6 ms·

Most of the management of government services could be automated, direct democracy is possible with software. There's certainly a lot of potential for AI augmen

Most of the management of government services could be automated, direct democracy is possible with software. There's certainly a lot of potential for AI augmentation in day-to-day governance. It certainly wouldn't be perfect at first. I'm sure some people would suffer from the mistakes made in the prototyping phase. But this isn't about Brexit. The question I believe will be answered is would someone living in an already failed state suffer worse with an augmented or fully automated administrator, or under current horror of war, starvation, fear? Civil wars start over ideological differences, political failures are the result of corruptions of government officials, power plays of individuals, mistakes of management. Nations split and crumble against the best interest of the people because leaders have human agendas. The cold impartiality of software takes the politics out of governance, it favors sustainability over the long term. We in the tech world see things from a position of privilege, protected by relative wealth, a large seemingly stable system of governance, but in many places in the world a new system of administration would be a gift to cherish.

kbenson 10y ago
The problem with immediate and direct democratic representation is that humans have many mental foibles due to our evolution which can combine into feedback loops which have disastrous consequences (lynch mobs are inherently democratic in nature). Non-representation or representation that is delayed too long is problematic because it disenfranchises people and they eventually resort to violence to reassert their representation. These are the extremes we need to walk between, Democracy and the Dictator. If you think just allowing people to vote directly will somehow circumvent the natural tendency of people for herd mentality, confirmation bias, and valuing anecdotes and personal experience over statistics, then I think you're vastly underestimating those problems. The role of technology is to help is increase the positive aspects of whatever point along that spectrum we've settled, and decrease the negative aspects. For example, making it very easy for representatives to quickly and accurately poll their constituents (and making these polls and their results public) rather than relying on lobbying and the varying abilities of special interest groups to activate advocates would probably yield much better representation for the average person. A very vocal 2% should not drown out the other 98%, but they should not necessarily be ingored either. Good polls would likely have a question and a scale of how much you strongly agree, somewhat agree, abstain, somewhat disagree, strongly disagree, and have the question asked at least 2-3 different ways, from different perspectives.
- rwhitman 10y ago
  >> If you think just allowing people to vote directly will somehow circumvent the natural tendency of people for herd mentality, confirmation bias, and valuing anecdotes and personal experience over statistics, then I think you're vastly underestimating those problems. That's why software makes it possible. Large scale polling systems are well understood at this point. Abuse detection is a solved problem, correcting for bias is well understood too. Giving people access to information, while at the poll, will increase the number of truly informed constituents when voting. This doesn't completely reinforce my point, but if you're interested in the topic it's worth the listen: http://www.flashforwardpod.com/2016/04/19/episode-11-swipe-right-for-democracy/ http://www.flashforwardpod.com/2016/04/19/episode-11-swipe-r...
  - kbenson 10y ago
    > Abuse detection is a solved problem, correcting for bias is well understood too. What about the bias introduced from the initiative name and/or description?[1] Who decides this? Who writes the laws that are proposed by initiatives? What if there's no counter side to a law proposed, or it's very complex, and hard to understand? Right now we have people whose job it is to keep track of these laws, make sure small provisions don't overstep, and come to an agreement on what should be acceptable to vote one. We elect those people (or we elect the people that employ and control those people). Who does this in a direct democracy? Do we still elect officials, and they do everything except they don't vote? > Giving people access to information, while at the poll, will increase the number of truly informed constituents when voting. If we have a direct democracy, there will be a lot more initiatives. 15 things every day, if we are keeping up with congress, according to to the flashforward link you supplied (thanks, I'm listening to it in the background right now). Looking up all the relevant information for each initiative to make an informed decision will take quite a while. Assuming a minimum of 15 minutes per item to research and consider each item, that's close to four hours, each day. Even halving that is a significant time investment. I think people will naturally defer to authorities they suggest. There's even less incentive for the random authorities people might choose to be accountable for their speech, and I'm also quite worried that people would choose non-qualified people for their queues on what to vote for very poor reasons (pick your Kardashian or their relation). If we want to defer to someone for our vote, why not our elected official? What I think would be interesting, was a way to wrest my vote from my representative on a by-vote basis. That is, we all get a chance to have a direct vote, but if we don't, our share goes to our representative. For example, if I'm in CA, and my senators are voting on a bill, and I vote specifically a certain way, then their votes count for (CA vote eligible population -1)(CA vote eligible population) each, and my vote counts for 2/(CA vote eligible population), one for each senator, and assuming nobody else votes directly. For important issues, we can wrest control from our legislators, but otherwise the system will perform as it has. I'm still not entirely sure it would be a good idea given our predisposition as a species for poor judgement in certain situations, but I think it's a lot more feasible than a direct democracy on everything (and it keeps people around to design and present bills, and still incentivizes them to do so). 1: http://www.spokesman.com/stories/2009/dec/10/states-ballot-language-rules-worth-imitating/ http://www.spokesman.com/stories/2009/dec/10/states-ballot-l...

kbenson 10y ago

The problem with immediate and direct democratic representation is that humans have many mental foibles due to our evolution which can combine into feedback loops which have disastrous consequences (lynch mobs are inherently democratic in nature). Non-representation or representation that is delayed too long is problematic because it disenfranchises people and they eventually resort to violence to reassert their representation. These are the extremes we need to walk between, Democracy and the Dictator. If you think just allowing people to vote directly will somehow circumvent the natural tendency of people for herd mentality, confirmation bias, and valuing anecdotes and personal experience over statistics, then I think you're vastly underestimating those problems. The role of technology is to help is increase the positive aspects of whatever point along that spectrum we've settled, and decrease the negative aspects. For example, making it very easy for representatives to quickly and accurately poll their constituents (and making these polls and their results public) rather than relying on lobbying and the varying abilities of special interest groups to activate advocates would probably yield much better representation for the average person. A very vocal 2% should not drown out the other 98%, but they should not necessarily be ingored either. Good polls would likely have a question and a scale of how much you strongly agree, somewhat agree, abstain, somewhat disagree, strongly disagree, and have the question asked at least 2-3 different ways, from different perspectives.

rwhitman 10y ago
>> If you think just allowing people to vote directly will somehow circumvent the natural tendency of people for herd mentality, confirmation bias, and valuing anecdotes and personal experience over statistics, then I think you're vastly underestimating those problems. That's why software makes it possible. Large scale polling systems are well understood at this point. Abuse detection is a solved problem, correcting for bias is well understood too. Giving people access to information, while at the poll, will increase the number of truly informed constituents when voting. This doesn't completely reinforce my point, but if you're interested in the topic it's worth the listen: http://www.flashforwardpod.com/2016/04/19/episode-11-swipe-right-for-democracy/ http://www.flashforwardpod.com/2016/04/19/episode-11-swipe-r...
- kbenson 10y ago
  > Abuse detection is a solved problem, correcting for bias is well understood too. What about the bias introduced from the initiative name and/or description?[1] Who decides this? Who writes the laws that are proposed by initiatives? What if there's no counter side to a law proposed, or it's very complex, and hard to understand? Right now we have people whose job it is to keep track of these laws, make sure small provisions don't overstep, and come to an agreement on what should be acceptable to vote one. We elect those people (or we elect the people that employ and control those people). Who does this in a direct democracy? Do we still elect officials, and they do everything except they don't vote? > Giving people access to information, while at the poll, will increase the number of truly informed constituents when voting. If we have a direct democracy, there will be a lot more initiatives. 15 things every day, if we are keeping up with congress, according to to the flashforward link you supplied (thanks, I'm listening to it in the background right now). Looking up all the relevant information for each initiative to make an informed decision will take quite a while. Assuming a minimum of 15 minutes per item to research and consider each item, that's close to four hours, each day. Even halving that is a significant time investment. I think people will naturally defer to authorities they suggest. There's even less incentive for the random authorities people might choose to be accountable for their speech, and I'm also quite worried that people would choose non-qualified people for their queues on what to vote for very poor reasons (pick your Kardashian or their relation). If we want to defer to someone for our vote, why not our elected official? What I think would be interesting, was a way to wrest my vote from my representative on a by-vote basis. That is, we all get a chance to have a direct vote, but if we don't, our share goes to our representative. For example, if I'm in CA, and my senators are voting on a bill, and I vote specifically a certain way, then their votes count for (CA vote eligible population -1)(CA vote eligible population) each, and my vote counts for 2/(CA vote eligible population), one for each senator, and assuming nobody else votes directly. For important issues, we can wrest control from our legislators, but otherwise the system will perform as it has. I'm still not entirely sure it would be a good idea given our predisposition as a species for poor judgement in certain situations, but I think it's a lot more feasible than a direct democracy on everything (and it keeps people around to design and present bills, and still incentivizes them to do so). 1: http://www.spokesman.com/stories/2009/dec/10/states-ballot-language-rules-worth-imitating/ http://www.spokesman.com/stories/2009/dec/10/states-ballot-l...