3 ms·
As much as there's ways that computing could be (and is unevenly) used to increase retail-level government, there are good reasons to continue to include human
by shubber 16y ago
As much as there's ways that computing could be (and is unevenly) used to increase retail-level government, there are good reasons to continue to include human beings in the governing process.
A few notions to look into:
* Condorcet voting systems
* Mechanism Design Theory
* Parliamentary systems
Finally, consider strongly the problem of the influence of capital on any system. In the US, there's the issue of campaign finance, for instance. But if you were to program a government, the problem would be worse: you'd only have to buy off the development team once, rather than bribing members of government repeatedly over time.
- nrkn 16y agoYou could make the development process completely transparent, with input from a wide variety of people, checks and balances etc. and not begin implementing it without plenty of debate, examination etc. The bias is that those who are most interested in the process and/or the most vocal could end up being those with the most influence. But there's no reason not to at least start examining the problem from many angles and for as many people as possible to start trying to pick holes in it.
- shubber 16y agoBelieve it or not, that's the premise of a republican democracy. There's just more prose and less code involved.
- smokeyj 16y agoI suggest regulation not be monopolized, which mitigates the effects of a malicious regulator. Could you help me understand why maintaining a single point of failure is the only way to govern.
- shubber 16y agoThere's two responses that spring immediately to mind: First, "what time is it?" There's lots of cases where a single centralized authority can provide definite and concrete advantages. Second, ultimately, every system of authority (and note that I'm using the word in a very different sense now) is built on a foundation of a threat of death. And that kernel should be held at as much of a remove, as carefully controlled as is possible. Introducing "failover" regulators means that you now have multiple parallel columns of authority, and each basic threat of death needs it's own system of control. That strikes me as nightmarish.