5 ms·

Well in that case even bits in a computer aren't binary. Because bits are represented by physical states we can measure (such as a voltage on a wire), and since

by umvi 6y ago

Well in that case even bits in a computer aren't binary. Because bits are represented by physical states we can measure (such as a voltage on a wire), and since said physical states are analog, there are certainly exceptions where you can't be sure if it's a 0 or a 1. A flip flop in a chip might be metastable and you can't tell if it's a 0 or a 1. So should you then conclude bits in a CPU's registers are not binary because there are edge cases where the flip flops are metastable? Or should you conclude they are "binary 99%+ of the time" which is to say "mostly binary"? It's useful to recognize the exceptions that exist, but let's not split hairs over the semantics of "mostly".

tomlockwood 6y ago
The thing is, the fact that even computer bits aren't binary is a very important piece of knowledge when you get deep into some computing topics. Heck, there are whole systems that were built solely to deal with this issue - see gray codes. 1 and 0 being discrete states is a convenience, for sure, but truth is often a matter of precise semantics.
eyelidlessness 6y ago
> Well in that case even bits in a computer aren't binary. Because bits are represented by physical states we can measure (such as a voltage on a wire), and since said physical states are analog, there are certainly exceptions where you can't be sure if it's a 0 or a 1. This is a baffling take. The analog nature of binary physical state is a fundamental limitation/impediment of hardware advancement. It’s part of why die shrinks are increasingly challenging. It’s part of why some techniques increasing density in SSDs make them less reliable. The closer one bit is to another, the more likely each bit's physical non-binary-ness can be ambiguous. Binary—in computers, in biological sex, in gender—can sometimes be a useful abstraction, and sometimes it can be harmful, and in no case is it technically correct.
- dolni 6y ago
  It's not really baffling. The underlying analog nature of bits becomes problematic when many bits become ambiguous. When nearly all of them are unambiguous, presenting bits as strictly on or off is perfectly serviceable. As almost everyone using a computer needn't care about the hidden analog nature of bits, most of society needn't care that sex rarely manifests in a non-binary way. All that needs to be said to most people is "be a decent human being."
  - tomlockwood 6y ago
    If we were reading an article about the nature of bits in a real computer system, I'd expect it to say they aren't discrete. I'm not even sure what you're proposing anymore.
  - eyelidlessness 6y ago
    > It's not really baffling. The underlying analog nature of bits becomes problematic when many bits become ambiguous. When nearly all of them are unambiguous, presenting bits as strictly on or off is perfectly serviceable. It’s baffling because it acknowledges then discards the ambiguity, and for the reasons that ambiguity is important which I already addressed. It has nothing to do with volume of ambiguity: a semiconductor product which misrepresents 0.01% of its bits is probably less serviceable than one which misrepresents 10%. > As almost everyone using a computer needn't care about the hidden analog nature of bits, most of society needn't care that sex rarely manifests in a non-binary way. All that needs to be said to most people is "be a decent human being." When both biological sex and social gender are non-binary for some of the population, insisting that they are is counterproductive to the directive to “be a decent human being”. Disacknowledging the reality of another person’s real experience is one of the most common tools used to neglect their needs, and ultimately to abuse and oppress them.
- samatman 6y ago
  The analog nature of binary sex is also a fundamental limitation of biology. From an information theory standpoint it's fair to model any deviation from the male/female poles as errors in the goal of reproduction with recombination through meiosis. Your idea of what's technically correct here doesn't survive scrutiny. Errors and ambiguities invalidating abstractions is the behavior of a mathematician or a philosopher, not an engineer.
  - eyelidlessness 6y ago
    Now this is baffling from even a lay perspective of evolution. Life evolves and survives based on success, not opinions. If seahorses favored a sex delineation where females gave birth they’d be extinct. Fortunately for seahorses they preferred mating.
    - samatman 6y ago
      Your objection is nonsensical. Seahorses have an unusual strategy for gestation, but they have a gender binary. As indeed they must, to reproduce through meiosis.
      - tomlockwood 6y ago
        So you'd disagree with the author of the article posted?
      - eyelidlessness 6y ago
        Yes, you’re right, how absurd for me to not recognize that they developed their biology in a vacuum, and obviously out of thin air unlike every other mating species.
- oska 6y ago
  The point they are making is that nothing in the physical world is going to be truly binary. Even when we try to represent abstract binaryness in specially designed ways, such as we continually do with computing equipment, there are always going to be edge cases in the physical representations used. So they are making a very pertinent point. Trying to deny a binary phenomenon in nature, such as the male/female sexual split, by pointing to various edge cases is just intellectually dishonest.
  - eyelidlessness 6y ago
    > The point they are making is that nothing in the physical world is going to be truly binary. That was in response to the point already having been made. The more important point is that when those abstractions are harmful to electronics it’s an engineering problem, but when they’re harmful to humans they’re deadly.

by umvi 6y ago

tomlockwood 6y ago

The thing is, the fact that even computer bits aren't binary is a very important piece of knowledge when you get deep into some computing topics. Heck, there are whole systems that were built solely to deal with this issue - see gray codes. 1 and 0 being discrete states is a convenience, for sure, but truth is often a matter of precise semantics.

eyelidlessness 6y ago

> Well in that case even bits in a computer aren't binary. Because bits are represented by physical states we can measure (such as a voltage on a wire), and since said physical states are analog, there are certainly exceptions where you can't be sure if it's a 0 or a 1. This is a baffling take. The analog nature of binary physical state is a fundamental limitation/impediment of hardware advancement. It’s part of why die shrinks are increasingly challenging. It’s part of why some techniques increasing density in SSDs make them less reliable. The closer one bit is to another, the more likely each bit's physical non-binary-ness can be ambiguous. Binary—in computers, in biological sex, in gender—can sometimes be a useful abstraction, and sometimes it can be harmful, and in no case is it technically correct.

dolni 6y ago
It's not really baffling. The underlying analog nature of bits becomes problematic when many bits become ambiguous. When nearly all of them are unambiguous, presenting bits as strictly on or off is perfectly serviceable. As almost everyone using a computer needn't care about the hidden analog nature of bits, most of society needn't care that sex rarely manifests in a non-binary way. All that needs to be said to most people is "be a decent human being."
- tomlockwood 6y ago
  If we were reading an article about the nature of bits in a real computer system, I'd expect it to say they aren't discrete. I'm not even sure what you're proposing anymore.
- eyelidlessness 6y ago
  > It's not really baffling. The underlying analog nature of bits becomes problematic when many bits become ambiguous. When nearly all of them are unambiguous, presenting bits as strictly on or off is perfectly serviceable. It’s baffling because it acknowledges then discards the ambiguity, and for the reasons that ambiguity is important which I already addressed. It has nothing to do with volume of ambiguity: a semiconductor product which misrepresents 0.01% of its bits is probably less serviceable than one which misrepresents 10%. > As almost everyone using a computer needn't care about the hidden analog nature of bits, most of society needn't care that sex rarely manifests in a non-binary way. All that needs to be said to most people is "be a decent human being." When both biological sex and social gender are non-binary for some of the population, insisting that they are is counterproductive to the directive to “be a decent human being”. Disacknowledging the reality of another person’s real experience is one of the most common tools used to neglect their needs, and ultimately to abuse and oppress them.
samatman 6y ago
The analog nature of binary sex is also a fundamental limitation of biology. From an information theory standpoint it's fair to model any deviation from the male/female poles as errors in the goal of reproduction with recombination through meiosis. Your idea of what's technically correct here doesn't survive scrutiny. Errors and ambiguities invalidating abstractions is the behavior of a mathematician or a philosopher, not an engineer.
- eyelidlessness 6y ago
  Now this is baffling from even a lay perspective of evolution. Life evolves and survives based on success, not opinions. If seahorses favored a sex delineation where females gave birth they’d be extinct. Fortunately for seahorses they preferred mating.
  - samatman 6y ago
    Your objection is nonsensical. Seahorses have an unusual strategy for gestation, but they have a gender binary. As indeed they must, to reproduce through meiosis.
    - tomlockwood 6y ago
      So you'd disagree with the author of the article posted?
    - eyelidlessness 6y ago
      Yes, you’re right, how absurd for me to not recognize that they developed their biology in a vacuum, and obviously out of thin air unlike every other mating species.
oska 6y ago
The point they are making is that nothing in the physical world is going to be truly binary. Even when we try to represent abstract binaryness in specially designed ways, such as we continually do with computing equipment, there are always going to be edge cases in the physical representations used. So they are making a very pertinent point. Trying to deny a binary phenomenon in nature, such as the male/female sexual split, by pointing to various edge cases is just intellectually dishonest.
- eyelidlessness 6y ago
  > The point they are making is that nothing in the physical world is going to be truly binary. That was in response to the point already having been made. The more important point is that when those abstractions are harmful to electronics it’s an engineering problem, but when they’re harmful to humans they’re deadly.