GNU+HURD Won't Win
04/13/2018 (Fri) 07:48:26
I'm very sceptical that HURD is developed enough to be released by the end of this year. Does GuixSD even have a newbie-friendly installer yet, like the major distributions?
Anyway, let's assume that the world is perfect. HURD is completed by the end of the year. Then what? Now you have Linux and GNU as two distinct entities, distinct operating systems. Now third party developers might have to develop for one additional system; lots of developers don't even port their software to BSD, as far as I've seen. How would they deal with the threat of incompatibility with software originally designed to run with the Linux kernel? Maybe that so many developers use glibc will mean that this is not a big issue at all. I don't know.
Okay, what if it wasn't an issue? What incentive would people have to adopt the GNU operating system, instead of or in addition to Linux? The BSDs already operate according to a centralized development model that is its prime distinction from Linux. GNU would be either be moving toward that model or it could be like Linux and allow for distributions of the GNU operating system built upon HURD. But what would GNU have to offer? Obviously, completely, 100% free software. Ideologically, that's a big win, but this wouldn't translate necessarily to practical benefits for the user. The driver support would be worse, the software repository smaller. The only way it could grow would be to convince would be Linux users, and even would be BSD users, to develop for GNU; but aside from the free software philosophy, which is not exclusive to but only purer in the GNU OS, what would be the draw?
With that question in mind, I just don't see HURD ever taking off even if it finally makes it out of the hangar and unto the airstrip.