On 05/04/2011 05:06 AM, Reuben Thomas wrote:
> Sorry for the confusion. I shall start a new thread.
>
> Bruno & Jim, at least, you've both written several replies to earlier
> messages which I have little hope of collating in a sane amount of
> time, so I'd be most grateful if you could just reply once, here.
>
> On 4 May 2011 10:14, Reuben Thomas <r...@sc3d.org> wrote:
>> $ ls
>> */{acosl,asinl,atanl,cosl,expl,logl,sinl,sqrtl,tanl,isfinite,tmpfile,fflush,futimens,nanosleep,getgroups,tzset,strtoimax,strtoumax}.*
>> |xargs git log --pretty=format:%aN |sort |uniq -cThanks for summarizing again. From my perspective on all of those, my work on the fflush and futimens modules are the most complex (that is, providing new code not available anywhere else, especially since in the case of fflush we are fixing bugs present in current glibc), if we thought that a stronger license could be reasonably used as leverage to improve freedoms of projects trying to use the code. However, I'm definitely of the opinion that relaxing the license to LGPLv2+, to match the fact that glibc provides those same interfaces under that license, is reasonable. Therefore, I add my consent to relicensing the above list. However, for gnulib interfaces not in POSIX or glibc (this discussion has already mentioned modules like the gl-list abstraction), or which call exit() and therefore don't belong in a safe library, we really are adding value that should only be available for projects willing to be compatible with the GPL. Thankfully we are not trying to shoehorn any of those gnulib modules into libposix. -- Eric Blake ebl...@redhat.com +1-801-349-2682 Libvirt virtualization library http://libvirt.org
signature.asc
Description: OpenPGP digital signature
