5 ms·

The "scumbags" most likely know the cops at their best and worst. Rating a cop is relative as they provides a wide ranges of services. It's all relative.

by ct0 4y ago

daenz 4y ago
Which is why their rating is still counted, but its biased by what society thinks of the person doing the rating. Perfect shouldn't be the enemy of better here. If ratings are something people want to do, you have to bias ratings with additional context, otherwise there is no mechanism to help separate ratings from honest and dishonest people.
- throwaway787544 4y ago
  But this is also a premature optimization and potentially YAGNI. Yes, you want down-weighting because of one potential complication. But you haven't even built and operated this system yet, so you don't know if this workaround you're applying will even be necessary or will end up causing larger issues. One example of this is "voter fraud". It is possible to commit voter fraud, and so certain parties want strong identity guarantees to vote. But voter fraud isn't happening in practice, and stringent ID requirements actually suppress minority votes. It is better to implement the simplest possible system first, observe it, and fix problems later as needed, after careful analysis of the running system.
  - thfuran 4y ago
    >YAGNI But not the ordinary YAGNI. You're Absolutely going to need some way of dealing with low-quality input.
    - throwaway787544 4y ago
      Why? What's the worst thing that could happen from poor quality input? Unreliable scores? These reviews don't mean or do anything in the real world, and it's much less serious than an actual IA complaint. There's also no evidence that there would be enough poor quality inputs to make a meaningful impact on scores. You really should wait and see before you invest a ton of time and money into solving problems that end up not being problems.
      - thfuran 4y ago
        >Why? What's the worst thing that could happen from poor quality input? Unreliable scores? These reviews don't mean or do anything in the real world If you are actually operating under the assumption that the output doesn't matter at all, then implementing the review system in the first place is a waste of effort, let alone any features thereof. But assuming that we do care about the output, its quality does matter. >There's also no evidence that there would be enough poor quality inputs to make a meaningful impact on scores. In pretty much the same way that there's no evidence that any given new website needs to worry about sql injection.
- ryandrake 4y ago
  This encourages an undesirable outcome: Police now know that Group A's ratings will be more heavily weighted and Group B's ratings will not be heavily weighted. They will then be incentivized to treat Group A with kid gloves and mistreat Group B, regardless of what caused the police interaction. I'll leave it as an exercise to the reader to consider what the police might use as a proxy/heuristic to figure out what group a suspect is in.
  - daenz 4y ago
    I think you meant they would mistreat Group A more, if they are more heavily weighted, but I do see your point. I agree that this is a danger. Fortunately, with additional data available, data scientists will be able to show any biases more clearly. If it happens with a rating system, it was happening before the rating system, except now there is the benefit of more data. Similarly, you could also show if people from Group A (the more weighted group) also rate the police lower, despite similar treatment as those of Group B. The data will cut both ways. EDIT>> Nvm, you were using "weighted" to mean "the weight of their vote", not "the weight applied to their vote."

The "scumbags" most likely know the cops at their best and worst. Rating a cop is relative as they provides a wide ranges of services. It's all relative.

by ct0 4y ago

The "scumbags" most likely know the cops at their best and worst. Rating a cop is relative as they provides a wide ranges of services. It's all relative.

daenz 4y ago

Which is why their rating is still counted, but its biased by what society thinks of the person doing the rating. Perfect shouldn't be the enemy of better here. If ratings are something people want to do, you have to bias ratings with additional context, otherwise there is no mechanism to help separate ratings from honest and dishonest people.

throwaway787544 4y ago
But this is also a premature optimization and potentially YAGNI. Yes, you want down-weighting because of one potential complication. But you haven't even built and operated this system yet, so you don't know if this workaround you're applying will even be necessary or will end up causing larger issues. One example of this is "voter fraud". It is possible to commit voter fraud, and so certain parties want strong identity guarantees to vote. But voter fraud isn't happening in practice, and stringent ID requirements actually suppress minority votes. It is better to implement the simplest possible system first, observe it, and fix problems later as needed, after careful analysis of the running system.
- thfuran 4y ago
  >YAGNI But not the ordinary YAGNI. You're Absolutely going to need some way of dealing with low-quality input.
  - throwaway787544 4y ago
    Why? What's the worst thing that could happen from poor quality input? Unreliable scores? These reviews don't mean or do anything in the real world, and it's much less serious than an actual IA complaint. There's also no evidence that there would be enough poor quality inputs to make a meaningful impact on scores. You really should wait and see before you invest a ton of time and money into solving problems that end up not being problems.
    - thfuran 4y ago
      >Why? What's the worst thing that could happen from poor quality input? Unreliable scores? These reviews don't mean or do anything in the real world If you are actually operating under the assumption that the output doesn't matter at all, then implementing the review system in the first place is a waste of effort, let alone any features thereof. But assuming that we do care about the output, its quality does matter. >There's also no evidence that there would be enough poor quality inputs to make a meaningful impact on scores. In pretty much the same way that there's no evidence that any given new website needs to worry about sql injection.
ryandrake 4y ago
This encourages an undesirable outcome: Police now know that Group A's ratings will be more heavily weighted and Group B's ratings will not be heavily weighted. They will then be incentivized to treat Group A with kid gloves and mistreat Group B, regardless of what caused the police interaction. I'll leave it as an exercise to the reader to consider what the police might use as a proxy/heuristic to figure out what group a suspect is in.
- daenz 4y ago
  I think you meant they would mistreat Group A more, if they are more heavily weighted, but I do see your point. I agree that this is a danger. Fortunately, with additional data available, data scientists will be able to show any biases more clearly. If it happens with a rating system, it was happening before the rating system, except now there is the benefit of more data. Similarly, you could also show if people from Group A (the more weighted group) also rate the police lower, despite similar treatment as those of Group B. The data will cut both ways. EDIT>> Nvm, you were using "weighted" to mean "the weight of their vote", not "the weight applied to their vote."