[pp.int.general] LQFB: status quo in Germany // was: liquid feedback papers and/or data?

[carlo von lynX]

Since I am active in both German and Italian PP I see how
the German bias is offuscating some thinking, also some
fiction has established itself as mainstream party thinking.
So let me add some counter-information:

On Sun, Apr 27, 2014 at 03:32:40AM +0200, hyazinthe at emailn.de wrote:
> Currently the area "binding online resolution system" is heavily under construction in Germany.
> We used to have an unbinding LQFB, but for various reasons it is not accepted to a worth mentioning degree.

The unbinding LQFB worked exactly once, in the year 2010 when the
Berliners wrote their election programme. Almost everyone was
convinced of it, everyone was motivated, the proposals were
astoundingly more advanced then what a regular editorial process
could have come up with so the rhethoric of "swarm intelligence"
was born - because it described best what we actually experienced.
The regular assembly just ratified the choices made in LQFB and
we had an amazing programme that had a relevant role in winning
the elections. In a survey Berliner pirates said that the liquid
democratic principle was the most important ingredient to their
success:   https://lqpp.de/be/issue/show/635.html

I would add there was another essential ingredient to our 2011
success: humility. We were humbled by our collective intelligence
and by the electorate who actually started to believe in us. We
have quickly lost that humility after the elections and it has
impeded a constructive use of LQFB ever after. I have never again
seen an LQFB project work so well as the original deployment.

> Right now in April 2014 probably not more than 30 people use it; it easily could be thousands
> of people in the scope of the german pirate party.

Thirty is such an exageration I must assume you are biased by
your own agenda.

> At the same time it got obvious, that we urgently need a binding online resolution system:
> When you have special success, thousands of members and a couple of them in parliaments,
> there is a strong structural need for making official basicdemocratic (!), fast resolutions

"Basic" or direct democratic voting has two very serious problems:
- When used in small scale, loud minorities can take over the votes
  (so called domination of the extroverts or dictatorship of the actives)
- When used in large scale, you introduce populism since the majorities
  will decide by whatever is their current thinking without any obligation
  to fully understand the issue at hand or listen to other opinions.

These problems are both addressed by liquid democracy. As much as you
don't like some aspects of LQFB, you can't deny that it is a superior
model of democracy to either direct or representative democracies with
their long history of respective failures.

> between party conventions; so being able to do that any time. You all better start early
> establishing such a structure; you gonna need it. Most striking reason for needing this is the
> history of the german pirate party: With success small, but very loud and aggressive groups came
> to infiltrate the pirate party germany and try to set the tone by being loud and radical, and you know what ?
> It actually works pretty well – unless there is a way to make official, basicdemocratic, fast resolutions.
> If a system is basicdemocratic, such destructive groups have no chance to dominate by simply being
> very loud and aggressive.

You just described the problem of the loud minorities.
You may want to check out http://my.pages.de/activerts
which elaborates on that. Yet the solution you are 
proposing is clearly worse than liquid democracy.

> We're in the process of establishing the "Basisentscheid" (translated "basis/grass route resolution"); that's
> the name for a different online resolution system, which is binding. A lot of issues, which LQFB has, the
> "Basisentscheid" doesn't have. I can't say, when it's going to be ready, but don't expect that to
> be fast, because there are very, very less people actually indeed working on realizing it.
> The problem with wishing and complaining on the one hand and actually indeed doing something on the other hand, you know ?
> But on one of the last party conventions all gathered german pirate party members already voted for the resolution of
> implementing the "Basisentscheid"; so, it's not just a theory or exercise – it is seriously coming for sure.
> 
> Here are the TOP 5 reasons why LQFB fails in Pirate Party Germany:
> 
> 5. No home developer team
> Relying on LQFB without an own home developer team means making oneself independant from a problematic software, which
> development is not in one's hand.

The Austrian Pirates have a very effective developer team.
I recommend to everyone to check out the work they have done
on the LQFB code. Austria also has a permanent assembly, so
LQFB is much more important to them than to the Germans.
The Austrian team also includes the excellent work done by
a developer in Bavaria, the man who wrote "Pirate Feedback."
So it's plain untrue that there are no developer teams,
it's just that everyone is afraid of running a different
codebase than the official one. It's like being scared to
replace PGP with Pond, but instead everyone criticizes how
PGP isn't secure enough - and keeps on promoting it.

> 4. Not so good verification process
> Establishing a verification process is an important preparation of making LQFB binding; you need to make sure that behind every
> vote there is only one real person. The current verification process is not implemted, yet, but is really not the best and not really data reductive
> One can fix it, but the attitude of the people behind LQFB is "this is not a problem, YOU are the problem".
> Whenever there is any problem regarding LQFB, you instantly meet unwillingness to accept the issue as a problem and additionally solve it.

This is a problem specific to Germany. Other pirate parties
have understood that the participation in a voting platform
is a political act which deserves to be transparent like we
expect other political groups to be transparent. It's ridiculous
that Germans expect their privacy to be respected when they are
acting as politicians.

> 3. The interface is very unattractive and not very functional
> The current interface scares a lot of people away, is very difficult to access, and doesn't make lust to deal with the system.

An alternative UI has been developed and deployed, yet people
stick to the old one. In Italy the UI has never been an actual
problem: If decisions in LQFB are binding you have a motivation
to deal with possible imperfections and just live with it, and
that's what happens. People use LQFB day-in day-out. Of course
you should contribute your ideas for improving the UX and UI,
but still - democracy is not simple, there is only so much
that you can improve - and do you stop riding the train just
because the ticket machines have horrible usability?

> 2. Super delegation problem
> A super delegation means, that extremely much votes are concentrated on one single person what enables this person to decide alone

No doubt after the success of 2011 Berliners were so enthusiastic
about the delegation principle, they used it abundantly also in the
national LQFB with a side effect of dominating the votes over other
regions of Germany. I'd say that was a bit over the top.

It is important to understand however that the problems with
direct and representative democracy are serious and big and
cannot just be ignored and denied as you are suggesting to do.

Liquid democracy in principle addresses those problems, and even
if a single person collects a particular high number of delegations
it is still less bad than regular representative democracy: because
you can remove that delegation at any time.

That is also why the notion that superdelegates should be elected
in secret is wrong: Secret vote is necessary in an election of a
person because after that election the person stays in power for
a certain number of years. In liquid democracy, if there is any
doubt on how a person achieved her high number of delegates (say
she bribed a certain percentage of it), then the remaining people
can remove their delegation, effectively de-electing the person.
(And you should have a party big enough that it is impossible to
bribe all members ;-)).

> 1. Unlawfulness & Voting machine problem
> The Voting machine problem means, that technically a democratic election basically only can have 2 of the following 3 qualities:
> Electronic, secret & open to scrutiny.

That is why it is absurd to expect digital voting to allow for
privacy of the participants. If you don't like being a politician,
stay out of the steering permanent assembly of your pirate party,
but don't expect the party to either have no transparency and
accountability, or to revert to more primitive forms of democracy
with their respective huge problems.

> the time elite problem is solved by cyclic voting phases,

LQFB doesn't have a time elite problem.

> in contrast to the sMV the "Basisentscheid" doesn't exclude offliners,

If offliners actually exist, then they usually are very social
people. All they need to do is go to the nearest weekly meeting
and have a friend pirate help them participate in the voting tool.
This is rhethoric that hasn't seen a decent debate or is kept up
even if the counter-positions are known.

> in short or mid term the "Basisentscheid" will be staffed with an own
> software, which is in our hand, so that we don't depend on the
> LQFB developers, who are not favorable to us,

All open source software is in our hands, and there are already
tons of solutions for populistic direct democracy out there.

> to fill the hole, which removing delegations left, with a kind of advice/recommendation system;
> so, you can't give your power to another person, who then technically decides for you, but
> you can follow people you trust regarding certain topic areas or concrete topics, if you want;
> when you follow them, that means, that you can read what they think of certain
> proposals, if they make a comment on certain proposals.

Which is exactly how LQFB is actually being used everywhere where
a binding permanent assembly has been implemented with it. You could
spare yourself the work by simply enforcing that delegations are
disabled if a person hasn't been participating frequently enough.

> In the course of the years another insight raised:
> Discussion is more important than making a resolution.

Yes, that much we agree upon. LQFB has a very evolved debating
system compared to plain direct-democratic issue & checkbox
voting tools, but it sure could use some rehaul.

> And anologue speaking in systems:
> A binding web-based resolution system is not enough,
> we also especially need a web-based discussion system, which
> is made for optimizing discussing instead of just enabling discussing.

That is correct, so propose something. LQFB indeed lacks
in that area. I remember the Austrians making some experience
with reddit-like forums... I think a slashdot approach could
be helpful. Would be best if your new Meinungsfindungstool
was fully integrated with LQFB.