It seems groups are much fuzzier than expected. Individual may accept to:
- Improve, non backward-compatible way (= anything)
- Bug-fix, non backward-compatible way
- Improve, backward-compatible way (including additions)
- Do nothing at all
These choices may apply to logic and grammar, leading to 4^2 = 16 possible combinations, even though some combinations are highly improbable and some others contain simply very few people (as la .guskant. said about former "group 2"). If we add "semantics" to logic and grammar, we get to 4^3 = 64. And if we add "No idea on the topic" as a choice, we got 5^3 = 125.
And inside some combinations, we find people in favor of X or Y proposition (Zasni Gerna of Xorxes, Solpahi's connectives, Zantufa...). which increases again the number of cases.
(Please note that I did not add "Bug-fix, backward-compatible" because from what I read, it seems merely impossible (correct me if I'm wrong). Else we'd have 6^3 = 216.)
So better not trying to but people in 3 "labeled compartments", I guess! That leads me to the conclusion that submissions must be evaluated on a per-case basis, with a stable and well-known evaluation grid.
***
Now about organizations.
I feel like the separation between BPFK and the GIT repositories maintained by the Coders' Group is nonsense (from an pure organizational point of view). I foresee multiple possible outcomes:
- Nothing changes: BPFK discusses/votes things that will never be included, and the Coders' Group include things that will never be discussed/voted. Lojban dies.
- Pure schism: each group "wakes up" (=becomes more active) and decides to build its own language. What will emerge is unclear to me. One sure thing is that the small community will be split into two (or more) weaker ones.
- Put in common:
- By cooperation: groups (or some members) agree to work together (or merge) with proper means, common rules and common tools. That requires adhesion and (good) tools.
- Cooperation may happen with renewal: groups may agree to create a new entity with new (or updated) rules.
- By dissolution: one group may simply dissolve, leaving the other one the only "official". I personally think this is dangerous because we surely need the point of view of everybody.
- By forcing: One group may force the other to accept its own way to work. The most obvious case would be preempting ("pull the rug out to"): BPFK could fork repositories and tag them as "BPFK Official" or whatever. This is unfair, but perfectly legal.
Of course, cases may be partial and mixed: some members may join the other group, while other create a new entity or a new language, etc. I just hope people won't be dumb enough to create a worse situation.
(The terms you were looking for are "compromises" and "trade-off"... ;-) )
That leads me to the conclusion that submission protocol/rules are to be proposed, discussed and accepted by a wide range of people. I won't enter pure language discussions, as I don't feel legitimate for this. But I'll try to propose solutions to help about rules and protocols. Any idea is welcome of course.
la .sykyndyr.