From octave-maintainers-request at bevo dot che dot wisc dot edu Tue Dec 16 21:57:55 2003 Subject: octave-forge -> octave From: Paul Kienzle To: octave-maintainers at bevo dot che dot wisc dot edu Date: Tue, 16 Dec 2003 22:57:29 -0500 [Continued from bug-octave list] On Dec 16, 2003, at 10:35 PM, John W. Eaton wrote: > | Okay. I can't remove the functions from octave-forge since > | other octave-forge routines depend on them, so I may include > | yours instead, or better yet, have a facility to install them only if > | they are not already present. > > Yes, I understand the transition problem, but I hope that if Octave > starts to include code from octave-forge, that the octave-forge > routines will eventually be removed (assuming the Octave versions do > everything we need for octave-forge). That's what I did for the last big batch (around 2.1.35?) In future though, I think I will move them to an 'old' directory and make more use of configure to see what needs to be installed. > At this point, my intent for adopting functions is that if they are in > octave-forge for compatibility with the base Matlab, I would like to > see them moved into the core Octave distribution. There are a number of them. Particularly the interpolation routines. I have a small concern that gutting octave-forge may lead to it becoming irrelevant, but so long as major packages like signal, image, communications, and maybe one day statistics and control systems are host there, I guess there is little danger of that. BTW, you may want to consider vertical rather than horizontal integration for things like the differential equations solvers. You can't split them off into a specialized package when they depend on the base libraries in libcruft, the C++ interface in liboctave, and the glue layer in src/DLD-FUNCTIONS. Or to put it another way, how can liboctave be constructed from a separable components? Any idea how much liboctave is used outside of octave itself? > Likewise, some of > the current functions in the core Octave distribution should be moved > out to separate packages. We are happy to host them. Paul Kienzle pkienzle at users dot sf dot net