00:00SPEAKERI mean, you know, it's October, I feel like it's time to talk about a terrifying subject, right?
00:04SPEAKERYeah! Do you think this is terrifying? I think it's probably going to be okay.
00:07SPEAKERI don't know, I feel like everything is terrifying now, so yes, by default.
00:12SPEAKERI mean, that is the defining characteristic of our time, is a sense of impending dread.
00:16SPEAKERI mean, I don't know, I'm like half optimistic. I kind of think that maybe this might be good for
00:20SPEAKERus. Maybe we shouldn't be in charge of ourselves. I've read a lot of speculative fiction where
00:25SPEAKERthey're like, well, we let the computers take over and then shit started going really well.
00:28SPEAKERSo, you know, it's not all terminators until everybody forgot how to maintain them and then
00:33SPEAKERthey ran away with themselves. Look, if the computers can't maintain themselves,
00:36SPEAKERthen they shouldn't fucking be in charge. That's number one criteria.
00:39SPEAKERMay you live in terrifying times, that's the saying, right?
00:42SPEAKERIt's fucking Confucius, what a dick.
00:58SPEAKERHey, Brad.
01:06SPEAKERHello. Good morning.
01:07SPEAKERWe made a tech pod. This is the seventh episode.
01:10SPEAKERThis is a monument to our technical expertise.
01:13SPEAKERWe have a week worth of podcasts now. If you start on Sunday,
01:15SPEAKERyou can finish on Saturday. We're done. This whole run is complete.
01:18SPEAKERJust put them on a loop. 24/7.
01:20SPEAKERThis is Brad and Will Made a Tech Pod. We talk about one topic every week and then
01:24SPEAKERsometimes some emails and maybe other things. This week's topic is the Singularity.
01:33SPEAKEROh my God. This is terrifying.
01:34SPEAKERNo, no, no. Look, man, everybody thinks this is terrifying, but it doesn't mean...
01:45SPEAKERAnd then a bunch of walking on skulls. Although, you know,
01:47SPEAKERI heard something really interesting yesterday. Did you know that the theme for the Terminator
01:53SPEAKERwas... It has that weird cadence because the guy who made it was an improv guy and
01:59SPEAKERhe set up the synthesizer wrong, that did the beat for the...
02:03SPEAKERWait, is that really why?
02:04SPEAKERUnintentional.
02:05SPEAKEROh my God. I had no idea. Yeah. It's got some insane time signature. It's like, what is it?
02:09SPEAKERIt's like 1514 or something like that.
02:12SPEAKERHe literally started playing like banging pots and stuff together to make the music sound,
02:18SPEAKERthe clinging metal sounds. And by the time he had finished recording it, he didn't do notation
02:24SPEAKERand stuff. And apparently the people who got it listened to it and were like, what the hell is
02:28SPEAKERthis? Oh, wow. This is... But it really works because it's like the combination of analog
02:32SPEAKERand digital, like working together and not working together. I don't know. I thought it was...
02:37SPEAKERHe shouldn't have told that story because I thought it was an intentional stroke of genius,
02:40SPEAKERthat he made it sound like this kind of like thudding mechanical assembly line,
02:45SPEAKERindustrial kind of sound. Well, I mean, that part was intentional for sure,
02:49SPEAKERbut like the weird cadence is the accidental... I love it when I hear stories like that,
02:55SPEAKER'cause it's just like, hey man, that's art, right?
02:58SPEAKERIt is.
02:59SPEAKERAnyway.
02:59SPEAKERIt happens.
03:00SPEAKERThe singularity.
03:01SPEAKERArt is not a thing you do. Art is a thing that happens to you.
03:05SPEAKERYeah. Yeah. Okay. Yeah. Artist intent, straight to hell. Go straight to hell, artist intent.
03:10SPEAKERThis is a death of the author style podcast, just so people know.
03:14SPEAKERSo the singularity, Brad.
03:16SPEAKERYeah. Speaking of authors, I guess that's the place to start,
03:18SPEAKER'cause this topic is very speculative and that may or may not have happened already,
03:22SPEAKERand it may happen again.
03:24SPEAKERYeah. So, well, okay. So the idea of the singularity is that it's just...
03:28SPEAKEROf any singularities, it's a point in time beyond which you can't predict what's going to happen
03:33SPEAKERafterwards, the future beyond that point. So like black holes are singularities 'cause we
03:38SPEAKERdon't know what happens on the other side of them, right? If we fire you into a black hole right now,
03:42SPEAKERBrad, what's gonna happen to you?
03:44SPEAKERYou're right about that. I always thought that that was more a reference in that case to the
03:48SPEAKERphysicality of them, which is a singular point in space.
03:51SPEAKERWell, but they destroy the information,
03:53SPEAKERthe structure and information. They're like infinite entropy too.
03:56SPEAKERSo I guess it works both ways, right? Like it has kind of a double meaning.
04:00SPEAKERBut that's not the kind of singularity we're talking about today.
04:02SPEAKERNo.
04:03SPEAKERNo. We're talking about the computer singularity, the technological singularity.
04:06SPEAKERThat's the actual classical term, right? Technological singularity. I mean,
04:09SPEAKERthat's the tech is the context in which it is used typically.
04:12SPEAKEROriginally, according to Wikipedia, this was a concept originally created by John von Neumann.
04:17SPEAKERI'm gonna fail my sixth grade book report on this one, citing Wikipedia.
04:21SPEAKERI had always heard Raymond Kurzweil attached to this, but did he just pick it up?
04:25SPEAKERYeah. He's the guy who got really into it. He's the one who's changed. Like,
04:29SPEAKERI think probably Kurzweil is maybe the first person who... Well, he wrote a book called
04:34SPEAKERThe Singularity is Coming about 15, 20 years ago now. And he has actively adapted his life to try
04:43SPEAKERto live long enough to reach the singularity. And then in his conception, the singularity is
04:48SPEAKERlike, why not? We'll be able to upload brains.
04:50SPEAKERSo is he just one of these, I'm terrified of mortality and I want to digitize myself
04:54SPEAKERand live forever types?
04:56SPEAKERAren't we all really at the end of the day?
04:57SPEAKERI mean, I guess.
04:58SPEAKERWould you upload yourself?
05:00SPEAKERIs that really you in there or is that just a copy?
05:03SPEAKERLook, man, when I go through an elevator, is that me on the other side? I don't know.
05:08SPEAKERWhat about every time you go to sleep?
05:10SPEAKERYeah. Am I dying every time I go to sleep?
05:12SPEAKERI definitely am.
05:13SPEAKERIs this even my hand I'm seeing, man? I don't know. Is that red? Do you see red the same as me?
05:18SPEAKERMaybe. I don't know. We know people who don't though. Brains are weird.
05:22SPEAKEROkay. So if the singularity is a concept in technology where everything changes
05:28SPEAKERunpredictably on the other side, I mean, could you credibly argue that multiple
05:32SPEAKERsingularities have already happened throughout history?
05:35SPEAKERLike say the printing press. I don't know. Would that rise to that level?
05:38SPEAKERWell, so printing press is a weird one. I mean, because it did dramatically expand
05:41SPEAKERthe amount of human knowledge in the world, right?
05:43SPEAKERRight. Not just the amount of knowledge, but the way it was disseminated into the number of people.
05:47SPEAKERYeah. The printing press is a tough one. I think you could argue that the rise of
05:52SPEAKERthe modern smartphone has maybe created a couple of singularities, right?
05:56SPEAKERDefinitely. I think we're absolutely in that territory in the last 20, 25 years,
06:00SPEAKERbut I'm trying to broaden our conception of what this term may actually mean.
06:05SPEAKERWhat was the first one?
06:06SPEAKERLike maybe the printing press, like maybe, I don't know, the steam engine or something around the
06:12SPEAKERindustrial revolution. I don't know. Before those events occurred, I don't think you really could
06:16SPEAKERhave foreseen what the world would look like after, right?
06:18SPEAKERMan, that's a really good question. Yeah. The invention of the steam engine and the
06:22SPEAKERindustrial revolution created cheap goods and services.
06:26SPEAKERRight. It created a commodification of goods and services. It also created a level and a type of
06:33SPEAKERpollution that probably nobody saw coming.
06:37SPEAKERYeah. Could you predict global warming from the first steam engine? No, absolutely not.
06:41SPEAKERAnd it probably created a level of wealth inequality that was previously
06:45SPEAKERunforeseen, right? I mean, obviously you had monarchies and divine right of kings and conquerors
06:50SPEAKERand stuff, but I think the ability of industry to concentrate wealth in individuals was exponentially
06:56SPEAKERgreater than anything before that, right? Yeah. I mean, I think so.
07:00SPEAKERI'm just, this is just off the top of the head. I'm not an expert here, but.
07:03SPEAKERNo, but I mean, so in that context, can like the invention of antibiotics,
07:07SPEAKERis that a singularity too? That's a good question.
07:09SPEAKERBecause it dramatically changed lifespan. It did, yeah.
07:12SPEAKERWhich increased population, which had all sorts of other unintended side effects.
07:15SPEAKERThat's totally, yeah. I saw a breakdown, like an infographic kind of thing of not just life
07:19SPEAKERexpectancy and mortality rates from the beginning of the 20th century, but also like the breakdown
07:24SPEAKERby cause of death. And not only did way more people die back then, but also the majority of
07:28SPEAKERthem were of what we now consider like easily curable diseases, you know, infections and stuff
07:33SPEAKERlike that. Like the consumption has to be high on that list, right?
07:36SPEAKERYeah, probably. I'm sure. Yes, I'm sure that was.
07:38SPEAKERBrad, I've got the consumption.
07:42SPEAKERSorry, we're not doing voices on this podcast. But in the modern context, I mean,
07:47SPEAKERI think you could argue that the smartphone and maybe even the addition of like the moment
07:53SPEAKERthat everybody started carrying cameras around in their pocket have both been singularities
07:57SPEAKERin that sense. But I think when we talk about the modern technological Kurzweilian-Werner-Vinge
08:03SPEAKERsingularity, we're talking about the moment at which like there's more brain power in computers
08:12SPEAKERthan in brains. It's something more outlandish than we've seen yet, right?
08:16SPEAKERWell, kind of. I mean, I don't know.
08:18SPEAKERBut just to go chronologically, I guess, just to work up to the future.
08:22SPEAKERYeah.
08:23SPEAKERLike if you're trying to identify, excuse me, the A or the singularity that has already
08:27SPEAKERoccurred in modern times, like where's that line? Because you could say it was the microprocessor
08:33SPEAKERor the internet or the, you know, the smartphone or social media, like all the...
08:39SPEAKERWas it AOL?
08:42SPEAKERIt was definitely all those CDs, the free trials and cereal boxes that destroyed society.
08:47SPEAKERYeah. They got more people online than anybody else for the first 10 years. I think it's the
08:53SPEAKERphone.
08:53SPEAKERThat's totally a valid argument. I guess the only thing I'm saying is that like the modern
08:56SPEAKERinformation age is kind of a continuum. Like all of those things contributed to and built
09:01SPEAKERon each other, right? Until it brought us to the society destabilizing point that we're at now.
09:06SPEAKERWell, yeah. I mean, if you look at the last 10 years, starting in around in 2009,
09:10SPEAKERlike the iPhone was a little bit of a slow roll and it wasn't until 2010 that you probably
09:16SPEAKERgot a billion people using smartphones, right? Maybe even 2012. But I mean, think about the
09:23SPEAKERamount of compute, think about the impact of everybody having like a really dope 2002
09:28SPEAKERera computer that they're carrying around in their pocket that also has access to the
09:32SPEAKERentirety of human knowledge all the time. And then we use that to send cat pictures
09:36SPEAKERaround basically. And yeah, be shitty to each other.
09:40SPEAKERYep. Animated kiss.
09:42SPEAKERYeah. Jeff Simeon, I think.
09:43SPEAKEROh, so yeah. I mean, we're clearly not... So one of the things that's interesting about
09:49SPEAKERthis topic to me is that it's one of those 10 year off topics. Everybody always is like,
09:54SPEAKERoh, well, for example, in 1993, I think, Vernor Vinge, who's a science fiction author,
10:01SPEAKERhe wrote a fantastic book called Rainbow's End about AR, augmented reality, which if
10:08SPEAKERyou haven't read, you should read. But he wrote an essay in 1993 called The Coming
10:13SPEAKERTechnological Singularity, which he basically was like, look, when this happens, it's going
10:18SPEAKERto be the end of the human era. And once the super intelligence comes online, it will start
10:25SPEAKERto upgrade itself and advance technologically at an incomprehensible rate.
10:29SPEAKERSo is he and that piece, are they the reason that Runaway AI is thought of as the classical
10:37SPEAKERexample of this concept?
10:38SPEAKERI think he's probably... If you think about computers gone bad in science fiction, there's
10:44SPEAKERlike 2001, right? The thing that happens in 2001 is that there's a logic gate that basically
10:53SPEAKERmakes the computer divide by zero and it goes crazy. This is him saying, hey, maybe the computer
10:59SPEAKERis getting smarter, isn't great for us. And I think he was probably the first one who
11:02SPEAKERdid that. But also it's worth saying, Vernor Vinge's... I don't think that in science
11:10SPEAKERfiction, especially the idea that the end of the human era is a bad thing is universal.
11:17SPEAKERSo I think it's Neil Asher wrote a whole series of books about basically human controlled
11:24SPEAKERcivilization where humans ceded day-to-day governance and stuff to the AI, because they
11:30SPEAKERcan take the long view in a way that humans are notoriously bad at. Neil Asher's English,
11:35SPEAKERI think, which maybe has some impact on that, because they have a kind of complicated relationship
11:40SPEAKERwith government. But yeah, like I said in the beginning, it's not all stomping on skulls
11:49SPEAKERand getting the machine gun cars out.
11:52SPEAKERIt's not a guaranteed apocalypse, but it might be.
11:54SPEAKERNo, it could be a gentle fun apocalypse. What if the computers just plug us into
12:00SPEAKERpods and they use us to generate electricity and they just put us in constant pleasure
12:05SPEAKERmachines, a constant state of full, just full serotonin release. So we're like doped out
12:13SPEAKERof our gourds the whole time and they give us doom 2020 or 2019. Like we get doom early.
12:20SPEAKERWhat's not to love there. I liked the version of that script better where they were using
12:24SPEAKEReverybody's brain power to make a gigantic parallel organic supercomputer and not just
12:28SPEAKERusing it for body for their electrical as batteries. Man, why didn't they leave that
12:33SPEAKERin there? That was so good versus really being dumb. Violating thermodynamics.
12:38SPEAKERAlso, maybe they wouldn't have been sued then. But anyway.
12:42SPEAKERSo I mean, but the other thing that's happening here is we are at a place where
12:47SPEAKERjust this morning, actually Google published that they've solved one of the big supercomputer
12:54SPEAKERchallenge in the form of a quantum computer, which I think we, I can't remember if we talked
12:59SPEAKERabout that on a show or off air recently, but I had thought until very recently that quantum
13:04SPEAKERcomputing was mostly theoretical and hadn't actually been put into practice. And then here's
13:07SPEAKERGoogle saying like, Hey, we've done it. And we're putting it to work crunching, you know,
13:11SPEAKERsolving algorithms, computations that would take millennia on conventional computers.
13:17SPEAKERIt's yeah, it's a, I can't remember the term, but basically they solved a problem of quantum
13:23SPEAKERsupremacy, which is a situation where the quantum computer is able to solve something that would
13:28SPEAKERtake an inordinate length of time on a traditional supercomputer. So they solved a problem in a few
13:34SPEAKERminutes that would have taken a normal supercomputer 10,000 years. Right. Right.
13:38SPEAKERIt's not a useful problem, but it's a problem that they could solve.
13:41SPEAKERSo it was a proof of concept, but I mean, not to get too far down the quantum rabbit hole, but
13:45SPEAKERis it the case that quantum computing excels at, I don't know how to put this like types of
13:51SPEAKERcomputation that are not really like practically useful for what we do day to day on computers?
13:55SPEAKERLike it's more of a like high level math kind of situation. It's it's that that part is unclear,
14:01SPEAKERlike right now they're not really useful for anything. Yeah. That's what I mean. Yeah.
14:05SPEAKERLike the solution that they solved is a problem that you don't need to know the answer to.
14:09SPEAKERRight. It's just a really complicated mathematical thing. Like it's not, it's not a, it's not a type
14:14SPEAKERof computation that's directly applicable to like your average desktop PC. Right.
14:19SPEAKERI, I, I would not expect us to have quantum computers in our desktop PCs in the foreseeable
14:23SPEAKERfuture. Yeah. That's kind of, that's what I meant. You're not going to need quantum computing to run
14:28SPEAKERray tracing on the PlayStation six is what I'm saying. But you might, it might be utilized in a
14:34SPEAKERself-sustaining learning, self-improving artificial intelligence. Maybe there you go.
14:40SPEAKERThat's the thing that you would use quantum computing for. Maybe it's if you needed it,
14:45SPEAKERI'll actually, I'll save this when we talk about quantum computing, but I, which I think we should
14:48SPEAKERdo soon because it's interesting and it's weird and it's really, really confusing and nightmarish.
14:55SPEAKERBut yeah, like the idea is that the quantum computer is presumably something that happens
14:59SPEAKERin our brain is quantum and is making is happening at a quantum level. And that's why we're conscious
15:04SPEAKERand you know, things without brains are not so. I see. It's one of the theories, I think that might
15:10SPEAKERjust be a science fiction theory. This is very heady stuff for this time of the morning,
15:14SPEAKERbut here we are. Yeah. I should have made more coffee. So, so yeah, like AI learning AI,
15:19SPEAKERlike self-aware sentient AI that has a sense of, you know, it's, it's sapient or sentient or
15:25SPEAKERwhatever has a sense of self, but also has the like insane amount of precision and speed
15:31SPEAKERas any other computer. Like that's the nightmare. Well, maybe not nightmare, but that's the
15:35SPEAKERclassical technological singularity, right? Yes. As soon as I guess what I'm saying is like
15:40SPEAKERto put it in plain language, like as soon as it becomes aware enough of itself to direct its own
15:45SPEAKERimprovement, the directing its own improvement, I think is the key. That's when we're through the
15:49SPEAKERlooking glass to use the metaphor. My immediate inclination is to say, what do you think would
15:53SPEAKERhappen after that? But the whole point of this conversation is that nobody knows, right? Well,
15:56SPEAKERit's, it's, it's unpredictable. I think. I mean, so one side of this is like the war games,
16:05SPEAKER20, I always tend when I talk about this stuff, I always end up framing it in terms of popular
16:10SPEAKERculture, because that's what people are familiar with. But if you think about like the Whopper,
16:14SPEAKERthe war games computer, it, it, it didn't go out and, you know, grab all of the other computers
16:21SPEAKERon earth and, and root them and make them its own and use them for its, for its nefarious purposes.
16:27SPEAKERThat's a, that's, that is a thing that seems difficult to impossible from, from like a modern
16:33SPEAKERstandpoint, but on the other side of it, like, I wouldn't be surprised if the singularity ends
16:39SPEAKERup because somebody wrote a custom ad algorithm that goes out and learns everything about
16:45SPEAKEReverybody. And as a result, gains consciousness and destroys the world. I mean, the whole,
16:49SPEAKERthe whole problem with an AI like that is that it has the amount of, it has, it has the technical
16:53SPEAKERwherewithal to like solve any problem it encounters almost instantly. Right. Like presumably
16:57SPEAKERif it has enough computational power to become self-aware in the first place, then it can like,
17:01SPEAKERyou know, we can, it can write custom interfaces to take control of any other system it encounters
17:06SPEAKERin like a split second. Right. Like it doesn't need, does that make sense?
17:10SPEAKERWell, that's the theory, but, but like the, I mean, so the thing about brains and the
17:14SPEAKERthing about our intelligence that's become clear over the last 10 years is that we're singular,
17:19SPEAKERlike our brains are, or maybe not our brains, but high end mammal brains are singular in evolution,
17:28SPEAKERlike in the universe that we're aware of. They don't - the amount of computing we have capable
17:35SPEAKERas a species, we have available as a species is not - we still aren't sure that it's comparable,
17:43SPEAKERlike I'm talking about not just in supercomputers, I'm talking about in every device and every IC
17:47SPEAKERwe've built as a species, we're not sure that it's equivalent to like one human brain.
17:51SPEAKEROh wow.
17:52SPEAKEROr even a dolphin or a monkey brain.
17:54SPEAKERYeah.
17:55SPEAKERRight. Because it turns out we're building abacuses and on the other hand, there's this,
18:00SPEAKERand we're trying to, we're trying to turn a bazillion abacuses and abacai, sorry,
18:05SPEAKERinto one big, big thinking thing. You know, whereas the thing with a few trillion neurons
18:13SPEAKERin our melons, it turns out is just as good as all of the abacai we've built as a species.
18:20SPEAKERSo there's no direct correspondence between a neuron and a transistor is what you're saying?
18:24SPEAKERYeah, no, because neurons are multipoint, like transistors are point to point.
18:28SPEAKERYeah, it's on off.
18:30SPEAKERSignal goes in, it's either a one or a zero. Neurons, honestly, there's so little we understand
18:35SPEAKERabout how the brain works on the day to day that there's still questions about what the role of
18:41SPEAKERneurotransmitters are and whether the electricity is the thinking, the neurotransmitters are the
18:45SPEAKERthinking, the hormones, like what all is happening inside your brain? We don't know.
18:49SPEAKERFrom what I know, neural pathways are like created all the time, right? They're based on behavior and
18:55SPEAKERexperience and so forth. And it's not like a computer can sit there and create new silicon
18:58SPEAKERpathways and it's hardware and so forth. Well, so I'm almost done with Neal Stephenson's new book,
19:05SPEAKERThe Fall or the Dodge in Hell, I think is what it's called. And it's basically the story of
19:12SPEAKERuploading people and what happens and how a brain would make sense of just being in a computer.
19:21SPEAKERHe doesn't take, as Neal Stephenson is wont to do, he does not take the simplified,
19:26SPEAKER"Well, Jeff Bridges got sucked into the computer and now he's in Tron"
19:33SPEAKERapproach. He starts with literally the first person being uploading, having to recreate a world
19:40SPEAKERthat their brain can make sense of. Because at the end of the day, our brains are just a collection
19:45SPEAKERof senses and a processing engine to turn those senses into concepts and thoughts. And it's a
19:53SPEAKERfascinating read. I highly recommend it. I was just going to say, it's like so Neal
19:57SPEAKERStephenson for you to sit here and be able to articulate this abstract concept that the book
20:01SPEAKERis about, but not fully remember what the title is. That's very Neal Stephenson.
20:10SPEAKERBasically, the mind, the human mind, the consciousness is such an exotic and mystical
20:16SPEAKERconcept that it's hard for me to imagine how that can be directly translated into something as rigid
20:20SPEAKERas software. No, no, I was just going to say, but we're literally at the dawn of a new era of
20:26SPEAKERcomputing. The quantum computers work. There's some, just to be clear, there is some question
20:32SPEAKERfrom other companies making quantum computers, whether Google's result is valid or not. We'll see
20:37SPEAKERthat'll work itself out in the court of public opinion. But there hasn't been a major change in
20:42SPEAKERhow computing works in 40 years at this point. We've been on a basically through line from the
20:50SPEAKER8086 on up to your 9900K and GPU, and we're all using the same underlying technology. The way
20:58SPEAKERthe quantum computers, like literally the same lingering instruction set, right? Still hanging
21:03SPEAKERaround in there. Yeah. You can run code that ran on a computer that I would have bought when I was
21:07SPEAKEReight years old on a computer that runs today. This is a completely new thing, right? So who
21:16SPEAKERknows what the other side of this looks like? If the growth rate for quantum computers is even
21:25SPEAKERclose to similar as it was for silicon based computers, who knows where we'll be in 10 years.
21:30SPEAKERSo you hear, I don't know if this is much of a digression, but you hear a ton about like neural
21:34SPEAKERnets and learning algorithms and that kind of computing these days. Does that have any direct
21:40SPEAKERcorrespondence to the type of stuff we're talking about? I have a vague kind of intuitive understanding
21:46SPEAKERof what that stuff is just from kind of the way it's described that I assume it's like the type
21:49SPEAKERof algorithm that can intelligently examine what it's doing and self-improve to some degree, but
21:54SPEAKERdoes that have any relation to the topic at hand? So the self-improvement is complicated. Usually,
22:03SPEAKERwhen we talk about machine learning or AI now, what we're talking about is just fancy pattern
22:08SPEAKERrecognition. So if you think about the things that computers are traditionally bad at, but brains
22:14SPEAKERare really good at. So if I throw a baseball to you, your brain can pick out the movement and it
22:21SPEAKERcan pick out the object and it's like, "Oh, that's a baseball." And then it can put your hand up in
22:24SPEAKERthe position to catch it and how fast it's coming at you, et cetera. And there's terrifying math
22:30SPEAKERinvolved there that your brain just does instinctively in the reptile parts and you don't
22:34SPEAKERhave to think about it. You just know, "My hand goes here and then I close it and the ball will be
22:38SPEAKERthere." Because if you weren't able to do that, we would have all been eaten by lions hundreds of
22:42SPEAKERthousands of years ago. So if you put a 386 in front of that problem and feed it a camera feed,
22:51SPEAKERit's not even able to tell what the video is. Its image recognition is hard. So when we have
22:56SPEAKERcomputers do stuff like that, we use more simplified inputs than just straight RGB vision.
23:03SPEAKERWe use sonar or radar or like a motion detector or something like that. And then the robot hand
23:09SPEAKERgoes up in simple signals that a computer is able to easily process. Yeah. So the machine learning
23:14SPEAKERstuff we have now is just better at pattern recognition and signal processing than the other
23:21SPEAKERstuff. And while the algorithms, the reason they get better is they can improve themselves at that
23:27SPEAKERparticular task. We haven't built general purpose pattern recognition algorithms. So if you build,
23:37SPEAKERfor example, what's that Amazon thing that takes a picture of you? It's an Echo and it also has a
23:44SPEAKERcamera and you put it in your closet and it'll tell you if your outfit looks good because it
23:47SPEAKERhas a camera. It's the Amazon show or something. I don't know if I trust Amazon's opinion on that,
23:51SPEAKERbut look, don't trust the computer. But it's not like it has one universal input system and it's
24:00SPEAKERlike, well, okay, my algorithm understands both human language and what fashion is currently
24:08SPEAKERdoing. It has multiple things running at the same time. It's just two systems talking to each other.
24:13SPEAKER- Okay. So what I'm getting from this, it sounds like facial recognition is maybe the kind of
24:18SPEAKERprototypical example of machine learning, but again, it's just training a dumb algorithm
24:23SPEAKERon the general parameters of what a face looks like.
24:26SPEAKER- Well, and it's funny because that's one that we've been able to do for a long time because
24:29SPEAKERfaces have shading and stuff. So you can take a black and white image of a face and you can see
24:34SPEAKERdark patches where the eyes are and light patches where the nose is and a dark patch where the mouth
24:37SPEAKERis. And your point and shoot camera from 2005 was able to figure out where faces were in the picture
24:45SPEAKERand focus accordingly. - Now we're getting to the point where you can track it in real time and apply
24:50SPEAKERgraphics, 3D models that rig to the facial points of articulation and have that move around in real
24:56SPEAKERtime. - I do that for work. We have an algorithm that does face tracking for mapping for when we
25:03SPEAKERdon't have access. We can do that with a 2D camera instead of having to use a depth camera or
25:07SPEAKERsomething. - Basically machine learning is just starting to get our feet wet replicating the dumb
25:11SPEAKERlizard brain part of the human mind. - Kind of, maybe. - So what you're saying is that the term
25:17SPEAKERmachine learning is perhaps a bit self-important and overblown. - Remember when people used to
25:22SPEAKERcomplain about cloud computing being a dumb phrase because the computer is just talking about servers.
25:28SPEAKER- Okay, it's that. - Yeah, it's just a server. Machine learning is just, AI in the current
25:34SPEAKERcontext is just, hey, this is a way for people to understand really complicated pattern recognition.
25:40SPEAKER- Are you saying that Silicon Valley came up with a self-important and fanciful term
25:46SPEAKERfor a very mundane technology? - Well, I don't want to undersell this. It's a hugely beneficial
25:53SPEAKERtechnology and also a hugely detrimental technology if used improperly. - I'm not
25:57SPEAKERsaying it's meaningless or useless. It's just that it doesn't sound like it necessarily
26:00SPEAKERdoes as much as the name would imply. - It is the very first step on the road that the name implies.
26:06SPEAKER- That makes sense. - I was gonna say, I think it's the difference between pattern recognition
26:14SPEAKERand being able to recognize that you are you and knowing that you are you, right? - Sure.
26:22SPEAKER- So I know that you are Brad Shoemaker. - I'll take your word for it. - I know that
26:26SPEAKERyou're a person. I knew that you grew up in North Carolina. I know that you play video games and I
26:34SPEAKERknow that you are an ineffable essence of a human. The computer just knows that your face lines up
26:40SPEAKERwith a dude named Brad Shoemaker. - So it's just a shape broken down into data? - Yeah. - All right,
26:45SPEAKERthat's a lot to chew on. I could maybe use a couple minutes to turn the pressure valve in my
26:50SPEAKERhead and let some of the steam out. - Yeah, that sounds good. Are you optimistic coming out of this
26:53SPEAKERor are you feeling bad? - I don't know. Why don't we talk about that on the other side of the break,
26:57SPEAKERlike where you think things are going and how the future of human civilization is going to develop.
27:05SPEAKERYou know, just light airy topics. - There we go. - For eight in the morning. - Yeah, the lightest.
27:17SPEAKERHey, we're back. - Hi, Will. After careful consideration and releasing a fair amount of
27:22SPEAKERsteam, I'm feeling, let's say, ambivalent at best about where things are going. - I think that's
27:28SPEAKERkind of the right, like, I honestly think that's the right response. Like realistically, is this
27:33SPEAKERthe thing we're going to see in our lifetime? When Werner Vinge wrote the thing in 1993 that was the
27:38SPEAKERkind of kickoff for all this, I think he gave a window of 2005 to, he would be surprised if it
27:44SPEAKERoccurred before 2005 or after 2030. So, I mean, we're there now. - Almost, yeah. Well, there's a
27:51SPEAKERlot of things that sound like they're going to be coming to a head around that time, so. - Yeah,
27:56SPEAKERand realistically, you know, we've kind of mucked this up pretty well consistently for the last
28:00SPEAKER150 or 2000 years, depending on how you count it. Why not give the computers a chance? That's what
28:05SPEAKERI say. - Yeah, I can't say that we have the best track record necessarily, but I don't know. My
28:10SPEAKERimmediate temptation is to just say, "Hey, what do you think is going to happen? Like, tell me the
28:15SPEAKERnext, like, civilization-shaking technological development that's going to occur and what will
28:20SPEAKERhappen afterwards?" But that's an absurd question because nobody can answer it. - Well, I mean,
28:25SPEAKERyeah, and in a lot of ways, we have a bunch of other singularities coming too. You know,
28:30SPEAKERwe have the rise of the global warming stuff that's happening is impossible to predict what's
28:37SPEAKERon the other side of at this point. I mean, we're, as technology has advanced over the last,
28:45SPEAKERso the thing we didn't talk about is all the singularity stuff is about the hockey puck part
28:49SPEAKERof the curve, right? It's the intelligence is increasing and then all of a sudden it does the
28:54SPEAKERhockey stick thing and goes straight up. - Spikes upward, yes. - And we're, as technology in general
29:00SPEAKERhas advanced, we've reached that hockey puck place faster and faster. So, you know... - You mean just
29:07SPEAKERwith general technological development, not on any one specific field? - Yeah, I mean with everything.
29:12SPEAKERLike, if you look at our manufacturing capability, the amount of time between, "Hey, we think we
29:17SPEAKERshould make more batteries," and, "Oh, we're making, you know, a million times as many batteries
29:23SPEAKERin a year as we did 20 years ago," that time has shrunk down dramatically. - Right. So maybe just,
29:30SPEAKERfor example, with respect to climate change, like a wildly optimistic view or prediction might be
29:36SPEAKERlike, "Oh, maybe perhaps some form of carbon scrubbing technology will come online in a way
29:40SPEAKERthat we cannot in any way foresee what the effects will be of on the other side." - Yeah, that's the
29:45SPEAKERRepublican talking point. My feeling on the climate change is that, once the methane and
29:52SPEAKERthe permafrost of the tundra melts, we have a real big problem. But, you know, what do I know? - I
29:58SPEAKERthought some of those theories had been debunked, but then again, I don't know who it is that's doing
30:02SPEAKERthe debunking, so what can you trust? Information is dead. Knowledge is dead. - Look, when the
30:07SPEAKERsalinity of the North Atlantic gets to the Gulf Stream, it stops flowing in the direction it's
30:11SPEAKERflown for the entirety of human civilization. There are multiple singularities in the climate
30:20SPEAKERchange world. It's great. It's a great place to think about. - This is a cheery topic. What else
30:25SPEAKERyou got? Give me something else. - Well, okay, so then there's this guy named Daniel Suarez,
30:30SPEAKERwho wrote a series of books about the singularity that were not of the optimistic variety. His point
30:36SPEAKERwas that the most dangerous weapon in the computers rise against us isn't armored drones
30:44SPEAKERand robot tanks and stuff like that. He's like, "It's just self-driving cars and maybe elevators,
30:49SPEAKERcomputer-controlled elevators." - Are we talking about a nefarious AI that wants to kill us and
30:53SPEAKERis looking for ways to do so? - I'm talking about the Skynet situation, yeah, you know.
30:56SPEAKER- Okay, any just heavy moving objects it can control to throw at people? Is that what you're
31:02SPEAKERsaying? - Yeah, the first third of his first book, and I can't remember what it's called,
31:06SPEAKERis basically about how the robots wipe out most of humanity.
31:13SPEAKERIt's mostly self-driving cars, you know. One self-driving car can kill easily 50,
31:18SPEAKER60 people before it stops working. So yeah, good stuff. - Well, I don't know. Can we broaden this
31:24SPEAKERtopic and try to think of slightly cheerier examples of trends that you see culminating?
31:28SPEAKER- Yeah, here's a good one. Industrial farming in the 60s is a good one. We developed technology in
31:35SPEAKERthe 60s to basically solve the hunger problem for the entire world, applying science to farming in a
31:46SPEAKERway that we never had before, and farming at scale. As a result, our food output has dramatically
31:52SPEAKERincreased. I mean, there were some other unintended side effects of that. But in 1960,
31:58SPEAKERthey were talking about not being able to feed 3 billion people, and we have since blown through
32:04SPEAKERthat without having to start eating people or making soylent. I mean, I guess we do make
32:08SPEAKERsoylent, but that's all different, sure. - I feel like this raises the real problem that we
32:13SPEAKERare generally pretty capable of developing the means to solve civilizational problems,
32:18SPEAKERbut we don't necessarily have the will to implement them. - Yeah, I think that's fair.
32:23SPEAKERUsually, we wait until it's an emergency rather than being active beforehand. Realistically,
32:31SPEAKERthough, I feel like, you know, barring weird advances in quantum computers, I don't think
32:35SPEAKERthat this is a thing that should keep people up at night. Are the computers going to take us over?
32:41SPEAKERYeah, we're probably okay for our lifetime. - There are much more pressing problems to
32:45SPEAKERlose sleep over than that, I would say. - Yeah, as long as we have an axe,
32:49SPEAKERwe can always reprogram the computer the old-fashioned way. That's what I say.
32:51SPEAKER- So does that pretty much do it for your list of trends you see culminating in kind of massive,
32:57SPEAKERmassive forms of upheaval? - Oh, they kind of all do, is the thing. That's the weird thing about
33:02SPEAKERthe future we live in. Everything culminates in a place where you can't predict what's going to...
33:08SPEAKERLook, the Oreo selection of the consumer product diversity is approaching a singularity. I have no
33:14SPEAKERidea what flavor Oreo is coming next. Like, is it going to be... They had Pop Rocks in Oreos.
33:20SPEAKER- No thanks. - They have Sriracha Oreos,
33:23SPEAKERprobably, now. I don't know. Madness. - Human civilization has peaked. Let's
33:27SPEAKERjust shut it down. Speaking of consumer product diversity, Logitech sent me a care package the
33:32SPEAKERother day. They have a brand new keyboard called the G Pro X. - Is this a hands-on report?
33:40SPEAKER- This is a hands-on report. This is our first hands-on report. - Hands-on in the most literal
33:44SPEAKERsense. Fingers on. - Yeah, I've been typing on it. So they did something that I didn't think was
33:48SPEAKERpossible, but they, as a large scale... It's a 10-keyless keyboard, which means it doesn't have
33:52SPEAKERthe numpad, which I like. - Man, this is a great chance for me to get up on my soapbox about
33:59SPEAKERwanting a numpad. - Oh no. People don't use the numpad? - Look, when I'm typing, when I'm writing
34:04SPEAKERor I'm playing games, I want my mouse hand and my ASDF hand as close together as possible so I don't
34:10SPEAKERhave wrist problems. Because if they spread out too far, it messes with your wrist. - I guess.
34:15SPEAKER- You can get, for like $10 from Amazon, a numpad that's USB. Just plug it in another USB port. You
34:22SPEAKERhave plenty of them now. It's the 21st century. And when I need to do spreadsheet shit at tax
34:26SPEAKERtime every year, I get out the numpad and I... And I'm good. I mean, I use my numpad to enter my pen
34:33SPEAKERon my computer every time I sit down. It's like using a phone. There's perfectly suitable number
34:37SPEAKERkeys for the four to six digits that you have on top of the computer, Brad. Anyway. Okay, so the
34:43SPEAKERthing about this Logitech keyboard that blew me away is it's a mechanical keyboard. They've been
34:49SPEAKERdoing that for a while. They used to use these Romer-G switches that they designed themselves
34:55SPEAKERthat are kind of okay, but not great. - Just to clarify, I assume this is the case, but the switch
34:59SPEAKERin a mechanical keyboard is just the action underneath the key, right? The literal mechanical
35:04SPEAKERpart of the key. Okay. - Yeah, it's the thing. So... - Because I see people talk about switches
35:08SPEAKERin their keyboards constantly to the extent that they know what they're called. And that seems like
35:12SPEAKERa level of obsessiveness. That's kind of nuts, but... - Oh, Brad. - Go ahead. - I'm about to open
35:17SPEAKERa whole new world for you. I mean, I've been using the same Apple aluminum keyboard for the last 10
35:22SPEAKERyears or more, like just so people know. - So that's a membrane keyboard. That is a keyboard
35:27SPEAKERwith a piece of plastic underneath it. And that one might have little switches. I can't remember,
35:32SPEAKERbut the upshot is it presses down. It pushes a piece of plastic into a contact on a circuit board,
35:38SPEAKERand that's how the key is pressed. With a mechanical keyboard, there's a physical switch
35:44SPEAKERunder the key cap that when you push it down, it makes contact with... It pushes two contacts into
35:52SPEAKEReach other, and then when you release it, it comes off. And the mechanics of those switches can
35:57SPEAKERdetermine how much pressure you have to push on the key to make it depress, where the actuation
36:03SPEAKERpoint is, whether it's like right at the top where you can just barely, barely rest your finger on
36:07SPEAKERthe button, or you have to press all the way down and hit all the way to the bottom to get it to
36:11SPEAKERpress, whether it clicks, whether there's a tactile bump when the key press happens,
36:17SPEAKERor whether it's just a smooth linear action all the way down. - So is feel the reason that people
36:21SPEAKERtypically prefer mechanicals? - It's why I like them better. It turns out for me, for my wrist
36:27SPEAKERstuff, having a little bit more resistance against the key is better than having no resistance,
36:32SPEAKERbecause it means I type less hard. I type them out of hard that I need to versus just pounding
36:36SPEAKERthe keys like a ham-fisted tyro. And yeah, that's it. So this Logitech keyboard, the G Pro X,
36:45SPEAKERwhich they sent me for promotional purposes, they didn't pay us to talk about this,
36:51SPEAKERbasically it has removable switches, so they have a little puller, and you can pop the keycaps off.
36:57SPEAKERIt uses Cherry MX keycaps, which is a pretty standard keycap type, so you can get like...
37:02SPEAKERDo you know about the keycap scene, Brad? - I've certainly heard of the term Cherry MX before,
37:07SPEAKERbut is that the shape of the keycap? Is Cherry a brand? - Yeah, it's like a brand. It has a
37:12SPEAKERlittle plus on the top, and basically the switch has a plus on the top, and the keycap has a plus
37:17SPEAKERhole in it, and you push the keycap into the plus. But the upshot is because the keycaps are
37:22SPEAKERstandard, there are people, bazillions of people that make keycaps, and you can get keycaps that
37:27SPEAKERlook like a rainbow, you can get them that are a gradient from the top left to the bottom right.
37:32SPEAKERSo it's like largely an aesthetic thing? - Yeah! - Yeah, I guess I could see it. - But the typing,
37:38SPEAKERthe feel is the reason you start here. The aesthetics are just an added bonus. - So I mean,
37:42SPEAKERI like the feel of a mechanical keyboard. I mean, I learned to type on an IBM 286 with the most
37:47SPEAKERmechanical of keyboards in existence, but noise is the big thing that has always really stopped
37:53SPEAKERme from wanting to explore that scene. Like, is there such a thing as a quiet mechanical keyboard,
37:57SPEAKERbecause I just, not just for being on a mic with people on Discord playing games, but partially
38:03SPEAKERbecause of that, but also just for my sanity, I just don't want a lot of clackety-clack all
38:06SPEAKERthe time. So like, is that just part of having one of these keyboards? - No, so, okay, so there's
38:11SPEAKERalways going to be the, well, okay, there's stuff you can do. You can put these little shock
38:16SPEAKERabsorber things on there that basically make the click when the key hits the bottom of the throw.
38:23SPEAKERYou know, when you mash down your key, it makes a little click when the plastic hits the plastic.
38:27SPEAKERYou can put a little rubber thing in there that'll slow that down and make that sound better.
38:31SPEAKERFor the most part, you can get - one of the characteristics of the different color switches,
38:37SPEAKERyou've heard of like Cherry MX Red and Cherry MX Blue and Cherry MX Brown and all that.
38:43SPEAKERTypically, blue means it's clicky, red means that there's a tactile bump and it's a little bit
38:48SPEAKERclicky, and brown means it's linear and quiet. - Okay, well, how quiet is quiet though? Like,
38:54SPEAKERquiet by mechanical standards or just quiet? - I'm going to put my - I have - this keyboard
38:59SPEAKERhas blues and browns on it right now and I'm going to let you hear. - Okay. - That's a blue.
39:04SPEAKERThat's a brown. - Okay, I'm familiar with that sound. - Yeah, so it's like normal keyboard sound.
39:08SPEAKERIt's not super loud. - Okay, so like I had - this is, again, like I swear Logitech is not paying us
39:15SPEAKERanything. I just happened to have been looking at the, I think it was the G915 recently. - Yeah,
39:19SPEAKERthe G915 is a - that's their low profile, it's like low profile mechanical keys. So they don't
39:27SPEAKERhave quite as long a throw as a normal mechanical key. - Okay, well, so like looking at it on - I
39:30SPEAKERwas looking at it on Amazon recently and they let you select when you buy it between what they
39:34SPEAKERdescribe as clicky, linear, and tactile, or is that just - are those just other terms for the
39:39SPEAKERreds, blues, and browns that you're talking about? - So I think when they had that one,
39:42SPEAKERthey didn't brand them as red, blue, and brown, but red, blue, and brown is what everybody
39:47SPEAKERis - is kind of the standard because Cherry MX was the first one who was selling lots of mechanical
39:53SPEAKERkeyboard switches. - I see. Okay, that makes sense. - I was gonna say the thing that makes
39:56SPEAKERthis special is it's the first mass market keyboard I've seen that has removable switches.
40:00SPEAKER- Okay. - Like this is - like Corsair and Steel Case, SteelSeries, and a bunch of other companies
40:07SPEAKERthat make like PC peripherals have been doing this for a while, Cooler Master, but they're selling
40:12SPEAKERthousands and maybe tens of thousands of keyboards I think on the outside. When somebody the scale of
40:17SPEAKERLogitech comes in and starts selling mechanical keyboards, like it's a real moment for the
40:22SPEAKERmechanical keyboard community. - Yeah, for the markets who broaden dramatically. - Yeah,
40:26SPEAKERit's like the number of people that Logitech reaches versus the number of people Corsair
40:31SPEAKERor, you know, like this is - just to be clear, this is literally a thing that started on like
40:36SPEAKERMassdrop and on message boards where people who are fans of mechanical keyboards were
40:44SPEAKERlike designing circuit boards and buying switches from China and assembling them themselves. - As
40:50SPEAKERgrassroots as you can get. - Yeah, and now Logitech does it, which is like - it's one of the
40:55SPEAKERthings, it's a singularity for the hobby. - There you go, there's singularities all around us. - Yeah,
41:01SPEAKERwho knows? Who knows what the other side of this looks like? Is my mom gonna be putting a mechanical
41:05SPEAKERkeyboard on her laptop? I don't know, maybe, probably not. - Probably not. - It's pretty good
41:10SPEAKERthough, that's the TLDR on the keyboard, I like it a lot. - And then is any of the stuff that they
41:14SPEAKERsent you wireless? The whole reason I was looking at keyboards on Amazon recently is that my keyboard
41:19SPEAKERat work broke and I got as a replacement one of the - I'm trying to remember what Apple calls them,
41:24SPEAKERit's the current model of their aluminum keyboard but it's wireless. - Oh, does it have the butterfly
41:29SPEAKERkeys? - I can't remember, I don't think it has like the objectionable key action but the reason
41:33SPEAKERI bring it up is it's - I think they call it the magic keyboard or something. - Oh god,
41:37SPEAKERgross. - Typically Apple nonsense but the point is it's wireless and it's wireless and like I had
41:42SPEAKERalways been pretty skittish about getting wireless input devices because I just didn't want any
41:46SPEAKERlatency but this thing has got a huge battery in it, it's rechargeable and it seems to work just
41:50SPEAKERas well as a wired keyboard so maybe start thinking about looking into getting some wireless
41:54SPEAKERstuff for home but I don't know if that stuff's good yet. - So I have one of their G Pro mice
41:59SPEAKERwhich is the light - it uses their light speed wireless thing. The white - it's like - it's
42:05SPEAKERlower latency, higher update rate but it uses a lot more battery than say Bluetooth or one of the
42:13SPEAKERold 2.4 dedicated dongles. - It comes with its own dongle, that's not standard Bluetooth or anything?
42:19SPEAKEROkay. - Well, so it can do standard Bluetooth or it can do light speed, you kind of switch back
42:23SPEAKERand forth depending on which device it's connected to. - I see. - So if you want to pair it with your
42:28SPEAKERMac then you can or if you want to pair it with a Steam Link you can or you can plug the light
42:31SPEAKERspeed dongle and it shows up as just a normal mouse to the OS. - And then can you just use one
42:36SPEAKERdongle if you get multiple devices that use that standard? - So you use a different dongle for each
42:41SPEAKERdevice. - Ah okay. - Which I assume is a limitation of the light speed tech but they provide like an
42:47SPEAKERextender thing that lets you plug a micro USB into the dongle so you can put the dongle on your
42:51SPEAKERdesk. I'm uncomfortable talking about my dongles, we're all friends here. My dongles are on my
42:59SPEAKERdesktop underneath my monitor so they're kind of out of the way but close so that there's no line
43:04SPEAKERof sight issues. The G Pro is a wired keyboard though, it's old school. - Ah, it's the G Pro and
43:10SPEAKERG Pro X. - Okay, I've just gotten, I mean you've seen my desk, like I've gotten really into very
43:15SPEAKERminimal sort of desktop layout. - It's very clean. - Yeah, like I've just found that my anxiety
43:21SPEAKERtends to go down proportionate to the amount of clutter on my desk so I've eliminated almost
43:26SPEAKEReverything else and now I'm just looking around at the cords going to my input devices and thinking
43:30SPEAKERmaybe these can go too. - You shouldn't see my desk, it would make you have hives. - I've seen,
43:35SPEAKERI tried, I blocked it out, I repressed the memory. - Yeah, yeah, that's probably best. Do you want to
43:42SPEAKERdo an email? - Yeah, let's read an email or three. - Let's see, we have, oh this is, I thought this
43:49SPEAKERwas a good one. Let's see, this person asked us not to use their handle, so I'm not going to say
43:54SPEAKERany names, but they say my workplace is remote. On a good day we get two hundredths of a megabit
44:01SPEAKERdown, megabit per second down and five megabits per second up of internet. I'm not kidding,
44:06SPEAKERit is not a laughing matter when the internet is being used frequently in the morning,
44:09SPEAKERit is so slow that speedtest.net won't even load. So then they talk a lot about the phone that they
44:16SPEAKERhave, which we don't even need to get into. But they said, I've just started using tethering while
44:23SPEAKERI'm at work because I have unlimited data to send emails and such. Tethering doesn't work that well
44:30SPEAKERand they're not sure if it's their phone or their carrier or what. And they were like,
44:33SPEAKERwhat's the best option for tethering in 2019 if cost isn't an issue? Is it Google Fi because it
44:38SPEAKERuses multiple networks or is this a software problem rather than a hardware problem? - That's
44:43SPEAKERit. - Do you have an answer? - Yeah. So the tethering, the thing that's going to get the
44:48SPEAKERtethering on most phones these days is the network connection. Like which carrier you're on. A,
44:55SPEAKERyou're using the internet for work, work should provide internet that doesn't suck and that you
44:58SPEAKERcan use if you need internet for work. Workers unite. - Are you saying I can bill my, are you
45:03SPEAKERsaying I can send my ISP bill to my company from now on? Because sold. - I mean, my company pays my
45:09SPEAKERISP bill, but I also send the checks. So it's a little bit different situation. - You've probably,
45:13SPEAKERyou've got a little more corporate influence than I do. - Look, I am, look, who's going to yell at
45:18SPEAKERme? Nobody. So yeah, like the gate is which carrier gives you the best signal at work.
45:23SPEAKERFirst off, work should provide internet that doesn't suck if you need internet for work.
45:27SPEAKERSecond, the carrier is going to be the gate, not the hardware and not the device. - So any decent
45:34SPEAKERsmartphone should be able to tether pretty efficiently provided the cell signal is good
45:38SPEAKERand everything. Yeah. - I mean, I think the, when you get on the real low end and like the 150,
45:43SPEAKER$200 phones where they maybe don't have enough CPU, they're running older versions of
45:49SPEAKERSnapdragon and stuff like that. Maybe you'll have problems. But usually in my experience,
45:55SPEAKERthe problem is my cell phone provider's network doesn't provide enough coverage. So
46:00SPEAKERthe options are to go to different cell phone stores for T-Mobile and Sprint and Verizon and
46:05SPEAKERAT&T and get loaners, take them to work, see which one works best. Or the Google Fi is actually not a
46:12SPEAKERbad, like because of the way Google Fi works, it's an interesting choice because they buy bandwidth
46:20SPEAKERfrom multiple carriers. I don't think they buy from Verizon. I think it's only the GSM carriers.
46:25SPEAKER- Wait, does this person work in an office or at home? - They said they are in an office.
46:31SPEAKER- Oh, wow. I thought this was a work at home scenario. This is utterly insane.
46:36SPEAKER- No, no, no. Oh, it says my workplace is remote. So I think that means like not working at home,
46:43SPEAKERbut like a remote. I took that to mean like they work on top of a mountain someplace or something.
46:47SPEAKER- I see. - Out in the hinterlands.
46:49SPEAKER- So are there, what am I trying to ask here? Are there non-phone solutions available for tethering?
46:56SPEAKERI mean, like a kind of a desk mounted or like a stationary equivalent of what tethering would be,
47:02SPEAKERlike just a box that you buy and plug into a machine?
47:05SPEAKER- Yeah, yeah. I guess those little, what are they called? They're called portable Wi-Fi hotspots,
47:12SPEAKERbut that's all that is. It's like a little credit card size box with a battery in it that has a cell
47:17SPEAKERmodem and a Wi-Fi hotspot basically. So yeah, find out which carrier has the best coverage in your
47:25SPEAKERworkplace. That's the TLDR. So if one of your, if one of your, get your coworkers who have different
47:31SPEAKERcell plans than you to all run speed tests on their phones and see who does best.
47:34SPEAKER- Yeah. Crowdsource your internet.
47:36SPEAKER- And then switch to that network. Yeah. I think that's it. I mean, we have more,
47:40SPEAKERwe have more, I would say let's get more letters. The thing we've gotten the most letters about is
47:45SPEAKERwhether you've bought monitors yet, Brad. Still using the 22s, huh?
47:49SPEAKER- No, they're 20s. I didn't think to correct you last week. They are 20 inch, 1680 by 1050.
47:54SPEAKER- It's an audio podcast. You can't see this, but my head is in my hands.
47:58SPEAKER- Will's face is in his hands right now. I'll just buckle up for this ongoing saga.
48:03SPEAKERI got some good suggestions from people on Twitter of monitor models that they like.
48:08SPEAKERRazer just announced a high refresh IPS monitor, but it's like seven, six or $700. And I'm looking
48:14SPEAKERto buy two. - Is it 120 or 144?
48:16SPEAKER- It's 144, but I'm looking to buy two monitors. So that's kind of not
48:20SPEAKERan option for pricing reasons. - A couple of people asked where you
48:26SPEAKERread, where you go for monitor panel reviews. - Like I mentioned last week, I have been looking
48:32SPEAKERat tfccentral.co.uk, but I've only recently started really digging into their stuff or
48:39SPEAKERanybody else's stuff. They seem incredibly thorough because they will dig into a specific,
48:44SPEAKERlike they'll dig into the model that they're reviewing and look and see who makes the panel
48:47SPEAKERin it and list all of that information out. So you can get like very good detailed info
48:53SPEAKERon the innards. And then also the monitor subreddit seems to be obsessive about tracking
48:59SPEAKERall that stuff too. So that subreddit is how I found out about these new panels that are on
49:03SPEAKERthe way that hopefully will be on the market in the next couple of months. I'm excited to
49:08SPEAKERbuy some monitors on top of these things being ancient and using a ton of electricity. They're
49:14SPEAKERvery heavy. And I've got them on this, EchoGear sent me this two arm monitor mount a while back
49:21SPEAKERthat I've been using. And I found at a certain point that it's digging into the particle board
49:26SPEAKERof my desk. Like the monitor is so front heavy. It makes that assembly so front heavy that it
49:30SPEAKERwas literally like bowing tilting forward and bowing the desk. So I went out and bought a
49:35SPEAKERcouple of cheap pieces of steel, sheet metal kind of stuff and spray painted it and put it under
49:42SPEAKERthere. So yeah, I'm also looking forward to lighter monitors that are maybe not going to
49:46SPEAKERcollapse this desk that I'm using. It's time, Brad. Yeah, this is beyond past time. It's time,
49:53SPEAKERbut I'm going to ask you every week. That's that's where we're at. Even when I buy new monitors,
49:57SPEAKERthere's part of me wants to set these up in a corner somewhere hooked up to like, I don't know,
50:01SPEAKERan Apple TV or something, then just let them go forever and see how long it takes them to die.
50:06SPEAKERLook, you could probably get five or six of those from eBay. You can make like a little,
50:10SPEAKERlike a Brazil Terry Gilliam like monitor statue with all that would be really cool.
50:16SPEAKERI mean, it'll be literally will be 15 years in February and I've been using them almost nonstop
50:20SPEAKERsince then. And they're just, and they're not showing any signs of slowing down. Like what,
50:24SPEAKERmaybe I could set the Guinness record for longest monitor use. No, no, no, no, no, no. Nope. That's
50:30SPEAKERdangerous. Look, those cold cathodes are so dim now in there that there's there they're losing
50:36SPEAKERjuice. You're gonna have to replace them before long. Yeah. Like that I may have mentioned before
50:40SPEAKERthat these things are totally fine when I'm only using them next to each other. But the second I
50:44SPEAKERopen a laptop or get my iPad out or any, the second, any other LCD appears in my field of view
50:49SPEAKERwith these monitors, I just want to throw them out the window because they look like dingy garbage.
50:53SPEAKERYeah. It's time. Yeah. They're they're dim. They don't have the nits. No, they definitely, they're
50:59SPEAKERpretty short on nits. I think that's as short on nits as good a place as any to wrap this up.
51:07SPEAKERIf you have an email, if you have a question, send it to techpod@content.town.
51:12SPEAKERYeah. We would love to get your questions. We've gotten a bunch of good ones. I think we're,
51:17SPEAKERthey don't necessarily have to do that stuff we've talked about. That's what I wanted to say. I would
51:20SPEAKERlove to get emails from people about things we haven't talked about. Yeah. That would be a nice
51:24SPEAKERway to explore some other, other topics. We got a bunch of great topic ideas over the last couple
51:28SPEAKERof weeks. So we're always, we love hearing those. But send your questions too. We'll, we'll, I think
51:35SPEAKERwe almost have enough to do an all questions episode. So if you have weird technology questions,
51:39SPEAKERsend them in. Yeah, that'd be fun. And as always, reviews on iTunes are hugely helpful. You can find
51:45SPEAKERus on Spotify, pocket casts, all the places that podcasts are found. Yeah. Thanks for listening in
51:50SPEAKERfor your feedback. Yeah. It's been people really liked last week's episode. It seems like, so tell
51:54SPEAKERyour friends screens. We all spend a lot of time looking at them. It turns out it's funny that the
51:58SPEAKERone that we recorded first and we're like, I don't know if we should post this one. I don't know if
52:02SPEAKERit's good enough. And then people are like, that's a great episode. People love, people love a tale
52:07SPEAKERof struggle and overcoming the odds of buying monitors. Brad, are we still gonna be talking
52:13SPEAKERabout your monitors next year? No. In 2020, the future? We shouldn't talk about this anymore ever.
52:17SPEAKERLet's end this thing now before people turn it off in disgust. Bye everybody. See you next week.
52:22SPEAKEROkay. Bye. Thanks.