Comment by winchling | Hacker News Reader

Comment by winchling | Hacker News Reader

Parent Full thread

What's the ideal size of a polis? Too small: there's no privacy. Too large: also no privacy.

In the first case, for example a roving band of stone-agers, say N = 30 with all males related and everybody knowing the leader. Which is good because whatever the political system, leadership entails consent of the led. But bad because the leader knows everyone so it's harder to act or think differently. Errors in understanding the world are firmly entrenched.

The second case might be for example a modern state of N = 100M, with IT. There are notional leaders but they are partly controlled by unseen oligarchs, possibly residents of other states. The bureaucracy and businesses collect and sell information about everybody. Voting, if it exists at all, generates little consent, since only a negligible fraction of people know the official leaders, let alone the oligarchs, let alone what sort of decisions will be consequently be made.

The ideal is therefore presumably where N is some fairly low multiple of Dunbar's Number. But even if we wanted to we don't have the technology or knowledge to implement this yet.

It's almost as if decentralized federalism of the United States, with a limited federal government, states which are essentially 50 laboratories to experiment with which ideas work, which don't, and which need to be different, with lots of county and town level control as befits the granularities of each level of affiliation, starts to make sense.

>states which are essentially 50 laboratories to experiment with which ideas work, which don't

I've been noticing an edge case where a bunch of people sharing certain beliefs about how government should work manage to ruin one state with their belief but because they're fairly well off they can then flee to several other states and put them on the same path before the experimental results from the first state can be analyzed and published. Of course no raindrop feels responsible for the flood so nobody ever has to think about their personal impact on society.

Freedom of travel and inability to restrict cross(-state)-border commerce mean that whole classes of potentially-very-nice policies can't be part of those "experiments", though, or can be but only at much higher cost than they'd be if states could control those things. Others don't make much sense at a state level, like environmental regulation—you can regulate pollution all you like, but the states up-stream or up-wind might choose "defect" on that particular dilemma and then you're screwed anyway.

I think you make a great point here. Certain aspects of interstate commerce law in its current form shouldn't be off the table for reconsideration. The experiment is yielding results that merit discussion, particularly around guns.

Now imagine this applied to an entire planet.

All countries doing their own experiments, no wars, and people being free to travel across borders at any time, to live in and support the ideals they prefer.

Doesn't work that well if people move to other countries to change them to become like their original country. Would only work if travel across borders with the intention of staying was forbidden. Would also incentivize people to change their own country for the better when they can see other successful countries they just have to copy.

> Doesn't work that well if people move to other countries to change them to become like their original country.

You can make the exact same argument against 'states as laboratories'.

That seems like a non sequitur. If each country were left to decide on its own, it seems that many, in fact historically nearly all, would not have open boarders. It seems like you're assuming the freedom of a group to govern themselves would result in them choosing what you personally find ideal. The world as it is now is the null hypothesis. People choose to form countries, and countries choose to have boarders.

If you want a sci-fi description of something like this, consider Too Like the Lightning by Ada Palmer. Enjoyed her take on “hives” which form as groups of people choose how to govern themselves and are relatively free to join. Some choose autocracy, others full democracy.

> to experiment with which ideas work, which don't

And this only works if people care about the results of the so-called experiments and use them to improve things. The GOP doesn't seem to care that trickle-down economics has never worked, and yet they still keep implementing it:

* https://en.wikipedia.org/wiki/Trickle-down_economics

Most recently Trump tax cuts and the Kansas experiment:

* https://en.wikipedia.org/wiki/Kansas_experiment

See also education cuts (e.g., Oklahoma) and economic growth:

* https://www.investopedia.com/articles/economics/09/education... * https://en.wikipedia.org/wiki/Education_economics#Returns_on...

No, it is not “almost as if” that at all because the mode of that interaction determines everything. You can have all the things you describe and still fail. India’s system, for instance, is modeled off the US in part and it has different problems. Other governments are modeled differently and work well. This is a facile statement that is simply appealing to national pride.

I agree but I'm not quite sure what you mean with your last statement. Lacking knowledge of "perfect" size for cities (and countries) is one thing, and iirc there are people studying just that (looking at stability, self-reported happiness of citizens, crime, mental health etc), but what technology would we need to implement the results? "Let's split this country into five smaller ones" doesn't sound like a technological problem ;)

Reply on news.ycombinator.com