[copsarebastards]

> You say that like it's some huge number. 70 million or not, it's still 1%. Numbers get their significance relatively, not in themselves.

Relative to what? I can't believe I'm having this conversation.

70 million people is more than the number of people killed in both world wars. 70 million people is more than the population of the UK or Canada. Are you claiming that the world wars caused negligible deaths, or that the populations of the UK or Canada are negligible?

You can't just say 1% is a negligible amount without context. Whether 1% is a negligible margin of error is entirely dependent on context. 1% blood alcohol will probably kill you, 1% error on your taxes, if intentional, is enough to put you in jail, 1% error in floating point arithmetic is the difference between a missile that hits its target and a missile that lands in a civilian residence. But numbers get their significance relatively, right?

70 million people is a lot of people.

[esrauch]

> 70 million people is more than the number of people killed in both world wars

Not that I necessarily disagree with your sentiment, but the world population was lower when those wars happened, so the the same number of people dying was far more than 1% at the time. Looks like WW2 alone was 3-4% of the global population. The world wars were also important for reasons other than number of deaths anyway.

[copsarebastards]

That's a fair point on the change in population since WW2.

However, I think that the atypically-sexed population is important for reasons other than numbers, too.

[coldtea]

Sure, but that's a totally different argument.

[copsarebastards]

It's the argument esrauch was making.

[ajkjk]

I can't believe you're having this conversation either. You're correct, obviously; you've thrown out tons of trivially correct examples to make it clear to everyone just how right you are, none of which have anything to do with this. You are of course aware that 1% out of a population has nothing to do with the concept of a 1% error on your taxes or 1% blood alcohol. Alcohol does not affect you 1% of the time if you have a 1% BAC.

Everyone here knows that if you make the population arbitrarily large, the 1% sample becomes large too. But can you really argue that being able to represent 99% with a binary spectrum isn't a pretty good approximation? What percentage would be good enough for you? Or are you going to say "99.9% isn't good enough because 7 million is a lot of people. that's more than died in X'?

[copsarebastards]

> I can't believe you're having this conversation either. You're correct, obviously; you've thrown out tons of trivially correct examples to make it clear to everyone just how right you are, none of which have anything to do with this. You are of course aware that 1% out of a population has nothing to do with the concept of a 1% error on your taxes or 1% blood alcohol. Alcohol does not affect you 1% of the time if you have a 1% BAC.

So you agree then that whether 1% is negligible is based on context?

> Everyone here knows that if you make the population arbitrarily large, the 1% sample becomes large too. But can you really argue that being able to represent 99% with a binary spectrum isn't a pretty good approximation? What percentage would be good enough for you? Or are you going to say "99.9% isn't good enough because 7 million is a lot of people. that's more than died in X'?

Yes. In case you didn't notice, 7 million people is a lot of people.

[coldtea]

>So you agree then that whether 1% is negligible is based on context?

Yeah, or as I put it: "Numbers get their significance relatively, not in themselves".

Whereas you repeatedly stated how 70 million people is a huge number in itself.

E.g. If I told you there are 70 million people that have blue eyes, is that "a huge number?" No, it's actually a small number. One would expect blue-eyed people to be in the 100s of millions or billions.

[copsarebastards]

> Whereas you repeatedly stated how 70 million people is a huge number in itself.

I haven't claimed that at all. I've said over and over that context indicates whether it's important.

> Whereas you repeatedly stated how 70 million people is a huge number in itself.

No, I've said 70 million people is a lot in terms of medical and social policy. It would be very possible, for example, for 1% of people to account for 10% of medical expenses--an amount you would probably care about at tax time. I think that's a number that matters to almost anyone's political goals.

In contrast, you've been repeatedly stating how 1% is not a large number. Based on what?

> E.g. If I told you there are 70 million people that have blue eyes, is that "a huge number?" No, it's actually a small number. One would expect blue-eyed people to be in the 100s of millions or billions.

Science doesn't give a shit about your expectations. "Expectations" are entirely irrelevant to whether a number is big or little. A number is big or little depending on what effects it causes and what effects you're trying to achieve.

You're accusing me of arguing that 70 million is inherently a large number, but you're arguing that 70 million is inherently a small number, completely arbitrarily. I'm not even saying 70 million people is a big or small number inherently, I'm saying that 70 million people is a huge number when the properties of that group have medical and social implications. 70 million blue-eyed people isn't a small or large number, it's an irrelevant number, because whether or not someone's eyes are blue has almost no implication that I care about. If you understand why it's not a big or small number, but an irrelevant number, you'll understand my point.

[coldtea]

>Science doesn't give a shit about your expectations.

Language.

Also, I didn't say it's about "MY" expectations. It's about what the expected distribution is, which is the whole context that makes something big or small.

"Expectations" are entirely irrelevant to whether a number is big or little.

Actually, it's all about that. Bringing 10,000 times 6 by throwing dice 20,000 times is too big, because the expected outcome is about 1/6 throws to be 6.

[Retric]

Context is important, a 1% rate of serial killers would be a ridiculously huge issue for a country.

[dwaltrip]

Example (with made up numbers):

0.1% of people fart right before falling asleep. 0.1% of people will commit murder in the next week. Now change the numbers to 20%.

In once case it matters a lot, another case it doesn't. Context matters when discussing populations also..

[threatofrain]

Yes, but so what? How do you link this argument back? 70 million is a lot of people in MORAL terms. What about scientifically?

We aren't talking about killing 70 million people. We're talking about the strength of constructs in terms of scientific utility.

This is also the problem with huge numbers. It's very hard to process and we are intuitively intimidated by the largeness, such as with numbers from the state budget. $70 billion? Oh my god. How am I supposed to process that number?

Also note that 1% is a figure arising from the most inclusive definitions.

[copsarebastards]

> Yes, but so what? How do you link this argument back? 70 million is a lot of people in MORAL terms. What about scientifically?

What about scientifically? Scientifically, there's no concept of negligible or not negligible. On what scientific grounds did you decide 1% was negligible?

The negligibility of a percentage is only choosable based on your values and how much you value what exists in that percentage. My argument is that in most contexts, you probably care about 70 million people. If that's not the case, you may be a sociopath. But my guess is that you aren't a sociopath--you're just operating under some temporary delusion that because you've decided to say 1% of people instead of 70 million people, your decision that the group of people in question is negligible is scientific.

> We aren't talking about killing 70 million people. We're talking about the strength of constructs in terms of scientific utility.

If you're claiming that 70 million people have no scientific utility, I'd like to see what utility function you're using.

> This is also the problem with huge numbers. It's very hard to process and we are intuitively intimidated by the largeness, such as with numbers from the state budget. $70 billion? Oh my god. How am I supposed to process that number?

I'm not sure how the fact that large numbers are hard to process means that 70 million people is negligible. Certainly saying 1% instead of 70 million makes it easier to process, but playing to human mental limitations isn't a particularly good source of truth.

> Also note that 1% is a figure arising from the most inclusive definitions.

I'll happily make similar arguments about 7 million people instead of 70 million.

[threatofrain]

Actually, there is a way to decide if something is scientifically better. All you have to show is that your construct is competitive within the ecosystem of constructs. You can weakly improve upon an existing model by adding tons of domain-specific complications, which is what should've happened. But instead of saying that there are two predominant sexes, along with many abnormal and discrete bins, they propose a "continuum".

The researchers found that, in the most inclusive definitions, 1% of the population isn't sufficiently accounted for by traditional constructs.

But there's no new theory here. How do I predict complications based on what factors? What's the new model? The "spectrum"? A spectrum is a scale with escalating and deescalating values as you travel up and down, where jumps in the spectrum are connected to jumps in prediction. As for abnormal and discrete bins, well, the scientific community already has that. What's new to the table? A reformation of language so that we avoid the word "abnormal"? But where's the improved model?

Also note that you propose that there's no way to think about scientific or construct "betterness". Yes there is. You can measure by complexity, prediction, explanation, or generalizability. These are just a few ways. But you waved away scientific discussion, and instead choose only to use the moral lens, and bring up sociopathy.

Also, the reason I am talking about human limitations in processing large numbers is because I am accusing the opposition of abuse. I am not saying you should believe me because of X, I'm saying beware of opposition arguments because they are abusive to human minds.

And on the matter of using percentages to interpret numbers, I return to my example of state budgets, because that is a place where politicians often abuse psychology by stating what appears to be extravagant numbers. By extending your statements, I might say that not only is $70B a lot of money, but so is $7B. But then what if you told me that $70B is less than 1% of the state budget? What did you just do to that number?

Honestly, 10,000 people dying is a lot. Therefore, let's not talk about construct validity?

[copsarebastards]

> Actually, there is a way to decide if something is scientifically better.

Not in a general sense, there isn't. "Better" can only be scientifically defined in terms of a utility function, a goal. If you're trying to conduct, copper is better than rubber, if you're trying to insulate, rubber is better than copper. If you're trying to provide adequate healthcare and social protection to people, then a lower margin of error would be better.

> All you have to show is that your construct is competitive within the ecosystem of constructs.

Competitive based on what utility function?

> But there's no new theory here. How do I predict complications based on what factors? What's the new model? The "spectrum"? A spectrum is a scale with escalating and deescalating values as you travel up and down, where jumps in the spectrum are connected to jumps in prediction. As for abnormal and discrete bins, well, the scientific community already has that. What's new to the table? A reformation of language so that we avoid the word "abnormal"? But where's the improved model?

I think an admission that the current model is inadequate goes a long way towards motivating the discovery of better models.

> Also note that you propose that there's no way to think about scientific or construct "betterness". Yes there is. You can measure by complexity, prediction, explanation, or generalizability.

Okay, so you've named a bunch of utility functions. Now do you really want to apply those to this situation? How do we apply these to the question of whether 1% is a negligible margin of error. Let's optimize for those:

1. Lower complexity: "everyone is a man" seemed to work back in the day. 2. Higher complexity: let's subdivide male and female. There are certainly other genetic traits besides X and Y chromosomes that we could include in our definition of sex. (Hint: It's silly to optimize for higher complexity, but why? I propose that the answer is based on your values.)

[panzagl]

People are reading too much into this article- it's more like "1% of the population has this interesting genetic condition that may have impacts under some conditions" not "1% of the population is being discriminated against"

[threatofrain]

Yes, I think people are playing border politics without realizing it. They want to be more inclusive, but this is about construct validity and adding something new to the table, not talking about how 70 million people dying is tremendous.

[coldtea]

>Relative to what? I can't believe I'm having this conversation.

Relative to the total population. Whether it's Canada, the US, France, or the Whole World you're taking into account, it's still 1% of it.

>You can't just say 1% is a negligible amount without context.

Probably you missed TFA and the whole conversation thread you're answering to?

The context was if only 1% of the population doing them is enough to call sexual preferences "a spectrum" (with regard to those "atypical sexual practices"). Something divided in 99% and 1% is not a "spectrum" by any stretch of the imagination. In fact there's a word for that 1%, outliers.

>1% blood alcohol will probably kill you, 1% error on your taxes, if intentional, is enough to put you in jail, 1% error in floating point arithmetic is the difference between a missile that hits its target and a missile that lands in a civilian residence. But numbers get their significance relatively, right?

Of course. 1% blood alcohol gets its significance not in what it is ("1% oh, so much") but RELATIVE to the amount that's OK for a human to stand.

1% error in missile calculations gets its significance RELATIVE to the target area it has to hit and the acceptable margin of error.

[copsarebastards]

> Probably you missed TFA and the whole conversation thread you're answering to?

TFA and whole conversation are exactly the context which makes it ridiculous to claim that 1% is an acceptable margin of error.

> Of course. 1% blood alcohol gets its significance not in what it is ("1% oh, so much") but RELATIVE to the amount that's OK for a human to stand.

> 1% error in missile calculations gets its significance RELATIVE to the target area it has to hit and the acceptable margin of error.

Agreed. 1% error in judging the gender of people is significant relative to medical and social policy targets. On what grounds are you claiming that 70 million people are ignorable in medical and social policy?

Ironically, the only argument from you I've seen so far against sex being considered a spectrum is basically, "1%, oh, not so much". You said: "You say that like it's some huge number. 70 million or not, it's still 1%."

And ultimately, this is in research before we are even talking about medical and social policy. I'm not sure why we should just discard that 1% of data at all--there's no reason to artificially create error in reasoning that isn't imposed by data collection methods.

[Zancarius]

> 70 million people is more than the number of people killed in both world wars.

I've seen some estimates that suggest 72 million were killed in WWII alone, either from direct involvement or as civilian casualties. Considering the world population was ~2 billion at that time, close to 3.6% of the total world population died as a consequence of the war.

Oftentimes, it's helpful when comparing approximate statistics from different eras that you use the same relative baseline--in this case world population at the time those statistics were estimated rather than now.

Edit: Didn't see esrauch's sibling comment. Give 'em an upvote.

[maratd]

> 70 million people is a lot of people.

So one person isn't a lot of people ... are you saying one person is negligible? That one person's life doesn't matter?

Two can play this game.

1% is 1%. Every life is important, but 1% is still 1%. And 1% is not a lot. Whether it's people, apples, or pencils doesn't matter. It's a ratio.

[copsarebastards]

> So one person isn't a lot of people ... are you saying one person is negligible? That one person's life doesn't matter?

Can we make laws, do medical research, etc. that will effectively help 1 person? I don't think so. If I see one person by the side of the road with a flat tire, I'll help that 1 person, but in the context of policy and research, 1 person usually doesn't matter because policy and research can't usually create a meaningful impact.

We can, however, make laws and do medical research that has an impact on 70 million people. I present as evidence for this the fact that life has gotten better (according to a variety of shared values which we could agree upon--fewer suicides, less violence) in the last few decades for people of atypical sexes.

> 1% is 1%. Every life is important, but 1% is still 1%. And 1% is not a lot. Whether it's people, apples, or pencils doesn't matter. It's a ratio.

No, context matters. If you don't think 1% is a lot in any context, maybe let's get you up to a 1% blood alcohol and see how you feel (hint: you won't feel).

[coldtea]

>We can, however, make laws and do medical research that has an impact on 70 million people.

Which is nothing this thread of discussion was about.

Nobody said not to study or legally hep those people.

Just that 1% is not enough to describe the total of cases as a "spectrum".

[copsarebastards]

> Which is nothing this thread of discussion was about.

> Nobody said not to study or legally hep those people.

> Just that 1% is not enough to describe the total of cases as a "spectrum".

Assuming that your medical and social policies are at all data-driven, failure to include people in your data is a guaranteed way to ensure that they are not studied or legally helped.