# Re: Supermajority requirement off-by-one error, and TC chairmanship

```On Sat, Feb 16, 2008 at 08:31:12PM -0800, Don Armstrong wrote:
> On Sat, 16 Feb 2008, Bas Wijnen wrote:
> > Yes, that too. :-) But as I wrote, for the 50% situation, there is a
> > reason we want that. We want to say "there are more people in favour
> > than against". With the supermajority, we want to say "there are
> > many more people in favour than against".
>
> Right. My main point is that for pathologically small values of voters
> such as this, changing the meaning of super-majority to include
> equality means that there is no effective difference between majority
> and super-majority. Perhaps this is a bug that should be solved by
> increasing the TC membership instead. [If 7 people are voting,
> suddenly all of these issues go away; 4/3, 5/2, and 6/1 become the magic
> numbers for N of 1, 2 and 3.]

I don't think we should bother people with TC work for statistical
reasons. ;-)  I mean, if the TC is too small, and it would be good for
it to be bigger, sure.  But just because we can't figure out how to
count the votes, doesn't seem like a good reason to add members. :-)

> > When the actual value is arbitrary anyway, it makes sense to solve
> > it.
>
> All of the values we pick are going to be arbitrary to at least some
> degree, so this isn't terribly convincing to me.

I think we should see that for extremely small samples, such as the TC,
statistics don't really work.  The whole idea of talking about "a
percentage of the voters" stems from the approximation that one voter is
an negligible part of the total.  With 6 people, it's 16 2/3%.  That's
obviously not negligible. ;-)

However, there's not much we can do about that.  Even with 7 people, you
still have the same problem if only 6 are voting.  But it seems
reasonable to me to require supermajorities in clear terms.  That is,
when we want to say "at least 5/6", we should not say "more than 2/3".

If we don't really want to say "at least 5/6", because "at least 2/3" is
good enough, then we can say that.  That was the original proposal.  I
don't really have an opinion on what the value should be, but I do think
we should say what we mean.

And of course, "at least 5/6" isn't exactly the same as "more than 2/3",
when some people don't vote.  So it is also possible that we do actually
want to keep it like this.  But I think it's good to do this conciously,
because what's being said doesn't seem to be what it really is.

Thanks,
Bas

--
I encourage people to send encrypted e-mail (see http://www.gnupg.org).