It sounds like the only way to "beat" this is to try to find some unrelated feed of noise to base the key presses on. Because when you actually do think about it deliberately, you fall into a pattern.
One way to do this without an external data source might be to just visualize (or look at) each object in turn clockwise of a room you are familiar with. Or use any other random source of data. Then come up with a rule that applies about half the time to some objects and half to the other. Or rather, come up with the rule before thinking of the room.
Maybe count the number of letters in the name of the object. If it's even, hit the f key, and if odd, hit d.
Granted, I only tested this for a few minutes, but I found that if I thought about "repeat" vs "switch" instead of "d" vs "f", then I could reliably get under 50% prediction. Sure, the prediction rate was high 40's, but that at least means I won!
A lot of it comes down to avoiding auto-piloting into a 50/50 distribution. Subconsciously, "random" means approaches 50/50 since that's how it works, right? I'm sure given enough time, it'll be able to predict me with >50% accuracy, but that's entirely okay. Random doesn't have to be fairly distributed. As long as it doesn't get to 100% then I still have personhood :^)
From the Github description, a quote: "In a class I taught at Berkeley, I did an experiment where I wrote a simple little program that would let people type either “f” or “d” and would predict which key they were going to push next. It’s actually very easy to write a program that will make the right prediction about 70% of the time. Most people don’t really know how to type randomly. They’ll have too many alternations and so on. There will be all sorts of patterns, so you just have to build some sort of probabilistic model. Even a very crude one will do well. I couldn’t even beat my own program, knowing exactly how it worked. I challenged people to try this and the program was getting between 70% and 80% prediction rates. Then, we found one student that the program predicted exactly 50% of the time. We asked him what his secret was and he responded that he “just used his free will.”"
I can get it to about 60%. Given the algorithm used, it seems like it should be trivial to "trick" it by just cycling through novel 5-grams over and over, but to be honest I can't be bothered to check. And that would be rather contrary to the point of the exercise, if somewhat undercutting of the premise that free will is equivalent or comparable to randomness.
> I can get it to about 60%. Given the algorithm used, it seems like it should be trivial to "trick" it by just cycling through novel 5-grams over and over, but to be honest I can't be bothered to check. And that would be rather contrary to the point of the exercise, if somewhat undercutting of the premise that free will is equivalent or comparable to randomness.
Right; if you guess completely randomly it will hover around 50%, but if you choose the least-likely 5-gram from your history, then it should end up less than 50%.
yes, I've done some of these before. Ive gotten up above 200 key presses at ~50% before getting bored.
my secret was decidedly less sexy than free will though. i just had XX years of working as a statistician and having a better intuition as to what random strings and numbers actually look like :/
Same here, though I don't have any years of experience working as a statistician.
I did have a job where I had to implement a bunch of cognition tests for scientific studies. Initially I worked purely off the research papers describing how the some of the tests work - apparently by presenting a series of random numbers.
After I had the first few up and running, the feedback from the researchers was that my random numbers weren't random enough. "Weird", I thought... I had seeds for their sessions so I could reproduce the sequences they'd seen and they looked pretty random to me.
In the end it turned out my sequences were "too random", they didn't like any number being repeated, having more than 3 instances of the same number within any 6-digit sequence, and a few other "tweaks". Turns out actual randomness is confusing even for some very experienced and well trained people!
It's easy to be random if you have memorized random (or normal) data that your adversary isn't focused on. For example, early digits of pi work in most cases. Or use a generator to create a sequence to memorize.
quite slow and considered actually. i doubt i could pump them out rapid fire, but maybe i should try. i don't have anything memorised and just go off of feel of runs of numbers.
i have considered trying to memorise some algorithm, but never done it.
i assume that if anyone tried to get me to do these things with a non- uniform distribution I'd probably be screwed, but never tested it.
> Free will is equivalent to randomness with many (asymptotically infinite) degrees of freedom.
I don't think this is obviously true. I can make a random number generating device with as many degrees of freedom as you like by producing a bunch of vertically polarised photons and measuring their horizontal polarisation. I don't think anyone would describe this system as having "free will".
Conversely humans with "free will" don't usually behave very randomly. They often have reasoning (good or bad) and are able to explain why they took whatever actions they took.
Free will seems to be qualitatively different to randomness.
Interestingly, this book is from 1971, whereas a famous duck went through a similar plot already in 1952 [1]. Now, how to decide if I should add this to the Wikipedia entry of "The Dice Man"?
I think that's true, I think free will is more reason (of the general kind that's not only logical but takes into account your mind), than more randomness. Rather, the more your actions are aligned with your own good, brought by understanding yourself and reality, the more free I think you are. (to give an example: is an addicted person free, even if he is in principle doing "what he wants?", or is he trapped by his own desires?) In this sense some people would say I'm a 'compatibilist', because I think there's no great tension between determinism and free will.
Following reason is as deterministic as following your addiction.
Sure determinism and free will are compatible as long as you believe that free will is your emotional reaction to discovering what your mind has determined.
I don't agree that reason correlates with free will. Many people act irrationally. Behavioral economics was developed because "rational actor" predictions applied to large groups of people turned out to be consistently incorrect.
One can willfully be irrational.
I do agree with you though, that free will doesn't necessarily mean "doing what I want" (which is a libertarian principle from the French Revolution), but instead, is freedom from desires. That's true freedom, but is rarely found and achieved only after years of ascetic practice.
Don't confuse "having reasons" with "being rational". Free will is a freedom to act according to your internal reasons, with that process of deliberation over your reasons being responsive to various types of feedback. Those reasons don't have to be rational.
One step back please, the "free will" was an ill-conceived start. The core claim is about how random is human will. Throwing in the word "free" adds confusion.
The will of an addict is a bit less random, okay. Is there anything quantitative going on here besides "some stuff can increase randomness" and "some other stuff can decrease randomness"?
My claim is that randomness is a dead end. It's irrelevant. But I do think free will can be a useful concept (with little relationship to randomness), although it's too complicated for a short comment I think (a complex quantification of how you can realize good things for yourself and the world).
That said, I think this is a nice experiment exactly for surfacing this kind of questioning.
> Free will seems to be qualitatively different to randomness.
There is no such thing as free will, so any attempt to define it will fail.
If we naively define it as something that isn't just pure randomness, but also something that isn't pure causality, then what is it?
Either things are causal, in which case, you can replay everything and arrive at the same point. That doens't fit an intuitive definition of free will.
Or there is randomness, play the same sequence of events again and some inherent randomness will cause a different outcome. That also doesn't fit any intuitive sense of free will.
So, since we can't even define it, it is meaningless to assume there is free will in the first place. We can talk in terms of causality or randomness.
> Either things are causal, in which case, you can replay everything and arrive at the same point. That doens't fit an intuitive definition of free will.
I think this is somewhat inaccurate. I'm not sure if I personally agree with it, but at the very least compatibilism is a serious philosophical viewpoint and shouldn't just be dismissed like this.
In any case I think the concept of "free will" is pretty much independent of randomness. If you provide me with some source of randomness (e.g. some quantum random number generator) I don't think I get more free because of it.
There's an incongruity between a "mechanistic" model of causality and how we actually live, and actually should live.
Your explanation is a good response to the posted experiment, but I don't think it's a good response to common sense, especially when we consider important (largely pragmatic) concepts like agency and responsibility.
> If we naively define it as something that isn't just pure randomness, but also something that isn't pure causality, then what is it?
Read up on Compatibilism. Free will can be defined in a way that's compatible with determinism. A deliberation over values resulting in a justification for a choice, aka "reasons" or "will". When you are "free" to act in accordance with those reasons, ie. you haven't been coerced into acting against your own reasons, then you've made a choice of your own free will. Simple and totally compatible with determinism.
> I can make a random number generating device with as many degrees of freedom as you like by producing a bunch of vertically polarised photons and measuring their horizontal polarisation.
Unless I misunderstand you, this is simply not true. Your device will be exceedingly easy to model, it's just a simple sum of i.i.d. random variables and subject to central limit theorems. (While humans definitely aren't.)
While the sum of many i.i.d. random variables may be easy to model, that doesn't mean that everything involving i.i.d. random variables is easy to model.
I just typed randomly and got 50%. Does that mean I've got more free will than you? Is my will random? I've got my doubts, but I don't have anything better either.
Careful. If you get significantly below 50%, then you could be predicted by the negation of this algorithm. To truly beat it, I think you want to be right around 50%.
The prediction algorithm is actually very straightforward[1]. It's a fun exercise to write a sequence generator specifically to defeat it. Here is a sequence that can get the prediction rate down to 13%:
You can improve that by around 13% if you soften the inequality (your > should be >=, I believe)
table = {}
s = 'ddddd'
for i in range(100):
if i%5 == 0:
print(s)
state = table.setdefault(s, {'f':0, 'd':0})
s = s[1:] + ('f' if state['f'] <= state['d'] > 0 else 'd')
state[s[-1]] += 1
>The prediction algorithm is actually very straightforward[1]. It's a fun exercise to write a sequence generator specifically to defeat it.
if you are inspecting the prediction algorithm, and you want to specifically defeat it, doesn't that mean copy the prediction algorithm and throw a "not" on it? It's the old halting problem trick...
maybe I don't understand, 50% is the goal? when I played I thought % was how often it predicted what I pressed, and even though 0% is 100% wrong which is 100% right in a parallel universe, it would still represent me winning.
in any case, my post was serious, regardless of the % number, if you are inspecting their algorithm and writing your own, I can't understand why you wouldn't use theirs in yours. Finding an isomorphism might be fun, but it's not necessary.
50% means it is "no better than random" while 0% means it is accurately predicting what you won't press and 100% means it is accurately predicting what you will press.
> it is accurately predicting what you won't press
That is not what it's predicting. You have changed the definition of predict so that an incorrect prediction can be misinterpreted as a correct prediction.
If you have two choices ... and I always predict the wrong one ... another way to say that is: "I always predict the one you won't choose." If I can't actually predict which one you choose, then statistically, I should be right 50% of the time.
So? If I can reliably lose every basketball game, that doesn't make be perfect at basketball.
To put it more plainly: if you were wagering on the outcome, would you rather win 50% of the time (EV=0) or 100% of the time (EV=max)?
100% misprediction is only relevant if a 3rd party adversary is observing the game. In the absence of such an adversary, being reliably wrong it worthless.
You are comparing to things that don't even make sense in the current context.
You have two choices.
I predict correctly or I predict incorrectly. If I'm randomly guessing, I will be right 50% of the time (effectively, this is a coin flip). If I always guess incorrectly, then I'm correctly guessing the incorrect choice. If I always guess correctly, then I'm correctly guessing the correct choice.
I'm trying to explain why 50% is your target for "fooling the algorithm" and not 0%, which is not fooling the algorithm at all. There's no reason to be obtuse here, I'm legitimately trying to help you understand some statistics... this isn't an argument, it doesn't matter "what you believe" ... this is math.
Another way to put it is that there obviously exists an algorithm that can predict what you will choose. It just isn’t this one, but the opposite. The entire point of this exercise is to prove that you have free will. If there obviously exists an algorithm that can predict what you will do, you don’t have free will, do you?
Thus “random,” aka, 50% is the target: this algorithm cannot predict what you do; you have free will.
50% or 0% are both interesting goals, but the headline implies that the website is some test of free will. If you get 0%, then it shows there is an algorithm that predicts your performance, and the Kolmogorov complexity is finite. That algorithm is the inverse of the one the website is using.
If we conflate "free-will" with "ability to generate truly random sequences" then the goal should be to generate a completely incompressible, unpredictable sequence, which 50% would probably be closer to.
> If I can reliably lose every basketball game, that doesn't make be perfect at basketball
What they are saying is that when you have an algorithm you can reverse it to do the opposite. If you could reverse your basketball skills exactly that would be true for your statement.
People who are good at chess sometimes like to play reverse chess where the goal is to get checkmated. If you are playing against another good player it can be quite difficult -- as difficult as winning under the normal conditions.
Maybe the issue is that we're being too discrete, thinking in terms of absolute predictability or absolute unpredictability.
Ultimately a 50% prediction rate means that you are as unpredictable as is possible, the state of maximum entropy. Any deviation from 50%, towards 100% or 0%, is a state of lower entropy.
You can forget the free will and just consider a fair coin toss. If someone had the ability to always guess a coin toss incorrectly, every single time, then you'd know that this person possesses something absolutely incredible and unique, rather magical. Similarly if they guess a coin toss correctly every single time they would also possess something magical.
It's the person who can only correctly guess a coin toss 50% of the time that is uninteresting, ordinary and possesses absolutely no special or magical insight. But deviating from 50% would require having some knowledge about the coin. A 0% prediction rate would require having total knowledge of the coin, so would a 100% prediction rate. It's 50% that requires having no knowledge whatsoever.
If we forget free will, then I will grant your position regarding statistics.
But suppose you could collect every bit of evidence in the universe and use it in a massive calculation to guess what my next decision would be.
This would be a test of whether I had free will, or whether I was simply doing what the universe made me do. If I wanted to demonstrate free will, I wouldn’t be aiming for random. I’d be aiming to make your guess wrong as often as I could.
If I could make your guess wrong 100% of the time, this would not mean that you could simply flip the bit on your guess next time. You can modify the calculation after every guess and it wouldn’t matter. I’d still do what your calculation thought I wouldn’t based on previous history. If I could do that, that would be a magical thing. Yes. I would have demonstrated completely free will according to this test.
This is why we call rand(), not freewill().
Free will is not random at all. It means you wear the shirt you want to wear, not a completely random shirt.
Oh, and before somebody objects, yes this is a terrible test of free will. Because whether I want to score high or low on the test or in the middle is up to me if I have free will. ;)
There really isn’t a “yes but” here, this is stats 101. Welcome to the class.
A 0% prediction rate means that you are 100% predictable by simply taking the opposite of the algorithm’s input or output. Thus, you have no free will (according to this tool).
The 50% goal is against an adaptative adversary. For some reason, Zero Knowledge Proof[1] came to my mind, so I interpreted the goal as reaching a 50% accuracy on the oracle side.
Your interpretation of the goal is also valid, but not in a cryptography context, for example.
Although I do believe we have free will, I don't agree with this specific example: looking at the algorithm used to predict key presses and just doing the opposite of what would the algorithm predicts, is letting the algorithm tell you what to do. This is related to inverse psychology.
Even a Turing machine has free will. You can’t find out what it will do without running it, in a way beyond all questions of technical capability. Mechanistic attacks on free will died with godel.
I would say it's a measure of probability. Of course measuring probability in non repeatable situations gets tricky.
But, if I wrote there is a 90% probability that lupire will write an application to 'beat' the free will program, and then you did so, it should at least give you pause.
I say that I have free will because my decisions come from myself and not from something outside me. Sure, the context in which I live can influence my decisions, but the final choice is mine and it can happen that I take a decision that goes against what the context (society, family etc.) would expect from me.
Yes, I had the same discussion a couple years ago and I concluded it by saying that it boils down what's myself: is it my brain? My consciousness? My soul? But this is a related but different problem.
I call this the "subjective experience of free will": that is, it does seem to be a thing that people perceive, and it's such a convincing effect that we often turn it from a subjective experience to an objective belief (that free will exists).
Sure, I never said I have an objective proof. But there's no proof of the contrary either, so the two positions are at the same level, objectively speaking.
> With sufficient technology, we could take a snapshot of your brain and predict exactly how you respond to any given stimulus with 100% accuracy.
This is assuming that intelligence is computable, but no one has proved it yet. And doesn't even get into the problems of quantum physics. Unfortunately, the real illusion is determinism.
I mean you've got a body, too, right? So the moment the environment changes so does the mind state. So unless you've come up with a method of modeling all of those influences... And food, and hormones, and... And then you've got stuff like cosmic radiation slamming into your DNA causing dsDNA breakage and that precipitates into apoptosis and suddenly the model is off due to some completely unpredictable shit, 'cause were talking about interactions that are occuring on the scale of 6.022*10^23 per mol, and a lot of mols and a lot of different species of matter themselves subject to probability moreso than determinism at any reasonable scale.
If sufficient technology were to emerge, I expect that under no circumstances would it ever be considered as a worthy project to invest in. But the idea of such technology ever emerging itself is frankly worthy of ridicule.
Does it count if you allow the test to be accurate but you still maintain control over when it is wrong?
I just spammed F and randomly hit D and it pretty much never predicted the D.
Which I think captures a fundamental flaw in algorithms vs humans. Humans know how to mask their signals within noise. I'm inclined to believe the generalization that algorithms that aim for accuracy are highly susceptible to being manipulated by humans who are allowed to have free will over the inputs.
It's not really a flaw in algorithms nor a skill in humans.
It's (also?) a weakness in humans: someone mostly does X, we lean very strongly into the assumption that they will never do !X. This is one of the most basic elements of human-to-human deceipt since ... well, since the dawn of any kind of recorded/remembered human history.
This is one of the kind of tests we can perform on an individual in the future if we suspect they are artificial. A standard human should be fairly predictable, but an artificial person is more likely to produce true random patterns when asked to type randomly.
I feel like there's an "aesthetic" to randomness, which I see in examples of Japanese wabi sabi. I don't know how to describe what that looks like in keypresses exactly, but if I imitate that, I'm able to keep the rolling average at <52% over a few hundred keypresses done with my eyes closed. I hit 51% three times in a row.
I question the relation to free will however--will seems more related to preference than randomness to me, and a better example of that would be something like "I prefer the f key". Conflating randomness with free will has always seemed like a weak attempt to argue for the existence of free will to me.
Maybe I've just seen way too many random sequences, but I was able to hold it around 40% for a while. True randomness tends to generate sequences that are slightly "uncomfortable" or "inhuman" -- I can't think of a better way to describe it. So I just kept trying to generate the least-comfortable sequences I could think of, and it worked alright. There should be a lot more long runs than you'd imagine.
I can look at the digits of PI and press d and f depending it's odd or even. So what does it prove? Nothing. Being able to beat this algorithm doesn't mean that I have free will, and choose to fall into patterns (knowing the PI solution exists) doesn't mean I don't.
If your behaviour can be predicted then you don't have free will.
Obviously you can argue whether this really predicts human behaviour or even the above implication itself (for example I think compatibilism kinda disagrees with it) but I'd say it is related.
From my point of view, that someone can predict what I will do from past behavior does not imply I don't have a free will.
Only way that would prove I don't have a free will is someone who would predict my behavior without knowing anything about my past, like a fortune teller that would tell my friend (because if fortune teller tells it to me it might be self fulfilling prophecy :D) something I would do; without having any interaction with me.
I think it is true. If something could predict all your actions with 100% certainty then you have no free will - you have no ability to decide for a different outcome than the one that is predetermined.
Although if it wasn't like that and you could do truly random decisions, I'm not sure what that says about your free will either.
If I had to guess, I don't think I have a free will. In my opinion my actions are a deterministic function of the state of my mind (including memory, personality) + input. I have no choice in the matter, only an illusion of one.
The prediction machine in the though experiment would make the prediction after it knows the result of the coin flip and before the action is done by the subject.
> In my opinion my actions are a deterministic function of the state of my mind (including memory, personality) + input. I have no choice in the matter, only an illusion of one
It's important to note that as far as materialist physics goes, "you" are literally "your state of mind (including memory, personality) and input". The fact that this system (i.e. "you") produces an apparently deterministic result doesn't mean that "you" didn't do it.
The common confusion comes from the fact that systems cannot in general predict their own outputs without going through the required computational process. I think it's related to the concept of "computational irreducibility". So "you" as a system needs to go through the set of motions to figure out what you eventually decide to do. Before the deliberation you can't predict the result, and thus it "feels as if" "free will" is a messy indeterministic process. But that's not a given.
Weird things happen when some other system can predict "you". The problem there is that if "you" can be simulated, you don't know which "you" is "real". But the mere fact that your state of mind and inputs will deterministically produce decisions doesn't mean "you" don't have a choice. In fact it is opposite -- "you" do have a choice, and _only_ the system "you" makes the choice (and thus it is determined solely by "you", hence, deterministic!)
I think that is exactly my stance. You are a system that is complex, iterative, and chaotic (in the mathematical sense). You are probably deterministic, but the only way you can be accurately predicted is by simulating your entire brain and sensory experience, along with memories, to the degree it's no longer a simulation; it's just building a copy. At that point, it's not a prediction; it's creating a copy of you and seeing what they would do.
I'd add that the devil is in the details of the "copy".
To accurately simulate/copy the "sensory experience", you'll have to simulate/copy the entire "observable universe" of the person. Unless you just want to predict the behavior of a person locked inside a solitary, dark, sensory deprived room, the practical implication is that you need to simulate/copy the entire observable universe to run an accurate simulation/copy.
Which is why I always tune out when Laplace's demon is brought up. Granting such an obviously impossible premise means all kinds of insane/wrong conclusions can pop out the other end.
There's admittedly a relationship, but it's very subtle and nuanced.
For example, you can predict that I'm going to pay my rent with > 90% accuracy, but that has nothing to do with whether I'm paying out of my free will or not.
A Marxist will argue that it's the oppression of the bourgeoisie that makes me pay, and a free market capitalist would counter that I made the choice "freely". But the truth is somewhere in between, and probably depends "subjectively" on how the person "feels" about doing it.
Of course if the prediction for all actions (not merely within an experimental setting) is close to 100% then there's probably something weird about it, but I don't think that's remotely possible. Not with current tech, and unlikely with future tech either.
The question is never if we have free will or not. The question is if we are in a state where free will is available or not.
For example, when I’m aware that there is space around myself, then I’m also aware that there are thoughts, feeling in the body, breath etc. In this state, I believe, it’s possible to choose. In this test I can choose the only press the other key and this algorithm doesn’t do so well.
Then something triggers my attention, I react, I forget that there is space, breath, thoughts etc. and I start to execute my conditioning, which is highly predictable and free will simply isn’t available at that point.
Then eventually something pulls me back to the presence, awareness. I remember there is space, breath, mind, feelings, emotions. I’m home. I’m free.
The universe led up to this moment where you fool yourself into thinking you have free will just cause you sat still for a bit. Doesn't mean much.
Think about it, if the universe is all predetermined, how peaceful your mental state is or isn't has no impact on it. We all either have free will or not, but breathing deep isn't gonna trigger it.
Free will is any random process with an asymptotically infinite number of degrees of freedom, by definition.
Are humans such a process?
Who knows, but attempts to model individual humans have certainly failed before. (We need lots of humans in aggregate to successfully model them, once the central limit theorem starts working.)
I challenge you to get any more than a single digit percent of any representative sample of people to agree a random process is free will by definition. It's one of exceedingly few definitions I've heard in decades of discussing the subject that I'll agree doesn't immediately make free will impossible, but at the same time it's one that too me at least exemplifies something that is neither free nor will.
No, I don't. I can reject the notion of free will on the basis that I would never consider a "random process" either "free" nor an expression of "will".
When you say "you fool yourself" then I have to admit that I don't know who am I or how I could define myself anyways. I have been sitting with that question every day for about 12 years now.
Only reason you believe there is a mind is because the mind says it exists. But have you investigated where is that mind and who is that which mind if speaking to?
The thing is for this discussion, analysis of the ego and the role of "the mind" or however you want to call it, I simply don't think it matters for the question of free will. It's one of causality at the beginning of the universe for me. The mechanism through which humans would drive their thought process or behavior is besides the point if the cause is the initial state of the universe and how all the particles and forces interact from that point onwards. A rock on the ground's "life" (path through universe) would be as predetermined as ours, it wouldn't change it if the rock had thoughts or an ego.
The discussion about the ego and "the mind" and the "who you are" in that sense are then still useful for the human to feel comfortable in feeling like they have agency, and I am aware of Buddhist philosophy in finding peace through detaching from our inner voice and gaining wisdom from not letting ourselves being driven purely by what we think we are or "what the ego says we are", but I think it's besides the point in this context (but very useful to include in a balanced "diet of thoughts").
Maybe the ego is an evolutionary defence against nihilistic traits so we all don't kill ourselves by realising none of what we do is any agency and it's all predetermined by the start of the universe. But anyway this part of the discussion enters into a rabbithole detached from the core discussion of there being free will or not, at least to me, because I cannot accept that a core property of the system "universe" and the definition of it's causality or not could depend on the thought process that goes through minds of humans in some remote part of that very universe. A detail about the higher level thought process of some "blobs of particles" in the universe don't define the way the universe itself works. Our relationship to the self doesn't have anything to do with physical reality of if-this-then-that. In other words the causality doesn't care what we think of it, it's either all pre-determined or not, our opinion doesn't matter.
If I ask you to choose A or B, and let you take as long as you want to make this decision, at the point where you make the decision, you still have no idea why you chose what you chose. It is ultimately mysterious to you and anyone else why you chose what you chose. If you had any reasoning, it is still mysterious why that reasoning caused you to choose what you chose and why it didn't have the opposite effect?
The impetus to choose A or B in this case arises from the dark pool of the mind, sneaks up behind you and you mistakenly identify with this decision. You simply cannot think to think what to think.
You say “if I ask you to choose” but if there is no free will then I cannot choose.
Also if there is no free will then we cannot judge anyone, no matter what they did, because they didn’t have choice.
Actually my argument is that the best evidence for free will is that the world is such a mess. Without free will we would live in perfect harmony with nature, like all the other species do. We are so free that we can even deny the existence of our freedom, hence avoid taking responsibility of our actions.
> Also if there is no free will then we cannot judge anyone, no matter what they did, because they didn’t have choice.
This doesn't make sense. If there is a rock in the middle of the road, it would make sense to remove the rock from obstructing the road. Are we "judging" the rock in this case?
OTOH, if the rock happen to cause a massive accident, it seems absurd to punish the rock and try to cause as much suffering to the rock to compensate for all the suffering it caused.
> Without free will we would live in perfect harmony with nature, like all the other species do.
Harmony how? Like all the viruses, parasites, do you know what dolphins to pufferfish? What about cannibalism? Or infant mortality in nature?
> We are so free that we can even deny the existence of our freedom, hence avoid taking responsibility of our actions.
Realizing that free will does not exist does not mean that we don't have to live life anymore. We still have to live it. What it does it remove a lot of unnecessary baggage. Is it necessary to punish a rock for falling over a person? An elephant going on a rampage? What about murder?
When you realize that murder (or whatever immoral actions) is just the totality of the universe, you can think of any bad actions as just a very complex case of a brain tumor. You wouldn't blame a person for committing homicide once we learn that they went on a rampage because they have a brain tumor. But if we know enough about the universe, we will realize that pretty much anything that a human ever do is just another complex case of a brain tumor. Why did they do what they did? Well, here's a trillion page explanation of exactly why it happened starting from the Big Bang. Does it make sense to blame them now, still?
My experience is the converse: when distracted, I assume I am the author of my thoughts, the doer of my actions, but when I really pay attention, I notice that thoughts and intentions arise unbidden.
I identify with consciousness, and the origins of these intentions is unconscious, therefore I conclude that the intentions do not originate from “me”. It’s still my action, my “choice”, my kamma, but it’s not my “free will”.
What exactly do you mean by choose? Are you saying there are actions that have no discernable cause? That there's a state in the brain that is unrelated to the state of the brain a brief moment ago? If so, is that a type of telekinesis?
Yeah, I think what we experience as free will takes place mostly in what we think of as our brain pretty much by definition. We might be a brain in a vat or even a Boltzmann brain, but whatever it is, the organ that constructs our reality is a brain.
Yea I also like to say that if my will is truly free, it's not under anyones control.
One thing I want to add: I think will and intention is also a question of scale or perspective. Bacteria is capable of making "their own" decisions but they probably cannot dictate which direction body is going to walk.
In similar way Planet Earth is under influence of the Sun, and the Sun is under influence of larger forces of the Milky Way and so on.
You could say that every level of existence is probably under a will of some level above and maybe each level can impose their will on levels below them.
From this perspective we could say that pure awareness is beyond physical body, brain and mind, so then if you want to really will your body it must be done from the state of pure consciousness - and then the body, brain, mind must follow as they're level below.
Anyways, don't take me seriously, this just entertainment.
I remember reading Rand where she argued that free will existed solely in the decision to focus or not. When you focused, what you thought was determined by your characteristics and history; and when you didn't, the same, although not in the same way. But free choice manifested in the decision to intentionally focus attention or not.
I don't know that I agree, but thought it was interesting.
"Hacker's Keyboard" was supposed to let you pull up the keyboard anywhere but some system update broke it. I got around it with Teamviewer to my desktop.
I also got under 50%, started out with 10% actually, which finally went to 36%. But I feel like it's just because I threw in many of single keys in a row there.
That's what you need to do. Most people think streaks are unlikely, which biases people to void even small streaks. The algorithm expects that, so to fool the algorithm initially, you need to tilt towards streaks, and then when begins to pick up on that, tilt away.
Something about the feedback of being able to see when it was guessing wrong made me beat way more often than 50%, just by mixing it up or being a little unpredictable when I thought it was going to catch on to me or guessed a few in a row. I was consistently in the 20-30% for a large amount of guesses.
But when I started trying to type random keys as fast as possible it started getting a lot more right.
I chose not to press a key, but may change my mind at some random nexus in the future. Thus, we are at an stalemate for now, as the input event is yet to be observed. =)
Examining the code, my first thought was to throw it a de Bruijn sequence[1]: a cyclic sequence of length 2^n that contains every string of length n as a subsequence. It should (and does) learn a length-32 de Bruijn sequence after only a couple of repeats, but performs miserably on a length-64 sequence.
With a little (computer-)elbow grease, one can find a sequence with perfectly free will. After a short time, it starts repeating every... that's right, 64 characters. So of course, we want to find the shortest possible prefix.
Note that the last two lines are a de Bruijn sequence. Repeat those until you get bored. That's how you exercise your free will.
There is a deterministic bias: when in doubt, the page guesses "f". If it flipped a coin instead, a few experiments show that the greedy algorithm I and omoikane use cannot beat 25% for long.
278 comments
[ 3.3 ms ] story [ 349 ms ] threadOne way to do this without an external data source might be to just visualize (or look at) each object in turn clockwise of a room you are familiar with. Or use any other random source of data. Then come up with a rule that applies about half the time to some objects and half to the other. Or rather, come up with the rule before thinking of the room.
Maybe count the number of letters in the name of the object. If it's even, hit the f key, and if odd, hit d.
A lot of it comes down to avoiding auto-piloting into a 50/50 distribution. Subconsciously, "random" means approaches 50/50 since that's how it works, right? I'm sure given enough time, it'll be able to predict me with >50% accuracy, but that's entirely okay. Random doesn't have to be fairly distributed. As long as it doesn't get to 100% then I still have personhood :^)
I can get it to about 60%. Given the algorithm used, it seems like it should be trivial to "trick" it by just cycling through novel 5-grams over and over, but to be honest I can't be bothered to check. And that would be rather contrary to the point of the exercise, if somewhat undercutting of the premise that free will is equivalent or comparable to randomness.
Right; if you guess completely randomly it will hover around 50%, but if you choose the least-likely 5-gram from your history, then it should end up less than 50%.
my secret was decidedly less sexy than free will though. i just had XX years of working as a statistician and having a better intuition as to what random strings and numbers actually look like :/
I did have a job where I had to implement a bunch of cognition tests for scientific studies. Initially I worked purely off the research papers describing how the some of the tests work - apparently by presenting a series of random numbers.
After I had the first few up and running, the feedback from the researchers was that my random numbers weren't random enough. "Weird", I thought... I had seeds for their sessions so I could reproduce the sequences they'd seen and they looked pretty random to me.
In the end it turned out my sequences were "too random", they didn't like any number being repeated, having more than 3 instances of the same number within any 6-digit sequence, and a few other "tweaks". Turns out actual randomness is confusing even for some very experienced and well trained people!
Do you have a memorized random string?
It's easy to be random if you have memorized random (or normal) data that your adversary isn't focused on. For example, early digits of pi work in most cases. Or use a generator to create a sequence to memorize.
i have considered trying to memorise some algorithm, but never done it.
i assume that if anyone tried to get me to do these things with a non- uniform distribution I'd probably be screwed, but never tested it.
Not sure if the experiment tests this. (Pressing a different key or closing the browser tab are also variables in this probability distribution.)
I don't think this is obviously true. I can make a random number generating device with as many degrees of freedom as you like by producing a bunch of vertically polarised photons and measuring their horizontal polarisation. I don't think anyone would describe this system as having "free will".
Conversely humans with "free will" don't usually behave very randomly. They often have reasoning (good or bad) and are able to explain why they took whatever actions they took.
Free will seems to be qualitatively different to randomness.
I was going to suggest reading "The Diceman" by Luke Rhinehart [0].
But then I don't want to do that, because it really messes with some people.
[0] https://en.wikipedia.org/wiki/The_Dice_Man
[1] https://en.m.wikipedia.org/wiki/Flip_Decision
https://en.wikipedia.org/wiki/Two-Face
Sure determinism and free will are compatible as long as you believe that free will is your emotional reaction to discovering what your mind has determined.
One can willfully be irrational.
I do agree with you though, that free will doesn't necessarily mean "doing what I want" (which is a libertarian principle from the French Revolution), but instead, is freedom from desires. That's true freedom, but is rarely found and achieved only after years of ascetic practice.
Don't confuse "having reasons" with "being rational". Free will is a freedom to act according to your internal reasons, with that process of deliberation over your reasons being responsive to various types of feedback. Those reasons don't have to be rational.
The will of an addict is a bit less random, okay. Is there anything quantitative going on here besides "some stuff can increase randomness" and "some other stuff can decrease randomness"?
That said, I think this is a nice experiment exactly for surfacing this kind of questioning.
There is no such thing as free will, so any attempt to define it will fail.
If we naively define it as something that isn't just pure randomness, but also something that isn't pure causality, then what is it?
Either things are causal, in which case, you can replay everything and arrive at the same point. That doens't fit an intuitive definition of free will.
Or there is randomness, play the same sequence of events again and some inherent randomness will cause a different outcome. That also doesn't fit any intuitive sense of free will.
So, since we can't even define it, it is meaningless to assume there is free will in the first place. We can talk in terms of causality or randomness.
I think this is somewhat inaccurate. I'm not sure if I personally agree with it, but at the very least compatibilism is a serious philosophical viewpoint and shouldn't just be dismissed like this.
In any case I think the concept of "free will" is pretty much independent of randomness. If you provide me with some source of randomness (e.g. some quantum random number generator) I don't think I get more free because of it.
Your explanation is a good response to the posted experiment, but I don't think it's a good response to common sense, especially when we consider important (largely pragmatic) concepts like agency and responsibility.
Read up on Compatibilism. Free will can be defined in a way that's compatible with determinism. A deliberation over values resulting in a justification for a choice, aka "reasons" or "will". When you are "free" to act in accordance with those reasons, ie. you haven't been coerced into acting against your own reasons, then you've made a choice of your own free will. Simple and totally compatible with determinism.
Unless I misunderstand you, this is simply not true. Your device will be exceedingly easy to model, it's just a simple sum of i.i.d. random variables and subject to central limit theorems. (While humans definitely aren't.)
I have free will to eat the burger, or not.
The biggest impacts on that decision are whether I like burgers and if I'm hungry.
Whether I'm 'in the mood' is the only random component.
And the outcome isn't random. I can fairly accurately predict that after 20 times of being offered that choice, most people will say no.
I lined my finger's up to the keys, then looked away from the screen while I just "randomly" tapped to the beat of some music I was listening to.
it still guessed 70%
[1] https://github.com/elsehow/aaronson-oracle/blob/master/src/i...
if you are inspecting the prediction algorithm, and you want to specifically defeat it, doesn't that mean copy the prediction algorithm and throw a "not" on it? It's the old halting problem trick...
in any case, my post was serious, regardless of the % number, if you are inspecting their algorithm and writing your own, I can't understand why you wouldn't use theirs in yours. Finding an isomorphism might be fun, but it's not necessary.
That is not what it's predicting. You have changed the definition of predict so that an incorrect prediction can be misinterpreted as a correct prediction.
To put it more plainly: if you were wagering on the outcome, would you rather win 50% of the time (EV=0) or 100% of the time (EV=max)?
100% misprediction is only relevant if a 3rd party adversary is observing the game. In the absence of such an adversary, being reliably wrong it worthless.
50% misprediction cannot be improved upon with trivial methods.
You have two choices.
I predict correctly or I predict incorrectly. If I'm randomly guessing, I will be right 50% of the time (effectively, this is a coin flip). If I always guess incorrectly, then I'm correctly guessing the incorrect choice. If I always guess correctly, then I'm correctly guessing the correct choice.
I'm trying to explain why 50% is your target for "fooling the algorithm" and not 0%, which is not fooling the algorithm at all. There's no reason to be obtuse here, I'm legitimately trying to help you understand some statistics... this isn't an argument, it doesn't matter "what you believe" ... this is math.
No, you aren't, because you aren't saying that it is the incorrect choice. You are saying it is the correct one.
Thus “random,” aka, 50% is the target: this algorithm cannot predict what you do; you have free will.
If we conflate "free-will" with "ability to generate truly random sequences" then the goal should be to generate a completely incompressible, unpredictable sequence, which 50% would probably be closer to.
What they are saying is that when you have an algorithm you can reverse it to do the opposite. If you could reverse your basketball skills exactly that would be true for your statement.
50% means (if you're predicting their prediction) means you're achieving random.
"Free will" is a bit strong, it's more about whether people can produce randomness, which they generally can't.
50% prediction rate would indicate that half of the time you are predictable, meaning you don't have much free will.
0% prediction rate would indicate that you have 100% free will.
Randomness is not a good measure of free will.
Ultimately a 50% prediction rate means that you are as unpredictable as is possible, the state of maximum entropy. Any deviation from 50%, towards 100% or 0%, is a state of lower entropy.
You can forget the free will and just consider a fair coin toss. If someone had the ability to always guess a coin toss incorrectly, every single time, then you'd know that this person possesses something absolutely incredible and unique, rather magical. Similarly if they guess a coin toss correctly every single time they would also possess something magical.
It's the person who can only correctly guess a coin toss 50% of the time that is uninteresting, ordinary and possesses absolutely no special or magical insight. But deviating from 50% would require having some knowledge about the coin. A 0% prediction rate would require having total knowledge of the coin, so would a 100% prediction rate. It's 50% that requires having no knowledge whatsoever.
But suppose you could collect every bit of evidence in the universe and use it in a massive calculation to guess what my next decision would be.
This would be a test of whether I had free will, or whether I was simply doing what the universe made me do. If I wanted to demonstrate free will, I wouldn’t be aiming for random. I’d be aiming to make your guess wrong as often as I could.
If I could make your guess wrong 100% of the time, this would not mean that you could simply flip the bit on your guess next time. You can modify the calculation after every guess and it wouldn’t matter. I’d still do what your calculation thought I wouldn’t based on previous history. If I could do that, that would be a magical thing. Yes. I would have demonstrated completely free will according to this test.
This is why we call rand(), not freewill().
Free will is not random at all. It means you wear the shirt you want to wear, not a completely random shirt.
Oh, and before somebody objects, yes this is a terrible test of free will. Because whether I want to score high or low on the test or in the middle is up to me if I have free will. ;)
A 0% prediction rate means that you are 100% predictable by simply taking the opposite of the algorithm’s input or output. Thus, you have no free will (according to this tool).
Your presupposition seems to be that randomness = free will.
I propose that being forced to choose a random letter is the opposite of free will.
Your interpretation of the goal is also valid, but not in a cryptography context, for example.
[1] https://en.wikipedia.org/wiki/Zero_knowledge_proof
But, if I wrote there is a 90% probability that lupire will write an application to 'beat' the free will program, and then you did so, it should at least give you pause.
With sufficient technology, we could take a snapshot of your brain and predict exactly how you respond to any given stimulus with 100% accuracy.
Free will is an illusion. Consciousness is undefinable and unprovable.
This is assuming that intelligence is computable, but no one has proved it yet. And doesn't even get into the problems of quantum physics. Unfortunately, the real illusion is determinism.
I mean you've got a body, too, right? So the moment the environment changes so does the mind state. So unless you've come up with a method of modeling all of those influences... And food, and hormones, and... And then you've got stuff like cosmic radiation slamming into your DNA causing dsDNA breakage and that precipitates into apoptosis and suddenly the model is off due to some completely unpredictable shit, 'cause were talking about interactions that are occuring on the scale of 6.022*10^23 per mol, and a lot of mols and a lot of different species of matter themselves subject to probability moreso than determinism at any reasonable scale.
If sufficient technology were to emerge, I expect that under no circumstances would it ever be considered as a worthy project to invest in. But the idea of such technology ever emerging itself is frankly worthy of ridicule.
I just spammed F and randomly hit D and it pretty much never predicted the D.
Which I think captures a fundamental flaw in algorithms vs humans. Humans know how to mask their signals within noise. I'm inclined to believe the generalization that algorithms that aim for accuracy are highly susceptible to being manipulated by humans who are allowed to have free will over the inputs.
I think this is a sound strategy for “getting away with it” in society.
It's (also?) a weakness in humans: someone mostly does X, we lean very strongly into the assumption that they will never do !X. This is one of the most basic elements of human-to-human deceipt since ... well, since the dawn of any kind of recorded/remembered human history.
ddddd dfddd dffdd dfdfd ddfff ddfdd fdffd
That's only 10%.
I question the relation to free will however--will seems more related to preference than randomness to me, and a better example of that would be something like "I prefer the f key". Conflating randomness with free will has always seemed like a weak attempt to argue for the existence of free will to me.
I can look at the digits of PI and press d and f depending it's odd or even. So what does it prove? Nothing. Being able to beat this algorithm doesn't mean that I have free will, and choose to fall into patterns (knowing the PI solution exists) doesn't mean I don't.
Obviously you can argue whether this really predicts human behaviour or even the above implication itself (for example I think compatibilism kinda disagrees with it) but I'd say it is related.
From my point of view, that someone can predict what I will do from past behavior does not imply I don't have a free will.
Only way that would prove I don't have a free will is someone who would predict my behavior without knowing anything about my past, like a fortune teller that would tell my friend (because if fortune teller tells it to me it might be self fulfilling prophecy :D) something I would do; without having any interaction with me.
Although if it wasn't like that and you could do truly random decisions, I'm not sure what that says about your free will either.
If I had to guess, I don't think I have a free will. In my opinion my actions are a deterministic function of the state of my mind (including memory, personality) + input. I have no choice in the matter, only an illusion of one.
It's important to note that as far as materialist physics goes, "you" are literally "your state of mind (including memory, personality) and input". The fact that this system (i.e. "you") produces an apparently deterministic result doesn't mean that "you" didn't do it.
The common confusion comes from the fact that systems cannot in general predict their own outputs without going through the required computational process. I think it's related to the concept of "computational irreducibility". So "you" as a system needs to go through the set of motions to figure out what you eventually decide to do. Before the deliberation you can't predict the result, and thus it "feels as if" "free will" is a messy indeterministic process. But that's not a given.
Weird things happen when some other system can predict "you". The problem there is that if "you" can be simulated, you don't know which "you" is "real". But the mere fact that your state of mind and inputs will deterministically produce decisions doesn't mean "you" don't have a choice. In fact it is opposite -- "you" do have a choice, and _only_ the system "you" makes the choice (and thus it is determined solely by "you", hence, deterministic!)
To accurately simulate/copy the "sensory experience", you'll have to simulate/copy the entire "observable universe" of the person. Unless you just want to predict the behavior of a person locked inside a solitary, dark, sensory deprived room, the practical implication is that you need to simulate/copy the entire observable universe to run an accurate simulation/copy.
And that's not very feasible.
Does a family member knowing you well deprive you of freedom of choice?
No, it's not true. Read up on Compatibilism for a definition of free will that's compatible with determinism.
For example, you can predict that I'm going to pay my rent with > 90% accuracy, but that has nothing to do with whether I'm paying out of my free will or not.
A Marxist will argue that it's the oppression of the bourgeoisie that makes me pay, and a free market capitalist would counter that I made the choice "freely". But the truth is somewhere in between, and probably depends "subjectively" on how the person "feels" about doing it.
Of course if the prediction for all actions (not merely within an experimental setting) is close to 100% then there's probably something weird about it, but I don't think that's remotely possible. Not with current tech, and unlikely with future tech either.
It's easy to create a more complex example to defeat a more advanced adversary. (Many exist.)
For example, when I’m aware that there is space around myself, then I’m also aware that there are thoughts, feeling in the body, breath etc. In this state, I believe, it’s possible to choose. In this test I can choose the only press the other key and this algorithm doesn’t do so well.
Then something triggers my attention, I react, I forget that there is space, breath, thoughts etc. and I start to execute my conditioning, which is highly predictable and free will simply isn’t available at that point.
Then eventually something pulls me back to the presence, awareness. I remember there is space, breath, mind, feelings, emotions. I’m home. I’m free.
Think about it, if the universe is all predetermined, how peaceful your mental state is or isn't has no impact on it. We all either have free will or not, but breathing deep isn't gonna trigger it.
Are humans such a process?
Who knows, but attempts to model individual humans have certainly failed before. (We need lots of humans in aggregate to successfully model them, once the central limit theorem starts working.)
Only reason you believe there is a mind is because the mind says it exists. But have you investigated where is that mind and who is that which mind if speaking to?
The discussion about the ego and "the mind" and the "who you are" in that sense are then still useful for the human to feel comfortable in feeling like they have agency, and I am aware of Buddhist philosophy in finding peace through detaching from our inner voice and gaining wisdom from not letting ourselves being driven purely by what we think we are or "what the ego says we are", but I think it's besides the point in this context (but very useful to include in a balanced "diet of thoughts").
Maybe the ego is an evolutionary defence against nihilistic traits so we all don't kill ourselves by realising none of what we do is any agency and it's all predetermined by the start of the universe. But anyway this part of the discussion enters into a rabbithole detached from the core discussion of there being free will or not, at least to me, because I cannot accept that a core property of the system "universe" and the definition of it's causality or not could depend on the thought process that goes through minds of humans in some remote part of that very universe. A detail about the higher level thought process of some "blobs of particles" in the universe don't define the way the universe itself works. Our relationship to the self doesn't have anything to do with physical reality of if-this-then-that. In other words the causality doesn't care what we think of it, it's either all pre-determined or not, our opinion doesn't matter.
https://www.gocomics.com/calvinandhobbes/1988/06/07
The impetus to choose A or B in this case arises from the dark pool of the mind, sneaks up behind you and you mistakenly identify with this decision. You simply cannot think to think what to think.
Also if there is no free will then we cannot judge anyone, no matter what they did, because they didn’t have choice.
Actually my argument is that the best evidence for free will is that the world is such a mess. Without free will we would live in perfect harmony with nature, like all the other species do. We are so free that we can even deny the existence of our freedom, hence avoid taking responsibility of our actions.
This doesn't make sense. If there is a rock in the middle of the road, it would make sense to remove the rock from obstructing the road. Are we "judging" the rock in this case?
OTOH, if the rock happen to cause a massive accident, it seems absurd to punish the rock and try to cause as much suffering to the rock to compensate for all the suffering it caused.
> Without free will we would live in perfect harmony with nature, like all the other species do.
Harmony how? Like all the viruses, parasites, do you know what dolphins to pufferfish? What about cannibalism? Or infant mortality in nature?
> We are so free that we can even deny the existence of our freedom, hence avoid taking responsibility of our actions.
Realizing that free will does not exist does not mean that we don't have to live life anymore. We still have to live it. What it does it remove a lot of unnecessary baggage. Is it necessary to punish a rock for falling over a person? An elephant going on a rampage? What about murder?
When you realize that murder (or whatever immoral actions) is just the totality of the universe, you can think of any bad actions as just a very complex case of a brain tumor. You wouldn't blame a person for committing homicide once we learn that they went on a rampage because they have a brain tumor. But if we know enough about the universe, we will realize that pretty much anything that a human ever do is just another complex case of a brain tumor. Why did they do what they did? Well, here's a trillion page explanation of exactly why it happened starting from the Big Bang. Does it make sense to blame them now, still?
I identify with consciousness, and the origins of these intentions is unconscious, therefore I conclude that the intentions do not originate from “me”. It’s still my action, my “choice”, my kamma, but it’s not my “free will”.
What exactly do you mean by choose? Are you saying there are actions that have no discernable cause? That there's a state in the brain that is unrelated to the state of the brain a brief moment ago? If so, is that a type of telekinesis?
Yea I also like to say that if my will is truly free, it's not under anyones control.
One thing I want to add: I think will and intention is also a question of scale or perspective. Bacteria is capable of making "their own" decisions but they probably cannot dictate which direction body is going to walk.
In similar way Planet Earth is under influence of the Sun, and the Sun is under influence of larger forces of the Milky Way and so on.
You could say that every level of existence is probably under a will of some level above and maybe each level can impose their will on levels below them.
From this perspective we could say that pure awareness is beyond physical body, brain and mind, so then if you want to really will your body it must be done from the state of pure consciousness - and then the body, brain, mind must follow as they're level below.
Anyways, don't take me seriously, this just entertainment.
I don't know that I agree, but thought it was interesting.
(And no, I didn't write a Perl script, or have multiple goes to game the score, I just “tried not to do what I thought it'd think I'd do”.)
But when I started trying to type random keys as fast as possible it started getting a lot more right.
Being able to generate randomness is something only nature can do so by mimicking it you blend in with it.
With a little (computer-)elbow grease, one can find a sequence with perfectly free will. After a short time, it starts repeating every... that's right, 64 characters. So of course, we want to find the shortest possible prefix.
Note that the last two lines are a de Bruijn sequence. Repeat those until you get bored. That's how you exercise your free will.There is a deterministic bias: when in doubt, the page guesses "f". If it flipped a coin instead, a few experiments show that the greedy algorithm I and omoikane use cannot beat 25% for long.
[1] https://en.wikipedia.org/wiki/De_Bruijn_sequence