DecaDeciHuman
No bio...
User ID: 3518
DecaDeciHuman
3mo ago
"The last ten times I played Russian Roulette I was fine". Beware selection bias. If a prior mass intervention had caused this, you likely wouldn't be here to make that statement.
(Also, "drop dead in 20 years" is an obvious extreme example to illustrate the point. I am far more concerned in practice about e.g. long-term impacts on fertility, as that legitimately can have lags of 15-20 years.)
DecaDeciHuman
3mo ago
You're assuming COVID in this scenario was the intended bioweapon and not an intermediate they were e.g. gene-harvesting.
DecaDeciHuman
3mo ago
Eh, it really wasn't my cup of tea.
Like so many other of his novels, it felt very much like two or three novels roughly grafted together, and where only the first one was interesting. Not as egregious as Seveneves, but that's not saying much.
DecaDeciHuman
3mo ago
You can find papers with actuarial analysis
Agreed. Pop quiz: substance X causes you to drop dead in 20 years with no side effects before then. It has been 10 years since substance X has been introduced. What does actuarial analysis show on the effect of substance X?
DecaDeciHuman
3mo ago
COVID was actually very bad
In which regard? I strongly suspect I agree with you here; this statement is too ambiguous without clarification to be able to truly tell.
- Medicine is obviously politically compromised when it comes to culture war topics.
Agreed.
The correct response to 2 is to [...] not ignore general and uncontroversial medical advice.
Agreed. Now define uncontroversial.
A reasonable middle ground is to do things like actual independent high quality research
If only it was legal to do so.
ask someone who is not politically compromised (me! me!).
P(politically compromised | states is not politically compromised) > P(politically compromised | does not state is not politically compromised)
Just because someone was wrong one time or on one category of things doesn't mean you stop listening to them for everything.
It does mean I trust everything they say less, yes.
DecaDeciHuman
3mo ago
I think RLHF helps a bit with this
RLHF tends to make a model less calibrated. Substantially so.
DecaDeciHuman
3mo ago
The amount of harm done by bad faith actors won't stay the same once you've settled on a policy about false negatives and false positives.
Agreed.
The bad actors will see your policy, and act in ways that the policy incentivizes.
Ditto for good faith actors, for that matter. And "ways that the policy incentivizes" is not the same thing as "ways that the maker of the policy publicly announced that the policy would result in", nor is it the same thing as "ways that improve the outcome compared to without the policy".
If you put a policy in place that results in, oh, severe consequences for using a persons non-preferred gender, you will likely get many people to stop using said person's preferred gender, as they went through a logic chain along the lines of "I am too forgetful / do not have the memory to 100% guarantee that I always get 100% of people's preferred gender correct 100% of the time, and the consequences are severe, so I will stop using gender altogether to refer to people".
DecaDeciHuman
3mo ago
Not sure about that one.
There are definitely single identities that have been targeted by sociopath copycats. Admittedly, typically prominent identities, as otherwise there isn't much upside.
What matters here is more how much sway / visibility / power / soft power / whatever you wish to call it the identity has, not really the size of said identity.
A community is really just a shared (partial) identity - and the flipside of that is that an identity is really not all that different from a community that currently has one member.
DecaDeciHuman
3mo ago
Could someone who does not like another person's culture buy all their offspring and raise them in a different culture?
DecaDeciHuman
3mo ago
Another more subtle selective pressure in this scenario is for egregores to encourage other woman to pass around barren egregores, and to discourage other woman to pass around fertile egregores.
One might suppose that this also results in selective pressures on the human genome to be asymptomatic carriers of barren egregores (i.e. pass them on to other woman, but to not be actually affected by it themselves). And the opposite for fertile egregores - i.e. to be heavily affected by them but not pass them on to others.
Again, all of these are more subtle effects in the presence of limited total resources.
DecaDeciHuman
3mo ago
I'm fine with it being defined by a common law court system.
Ah, here's our disagreement. Or a disagreement, at least.
My prediction of such a system is that the hole of vetoing contracts by way of declaring them coercive would grow over time until the state regained full control.
It is competition with the state, which cannot stand without support for the reasons @MaiqTheTrue highlighted.
DecaDeciHuman
3mo ago
I would also presume so offhand.
The grandparent comment came across as attempting to trivialize the subject by mentioning something mundane that can - technically - trigger the same effect.
DecaDeciHuman
3mo ago
Having worked with Prolog before - what the parent describes is the siren song not the actuality.
In actuality Prolog ends up with you quickly writing something that describes what you want it to do... and then spending most of your time trying to reword it in a way that it will be able to interpret before the heat death of the universe.
DecaDeciHuman
3mo ago
it should drive down the price of all goods to near 0.
a) it could drive down the price of all goods that do not intrinsically require human labor to near zero.
b) [thing A which limits to 0] / [thing B which limits to 0] does not necessarily limit to 0.
c) if 99/100 necessities limit down to near zero... the funny thing about necessities is that they are necessary. All of them, not just a subset thereof.
DecaDeciHuman
3mo ago
Government A: takes in $1B in taxes and distributes it out in UBI.
Government B: takes in $1B in taxes and distributes it out in welfare.
What is the difference between these two in terms of inflationary pressures?
(If you consistently print money to be able to give out more money than you take in you'll likely get inflation, of course.)
DecaDeciHuman
3mo ago
Inertia.
Like so many other things, phone numbers went through a series of phases:
- Build something that worked, with many distinct mostly-accidental features and quirks.
- Everything seems fine...
- Strip out a seemingly-unnecessary feature or quirk.
- Everything seems fine...
- Strip out a seemingly-unnecessary feature or quirk.
- Everything seems fine...
- Strip out a seemingly-unnecessary feature or quirk.
- Everything seems fine...
- Coast along for a while on inertia [we are here].
- Realize that whoops, some of those quirks were load-bearing.
DecaDeciHuman
3mo ago
My prediction is that a nontrival portion of the automation goes into making more (make)work for others.
DecaDeciHuman
3mo ago
This is the sort of thing that - if one is canny - can be very lucrative. Find something that was in a glut and get into it as early as you can afford to, hoping that you'll make bank in the latter portion of your career (...preferably, and retire before it goes bust again).
Of course, it can also be a trainwreck if you time it wrong or if the market goes under. Such things are never without risk.
DecaDeciHuman
3mo ago
What/who currently guarantees student loans in the US anyway?
DecaDeciHuman
3mo ago
In Haiti the life expectancy is currently 64 years on average. That’s about 60 years higher than the four year life expectancy of the slaves that were shipped to Haiti to work there.
By this definition hospice homes are the worst hellholes on the planet.
(Or, to put it another way: this fails to take into account the mean age and health of said slaves that were shipped - it was not 'a representative sample of newborns from the population'.)
(I am well aware that even when taking this into account conditions were still hellish. Just making the point that this is not an apples-to-apples comparison.)
DecaDeciHuman
3mo ago
The tragedy of all of this:
I am - and pretty much always have been - firmly in favor of people being able to resleeve in the classic scifi style. I used to be open about this.
The current discourse around the subject? Has been actively counterproductive, and strongly so: I will not state the above in a non-anonymous setting. Full stop. (And even in an anon. setting I will not state the above without the full explanation, and am typically loathe to do so even then.)
Why?
Because instead of acknowledging - and working to overcome - the limitations and risks of the current technology, the discourse has largely been to:
a) suppress any research or studies showing the limitations and risks of the current technology.
b) suppress any research into improving said technology (mainly indirectly via the prior point)
c) attempt to gaslight others around them into pretending the current technology is resleeving in the classic scifi style.
d) attempting to press the current technology onto people who do not yet have the ability to fully grok the risk involved in said current technology.
If we had sex-change technology such that it was externally indistinguishable, fully functional, and fully reversible? I would be 100% for it.
If it wasn't, but was still fully reversible? I'd still be for it.
If it was externally indistinguishable, and fully functional, but not fully reversible? I'd be for it for consenting adults. (Though this scenario would be somewhat odd: if you had the tech to transition both ways why couldn't you do both in sequence?)
Unfortunately, what we have now is none of the above, to people who our society has otherwise deemed incapable of making irreversible decisions.
DecaDeciHuman
3mo ago
I simultaneously strongly agree with portions of this comment, and strongly disagree with others. I suspect I may not be the only one, although I cannot know that.
This highlights one of the limitations of a comment system like this: there is one up/downvote for the entirety of a comment.
DecaDeciHuman
3mo ago
I think that's a reason to think less of the Catholic Church as an organization, but not of random Catholic laypeople [...]
Would you say the same about all organizations? Or is this specific to a subset thereof? If so, what is the criteria for an organization to be in said subset?
DecaDeciHuman
3mo ago
Interesting that you phrase all of these as requirements/restrictions/what-have-you as opposed to permissions/freedoms/what-have-you. Interesting that you have 9 for children & adults, and 10 for children, but no 'for adults'.
I suspect you'd get different responses between the following categories:
- Freedom to do X, with no stance on later potential additional permissions/freedoms/what-have-you down the line.
- Freedom to do X, and a flat no to additional permissions/freedoms/what-have-you down the line.
- Ban on more than X, with no stance on later potential additional requirements/restrictions/what-have-you down the line.
- Ban on more than X, and a flat no to later potential additional requirements/restrictions/what-have-you down the line.
This suffers from precisely the same issue as talking about e.g. refrigerants. There have been those that stood the test of time - that you can no longer purchase.
Or to put it another way: let's talk excipients.
"drop dead in 20 years" is a blatent example to illustrate the point.
I am far more concerned in practice about things like, oh...
"This causes you to express a particular IgG antibody. Oops, said antibody breaks oocytogenesis. Oops, now any of your subsequent female children have nonviable oocytes". As this is something that:
a) has massive longterm effects (infertility of female offspring).
b) is essentially completely invisible for ~15 years or so.
c) is nontrivial to figure out even down the line.
More options
Context Copy link