Re: Restated Amendment: We Choose Wording of the Day

2015-09-04 Thread Jakub Wilk

* Sam Hartman , 2015-09-04, 14:28:

  - GENERAL RESOLUTION STARTS -


  Constitutional Amendment: TC Supermajority Fix

  Prior to the Clone Proof SSD GR in June 2003, the Technical
  Committee could overrule a Developer with a supermajority of 3:1.

  Unfortunately, the definition of supermajorities in the SSD GR has a
  off-by-one  error.  In the new text a supermajority requirement is met
  only if the ratio of votes in favour to votes against is strictly
  greater than the supermajority ratio.

  In the context of the Technical Committee voting to overrule a
  developer that means that it takes 4 votes to overcome a single
  dissenter.  And with a maximum committee size of 8, previously two
  dissenters could be outvoted by all 6 remaining members; now that
  is no longer possible.

  This change was unintentional, was contrary to the original intent
  of the Constitution, and is unhelpful.

  For the avoidance of any doubt, this change does not affect any
  votes (whether General Resolutions or votes in the Technical
  Committee) in progress at the time the change is made.

  Therefore, amend the Debian Constitution as follows:

Index: doc/constitution.wml
===
--- doc/constitution.wml(revision 10982)
+++ doc/constitution.wml(working copy)
@@ -913,7 +913,7 @@
 
 
  An option A defeats the default option D by a majority
-  ratio N, if V(A,D) is strictly greater than N * V(D,A).
+  ratio N, if V(A,D) is greater or equal to  N * V(D,A) and 
V(A,D) is strictly greater than V(D,A).
 
 
  If a supermajority of S:1 is required for A, its majority 
ratio






  Constitutional Amendment: Fix duplicate section numbering.

  The current Debian Constitution has two sections numbered A.1.
  This does not currently give rise to any ambiguity but it is
  undesirable.

  Fix this with the following semantically neutral amendment:

   - Renumber the first section A.1 to A.0.


  - GENERAL RESOLUTION ENDS -


Seconded.

--
Jakub Wilk


signature.asc
Description: Digital signature


Re: Restated Amendment: We Choose Wording of the Day

2015-09-04 Thread Wouter Verhelst
On Fri, Sep 04, 2015 at 05:34:44PM +0200, Wouter Verhelst wrote:
> However, as I've said in <20150903164145.gb23...@grep.be>, I think the
> better fix is to update 6.1.4 as follows:
> 
> -4. Overrule a Developer (requires a 3:1 majority)
> +4. Overrule a Developer (requires a 2:1 majority)
> 
> This way, the off-by-one change that was introduced with CSSD is
> reverted *for the TC*, while the supermajority definition isn't.

Only after I sent this mail did I notice that there's actually a second
reference to the 3:1 number in the paragraph below. Not very relevant
since this isn't a formal amendment, but if it turns into one, that
would need to be changed as well...

-- 
It is easy to love a country that is famous for chocolate and beer

  -- Barack Obama, speaking in Brussels, Belgium, 2014-03-26


signature.asc
Description: Digital signature


Re: Restated Amendment: We Choose Wording of the Day

2015-09-04 Thread Wouter Verhelst
Hi Sam,

On Fri, Sep 04, 2015 at 02:28:20PM +, Sam Hartman wrote:
>- GENERAL RESOLUTION STARTS -
> 
> 
>Constitutional Amendment: TC Supermajority Fix
> 
>Prior to the Clone Proof SSD GR in June 2003, the Technical
>Committee could overrule a Developer with a supermajority of 3:1.
> 
>Unfortunately, the definition of supermajorities in the SSD GR has a
>off-by-one  error.  In the new text a supermajority requirement is met
>only if the ratio of votes in favour to votes against is strictly
>greater than the supermajority ratio.
> 
>In the context of the Technical Committee voting to overrule a
>developer that means that it takes 4 votes to overcome a single
>dissenter.  And with a maximum committee size of 8, previously two
>dissenters could be outvoted by all 6 remaining members; now that
>is no longer possible.
> 
>This change was unintentional, was contrary to the original intent
>of the Constitution, and is unhelpful.
> 
>For the avoidance of any doubt, this change does not affect any
>votes (whether General Resolutions or votes in the Technical
>Committee) in progress at the time the change is made.
> 
>Therefore, amend the Debian Constitution as follows:
> 
> Index: doc/constitution.wml
> ===
> --- doc/constitution.wml  (revision 10982)
> +++ doc/constitution.wml  (working copy)
> @@ -913,7 +913,7 @@
>
>
>An option A defeats the default option D by a majority
> -  ratio N, if V(A,D) is strictly greater than N * V(D,A).
> +  ratio N, if V(A,D) is greater or equal to  N * V(D,A) and 
> V(A,D) is strictly greater than V(D,A).
>
>
>If a supermajority of S:1 is required for A, its majority 
> ratio
> 
> 
> 
> 
> 
> 
>Constitutional Amendment: Fix duplicate section numbering.
> 
>The current Debian Constitution has two sections numbered A.1.
>This does not currently give rise to any ambiguity but it is
>undesirable.
> 
>Fix this with the following semantically neutral amendment:
> 
> - Renumber the first section A.1 to A.0.
> 
> 
>- GENERAL RESOLUTION ENDS -

I second this amendment.

However, as I've said in <20150903164145.gb23...@grep.be>, I think the
better fix is to update 6.1.4 as follows:

-4. Overrule a Developer (requires a 3:1 majority)
+4. Overrule a Developer (requires a 2:1 majority)

This way, the off-by-one change that was introduced with CSSD is
reverted *for the TC*, while the supermajority definition isn't.

I say "change", because I don't buy the argument that it's a "bug"; I
think the removal of the "X + one" requirement would be a bug. However,
I do accept that requiring 3 + 1 votes in favour per opposing vote is
problematic for the TC, so I do agree that reducing the required number
of votes is desirable. The above does that.

To clarify, I've always interpreted a "majority" to mean "One vote more
than X over Y". This definition is easily proven correct when one
considers a simple majority: to get a simple majority, strictly more
than 50% of the vote is required (otherwise you don't have a majority,
you have an equilibrium). While the constitution doesn't specifically
refer to 1:1 simple majorities (it doesn't need to, since a result that
doesn't manage to reach simple majority wouldn't be the condorcet
winner), I think it would be inconsistent and wrong for us to change the
constitution in this manner. If that argument manages to convince you, I
would appreciate it if you could update your proposal in that manner.

Having said that, I don't feel strongly enough about it to make this a
formal amendment; while I think the change would be undesirable for
regular GRs, I doubt it would make much difference in practice, so I'm
not going to pursue this unless someone pokes me.

Regards,

-- 
It is easy to love a country that is famous for chocolate and beer

  -- Barack Obama, speaking in Brussels, Belgium, 2014-03-26


signature.asc
Description: Digital signature


Re: Amendment: We Choose Wording of the Day

2015-09-04 Thread Sam Hartman
Fixed, I hope; thanks.



Re: Restated Amendment: We Choose Wording of the Day

2015-09-04 Thread Sam Hartman

Restated to fix comments received.
For formality, to the extent that I am able, I withdraw my previous
amendment.

As I discussed, in Andreas's resolution, I think that the strategic
voting fix introduces more problems than it serves.  INstead, I propose
that we don't fix that, but trust ourselves to propose ballot options
that are statement-of-the-day-like ballot options not requiring a
super-majority when doing so is wise.  I think that doing so is
generally a good idea when you have a super-majority option and its
opposite on the same ballot--when there is substantial contraversy about
whether to move in the direction of the super-majority option or some
other option on the same ballot.

I have chosen to retain the preference for the default option in the TC.
If four members of the TC really cannot live with an option, we're
better off with more discussion or taking it to a GR.

Even in the Init system discussion, which I think is the most
controversial decision to come before the TC, several of the TC members
who preferred options that did not win explained what changes would need
to be made for them to consider options similar to the one that won to
be acceptable (ranked above FD).
As it happened, four TC members didn't think no decision was better than
the decision we got: if four members had ranked the winning option below
FD, the chair would not have had the opportunity to use his casting
vote.

We also have some strong evidence from emails where some TC members
explained their balloting decisions including what they ranked above FD
that the tactical voting people were afraid of didn't happen.

We're actually quite good at deciding whether another round of painful
discussion is worth the cost or not, and when people we've appointed to
make these decision decide that it is, I'd rather not second guess them.

Specifically, I formally propose to replace the GR text with:

   - GENERAL RESOLUTION STARTS -


   Constitutional Amendment: TC Supermajority Fix

   Prior to the Clone Proof SSD GR in June 2003, the Technical
   Committee could overrule a Developer with a supermajority of 3:1.

   Unfortunately, the definition of supermajorities in the SSD GR has a
   off-by-one  error.  In the new text a supermajority requirement is met
   only if the ratio of votes in favour to votes against is strictly
   greater than the supermajority ratio.

   In the context of the Technical Committee voting to overrule a
   developer that means that it takes 4 votes to overcome a single
   dissenter.  And with a maximum committee size of 8, previously two
   dissenters could be outvoted by all 6 remaining members; now that
   is no longer possible.

   This change was unintentional, was contrary to the original intent
   of the Constitution, and is unhelpful.

   For the avoidance of any doubt, this change does not affect any
   votes (whether General Resolutions or votes in the Technical
   Committee) in progress at the time the change is made.

   Therefore, amend the Debian Constitution as follows:

Index: doc/constitution.wml
===
--- doc/constitution.wml(revision 10982)
+++ doc/constitution.wml(working copy)
@@ -913,7 +913,7 @@
   
   
   An option A defeats the default option D by a majority
-  ratio N, if V(A,D) is strictly greater than N * V(D,A).
+  ratio N, if V(A,D) is greater or equal to  N * V(D,A) and 
V(A,D) is strictly greater than V(D,A).
   
   
   If a supermajority of S:1 is required for A, its majority 
ratio






   Constitutional Amendment: Fix duplicate section numbering.

   The current Debian Constitution has two sections numbered A.1.
   This does not currently give rise to any ambiguity but it is
   undesirable.

   Fix this with the following semantically neutral amendment:

- Renumber the first section A.1 to A.0.


   - GENERAL RESOLUTION ENDS -


pgpUpgzfDuSSS.pgp
Description: PGP signature


Re: Amendment: We Choose Wording of the Day

2015-09-04 Thread Edward Allcutt

On Fri, 4 Sep 2015, Sam Hartman wrote:

--- doc/constitution.wml(revision 10982)
+++ doc/constitution.wml(working copy)
@@ -913,7 +913,7 @@
 
 
  An option A defeats the default option D by a majority
-  ratio N, if V(A,D) is strictly greater than N * V(D,A).
+  ratio N, if V(A,D) is greater than N * V(D,A) and v(A,D) is 
strictly greater than v(D,A).
 
 
  If a supermajority of S:1 is required for A, its majority 
ratio



Should that be "greater than or equal to", and should those latter be 
capital Vs?


--
Edward Allcutt



Re: Amendment: We Choose Wording of the Day

2015-09-04 Thread Jakub Wilk

* Sam Hartman , 2015-09-04, 12:03:
Unfortunately, the definition of supermajorities in the SSD GR has a 
fencepost error.


Please say "off-by-one" rather than "fencepost". See this sub-thread for 
rationale: https://lists.debian.org/20150831082617.gs9...@sym.noone.org



if V(A,D) is greater than N * V(D,A)


For me, "greater" and "strictly greater" means the same thing.
I think you wanted to say "greater or equal" here.


and v(A,D) is strictly greater than v(D,A).


Here both "v" should be uppercase.

--
Jakub Wilk



Amendment: We Choose Wording of the Day

2015-09-04 Thread Sam Hartman

As I discussed, in Andreas's resolution, I think that the strategic
voting fix introduces more problems than it serves.  INstead, I propose
that we don't fix that, but trust ourselves to propose ballot options
that are statement-of-the-day-like ballot options not requiring a
super-majority when doing so is wise.  I think that doing so is
generally a good idea when you have a super-majority option and its
opposite on the same ballot--when there is substantial contraversy about
whether to move in the direction of the super-majority option or some
other option on the same ballot.

I have chosen to retain the preference for the default option in the TC.
If four members of the TC really cannot live with an option, we're
better off with more discussion or taking it to a GR.

Even in the Init system discussion, which I think is the most
controversial decision to come before the TC, several of the TC members
who preferred options that did not win explained what changes would need
to be made for them to consider options similar to the one that won to
be acceptable (ranked above FD).
As it happened, four TC members didn't think no decision was better than
the decision we got: if four members had ranked the winning option below
FD, the chair would not have had the opportunity to use his casting
vote.

We also have some strong evidence from emails where some TC members
explained their balloting decisions including what they ranked above FD
that the tactical voting people were afraid of didn't happen.

We're actually quite good at deciding whether another round of painful
discussion is worth the cost or not, and when people we've appointed to
make these decision decide that it is, I'd rather not second guess them.

Specifically, I formally propose to replace the GR text with:

   - GENERAL RESOLUTION STARTS -


   Constitutional Amendment: TC Supermajority Fix

   Prior to the Clone Proof SSD GR in June 2003, the Technical
   Committee could overrule a Developer with a supermajority of 3:1.

   Unfortunately, the definition of supermajorities in the SSD GR has a
   fencepost error.  In the new text a supermajority requirement is met
   only if the ratio of votes in favour to votes against is strictly
   greater than the supermajority ratio.

   In the context of the Technical Committee voting to overrule a
   developer that means that it takes 4 votes to overcome a single
   dissenter.  And with a maximum committee size of 8, previously two
   dissenters could be outvoted by all 6 remaining members; now that
   is no longer possible.

   This change was unintentional, was contrary to the original intent
   of the Constitution, and is unhelpful.

   For the avoidance of any doubt, this change does not affect any
   votes (whether General Resolutions or votes in the Technical
   Committee) in progress at the time the change is made.

   Therefore, amend the Debian Constitution as follows:

Index: doc/constitution.wml
===
--- doc/constitution.wml(revision 10982)
+++ doc/constitution.wml(working copy)
@@ -913,7 +913,7 @@
   
   
   An option A defeats the default option D by a majority
-  ratio N, if V(A,D) is strictly greater than N * V(D,A).
+  ratio N, if V(A,D) is greater than N * V(D,A) and v(A,D) is 
strictly greater than v(D,A).
   
   
   If a supermajority of S:1 is required for A, its majority 
ratio






   Constitutional Amendment: Fix duplicate section numbering.

   The current Debian Constitution has two sections numbered A.1.
   This does not currently give rise to any ambiguity but it is
   undesirable.

   Fix this with the following semantically neutral amendment:

- Renumber the first section A.1 to A.0.


   - GENERAL RESOLUTION ENDS -


pgpBrum4Qz7ha.pgp
Description: PGP signature