Agnosticism and Intelligent Design

[Keith&Co.]

Same response as to Intelligent Design, a supposed intelligent force did not do such a great job....

You're just not looking at it right.
Sure, i mean, infants are all-the-time choking, sometimes to death, because food and air intake access ports are co-located, but just look at the beauty apparent in a butterfly's wings. Huh? Huh?

 

[steve_bank]

Same response as to Intelligent Design, a supposed intelligent force did not do such a great job....

You're just not looking at it right.
Sure, i mean, infants are all-the-time choking, sometimes to death, because food and air intake access ports are co-located, but just look at the beauty apparent in a butterfly's wings. Huh? Huh?

I believe chimps have separate tubes for food and air, they can not choke.

 

[phands]

I believe chimps have separate tubes for food and air, they can not choke.

So the intelligent designer made a better job of chimps than humans???? Wow...gawd is truly a fuck-up.

 

[excreationist]

Same response as to Intelligent Design, a supposed intelligent force did not do such a great job....

You're just not looking at it right.
Sure, i mean, infants are all-the-time choking, sometimes to death, because food and air intake access ports are co-located, but just look at the beauty apparent in a butterfly's wings. Huh? Huh?

I believe chimps have separate tubes for food and air, they can not choke.

https://biology.stackexchange.com/questions/1133/can-any-other-animal-choke-on-food
"Humans are the only mammal that cannot breathe and swallow at the same time, and we are the only species that can choke on its own food. The reason? The lowering of the voice box in our throats (during infancy) enables us to create the enormous range of sounds used in producing language; but this lowering of the voice box comes at a big cost in adulthood."

So that would be partly why chimps don't talk.

 

[excreationist]

.

That pattern by the puffer fish is not nearly as impressive as our ability to catch a ball thrown at us,

Yeah.

Thr BIG thing my first bio teacher insisted on was that humans use tools. He got testy about any example of animals using tools, because to him, that set US apart.
So, if we're talking complicated BEHAVIOR, a toddler with a bucket and shovel making sand piles is probably more complicabted than the efforts of a fish. Or not.

Still need details on how to measure complicatery.

When I was thinking about the toddler I was imaging them picking up a handful of sand then dropping it at a spot making a messy pile vs making a fairly precise geometric design. BTW you could train a dog to make a pile of something.

 

[excreationist]

How do you determine this? How is the toddler's pile of sand automatically less complex than the puffer fish's pile? You haven't seen it, but you KNOW it's less complex?
How? How are you so sure?

Extremely intricate geometric patterns is not particularly impressive if you know how recursive programming works.

I do know how recursive programming works. I've made recursive functions that can convert numbers into their worded equivalents even if the number is in the trillions. The design would partly be made by its sense of position and then involve its visual inputs starting from an empty area of ground.

That pattern by the puffer fish is not nearly as impressive as our ability to catch a ball thrown at us,

A dog can be trained to catch a frisbee and a dolphin can throw and catch balls. I'd like to see you train a dog to make a design like this:

or a sparrows ability to tell worm from twig or a squirrels ability to hide and remember where they've hidden hundreds of nuts. Some things in nature are truly impressive.

Those behaviours have obvious survival value. I'd say it is easier to recognise a worm from a twig than to constantly know which part of the design you're in when some or none of the design is currently visible.

The intricacies of the puffer fish is probably sexually selected for. Because only a healthy puffer fish male would have the energy and skill to pull it off. So same deal as with peacock feathers. Which would constantly push puffer fish evolution towards greater and greater intricacies. Until we reach the limit. Which is the point where making the pattern will help kill the fish, due to over expenditure of energy.

BTW you can also train a chimp to do complex things but I don't think people are able to get a chimp to draw a reasonable circle. And drawing a circle is a fairly simple behaviour. (though it takes a lot of training to get right)

 

[excreationist]

Well said. Plus many of these behaviors result from a general principle seen in many places in nature - "Simple rules applied locally". Conway's Game of Life is non-natural example. Flocking behaviour in Starlings looks remarkably complex, but is a small set of simple rules applied to the nearest 7 neighbours.....it also looks beautiful.

ETA: the Mandelbrot set looks very complex too, but is a very simple rule set.

The design does look a bit like the Mandelbrot set. I guess you think that a little bit of data, like the Mandelbrot set formula, appeared in the fish's brain from a mutation and tada it could create the design.

 

[steve_bank]

I believe chimps have separate tubes for food and air, they can not choke.

So the intelligent designer made a better job of chimps than humans???? Wow...gawd is truly a fuck-up.

I think Earth was a test run.

 

[steve_bank]

I do know how recursive programming works. I've made recursive functions that can convert numbers into their worded equivalents even if the number is in the trillions. The design would partly be made by its sense of position and then involve its visual inputs starting from an empty area of ground.

That pattern by the puffer fish is not nearly as impressive as our ability to catch a ball thrown at us,

A dog can be trained to catch a frisbee and a dolphin can throw and catch balls. I'd like to see you train a dog to make a design like this:

or a sparrows ability to tell worm from twig or a squirrels ability to hide and remember where they've hidden hundreds of nuts. Some things in nature are truly impressive.

Those behaviours have obvious survival value. I'd say it is easier to recognise a worm from a twig than to constantly know which part of the design you're in when some or none of the design is currently visible.

The intricacies of the puffer fish is probably sexually selected for. Because only a healthy puffer fish male would have the energy and skill to pull it off. So same deal as with peacock feathers. Which would constantly push puffer fish evolution towards greater and greater intricacies. Until we reach the limit. Which is the point where making the pattern will help kill the fish, due to over expenditure of energy.

BTW you can also train a chimp to do complex things but I don't think people are able to get a chimp to draw a reasonable circle. And drawing a circle is a fairly simple behaviour. (though it takes a lot of training to get right)

A thread for science. Tool making and problem solving exists across the spectrum of critters.

Humans just do it better. In the 50s we were taught in school that humans were unique because of speech, opposing thumb and forefinger, and tool making. Commination's exist in non humans as well as tool making. Chimps quarry and fasion stones by chipping into nut cracking tools. There are many examples.

We have Cormorants in Puget Sound, birds who dive for fish. They line up in a straight line on the water and dive together. I watched a seagull line up ion the end and try to dive with them. Birds across the spectrum are problem solvers.

your tag is excretionist but I get an anti evolution vibe. Where do you stand on evolution?

 

[DrZoidberg]

That pattern by the puffer fish is not nearly as impressive as our ability to catch a ball thrown at us,

A dog can be trained to catch a frisbee and a dolphin can throw and catch balls. I'd like to see you train a dog to make a design like this:

Anticipating the parabola of a hurled object directed towards us is advanced trigonometry in real time. It's an absolutely amazing piece of evolutionary design.

The puffer fish pattern is just a set of pre-programmed instructions that it has plenty of time to draw. Fish have a swim bladder which detects movement and direction. Fish have a sense of where they are and which direction they are facing at all times.

If you use evolutionary principles to create a programme that paints a similar pattern, we could bang that out in minutes. It's pretty basic.

If you use evolutionary principles to create a programme to make a robot catch a ball thrown at it... good luck with that. Our top minds have been trying for decades. Its not going so well.

or a sparrows ability to tell worm from twig or a squirrels ability to hide and remember where they've hidden hundreds of nuts. Some things in nature are truly impressive.

Those behaviours have obvious survival value. I'd say it is easier to recognise a worm from a twig than to constantly know which part of the design you're in when some or none of the design is currently visible.

You're thinking about it backwards. You are thinking in terms of how useful something is to the species. Evolution doesn't care about that.

Human males aren't excited about female breasts because of great design. It's just an absurd part of genetic drift. A woman with big breasts have a much easier time finding a strong mate in spite of having a feature that is 100% a pure handicap.

Real time pattern recognition is hard to programme.

The intricacies of the puffer fish is probably sexually selected for. Because only a healthy puffer fish male would have the energy and skill to pull it off. So same deal as with peacock feathers. Which would constantly push puffer fish evolution towards greater and greater intricacies. Until we reach the limit. Which is the point where making the pattern will help kill the fish, due to over expenditure of energy.

BTW you can also train a chimp to do complex things but I don't think people are able to get a chimp to draw a reasonable circle. And drawing a circle is a fairly simple behaviour. (though it takes a lot of training to get right)

Chimps are much stronger than humans. But here's the cool part. Humans and chimps have about the same amount of muscles. A muscle isn't just a muscle.

Chimps have muscle attachments that are located farther from the joints point of rotation. It gives them more torque when activating their muscles.

The price they pay for it is with dexterity. Chimps don't have nearly the same fine motor control humans do. Humans are specialized tool users.

It was a long way of saying that chimps can't draw for shit, and never will. They don't have nearly enough motor control to paint anything.

Elephants in contrast have amazing motor control in their trunks. It's way easier to teach an elephant to draw.

 

[DrZoidberg]

Well said. Plus many of these behaviors result from a general principle seen in many places in nature - "Simple rules applied locally". Conway's Game of Life is non-natural example. Flocking behaviour in Starlings looks remarkably complex, but is a small set of simple rules applied to the nearest 7 neighbours.....it also looks beautiful.

ETA: the Mandelbrot set looks very complex too, but is a very simple rule set.

The design does look a bit like the Mandelbrot set. I guess you think that a little bit of data, like the Mandelbrot set formula, appeared in the fish's brain from a mutation and tada it could create the design.

The Mandelbrot set is the basic algorithm for anything in nature. It's repeated everywhere around us. Because evolution/DNA uses recursion to program us the Mandelbrot set is the most efficient way of doing that.

If you program a tree following the Mandelbrot set each leaf will have maximum exposure to sunlight.

This fact is yet another piece of evidence for team evolution

 

[excreationist]

....The Mandelbrot set is the basic algorithm for anything in nature. It's repeated everywhere around us. Because evolution/DNA uses recursion to program us the Mandelbrot set is the most efficient way of doing that.

If you program a tree following the Mandelbrot set each leaf will have maximum exposure to sunlight.

This fact is yet another piece of evidence for team evolution

Thanks for informing me about various aspects of evolution but it seems sometimes you are a bit mistaken.

The Mandelbrot set algorithm involves f(z) = z² + c and you put in complex (part imaginary) numbers and see if they diverge. I'm not aware of this complex number algorithm being in nature although it is a "fractal" and fractals can be related to nature. e.g. ferns and cauliflower where it keeps looking similar when you zoom in. (and many others)
The algorithm for a "Barnsley ferm" seems to based dozens of initial values
https://en.wikipedia.org/wiki/Barnsley_fern
https://www.chradams.co.uk/fern/maker.html (as supposed to the simple formula for the Mandelbrot set)

 

[excreationist]

Anticipating the parabola of a hurled object directed towards us is advanced trigonometry in real time. It's an absolutely amazing piece of evolutionary design........
.....If you use evolutionary principles to create a programme to make a robot catch a ball thrown at it... good luck with that. Our top minds have been trying for decades. Its not going so well.

Well this is from 4 years ago: (which is better than a human)
https://www.youtube.com/watch?v=EqMPLnIRUvQ
Note that it is "taught" how to catch. I'd assume it is using a neural network rather than it using trigonometry. That would be a similar method to how a dog or a dolphin would learn to catch an object. BTW I've studied two AI subjects at university. Neural networks would be based on pattern recognition / classification and prediction. It would be a bit like how a neural network can be taught whether a picture is a cat or a dog. (note the AI I learnt hardly covered neural networks since it was from the year 2000 and before)

About trigonometry:
https://en.wikipedia.org/wiki/Trigonometric_tables
"...Modern computers and pocket calculators now generate trigonometric function values on demand, using special libraries of mathematical code. Often, these libraries use pre-calculated tables internally, and compute the required value by using an appropriate interpolation method. Interpolation of simple look-up tables of trigonometric functions is still used in computer graphics, where only modest accuracy may be required and speed is often paramount....."

Computers can have lookup tables but I don't think evolution would have a lookup table. The pattern that the neural network learns would give approximate values rather than manipulate numbers like a calculator.

....Real time pattern recognition is hard to programme....

It can be done with neural networks. Are you familiar with things like "DeepMind"?
https://www.youtube.com/watch?v=V1eYniJ0Rnk

.....Chimps don't have nearly the same fine motor control humans do. Humans are specialized tool users.

It was a long way of saying that chimps can't draw for shit, and never will. They don't have nearly enough motor control to paint anything.

Elephants in contrast have amazing motor control in their trunks. It's way easier to teach an elephant to draw.

Wow...

 

[DrZoidberg]

....The Mandelbrot set is the basic algorithm for anything in nature. It's repeated everywhere around us. Because evolution/DNA uses recursion to program us the Mandelbrot set is the most efficient way of doing that.

If you program a tree following the Mandelbrot set each leaf will have maximum exposure to sunlight.

This fact is yet another piece of evidence for team evolution

Thanks for informing me about various aspects of evolution but it seems sometimes you are a bit mistaken.

The Mandelbrot set algorithm involves f(z) = z² + c and you put in complex (part imaginary) numbers and see if they diverge. I'm not aware of this complex number algorithm being in nature although it is a "fractal" and fractals can be related to nature. e.g. ferns and cauliflower where it keeps looking similar when you zoom in. (and many others)
The algorithm for a "Barnsley ferm" seems to based dozens of initial values
https://en.wikipedia.org/wiki/Barnsley_fern
https://www.chradams.co.uk/fern/maker.html (as supposed to the simple formula for the Mandelbrot set)

It's still just expanded iterations of the same algorithm. Which is how we would expect if evolution is true

 

[DrZoidberg]

Computers can have lookup tables but I don't think evolution would have a lookup table. The pattern that the neural network learns would give approximate values rather than manipulate numbers like a calculator.

I'm pretty sure nobody has cracked that. It's still a mystery how we do it.

We have a good handle on what information is passed to the brain. It's very little information. It's a mystery how the brain can make sense of any of that, let alone track objects in flight.

We do know that nearly everything we see isn't actually in front of us. Its memories. The brain looks for some triggers in the data stream from the eyes and if there's a match it brings that image from storage.

That's why eye witness testimony is basically worthless. If you believe something you will see it everywhere. If you're a racist and you see a robery you will see a black or Arab person doing it.

It can be done with neural networks. Are you familiar with things like "DeepMind"?
https://www.youtube.com/watch?v=V1eYniJ0Rnk

Yes, but even if they make it work it doesn't prove anything.

Humans have a bad habit of describing the brain as whatever is the most advanced technology they can think of. In the early 20'th century it was a switchboard. To the Victorians a steam engine. And to the Romans a catapult. No joke. So we'll talk about the brain as a computer. It's not.

At this point we have no idea how the brain works. We're still just guessing.

 

[excreationist]

...It's still just expanded iterations of the same algorithm. Which is how we would expect if evolution is true

No Barnsley's fern involves a probability that there will be something at a point:
https://en.wikipedia.org/wiki/Barnsley_fern#Syntax_Examples

The Mandelbrot set involves a simple but fairly CPU intensive part-imaginary formula that "complex" (real & imaginary) numbers are put into it you see what happens when you keep on putting the output back into the formula. When you zoom in there is a huge amount of variety while in the fern algorithm it just looks like the simple fern pattern repeated.

 

[excreationist]

....Humans have a bad habit of describing the brain as whatever is the most advanced technology they can think of. In the early 20'th century it was a switchboard. To the Victorians a steam engine. And to the Romans a catapult. No joke. So we'll talk about the brain as a computer. It's not....

I said "neural network" not computer. You don't seem to know what a neural network is. A computer traditionally does one thing at a time, step by step (though now there are threads and multiple cores). A neural network has large numbers of neurons that each connected to large numbers and many can fire simultaneously. A neural network has weights/thresholds where it is hard to understand what is going on.

This information can be from training where you train it to tell the difference between a picture of a cat and a dog. On the other hand it is easy to understand how a computer works. It has instructions that operate on data.

....It's a mystery how the brain can make sense of any of that, let alone track objects in flight.

Since they can teach/train that AI with a robot arm to catch objects, I think they've got a good idea of the process involved.

It can be done with neural networks. Are you familiar with things like "DeepMind"?
https://www.youtube.com/watch?v=V1eYniJ0Rnk

Yes, but even if they make it work it doesn't prove anything.

The idea behind artificial neural networks are inspired by what we know about how brains work. And it turns out that it can allow AI that can beat humans - e.g. in the game of "Go"
https://en.wikipedia.org/wiki/AlphaGo#Algorithm

At this point we have no idea how the brain works. We're still just guessing.

Making some guesses doesn't prove we have "no idea". Maybe you have no idea... after all you think catching a ball (which you can train a dolphin to do in a fairly short amount of time) is "an absolutely amazing piece of evolutionary design" while you say:

"The puffer fish pattern is just a set of pre-programmed instructions that it has plenty of time to draw. Fish have a swim bladder which detects movement and direction. Fish have a sense of where they are and which direction they are facing at all times.

If you use evolutionary principles to create a programme that paints a similar pattern, we could bang that out in minutes. It's pretty basic."

Note that the design is three dimensional and is worked on for about a week. There are also water currents that it needs to fight. But unless you understand neural networks I don't think you will have a hope of understanding what I'm trying to say.

 

[DrZoidberg]

...It's still just expanded iterations of the same algorithm. Which is how we would expect if evolution is true

No Barnsley's fern involves a probability that there will be something at a point:
https://en.wikipedia.org/wiki/Barnsley_fern#Syntax_Examples

The Mandelbrot set involves a simple but fairly CPU intensive part-imaginary formula that "complex" (real & imaginary) numbers are put into it you see what happens when you keep on putting the output back into the formula. When you zoom in there is a huge amount of variety while in the fern algorithm it just looks like the simple fern pattern repeated.

So that's your answer. That's not how nature did it. The Barnsely equation can be a description of how a fern looks. But it can't be a description if how nature did it.

 

[excreationist]

....Real time pattern recognition is hard to programme.....

More neural network stuff - it describes the photos after being thoroughly trained:
https://machinelearningmastery.com/how-to-caption-photos-with-deep-learning/

Note this doesn't involve the computer being explicitly programmed by a programmer how to recognise all those objects - it is just "trained"

 

[excreationist]

....So that's your answer. That's not how nature did it. The Barnsely equation can be a description of how a fern looks.

It's a group of equations really:

But it can't be a description if how nature did it.

The fern wouldn't use the exact same algorithm (using matrix maths) but they'd have some similarities since the output looks pretty much the same.
My point is that the nature of the Mandelbrot set algorithm is completely different. Also ferns have a lot of self-similarity while the Mandelbrot set has a huge (almost infinite?) amount of variety.