[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Our supermajority requirement has changed !



In the original constitution, which can be seen here:
 http://www.debian.org/devel/constitution.1.0
in order to overrule a maintainer we need a 3:1 supermajority
including the chairman (because it counts as a tie).  I think the need
for the chairman to agree in this case is a mistake.

In the `tidyups' which included switching to Schwartz Sequential
Dropping, seen in the current constitution:
 http://www.debian.org/devel/constitution
this was changed.  See A.6(3), and note the use of the words `strictly
greater'.  The effect of this with large electorates is largely
irrelevant, but with a small electorate like the Technical Committee,
it effectively increases the supermajority requirement.

The plain language of the committee's power to overrule a developer,
in 6.1(4), says `this requires a 3:1 majority'.  However if one tech
ctte member dissents the current wording of A.6(3) would _four_ other
ctte members are required, ie a 4:1 majority, as otherwise the number
of yay-sayers would not be _strictly greater_ than 3 times the number
of nay-sayers.

The change also had the effect of remvoing the casting vote for
choices between the default option and another option, so for example
if we have 3 votes A:FD:B and 3 votes B:FD:A, then FD would win,
whereas previously the casting vote would decide.

So, I think this is a bug which should be fixed.

Manoj, as Secretary, can you confirm that you agree with my
interpretation of A.6(3) ?  Do you have an opinion about the apparent
conflict with the plain language of 6.1(4) ?

I think the right fix is to delete the word `strictly'.  If we do
delete the word `strictly' it might be better to invent a different
phrase for `defeat[s] the default option' in A.6(3) because `defeats
the default option' in A.6(3) means something different to `defeats
[an option which happens to be the default option]' in A.6(4) onwards.
Perhaps replace `defeats' with `matches'.

Ian.



Reply to: