The Unknown « Robot Claw

Archive for the ‘The Unknown’ Category

Is This Real Life?

Tuesday, May 3rd, 2016

I watched an extraordinary film recently called Welt am Draht or World On A Wire.

It’s a 3.5 hour two-part film made for German TV in 1973. It’s beautifully shot and acted, and one of the more watchable films I’ve seen in a long time. To say I enjoyed it greatly would be an understatement, and I believe I would have done so even if it didn’t have such a remarkable story.

Spoiler alert! I’m going to spoil the film now, so skip to after the next image if you want to watch it without knowing what it is about. Note however that the remainder of this blog will deal with the same issues covered by this remarkable film.

The film is about an IBM-like company (named IKZ) which has created a simulation of reality (Simulacron) which contains 9000 individual simulated humans that are used for consumer trend analysis (such as to determine what fashions may be in vogue years later). One of the programmers kills himself under mysterious circumstances after claiming to have learned a terrible secret, and his replacement (Stiller) finds himself in increasingly bizarre situations as he attempts to determine exactly what led to his predecessors suicide.

The truth is – and here’s the big spoiler – that the world of IKZ is itself a simulation, and none of the inhabitants knows. The dead programmer found out, and in time Stiller finds out himself, and races against time (and the clever and subtle changes being made to the world by whoever is simulating it) to somehow find a reason to keep living now that he knows his life is nothing but data in some computer ‘above’ his reality.

As I said, this film was made in 1973. Even more remarkable was that it was based on a novel from 1964. It’s fantastic: watch it.

Ask yourself: “Am I real, or just a simulation?”

The idea of our reality not being real is hardly new, and for centuries philosophers have debated the nature of reality and whether anything actually exists at all. But in the last few decades the idea that our reality may be a simulation has slowly been gaining some form of credence. This paper in particular, has helped drive the argument. This is not crazy-talk, as recently as a month ago the American Museum of Natural History had a well-attended talk (moderated by Neil deGrasse Tyson) on the topic (only one of the five debaters rejected the possibility).

I’ll skip to the end before I throw out a few considerations: the answer seems to be ‘maybe’. We may one day even be able to prove it is true. In fact it is unlikely – even impossible – that we could ever prove it isn’t true. So we may have to accept the possibility that we are living in a simulation and consider from what effect – if any – this has on our lives.

Bostrom’s argument supposes only one of these is true:
1) No civilization ever becomes able to simulate the universe
2) No civilization that can ever does simulate the universe
3) We are living in a simulation

One of the logical conclusions of his argument is that if we fit into one of the above criteria, that one must be true. Right now we’re in catgeory 1, with neither the ability nor the means to simulate reality. And that’s supposing that reality is even calculable, which is a big unknown due to the probabilistic nature of quantum mechanics and the fact we currently don’t know much about things like gravity, and dark matter (ie. fundamental parts of reality).

But humans are smart, and we learn how to build more powerful computers every year. One day, presumably, we may have the means to simulate ourselves, at which point we’ll enter category 2. When we’re at the point where we can simulate reality, will we? It seems to me this is more a sociological question than a scientific one, but I can’t think of many times in the history of humanity where we’ve developed the ability to do something (non destructive) and not done it. So it is at least likely that we will build the simulation and then turn it on.

At which point we must face the fact that if we can simulate reality perfectly, there is no logical argument against us being simulations ourselves. So if we ever get beyond category 2 and create out own reality sim, we are simultaneously proving that we ourselves are simulated.

There are other possible proofs. One idea is that a 3D reality simulation must have some discrete co-ordinate system (x/y/z) which would have an effect on EM wave transmissions and background cosmic radiation. There are claims that evidence to support this has already been found.

A wilder consideration is that the second law of thermodynamics when applied to the closed system that is our universe suggests that disorder should always be increasing. And yet we live in a universe full of order – in fact we see more order than disorder (ie. atoms are arranged in planets and not distributed randomly throughout the universe). This itself has led to endless philosophy (this will bend your mind) but one theory may be that our simulation has been tweaked to accommodate us in (macroscopic) violation of the second law. In other words the simulation agrees entirely with the laws of physics only as far as it must to allow our universe to exist in the first place.

Going further down the rabbit-hole, the logical question is ‘how’? How would it ever be possible to build or power a computer able to do this, much less code the simulation itself? We can’t know the answers, and if we are simulated likely never will, but from the point of view of a simulation we may create certainly one must ask “Why simulate everything when you can just do one person?”

Consider that for a moment. Maybe reality is not a simulation. Maybe just you are? Maybe just I am. Maybe none of you exist, and the comments you leave on my blog (or texts you send me) were just generated by a computer to make the world I perceive seem real? Maybe everyone I interact with only exists while I interact with them? It may be that reality isn’t simulated, only my reality is? (You may want to read up on the Bishop George Berkley at this point.)

Which brings me to: If we are simulated, so what? The answer is it doesn’t matter. Assuming the simulators continue to let the simulation play out without interference (or at least tweak the code to hide their interference) then it has no affect on our lives at all. What we don’t – can’t – know won’t hurt us, you could say.

My Dad at this point is thinking another obvious question: who are the simulators? If we are simulations who built the computer and who wrote the code? This is for some an uncomfortable element of the argument, since it blurs the lines between science and faith.

Many years ago in a quantum physics text I read a quote that I will paraphrase here: “Science should be careful of looking too closely else they may find God”. It seems to be this is just as relevant to the simulated reality argument.

(If this topic interests you, you may also want to read about the idea that our universe itself is a computer and another theory explaining the apparent 2nd law violation.)

Posted in Art, Movies, Puzzles, Science, Society, Tech, The Unknown | Comments Closed

Icehouse

Tuesday, April 26th, 2016

It’s always cold inside the icehouse
Though the rivers never freeze
There’s a girl inside the icehouse
I can see her clearly through the trees

And now she’s dreaming of a new love
And she hopes he’ll be there soon
But she’s got so long to wait for him
Because he needs another year to get there

There’s no love inside the icehouse

The devil lives inside the icehouse
At least that’s what the old ones say
He came a long time ago
He came here in the winter snow
Now it’s colder every day

She’s still dreaming through the summer
And she’s hoping through the spring
She says she’s got no time for winter nights
She doesn’t notice as the days grow darker
She can’t remember getting any older

There’s no love inside the icehouse

Words (c) Iva Davies

Posted in Art, Music, The Unknown, Time | Comments Closed

In The Cards

Sunday, March 6th, 2016

Go and get a deck of cards, shuffle it, and deal it out in the order you shuffled it. You may find this hard to believe, but it’s extraordinarily likely that no one ever has shuffled cards into the same order you just did.

You’ve probably heard this before, since it’s one of those quite interesting facts that does the rounds. But in case you haven’t, the reason is that the amount of different ways a 52 card deck can be ordered is astonishingly high. The number is so big it’s difficult to parse:

8065817517094387857166063685640376
6975289505440883277824000000000000

Yes that’s one number written across two lines. To get an idea how large this is, it’s significantly higher than the number of stars in the universe, the number of atoms in your body, and the number of seconds that have elapsed since the universe was created in the big bang.

I’ll take it one step further. Depending on who you ask, between 60 and 108 billion humans have ever lived, so we can use an average of 80 billion. Applying the Doomsday Argument to this average suggests that about 1.2 trillion humans will ever live (and there’s a blog post on that topic itself!).

So in the entirety of human history, if every human ever lived to an average of 70 years and spent every single second of their lives shuffling cards the entire output of humanity would only correspond to a miniature subset of the total possible permutations of a 52 card deck (3.3E-45% to be accurate).

So shuffle that deck, deal it out, and be impressed with a creation that only you have made.

This has an interesting relevance in the field of gambling, which requires randomized deals lest the player guess the card order. Using human dealers, decks can be shuffled in a way that makes them almost completely random (although studies have shown that a virgin deck must be shuffled anywhere from 4 to 7 times to eliminate the order inherent in the way it was packaged). But these days the vast majority of deck shuffling is done by computers, and it’s not trivial to make computers do things truly randomly.

The very first computer games that included card shuffling had extremely primitive random number generation and could only return limited unique decks. Random number generators require a ‘seed’ (ie. a start value upon which all others are based) and every sequence based on the same seed is identical. Games on the Atari 2600 and Intellivision (shown above) typically used hardware values or player input (such as the number of frame refreshes that occurred before the player pushed the start button) as seeds, but even then were limited to usually only a couple of hundred unique decks. Given enough time and effort therefore, you could know the entire order of cards based upon the first few dealt.

As time moved on the algorithms became more sophisticated, and so too did the random number generators, but even then it was possible to predict deck orders if you had enough information. In 1999 an online casino, in an attempt to demonstrate their games were not rigged, actually posted their RNG code online. Someone got it, worked out how they seeded (based on the clock time, as I did in my polycap simulation), and actually wrote their own code that was able to reproduce perfectly the shuffling of the games they were playing online.

So we get to today, where RNG’s use very creative ideas to seed themselves with truly random seeds (such as using code to convert video frames captured from random Youtube videos or 1 second of white noise from a radio into seeds). But there is still a problem in that the range of randomly generated numbers is still limited to about 4E38. In short, you can’t generate a number between 1 and 8.06E67, which means you cannot generate one number for each possible deck permutation.

There are ways around this (hint: using only a single coin you can generate two random values) but it makes the task of writing a deck shuffling simulator that can account for every possibly permutation non trivial.

I think.

So as a result of this thought experiment, Bernard’s going to do it! Here’s my design document:

1) Assign all 52 cards a random number
2) Sort them
3) Output shuffled deck

It’s trivial stuff, and should only take him a femtosecond or two to implement. But the true fun is in the testing! For what I’m really interested in is how many unique shuffles are completed before a repeat occurs. Therefore the output (deck order) will need to be saved as well as the time it takes for each shuffle to be completed. Plus, since 52! is insanely large (the world will end before his computer shuffles that many times) I’d say saving the first 15 cards + the time the shuffle occurred is sufficient to do some statistical analysis.

So there you go Bernard, there’s your challenge. Write the code, run it 1.3 trillion times*/**, save the first 15 cards in each deck and the time the shuffle was performed and then analyze it to see if any sequence repeated.

Let us know the results 🙂

* I’ll assume you have a modern Pentium running about 100k MIPS, and that this code requires maybe 1000 operations to execute (a big guess there; the sort could take many more), which means about 12000 seconds or 3.5 hours per experiment. However writing results will slow it down a lot I suspect. Good luck!

** A very rough mental calculation tells me this may be a file size in the order or 17Tb. I hope you have a lot of space! Even more luck to you sir!!

Posted in Otaku, Science, Tech, The Unknown | Comments Closed

Bunyips!

Tuesday, October 27th, 2015

This time I’ll detail a uniquely Australian cryptid: the ‘bunyip’.

The origins of this creature are somewhat murky, but it is believed that in the early 1800s, as settlers migrated away from Sydney inwards from the coast and down toward Victoria, tales began to emerge about a large water-creature called the bunyip by the Aboriginals. While the descriptions seemed fantastic Europeans had already been so surprised by other unusual Australian fauna – especially the platypus – as to take them seriously.

The bunyip was said to be quite large, and while mostly docile could threaten a man and was indeed responsible for some Aboriginal deaths. Early settlers were suitable concerned about running across such a beast, especially since the Aboriginals were very scared of them. The actual appearance of a bunyip though was unclear, despite efforts by scientists (including Banks) to pin them down. It was generally believed to be semi-aquatic, large, and somewhat mammalian but with birdlike features (especially the head). What was agreed on was that the bunyip could produce a loud and alarming moan, which could be heard at night from great distances.

In 1845 a Victorian newspaper reported the discovery of bones believed to be those of a bunyip. A couple of years later the Australian Museum in Sydney even put a bunyip skull on exhibit, but it was later discovered to be a deformed horses skull. By this time, with none having actually been seen by reputable witnesses, the creature was transitioning into folklore, and it’s status as an actual living creature was fading fast.

But sightings continued including a widely publisized (at the time) account in 1852 by an infamous escaped convict who lived with an Aboriginal tribe for decades. He claimed to have seen bunyips several times, describing them as timid but dangerous creatures that inhabited lakes and preferred to eat women. He had only seem them half-submerged, and said they were covered in feathers. Aboriginals still insisted the creature was real, although accounts of it having supernatural powers made these claims increasingly difficult to believe.

By the depression, the word ‘bunyip’ had become synonymous for ‘impostor’ in Australia, and few seriously believed the creature existed. Scientists and anthropologists had come up with several explanations for the origin of the creature, including:
– Fur seals, which were known to travel far inland in some Australian river systems
– Crocodiles, which can grow to be enormous especially in northern Australia
– An as-yet-undiscovered species of otter or giant eel
– A surviving Diprotodon, which is an extinct aquatic wombat-like creature bigger than a hippo (this was a prevalent theory apparently)
– An ancestral memory of a duck-billed or other aquatic dinosaur that had somehow survived into the early Aboriginal era

Even these explanations faded in time, and these days the bunyip is considered no more real than other Aboriginal Dreamtime fauna such as The Rainbow Serpent or the great frog Tiddalik.

Those two stamps contrast the bunyip of myth with the (presumed) origin of the creature. Bunyips today exist only mostly in the world of childrens books and movies or advertising, and look a bit like this statue of one in canberra:

While still included in the ranks of cryptids, recent sightings of bunyips – or even faked sightings – are almost nonexistent. This is a creature that seems to either have never existed at all, or be so good at hiding in the hidden parts of Australia that no one believes it ever existed at all. Which theory do you prefer?

Posted in Animals, Art, Miscellaneous, Nature, The Unknown | 2 Comments »

Wildlife XI: Return of the Robo Camera

Monday, September 14th, 2015

This time I left the camera out for about 6 weeks, and it took a staggering 800+ photos. Incredibly the batteries still had 30% charge left after all this time (which included a series of extremely hot days). I’m really pleased with this camera overall.

Looking through all those photos for hidden gems was a bit of a drag though, since about 775 of them had nothing of much interest 🙂

Let’s start with the squirrel shots for mum:

Doing a quick tail comparison tells me the first and last shots are of different squirrels, which is entirely likely given the number of them living around here. The selfie shot is funny isn’t it?

You can see I switched the position of the camera after the first month. This was to catch a very particular target on film, but I’m sad to say I failed. I’ll try again though, and I’ll leave you in suspense as to what I’m after…

The rest of the shots were of the usual cats: