1,591 karma · joined April 26, 2020
When I want finical advice I hire an expert that gets paid the same no matter where I invest my money, because his incentive is to give me the best advice so I stick with him. If I get finical advice from someone that gets commission based on selling specific investments then the advice is more likely to be less than the best.
1. It’s very hard to shop around to find better prices or better service.
2. People often make choices like “Do you want this expensive life saving procedure or not?”
3. The average person isn’t informed enough to understand what their medical treatment needs are. Patients ask doctors what their recommendation is. Doctors often have an incentive to sell the most healthcare they reasonably can.
4. My ability to pay for healthcare is directly tied to my employer. There simply are less jobs that provide good healthcare than people that would like them. My ability to change positions in society is limited by my desire to continue to have access to affordable healthcare. This is practically a freedom limiting mechanism.
5. I’m limited in my choice of which healthcare to purchase based on my Insurance policy. My choice in insurance policy is similar to a suburbanite’s choice in an ISP. I have very few choices.
If resolving the adversarial relationship between citizens and police would improve that then it’s something we should explore. Right now I’m not sure how that could happen, given the deep problems with American Law Enforcement, Correctional, and Justice systems.
The media plays a critical role in encouraging people to get a vaccine.
If effective moderation can occur at smaller levels, like discord channels or subreddits, then those communities won’t been to be removed by the larger platform. This would also be helpful for startup social media platforms that have yet to bring enough revenue to afford a facebook sized moderation team.
Technically speaking it can do things like:
* Flag posts that contain a blacklist of words, including non-obvious spellings of said word (using non-standard Unicode characters in place of letters)
* Cross reference IP addresses or user names with banned users in other communities
* Notify moderators of trending slogans, phrases, or hashtags that have non-obvious extremist roots.
* Identify images containing extremist/hateful content
* Flag content that contains any political discussion for communities that want to be completely apolitical.
* Flag pornography
EDIT: To be clear, the target audience for this would be community moderators/admins or startup social networks that haven’t built their own moderation infrastructure.
Yeah, an opiate addiction now is ~10x the cost of prerolled cigarettes addiction. It it was closer to 1x then the property crimes and desperation associated with it would be significantly decreased. In the book Cherry by Nico Walker the author writes about his experience of robbing banks in order to fund he and his wife’s heroin addiction.
Being around a group of people is easily 20x more stimulating to me than any VR experience I’ve had, and I’ve experience the state of the art in VR. Spending day after day, year after year, going to classes with the same group of peers was very socially validating and helped me be more engaged with my time at school.
This only works if all governments agree on this rule. AI is an economic competitive advantage, and it’s in every countries best interest that advances in AI are developed in their borders.
If we had generic facilities that could produce any mRNA vaccines, and an mRNA vaccine approval process that takes 30 days then we could have had covid vaccines ready to go before March 2020, and those whole story looks very different.
I have an idea of creating a plugin to a decentralized social network that acts as a moderator. The plugin will work on behalf of the user, compared to current moderation which acts on behalf of the platform. Different users can select either different moderators or different settings for the same moderators. Moderating could be a mix of AI and human flagging. Moderating could also include more editorial work such as suggesting people to follow.
This tactic would best be against a force that doesn’t know about it, or an organized force.
A robotic tape library might fit that definition, and it might not.