First off, thank you, Blogger, for giving us advance warning of the outage. Good, one less distraction to distract me from the stuff I need to do this evening.

Second: I have been thinking about how to solve the problem Proposition 77 was ostensibly written to solve but unable to solve. We do need a way of addressing the horrible situation that is the gerrymandering of California State Senate/Assembly/Federal House of Representatives districts. I will agree with those who were pro-Prop 77 (Including Kos from DailyKos) on one thing: the status quo sucks. However, appointees or elected commissioners can be unduly swayed by partisan considerations to commit a screw job on the public and give us yet another tortured map.

Here's my idea. Empanel a group of programmers, who pledge to remain non-partisan, to build a networked cluster computer system (it would be isolated from local area, wide area and internetworks and only communicate within the cluster) and write Free/Open Source software to create a completely unbiased system for creating districts which will be the best compromise between these factors:

1.) Diversity: all districts should be as integrated as possible.
2.) Political Partisan Diversity: all districts should be as balanced as possible between the Democratic and Republican Parties. Let's get real folks, adding third parties into this bipartisan mix would serve only to confuse.
3.) Geography: all districts should make as much geographic sense as possible.

The F/OSS solution would run on a F/OSS operating system like Linux. Linux currently is the operating system of four of the top five supercomputing networked computer clusters so it is definitely up to this task.
Because the solution for crunching demographic data to create districts will be licensed with a F/OSS license like the GPL v.2 and written with F/OSS tools and run on a F/OSS clustering operating system, all aspects of the new system could be audited by teams of computer scientists picked by the political parties and by non-partisan political organizations like the California League of Women Voters and California Common Cause.

There would be plenty of time to shake out the bugs before it does its job: crunching the California results from Census 2010 to establish the new districts.

After the Census 2010 results are processed, and three alternate versions of a California redistricting map were created, each one would be put up to a vote by duly registered voters of the State of California. The version of the map that gets the most votes would be the version that would prevail until 2021, when the next redistricting cycle would run. The process of creating a new supercomputing cluster and writing new software for the purpose would begin anew a few years before the fact, and will be ready for the next round of redistricting. As such, the system would never run more than a few years behind the state-of-the-art and would be continuously improved between redistricting cycles.

So basically this process:

• would be completely nonpartisan


• rely on impartial computers, not biased human beings


• would use objective criteria for creating districts


• would be continuously improved
  

Sounds good, doesn't it? Email me to let me know what you think.