Login
You're viewing the functional.cafe public feed.
  • Nov 18, 2026, 10:06 PM

    It's like what if a century ago, a bunch of sex pests formed a conspiracy to teach people improper kerning techniques, and now it's just everywhere.

    💬 1🔄 1⭐ 0

Replies

  • Nov 18, 2026, 10:07 PM

    Anyway, once you learn to see the matrix, either in the linear algebra or typesetting sense, then you have been afflicted with the curse. It is too late. You understand, and that understanding will never leave you in peace.

    💬 1🔄 1⭐ 0
  • Nov 18, 2026, 10:13 PM

    If you are not cursed in this way, a few hints to help the afflicted in your life:

    • "Two places at once" is complete nonsense. (Much more accurate would be "place is the wrong way of describing this system.")
    • Schrödinger's Cat was intentional disinfo put forward by a serial rapist. Invoking it helps no one.
    • Entanglement isn't spooky, it just isn't. It's a kind of correlation, and while it's very cool, it's not spooky.

    💬 5🔄 6⭐ 2
  • Nov 18, 2026, 11:27 PM

    @xgranade So much of the mystery and incomprehensibility of QM melts away when you discard all the incorrect interpretations of things.

    I too have this curse. Fortunately my kerning curse is weak and only manifests in the worst cases, I'm sorry you must bear both full-force.

    💬 0🔄 1⭐ 1
  • Nov 18, 2026, 11:38 PM

    @xgranade the gist seems to be everything you hear about quantum physics is someone talking about how they don't understand it (either sincerely or pretending not to), never about how it actually works.

    💬 2🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • Nov 18, 2026, 11:43 PM

    @kevingranade @xgranade This has certainly frustrated my attempts to learn what quantum computers are!

    I hear some people who don't know better assert that they're just faster computers. But as a computer scientist I know that it can't be that simple, or noone would be interested in them.

    💬 2🔄 0⭐ 0
  • Nov 18, 2026, 11:44 PM

    @alcinnz @kevingranade Yeah, no, that's absolutely not the case. There's some problems for which quantum computers exhibit different asymptotic scaling such that for large enough instances of those problems, quantum computers necessarily have an advantage.

    But it absolutely isn't a constant.

    💬 0🔄 1⭐ 0
  • Nov 18, 2026, 11:49 PM

    @alcinnz @kevingranade @xgranade I would say they're not computers in the Church-Turing sense, but machines that look to be useful for running digital simulations of statistical phenomena that would be prohibitively slow to simulate with computation.

    💬 1🔄 1⭐ 0
  • Nov 18, 2026, 11:52 PM

    @alcinnz @kevingranade @xgranade And the formalism for QC leads to algorithms that, if realized, would be able to do very bad things like factoring large numbers in tractable time, but bridging the gap between mathematical fantasy land and physical reality is very very hard (and in my somewhat educated opinion, impossible).

    💬 2🔄 1⭐ 0
  • Nov 18, 2026, 11:55 PM

    @dalias @alcinnz @kevingranade

    > and in my somewhat educated opinion, impossible

    From someone who worked in QC for twenty years, and who invented some of the core things that an eventual QC would need, a universe in which QC is impossible would look very different from what we currently understand the universe to look like. That's not to say that QC is absolutely possible but it would in and of itself upend much of current understanding about physics were it to turn out that QC is impossible.

    💬 2🔄 1⭐ 0
  • Nov 18, 2026, 11:57 PM

    @dalias @alcinnz @kevingranade Anyway, quantum computers are provably computers in the sense of the *weak* Church–Turing thesis (that is, in terms of computability), but aren't identical to classical computers in the sense of the *strong* Church–Turing thesis (that is, in terms of complexity).

    💬 1🔄 1⭐ 0
  • Nov 19, 2026, 12:17 AM

    @xgranade @alcinnz @kevingranade What I mean is that a QC does something different than the Church-Turing sense of computation. It does not answer a rigorous logical yes/no condition, accept/reject candidate members of a language, etc. Its results are qubits that need to be measured. This is a very different sense of computation.

    💬 3🔄 1⭐ 0
  • 💬 0🔄 0⭐ 0
  • Nov 19, 2026, 12:24 AM

    @xgranade @alcinnz @kevingranade In some sense real physical computers are like this too. They are subject, albeit with extremely low probability, to flipped bits from existing in a physical reality, on top of having some degree of finitude. But at least in an idealized form they are equivalent. I don't see any way a QC is equivalent except possibly in a limit as you approach infinitely many measurements. OTOH a QC is much better at other things.

    💬 0🔄 1⭐ 0
  • 💬 0🔄 0⭐ 0
  • 💬 0🔄 1⭐ 0
  • Nov 18, 2026, 11:57 PM

    @dalias @kevingranade @xgranade Yeah, between QC & homomorphic encryption... I'm finding it useful to distinguish between computers (in the Church-Turing sense) & what I'll call "calculators".

    Btw can you explain what sorts of operations QCs optimize?
    I have some idea, but as you say most explanation are quite poor.

    💬 0🔄 0⭐ 0
  • 💬 0🔄 0⭐ 0
  • Nov 19, 2026, 4:47 AM

    @xgranade correlation as in two signals in the same direction and seeing how well they match instead of convolution where the time signals are moving in opposite directions?

    💬 0🔄 0⭐ 0
  • Nov 19, 2026, 5:07 AM

    @xgranade

    Schrödinger's Cat was intentional disinfo put forward by a serial rapist. Invoking it helps no one.

    And now I'm wondering if Hellsing's author knew that because it would make a lot of sense.

    💬 0🔄 0⭐ 0