Re: [Ag Meinungsfindungstool] Abschied

[Scott Raney <scott AT metacard.com>]

On Sat, Sep 6, 2014 at 2:59 PM, Alexander Praetorius
<alexander.praetorius AT serapath.de> wrote:
> one rule of thumb would be to build fast feedback loops, so that you spent
> less time debating and speculating, but getting more real feedback, so the
> smaller the "MVP" (Minimum Viable Product), which already brings at least
> some "value" to whatever kind of users (in our case, that would be the
> pirate party and their internal communication processes), the better.

Agreed: We simply don't know enough about how human social interaction
to actually "engineer" most parts of these systems, which means doing
a lot of trial-and-error work instead.  Rapid Prototyping of the UI
(or "UX" as they seem to call it nowadays) to try different things
must be a part of overall plan.

> Just try to boil it down to the minimum feature set, of which you are kind
> of very sure, that those will be needed anyways and push it out.
> Then, when there will be actual usage and other kinds of feedback, the
> discussion is less exposed to waste time arguing about things that cannot be
> known for sure yet.

Agreed, again!

> this "ontology" we are building sounds nice in theory, but makes it also
> more complex to develop real systems that make use of it.

I'm not proposing a full understanding of authoritarianism is
necessary before we can start building decisionmaking systems that
don't fail when faced with authoritarian behavior.  But to continue
with Martin's analogy, the Wright brothers had a very rudimentary
understanding of airflow over a wing and made some serious errors in
their design (turns out their wing sections were too thin, a flaw
latter addressed by Fokker).  Still, what little they did know was
essential for them to get as far as they did, and IMHO unless we use
what little we do know about authoritarians and social dominators to
inform our designs, we're going to build a "crashing machine" instead
of a flying machine ;-)

> OWL (Web Ontology Language) was broken down into a light, medium and heavy
> weight version and you could argue, that RDF is even a more "lightweight
> version", because its all about other programmers understanding what it is
> about and building programs, that make use of all those features.
>
> So, even though, in a more "evolved future" all those "details" of the
> ontology make sense, but the initial version has to be radically simplified
> and should aim for just pushing out something that people can actually use
> to easily implement there own discussion systems that make use of it...

This brings up another issue: Horses for courses.  A system that is
designed, as you propose, to facilitate communication among a
relatively homogeneous group (which I'd assume describes the Pirate
Party) has a different set of requirements from one intended to be
used by the general population.  Although to some extent you can adapt
some of the ideas from one type of system to the other, IMHO it should
be a design goal to generalize the system as much as possible.  For
example, relying a robust moderation system might be optimal if you
have a couple of hundred active participants, but it becomes a
nightmare to manage if you have thousands or potentially millions of
users.  My philosophy of course is that representative government is
fundamentally flawed, and although things like Liquid Democracy
address the worst of the problems, the system must be designed to
support true direct democracy in those (hopefully few) cases where it
is essential to do so.
  Regards,
    Scott