Better would be to let anyone prepare a map. Draw a random sample of voters from throughout the State (say 5,000) as a redistricting panel – they can meet in their county seats. Have them rank all the maps, and determine the winner using the Condorcet method.

Then divide the ballots into groups based on the districts of the statewide winner. Recalculate a Condorcet winner for each district. If a district winner coincides with the statewide winner, then that district is completed.

Repeat the process in the rest of the State with a new panel.

The U.S.A. has its STONE AGE minority rule gerrymander regimes (inherited from the de facto gerrymander formation of the English House of Commons in the 1200s — a mere 700 plus years ago)

– half the votes in half the gerrymander districts = about 25 percent indirect minority rule — with the resulting EVIL powermad Donkey and Elephant control freaks in legislative bodies (esp. in 1 party *safe* seat areas).

Primary math is much worse — about 10 percent of the voters nominate the party hack gang that later gets in control in general elections.

For example, that would make Berkeley voters able to elect and OC state legislator and vice versa. It would never fly with those constituents.

OTOH, a “vote for 53″ on a mass at-large Congress ticket is an idea. It jives with Thornton and allows true at-large voting and eliminates in-state carpetbagging arguments.

Yes, lots of those. But nothing I’m aware of that says “the two largest parties …” rather than saying “all parties that meet some threshold condition”.

Thanks to Michael Seebeck for correcting my numbers.

Total Votes / Total Seats = EQUAL votes needed for each seat winner — via vote transfers using pre-election candidate rank order lists.

ALL voters elect a legislator — majority rule (aka Democracy) and minority representation.

Also minor parties get a better deal on petitions in lieu of filing fee, than major party members do.

Had the partisanship been totally removed and the commissioners been selected at random without regard to political labels, it might be useful.

Reference: the unusual death of a merchant seaman in East Saint Louis {Illinios] in 1966, the bicoastal bisexual adventures of an Air Canada steward in the 1970s……

In addition, there’s the problem that Prop. 11 puts the number two — as in two political parties — in the state constitution in a way that hasn’t been the case before. Rather than saying, for example, that each party that got X% in the last gubernatorial election has four seats each, or y% of total voter registration, it says the two largest parties get four seats each. See the difference?