On Fri, Nov 22, 2002 at 12:06:01PM -0600, Manoj Srivastava wrote: > I think this is not really a matter of screwing up, this is a > matter of, in some cases, avoiding the tyranny of the majority;
...a platitude directly rebutted by the paper to which John Robinson
linked.
> My contention is that there are a number of documents that
> define what the project is; and that we agreed to follow when we
> signed on to the project, and any changes to these documents, which
> cut to the heart of not just the developers, but the whole free
> software community (the DFSG is known as the gating criteria for free
> software far beyond the extents of the project), ought to be signed
> on by _most_ of the developers.
Yes. Anything more than half is "most", by definition.
> So, supermajorities are, in my opinion, still needed for cases
> where we want a (very) rough consensus, where mere majority ought not
> be the sole criteria for adopting a measure.
consensus
n : agreement of the majority in sentiment or belief [syn: {general
agreement}]
Hear, hear. I agree that we should have a rough consensus before
changing such documents.
We just don't need a supermajority to have it.
(Technically, I suppose, we'll always have a supermajority except in
cases where the winning option does so by only one vote.)
--
G. Branden Robinson | Optimists believe we live in the
Debian GNU/Linux | best of all possible worlds.
[EMAIL PROTECTED] | Pessimists are afraid the optimists
http://people.debian.org/~branden/ | are right.
msg02151/pgp00000.pgp
Description: PGP signature
