Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The impossible predicament of the death newts

crookedtimber.org

Post reply on HN
Print view View original

181–190 of 192 posts

Re: The impossible predicament of the death newts

#181

Post by odyssey7 »

mattigames wrote: post #147
odyssey7 wrote: post #142

Earlier quoted context omitted.

You’re comparing the intelligent design of your job search to evolution.

Luck is defined as something happening that was unlikely to happen, so having that in mind that evolution kept happening that is no luck, happens on every living organism (last specimens before extension being the exceptions), that the smarter apes got to reproduce is not luck, it is predictable, that they had to use less muscles if they are smarter is also predictable, that muscles are something primates can give up…

There’s a practically infinite space of possible innovations. The fact that any part of it has ever been explored is just down to luck—and that’s ignoring the fact that once evolution discovers an innovation, survival of the fittest is still itself not a rule but rather a likelihood. Maybe there was a fish who could communicate telepathically and fly at the same time there was a fish with feet, but the latter ended up devouring the former in the nursery.

Once you reach a particular point, things might tend to play out in ways that look more deterministic in specific places, but fate is still hard to predict. Consider the Vaquita. A species that has thrived for ages has been nearly wiped out of existence because a random primate species evolved to invent plastic fishing nets, and now that same primate species might altruistically manage to govern itself out of destroying the Vaquita. Really, nothing in that story was guaranteed to happen based on the frontier of the search process 1,000,000 years ago, it was just how the dice landed.

The Vaquita’s survival or loss is playing out in some ways as an international diplomacy story, where the first Trump administration saw declines in Vaquita numbers and the Biden administration took steps to improve their chances, and then Trump was re-elected, with some people believing that came down to Biden—a single human man who does not live in Mexico—experiencing age-related declines, and with others believing oligarchs bought the election to support causes like Russia’s pursuit of Ukraine. Really, this is about as random and as divorced from survival of the fittest as it gets. But it is no more random than the fact that the person who might have ushered in 1,000,000-year era of world piece and cured all suffering in all species would be just as likely to die as a child in a car accident as anyone else.

In ML, evolutionary algorithms are classified under randomized optimization, due to the way that they take random steps to forge random paths into vast combinatorial spaces that could never be completely understood or completely explored.

Re: The impossible predicament of the death newts

#182

Post by jkman »

mekoka wrote: post #135
xenadu02 wrote: post #130

Earlier quoted context omitted.

This is another case of a huge fallacy humans seem endlessly afflicted with: The Root Cause Fallacy. You are assuming there is but one cause for development and/or loss of resistance. There may not be much pressure to develop resistance to tetrodotoxin for most species. Simultaneously there might be a higher metabolic cost to retaining it for some species but not for others. It is also possible that resistance with l…

> You are assuming... I made no assumptions. As I pointed out to another commenter, you might be in too much of a haste to play at being a contrarian. It might be more useful to pay closer attention to what you're objecting to. Evolutionary game theory demonstrates that evolution is a matter of fitness payoffs. If cost of a trait increases, fitness is reduced. The prevalence of a trait in a fit population is indicati…

Jesus man, your hubris is astounding. 'I made no assumptions'? Ridiculous lol

Behave yourself

Re: The impossible predicament of the death newts

#183

Post by thaumasiotes »

amanaplanacanal wrote: post #99
aqme28 wrote: post #77

Earlier quoted context omitted.

But a nonzero number of animals and people die of tetrodoxin poisoning, so there is some pressure. Therefore if it were cheap and easy enough, it’s likely we all would have evolved it. That cheapness threshold might just be incredibly high.

If it is rare enough it probably doesn't exert much selection pressure. Has anybody modeled what percent of a population has to die from something for the protective gene to become widespread?

> Has anybody modeled what percent of a population has to die from something for the protective gene to become widespread?

The question is incoherent. The gene spreads if the organisms carrying it average more children. It unspreads if they average less. All of them could be dying of the same thing, and it wouldn't matter.

The rate of spread is given by the https://en.wikipedia.org/wiki/Selection_coefficient , but cause of death isn't relevant.

Re: The impossible predicament of the death newts

#184

Post by pfdietz »

arp242 wrote: post #5

> Meanwhile, evolving resistance also comes at a cost. We don’t know that directly, but we can infer it pretty well. If resistance to tetrodotoxin were cheap and easy, everything would evolve it. [..] We don’t know, but we’re pretty sure there must be something. We know that garter snakes outside of the Pacific Northwest are much less resistant to tetrodotoxin. They’ll drop dead from doses that their Oregon cousins s…

> There is no disadvantage to being able to synthesise vitamin C,

Synthesizing vitamin C takes energy, energy that could be used for other biological processes. It's also possible excess vitamin C has some minor deleterious effect. For example, it's an antioxidant, and these render immune cells somewhat less effective against certain threats (which they use oxidizing chemicals to destroy). It's been found larger doses of the ACE vitamins causes increased growth of lung cancer, probably due to reduced immune attack.

Some have argued against this idea, though, although I'm not convinced by the argument (see if you can spot the problem.)

https://pmc.ncbi.nlm.nih.gov/articles/PMC3145266/

Re: The impossible predicament of the death newts

#186

Post by the_af »

thaumasiotes wrote: post #159
the_af wrote: post #145

Earlier quoted context omitted.

Interesting. Let me see if I follow: once the snake population has the warning coloration, and predators know not to eat them, then individual snakes being successful at eating poisonous newts is unrelated to the snakes living long enough to pass their genes (i. e. being successful in terms of natural selection). So a snake which has the right colors will be successful, regardless of its diet. I wonder, is there a po…

> Let me see if I follow: once the snake population has the warning coloration, and predators know not to eat them, then individual snakes being successful at eating poisonous newts is unrelated to the snakes living long enough to pass their genes (i. e. being successful in terms of natural selection). So a snake which has the right colors will be successful, regardless of its diet. Sort of. Whether the snake is pois…

> This does not reflect a good understanding. Success means having more children, not having any children.

I never said "any children", so your remark doesn't reflect a good understanding of what I wrote.

Try not being pedantic for pedantry's sake, and engage with a question asked in good faith without playing games of one-upmanship.

Re: The impossible predicament of the death newts

#187

Post by mbonnet »

mekoka wrote: post #135
xenadu02 wrote: post #130

Earlier quoted context omitted.

This is another case of a huge fallacy humans seem endlessly afflicted with: The Root Cause Fallacy. You are assuming there is but one cause for development and/or loss of resistance. There may not be much pressure to develop resistance to tetrodotoxin for most species. Simultaneously there might be a higher metabolic cost to retaining it for some species but not for others. It is also possible that resistance with l…

> You are assuming... I made no assumptions. As I pointed out to another commenter, you might be in too much of a haste to play at being a contrarian. It might be more useful to pay closer attention to what you're objecting to. Evolutionary game theory demonstrates that evolution is a matter of fitness payoffs. If cost of a trait increases, fitness is reduced. The prevalence of a trait in a fit population is indicati…

You need to accept that you don't know what you're talking about.

Re: The impossible predicament of the death newts

#188

Post by prova_modena »

titanomachy wrote: post #19

Newt poisoning of humans must be rare, I’ve lived in this region my whole life and I don’t think I’ve ever heard of this!

I did several semesters' worth of biology fieldwork in college that involved handling many of these guys, under the guidance of a professor who was obsessed with them. Their toxicity was discussed more as a point of interest rather than an immediate danger. We always wore gloves when handling them, but the rationale was more for the newts' protection rather than our own. I never heard of there being a newt poisoning incident during the past decades of this professor overseeing similar fieldwork. Of course, eating them was out of the question.

Re: The impossible predicament of the death newts

#190

Post by thaumasiotes »

the_af wrote: post #186
thaumasiotes wrote: post #159

Earlier quoted context omitted.

> Let me see if I follow: once the snake population has the warning coloration, and predators know not to eat them, then individual snakes being successful at eating poisonous newts is unrelated to the snakes living long enough to pass their genes (i. e. being successful in terms of natural selection). So a snake which has the right colors will be successful, regardless of its diet. Sort of. Whether the snake is pois…

> This does not reflect a good understanding. Success means having more children, not having any children. I never said "any children", so your remark doesn't reflect a good understanding of what I wrote. Try not being pedantic for pedantry's sake, and engage with a question asked in good faith without playing games of one-upmanship.

You said that living long enough to reproduce constituted Darwinian success. That idea is (1) extremely common, and (2) dead wrong.
Post reply on HN