Comments by "Mikko Rantalainen" (@MikkoRantalainen) on "Veritasium" channel.

  1. I hope the original study at least randomized the order of answer choices. I wouldn't be overly surprised to find out that the order of choices was more important than anything else. That said, I totally know how non-natural bayesian statistics is human beings.
    52
  2.  @williamherd7251  I agree, if you use all fans on constant power using a single big fan would be more effective. However, if you actually used the possibilities of having different wind patterns using different speeds for different fans, the design shown here is actually pretty clever. Plus it seems to use off-the-shelf parts so it will be much cheaper than fully custom setup using lots of fans.
    40
  3. I agree, Veritasium, Vsauce, SmarterEveryDay and Sabine Hossenfelder are prime examples of channels that make YouTube worth using even if you wouldn't like all the ads and random stuff.
    26
  4.  @KyleTheFolf  I would guess the decision makers (maybe called product owner?) has just prioritized other features that they focus on. Software development faces the problem that you have way more ideas about how to improve the product than you have actual resources to implement. As a result, you have to drop lots of great ideas and implement only the features you think are most important. That said, I still cannot figure out why they decided that doing the extra work to change behavior of dislike button was worth doing. Or why they removed the feature to allow community to offer subtitles to the channels.
    19
  5.  @Beanbean1313  I would recommend looking up what Wozniak did when Jobs refused to pay the original engineers bonuses. It's clear from the start that Wozniak wasn't in the game for money. And Jobs was all about money and secondarily about other stuff. And yes, Jobs was good at making money whenever Apple had a product good enough.
    14
  6. 10:25 I think one way to think about this is to take the Newtonian equation F = ma and simply observe that the acceleration ("a") is also defined by change in speed over time. And if time is warped by gravity, it obviously follows that F=ma is not that simple anymore.
    13
  7. ... and in addition, assume 10% change that any deceiving behavior is a misunderstanding.
    12
  8. The explanation and animation was great indeed! It made me immediately understand why applying it for the wave function obviously makes sense.
    11
  9. I have the bell enabled and still I got info about this video just today. Now is 20th day and this video was released in 16th.
    9
  10. 29:05 It's much easier to learn to tie the underlying crossing the other way and keep doing the loop the same way you always did. The simplest way to debug the issue (that is, identify incorrectly tied knot) is to notice that one of the loops points forward on your shoe and another points backwards. Correctly tied knot will always point the loops towards the sides under heavy tension (e.g. after walking for a while).
    8
  11. The results will be used nefariously by the commercial company. As you saw in the end, this researcher wasn't after money but pure knowledge.
    8
  12.  @harrybhalerao8315  Get-there-itis was strong in this story. If the plan was to use radio to locate the only landing option and there's any problem with your radios (such as never hearing any response to any voice communication) it's time to return instead of keep going.
    7
  13. 4:22 I think it would have been even more interesting to have a ratchet mechanism so that once you're supposed to make left turn, you no longer turn the steering towards right no matter which position it currently has. That would probably have prevented even continous turn to left because you cannot (easily?) balance the bike during the cornering anymore. Also, having a driver that can do something like Fabio Wibmer or Danny MacAskill would have been interesting. Though, those drivers could have just raised to front wheel of the ground in case the steering doesn't work anymore...
    7
  14. 22:10 Make no mistake: meaningful "relationship" can mean something else but a marriage and kids.
    7
  15.  @fullaccess2645  The problem with flag is that it will the influenced by the fan but you still cannot pass the flag though the fan to see the turbulence after the fan. A good demonstration would be using some kind of RC plane to draw horizontal smoke lines orthogonal to direction of the vehicle movement and the fan would then run through those smoke lines while another drone high enough (and maybe somewhat aside not to cause vertical air movement towards the smoke) would shoot the lines. If Derek's explanation is the correct one, this vehicle should be able to run through the smoke lines and the part of the smoke line that is hit by the vehicle will be delayed next to all the other areas. The only question is, will the smoke stay in visible clusters long enough when released into the wind?
    7
  16. This just underlines how bad methods our even best educational systems currently have.
    6
  17. Right, I actually wrote the same thing and then deleted the comment when Derek said it immediately after I had written the comment.
    6
  18. I guess the game of life simulating game of life is nicer demo because the actual rules that game of life executes are REALLY simple compared to minecraft. Imagine how complex the starting state of game of life would need to be to correctly simulate minecraft engine!
    6
  19. What does surprise me is not using ECC memory for any critical process. For avoiding problems in CPU or caches, you can just execute all logic twice in series and report error if second result does not match the first result. So the max to avoid this whole error class is 2x CPU load plus the price of ECC over regular RAM. Modern CPUs have multiple cores so you can (and should, to avoid unnoticed repeatable errors in CPU) run both the "original" and "verification" tasks in parallel so there's no extra latency needed.
    5
  20.  @freshrockpapa-e7799  I guess it depends on where you live. The difference between quality of education between countries is enormous.
    5
  21. If you were hurt with the church organ sounds, you're listening all audio with too high volume and you should reduce the overall listening volume if you want to preserve your hearing. I thought that the church organ sounds were actually mixed a bit quieter than they would appear in real world.
    5
  22. i just realized that you could actually make the thumbnails to stand out more simply by having a blue sidebar like in this video at 17:19. When most thumbnails at Youtube do not have single color siderbars, using one could make it seem different from everything else!
    5
  23. Also, the current education system accepts that children don't understand maths of previous years before attempting new stuff. Maths builds on previously learnt stuff so trying to skip stuff (you didn't learn this last year, try to cope without) is not going to work and will just make people hate maths. The minimum bar should be raised so high that all children understand all the things during the math classes but they may still make lots of mistakes while applying those concepts. However, no children should be left without support when they fail to understand some building block that they're expected to know next year.
    4
  24. Chernobyl doesn't look like shown in this video either nowadays because the New Safe Confinement structure was finally transported over it in 2016.
    4
  25. The actual disaster site was inside Shelter Structure (sarcophagus) which was visible in the background while measuring the radiation at Chernobyl area. Nowadays there's additional New Safe Confinement structure around the sarcophagus, too, to reduce radiation levels even more. The radiation inside the sarcophagus is still too high for humans to enter (nearly sure death due the radiation poisoning in a couple of days if you ever visited the inside for longer than 15 minutes). I recommend reading the Wikipedia page about the "Chernobyl Nuclear Power Plant sarcophagus" because that old structure was already huge and it had to be built in insane radiation environment. The situation at Fukushima was really bad, too, but the end results are not nearly as bad as at Chernobyl. The Fukushima site haven't needed any additional extra structures to reduce the radiation similar to sarcophagus at Chernobyl.
    4
  26.  @UristMcKerman  When did astronauts end up on ISS without ability get back to Earth? Redundancy doesn't guarantee original timeframes.
    3
  27.  @somethinglikethat2176  Sure, I think the NASA management was thinking that the shuttle has internal redundancy so much that you don't need a redundant shuttle (that is, another one for backup). NASA had a huge problem with internal communcations. At one point the engineering estimated that there was a one in thousand change for a catastrophic failure during a shuttle mission. And the management was thinking there was a one in million change for the same thing! If management was this far off from the reality (if these groups disagree on probablity, I'd assume engineering is closer to reality), it's not a surprise mistakes were made.
    3
  28. I think a simple fix could be that publishers would be held accountable for any research they publish. As a direct result, they would be forced to publish any replicate research that results in different results from previously published research. If publications do not want to do lots of replicate research publishing, they should improve their own quality checking (basically more elaborate peer review) before publishing.
    3
  29.  @error.418  If I undertood the story correctly, they had only the radio system to locate the actual place to land. Dead reckoning and celestial navigation were just good enough to get into distance where radio should work for sure. Of course, that assumes that the antennas are intact and the radio is working, neither of which did they check after departure.
    3
  30. And on such a scale, typical human life is between one and two minutes! Another way to understand scale of things is to use 1m x 1m x 1m cubes and represent a year with a small 1mm x 1mm x 1mm cube. If you take a typical human life and organize it in a row at the bottom corner, it would be a narrow line between 50–100 mm from a corner (about 2–4 inches if you don't know SI units). The time from all the stories in the Bible would two a couple of straight lines on the bottom, the time from the asteroid that killed all the asteroids would be 65 mm (~2.5") thick layer of small cubes at the bottom of a bigger cube. And the time when life first appeared on Earth would be about 4 full big cubes.
    3
  31. The interesting question is, what are your requirements? If you want an everlasting light bulb and don't mind the efficiency, it's really easy to do. However, if you want spectrum distribution identical to black body radiation (that is, quality wise matching an incasdescent light bulb) and efficiency above 100 lm/W (better than CFL), then creating everlasting light bulb is really really really hard. And those white LEDs? The failure point of those is either too thin phosphor coating (white LEDs use phosphor to convert UV light to visible light and that's a wear part) or electrolytic capacitors having limited lifespan inside the inverter needed to power those LEDs.
    3
  32. Knowledge and culture (e.g. music and art) is the only thing that remains from humans in long timeframes. If you don't do neither, your job is to provide support for those that do.
    3
  33.  @talosgak1236  Speak for yourself. I've learned a lot from many videos I watch. However, that requires active watching, not thinking about other things at the same time.
    3
  34. To be exact, the CIA method was shining something with a laser and measuring the reflection. It obviously works but it's an active method which means it can be detected. The method in this video is similar to passive sonar in submarines - it cannot be detected because the listener doesn't need to transmit anything. And just like passive sonar is much harder to create than active sonar, the problems are similar in this version, too.
    3
  35. Why walk on two feet if you can use three and automatically result in more stability on solid surfaces?
    3
  36.  @antonvlaskin8496  Yes, "site" is unclear. Some people use it as a synonym for being at the location, e.g. the closest parking lot. Obviously you cannot visit the actual reactor because that would be way too radioactive still.
    3
  37. Autorotation is more like using the rotor as wings with really poor connection to the plane body. It's closer to airplane with a broken engine.
    2
  38. Any device with an integrated battery that is not easily user replaceable is an example of planned obscolence. Manufacturers will keep telling lies about why they supposedly cannot create thin waterproof devices with 3.5 mm audio plug and user replaceable battery even though such devices have already existed in the past.
    2
  39. I think most people do not understand what "random" means in this context. Maybe ask for "Pick a number between 1 and 100 you think other people wouldn't pick" instead? There a truly random value would be the best option because you assume most people are not able to generate real random numbers at will. And the least "random" number would be "Pick a number between 1 and 100 you think most people will pick" – I'm pretty sure 7 would win that one in Western countries and 8 or 88 in Asia.
    2
  40.  @vishwajeetprasad1729  I'm happy to buy last years products from mindless people following this year's trends as long as I can fix the stuff that gets broken. Currently, even if you can solder stuff like Louis, there's no way to get software upgrades to an iPhone once Apple stops supporting it. With Android, there's e.g. LineageOS.
    2
  41. The interesting part about iPhones is that many people rationalize their high purchase price by the "fact" that Apple supports those devices longer than Android manufacturers. However, if you actually check the Apple documentation, they promise zero support for their products. In fact, they could never release a single software update nor any spare parts and they would still not break any promises they have made to the public.
    2
  42.  @williamzame3708  Challenger disaster was NASA management misunderstanding about the amount of redundancy in the system. The design used redundant o-rings (there were two o-rings and the design was calculated to be safe even if one failed). However, the cold temperature caused common fault mode which meant that the probability of failure of single o-ring was no longer independent and redudancy was lost because another o-ring would fail simultaneously with high probability.
    2
  43.  @Saigonas  The original Cosmos was scientific TV series hosted by Carl Sagan. The newer variant is a TV series hosted by Neil deGrasse Tyson. I think this episode was closer to the newer series.
    2
  44. Zac doesn't need to have extreme hobbies because his body will produce all the adrenaline he can ever need during normal business hours.
    2
  45. Great video! This is easily the most clear explanation of FFT and it also explains the history of it. And it again makes one wonder just how awesome mathematician Gauss was!
    2
  46. Superb episode!
    2
  47. "Nothing to hide; nothing to fear" is another way to say "I don't understand what you guys are talking about but I guess it cannot be anything important."
    2
  48. The example with coins was great but it made you think that to go around the other circle, you must rotate double the "expected" amount instead of "expected" + 1.
    2
  49. I like to say that I'm expecting time after death to feel exactly similar as the time from the start of the universe until my birth.
    2
  50. Yes, this wouldn't work for plane because that cannot use the friction between the vehicle and the ground.
    2
  51. 7:15 The image is made out of round blobs because your light source is a round thing called the Sun instead of a true point light that exists only in mathematics. The slit works a bit like pinpoint camera and you end up with the image of the sun (multiplied by the interference of the lightwaves).
    2
  52.  @shashwatsen7150  Sunlight is definitely not coherent light. It has photons with different frequencies.
    2
  53. I don't think Red Bull will do that unless they can send live video back. Our current understanding of general relativity suggests that there's no way to communicate any information back over the event horizon. Red Bull seems to be a firm believer of "if nobody is seeing the tree falling, the tree is not falling".
    2
  54.  @timoord1611  Supposedly this vehicle can do faster than wind so any smoke in front of the vehicle should go through the fan because the vehicle wouldn't run faster than wind. As such, a good demonstration would be to have the smoke quite a lot in front of the vehicle and when the vehicle catches the smoke (which is moving with the wind!) it's both obvious that this is indeed faster than wind movement and we would be able to see the wind movement near the fan.
    2
  55. I think you should do similar video about every great scientist. For well known ones such as Einstein, you could obviously focus on something less well known aspect of his inventions for that episode.
    2
  56.  @WlatPziupp  As I see it, it's a valid research platform for teaching AI to keep the robot upright. However, I don't agree with the design limitation of having two feet with multiple joints when three more simple legs would be more stable and easier to manufacture.
    2
  57. I would guess you could fake a poor test result by forcing yourself to give answers faster than the test requires and then trying to do your best. For example, after every test case, take a 30 second wait time without looking at the next test case. Because g-factor is mostly about how fast you can execute tasks, that should be enough to bring your totals down.
    2
  58. You can think it as a lot of matter on the surface of the start falling towards the core. When that fall is long enough and the star is massive enough (meaning the acceleration of those atoms is very high) the collision of the atoms in the middle is hard enough to fusion elements that the star was not able to fuse while it was going normally. And every time fusion happens with elements lighter than lead, you get extra energy. And the extra heat from that fusion getting hit with more atoms still falling from the surface will cause even more heat and fusion with very rapid chain reaction. I would guess that the reason supernovas always look similar is related to the fact that elements heavier than lead do not release energy when fused.
    2
  59. And look at all those trees! It's pretty clear that the slide happened pretty deep in the soil.
    2
  60. 13:15 I think you could extract even more information from the GoPro video if you normalize every frame to have range from 0 to 255 (assuming you had 8 bits per channel recording). I don't know if it matters if you normalize the values in log or linear colorspace, though, so maybe check both?
    2
  61. You're so going to hate HDR TV sets in the future if you watch movies in dark room...
    1
  62. And everybody agreed that Earth goes around the Sun, not the other way around.
    1
  63. Can you use a laser (with a lens) instead of led as the light source? That should allow getting more photons for the high speed camera.
    1
  64. Considering that one average banana has mass of about 0.2 kg, 20 million bananas would have total mass around 4000 tons or about a hundred big semi trailers fully loaded with bananas, so you would be crushed under the bananas even before you could eat those!
    1
  65. If I understand this correctly, basically the wheel breaking force is used to rotate the fan which then moves air faster than wind through the fan. As a result, the forward speed can be up to wind speed + the speed of air moved by the fan. The video failed to explain what the left side control actually did; was it similar to helicopter collective control?
    1
  66.  @Nabeelco  The fact that rotor spins during the autorotation simply increases the wing speed and therefore causes more lift. Of course, if the rotor stops spinning, the wing speed is so low (the wings being the blades of helicopter rotor) that the helicopter is going to fall.
    1
  67. 6:00 Pro tip: if you are trying slackline for the first time, setup it around 30 cm (1 ft) above the ground, not high.
    1
  68. Superb video and the first time I really understood how the bottom part of a modern sewing machine works!
    1
  69. To get pretty good idea how different stitches are made, try search "groz-beckert stitch type slow motion". That should result in multiple animations by a German sewing machine manufacturer.
    1
  70. These are not the astedroids you're looking for.
    1
  71. 16:26 Big amount of snow moving on the hill looks surprisingly similar to foam on top of the sea.
    1
  72. "Science, like magic but real"
    1
  73. Great video! I would have liked to know more about the artificial snowflake initial seed - I think the video talked about saphire in some points. Why does that work?
    1
  74. That's a great example of human vision system trying to detect faces everywhere.
    1
  75. THIS! I can proudly say that I have never owned a smartphone that I cannot replace the operating system. Sadly, I cannot say the same about all the cars even though I do most repairs to those too by myself.
    1
  76. Imagine travelling to all these place only to make this video!
    1
  77. To capture content with really high luminocity scale you could use two cameras with one shooting through partially reflective mirror at 45 degree angle and another shooting the image via the mirror surface. If you use e.g. 90% reflective mirror, you can use the same exposure for both cameras and have 10x more light for one camera. Of course, this requires very solid stand for the cameras because you want to make the cameras pixel perfect to the target even though the cameras are orthogonal to each other. You also want to use prime lens for both so that differences in zoom do not mess your resulting image. And of course, you can use more mirrors and cameras to capture more exposures. When every camera uses identical settings (aperture, shutter time, ISO, etc) you get perfect image when you combined all the resulting videos from all the cameras.
    1
  78. There have been claims that bears have actually much better ability to detect tinyest amounts of smells – even better than dogs. It would be interesting to know if this is actually true.
    1
  79. Great video! I didn't know that graphite in pencil is hold together only by Van de Waals forces.
    1
  80. Have you already seen videos by Vsauce? He, too, is pretty good in mind blowing. HIs "How Earth Moves" is a nice counterpart for this video.
    1
  81. I think we would need worldwide legislation that if you publish some piece of news on your news website or other distribution site, corrections to said piece of news must always have equal or greater visibility on your platform and you're required to publish such corrections if you published the original piece of news. That would put at least some incentive to publishing probably wrong stuff on the front page of your publication because you would highly probably end up publishing the corrections later with similar visibility. If you would continue to publish crap, at least half of your content would be corrections only pretty soon. And note that this should include all news, not just science related news. Obviously yellow journalism wouldn't want any of this.
    1
  82. Am I the only one that doesn't get clickable objects over the video at the end?
    1
  83. Would it be possible to grow actual Gecko hairs on some artificial material using live cells from a gecko? If I understood correctly, once the hair has grown, it's not a living thing (similar to human hair).
    1
  84. I think this would have been better without the music in the background while you're directly in the camera. Background sounds in the clips are great. You should not be afraid of having some silence in the video. Other than that, very good story telling and timing.
    1
  85. It seems to me that the wave collapse on measurement is such a huge sudden change in the system that it doesn't make sense. I've yet to understand why Pilot wave theory is not assumed to be the more accurate description of the universe. I do understand that Copenhagen interpretation and wave function collapse are mathematically easier to handle than Pilot wave theory – in the same way as Newtonian physics are easier to understand than the general relativity. Pilot wave theory explains the same real world measurements without a need for sudden collapse of the system on measurement and without a need for parallel worlds interpretation (many worlds theory). My guess is that Pilot wave theory is closer to reality but many worlds theory may still be usable because it results in the same predictions for all practical experiements and the mathematics are much easier to handle. Thatmakes many worlds theory the metaphysical truth: not exactly the real truth but pretending that the world behaves according to that description still results in practically flawless results.
    1
  86. Good video and I think you could have underlined the fact that science works this way in general – some research results are incorrect or incomplete and other researchers will comment about those details and then additional research is done to measure if the difference is important. Another thing that would have been worth explaining more is the situation where the circuit is not complete. The power is delivered over the gap in 1 m/c but the power will drop back to zero after some time (depending on the impedance and capacitance of the whole circuit combined with the resistance of the load).
    1
  87. I understand that gyros allow the mouse to understand the true orientation even if a wheel slips. However, how do they track the distance moved because even with the suction effect, wheels may slip and the true movement is less than the wheels rotate. Are the accelerometers accurate enough to integrate the distance travelled?
    1
  88. Try watching Cosmos: A Spacetime Odyssey hosted by Neil deGrasse Tyson, too.
    1
  89. I think you should have used land survey telescopes (or whatever those things are called which are used to measure building foundations using trigonometry and high quality optics). Those would have made it possible to accurately position the helicopter over the target area. Another option might have been high quality cross lasers and sensors in the helicopter. Again, use the stuff certified for building bridges and you should be fine automatically. Cell phone GPS in a partial metal container (helicopter) is not going to be accurate at all. And there exist helicopters with computer stabilized hover which should reduce the pendulum effect a lot. As would shortening the pendulum arm/rope.
    1
  90. 10:25 The face you make for "Oh, that aged well..."
    1
  91. Considering that atoms do not scale, your skin and walls of your veins would be about 100x thinner and would probably burst wide open from smallest forces so you would probably bleed to death even without the blender starting. Even with MSc in software engineering I think the biological limits would be the real issue here. And yes, I agree that this would be pretty bad interview question if you actually measured its effectiveness vs long term performance.
    1
  92.  @MthaMenMon  Wow! Combination of poor educational methods with limited number of tries. I haven't heard about anything equally stupid in education for a long time.
    1
  93.  @physics77guy  Here in Finland deriving formulas is not really done until university studies. And the skills of people suck a lot as a result.
    1
  94.  @freshrockpapa-e7799  Are you honestly claiming that education in schools has identical quality in different countries? I would recommend checking PISA research results one day.
    1
  95.  @freshrockpapa-e7799  I was replying to comment "no it isn't". How do you know what was actually meant and my interpretation as "identical" was incorrect?
    1
  96. 22:05 The beating effect from 261 Hz to one ear and 263 Hz to another ear with fully separated closed back headphones is nearly nonexistant for me.
    1
  97. I think the biggest problem with this video was that it started with electric car history (which was great, by the way). However, you didn't continue with the evolution of electric cars to current day as would have been usual for such content. As a result, many viewers felt that this was done to avoid mentioning Tesla. You even sensored the Mercedes logos in the beginning. You explain that the decision was yours and I believe it. Compared to your video about Mercedes-Benz Active Brake Assist and other Mercedes episodes, this failed to explain the science/tech instead of feeling like paid ad. The sponsored content is not bad itself as long as the video is and feels objective.
    1
  98. That's his April 1st prank – desyncronize with the rest of the world by a little bit.
    1
  99.  @eliasbergstrom5300  People have trouble truly understanding even 10000 years. Even 200 years is hard to grasp for real and that's only about 10 generations.
    1
  100. Stock markets cannot be truly random as long as people without understanding of statistics are buying any meaningful amount of stocks. I've been thinking for years that stock markets are about betting about what other people believe to be correct value for the stocks in the future. There's a lot of psychology in that equation and a lot of superstitious behavior (e.g. stock value graph having 3rd spike changes the overall direction of the stock because many enough people believe that 3rd time the charm; similarly, many people believe in special numbers such as 10, 100, 1000 and if stock value ever gets close to those numbers, many enough people suddenly change behavior again).
    1
  101. I misunderstood the start of your comment to mean that you had IQ 71 in your latest IQ test, not as 71 years old. As a result, I was wondering that you're writing flawless English and switched to computer programming despite such a low IQ result.
    1
  102. It's a damn shame that the leadership in Nichia seemed to had vanity pride instead of willingness to improve. I guess it goes well with the Japanese work culture where the "boss is always right" and nobody is allowed to question any decision the boss makes and nobody is even allowed to deliver any bad news to the boss. If you deny all negative feedback, how do you think you can ever understand the reality correctly?
    1
  103. So this is how life based on Intelligent Design would really look like.
    1
  104. Great video! I didn't know that the energy from the avalanche was enough to partially melt the snow but it indeed explains why it's next to impossible to get out from the snow after getting buried into the snow. Of course, if you're buried deep, snow weights at least 100 kg per 1 m³ so being buried a couple meters deep would keep you in place even if it didn't re-freeze.
    1
  105. I think you missed to opportunity to make this video 37 minutes long.
    1
  106. I think the way you introduced the older clip around 10:42 made it appear that you're going to explain some additional details that were not explained in the previous video. However, it turned out to be just the same clip without any additional explanations. It would have been nice to say before the clip that you'll not comment about it any further.
    1
  107. Superb video as usual!
    1
  108. You should repeat this demonstration in a freezer because one could argue that this was about metallic string tranferring heat from the room into the ice and the melting of the ice happened around the string because of that. And obviously the water will freeze again after the string has gone away because there's a big block of ice nearby. If the room had been below 0 °C this cannot happen because the whole string would have been sub zero for the whole experience. I see you already did this: https://youtu.be/qQCVnjGUv24
    1
  109. So, if you're on board of a really fast spaceship and turn on your average flashlight, the emitted light to observer in rest will look like gamma radiation, right?
    1
  110. I definitely recommend watching the Cosmos series hosted by Neil deGrasse Tyson.
    1
  111. It's important to understand that when man went to the moon, we still didn't have a technology to fully store the analog video sent back to the earth. And we have practically lost the abilitity to go to the moon and yet anybody can record 4K video today.
    1
  112. The pale blue dot speech is always a nice thing to include in any video. I think it would have been worthwhile the include the next paragraph, too, about fighting against other people for temporary control of a small corner of the pale blue dot. If we teach the next generation nothing more than basic math, reading, writing and full understanding of the pale blue dot speech, the next generation will be much better than the previous one.
    1
  113. Yeah, imagine the amount of shaking after your first day of work.
    1
  114. Great video! I would liked to see some mention about how though this metal is and how much friction you can get without having rubber on the surface.
    1
  115. I think that's actually hysteric laugh caused by the fear. If that snake were to bite Zac causing his grip to release, he would be the next one in the line.
    1
  116. Hydrogen is not going to fly because hydrogen is too hard to produce which causes overall effiency of hydrogen be lower than charging and discharging Li-on batteries. Hydrogen may be used for rural areas because pumping hydrogen is faster than charging batteries but it will not be the big picture.
    1
  117. 8:18 Nice camera trick!
    1
  118. According to the info I could find, you would need ridiculous heavy weights to cut ice colder than -3.5 °C using this method assuming the air temperature outside the ice is also -3.5  °C or lower. And if you try to hang e.g. 100 kg weights to that tiny wire, it will snap for sure.
    1
  119.  @josephbrennan370  Netflix used to have Cosmos: A Spacetime Odyssey but it wasn't their original series.
    1
  120.  @vorname1485  I think the original response was about the fact that Netflix probably wouldn't finance creating such a science series. They might just distribute it if it turns good.
    1
  121. I think that the rated voting system (approval voting) is simple to understand but it would result in strategic voting because if you have one good candidate X and one barely acceptable candidate Y, you don't want to select X+Y in the election because that would give too much power to Y. Many would rather just vote for X to avoid ending getting Y selected in case it would be the logically 2nd best option for many people but never the best option for anybody. If you actually wanted to have regression to mean though, this would be the way to go. I'd rather want a voting system that can switch to radically different candidate if majority of the people voted for him or her. It seems to me that the least bad option would be to use Condorcet voting system with Schulze ranking. The bad part of that system is that it's next to impossible to easily calculate by hand. In practice, we would need to use computers to calculate the initial result and hand calculation could be later used to confirm the result. However, you wouldn't need to trust random computer but every interested voter could take the whole list of votes (this would obviously need to be public) and run the program of their choice on a computer of their choice to verify the official voting results.
    1
  122. I'm actually a halfer but I clicked "like" anyway because I'm trying to train the algorithm for my preferences correctly. The the sleeping beauty question, I understood the question as "what's probability for the coin toss result for this wakeup" and for that, the probability must definitely be 50%.
    1
  123. The search for Mersenne primes (the program called prime95) has provided us the best known software to verify your system is stable. The software has "torture test" mode which re-computes some of the really hard tests with known results and if your system fails to compute the exact same result, there's something wrong with it. The interesting part is that a single bit error in any part of your system will result in incorrect result and the computation can be configured to require as much RAM as you wish. As a result, if you configure it to use all the RAM in the system and it can successfully compute many numbers, your system is running bit perfect. And if you test it with any overclocked system, the changes are that the system is not bit perfect. Overclockers call system "prime-stable" when it can run prime95 torture test for 24 hours without any errors. So you could argue that this weird mathematical process has already provided us the best known method for verifying computer hardware correctness! As a bonus, the software is so well optimized that it can typically generate maximum heat for the CPU thanks to running on all parts of the CPU in parallel. As a result, it will also test the cooling solution of your computer in addition to checking it's running bit perfect.
    1
  124. I think they should introduce a bridge or two to the track to make the map making part much harder because true 2D algorithm wouldn't be good enough anymore. Of course, that would require setting max height for the mouse to make it fit under the bridge.
    1
  125. [insert standing ovation emoji here once it's released]
    1
  126. My thoughts exactly! I remember listening a lecture about FFT during my university studies and the teacher could as well not be there because the way he tried to teach it provided zero value. If this video had existed then, I could have watched this video just once and understood in less than 20 minutes the whole thing.
    1
  127. I wish you had deinterlaced the video clips used in the intro. Some of those clips had hideous combing artifacts because of using interlaced video clips mixed within progressive video. You should use something like FFMPEG "bwdif" algorithm or better. Other than that technical issue, great work as again!
    1
  128. Moving into headwind is obviously possible but not with this exact same design because this design requires that the vehicle can move slower than wind into downwind even if the propeller is anchored in place. Here's a very simple setup for moving directly into headwind: imagine a wind power station mounted on top of a tank where the tank has an electric motor. The wind is used to generate electricity which is then used to move tank. It's obvious that the tank can move (slowly) to any direction as long as the windmill is rotating. In this case, too, the tank is hold by the friction and the windmill extracts the power from wind movement relative to ground.
    1
  129. I think the "pivotal voter" example is an artifial problem that doesn't happen in reality. It basically argues that if the voting results were otherwise perfectly tied, there must not be a new ballot that tips the balance one way or another because the voter with the new ballot could be declared as "dictator" because it broke the tie. If you want to avoid nearly arbitrary results for near perfect ties, just compute the pairwise ranking between the resulting ranking and if the winner doesn't win with at least 1% margin, organize a new round of voting. This would force another round of voting until enough people adjust their ballots so that at least 1% margin between the best and second best candidate can be found. The change of resulting in more than two rounds would be practically zero with real world humans as the voters as long as you make the results of previous round public.
    1
  130. See also: https://www.ted.com/talks/nick_bostrom_what_happens_when_our_computers_get_smarter_than_we_are
    1
  131. Such a vaccine is called "education". It turns out that the people that would most benefit from such a vaccine are denialists that will avoid it in any case.
    1
  132. To me James Clerk Maxwell is even more important than Einstein. As far as I know, Special Relativity and General Relativity directly follows from Maxwell's equations when you apply the single assumption that time is relative and speed of light is always contant. Obviously, that's a pretty bold assumption to do and you have to solve lots of complex math about the results which is where Einstein did most of the work. It could have been turned out that the universe doesn't work that way and the bold assumption is not true, but luckily for Einstein, his assumption was the correct one.
    1
  133.  Science Revolution  Walk away from the computer and breath. You might have your brain overheating slightly. Photon is not a particle in sense that you could stop it and measure its mass. However, some effects that photons cause to matter behave like it were a particle. And some other effects behave like it was a wave instead. This is called wave particle duality. For more information, see https://en.wikipedia.org/wiki/Wave%E2%80%93particle_duality As for the stars and galaxies billions of light years away - we don't know if they still exist or not. There's no known reason to believe that they wouldn't still exist but if something happened to them say 100 million years ago, we still wouldn't know about it because according to Einstein, there's no way any information can travel such distances faster than light.
    1
  134.  @TeguhNugrahaPratama  It's also interesting to note that Maxwell was able to correctly measure the speed of electromagnetic force fields caused by electricity and it was only much later figured out that light is actually electromagnetic radiation, too.
    1
  135. Except that origami is like texture unwrap with zero seams allowed and no strecthing either.
    1
  136. Yet another example about why patents and copyrights are detrimental to society.
    1
  137.  @leftear99  What kind of lifespan do you think would be good for society? It's clear that the current 75–150 years for the copyright and 20 years for a patent is way too long and just hinders innovation but I have trouble imagining that even short timeframes would improve things. If we both agree that the current time periods are problematic, why do you think a shorter but still fixed timeframe would be better than zero timeframe?
    1
  138.  @leftear99  Sure, but you could create content and inventions that we currently based on pre-arranged contracts. Electrician is not going to randomly build stuff in my house and then request payment from me after the fact when I try to use the installed wires. Instead, I agree on payment with the electrician before he or she starts the work. I see no reason why creative work wouldn't work the same. (And as an another example I'm writing software for living and I know I get paid before starting the actual work.) As such, there's no need for altruism to create stuff even if copyright nor patents existed. The modern business models have been created based on copyright and patents but it seems to me that this solution (copyright + patents) causes more problems than it provides good. The only big question is how to convert to better system without collapsing too many businesses?
    1
  139.  @jesperlett  I think it would be incorrect to say that slipperyness of ice has "nothing to do with pressure". Surely the pressure also helps when the temperature range is between 0 °C and -3.5 °C but outside that range, the pressure alone does very little.
    1
  140.  @ExtantFrodo2  Don't trust GSM (or any other public phone standard) to connect phonecalls or text messages correctly and never ever trust the caller id for phonecalls nor text messages. How do we communicate over untrusted network called the Internet? With encryption on top of the untrusted data transmission channel. When you use something with end-to-end encryption, you'll be safe. Something like Signal or Whatsapp (assuming you can trust FISA secret court order has not been forced Whatsapp to install backdoor to your phone) would be safe. This is why some countries are actively trying to ban encryption from their citizens. That would prevent them from controlling/monitoring your communication channel over phone at will.
    1
  141. I think a great way to think about possible roadblocks in your job is to stop, think about two possible solutions for it and then go to your manager to ask if either one would be a good solution and ask them to explain why. This accomplish two things: you communicate to your manager that you can invent solutions by yourself and you can validate your own solutions. At first, it might be that you fail pretty often but once you do this for a little time you'll notice that your solutions might actually get better than what your manager could have come up with.
    1
  142. I think the speed of light must be at least very nearly symmetrical because otherwise cosmic background radiation should be skewed towards the direction of slower speed of light.
    1
  143. Great video! Definitely in top 3 of all the videos you've done this far!
    1
  144. See also: https://youtu.be/sCuKuUgNfjA
    1
  145. Would it be possible to use protective gloves and clothing while handling those highly dangerous snakes? I think we have materials that would withstand the teeth of the snake.
    1
  146. If you want to find more about sidereal day and Earth movement, try video called "How Earth Moves" by Vsauce.
    1
  147. See also: https://xkcd.com/1877/
    1
  148. For more details about USB superposition, see https://www.reddit.com/r/ProgrammerHumor/comments/226qum/usb_superposition/
    1
  149. It's often easy to forget that in long timescales, DNA is more permanent than rocks. And that's because it self-replicates.
    1
  150. 11:15 Imagine building 1000x heavier jumper that still pushes 300g to the launch point during the jump. I would guess that the jump surface would need to be made out of hardened steel plate on a concrete platform.
    1
  151.  @20firebird  I think the cat is an important part of the thought experiment because it's used to extend the dynamics from the single atom to macroscale. As a result, the cat observing its own state is a huge problem for the experiment.
    1
  152. Are you saying that if he had been American, he would been called e.g. "Oliver Stone" instead?
    1
  153. Great video! Not only it told an interesting story about history of blue LEDs, it also helped me to understand the band gap, well, and the hill used in blue LEDs.
    1
  154. Honestly, I was expecting at the start that you're going to allow the concrete to harden and then demonstrate that since humans are softer than concrete, you can break the concrete around you without breaking you.
    1
  155. 1:20 "We are not astrology majors." I wouldn't been able to not respond "That's cool because this is an astronomy question."
    1
  156. 2:20 A great demonstration of a chess master figuring out the pieces on the board can be found in YouTube video titled "Magnus Carlsen's Mind-Blowing Memory! World Chess Champion tested". The idea is to who arrangement of pieces from middle of the game from various chess master games over decades. For example, Magnus Carlsen (born in 1990) was able to idenfity "24th game Kasparov vs Karpov from Sevilla 1987" before all the pieces were even positioned on the board.
    1
  157. I think the examples you provided at the end were definitely legitbait – show the most interesting thing about the video as the title and thumbnail. A clickbait (for me) is when the title or thumbnail intentionally witholds the only real piece of information provided just to get people to click it, or the title or thumbnail is intentionally misleading and promises stuff that the content cannot deliver. That said, I'd prefer longer titles with more details. For example, you could have used "The simplest math problem no one can solve: the Collatz conjucture" where the start of the title tells what's it about for most people and mathematicians can skip it if the already know everything about the collatz conjucture. I think the thumbnail with "3x + 1" was definitely legit. And as your videos nearly always explain "why" I think it should be included in most titles. For example, I think "Why backspin basketball flies off dam? Magnus effect!" would have been more descriptive title and I would guess that this kind of QA style would work for video titles, too, similar to Quora or Stackoverflow.
    1
  158. Great video! It really makes me think that normal CPUs would need a couple of analog matrix multiplication units where loading the matrix constants would be a slow operation but computing output for a big array of input would be fast. Because normal CPUs already run fully digital otherwise, it would make sense to return to fully digital domain after each such analog operation. In addition, the variable resistors wouldn't need to be accurate but only stable – you could compute calibration values for each resistor by inputting known input arrays with known output values and compute the suitable calibration values (basically input identity matrix multiple time with rows and columns mixed for each retry).
    1
  159. Very interesting riddle and I definitely wasn't able to figure out the correct solution. However, as you can change the loops simply by offsetting the box numbers, would following be a better or worse strategy? After opening 26 boxes, do not follow the sequence. Instead go to your own box and add 1 to the number in the slip and open that box. Continue doing that for the next 24 boxes. If you end up with a box number that you've already opened, add one to the slip inside that box. Is the probability of being in a loop shorter than 26 boxes in two different loop configuration higher or lower than being in a single loop shorter than 50 boxes?
    1
  160. Did your computer use ECC memory? Most common problem for random crashes is slightly faulty memory and without ECC RAM you only see the consequences instead of getting a warning or error message about faulty RAM module.
    1