Kempf & Kunhya on Lex Fridman: How FFmpeg runs 90% of video

Jean-Baptiste Kempf 0:00 The important is, is your code good? We care about excellent code. We don't care who you are. Like maybe you're a dog. I don't care, right? I don't care where you come from. I need to look at your code. Oh, yeah, but I'm engineer at this very large company in, in, in Italy, in Germany, in the US. We don't care. We care about the quality of your code because this is what defines our community and which means that we have a lot of people who contribute who are some very different backgrounds and, and, and very introverts. Sure. But that's okay, right?

Lex Fridman 0:31 FFmpeg is probably one of the biggest CPU users in the world. Everything we've just said in the past couple of minutes, every sentence is someone's lifetime's work. There are books about every sentence. So the level of complexity in many cases is inordinate.

Jean-Baptiste Kempf 0:45 FFmpeg has one hundred thousand lines of assembly for all the codecs.

Lex Fridman 0:50 For all codecs. Mm-hmm.

Jean-Baptiste Kempf 0:50 And just this one has two hundred and forty thousand. Every cycle matters. We are talking about probably three billion devices which are going to decode video nonstop because, for example, thirty percent of the video from Netflix are now in AV1, fifty percent of YouTube.

Lex Fridman 1:07 This is what peak video codecs should look like. Seventy-nine point nine percent assembly, nineteen point six percent C, and, uh, zero point five percent other.

Jean-Baptiste Kempf 1:18 And what's incredible is with those tweets, which is factual, people get crazy.

Lex Fridman 1:25 For the last two years, they go crazy. No, intrinsics is fine. The compiler-

Jean-Baptiste Kempf 1:28 You can optimize your compiler. Auto-vectorization is your fault. You don't understand. And we've tried that forever, right?

Lex Fridman 1:35 For two years and two years later, showing hundreds of examples of, of handwritten assembly. No, no, no, you're doing it wrong. The compiler can do this. The intelligence agencies tried to like say, "Can you put a backdoor in VLC?"

Jean-Baptiste Kempf 1:48 Yes. Two of them.

Lex Fridman 1:50 Well, what do you say?

Jean-Baptiste Kempf 1:51 No.

Lex Fridman 1:52 [laughs]

Jean-Baptiste Kempf 1:52 Well, I was a lot less polite.

Lex Fridman 1:54 Basically saying, "Hell no."

Jean-Baptiste Kempf 1:56 Like if we had to compromise our software, we would shut it down. This is clear.

Lex Fridman 2:00 Any tweets, uh, Kieran, you regret?

Speaker 2:03 Tweets I regret?

Lex Fridman 2:05 Or is it like that, how does the French song go? Regret nothing.

Jean-Baptiste Kempf 2:08 Don't regret anything. No, it's because regrets are attacks on your mind.

Lex Fridman 2:17 The following is a conversation all about FFmpeg and VLC with Jean-Baptiste Kempf and Kieran Kunhya. FFmpeg is an open source software system that is the invisible backbone behind YouTube, Netflix, Chrome, VLC, Discord, and basically every platform that touches video or audio on the internet. It can decode, encode, transcode, stream, and play almost any video or audio format ever created. To me, it is one of the most incredible software systems ever developed, and it's all done by volunteers. VLC is also a legendary piece of software. It is an open source media player that plays basically anything you throw at it, any format, any platform, no ads, no tracking. It has been downloaded over six billion times, and again, for me, it has been one of my favorite pieces of software ever with the most legendary logo, which I of course had to honor in this conversation by wearing the VLC traffic cone hat the whole time. So again, above all else, thank you to the incredible volunteer engineers who put their heart and soul into this code that has been used and loved by billions of people. Thank you. And about the two great engineers and human beings I'm, uh, talking to in this episode, Jean-Baptiste is the president of VideoLAN and is a key figure behind VLC and FFmpeg. Kieran is a longtime codec engineer, FFmpeg contributor, and the man behind the now infamous FFmpeg account on Twitter/X that I recommend everybody follow for the memes and for the unapologetic celebration of open source and great low-level software engineering. Let me also say that it's inspiring and humbling that so much of modern civilization rests on software built by people who are not chasing fame or money, but are obsessed with the craft of engineering. We live in a world where billions of people consume video every day without ever thinking about the invisible machinery underneath it. But that machinery matters. Open source infrastructure matters. It is one of the great examples of human beings quietly collaborating across borders to build something useful, durable, and elegant for the rest of us. And so this conversation is not just about codecs and media pipelines. It is also about the deeper spirit of engineering and generosity that makes projects like FFmpeg possible. Again, I can never say it enough. Thank you. This is the Lex Fridman Podcast. To support it, please check out our sponsors in the description where you can also find links to contact me, ask questions, give feedback, and so on. And now, dear friends, here's Jean-Baptiste Kempf and Kieran Kunhya.

Lex Fridman 5:35 So the legend goes VLC can open everything. What's the weirdest thing?That you know that it can open

Jean-Baptiste Kempf 5:42 You know, there is a ton of people who are using VLC to record VHS videos, right? Like, it's just like you plug it with a capture card and you can basically record VHS video.

Lex Fridman 5:52 Well, how does that work?

Jean-Baptiste Kempf 5:53 Basically, it's, you know, those type of capture card where you can put a Peritel in or, or RGA, and you put that, and actually VLC can play those type of cards, and there is a module which allows to control directly some, those VCR camcorders. We support DVD audios lately, right? We spent the, the summer working on DVD audio support, and like there is no, no one's making any DVD audio support. There is a custom encryption schemes.

Lex Fridman 6:19 What about Lucasfilm?

Jean-Baptiste Kempf 6:20 Oh, yeah, and there is of course all the weird codecs support, game codecs supported by, by FFmpeg.

Kieran Kunhya 6:25 The one Star Wars video game, the first ten-second opening sequence, someone has gone and implemented that and made sure that's bit exact on one disc that existed at one time of one little sequence in the game.

Jean-Baptiste Kempf 6:36 A- and then funnily, there, there was a, at one VideoLAN conference, we made a competition to me- make the weirdest and most horrible file ever-

Lex Fridman 6:44 Mm-hmm

Jean-Baptiste Kempf 6:44 ... and see if VLC could play it.

Lex Fridman 6:46 What did it end up being? What's the file?

Jean-Baptiste Kempf 6:48 It was an MKV file made by Derek-

Lex Fridman 6:52 Yeah

Jean-Baptiste Kempf 6:52 ... which each of the frame was changing resolution, aspect ratio [laughs]

Lex Fridman 6:57 [laughs]

Jean-Baptiste Kempf 6:57 ... rotation, and, and, and it was like-

Lex Fridman 6:59 Did it work?

Jean-Baptiste Kempf 7:00 Yes. And there was another one where the whole video was actually animated subtitles, right? SSA, right? So-

Lex Fridman 7:08 Yeah. I remember that, yeah

Jean-Baptiste Kempf 7:08 ... each, each, this one was-

Lex Fridman 7:10 [laughs]

Jean-Baptiste Kempf 7:10 And so each, each frame was a black frame, but on top of that, there was a, a, a subtitle that was animated for each frame.

Kieran Kunhya 7:16 There was a file that's a valid zip and a valid MP3 at the same time or something like that, so [laughs]

Jean-Baptiste Kempf 7:20 So yeah, we'd made a competition of stupid files.

Lex Fridman 7:22 [laughs] And it worked. It opened all of the stupid files.

Jean-Baptiste Kempf 7:27 Yes.

Lex Fridman 7:27 By the way, uh, for people who are not familiar, I am wearing a hat. Would it be fair to say this is the best worst logo of all time, the cone?

Jean-Baptiste Kempf 7:36 Yeah, by far, right? The logo of VLC is so iconic, right? Like we are a team with a small number of people, and, and the icon is known everywhere. I go to middle of nowhere in India or in China, people know the cone, right? And, and twenty-five percent of the website traffic that comes to, uh, our main website is cone player, right? So, so many people don't know VLC, right? They know the cone player.

Lex Fridman 7:59 That's the thing that Google for is cone player.

Jean-Baptiste Kempf 8:01 Yeah. They, they go on Google and they, they put cone player and, and they download VLC, right? So that's iconic. A- and once we tried to change it as a joke, right? We said it was going to be a, a type of, um, uh, caterpillar construction, and, and we said that during April first. Uh, and-

Lex Fridman 8:17 Mm-hmm

Jean-Baptiste Kempf 8:17 ... we had around ten thousand emails saying, "No, don't change the logo," and so on, right? So it's so iconic, right? It's so distinctive, right? If you want to do a video player, you're going to put a play button on a TV, right? And that's a YouTube, YouTube logo, right? It's unoriginal. This one is orange, right?

Lex Fridman 8:33 Yeah.

Jean-Baptiste Kempf 8:33 It's very bright. Uh, and, and it's weird.

Lex Fridman 8:37 And it's ridiculous, and it's absurd, and it's hilarious. It becomes meme, and meme becomes culture.

Jean-Baptiste Kempf 8:41 And you keep it, and you know about it, and you know that in twenty years, like you still have, going to have the cones and remember, oh yeah, that was a video player.

Lex Fridman 8:49 Yeah. And we'll talk about, you know, the, the mission of FFmpeg being a kinda the archival aspect of it. So you can think about a thousand years from now we'll have all these videos that only VLC can open. Humans, human civilization has already destroyed itself multiple times, and the only thing that will remain is this like, you know, the, the cockroaches will be crawling around and it'll be the VLC logo-

Jean-Baptiste Kempf 9:12 [chuckles]

Lex Fridman 9:12 ... with our, some of the archival footage that VLC can open. And the aliens will show up and they'll p-press play and they'll get to see it all.

Jean-Baptiste Kempf 9:18 Well, really, really hope so, right? But there is also so many memes where people say, "Well, I'm sure I can put a pancake inside my DVD drive and VLC will play it." And they-

Lex Fridman 9:25 Can they?

Jean-Baptiste Kempf 9:26 No, we tried. It doesn't. Um-

Lex Fridman 9:27 Doesn't

Jean-Baptiste Kempf 9:28 ... but we actually have a video of us trying that. Uh, it didn't work.

Lex Fridman 9:31 A codec for physical reality, I don't know what that would even look like.

Jean-Baptiste Kempf 9:34 There was a guy who did that, right? He, he, he, he printed a small cone, right? Like the, the ones we, we distribute as goodies, and inside he put an RFID chip, um, which was his way of playing a movie, right? And so he-

Lex Fridman 9:46 Yeah

Jean-Baptiste Kempf 9:46 ... put this on a RFID player, and when he put that it was playing like the last Star Wars and so on. So instead of having like DVD boxes, he had like VLC cones all around and he plugged that a- and that was like physical objects.

Lex Fridman 9:59 So the thing that we're talking about is everything around video codecs, video encoding, video decoding, video streaming, video player client that I'm wearing on my head, the entire ecosystem enabling free media. Uh, we'll talk about FFmpeg, we'll talk about VideoLAN, VLC, and all the other incredible video technology, uh, that is used probably by billions of people. So JB, you're the lead developer behind the legendary VLC player. Kieran, amongst many other things, you're lead developer behind the legendary FFmpeg handle on Twitter. And, uh, both of you have spicy opinions, I would say. So today we wanna talk about FFmpeg and VLC. Uh, for context for people who are not aware, and I'm sure basically everybody listening to this have used these two technologies probably regularly without knowing it. So FFmpeg underlies basically most video on the internet, including YouTube, Netflix, Chrome, Firefox, of course VLC, and countless other video platforms. It is estimated that over ninety percent of video processing workflows online and offline involve FFmpeg. VLC has been downloaded at least six point five billion times. But likely that number, 'cause it's impossible to really count the number, uh, is much higher than that. Virtually any operating system supports virtually any media format, the limitation being it can't open pancakes. So-

Jean-Baptiste Kempf 11:39 [chuckles]

Lex Fridman 11:39 Uh, can we just lay out some of the basics to, to help people understand what's involved in all of this? So when we press playOn a video player like VLC, what happens? What-- How does it go from the, the file or the stream to the pixels on the screen and the sound on the speaker? What are the big stages to be aware of?

Jean-Baptiste Kempf 12:01 So there are several stages, right? The first stage is to get from an address, right, which is the ca- type of URL, to give you a byte of streams, right? So this would be, for example, HTTP, file, DVD, right? You give the path to the media, and it gives you a, a stream of data.

Kieran Kunhya 12:19 The stream needs to be cut up by what's known as the container, the demultiplexer or demux. Um, we'll try and keep the jargon light throughout this, but, um, it needs to go and start demarcating video and audio frames. So it just gets data from the operating system blocks at a time and needs to start cutting these frames up into compressed data. It then needs to start doing simple parsing of the video frames-

Lex Fridman 12:40 Mm-hmm

Kieran Kunhya 12:40 ... mainly to figure out whether that codec is GPU decodable or needs to fall back to software. We're very sort of used to assuming the GPU will play all of these things. There'll be hardware acceleration. I think it's up to forty-five percent of files are not GPU decodable. So these need to be probed. They need to be detected. There can be variants of a given codec, some of which are decodable on the GPU. Different vendors of GPU might have different capabilities, so those need to be detected. So if, if it's GPU capable, you pass it through to the GPU black box. So now if there's a software fallback, that means in the beginning is, is to first do deentropy coding, so removing the mathematical coding of the bit stream. So this uses capabilities such as Huffman coding or arithmetic coding to actually decompress the mathematical layer of the bit stream. We then need to start reading the syntax elements for intra prediction. So intra prediction are like still images of the video, so your iframes.

Lex Fridman 13:39 Mm-hmm.

Kieran Kunhya 13:40 So this works and operates in the spatial domain. So you do your intra prediction spatial domain. You, you have a residual because your prediction isn't quite matching that of reality. So you've made a prediction, but then there's a little bit left, and that's what's known as the residual. This is stored in the frequency domain, and these are quantized to decompact their space. We then need to do the inverse transform to bring them back to the spatial domain and apply these residuals.

Lex Fridman 14:05 So a lot of the process of the decoding is this thing is compressed.

Kieran Kunhya 14:09 Yes.

Jean-Baptiste Kempf 14:10 Yes.

Lex Fridman 14:10 And you have to predict the highest quality thing that's supposed to go there. Iframe-

Kieran Kunhya 14:16 Yeah

Lex Fridman 14:16 ... is the best representation you have spatially.

Kieran Kunhya 14:18 Yes.

Jean-Baptiste Kempf 14:19 Yes.

Lex Fridman 14:19 And then you, uh, and then there's a lot of temporal compression that can happen depending on the codec, and then you're predicting. You're predicting what the reality that was captured in this rawest form.

Jean-Baptiste Kempf 14:29 Yeah, because what, what people don't realize is that the compression on video and audio is one hundred times, right? Like, people don't realize how compressed we, we do, right? For audio, you move, you compress by when you go from normal audio to MP3, you compress by ten times, right? When, when you move to video, you need one hundred times, two hundred times, right? So you need to remove all the details, but that you don't care about. Because all the compressions that we do, and that's very important, people forget about that, is to be viewed by humans, right?

Lex Fridman 15:02 Mm-hmm.

Jean-Baptiste Kempf 15:02 So all the codecs, either for audio, mimic basically how your ear works, right? And, and a lot of things about, like, the, the, the response on the ear and same for, for your eyes, right? And, and so, for example, on video, we don't work on RGB, right? Everyone expect to work in RGB. We don't, right? We move to YUV, which is basically one is luminance, brightness, and the other are colors. And this matches your eyes, where inside your eyes you have the cones and the buttons, right? With some of them look on brightness and more on, on the other on colors, right? So we need to compress a lot, and so we need to degrade. But in order to degrade, we need to match the human perception, and this is why it's so difficult. And then we need to use the maximum power, mathematical power, uh, very complex technologies. Uh, we move to the frequency domain, as Kieran said. We do a ton of dequantizing and, and in order to get the best compression, but it still looks good.

Lex Fridman 15:58 You're trying to compress in order to maximize the highest quality thing for human perception.

Jean-Baptiste Kempf 16:05 That is correct, and that is correct. And this is very important, right? Compression is not like a zip, right? A zip, you have data in, you get data out, right? And you try with all the, the zip compression to arrive with, uh, uh, the limit. Here we are degrading the signal, right? And so we need to degrade both the audio and the video signal in the best way possible. A- and we can do that, but it involves, first, a, a lot of theoretical knowledge about how it works, the eye works, but it, a lot of mathematical, uh, change, a lot of mathematical tricks, right? For example, when you move to RGB and you do go to YUV, for example, what we do very often is that we scale down the resolution of the color compared to the brightness. And most of the time-- And just this, without compression, it divides the size by two. But most people don't see it, right? Um, and so on and so on, right? Uh, and then you go to very complex mathematical, uh, change. So of course, uh, Fourier's transform, which de facto are not Fourier transform. They are like, um, discrete cosinus transform, but that's the same idea. So frequency domain, um, we split the video by blocks, right? So that's why when it's wrongly decoded, you see those blocks and badly encoded, you see those blocks, a- and so on, to arrive to compression states that are insanely high, right? And each generation of the codec is like thirty percent less-

Lex Fridman 17:28 Mm-hmm

Jean-Baptiste Kempf 17:29 ... for the same quality, right? And this requires amount of power, um, of computational power that are huge.

Kieran Kunhya 17:35 No, maybe you should, you should elaborate. It's thirty percent better, but an order of magnitude, perhaps, perhaps even two orders of magnitude more compression power. That, that's the big difference.

Lex Fridman 17:45 What do you mean by compression power?

Kieran Kunhya 17:47 Sorry, CPU power to achieve that level of compression.

Lex Fridman 17:49 Oh, yeah. So and you have to be able to leverage the CPU and sometimes GPU, like you mentioned. And then we should mention that a lot of this programmingUh, is done at the lowest possible stack-

Jean-Baptiste Kempf 18:01 Yeah

Lex Fridman 18:01 ... whether it's C and of course, as, as the legendary-

Jean-Baptiste Kempf 18:05 Yeah

Lex Fridman 18:05 ... Twitter handle, um, re-emphasizes over and over, a lot of assembly.

Jean-Baptiste Kempf 18:10 So what happens is globally is that you have an address, right? Which gives you, uh, with the operating system, a stream of bytes, so a stream of data, right? And this is the first step. And the second step arise with demuxing, where you're going to separate audio, video, subtitle in type of different tracks. And then on each of those tracks, you're going to decompress them, decode them, either audio with an audio codec, video to video codec, and subtitle to subtitle codec. Um, and once you've decompressed those type of things, you have raw images, raw, and then you're going to talk with your, uh, graphic card in your screen and display that. And same for the audio, you're going to talk to your audio card, which then is going to go, um, in analog to, to your audio speakers.

Kieran Kunhya 18:50 And everything we've just said in the past couple of minutes, every sentence is someone's lifetime's work. There are books about-

Lex Fridman 18:56 Mm-hmm

Kieran Kunhya 18:56 ... every sentence. So the level of complexity, in many cases, i- is inordinate. You know, it's, it's, it's... Every sentence has thousands of people working on this-

Lex Fridman 19:06 Mm-hmm

Kieran Kunhya 19:06 ... in, in the industry as a whole, books written about it. So there's a lot of detail, there's a lot of subtleties, there's a lot of both academic and practical realities, um, both of which matter.

Lex Fridman 19:20 Uh, we mentioned codecs, but I don't think you mentioned, uh, containers. So what, what's the actual containers for some of the stuff we're talking about? So people are familiar with MP4, uh, MOV, MKV. So anyway, what, what are containers versus, uh, the thing that goes inside?

Jean-Baptiste Kempf 19:41 So the container is what we call also the muxer, right? When I say demuxing, it means decontainerizing, right? So a- actually, if you look, mux, mul- multiplexer, and demultiplexer, right? Mux and demux are those. And same, a codec is actually coder-decoder, right?

Lex Fridman 19:58 Mm-hmm.

Jean-Baptiste Kempf 19:59 Um, and, um, so containers are this collection of multiple tracks, right? So it's a, what normal people call the file format, but it's a bit more, um, um, subtle than that. But the most known one, of course, is MP4, but, uh, when I started, it was AVI, right? AVI was the, the video format from-

Lex Fridman 20:17 Mm-hmm

Jean-Baptiste Kempf 20:17 ... from, uh, Microsoft, and, uh, MOV, M-O-V, which became MP4, was a format from Apple. Um, in the open source community, uh, one of the person that is still active on VideoLAN is called Steve Lhomme and sta- started the, this Matroska format, which is, like, a bit more complex and, and, and more future, uh, proof. Um, and, um, there are so many others.

Lex Fridman 20:38 So, I mean, there's a, it's a pretty common thing, and m- maybe it'll even happen in this conversation, that people confuse container and the codec, right? So confuse MP4 and H.264, for example. Is that a horrible violation?

Jean-Baptiste Kempf 20:51 No, it's not, because technically, the name of H.264 is MPEG-4 Part 10. Because MPEG-4 is actually a meta specification which has several things in it, right? There is the Part 2. Uh, so there is, like, audio codecs, right? AAC de facto is MP4 audio-

Lex Fridman 21:11 Mm-hmm

Jean-Baptiste Kempf 21:11 ... something. There is, uh, actually several video codecs, right, inside the MPEG-4 specification. One of them is MPEG-4 Part 10, called also AVC, called also H.264.

Lex Fridman 21:23 Mm-hmm.

Jean-Baptiste Kempf 21:23 Right? So it's completely the fault of the industry to, to, to, to make things difficult to understand. So that's very difficult so that people then don't understand why sometimes you talk about MPEG-4 Part 10, where you mean H.264, and why it's not MP4.

Lex Fridman 21:38 So you can technically shove in all kinds of different codecs inside containers and horribly so.

Kieran Kunhya 21:44 But broadly speaking, though, MP4 is understood to generally be H.264 plus AAC audio. 99% of the time, that's that, and that, the, the rest are de minimis, the small effects, you know, edge effects really compared to that. So it's not the end of the world. There, there are people who do get annoyed by that. But also in reality, something like VLC, just to point out, the file may say .MP4, but it may be something completely different, and that's one of the challenges both FFmpeg and VLC have is the real world is a completely different place to a three-letter file format.

Jean-Baptiste Kempf 22:17 And this is very important to say, right? Like, for example, in VLC and in FFmpeg, we discard the file format, right? We, we look into the file to understand what's in it, because so many people, like, they say, "Oh, it's a video, it should must be MP4," but technically it's an MOV or maybe it's a MKV, right? So we analyze in real time everything that we have, and we don't trust, uh-

Lex Fridman 22:42 Mm-hmm

Jean-Baptiste Kempf 22:42 ... the, the, the format.

Lex Fridman 22:42 So what information does the fact that it's .MP4 give you?

Jean-Baptiste Kempf 22:45 It helps, right? It gives you a hint, right? Just like, oh, it, it's finished by .MP4. I will start first by opening, probing it with the MP4 container demuxer to see, well, it should be that. But I don't trust it, and if I'm lost, I say, "Okay, maybe I'm going to try it." So it bumps the priority of the module.

Lex Fridman 23:06 So how do you get to, uh... Just to take a bit of a tangent there. You know, the dumb thing is if you try the MP4, but it turns out it's a different codec than you would have expected, uh, most players just break there.

Jean-Baptiste Kempf 23:21 Yes.

Kieran Kunhya 23:22 Yes.

Lex Fridman 23:22 And so how do you not break? There's just, uh, philosophically, I'm sure there's a bunch of stumbling blocks along the way where you, y- it's easier to just break and stop, freak out. That's it. How does VLC not?

Jean-Baptiste Kempf 23:34 This is why VLC is popular. Um, but the reason is because actually VLC was, is just a client of a streaming solution called VideoLAN from, from, from very long time ago, from the late '90s. And when you're playing video, which are on UDP, right, in network, they might be damaged, right? So you don't trust your inputs, and this is very important into the security, is that you don't trust your inputs. SoEverything in VLC is prepared to, um, work with broken files.

Lex Fridman 24:05 Mm-hmm.

Jean-Baptiste Kempf 24:06 And it's a philosophical idea from the beginning, and everything is engineered into that. A-and it's a culture, right? And so for example... And VLC became very popular on that because a long time ago when people were, uh, pirating content, um, which they do a lot less, uh, today, um-

Lex Fridman 24:23 And none of us ever have done so.

Jean-Baptiste Kempf 24:24 No, of course not.

Lex Fridman 24:25 Yeah.

Jean-Baptiste Kempf 24:25 Um, the metadata to play some files like AVI is at, at the end of the file, right? And when you're downloading, you don't have that, right? So VLC was just like, "Hey, this file is broken, but I'm still going to try to interpret it," and this was very useful.

Lex Fridman 24:41 We hinted at the awesomeness of the various different stages. We hinted at the awesomeness of codecs, the depth and the richness and the complexity of everything involved there. What... Let's [chuckles] try to define what is a video codec? What, what's involved there? What, what does it mean to compress something? You already started to hint at it-

Jean-Baptiste Kempf 24:59 Sure

Lex Fridman 24:59 ... but can, can we elaborate a little bit more?

Kieran Kunhya 25:01 So there's a huge amount of redundancy in any video, uh, both spatial and temporal, and the point of any video codec is to remove this redundant data, use mathematical properties as part of this reduction process. So more often than not, using several orders of magnitude more compute to compress because that's more costly versus both costly both financially and in CPU resources-

Lex Fridman 25:23 Mm-hmm

Kieran Kunhya 25:23 ... versus the decompression. So it's asymmetric in that respect. Uh, often the case because compression is done once, but there could be lots of viewers of another file.

Lex Fridman 25:32 Mm-hmm.

Kieran Kunhya 25:33 So to take that information and compress it by 100X, 200X, removing redundant information and using mathematical properties to make that small, but also have properties such as error resilience. So as, as JB suggested, VLC in the beginning was, was used to play UDP network feeds, and UDP network feeds lose packets. And so some of the design goals of a codec is also to be recoverable.

Lex Fridman 25:57 Mm-hmm.

Kieran Kunhya 25:57 You, you need to actually be able to join a stream. It's not necessarily a file. You need to join, get on the decoding process, and start decoding.

Jean-Baptiste Kempf 26:06 A-a-and, and to give a more image to, to, to, to people who are not familiar, right? Like, when you're going to see any type of movie, right? You're going to see the camera is going to pan, right, and, and, and travel, and you realize that, for example, all the background is the same from, for like a minute, right?

Lex Fridman 26:23 Mm-hmm.

Jean-Baptiste Kempf 26:23 Or 30 seconds, right? So you can reuse the cloud that you see, uh, uh, on the background, you can reuse that from a frame to another, right?

Lex Fridman 26:30 Mm-hmm.

Jean-Baptiste Kempf 26:31 And so it's gets the more, the more memory you have, the more power, the more comparisons you can make, right? And so the more compressed you can be. And most of the modern codecs are basically doing that.

Lex Fridman 26:44 So just to make it even more explicit, so what is video? Video is a bunch of pixels off an RGB. You have three values, and you have a grid of pixels, and you have, let's say, 24 or 30 or 60, uh, frames a second, and you just have all these pixels repeating and showing different stuff-

Jean-Baptiste Kempf 27:08 Mm-hmm

Lex Fridman 27:08 ... 30 times a second. And so the question, the philosophical, the technical question is, how can I compress all of that, store all of that at 100X?

Kieran Kunhya 27:20 Yep.

Jean-Baptiste Kempf 27:20 Or 1000X, right?

Lex Fridman 27:22 1000X.

Jean-Baptiste Kempf 27:22 The, the target is 1000X, right?

Lex Fridman 27:24 And the goal is when you say redundancy, what is redundant? Meaning stuff at best that humans wouldn't notice if it was missing.

Jean-Baptiste Kempf 27:35 So for example, you have a picture of a cloud, right? And from the next frame, they're still going to be the same cloud, so it's redundant. You could just put it once and not do it, right? Or you have a, a, a black background behind me, for example. The black is the same on the whole picture, right? So you can say, "Well, you know, in this picture, take the pixels that you have on the top left and the one on the top right. I'm not going to give the value. I'm just going to tell you it's the same at the top left." And then you can say for frame one, um, reuse something from the previous frame or the previous, previous frame, and so on and so on, right? So th- you could basically, it's unlimited, but then it's limited in terms of memory or in terms of compute power. Because, for example, if you need to compare pixels on 200 frames in the past on 4K resolutions, it's a huge amount of compute.

Lex Fridman 28:26 And then when you're showing it, you have to do the decompress of all of that. So you... Is it the codec, the, has the encoding and the decoding is a, is a coupled process that you're developing?

Jean-Baptiste Kempf 28:37 Yes. Exactly, right. And those are two different, um, uh, trade-offs, right? Are you going to compress more? Uh, but then it might be more difficult to, to, to decode. Um, are you going to comp- to make it a codec that is more complex to encode and easier to decode? Are you going to make a codec that is easier to encode because you need to be fast, but then the, the client side, the, the player is going to spend more time? That's why you have so many different type of codecs, is that it's not always easy. A-a-and to make it even more complex, modern codecs like AV1, AV2, or VVC are actually not codecs. They are a collection of tools, right? There are multiple tools, multiple codecs in the same codec to, depending on the image, get the more compression.

Kieran Kunhya 29:21 So just to elaborate, codecs like AV1, VVC have a much wide, have a wide audience. It could be a screen share content, it could be video, it could be animation. All of these require different coding tools. So what happens these days is a collection of tools are put in and called AV1 and called AV2, called VVC to allow for different use cases. So you may be on Zoom and sharing your PowerPoint, and then you need to show the audience a video. That codec needs to start changing its tool set depending on the content to compress in a different way.

Lex Fridman 29:59 And like you said, there's a, a bunch of incredible engineers behind each part of that, each part of the tools that make up AV1, for example.

Kieran Kunhya 30:05 Sure.

Lex Fridman 30:06 Uh, so

Lex Fridman 30:07 we've, uh, kind of danced around it. We talked about VLC, the logo, the hat. Uh, let's talk about FFmpeg. What, what is FFmpeg exactly?

Jean-Baptiste Kempf 30:17 FFmpeg is basically the low-level libraries for codec, so compressions and decompression, muxers and demuxers, and filters. It's-- The core is this, and then you have a several tool which allow you to create a type of pipeline to process any type of video files. Uh, and it's used as a library absolutely inside everything from VLC to Chrome to your smart TVs to basically any video that you see online, um, you usually use FFmpeg. And FFmpeg in it has all those type of tools and sometimes depend on other libraries like x264, uh, Libvpx, and others, right? So it's really now the de facto, um, tool to process images.

Kieran Kunhya 31:06 From a philosophical level, I think it's incredible that your home videos, your, your grandmother's home videos and trillion-dollar corporations effectively are on a level playing field using the same technology stack.

Jean-Baptiste Kempf 31:19 Mm-hmm.

Kieran Kunhya 31:19 Uh, it's, it wouldn't be a surprise, you know, these big companies just have 3,000-line FFmpeg commands.

Jean-Baptiste Kempf 31:27 Mm-hmm.

Kieran Kunhya 31:27 There are some that use the API, but there are some that just have long command lines.

Lex Fridman 31:31 So yeah, there's a bunch of tools, like literally command line tool, FFmpeg, of course, FFprobe. There's libraries, Libav codec, Libav format, Libav filter. But the FFmpeg on the command line-

Jean-Baptiste Kempf 31:47 Yeah

Lex Fridman 31:47 ... is, like, legendary because you can cut-- Like there's so many parameters. You can customize everything to hell.

Jean-Baptiste Kempf 31:53 It's a language. It's an actual language.

Lex Fridman 31:55 It's an actual om- yeah, you could think of it as a programming language.

Jean-Baptiste Kempf 31:57 Yeah, of course, I'm sure. Because so most of the people, they're going to take FFmpeg file in, file out, and specify the format, right? But you can-- We've seen thousands of characters, and we've seen also, like, people, like, doing, um, programming, um, uh, generation of command lines to, to, to make FFmpeg. There is a ton of people who are using AI to generate command lines for FFmpeg because you have no idea what it is. But you can do- specify so many filters, right, uh, on command line, right? So, um, FFmpeg is this collection of toolbox for multimedia processing that everyone, everyone uses, and everyone that is watching your videos are also using, right? You, you're on YouTube. Well, it's FFmpeg on, on the client side. Well, the, your server side, on the server side. The client side is probably Chrome. Well, you're using FFmpeg also.

Lex Fridman 32:47 Mm-hmm.

Jean-Baptiste Kempf 32:47 And, and you're using OBS to record. Well, it's FFmpeg, right? You're using a ton of important, like, big box, professional boxes. Well, it's very possible that inside some part of FFmpeg is running.

Lex Fridman 32:57 I mean, there, there's like so many just to give people an idea. Like I, I use FFmpeg a lot on, on everything, just trivial stuff like, uh, take a video, add an intro video and an outro video, and fade one into the other, like, uh, what is it called? Dip to black, like where it, it dips and then shows the next video and does the same thing with audio. Uh, there's like a cross dissolve of the audio. It's a quiet-- It quiets the audio and makes it loud again. And then there's a bunch of stuff like, uh, showing the captions on screen hard, like baking the captions in. You can customize the font. You can do all kinds of layering of audio and video. There's a million things and, of course, all of that works, like, magically with basically any codec. Like anything you, you can shove in on the audio and the video side, it works.

Jean-Baptiste Kempf 33:53 But it's, it's like if you, if you look at, for example, you can do things that you would do with Adobe After Effects-

Lex Fridman 33:59 Mm-hmm

Jean-Baptiste Kempf 33:59 ... in command line on FFmpeg, right? It's, and it's very interesting because, for example, for, for, for imagings, there is not such tool. There is a few tools, but not with the breadth of FFmpeg.

Lex Fridman 34:10 So ImageMagick has a similar kind of-

Jean-Baptiste Kempf 34:13 Yes, but you will not-

Lex Fridman 34:14 ... spirit, but-

Jean-Baptiste Kempf 34:14 ... do some filters, complex filters

Lex Fridman 34:16 Yeah.

Jean-Baptiste Kempf 34:16 You don't have the equivalent of Photoshop-

Lex Fridman 34:18 No

Jean-Baptiste Kempf 34:18 ... in command line, right? But you, for video, you have FFmpeg in command line.

Lex Fridman 34:22 Yeah. It's incredible. I mean, it's like an, it's an example of a thing when a bunch of great people get together and they get a vision, and they stick by that vision for many years, which is incredible.

Jean-Baptiste Kempf 34:33 And the vision behind, and the same for VLC and FFmpeg, is that we make every- everything that is very complex easy to use for the normal people, for everyone.

Lex Fridman 34:45 Mm-hmm.

Jean-Baptiste Kempf 34:45 Right? Our goal is to make something that is insanely complex technically and make it easy to use, right? And, and people, they use VLC, they drop a file. They don't realize how complex the file is, but they, they play it. Or, or people put any type of thing inside FFmpeg with complex filters, and it just works like magically, right? And people-

Lex Fridman 35:04 Mm-hmm.

Jean-Baptiste Kempf 35:05 And this is our mission, right? Make very complex things.

Kieran Kunhya 35:08 We wouldn't be here and, and you wouldn't be here if this required, you know, a traditional television studio setup.

Jean-Baptiste Kempf 35:15 Mm-hmm.

Kieran Kunhya 35:16 It, it's tools like FFmpeg that democratized this. The, the podcast and streaming revolution, the YouTube revolution-

Jean-Baptiste Kempf 35:21 Mm-hmm

Kieran Kunhya 35:22 ... was caused... You know, FFmpeg was a big player in that because it, it democratized this technology that was once in the, in the '90s, for example, you needed equipment that cost hundreds of thousands of dollars to do compression. It was the size of a car, and now everybody has that at almost an exact level playing field, and that's something that's so remarkable.

Lex Fridman 35:42 It gave voice to a lot of people, and we, uh, just to clarify, we say you, you wouldn't be here, not the human, but the podcast. [laughs]

Kieran Kunhya 35:49 The podcast. Sorry. You, you as the... Sorry.

Lex Fridman 35:50 I would still... VLC did not have anything to do on a biological level-

Jean-Baptiste Kempf 35:54 [laughs]

Kieran Kunhya 35:55 [laughs]

Lex Fridman 35:55 ... at creating me as a human.

Jean-Baptiste Kempf 35:56 But, but it's like, uh, you realize also everything move from text to images and images to video, right?

Lex Fridman 36:02 Yeah.

Jean-Baptiste Kempf 36:02 Look, look at social networks. Video is everywhere. It's the most powerful, um, medium there is, right? And, and when you see shorts and, and ins- reels and, and TikTok, right? It's amazingly powerful to, to, to give... Video is amazing for that, right? But the complexity is important.

Lex Fridman 36:20 It's what people don't realize. I mean, this is really-It gave power to the individual all across the world. That's real freedom. And I think, I can't believe it, but we still haven't mentioned the actual obvious thing for people who are not familiar, which it's open source, and there's a open source community of users and developers behind it. So it's really, it's a movement. So, like, we'll talk a bunch in a bunch of different ways about the community behind it. But, uh, can you speak to the open source element? So when we say what is FFmpeg, it's an open source project.

Jean-Baptiste Kempf 36:57 Yeah. So FFmpeg, VLC, x264, VideoLAN, everything we do is fully open source. And for the people who don't understand how open source is, my usual analogy is about a chocolate cheesecake, right? Um, usually for you, when you want to buy your cheesecake, you go to a bakery, they give you the cheesecake. Um, the other one way of having a cheesecake is have your grandma give you a recipe of how to make that. When we do open source, we give you the chocolate cake, and we give you the recipe to actually remake the same cake, but at the same times tell you how to build the oven and also how you're allowed to modify the recipe and resell it to someone else.

Lex Fridman 37:35 Yeah.

Jean-Baptiste Kempf 37:35 And this is because software is just a very long recipe of small instruction. Computers are not very clever. They go very, very fast. So a normal program has tens of billions of instruction instead of the tens when you have your chocolate recipe. So a lot of the software industry was about selling software, like, like where you just have, like, the final cheesecake. In open source, we make- we give you everything, and that manage to get a lot of people work together, right? Because then you decide that you're going to make the best program, the best recipe for video, and you create communities. In, um, FFmpeg, since the beginning of FFmpeg, probably two thousand to three thousand-

Lex Fridman 38:17 Yeah. In the thousand, yeah

Jean-Baptiste Kempf 38:17 ... people have contributed from, from the beginning, right? And then it's exactly like the Linux kernel, right? The Linux kernel has probably ten thousand people contributing everywhere, and they get together, well, mostly online, right? So they v- virtually get together to create the best tool for something. And on FFmpeg and VLC, it's just like, well, this codec doesn't work, so I'm going to work on the codec, and I'm going to add the support for this file inside FFmpeg, so it will be beneficial to everyone. Because again, we work for the greater good. We work for everyone, and that is what open source is.

Lex Fridman 38:54 And we, we should mention, depending on the licensing, uh, you could probably build a billion-dollar, maybe even a trillion-dollar company around basic- as a wrapper [chuckles] to-

Jean-Baptiste Kempf 39:07 Well, yes-

Lex Fridman 39:08 Yeah

Jean-Baptiste Kempf 39:08 ... people do.

Lex Fridman 39:09 Yeah.

Jean-Baptiste Kempf 39:09 People do, right? Uh, there was a lot of, uh, um, problems with, uh, mostly, uh, cloud providers who are basically running some open source tools, um, in, in the cloud and just give you, uh, the API to access to that. And there was a lot of, uh, um, databases like Mongo or Elastic who changed their license in order to avoid those type of, uh, of scenarios.

Kieran Kunhya 39:31 This is a question we get a lot in FFmpeg is, "Why don't you do that?" And you can't. We have, we have thousands of contributors, some of whom aren't even alive anymore.

Lex Fridman 39:40 Mm-hmm.

Kieran Kunhya 39:40 It, it- you would need all of their agreement to do that, and, and JB will go maybe a bit later and talk about how challenging that process was in VLC to do the re-licensing.

Jean-Baptiste Kempf 39:49 The license is a social contract in terms of Rousseau de facto of the community. The community does not agree on much beside the license. People go around, discuss around because of the license, and that also allow those license fork, right? Sometimes the community splits, but it's possible because of the license then to merge back.

Lex Fridman 40:11 Mm-hmm.

Jean-Baptiste Kempf 40:12 And we've seen that so many times, right? GCC and GC, um, and EGCC in the past. Uh, we have seen, for example, all the web browsers, right? They started as web, like KHTML, which becomes WebKit and then which becomes Blink, right? So open source license is like the, the core of the community and, um, people are coming from all around the world, very different type of, uh, religion, um, political borders. They, they work in the same way on a project to solve a specific problem, and the specific problem we're working on is to make multimedia easy for everyone.

Lex Fridman 40:50 Uh, looking it up on Perplexity here, looking at the different open source licenses. Most major open source licenses fall into two buckets: permissive, very few conditions, and copyleft, share-alike requirements for derivatives. Below is a brief practical summary of the main ones you'll see in the wild. MIT license, BSD, ISC, uh, Apache, uh, GNU GPL, GNU AGPL. Where's LGPL? Yeah, LGPL. Let's see. There's the Mo- the Mozilla Public License. There's Eclipse Public License. It goes on. There's a lot of variety. I mean, I think, I think the really popular ones is MIT, GPL, LGPL-

Jean-Baptiste Kempf 41:34 Yeah. And BSD. BSD

Lex Fridman 41:36 ... and BSD, Apache. Sometimes you'll see-

Jean-Baptiste Kempf 41:38 Apache as well. Un licensed, that's an option. Attempts to dedicate code to the public domain with a fallback permissive license. There are many licenses for many different thing. What people don't understand that public domain is something that doesn't exist worldwide, right? So it's, um, all the open source licensing u- use the copyright law, right, the international copyright law, in order to give rights on how you use the software or how you modify. It's de facto a, a copyright license contract that you give to the end user or to the developer. And so you have like the first one, which are basically very permissive, MIT, BSD. You give the code and basically you do whatever you want, right? You take it, you want, you modify, you do what you want. Um, and this is, uh, popular for, uh, uh, JavaScript and the type of, uh, BSD operating system.

Lex Fridman 42:29 So some of them, one, one of the parameters is whether they require attribution, meaning if you use the code, you have to say-

Jean-Baptiste Kempf 42:35 Yes. So in those type of permissive license, some you need to say if you use it, which is called attribution, and some you don't. And then there is aThe other part of license which are copy left, where you need to give back to the community your modifications and with different string attached. Um, some weak, uh, copy left license, like, um, the Mozilla Public License, to some which are a bit stronger, uh, like a GPL, or even very strong like AGPL. So all of those are different, um, type of licensing that depends on what your goal are and how you want to structure your community, which is why I spoke about social contract, because this is very important to understand. Um, FFmpeg and VLC are mostly GPL or LGPL. Uh, the Linux kernel is GPL, uh, but Android is Apache. Uh, a ton of, uh, JavaScript framework that are using are mostly MIT. Um, all the BSD kernels, OpenBSD, NetBSD are of course BSD. Um, and, and so the-- it's philoso-philosophical change on how you want people to contribute back basically.

Lex Fridman 43:44 Mm-hmm. So there's, uh... I think you talked about that, uh, you've moved at one point from GPL to LGPL on certain parts of the project. What, what-- can you describe the difference between the two, and what does it take to move to, I guess, a more permissive... So that direction is more permissive. LGPL is more permissive than GPL.

Jean-Baptiste Kempf 44:04 Yeah. So you have to realize that you can always go from more permissive to less permissive, right? Because of course, those license are basically statement. So if you restrict, you can always restrict more, right? So i-in a GPL project, you can take MIT code, but you cannot do the opposite, right? Because there are more constraint to, to match. Um, uh, indeed, in fact, I changed the core of LibVLC, which is the engine of VLC-

Lex Fridman 44:33 Mm-hmm

Jean-Baptiste Kempf 44:33 ... from, um, GPL to LGPL. Uh, and there were two reasons, uh, to do that. The first one is that so people can use, um, the VLC engine, LibVLC, into third-party applications. So a lot of applications which are playing video on your phone or on your tablet are actually VLC engine in it, which is calling FFmpeg in it.

Lex Fridman 44:54 Yeah.

Jean-Baptiste Kempf 44:55 Um, so that was one of the way to create, uh, one of the company I created, which is doing consulting and integration of those types o- of application, where you integrate VLC into third-party solutions, um, like inside game engines or stuff like that. With GPL, you couldn't do that because that means you needed to open source everything, and those are, uh, for a lot of, like, commercial companies who don't want that.

Lex Fridman 45:18 So you can create a company with LGPL, you can create a company around it.

Jean-Baptiste Kempf 45:22 Yes.

Lex Fridman 45:22 You can do a commercial thing. You don't have to open source it.

Jean-Baptiste Kempf 45:24 Yes.

Lex Fridman 45:24 So that's a big, big leap.

Kieran Kunhya 45:25 So you could play video in your game.

Jean-Baptiste Kempf 45:27 Yes.

Kieran Kunhya 45:27 The pro- the problem is I, I'm a game developer, and I want to play some videos-

Jean-Baptiste Kempf 45:32 Mm-hmm

Kieran Kunhya 45:32 ... and I don't want to be forced to open source the entire game just to play those videos. So that's where the, the consulting business, the LibVLC LGPL-

Jean-Baptiste Kempf 45:39 Mm-hmm

Kieran Kunhya 45:40 ... allows you to do that. The LGPL, the Library GPL, as it used to be known, allows you to do that.

Jean-Baptiste Kempf 45:44 And FFmpeg is exactly desi- the same. It force-- LGPL forces you to give back what you change on this component, this-

Lex Fridman 45:52 Yeah

Jean-Baptiste Kempf 45:52 ... library, which is why it's L- Library GPL.

Lex Fridman 45:55 Mm-hmm.

Jean-Baptiste Kempf 45:55 And so you can use FFmpeg as LGPL into, like, any type of application, even non-open source, but you need to d- give back the modification you did on FFmpeg. Same on LibVLC.

Lex Fridman 46:06 Is it limiting from an open source perspective to go GPL? Because if you-- if your, uh, library, if your code is GPL, it means you're not... You, you, you're basically discouraging companies from building a business around it.

Jean-Baptiste Kempf 46:22 Yes.

Lex Fridman 46:22 Right? Is that... Is that fair to say?

Jean-Baptiste Kempf 46:23 It depends on... It depends on the company, but the company whose business model requires the source, the, the application to be closed source, yes, it's limited. So that's why, for example, I moved to LGPL. The second reason, um, is a, a bit more obscure, is that the, the terms of condition of the, uh, app store, the Apple App Store for iOS makes it very complex to have GPL application on it, while it's easier to have LGPL applications on it. So VLC on Windows and on Mac, uh, and on Linux is GPL. The core is LGPL. But on iOS, uh, the iPhone version and the Apple TV version is a type of different license called the MPL. Um, and yes, I, I went and changed the license, and it was a long story.

Lex Fridman 47:08 Yeah. So I, I think basically to change the license, you have to contact all the contributors.

Jean-Baptiste Kempf 47:12 Yes. It, it, it's very important to understand that open source projects are what we call in the US copyright, uh, law, joint work, or in, um, civil law, um, collective works or collaborative works, is that you work all together in term the same goal, and then it creates one software, which is one release. Um, but the copyright is kept by all the individuals. Some open source project don't do that. They force copyright assignment, but this is not what we do. We're communities. So everyone has basically copyright on what they changed. Um, and this copyright stays even if at the end your, your contribution was deleted because the new contribution was based on your previous one, right? So if you want to properly re-license, you need to find all the contributors. A-and at that time, I had to contact more than three hundred and fifty people. And sometimes, well, they're just an email, right? So it's-- you need to actually track down. I, I actually, like, traveled to some place to go someone that I was like, sorry, that I'd found online to see a... to go to their job and say, "Well, you l-licensed that. Can you-- do you want to change from GPL to LGPL?" Most of the times they don't even care. They wanted to help VLC. But also it, it brought me to very complex situation. I arrived to the work of a, a person who was a factory vo- worker.

Lex Fridman 48:30 Mm-hmm.

Jean-Baptiste Kempf 48:30 Um, and I said, "Well, I need to you to sign that," because it was his son who died who actually wrote the code, right?

Lex Fridman 48:39 Mm-hmm.

Jean-Baptiste Kempf 48:40 So I had to explain all those type of open source meaning, and no, I was not a company trying to rip out the two line or five line that that guy did-

Lex Fridman 48:49 Mm-hmm

Jean-Baptiste Kempf 48:49 ... but was useful, and the whole community agreed on that, and he had no idea I was a factory worker. This com-- A-and I was a lot younger, right? Like, uh, it was m- fourteen years ago, andLike, like I was almost in tears, right? It's ve-very difficult, right? We are talking about lives of people, and he explaining, and we went talk about the photo of this guy, right? So it's important to do it right and to do it correctly. Um, but yes, that means tracking down everything because every contribution works.

Kieran Kunhya 49:18 Mm-hmm.

Jean-Baptiste Kempf 49:18 There are some project who don't respect that, and we do re-licensing a bit, like aggressively.

Kieran Kunhya 49:23 Mm-hmm.

Jean-Baptiste Kempf 49:23 But as I said, it destroyed the whole heart of the community because it's-- we o-only agree on the, on the license, so that's important.

Kieran Kunhya 49:31 I would emphasize the community is such a wide-ranging group of people. There's people in a Syrian war zone with electricity part-time. There's, there's all people from all walks of life-

Jean-Baptiste Kempf 49:44 Mm-hmm

Kieran Kunhya 49:44 ... rich, poor, young, old. Um, so it's quite remarkable to get, you know, a group of people aligned on something. And that's an achievement in itself.

Lex Fridman 49:56 Yeah, it's incredible. And a lot of them are introverts, so you coming, coming to find them and getting them and getting them to answer an email might be quite, quite difficult.

Jean-Baptiste Kempf 50:05 Most of us are introverts, right? You, you need to be more precise. You have extremely introverts, extremely, extremely introverts and introverts, right?

Lex Fridman 50:12 Yeah.

Jean-Baptiste Kempf 50:12 It's just like a whole spectrum of different people. It doesn't matter. The important is, is your code good?

Lex Fridman 50:19 Mm-hmm.

Jean-Baptiste Kempf 50:19 Is your code great? Is your technology great? We care about excellent code. We don't care who you are. Sorry, it's just like we have no idea to check. We cannot check, right? Like, maybe you're a dog. I don't care, right? I don't care where you come from. I need to look at your code. And this is important because people don't understand that, and they come to the community and send them some patches, and they get rejected, and they don't like that because, I mean, you're just like, "Sorry, it's not up to our standards." "Oh yeah, but I'm engineer at this very large company in, in, in Italy, in Germany, in the US." We don't care. We care about the quality of your code because this is what defines our community, and which means that we have a lot of people who contribute who are some very different backgrounds and, and, and very introverts, sure. But that's okay,

Jean-Baptiste Kempf 51:07 right?

Lex Fridman 51:07 So one, one of the legends of the community is, uh, of course, Linus Torvalds, who, uh, created Linux and is a longtime maintainer of the Linux kernel. As the legend goes, he can be pretty harsh on this meritocratic process of reviewing the code and saying it's not good enough. Can, can you just speak to, to the legend of Linus Torvalds?

Jean-Baptiste Kempf 51:29 Linus is one of a kind, right? And, and I would even go and say that what he did on Git is more interesting than what he did on the Linux kernel.

Lex Fridman 51:39 Mm-hmm.

Jean-Baptiste Kempf 51:39 He's very harsh, but what people don't see is usually when he's harsh to, it's people who are maintainer of part of the kernel, right? So they know him, right? So he's not very harsh like that to everyone. The thing is, what he created in his room is basically powering every server online, right? Even at Microsoft cloud called Azure, I'm quite sure seventy, eighty percent of the servers are running Linux. All your Android phones are running Linux. What he did, um, with the power of open, of open source, sure, is amazing. Um, and yes, the quality of the Linux kernel is very high, and yes, it's difficult, but we cannot compromise on that. We cannot compromise on quality because in the end, and you have to understand that, is the core community of VLC is five people. The core community of FFmpeg is ten to fifteen, and we are the ones who are going to maintain your code, right?

Lex Fridman 52:37 Mm-hmm.

Jean-Baptiste Kempf 52:38 Because one thousand contributors in the timeline and just ten staying, it's one percent chance that someone comes and stays. One percent. So you will have change of job, change of wives, you have children, you have accident in life, you're going to change jobs, whatever. You're not going to come back. It's most likely. So we are the one going to man-maintain your code. It needs to be maintainable. It needs to be excellent. And yes, sometimes that means that you need to rework your work because it was good, but it's not excellent. And we need excellence because we are very few to maintain something that is critical for the whole.

Lex Fridman 53:15 But we should also mention that there is some, some spiciness, some harshness to the language that's sometimes used when you're, uh, keeping this high bar of excellence. Uh, is there something to say to that?

Jean-Baptiste Kempf 53:27 It's, it's true, right? Um, it's also the fact that, for example, what we're doing is low level, it's extremely technical. Uh, you get into this community, the tone gets, um, very like a type of-- It's a subculture, right? So people who arrive from the external are basically not known to the subculture. Most of those people around FFmpeg and VLC, we do VideoLAN DevDays, VDD, uh, every, every year. They are so fun in real life, and they love it.

Lex Fridman 53:54 Mm-hmm.

Jean-Baptiste Kempf 53:54 But it's true that you're online and sometimes, like, the, the tone, you don't realize how it is. Uh, but that's okay.

Lex Fridman 54:01 It's a culture. I mean, you, you get this in the gaming culture. There's pretty harsh, intense, the way people communicate, and it's-- everyone understands that the way you show love and respect just looks different in different communities. Sometimes people... [chuckles] It de- it depends. If it's a book club, usually people are going to be much sweeter. If it's an open source [chuckles] project that's very high stakes and used by millions of people-

Jean-Baptiste Kempf 54:24 But it's very not often insults that you see, for example, in the gaming, right? And so Linus' tone is a bit unusual, even for the open source community. It's more like it's more harsh on the result, saying, "No, this is not good. This is crap." Those type of things that you will see.

Lex Fridman 54:40 Try not to make it about the person, make it about the code.

Jean-Baptiste Kempf 54:42 Yeah.

Kieran Kunhya 54:42 Yes. It's very, very matter of fact, and I think you've got to look at it in terms of, you know, the famous FFmpeg is developed almost entirely by volunteers, and that's true. And you've got to imagine someone's done a hard day's work at their day job. They come home. You know, terseness might be a thing, you know, it... And, and that's not something to take personally.

Lex Fridman 55:01 you're tired, you're busy, but you still care about this open source stuff.

Jean-Baptiste Kempf 55:05 Yeah.

Lex Fridman 55:05 But you may not be able to explain and, and handhold someone on every subtle detail.

Jean-Baptiste Kempf 55:10 And also you have to realize that most people don't speak English as native language.

Lex Fridman 55:16 Mm-hmm.

Jean-Baptiste Kempf 55:16 And, and this is especially for, for open source projects like FFmpeg and VLC, which are mostly centered out of Europe. Sometimes like, um, people who are from the US or, or just like are very not happy about the tone, but most of the time it's also like they don't know better, right? It's difficult. Language is-- English is a difficult language. There is so many subtleties and tones and so on that you don't have, right?

Lex Fridman 55:40 Mm-hmm.

Jean-Baptiste Kempf 55:40 So, um, often it's also difficult in those type of community about like different cultures and languages.

Lex Fridman 55:46 So as the legend goes, JB, you, uh, repeatedly turned down millions of dollars to, uh, keep VLC open source free for everyone without ads. So take me through the reasoning behind that decision of leaving millions of dollars on the table.

Jean-Baptiste Kempf 56:06 Yeah, that's like almost a meme, right, on, on Reddit or-

Lex Fridman 56:09 There literally is a meme on Reddit. [chuckles]

Jean-Baptiste Kempf 56:11 9GAG and yeah, yeah, I see. There's-

Lex Fridman 56:14 You [chuckles] looking like a wizard in the, in the VLC hat on Reddit. This is, uh, JB, the creator of VLC media player. He refused tens of millions of dollars in order to keep VLC ad-free. [chuckles] Thanks, Jean-Baptiste Kempf. You can even summon him on Reddit. [chuckles]

Jean-Baptiste Kempf 56:33 Yeah. And usually if you see, right, it's usually like people tag me, right? And, and, and then there is me, and then like I say, "Good morning."

Lex Fridman 56:40 Good.

Jean-Baptiste Kempf 56:40 I got twenty-four K upvotes, which is great, right? My karma on Reddit is amazing. [both laughing] At least on that account. Um, so the question is, needs to be answered firstly, what is a story about VLC, right? Um, because yes, this is true, I refused dozens of millions of dollars, yes, several time. Yes, I could be a multimillionaire and be somewhere on the beach. Um, but I did not do it because I thought it was not moral, and it was not the right thing to do. And this is very important for myself, is to be like, I work for the greater good, I work for people, and I don't want-- It's not just by my, by myself. But the reason is also because I, I did not feel that I'm completely legitimate to do that, and let me explain to you why. VLC story is a very weird story. In France, we have university, and we have a type of top colleges, um, and those top of excellency schools are engineering school, business schools, and basically lawyers a-and medical, right? But they're outside of university, and in order to enter those, you spend two years working like crazy math, physics to enter those best engineering school. One of the school is called the École CentraleParis. It has changed name since, but it was called the École CentraleParis. And because it was Centrale, they had to move it because it was too small after the World War II, and, and they moved it-- They wanted to move it to the central of France in a place called Clermont-Ferrand, and the alumni decided that this was not okay, right? It is the, the school that Eiffel, right, the, the one who did the Eiffel Tower, attended to, right? So they said, "No, no, we are amazing, great school. We cannot do that." And so they bought a piece of land south of Paris, uh, very near Paris. Um, and it was a campus managed by a nonprofit of the alumnis, okay? Because of that, everything on the campus was managed by student. M- The university did nothing, right? So, uh, radio, uh, TV, uh, supermarket, uh, library, um, defining who was going t- into which rooms. Um, everything was managed by the student.

Lex Fridman 58:48 That's amazing. That's an amazing experiment, that it all, all didn't go to hell quickly. It somehow flourished.

Jean-Baptiste Kempf 58:55 It worked great, and I learned so much, um, in my life doing those side activities, right? Because you're, you're twenty-two, and you need to run your campus, else you don't have electricity, right?

Lex Fridman 59:05 Mm-hmm.

Jean-Baptiste Kempf 59:05 So you care about that, right? But anyway, in the '80s, um, they did a full experiment of deploying a network, uh, mostly sponsored by IBM and 3Com, which was a token ring network. So token ring is something that probably almost no one knows about anymore. It's a networking technology where you don't have routers, right?

Lex Fridman 59:26 Mm-hmm.

Jean-Baptiste Kempf 59:26 Everyone is linked. Um, it's type-- like really a ring, and when you want to send a message, you talk to your neighbor, who's going to put the message to the next one, who's going to put the things to the next one in terms of ring. The issue with token ring is, of course, is that it's very slow because every computer on the network needs to open the message, see if it's okay. Is it for me? No, it's not, and then send it back like a token, which is traveling around the ring. In the '80s, you're doing some Telnet and sending mails as university. That's okay, right? But starts the '90s. I-In the '90s, um, and start video games, and when you have li- high latency in video games, basically you die, right? So in nineteen ninety-four, nineteen ninety-five, around Doom and Duke Nukem coming around, they want a faster network. So the students go and see the university and say, "You know what? We want a faster network. We need to work," which, and also play video games. Um, and the university tells them that basically, "Oh, I'm sorry. We cannot help you because you understand the campus is not ours. You manage it, so do something. And you should see some basically partners of this university and basically go away." And they go, and they actually go and see, um, the, the, the CIO of, uh, Bouygues, which is a large, uh, French company and who's doing some TVs, uh, in France. And he says, "Well, you know what? The future of video is satellite." Well, today we know it's not, but at least it was a good idea. In nineteen ninety-five, the first of satellite d-dish, and he says that instead of having like one satellite dish and a big decoder for each of the students, which are one thousand and five hundred, what about you build, like you put a n-Enormous dish and only one decoder, and you send the video directly on the network.

Lex Fridman 1:01:15 Mm-hmm.

Jean-Baptiste Kempf 1:01:15 And that required a very fast network. Today, it's obvious, but at the time was, like, the first to do video streaming. So they built this project, which was called Network 2000. Of course, right, we are in the '90s, right? Everything is-

Lex Fridman 1:01:29 [laughs]

Jean-Baptiste Kempf 1:01:29 ... futuristic is called 2000, like-

Lex Fridman 1:01:30 Yeah, 2000, yeah.

Jean-Baptiste Kempf 1:01:32 And so they do the Network 2000 project. It's completely hacked. It crashes after 45 seconds. That's okay. The demo is 40 seconds. It leaks memory. That's okay. They put 64 megabyte of RAM instead of the 8 or 16 you have, and the demo should have stopped there, and that was the Network 2000 project by the students.

Lex Fridman 1:01:49 What was the format of the video that they had to work with?

Jean-Baptiste Kempf 1:01:52 MPEG-2 because satellite is MPEG-2 TS for transport, MPEG-2 video, and MPEG-2 audio at that time. And the project should have stopped there. Everyone was happy. They had, like, amazing ATM network at 155, 55 megabits per second. They had probably one of the best network in Europe at that time, and they stopped the project. Six months or a year later, two students arrive and say, "Well, you know what? Maybe other people care about video streamed on a local network," and they create the VideoLAN project, VideoLAN. And one of them is called Christophe Massot, that is a good friend of, uh, both Kieran and me, and they start the project. It's not even open source yet, and they spend around three years to get the school to agree to make it open source. Because the, the university wanted to get some-- because of the IP and copyright of the students, wanted to basically monetize these MPEG-2 decoders.

Lex Fridman 1:02:47 Just to be clear, so what's the main application? Streaming on a local network?

Jean-Baptiste Kempf 1:02:50 It was streaming on a local network.

Lex Fridman 1:02:52 By the way, that's just, like, to state the obvious. This is before YouTube. This is before-

Jean-Baptiste Kempf 1:02:56 Ten years before YouTube. You have a Pentium 60 or 75, right? Y-y-you, the main machine was 48, 86DX at 33 megahertz, right?

Kieran Kunhya 1:03:04 Bear in mind, tele-television was the main form of video at the time. You could get new channels. In the '90s, having even one new channel when you, you grew up with four channels, having a fifth or a sixth-

Jean-Baptiste Kempf 1:03:14 Mm-hmm

Kieran Kunhya 1:03:14 ... was a big deal, and so having this satellite service with, you know, dozens, even hundreds of channels was so groundbreaking.

Jean-Baptiste Kempf 1:03:22 Especially because this is university where you had ton of different nationalities, right? So there was a ton of people who wanted-- So the-- in the end, they had, like, several dishes on different type of satellite, right? Because, for example, a lot of people were coming from, uh, the Maghreb or the Middle East and, and they, they, so they went to different type of, uh, uh, of satellites. Anyway, the solution w-worked great, and they started the VideoLAN project. The VideoLAN project has several s- and, and some are completely crazy solutions, like one how to create multicast on an unicast network, but let's not come to that.

Lex Fridman 1:03:55 [laughs]

Jean-Baptiste Kempf 1:03:55 It's too, too complex. But VideoLAN client part is what became VLC.

Lex Fridman 1:04:01 Mm-hmm.

Jean-Baptiste Kempf 1:04:03 Actually, they basically strong-armed the, the university to force it to open source because university did not understand that. And in 2001, it's still early. But basically, yes, the university agreed early 2001 to make it open source. I joined the project in 2003 because that's when I joined the university. So the first thing is I'm not the one who created VLC because actually no one did, right?

Lex Fridman 1:04:26 Just kind of naturally emerged from the VideoLAN project. And we should mention that, like, again, you, you, you said it just, but to make it clear, VideoLAN, uh, as what it became was at the time was a set of technologies around video, and the VLC, what you called the client, that's the thing that most normies, uh-

Jean-Baptiste Kempf 1:04:47 That is correct, and-

Lex Fridman 1:04:48 ... think of, like, as the thing, which is, like, the thing that pops up when you click on a video and you play it.

Jean-Baptiste Kempf 1:04:53 So I arrive in 2003, and then I will create, uh, the open source, um, nonprofit organization called VideoLAN, and I took everything out of the university to create it an, a nonprofit project and something sustainable. It's, yes, it's true that I spent more time than anyone on VLC and VideoLAN. That is sure. Um, but it's a continuity of a previous project, VideoLAN, the student project, which is a continuity of the Network 2000 project, which is a continuity of that and that.

Lex Fridman 1:05:22 I'm sure there's moments along the way there you were thinking of, like, what is the future of this from an open source perspective? 'Cause as, as the internet is blowing up, and there is companies-- I mean, for people who don't remember, like, there's companies making huge amounts of money.

Jean-Baptiste Kempf 1:05:39 A-and I can tell you that in 2005, the project should have died, um, and I made it to continue the project. Uh, at some point, we were only two active developers. Um, and I thought it was great technology and was useful, and it will be useful and, and, and I made that my, my, my life and my, my time. Um, and I made that grow from a few hundreds of thousand of users, millions of users to what we have now, which is probably billions of, uh, version of VLC around the world and used everywhere. Um, so that's a bit the story of VLC. There is ton of very funny story around that. Many people from, uh, around the world working on it, like you said, in Syria or middle of nowhere in India. But along the way, I got several offers which were either to bundle, uh, toolbars, right? You remember those horrible toolbars-

Lex Fridman 1:06:35 Oh, yeah

Jean-Baptiste Kempf 1:06:35 ... which were basically spyware or changing your web browser or your search engine or even, like, advertisement inside VLC. And I didn't like that, right? I am-- And, and people don't understand that. It's not, I'm not against money, right? I'm very happy to make money. I created several startups and, and one I hope that is going to work very well. It's the fact that I believe that you need to win money ethically. There is a right way of doing that, and, and doing sneaky advertisement or stealing data is not the correct way, right?

Lex Fridman 1:07:10 Mm-hmm.

Jean-Baptiste Kempf 1:07:10 For example, if Netflix arrived at some point and say, "Well, we want to put Netflix inside VLC," probably the story would have been different, right? But they didn't. The only people who came to us wereshady ads company.

Lex Fridman 1:07:22 Mm-hmm.

Jean-Baptiste Kempf 1:07:22 And if I do that, right, I would have a ton of money, right? And then three years later, project is gone, right? Someone forks it and, and something else happens.

Lex Fridman 1:07:31 So it's not even necessarily ads or any of that, it's the shadiness of the-

Jean-Baptiste Kempf 1:07:36 Yeah

Lex Fridman 1:07:36 ... dishonesty of the-- So you had a good radar, you had a good threshold of like, no, this compromises the spirit of what this is supposed to represent.

Jean-Baptiste Kempf 1:07:46 But also it's, it's for me, right? I'm, I'm like very selfishly, I need to go to bed at night and be happy about what I've done, right? Maybe it's my upbringing, um, uh, maybe it's my parents' fault or, or whatever, right? But I believe there is right and wrong, right? And this was the right decision at the time. It still is. I want to be proud of what I've been doing. And like, if I had sold out, I would have betrayed so many other people's work.

Lex Fridman 1:08:14 Yeah, well, I should say me and, uh, most of the internet thank you for that decision. It's inspiring, uh, for others, uh, I think that, um, that are pushing the open source movement forward, that it's okay to do these kinds of huge sacrifices if you believe it's right. And I think in that case it was right and it was the reason that VLC became as successful as it was, 'cause it, it em- it's an embodiment, it's a symbol of like, you know, freedom and what the open source community can create.

Jean-Baptiste Kempf 1:08:46 Yeah, and be a service for so many people around the world, and this is important.

Kieran Kunhya 1:08:50 We, we should emphasize in the 2000s it was really normal to download a program and it secretly installs some spyware.

Lex Fridman 1:08:56 Yeah.

Kieran Kunhya 1:08:56 It was, it was buried in very faint text or in the license text box that nobody reads that at the bottom-

Lex Fridman 1:09:03 Yeah

Kieran Kunhya 1:09:03 ... oh, I will be installing this toolbar-

Lex Fridman 1:09:04 Yeah

Kieran Kunhya 1:09:04 ... and changing all these things, and it was, it was very common to have to, you know, you install a program to do something at the time of any sort.

Lex Fridman 1:09:11 To put yourself in the mind of a developer at that time, I think it's very easy, to everybody listening to this, it's very easy at that time to convince yourself to take a few thousand dollars-

Kieran Kunhya 1:09:24 Mm-hmm

Lex Fridman 1:09:24 ... a few thousand dollars to do it.

Kieran Kunhya 1:09:25 Yeah.

Lex Fridman 1:09:26 To, to say no to much more money-

Jean-Baptiste Kempf 1:09:30 Yes

Lex Fridman 1:09:30 ... takes guts and takes vision.

Jean-Baptiste Kempf 1:09:34 The last offer I had was obscene, and they say, "Yeah, but imagine with all that money you could build something new, open source," right? It was like the mind trick was, it was difficult.

Lex Fridman 1:09:47 Yeah.

Jean-Baptiste Kempf 1:09:47 But for me it was just like, no, this doesn't work like that or this is not the right thing, so I don't do it. Uh, uh, uh, and again, right, it's not that I don't like money or, or whatever, it's just like it wasn't right.

Lex Fridman 1:10:01 Well, once again, thank you from me and from the rest of the internet.

Lex Fridman 1:10:05 Let me talk a little bit more about the open source movement, about the fact that, as you say over and over and over and over, FFmpeg, uh, is, uh, and many open source projects are built by volunteers. So, uh, there's a bit of drama recently, uh, Kieran, on the interwebs, on Twitter. Uh, you have a spicy style on Twitter that I think articulates and celebrates all the incredible developers and, uh, development and the, uh, the code, especially Assembly that's involved in, uh, building some of these codecs and building some of this incredible technology. But that brings us to the, a bit of a debacle that happened. Tell me the full saga of what happened with the Google security engineers.

Kieran Kunhya 1:10:50 Just to be clear, Google are one of the biggest supporters of open source out there. They have been for a long time. It's just, I think some things kind of went a bit overboard this time. So FFmpeg itself, and, and this is not like a secret, it's on the homepage, you know, the, it processes untrusted data. There can be security issues when you parse untrusted data. That's very normal. But recently what changed was Google started using AI to create security reports on an open source project, FFmpeg. Volunteers had to deal with that. They did-- They provided very limited funding and they even went to the media first announcing how good their AI was before the issues could be fixed.

Lex Fridman 1:11:29 And this is in the public forum.

Kieran Kunhya 1:11:31 Yeah, this is all public.

Lex Fridman 1:11:31 So reporting, reporting an issue using AI to find an issue in the code, which is a security vulnerability, and then reporting that publicly before you're able to fix it.

Kieran Kunhya 1:11:40 Yeah. It's announcing how good their I- AI is, that they provided a standard 90-day industry deadline, um, without poten- without really understanding the nature of volunteer-driven development. In addition, this vulnerability was on an obscure 1990s game codec. Um, the way-- And f- and let's look at it from their standpoint to begin with. Let, let's, um, you know-

Lex Fridman 1:12:04 Yeah. Can you steer me in their case?

Kieran Kunhya 1:12:06 Yeah, sure. They have substantial resources working on the security of open source projects that, you know, are ubiquitous and they've used, you know, a lot of compute to do that and very expensive and very capable security researchers, um, to do that. And that, that's their viewpoint is they are contributing by doing that. But I think that's where opinions differ. Um, it, it, it opened up a lot of interesting fissures, uh, I would say. Um, it does seem that there's a portion of the security community that, um, look at themselves a bit like building architects that never have to go to site. You know, going to site is something that is a little bit beneath them, the actual day-to-day construction. They're there to do their security things and it's someone else's problem. The security industry also kind of has a very aggressive tone towards things. The, the language they use is extremely aggressive. They use very strong language like, "You will get popped." Um, so and to, to Joe Public, get popped, you know, means something quite bad. Um, for them it means to get hacked. The way I would look at it personally is a little bit like the padlock on your home. Um, not everyone-- The padlock on your home or, you know, the lock on your home is there to, to protect against thecapabilities of, of what it's there to protect. It's not there to protect nuclear secrets. It's not there to protect Fort Knox. And it could be looked at that they're using AI at a level of scale to go and pick those locks and then say, "Hey, your locks, your lock's not secure. You need to deal with this." Whereas actually they're the ones with resources to be able to, to fix this. But that seems to not be something either they'll contribute to in terms of patches or in terms of financially. And the scale of AI is kind of the issue. The, the, the bug reports are very wordy. They're very, very... It's almost a denial of service by AI-generated bug reports on very niche codecs. Um, and the other issue the security community has is everything is marked high priority. You're going to, you know, "This is the most important thing in the world, and you need to deal with this. High, high, high, vulnerable, scary, scary, scary," on a game codec used on one disk in 1993.

Lex Fridman 1:14:23 Yeah.

Kieran Kunhya 1:14:24 And that, that's where the dichotomy lies. Going around telling everyone that their padlock's not safe, well, that's a hobby project of somebody. The, the safety of that codec is consummate to what that person thinks. It's their hobby. It's good that they're security analyzing it, but it doesn't need a big scary warning, "This is a critical vulnerability." Um, may also, may, may recently also, um, see that, um, there was another quote-unquote vulnerability. It wasn't at Google in this case, but, um, a filter could overflow and have an integer overflow, and one of your pixels could be the wrong color, and this was marked high, 7.5 severity in red. And at some point, the security industry needs to realize you can't keep crying wolf like this because this just leads to people, you know, the equivalent thereof of putting password stickers on their PC. You know, you can't just keep crying wolf every day. And, and I appreciate, you know, that's their modus operandi is to create as much scare and fear. But from the Google standpoint, at the end of the day, they need to contribute either financially or with patches. Google uses FFmpeg at, at a scale probably you or I couldn't even contemplate, millions of CPU cores. And yes, they contribute in areas mostly regarding their own products, so VP9, AV1. But in a wider sense, there's a disproportionate level of contribution. Yes, they fund students. Yes, they fund Summer of Code. And I think, uh, so Alex Strange is a former FFmpeg developer, um, I think posting in a personal capacity.

Lex Fridman 1:16:02 So he, uh, posted about security engineers on, uh, Hacker News. Uh, his post reads, "The problem with security reports in general is security people are rampant self-promoters," in parentheses, "Linus once called them something worse. Imagine you're a humble volunteer open source developer. If a security researcher finds a bug in your code, they're going to make up a cute name for it, start a website with a logo. Google is going to give them a million-dollar bounty. They're going to go to DEF CON and get a prize, and I assume go to some kind of secret security people orgy where everyone is dressed like they're in The Matrix. Nobody [chuckles] is going to do any of this for you when you fix it." [both laughing] So basically commenting on the sort of the incentives for the different, uh, people involved and misaligned.

Jean-Baptiste Kempf 1:17:01 The problem here is, is a disproportion of means on discovery compared to patching it, right? And this is the biggest issue, right? Um, a- and after that debacle, Google did some changes-

Kieran Kunhya 1:17:15 They are now starting to send patches, which is, yeah

Jean-Baptiste Kempf 1:17:17 ... And they also now have reward tools for fixing issues. So it, it has changed a bit because of that, uh, debacle. So it's good, right? But we've seen, and we talk about Google, but we have seen like some, some other large companies saying, "Oh, you need to fix this bug because it's critical in our product."

Lex Fridman 1:17:33 Can you explain the XZ fiasco? The FFmpeg tweet reads, "The XZ fiasco has shown how a dependence on unpaid volunteers can cause major problems. Trillion-dollar corporations expect free and urgent support from volunteers. Microsoft, Microsoft Teams posted on a bug tracker full of volunteers that their issue is high priority. After politely requesting a support contract from Microsoft for long-term maintenance, they offered a one-time payment of a few thousand dollars instead. This is unacceptable. We didn't make it up. This is what Microsoft, Microsoft Teams actually did." And then they, uh, you give the image and the details and all that kind of stuff, showing that these trillion-dollar companies are not giving much money, not giving much support.

Kieran Kunhya 1:18:24 They think, they think an open source project is a traditional vendor that they have an SLA. They think a public bug tracker is actually, you know, a, a third-party vendor's Jira where you can do all of these things. It, it, it's not. It is there to report bugs. I think the thing that made this particularly heinous was the name-dropping of Microsoft, the name-dropping that this is a visible product. If this was, if this was a, just a general bug report, I think that would have made it a lot better.

Lex Fridman 1:18:51 Yeah, so they literally said, like, "This is a big deal because a lot of people are using it in Microsoft." I wonder what happens psychologically. So I think what happens in these companies, maybe you can correct me, is they... Y- you're, you're right. They just think of FFmpeg as like a vendor that Microsoft surely is paying a huge amount of money to. They kind of assume that in their interaction, and nobody anywhere on the stack is going like, "Wait a minute. Shouldn't we be giving like millions of dollars to FFmpeg?"

Jean-Baptiste Kempf 1:19:25 And this is a very big problem in, in large... Like we're talking about some companies, but it's the same everywhere, right? Um, a lot of those companiesLike, the, the-- when we talk to that person, right? He was just like a, a manager on one project in Microsoft Teams, right? He had, like, never really, mm, discussed with open source community. He had no idea, right? It was like... A-and but the problem is that usually there is what we call OSPOs, right, open source program offices, in those type of companies, and they are the ones who are supposed to discuss with open source vendors-

Lex Fridman 1:19:58 Mm-hmm

Jean-Baptiste Kempf 1:19:58 ...um, or, or v- open source communities. But, like, they often don't explain that correctly internally, right? And here it's just like, "We are not your supplier. If you want me to be your supplier, I'm very happy, right? I will send you a contract and SLAs." Like, I created f- five companies who are doing that around open source projects, so that's okay.

Lex Fridman 1:20:19 We should say that some, some of the spicy tweets that Kieran, you're behind, and some of the debacle produced results.

Kieran Kunhya 1:20:28 Yes.

Lex Fridman 1:20:28 Positive results.

Kieran Kunhya 1:20:29 Donations have increased substantially. They're still not enough to cover even a single full-time developer, but on both a, you know, awareness level and a technical level, there's substantially more technical awareness and sort of awareness of the importance of FFmpeg as a result, as a result of X and, and what's happened. I can say, you know, it, it, it solved its purpose. People realize the level of importance FFmpeg has.

Jean-Baptiste Kempf 1:20:55 And on VideoLAN, it's the same, right? Like, for example, a, a very simple example. For more than a year, we couldn't update VLC on Android because of a bug on the Play Store, on Android Play Store, right? The only way we got someone to answer was to put an very spicy, as you say, um, tweet saying that we are going to stop distributing VLC for Android, right? And we have around one hundred million people using that. Um, and now then someone from Android actually came and discussed to us, right? Um, we had the same issue with, with, uh, Microsoft or, or like saying that we were going to stop distributing VLC on the Windows Store. And unfortunately, we are so small that the only very strong power we have to solve those issues is blaming on social network because it snowballs and now they listen to us. But so as large companies often have difficulty talking to us. Like, for example, VLC, right, is probably one of the top 10 software used on Windows. I am not part of Microsoft ISV programs, right? I don't have a point of contact at Microsoft, right? Well, I'm sure any other software, Adobe, Spotify, has a co- point of contact. I don't have that, right? So raising awareness works. It's sometimes very spicy, lot of drama. Well, X and Twitter are okay for that, but it's efficient.

Lex Fridman 1:22:30 Uh, so everybody listening to this should go, uh, follow FFmpeg on Twitter, on X, follow VideoLAN on Twitter, on X. Uh, go donate. Donate-

Kieran Kunhya 1:22:44 Thank you

Lex Fridman 1:22:44 ...to FFmpeg.

Kieran Kunhya 1:22:45 And thank you, Lex. Over the years, several years, you've been a supporter of, you know, FFmpeg and VideoLAN on X, you know, sh- giving us shout-outs, appreciating, you know, what we do.

Lex Fridman 1:22:55 FFmpeg for life.

Jean-Baptiste Kempf 1:22:57 And for example, like, um, Tim Sweeney, Carmack, and a few others, like very high-level people have raised also the awareness on, on our, uh, X accounts, and that helped a lot also.

Kieran Kunhya 1:23:09 Karpathy as well.

Jean-Baptiste Kempf 1:23:09 Karpathy, yes.

Kieran Kunhya 1:23:10 Karpathy as well, yeah.

Lex Fridman 1:23:12 Yeah. I mean, also, you know, outside of the fact that so many people use it, it's so impactful on the world, it's also a great representation of a great open source project. Like the value of Assembly and C and making sure that, like, you take programming seriously for real world systems.

Kieran Kunhya 1:23:30 It's not just that. We, we'll talk about Assembly later, I'm sure, because that's its whole topic in itself. But it's also celebrating people like Andreas Reinhardt who do maintenance. It is, I believe, unpaid, as I believe, as a volunteer. He's doing massive refactorings. Uh, Andreas Reinhardt and Anton Kern of rewriting ffmpeg.c with threading, celebrating those guys, celebrating the untold labor that's gone into this that actually doesn't change anything from the user standpoint. The files are exactly the same, but wow, the, the, the airplane has been rebuilt whilst it's in the air.

Lex Fridman 1:24:03 Christian Garcia said, "As a teenager running this account," referring to the FFmpeg a-account, and you responded, "Teenagers have written more Assembly in FFmpeg than Google engineers," but also just pointing out that there's a lot of incredible contributors who are teenagers.

Kieran Kunhya 1:24:19 Like JB said, we don't care who you are, where you're from, what you do. Teenagers have written thousands of lines of Assembly, um, over the years. Give a shout-out back in the days to Daniel Kang. So also highlighting the work of people like Ruikai Peng. This is a sixteen-year-old, some of his first contributions to FFmpeg, actually doing and putting some of these quote unquote security researchers to shame by act- by actually finding issues and fixing them and being sixteen. There's no barriers. There's no barriers to you have to study on, at college under this person and understand these. It's you can learn C, and let's be honest, it's from, it's from the K&R book. Learn C. Uh, you can learn Assembly. We'll talk about that maybe a bit later. You can contribute to world-class technologies.

Jean-Baptiste Kempf 1:25:06 In VLC, um, one of the oldest contributors called Felix, he's the one doing everything on Mac and iOS. He's starting working on VLC. He was sixteen. Um, we had a guy called Edward Wong, who used to be a Google Summer of Code student who stayed for three years around VideoLAN. He was fourteen, right? Um, and, and part of Google Summer of Code and Google Code-in, which were programs where, uh, basically we have students or high school, um, we wrote a ton of Assembly for x264 a-and for VLC and for FFmpeg, right? So everyone can contribute.

Kieran Kunhya 1:25:41 And he also did a good job because he didn't play the alarmist, um, CVE heist, create a CVE, which is like, um, a public exposure of security and do these big, scary red 7.5 high priority. He just fixed an issue in Git after three days and just fixed it. He didn't need to go and play a big security drama about it. And I think, I posted, you know, the kids are all right, whereas-

Jean-Baptiste Kempf 1:26:05 [laughs]

Kieran Kunhya 1:26:05 ... there's, there, there's, there's, you know, there is a por- th- I'm not saying all security people do this, but there is a portion of the security community, as Alex said, that likes to hype themselves up by creating drama. They would've happily raised, "This is a high priority CVE 8.0" or whatever on a, on a issue that actually was in Git. It wasn't even in a release, it was in development, and three days later was fixed.

Jean-Baptiste Kempf 1:26:27 Well, I just wanna put a little bit of love out there, even to the bigger community. Uh, much love and respect to Google engineers. Like you said, they're, uh, uh, some of the, the best software engineers in the world, and they do contribute a lot-

Kieran Kunhya 1:26:43 Yes

Jean-Baptiste Kempf 1:26:43 ... even on the security front. And also, you know, I'm a big fan of Theo. Much love to Theo. He was part of this, uh, debacle and drama a little bit. I think when you just zoom out on the grand arc of human history, the drama contributed positively to everybody involved. Donations went up. It brought more attention to the topic-

Kieran Kunhya 1:27:04 Yeah

Jean-Baptiste Kempf 1:27:04 ... allowed, uh, everybody to bicker in a way that ultimately, uh, got them to figure out what FFmpeg is all about.

Kieran Kunhya 1:27:11 So the way, the way we looked at this is, like, it's a rap battle at the end of the day, you know?

Jean-Baptiste Kempf 1:27:16 [laughs]

Kieran Kunhya 1:27:16 No, but it is, it is.

Jean-Baptiste Kempf 1:27:17 It is. It is.

Kieran Kunhya 1:27:17 We say stuff. We say stuff.

Jean-Baptiste Kempf 1:27:19 Yeah.

Kieran Kunhya 1:27:19 But we can, we can leave it on, X is a perfect place for, you know, international rap battle. We, you say stuff, I say stuff about your mama, but it doesn't mean, you know, I'm g- I have an actual personal issue with her.

Jean-Baptiste Kempf 1:27:29 Yeah.

Kieran Kunhya 1:27:29 A- and that's what it looks like. The, the Theo situation, you know, JB can maybe expand, went a little bit too far and there was a little... But, you know, it, it's just a bit of fun. It's just a bit of rap battle. It's a bit, it's WWE. You know, everyone's having a bit of fun on X.

Jean-Baptiste Kempf 1:27:42 Mm-hmm.

Kieran Kunhya 1:27:42 It, it doesn't need to be taken seriously. You know, the, the teenagers thing, you know, that, so that, that guy was a Google employee saying, "Hey, you know, there are other ways to run an open source business." You know, go, and then it's like, oh, man, just have a bit of fun, you know?

Jean-Baptiste Kempf 1:27:54 Yeah.

Kieran Kunhya 1:27:54 That's what the point of this account is. And, and furthermore, if you can teach people about the ways of open source projects, assembly, et cetera, by doing that, I think there's a lot to be offered here. It's not dunking on people for dunking's sake. It's showing actually the story that I think X learnt is these are not big corporate open source projects. This is not Kubernetes where there's, you know, hundreds, maybe thousands of people-

Jean-Baptiste Kempf 1:28:15 Thousands

Kieran Kunhya 1:28:15 ... paid to develop this stuff. These are just people in their basements in their spare time, and if you can address that topic in a fun and entertaining way-

Jean-Baptiste Kempf 1:28:23 Mm-hmm

Kieran Kunhya 1:28:23 ... I think that, that, that's the good thing, and that's, that's the value of X and then the reach we have.

Jean-Baptiste Kempf 1:28:28 A- and, and to be honest, right, like even on, at Google, Google is one entity, but so many different people, right? And, and some, a- and there is a ton of Google engineer we work with, um, all the time, and even, like, Google from YouTube to Chrome to Chrome Media to the rest of Google, those are very different type of entities. But what we do is efficient. And, and for example, uh, for, for, for Theo, right, it went a bit too far. I had him, like I, I calm everyone down. I had him on the phone. We say, "Okay, like this goes too far," and so on. But in the end, um, yeah, it's a rap battle, but it's positive for the project. It, like the awareness we have on open source and, and I mean true open source from communities right now is increased dramatically in the last two years, and this is useful.

Kieran Kunhya 1:29:19 Uh, what do you think motivates all the incredible contributors that we've been talking about? Like, what's the, what's the engine? It's so interesting to see.

Jean-Baptiste Kempf 1:29:26 So-

Kieran Kunhya 1:29:26 Like you said, they're sitting in the basement. What's the driver? What's the engine there?

Jean-Baptiste Kempf 1:29:29 There are many, uh, drivers, but weirdly the main one is that what we do in multimedia plays videos, and video is cool, right? And, and for example, we have so many people in the community who arrive because they loved watching anime, right? A- and this is like the advice when people ask me, "What should I work on in open source? How do I start?" And my quest- my answer is always the same. Work on something you love.

Kieran Kunhya 1:29:56 Mm-hmm.

Jean-Baptiste Kempf 1:29:56 I am working on VLC because I love movies, right? And I love watching the same movies over and over, even if my wife hates me when I do that, right? But because it's interesting, right? Because it's a topic that you like, right? The first, that's the first thing where people come to usually to VLC and FFmpeg. The second thing is that technically we, because we search for excellent, this is the best school ever, right? This is the best school ever of programming. If you're good in C, in FFmpeg, if you know how to write Assembly, I assure you you're going to be one of the best programmers ever, even if you're working on writing TypeScript, because this is the most amazing thing to do. And you will, like, have to get reviews by some of the most seasoned programmer ever who are going to look at every part of your code and tell you why it's not great. It's like we are the best teachers that you've ever had in programming, right?

Kieran Kunhya 1:30:52 Andrew Kelley started Zig. He was an FFmpeg developer and started Zig after-

Jean-Baptiste Kempf 1:30:57 Mm-hmm

Kieran Kunhya 1:30:57 ... his FFmpeg school. I mean, it, it's the place to learn so many aspects of programming in the real world, in a thing used by billions of people. You have nowhere to hide. You have to be open and honest about your flaws and, and how you can learn and be better.

Jean-Baptiste Kempf 1:31:12 And what is also interesting in multimedia is that you have 16 milliseconds to display a frame. It's not like a game engine where you can basically slow down and wait a frame. Like, so it's, you need to be good, right? There is no choice, else you don't have your video. It, and, and because of how codecs, if you miss a frame, you're going to destroy the look of the video, right? So you need to be good. You need to be perfect to have the right thing. But also is that it's not just pure programming in the mathematical fence, right? A lot of people don't understand, but, um-In order to program correctly on the open source multimedia community, you need to understand how computer works. And, and when you write assembly, you need to understand about, uh, CPU pipelining, right? You need to understand how SIMD works, how the ALU works, right? You need to understand what, how IO works, right? And this is what I think that is missing to a lot of engineers and software engineers today, is understanding what we call computer architecture. Um, and, like, seriously, like, some of the debates is like, "Should we use this assembly call or this one?" And people say, "Well, no, it's going to be, like, three cycle on this, uh, type of CPU and this one," and has massive impact on the output, right?

Kieran Kunhya 1:32:27 We should expand, FFmpeg is probably one of the biggest CPU users in the world. There's proba- it's probably running-

Jean-Baptiste Kempf 1:32:32 Mm-hmm

Kieran Kunhya 1:32:33 ... as we speak easily 100 mil- order of magnitude 100 million, maybe even a billion CPUs as we speak. So every instruction matters. There's not... The, the, the impact, at least in terms of CPU, is massive for everything that we do.

Jean-Baptiste Kempf 1:32:51 So first you come because it's an interesting subject, then you stay because it's excellent, and in the end, you're very proud of it because it's on the end of everyone. Like, so many people like, "Oh, I'm working for whatever consulting company, and I'm doing some portal to download invoices for your PG&E." Wow, great. Like, so many w- jobs are like that. You're not going to, to tell that to your grandma. But if you go to see your grandma and say, "I do this so that you can play video on your laptop," they understand. And this is very important, right? Because you're working on VLC, FFmpeg, x264. It's in the end of hundreds of million of people, and you have an impact. Um, and so you can be proud of yourself. And so I think that in addition to great, doing a great resume, all those things are why people contribute.

Kieran Kunhya 1:33:42 Yeah, th- those are side effects. My favorite quote on this topic is, uh, John Collinson. He said, "The world is a museum of passion projects." You know, everything out there is, is a passion project. And open source multimedia and open source in general, you can just do that so much faster. There's such a faster network effect, you know?

Jean-Baptiste Kempf 1:33:58 Mm-hmm.

Kieran Kunhya 1:33:59 I can open a cafe, and that can be my passion project, but I have to get building codes, I have to build a building, I have to find a th- a, find a location, I have to do all the, you know, all sorts of things. Well, in the software world, that passion project can be, can move quickly, it can be amplified by the network effect, and that amplification can be more than the, the, the sum of, the sum of the parts. You know, you can be, you can find people interested in extremely obscure things and have a network effect and make something that is truly amazing.

Lex Fridman 1:34:31 And, uh, on that topic of passion projects, uh, Tim Sweeney actually said in a reply to a tweet that was, uh, complimenting JB, he said, quote, "Many things in the world only happen because an awesome person decides to do it. This is the case with VLC." And that speaks to something interesting to me, that it does seem that a small number of people, sometimes one person, can create something incredible in the software world. Like you said this over and over and over. I think JavaScript is an incredible thing created by, by, uh, initially a single person. The, some of the programming languages like Python and C and Java, like just one person has this vision, has this design, and brings it sometimes over a weekend is the initial spark.

Jean-Baptiste Kempf 1:35:18 Yes, Linus built Git in two weeks.

Kieran Kunhya 1:35:21 Yeah.

Jean-Baptiste Kempf 1:35:22 Wow.

Kieran Kunhya 1:35:22 Yeah.

Lex Fridman 1:35:23 Yeah, it changed the world, Git. I mean, it really changed the world.

Kieran Kunhya 1:35:25 Linus' passion project. "Hey, I- uploading this tarball to an FTP, like, deal with it." [chuckles]

Jean-Baptiste Kempf 1:35:29 But for me, it's not just in software, right? And I believe in, in, in individuals that are going to change the world, right? And it's with a good, as you said, vision, right? I want to do that. Um, it is useful, it will be useful. And whether it's going to, like, build train or cars or rockets or something, like I, I believe people who believe in themself and have a vision can have a huge impact for humanity.

Lex Fridman 1:35:56 Let's actually s- uh, zoom out before we zoom back in. We'll just keep going up and down the, the stack. So, uh, you know, we've been talking back and forth VLC and FFmpeg. Kieran, you said that FFmpeg and, uh, VideoLAN, VLC coexist, and there's no central point of importance. It's a kind of what you call the binary star system.

Kieran Kunhya 1:36:17 Yeah.

Lex Fridman 1:36:18 Uh, they succeed because of each other. Can you explain the difference, how they interact? What is the-

Kieran Kunhya 1:36:24 Sure

Lex Fridman 1:36:24 ... Are they competitors?

Kieran Kunhya 1:36:26 I don't, I don't think they're competitors. I think, I think the simple answer is, the short answer before I go into detail is VLC is to FFmpeg as Android is to Linux. So they depend on each other, but they, they coexist because of each other. So they are a binary star system is the analogy I used.

Lex Fridman 1:36:43 By the way, I feel horrible that I just recently learned that Alpha Centauri, the closest star system to us, is a triple star system.

Jean-Baptiste Kempf 1:36:50 And, and when you start doing the physics, it's a nightmare, right?

Lex Fridman 1:36:53 Yeah.

Kieran Kunhya 1:36:53 Oh, yeah.

Jean-Baptiste Kempf 1:36:53 But, but, but like-

Kieran Kunhya 1:36:55 Hence the three-body problem. But anyway. So a lot of FFmpeg pipelines involve the x264 project, which is a VideoLAN project. I w- I would put a finger in the air that's, and say 80-plus percent of those pipelines are dependent on a VideoLAN project. VLC, obviously, as we've discussed, a VideoLAN project, uses FFmpeg, gives it reach, exposure to weird files, uh, historically used some donation money to fund FFmpe- FFmpeg development, and we'll talk a bit maybe about some of the reverse engineering later. So it's a binary star system. They, they work and feed off each other. Many of the developers are shared. There's no central location. It's a virtuous cycle working together.

Lex Fridman 1:37:36 And we should mention that x264 is the encoder for H264 video standard. So H264 is the standard.

Kieran Kunhya 1:37:44 Yes.

Lex Fridman 1:37:45 x264-

Kieran Kunhya 1:37:46 Is the implementation, open source implementation of the standard

Lex Fridman 1:37:49 ... that's used by basically everybody-

Kieran Kunhya 1:37:51 Yeah

Lex Fridman 1:37:51 ... for everything.

Kieran Kunhya 1:37:52 Yeah

Lex Fridman 1:37:52 It's that, that is the main driver of this. When you think of an MP4 file that has H264 codec in it

Kieran Kunhya 1:37:59 If it, if it came from a software environment like a data center or somewhere, the chances are it was created with x264.

Lex Fridman 1:38:06 And that's under the flag of, uh, VideoLAN.

Kieran Kunhya 1:38:09 That's a VideoLAN project. So in the VideoLAN graphic, it sits in the VideoLAN world.

Lex Fridman 1:38:14 And VideoLAN has a, says a bunch of stuff in it. Go to the VideoLAN website, there's a bunch of icons.

Jean-Baptiste Kempf 1:38:21 Like if you look, there is so many libraries, right? Um,

Lex Fridman 1:38:24 Lib DVD CSS.

Jean-Baptiste Kempf 1:38:27 Yes.

Lex Fridman 1:38:27 Lib DVD Nav, Lib DVD PSI, Lib VLC of course, VLC Unity, Lib blu- [laughs] Lib Blu-ray.

Jean-Baptiste Kempf 1:38:36 Lib Blu-ray.

Lex Fridman 1:38:36 [laughs] Uh, yeah, there's many more.

Jean-Baptiste Kempf 1:38:39 A- and there is so many more, right?

Lex Fridman 1:38:40 Yeah.

Jean-Baptiste Kempf 1:38:40 Like l- lately, the David project that we might talk about is the last project from VideoLAN. It's everywhere, right? And we do- we have a Lib Spatial Audio lately that we announced. We have a-

Kieran Kunhya 1:38:51 CheckASM.

Jean-Baptiste Kempf 1:38:52 CheckAS- CheckASM-

Kieran Kunhya 1:38:53 We'll talk about that later

Jean-Baptiste Kempf 1:38:53 ... which is like an insane project-

Lex Fridman 1:38:55 Nice

Jean-Baptiste Kempf 1:38:55 ... but amazing. Um, so, a- and x264 is one of those VideoLAN project. And my opinion, for example, is that x264 was, is the most amazing encoder ever designed, and this helped the adoption of FFmpeg. A lot of people and large companies went through FFmpeg because they wanted to use x264, and x264 increased the popularity on FFmpeg. But also VLC had its popularity because it's played so many files that were done by FFmpeg, right? So it's, it's many project that are interwind a- and work together.

Kieran Kunhya 1:39:32 Yeah. Unfortunately, there's a, there's a thing on X where, um, VLC is mentioned and there's people, "A quick reminder that it's FFmpeg inside doing the actual work." And that, and that's like I said, it, it's not... That, that's not the case. The, the, we work together.

Jean-Baptiste Kempf 1:39:46 And, and to give you an idea, right? When I compiled VLC for Windows, I compiled around 16 million lines of code, right? One million of those are inside the VLC repository, and FFmpeg in total is probably two, around two, right? But so it means that so many dependencies are outside. And if you also look at FFmpeg per se, FFmpeg also is integrating third-party libraries like x264, but Lib Opus and so many others, right? So we all depend on each other.

Lex Fridman 1:40:15 Uh, yeah, that's why I was, uh, hoping to do this episode as we are doing that just, uh, kind of joins FFmpeg and VLC-

Kieran Kunhya 1:40:23 Yeah

Lex Fridman 1:40:23 ... 'cause it's really, it's, it's really two, two of the same, like you said, b- uh, binary star system and we're all just orbiting

Lex Fridman 1:40:29 it. Uh, can we give a shout-out to some of the people along the way? We th- didn't really quite talk about the history of FFmpeg, so, uh, m- maybe can, can you tell me about Fabrice? Can you tell me about Michael Niedermayer? Can, can you tell me about some of the key figures here?

Kieran Kunhya 1:40:46 Let's just talk about the, the eras of FFmpeg, 'cause there, there's key eras that, and key people that made this possible. Um, Fabrice Bellard, as you mentioned, creating the concept, and then probably in the 2000 era, I would call the era, eras tour of FFmpeg is the 2000 era was Michael Niedermayer. So key things he got done was exhaustive support for DivX and Xvid at the time, and all sorts of weird variants of what's known as MPEG-4 Part 2. So this predates the MPEG-4 Part 10 that we're used to. So this was 2000 era video codecs where there were, oh, flavor after flavor of weird, weird decoders.

Lex Fridman 1:41:26 Mm.

Kieran Kunhya 1:41:27 At the time in the 2000s, you needed a new player to play every different type of file format. So there was Windows Media Player to play Windows Media formats. There was RealPlayer to play RealMedia formats, and those were the other, the other key thing in FFmpeg at the time were native decoders for those. I actually do remember being a teenager, I must have been, figuring out there was this one player that could play, could decode these files without having separate bloated players. 'Cause at the time when you downloaded RealPlayer, there was a ton of other stuff in there, a ton of ads, a ton of other things, and just having a simple library that was fast led to that. And then I think 2008 was a, 2008 onwards was a big change because that's when H264 got its maturity and, and I think something hopefully we'll talk about a bit more. This was the beginning of high definition video. So H264 was the key decoder of that. So I'd call that the late 2000s and 2010s, and that's when the big reverse engineers came along and really did astonishing work. The beginning was a single player that could play Xvid, DivX, Windows Media, and RealPlayer was already in a massive achievement in itself without codec packs, without weird stuff you had to download that had weird ads and weird spyware.

Jean-Baptiste Kempf 1:42:43 VLC 1.0 was out on those times, 2000, 2009, 2010.

Kieran Kunhya 1:42:49 Yeah.

Jean-Baptiste Kempf 1:42:49 And this is, like, where it exploded.

Lex Fridman 1:42:53 Yeah, without codec packs, it just works-

Kieran Kunhya 1:42:55 Yeah

Lex Fridman 1:42:55 ... across all these different-

Jean-Baptiste Kempf 1:42:57 It, the fact though, it's just like all the codecs packs are FFmpeg inside VLC, plus we have other modules for all the type of codecs.

Kieran Kunhya 1:43:04 But at, at the time that wasn't, is there were weird... In the 2000s, there were weird codec packs with DLLs coming from this place, DLLs coming that-

Jean-Baptiste Kempf 1:43:11 With a lot of spyware

Kieran Kunhya 1:43:11 ... with spyware, with you know what. You, it wasn't reliable, you didn't know, and having a single player that was open source or single playback module/player that could do this, that was open source. But I think the thing to emphasize is this task in the 2000s that Michael did was, was Sisyphus, and it was really the number of edge cases are, were beyond comprehension in terms of you could have a Chinese CCTV system that did one weird variant of MPEG-4 Part 2, mem- what's known as MPEG-4 ASP, and that was a weird variant, and you had to fix that without breaking everybody else-

Jean-Baptiste Kempf 1:43:44 Mm-hmm

Kieran Kunhya 1:43:44 ... times a million.

Lex Fridman 1:43:45 So that's, uh,

Lex Fridman 1:43:46 so you said that- that's where a lot of the reverse engineering was happening.

Kieran Kunhya 1:43:49 It start, it started in the 2000s with the Windows Media stuff-

Lex Fridman 1:43:52 Yeah

Kieran Kunhya 1:43:52 ... 'cause that was proprietary. It started with the RealMedia, started with Benjamin Larsson.

Jean-Baptiste Kempf 1:43:56 Kostya Shishkov.

Kieran Kunhya 1:43:57 Kostya Shishkov, that era. Those were the key, that was the key groundwork.And then in, in the 2010s was kind of the Paul Maheral, Kostya era building, doing, doing some of the most difficult codecs. Um, JB maybe can talk about GoToMeeting 4 and GoToMeeting 5, and-

Lex Fridman 1:44:13 What? What's the, what's the GoToMeeting?

Jean-Baptiste Kempf 1:44:15 So, so, so, so, so, like, let- let's talk about this amazing Ukrainian guy called Kostya, who was, uh, at that time living in Germany and who was in love with Sweden, right? He-- And the guy was the most... He, he's like, like, like a lot of the people in the community are very clever. He's one of those who are like borderline geniuses, right?

Lex Fridman 1:44:40 Mm-hmm.

Jean-Baptiste Kempf 1:44:41 He was able to reverse engineering extremely complex codecs. Um, and he, he does that, and we do a to- a bit of engineers with Kieran, but clearly not at this level. [chuckles]

Lex Fridman 1:44:50 No, no, no.

Jean-Baptiste Kempf 1:44:51 Um, he reverse engineered binary blobs, which are 20 megabytes?

Kieran Kunhya 1:44:56 Yeah, so just for reference, uh, um, one megabyte binary blob to reverse engineer is probably order of magnitude a month of work, and this guy is doing 20, 30 megabyte blobs. Maybe we'll talk about that in a minute, about the subtleties of how you do that. But this guy is doing it for very difficult and very obscure codecs.

Jean-Baptiste Kempf 1:45:13 And, and did that for fun, right?

Kieran Kunhya 1:45:15 Yeah.

Jean-Baptiste Kempf 1:45:15 And, and, um, so GoToMeeting, uh, was a big problem with VLC because that was like the number f- number one, uh, feature request for a long time, so I put a bounty. A-and the guy at some point said, "Okay, JB, I'm going to do it." And in a matter of two months, and then he explained how he did it. He was just like, "Oh, I looked at the code, like this looked like a, a, a DCT that I used to see on WMV and so on." He did that, and the funniest part is that the code he's wr-written is a ton of jokes. And there is, there is a ton of JB, right, my name, and, and Kempf, um, and Kempf and Kostya, uh, jokes inside the code. The code is beautiful, right?

Lex Fridman 1:45:57 So one of the things I, I, I wanna comment is, uh, I've gotten a chance to speak to some of the developers, some of the as- assembly language level, uh, uh, people, and they all always make everything sound like it's kinda easy. There's a, a kind of humility because, I mean, maybe [chuckles] just the level of what's required to do this stuff is so high that everything else seems easy, I guess is the lesson to take away from that.

Jean-Baptiste Kempf 1:46:23 So in the community, like some of the most impressive people are the ones doing reverse engineering-

Lex Fridman 1:46:28 Yeah

Jean-Baptiste Kempf 1:46:28 ... and the other ones doing the assembly folds, right?

Lex Fridman 1:46:30 Mm-hmm.

Jean-Baptiste Kempf 1:46:31 Um, and both of those type of people are amazing. Um, x264, for example, became amazing because, uh, of a, a guy called, uh, Lorraine Merritt-

Kieran Kunhya 1:46:42 Yeah

Jean-Baptiste Kempf 1:46:42 ... who is, was from University of Washington, I think.

Kieran Kunhya 1:46:44 At the time, yeah.

Jean-Baptiste Kempf 1:46:45 And who was like, who made everything great and fast doing a ton of assembly. Um, uh, yeah. So this is like the, the, the, the golden era, I guess, where so many things got done.

Kieran Kunhya 1:46:57 Yeah, if you look at Kostya, for example, he looked at the world as a binary specification. He, he didn't need documentation or anything. It's, "I have a binary and I can figure all of this out." And, and he, he regularly used the phrase binary specification. Ah, you know, it's not a problem. And he went and he would go away, and he would come back, and he, he would do interesting stuff.

Lex Fridman 1:47:15 Can you, uh, actually speak to the details or any, any add color and texture to what it takes to reverse engineer a blob?

Kieran Kunhya 1:47:22 Yeah. So, so let's look at GoToMeeting, for example, is a good one because, um, I record a meeting, uh, on GoToMeeting, for example. How do I play it back without needing this GoToMeeting player? There may not even be a player. I may, I may need to send a recording of a meeting to someone that doesn't have a player or whatever. So fi-first of all, there's a ton of other stuff there. There's a, a actual video conferencing client. You need to go and find... It may be easy, it may not be easy to find the actual module doing the decompression. You need a way to actually dump the YUV data from the module. So often it rever- it involves opening in a disassembler, trying to guess where the hooks are to incorporate that module and run that module natively to decode a sample file. So figure out where this module is doing the decoding process and, and find a way to hook in and output the raw YUV data, 'cause you will need that-

Lex Fridman 1:48:16 Mm-hmm

Kieran Kunhya 1:48:16 ... as a point of comparison for when you actually do the reverse engineering, 'cause you'll need to be bit exact or in some cases close to bit exact. And then you, you open up your disassembler, use a lot of intuition to go and figure out, you know, where the DCT is, where's entropy coding. There, there is a kind of, not a rule book, but there's always a pattern of some sort. Um, for example, GoToMeeting, you know it will be a s- a lot of screen codec tools. There's also different variants. So often I think there's what's GoToMeeting 4, 5-

Jean-Baptiste Kempf 1:48:46 Well, 2, 3, 4, I think.

Kieran Kunhya 1:48:47 2, 3, 4.

Jean-Baptiste Kempf 1:48:48 Yeah.

Lex Fridman 1:48:49 So as you mentioned here, going to Perplexity, GoToMeeting uses its own proprietary codec for older s- recorded sessions historically stored in WMV files that require a special decoder to play properly on Windows. Without this decoder installed, Windows Media Player and some editors cannot decode the video track, so you may only hear audio or see a black screen. Boy, do I remember that. But this is reverse engineering that.

Jean-Baptiste Kempf 1:49:16 This is key, right? Because the GoToMeeting is something that not many people know anymore, right? Well, you know about Zoom and, and Teams and so on, but like now let's fast-forward 10 years, 15 years, and like this is a Gotomeeting.exe for Windows 32 bits, right? Which is like, oh yeah, but I'm on Android, I'm on an iPad, I'm somewhere else, right? How are you going to do that? I'm going to be on RISC-V, on Arm. Those are blocked, but there are tons of files we need to f- support for the future. And this is why those type of work are in- exceptionally useful for humanity.

Lex Fridman 1:49:51 I just have to say though, that reverse engineering process is mind-blowing.

Kieran Kunhya 1:49:55 Yep.

Lex Fridman 1:49:56 It's crazy. It's like, it's a kinda like, you know, I've been reading a lot in the interview archeologists. I mean, you, you just have so little signal. Yes, yes, you know over time you get so much experience, you understand the structure of the original code, so you can kinda start inferring basics.

Kieran Kunhya 1:50:14 Yeah.

Lex Fridman 1:50:14 But you're like

Speaker 1:50:16 And we, like, like, like archeologists with a little brush trying to reconstruct the entire human civilization

Kieran Kunhya 1:50:22 Kieran is too humble, but, uh, Kieran has done some reverse engineering also

Speaker 1:50:24 Of, of CineForm, yeah, at the time, um-

Kieran Kunhya 1:50:26 CineForm, nice

Speaker 1:50:27 ... yeah, at the time before, um, actually led to the open sourcing of that work. Um, so in parallel to doing the binary side, you obviously have samples. In many cases, you don't have many samples, um, so you have to figure out what all the different flavors are, and you may have a s- So CineForm, for example, is actually a collection of different approaches and toolkits within that codec 'cause often it, it grows naturally. And, and the hard part is finding the sample that gets you kind of somewhere to start without having to implement 10 different other things. So start there. I think thankfully at the time I found a sample by pure chance that had a lot of flat blocks. It was animation, so that really helped a lot because it wasn't using particularly complex coding tools, et cetera, and you could kind of get somewhere and then, and then build up and build up until you figure, "Hey, here's a few bits here. I missed this. I missed this, this if branch that it does," and go, "Oh." So when we say samples, you mean sample videos-

Kieran Kunhya 1:51:19 Yes

Speaker 1:51:20 ... and then, and then you're tracking, trying to infer, like, what is this codec doing-

Kieran Kunhya 1:51:24 Yes

Speaker 1:51:25 ... by observing the sample and then looking at what, at the l- at, at the machine l-

Kieran Kunhya 1:51:30 The machine code saying-

Speaker 1:51:31 At the machine code

Kieran Kunhya 1:51:32 ... "Ah, I have byte, this byte is six. Take this branch." And in a different sample, oh, it's-

Speaker 1:51:36 That's nuts, man

Kieran Kunhya 1:51:37 And, and, and, and, and-

Speaker 1:51:38 That is nuts

Kieran Kunhya 1:51:39 ... so you see, this is nuts. Then you go to things like GoToMeeting.

Speaker 1:51:42 Yeah, yeah.

Kieran Kunhya 1:51:42 It's like-

Speaker 1:51:43 Mine was easy, right? [laughs]

Kieran Kunhya 1:51:44 Ima- imagine-

Speaker 1:51:45 Yeah, right

Kieran Kunhya 1:51:45 ... two order of magnitude of more complexity. A guy alone somewhere in Germany doing that.

Speaker 1:51:53 Yeah.

Kieran Kunhya 1:51:53 A- and for a long time, you work, you're in a black box because a decoder, for a long time, because there is so many step from the entropy decoding, the intra prediction, the motion prediction, the IDCT, and so on. For a long time, you don't see anything, right? So you're debugging purely in memory.

Speaker 1:52:10 Debugging guesswork. And you may have the buffer that the coefficients are stored in completely wrong, and so you may be going down a complete rabbit hole thinking it's this, and then, oh damn, that's not, that's, that's something else, and-

Kieran Kunhya 1:52:21 And you're doing that on binaries that are tens of megabytes, millions of instructions, right?

Speaker 1:52:27 So you're, you're stepping through the debugger, like one by one, you know, instruction by instruction going, "Hey, this instruction changes this. This does this."

Kieran Kunhya 1:52:34 Pausing the program on the CPU level. Like it's-

Speaker 1:52:36 Pausing it, yeah, on the CPU level, watching what's going on, trying to figure out-

Kieran Kunhya 1:52:39 S- sometimes you need to, like, be in a VM, so yeah, that you can pause the VM.

Speaker 1:52:43 Yeah, pause the VM, dump the memory, 'cause there could, some of the codecs could have encryption. There could be, um, like a DRM on there. So you need to dump the memory from a virtual machine.

Kieran Kunhya 1:52:52 Like when I joined Décodage Central Paris in 2003, Jon Lech Johansen basically broke the DVD specification and created DCSS, showed us how he was breaking a DRM, which was MP4 FairPlay from Apple. What he did on his laptop, and I was young, I was 21, was just like mind-blowing because he was basically debugging Windows i- inside a type of VM with ex- Like, wow.

Speaker 1:53:17 It's incredible. It's mind-blowing and inspiring. Does it get, like from your experience and from what you've seen in the community, does it get discouraging? Does it get-

Kieran Kunhya 1:53:26 People help you. People send you samples. People are keen. Sometimes you don't have access to an encoder, so this is even more difficult because you just, you just ask and you have to ask for samples. I remember VideoLAN used to tweet for samples at one stage. "Hey, I need this obscure sample," and-

Speaker 1:53:42 For a long time it was, "Oh, I need this codec and I need this codec."

Kieran Kunhya 1:53:45 And, and, and if you were really lucky, you would find like... If you were un- if you were unlucky, you'd get like one or t- you'd get nothing or you'd get one or two, and then they would... Sometimes you'd find a goldmine. It's like, "Yeah, my company has 100,000 of these files 'cause we're dependent on it for some reas-" Um, and so those are the, those are kind of the best if, if... Because then they can test bit exactness across the huge range of coding tools.

Speaker 1:54:06 Can you explain bit exactness?

Kieran Kunhya 1:54:07 Bit exactness, so, um, most but not all video codecs, certainly from about the 2000s onwards, have a bit exact definition, so every implementation must produce exactly the same bits, bit for bit, in exactly the same data that comes out of a decoder.

Speaker 1:54:26 For like a large number of samples?

Kieran Kunhya 1:54:28 For a given sample. So Lex's implementation, JB's implementation, and my implementation of H.264 must match bit exactly. That wasn't the case in the '90s of MPEG-2. Probably fair to say one of the biggest mistakes the video industry made, and I think people who were in the room in '92, I don't think, most, both of us were in diapers, I suspect-

Speaker 1:54:47 [laughs]

Kieran Kunhya 1:54:47 ... but, um, have acknowledged, um, I, I would give a shout-out to Yuri Resnik. He's acknowledged that was one of the big mistakes of the era.

Speaker 1:54:56 And you're saying the encoders needed to be able to run tests and then the, the, the bit exactness. I mean, that's a nice thing to guarantee.

Kieran Kunhya 1:55:04 Yeah.

Speaker 1:55:04 Like there's a parallel sort of development here on, uh, the way the, the web browser works, which is, you know, takes HTML and displays it, and there's no bit exactness there across the different engine.

Kieran Kunhya 1:55:15 I would, I would point out actually FFmpeg is unique in the sense that it's, it has been a winner-takes-all scenario. You have... Brow- browsers is a good analogy because it has to parse a lot of different content and render it in a particular way, like a decoder. But there still are multiple browser engines. There's Firefox's one, there's Chrome's one, there's a few Japanese ones that are pretty decent. That's not been the case in multimedia in general, um, across a wide range of codecs. FFmpeg has kind of won it all, I suppose, in a sense, because of, because of the fact that you can get every new codec added is actually worth more than the value of that codec itself because it makes the whole thing better.

Speaker 1:55:52 Man, this is really cool. Uh, going to Perplexity. Yuri Resnik is a multimedia and signal processing researcher, got his PhD in computer science from Kyiv University with over 150 papers and more than 80 granted US patents, contributor to major multimedia standards including H.264, MPEG-4, AVC-H265, uh, MPEG-4 ALS, G.718, and-

Kieran Kunhya 1:56:17 G.71 is telco stuff. Telco.

Speaker 1:56:19 Oh. And so he was more connected to company-

Kieran Kunhya 1:56:21 Real audio, real video, right?

Speaker 1:56:22 Yeah.

Kieran Kunhya 1:56:22 That was-

Speaker 1:56:23 Oh, yeah

Kieran Kunhya 1:56:24 ... very important at that time

Speaker 1:56:25 ... Xencoder, Brightcove contacts. This, man, I need to hang out with Yuri. He's legit.

Kieran Kunhya 1:56:30 Yeah. And he's like one of the most nice person-

Speaker 1:56:33 Slack guy, yeah

Kieran Kunhya 1:56:33 ... ever, right? Like, uh, like, um, for example, for, for my n- for my startup, uh, that I'm doing right now called Kyber, right? I met Yuri, uh, because I met him every year at the Mile High Video, uh, Conference, which is in Denver. Um, and he w- he gave me, like, so much good ideas and, and good things. He's, like, really amazing person.

Speaker 1:56:52 He tell, he tells us how, how, you know-

Kieran Kunhya 1:56:55 How great it is to be, you know, even know us. And then we just like, you know, you look at that and it's, I think it's the, I think it's the other way around, Yuri

Lex Fridman 1:57:01 That reminds me of a thing that you, uh, mentioned to me about FATE testing and, like, the insanely rigorous process that's used to test everything that's incorporated into, um, into FFmpeg. Can you take me through the testing process?

Kieran Kunhya 1:57:14 Yeah. So FFmpeg has a system called FATE, FFmpeg Automated Testing Environment.

Lex Fridman 1:57:18 Mm-hmm.

Kieran Kunhya 1:57:19 Because FFmpeg runs on so many different OSs and can be compiled with so many different compilers, there's been a crazy number of configurations. So you can see the absurd combination of, um, compiler variants, operating system variants, instruction sets. You can see at the top macOS has tons of different variants because it has iOS, it has tvOS.

Lex Fridman 1:57:42 Well, I'm looking at a page fate.ffmpeg.org 81 minutes ago, 76 minutes ago, looking at the different architectures, the operating systems, the different compilers, Apple Clang version, uh-

Kieran Kunhya 1:57:57 Combinations are crazy

Lex Fridman 1:57:58 ... the combination is insane. RISC-

Kieran Kunhya 1:58:00 So these are all run by volunteers, so these are all volunteer systems. The, the ones at the top, for example, the Macs I host in my office, for example, uh, host all sorts of different stuff. Other people host other things. So it's really there to make sure, um, c- because FFmpeg does quite complex C code, for example, you do have miscompilations. So the compiler will sometimes compile C code incorrectly. For example, this happens once in a while.

Lex Fridman 1:58:25 Oh, there's like, there's a log of all the compilations.

Kieran Kunhya 1:58:28 Yeah, log of all the compilations, all the tests. I think one of the other ones will show all the tests passing.

Jean-Baptiste Kempf 1:58:33 If you click, you can see all the tests-

Kieran Kunhya 1:58:35 Yep

Jean-Baptiste Kempf 1:58:35 ... back. All tests successful.

Kieran Kunhya 1:58:37 In logs test, yeah. So you see all those tests are passing of all the different codecs, all the different filter transformations, all the-

Lex Fridman 1:58:47 [laughs]

Kieran Kunhya 1:58:47 The, the level of scale is, is quite crazy.

Lex Fridman 1:58:50 Oh, that's nuts.

Kieran Kunhya 1:58:50 On all the combinations. It's, it's not just a matrix at this point. It's like a pivot table of different combinations.

Lex Fridman 1:58:56 That's nuts.

Kieran Kunhya 1:58:57 And it, it's a key part of what we do because you, you may be able to test something locally, you make a change, but actually that breaks GCC version 11 on Mac or something like that, and you're able to then fix that. We also have miscompilation, so the C code, sometimes the compiler can have a bug in it where it creates the wrong output, and that can have quite a big effect sometimes on a vid- on a video because of the way frames have dependencies. Even a small change in the output can, can cascade to actually quite big glitches.

Jean-Baptiste Kempf 1:59:27 You see PowerPC, you see RISC, you see ARM.

Kieran Kunhya 1:59:31 There was PowerPC, there was RISC, there was weird stuff in the past like DEC Alpha. There was-

Jean-Baptiste Kempf 1:59:35 You see Visual Studio, different versions of Clang or GCC

Kieran Kunhya 1:59:37 ... Visual Studio, Intel compiler, Apple Clang, you name it.

Lex Fridman 1:59:40 What are some of the pain points? Like maybe do you have emotional triggers, uh, maybe nightmares about a particular operating system, a particular container, codec combination of-

Kieran Kunhya 1:59:53 I mean, for, for me it's really easy because, uh... So I, I have a day job. Um, my company builds, the company I started builds, um, equipment for broadcasting sports matches between TV stadiums and studios, for example. We have to work with 10-bit video, and 10-bit video has, um, a set of challenges that you can't process 10-bit data natively on a CPU. So that means you have to stick it in, um, 16 bits. So that means you have six wasted bits. So there's different packing formats to actually pack the data more efficiently because when you send that over a network, you lose... 'Cause you need to, you need to save that 40%. For example, on PCI Express, you may only have bus bandwidth to do that. And so I think internally we have about... Some, some are, some are industry ones and some are internal to our own hardware that we build. We have a, I think a five by five or six by six matrix of every single format to every single other format conversion. In fact, one, one of them I sent you, and they're all written in handwritten assembly, and they're all written, and they all support different CPU generations. So this is really traumatic, handling all these different combinations times a million.

Lex Fridman 2:01:02 By the way, the company you're talking about is Open Broadcast Systems.

Kieran Kunhya 2:01:04 Yeah, so no, no relation to the free OBS streaming service.

Lex Fridman 2:01:07 Yeah.

Lex Fridman 2:01:08 Yeah.

Kieran Kunhya 2:01:08 Um, but JB and I have started companies broadly speaking around the FFmpeg VLC ethos, so that's really low-level work. So, so in most companies, this, this wouldn't be written in assembly. C- It would be accepted that C's, C is fast. Um, as you can see from that, C is not fast. Um-

Lex Fridman 2:01:26 Uh, so here it says 62 times faster than C.

Kieran Kunhya 2:01:31 Yeah. So it's taking those, the ethos of doing low-level programming, real-time programming, and take- using that for commercial applications, and JB and I have started companies around that, in many cases hiring developers from the open source community to use that ethos. And, and so that, that's a great example of, of some of the things we're doing. In most companies it would be, say, "Oh, I'll write this in C and it's fast and we're done." But actually you can get a lot better.

Jean-Baptiste Kempf 2:01:59 For me, like some of the headaches we have is around some OS that are difficult to support, right? Because, um, uh, if you look at VLC and thanks to FATE and FFmpeg, we run on, the last version of VLC runs on Windows XP and still run there and runs on Windows 11. Um, we work on macOS 10.7 to the la-latest macOS, whatever it is, right, uh, 26. Um, we work on iOS since iOS 9, well, we are actually iOS 26, right? We support, um, we support, uh, many types of Linuxes, BSD, Solaris. The last version still runs on OS2, right? Like there is maybe 10 users of OS2 in the world, and one of them is maintaining VLC. Then you realize that this very small team around VLC and using FFmpeg codecs and, and all the other ones support more OSs thanMicrosoft or Google or Apple, and they have infinite amount of, uh, of, uh, power and resources. But for example, the worst is iOS. Um, for, in order to build on iOS 9, we need to do some very clever mixing of several version of the Xcode IDE and SDK from Apple, from several version, and do a type of Frankenstein version of that so that we can still support iOS 9, which is not supported at all by the, the compiler of, uh, uh, of, uh, Apple in order to still run on Arm32, um, on iOS 9. And you've seen on Fate that it was still supporting iOS 9, right? So, so, um, my headaches are mostly related to the, the support of so many OSs. And it's important because, like, we receive so many people saying, "Hey, thank you, I still have my iPad 2 to watch movies," and it still works on iOS 9, right? And it's also an impact of, like, not forcing people to buy new hardware when it works fine, if you optimize it correctly, which brings us to what we were saying about Assembly. It's also fighting, like, the fact that you need to buy something new nonstop, uh, while you could optimize more, which is a lost art.

Lex Fridman 2:04:18 You gotta tell me about this lost art or this, uh-

Jean-Baptiste Kempf 2:04:23 Yeah

Lex Fridman 2:04:23 ... the, the carriers of the flame of Assembly. What, uh, what is, what, what is Assembly? Why is it beautiful?

Jean-Baptiste Kempf 2:04:31 Yeah.

Lex Fridman 2:04:31 Why is it challenging? How does it work?

Kieran Kunhya 2:04:34 So when you write Assembly code, you write this using the instructions the actual processor is using directly. So most of the time you would write in a language, let's take C as a good example. The compiler would use that to create, um, Assembly language and machine code instructions for you based off your C code. And there's a specific flavor of Assembly that we use in FFmpeg that's called SIMD, SIMD, single instruction, multiple data. So this means, for example, say, say I want to add five to a number in scalar Assembly, so this is what's known as you work on an individual element. So I wanna have a number of-- I have the number 10 and I want to add five. I use the add instruction, and I add five to 10, and I get 15. With SIMD, with SIMD, I can have a whole vector of 16 different numbers. They could all be different. If I want to add five to that, I can, I can run one instruction, and that one instruction sums all 16 elements. And that, as you can imagine, lends itself very well to video. Video, uh, video is, you know, pixel grid, so I can perform operations on multiple pixels at the same time. The key thing that we do differently in FFmpeg is we don't use any abstractions or any major abstractions on top of that. So there's a part of the world that uses what's known as intrinsics. So these are C functions that behave very similarly but not quite the same to writing Assembly by hand. So the registers that data i- uh, is stored in on the CPU, the compiler allocates those for you. And so the key thing to understand was when we write SIMD is we have a 10x, and not percentage, 10x to 50x speed improvement. That, that function is 62x, um-

Lex Fridman 2:06:21 That's nuts

Kieran Kunhya 2:06:22 ... on the FFmpeg account, as you know, posts and tweets a lot about that to try and say, "Hey, we are doing this stuff."

Lex Fridman 2:06:29 You are a person who sees the beauty in Assembly, but it's also extremely useful for these kinds of application to actually-

Kieran Kunhya 2:06:35 Yes

Lex Fridman 2:06:35 ... significantly outperform even C, which is crazy.

Jean-Baptiste Kempf 2:06:40 It is necessary.

Kieran Kunhya 2:06:41 Yes.

Jean-Baptiste Kempf 2:06:42 Right? Because, like, one of the projects, um, that we need to talk about is called David, right? So David is a decoder for the format, uh, that was done by Alliance for OpenMedia, uh, which is an, a video decoder called AV1.

Lex Fridman 2:06:58 So it-- for people who don't know, we've been talking about H.264. AV1 is another hugely popular standard and codec that is increasingly taking over, uh, the internet.

Jean-Baptiste Kempf 2:07:12 And when this format was launched, uh, many people said, especially even from the Alliance for OpenMedia, right, which is Google, Netflix, Amazon, uh, Mozilla, say, "Well, this format is so complex, it must be done in hardware to do decoding," right? And well, I arrived, uh, with a few other people, uh, mostly, um, Ronald, uh, Henrik, and Martin, and we said, "We need to have an extremely good software decoder, uh, because it's going to take time to have hardware." And so we wrote this project, which is beyond insane. Um, we are talking about 30,000 line of C, but 240,000 lines of handwritten Assembly, right?

Lex Fridman 2:07:56 Handwritten Assembly, 240,000 lines. That's incredible. That mean-- I mean, some of the stuff we're talking about is probably the biggest Assembly, uh, code bases.

Jean-Baptiste Kempf 2:08:09 To give you an idea, and Kieran can correct me, but I think the FFmpeg has 100,000 lines of Assembly for all the codecs.

Kieran Kunhya 2:08:17 For all codecs. Mm-hmm.

Jean-Baptiste Kempf 2:08:17 And just this one has 240,000. Um, it's a VideoLAN project, of course. Um, and it is optimized at the maximum because the motto when we're starting the project is every cycle matters, right? Every cycle matters because David is used in VLC and in some software AV1 playback stacks. We are talking about probably three billion devices which are going to decode video nonstop because, for example, 30% of the video from Netflix are now in AV1, 50% of YouTube, right? So, and you often don't have a hardware decoder because not many devices have a hardware decoder. And with David, we realized that with one or two cores, you were able to decode 720p correctly. So it is, like, literally-

Lex Fridman 2:09:07 [laughs] Yeah, that's David

Jean-Baptiste Kempf 2:09:08 ... incredible, right?

Kieran Kunhya 2:09:09 That's David. Look at that Lex.

Lex Fridman 2:09:10 Yeah. So this is, uh, another spicy tweet from you. "This is what peak video codecs should look like."

Kieran Kunhya 2:09:18 79.9% assembly, 19.6% C, and, uh, 0.5% other

Jean-Baptiste Kempf 2:09:25 And, and what's incredible is with those tweets, which is factual, people get crazy. They are unhappy, right? They say-

Kieran Kunhya 2:09:33 For a year-- For the last two years, they go crazy. "No, intrinsics is fine. The compiler is..." Oh, they go, "I am never-"

Jean-Baptiste Kempf 2:09:37 "You can optimize your compiler. Auto-vectorization, it's your fault. You don't understand." And we've tried that forever, right?

Kieran Kunhya 2:09:44 For two years, and two years later, showing hundreds of examples of, of handwritten assembly. "No, no, no, you're doing it wrong. The compiler can do this." So w-we should actually just articulate a little clearer. So the intuition there from the software engineering folks, when you have code like, okay, let's just take an example, C++. There's a compiler that's doing a lot of the optimization.

Jean-Baptiste Kempf 2:10:05 Yes.

Kieran Kunhya 2:10:06 And the presumption is if you have a good enough compiler, if you continue to improve the compiler, you're going to generate code-

Jean-Baptiste Kempf 2:10:13 Yeah

Kieran Kunhya 2:10:13 ... that can p-perform like optimal performance. You cannot possibly beat it.

Jean-Baptiste Kempf 2:10:18 Yeah. Yes.

Kieran Kunhya 2:10:18 And you're consistently challenging that thought, that if you do-

Jean-Baptiste Kempf 2:10:21 By orders of magnitude.

Kieran Kunhya 2:10:22 By orders of magnitude-

Jean-Baptiste Kempf 2:10:23 Yeah

Kieran Kunhya 2:10:23 ... handcrafted assembly can outperform C.

Jean-Baptiste Kempf 2:10:27 The two things that they tell us is, "Yeah, but modern compilers have auto-vectorization," right? Because SIMD that we're doing is vectorization, and like it's not even close, right? It's not even close, right? It's not like 5%, 10% slower. It's multiple times slower.

Kieran Kunhya 2:10:43 So can we-- I don't know if you can say something philosophically, 'cause there's a lot of, there's a lot of great software engineers, great engineers, great machine learning people. Karpathy will listen to this and say, "What's the intuition he's supposed to get from this? What are we supposed to..."

Jean-Baptiste Kempf 2:10:57 Karpathy learned assembly because of the tweets, by the way. I just... He start, he, he went, he's like, "Oh, I think this is a movement."

Kieran Kunhya 2:11:01 He's like, "Let me figure out what's happening here."

Jean-Baptiste Kempf 2:11:03 No, no, he, and, and you know the way he documents his work, and so. Philosophically, what's important to realize is that we passed the time where hardware was going so much faster, right? We're at the end of the Moore law. We have limitation for, uh, for AI, for memory. You need to go down in the stack and optimize more to get more power from what you have because our request for power, CPU power, GPU power are exploding while the hardware is not exploding in speed, right?

Kieran Kunhya 2:11:34 Mm-hmm. Yeah.

Jean-Baptiste Kempf 2:11:35 So you, what people do is that they add more cores, right? But that's basically like at some point you can have a 250 cores, right? So what we do is to take every inch of the machine-

Kieran Kunhya 2:11:46 Not just that, not just that. We, we abuse the machine. We, we, we go and use, we use the machine in ways that the, that the creator didn't expect. Sometimes we use an instruction that's completely unrelated to what we do. We use a cryptography instruction in video processing to do nothing related.

Jean-Baptiste Kempf 2:12:01 And one of other things that we do, for example, uh, in David, which is a bit crazy, is that we don't use the function calling convention from the operating system.

Kieran Kunhya 2:12:13 We should explain that.

Jean-Baptiste Kempf 2:12:14 That is extremely-

Kieran Kunhya 2:12:15 Yeah

Jean-Baptiste Kempf 2:12:15 ... complex, but basically, usually when you do move from one function in, in code to another, there is a way to save the registry, the state of the CPU to enter another function. Um, and this is like standard.

Kieran Kunhya 2:12:28 It's a bit complex. I would, I would simplify this a bit. So, so David does things to abuse the calling convention. You could define the calling convention as I've written a function, and I want to call another function. How is the data shared between the functions? Because there, there's a convention, what's known as a calling convention, and what David does for optimal reasons is create its own calling convention sometimes. So if I wanna call Lex Fridman's library, we got, we've got to agree on a convention so that I can share data with you in the assembly language space. And one of the challenges in assembly is, uh, every operating s- well, not every operating system, but there are, well, at least four that I can think of on X86. Li-Linux 32-bit, Windows 32-bit, Windows 64, Linux 64. They all have their own calling conventions, and so one of the amazing things Loren Merritt did, who we talked about before, was create a very lightweight abstraction layer, so you could write your assembly code once, and it handled all the calling convention stuff for you. Which was always a problem because you had to manage four different variants. But David takes this even further that for speed reasons, it does its own calling convention, um, within itself to bypass the kind of rules, the rules of, the rules of sort of, of functions and say, "Okay, actually, I'm gonna call a function this way because I know it's within my library." Does it have to be special to every single operating system? Well, if it's custom, no, but the, the, the challenge is in general, yes, and in terms of, in terms of each instruction set. So the thing to also emphasize is we do this on every instruction set. So every instruction set has its own handwritten assembly, which is even more crazy, and that, that, that has, um, that matrix has got bigger in recent years because of RISC-V, because of ARM64, because of the new, uh, SVE. There's SME. X86 has AVX-512, AVX. So we do runtime processor detection. We see what the machine FFmpeg is running on or David's running on is capable of because you could be on a laptop from 2008 where this isn't there. Runtime detection, we set function pointers accordingly.

Jean-Baptiste Kempf 2:14:31 Mm-hmm.

Kieran Kunhya 2:14:31 And then from then on, off you go. Or you could be on a machine with RISC-V.

Jean-Baptiste Kempf 2:14:36 Yes, and in all that, we don't even respect the calling convention of the operating system in order to be faster because we know that we are going to be called from within our binary, so we can share data without saving all the registry in the common way because that can w-lead to loading and saving registry on the L1 and L2 CPU and gets us faster. So that's why I said that understanding CPU architecture, computer architecture is key, and this is also why it's handwritten. I don't know anyone. I've never heard any other project than David doing that. This is what Kieran calls it, calls it an art, right? It is an art.

Kieran Kunhya 2:15:14 I think, I think in a, in a mass world, there isn't something on billions of devices. I, I know there are some specialist industries. I know in high-frequency trading they take this really seriously, where they're receiving feeds from a market, and they need to react within X number of microseconds, and so the instructions matter. But that's not a mass, you know, a mass-produced thing that's on a billion devices that's hyper-specialized, running on hyper-specialized hardware. We're running on all hardware from- Sorry to linger on it, but like th-that's a really counterintuitive, almost like revolutionar

Lex Fridman 2:15:45 The idea here that there's a huge amount of value to assembly. Like what are we supposed to take away from that? Like what-- You know, there's a bunch of people listening to this, they're basically like, sorry, f-for myself included, you know, I, I programmed for many, many years in C, C++ going up the standards of C++, fell in love with C++ even in the meta program and so on, and then transitioned more and more because of machine learning about 15 years ago to Python.

Jean-Baptiste Kempf 2:16:12 Yeah.

Lex Fridman 2:16:12 And so like for me in this Python world, JavaScript world, now vibe coding where I'm just using natural language sitting in my jacuzzi drinking a drink and just talking to the computer, re- like, like [chuckles] record stops, why is the value we to go back all the way down to the low level? Like what's the intuition?

Jean-Baptiste Kempf 2:16:31 Because you can get more power per dollar invested, right? And sometimes it's going to be a problem that is limited by your hardware. A, a, a good analogy is what you see in quanti- quantization in, in LLMs, right? And people are doing, "Oh, I'm going to do that in FP8 or FP, FP4," or some c- some crazy things like Microsoft Phi who did in 1.5, because you're constrained by memory, because you're constrained by the machine you can run. Because at some point, uh, we are doing real time, and I believe this is going to happen on AI inference also, is that at some point you need to ge-get faster, and you cannot always get harder, uh, more powerful hardware, right? So you need to analyze code and see where, like where is the mission critical, where is the things that are called nonstops. And for example, David is a good example. It's going to be run billions of hours per day.

Lex Fridman 2:17:28 Yeah.

Jean-Baptiste Kempf 2:17:28 That makes sense. It doesn't make sense to be on the glue of FFmpeg-

Lex Fridman 2:17:33 No

Jean-Baptiste Kempf 2:17:33 ... uh, CLI. It makes sense over there.

Lex Fridman 2:17:35 Yeah, and this has to do, also we'll talk about it more, but, uh, your, uh, new effort, your new company, Kyber, is doing that kind of thing for ultra low latency, so the slogan being every millisecond counts. And when you actually extremely highly constrained in some dimension-

Jean-Baptiste Kempf 2:17:53 We are also arriving at a point where we've done so many great things, but the hardware is getting back to us, right? Because cost is increasing, because we need more power, and so you're limited by either, uh, your CPU, your RAM, or your networking, and you need to optimize, and this is where value is going to be. Especially because like doing AI is going to help do the programming of like business, right? And so the core thing that you will not be able to, uh, vibe code are optimization for the hardware to be as fast as is possible.

Lex Fridman 2:18:31 I'd love to talk to you about who and how should learn Assembly, but first I think we need a bathroom break. Quick ten-second thank you to our sponsors. Check them out in the description. It really is the best way to support this podcast. Go to lexfridman.com/sponsors. And now back to the episode. All right, and we're back. There's this, uh, nice repo with the Assembly lessons. First of all, do you think, uh, developers should learn how to program in Assembly, and how would you go about learning it? What is this ASM-lessons?

Kieran Kunhya 2:19:08 So I, I personally wasn't happy with the way Assembly is taught in books and online, 'cause it's very grammar focused, and you don't in general learn a language from learning the grammar and the structure. Y-you learn a language by asking someone what their name is, and you start from there, and you go and solve real problems, uh, that you have when you want to communicate. You don't, you don't learn sentence structure, and this is the interrogative and the adverb, and all, all the Assembly books seem to be doing like that, going through every instruction, even ones that aren't really relevant, explaining what they all do and how they... It, it actually doesn't really change much. So, and the other problem that we have in our community is Assembly is taught sort of hand to hand, like person to person, like blacksmithing one by one. That's, that's the only logical sort of analogy, and that doesn't really scale online. It doesn't, um, do other things. So this, this le- this, this-- I started a set of Assembly lessons in the, in the way it's done in FFmpeg, which is a little bit different to the way Assembly in general for... I don't know. I'm trying to think the other good big use case of Assembly is in embedded devices, in really, really low power, cheap devices, and that's completely different to what we're doing here. I think it would be good if you could highlight the requirements, which are quite simple. It's high school mathematics and C. And actually not even C, really, really it's pointers. To emphasize, yes, we've talked about how brilliant this stuff is, but high schoolers like Daniel Kang have written, um, Assembly in FFmpeg. I think there's been contributions because of these lessons. So it's really about trying to s- to get this dying art to continue, because we've shown it's possible with David to produce something amazing.

Lex Fridman 2:20:46 Mm-hmm.

Kieran Kunhya 2:20:46 There's still a lot of codecs in FFmpeg that are only maybe partially Assembly, um, Assembly optimized. And so it really, it really starts with basics and continues, explains a lot of the jargon, a lot of the syntax. It doesn't really try and explain to you, you know, interrupt handlers and interrupt instructions and all of these different jump targets. It actually makes this really vector focused.

Lex Fridman 2:21:08 And, uh, describes all kinds of registers, uh, general purpose registers, vector registers. Uh, really nice examples. Oh, this is cool.

Kieran Kunhya 2:21:18 It's a classic. Yeah, it's a classic example of FFmpeg. But, um, some of this Assembly language is, is, is really beautiful and I, and I think it's beautiful because it's kind of like flying a Spitfire. It's really aviation at its purest, but also pushing the aircraft beyond what the designer thought was possible. So we're abusing, for example, sometimes cryptography instructions to do certain things, and there's a level of beauty in art where it's really you and the processor.

Lex Fridman 2:21:45 Mm-hmm.

Kieran Kunhya 2:21:46 There's, there's, there's nothing in between. It, it's you and the, the, the joystick of the cockpit and y- and you, you move that joystick and it's physically connected to the ailerons, and you can push that plane beyond what it can normally do, and there, there's a level of, yeah, beauty and am-amazingness to go that. But I don't thinkthe sort of person by person assembly that someone taught me and I've taught multiple people is gonna work long run because of the particular flavor and the way that we do it.

Lex Fridman 2:22:16 It's literally, uh, no, I should-- [chuckles] I was gonna say wizards handing it down. Um, I realize I look like a wizard wearing this hat.

Kieran Kunhya 2:22:24 [laughs] Yeah.

Lex Fridman 2:22:24 But you're basically just like the sages, the wise sages handing down the craft.

Kieran Kunhya 2:22:29 Yeah.

Lex Fridman 2:22:29 Can, can I ask about LLMs? Like-

Kieran Kunhya 2:22:31 Yeah

Lex Fridman 2:22:31 ... can they help?

Kieran Kunhya 2:22:32 They had more of an understanding than I expected, but they are still-- I've asked it questions, and it still goes and starts hallucinating. Not hallucinating, but making modifications, and then I go: "Is it bit exact?" "No." "Fix it." And then it just goes and does the same thing, and i-it's going, it... There isn't the corpus of information like Stack Overflow to work on.

Jean-Baptiste Kempf 2:22:52 There is not enough data to train on. Um, and this is the biggest issue. Um, I, I started my career actually doing some, uh, assembly for Itanium, right? So the Itanium is a dead, um, processor type, right, which was done by Intel and HP a long time ago when they wanted to do sixty-four bits. Uh, well, they lost, and then we got AMD, who did it, AMD64, which became x86-64. But Itanium was extremely interesting in the sense that, um, those were processors who had a ton of computing power to do floats, FMAs, which is similar to what we need now for, for LLMs, right? And you could pack three operations per line that could be loaded. So basically, you had an output of basically six billion of operation per second, but the bus, um, the, the memory bus only allowed one point five, right? So your, your CPU was four times faster, so you had to do crazy things to, to pack things in memory, reuse the registered, and those type of semantics, no language could do that, right? So, um, like I, I, I have the Itanium programming, uh, book because Intel did amazing books, but that's exactly what Kieran says. If you don't know what you're, you're, you're going to do, it's impossible to read, right? It's a ton of jargon and so on. While those lessons are amazing because they are targeted to a real problem, and you can do it yourself.

Kieran Kunhya 2:24:21 And people have. People have. There are patches, and they said, "Oh, I studied your lessons, and, um, here's my first changes."

Lex Fridman 2:24:26 That's amazing.

Jean-Baptiste Kempf 2:24:26 And part of that, um, in the lessons is, is a framework called x86inc, written by, uh, Lorraine when he was working on x264, and it allows you to do more things about that to create a type of like, uh, not caring too much about different calling convention. And we had a lot of students who, uh, gave code to x264 using, uh, that a long time ago, right? So it's really doable, and I believe it's, um, necessary to understand, uh, assembly language, even if you don't do it much, to understand what's going on inside your computer, and that will make you a better programmer. And I assure you that because doing that, you will understand some of the architecture of the memory inside your, your computer, right? Understanding register, L1, L2, L3, RAMs, uh, SSDs, disk, and so on, um, which are very important because then you have a good programming culture that will make you

Jean-Baptiste Kempf 2:25:26 a better programmer.

Lex Fridman 2:25:27 Uh, what do you think about the Rust programming language? 'Cause that's a bit of a meme.

Jean-Baptiste Kempf 2:25:31 We have very different opinions with Kieran.

Kieran Kunhya 2:25:33 I, I think it's valuable what they're doing in terms of memory safety as a concept.

Lex Fridman 2:25:37 Can it achieve the s- some of the speed up that assembly achieves?

Kieran Kunhya 2:25:42 Oh, not, not assembly by hand, no. I think that that's a given. C potentially, but I see it very... It has a very big Esperanto vibe, uh, about it. It's like we're gonna solve this, and we're doing this in a particular way.

Lex Fridman 2:25:54 Meaning it's a bit too utopian?

Kieran Kunhya 2:25:56 There's a lot of focus on the self-importance rather sol- than solving real world problems. It reminds me of the Sinclair C5. Sir Clive Sinclair of Sinclair Computers built a car, and he said, "Oh, everyone will be traveling around in, in, uh, one of these electric cars." And it was, um... Rust reminds me of that where I think the community doesn't, the community doesn't quite understand that in order to get people to move, you have to build something that's as good as, if not better than what you have now. Um, yes, people are doing Rust rewrites, but if they're, if they only, if they only do eighty-five, ninety percent of the feature set of what, what we need, like things like core utils, that last one percent takes ninety-nine percent of the time. Um, to, to use Elon's famous quote, "Prototypes are easy." Like this kind of stuff is easy, but this... To get a real electric car, you have to make a car as good as, if not better than what we have now, and Rust isn't in that stage yet. I think we-- I don't think anyone would object to seeing Rust code in FFmpeg, but it needs to work as well and support the same unit testing as everything else. It needs to be flawless. It, it can't just randomly break. They can't just randomly break ABI when they want to. It needs to, it needs to have, I think, more... I think it still has only one compiler implementation.

Jean-Baptiste Kempf 2:27:11 Yes.

Kieran Kunhya 2:27:12 So i-it's got to be as good as, if not better. And saying, "Hey, here's my utopia of memory safety," isn't enough, even though we all, we probably all agree that that's the goal.

Jean-Baptiste Kempf 2:27:24 So, um, I've done a ton of Rust, and the two major topics I had was adding Rust modules inside VLC. One of the reasons VLC got popular, um, and which was one of the main architectural decision, is that VLC is a very small core and a ton of modules, right? And so you can write modules in C, in C++, in Objective-C, and anything that is basically interoperable with C. And, and, and so we t- we, we did some Rust modules, and so I have experience on that, and, and I wrote some of it. And also, like my, my new startup, uh, called Kyber, is an open source project mainly done in Rust. Um, what Rust is extremely good in, in the sense that, um-It's a better C++ that cares about memory and allows you to do things about memory ownership that no one else can do so far. However, it's great when you start a new project from scratch and you do everything in Rust, but it's very not good when you interop with ex-existing part. And some part of the Rust community believes that they need to rewrite everything, and everything will be better with Rust. And, and the answer is like, no. Like, I'm almost always in all my years of being engineer, manager, CTO of startup and so on, don't rewrite, right?

Lex Fridman 2:28:47 Is that-- That's the, that's the initial instinct for a lot of people when they show up to a code base, uh, probably before LLMs, is like b- probably because they don't understand the, the, the wisdom of the way things have been done in the past. They say, "Well, we need to rewrite it." Hence why there's a thousand JavaScript frameworks.

Jean-Baptiste Kempf 2:29:06 But the reason is the following, and this is very important to understand. It is an order of magnitude easier to write code than read code.

Lex Fridman 2:29:16 Yeah.

Jean-Baptiste Kempf 2:29:17 And you see that also with LLM. They can write code, but analyzing is a lot-

Lex Fridman 2:29:22 Mm

Jean-Baptiste Kempf 2:29:22 ... more difficult. And so when you arrive a-and when you arrive to a very complex piece of code, right? You, you don't understand it, right? Um, because it's so much more effort to understand the code from someone else because you don't have the thought process. Um, a-and often I joke about, um, some, some languages, uh, mostly, uh, Perl, for example, um, which has very complex syntax. And imagine I am at my maximum intellectual efficiency in programming, right? And I write the best code ever. I will not be able to understanding myself six months later, right?

Lex Fridman 2:29:58 Mm.

Jean-Baptiste Kempf 2:29:58 Because reading code is more difficult. So very often you arrive, you don't understand all the wisdom, all the business logic, the reasons that were done that is maybe not documented. And you say, "Well, I'm going to rewrite it." And the thing is, no, you don't, right? Because that's, as Kieran said, right? I'm going to rewrite core utils in Rust, and then of course you arrive very quickly at 80%, um, then 90% takes a bit more time, and then you got the last ones, right? On the other side, right? So for new projects, it's great. Everything related to parsing files, uh, networking, uh, because of the memory, uh, checker, memory checker, it's amazing, and there is nothing else. To answer a bit differently for us, imagine I take a, a piece of software like, uh, DaVinci or x264, right? Which has a ton of runtime in assembly, right? Um, I rewrite the C part in Rust, right? So it's more secure. Yes. But then you arrive into the assembly, and you can jump anywhere in the memory because we are doing handwritten assembly. So even if I rewrite the C part in Rust, for security reason, you, you break all the security when you rewri- you write handwritten assembly because we can jump anywhere. So in my opinion, we need to do something that is, um, secure assembly, right? So which is compile time, check the assembly, which is similar to the check assembly projects that we're doing on, on DaVinci and x264 with VideoLAN, is to start instrumenting your assembly at compile time to check that it's not jumping anywhere in the memory. Because else you might rewrite a part of C in Rust, but if you want to have the same performances, you're going to have inline assembly, and so you destroy your whole security model. Um, so that's a bit what I think about Rust.

Kieran Kunhya 2:31:51 No, I, I just wanna... I, I would say on a personal level, I'm so in awe about the assembly. I actually-- Once in a... It never gets old, the speed improvements to show 62X. Uh, so there are months p- on a personal level, I, I run our internal test suite at work and just see I'm still in awe at the gains we have.

Lex Fridman 2:32:10 Well, there, there's a source of joy and happiness for programming for different reasons. But I think one of the gr-greatest happiness is, is in the optimization of code.

Kieran Kunhya 2:32:21 Yeah.

Lex Fridman 2:32:21 And it sounds like y-you're, like, at the cutting edge of that.

Kieran Kunhya 2:32:25 I was like, "Whoa, that was cool."

Jean-Baptiste Kempf 2:32:25 And in the community, I want to speak about two people who are wizards of assembly, right? The two of them are actually working, uh, living i-in north of Europe, uh, Sweden a-and, uh, and Finland. And, um, Henrik Gramner knows so much about Intel x86 assembly that when we ask questions at Intel about things, they tell, like, "Why are you asking us Intel? You have Henrik. Henrik knows better." He knows all the cycles of almost all the SIMD instruction by all the CPU generation. "Oh, yes, this is a P4, this is a Nehalem, this is a Core 2," et cetera. That person is, like, the best person on assembly in the world.

Kieran Kunhya 2:33:12 Yes.

Jean-Baptiste Kempf 2:33:12 And he's the nicest person that you've seen, like very... He, he arrive, you don't see, uh, he's amazing. And the other one is, is called Martin, Martin Storsjo, and he's... They're doing mostly the same on Arm, right? So Neon, right? And iPhones and Androids and so on. And he codes in assembly on his phone, editing it with the crappy, uh, uh, keyboard, like virtual keyboard you have while watching his kids play in the playground, right? Like, like this is just like wizard level. So those two people are like-

Kieran Kunhya 2:33:55 Yes.

Lex Fridman 2:33:56 So a part when you're programming assembly at that high level, a part of that is knowing the architecture that you're programming on.

Jean-Baptiste Kempf 2:34:03 Yes.

Lex Fridman 2:34:03 So x86 and Arm.

Jean-Baptiste Kempf 2:34:03 On Arm in particular, yes.

Lex Fridman 2:34:04 Arm in particular. But x-x... I mean, these are complicated architectures, right?

Kieran Kunhya 2:34:09 Yeah. But, uh, Arm, um-In some ways it's more co- X86 with, uh, out of order execution is not so bad. Arm, you really need to understand all the different generations of Arm processor 'cause they're all different. There's A72, uh-

Lex Fridman 2:34:22 Yes

Kieran Kunhya 2:34:22 ... et cetera, et cetera, and there's the Apple variant, there's this variant, there's that, and you need to write code that works efficiently on all of them. X86, well, broadly speaking, you have Intel, AMD, and you have sub-variants, but generally speaking, there's... Something fast is gonna remain fast on all of the variants, whereas an Arm is a completely much more complicated

Kieran Kunhya 2:34:42 ball game.

Lex Fridman 2:34:43 We're, uh, taking a non-linear journey through history here, but we're talking about, uh, Michael Niedermayer. Uh, and I, I wanted to ask about this. Uh, for a time there was a split in FFmpeg and, uh, Libav.

Jean-Baptiste Kempf 2:35:00 Yes. So, um, in open source projects, um, sometimes you disagree, right? Um-

Lex Fridman 2:35:08 [laughs] You have such a nice way of putting it, yeah.

Jean-Baptiste Kempf 2:35:12 And the good thing is, because of the license, you're allowed to basically do your own, right?

Lex Fridman 2:35:17 Yeah.

Jean-Baptiste Kempf 2:35:17 Um, and this is normal, and this has happened all the time, right? For at the point there was a, a GCC at the time of GCC2 and EGCS, uh, which became then GCC3, right? There is what we told KHTML with WebKit, with Blink. Um, it is a same process. And also, like, when I want to do a new feature today in VLC, I fork, I do my thing on my own, and then I merge back to the community. So there was a, a, a split in the open source community on FFmpeg, which become Libav and FFmpeg, and after a few years, well, the community merged back and people moved on. It's a bit, um, drama that is normal in open source community, but forks are even... They're important because they, they, they change the, the status quo of a community. Um, um, not talking about FFmpeg a-a-and, and Libav here, but the, or the GCC fork made GCC a ton better because the, some people wanted to change the architecture fundamentally to make it faster. And of course, it's always question of people and so on, but in the end you realize that FFmpeg today is better than it was, uh, before the fork. Um, and now, well, we're back all together, right? Uh, and I spent a lot of time, and, and Kieran can say, uh, in the, in, in the community. Um, it's not often, to be honest, uh, very well explained, uh, because a ton of the reasons are not very public. Uh, but I think that's, that's normal, and that's good.

Lex Fridman 2:36:49 Yeah, I mean, you're making it sound really nice, but there, there is battl- there's pretty heated battles inside open source projects. I mean, it is a very passionate community and you're kind of in a distributed way have to define the direction of things.

Jean-Baptiste Kempf 2:37:01 Yeah.

Lex Fridman 2:37:01 So here looking at, uh, Perplexity, "FFmpeg and Libav split in 2011 mainly over project governance, leadership style, and development processes, not because of a fundamental technical disagreement. Uh, FFmpeg effectively absorbed Libav's work [chuckles] while Libav withered and most distributions and developers moved back to FFmpeg." Yeah, that was a, that was a weird, that was a we- from a user's perspective, that was a weird experience 'cause, you know, I'm a Linux user, so, you know, whether it's Ubuntu and so on, all of a sudden, I think for, for a c- for, for a little bit, Ubuntu, I feel like, am I remembering correctly, switched to Libav and-

Jean-Baptiste Kempf 2:37:39 12, 14, something like that.

Lex Fridman 2:37:40 Yeah.

Jean-Baptiste Kempf 2:37:41 Something like that.

Lex Fridman 2:37:41 And then they switched back to FFmpeg.

Jean-Baptiste Kempf 2:37:43 They switched back, yeah.

Lex Fridman 2:37:43 I was like, "What is happening?" [chuckles] So on the sort of, um, you f- you get to feel the ripple effects of the different internal debates that are happening.

Kieran Kunhya 2:37:53 To be fair, on Apple, when you type GCC, you get Clang. Like they, they did something like that as well, so.

Jean-Baptiste Kempf 2:37:58 Yeah. So, so to me it's like the fork was like heated drama, but most of the development from Libav was merged back into FFmpeg, right?

Lex Fridman 2:38:08 Yeah.

Jean-Baptiste Kempf 2:38:08 So de facto FFmpeg got a s- a, a superset around Libav, and so that gave the user, because in the end we work the user- for the users, a, a larger set of features and a ton of things that were, um, discussed.

Lex Fridman 2:38:22 Mm-hmm.

Jean-Baptiste Kempf 2:38:22 For example, the debate on reviews, on, on how we push are something that now is completely settled in FFmpeg and is following what mostly what everyone in the community agrees, right? So de facto, everyone who was active on Libav came back i, in work on FFmpeg because l- the disagreements were fixed, and in the end, FFmpeg is stronger, uh, than it, it was before, right? And-

Lex Fridman 2:38:49 Mm-hmm

Jean-Baptiste Kempf 2:38:49 ... I know people love drama, but, um-

Lex Fridman 2:38:52 Well, my, my main concern, I, I understand, and I think looking at the, the long history, it's all for the good. But I do... I am concerned because there's so few humans that are critical to the success of open source projects that I have seen it, uh, be a psychological toll on folks and, you know, sometimes leads to burnout. So you have these incredible people that are at the core of open source projects. There is a moment that happens 'cause, like, what is the motivation of doing it? Ul- ultimately it's because you're passionate about it and it makes you happy. Then a certain point you wake up and it's like, "This's been a bit too much heat from the drama." So, like, at the, at the project level, the project continues and often flourishes, but sometimes there's these individual humans that are just like-

Jean-Baptiste Kempf 2:39:44 But-

Lex Fridman 2:39:44 "I've had enough." [chuckles]

Jean-Baptiste Kempf 2:39:45 Yeah, but it's not just about forks, right? So there's a g- very-

Lex Fridman 2:39:48 Yes

Jean-Baptiste Kempf 2:39:48 ... what, what you, what you are, um, referring to is, uh, one of the most challenging and most interesting part of, uh, open source today is maintainers burnout, right?

Lex Fridman 2:40:00 Mm-hmm.

Jean-Baptiste Kempf 2:40:00 And AI is a problem because of that. And Daniel Steinberg, which is, um, the maintainer of Curl, uh, who's probably one of the best promoter of open source in the world. He's, by the way, a member of the European Open Source Academy with me, so I'm very, like, humbled to be on the same community as him, right? He's against what he call AI slop, right? Because it gives a ton of, um, uh, fakereports or bad reports, bad patches. And then a lot of maintainers have a lot of burden to maintain the software. And this is, uh, straining the, um, the, the, the mind of open source developers much more than forks.

2:40:42 Mm-hmm.

2:40:42 And, and for example, the XZ fiasco was because there was one guy maintaining it, and he got basically hammered by two attackers who were asking him questions nonstop at weird times at night to block him, and at some point he got fed up and says, "Okay, I can't do that," and gave the commit access to the attacker.

2:41:01 Mm-hmm.

2:41:01 Um, so burnout in open source community is something that exists, uh, but mostly it's about maintaining things, right?

2:41:11 No, for sure. But I, I wonder how do we, we help that, 'cause those people are so important. The, the human beings are so important, the core of these pro- projects.

2:41:19 So, so for example, now I am maintaining a ton of multimedia and non-multimedia library-

2:41:23 Mm

2:41:23 ... as maintainer, uh, because the maintainers f- got fed up, right? Um, some on VideoLAN, some outside of VideoLAN, uh, because it's sometimes you need a, a, a tough skin, right? Because you get, like, it's not really attacks, but, uh, oh, this is not working, this is not working, and you feel it personally. Um, and this is also why resources or the, the Google fiasco, um, is, was a problem, right? They don't realize that in the end you have, um, you know, it's like the same graph where you see, like, everything and it's just like one random open source project that is maintaining the whole-

2:41:59 The Nebraska thing, yeah

2:41:59 ... in-internet. You see the one, right? The-

2:42:01 Yeah, this is the meme. I mean, it applies to, to a lot of open source projects.

2:42:05 So many things, yeah.

2:42:05 But this is the all modern digital multimedia infrastructure, and then that thing at the very bottom that everything relies on is FFmpeg. It's true. And then there's usually, you know, a handful of folks that are maintaining that.

2:42:19 And FFmpeg or VLC, right? You have a community of 10, 15 core developers, are not the worst open source project. XZ, which is even in more installations, is one person, right? There is one guy-

2:42:32 LibXML is, uh-

2:42:34 Yeah, LibXML, right? There was a big stop. No one is maintaining-

2:42:37 Yeah

2:42:37 ... LibXML anymore, which is like pass- the only library that is able to pass XML everywhere.

2:42:41 All the crazy edge cases of XML under ridiculous circumstances, and they get attacked by security researchers because there's one other crazy edge case that they haven't thought of and say, yeah, but the body of knowledge to actually resolve that is massive.

2:42:56 There is one guy maintaining all the time zones for everyone who is in the middle of, I think, was it Nebraska or South Dakota?

2:43:03 Yeah, it could be, yeah. [laughs]

2:43:03 Like,

Jean-Baptiste Kempf 2:43:04 the mental health of the open source maintainers is something that large corporations don't care or don't see, right? It's just like, oh, yeah, I'm just doing an open source, uh, report and so on.

2:43:16 Some of it is financial, but some of it-- And b- people should definitely support open source financially-

2:43:22 Yeah

2:43:23 ... all across the board.

2:43:24 Yeah.

2:43:24 But some of it is also, like, spiritual on a basic human level. There's something that happens, like, with this image of F- of FFmpeg and so much of the internet depending on it, where people almost, like, talk down to the folks who are carrying these projects forward and maintaining it. In the security community, they certainly did. That was one of, that was one of the things I think that argument came out is there was, there was a portion of the security community who's like, "No, these guys write crap code. They need to fix their crap code." I'm like, "No, no, no, no. This is a guy's hobby project. You've, you've have a security bot that's gone and found some AI-generated stuff. That guy didn't write crap code. It's just an edge case to the 99th point nine nine nine nine nine percentile he didn't think about because it's ho- his hobby project decoding Star, Star Wars games."

2:44:10 Forget the hobby project aspect of it. It's, it's just hard work and it's b- it's beautiful, and it's like the, the right approach there is to celebrate people-

2:44:18 Sure

2:44:18 ... for doing incredible, incredible work. It's, it's just incredible that humans step up-

2:44:25 Sure

2:44:25 ... not getting really paid at, at first or maybe ever, and then they're doing it out of the love of it, and we need to... Like, human civilization runs on people like that. We need to celebrate them.

2:44:36 To, to, to give you an idea, I received death threats o- on VideoLAN, right? And, um-

2:44:40 You mentioned that to me. Like, what, what is, what is behind that?

2:44:43 So that must be, what, 2009, 2010, right? Um, uh, Apple is moving from PowerPC to Core Duo, s- um, that's probably in 2006. And by 2009 or 2010, I decide that we are not going to do new versions of VLC for PowerPC.

2:45:01 Mm-hmm.

2:45:02 At that time, like VLC, we were close to the number 1.0 release. We were four of us, right?