Haven't listened to this one yet but for people with sticker shock over the time commitment, Lex consistently does amazing podcasts with really sharp people over the course of a few hours and imo (being an avid podcast listener) it's one of the best out there.
Edit to add: His recent episode with Jack Barsky (former KGB spy) was exceptional and worth a listen if 5hrs is too long of a test case
I tried multiple times to give a fair shake, but Lex seems to never to any research on the guest, almost never has anything interesting to go back and forth on with guests, and seems mostly there to have a wall for guests to talk to, but somehow is less interesting than the wall.
There is no exchange of ideas, its just the guest being able to say whatever they want without any pushback.
I have no idea why interesting guests get called onto this show, other than just having a place to talk to a lot of listeners
These are exactly my feelings with Lex Fridman Podcast.
I tried with so many of his guests, the lineup is amazing! But every time I end up dropping early on the conversation. Surprisingly, I find him even worse than Joe Rogan as a host (surprising because Joe doesn't have a technical background, whereas Lex does). Joe at least is an apparently neutral host with some charisma and guests get to portray their ideas clearly. Lex somehow sucks all the fun out of the conversation.
Contrast with, for example, Tyler Cowen which also hosts a wide range of interesting guests, and still is able to pose interesting questions and guide the conversations in way that are insightful for the listener.
My impression is that he often tries to 'steelman' the other side by purposely putting forth more simple-minded positions or questions than he actually holds. At least in AI territory it often seemed that way, noticeable when a guest mispeaks or mislabels a theory and Fridman corrects them, suddenly knowing the exact scientific term or implications.
I don't think he's necessarily doing a great job with this approach, but I'm guessing that there's some attempt of method behind it, rather than him being negligently oblivious to the subject at hand.
> noticeable when a guest mispeaks or mislabels a theory and Fridman corrects them, suddenly knowing the exact scientific term or implications.
In my experience, (Elon interview) they are _both_ misspeaking. Just one (Elon) more so than the other (Lex).
Of course, you wouldn't know this as an outsider to that field, necessarily - and it does have the effect of making both parties seem smarter than they actually are.
But that's not generally what news anchors do - frequently they go in with a pre-established narrative; and whether or not they are agreeable or combative will depend upon that narrative.
Joe pokes the bear with his genuine (yet curious) ignorance, and listens to some very smart people with a child like wonder as they respond.
Cowen is absolutely one of the best out there too, but I do wish he got in deeper with his guests! His blog Marginal Revolution is also consistently interesting and he has a good map of great spots to eat across the DMV
The most annoying habit he's picked up recently is asking people to define words. It's never really interesting. For instance, during an interview with an economist, "let's start with what is economics"? He was interviewing CEO of Coinbase and Armstrong will try to give a normal answer and Lex will just interrupt "what is a wallet" or ask questions like "do you consider Coinbase layer 1, 2 or 3". They're just bad questions and interrupt a normal conversation.
On the other hand someone like Tyler Cowen doesn't ask for clarification unless he legitimately doesn't know what some is talking about. Some guests even said "maybe I should explain that", to which Cowen replies, "this conversation is for us, don't worry about the audience", and that leads to a much better natural and interesting conversation. If the listener doesn't understand anything they can always pause and look up what they don't understand
The "is Coinbase layer 1, 2, or 3" is a particularly dumb one and a good example of how eye-rollingly bad his questions can be. It's hard for the guest to get into the interesting details when interrupted with a question that's wrong in a bunch of ways that betray he has no idea what he's talking about. It just feels like he's full of shit.
If I were asked this question, I would actually consider this a good question, because:
- it probably is something people are wondering
- it's a good moment to talk about layers and where coinbase fits in the ecosystem
the answer for the curious: none of them. Coinbase is just a wallet using the layer 1 of most cryptocurrencies, and can also use apps and wallets from other layers (I'm not sure it does support any other layers at the moment though). A layer 2 can be seen like another cryptocurrency built on top of a cryptocurrency (layer 1), and a layer 3 is the same built on top of a layer 2, and so on.
It's a different kind of investigation into a subject. Perfectly valid. Any textbook on a subject, will often start with the most basic question. The Who/What/Where/Why/When of questions. Another way to describe First Principles Thinking. I find it extremely foundational to know the basics in developing higher knowledge in a field.
Interviewing is a skill and Joe had a long time to perfect it. That said after 300 podcasts Lex should’ve improved but he hasn’t.
He really needs to get a producer to provide him with constant feedback and to guide him in general; every great interviewer/newsperson had an even better producer behind them.
I'm amazed at all the people here criticizing Rogan's interview skills. I think it's an instance of disliking a person in general and applying it unwittingly to everything they do.
Rogan may be a shitty person but he is a phenomenal interviewer. He manages to connect with the people that he is interviewing on a personal level in a way that makes them open up like they have known him as a childhood friend.
That just makes him a stable, consistent sounding board for his guests and listeners.
This "uniquely individual, but authentic and consistent persona" factor is a great trend for public persons. The reason why it works is because even if you don't agree with or like the host, say Lex or Joe, they are authentic and consistent enough that you can 'subtract' the difference between your average opinion and theirs from a conversation with a guest and still get back a reasonable read on the guest from your perspective.
It's like they're a human guage block, constantly measuring their guests with their own personality and reporting the relative measure. Then you slap your Rogan-to-infogulch guage block on top and now it's like you're measuring their guests yourself by proxy. (I'm tired but this seems like a reasonable analogy.)
It’s funny how all of this works. I’ve tried many times to listen to Tyler Cowen after being introduced to him on Lex’s podcast, I’m still subscribed even. I just can’t do it, something about how he approaches the interview process grates on my nerves.
On the one hand, I appreciate his breadth (but not depth, next point) of knowledge, his charisma, his pace, and his energy, all very admirable traits and qualities. I always listen to his podcasts.
On the other hand, there are many things that bother me about him. What I write may appear overly critical, but he is a public figure and it goes with the territory. Nobody criticizes me -- I am a nobody.
Some, perhaps more than a few, of his questions -- "I ask the questions I want to ask and not the questions you want to ask" -- more than curve balls seem like balls thrown off the field: "Try to hit this one!" To the point that his guests are puzzled, but they do not want to look ignorant or poor guests and do not react.
He greatly overestimates his knowledge of science, technology, art, cooking and everything else excluding, maybe, economics. Let's take languages. He says he learned to speak Spanish, then I hear him speak a few words of Spanish, and what I get is that he knows a few Spanish words and phrases, which in my opinion is not the same as speaking the language. I throw together a few ingredients, but I am not a chef.
When he talks with pace, conviction, determination and apparent curiosity -- I say apparent because I have never heard him change his outlook on anything, whatever he proposes is doable and there must be some low IQ person in charge or some sort of rent-seeking behavior to make it not happen -- about what I know (biology, AI, my country, women's rights), he is far from having a solid grasp of the subject, despite his conviction, tone and determination.
But no one objects because he interviews and is not interviewed -- sometimes he is, next point -- and his guests usually shy away from replying to questions and opinions, informed or uninformed, about classical music, paintings, Chinese cuisine and obscure African poets.
He maintains his intellectual position at the top of the mountain through fog, smoke and utter confusion, which is admirable for the audacity, but also rather unsettling. When interviewed, he has a ready answer for any question, which for some may demonstrate his knowledge, wit and intelligence, to others, like me, he seems to be pulling answers out of his backpack.
He takes positions that he does not follow and marries himself to "causes" because they suit him. For example, he often speaks out against alcohol and legitimately so, but my impression is that it is convenient for him because he does not like to drink: would he follow his own recommendations if he liked alcohol instead?
In an interview about his latest book (which, like others he has written, are rather forgettable, a pot-pourri of whatever crosses his mind on the subject), he said that "on average" exercise is a net benefit to longevity and mental capacity. But he doesn't seem to exercise, and he doesn't talk about exercise or moderation at the dinner table because, as far as I can tell, he doesn't like to exercise and likes to eat plenty. But it is clear that a glass of wine a day is not a "bad thing" (some say it has positive effects on longevity) and that being 50 pounds overweight (all fine, it is his choice and my point is not about his lifestyle choices) puts one at risk for diabetes, poor quality of life, and early death. However, thanks to his assertive way of speaking and presenting his ideas, and his position among the "intellectual class" no one ever objects or asks though questions.
I still think Cowen is about the most interesting interviewer out there, but I also think this is all spot on. I've absolutely had this skepticism for having ready answers for everything. He'd be more interesting if he said "I don't know" more often.
Lex is a fantastic host. Probably the best one in podcasting right now. He pushes back all the time, but knows when to move on when there is disagreement. There's nothing worse than a host who needs to win arguments.
I appreciate that he lets guests speak without any need to conform to a narrative. I can make my own opinion. I don't need the host doing it for me.
I truly don't know where you get the impression that there is no exchange of ideas. I've experienced him going very deep on many, many different topics.
I like Lex, but this is way overstating his pushback. He's mentioned himself how Rogan tells him he needs to push his guests more. He just prefers not to, because of an idea of love of the person...
I also think he can push back more, but on the whole he has interesting people and it's kinda nice to see him ask some basic open ended questions and see how different guests react to it.
I would say it's a strength of his that the guests just get to say what they have to say. He always prepares a couple of questions in case the conversation stalls, but other than that, he let's the guest talk about their ideas, instead of sticking to a specific story he wants to cover.
I concur with this. While Lex, IMHO, has FAR less personality than, for example, Joe Rogan, I’m not watching because of Lex; I’m watching because of the guest. I’m very happy to have the guest take over and, for the most part, lecture their thoughts and opinions. Lex allows this to happen.
>He always prepares a couple of questions in case the conversation stalls
A lot of people underestimate how much effort it takes to keep a relatively smooth flowing conversation for a matter of hours. Ultimately if the words are flowing the host is doing a good job.
In many cases for some of the more controversial guests Lex adds some disclaimers at the beginning, but it's not the host's job to argue every single statement.
It's interesting that Lex is so polarizing, I personally find him a great host.
Him almost never doing research is objectively false though. He often reads books/articles/tweets written by the guests and asks about them for example.
For me it's not that he doesn't push back, it's that he often interrupts with some trite nonsense. So the guest is getting in the details of something specific and interesting and Lex will interrupt with something he thinks is profound that's meaningless and derails the explanation. The questions he asks I find similarly frustrating - like he's not really listening, or thinking about it in any depth.
I get the sense he's constantly trying to prove how smart he is, and at least to me - it backfires badly. Hard to describe, but I guess it comes across mostly as shallow bullshit and it's tedious to listen to him despite his great guests.
> I get the sense he's constantly trying to prove how smart he is
I've listened to hundreds of hours, and I just don't get this at all. I don't think he has a selfish motive for the "interruptions". I think he's just saying what's on his mind, in a somewhat vulnerable way.
Yeah same. He might want to prove his existing knowledge (natural, to show the guest that he understands what he's talking about) but doesn't go about trying to outsmart people.
Couldn't have said it better myself. It's especially eye-rolling when he tries to inject comments or questions about how "it's all about love" straight out of nowhere. For someone with a scientific background, he sure produces a lot of fluffy nonsense, to the point where Carmack had to stop him there once or twice.
Reminded me of Jordan Peterson getting shut down by Richard Dawkins once he started going down yet another Deepak Chopra-style mystification rabbit hole.
Not so impressed by the research: I forgot the exact wording, but it took me a lot of effort to keep listening when he asked Neal Stephenson something to the effect of if he had ever heard of the metaverse.
I think I remember better now. It was Friedman discussing the notion of some iPad app that could be used to learn and educate, adapting to the user, etc. Stephenson replied that he had indeed written a whole book about it (Diamond Age).
Yeah, I was surprised too. I thought a technical community would appreciate the amount and breadth of technical knowledge that Lex explores with guests and allows the public domain to gain.
100% agree. But if you mention this, he will tell you that he is a college professor, and therefore you don't know what you're talking about.
In 20 years, with luck, he will look back on himself today and just sigh.
He simply does not follow what the majority of his guests are saying sometimes to the point of argument. Maybe "strong disagreement" is a better phrase. He will not understand what a guest says, then challenge it as if he did. It's infuriating to listen to.
I mostly avoid his podcast now, because of it. I may listen to this one because John always says a lot of things that are interesting to me. Unless it's VR. Then, I don't give a hoot.
I wish he was unprepared. The creepier moments of the show are when he starts talking of his own ideas which sound like someone made up a parody on naive technocratic optimism blended with teenage babbling of how all humans are going to love each other etc etc.
His more recent "it's all love" trend is the cringiest of the bunch, couldn't agree more. I truly wish you could have a Lex-free version of his podcast with just the guests talking.
Yeah it drives me crazy and I'm unable to listen to them.
It's maybe mean to say, but I just don't think he's that smart and unfortunately it shows. It also doesn't help that he doesn't seem to be aware of it.
Curious, why do you think all those great guests come to his podcasts? Wouldn't they politely decline if they thought he's not a very smart guy and a terrible host? I really think John Carmack, Mark Zuckerberg or Kevin Systrom don't need the publicity via his podcast, so what other reason would they heave?
He has a large audience and is anodyne enough to not matter as part of the interview. He may ask some dumb questions or say some dumb stuff, but he won’t be hostile.
It’s why someone may go on tv for a few minute spot, it’s good for getting your message out to a large audience despite the interviewer being in the way of that.
I'm guessing it's because of his massive reach towards a specific type of audience that may be somehow beneficial to them. It's most certainly not because of the profundity of the intellectual exchange they have on these podcasts.
Same. I really want to like the guy, it feels like all of the credentials are there, all the stars are aligned, but his contribution always comes off as anemic and uninspired. There are plenty of times where I wish he would give more space to the guest instead of injecting something of his own that adds zero value.
Not to hate on him but the credentials are not really there. He taught a fluff seminar course on self driving at MIT and used the brand of the university to line up a ton of great guests for an AI podcast that he then pivoted into a personal brand. From what I understand his research at Drexel was focused on HCI and from the interviews that I've seen it looks like he has a pretty shallow understanding of ML.
I actually feel the complete opposite is occurring, the guests are free to go really deep on their own and Lex provides a simple sounding board without getting his ego involved he teases more out of his guests and doesn’t try to appear clever. If Lex is so uninformed, unresearched an unengaging why does he keep getting exceptional guests and have such an incredibly popular podcast. I can understand you not personally liking his style but he hardly ever interrupts his guests or gets in the way apart from smoothly moving onto the next topic. Maybe his philosophy is just enjoy the speaker and let them shine, rather than worrying his own ego to want to be constantly noticed.
Same. Likewise, the interviews with Roger Penrose, Jack Dorsey, Stephen Kotkin, Lisa Feldman Barrett, and his own father have stayed with me.
All for very different reasons.
I stopped binging, with great sadness, when it looked like Lex was going the down the intellectual dork web path (Malice, Rogan, Weinstein, Harris, Haidt, Ferguson). Hopefully that was just a phase.
This is probably his best interview, IMO. Keller was allowed to go very deep into interesting topics and by the end of it you get a solid perception of what CPU development was like in the earlier days (DEC, AMD Athlon) to how it has evolved with modern synthesis tools and large teams. As well as the business overall.
I saw it is as they both had a good intuition about each other, so they were both free to talk without the need for some "translation" barrier, which appear when two people are constantly struggling to translate things in a frame of reference that they think the other requires.
Lex has a special skill of making me roll my eyes every time he speaks. I hate that I always return to his episodes, though (he has amazing guests). The way I approach it is to jump the parts he speaks.
It seems like you just prefer a different interview style, but that doesn't make him a bad interviewer. A good interviewer asks questions that they know a guest can speak at length in answering. An interviewer's job isn't necessarily an archaeology quest, it's to present to an audience.
> There is no exchange of ideas, its just the guest being able to say whatever they want without any pushback.
This is precisely why I listen to his podcast. I am not interested in Lex's ideas (beyond maybe AI), I am interested in the guest's idea. The back-to-back Oliver Stone and Stephen Kotkin episodes where really interesting to that effect. Here are two people with diametrically opposite points of views on Russia and Ukraine, and both had ample time to explain their positions in depth, without a host trying to trick or confront them. Let the listener make up his own mind.
And when I'm not interested in a guest (Joe Rogan recently), I just skip that episode.
There are many small things that irritate me when listening to his podcast, but his ability to let his guest speak in depth about his ideas trumps them all.
Lex Fridman has his ups and downs as a host but how on earth can you call him terrible? He made Zuckerberg prove his humanity by solving a cardboard captcha ffs
The interviews with academics and technical folks he does are fine because it's a domain where you can tell that he knows his stuff but he seems to have increasingly become an off brand Joe Rogan with political guests or nutritionists or even stranger guests on the channel, whose quackery goes largely unchallenged.
The worst offender was probably Oliver Stone. That episode became a straight up megaphone for propaganda.
In the first few minutes of the KGB interview he asks for an explanation of the Stasi. It was clear he didn’t know what the Stasi was. He didn’t just come across as unprepared he came across as someone who isn’t even interested.
>In the first few minutes of the KGB interview he asks for an explanation of the Stasi. It was clear he didn’t know what the Stasi was.
I think your perceptions of Lex will mislead you to assume he's asking out of personal ignorance instead of asking simple questions he already knows the answer to for the sake of the audience.
As a result, you happened to pick a bad example as Lex already knew what the German Stasi was before that interview because he discussed it over a year ago with Michael Malice back in December 2020 (deep link to "Stasi" conversation):
https://youtu.be/uykM3NhJbso?t=56m39s
Lex does not tire out listeners with constant disclaimers in every interview by saying, "I already know about <X> but for the sake of my audience, what does <X> mean?"
Thankfully, he just shortens all that to : "What does <X> mean?"
... but then that makes him look unprepared. But if he puts in that disclaimer every time, we'd complain he tries too hard to let us know he's smart about the topic. Either way, it seems you can't please everybody.
He clearly asks these 'entry-level' questions for the audience when he thinks it's helpful. I've been thankful of that a few times when a guest, speaking about a field that is far from my own, is asked to clarify something briefly. So I'll disagree with you here that it 'is clear he didn't know what the Stasi was'.
I'm glad he does. Apparently I'm less familiar, than you, with the details of the ridiculously wide range of topics he discusses. I enjoy not having to prepare to listen to a podcast, or go through the whole thing confused, with a list of topics I need to read about to understand what I just heard.
Agreed - he somehow gets the most amazing guests, but it's all un-interesting, softball questions that don't challenge the guest at all, delivered at a glacial pace.
Yes, it's designed that way imo. More of an exploration into a subject, letting the guest shine and transmit their thoughts.
Would you have preferred someone like Charlie Rose, constantly interrupting the guests. A 2 way dialogue might be more normal, but you lose a lot of signal. These aren't confrontational interviews, but more of an educational one hosted by a non-expert.
If people prefer a discussion of 2 experts, this is clearly not it.
Ariel's ability to clearly express and convey such an amazing vision for space colonization... Talking about how to make self assembling space station modules.
It's interesting that Lex not injecting his own ideas is a con to you. For me, that's a huge pro. He's one of the few content producers I enjoy because he doesn't seem to push any particular agenda. He just provides a platform for really high quality guests to talk. Occasionally he'll ask them fun questions that you wouldn't expect for his guests, like asking Brian Armstrong about the meaning of life.
I wouldn’t say hes a terrible host but he is a certain kind of host which is like the opposite of a Charlie Rose- who mostly glossed over the technical details of a person’s accomplishments to get to their more personal motivations and universal human interest type stuff. From the 3 or 4 episodes I’ve seem focuses entirely on the technical and is good at that but not very good at drawing the guest out emotionally. But this method still has a lot of value for a technical podcast.
Yeah I agree, listening to this episode I was thinking "Carmack loves talking, and Lex really lets his guests talk". He often just asks things that are useful to him, and not necessarily relevant to the guest. Yet, it seems to work, I end up listening to a lot of the content and really liking it. It's not a bad formula.
This reminds me a bit of Joe Rogan who also ends up interviewing a lot of interesting people but doesn't seem to really care about where the conversation goes. Rogan talks much more though, and I think is more into discovering what kind of person he's talking to (sometimes using substances).
I agree with this take mostly. There are many times I expect Lex to guide his guest to a particular conversation and he fails to do it. Rogan is better at this, to my surprise. Though Rogan has his bad days too, but generally he's better at it. I admit though, I am not an avid listener of either shows. I think a lot of this comes down to personal preference. The beginning of Lex's shows make me cringe sometimes.
Coming back to this. Just want to say I disagree, while it would also be a nice listen to have experts discuss topics with one another equally. It is refreshing to gain knowledge on a subject from an expert in this conversational format to basic questions.
The fact is Lex isn't an expert in 99% of the subjects his guests are coming from. And neither are we. But, the format is a great public lecture with a soft conversation in between. And it should be increasing the general knowledge of the public who are watching. And that's one of his goals. And why the podcast is popular with many not from technical backgrounds.
I feel like every technical person will surely agree that the average intelligence of most of the public is going downhill, at the same time as the complexity of science & technology increase. This kind of general intelligence podcast can only be a good thing.
Thank you for this comment. The host took me by surprise.
I saw John Carmack and 5 hours and just thought: "Wow, those two must have really hit it off somewhere".
Listening to his questions was a struggle. I wish I could have only listened to Johns answers.
He should keep his opinion to himself more. He hasn't yet accomplished anything significant in the world, so nobody really cares what he thinks about particular subjects. He comes off as pompous. I also don't like that he takes crypto seriously.
I totally agree. His guests are great, he stinks. I'm sure he's a perfectly nice and amiable person IRL, but as a podcast host I am not a fan. I can't really watch it.
I am also confused how he became so popular, he seems pretty uninformed, my best guess is that his suit fools people into thinking that he must know what he is doing.
I find Lex Fridman to be wooden, robotic, non-sequitur, not smooth or charismatic, and (I admit, shallowly) not a pleasing speaking voice to listen to, and not that articulate actually. The main reason he gets famous guests is that he has a reputation for getting famous guests. He's like a Kardashian.
I actively have to tune him out and suppress my wish that someone, anyone else were asking the questions.
But it is what it is. We don't have voting rights on who random walks their way to being famous in this world.
There are plenty of more talented podcasters than him, but the appeal of his interviews is that he asks simple, non-leading questions, often stupid ones (e.g. “what is a compiler”), but encourages the guest to talk and give really full answers. And he catches details they say, and follows up with questions that go deeper into the specifics, rather than just returning to a script of pre-planned questions.
> follows up with questions that go deeper into the specifics, rather than just returning to a script of pre-planned questions.
Maybe I just saw the wrong interviews, but I got the exact opposite impression from those I've seen. He was literally reading off a script of questions and not following through on interesting things the interviewee says or asking them to develop ideas or explain themselves. Can you point to any particularly good episodes that I can use to give him another chance?
I liked his early programming-related ones the most: Chris Latner, Jeremy Howard, Bjarne Stroustrup, Jim Keller, Brian Kernighan. I actually haven’t listened the bulk of his interviews (nothing related to crypto, philosophy, history, physics). Maybe those are done better by other podcasts.
It sounds like you’re evaluating the tone and charisma while others are evaluating the content and dialog.
He is very wooden, but I agree with the others, his questions are incredible and cut straight to the point. For example while others (Tim
Ferriss) might have asked Zuckerberg “How do you deal with negative sentiment around you and your work?” Lex asked something like “Many people believe that your products are destroying democracy, that social networks polarize us into bubbles, and you are one of the most hated men in America. How do you contend with that?” Wow and ouch! Amazing conversations
I tried listening to his conversation with Brian Armstrong.
He insisted multiple times that being "canceled by the woke mob" is something that the government does to you because they are afraid of losing power to common people on the internet. That's not just kind of wrong, or kind of skewed, that's a complete inversion of reality.
You could replace him with a markov chain trained in the right wing conspiracy media bubble and get basically the same quality questions.
He = Armstrong or Fridman insisting on being canceled by the woke mob? Unclear if you're saying that was Fridman's view or Armstrong's view and Fridman didn't push back.
He = Fridman. Armstrong was much more clear and eloquent.
Fridman really seemed to be going for applause lines from a very specific crowd, not anything particularly coherent. I think he's realized that his growth as a podcaster comes from the demographic that frequents hosts like Jordan Peterson, Joe Rogan, and Ben Shapiro, and the more he can get them engaged the higher his numbers will get. But the topics that demographic cares about aren't AI and ML.
Just as a point of pushback on the audience, I'm not a listener of any of the shows that you mentioned; my most listened to other podcasts include Tyler Cowen, Vergecast, Trade Talks, Vox, and the Charter Cities Podcast
I think he's pivoting pretty sharply. If you look at the first podcasts he has listed on IMDb:
* Andrew Ng: Deep Learning, Education, and Real-World AI
* Michael I. Jordan: Machine Learning, Recommender Systems, and the Future of AI
* Marcus Hutter: Universal Artificial Intelligence, AIXI, and AGI
* John Hopfield: Physics View of the Mind and Neurobiology
* Alex Garland: Ex Machina, Devs, Annihilation, and the Poetry of Science
* Ann Druyan: Cosmos, Carl Sagan, Voyager, and the Beauty of Science
Compare to his last few episodes:
* Steve Keen: Marxism, Capitalism, and Economics
* Richard Haier: IQ Tests, Human Intelligence, and Group Differences
* Susan Cain: The Power of Introverts and Loneliness
* Douglas Murray: Racism, Marxism, and the War on the West
* Richard Wolff: Marxism and Communism
That's a pretty hard pivot in topics that he's covering. If discussion about whether BLM is Marxism intent on destroying the West isn't the content you're looking for, I don't think you're the audience he's targeting anymore.
The pivot happened ages ago, his podcast was originally called something like the Artificial Intelligence Podcast, and then changed to the Lex Fridmin podcast along with a change in more general guests. I do wish he focused more on computer related topics, but his guests still tend to be very interesting.
That's a good point, I do find myself just skipping over those episodes though rather than throwing out the podcast as a whole (I'm also trying to be better about that in general including with books which I would religiously read from beginning to end no matter the quality which is silly).
I know what you mean but he has improved over time. His main fault is talking too much sometimes, I'd rather hear the guest. While some of his questions are lame, he usually makes up for it with some well researched and interesting questions. Not every question can be a winner in a five hour interview.
Really? That's fascinating, if anything I usually think he's not rigid enough and too easily goes along with guest claims (the Oliver Stone episode stands out for not pushing back).
Even though I didn't align or agree with Stone. It is refreshing not to hear the constant back and forth you see on TV's or just each network running their own agenda. Truth is a fine thing, nothing is black or white. I feel in these times it's more important that we stay open minded to ideas, speech and discussion.
The act of pushing back, retaliation, or suppressing information actually increases the desire for said info and notoriety/fanfare of many people and ideas. In the Oliver Stone interview, they do discuss the best way to interview someone is with a blank canvas, and that is honestly refreshing. Let viewers develop their own opinions. But, a little more pushback or rather counter-arguments might have been welcomed, but that might change the tone of the interview into a more adversarial one.
People need to get out of closed off echo chambers, be respectful of each other, and just agree to disagree, but still look at each other as a human being. The only way we progress through the 21st century.
And Carmack is known for making multi-hour extemporaneous keynote speeches that somehow remain interesting the whole way through. Should be a good one.
Carmack is literally a legend from a completely meritocratic perspective. His respect is 100% something he’s earned by being a master at his craft and a deeply knowledgeable person who has continually pushed his field forward for decades while being an accessible and mentoring individual.
In short, he’s a “nerd’s nerd”.
It’s hard not to be enraptured by someone like that. It would be like listening to DaVinci discuss theoretical aerodynamics for three hours.
For those who do not know Lex, just to nuance this, I think it is interesting when dealing with technical subjects (like recently with Demis Hassabis). Outside of technical subjects, it is a bit of a mixed bag, from interesting (like Roger Reaves) to appalling (like entertaining Oliver Stone's bullshit conspiracy theories).
This one is likely to be a technical one and certainly something I am keen to listen.
One thing I don't like about Lex is that he doesn't really challenge his guests (which is one of the reasons why he can bring many well known names, as they know they can mostly give their talking points).
That's not a ding against him, as he has very wide range of guests, so he lacks expertise to challenge them, but something to keep in mind.
He's not perfect. Though I find it pleasant that he has a mostly neutral point of view as that's when you can get the most information out of a discussion. I think in many traditional media formats, the opposite is the case, which is what we're used to. Like for e.g. Charlie Rose, or any of the biased media outlets today.
One can possibly argue that information has always been manipulated and public discourse as well since the dawn of civilization. So, I find having an interviewer be open to multiple possibilities or interpretations is refreshing.
He's not completely neutral though, he has a slight liberal bias, and at the same time doesn't address criticism of said guest many times especially when they are a friend, like Joe Rogan, or Elon Musk. That being said, it is a scientifically progressive podcast that is very illuminating in many fields for non-experts, and the public. And I have learned a lot and enjoyed listening to it.
Yeah, his interview with Pfizer CEO was essentially an advertisement. He also advertises cryptocurrencies a lot, like they are the next money system in the world. Largely downgrading all their harm and inabilities. The man just making his own fortune.
On the other hand, the freedom that guests have is worthwhile.
I default to 2.3x plus silence skipping and I know people who go faster. Most conversational podcasts run effectively at 3x with the silence skipping. It sounds like gibberish for a second or two and then once the focus comes in it's normal. I can't listen to things are normal speed. I started out at 1.5 and then went up in .1 increments over time until it was too much. I still tune some faster speakers back to 2 or 1.1, but that is the minimum. I was always a piss-slow reader so I prefer to listen to things, but anything close to 1x just takes forever.
It is also important to note that Fridman’s conversational and interview style is very dry and information focused. About every question is interrogative with dense, deep answers; but lacking much charisma. He’s almost the inverse of Joe Rogan.
I really love one of the points Carmack makes: You are leaving productivity on the table by not using an IDE. IDEs bring you closer to the Lisp/Smalltalk ideal of live-coding a running system, but with programming languages people are likely to use. Text editor based edit/compile/run workflows are pointless grognardism by sticking with "stone knives and bearskins" tooling.
Do they use a debugger as a tool of first resort? Do they step through their code line by line?
One of Carmack's points was that gdb's terrible interface has made debuggers a tool of last resort in the Linux/open source world, when they should be a tool of first resort, and code should be stepped through even before it's run normally because that's how you get a clear picture of how it runs. Visual Studio's excellent debugger integration alone has made the open source development experience truly "stone knives and bearskins" by comparison until relatively recently.
Back in the day, from the people who brought you commercial Smalltalk:
July 10, 1989 "ParcPlace Systems Inc, of Mountain View California, has launched its Objectworks software development system for AT&T’s C++ Release 2.0: the integrated set of object-oriented tools include an incremental compiler and linker, source level debugger and source code browsing, and will be available in late August for Sun Microsystems’s Sun-3 workstation, selling for $2,500."
(in reference to the pain of using multiple languages on a project)
"At Meta we have a lot of projects that use React frameworks, you've got javascript here, and then you have C++ for real work, and you may have Java interfacing with some other part of the Android system, and those are all kinda horrible things."
Of course just a few minutes later Carmack says that garbage collection is unequivocally a good thing for most programs, so C++ pros shouldn't puff up their chests too much!
I think you're misinterpreting that and/or quoting it out of context.
He was going on about how many different languages were in use. The "horrible thing" he's referring to is the variety, not that they aren't all C++.
The "real work" he's referring to is performance intensive work and I think his phrasing makes it clear that he like C++ but he gives a lot of credit to modern languages for their ease of use as well.
Agreed - I hoped my parenthetical would get that across but it's worth making it clear that Carmack isn't against different programming languages. He rather feels that being forced to use multiple languages on the same project brings extra pain.
The thing is, he also complains about a time at Oculus where he needed C++ experts to right the ship but Facebook didn't have the talent base for it. Since he says 99% of code should be written in GC'ed languages, it shouldn't be surprising there's a talent vacuum in C++ among modern programmers.
I listened to a few episodes but can’t help sensing some form of shallowness. The oliver stone episode was particularly disappointing. He basically talks to all celebrities within his reach about whatever he feels the want to talk about.
John Carmack is and will always be one of the greatest programmers for me. His drive, motivation and work ethics are also very inspiring. Whenever I feel low the first thing I tried (sometimes failed) to pull me out is reading a few chapters of "Masters of Doom".
I'll probably never be as productive as he has been, but I appreciate the games and impact he made.
I also enjoy this podcast because this is one interview of Carmack that I can almost fully understand :P
As someone who grew up playing Doom II along with Quake I/II/III I can't thank you enough for name dropping that book.
I've always followed id software's games but never really knew their origin story or the full history of how it all came together. I'm about 2/3rds of the way done with the book which I've read in 1 sitting so far. I felt compelled to stop to make this reply. This is one of the best books I've ever read in my entire life.
Thanks. I am into programming but it's more focused on web development. Gaming is how I got introduced to web development tho, specifically making Quake clan sites and being a part of a Quake III ladder site.
I finished "masters of doom" a month ago from a HN recommendation, was fantastic and I see a lot of their organisational patterns repeat in the startup space
Yeah exactly, I do think that early ID (before Romero left or maybe before Quake) the golden standard of startups: Small, efficient, extremely smart, hard working and solving real, top-notched problems.
It's not about being a wall, or being uninteresting that gets me. Unlike most boring interviewers, he also positively makes me cringe. Like, I just can't.
I believe he is trying to emulate Prince Myshkin from The Idiot, but he has a profound misunderstanding
of what it means to be an Idiot in the way of the Prince, and so, he is only able to emulate the surface level behaviours but not the underlying identity.
I was watching his episode with Jaron Lanier, and because I loved what his guest had to say, I really had to push through all the cringe and force myself to watch it. The same thing has happened multiple times. Whenever I see an amazing guest on his podcast, I feel sad that I'll either have to skip it, or go through a painful experience.
It would be amazing if there was an alternate edited form of his podcast, where Lex is replaced with someone else who fills in whatever Lex is talking about in a concise way, just to keep the flow going. I would definitely watch something like that, solely because of his guest lineup.
Edit: I feel like this comment might be interpreted as too harsh, but just to clarify, I am not attacking his person. Interviewing is a skill, and I'm only criticizing his skills. If I tried interviewing, I might suck as well, but I wouldn't expect the kind of guests he gets.
I thought this was a fantastic podcast. I'm not sure how so many people can hate Lex and his (free!) work. I've been waiting for some of these interviews (with a technical interview for years/decades). I'm a bit confused why you'd want to speed then up or skip them, surely this is missing the point of settling into a thought provoking conversation? I happened to have a longish drive yesterday for a day of walking in the hills and listened to the whole thing in 2 sections. Never felt bored and had quite a few interesting thoughts come up during the conversation. Thanks Lex!
I never get that argument about not listening faster. Do you slow things down to .75x when you want to focus even more? I will get far more out of the conversation by listening to it 2 times at 3x speed than I will listening to it once at 1x.
Can't wait to find out what the heck he's been working on in AI! Seems to me like he would be best at scaling to extreme model size and making efficient use of available hardware, but it would be too expensive to keep up with the corporate labs as an independent researcher. And from what I've seen before it seems like his opinion is that scale might not be required.
FWIW, Lex is good about including a ToC, and when you have that, you can use the dropdown menu to open up YouTube's neural-net-generated transcripts (which are surprisingly good but also deliberately extremely literal and badly formatted) to read that instead of spending 5 hours listening to it. (This is what I did to quickly check out his comments on Gato etc.)
5 hours?! I'm diving in (I try to fastforward when Lex talks). Carmack is one of the few people that actually inspires me a bit - like I listen to him and think "okay, I should be writing way more code than I do".
Lex Fridman misrepresents his role at MIT to make himself sound more accomplished than he really is, and the way he went about putting out his self driving car research directly to the press instead of going through peer review is shady. The guy is a cringeworthy grifter. He’s trying to be viewed as some AI expert but he’s totally full of shit. Elon made him famous and he got the Joe Rogan push. Absolutely droning interviewer in need of media training.
The MIT directory lists him as a "Research Scientist"[0] and he has solid set of papers he's published on self-driving cars and AI[1]. This is completely consistent with the way he describes himself: "I'm an AI researcher working on autonomous vehicles, human-robot interaction, and machine learning at MIT and beyond"
Attacking someone's character just because their name is mentioned and you don't like them is pretty boring conversation.
If you're going to drive-by crap on someone, at least recommend podcasters that you do like assuming there are any, especially those that might appeal to Lex's audience.
We've banned this account for repeatedly breaking the site guidelines (and I do mean repeatedly). Please don't create accounts to do that with—it will eventually get your main account banned as well.
If you don't want to be banned, you're welcome to email hn@ycombinator.com and give us reason to believe that you'll follow the rules in the future. They're here: https://news.ycombinator.com/newsguidelines.html.
How does he misrepresent his role at MIT to sound more accomplished? I have watched a ton of his episodes and he hardly mentions anything about himself. He's fairly humble overall, especially compared to other podcasters in this space.
Yes. He may be a bit of a gray museum wall, but I don't go to the museum and think "gee, a pink polka dot background would really set off this Rembrandt painting."
He doesn't get in the way, and that's more than can be said about a lot of people vying for attention these days.
How did he misrepresent his role at MIT? He never claimed to have gotten a degree there and he never claimed to be a professor there. He taught a course there though, as an independent researcher.
I don't think he got famous through Elon Musk, but maybe through Joe Rogan to some degree (he already had hundreds of thousands of views before his first episode with Elon Musk.)
How exactly is he a grifter? Sure he runs ads, like almost any podcast, but I don't think he's trying to sell anything or so.
His interviewing style is peculiar, but that's a matter of opinion, and you're free to not watch his show. He undeniably has a lot of very interesting guests on, and that's probably the main reason why people watch his show.
I agree with you. I think OP is way off the mark here. I actually think Lex is super thoughtful and humble. Always giving praise to people and never bringing much of his academia efforts.
He turned his youtube lectures channel on self driving into "MIT AI Podcast", which helped him get a ton of top ML people on and then pivoted that into "Lex Fridman Podcast" with a lot more IDW style crackpots mixed in, lending them credibility that they don't deserve.
No, you are not talking about the institution here.
Lex just used the institution as a platform to self-promote
himself. Nothing intrinsically wrong with this, but Lex
isn't the most sophisticated brain at MIT.
There are smarter people there, less loquacious, but more powerful.
Then again, MIT also has evil people in power,
like previous MIT Chancellor Eric Grimson, who killed Aaron Schwartz
and betrayed the principles of the institution by secretly passing information on to the secret service. Grimson was still able to hold on to a title of Chancellor that pays him more than 400K USD per year!!! INSANE!
You are talking about Stephen Wolfram and his pothole self promoting
things. Wolfram is also a very dangerous individual. he is feeding off ignorance, and corrupting young minds.
Could you point me to an example where they did a press release about a model from the (approx) 70s which they reimplemented in PHP? I mean anything like that, not this specific example. Because that just sounds so outlandish to me.
They can transmit thoughts directly into your head... via a computer monitor.
They call it a "a brain-to-brain interface", doesn't that imply one brain is wired directly to another electrically?
It opens with "The ability to send thoughts directly to another person’s brain is the stuff of science fiction. At least, it used to be." ... again, via a computer monitor.
I remember when they billed kids fucking around with shape memory alloy as something really grandiose, I'll see if I can find that and come back and edit this post to include it if I do.
> They can transmit thoughts directly into your head... via a computer monitor.
> They call it a "a brain-to-brain interface", doesn't that imply one brain is wired directly to another electrically?
The actual paper summary is better:
> Two of the three subjects are "Senders" whose brain signals are decoded using real-time EEG data analysis to extract decisions about whether to rotate a block in a Tetris-like game before it is dropped to fill a line. The Senders' decisions are transmitted via the Internet to the brain of a third subject, the "Receiver," who cannot see the game screen. The decisions are delivered to the Receiver's brain via magnetic stimulation of the occipital cortex.
I don't think "a brain-to-brain interface" implies wiring the brains together at all. To me it means transmitting information that bypasses the traditional human communication channels - and this certainly does because it's read via the EEGs and then sent to the reciever via magnetic stimulation.
There is no information sent between people via the screen. The senders see the full game, and send their decision via magnetic stimulation to the receiver.
The "flashing lights" are ONLY seen by the senders, and used to stimulate the senders brains to send signals that can be read by the EEG. From the paper:
Senders focused their attention on a “Yes” LED light flashing at 17 Hz placed on the left side of their computer screen; to
convey a "Do Not Rotate" decision, they focused on the “No” LED light flashing at 15 Hz placed on the right side. These LEDs
are depicted as circles attached to the screens in Figure 1. The cursor position provided real-time visual feedback to the Senders.
The direction of movement of the cursor was determined by comparing the EEG power at 17 Hz versus 15 Hz, with a higher
power at 17 Hz over that at 15 Hz moving the cursor towards the left side near the “Yes” LED, and vice-versa for the "No"
LED
The decisions of the two Senders were sent to the Receiver’s computer through a TCP/IP network and were further translated
into two pulses of transcranial magentic stimulation (TMS) delivered sequentially to the occipital cortex of the Receiver. Each
TMS pulse lasted 1 ms. An eight-second delay was enforced between the two pulses to remain within the strictest safety
guidelines of TMS stimulation29. The intensity of the stimulation was set above or below the threshold at which the Receiver
could perceive a flash of light known as a phosphene: a “Yes” response was translated to an intensity above the threshold, and
“No” was translated to an intensity below the threshold. During each round of trials, the Receiver always received the decision
from one Sender first, then the other. The screen the Receiver saw also had visual prompts to remind them whose decision
the current TMS stimulation was conveying. Receivers made their decision based on whether a phosphene was perceived and
conveyed their decision (rotate or do not rotate) to the game using the same SSVEP-based procedure used by both Senders.
After the game state was updated, the trial moved into the second round and the above process was repeated.
So there was no physical light flash for the receiver - their brain was just stimulated so they would perceive one.
They created a device which they can successfully and safely move through blood vessels in the brain without causing injury and you characterize it as "Some college guys move coated wire with magnets"?!
I mean yes. Sort of like saying a covid vaccine is "some people mixing chemicals in test tubes".
The researchers also tested the thread in a life-size silicone replica of the brain’s major blood vessels, including clots and aneurysms, modeled after the CT scans of an actual patient’s brain.
So no, not a brain. Silicone tubes. On a table somewhere.
They're happy to talk it up like its an actual in-vitro human brain though.
It's much easier to criticize someone rather than to put one's self in their shoes and create something that was not there before, which many people consider to add value to their lives. Have you stopped to think why that might be?
Facebook is dying, so it seems less important now, but I'd have liked to hear in his own words how he felt supporting a company with such questionable ethics.
He can meaningfully retire on facebook money. Good for him and I have no requirement for him to publicly criticize them. Also, facebook is the main company plowing money and top engineers into AR/VR. They also created react and pytorch. Facebook has had a massive impact on software.
Sure but the question isnt how did he decide to fortify a successful company with his reputstion, but an unethical one. Successful does not mean ethical. They're so unrelated as to be orthogonal.
The poster suggested joining facebook was a new low in character for Carmack. So this is hardly "whataboutism"
> are you seriously trying to argue that Facebook is as unethical as id software
I guess it depends on what your concerns about Facebook are. Did id treat employees well? Did it produce products that made that world a better place? In the 90s there was just as much bad media and controversy surrounding Doom and violent video games as there is around Facebook.
Just to clarify, I am not trying to compare them point for point. Just surprised that anyone would take issue with Carmack participating at FB. The guy is primarily known for making edgy entertainment products, not elementary schools. I don't have a problem with either.
Do you honestly think there will be an explanation other than "they offered me a bunch of money to do $X"? Smart people aren't required to be self-righteous saints sacrificing themselves for the good of everyone else.
I don’t think this will actually be his legacy, many of the world’s famous scientists and engineers worked for companies, empires, and a variety of unsavory groups. I mean Charles Darwin was part of a scientific expedition for the British military, which at the time contributed real world misery and suffering on a global scale but this rarely comes up when discussing evolution.
Maybe this time will be different or maybe after a few hundred years we trim details until it becomes a digestible snippet where you never mention the bad side.
Zuckerberg is probably the guy who will get figuratively crucified for the abusive shit Facebook did. Which is why it's crazy someone bright would be like, 'oh me too!'
I assume he sees it differently. I'd like to understand.
For 99% of all people in the world, Zuckerberg will always be known as a brilliant mind who created Facebook - the social network. Right up there with Einstein and Elon Musk. They don't know about any abusive shit, don't want to know, and will never know.
Ford and Edison are universally recognized as American greats - a genius businessman and a genius inventor. I don't know anything about any abuse they might have done, and I suspect 99% of Americans don't either.
Edison publicaly electrocuted elephants in an attempt to paint AC, his competitor, in a negative light. He also colluded with other light bulb makers on the planned obsolescence of light bulbs.
Ford wrote the International Jew and received the highest honour Nazi Germany could give. Like Facebook, he is responsible for inspiring hatred in a large part of the world. https://en.m.wikipedia.org/wiki/The_International_Jew
You're missing the point - 99% of people don't know this and will most likely never know. These two men will be forever known as greats. In fact, Ford's legacy is probably so significant that learning about his anti-Semitic views might actually influence some fraction of American population - not in the direction you hope for. You don't know my friends, sure you want me to share this with them? Be careful what you wish for.
I've seen Edison come under some criticism for taking sole credit for what was largely the work of his employees. That's a story as old as time though, and a good number of people think that's the way things should be, so probably not too much of a detraction.
To me his legacy might be something like “Godfather of VR. He set the bar high and pushed the market to strive for it. For some reason he handed the keys to the wrong people before it really hit big, but thankfully he also gave us _____ and _____.”
While I hope we get the closure of finding out why he sold to FB, I think he still has much to bring to the world.
He certainly can be too. There’s no rules about there having to be a single grandfather.
What’s important to me about Carmack is that he was unwavering about the need for high fps and low latency through the entire pixel pipeline (as well as a lot of other things), and that he brought the skills to make it possible.
A VR industry based on a common denominator of 60fps with 10ms panel response times was very close to being the reality (and IMO that industry would have killed VR’s adoption for at least another decade)
It's not really comforting to know that Facebook is the reason the browser's find feature doesn't work on any web site, no computer is capable of running google chrome to browse a microblogging web site at 60fps, and modal forms commonly disappear losing the data you've entered into them if you resize your browser so you can see another window from which you are retrieving the data.
Implementing "responsive UI" using exclusively JavaScript and CSS-in-JS (rather than using CSS or not doing anything) was naughty and kludgey in 2010. The fact that it's now the "best practice" seems to have increased the rate at which forms delete themselves and the data users have typed in when the user makes the mistake of resizing the window (causing the user to briefly wonder whether computers have on the whole been worthwhile compared to the alternatives offered by mechanical printing presses, abacuses, papers, quills, and if one is feeling provocative, a typewriter, before repeating all the labor that they previously performed for their silicon master before it decided to unceremoniously discard it)
Preventing the browser's "find" feature from searching the text on the page (causing the user to tear their hair out, bash their computer with a rock, borrow a wilderness survival book from the local library, drive to the countryside, abandon their vehicle at a rest stop, and renounce civilization once and for all) is a technique called List Virtualization and you can learn how to do it in React here: https://www.patterns.dev/posts/virtual-lists/. react-virtualized has over 1 million weekly downloads on npm.
The benchmarks I was able to find comparing identical operations in different web frameworks look pretty bad for React as well.
Regardless, react "solved" front end UI development. It has reduced churn in frameworks, and essentially determined how to handle state, render components on a loop. It was a breakthrough in web technology.
I don't like Facebook Messenger but it is still used as a top or main communication by so many Silicon Valley friends and network of mine. All products are dying but not instantly right.
Carmack can go wherever he wants, do whatever he likes. So I just assumed he went to Facebook for the opportunity to work alongside other smart people, to work on his personal passion project. Who wouldn't do that, if given the chance?
Lex Fridman's interviews are the only reason I'd put up with a longer commute. They are super dense and super long. Like listening to an audiobook every few days.
The Tony Fadell interview was also interesting. A great review on the history of iPod and iPhone and what he was up to at Nest.
I've never understood sense of those talking heads type of podcast. You won't get anything useful out of it and it's not even entertaining. Listening to smart people talk doesn't make you smarter.
Why would it not make you smarter? Having a hard time following this logic. Perhaps not significantly, but if they say something you don’t know, then by definition you just got a little smarter, if only by one fact about one topic.
Also, I think a lot of ideas are presented black and white but podcasts like these show the intricacies of the gray area
> Listening to smart people talk doesn't make you smarter.
I don’t listen to standups thinking it will make me smarter, I don’t watch football thinking it’ll make me more athletic. Why would you assume people listen to smart people to become smarter and not just for the entertainment and informational reasons that are straight forwards reasons?
Likely they will say something that happens to make contact with some other thought you picked up somewhere else, and you'll make a new mental connection between concepts and that expands your mental map of the world.
It's always good to understand the world a bit better.
I just discovered this when I went to the store, and will probably take a nice long walk in the evening/night heat to devour at least part of this delicious release. What a nice gem to randomly receive. I'm curious to know what Carmack has been doing especially in regard to AI.
I've thought about hosting a podcast with a similar format to Lex's and yet I don't listen to podcasts much so I don't know what's missing from the scene.
If there were another one like Lex's, how would you want it to be? What would you want to be the same, what different?
Like many here, I have mixed feelings regarding Lex.
Last time I heard him, he started the podcast saying that he “fears nothing and no one”, which of course means he's either extremely immature or a psychopath. I'm betting in the former.
For whatever reason, he manages to get great guests, so I keep listening. I wouldn't say I tolerate him because he mostly doesn't get in the way of the interview much. Which is good, but the bare minimum for a good host.
I used to listen to Rogan a few years ago, but it has become unbearable to me and the Spotify exclusivity deal was the push that I needed to stop completely.
With Lex, I keep coming back, reluctantly, so I guess he's got something going with his show.
Edit: oh, and Carmack is a legend, I'll listen to him with almost any interviewer.
The fact that he's opening a podcast with that, lets you know that he's talking about "fear" in a very narrow sense: of how he questions conversationally, and how he chooses podcast guests. It doesn't mean he's a psychopath.
There’s no such thing as no fear, no matter how narrow you constrain its scope. It’s ignorance, hubris, naïveté, dishonesty, etc. Anything but bravery.
At around 17:30 John talks about Go in a way that convinces me that he understands the grugbrain mentality. He's definitely a pragmatic person who sees the value of simplicity in language design, particular when programming at scale.
Not so sure. On a technical level at least, everything that I have read from him has been very concise and pragmatic. His programming advice is always to be as simple as possible because simple is fast. For example, the one I always remember is that in order to optimize that you should try to remove as many "edge case optimizations" as possible, so that everything runs always the same way, and then optimize _that_.
I do value simplicity very much, but this doesn’t seem true. There’s a reason why musl is slower than glibc, st is slower than xterm, OpenBSD is slower than Linux etc. It’s great if the optimization via simplification works out like that, but usually it doesn’t.
id tech 3 is pretty elegantly organized, not to mention Carmack's contributions with bsp, netplay, finvsqrt, sss, megatexturing, presumably the current iteration of predictive rendering used by oculus kit. He's someone who knew the technology and the business top-to-bottom, and likely still does.
Carmack is more like an academic than a smoke and mirrors exec.
Having sat through as many as 9 keynotes in-person before that relationship went up in a blinding flash, I'll say I look forward to listening to this podcast in full.
What exactly do you mean? I strongly feel that Lex Fridman is a canonical anti-intellectual (or pseudo-intellectual, "in a sense" :P ), and I don't think that's unreasonable to point out on HN.
There is a lot of -ve feedback on Lex.
I dont want to dwell on it, instead, provide another end of the spectrum.
Stephen wolfram. The wolf man.
completely insane. So much arrogance, hubris and self promotion makes me stop watching instantly.
Edit to add: His recent episode with Jack Barsky (former KGB spy) was exceptional and worth a listen if 5hrs is too long of a test case