Re: LuaJIT project governance

  • From: demetri <demetri.spanos@xxxxxxxxx>
  • To: luajit@xxxxxxxxxxxxx
  • Date: Tue, 18 Aug 2015 06:50:35 -0700


We would appreciate the thoughts of people in this group on the best

way to manage the project.


Speaking as a non-expert it seems to me the biggest "design choice"
for the group is having some kind of plan for how major new features
will be approached.

I'm very confident that we can put together a team of people to review
bug reports and bug fixes. I'm less clear on how we go about a major
feature like the GC that has a deep impact on all users and is full of
subtle design tradeoffs. In the absence of Mike's authorial guidance,
it's hard for me to imagine how (e.g.) a GitHub Issues thread is going
to suffice in exploring all the tradeoffs and making a decision that is
"best" in some holistic sense.

As Luke mentions Snabb adopting a model like the Linux Kernel, I
suspect the Linux Kernel style works because of a strong authorial
voice at the top who can provide macro-level quality control.

I don't have any particularly good suggestions. Perhaps Mike would
be willing to occasionally endorse a high level description of a plan
for a major feature (at the level of say the current wiki description of
a new GC). Given such a "blessed" outline of a feature I can imagine
a distributed team of volunteers making progress on an implementation.
And having Mike's opinion on what he considers the major design
tradeoffs would make it easier to keep the team on point. I realize
that this proposal is probably not ideal for Mike in terms of being
"free" of a management role the project. So perhaps it's a
non-starter.

But I do think that having some idea for how we approach major
feature changes is at the top of the list of problems to address.

Demetri

Other related posts: