I agree with you in principle, and I understand the existing constraint. But I don't see developer time scaling up without those operating systems reaching (aga

I agree with you in principle, and I understand the existing constraint. But I don't see developer time scaling up without those operating systems reaching (again) critical mass, and then having upstream wanting to write drivers for even more operating systems. Which is fine per se. But I wish people could buy relatively new hardware (say, last 3 years) and be able to run FreeBSD or Illumos with most things working.