* Wouter Verhelst <wou...@debian.org> [2019-04-01 13:55]: > One thing that Debian has historically been good at, is to produce ports > for various architectures. However, we're not the most widely ported; ... > macOS, HPUX, IRIX, AIX, QNX, and Solaris[2]. > > Should we try to catch up with these other systems in terms of ports? > Specifically today, should we try to make Debian usable on any of the > operating system kernels that I quoted above?
No. Debian doesn't have to be "the most X" in every area. While it's true that Debian has historically played an important role with ports, I don't see how combining Debian with a proprietary kernel helps advance our mission to produce a free OS. (Maybe I misunderstood what you're proposing.) Of course, if someone wants to do the work we shouldn't stand in the way unless it puts too many costs on others. But is this an effort the DPL to rally around and encourage? Based on your email, I don't see why (but again, maybe I don't see the full proposition). We have to acknowledge that our resources are limited and use them where they have the biggest impact. I'd rather see more efforts spent on figuring out how Debian could be useful in a mobile environment. > [1] https://en.wikipedia.org/wiki/Gentoo/Alt#Gentoo_Prefix I should add one caveat to what I wrote about: if for example we improve APT/dpkg/whatever to come up with some improved solution for today's packaging systems, I could see an argument to be made to port this to Windows and macOS to get developer adoption and make it easier to switch to Debian proper (similar to how GCC used to be used on proprietary Unixes before GNU/Linux was usable). -- Martin Michlmayr https://www.cyrius.com/
