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

Re: Discussion: Possible GR: Enhance requirements for General Resolutions

On Mon, 05 Jan 2009, MJ Ray wrote:
> Ron <ron@debian.org> wrote:
> > Do you really think it would have been difficult to obtain 2Q seconds
> > for a resolution to recall the previous vote, and postpone it until
> > some of the more obvious glitches had been better ironed out?  [...]
> Yes, based on the summary of other votes by Wouter Verhelst and others.

I don't really think you can extrapolate from examples of past votes
that got considerably more seconds than they required to suggest that
proves we'll have trouble getting enough seconds for important issues.

In fact if past votes had regularly got 600+% more seconds than they
had required that would suggest to me that some large proportion of
people didn't actually understand the system they were participating
in.  It makes _no_ difference how many people second once the required
number is reached, so except in a few rare cases I _would_ quite expect
most thinking people (the type I'd most prefer to be involved in any
vote) to resist the urge to post more of them just for the "me too"
value of it.

Those results are not surprising, and if anything make it clear we
can easily get more seconds for notable issues than is currently
required.  How many more is debatable, but this isn't very good
evidence for your assertion that 30 people is a "very high" bar.

> So, are supporters hoping this situation will change, only a few
> well-connected DDs will be able to propose GRs, or what?

I don't consider myself "well-connected", but I don't really have
much doubt that I'd be able to get 30 people to put their hand up
if I had an issue of project-wide importance that we needed to
decide upon in a way that would be suitable for a vote.

If I couldn't do that, I'd expect to lose anyway, and approach the
problem in some other manner.

> > We seem to have totally lost the goal of making decisions that affect
> > many or all developers by consensus.  The process of building consensus
> > revolves around satisfying the concerns of people who see problems with
> > your planned course of action to arrive at a Better Solution.  If you
> > can't get the consensus of around 30 people to begin with, it doesn't
> > take a degree in advanced math or political science or military strategy
> > to arrive at the conclusion that you are a LONG WAY from having the
> > consensus of the whole project.
> In general, that's correct.

In general?  You have some specific counter example in mind where just 5
people really do represent a consensus of the project??

> In particular, if you need 30 people just to *start* the discussion period,

Other people already answered this.  It only needs _one_ person to post
about something to start a discussion.  The _formal_ discussion period
is limited in length, and IMO quite short.  Far too short in fact to
actually achieve a real, well considered, consensus in that time.
You need to start this process WAY BEFORE it ever gets to a formal
"last chance for discussion" period.  Else you are just certain to
still have disagreements by the time the vote _must_ be held, and the
looming vote can only further polarise opinions.  With a vote imminent,
the incentive to find workable compromise is minimal indeed.

> that's going to kill many potential options before they have any chance of
> building consensus

Many _potential_ options need to be 'killed' or integrated with other
options to build the larger consensus.  This is an absolutely essential
part of the process.  If we have 1021 potential voters and they all
separate into groups of 5 to push their own option or else, then we
will have ...  wait for it ...  204 options on each ballot.

And that's before some people second more than one option ...

I have lots of stupid ideas every day.  I don't for a minute believe
that all of them are worthy of inflicting on other people.
But sometimes I do need to inflict them on a just a couple of people
before their true stupidity is apparent and they can be shelved.
And occasionally, surprisingly, one tiny little part of that idea
will be the seed for something bigger.  Learning when to let go of
the crappy bits is a valuable lesson.

Ideas are like children, most people think their own are completely

> and others will be far too entrenched by the time public
> discussion starts; also, it's 30 DDs, not 30 people.

I'm not sure what you aim to imply there?  Are DDs more like sheep
than 'people' are or vice versa?  If people can't recognise a truly
Better idea just because it wasn't the first one to gain some measure
of support, then we are already doomed and the rules for voting don't
matter squat anyhow.

> so I've moved from seeking amendments, to emphasising
> the profound problems in the proposal
> Here's a summary list of concerns I mentioned in other emails:-
> 1. 2Q is unjustified and excessive;

The justification (or perhaps 'last straw') is the poor quality
of recent vote options, where many people even had quite some
difficulty figuring out what the difference between any two
options were.  We don't need 'vanity' options on a poll that
are essentially identical except for some disagreement between
two groups of 5 people that isn't clear at all to anyone else.
Condorcet may be a lovely system for presenting many options
but if people can't figure out what the difference is between
them, or what the effect of them will be, then it still isn't
going to work very well for making decisions.

The exaggeration about how big a change this is seems excessive,
but I don't think 30 / 1000 is by most normal scales of excess.

> 2. the obvious spoiler effect may exclude consensus options
> prematurely (interaction of thresholds and Condorcet voting);

Sorry, but that sentence is just entirely self-contradictory
and unparseable to me ...  Whatever effect you speak of is
not 'obvious' to me, and if options _had_ consensus clearly
there'd be more than 30 people supporting them and they
wouldn't be excluded ...

The only thing that might be excluded is options that were
prematurely pushed toward a vote ...  and if they have the
potential for real consensus, they can always be put to a
vote later once that is clear.

> 3. it favours organised campaign groups who gather in secret before
> springing discussion on debian lists;

It's harder to keep secrets among a group of 30 than it is among
a group of 5.  I don't see how this works in the favour of that
sort of thing.  Either they are a focus group that comes up with
something worthwhile that people will support, or a bunch of
dangerous wackos that we'll vote down when it gets that far.

> 4. it encourages defending proposals too early, during the discussion
> period.

I don't follow that either.  My whole point is the _real_ discussion
should all be taking place well before any formal process is started.
By the time the discussion period for an actual vote begins, it should
all be about honing the last few little details, not about "let's start
from a position of polar opposition and hostile antagonism and give
ourselves two weeks to sort that out".

You don't have to be a genius among profound thinkers to place a good
bet on how that will turn out.  We want to STOP this sort of behaviour,
not entrench it more deeply than it has already dug itself.

So I guess I'm sorry.  As much as I really do want to understand and
resolve your concerns, I don't find them particularly 'profound' as
stated here.  I won't discount that I'm just too stupid to understand
them, but in that case the onus would be on you still to convince me
of their merits.  Loaded explanations like "unjustified and excessive"
only work if you are preaching to the choir.  For the rest of us, that
will need to be backed up with some justification of your own if we
are to understand what injustice and excess really concerns you here.

> I don't think a 600% increase is a conservative step.

Fortunately this is just an error in your math :)  Let's see:

5  / 1021 = 0.004897, let's say 0.5% of potential voters.
30 / 1021 = 0.029382, let's say 3%

I don't think a < 2.5% change in the proportion of DDs required
to hold a vote and declare the result a "General" resolution
really puts us in the scope of some sort of radical revolution.

Claiming this is a 600% increase is, well, unjustified and
excessive I would say :)  The actual change itself seems very,
very, moderate without the application of creative accounting.

> don't you think there's a risk of this moving the project to
> votes where the options have simply been composed by *larger*
> warring factions?

Well if you really believe that might be a problem, then surely
you'd be in favour of my actually radical suggestion to raise
this threshold to something like 80% of people in the keyring?

The larger we make the factions, the more they'll have to solve
their _own_ infighting before they make their dispute everyone
else's problem to decide.  Funnily enough that's EXACTLY how
building a consensus does, should, and must start.

Can I count on you to second this if I propose it?  Or do you
think this should be shot down summarily if I can't find 4 other
people who agree?  Would it be better if I had to find 30 people,
or 50%, or 80% to do that?

> Alternatively, would it make the path of least resistance "ignore
> everyone else whenever possible because they'll never get 30 or 60
> DDs together"?

Are you saying that if I ever vote with some faction I will never
be able to "cross the floor" and vote with a different group of
people who I agree more with on some totally different topic?

> I wonder if there is a need for a more radical reform where once a
> proposal reaches the required number of seconds to trigger a vote, a
> ballot jury takes over and runs a vote which includes the proposal,
> the status quo, and a set of alternatives where the whole population
> would vote *for* at least one option, and reasonable combinations
> thereof.  Isn't that how Condorcet voting works best?

Wait, I'm confused again ...  if you are worried about secret groups
of 30 people having too much power to influence the project, where
are we going to get this jury from, and who will watch the watchers?

Unless you are gunning for a seat on such a body I don't see what
this gains us over putting the onus on proposers to establish that
their ideas are sufficiently refined to have a reasonable level of
support among the ordinary rank and file developers.

We don't need _more_ power structures, we need to limit the abuse
of them by small groups of people with an axe to grind or an agenda
to push.  We especially need to limit the people who spend more time
on this sort of busywork politik than they do on making actual useful
contributions to the project.  We AREN'T a democratic body, we are
an almost entirely disparate group of software developers trying to
build a common body of useful software.  We DON'T need a layer of
middle management to succeed at that.

If we NEVER have another vote for anything more significant than
who gets to wear the DPL hat until they disgrace themselves, then
I would consider that a MAJOR SUCCESS in refining our collaborative
process.  But I can accept that I'm dreaming the impossible dream
on that one too, and there will be times when AS A WHOLE we must
make some statement or decision on some issue.  Any vote that
passes by 50.01% cannot be said to have achieved that aim.  Every
time we have one of those we divide ourselves down the middle,
and a few more people unsubscribe permanently from one or more of
our mailing lists, and put a few more people on their personal

If you are worried about the vast majority of people who ignore
what goes on in -vote most of the time, then the answer is simple.
Make what you do there more relevant.  The bickering and the
wrangling is not attractive to people who care about the code or
the project.  Once that starts leaking out to affect us again
anyway, should you really be surprised that we'll build our own
consensus to rise up and stop you from doing that?

We may not have that consensus yet.  But it's growing.  And the
more thoughtless, ugly votes we have, the more people will refine
that consensus until we either get it right, or we splinter
entirely and leave the vote-mongers behind.

It's not really rocket science, not once you've seen it once
or twice before.  The interesting question is can we stop it
before it metastasises and rots a perfectly good project to
the core.  If we can, I think we WILL have done something
truly unique in the face of history.  So I'm hopeful.  And
telling it like I see it here in the hope others will have
the clues that I'm missing and we'll finally get it right.

But this isn't something we can get right with a straight up
or down vote, no matter how many options we put on the ballot.
We need to talk this through until either everyone's strong
objections can be assuaged, or it becomes perfectly clear that
their objections are about personal power struggles and not
the best interests of the project as a whole.

If we don't have the patience to do that, well, I guess we'll
just have another fucked up vote, and everyone will get nudged
just a little bit further in one direction or the other, and
whoever is left will do this all again next year ...

So that's about it from me, unless someone has something new
and particularly interesting to add I'm just going to get on
people's nerves from here on too.  I'm happy to talk to people
in private if they wish (and they can summarise that later if
it's worthwhile), but I don't want to join the ranks of people
who just repeat themselves over and over and over in the vain
hope that this will win people over to their way of thinking.

If enough people disagree with me that strongly, clearly we
are in no position to put it to a vote, so I'll just wait
patiently until they or I wise up sufficiently to find some
common ground about this problem.  It clearly won't go away
now until we do something proactive about it.


Reply to: