Gerrymandering

[gfish]

As it does on a regular basis, the subject of gerrymandering has come up again. And, as always, I'm seeing people make the perfectly reasonable suggestion that we deal with it algorithmically. I'm all for that... until it is claimed that this would somehow make it non-political. And that's just bullshit. Dangerous bullshit.

Comments

[randomdreams]

Huh, that's a really good, thoughtful short essay. Would you post this on g+ so I could reshare it?

[gfish]

Done!

[sistawendy]

Arg. See my comment in randomdreams's repost about the Iowa solution. It's crude, but reasonably effective.

[andrewducker]

Alternatively, switching to a proportional system would fix this remarkably well...

[gfish]

Sure, that's always an option. It gives political parties way too much power, for my tastes. You can't mount primary challenges if the politicians are just chosen off a ranked list defined by a central party process. And I like representatives having a connection to a limited geographic area. I want them to have a sense of who their constituents are, beyond "the entire nation generally, and the party process that put me on the list specifically".

But that's all fiddly details. If proportional systems are the easiest way to create a system that doesn't inspire distrust in the voting public, I'm all for it.

[andrewducker]

A purely proportional system across the whole company would certainly have that effect.

But AMS (as used for electing representatives across Scotland) gives you local representatives for each area elected under FPTP and then a top-up across larger regions to make it proportional overall.

And STV (as used for local elections in Scotland) gives you something in-between with multi-member seats and ranked voting, which is proportional-ish but removes most of the party control.