• 0 Posts
  • 154 Comments
Joined 1 year ago
cake
Cake day: May 7th, 2024

help-circle

  • That’s war. That has been the nature of war and deterrence policy ever since industrial manufacture has escalated both the scale of deployments and the cost and destructive power of weaponry. Make it too expensive for the other side to continue fighting (or, in the case of deterrence, to even attack in the first place). If the payoff for scraping no longer justifies the investment of power and processing time, maybe the smaller ones will give up and leave you in peace.



  • When referencing another person’s comment, it can be helpful to link to that comment or the article you mentioned.

    I’d also like to point out that many Wikipedia articles, particularly those written by experts on a given scientific subject, tend to be daunting rather than helpful for people not yet familiar with that subject.

    Explanations like the one you offered in this comment and the next reply can help make topics more approachable, so I very much appreciate that.

    To illustrate my point:

    In this case, the article first describes the principle as “pertaining to a lower theoretical limit of energy consumption of computation”, which doesn’t directly highlight the connection to information storage. The next sentence then mentions “irreversible change in information” and “merging two computational paths”, both of which are non-trivial.

    From a brief glance at the article on reversible computing linked further on, I gather that “irreversible” here doesn’t mean “you can’t flip the bit again” but rather something like “you can’t deterministically figure out the previous calculation from its result”, so the phrase boils down to “storing a piece of information” for our context. The example of “merging computational paths” probably has no particular bearing on the energy value of information either and can be ignored as well.

    Figuring out the resulting logic that you so kindly laid out – again, thank you for that! – requires a degree of subject-specific understanding to know what parts of the explanation can be safely ignored.

    Of course, experts want to be accurate and tend to think in terms they’re familiar with, so I don’t fault them for that. The unfortunate result is that their writings are often rather intransparent to laypeople and linking to Wikipedia articles isn’t always the best way to convey an understanding.






  • I’m autistic, which results in me deconstructing and analysing jokes instead of laughing (often to the displeasure of the people who think I didn’t find their joke funny – I promise, if I’m taking the time to disassemble your joke that means I found it funny and want to understand why).

    The flipside is that I occasionally crack out carefully engineered bangers, because I understand the importance of a setup, building expectations and putting the brain on one track of thought, then capping it off with the “derailing” of those expectations. The shorter you can get it, the less time the brain has to get off track on its own, diminishing that derailing effect.

    Of course, getting the inspiration and figuring out a way to put that into practice is it’s own unpredictable beast, and some jokes just fall flat despite my effort. Sometimes I misread the room or the audience too. I’m not a particularly talented comedian.

    But at least I’m not a setup without a punchline.



  • A definitional concession to make exponential series work. xn for n ∈ (0, 1) is the nth root of x, which gets ever closer to 1, while x^n for n < 0 equals 1÷ (xn). Between them lies the neutral element with respect to multiplication 1 (neutral meaning that x × 1 = x; a factor of one doesn’t actually change anything). Hence, x0 = 1.

    That rule breaks down for x = 0, obviously. Negative exponents don’t work at all because they’re division by zero, while all exponents > 0 result in 0. Semantically, 00 probably should be undefined, but the neutral element rule does provide a definition. There also isn’t really any reasonable use case where you’d need that to be consistent with anything else.










  • Short version: If we’re talking national level (that is, electoral votes), then Congress elects the president (House for President, Senate for VP).

    If we’re talking state level however, for most states the 34% will win and take all of the state’s electoral votes.

    This is the cornerstone of the two-party system, which emerges naturally as a consequence of plurality voting systems. If you have two left-wing parties, one of which gets 10% and the other 42%, they both loose to the 48% of the single right-wing party. Hence, it’s strategic for the left wing to unite, which would theoretically earn them 52% of votes (practically, voter disillusionment makes it more complicated).

    This is called the Spoiler Effect: A left-wing party would end up splitting votes off the Democrats, leading to a plurality victory for the Republicans. And in winner-takes-all systems, that plurality is enough to get the respective state’s electoral votes.