YouTube2Text

Tickets, A Necktie & 26 Years of Footage: Long-Form 3 — Transcript

by Van Neistat · 3,585 words · 557 segments · language en · Watch on YouTube

Full transcript

  1. 0:05Long form episode 3, a vehicles episode
  2. 0:08of sorts. I am being sucked into the AI
  3. 0:11rabbit hole. I know that will disappoint
  4. 0:13a lot of you, but more will be revealed.
  5. 0:16A build about [music] survival for a
  6. 0:18Patreon subscriber. Finding the
  7. 0:20connection between a necktie, traffic
  8. 0:22tickets, Ariel Schulman,
  9. 0:24>> hi.
  10. 0:25>> And Catfish the movie [music] using AI
  11. 0:27to fight the man. See what this little
  12. 0:29bad boy businessman is all about.
  13. 0:31>> We made an archive footage crawler. It
  14. 0:33works. Hallelujah. Plus a few spices
  15. 0:36[music] to keep things interesting.
  16. 0:43>> All cars today are painted
  17. 0:47in grayscale.
  18. 0:51Look, they're all grayscale.
  19. 0:56They're between the colors of black and
  20. 0:58white.
  21. 1:00Car in front of me.
  22. 1:02Car to the left of me. The oncoming car.
  23. 1:07And here's the break as a Mercedes right
  24. 1:09here.
  25. 1:12Just one exception. Look. Grayscale.
  26. 1:15Grayscale. Gray scale. Gray scale.
  27. 1:17Grayscale. Grayscale. Gray scale.
  28. 1:21Gray scale. Grayscale.
  29. 1:23Blue. Ding ding. Grayscale. Grayscale.
  30. 1:26Grayscale. All three of our cars are
  31. 1:28grayscale. Grayscale. Grayscale.
  32. 1:30Grayscale.
  33. 1:33Gray scale. Gray scale. Gray scale.
  34. 1:34Grayscale. Blue. Red.
  35. 1:41All right. So, it was somewhere around
  36. 1:43here.
  37. 1:45I think that Morgan's dad got demolished
  38. 1:48by a car on a bike ride.
  39. 2:33The thing that I found most interesting
  40. 2:35about the story of Morgan's dad's bike
  41. 2:37crash
  42. 2:39is that he didn't tell anyone about it
  43. 2:43until he was asked.
  44. 2:46So,
  45. 2:50you really have to kind of look for this
  46. 2:52date. You have to catch the
  47. 2:55light just right or else it doesn't show
  48. 2:57up.
  49. 3:00And this will mount in their beautiful
  50. 3:03house in Alpine.
  51. 3:08And then a lot of people are going to
  52. 3:10ask them about it.
  53. 3:15We're coming up on the 20th anniversary
  54. 3:18of the Respectability Tour. [music] And
  55. 3:21for some reason, while I was laying out
  56. 3:24the modules for this episode, I got the
  57. 3:28urge to include the ironing my tie
  58. 3:31sequence from the Respectability Tour.
  59. 3:34[music]
  60. 3:34And what's special about the ironing my
  61. 3:38tie sequence [music]
  62. 3:39is who made it. My brother Casey didn't
  63. 3:43make it. I didn't [music] make it. Ariel
  64. 3:47Schulman made it.
  65. 3:49>> Hi, how are you?
  66. 3:51>> Who is Ariel Schulman? I'll tell you
  67. 3:54after we watched Ironing My Tie from the
  68. 3:58Neistat Brothers Respectability Tour
  69. 4:002006.
  70. 4:12[music]
  71. 4:14I bought this tie on the Salvation Army
  72. 4:19on Bank Street in New London,
  73. 4:21Connecticut. [music] And I like to
  74. 4:22imagine it belonged to some like old guy
  75. 4:25who is a Yaley [music] or something, but
  76. 4:26it's probably not that old.
  77. 4:29Well, I'm not considering. I think my
  78. 4:31plan [music] is for 2007, I'm going to
  79. 4:33switch from a casual jeans and
  80. 4:36t-shirtbased
  81. 4:38um wardrobe [music] to a suitbased
  82. 4:41wardrobe.
  83. 4:43Yeah. Respectability this tour. Like the
  84. 4:46people that the producers of this
  85. 4:49project, you know, they said we're going
  86. 4:50to have people do us a lot of favors,
  87. 4:52but [music] don't try pull any stunts.
  88. 4:54And I think people have the wrong idea
  89. 4:56about us that we're like bad and mean to
  90. 4:59people. We just screw people over all
  91. 5:00the time. [music] But really, we're not
  92. 5:02like that. Um,
  93. 5:04you know, and what [music] we've tried
  94. 5:05this trip is just to be really
  95. 5:08respectable to everybody. and it's been
  96. 5:09really really beneficial.
  97. 5:17Why did I want to put this in this
  98. 5:19episode? What's the connection? This
  99. 5:21episode is about using AI to beat
  100. 5:23traffic tickets. But without knowing the
  101. 5:26connection, I decided to put ironing my
  102. 5:30tie in there and figure out the
  103. 5:32connection later. And I think I know the
  104. 5:37ironing my tie connection to this
  105. 5:40episode.
  106. 5:42And the connection is three-fold. First
  107. 5:47and loosest connection, traffic
  108. 5:49violations and the DMV. Casey and Mickey
  109. 5:53go to battle with the DMV 20 years ago
  110. 5:56in the respectability tour. I go to
  111. 5:58battle with the DMV in this episode. I
  112. 6:01got a ticket. I deserved in the
  113. 6:03respectability tour 20 years ago. I get
  114. 6:06a ticket I don't deserve in this
  115. 6:09episode. The second connection between
  116. 6:13this episode and the ironing my tie
  117. 6:16sequence is decency.
  118. 6:21That's what I think the ironing my tie
  119. 6:24sequence is about. I think it's about
  120. 6:26decency. And I think it was also
  121. 6:30indecent of the park rangers back in
  122. 6:34March of this year to drive around the
  123. 6:38OV area parking lot to ticket the dads
  124. 6:43there with their kids just trying to
  125. 6:45keep them off of screens. Ticketing the
  126. 6:47dads with hundreds of dollars of
  127. 6:50violations for not having green stickers
  128. 6:54on their bikes and trucks. Those rangers
  129. 6:57could have given us warnings and
  130. 6:59achieved the same results.
  131. 7:02And the third connection is catfish.
  132. 7:06The movie catfish and why catfish was
  133. 7:10titled catfish. What's the source of the
  134. 7:14newest meaning of the word catfish? And
  135. 7:18the newest meaning is in the dang
  136. 7:20dictionary. Miriam Webster's catfish. A
  137. 7:25person who sets up a false personal file
  138. 7:29on a social networking site for
  139. 7:32fraudulent or deceptive purposes.
  140. 7:37And it was Henry Just, Nev Scholman, and
  141. 7:41his brother Ariel Schulman.
  142. 7:43>> Oh, hi. How are you? who made Catfish
  143. 7:47the movie and coined the new definition
  144. 7:50of catfish by titling the movie Catfish.
  145. 7:55But it was Vince
  146. 7:58in the movie Catfish who gave us the
  147. 8:01beautiful explanation of where the
  148. 8:04catfish archetype comes from.
  149. 8:09They used to tank uh
  150. 8:12cod from Alaska
  151. 8:14all the way to China. They'd keep them
  152. 8:16in vats in the ship.
  153. 8:20By the time the codfish reached China,
  154. 8:23the the flesh was was mush and
  155. 8:27tasteless.
  156. 8:29So this guy came up with the idea that
  157. 8:33if you put these cods in these big vats,
  158. 8:36put some catfish in with them and the
  159. 8:39catfish will keep the cod agile.
  160. 8:44And there are those people who are
  161. 8:46catfish in life and they keep you on
  162. 8:49your toes. They keep you guessing. They
  163. 8:52keep you thinking. They keep you fresh.
  164. 8:54And uh I thank God for the catfish
  165. 8:56because we'd be droll, boring, and dull
  166. 9:00if we didn't have somebody nipping at
  167. 9:01our fan.
  168. 9:05And in that respect, me using AI to
  169. 9:08fight these tickets and me using AI to
  170. 9:12make trouble at the DMV, maybe that
  171. 9:16makes me a catfish.
  172. 9:33>> [music]
  173. 9:35>> This is the current way I fight a
  174. 9:38traffic ticket. [music] Google traffic
  175. 9:40lawyer. Hire traffic lawyer. Traffic
  176. 9:43lawyer beats ticket. I pay traffic
  177. 9:45lawyer hundreds of dollars. But for
  178. 9:47these, I'm going to try a new way. New
  179. 9:50way. What's this about? A new way.
  180. 9:52>> You know what?
  181. 9:56>> Let's use Claude for this.
  182. 9:59See what this little bad boy AI is all
  183. 10:02[music] about.
  184. 10:08>> [music]
  185. 10:10>> This next bit is about using an AI model
  186. 10:13as a sort of lawyer, which involves a
  187. 10:15lot of back and forth that this episode
  188. 10:17sponsor, Whisper Flow, drastically
  189. 10:20accelerated. Whisper Flow is a
  190. 10:22voicetoext tool. Works on my computer,
  191. 10:24works on my phone, works inside all the
  192. 10:26apps I actually use. You talk, it types.
  193. 10:30I bought Whisper Flow before they
  194. 10:32sponsored me and I've been pushing
  195. 10:34Whisper Flow word of mouth since I
  196. 10:36started using it. Okay, watch this. Can
  197. 10:38you investigate whether or not the DMV
  198. 10:40will issue OV green [music] stickers for
  199. 10:42electric dirt bikes like mine? Typing
  200. 10:44that out took me 26 seconds. Speaking it
  201. 10:47took about 7 seconds. My phone's builtin
  202. 10:50talk to [music] text makes me sound
  203. 10:52janky. Whisper flow makes me sound
  204. 10:54eloquent. [music] Watch. Um, so three
  205. 10:59filmmakers I have a tendency to [music]
  206. 11:01steal from are uh one Verer Herdzog,
  207. 11:07um, two, uh, Adam Curtis and, uh, three,
  208. 11:13um, Agnes Varda. Whisper Flow strips out
  209. 11:16the ums and the H, builds the list
  210. 11:18right, and spells the hard names
  211. 11:20correctly. Herzog, Agnes Varda. Somehow
  212. 11:23this state-of-the-art [music]
  213. 11:24technology lets me live more analog. Use
  214. 11:27the code spirited man for 1 month of
  215. 11:29Whisper Flow Pro free. Links in the
  216. 11:32description. Thank you, Whisper Flow.
  217. 11:43>> Take a picture of the tickets with the
  218. 11:44phone. Drag it into Claude. Claude has
  219. 11:48the plan.
  220. 11:48>> Okay, this is what you're going to do.
  221. 11:50immediately pay for the permits
  222. 11:54>> then I'll write a letter to the judge
  223. 11:56>> and that's where it up because
  224. 11:58you can't send a letter directly to the
  225. 12:00judge but I don't know that so I just do
  226. 12:03what Claude says tells me to get the
  227. 12:05permits right now I get the one for the
  228. 12:08Tacoma no problem [music] but then I
  229. 12:11tried to get [singing] the one for the
  230. 12:13bikes I tried to get the stickers for
  231. 12:14the bikes
  232. 12:16and that's what started my war with the
  233. 12:19DMV and That's what started my war with
  234. 12:21the DN.
  235. 12:38I've got dozens of hours putting
  236. 12:40together paperwork in preparation for my
  237. 12:42DMV visit to get green passes so that I
  238. 12:45can use the OV. Despite hours and hours
  239. 12:48of research and care and consulting with
  240. 12:51multiple AI frontier models,
  241. 12:57there is no way to do it. There is no
  242. 13:00way to get offhighway vehicle green
  243. 13:03stickers for this style of bike.
  244. 13:07And I'm learning that at the Department
  245. 13:09of Motor Vehicles right now.
  246. 13:12I put dozens and dozens of hours of
  247. 13:16effort and work based on your assessment
  248. 13:20that these electric bikes could be
  249. 13:22authorized with green stickers to use on
  250. 13:24OV property when that is not true. Why
  251. 13:28wouldn't you advise me of this from the
  252. 13:30beginning and save me all the trouble
  253. 13:33and hours and hours of work that could
  254. 13:40have gone into me making money to make
  255. 13:42it irrelevant as to whether or not I was
  256. 13:44in compliance.
  257. 13:47[music]
  258. 13:48Don't you just love DMV storylines?
  259. 13:54Here's
  260. 13:55[music]
  261. 14:00the paradox. You need OV sticker to ride
  262. 14:04in the OV area. You get the OV sticker
  263. 14:07at DMV. The DMV won't give OV stickers
  264. 14:12to electric dirt bikes. However,
  265. 14:16thanks to AI, I found out that something
  266. 14:20called SB586,
  267. 14:23new law, went into effect January 2026.
  268. 14:26The DMV has to issue OV stickers to
  269. 14:31electric dirt bikes. So, the DMV is not
  270. 14:35in compliance. That sort of helps me
  271. 14:38because I can go to the judge and say,
  272. 14:41"Look, I tried to remediate this
  273. 14:43problem, but [music] the DMV is not
  274. 14:45issuing the the stickers even though
  275. 14:49this law SBA 586 [music] says that they
  276. 14:52they must that they're compelled to."
  277. 14:54So, I had AI draft up the letter to the
  278. 14:58judge, put the whole case together. It
  279. 15:00told me how to put together all of the
  280. 15:02all of the it's called a mitigation
  281. 15:04packet and it's so it's like exhibit A,
  282. 15:07exhibit B. Here's me buying the pass for
  283. 15:10the Tacoma. Here is my testimony of
  284. 15:13being denied the stickers at the DMV.
  285. 15:15The mitigation packet says, "Judge,
  286. 15:17look, I know I broke the rules, but I
  287. 15:21tried to fix them. The DMV won't let me
  288. 15:23fix them. Can you please cut me a break
  289. 15:25here?" And I did fix one of them. I got
  290. 15:27the sticker for the truck. And this is
  291. 15:29called curing the tickets. [music] AI
  292. 15:32came up with this strategy. I broke the
  293. 15:34law, but subsequently I've cured the
  294. 15:37infractions. Well, one of them [music]
  295. 15:44[music]
  296. 15:45Zoom court preparation. My entire case I
  297. 15:49put together with Claude
  298. 15:52knowing that the law and the court hates
  299. 15:55that. Let's see if I win.
  300. 15:58I'm not allowed to record the
  301. 15:59proceedings. It's illegal.
  302. 16:02>> Thank you, Court Deputy Jerry, for
  303. 16:04forwarding my mitigation packet to the
  304. 16:06prosecutor, as I found out I wasn't
  305. 16:09allowed to send it to the judge, which I
  306. 16:11had done under the advice of my counsel,
  307. 16:14Claude. [music]
  308. 16:19So, here's what happened. They offered
  309. 16:22me a deal.
  310. 16:25don't have to pay for the Tacoma not
  311. 16:27having the sticker cuz I got I got one
  312. 16:31subsequent to the violation. Then I get
  313. 16:34to pick one of the two bikes to pay the
  314. 16:37violation for. The prosecutor offered me
  315. 16:40this or take my chances with a judge at
  316. 16:44a later date. So, I took the deal. Even
  317. 16:48though the DMV makes it impossible
  318. 16:53to register these bikes, even though the
  319. 16:56California state law says that the DMV
  320. 16:59has to make it possible for us to
  321. 17:01register these bikes.
  322. 17:04My next battle is with the DMV.
  323. 17:08It's not going to work, but maybe I can
  324. 17:10light a fire under their ass.
  325. 17:23I dreamed of a footage crawler that
  326. 17:25lived on a Raspberry Pi inside my
  327. 17:27Odyssey book. Plug it in, tell it what
  328. 17:29shots I'm looking for. Motorcycle night
  329. 17:31city. Crawler searches through my
  330. 17:3326-year archive, finds me the shots, and
  331. 17:35gives me the choices. [music] 9 terab of
  332. 17:38footage. That's 9,000 gigabytes.
  333. 17:42Now, how the do you make something
  334. 17:44like that?
  335. 17:46Where do you even start? Where do you
  336. 17:49even start? If you're lucky like me,
  337. 17:52Isabelle will connect you with Jake Van
  338. 17:54Clee, who is the absolute best person on
  339. 17:58planet Earth to teach you how to build
  340. 18:01something like this using AI. So, having
  341. 18:04gone through this, I am as dumb as you
  342. 18:07can be and still be able to maybe work
  343. 18:10with this stuff. I want to try to
  344. 18:12explain [music] to a non-technical
  345. 18:14person like me. I don't really know
  346. 18:16computer stuff beyond editing. I want to
  347. 18:18explain [music] how Jake envisioned this
  348. 18:21process such that I without knowing any
  349. 18:25of the computer words like data cleaning
  350. 18:28or vector database. Jake was able to
  351. 18:31explain to me what I would in turn have
  352. 18:35to explain to an AI model in order for
  353. 18:37the AI model to make this footage
  354. 18:39crawler. I want to try to explain that
  355. 18:41process to you
  356. 18:43without boring you to death. God help
  357. 18:46me.
  358. 18:47>> But realistically, the three things that
  359. 18:48you need to know,
  360. 18:50>> I need a bunch of frames and I need
  361. 18:53descriptions of those frames attached to
  362. 18:58a video in a text file. Overall, the
  363. 19:01problem we're trying to solve is to
  364. 19:03reduce the nine terabytes of video data
  365. 19:06into data small enough for a computer to
  366. 19:09search through quickly. Video is
  367. 19:12[snorts]
  368. 19:14huge. One minute of video shot on shot
  369. 19:18on this Canon 1DX is five gigabytes. One
  370. 19:23minute, five gigabytes.
  371. 19:27I have 451
  372. 19:29hours of footage. Granted, it's not all
  373. 19:32in 4K. So Jake understands the strategy
  374. 19:36of how to reduce this amount of data and
  375. 19:40it's absolutely ingenious.
  376. 19:43>> So if you say I want this video, well it
  377. 19:47needs instead of it going through
  378. 19:49searching through every video figuring
  379. 19:50out these images, you're doing that hard
  380. 19:52work up front. You're freezing what
  381. 19:56those description are in time, taking it
  382. 19:58away from visual data and turning it
  383. 20:00into text data. And now text data much
  384. 20:04easier to search through, much simpler,
  385. 20:06much more efficient.
  386. 20:07>> Instead of making the computer search
  387. 20:09through gigantic video files, we're
  388. 20:13going to create a database of text
  389. 20:16descriptions of the video files. So the
  390. 20:20computer will be searching through tiny
  391. 20:22word data, not enormous video data. A
  392. 20:26picture is worth a thousand words. A
  393. 20:29word is worth 1 1,000th of a picture.
  394. 20:32It's actually less than a thousandth in
  395. 20:35computer data. My 9 terabytes of video
  396. 20:39is only 12 megabytes of description.
  397. 20:43750,000th
  398. 20:46the amount of data to describe videos
  399. 20:50rather than watch them. In mass, it's
  400. 20:53the difference between a paperclip and
  401. 20:56an F1 car. A paperclip weighs 750,000th
  402. 21:01the amount of Max Versstappen's Red
  403. 21:04Bull. We're going to reduce the 9
  404. 21:06trillion bytes of data the computer has
  405. 21:08to look through to 12 million bytes of
  406. 21:13data of information the computer has to
  407. 21:15look through by creating a database of
  408. 21:18text that labels each video. Computer
  409. 21:22don't got to look through the video.
  410. 21:24Computer just got to look through the
  411. 21:25words. But how do we get the words? How
  412. 21:29do we label the 48,000
  413. 21:32videos? motorcycle,
  414. 21:35city, night, ladder, hammock, snowfall
  415. 21:40times 48,000. How do we do that? We use
  416. 21:44the doomsday machine. AI, baby. An AI
  417. 21:48model is going to look at the videos and
  418. 21:50describe them. And in the description is
  419. 21:53the name of the video file, too. So, now
  420. 21:55the video is searchable by file name and
  421. 21:58descriptive words. Okay. So, we use AI
  422. 22:01to watch the videos and create a giant
  423. 22:04database of video labels or tags or
  424. 22:07descriptions, whatever you want to call
  425. 22:09them. So, are we going to make this AI
  426. 22:12watch 450 hours of video and write
  427. 22:16descriptions? How many tokens is that
  428. 22:18going to take? How much is that going to
  429. 22:20cost? Tokens is like the electricity the
  430. 22:24AI uses when you use Open AI or Claude.
  431. 22:29That's what they're charging you for,
  432. 22:31the tokens. Probably tens of thousands
  433. 22:35of dollars in tokens to have one of the
  434. 22:37Frontier models watch and label 9
  435. 22:41terabytes of video. Jake had two
  436. 22:44brilliant ideas for that.
  437. 22:46>> And you actually just pull like two or
  438. 22:49three frames from every single video.
  439. 22:51>> And that's brilliant idea number one.
  440. 22:53Yeah. Yeah. I ain't got to watch the
  441. 22:55whole video, just a few frames. So, we
  442. 22:58told this software called FFmpeg to pull
  443. 23:02the 50th frame, which is 2 seconds in,
  444. 23:06pull the frame that's 3 seconds from the
  445. 23:09end of the video, and then a frame every
  446. 23:1130 seconds in between them. Most of my
  447. 23:14clips are less than a minute long. So,
  448. 23:16FFmpeg
  449. 23:18only got to pull a few frames per video.
  450. 23:22So, that's half the solution to the
  451. 23:25token problem. We're reducing the frames
  452. 23:29with fffeg.
  453. 23:31>> Um, and then what you could end up doing
  454. 23:33is just having like a decent vision
  455. 23:35model. Go through, look at the image,
  456. 23:38give a description, throw that
  457. 23:40description onto um, uh, just a little
  458. 23:43markdown file or something simple like
  459. 23:45that.
  460. 23:46>> But now we need a model to watch and
  461. 23:49label those frames. And that's a lot of
  462. 23:52tokens. You're going to get probably
  463. 23:54like 20,000 minutes of use time, right?
  464. 23:58Billing minutes is a lot of them build
  465. 24:00by minutes. I mean, that's going to be
  466. 24:02four figures.
  467. 24:03>> Still a lot of data to sort through. Got
  468. 24:05to look through all of them frames and
  469. 24:07describe them. About two or three days
  470. 24:10of compute, letting the computer just
  471. 24:12run and chew and chew. Two or three
  472. 24:15days. That's a lot of tokens.
  473. 24:17>> Um, so the local tool process is
  474. 24:19definitely what I would recommend.
  475. 24:21Sovereignty is part of the narrative.
  476. 24:23>> Yeah. Okay. Then yeah, you're going to
  477. 24:25want to do it local.
  478. 24:26>> So the second brilliant idea to save
  479. 24:29tokens that Jake gave us, a local model.
  480. 24:33Download a local model. It's just
  481. 24:35software. It was called Moon Dream. A
  482. 24:38local model is a model that lives on my
  483. 24:41computer. Download it for free. My
  484. 24:44computer does all the work. No tokens.
  485. 24:48So, I asked Claude, "Findy me a free
  486. 24:50local vision model." Claude had me
  487. 24:53download
  488. 24:54to host the model and Moonream, the
  489. 24:57vision model itself that identified and
  490. 25:00labeled the video files. Don't use the
  491. 25:03Frontier models data centers.
  492. 25:06Took two days of constant compute and I
  493. 25:09used Isabelle's Mac Mini for it. And
  494. 25:11then I was basically done and Claude
  495. 25:15walked me through the REST OF IT.
  496. 25:17>> [screaming]
  497. 25:18>> OKAY. SO, I'M HAVING THIS feeling and I
  498. 25:21don't know if you remember this feeling
  499. 25:22and maybe you get this feeling doing
  500. 25:24something different, but it's the
  501. 25:26feeling
  502. 25:27when
  503. 25:29you're learning how to ride a bike and
  504. 25:32who's ever teaching you pushes you and
  505. 25:35lets go of the seat and you're doing it.
  506. 25:38You're like, "Oh my god, I'm I'm riding
  507. 25:39a bike. I'm riding a bike." And then you
  508. 25:41crash. [laughter]
  509. 25:42But it's that feeling. Okay? And that's
  510. 25:45the feeling I just got. I did one test
  511. 25:48of the footage crawler and I have to
  512. 25:51tell you how it like what happened. So
  513. 25:53it had me test dog. It said type in dog
  514. 25:57and run it and then test one of the
  515. 25:59clips. So this is what happened. See
  516. 26:02where the arrow is? I typed in dog there
  517. 26:04and then I pushed play. I pushed play.
  518. 26:08So I saw this, right? And I said, "Ah,
  519. 26:11there's no dog.
  520. 26:13There's no dog in this frame. It saw
  521. 26:17Lua's dress
  522. 26:20and it kind of looks like a dog and I'm
  523. 26:23like, "Ah, I'll just change the tags and
  524. 26:25I'll like figure out how to correct
  525. 26:27that." And it's a time lapse. And so I
  526. 26:29kept watching
  527. 26:33and look, I mean, it it went by so fast
  528. 26:36you won't even be able to see it right
  529. 26:39here for like How many frames is that?
  530. 26:42like five frames. Unbelievable. So, like
  531. 26:46when I clicked that, I got that little
  532. 26:48taste, like a little eyropper of that
  533. 26:51letting go of the bike. I got it. The
  534. 26:54best I've ever gotten it as an adult was
  535. 26:56when I when my son, it was so weird.
  536. 26:58Mirror neurons. I let my son go and he
  537. 27:01got it for a second and it was just
  538. 27:02like, it was that exact sensation. It
  539. 27:05was so crazy. It was like asset or
  540. 27:08something.
  541. 27:10baseball field. Boy riding a bike.
  542. 27:14[music]
  543. 27:20Wow, this is better than last time.
  544. 27:24You're getting it.
  545. 27:27You're supposed to wipe out, so don't
  546. 27:28worry about that. [music]
  547. 27:30And you crashed very well.
  548. 27:35[music]
  549. 27:47underlined years ago, chosen at random
  550. 27:50today.
  551. 27:56Art doesn't get made on the clock, but
  552. 27:59it can get finished on the clock.
  553. 28:06Join a paid tier of the Patreon to get
  554. 28:08in touch with me or to avail yourself of
  555. 28:10this channel's alternative universe. You
  556. 28:13can click this. It'll take you right
  557. 28:15there.

About this transcript

This page contains the full transcript of Tickets, A Necktie & 26 Years of Footage: Long-Form 3 by Van Neistat, generated from the public captions YouTube serves with the video. The transcript has 3,585 words across 557 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.