There was a news report last week that House Majority Leader Eric Cantor had prompted a thread on Quora about how Congress could use social media to enhance the legislative process. That topic is dear to me, so I posted an answer. Unfortunately, as is far more often true than not in these kinds of situations, it seems that people are just raising questions without showing any real concern for the answers. In any case, here’s the link to the thread on Quora, and here’s my answer:
The time is ripe for building an Online Town Hall worthy of the name.
It would need to be:
1) Undeniably useful for soliciting and bubbling up good ideas;
2) Impeccably fair at ensuring a fair hearing to all comers, and;
3) Massively scalable.
There are several obvious and consistent shortcoming in the current generation of online voting/crowsdourcing/Q&A systems (Quora, 10Questions, Google Moderator, Mixed Ink, Microsoft Town Hall, Liberty Lab, et al).
1) They leave themselves open to “Power Law” biases which favor suggestions made by early entrants and celebrity participants, bubbling up content in ways that make the popular ever more popular, leaving many potentially valuable contributions buried in a virtual haystack.
This problem could be rectified by various queuing innovations, requiring participants to rank or rate a set of existing contributions (say, the immediately prior 10 or 20) before their own ideas are eligible for consideration. A practice which insists that people must listen before they speak would guarantee a fair round of initial vetting.
2) The systems mentioned generally use plurality voting or extremely simplistic rating systems which fail to capture the comparative sentiments toward a multiplicity of competing ideas.
This problem could be rectified by implementing ranked choice voting tools within the online townhall. RCV has been shown to be quite good at illustrating how opinion can coalesce around consensus ideas. I’ve built a prototype at http://apps.facebook.com/we-vote/ to demonstrate the power of an interactive ranked choice/Instant Runoff balloting system in a social media environment. (Note that there is a straw poll ballot available there for the GOP 2012 presidential nomination.)
3) Members of the general public become reluctant to participate when they conclude that a venue is being overrun by astroturfers, noisy idiots, and silly swarms. Publics also become disheartened when they feel they have legitimately organized to be heard, and the venue owner nevertheless ignores or downgrades their contributions.
This problem can be rectified by engineering the contribution vetting process in ways that will ensures every coalition (pre-arranged, or self-discovered in process) gets an opportunity to be seen putting its best foot forward, so to speak. The technology used to reduce question duplication here at Quora could be implemented in ways that reduce answer and argument duplication in an online town hall system. This would reduce the load on consumers of information and steer generators of information into making refinements rather than essentially repeating content already uploaded by potential friends and allies.
4) Many online straw polls/ political crowdsourcing experiments have devolved into echo chambers for narrow coalitions. The Contract From America was a premier example of this.
Here again, ranked choice voting tools that provide rich visualization of result sets can rectify the problem. They would need to be combined with: a) clear delineation between nomination and voting phases on an issue, and b) tools that allow ad hoc subqueries within the participant datasets for the purpose of surfacing minority coalitions. Tools could be built along such lines that would make it possible for seemingly polarized oppositions to negotiate bottom line shared consensus points.
Again, please see my RCV/IRV demo at http://apps.facebook.com/we-vote/