5 ms·

[7] You can tell me Heinlein wrote island and I could believe it but I perhaps shouldnt [-1,4] imagine, as a machine, your opponent knowing (suspecting?) you c

[7] You can tell me Heinlein wrote island and I could believe it but I perhaps shouldnt [-1,4] imagine, as a machine, your opponent knowing (suspecting?) you can watch him from behind (or from inside his mind, more realistically/usefully, but from behind could be terrifying) [-4, maybe -3] tech debt is not entropy, but sophistication?

082349872349872 2y ago
Huxley and Heinlein were born only 13 years apart, so now, over a century later, maybe we can lump them into, if not the same generation, at least the same Zeitgeist. Somewhere I'd read about a fighting-game AI which had access to all of the human player data (so not the spirit, but knowing the joystick inputs would be a good proxy for "pwn the technique") and was unbeatable, so the programmers nerfed it by giving it only a noisy representation of the human player's avatar's outputs, at which point it returned to satisfyingly mortal. Doesn't 三体 touch upon (sophons/wallfacers) a watch-from-inside/behind opponent? Tech debt as sophistication would explain its sign flip over time! It also gives a metaphor perhaps better than "paying off tech debt": it's good if a program is a bit witty, but bad when it starts smelling of the lamp? > "In the long run, every program becomes rococo, and then rubble." —AJP EDIT: Re: [-1,4] Just as implication allows only 3 possibilities out of 4, my wife likes to say (see also https://news.ycombinator.com/item?id=42329631 https://news.ycombinator.com/item?id=42329631 ) that when riding, there should be 2 bodies but 3 minds: what you know, what the horse knows you know, and what you know the horse knows. The rider should have an accurate model of the horse's attention and intentions, but not the other way around. (in particular, you might be very worried about an emergency situation, but the horse should never realise this) I believe this habit helped her with her fencing, because there also we are attempting to hide some aspects of our attention and intentions from someone who is very closely monitoring our posture, in order to discern them. EDIT2: just as a prelude to a comment on distance: napoleonic sabre cavalry was terrified of an opponent coming up from behind left (strong to their weak) — that's why all the dashing portraits with jackets/cloaks "jauntily" thrown over the left shoulder; ultimately the fashion was very practical, both for defence and for impressing the ladies. Lermontov's karakul nabadi is a particular instance. Налетели ветры злые. Да с восточной стороны И сорвали тёмну шапку Ой да с моей буйной головы. Lagniappe: https://www.youtube.com/watch?v=urjAwPfFPqU https://www.youtube.com/watch?v=urjAwPfFPqU
- gsf_emergency 2y ago
  The setup of opposing forces leads one to think in terms of entropy ("the fog of war", the manifold of possible mistakes, etc etc).. Then there's that story about the break even point between French and Mamluks iirc? (And what was the break even point between LGA and the Coalition?) Somewhere in between we might characterize tech debt; rococo doesn't turn to rubble overnight, & most of it is still around. Or maybe there are some relatively cheap ways to mitigate tech debt. Always backpropagating debugs to spec, for instance. So we could kick the (meta)grokking/turning point down the road https://www.science.org/content/blog-post/nassim-taleb-scientific-discovery https://www.science.org/content/blog-post/nassim-taleb-scien... >There's no built-in payoff for complexity (or for making things complex). "In academia," though, he says, "there is". [Make sure your phonebooth expands at around the rate that the range of your (melee) weapons/number of participants increase] And those promised tubes are annoyingly hard to find http://wavefunction.fieldofscience.com/2015/02/derek-lowe-to-world-beware-of-von.html?m=1 http://wavefunction.fieldofscience.com/2015/02/derek-lowe-to... Thankfully, Derek has somewhat identified the opponent as, not other scientists, but biology/evolution..
  - 082349872349872 2y ago
    > There's no built-in payoff for complexity (or for making things complex). Says the hideously (when compared to single-celled organisms) complicated h sap. Then again, I think social darwinists ought to admit that it's the single-celled organisms who have the fitness they claim to be looking for?
    - gsf_emergency 2y ago
      Vonnegut has admitted it?
- gsf_emergency 2y ago
  To riff on your wife's grokking, the number of effective participants increases faster than what one would usually expect: there's what you know,what you know what other academics know, and what other academics know you know Ditto for "industry" but there the default data structure (nominally) is ____, so perhaps sophistication is monotonic (not accounting yet for immigration, oligarchal or otherwise psychopathic/imbecilic, modelled as using weapons of uncertain range) https://news.ycombinator.com/item?id=42453789 https://news.ycombinator.com/item?id=42453789

082349872349872 2y ago

Huxley and Heinlein were born only 13 years apart, so now, over a century later, maybe we can lump them into, if not the same generation, at least the same Zeitgeist. Somewhere I'd read about a fighting-game AI which had access to all of the human player data (so not the spirit, but knowing the joystick inputs would be a good proxy for "pwn the technique") and was unbeatable, so the programmers nerfed it by giving it only a noisy representation of the human player's avatar's outputs, at which point it returned to satisfyingly mortal. Doesn't 三体 touch upon (sophons/wallfacers) a watch-from-inside/behind opponent? Tech debt as sophistication would explain its sign flip over time! It also gives a metaphor perhaps better than "paying off tech debt": it's good if a program is a bit witty, but bad when it starts smelling of the lamp? > "In the long run, every program becomes rococo, and then rubble." —AJP EDIT: Re: [-1,4] Just as implication allows only 3 possibilities out of 4, my wife likes to say (see also https://news.ycombinator.com/item?id=42329631 https://news.ycombinator.com/item?id=42329631 ) that when riding, there should be 2 bodies but 3 minds: what you know, what the horse knows you know, and what you know the horse knows. The rider should have an accurate model of the horse's attention and intentions, but not the other way around. (in particular, you might be very worried about an emergency situation, but the horse should never realise this) I believe this habit helped her with her fencing, because there also we are attempting to hide some aspects of our attention and intentions from someone who is very closely monitoring our posture, in order to discern them. EDIT2: just as a prelude to a comment on distance: napoleonic sabre cavalry was terrified of an opponent coming up from behind left (strong to their weak) — that's why all the dashing portraits with jackets/cloaks "jauntily" thrown over the left shoulder; ultimately the fashion was very practical, both for defence and for impressing the ladies. Lermontov's karakul nabadi is a particular instance. Налетели ветры злые. Да с восточной стороны И сорвали тёмну шапку Ой да с моей буйной головы. Lagniappe: https://www.youtube.com/watch?v=urjAwPfFPqU https://www.youtube.com/watch?v=urjAwPfFPqU

gsf_emergency 2y ago
The setup of opposing forces leads one to think in terms of entropy ("the fog of war", the manifold of possible mistakes, etc etc).. Then there's that story about the break even point between French and Mamluks iirc? (And what was the break even point between LGA and the Coalition?) Somewhere in between we might characterize tech debt; rococo doesn't turn to rubble overnight, & most of it is still around. Or maybe there are some relatively cheap ways to mitigate tech debt. Always backpropagating debugs to spec, for instance. So we could kick the (meta)grokking/turning point down the road https://www.science.org/content/blog-post/nassim-taleb-scientific-discovery https://www.science.org/content/blog-post/nassim-taleb-scien... >There's no built-in payoff for complexity (or for making things complex). "In academia," though, he says, "there is". [Make sure your phonebooth expands at around the rate that the range of your (melee) weapons/number of participants increase] And those promised tubes are annoyingly hard to find http://wavefunction.fieldofscience.com/2015/02/derek-lowe-to-world-beware-of-von.html?m=1 http://wavefunction.fieldofscience.com/2015/02/derek-lowe-to... Thankfully, Derek has somewhat identified the opponent as, not other scientists, but biology/evolution..
- 082349872349872 2y ago
  > There's no built-in payoff for complexity (or for making things complex). Says the hideously (when compared to single-celled organisms) complicated h sap. Then again, I think social darwinists ought to admit that it's the single-celled organisms who have the fitness they claim to be looking for?
  - gsf_emergency 2y ago
    Vonnegut has admitted it?
gsf_emergency 2y ago
To riff on your wife's grokking, the number of effective participants increases faster than what one would usually expect: there's what you know,what you know what other academics know, and what other academics know you know Ditto for "industry" but there the default data structure (nominally) is ____, so perhaps sophistication is monotonic (not accounting yet for immigration, oligarchal or otherwise psychopathic/imbecilic, modelled as using weapons of uncertain range) https://news.ycombinator.com/item?id=42453789 https://news.ycombinator.com/item?id=42453789