I gotta say. It takes an inspirational amount of balls to decide to talk for a living when you have a speech impediment. And it’s even more inspirational to see your incredible success despite said impediment. Hats off to you, sir.
@@211212112 I disagree: i think it actually enhances his channel, making it more distinctive and unique!! He is incredibly easy to understand, makes almost no pronunciation mistakes, and is very delightful to listen to in general.
I just found your channel yesterday and have viewed a few if your videos. Fascinating topics that you deliver in clear concise terms. Great job. You have earned a new subscriber. I can't wait to catch up viewing your other vids and look forward to your content in the future. Thank you.
I just noticed Isaac now has a patreon page link in his video description. All his subscribers should donate a few dollars to help him keep this channel going! I see alll the time in these comments how much everyone appreciates these videos! Let's support Isaac! thumbs up this comment so it stays at the top! Keep them coming Isaac!
I was born in a town in England, 7000 people live in that town. Of course, I could have been born outside the town. According to the doomsday argument there’s probably not much more than 7000 people living outside of this town. After all if there are 7000 people living outside the town and 7000 in it, there are 50 50 odds I would have been born outside. If there are only 7 people living outside the town then I had pretty close to 99.9% chance of being in the town, which seems a perfectly reasonable possibility, since I was born here. If there were 7 million people outside my town then the odds of me being born in my town is about 1/1000. That seems very unlikely but is not an absurd number of people. Now 7 Billion people living outside my town, that would be absurd. After all, I was born here, so the idea that my being born here was a literal one in a million chance is ridiculous, it’s just too unlikely. No, the mere fact I was born in this town of 7000 means that the global population is probably between 7000 and 14000, maybe a little higher. I’m sure if I go look up a global census the numbers will back me up.
Crazy Idea, but... maybe we should consider more then one town in the world? Assuming the town is like 14km² in size I would rather guess on a world population 253.000.000.000 people.
That isn't how the arguement works... the odds of ANYONE being born in that town vs the odds of YOU being born in the town. There was a 100% chance someone would be born there.
You made a mistake. You need to compare your town to other towns and not to the whole world. So you must reason with 7000 people against 20 Million, if thats the most populated town in the world.
I so wish I had you as a science teacher. I love this at last a true scientific videos that leave me asking more questions and yet want to know more thank you for your these videos.
The problem with the doomsday argument I feel, is that it equates being a person to being a random chance. It is not. Your mind is not randomly assigned a body. Things like lobotomies and brain damage sufficiently prove, in my eyes anyway, that the mind is physical and fundamentally tied to the brain. Since each human is assigned a number, the n'th human must always exist for n =< n(lim t -> ∞ ). Though the chance of "you" being an early human are small, they must exist. Also, don't forget the individual odds of existing as "you" are also astronomically tiny. All the different guys your mother could've met, all the different genetic content of the sperm cells and egg cells, the ~350 million other sperm cells released at the same time... All in all it's an astronomically small chance for "you" to exist. The chance of your mother and your father to have a child, however, is actually very high.
I'm not convinced that loss of function due to brain damage proves locallity just yet. The same argument could be used to claim youtube is local on a smart smartphone or PC. Damage the radio or logic components and you have loss of functionality. Now we know thats not the case due to our understanding of transmitting information over radio. One would have to presume locality to dismiss the radio as evidence for non locality. Simular with the mind and the brain. There is increasing data to support the non locality of the mind. It is commonly dismissed by presuming the mind is fundamentally tied to the brain. Circular.
That's brilliant and underappreciated. The fact is that we have very little idea how the mind could possibly emerge from the brain. We do know, however, through dreams especially, that the mind is capable of recreating an entire universe of matter. Mind may be creating our brains, rather than the other way 'round.
The first time I watched one of your videos, I did find the speech issue quite distracting. After watching dozens of your videos in the last couple of weeks, however, I've gotten so used to it that I didn't even notice it in this video until you mentioned it. Your content is 1000% good enough to wash over such issues.
I actually came up with the doomsday argument independently on my own, good to know it has a name. I didn't consider a lot of those alternative viewpoints though, this goes much deeper into it than I ever did. One counter argument I came up with is that all intelligent civilizations would probably come up with this exact argument regardless of whether they go on to die in a nuclear war or go on to colonize entire galaxies.
ROFL, that may be the most personally amusing rebuttal I've ever heard on the matter. Yes, yes, that would be exactly what is being claimed, and is basically the reasoning behind the time-dependence rebuttals.
The Doomsday Argument seems to be trying to create an artificial paradox. It's like saying "What does my personal existence tell me about my Dads sperm count?"
i been watching your videos for the last couple of days, "man" you are a true science head. As a visual artist and a sci-fi writer you made me rethink about the whole genre and the fact they been doing doing it all wrong. thx u your series will definite make me a better writer.
What an amazing explanation! I went from not knowing anything about this, to being confused in the first few minutes, to understanding the concept before your explanation was even over!
Three points I can think of: 1) This implies a catastrophe that is 100% complete. If we manage to get off-planet in the next 100 years, then it is difficult to see how something could wipe out all of humanity. Even if we aren't off-planet, if there are enough survivors then ironically we have bought many many more years until that "1 trillion" threshold is reached again. 2) Why are Adam and Eve counted as 1 and 2? They had billions of ancestors before them. Technically this should start with the first life form and go forward to today. 3 3) If a philosopher about 10 thousand years ago contemplated this (at a time of about 2M people world wide), should he have concluded that the human race had only a few thousand years left until it hit 10-15M? And then at 8 thousand years ago, when the human race didn't end, should they have concluded that they somehow got lucky in the past and now only have another 3-4 thousand years? Etc etc. If you want to use the human race as stating at about 50,000 years ago, then each generation since then has somehow avoided the Carter Catastrophe. That's either some insane luck, or (more likely) Carter doesn't apply to dynamic and evolving systems. Seeing as the universe is in a fairly young state (just wait until the stars die out and we are black hole farming!) I see no reason to question that the human race is in a fairly young state as well. All times are not equal. we are getting each generation at staying alive. I have no problem assuming, statistically or otherwise, that we are at the beginning or our overall existence as a species. In fact once we get off planet, that fact will be nearly assured.
taking this seriously from someone named "raven lord" with a Hello Kitty skull and crossbones avatar, is difficult. Also I think you may be confusing "off planet" with "off Sol system". If we colonized Jupiter's moons and some cataclysm rocks the sun, nothing is assured.
+Curvtixo D what could possibly mess up the sun so much that it messes all of us up as well? Obviously going further away is always better as time goes on, but being worried about that is like being worried about what we'll do when Andromeda collides with the milky way. Even tho the former is much more likely to occur, it's unlikely to occur in the scale of time in which it would be a huge issue for the human race.
Establish an off planet base who could survive without earth would take more than 100 years, you not only need to make all your food but also all your spare parts for life support and other critical systems, stuff its far cheaper to import. 300 years years then its likely.
I like it when you mention Nick Bostrom. I loved Superintelligence: Paths, Dangers, Strategies. I'll need to re-read it about 12 more times to really grasp it though.
I need to re-read it myself, though I can't recall if I finished it or not, a lot of it was familiar territory. Bostrom's a fun thinker, he does a good job bridging the science-philosophy divide that's sadly cropped up in recent years. I don't always agree with his reasoning, unsurprising since he often doesn't himself, but he's quite out-of-the-box and always constructs his arguments in a very detailed yet approachable fashion.
Awesome channel! I've been binge-watching your videos since came across them about a week ago. This is the first one where I can come up with some descent critiques. And was not getting my mind blown. The comparison between your 100 door example and our existence in this moment in time only makes sense if you believe that our souls (replace this term with what you want: minds, consciousness, etc) are randomly assigned to each person. Like each person is randomly assigned to each room. If you don't make that assumption your line of reasoning breaks down. Regardless of how you would classify trans-humans or our very old ancestors. If you, for example believe that our consciousness and self-awareness is created by our brains alone. And could possibly even be an illusion. Then there is no random assigning of anything involved. Rather our consciousness is the result of the universe just existing and evolving. But even if granted that assumption we still need a good hypothesis of how consciousness gets assigned and a strong definition of human. Let's say there is a k3 civilization that has existed for a billion years in a far away galaxy, and their light hasn't reached us yet. Can we include them in this example? I could likely cherry-pick conscious beings from their civilizations timeline and ours which are more similar than two from the same timeline. Can we include them? Or maybe expand the comparison to include every conscious being in the universe. In which case we also need to assume that we are the only ones. I could ramble on some more on these, and other assumptions you need to make in order for the 100 door comparison to work. But i'd rather go watch your video on dark matter instead. Keep up the good work
Well as I remind folks, this isn't my argument, I'm reciting a classic example with some flavor changes I felt better explain it, though substantively its the same. It and SimHyp are both classic arguments of the Anthropic Principle, which is one of the more legit approaches to looking at things form the standpoint of not knowing much, Mediocrity Principle being the other big one. The episode on AP is probably a good one to watch to follow this one up with, it shows some of the strengths and weaknesses of AP as a reasoning tool.
stats is simple maths even for some one who can't handle calc or algebra - ( i found :) a couple of simple formulas, some easy to understand graphs, -- it is just a matter of getting the training and practise - i am a words guy, every thing has to be done in words, i have to translate maths to words to do anything. Stats translates easy whilst algebra and calculus do not translate at all for me. :)
I’m terrible at algebra. But great at geometry and stats. I can visualize them way better than the quadratic formula for instance. However, a lot of folks who are way better than me at algebra lag in stats and geometry. I think stats and geometry rely more on intuition than algebra.
Jack Johnson: It's time someone had the courage to stand up and say, "I'm against those things everybody hates!" John Jackson: Now I respect my opponent, I think he's a good man, but quite frankly, I agree with everything he just said!
Subs like submarines or subs like sub sandwiches or subs like substitutions or subwoofers or subroutines or subtitles or the Seafarers' Union of Burma's?
The doomsday argument has never really sat well with me. I think it boils down to this: someone has to fill those first 10 cells in either case, but the same argument will lead to the wrong conclusion for those 10 prisoners in one of those cases. The thousandth human born on Earth would have had every reason to bet that he was nearer to the end of humanity than the start of it, despite being wrong (obvious only from our position in time). Now sure, the point of a statistical argument is that it will give the right answer most of the time, but that it will sometimes be wrong. But I think it's meaningless to draw any conclusions from the doomsday argument, and that we should proceed on the assumption that we are simply early in history until we can know otherwise. This assumption neatly solves the Fermi Paradox as well; we don't see other intelligence out there because we're first. Because someone has to be.
Honestly it is my hope that the solution to the Fermi Paradox is that we are the first or at least one of the first. Maybe it is just my hope for a possible epic space opera, but the optimist in me wants to see a huge expansion outward which would be likely if we were one of the first. Statistically we are probably nearing the end, but humans have beaten statistics before if you think about the evolution of each of our prior ancestors and determine what the chance was of them eventually evolving into a race that visits the moon before any other species accomplished the same goal.
what defines as a "name"? Real name, or online nickname? or maybe the IP adress of a computer, or the actual *name* of the computer? is it the name of the _model_ of the computer? I could go on, but you get the idea
I like the analogy of a cave men looking at New York and asuming there is no intelegent life because he does't see smoke signals. We have no reason to assume an advanced civilization would even have a physical form.
There are two main problems I can think of for this: 1) Someone had to be first, and it just so happened to be us, no matter who was first, they would've thought of this hypothesis. 2) the start and end of something is a very loose term. Let's say that hypothetically, the first child of someone younger than the WWW is born in 2019. 14 years later they will then think to themselves "it's far more likely that I'm currently in the middle of an interconnected civilisation than at the very start of one. Therefore no one will use internet in 34 years from now." And this logic could be applied to any invention if it worked, our civilisation would go in a pyramid of progression with us at the top.
Yes, really. I don't really see how there is any meat behind these probability arguments. In both scenarios, there was a person number 1, and there would have been extremely early persons. If they were able to calculate probability, they might conclude their existence was unlikely - but nevertheless they exist. The Fermi Paradox is whole different deal. It's not that it's impossible for humans to be first - someone must be. It's that there's a rather large universe out there, and it's been an awful lot of years since the start of the universe, so if humans are the first, why did it take so many billion years to even reach this point. Why not intelligence life at the 9 billion year mark? or the 11 billion year mark? Why now? This is tied up the doomsday argument because if you flip it the other way, the most _likely_ outcome of intelligent life comparable to our own is eventual external expansion in antimatter fueled starships (or better methods if possible) of the civilization that launched it. Though, if the observable universe is already 93 billion years across, well that explains part of it. Only these exponentially growing civilizations from inside our own galaxy and immediately adjacent galaxies have had enough time that they could have reached us already.
I'm not sure I'm enough of a statistician to credibly weigh in on this subject but it seems that these arguments suggest that with the potential of perhaps infinite future time, nearly any occurrence with a known origin is prohibitively unlikely to occur this early. So much of what's happening now, when viewed in isolation, are 1-in-a-zillion long shots, but so would be anything else that could be happening now. At some point your probability has to spike into a certainty and you have to accept that the incredibly unlikely is as likely as any other similarly specific possibility take the reality at face value.
Universe age 13 billion vs solar system 4 billion. Now that is billions and billions of planets in the habitable zone to have come and gone before Earth even existed. Makes us being the first very unlikely given those numbers if life is not exceptionally rare.
Gosh, you were smart to not include that part of wild theories where by knowing about the Doomsday Argument we somehow influence the reality and chances even more. I was so generously enraged when i read about this speculation.
As interesting as always, caught this one a week late. Strangely hoping to wait another week for a long video. The 30+ min ones are by far my favorites! Keep it up man!
Well you're sort of in luck, the one for this Friday is about 30 minutes, probably just shy. :) It's two piece on following up on the the last few videos to hit a couple topics I felt needed more explanation.
The Carter Catastrophe uses a probabilistic calculation formulated for a time-independent situation, and applied to a scenario that is very much time-dependent. In our captive scenario, the door numbers do not change with time, which is why the calculation is valid for that example. You could poke your head out at any moment and observe that your room number is 7, and if you managed to get your head out again, you will see that it is still 7. If instead the door numbers periodically ticked up from 0 to some maximum value (10 or 100) chosen individually by coin flips, and each person is shoved into their room at different points in that sequence, then when you poke your head out the room, the number you observe depends very strongly on when you did this. There's an additional parameter we need to keep track of, and it is the addition of this new parameter that destroys the Catastrophe. Every civilization, so long as it lives to the Doom Soon point, is going to look like a Doom Soon scenario at some point in its lifetime. Every single one. Ergo, the prior on this new parameter has to be monotonically decreasing. A careful consideration of that parameter reveals that it transforms like a scale parameter, which strongly implies that the Jeffries prior on it is some function 1/s. This provides the vital scale invariance we need to express the prior that we do not know ahead of time how long our civilization will last, ergo we do not know what proportion we have already burnt up. Thus the Fisher information of your birth ordinal is not very high, and simultaneously censors much of the observations available to you, which further restricts the Fisher information of any human you can observe and verify. (Even if our civilization lives for millions of years, we are not going to observe the birth ordinal of a baby born even a thousand years hence unless we get very extended lifespans PDQ.) This results in the posterior distribution on this parameter not being much different from its prior, and since this prior is very biased towards young civilizations, you should not be at all surprised to find yourself in the beginning of your civilization and not the end. Of course, how big this bias is depends on the particulars of the situation, but it does reveal what I believe to be a serious problem in this Doomsday argument: the argument fails to properly condition on all relevant parameters. Time is a relevant parameter in the lifetime of a civilization (indeed, I would argue it to be the _only_ relevant parameter), and as such any argument that seeks to answer how long a civilization would last without consideration of time scale invariance is junk.
I'm not following the point of your first paragraph but I agree with the fundamental issue. Humans communicate--we are not independent observers noting our room number, but rather a tiny fraction of the population will come up with this argument and we are observing the first person to do so, not a random person to do so. Being early in an ordered list proves nothing.
@@LorenPechtel The main crux of the first paragraph in the sequence where people are being progressively accumulated, it takes time for the numbers to build up. We know that, as mortal beings, that we have about 100 years of life before we kick it, so we have a very limited time to make these observations, and that fact is important because we will not live to see the day the last child is born (we hope). It's like if you still had your watch in this experiment and know that only a limited amount of time had passed between you being knocked out and you waking up and finding out that you're in room 7, and that amount of time is too little for 10 rooms to be filled, let alone 100. We've woken up and figured out that we're in room 7 _before_ the experiment is actually ready. That's where the time-dependence comes in. Supporting Doom Soon over Doom Late has to come from the difference between what these two hypotheses predict concerning what we can observe. It is not clear to me that Doom Soon makes any predictions in what we observe different from that of Doom Late. No consistent theory of inference will allow you to make the conclusion of Doom Soon over Doom Late in that case. No difference in predicted observables; no support for either. Period. What really cued me in on the problems with this argument is this absolute symmetry of the observation. A tribal Cro-Mangon might have made this observation and concluded that the most likely outcome would be that the human population would cap out at about a million and that would be it. And he would have been dead wrong. Now I don't know if I have pinpointed the actual flaw in the argument, but that symmetry does strongly indicate that there is some flaw in there somewhere. Futher, I didn't spot this four years ago, but at 7:36 the Bayes Rule calculation is wrong. The priors are lacking, and the marginal (denominator) is not correct. When I carried out the calculation myself using all the required figures, I came up with a value of 1/2 either way - even odds. I found that really interesting, so I generalized the calculation for any number N and M for the two groups, and any observation of the room number (as long as that room number is comparable with N and M), and same result: 1/2 - complete insensitivity to the sizes of the groups, and suspiciously equal to the naive weighting of the two groups. Something to think about, as Bayesian statistics is critically dependent on proper conditioning of your probabilities.
Considering the number of unknowns, unknowns unknowns, and the probability distributions being unknown, I doubt the doomsday argument or the anthropic principle can tell us much. They're interesting arguments to hear and contemplate, but at the end of the day they don't tell us anything new.
Yes I talk about that a bit more in the Anthropic Principle video, its more about how to view a situation and Doomsday, SimHyp, Fine-Tuning etc are just extreme uses of it.
Blade of grass fallacy. A golf ball is hit into the green, it hits a blade of grass. That blade of grass says it has the worst luck to be crushed when there are a hundred billion other blades of grass in this massive lawn. However the ball MUST have landed on one of them. Someone had to be born early, and no matter who they are they will think they are insanely unlucky to be born so early, when they could have been born as a hundred trillion other humans. However you MUST have been born as one of them.
I am still trying to figure out the best way to do a blog or podcast, though I did open a Patreon a few days back finally but still am a bit unclear on how to use it. It says I should have 'rewards' built in, but other than putting donors in the credits, which seems like a nice idea especially since its nothing but me and the musicians right now, I can't really think of a decent reward. www.patreon.com/user?ty=h&u=3365828
This seems to touch on the Paradox of the Ravens a bit. More importantly, though, *someone* had to be born now. Even if there were a much larger Doomsday Late group in the future, someone still had to exist now for that group to grow from, and someone needs to be inside these brains.
Oh sure, and as I recall Paul Franceschi argued that connection pretty strongly though his stuff was a bit deep for me. I probably should try it again sometime, been a while. Anyway, re: Raven/Hempel Paradox the usual Bayesian approach would be to say that, in this case I suppose, a black raven flying down the hallway does provide some evidence about which doors are blue or red but its very, very minimal. On your other point, of course someone has to be born now, same as someone in 100 group had to be in room 7, but that doesn't really alter the Bayesian approach on it.
Definitely love this channel. My mind had already lost its virginity to existentialism before I came here, but I still crave a good conversation about these kinds of things!
As a mind you aren't randomly placed in some body at any time. You are the mind of the person that was born. You have no probability to be anything else.
It is amazing how often new information changes the odds. For example, if I'm locked in a dark room and told to guess the color, red, blue, or green I have a 1 in 3 chance of guessing right. If after making my guess (say my guess is red) I'm told that my room is *NOT* one of the colors I did not guess (say it's not green), and given the option to change my guess, I still have a 1 in three chance that I guessed wrong, but now I know the room is not green. I should change my guess to blue, because 2 out of three times it will be blue.
Great vid! Instead of assuming we are in the doomsoon universe becouse of logic, i argue we as a species better start to make it a massive effort to force our tec into the direction of a galaxy colonizing species and enforcing a doom-late universe! Free will you know
No it is not, if you consider that TRUE (not pseudo) RANDOMNESS exists. If you can randomly go left or right on unknown path, that is basics of what with evolution developed into (true) free will. As we (i mean scientists) understand, there would be no universe (as we know it) without randomness, only a soup of evenly distributed mass/energy).
That's Heisenberg's uncertainty principle right? randomness is fundamental so the universe is has no determinestic structure. Although one can agrue that biological systems do not require quantum mechanics. Our brain may be wired with an extreme bias to certain choices. And from anecdotal evidence i can tell you that human behaviour is mostly as flexible as a brick wall.
That argument implies events that are governed only by random chance and statistics. After all, a coin cannot influence if it will land heads or tails. But humans can make decisions and take actions that will radically alter their odds of survival, either as individuals or as a group. Consequently, I think the Doomsday Argument is an interesting intellectual concept but has very little relevance in the real physical world.
Isn't the fundamental problem that the argument assumes that personal identity is separate from the situation of our existence. To work, it requires that swapping me with a person born in 3000AD changes the universe in some way, but it doesn't. Since the two individuals will be in exactly the same circumstances they will develop into exactly the same person. The swap is not a swap at all, and hence the possibility of 'me' being born in 3000AD is meaningless since 'me' is totally defined my existence here in 2017
I think the freakish odds thing can be answered at least in part with "any system despite the odds must invariably find itself in a state" in other words the odds may be freakishly low but someone had to be born during this time period. and someone had to be born in the hunter gatherer period. i mean you can literally extend this argument back and say that humans should have died out before getting to this point because it was very unlikely for someone to be born as a hunter gatherer pre-farming human then to be born as a post farming human.
I always find it weird to do statistics on one bit of data. Taking a 100 sides dice and throwig a 1 is unlikely but can happen. And if it happens, arguing, that this can't be because there is only a 1% chance of this happening is weird. Or even worse: Have a continuous distribution and argue: the probability of this event that just occured would be zero if we have a continuous distribution, therfore there have to be less possible outcomes to make this outcome more likely. At the same time you can't really argue against it.
Which is why the argument is not just taking the absolute chance of an event, but comparing two chances of two sets of events. Though a single bit of data still is weird. "At the same time you can't really argue against it." You cannot really argue in favor of it either. The only arguments which strongly support doing Bayesian inference/Solomonoff induction rely on repeated observations. With a single point of data, you cannot have more than a vaguely defined likelihood I think.
Yay a new Isaac Arthur video! Just what I needed to start my day with. :) By the way, your videos have been the topic of discussion between my fiancé and I lately. Last we talked about what was discussed in the Simulation Hypothesis video. Also, your videos have really helped turn my days around. I feel like I actually have something to look forward to - something fascinating and in-depth. It might seem silly but your videos have really made my days more tolerable.
A presumably-belated congratulations on your engagement, by the way. I think that may be the first time I've ever heard someone say they felt upbeat after hearing about the Simulation Hypothesis or Doomsday Argument, normally they tend to be so depressing I was actually reluctant to cover them :) Needless to say I'm very glad to hear people are enjoying them.
I'm also always looking forward to your new (lately weekly) videos. Always really fascinating topics for my weekend. Especially that they are so in-depth. Actually because they are so in-depth. "Normal" more shallow RUclips-Videos about such topics, just don't cut it for me anymore :) I really like your scientific approach to everything! Saying this as a physics student in the second semester :)
I started reading Nick Bostrom's material about 18 months ago. When I came to his analysis on simulation argument / simulation hypothesis I was very much in the doldrums for a few months. :) Who wouldn't be?
There's something about this argument that I don't like, but I can't pin it down. I think I need a better understanding of probability first. Good video, Isaac. Something less depressing next time? :P
The Doomsday Argument is like that, it takes a while to really absorb and doesn't 'feel right' to anybody, at least not anybody I know myself included. But the end of the world depressing? Nah ;) I'm actually considering an entirely unplanned, non-queue video on an upbeat topic next time around just to clear the air.
I think it does not 'feel right' because one of the premises it poses is wrong. The logical conclusion is irrefutably right, but it makes an assumption which in my opinion is not. This assumption is that your consciusness is in a 'pool' of consciousnesses that are randomly assigned to physical beings existing at a set time. I would bring the counter argument that the physical state you are in gives rise to your current consciousness (and yes, that could mean that you are just existing in this one single instance, after all we can never be certian if we experince passage of time). Only universes (if there are multiple) that lead to your exact physical state, or other locations in our universe that lead to the exact same physical state (assuming our universe is extremely large, or infinite), are in the 'pool' of selection. Now, the chance of you appearing in a certain civilization/culture in the same universe multiple times is extremely unlikely anyways, thus you can not make a statistical conclusion about the faith of your civilization/culture based on the observation of your existence.
Personally I think trying to apply too much math and probability to problems such as these implies too much of an assumption of strict order, while the universe itself appears to be a mix of order and chaos, especially when life is concerned. Let alone in sequence of events probability breaks down as reliable, since there being a miniscule chance of you being born earlier in history doesn't change that somebody had to be that person and all the earlier people that led to them had to exist.
I think the Doomsday Argument is perfectly sound and probably correct, but only if we stay on Earth. The second we have a self sustaining colony off Earth, be it on Mars, one of the various moons in the solar system, or even something like an O'Neal cylinder with good enough hydroponics to be able to survive without Earth the odds of our survival increases drastically. That 1% per year chance of us all dying each year plummets massively if we're on two planets and it continues to plummet the more we colonize other bodies in the solar system. If ever we get colonies to other stars (I'm extremely pessimistic about that, but not completely convinced of its impossibility) then the odds of us being wiped out plummet to a ridiculous low. Also I'd like to point out that the core assumption in the Doomsday argument, that we're about in the middle, would lead to incorrect conclusions about how many total humans will ever be born for about three quarters of all humans who will ever live. It really only works for the quarter or so who really are about in the middle. Since we have no idea whatsoever where we sit in that order, and since it's entirely possible that the total number of human could fall ANYWHERE on the scale between 100 billion and several quintillion, it really messes with the odds. The 91% only works if your reduce it to two options. If you change the door scenario so that you don't know how many doors there are but only that it is no less than 10 and no more than 100, which is much closer to our real situation, then knowing you're in door seven doesn't make it so massively likely that there are only 10 doors. It's still pretty unlikely that there are 100, but anything up to (estimating here because I'm too lazy to do the math) around 25 or 30 is pretty likely and anything up to around 60 or 70 remains reasonable.
Isaac at the start of your videos you warn people of your speech impediment equating it to making it harder for people to understand what your saying. But in full honesty even from day one you've been quite easy to understand and at least to me are very well spoken more so than others that have no such impediment. I'd prefer listening to you talk about these subjects than most anybody lol so keep up the good work my friend, because your videos make my 45 minute commute to work much better
Calling bullshit on something here. There is NO improbability of being born early in a doom late group because you have to be born somewhere and any given position is equally probable and improbable as any other. Example: Drawing a specific hand of cards in a five-card poker game is about 1 in 660,000. Thus, getting a specific hand is very unlikely, but no matter what hand you DO get, it was a 1 in 650,000 event. If you drew a million hands in a row you would get a 1 in 650,000 result each and every time.
I'm simply amazed at how well you covered this large topic. It's all fascinating really to think about these kinds of problems at hand; Involving these concepts, with largely debated logical reasoning instead of how people generally cover these terms. ohh by the way I don't know if you've covered this or not but I'd love to see a video based on Zero point energy!
How can we know if we can count ourselves as a random sample? And what about the gambler fallacy? If you have a 1% chance of extinction each year, surviving 300 years is unlikely, but if you survive 299 years, then you have a 99% chance of surviving, not 5%. Thus, we can only make the doomsday argument from the current day. Tomorrow, the human deadline with 95% confidence gets pushed back, yet the argument remains the same, even though the doomsday has changed.
So, I've more or less just discovered the channel, and I've been voraciously consuming videos, so thanks for the content! I wanted to add, I wrote a small Matlab script because I was curious about the sleeping beauty problem. My results are coming up that the thirder position is clearly correct, so I wanted to see if anyone can poke a hole in my thinking on the script. I'll explain each step for people that don't know Matlab. First, I initialize my array for the 3 possibilities of when SB is awakened (Heads and Monday, Tails and Monday, Tails and Tuesday respectively) awake = zeros(3,1); for ii = 1:100000 (starting a loop of 100000 trial runs) if rand > 0.5 (testing a random number between 0 and 1, inclusive, pulled from a uniform distribution ... I've decided that the number being over 0.5 means heads came up) awake(1) = awake(1) + 1; (Heads and Monday counter is incremented) else awake(2) = awake(2) + 1; (Tails and Monday counter is incremented) awake(3) = awake(3) + 1; (Tails and Tuesday counter is incremented) end (close if statement) end (close for loop) What my results were was: awake = (50028, 49972, 49972) The numbers are close enough that I feel pretty comfortable in saying that the difference is due to statistical noise, and the way I'm reading the results is that, over 100000 trials of this experiment (Sleeping Beauty has a true dedication to science!), that she was awake on a Monday due to heads 50028 times, Monday due to tails 49972 times, and Tuesday 49972 times. The means the odds of her being in a situation where she's awake and heads came up is clearly 1 in 3. Is my thinking wrong here? Am I missing something?
Personally I believe‘the doomsday argument’ should be taken seriously. We can reduce the chances of being wiped out as time passes by identifying the threats and dealing with them appropriately.
Here's the issue with the Doomsday Argument: you must make the assumption that conscious existence in some form, either human or some direct or divergent successor, is finite. Making that assumption puts us in a precarious situation. It's an interesting thought experiment, but assuming our species advances to such an extent that a critical mass comes to be a) aware of the argument and/or b) are sufficiently intelligent and educated to understand it, it could trigger a dangerous brand of nihilism and an existential crisis. In short, i'm not sure this is a useful idea to be sinking time into. This was, however, a well made video. Kudos. I've just subscribed.
You are in a hallway full of doors. Someone slips a piece of paper under the door that says, "You are in room 7, How many rooms do you think there are?" There is still the possibility there is only 1 door... I never said the slip of paper was definitely telling the truth. Of course it is completely possible that there are thousands of rooms and they only put people in the first 10 to see what they predict.
That was pretty much my argument. Because we are aware of our past and our history, we are not "born" in room 7. "We" as civilization walk a sequence of rooms starting from room 1, and it happens that at room 7 "god" gives us the doomsday argument. The probability of this to happen somewhere around room 7 is 1, and gives no power of predicting how many rooms are left. I suspect Nick Bostrom inflated this doomsday argument story out of nothing, and it was debunked long ago.
it seems a common assumption that in probabilities when there are several indications or hints contraddicting each other it's some kind of paradox - when really it should just essentially "even out" - that is a bit simplified - but if there are two options and we know nothing we can say it's a 50/50 guess - if now we find a hint or a solid argument indicating the chances are 90/10 in one direction - and another indication that they are 10/90 in the opposite direction - that's not a paradox, that's just an unlikely coincidence having happened and the indications we have even out back to 50/50 - or 60/40 or something like that depending on the exact example
This is the one and only time I have ever and probably will ever say this. Thank you for reminding me about the subscribe button. I'm a space cowboy, so I would have just kept looking up your video's the hard way.
Did anybody notice 83,000 views, 2,000 thumbs-up & only 49 thumbs down!?! What a successful ratio. Isaac Arthur, your channel is great. You have replaced 100% of my conspiracy video watching with your shows. What are the chances of that.
Isaac, I can't say this enough, but your videos are amazing! I don't useually subscribe to 'homemade' channels, but yours is just so informative. + love that i can watch this at night with captions on, so I won't disturb my roomies. Hope you don't mind me asking, but why do you not have a larger reddit presence? - I'm sure that /science or /documentaries would love both your videos and your input.
Oh Reddit is something I've been trying to transition into using, I really didn't know of it till recently, but it's not quite clicking and I tend to feel time spent figuring it out is time better spent on the next video. A few of my subscribers have posted videos over there and asked me to pop in for Q&A but that's about as far as I've gotten yet.
this is the best channel (your great content-you) I've yet seen for realistic science! I cannot thank you enough.!! A thought.. what if, as with variously accepted 'laws' the application of exotic forms of energy could cancel well enough any of them out? so the higgs field can be suppressed with countervailing energy so too could be other forces, couldn't civilizations using such understanding & with an enlightened viewpoint about equitable resource allocation effectively become invisible by choice? as much as I love your arguments in the main, the idea of actively choosing an upper limit to ones own resource usage seems totally alien to all of your views?
Hey Isaac, another great video. I wanted to add 2 things. There is a point that a civilization could reach, where they would become indestructible. The civilization destroying scenarios are generally only planet wide, so once a civilization spreads beyond then it can effectively become unkillable. Or at least impossible to wipe out 100%, and remember that we need 100%, because anything less and they can rebuild almost instantly. It's like completely sterilizing a planet of all life, it's incredibly hard to do. Now imagine completely sterilizing all the planets/stars/systems/backups that a civilization could potentially spread through. Type II civilizations are generally considered indestructible because they can always rebuild and if there were super weapons capable of doing system wide complete sterilization, then we'd definitely see evidence of its destruction. Also, imagine a superintelligence that creates backups, puts them onto nanobots, and sends those nanobots off at 99.9999999% the speed of light to seed the rest of the universe. No matter what ended up happening locally, the entire universe could be seeded almost instantly and irreversibly with backups. Second, have you considered the possibility that we'll actually get to live out both of the scenarios? Meaning that what if we're born at the end of human civilization and then we continue until the end of the universe as the superintelligence that we uploaded/merged with. The current assumption is that more biology will exist in the future (more numbers), but remember that our biology is already obsolete compared to what computers can do in almost every category. So if we agree that we'll eventually get an SI than it'll be a singular being and not a bunch of individual Ai's. Since it'll be singular, then the odds become even more impossible since we basically just multiplied the two scenarios together, oh and don't forget to multiply by the possibility of the universe even existing. We're both at the end and the beginning. Our universe is clearly a winner take all system, and the first winner will have it all to themselves. I think the answer to the Fermi Paradox is that we really are the first... and the last.....
I always wonder what protocol RUclips uses for it's filters, I just had to rescue this one from the 'likely spam' review folder :) No links, no foul language, no ALL CAPS, just regular text. I should probably make that comment-relevant with some remark about better AI for review systems but I just find it vaguely amusing and weird. Anyway I don't think there's any 'totally safe state' for a civilization, a pissed off xenocidal AI could probably send of murderbots at a slightly faster speed then civilization preserving bots. I should probably also add that I would not consider an AI non-biological, biology is a pretty arbitrary term especially when discussing aliens at the same time. Biology is an emergent property of chemistry where self-replicating systems is concerned, I think that would give us a pretty broad leeway in that regard.
Yes, but that xenocidal Ai would still be left alive. We're talking total sterilization requirements. If you send off a backup copy at 99.9% the speed of light, in a matter of months it will be completely unreachable since it'll require more than 100% the speed of light to catch up to it. And even if you include FTL travel, then we're still in an infinite universe, you'd be able to send infinite backups out and seed the universe in minutes. The main point is that once life starts, it's WAY harder to wipe it out. It's resilient as hell, so if dumb life is like this, imagine what a god-like life will be capable of. Another analogy is like comparing it to the internet. When the entire internet was a few computers, we could have killed it, but at this point it's virtually impossible to shut it down because it's such a redundant, decentralized system. It'd honestly be easier to kill off all humans vs the entire internet. So more intelligence gives you the risk of developing your own doom, but your ability to survive/repair advances far faster than the potential damages. Another analogy, the Terminator from T1 to T2, the T1000 was much tougher, much more unkillable since it was significantly more adaptable. Also, imagine a Grey Goo scenario and think about how hard it would be to wipe out every last nanobot. The whole point is that, as you advance/evolve you become more adaptable and significantly more capable of surviving greater and greater threats. When was the last time you were worried about being eaten by a lion or going extinct because your habitat is being wiped out? You're right on the non-biological part, I use it more to describe things that are driven by biological evolution vs self evolution. So people are biological because they're stuck being people, and robots/Ai's have no such limits. The reason everything trends to a singular being is that once you have the ability to connect different minds together, then you have a hive mind which is a new singular entity, and all the individuals eventually become one thing. It's always hugely more efficient/beneficial to do all your thinking together as one vs individually. You can have individual minds, but they simply won't be able to compete with the hive mind (singular). I think this got marked as spam is because I got involved in a bunch of other conversations below and that made me look like a spammer. I've been working on this for the last few years, so this is a crazy interesting topic for me and you're the only person on RUclips talking about the topic in an intellectually rigorous way. So sorry in advance for my long ass replies ;)
You were right, this really isn't intuitive at all! I loved it nonetheless. This is one of the best channels on youtube, please keep it up! I enjoy writing, and I'm in the process of writing a sci-fi story right now, and the explanation of this concept really inspired me and helped me incorporate more complex ideas in my story, so thank you for that! Also, the whole argument seems a bit arbitrary to me. I haven't watched the "Apocalypse How" video yet, so I don't know if that video covers this problem, but wouldn't a "Doomsday Soon" scenario involve a lot more variables that aren't really taken into consideration for this argument? Such as different kinds of probability for different kinds of "Doomsday" scenarios (among other things)? What I find really counter-intuitive about the argument, however, is the emphasis on the improbability of us existing within the first 100billion people born. Sure, it might be improbable, but those 100billion people must have existed first for the rest to be born as well. Now, I don't know much about probability or statistics, and I did struggle to understand the concepts in the video, but wouldn't the existence of a trillion trillion trillion trillion lives presuppose and, in fact, assure the existence of those first 100billion people? I understand, though, that the problem lies in the probability of us, of me, let's say, as an individual, being born among those first 100billion people. Again, though, those first billions must inevitably exist for the trillions to come later, and that's what seems very counter-intuitive to me. All in all, I think I understood at least the very basic premise behind the argument. EDIT: So, in both scenarios those first 100billion people must exist, I guess. I mean, it's counter-intuitive, it really, really is, but I think I understand the reasoning nevertheless; especially when I parallel the situation with the 100-10 door experiment. And one last thing, if we suppose that transhumanism is inevitable, wouldn't the fact that we're not in any way "transhuman" increase our possibility of being born prior to the invention of any such technology, therefore making it more probable that we're born now? What if we suppose that technology that significantly alters the state or physical form in which people are born is not far off into the future, how does that change our probability, if at all, of being born now? Although, if I understood this right, this "Doomsday Argument" works with the assumption that "transhumans" and "natural" humans, like us, are grouped together in a single group, and, perhaps, my aforementioned questions actually divide the two into two separate groups? Perhaps it's not two groups, though, perhaps, us being born as "natural" humans is akin to looking at the door number? Nevertheless, isn't that kind of an arbitrary definition or an arbitrary distinction or an arbitrary decision not to make that distinction? In fact, this whole argument seems to be based on arbitrary definitions and assumptions? So it makes me wonder how viable it is for the real world, or is it simply just a thought-experiment? EDIT: Also, when it comes to the 1% death per year argument, that doesn't make any sense to me since the results of the probability would depend wholly on the year you start counting, which seems completely arbitrary to me. Sorry for using the word arbitrary so much.
Isaac Arthur It won't be out for years to come, since I don't really plan on publishing anything until after I finish my studies, but I will post on one of your videos when the time comes if you're still active. Although I could post it on the internet somewhere, I guess. I'll be incorporating an altered version of the sleeping beauty experiment in the story I'm writing, although it won't be the main focus of the story. At any rate, I only discovered the experiment thanks to this video, so thank you for that!
Fortunately 'arbitrary' happens to me right there with 'indeed' as words I use way too often and am trying to ctuback on, so you're good :) The 1% death argument didn't have an arbitrary start though, as I recall anyway, I sometimes misremember the specific examples I make up for these and I generally dislike rewatching my own videos after I post them. As I recall that one begins with someone cobbling together a doomsday tech design and mass distributing it, with an assumption they can't purge the knowledge and it will be flowing around occasionally making it into the hands of nihilistic folks, some small number of whom have the resources to produce it and will luck out and not be caught. But it wouldn't matter anyway, if you and I could only exist because a fair coin flipped heads ten times in a row, it doesn't make that coin anymore or less likely to come up heads next time. Even if there always had been a 1% chance per year of us getting wiped, all that means is we've been very lucky so far or we botched our model. It wouldn't mean, if the former, that we'd keep being lucky. Only those events still to come matter, if you make it 50 years ahead in time when you had a 1% chance, the clock is reset to then, every year you survive negates that probability and starts it over again. Doesn't change the odds.
One counter-argument for the 1% per year doom scenario is technology shifting the odds. For example, I'm sure that 1% change includes the hidden assumption that everyone is on Earth, so a planetwide disaster can hit everyone. As soon as we have the first self-sustaining colony anywhere else, a whole class of threats become invalid as 100% kill outcomes, and that 1% drops. I'm sure there are other, more subtle versions of this argument that could be made.
Another very well done video. I've heard this argument before and I find it a better way to explain how statistics work to people rather than a good Fermi paradox argument. If anything it's an excellent example of how useless statistical analysis can be in the absence of the scientific method. The Carter argument would support every apocalypse cult and prophet-of-doom from the Mayans to the Bible's prophets which means they should have been right and the world should have ended a long time ago and you and I shouldn't exist. And yet here we are.
Oh yes, it definitely opens the mind to the statistical traps one can get one's self into the further you get from hard data and experiments. However I think 'perilous' might be more accurate than 'useless', though I know what you mean of course.
Random Sampling breaks down when the coin flip algorithm is designed because the algorithm is not Random. Head tweaker isn't it? Kind of like altering the behavior of a quantum particle just by looking at it or being bitten in the Hand by a dead Schrodinger's Cat even though you can't put your hand in a sealed Box without opening it. These questions are very fun to play with. TY for the Show.
Offtopic: I already watched lots of your videos and I had no problem understanding your english. And I am not a native speaker. I believe you underestimate yourself. Keep up the good job
Even if a 1% annual chance of extinction is accurate now, that number will decrease as we become a spacefaring civilization. Just having a sustainable colony on 1 additional planet will probably decrease extinction probability by one order of magnitude.
The biggest problem with the Doomsday Argument is that we have lots of data on life forms that have and have not become extinct over the last 100 years, and the Doomsday Argument does worse than a coin flip at predicting which ones will go extinct. It predicts that older species that had more individuals in the past and that have low birth rates (sharks and rays, for instance) are more likely to survive longer than newer species with high birth rates (mice, rats, and rabbits, for example). Would anyone be willing to bet that hammerhead sharks are going to outlast brown rats? I will reluctantly take your money. There are two northern white rhinos left in the world, both female. Statistically speaking, what is the likelihood that these two individuals "selected at random" are the last two individuals in the species? It's millions to one! In terms of the Doomsday Argument, if five million northern white rhinos lived in the past, there is a 50% chance that another five million northern white rhinos will be born, and at their current birth rate, they will likely last... uh, forever, since no new ones are being born? An ecologist would give them much worse odds. This is a theory that doesn't survive contact with facts. Thank you for the great content, Isaac. I'm a huge fan.
Thanks for your work! I love your videos even when I disagree with some fiddly details or points of view. 98% of the time when I start to think, “Ah, but what about this?!” You address that issue next. On topic: People often whine about not living in the Better World they can imagine. This doesn’t make the universe unfair. It just means that they have at least some imagination, and Nature is impartial. The point isn’t to live in that world, but rather to build it around you, and keep imagining improvements. That’s the function of life itself. You can’t use Bayesian to decide what “could” be true from an open field. It’s for deciding which is likely to be true among actually known alternatives using feedback analysis. With no current data ping, nothing to analyze, it’s simply the wrong tool for Fermi’s Paradox. After you meet a bunch of alien civilizations and have some feedback to work with, then you can use Bayesian to close in on what the likely civilization content of the universe will be. For the FP, Bayesian is an aftermarket accessory when you have already answered the essential question. Similarly, you also can’t win the “iocaine powder battle of wits to the death” using Bayesian. Both cups could be poisoned, and you can’t predict that with math or logic. Bayesian works better in the Doors of Doom red/blue problem because at least you have some known parameters to work with. Even that assumes you trust those notes and door-numbers as reliable feedback, which I wouldn’t, because what kind of lying sadistic assholes lock people up in mysterious rooms and chase them around with monsters? Just psychologists and mass murderers, and I don’t trust either. There are simple answers to the F.P. that allow for a lot of sensible alien civilizations out there, while also telling us some of the things these beings DON'T do. (You could apply some Bayesian to sort out what those things are, so you can concentrate on the better options.) Humans just don’t like it because we aren’t one of them yet, and we are still doing the wrong things. When we mature and learn to behave, then we can sit at the Grownup’s Table and drink the ambrosia. Until then, here’s your ginger ale, go to your desk and study, and don’t badger the whole neighborhood with your tantrums. As Zed noted in Men in Black, "Sucks, huh?" There's a lesson in the film clip: "The most unlikely threat" from Oscar Koeroo channel.
This argument gives me the same vibe as that paradox about guessing which day of the week something will happen on when you’ve been told the day will be chosen so that you cannot predict which day it will be. Can’t be Friday! Then I’d know because that’s the last day it could happen. We can’t possibly be among the first and only humans, then we would be too lucky!
No closed captioning I love your voice. and I'm not gather is I'm well aware that they can understand your voice but I can give it time don't worry about it buddy
who would down vote this ? Arthur is making some really good, interesting discussion videos here! I am hooked :D Its the perfect mix between scifi and philosophy and maths :D
Man, I just found this guys channel, I love it. He is funny and so good at explaining stuff in a way I can understand, hahahaha... I just thought he had a funny accent and once I got used to it, I don't have any problem understanding him. So very insightful and thought-provoking, did I mention I love it! He is deep......
I like this video quite a lot. I am not sure how others feel about it, but I find some aspects of the doomsday containing a macabre sense of humor, albeit at the abstract level. Parts of this video that made me laugh my ass off -- 'Doom-soon' vs, 'Doom-late' groups; 'the probability of staying alive at 99% per year'. (A fun way of refreshing how to run simple logarithmic calculations!)
I gotta say. It takes an inspirational amount of balls to decide to talk for a living when you have a speech impediment. And it’s even more inspirational to see your incredible success despite said impediment. Hats off to you, sir.
Your voice is better than a lot on the internet, and I don't need captions, but that's just me.
I prefer his speech impediment. His new videos it ain’t as bad, and I think in a way the channel is just a bit less for it.
@@211212112 I disagree: i think it actually enhances his channel, making it more distinctive and unique!! He is incredibly easy to understand, makes almost no pronunciation mistakes, and is very delightful to listen to in general.
@@altareggo he enunciates better than I, but I think u misread my comment since you make one of the exact same points. ie unique/distinctive
@@211212112 agreed. Thanks for the correction!
It didn’t occur to me that he sounded like Elmer Fudd until the image came up
I just found your channel yesterday and have viewed a few if your videos. Fascinating topics that you deliver in clear concise terms. Great job. You have earned a new subscriber. I can't wait to catch up viewing your other vids and look forward to your content in the future. Thank you.
Thanks Rene!
well said. agreed. i await your next video
One of the few youtube channels I watch every video of: quality like this is hard to come by.
I just noticed Isaac now has a patreon page link in his video description. All his subscribers should donate a few dollars to help him keep this channel going! I see alll the time in these comments how much everyone appreciates these videos! Let's support Isaac! thumbs up this comment so it stays at the top! Keep them coming Isaac!
Thanks Jason.
I'll support
I was born in a town in England, 7000 people live in that town. Of course, I could have been born outside the town.
According to the doomsday argument there’s probably not much more than 7000 people living outside of this town. After all if there are 7000 people living outside the town and 7000 in it, there are 50 50 odds I would have been born outside.
If there are only 7 people living outside the town then I had pretty close to 99.9% chance of being in the town, which seems a perfectly reasonable possibility, since I was born here.
If there were 7 million people outside my town then the odds of me being born in my town is about 1/1000. That seems very unlikely but is not an absurd number of people.
Now 7 Billion people living outside my town, that would be absurd. After all, I was born here, so the idea that my being born here was a literal one in a million chance is ridiculous, it’s just too unlikely.
No, the mere fact I was born in this town of 7000 means that the global population is probably between 7000 and 14000, maybe a little higher. I’m sure if I go look up a global census the numbers will back me up.
Crazy Idea, but... maybe we should consider more then one town in the world?
Assuming the town is like 14km² in size I would rather guess on a world population 253.000.000.000 people.
@@quazar5017 but that is just like saying, The reality does not fit the math so lets change the the reality because math is always right.
I knew why I hated this argument. But you put into words better than I could thought. Thanks.
That isn't how the arguement works... the odds of ANYONE being born in that town vs the odds of YOU being born in the town. There was a 100% chance someone would be born there.
You made a mistake. You need to compare your town to other towns and not to the whole world. So you must reason with 7000 people against 20 Million, if thats the most populated town in the world.
I so wish I had you as a science teacher. I love this at last a true scientific videos that leave me asking more questions and yet want to know more thank you for your these videos.
The problem with the doomsday argument I feel, is that it equates being a person to being a random chance. It is not. Your mind is not randomly assigned a body. Things like lobotomies and brain damage sufficiently prove, in my eyes anyway, that the mind is physical and fundamentally tied to the brain. Since each human is assigned a number, the n'th human must always exist for n =< n(lim t -> ∞ ). Though the chance of "you" being an early human are small, they must exist.
Also, don't forget the individual odds of existing as "you" are also astronomically tiny. All the different guys your mother could've met, all the different genetic content of the sperm cells and egg cells, the ~350 million other sperm cells released at the same time... All in all it's an astronomically small chance for "you" to exist. The chance of your mother and your father to have a child, however, is actually very high.
I would add that decisions made by parents to have a child are not random, at least in modern times.
Christian Libertarian it's one random sperm cell out of the 350 million. That's random enough.
But only one egg. And a decision, not randomly taken, to allow the sperm and egg to join.
I'm not convinced that loss of function due to brain damage proves locallity just yet. The same argument could be used to claim youtube is local on a smart smartphone or PC. Damage the radio or logic components and you have loss of functionality. Now we know thats not the case due to our understanding of transmitting information over radio. One would have to presume locality to dismiss the radio as evidence for non locality.
Simular with the mind and the brain. There is increasing data to support the non locality of the mind. It is commonly dismissed by presuming the mind is fundamentally tied to the brain.
Circular.
That's brilliant and underappreciated. The fact is that we have very little idea how the mind could possibly emerge from the brain. We do know, however, through dreams especially, that the mind is capable of recreating an entire universe of matter. Mind may be creating our brains, rather than the other way 'round.
The first time I watched one of your videos, I did find the speech issue quite distracting. After watching dozens of your videos in the last couple of weeks, however, I've gotten so used to it that I didn't even notice it in this video until you mentioned it. Your content is 1000% good enough to wash over such issues.
I actually came up with the doomsday argument independently on my own, good to know it has a name. I didn't consider a lot of those alternative viewpoints though, this goes much deeper into it than I ever did.
One counter argument I came up with is that all intelligent civilizations would probably come up with this exact argument regardless of whether they go on to die in a nuclear war or go on to colonize entire galaxies.
When I watched this video there where only 245 views. So I should expect on average a total amount of views of 490? Maybe more, maybe less?
ROFL, that may be the most personally amusing rebuttal I've ever heard on the matter. Yes, yes, that would be exactly what is being claimed, and is basically the reasoning behind the time-dependence rebuttals.
The Doomsday Argument seems to be trying to create an artificial paradox.
It's like saying "What does my personal existence tell me about my Dads sperm count?"
I'm in at 499. DISPROVEN.
I saw it when there was 545 views that means that it will probably only grow to 1090 views.
Thank you for this epic series! :-)
i been watching your videos for the last couple of days, "man" you are a true science head. As a visual artist and a sci-fi writer you made me rethink about the whole genre and the fact they been doing doing it all wrong. thx u your series will definite make me a better writer.
Thanks Lloyd, I appreciate that, it's always nice knowing when the episode are offering folks new insights and views.
Lloyd Green u should volunteer your artistic services
What an amazing explanation!
I went from not knowing anything about this, to being confused in the first few minutes, to understanding the concept before your explanation was even over!
I'm telling all my friends about your youtube. and our minds are exploding with thoughts.
Three points I can think of:
1) This implies a catastrophe that is 100% complete. If we manage to get off-planet in the next 100 years, then it is difficult to see how something could wipe out all of humanity. Even if we aren't off-planet, if there are enough survivors then ironically we have bought many many more years until that "1 trillion" threshold is reached again.
2) Why are Adam and Eve counted as 1 and 2? They had billions of ancestors before them. Technically this should start with the first life form and go forward to today. 3
3) If a philosopher about 10 thousand years ago contemplated this (at a time of about 2M people world wide), should he have concluded that the human race had only a few thousand years left until it hit 10-15M? And then at 8 thousand years ago, when the human race didn't end, should they have concluded that they somehow got lucky in the past and now only have another 3-4 thousand years? Etc etc. If you want to use the human race as stating at about 50,000 years ago, then each generation since then has somehow avoided the Carter Catastrophe. That's either some insane luck, or (more likely) Carter doesn't apply to dynamic and evolving systems.
Seeing as the universe is in a fairly young state (just wait until the stars die out and we are black hole farming!) I see no reason to question that the human race is in a fairly young state as well. All times are not equal. we are getting each generation at staying alive. I have no problem assuming, statistically or otherwise, that we are at the beginning or our overall existence as a species. In fact once we get off planet, that fact will be nearly assured.
That probably won happen in 100 years
#2 is actually mentioned in the video
taking this seriously from someone named "raven lord" with a Hello Kitty skull and crossbones avatar, is difficult. Also I think you may be confusing "off planet" with "off Sol system". If we colonized Jupiter's moons and some cataclysm rocks the sun, nothing is assured.
+Curvtixo D what could possibly mess up the sun so much that it messes all of us up as well? Obviously going further away is always better as time goes on, but being worried about that is like being worried about what we'll do when Andromeda collides with the milky way. Even tho the former is much more likely to occur, it's unlikely to occur in the scale of time in which it would be a huge issue for the human race.
Establish an off planet base who could survive without earth would take more than 100 years, you not only need to make all your food but also all your spare parts for life support and other critical systems, stuff its far cheaper to import.
300 years years then its likely.
I really appreciate your channel. There's no one else out there quite like it. Keep up the good work!
Thank you Ethan!
I like it when you mention Nick Bostrom. I loved Superintelligence: Paths, Dangers, Strategies. I'll need to re-read it about 12 more times to really grasp it though.
I need to re-read it myself, though I can't recall if I finished it or not, a lot of it was familiar territory. Bostrom's a fun thinker, he does a good job bridging the science-philosophy divide that's sadly cropped up in recent years. I don't always agree with his reasoning, unsurprising since he often doesn't himself, but he's quite out-of-the-box and always constructs his arguments in a very detailed yet approachable fashion.
@@isaacarthurSFIA
With "recent years" you mean the last 300 years?
11:30 "Cheap and easy access to coffee alone makes modern times better in my opinion" LOL!
Nah.... relatively cheap and easy access to a wide variety of herbal and green teas... now THAT does it for me!! Coffee... meh.
Awesome channel! I've been binge-watching your videos since came across them about a week ago. This is the first one where I can come up with some descent critiques. And was not getting my mind blown.
The comparison between your 100 door example and our existence in this moment in time only makes sense if you believe that our souls (replace this term with what you want: minds, consciousness, etc) are randomly assigned to each person. Like each person is randomly assigned to each room. If you don't make that assumption your line of reasoning breaks down. Regardless of how you would classify trans-humans or our very old ancestors. If you, for example believe that our consciousness and self-awareness is created by our brains alone. And could possibly even be an illusion. Then there is no random assigning of anything involved. Rather our consciousness is the result of the universe just existing and evolving.
But even if granted that assumption we still need a good hypothesis of how consciousness gets assigned and a strong definition of human. Let's say there is a k3 civilization that has existed for a billion years in a far away galaxy, and their light hasn't reached us yet. Can we include them in this example? I could likely cherry-pick conscious beings from their civilizations
timeline and ours which are more similar than two from the same timeline. Can we include them? Or maybe expand the comparison to include every conscious being in the universe. In which case we also need to assume that we are the only ones.
I could ramble on some more on these, and other assumptions you need to make in order for the 100 door comparison to work. But i'd rather go watch your video on dark matter instead.
Keep up the good work
Well as I remind folks, this isn't my argument, I'm reciting a classic example with some flavor changes I felt better explain it, though substantively its the same. It and SimHyp are both classic arguments of the Anthropic Principle, which is one of the more legit approaches to looking at things form the standpoint of not knowing much, Mediocrity Principle being the other big one. The episode on AP is probably a good one to watch to follow this one up with, it shows some of the strengths and weaknesses of AP as a reasoning tool.
Ok, this is really the only episode where i must admit, i'm too stupid to understand.
🙁
stats is simple maths even for some one who can't handle calc or algebra - ( i found :) a couple of simple formulas, some easy to understand graphs, -- it is just a matter of getting the training and practise - i am a words guy, every thing has to be done in words, i have to translate maths to words to do anything. Stats translates easy whilst algebra and calculus do not translate at all for me. :)
@@kparker2430 Algebra is easy and intuitive. I found Calc difficult when I was at uni.
Mickeyislowd, This gave me a headache.
I’m terrible at algebra. But great at geometry and stats. I can visualize them way better than the quadratic formula for instance. However, a lot of folks who are way better than me at algebra lag in stats and geometry. I think stats and geometry rely more on intuition than algebra.
K Parker same here I’m all pictures and words. I can’t remember maths I don’t fully understand
Jack Johnson: It's time someone had the courage to stand up and say, "I'm against those things everybody hates!"
John Jackson: Now I respect my opponent, I think he's a good man, but quite frankly, I agree with everything he just said!
You deserve more subs. Keep up the good work man!
absolutely. dude you are the man
Unfortunately I think his videos are on a wave length above that of the average viewer.
+IJustMadeAComment He had 5 k subs when i wrote this, now he's got 16 k subs. So the channel does grow.
265k a year after that/ Exponential growth :-)
Subs like submarines or subs like sub sandwiches or subs like substitutions or subwoofers or subroutines or subtitles or the Seafarers' Union of Burma's?
Watched this three times. Finally grasped Bayes theorem and now I somewhat understand it.
The doomsday argument has never really sat well with me.
I think it boils down to this: someone has to fill those first 10 cells in either case, but the same argument will lead to the wrong conclusion for those 10 prisoners in one of those cases. The thousandth human born on Earth would have had every reason to bet that he was nearer to the end of humanity than the start of it, despite being wrong (obvious only from our position in time).
Now sure, the point of a statistical argument is that it will give the right answer most of the time, but that it will sometimes be wrong. But I think it's meaningless to draw any conclusions from the doomsday argument, and that we should proceed on the assumption that we are simply early in history until we can know otherwise. This assumption neatly solves the Fermi Paradox as well; we don't see other intelligence out there because we're first.
Because someone has to be.
We are just the lucky bastarts to be born in the les then 1% ratio of the doom late.
Honestly it is my hope that the solution to the Fermi Paradox is that we are the first or at least one of the first. Maybe it is just my hope for a possible epic space opera, but the optimist in me wants to see a huge expansion outward which would be likely if we were one of the first. Statistically we are probably nearing the end, but humans have beaten statistics before if you think about the evolution of each of our prior ancestors and determine what the chance was of them eventually evolving into a race that visits the moon before any other species accomplished the same goal.
On a side note I wonder what the probability of two people with the same name watching the same video in the same approximate time.
what defines as a "name"? Real name, or online nickname? or maybe the IP adress of a computer, or the actual *name* of the computer? is it the name of the _model_ of the computer? I could go on, but you get the idea
I like the analogy of a cave men looking at New York and asuming there is no intelegent life because he does't see smoke signals.
We have no reason to assume an advanced civilization would even have a physical form.
Started watching your vids a few days ago, happened on one by chance. Great stuff and well presented, keep them coming!
There are two main problems I can think of for this:
1) Someone had to be first, and it just so happened to be us, no matter who was first, they would've thought of this hypothesis.
2) the start and end of something is a very loose term. Let's say that hypothetically, the first child of someone younger than the WWW is born in 2019. 14 years later they will then think to themselves "it's far more likely that I'm currently in the middle of an interconnected civilisation than at the very start of one. Therefore no one will use internet in 34 years from now."
And this logic could be applied to any invention if it worked, our civilisation would go in a pyramid of progression with us at the top.
thats what i was thinking. Whoever was first no matter the scenario would have been contemplating the improbability of their existence
Yes, really. I don't really see how there is any meat behind these probability arguments. In both scenarios, there was a person number 1, and there would have been extremely early persons. If they were able to calculate probability, they might conclude their existence was unlikely - but nevertheless they exist.
The Fermi Paradox is whole different deal. It's not that it's impossible for humans to be first - someone must be. It's that there's a rather large universe out there, and it's been an awful lot of years since the start of the universe, so if humans are the first, why did it take so many billion years to even reach this point. Why not intelligence life at the 9 billion year mark? or the 11 billion year mark? Why now? This is tied up the doomsday argument because if you flip it the other way, the most _likely_ outcome of intelligent life comparable to our own is eventual external expansion in antimatter fueled starships (or better methods if possible) of the civilization that launched it. Though, if the observable universe is already 93 billion years across, well that explains part of it. Only these exponentially growing civilizations from inside our own galaxy and immediately adjacent galaxies have had enough time that they could have reached us already.
The first argument is like saying that lottery winners contemplating the improbability of their situation means that winning the lottery is likely.
I'm not sure I'm enough of a statistician to credibly weigh in on this subject but it seems that these arguments suggest that with the potential of perhaps infinite future time, nearly any occurrence with a known origin is prohibitively unlikely to occur this early. So much of what's happening now, when viewed in isolation, are 1-in-a-zillion long shots, but so would be anything else that could be happening now. At some point your probability has to spike into a certainty and you have to accept that the incredibly unlikely is as likely as any other similarly specific possibility take the reality at face value.
Universe age 13 billion vs solar system 4 billion. Now that is billions and billions of planets in the habitable zone to have come and gone before Earth even existed. Makes us being the first very unlikely given those numbers if life is not exceptionally rare.
Gosh, you were smart to not include that part of wild theories where by knowing about the Doomsday Argument we somehow influence the reality and chances even more. I was so generously enraged when i read about this speculation.
Man I wish I knew what was so good about coffee it just make my stomach hurt like I'm sick.
As interesting as always, caught this one a week late. Strangely hoping to wait another week for a long video. The 30+ min ones are by far my favorites! Keep it up man!
Well you're sort of in luck, the one for this Friday is about 30 minutes, probably just shy. :) It's two piece on following up on the the last few videos to hit a couple topics I felt needed more explanation.
How dare you assume my door color.
The Carter Catastrophe uses a probabilistic calculation formulated for a time-independent situation, and applied to a scenario that is very much time-dependent. In our captive scenario, the door numbers do not change with time, which is why the calculation is valid for that example. You could poke your head out at any moment and observe that your room number is 7, and if you managed to get your head out again, you will see that it is still 7. If instead the door numbers periodically ticked up from 0 to some maximum value (10 or 100) chosen individually by coin flips, and each person is shoved into their room at different points in that sequence, then when you poke your head out the room, the number you observe depends very strongly on when you did this. There's an additional parameter we need to keep track of, and it is the addition of this new parameter that destroys the Catastrophe.
Every civilization, so long as it lives to the Doom Soon point, is going to look like a Doom Soon scenario at some point in its lifetime. Every single one. Ergo, the prior on this new parameter has to be monotonically decreasing. A careful consideration of that parameter reveals that it transforms like a scale parameter, which strongly implies that the Jeffries prior on it is some function 1/s. This provides the vital scale invariance we need to express the prior that we do not know ahead of time how long our civilization will last, ergo we do not know what proportion we have already burnt up. Thus the Fisher information of your birth ordinal is not very high, and simultaneously censors much of the observations available to you, which further restricts the Fisher information of any human you can observe and verify. (Even if our civilization lives for millions of years, we are not going to observe the birth ordinal of a baby born even a thousand years hence unless we get very extended lifespans PDQ.) This results in the posterior distribution on this parameter not being much different from its prior, and since this prior is very biased towards young civilizations, you should not be at all surprised to find yourself in the beginning of your civilization and not the end.
Of course, how big this bias is depends on the particulars of the situation, but it does reveal what I believe to be a serious problem in this Doomsday argument: the argument fails to properly condition on all relevant parameters. Time is a relevant parameter in the lifetime of a civilization (indeed, I would argue it to be the _only_ relevant parameter), and as such any argument that seeks to answer how long a civilization would last without consideration of time scale invariance is junk.
I'm not following the point of your first paragraph but I agree with the fundamental issue. Humans communicate--we are not independent observers noting our room number, but rather a tiny fraction of the population will come up with this argument and we are observing the first person to do so, not a random person to do so. Being early in an ordered list proves nothing.
@@LorenPechtel The main crux of the first paragraph in the sequence where people are being progressively accumulated, it takes time for the numbers to build up. We know that, as mortal beings, that we have about 100 years of life before we kick it, so we have a very limited time to make these observations, and that fact is important because we will not live to see the day the last child is born (we hope). It's like if you still had your watch in this experiment and know that only a limited amount of time had passed between you being knocked out and you waking up and finding out that you're in room 7, and that amount of time is too little for 10 rooms to be filled, let alone 100. We've woken up and figured out that we're in room 7 _before_ the experiment is actually ready. That's where the time-dependence comes in.
Supporting Doom Soon over Doom Late has to come from the difference between what these two hypotheses predict concerning what we can observe. It is not clear to me that Doom Soon makes any predictions in what we observe different from that of Doom Late. No consistent theory of inference will allow you to make the conclusion of Doom Soon over Doom Late in that case. No difference in predicted observables; no support for either. Period.
What really cued me in on the problems with this argument is this absolute symmetry of the observation. A tribal Cro-Mangon might have made this observation and concluded that the most likely outcome would be that the human population would cap out at about a million and that would be it. And he would have been dead wrong. Now I don't know if I have pinpointed the actual flaw in the argument, but that symmetry does strongly indicate that there is some flaw in there somewhere.
Futher, I didn't spot this four years ago, but at 7:36 the Bayes Rule calculation is wrong. The priors are lacking, and the marginal (denominator) is not correct. When I carried out the calculation myself using all the required figures, I came up with a value of 1/2 either way - even odds. I found that really interesting, so I generalized the calculation for any number N and M for the two groups, and any observation of the room number (as long as that room number is comparable with N and M), and same result: 1/2 - complete insensitivity to the sizes of the groups, and suspiciously equal to the naive weighting of the two groups. Something to think about, as Bayesian statistics is critically dependent on proper conditioning of your probabilities.
Considering the number of unknowns, unknowns unknowns, and the probability distributions being unknown, I doubt the doomsday argument or the anthropic principle can tell us much. They're interesting arguments to hear and contemplate, but at the end of the day they don't tell us anything new.
Yes I talk about that a bit more in the Anthropic Principle video, its more about how to view a situation and Doomsday, SimHyp, Fine-Tuning etc are just extreme uses of it.
Isaac Arthur Thanks, will check it out. By the way great job on the content. The work you put into your videos really shows.
Blade of grass fallacy.
A golf ball is hit into the green, it hits a blade of grass. That blade of grass says it has the worst luck to be crushed when there are a hundred billion other blades of grass in this massive lawn.
However the ball MUST have landed on one of them.
Someone had to be born early, and no matter who they are they will think they are insanely unlucky to be born so early, when they could have been born as a hundred trillion other humans.
However you MUST have been born as one of them.
funny fallacy there :) yeah, someone has to be me/you/that blade of grass ;)
These videos are the only reason I come on RUclips. Please start a blog / podcast or set something up so we can donate / subsidise!
I am still trying to figure out the best way to do a blog or podcast, though I did open a Patreon a few days back finally but still am a bit unclear on how to use it. It says I should have 'rewards' built in, but other than putting donors in the credits, which seems like a nice idea especially since its nothing but me and the musicians right now, I can't really think of a decent reward.
www.patreon.com/user?ty=h&u=3365828
This seems to touch on the Paradox of the Ravens a bit. More importantly, though, *someone* had to be born now. Even if there were a much larger Doomsday Late group in the future, someone still had to exist now for that group to grow from, and someone needs to be inside these brains.
Oh sure, and as I recall Paul Franceschi argued that connection pretty strongly though his stuff was a bit deep for me. I probably should try it again sometime, been a while. Anyway, re: Raven/Hempel Paradox the usual Bayesian approach would be to say that, in this case I suppose, a black raven flying down the hallway does provide some evidence about which doors are blue or red but its very, very minimal. On your other point, of course someone has to be born now, same as someone in 100 group had to be in room 7, but that doesn't really alter the Bayesian approach on it.
Yeah, but those would be very few wrong people then.
Definitely love this channel. My mind had already lost its virginity to existentialism before I came here, but I still crave a good conversation about these kinds of things!
I love your content im happy I found your channel a year ago.
wow, that's seems like an eternity ago, back when there were 3 or 4 vids and I did one maybe every other month... glad you've stuck around!
+Isaac Arthur Yup no problem buddy keep up the great work and I can tell u put effort into your videos.
Will do
Always enjoy watching your videos, they are very relaxing and easy to watch over a lunch break :)
Wait... who's on first and what's on second?
I don't care. Oh, wait, he's the shortstop.
Just got home from traveling to find one of your videos. Always such a pleasure to watch! :D
As a mind you aren't randomly placed in some body at any time. You are the mind of the person that was born. You have no probability to be anything else.
It is amazing how often new information changes the odds. For example, if I'm locked in a dark room and told to guess the color, red, blue, or green I have a 1 in 3 chance of guessing right. If after making my guess (say my guess is red) I'm told that my room is *NOT* one of the colors I did not guess (say it's not green), and given the option to change my guess, I still have a 1 in three chance that I guessed wrong, but now I know the room is not green. I should change my guess to blue, because 2 out of three times it will be blue.
Great vid! Instead of assuming we are in the doomsoon universe becouse of logic, i argue we as a species better start to make it a massive effort to force our tec into the direction of a galaxy colonizing species and enforcing a doom-late universe! Free will you know
But isn't free will an illusion
No it is not, if you consider that TRUE (not pseudo) RANDOMNESS exists. If you can randomly go left or right on unknown path, that is basics of what with evolution developed into (true) free will.
As we (i mean scientists) understand, there would be no universe (as we know it) without randomness, only a soup of evenly distributed mass/energy).
We are presented a choice but the universe is structured for your exact choice.
Does this mean that every random event, like exact timing of spontaneous decay of unstable atom or particle is predetermined and thus calculable???
That's Heisenberg's uncertainty principle right? randomness is fundamental so the universe is has no determinestic structure. Although one can agrue that biological systems do not require quantum mechanics. Our brain may be wired with an extreme bias to certain choices. And from anecdotal evidence i can tell you that human behaviour is mostly as flexible as a brick wall.
I's weird that Isaac thinks he is hard to understand, he speaks more clearly than I do. Great content!
That argument implies events that are governed only by random chance and statistics. After all, a coin cannot influence if it will land heads or tails. But humans can make decisions and take actions that will radically alter their odds of survival, either as individuals or as a group. Consequently, I think the Doomsday Argument is an interesting intellectual concept but has very little relevance in the real physical world.
Even if the universe is predetermined and there is no free will, the reality is that life is not perfectly random. This can already be seen even now.
not perfectly random, but predominantly random. Entropy is increasing as the universe cools. Expect things to get more random as time passes.
I have just started binging your channel. RUclips needs more like you!!!!
Isn't the fundamental problem that the argument assumes that personal identity is separate from the situation of our existence. To work, it requires that swapping me with a person born in 3000AD changes the universe in some way, but it doesn't. Since the two individuals will be in exactly the same circumstances they will develop into exactly the same person. The swap is not a swap at all, and hence the possibility of 'me' being born in 3000AD is meaningless since 'me' is totally defined my existence here in 2017
I think the freakish odds thing can be answered at least in part with "any system despite the odds must invariably find itself in a state" in other words the odds may be freakishly low but someone had to be born during this time period. and someone had to be born in the hunter gatherer period.
i mean you can literally extend this argument back and say that humans should have died out before getting to this point because it was very unlikely for someone to be born as a hunter gatherer pre-farming human then to be born as a post farming human.
I always find it weird to do statistics on one bit of data. Taking a 100 sides dice and throwig a 1 is unlikely but can happen. And if it happens, arguing, that this can't be because there is only a 1% chance of this happening is weird.
Or even worse: Have a continuous distribution and argue: the probability of this event that just occured would be zero if we have a continuous distribution, therfore there have to be less possible outcomes to make this outcome more likely.
At the same time you can't really argue against it.
Which is why the argument is not just taking the absolute chance of an event, but comparing two chances of two sets of events.
Though a single bit of data still is weird.
"At the same time you can't really argue against it." You cannot really argue in favor of it either. The only arguments which strongly support doing Bayesian inference/Solomonoff induction rely on repeated observations. With a single point of data, you cannot have more than a vaguely defined likelihood I think.
Yay a new Isaac Arthur video! Just what I needed to start my day with. :) By the way, your videos have been the topic of discussion between my fiancé and I lately. Last we talked about what was discussed in the Simulation Hypothesis video. Also, your videos have really helped turn my days around. I feel like I actually have something to look forward to - something fascinating and in-depth. It might seem silly but your videos have really made my days more tolerable.
A presumably-belated congratulations on your engagement, by the way. I think that may be the first time I've ever heard someone say they felt upbeat after hearing about the Simulation Hypothesis or Doomsday Argument, normally they tend to be so depressing I was actually reluctant to cover them :) Needless to say I'm very glad to hear people are enjoying them.
I'm also always looking forward to your new (lately weekly) videos. Always really fascinating topics for my weekend. Especially that they are so in-depth. Actually because they are so in-depth. "Normal" more shallow RUclips-Videos about such topics, just don't cut it for me anymore :) I really like your scientific approach to everything! Saying this as a physics student in the second semester :)
I started reading Nick Bostrom's material about 18 months ago. When I came to his analysis on simulation argument / simulation hypothesis I was very much in the doldrums for a few months. :) Who wouldn't be?
14:24 "Well I'm a gamblin' man says one of the committee" Fantastic government decision making xD
Been binge watching this series from the beginning very interesting
There's something about this argument that I don't like, but I can't pin it down. I think I need a better understanding of probability first.
Good video, Isaac.
Something less depressing next time? :P
The Doomsday Argument is like that, it takes a while to really absorb and doesn't 'feel right' to anybody, at least not anybody I know myself included. But the end of the world depressing? Nah ;) I'm actually considering an entirely unplanned, non-queue video on an upbeat topic next time around just to clear the air.
I think it does not 'feel right' because one of the premises it poses is wrong. The logical conclusion is irrefutably right, but it makes an assumption which in my opinion is not. This assumption is that your consciusness is in a 'pool' of consciousnesses that are randomly assigned to physical beings existing at a set time. I would bring the counter argument that the physical state you are in gives rise to your current consciousness (and yes, that could mean that you are just existing in this one single instance, after all we can never be certian if we experince passage of time). Only universes (if there are multiple) that lead to your exact physical state, or other locations in our universe that lead to the exact same physical state (assuming our universe is extremely large, or infinite), are in the 'pool' of selection. Now, the chance of you appearing in a certain civilization/culture in the same universe multiple times is extremely unlikely anyways, thus you can not make a statistical conclusion about the faith of your civilization/culture based on the observation of your existence.
Commenting to boost algorithm for this underrated channel.
Amazing channel.....
Personally I think trying to apply too much math and probability to problems such as these implies too much of an assumption of strict order, while the universe itself appears to be a mix of order and chaos, especially when life is concerned. Let alone in sequence of events probability breaks down as reliable, since there being a miniscule chance of you being born earlier in history doesn't change that somebody had to be that person and all the earlier people that led to them had to exist.
I think the Doomsday Argument is perfectly sound and probably correct, but only if we stay on Earth. The second we have a self sustaining colony off Earth, be it on Mars, one of the various moons in the solar system, or even something like an O'Neal cylinder with good enough hydroponics to be able to survive without Earth the odds of our survival increases drastically. That 1% per year chance of us all dying each year plummets massively if we're on two planets and it continues to plummet the more we colonize other bodies in the solar system. If ever we get colonies to other stars (I'm extremely pessimistic about that, but not completely convinced of its impossibility) then the odds of us being wiped out plummet to a ridiculous low.
Also I'd like to point out that the core assumption in the Doomsday argument, that we're about in the middle, would lead to incorrect conclusions about how many total humans will ever be born for about three quarters of all humans who will ever live. It really only works for the quarter or so who really are about in the middle. Since we have no idea whatsoever where we sit in that order, and since it's entirely possible that the total number of human could fall ANYWHERE on the scale between 100 billion and several quintillion, it really messes with the odds. The 91% only works if your reduce it to two options. If you change the door scenario so that you don't know how many doors there are but only that it is no less than 10 and no more than 100, which is much closer to our real situation, then knowing you're in door seven doesn't make it so massively likely that there are only 10 doors. It's still pretty unlikely that there are 100, but anything up to (estimating here because I'm too lazy to do the math) around 25 or 30 is pretty likely and anything up to around 60 or 70 remains reasonable.
Isaac at the start of your videos you warn people of your speech impediment equating it to making it harder for people to understand what your saying. But in full honesty even from day one you've been quite easy to understand and at least to me are very well spoken more so than others that have no such impediment. I'd prefer listening to you talk about these subjects than most anybody lol so keep up the good work my friend, because your videos make my 45 minute commute to work much better
Shared in a dozen or so transhuman-inclined groups.
Thank you Khannea!
Great video! You deserve more subs; you explain things very well!
Calling bullshit on something here. There is NO improbability of being born early in a doom late group because you have to be born somewhere and any given position is equally probable and improbable as any other. Example: Drawing a specific hand of cards in a five-card poker game is about 1 in 660,000. Thus, getting a specific hand is very unlikely, but no matter what hand you DO get, it was a 1 in 650,000 event. If you drew a million hands in a row you would get a 1 in 650,000 result each and every time.
Thats why you have to divide it into sets.
I'm simply amazed at how well you covered this large topic. It's all fascinating really to think about these kinds of problems at hand; Involving these concepts, with largely debated logical reasoning instead of how people generally cover these terms. ohh by the way I don't know if you've covered this or not but I'd love to see a video based on Zero point energy!
Thank you Jacob, and no I haven't covered Zero Point energy yet, except by a few casual or related mentions.
your welcome, I look forward to many more great videos! Ohh I forgot to mention this but have you ever heard of the temperatures below absolute zero?
Not in any mainstream theories, no, folks often accidentally confuse things like 10^-6 kelvin as negative though.
OMG I was so lost blue doors red doors 1 and 10 50/50 chance my damn brain is seriously smoking from ears amd the overflow is comming out of my ass.
I was like "Whatever, I will not play those games, if I ever find myself in such a cell, I will just flip a coin."
naphackDT but that coin is one of the infinite other coins, assigned a chance value immediately as it can't be superimposed.
New viewer. I have been enjoying these. Trying to start at the beginning of your series.
How can we know if we can count ourselves as a random sample? And what about the gambler fallacy? If you have a 1% chance of extinction each year, surviving 300 years is unlikely, but if you survive 299 years, then you have a 99% chance of surviving, not 5%. Thus, we can only make the doomsday argument from the current day. Tomorrow, the human deadline with 95% confidence gets pushed back, yet the argument remains the same, even though the doomsday has changed.
So, I've more or less just discovered the channel, and I've been voraciously consuming videos, so thanks for the content!
I wanted to add, I wrote a small Matlab script because I was curious about the sleeping beauty problem. My results are coming up that the thirder position is clearly correct, so I wanted to see if anyone can poke a hole in my thinking on the script. I'll explain each step for people that don't know Matlab.
First, I initialize my array for the 3 possibilities of when SB is awakened (Heads and Monday, Tails and Monday, Tails and Tuesday respectively)
awake = zeros(3,1);
for ii = 1:100000 (starting a loop of 100000 trial runs)
if rand > 0.5 (testing a random number between 0 and 1, inclusive, pulled from a uniform distribution ... I've decided that the number being over 0.5 means heads came up)
awake(1) = awake(1) + 1; (Heads and Monday counter is incremented)
else
awake(2) = awake(2) + 1; (Tails and Monday counter is incremented)
awake(3) = awake(3) + 1; (Tails and Tuesday counter is incremented)
end (close if statement)
end (close for loop)
What my results were was:
awake = (50028, 49972, 49972)
The numbers are close enough that I feel pretty comfortable in saying that the difference is due to statistical noise, and the way I'm reading the results is that, over 100000 trials of this experiment (Sleeping Beauty has a true dedication to science!), that she was awake on a Monday due to heads 50028 times, Monday due to tails 49972 times, and Tuesday 49972 times. The means the odds of her being in a situation where she's awake and heads came up is clearly 1 in 3.
Is my thinking wrong here? Am I missing something?
Personally I believe‘the doomsday argument’ should be taken seriously.
We can reduce the chances of being wiped out as time passes by identifying the threats and dealing with them appropriately.
This concept has always hurt my brain. Thanks for this video it was really well presented.
gotta say been loving your channel and this so far is wanna my fave videos ive watched.
This went over my head, I should re-watch it... I'm not very good with probabilities anyway.
It went over your head 100% of the times you watched it so far. What are the odds you'd understand it the 2nd time? :D
Great video again! I was really excited to see this one. Also great new profile icon! It fits the channel well.
Thanks, it seemed time for a change, and to standardize my various social media profile pics.
Here's the issue with the Doomsday Argument: you must make the assumption that conscious existence in some form, either human or some direct or divergent successor, is finite. Making that assumption puts us in a precarious situation. It's an interesting thought experiment, but assuming our species advances to such an extent that a critical mass comes to be a) aware of the argument and/or b) are sufficiently intelligent and educated to understand it, it could trigger a dangerous brand of nihilism and an existential crisis.
In short, i'm not sure this is a useful idea to be sinking time into. This was, however, a well made video. Kudos. I've just subscribed.
You sound all right to me, no impediment. I love your stuff.
You are in a hallway full of doors. Someone slips a piece of paper under the door that says, "You are in room 7, How many rooms do you think there are?" There is still the possibility there is only 1 door... I never said the slip of paper was definitely telling the truth. Of course it is completely possible that there are thousands of rooms and they only put people in the first 10 to see what they predict.
That was pretty much my argument. Because we are aware of our past and our history, we are not "born" in room 7. "We" as civilization walk a sequence of rooms starting from room 1, and it happens that at room 7 "god" gives us the doomsday argument. The probability of this to happen somewhere around room 7 is 1, and gives no power of predicting how many rooms are left.
I suspect Nick Bostrom inflated this doomsday argument story out of nothing, and it was debunked long ago.
it seems a common assumption that in probabilities when there are several indications or hints contraddicting each other it's some kind of paradox - when really it should just essentially "even out" - that is a bit simplified - but if there are two options and we know nothing we can say it's a 50/50 guess - if now we find a hint or a solid argument indicating the chances are 90/10 in one direction - and another indication that they are 10/90 in the opposite direction - that's not a paradox, that's just an unlikely coincidence having happened and the indications we have even out back to 50/50 - or 60/40 or something like that depending on the exact example
Isaac, your channel is amazing! Thank you for your work!
Thanks Bjorn!
HOLY F*CK. Only 10 minutes into it but this has got to be my favorite SFIA video ever.
This is the one and only time I have ever and probably will ever say this. Thank you for reminding me about the subscribe button. I'm a space cowboy, so I would have just kept looking up your video's the hard way.
Did anybody notice 83,000 views, 2,000 thumbs-up & only 49 thumbs down!?! What a successful ratio. Isaac Arthur, your channel is great. You have replaced 100% of my conspiracy video watching with your shows. What are the chances of that.
What conspiracy videos did you use to watch?
This guy is amazing if he writes and performs the content! Wonderful work! I hope you get the success you deserve.
Another great video, excellent work :)
Thanks!
Isaac, I can't say this enough, but your videos are amazing! I don't useually subscribe to 'homemade' channels, but yours is just so informative. + love that i can watch this at night with captions on, so I won't disturb my roomies.
Hope you don't mind me asking, but why do you not have a larger reddit presence? - I'm sure that /science or /documentaries would love both your videos and your input.
Oh Reddit is something I've been trying to transition into using, I really didn't know of it till recently, but it's not quite clicking and I tend to feel time spent figuring it out is time better spent on the next video. A few of my subscribers have posted videos over there and asked me to pop in for Q&A but that's about as far as I've gotten yet.
this is the best channel (your great content-you) I've yet seen for realistic science! I cannot thank you enough.!!
A thought.. what if, as with variously accepted 'laws' the application of exotic forms of energy could cancel well enough any of them out? so the higgs field can be suppressed with countervailing energy so too could be other forces, couldn't civilizations using such understanding & with an enlightened viewpoint about equitable resource allocation effectively become invisible by choice? as much as I love your arguments in the main, the idea of actively choosing an upper limit to ones own resource usage seems totally alien to all of your views?
Thanks Issac, hope you and yours' have a great holiday weekend!
You too Robert!
Hey Isaac, another great video. I wanted to add 2 things. There is a point that a civilization could reach, where they would become indestructible. The civilization destroying scenarios are generally only planet wide, so once a civilization spreads beyond then it can effectively become unkillable. Or at least impossible to wipe out 100%, and remember that we need 100%, because anything less and they can rebuild almost instantly. It's like completely sterilizing a planet of all life, it's incredibly hard to do. Now imagine completely sterilizing all the planets/stars/systems/backups that a civilization could potentially spread through. Type II civilizations are generally considered indestructible because they can always rebuild and if there were super weapons capable of doing system wide complete sterilization, then we'd definitely see evidence of its destruction. Also, imagine a superintelligence that creates backups, puts them onto nanobots, and sends those nanobots off at 99.9999999% the speed of light to seed the rest of the universe. No matter what ended up happening locally, the entire universe could be seeded almost instantly and irreversibly with backups.
Second, have you considered the possibility that we'll actually get to live out both of the scenarios? Meaning that what if we're born at the end of human civilization and then we continue until the end of the universe as the superintelligence that we uploaded/merged with. The current assumption is that more biology will exist in the future (more numbers), but remember that our biology is already obsolete compared to what computers can do in almost every category. So if we agree that we'll eventually get an SI than it'll be a singular being and not a bunch of individual Ai's. Since it'll be singular, then the odds become even more impossible since we basically just multiplied the two scenarios together, oh and don't forget to multiply by the possibility of the universe even existing. We're both at the end and the beginning. Our universe is clearly a winner take all system, and the first winner will have it all to themselves. I think the answer to the Fermi Paradox is that we really are the first... and the last.....
I always wonder what protocol RUclips uses for it's filters, I just had to rescue this one from the 'likely spam' review folder :) No links, no foul language, no ALL CAPS, just regular text. I should probably make that comment-relevant with some remark about better AI for review systems but I just find it vaguely amusing and weird.
Anyway I don't think there's any 'totally safe state' for a civilization, a pissed off xenocidal AI could probably send of murderbots at a slightly faster speed then civilization preserving bots. I should probably also add that I would not consider an AI non-biological, biology is a pretty arbitrary term especially when discussing aliens at the same time. Biology is an emergent property of chemistry where self-replicating systems is concerned, I think that would give us a pretty broad leeway in that regard.
Yes, but that xenocidal Ai would still be left alive. We're talking total sterilization requirements. If you send off a backup copy at 99.9% the speed of light, in a matter of months it will be completely unreachable since it'll require more than 100% the speed of light to catch up to it. And even if you include FTL travel, then we're still in an infinite universe, you'd be able to send infinite backups out and seed the universe in minutes. The main point is that once life starts, it's WAY harder to wipe it out. It's resilient as hell, so if dumb life is like this, imagine what a god-like life will be capable of. Another analogy is like comparing it to the internet. When the entire internet was a few computers, we could have killed it, but at this point it's virtually impossible to shut it down because it's such a redundant, decentralized system. It'd honestly be easier to kill off all humans vs the entire internet. So more intelligence gives you the risk of developing your own doom, but your ability to survive/repair advances far faster than the potential damages. Another analogy, the Terminator from T1 to T2, the T1000 was much tougher, much more unkillable since it was significantly more adaptable. Also, imagine a Grey Goo scenario and think about how hard it would be to wipe out every last nanobot. The whole point is that, as you advance/evolve you become more adaptable and significantly more capable of surviving greater and greater threats. When was the last time you were worried about being eaten by a lion or going extinct because your habitat is being wiped out?
You're right on the non-biological part, I use it more to describe things that are driven by biological evolution vs self evolution. So people are biological because they're stuck being people, and robots/Ai's have no such limits.
The reason everything trends to a singular being is that once you have the ability to connect different minds together, then you have a hive mind which is a new singular entity, and all the individuals eventually become one thing. It's always hugely more efficient/beneficial to do all your thinking together as one vs individually. You can have individual minds, but they simply won't be able to compete with the hive mind (singular).
I think this got marked as spam is because I got involved in a bunch of other conversations below and that made me look like a spammer. I've been working on this for the last few years, so this is a crazy interesting topic for me and you're the only person on RUclips talking about the topic in an intellectually rigorous way. So sorry in advance for my long ass replies ;)
You were right, this really isn't intuitive at all! I loved it nonetheless. This is one of the best channels on youtube, please keep it up!
I enjoy writing, and I'm in the process of writing a sci-fi story right now, and the explanation of this concept really inspired me and helped me incorporate more complex ideas in my story, so thank you for that!
Also, the whole argument seems a bit arbitrary to me. I haven't watched the "Apocalypse How" video yet, so I don't know if that video covers this problem, but wouldn't a "Doomsday Soon" scenario involve a lot more variables that aren't really taken into consideration for this argument? Such as different kinds of probability for different kinds of "Doomsday" scenarios (among other things)?
What I find really counter-intuitive about the argument, however, is the emphasis on the improbability of us existing within the first 100billion people born. Sure, it might be improbable, but those 100billion people must have existed first for the rest to be born as well. Now, I don't know much about probability or statistics, and I did struggle to understand the concepts in the video, but wouldn't the existence of a trillion trillion trillion trillion lives presuppose and, in fact, assure the existence of those first 100billion people? I understand, though, that the problem lies in the probability of us, of me, let's say, as an individual, being born among those first 100billion people. Again, though, those first billions must inevitably exist for the trillions to come later, and that's what seems very counter-intuitive to me. All in all, I think I understood at least the very basic premise behind the argument. EDIT: So, in both scenarios those first 100billion people must exist, I guess. I mean, it's counter-intuitive, it really, really is, but I think I understand the reasoning nevertheless; especially when I parallel the situation with the 100-10 door experiment.
And one last thing, if we suppose that transhumanism is inevitable, wouldn't the fact that we're not in any way "transhuman" increase our possibility of being born prior to the invention of any such technology, therefore making it more probable that we're born now? What if we suppose that technology that significantly alters the state or physical form in which people are born is not far off into the future, how does that change our probability, if at all, of being born now? Although, if I understood this right, this "Doomsday Argument" works with the assumption that "transhumans" and "natural" humans, like us, are grouped together in a single group, and, perhaps, my aforementioned questions actually divide the two into two separate groups? Perhaps it's not two groups, though, perhaps, us being born as "natural" humans is akin to looking at the door number? Nevertheless, isn't that kind of an arbitrary definition or an arbitrary distinction or an arbitrary decision not to make that distinction?
In fact, this whole argument seems to be based on arbitrary definitions and assumptions? So it makes me wonder how viable it is for the real world, or is it simply just a thought-experiment?
EDIT: Also, when it comes to the 1% death per year argument, that doesn't make any sense to me since the results of the probability would depend wholly on the year you start counting, which seems completely arbitrary to me. Sorry for using the word arbitrary so much.
Well you'll have to let us know when it's out, I alway like hearing about any fiction that springs from these topics. :)
Isaac Arthur
It won't be out for years to come, since I don't really plan on publishing anything until after I finish my studies, but I will post on one of your videos when the time comes if you're still active. Although I could post it on the internet somewhere, I guess.
I'll be incorporating an altered version of the sleeping beauty experiment in the story I'm writing, although it won't be the main focus of the story. At any rate, I only discovered the experiment thanks to this video, so thank you for that!
Fortunately 'arbitrary' happens to me right there with 'indeed' as words I use way too often and am trying to ctuback on, so you're good :) The 1% death argument didn't have an arbitrary start though, as I recall anyway, I sometimes misremember the specific examples I make up for these and I generally dislike rewatching my own videos after I post them.
As I recall that one begins with someone cobbling together a doomsday tech design and mass distributing it, with an assumption they can't purge the knowledge and it will be flowing around occasionally making it into the hands of nihilistic folks, some small number of whom have the resources to produce it and will luck out and not be caught.
But it wouldn't matter anyway, if you and I could only exist because a fair coin flipped heads ten times in a row, it doesn't make that coin anymore or less likely to come up heads next time. Even if there always had been a 1% chance per year of us getting wiped, all that means is we've been very lucky so far or we botched our model. It wouldn't mean, if the former, that we'd keep being lucky. Only those events still to come matter, if you make it 50 years ahead in time when you had a 1% chance, the clock is reset to then, every year you survive negates that probability and starts it over again. Doesn't change the odds.
You said Carter, and I automatically thought Samantha Carter from Stargate.....
Hahahahahahahahaha
I actually think Jimmy Carter.
If Socrates had come up with this argument, then he would have assumed our species would not have lasted to the Renaissance.
One counter-argument for the 1% per year doom scenario is technology shifting the odds. For example, I'm sure that 1% change includes the hidden assumption that everyone is on Earth, so a planetwide disaster can hit everyone. As soon as we have the first self-sustaining colony anywhere else, a whole class of threats become invalid as 100% kill outcomes, and that 1% drops. I'm sure there are other, more subtle versions of this argument that could be made.
Another very well done video. I've heard this argument before and I find it a better way to explain how statistics work to people rather than a good Fermi paradox argument. If anything it's an excellent example of how useless statistical analysis can be in the absence of the scientific method. The Carter argument would support every apocalypse cult and prophet-of-doom from the Mayans to the Bible's prophets which means they should have been right and the world should have ended a long time ago and you and I shouldn't exist. And yet here we are.
Oh yes, it definitely opens the mind to the statistical traps one can get one's self into the further you get from hard data and experiments. However I think 'perilous' might be more accurate than 'useless', though I know what you mean of course.
Random Sampling breaks down when the coin flip algorithm is designed because the algorithm is not Random.
Head tweaker isn't it? Kind of like altering the behavior of a quantum particle just by looking at it or being bitten in the Hand by a dead Schrodinger's Cat even though you can't put your hand in a sealed Box without opening it.
These questions are very fun to play with. TY for the Show.
Offtopic: I already watched lots of your videos and I had no problem understanding your english. And I am not a native speaker. I believe you underestimate yourself. Keep up the good job
Even if a 1% annual chance of extinction is accurate now, that number will decrease as we become a spacefaring civilization.
Just having a sustainable colony on 1 additional planet will probably decrease extinction probability by one order of magnitude.
I imagine it would, but to be clear, I just made up the 1% chance to demonstrate the mathematics.
The biggest problem with the Doomsday Argument is that we have lots of data on life forms that have and have not become extinct over the last 100 years, and the Doomsday Argument does worse than a coin flip at predicting which ones will go extinct. It predicts that older species that had more individuals in the past and that have low birth rates (sharks and rays, for instance) are more likely to survive longer than newer species with high birth rates (mice, rats, and rabbits, for example). Would anyone be willing to bet that hammerhead sharks are going to outlast brown rats? I will reluctantly take your money.
There are two northern white rhinos left in the world, both female. Statistically speaking, what is the likelihood that these two individuals "selected at random" are the last two individuals in the species? It's millions to one! In terms of the Doomsday Argument, if five million northern white rhinos lived in the past, there is a 50% chance that another five million northern white rhinos will be born, and at their current birth rate, they will likely last... uh, forever, since no new ones are being born? An ecologist would give them much worse odds.
This is a theory that doesn't survive contact with facts.
Thank you for the great content, Isaac. I'm a huge fan.
People will always be afraid of what they dont understand :) . And im from denmark i dont need captions. You are loud and clear
Maybe you talk a little fast. But no harm no foul :)
Thanks for your work! I love your videos even when I disagree with some fiddly details or points of view. 98% of the time when I start to think, “Ah, but what about this?!” You address that issue next.
On topic: People often whine about not living in the Better World they can imagine. This doesn’t make the universe unfair. It just means that they have at least some imagination, and Nature is impartial. The point isn’t to live in that world, but rather to build it around you, and keep imagining improvements. That’s the function of life itself.
You can’t use Bayesian to decide what “could” be true from an open field. It’s for deciding which is likely to be true among actually known alternatives using feedback analysis. With no current data ping, nothing to analyze, it’s simply the wrong tool for Fermi’s Paradox. After you meet a bunch of alien civilizations and have some feedback to work with, then you can use Bayesian to close in on what the likely civilization content of the universe will be. For the FP, Bayesian is an aftermarket accessory when you have already answered the essential question.
Similarly, you also can’t win the “iocaine powder battle of wits to the death” using Bayesian. Both cups could be poisoned, and you can’t predict that with math or logic.
Bayesian works better in the Doors of Doom red/blue problem because at least you have some known parameters to work with. Even that assumes you trust those notes and door-numbers as reliable feedback, which I wouldn’t, because what kind of lying sadistic assholes lock people up in mysterious rooms and chase them around with monsters? Just psychologists and mass murderers, and I don’t trust either.
There are simple answers to the F.P. that allow for a lot of sensible alien civilizations out there, while also telling us some of the things these beings DON'T do. (You could apply some Bayesian to sort out what those things are, so you can concentrate on the better options.) Humans just don’t like it because we aren’t one of them yet, and we are still doing the wrong things.
When we mature and learn to behave, then we can sit at the Grownup’s Table and drink the ambrosia. Until then, here’s your ginger ale, go to your desk and study, and don’t badger the whole neighborhood with your tantrums. As Zed noted in Men in Black, "Sucks, huh?"
There's a lesson in the film clip: "The most unlikely threat" from Oscar Koeroo channel.
Thank you so much for making and posting these videos. I love your mind AND your voice. Fuck the haters. Dude ... you rock.
Thanks!
This argument gives me the same vibe as that paradox about guessing which day of the week something will happen on when you’ve been told the day will be chosen so that you cannot predict which day it will be.
Can’t be Friday! Then I’d know because that’s the last day it could happen. We can’t possibly be among the first and only humans, then we would be too lucky!
No closed captioning I love your voice. and I'm not gather is I'm well aware that they can understand your voice but I can give it time don't worry about it buddy
who would down vote this ? Arthur is making some really good, interesting discussion videos here! I am hooked :D Its the perfect mix between scifi and philosophy and maths :D
Thanks, I think it's the topic, it's kinda depressing. Still a 37:1 like:dislike ratio ain't too bad :)
Man, I just found this guys channel, I love it. He is funny and so good at explaining stuff in a way I can understand, hahahaha... I just thought he had a funny accent and once I got used to it, I don't have any problem understanding him.
So very insightful and thought-provoking, did I mention I love it! He is deep......
I like this video quite a lot. I am not sure how others feel about it, but I find some aspects of the doomsday containing a macabre sense of humor, albeit at the abstract level. Parts of this video that made me laugh my ass off -- 'Doom-soon' vs, 'Doom-late' groups; 'the probability of staying alive at 99% per year'. (A fun way of refreshing how to run simple logarithmic calculations!)