[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#727708: multiple init systems: We have to see it for what it is: Lennart/Linux OS. Not.



These discussions of grand schemes to change the future of Linux
development are potentially interesting in the right forum, but this isn't
the right forum.  The goal of this discussion is to decide what Debian is
going to do for jessie, with some discussion of what Debian might do after
jessie.  There are many possible visions of the future, but right now we
have to decide what Debian is going to do in the near term.

It's basically impossible to translate grand schemes for forking large
amounts of software, changing major development models, or the like into
any sort of practically achievable advice for the Debian project in the
timeframe of the next release.  Similarly, the concept of init
configuration generators has come up in this discussion before, but a
working set of generators and universal syntax does not currently exist,
and we can't adopt something that doesn't exist.

If you all find these ideas interesting, I encourage you to go tackle
them!  Write code, test code, make the world a better place, and give
people more options.  I'm sure there are others who would be interested in
discussing them with you, and possibly collaborating.

But the role of the Technical Committee on these sorts of discussions is
reluctant and trailing-edge, not visionary and bleeding-edge.  We're a
decision-making body of last resort called on to make decisions when the
project needs to have a decision and the correct option isn't clear.  As a
result, we have to deal with the practical, concrete, and currently
available.  We are neither empowered to, nor in a position to, set grand
design strategies for what other developers might work on, only to decide
how to integrate the code we have available to us now in places where the
right decision is not clear.

This bug discussion is already quite long and protracted, and it would be
easier for those following it if strategic discussions of how to drive the
future of Linux could be moved to other, more appropriate forums where
they have a better chance of finding their audience.

-- 
Russ Allbery (rra@debian.org)               <http://www.eyrie.org/~eagle/>


Reply to: