[AFS3-std] AFS Standardisation

[Ken Hornstein]

>To explicitly address Ken's rather piquant "suggestion", the point IMHO
>is NOT to make this process depend on individuals.

Sigh.  I know that, and I was mostly kidding.  But let me speak to the IETF
process, since you brought that up (I don't think IEEE is necessarily that
close to the way OpenAFS runs or should be run, so let's not go there).

IMO, the IETF process kind of grew by necessity as the IETF itself
grew.  It went from a relatively informal and small group of people who
all knew each other to a giant group of people who didn't know each
other nearly as well.  Because the original process didn't really scale
at all, the IETF process had to get a lot more formalized.  I would
classify that as a necessary evil.

But in the AFS world, well, there AREN'T that many people.  You could
fit all of the major players in my living room (I speak from experience
there).  Everyone pretty much knows each other.  Now of course we should
develop new talent, but I don't see a HUGE influx of people arriving to
do OpenAFS work.  Including new people shouldn't be hard; there's plenty
of room for them.

I guess my point was that if people WANT to develop a process now to
insure fairness, well, I guess I have no problem.  I personally would
wait until the formal process was necessary before doing all of that
work.  And I think the TECHNICAL output of any process-derived document
would be equal to anything generated by a hypothetical "Steel Cage
Match" :-)

--Ken