Hacker News Reader

Comment by lazaroclapp | Hacker News Reader

The article makes the argument that intelligent life a couple orders of magnitude larger than human would think very slowly, since the time for a single message to cross a brain 10 times the size is 10 times longer. This necessitates that every thought traverses the entire brain.

But, taking not too much poetic license, we can argue that human society at a whole is a 10^6m scale intelligence (well, more like 10^3 if we are only counting the pile of human brains, but the distances are still in the 10^6 scale). It furthermore has incredibly slow thought transmission outside of very small "fast" nodes. Yet it solves the problem by not requiring most signals to propagate very far. Only a few ideas need to be globally known by humankind, the intermediate filtering is still mostly local.

So, a galaxy-sized "brain" can exist, it just needs to be very hierarchical, since any global consensus would take hundreds of thousands of years to formulate, even at the speed of light. In computing / distributed system terms: locality matters.

View on HN

You still can't call that a meaningful organism. The speed of light still means it can't have had more than a few dozen "clock cycles" since the galaxy stabilized enough to permit it, and it doesn't have all that many more "cycles" before the galaxy will be too dead to sustain it. There is no activity it could set itself out to do and then actually do it that would not be literally at least tens of thousands of years behind the stimulus that produced it, and that would be an incredibly quick reaction. There is no force of "evolution" or anything else that could drive this entity. It's just not useful to model it as an "organism".

Even human society can only be analogized to an organism. It isn't one either. It's not enough to point out there's vague similarities but ignore the multiple-orders-of-magnitude differences in the various scales.

And there's no need for "everything to be an organism". There's no physical reality to that idea. It's merely a reflection of our own predisposition towards weak analogical thinking when we should be dealing with things on their own terms. For human society, analogical thinking can easily lead us astray; for galactic scale society it's even harder to get away from analogical thinking because there isn't one (so far as we know) from which we can derive any observations at all, so there's no concrete facts in which we can ground our thinking.

As a stepping stone, "human society as an organism" may be the beginning of understanding. But when you start doing things like bemoaning its lack of self-preservation instinct, it's a sign you've gone too far with the analogy at the expense of dealing with what is really there.

> Even human society can only be analogized to an organism. It isn't one either.

Why not?

An organism can reasonably be defined as the minimal subset of a gene's phenotype that is actually capable of reproducing. On that view, an ant (for example) is not an organism, an ant colony is the organism. This neatly solves the puzzle of how it is possible for evolution to produce ants, since the vast majority of individual ants are sterile.

On that view, the minimal possible organism for a sexually-reproducig species is a mating pair. But a single pair of humans would not survive in the wild. At a minimum it takes a village (as they say). So a human village can reasonably be taken to be an organism, just as an ant colony can. But a village can't produce a technological society. At best, a village can subsist. So why is it unreasonable to consider (say) a city-state as an evolutionary advance on the village, and hence an organism in its own right even thought it isn't minimal? And so on through the Roman empire, the British empire, and the modern nation-state and multinational corporation, all of which are no less examples of the human genome's phenotype than the human body itself.

> An organism can reasonably be defined as the minimal subset of a gene's phenotype that is actually capable of reproducing.

Where did this definition come from?

> But a single pair of humans would not survive in the wild.

Not true; there are numerous examples of single families surviving in the wild.

> Where did this definition come from?

From the puzzle: how can it be that evolution produces a creature (ants) where 99% of the individuals are sterile? And the answer: ants are not the organism, the ant colony is the organism. This idea was introduced by Dawkins in "The Extended Phenotype".

> there are numerous examples of single families surviving in the wild

Not without some external support. At a minimum they went into the wild with some clothing and tools.

> From the puzzle: how can it be that evolution produces a creature (ants) where 99% of the individuals are sterile? And the answer: ants are not the organism, the ant colony is the organism. This idea was introduced by Dawkins in "The Extended Phenotype".

That has nothing to do with the definition of the word organism. Maybe you want another word for the unit of reproduction, but the word /organism/ already has a specific meaning, independent of reproduction.

That's news to me. AFAIK, the meaning of "organism" is inextricably bound to the concept of life, which in turn is inextricably bound to reproduction. But please do enlighten me: what is this established meaning of "organism" that has nothing to do with reproduction?

LMGTFY: https://www.google.com/search?q=define+organism

"an individual animal, plant, or single-celled LIFE form"

"the material structure of an individual LIFE form"

"a whole with interdependent parts, likened to a LIVING being"

So let's go on to look up "Life":

"the condition that distinguishes animals and plants from inorganic matter, including the capacity for growth, REPRODUCTION..."

So planets that host sentient life are a form of life themselves.

Not unless the planet itself reproduces somehow.

If we go and terraform Mars, for example, and make it just like Earth, won't Earth be said to have reproduced?

I don't think so. What really matters is the information that is being replicated. In the case of humanity, that information is encoded in our DNA and in our artifacts. But the planet exists independently of any information that is encoded on or in it, so I don't think it would be fair to say that the planet has reproduced. But it's arguable.

This mostly depends on whether the planets where life originates are anything like each other.

We just assume that every life-harboring planet will have water and trees and blue skies, but even the "dead" planets that we've seen so far are nothing like each other.

However if Earth is the only place with leafy green trees, for example, and we go on to plant these on Mars and other planets, we will essentially be making copies of Earth (which would otherwise be the only example of itself) by replicating its attributes and variables on other worlds as much as we can, for our own comfort.

Over a large enough scale of time and space, Earth could be seen as a successful organism that managed to replicate itself. It doesn't need to have any direct agency over its reproductive components (us) any more than we have any agency over our own sperm and eggs.

That's a good argument, but it depends entirely on whether we (earth life) adapt new planets to suit our preconceptions of life, or whether life from earth adapts to the conditions in finds on other planets and becomes very different from what it once was. In the former case, yes, I would totally accept that earth had replicated. But I think the latter scenario is more likely.

Thats not true, they have lone individuals who can live isolated in the wild. its just not a factually true statement.

First, the vast majority of people who survive in the wild take tools and clothing with them. And second, it's not enough to just survive. You have to reproduce. And it's not enough to just make a baby, you have to raise it to the point where it is capable of independent survival.

You have a very very mistaken view about the level of sophistication necessary for reproduction. Its just not based in fact. There are still multiple stone age level tribes on the earth. take a woman and man from such a society with no tools and clothing, and put them in a environment similar to what they are used to and they will build the tools and clothing they need and reproduce.

Most modern humans are completely disconnected from this lifestyle and probably would not be able to adjust quickly enough... but its not some impossibility of biology.

Do you think a gorilla or chimpanzee male and female couple placed in an appropriate habitat with a reasonable amount of luck could survive to reproduce?

> There are still multiple stone age level tribes on the earth.

Yes, of course. The key word there being tribe. Not isolated breeding pairs.

> take a woman and man from such a society with no tools and clothing, and put them in a environment similar to what they are used to and they will build the tools and clothing they need and reproduce.

Maybe. But even if this were true, the couple would still be relying on knowledge gained through their upbringing in the tribe. That's the reason doing this experiment with humans not raised in a stone-age tribe would almost certainly fail.

Isolated families surviving does happen indeed. A good example is the Lykov family who were completely isolated for over 40 years. [1][2] However they brought with them tools made by others before starting their isolated life, like steel axes. They also imported education & language. I haven't heard of any isolated family success stories that start without any external help.

[1] http://www.smithsonianmag.com/history/for-40-years-this-russ...

[2] https://www.youtube.com/watch?v=tt2AYafET68

> they brought with them tools made by others before starting their isolated life, like steel axes

Exactly. This is just a piece of an organism (society) "budding off" and surviving for a while on its own, not unlike an individual ant wandering away from the colony, or Henrietta Lacks's cancer cells (https://en.wikipedia.org/wiki/Henrietta_Lacks). AFAIK there is not a single example of a human mating pair completely cut off from all technological artifacts and successfully reproducing.

While I do believe this hasn't really happened, surely there's a strong observational bias involved ("AFAIK" - but why would you know? :) If they're isolated, chances of learning about them would be very slim by definition)

Actually, we do have some actual data:

http://www.discovery.com/tv-shows/naked-and-afraid/

It's not a well controlled experiment, but it's data. Even in that minimalist scenario they have some tools. Because it's just obvious that if you had no tools your odds of survival are pretty frickin' low. People die in the wilderness all the time even with tools.

Humanity survived with nothing more than stone hand axes for thousands of generations, and there are plenty of instances of individuals, even in modern times, surviving in the wilderness, alone, without significant tools for an extended period of time. People have been stranded and survived on deserted islands, alone, for extended periods of time. With a plentiful food source and decent shelter, it's been done many times. It's both harder and easier than you imagine.

> with nothing more than stone hand axes

And a community of other humans.

> it's been done many times

I am not aware of even a single instance where a single pair of humans with no technological support successfully reproduced (by which I mean: raised a child until it was old enough to survive independently. Just having a baby doesn't count.) If you know of such a case, please give me a reference.

What are you talking about? You are completely out of touch with reality. Society is great and is humans natural state... the vast majority of people are going to eventually experience psychological issues due to being alone and will lose the will to live. However being able to successfully hunt and gather enough sustenance to survive without a larger group, is not particularly 'impossible'...

It's not enough to hunt and gather and survive. You have to be able to reproduce. And you have to do it without any tools or clothing or shelter that was produced by civilization. That is very, very hard. It might not be impossible, but AFAIK it has never been done.

I like the metaphor that ants are "cells with legs".

Life really isn't constrained to any of our neat categorizations; life really is whatever works over time.

I think you'll find this interesting: http://accelerating.org/articles/transcensionhypothesis.html

It argues almost the opposite of your points re: "everything is an organism." One of the things the author talks about is the possibility that our universe (and just about everything in it) is exposed to evolutionary forces. It even deals with your "distances too large to form a feedback loop" issue (by taking the relativity of distance and time to logical extremes).

You seem to be assuming that an intelligent organism would have uniform thoughts across their entire body. Cephalopods, at least, don't entirely match that description. Different parts of their bodies act on their own, often with no need for central organization. A larger intelligence could perhaps work like that. Maybe then you would consider just a single section that is still intelligent when severed from the rest to be the "organism" and the rest to be some sort of "colony", but what if the borders aren't clear. What if any cubic meter chunk of alien flesh was sentient, no matter where you cut it from?

You still can't call that a meaningful organism. The speed of light still means it can't have had more than a few dozen "clock cycles" since the galaxy stabilized enough to permit it

Signals inside the cerebral cortex were cited as traveling as slow as 7mph. (I think I'm getting that from the old Cosmos series.) So a brain could be at least 95 million times larger. So Marvin the Paranoid Android's complaints referencing his "brain the size of a planet" could be realized in theory. In Michael Swannick's Vacuum Flowers the Earth had been taken over by a hive mind called "The Comprise" but it couldn't grow any larger because of lightspeed delays. Attempts to do so would result in "mitotic division" into rival sub-entities, one of which would kill the other.

(op cites 300 kph as neural transmission speed, but this is primarily true for long nerve fibers. In the brain, it's much slower, IIRC.)

You still can't call that a meaningful organism. The speed of light still means it can't have had more than a few dozen "clock cycles" since the galaxy stabilized enough to permit it

I'm curious how you define "organism", and what it means to be a "meaningful organism"?

(1) substrate -> (2) corpus -> (3) metabolic order.

Let's build one:

(1): plasma

(2): plasma cloud contained by EM

(3): Charged particle flows in context of a network of causality.

Does EM stand for electromagnetism? If so it seems that force is the most important factor in an organism?

> And there's no need for "everything to be an organism". There's no physical reality to that idea. It's merely a reflection of our own predisposition towards weak analogical thinking when we should be dealing with things on their own terms.

This. I look forward to a future where science is made by machines. What we know so far has the imprint of human values and limitations. I am very suspicious when a theory 'makes too much human sense'. Reality has no obligation whatsoever to be intelligible. What really is, outside from the human mind, has a structure that is unfathomable complex: human knowledge is necessarily a reduced representation of a tiny slice of such underlying reality. And most science is pragmatical science for human purposes, not fundamental work towards the comprehension of the inner workings of reality. This is a task for superior intelligent beings, not for us.

Yes, and also, who cares how long it takes to think?

The article points out:

> If our brains grew enormously to say, the size of our solar system, and featured speed-of-light signaling, the same number of message crossings would require more than the entire current age of the universe, leaving no time for evolution to work its course.

but this argues that stellar-scale life doesn't exist yet, not that it can't exist.

The universe is extremely young, ~10^10 years. Stars will continue to form for 10^14 years - ten thousand times longer than the universe has existed so far. And the universe will continue to evolve for many many orders of magnitude after that. There is plenty of time.

That argument also assumes that such an intelligence must evolve. If you were to build it out of individual cells, that seems reasonable. But such an intelligence could also be an emergent property of civilizations of intelligent beings, for example. We know too little of what sentience involves to be able to rule out such hive minds.

That's right. We humans are individualized at the organism level and tend to think entirely from the perspective of a single organism.

However it seems clear to me that networked intelligence, of the sort that has only existed on Earth for a couple of decades, is actually a new form of intelligence than individual intelligence.

Networked intelligence on Earth is extremely new and weak, however I am confident that in my lifetime we will recognize it as an "emergent intelligence".

> However it seems clear to me that networked intelligence, of the sort that has only existed on Earth for a couple of decades

Or 120 million years, if you think that ants fit the description.

I was referring to " the sort that has only existed on Earth for a couple of decades." That's why I said that ;)

Actually my mistake was here, where I wrote:

> actually a new form of intelligence than individual intelligence.

should have probably said "actually a new form of intelligence on top of human sentience." since I was referring specifically to the emergent intelligence of a sentient species able to encode our intelligence and transmit it throughout the network at near light-speed.

Or even billions of years, if you consider bacteria. They span enormous areas and reproduce so quickly that it does in effect act as a broad information network, just not in a way we would call sentient intelligence.

In fact they can have collective signaling networks as well -- consider the specialization and signaling that goes on in the biofilms of slime molds.

(some examples: https://www.biofilm.montana.edu/node/2438 & http://www.nature.com/articles/npjbiofilms20156 )

I'm reminded of Greg Bear's novel Blood Music.

Damn that was a scary piece of writing.

We humans are individualized at the organism level...

Are we, though? Isn't an individual human actually a large colony of co-operating cellular organisms?

... which are each just small bags of chemicals.

But it's only at the level of a brain that the illusion of consciousness appears.

I'd say our markets are a form of network intelligence.

We already have artificial intelligence, in the form of any sufficiently large organisation. Corporations are intelligences of a sort - they take in information and respond to it using defined rules, and by co-opting meat computers as needed.

With all the GPUs and cheap memory these days, I wonder why I am not reading about more attempts to evolve life in a computer than trying to design it.

If you want to read about evolving life in a computer, have you tried Greg Egan's short story Crystal Nights - http://ttapress.com/553/crystal-nights-by-greg-egan/ ? It's fiction, but it's good.

Just read this. Thank you for the recommendation. It was good.

Good story!

> would think very slowly

Very slowly for us humans, at our scale.

But "it" would think at its own, regular pace (which might, in turn, be very fast for an even larger entity). In regard, we humans may not even be a perceptible event on this time scale.

It's a matter of metabolism and of perception of time: we, as humans, think & react very slowly for smaller animals (and are even slower on the time scale of basic chemical reactions).

Well, the speed of thinking and control must be roughly related to the speed of actions. If you look at the giraffe, it's probably near the upper limit for an animal that still can run. Not about the muscles, but the brain couldn't be much further from the hooves or it would trip a lot, because the signals take so long to travel and there are limits to local "automation". Larger animals can exist, but they can't run, they must walk or swim slowly. Or the neural pathways would need to evolve into something faster.

If you look at human society, once we industrialized and developed things like strip mining, synthetic fertilizers, plastic bags, chlorofluorocarbons, trawling or nuclear technology, it's clear that our capability of doing things multiplied. Sometimes the thinking and values part to actually control what we're doing comes a little bit belatedly.

> * Larger animals can exist, but they can't run, they must walk or swim slowly. Or the neural pathways would need to evolve into something faster.*

Or the head-brain could farm out the running to a different control point - a second brain that controls running.

It seems like we may have already done this, ourselves: http://www.scientificamerican.com/article/gut-second-brain/

But some part of running requires vision since you must see obstacles and plan your footings. More the bigger and more fragile you are. So there would need to be another set of eyes and some basic processing then as well. And then, what about 3d perception, moving obstacles, predator recognition? You have to know what's just tall grass and what's a rock.

You'd be better off moving the brain down and just having a really long mouth / snout / tongue. And then you'd have an elephant...

You can only have very basic processing near the actual actuator.

But why couldn't you have specialized processing... "brains"... near the sensors, and then have these send a simple binary signal back to some central "ok this is what we'll do"-brain?

Human society is certainly a distributed intelligence. Elements of collective memory are external and persistent, and communication between individual units can be both remote and highly abstract.

Unfortunately it's not a very bright distributed intelligence. Memory has scaled successfully, but the strategic insight needed to guarantee self-preservation hasn't.

>So, a galaxy-sized "brain" can exist, it just needs to be very hierarchical

Or it could just think very slowly. For an entity that big there probably isn't any real hurry or much need to worry about external threats. Slow thinking also means that the speed of light stops being a problem.

It's a huge problem for humans because we don't live very long. If we lived for millions of years at one millionth of our usual pace we wouldn't find it limiting at all.

Human society is certainly a distributed intelligence.

That doesn't mean we should think of society as a singular "living creature".

What would mean that? What qualities is it lacking to be thought of as a singular living creature?

We are a collection of, among other things, much smaller living creatures...

Nobody is telling you how to think.

I like the idea.

I think there is a layer in between.

People's social circle, the subculture they belong to are the first meta-organism.

Society as a whole is then a meta-organism made from subcultures.

There's a nice sf short story, I think by Alastair Reynolds, which concerns worms in a glacier essentially acting as neurons in a vast, slow brain.

Found. "Glacial", Galactic North anthology.

Another Alastair Reynolds short story in the same vein is "A Murmuration" that I read (on my Kindle) in Interzone #257. In the story a flock of birds exhibits intelligence independently from the individual birds.

Well essentially thats how a hive mind works but I think it's a stretch to classify, idk say a bee colony, as constituting a single organism.

Considering my lacking but informed understanding of the fundamental requirements of life as it exists here on earth, I see a fundamental flaw with the idea that a galaxy sized organism could sustain the engine of life, protein folding.

There has to be some source of external energy that is fueling these organisms and galaxy sized anything make even the most massive stars in the galaxy look infitismally tiny in comparison.

Perhaps gravity of a super massive black hole could kick-start and spin some sort of electric turbine like biological organ or something to get things put into motion but considering life starts small such a scale would hit hurdles.

Of course this is assuming that life on distance planets all use the primarily carbon based deoxyribonucleic acid system of life.

> idk say a bee colony, as constituting a single organism.

Humans try to classify collections of matter into coherent objects. Otherwise there would be no way we'd be able to understand the world without that abstraction. But you live in that world long enough, and you begin to believe that reality neatly fits into the categories, however the categories exist only in our minds. But you can always make the categories break if you push them far enough. Plenty of philosophical discussions are really just about definitions of words and not anything 'real': Does life begin at conception? Is Theseus' ship the same? Is a bee colony a single organism?

From http://www.paulgraham.com/philosophy.html:

> Outside of math there's a limit to how far you can push words; in fact, it would not be a bad definition of math to call it the study of terms that have precise meanings. Everyday words are inherently imprecise. They work well enough in everyday life that you don't notice. Words seem to work, just as Newtonian physics seems to. But you can always make them break if you push them far enough.

We are colonies of cells.

Speak for yourselves. I'm a miniscule subcomponent of Gaia.

Their could be forms of life that we don't even have a clue about the biological nature of itz composition.

If you want to read a sci-fi exploration of a similar idea, pick up Ancillary Justice by Ann Leckie. (I can't say more than that without spoiling things.)

I can't help thinking about one of my favourite SF ideas - that the rules of computation vary depending on where in the galaxy you are (Vinge's "Zones of Thought").

That speed limit for biological activity might not exist in the universe if you consider the interesting natural "applications" of quantum phenomena. The ability for a particle to travel infinite paths simultaneously has been shown in photosynthesis, and fascinating examples of entanglement are in the literature. So it makes you wonder if distance has anything to do with the capabilities such a large brain.

It's hard to say whether those kinds of mechanisms could evolve for such a large organism. Photosynthetic molecules with interesting quantum effects developed while piggybacking on early bacterial evolution which was extremely fast because of the abundant radiation and their rapid reproduction. The chemical reactions in space might not be fast enough and may not have enough reaction energy (near absolute zero temperature) to evolve molecules with interesting effects.

Consensus is overrated. Humans agree to do stupid things all the time.

Being right is overrated, people agree about it all the time.

I don't think this exists without consensus.

Would it be possible for you to restate and/or expand on your comment above.

The concept of being right only makes sense across a dialectic of (perceived) truths. Obviously absolute truths about the state of the universe are still utterly inaccessible, and a priori truths never required consensus to begin with.

It also makes a dangerous assumption : that all forms of life is designed according to our idea of what life is. But life may very well exists outside of the concept of thoughts, or may not be constrained by space, time or causality. Or maybe span on severeal spaces/times.

For what we know anything could be alive, we just have now way to be verify it. Our concept of "life" is very limited, because it's limited to what we can perceive or imagine.

> or may not be constrained by space, time or causality

If something isn't constrained by causality, can we even classify it in the same "life" bucket that we define ourselves to be in? Can we even reason about it?

Well, the life bucket we define ourselves in may very well not be causal either. Maybe we just see it that way because of our limits. It's impossible to prove.

This is the limit of our condition: at a certain point, everything is a believe. I believe in causality because I have been perceiving the reality this way. And this believe is strong enough to prevent myself from jumping outside of the windows, fearing the consequence of my action.

But nothing proves that the universe is causal. You can imagine a non causal universe, and some mecanismes making us not see it.

You can imagine many totally randoms universe, but our conscienciouness only focus on the one we think make sense.

You can imagine that each individual live in it's own universe, but none of them are influencing each others.

You can imagine no universe and just a dream, or a virtual one projected to get a restult.

You can imagine a universe that is created and dying at every moment, replaced by another one in order to try to find the perfect combination or particules or to restore some kind of balance.

Thank you, Herbert Phillips Lovecraft.

Human nerve impulses travel at ballpark of 1e-7 c though, so limits of physics let you keep the relative signaling speed constant while making the system between enormously larger in radius. And the volume (and hence number of neurons) of course grows approximately as a cube of radius!

Something like this: https://en.wikipedia.org/wiki/Asynchronous_circuit#Asynchron...

It takes more than one traversal to make a thought. It probably takes a volume of brain to process a thought. So the speed might decrease geometrically with size? It could take 1000X as long.

Or maybe within the brain of this large organism, the signals do propagate one end to another but through wormholes, thus not being limited to speed of light.

A galaxy size brain with a quantum entanglement.

Which is not supposed to help in transmitting information at superluminal speeds.

Well if it's too local it kinda begs the question of whether it's one brain or multiple brains loosely linked together just as in your example it's not clear if the human society as a whole is a planet scale intelligence or a bunch of human scale intelligences bundled together.

a galaxy size brain would disagree with itself a lot.

My skull size brain is alreayd disagreeing with itself plenty.

Reply on news.ycombinator.com