in the last year, > > perhaps because VTK/mayavi meets a lot of people's needs for > > 3d plotting in python. > > mayavi2 is alive and in development, but I think the matplotlib and 3D > hasn't improved much. > Is one of the above substantially better than the other or does each have its strength and weaknesses? VTK/mayavi seems to be to be more impressive, I had a look at it a couple years ago to visualize data from finite element computations and for that it just seemed to be the best (open source) package out there - and it wipes the floor with quite a lot of the commercial code out there, too . The other question I have is whether we must use the Tk interface for VTK/mayavi or can we use it via some other way like Java? Tk would be yet another dependency. Another (for personally) important issue is that VTK/mayavi has good Windows support, so that the SageLite port to Windows remains a possibility. I have had discussions via #sage-devel with William about a pexpect port to python on Windows compiled with MSVC , but not code has been written yet (and probably won't be before SD5). > > If SAGElite is just the non-math parts of SAGE, then it won't be > > very much work, since that work is already mostly done. In fact, > > it's likely that it will make for *less* work for SAGE, because people > > who are very good at programming will use SAGElite (for the GUI > > interface to Python or program XXX or something, say), and will > > contribute back improvements. > > OK, that makes sense. The parts, like DSAGE and notebook should be > packaged as individual packages. This will be useful for packaging > SAGE in Debian in the future. > +1 The more packages we get into Debian unstable the less of a maintenance burden we will have because hopefully other people will help share the load. And that makes a modular Sage in the Debian universe more likely - with the previously discussed limitations in mind. Is also applies the other way around: Due to some issues with clisp on Solaris I have been looking at gcl. The last official release is a couple years old (2.6.7), but Debian unstable provides tarballs of 2.7.0pre since about 2003. So instead of going at it isolated we can profit from the expertise of the Debian people (who for example also do some testing on Itanium/Linux) and hopefully in time start to contribute back so that down the road everybody will profit. This is what Open Source is all about. > If we manage to integrate sympy well in SAGE, than I would be much > more motivated to create a proper SAGE packages in Debian. :) I think > it's very important to have Debian and thus Ubuntu packages - I just > read a recent survey [1], that Debian+Ubuntu use 44% of people using > linux on desktops. > Tonight at 11: Fedora & SuSE are dying. ;-) > Ondrej > > [1]http://www.desktoplinux.com/news/NS8454912761.html Cheers, Michael --~--~---------~--~----~------------~-------~--~----~ 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/ -~----------~----~----~----~------~----~------~--~---
