3 ms·

> It's absurd that such laws would need to be passed 50 times for all US citizens to benefit from it. It should be done at the federal level. You're missing an

> It's absurd that such laws would need to be passed 50 times for all US citizens to benefit from it. It should be done at the federal level. You're missing an important reason why regulating at the state level first is a good idea: because it allows you to test the implementation with a small fraction of citizens before rolling it out. Yes, basically everyone wants click-to-cancel, but actually writing good regulation is hard. Ideally, what would happen is that a few states would try things in different ways, and then when they figure out the best implementation, the federal government would pick up that implementation.

nkrisc 1y ago
That’s fair, that is a good point. But still in many cases I think a less-than-perfect law is better than none.
- throw10920 1y ago
  I agree! But it's actually easier to get a law passed at the state level than at the federal level. As you can see, congress has a hard time passing meaningful laws right now, and from people I know who work there, it's largely because there's too much stuff for them to do - they simply don't have enough bandwidth. At the state level, you have fewer signatures you need to get on petitions, you represent a larger fraction of the constituents, your representatives are more sensitive to your demands, etc. From a greedy/selfish perspective, having the states prototype laws before the federal government effectively offloads some of the regulatory burden onto the states. And, when the law works, the other states tends to notice - you get political momentum.
edmundsauto 1y ago
The problem here is that it’s testing the implementation details more than the generalized idea. It’s as if I wrote code to process data in a certain way, write it for an old mainframe and to process a specific set of data. There’s not a ton of generalizability to other data, and how you implement the code on other systems will impact the outcome. Especially because there are few objective measurements to evaluate the success of legislation
- throw10920 1y ago
  > The problem here is that it’s testing the implementation details more than the generalized idea. Well, the problem is that you can't test the generalized idea either. Even if a law is passed at the federal level, you're still only testing a specific implementation of a regulatory concept you want to implement. Having a bunch of entities (the states) try implementing the same concept in different ways allows you to explore more of the solution space than if you only have a single entity (the federal government) do it uniformly upfront!
  - edmundsauto 1y ago
    Yes, although this is expensive and harmful, there is little reason to think the bad experiments would adopt the winner, or that we can even measure the best outcome. Plus, for a lot of stuff - we don’t need experiments, we know what works! As an example, is there any reason to think we need to do experiments on whether children are fed at school for free?
    - throw10920 1y ago
      > expensive Citation needed > harmful Citation needed > little reason to think the bad experiments would adopt the winner What does this mean? Not a coherent sentence. > that we can even measure the best outcome So, exactly the same as when it's done at the federal level. > for a lot of stuff - we don’t need experiments, we know what works In terms of legislation? Factually incorrect. Legislation/regulation is extremely difficult to get right and it's incredibly rare that there's precedent that is universally-agreed-upon to be beneficial in general, let alone when the states don't try their hand first. > As an example, is there any reason to think we need to do experiments on whether children are fed at school for free? Again, what does this mean? This isn't a coherent sentence either.
      - edmundsauto 1y ago
        This is just not an appropriate response, but I'm happy you feel like you made your point. You can chalk up another internet point for p0wnage of someone you'll never meet. Was it your intent to shut down a conversation?

nkrisc 1y ago

That’s fair, that is a good point. But still in many cases I think a less-than-perfect law is better than none.

throw10920 1y ago
I agree! But it's actually easier to get a law passed at the state level than at the federal level. As you can see, congress has a hard time passing meaningful laws right now, and from people I know who work there, it's largely because there's too much stuff for them to do - they simply don't have enough bandwidth. At the state level, you have fewer signatures you need to get on petitions, you represent a larger fraction of the constituents, your representatives are more sensitive to your demands, etc. From a greedy/selfish perspective, having the states prototype laws before the federal government effectively offloads some of the regulatory burden onto the states. And, when the law works, the other states tends to notice - you get political momentum.

edmundsauto 1y ago

The problem here is that it’s testing the implementation details more than the generalized idea. It’s as if I wrote code to process data in a certain way, write it for an old mainframe and to process a specific set of data. There’s not a ton of generalizability to other data, and how you implement the code on other systems will impact the outcome. Especially because there are few objective measurements to evaluate the success of legislation

throw10920 1y ago
> The problem here is that it’s testing the implementation details more than the generalized idea. Well, the problem is that you can't test the generalized idea either. Even if a law is passed at the federal level, you're still only testing a specific implementation of a regulatory concept you want to implement. Having a bunch of entities (the states) try implementing the same concept in different ways allows you to explore more of the solution space than if you only have a single entity (the federal government) do it uniformly upfront!
- edmundsauto 1y ago
  Yes, although this is expensive and harmful, there is little reason to think the bad experiments would adopt the winner, or that we can even measure the best outcome. Plus, for a lot of stuff - we don’t need experiments, we know what works! As an example, is there any reason to think we need to do experiments on whether children are fed at school for free?
  - throw10920 1y ago
    > expensive Citation needed > harmful Citation needed > little reason to think the bad experiments would adopt the winner What does this mean? Not a coherent sentence. > that we can even measure the best outcome So, exactly the same as when it's done at the federal level. > for a lot of stuff - we don’t need experiments, we know what works In terms of legislation? Factually incorrect. Legislation/regulation is extremely difficult to get right and it's incredibly rare that there's precedent that is universally-agreed-upon to be beneficial in general, let alone when the states don't try their hand first. > As an example, is there any reason to think we need to do experiments on whether children are fed at school for free? Again, what does this mean? This isn't a coherent sentence either.
    - edmundsauto 1y ago
      This is just not an appropriate response, but I'm happy you feel like you made your point. You can chalk up another internet point for p0wnage of someone you'll never meet. Was it your intent to shut down a conversation?