5 ms·

> but in an antagonistic situation, one needs to have a high percentage of h/w-served hits. Can you explain "h/w-served hits"? (I am guessing other than consci

> but in an antagonistic situation, one needs to have a high percentage of h/w-served hits. Can you explain "h/w-served hits"? (I am guessing other than conscious?)

vinnyvichy 2y ago
I'm guessing it's about the general rule of thumb that hw is faster than sw, and so can react to urgent situations in a timely fashion. As for why sw might be shorthand for consciousness: sw can modify itself (and thus react to strategic issues) ... so yeah, hw is more or less defined as something that cannot modify itself.
- mistermann 2y ago
  Firmware is exceptional though.
  - 082349872349872 2y ago
    Fair enough, and we could get into microcoded architectures too. But basically the analogy (originally built around TLB handling) is supposed to be: software slow & flexible conscious response hardware fast & inflexible reflex response We're altricial, not precocial, so over our lifetimes we can (and hopefully do) alter many of our own "reflex" behaviours. In particular, details of the 4 F's are firmware configurable. Does that make sense? Lagniappe: https://www.youtube.com/watch?v=__KfrDZoBgQ https://www.youtube.com/watch?v=__KfrDZoBgQ (for mistermann)
    - mistermann 2y ago
      Yes but "4 F's"? Also, there's lots of potential for failure (and I'd say it is overwhelmingly observable that we have and do) in "can (and hopefully do)". Who knows what our absolute capabilities are, especially if we hardly or maybe even never try (at least certain things)? Writing the capabilities to firmware is a step after this, but only after the first step is accomplished.
      - vinnyvichy 2y ago
        Glad we moved the discussion into the Overton window of firmware. Maybe it's too early to start shipping metaphors in wetware, but developing general heuristics for productive+reproducible+survivable self-fuzzing? (*) "n X's" might still be a useful template to keep in mind, yes? Here are some good resources to explore the pain-gain frontier. (Or its baby versions, the supplies-surprise dilemma, shiprate quantification, etc.) [0-1] https://www.nature.com/articles/s41567-024-02525-w https://www.nature.com/articles/s41567-024-02525-w https://arxiv.org/abs/2309.15070 https://arxiv.org/abs/2309.15070 [2] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&query=Critical%20period&sort=byPopularity&type=story https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que... [0] caveat lector: "nature physics" is where economics that's almost certain to be rejected for very good proboscidean-tail-wiggling reasons would be dumped. (*) Going back to the "Blindsight tangent", the social-political plane might already be an ideal platform to begin quantitative experimentation. ("Bylaws as wetware")
        mistermann 2y ago
        > Maybe it's too early to start shipping metaphors in wetware I have some bad news for you, we've been at it for a long time. > but developing general heuristics for Oh God please no, we have too many already. I'm also generally speaking not a fan of apply principles from physics to metaphysics. I suspect I'm mostly misunderstanding you though, so take the above accordingly. (HN search can uncover/surface some very interesting phenomena, it is a very underappreciated tool.)
        vinnyvichy 2y ago
        >Who knows what our absolute capabilities are, especially if we hardly or maybe even never try (at least certain things)? Ego quoque, for misunderstanding what you meant by "certain things" (>) I stress that I manifestly do not want to see physicists and tech bros publish bio-meta-how-tos just yet, did you mean high quality professionally conducted BSL training? As for 082349872349872, I think exploring the Taoist metaphysics of marginally injurious mutual surprise as an epsilon over Buddhist tainted reinterpretations could be at most a waste of time. And what's more, I can vouch that he's not a physicist, not even vicariously. >though an ideology taking root in the mind is necessary to exploit its capabilities to their maximum. https://hn.algolia.com/?dateRange=all&page=0&prefix=true&query=Mistermann%20capabilities&sort=byDate&type=comment https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que... Okay so give me an example of an interesting ideology, as opposed to merely a conspiracy theory, such as "applying skepticism to your own beliefs" ) I could hit the News, but I am interested in how you think. E.g. did I misunderstand that you are interested in locating the precise quantitative boundaries of Systems I vs II? https://news.ycombinator.com/item?id=27489329 https://news.ycombinator.com/item?id=27489329

> but in an antagonistic situation, one needs to have a high percentage of h/w-served hits. Can you explain "h/w-served hits"? (I am guessing other than conscious?)

vinnyvichy 2y ago

I'm guessing it's about the general rule of thumb that hw is faster than sw, and so can react to urgent situations in a timely fashion. As for why sw might be shorthand for consciousness: sw can modify itself (and thus react to strategic issues) ... so yeah, hw is more or less defined as something that cannot modify itself.

mistermann 2y ago
Firmware is exceptional though.
- 082349872349872 2y ago
  Fair enough, and we could get into microcoded architectures too. But basically the analogy (originally built around TLB handling) is supposed to be: software slow & flexible conscious response hardware fast & inflexible reflex response We're altricial, not precocial, so over our lifetimes we can (and hopefully do) alter many of our own "reflex" behaviours. In particular, details of the 4 F's are firmware configurable. Does that make sense? Lagniappe: https://www.youtube.com/watch?v=__KfrDZoBgQ https://www.youtube.com/watch?v=__KfrDZoBgQ (for mistermann)
  - mistermann 2y ago
    Yes but "4 F's"? Also, there's lots of potential for failure (and I'd say it is overwhelmingly observable that we have and do) in "can (and hopefully do)". Who knows what our absolute capabilities are, especially if we hardly or maybe even never try (at least certain things)? Writing the capabilities to firmware is a step after this, but only after the first step is accomplished.
    - vinnyvichy 2y ago
      Glad we moved the discussion into the Overton window of firmware. Maybe it's too early to start shipping metaphors in wetware, but developing general heuristics for productive+reproducible+survivable self-fuzzing? (*) "n X's" might still be a useful template to keep in mind, yes? Here are some good resources to explore the pain-gain frontier. (Or its baby versions, the supplies-surprise dilemma, shiprate quantification, etc.) [0-1] https://www.nature.com/articles/s41567-024-02525-w https://www.nature.com/articles/s41567-024-02525-w https://arxiv.org/abs/2309.15070 https://arxiv.org/abs/2309.15070 [2] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&query=Critical%20period&sort=byPopularity&type=story https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que... [0] caveat lector: "nature physics" is where economics that's almost certain to be rejected for very good proboscidean-tail-wiggling reasons would be dumped. (*) Going back to the "Blindsight tangent", the social-political plane might already be an ideal platform to begin quantitative experimentation. ("Bylaws as wetware")
      - mistermann 2y ago
        > Maybe it's too early to start shipping metaphors in wetware I have some bad news for you, we've been at it for a long time. > but developing general heuristics for Oh God please no, we have too many already. I'm also generally speaking not a fan of apply principles from physics to metaphysics. I suspect I'm mostly misunderstanding you though, so take the above accordingly. (HN search can uncover/surface some very interesting phenomena, it is a very underappreciated tool.)
        vinnyvichy 2y ago
        >Who knows what our absolute capabilities are, especially if we hardly or maybe even never try (at least certain things)? Ego quoque, for misunderstanding what you meant by "certain things" (>) I stress that I manifestly do not want to see physicists and tech bros publish bio-meta-how-tos just yet, did you mean high quality professionally conducted BSL training? As for 082349872349872, I think exploring the Taoist metaphysics of marginally injurious mutual surprise as an epsilon over Buddhist tainted reinterpretations could be at most a waste of time. And what's more, I can vouch that he's not a physicist, not even vicariously. >though an ideology taking root in the mind is necessary to exploit its capabilities to their maximum. https://hn.algolia.com/?dateRange=all&page=0&prefix=true&query=Mistermann%20capabilities&sort=byDate&type=comment https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que... Okay so give me an example of an interesting ideology, as opposed to merely a conspiracy theory, such as "applying skepticism to your own beliefs" ) I could hit the News, but I am interested in how you think. E.g. did I misunderstand that you are interested in locating the precise quantitative boundaries of Systems I vs II? https://news.ycombinator.com/item?id=27489329 https://news.ycombinator.com/item?id=27489329