The tyranny of the majority sucks. The tyranny of the scientific method
sucks because the definition of the "real world" is a historical concept
that changes constantly with new information about it--i.e., there is no
"real world" but an approximative construct matching the oscillation
between theory and experience. Meritocracy sucks, because it's the
tyranny of the minority who can and does put the most time and effort in
the project.

You probably want epistemic community decision making instead: in the
case of systemd, an epistemic community would be formed with people who
can read the code, test it, provide quality assurance, and understand
the consequences of this code in the whole system, especially for
failing cases. Such epistemic community MAY be the same as the
meritocratic oligarchy--e.g., Debian's Technical committee, but not
necessarily. This is a critical point to consider. An epistemic
community must be capable both to know and to transmit that knowledge.
It must take on the role of teacher and not decision maker. Then such a
process would come with rational recommendations based on scientific
results and cultural understanding, so that informed consent can occur
within the community.

The difference with decision by committee is that in the case of Debian
and systemd, rough consensus not being reached, the decision to default
to systemd would have not happened, and may have become a Task or Blend
project instead to circumscribe its use to cases for which it works.

My 2 Satoshi

==
hk

-- 
 _ _     We are free to share code and we code to share freedom
(_X_)yne Foundation, Free Culture Foundry * https://www.dyne.org/donate/