On Tue, Feb 11, 2020 at 8:09 AM Fred Kiefer <fredkie...@gmx.de> wrote: > And most importantly, Ivan will you have time to cut all these releases? Of > course we could move that point around a week or two if that date fits better
I've only now seen this. This is ok and please coordinate with me closer to the release date. It's nearly zero effort for me if maintainers keep up to date Documentation/news.texi and other files you can see me updating by hand just before release. Most time-consuming is a) sorting out through 'what changed' and revising news.texi et al, b) getting the GPG signing key to work, c) coordinating what version we should have actually bumped to because we might break Debian automated checks etc. I'm ok doing all these, but more I have to do, more planning do I have to put into dedicating huge blocks of time to a release. (Note, having package maintainers and distro packaging maintainers online -- preferably in a realtime groupchat e.g. on IRC -- when cutting a release would not be a bad thing. This means we can sort out and test any issues within hours rather than days.) On Thu, Feb 27, 2020 at 9:10 PM Riccardo Mottola <riccardo.mott...@libero.it> wrote: > before a release, I would like these issues to find a solution. > > - understand why we don't work properly on my ThinkPad T23 neither with > the Cairo nor with the xlib backend with similar issues Did you date when things broke and at what change? I'm afraid that when I'm cutting a release I only use a single platform (Ubuntu or Debian amd64 with libobjc2 and clang) with usually one or two backend configs (usually cairo). Sometimes I build with GCC or with GCC runtime. I don't have other environments around, including a working environment on Windows. I don't know what should be release blockers, I'm leaving that to individual package maintainers. > - understand/fix/workaround the libobjc2 which impede me to test current > GNUstep on any FreeBSD 11.x and 12.x I currently have on i386 and amd64 Honestly, I would have expected that libobjc2 FreeBSD should work out of the box given David is/was working on FreeBSD...