Re: [apsa_itp] a quiet revolution in democracy

["David Bray" <David_Bray@xxxxxxxxxxxxx>]

GDSS started at the the University of Arizona, with the goals of: 

- enabling all participants to work simultaneously (human parallel
processing); 
- providing an equal opportunity for participation; 
- discouraging behavior that can negatively impact meeting productivity; 
- enabling larger group meetings which can effectively bring more
information, knowledge, and skills to bear on the task; 
- permitting the group to choose from a spectrum of structured or
unstructured techniques and methods to perform the task; 
- offering access to external information; and 
- supporting the development of an organizational memory from meeting to
meeting. 

... and while this does not include all the features of your e-democracy,
there are sufficient similarities that all you to extend your examination
of 'will this work?' from theoretical conjectures to empirical evidence --
that then might indicate yea or nay. If there's one thing that IS has
learned over-and-over, it's the human side of the equation that's much
more problematic than technology. People will always find work-arounds and
loopholes in an IS artifact if it's to their advantage. For some features
of your proposal, pre-existing research into GDSS (and later work on CSCW)
provide come concrete examples to test your conjectures; perhaps this
helps? 

Two other thoughts: on security, while I'd like to believe that security's
getting better all the time, the events in Estonia
(http://news.bbc.co.uk/1/hi/world/europe/6665145.stm ) seem to indicate
that it's actually the black hats that are progressing faster than the
preventive measures. And while you might think it odd that people might
use bots or trojans to abuse such an e-democracy, I doubt many in the
early 1990's could have foreseen the problems of spam and phishing that we
now confront; a significant portion of Wikipedia features user-created bot
programs -- both good and bad. Lastly, I'm still not sure you'd have an
adequate representation of the entire nation participating (what of the
digital divide?); what are the representative demographics of Wikipedia or
Digg.com? 

-david bray
http://www.oii.ox.ac.uk/people/visitors.cfm?id=112
Oxford Internet Institute

Michael Allan <mike@xxxxxxxxx>,Internet on Friday, August 17, 2007 at 6:04
AM -0500 wrote:
>On Thu, Aug 16, 2007 at 05:47:13AM -0400, David Bray wrote:
>> Michael - you may also want to search for Group Decision Support Systems
>> (GDSS) as the information systems field in the 1980's researched
>> text-based democratizing systems similar in several ways to what you
>> outline below. ...
>
>I did a quick search, looking for a connection with delegate cascades,
>and pyramidal meritocracies... (no luck).  Because that looks to be
>the critical dependency; it all rises or falls on that.
>
>>                Further, while I'd like whole-heartedly to see
>improvements
>> with democracies, Peter's comments are quite true and represent large
>> obstacles to the proposal: 
>> 
>> >People aren't political. 
>> 
>> Flaws include the fact that the e-democracy probably will have
>> companies/lobbyists doing pyramid schemes to "buy" people's votes, just
>> automated through the Internet. "Click here to earn 700 Linden dollars a
>> day to spend in Second Life" ($2 USD)... or "Free access to mp3's or
>> hi-res video if you align yourself with the XYZ party!" Instead of being
>> an e-democracy, it will become more of an "e-market" where voters will
>be
>> courted (duped?) like consumers in exchange for their votes. 
>
>But secret ballots prevent that. So, if we felt that laws against vote
>buying could not be enforced, then we could resort to secret ballots.
>It would be just like traditional electoral system, in that
>regard. (But it would be better if vote buying could be dealt with by
>other means. Secret ballots have disadvantages of their own, being
>harder to implement, and harder to police.)
>
>> Second, where will people get the time to wade through all these bills?
>> Automated programs ('bots') would probably emerge to do edits as proxies
>> for people/corporations/lobbying groups. No one will have enough time to
>> read all the laws that are proposed, so while you propose that people
>> could "defer" to the judgement of others -- how will you explicitly set
>> these controls (would you have to do it for each bill, again time
>> constraints). Also, such deference no longer remains a democracy, but a
>> sort of networked system of vassals pledging to their lords, almost like
>> democracy remixed with feudalism. 
>
>Hilarious! But seriously, I do not worry where people would find the
>time.  If a hundred or so full-time legislators can manage it
>(somehow), then a hundred-thousand or so part-timers would far surpass
>that effort -- even by contributing, on average, just one day a
>year. In fact, many of them (more than a hundred, I bet) would
>actually work at it full-time. (We might have the opposite problem:
>too many people taking time off from their jobs and other
>responsibilities, for politics!)
>
>> Third, an e-democracy would have profound information security
>> ramifications, how do you (1) know the person voting really is that
>> person, and (2) that the person's account hasn't been hijacked or
>> zombified? These same security problems confront VISA, Mastercard, and
>> corporate banks and have not been resolved. You could issue digital
>> certificates to everyone, but in practice that's a huge headache; still
>> probably would have 'phishing' for votes. 
>
>True, security is a serious design and operational constraint.  I am
>not qualified to say too much here (no direct experience in this) but
>I know that many other applications face similar constraints, so we
>would not have to invent that wheel.  We would use the current state
>of the art. If it is not yet good enough, it probably would be, soon.
>(Financial networks are among the apps with similar constraints.)
>
>> >People are twisted. 
>> 
>> An e-democracy assumes that giving more decision-making power to the
>> masses is a good things. But I'm not sure this tenet has been proven as
>> humans, in large groups, can do some fairly nasty or violent actions.
>Who
>> will prevent a majority from abusing (or enslaving) a minority? Recall
>the
>> Stanford Prison Experiment and the tragic results observed there? If you
>> can obtain IRB approval, you may want to run an actual field study to
>test
>> your ideas of an e-democracy, to see if human beings behave as
>rationally
>> as an e-democracy would require. This could contrast theory with
>observed
>> behaviors. 
>> 
>> On a more basic level, will everyone vote only for their own local
>> interests, and thus the good of the 'nation' is lost as laws and bills
>> that only benefit largely populated area passed? With any democracy --
>> electronic or no -- will enough people vote for the good of the state,
>> even when this may not align with their own immediate self-interests?
>> (Does this even happen now?)
>
>Peter makes similar points, in the other threads. I will try to answer
>there. It all seems to depend on the strength of the vote cascade, and
>the pyramidal meritocracy.  (But I am mostly out of my depth in
>this. I need to read more. And I agree too, it needs to be tested.)
>
>> ... one last thought on wikis-like activities: Wikipedia 'shows biased
>> page edits' http://news.bbc.co.uk/1/hi/technology/6947532.stm
>
>Wikipedia is an interesting battleground. :) But biased page edits can
>only occur with a centralized 'push' model of collabortation, where
>contributions are pushed into a shared, central text.  The
>collabortative model underlying community law-making is strictly
>'pull', which makes it immune from external biases. (Participants have
>their own texts, and nobody else can modify them.)
>
>-- 
>Michael Allan
>
>http://zelea.com/
>
>--
>apsa_itp mailing list served by Harvard-MIT Data Center
>List Address: apsa_itp@xxxxxxxxxxxxxxxxxxxxxx
>Subscribe/Unsubscribe: http://lists.hmdc.harvard.edu/?info=apsa_itp
>

--
apsa_itp mailing list served by Harvard-MIT Data Center
List Address: apsa_itp@xxxxxxxxxxxxxxxxxxxxxx
Subscribe/Unsubscribe: http://lists.hmdc.harvard.edu/?info=apsa_itp