Blogging for America

Hyperarchy: Open Decision-Making for Open Organizations

It was 9 p.m. on a Friday night and six Code for America Fellows were trying to decide where to go for dinner.  It went something like this:

  • Joel: Let’s go to Bindi.  They’re right across the street and they’re open late.
  • Scott: Bindi sucks!  Let’s go to King of Thai.  Their entrees are only $5.
  • Tyler: Yeah, Scott, but it’s a 15 minute walk and no one else lives in North Beach!  Let’s just order pizza.
  • Michael: I can’t eat anymore pizza.  Seriously, let’s just go to Subway.

And so it continued, with volume levels and personal insults on the rise.

The irony of the situation was not lost to us: as Code for America Fellows, we’re at the forefront of the Open Government movement, which strives to promote a more transparent, collaborative, and efficient form of government.  And yet, there we were, unable to make a simple decision, and shooting nerf guns at each other and calling each other names instead.

Fortunately for us, I had recently been introduced to Nathan Sobo of Pivotal Labs, who is working on a project called Hyperarchy.  Hyperarchy is an online decision-making tool that allows team members to pose questions, submit answers, and rank the list in their preferred order.  Under the hood, the system applies the Tideman method of ranked pairs to generate a consensus list from the sum of personal ratings.  By imposing a logical structure on the messy process of collaboration, the system allows teams to deliberate without falling victim to groupthink or gridlock.

We decided to use Hyperarchy to resolve our bickering over restaurants.  Within minutes of posing the question “Where should we eat dinner?”, all six of us were online, sorting the list of options in our preferred order.  As new preferences were registered, votes flashed across our screens and the consensus list resorted in real-time.  This elicited shouts, and threats to rank other people’s favorites at the bottom of our own lists, which led to further reshuffling.

Maybe because we were so unruly when we started the process, we were unable to abide by Hyperarchy’s hyper-logical result.  Instead, we reverted to proven political tactics: coalition-building; veiled threats; nerf weapons.  (For the record, we ended up at King of Thai, which, at the time, was our third choice.)

Even though Hyperarchy didn’t solve our immediate problem, the experience got us thinking about the democratic process and how this process is affected by technology.  Would election results vary significantly if voters were allowed to adjust their ballots in real-time?  Is geography still a meaningful component of representative democracy in today’s hyper-connected world?  Would the founding fathers have structured things differently if there had been a World Wide Web?

Nathan thinks that technology can and will transform democracy.  As he put it in a brown-bag talk at Code for America: “representative democracy is a ‘lossy’ algorithm that produces fuzzy, pixelated images — we can do better.”  His immediate goal for Hyperarchy is to help organizations achieve a more democratic and egalitarian process.  Hyperarchy is a finalist for 2011 Y Combinator funding, which will be announced at the end of the month.

Hyperarchy is currently in a beta version, but if you want to see how consensus-building can be aided by technology, log in to www.hyperarchy.com and try it out.

Code for America Labs, Inc is a non-partisan, non-political 501(c)(3) organization. Content is licensed through Creative Commons.