On Jun 21, 2009, at 8:21 AM, William Stein wrote: > 2009/6/21 gsw <georgswe...@googlemail.com>: >> >> On 21 Jun., 15:54, William Stein <wst...@gmail.com> wrote: >>> 2009/6/21 Bjarke Hammersholt Roune <bjarke.ro...@gmail.com>: >>> >>> >>> >>>> I quote from >>> >>>> http://www.sagemath.org/doc/developer/inclusion.html >>> >>>> which is about the inclusion procedure for new packages. The first >>>> requirement is written as: >>> >>>> "The license must be a GPL version 2+ compatible license. (This >>>> will >>>> be publicly revisited around Jan 15, 2009.)" >>> >>>> Whatever was decided or not decided around Jan 15, the date or the >>>> statement is in need of being updated. >>> >>> The GPL v2+ restriction wasn't publicly revisited. Therefore, >>> let's >>> publicly revisit it. >>> >>> Does anybody who cares a lot have a strong opinion on whether Sage >>> should start allowing in new libraries that are licensed GPLv3+? If >>> so, why? Please, no flamebait, unless you post only to the >>> sage-flame mailing listhttp://groups.google.com/group/sage-flame >>> where flaming is encouraged. >>> >>> -- William >> >> We can't hold back time, so sooner or later, some new libraries will >> appear that are GPLv3+, and which one would like to have in Sage, or >> some of the libraries present in Sage-4.0.2 will switch to GPLv3+. On > > In might be useful to list such libraries people know about, and also > what the implications have been for Sage: > > * libcocoa - it is GPLv3+; some people wish it were in Sage. I'm > pretty confident that the authors would change the license to GPLv2+ > if that were the only thing blocking inclusion of libcocoa into Sage. > > * GSL - it is now GPLv3+, and we *do* ship the GPLv3+ version in > Sage. > GSL is in "maintenance mode", so it doesn't really matter -- we > could easily put a GPLv2+ version back in for people who want a GPLv2+ > version. > > * GNUtls -- it is now GPLv3+, and we do ship the GPLv3+ version. -- > We *shouldn't* switch back to the GPLv2+ version, since this is > security software, and as such old versions typically have known > security vulnerabilities. > > * GMP -- it is LGPLv3+, but we forked it partly because of the > license change, and we provide an LGPLv2+ version. This turns out to > be valuable to a lot of our commercial competitors, since for various > reasons several of the MA's (maybe Mathematica, Magma, etc.) appear to > have reasons to not want to ship LGPLv3+ code. > > * MPFR -- I think it is now about to become officially LGPLv3+. > For people that require GPLv2+ only code, we'll ship an older version > of mpfr. This is probably reasonable, since very little seems to have > happened development with with MPFR in years, except for some speed > increases and minor bug fixes (correct me if I'm wrong). > > Is there anything else major on the horizon that we should be aware > of? I think listing libraries like I'm doing above is at least a > constructive way to discuss this topic. I hope nothing above is > considered flamebait -- that wasn't my intention (but if it is, please > be considerate and flame me on sage-flame, not here).
None of the above seem compelling enough to require a change in policy at this time. The new (not included) MPC library will have to follow MPFR (slower, but perhaps more accurate, than our own implementation on top of MPFR). Something else that might be of interest, http://cado.gforge.inria.fr/ , is also still v2.1. - Robert --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
