676: A Sternly Worded Instruction

I feel like we should do some Stoapocalypse follow-up. I was talking last week about how Richmond was due for like 30 inches of snow. Yeah, what happened? We did not get 30 inches of snow. Yeah, I saw your picture of your driveway. You've got barely enough to cover the ground. It was a little more than that, perhaps not by Boston standards, I will allow, but by local standards, it was somewhat significant. It was like three or four inches. I'm not going to do that computation for centimeters. It was a handful of centimeters. You've got to stop doing unit conversions. We need a moratorium on unit conversions, especially since you can't do them in your head and you're always just guessing. If people are interested in doing unit conversion, lots of ways that is possible after the fact. But I feel like what we should instead be doing is slowly acclimating people to our weird units. No, because our units, with these such a Fahrenheit, stink. They do, but there are units and we talk about them, and that's, you know, by listening to the show, people will become familiar with them. I'm pretty sure that people outside of the U.S. who consume U.S. media have plenty of other cases of other shows and things, you know, just using our units and making them figure it out. I don't think they need us to add to the pile. That being said, I also don't think it's our responsibility to keep converting everything. All right, well, let the record show I tried. Anyways, we got a small amount of snow, but on top of the snow, we got a pile of ice. And that ice, during the day, when it's sunny, which has been the last couple of days, it will melt a little bit at the top, and then refreeze and refreeze and refreeze and refreeze. And so it ends up that we have like a couple of inches of ice everywhere. That's not good. Which is not good. And the county, because remember in Virginia everything is by counties, the county has done an actually extraordinary job of keeping interstates clear, of keeping major, the bigger side roads, the bigger like non-interstate roads clear. you know for example my neighborhood dumps into one of the larger roads where where we are and that's bone dry it's been dry basically this whole time but when it comes to the neighborhoods they justifiably didn't get to them in time to like really clear them when it was workable and now we've got like two to three to four inches of ice on on all of these neighborhood roads and And the county has basically said, it'll melt one day, I guess. And so the kids, it's currently Wednesday as we record this. And at about noon today, the school system said, ah, screw it, we're out for the week. So that is the status of snowpocalypse. And then this coming weekend, we might get anywhere between two and 30 inches more snow. Although I think they've pumped the brakes on the 30 inches side of things. And I think it's going to be closer to like two to four. But we have been below freezing, as I'm sure you two very much have. We have been below freezing for a week almost. And we're not supposed to get above freezing until this coming Monday. So, you know, anything that hasn't been fairly well cleared, like, you know, Aaron and I and the kids went out and cleared about half of our driveway in order to make sure we could get out. And anything that isn't mostly clear is just turning into ever thicker layers of ice. And then we might put a couple to 30 inches of snow on top of it, which is going to be super fun. You went out and you shoveled half your driveway. Yes. See, that's how you can tell he doesn't get a lot of snow. How about you shovel the whole driveway? John, do you have any idea how f***ing hard it is to shovel under three inches of ice? You don't, because as much as you do, you don't know. You think I don't? Believe me, I do. John lives in Boston. He knows. Under ice. You get a metric a** ton of snow, but I don't think you get the ice. We get every possible combination of snow you can imagine, including heavy, slushy, icy, everything. Wintry mix, they call it. I'm not talking about slushy, icy snow. I'm talking about literal ice. We had to take a shovel to get through. I'm familiar with ice, Casey. And by the way, if you were familiar with winter, you would know that the correct time to shovel when it's less work is when the snow or ice or whatever is fresh-ish. Yeah. Instead of waiting for it to freeze overnight. Which is what we did. That's what we did. As soon as we could, we got out there. But you did half. You did half the job. Did you shovel your sidewalk, too? Yeah. Well, we don't have a sidewalk in the sense that you're expecting, but the walk from the driveway to the front door is shoveled enough for UPS to get there. Anyway, next time do the whole driveway. Can your kids play ice hockey on the street now? Oh, God, yes. People, they are skating. There are people skating around the neighborhood roads of Richmond on ice skates. Absolutely. And, yeah, it's a mess. And we don't have an ice scraper, which means we had to use, like, shovel, like yard shovels. You don't own an ice scraper? In the car? No. You know what I'm talking about? Like a hoe, but instead of a 90-degree angle at the end, it's just a straight shot, and so you can get under the ice and shit. I know what you're talking about, but, like, you shouldn't be using that on a driver. Anyway, you're going to mess it up. What am I choosing? I mean, I'm using a freaking shovel right now, John. What are you going to do? You need to take some, like, continuing education classes on how to live with winter. Jesus Christ. Oh, my God. I left the Northeast, so I didn't have to deal with your bulls**t winters, and yet here I am. Apparently, you didn't do any shoveling when you lived in Connecticut, either. No, because Dad had a snowblower. of a ridiculously overpowered snowblower, and he loved going out there. I have a ridiculously overpowered snowblower, and it's wonderful three times a year that I need it. But even then, you have to go out while the snow is fresh, before it has a chance to get packed down or have anybody walk on it and pack down more. And then what you do is you first clear the snow, and then you put down salt. Which I get. Because that will help keep it, you know, that will finish the job of any little remnants left behind. and then that will prevent it from getting ice covered all over it when everything starts to melt. If you do a good enough job shoveling, you won't have to worry about ice because the sun will be able to reach the darker pavement and it will either evaporate or sublimate off the ice and you'll get fresh, clean asphalt or pavers or whatever is underneath there that will slowly be exposed to the sun. And even if it's five degrees as it has been here, it will clear off the remnants of the snow. Anyway, I shoveled my entire driveway and sidewalk three times. Here's your cookie, John. Anyway, so it stinks around here. It's very icy. It's not fun. It's interesting in Erin's car because her car is rear-wheel drive when under electricity, and then it will turn on the gas motor to be all-wheel drive when necessary, so that's kind of fun. And I took my car out briefly to make a food run, and I didn't have the time or good spot to do any hooliganism, which was disappointing, but at this rate, I'll have plenty of times and perhaps can find a spot to do donuts and other, you know, ridiculous, childish things. But nevertheless, think happy thoughts for Virginia this coming weekend because it could get worse. Yeah, good luck. All right, let's talk about ATPFM, our website's passkeys. John, what have you been up to? I think it was last episode. I said that I had added passkeys. Not I had added passkeys support. Claude Code had to add a masking support to our website. And it was working when I talked about it on the show, even for the live listeners. But since the show was released, I got some feedback from users about how they expected it to work. One user, I forgot, I didn't say this person's name, sent me links. I asked for links like, show me some websites where you like how it works. And I forget what they sent, but they sent a bunch of websites, some of which I already had an account on. And, like, basically the sort of best practice way to do this is to not even wait for the person to enter anything. But if they have a passkey to essentially offer to log them in with their passkey, of course they can just say cancel and just, you know, manually log in because our website, atp.fm, has three ways to log in. You can log in with the magic email link, you can log in with a password, and you can log in with a passkey, and you can choose which one you want. But if you have a passkey and you land on the login page now, you won't even have to do anything. It will just pop out a thing that says, hey, do you want to log in with your passkey? And you can say yes or no. And if you say no, you can just do it manually. So that's working. And lots of people are trying out the system. I see them all logging in with their passkeys. And it is exciting. And I haven't got any bug reports yet. So fingers crossed. All right. With regard to private AI compute in Google, this was an announcement from November 11th of 2025. Google writes that Private AI Compute in the cloud is our next step in AI processing technology. It builds off the industry-leading security and privacy safeguards that we embed to keep you in control of your experiences and your data safe, guided by our secure AI framework, AI principles, and privacy principles. Private AI Compute is a secure, fortified space for processing your data that keeps your data isolated and private to you. It processes the same type of sensitive information you might expect to be processed on device within its trusted boundary. Your personal information, unique insights, and how you use them are protected by an extra layer of security and privacy, in addition to our existing AI safeguards. Remote attestation and encryption are used to connect your device to hardware-secured field cloud environments, allowing Gemini models to securely process your data within a specialized protected space. This ensures sensitive data processed by private AI compute remains accessible only to you and no one else, not even Google. Yeah, I think I had this in the notes ages ago, but this is essentially Google's answer to private cloud compute. Now, the tricky bits are always like how Apple did a bunch of stuff where they bent over backwards to say, and we'll give you this binary image of the thing, and this is the way you can prove to yourself mathematically that when you connect to our servers, what you're actually connecting to is the binary thing that we gave you, so we know we're not just giving you one thing and using a different thing on our servers. And lots of people looked at that and said, it's great and this is very helpful, but in the end, you do kind of sort of have to trust Apple to some degree. same thing with the Google private compute I'm sure they're doing all the same things but in the end you have to trust Google, trust that Google has correctly and honestly implemented the thing that they said they implemented which I'm sure they have because again you know they want to do this for Apple, they want to do it for themselves, being able to privately process stuff that is too expensive to do on devices, a useful service to offer and this is their offering that they announced in November so again no announcement from Apple about I mean, there's some rumors. I don't think we're going to cover them in the show. Maybe we mentioned last time, but like the bifurcation of the Apple's rollout of their LLM tech, where like 26.4 will have some stuff, but then like the chatbot thing will come later. And the rumors about Apple using Google's TPUs and their data centers was about the later chatbots and not the 26.4 thing. So we'll see. Maybe there'll be another weird joint statement sometime around the next WWDC to say, oh, and Apple's going to be using Google's private cloud computer. Maybe they just won't say anything, and they'll rebrand it. I think Marco was alluding to that last episode. Like, maybe they'll just keep saying private cloud computer. And the little asterisk will say either Apple's data center is using our thing or Google's data center is using their thing, which is basically the same thing. Oh, no, I was saying that it was very likely to just kind of be memory hold. Like, you kind of just never be talking, like AirPower, just kind of never be talked about again. You know, maybe a quick little mention somewhere. I think they will talk about it because they've been advertising it like crazy. They're definitely going to be using it. And Google's got the same thing. So why would they have a memory hole? Well, I think they will keep advertising their commitment to privacy. But the specifics of how that's achieved, I think, will become a little bit less specific. Because they have to – because, again, I think the outcome here is Apple's private cloud compute infrastructure is probably not going to be their long-term solution. It might not even be their next-year solution. I think ultimately they are going to most likely move to Google's large models exclusively for all their large model needs. I mean, I guess that's kind of what has already been reported. But I think that's going to be running on Google's infrastructure probably for the foreseeable future. For on-device stuff, I'm sure they're still going to do that. Yeah, but won't they just still call it private cloud compute? Maybe. I don't know. It's the same thing. It's just like they're not running Google is. But if it's the same idea that even Google doesn't have access to it and you can prove that you're running on the hardware that they say you're running on, I think they'll just rebrand it. But we'll see. Maybe. I do think, though, like this is a good example. You know, there's a lot of a lot of examples throughout, you know, recent tech history where Apple either, you know, designs their own complete thing, their own complete standard for something that doesn't yet have a standard. And then everyone else kind of copies it and kind of makes it the standard. see things like Qi 2 and Qi with magnets. MagSafe kind of became Qi 2 and that clearly influenced it. Apple kind of steered that. Or in some cases Apple participates in the design of a standard very heavily like USB-C where Apple takes a big role because they want it done a certain way and it's kind of done the Apple way and everybody gets to use that. I think private cloud compute set a framework in motion and a set of standards in motion that effectively Google was like, oh, yeah, that's probably a good idea. And then Google copied. And maybe that was in part because they were already talking to Apple about possibly hosting this stuff at that point whenever they decided to do this. Maybe that came afterwards. But whatever it is, I think without Apple having announced private cloud compute and put that stake in the ground the way it was back then, I think if that hadn't happened, Google would not have done their thing. However the actions came to be that Google did their thing that seems very similar, I think that is a direct result of Apple having announced private cloud compute the way they did. So even if Apple's version of this is kind of a temporary flash in the pan and isn't really a long-term thing, it was good that Apple did that because that caused Google, I think, to do their basically copy of it. And that's better for everybody, including Google's customers. And I don't think that's the kind of thing Google would have come up with on their own. That is not historically one of their strong suits. So I think overall, this was a very beneficial outcome, even if we never hear about Apple's version of it again. I wonder if Apple spent more money and time on private cloud compute or the supposed server chips that they made that are based on the M5. Because we know they were using, like, the M2 Ultras before that, but the rumor is that they actually made a, like, chip for their own servers that is not a chip that will go into any other devices. And if they just end up entirely on Google's data centers in, like, the next year or two, boy, that was a lot of money for those server chips that really, you know, went nowhere. But, you know, they can just join the Mac Pro with products that Apple lost interest in. Well, they can join the car and probably pretty soon VR. Well, these will be shipping, though. Your code will be running on them. You just won't know it. Yeah, I mean, obviously, I think Apple has a lot of things they could do with a large amount of their own processors and their own cloud. But ultimately, that is not Apple's wheelhouse. Google wins on infrastructure so much in so many ways across the Internet. Like, Google, for all of their faults in some of their weird privacy stuff sometimes and a lot of their, you know, kind of all-over-the-place product design, they have the best infrastructure in the world. They can scale to Apple's scale, and not a lot of other people can, including sometimes Apple itself. So while Apple does run a lot of their services, it's generally on other people's cloud hardware. It's not on a bunch of, like, Apple-branded data centers. I think there's some Apple data centers, but I don't think that's where most of their stuff is running. I think most of the stuff is running on, like, Google Cloud, Azure, AWS, like that kind of infrastructure, Cloudflare or whatever. Wasn't Cloudflare involved in private relay? Yeah, I think they did do something. Didn't they do, like, the private relay thing or whatever? Yeah, so anyway, to run large-scale, large-volume AI models is so top of the line in infrastructure. It's so cutting-edge. It's so demanding. It's so high scale. You need somebody like Google to do that. Like Apple was never going to serve that need at the right scale for them. So I don't see a situation in which it makes any sense for Apple to use their own AI chips right now. Maybe, again, 10, 20 years down the road, maybe things will be different. But right now, it makes total sense to go all in on a company that really specializes in this. And if you want a company that you already have a good relationship with and that has all the levels of the stack, that has first-party flagship models and huge infrastructure, the only answer is Google. So I think this – again, I'm very optimistic about this partnership. I think this is a very good partnership for both companies. And I think Apple made by far the best choice in choosing Google for this and investing in it like this because I think they're going to be way ahead of where they would be if they were on their own. And I can't imagine any other company having partnered with them that would have given them a better outcome. Speaking of Google and Apple, an anonymous ex-Googler wrote in and said, Apple is or was one of Google Cloud's largest customers. Google Cloud is the only public cloud with enough network bandwidth to serve iCloud Drive and backup, which was wholly on Google Cloud. Mobile Me was on Azure, contributing to its disaster. Shock fired. What a surprise that an ex-Google employee would say that. Maybe a slightly biased opinion there. But, I mean, having used these services in a professional capacity, I can say that Azure was, let's say, maybe not up to the standards of AWS and Google Cloud. Back when I was using Google Cloud, Google Cloud was also kind of the new entrant. But Azure was definitely weird and had, like, a weird Windows flavor to something. So I think there's probably a grain of truth in this. I mean, MobileMe had a lot of problems. Like, when you have that kind of service, usually, like, the server provider is usually not your problem. But it can contribute to it. It's basically, like, how easy is it to build a resilient system using these cloud services? And obviously, AWS is the biggest entrant in this market and is sort of number one for a reason. Google Cloud is slightly different in that Google's main expertise was building things for itself, and it only later decided we should be selling this to customers. But yeah, Azure was like, Microsoft's like, we can do this too. And it's like, yeah, I guess. Gracious. All right. And then continuing from the ex-Googler, the other main Apple product that runs on Google Cloud is Shazam, which is running on Google Kubernetes engine. Yeah, and they acquired Shazam, so I wonder if they just left it where it was originally running when it was an independent company. All right. We should talk with regard to using Alfred to remove the attribution stuff that was copied in text from macOS from the news app. And, John, you apparently have made a snarky comment, only on the Mac can you do these sorts of things, which at the time I agreed with. But Rowley writes in to say, you can also remove the attribution crud from copied text on iOS. I made a clean URL shortcut that removes the attribution text and also changes x.com links to xcanceled.com links. Yeah, I was also alluding to it. I don't know if Alfred does this, but the idea of, like, when you copy would run automatically and clean the thing out versus having to manually run a shortcut. But for all I know, shortcuts can do it automatically on copy as well. Anyway, sorry for slagging shortcuts. Just it's not my cup of tea, so I'm not always aware of all the capabilities. I'm sure the shortcut will work great for at least one year. All right. And then you had also snarked in episode 674, how long until we have screens on RAM? Well, Vasudev Baldwai writes, good news. You can buy DDR5 memory sticks with little OLED screens on them right now. The idea is that the screen would display details useful to know while overclocking your memory. It's only $590, which is not much more than current DDR5 prices. Should have known it existed because why wouldn't it exist? Why wouldn't it exist? Because it's a terrible idea and you don't want to put something heat producing on top of your RAM. You know, you can put a heat sink on your RAM. What's the opposite of a heat sink? Can we put a hot thing on your RAM? Yes, you can. Okay. But, okay. In the defense of this ridiculousness, back in the day, screens were expensive. Like, when we all came up with computers, screens were expensive and memory was expensive. Now, only one of those is true. What a wonderful world that we live in that an OLED screen, like, not even a crappy type of screen, an OLED screen is so cheap that they're shoving it in battery chargers and RAM sticks because, like, they cost basically nothing now. What an amazing world we live in. And that's, as a computer nerd, look, I know there's a lot of bad stuff going on in the world. We'll get through it. But there's a lot of good stuff in the world, too. And what an amazing world it is that we have created technology so good that we can waste screens on stupid crap that doesn't need them. It doesn't matter. Screens in places where, honestly, you shouldn't ever be able to see them because it's inside your computer. But I know everyone wants to be able to see everything on their computer. That's wild. The cases are made of glass. Remember that rumor from ages ago that we talked about on the show of, like, the Mac Pro case made of glass? Finally, Apple can now have screens on its RAM, too. But it'll be soldered to the motherboard. The screen is soldered to the motherboard, too. All right. Asus, whatever it's called. It's not Asus, I learned. It is not Asus. It has improved or is improving color accuracy for Mac users. Jacob Hahn on the CineD website writes, Asus has introduced new macOS Focus features for its ProArt displays aimed at improving color-critical workflows for photography and video professionals using MacBook, Mac Mini, and Mac Studio systems. Key updates include support for the Asus Display Widget Center on macOS, enabling software-based monitor control and MacBook keyboard brightness adjustment, as well as new M Model P3 color mode, designed to align ProArt displays with MacBook screens for consistent color reproduction. Yeah, that is hilarious that that is like, at all the companies that market like their, you know, sort of formerly PC monitors to Macs, they all seem to use this same phrasing or like same scenario to say like our monitor will exactly match your MacBook's like built-in screen. Because they, I guess they think that's what people want. Like we just want an accurate monitor. If my laptop screen is not accurate, don't match it. And they're just like, I think they just say like, if you have your laptop up on your desk, our monitor will match it and it will look right. But it's like there's more to, you know, like you never see Apple say, if you buy our monitors, they'll exactly match our laptops. They're always talking about the accuracy as measured objectively against some like color checker type thing. But they're like, no, it'll just match your laptop screen. Isn't that what you want? It's like, well, yeah, but that's not. Anyway, all this to say that more of these companies are aiming at Mac users with their monitors. You know, these sort of pro art ones and other product lines that are not gamery are more likely to, for example, work with the keyboard brightness controls on your Mac. You know, having to run a third-party driver, how long will that be supported, yada, yada. But, like, there's sort of a – it seems to be developing a gradient from Apple monitors, which you plug in and will work, in theory, with everything having to do with Mac OS. That'll work with the keyboard things. You know, does it even have a power button? Like, the Mac knows about them. Like that's at one end and at the other end is just the gaming monitor that you plug in. The back has no idea what the hell it is and hopefully you can get it working right. And these are kind of in the middle. Of course, these are a little bit more expensive and they tend not to have the all the fanciest features because those are still on kind of the gaming monitors first. So many LED backlights, super high refresh, you know, fancy OLEDs with fancy pixel structures. Give it time. So more of these will come out. But I'm glad to see some of these monitor makers realizing that there is a market here to cater to Mac users and also that Mac users are probably willing to pay what seem like astronomical prices because it's still cheaper than Apple's monitors. And speaking of, we live in this amazing age when screens are cheap, not Apple screens. Well, that was never the case. Apple screens have never, ever been cheap. I know. I'm just saying, like, they are steadily going against the tide. All right. And then we talked in last week's overtime about the Fender Mix Modular Headphones, and Graham Dobie writes, there's a Danish company called AIIAI, all caps, that's been making custom modular repairable headphones for 15 years or so. That brand name didn't age so well. It's all caps. Yeah, I think it's IAI, but it doesn't matter. I actually tried one of their pairs of headphones back when I was reviewing headphones during that brief time in my life before there were way too many of them, and they were fine. But that's all I can say about them. They were fine. but I wasn't impressed. Yeah, I mean, like the company, if you look at it like this, this does seem like a company that's really focused on this type of, you know, durable, rugged, no-nonsense, modular, you know. They're not aiming for the super-duper audiophiles. There is a little bit of like an aesthetic they're going for. But, yeah, it's good to know companies like this are out there. I've never heard of this company, but these products will affect it. If the Fender Mix appeals to you, these might appeal to you as well. And the prices seem similar. I looked at one or two of them. They're only like $300 or so. I mean, one of the ways that things like this have been modular in the past has basically been like you can clip on a cover, like a plastic cover of a different color where you want it. So you can customize your color scheme. And like there was a brief phase. I think I was part of that. I don't know if that's still what they're doing now. But there was a brief phase, like, in, like, the mid-2010s where, like, every headphone brand had some kind of, like, quote, modular design like that, where they would tell you, like, you can get any of these five different translucent plastic things you can clip on your headphones and make them unique. So that's one version of modularity. I don't think that version of it ever went anywhere, and for good reason. That's not modularity. That's just, like, skinning. That's, like, the faceplates that you get for your Xbox. The little clips are going to break off, and it's just going to fall off. Yeah, and all you're doing is adding. weight where there didn't need to be and adding bulk. And another creaky part. Yeah, exactly. So that kind of thing nobody's really asking for. But serviceability and modularity in wearable parts, that's the important thing. The important thing with the Fender headphones, things like replaceable batteries. Because when you're looking at modern headphones, like back when headphones were wired, there were two ways your headphones would die. One was the ear pads would rot out and you could replace those for $20, $30. The other was the cord would wear out where it meets the headphones, usually, like at that joint, or maybe at the plug side, but usually at the part where it meets the headphones. And for most people, most people couldn't service those tiny little wires, and, you know, so that usually you would just replace them at that point. Well, the good headphones would have connectors on both ends, so you just buy a new wire. Well, see, that came later. That was not common until relatively recently in the world of headphones, like the last 15 years maybe that's when that really became common. with wireless headphones. Once everything moves to Bluetooth, you still have the earpads being possible to wear out over time, but usually what wears out far before that is the battery. As we know, with any rechargeable battery-based device, those batteries have a certain lifetime, beyond which they don't really hold a useful charge. And if you're using a lithium-ion battery in some kind of device regularly every day or a few times a week, you're going to get a few years out of it maybe. Once you go past four or five years, you're probably not going to have a lot of battery life left. And headphones, like the rest of headphones, the actual speaker drivers, the enclosures, most of that stuff can last way longer than four or five years if it's not abused. But to have then the battery, which is the most easily worn out component, to have that be replaceable is great for extending the lifetime of headphones in the same way that, which I'm just saying, in the same way that when wired headphones switch to having socketed on both ends cables, that made those headphones last for long. Like the headphones, most wired headphones I have that I use in regular use, they're at least 10 years old. Some of them are even older than that. But, you know, I've had to replace the earpads a couple times, and that's it. And so, like, when you have, see, also monitors, when you have really good components that don't really go out of date, just kind of wear out, it's really good for longevity if you can replace easily the parts that will wear out. So this is all good. This is a long way of saying this is all good in the world of headphones, to have replaceable batteries in wireless headphones, because that is by far the earliest part to wear out, and the headphone can be useful a lot longer than that if you can replace with batteries. One of the things they're showing on this III website is that, I'm thinking of another thing that breaks on headphones, is that the plug for wired headphones, the wire that goes into your headphones connects to one of the ear cups, If you're lucky, you don't have those ones that connect to both. But anyway, if it kicks in one of your ear cups, that means there's a wire. It's somehow that takes the sound signal from that one ear cup over to the other ear cup. Usually that wire goes across the headband. And sometimes that wire that brings the signal from one ear to the other can be small and can wear out, especially since the headband is stretchy. And this video they're showing on their website for their TMA2DJ $200 headphones shows that their entire – little like whatever the 3.5 millimeter like headphone jacks oh yeah i think there are 2.5s up there but yeah yeah so the band is replaceable and if that breaks or that wire wears out or something else happens to it you check that away you buy another band and you plug the band into each of your ear cups and problem solve is it is very modular yeah that's hardcore um and yeah that that is a that's a great situation there also the the other ways the headphones most often break is they can crack right in the middle of the headband. If it's like a reasonably inflexible plastic headband, which most of the big over-ear noise-canceling ones, they have a mostly plastic construction, which is, again, largely good for weight, which is important for comfort. But over time, the stressing out of plastic where it bends on the headband, that can sometimes crack. And they've gotten better over time. Most of the models out there today have evolved with designs. That doesn't happen that often. but it used to happen a lot more. All right, and then Tyler H. writes, I tried the Fender Mix at CES. They're not heavy and sound decent. I think the draw should be the high-res audio, but customization is neat and battery replacement is fine. I've also used the Fairphone modular earbuds, and I hated them. They sound bad and are bigger, so why would you want to keep them around any longer? At least the mix looks fine and sounds good. Can I just go on a very, very, very quick rant about high-res audio? I'm timing you. Very, very quick. Okay, it sounds warmer in here. Bluetooth audio is not very good in its regular Bluetooth forms. To make it better, you don't need, like, 24-bit 96 kilohertz. You don't need that. You just need less compression. So people who are being sold, like, high-res audio, like, it sounds better because that's 24, 192 or whatever. No, it doesn't. It sounds better because they're using less compression. And they can use less compression on 1644, and it'll sound just as good, and you won't notice any difference whatsoever, especially in your not incredibly advanced Bluetooth headphone drivers. But anyway, that's it. I'm good. That was very fast. I'm proud of you. All right. And then way back in episode 672 in Ask ATP, we were asked about the difference between minimizing, hiding, zooming, et cetera. And Ben writes, here are two incredible free open source Mac OS apps that fix the where did my window go problem in my life. The first is Flash Space, a blazingly fast spaces replacement to find workspaces, switch instantly via hotkeys or gestures, and use floating apps across workspaces. Additionally, secondly, Alt-Tab, a Windows-style Windows switcher done right. Thumbnails that show the individual window content, separate keyboard shortcuts to show all Windows versus current desktop versus current app, quick access to quick close, minimize false screen, extremely customizable, and a drop-in replacement for the Mac app switcher. And these two apps pair absurdly well. Flashspace nails content switching, the right set of apps, then Alt-Tab nails window switching inside that context. If you've ever lost Windows, cross-spaces, or desktops, this combo is a lifesaver. These apps are kind of right in the sweet spot of things that are possible without too much effort on the Mac. As far as I know, they're all using the existing, very old, very creaky, very difficult-to-use accessibility APIs. and I believe they also both need like screen recording permission and you got to jump through all these hoops. They can't be sold to the Mac App Store. There's nothing to use the accessibility APIs. It can't be sold to the Mac App Store. Anyway, they take advantage of everything. You don't need, you know, it's not system integrity protection. It's not a HACSI or anything like that. It is just like use the accessibility APIs, use screen recording permission because once you have screen recording permission, now you can get little images of the windows so you can make little thumbnails out of them. And with accessibility, you can know where all the windows are. You can make them appear and disappear, bring them to the front, bring them to the back. You can do all sorts of stuff like that. You can't do very easily the kind of things that like stage manager does, but that's not what this does. Flash space and alt tab essentially like say, okay, so you're telling me I can make windows appear and disappear. And I know where they all are. I know what they'll look like. I know what apps they belong to. Fine. I'm going to build on that. And it is very, it is very straightforward. It's like you build up sets of apps just like spaces, but instead of, you know, with actual spaces, you're like swiping from side to side. And I don't know how it's implemented on the covers, but this is, think of it this way. It's like simulating spaces by just controlling what is visible at any given time on your one and only space as far as macOS is concerned. And they do it without animations. And so it seems really fast because, hey, hide these seven windows, show these four windows. Right. And having a floating thing of like, OK, these these these windows are seen in all of the flash spaces. But this context is just these windows. And this context is just these windows. And it is very configurable. And you could set up these little contexts. And if that's the way you want to work and you're sick of waiting for spaces to switch around, this is the thing for you. And alt-tab, same deal, if you want to be able to sort of alt-tab, command-tab. Anyway, through – They've infected you with all this Windows talk. I know. That's what they called their app. If you want to do Windows-style alt-tabbing instead of the Mac style where you have, like, where it's a hierarchy where you do command-tab through apps and then command, like, tilde to do the Windows within the app, this will bring you through all your windows and it'll show you thumbnails so you can identify them. And like, that's the way you want to work. You should check out these apps. They're pretty well implemented. This is not the way I work with windows at all, but I'm glad these apps exist for people who want them. And if you didn't know they exist, now you do check them out. We are sponsored this episode by Squarespace, the on one website platform designed to help you stand out and succeed online. Whether you're just starting out or scaling your business, Squarespace gives you everything you need to claim your domain, showcase your offerings with a professional website grow your brand and get paid all in one place So Squarespace makes it super easy to get a great website for you and your business From consultations to events and experiences you can showcase your offerings with a customizable website designed to attract clients and grow your business. Then you can get paid on time and easily with professional on-brand invoices, online payment support, and you can streamline your entire workflow with built-in appointment scheduling and email marketing tools, and so much more. No matter what your business is, whether it's digital goods, physical goods, they have, of course, integrations with things like sales tax and shipping services and things like that. And digital goods, you can do things like time slots if you're a consultant or a trainer, say. You can sell PDFs. You can sell private podcasts, membership newsletters. Anything you can think of to have a modern digital brand, you can sell all that with Squarespace. It's all backed by everything you might need to run your business, things like SEO tools, analytics, email campaigns, and so much more. And what's great about Squarespace is that you don't have to be a nerd or deal with software or packages or anything like that. Anybody can do it with Squarespace. So if somebody comes to you as a nerd and they're like, hey, I need to make a website for my business, you don't have to do it for them. You can just point them to Squarespace, using our coupon code, please. You can point them to Squarespace, and they can do it themselves, which is amazingly empowering. So you can go see for yourself at squarespace.com slash ATP. Use offer code ATP when you're ready to launch to save 10% off your first purchase of a website or a domain. But you don't have to purchase up front. You can try the free trial first. Start there, squarespace.com slash ATP. When you sign up, use code ATP for 10% off your first purchase of a website or domain. Thank you so much to Squarespace for sponsoring our show. New AirTags were announced. We have been waiting for this for months, some perhaps more than others. But Apple has announced new AirTags. They have a few very cool features. There's enhanced range and findability with a second-generation ultra-wideband chip. Precision finding works from up to 50% further away than in the previous generation. An upgraded Bluetooth chip expands the range at which items can be located thus. For the first time, users can use precision finding on Apple Watch Series 9 or later or Apple Watch Ultra 2 or later to find their AirTag. And the new AirTag is 50% louder than the previous generation. Also, apparently the chime has changed from F to G. And if I knew more about music, I would probably be more impressed. Yeah, follow that link. There's a video of someone playing like the old air tag versus the new one. It sounds different. I mean, I don't know if they did that on purpose. It was just a side effect of the louder chimey thing that they use to be louder. And it just happened to be a different note. But you will be able to tell the difference between the old ones and the new ones as they chirp. I was waiting for these because we have a couple of air tags that are just like, for whatever reason, just don't hold the charge anymore. Like, you know, I have a ton of these little cheap coin batteries, and, like, this one AirTag, like, every time we put a coin battery in there, it just eats it. So, like, okay, that one needs to be retired. Let's buy a replacement. And I said, no, don't. New AirTags are coming out. That was, like, six months ago. So, finally, new AirTags are out, and we can finally buy them. And we will. As much as I, you know, I appreciate the utility of AirTags, it's one of those products, like, the instant it was released. there were obvious problems with it. And after now, you know, what, five years is it? They've solved none of them. Like, the most obvious problem with it was there was absolutely no reasonable way to attach it to anything or attach anything to it. So the only way that people use AirTags is either showing them loose somewhere where they bounce around and fall out or putting them in some kind of accessory. but the air tags were not designed to accommodate accessories at all. Make a robotic air tag, Marco. But even the most simple thing, like a loop in it, something, like some kind of accommodation to recognize people are going to be using these in lots of different physical situations. Let's design it so it accommodates anything. Because instead what happens is, suppose you want to put it somewhere, on something small, like on a keychain or in a wallet or something, what has to happen is you have to have something that wraps all the way around all of its size just to keep it secure, which makes that whole thing bigger and bulkier than it otherwise would need to be. So it's a product that, like, the physical design of it has obvious problems. Everyone saw this. The instant it was released, see also the Vision Pro and the AirPods Max. Similar pattern. Like, Apple puts these out there, there's obvious problems from day one, and then years later they release a new version that addresses none of them. I got to give Apple credit. Like, when they half-ass the product, they gloriously half-ass it. Bravo. They probably make more money from the accessories that they sell for AirTags than they do from the AirTags. I don't – the accessories are definitely, obviously, very high profit for them. I don't know how many they sell. I mean, I see other people selling a bunch of AirTag accessories. Yeah, I mean, there's a million cheap things you can buy on, you know, everywhere for them. And speaking of that, we have an AirTag on one side of our keys, and we bought whatever the, like, you know, $2 little AirTag holder that lets you put it on a key chain from, like, Amazon or something many years ago. And it's made of, like, that rubberized plastic that, like, gets, like, sticky and tacky as it gets old. Yeah, you got to throw that away. You know that, right? Yeah, you got to what you pay for. That's really toxic. Yeah, that's gross. Yeah, service announcement, everybody. Once plastic starts emitting oil, you need to throw that away. That's not good for you. The problem with the Apple ones is they're bigger and bulkier than, like, the cheap third-party ones. These Apple ones are very expensive and also very large. Anyway, I'll be getting new AirTags. They should have a hole. You can find a million YouTube videos of people drilling holes in them. Have you seen all those? Like, you know, there's DIY ways to do this that I don't recommend this, but it's a thing you can do. Yeah, I mean, so the other thing is, like, I guess we will learn once we disassemble them. There were, so obviously, like, as soon as AirTags were launched, the other major problem with them is stuff like stalking. And Apple has done various mitigations over the years to try to make that less of a problem. It's still a problem, but, you know, this entire product category enables a level of stalking and surveillance that is, you know, very problematic in a lot of cases. And so for this product to exist at all, I think Apple does a pretty good job with mitigating that, if it's going to exist at all. That being said, I mean, back when it was released, we actually had conversations on the show about how, like, you know, maybe this is too much trouble to even be worth Apple having this product. But I guess they decided otherwise. But anyway, some of the rumors about this update before it came out were that maybe they might be making things like disabling the speaker more difficult. I guess we'll find out, like, as people get these and start, you know, taking them apart and stuff, I guess we'll find out if that actually came to pass. That would actually be a reasonable upgrade to have made. I just don't know, like, the physical realities of how possible that is. Yeah, I don't think it's possible. If someone has physical access to your AirTag, I mean, you can just put hot glue on the thing that moves to make the noise. Like, there's no stopping someone from, it's not like, you know, like, oh, is it possible to make it turn the camera on without making the green LED come on on your, like, MacBook Pro screen? Like, that is a tractable problem that you can solve in various ways. But, like, I'll give you physical access to this AirTag, but I'll make it so you can't disable the speaker. You can always disable the speaker. Like, it'll be very, very difficult to make it so someone with physical access couldn't disable the speaker. But, yeah, you know, Apple does what it can, and they're always refining the software approach to do it. And the thing is, it's a security convenience thing because, you know, we're using it within our family and we're not stalking each other. But we are constantly running into annoyances that are caused by the stalker mitigation. Right. So like like anything like this, trying to deal with the worst possible actors makes everybody's life more difficult. Because now you're like unknown air tag moving with you and blah, blah, blah. It's like, oh, it's just me. It's just, you know, you know, that stuff is there. But once again, the bad actors in the world make things worse for everybody. With regard to the environment, 85% recycled plastic in the enclosure, 100% recycled rare earth elements in all magnets, 100% recycled gold platy in all Apple design printed circuit boards. The paper packaging is 100% fiber based and can be easily recycled. And it's maintaining the same form factor as the original with this new AirTag being compatible with all existing AirTag accessories. They spun into the positive marker. You know why we didn't change it? The environment. That's why. It's not because we don't care. Yeah, well, and to be clear, like, in the bed they created for themselves, you can make an argument that, like, we shouldn't break accessory compatibility. Like, that is a reasonable argument to make, and that does have, you know, a cost to it if you wanted to break it. But, God, think of how, like, how much better the AirTag would be if it changed its shape to be a little bit more, to be at all accommodating of the physical reality of how this product was ever actually used. Remember when the very, very first iPad, the iPad 1, did not have any accommodations for a case. And so the case that went around it was this huge, thick rubber thing that had to go around all sides of the four sides of it just to hold on to it. And it added a huge amount of bulk. And the amazing innovation of the iPad 2, in addition to, you know, shedding a bunch of the thickness and weight and everything, the iPad 2 added magnets to the side, and they had that wonderful smart cover that gave you protection where most people needed it, the screen, without having to wrap around the entire iPad and bulk the whole thing up if you didn't want it to. So that was a great example of Apple recognizing the physical realities of how these products need to be used in the real world and designing something to accommodate that in the product itself to make the cases and accessories better. See also the Apple Pencil when it went from the weird, you know, excited to see you method to the magazine on the side of the iPad that hold it on there. It's not perfect, but it was so much more accommodating of the realities of using the product. Like, okay, we have a pencil. We need to, like, store it somewhere where we're not using it. Instead of having to, like, have it, like, be loose in your bag or wherever, let's have a few magnets in the iPad that are designed to accommodate the reality of this and have the pencil stick to it when people use it. You know, you can say, well, the iPad should just be nothing. It should be the Naked Robotic iPad and have no accommodations because what about the people who don't have a pencil? Those are wasted magnets. and that's an argument you can make. I think it's a bad argument for a very, very common use case of the product. Like, if it's something that, like, almost no one needs, that's a different story. But if it's a very common use case, the product should be designed with that in mind and to accommodate that as best as it can. And that's why every iPad, since the iPad 2, has had magnets to support cases because it makes them a lot better. The AirTag is the opposite of that. The AirTag has no accommodation whatsoever for any of the ways it is physically used. See, also the AirPods Max. I would say magnets inside the iPad are part of the make-a-robotic core thing because the idea is we don't put anything on it, but we provide places for you to put stuff on it. The magnets don't stick out, and they're in space that is otherwise unused inside it, so it's fine. And by the way, I feel like they could have, if they really wanted to, probably at additional cost of the product, put a hole in these suckers while still maintaining compatibility with all existing accessories because all people want is a little hole that you can put a little ring through. You know what I mean? I know it's difficult because it would be really thin. They have it in the AirPods case. Yeah. You know, there's the coin batteries in there. It's really close to the edge. You need some, like, very strong, like, metal ring thing around it so that little piece doesn't break through and blah, blah, blah. But anyway, they chose not to do it. I feel like the next time, if they keep selling this product and they eventually revise it further, the next time the only way they can move forward is they have to move to a smaller size coin battery. And that gives them a whole bunch of options because once you are, first of all, it would be smaller, period. And once you're moving to a smaller size with a smaller battery, you have more options for attachability instead of trying to put one very small hole with reinforcement around it on the very, very edge of the existing form factor. All right. Price and availability. It is available today for the same price as the AirTag 1, $29 for a single AirTag and $99 for a four-pack. You can also get free personalized engraving at Apple.com and the Apple Store app. And you should because you should engrave little emojis on them so you can call them by their names. All right, we are recording on the evening of January 28th, and I'm excited to tell you that we can't talk about new MacBook Pros because they haven't been announced yet, at least not as we sit here today. But apparently there's an event going on for creators yesterday, today, and tomorrow in Los Angeles. Apple has invited select creators for an Apple experience in Los Angeles, scheduled for January 27th through 29th. Creator Peter Mara shared an image of this, which we will put in the show notes. Save the date. Hi, Peter. Save the date for an Apple experience in such and such time in Los Angeles. Confirm availability is linked, and then please confirm or decline by January 9th. That's all. I'm assuming they're all there doing creator things. I mean, everyone thought, like, this would be a good time to announce the new M5 Pro and M5 Max MacBook Pros, but it seems like they're probably just showing them creator studio. I think the embargo just lifted for that because I saw Jason Snell's review of creator studio. So, I mean, I guess we'll find out. I mean, maybe they could release the MacBook Pros on the 29th, and maybe the people who are at this event are already using them right now. We don't know. But, yeah, we do feel like sometime early in this year we will see the M5 Pro and M5 Max MacBook Pros, just not today as we record. Yeah, that definitely seems like the answer here. You know, Apple has a lot of smaller events where they try to promote something to some audience, and they aren't always hardware releases, and especially when they're doing like creator events. It's usually to try to, you know, it's usually stuff that's less interesting to a podcast like ours. It's like we're going to try to get a bunch of like YouTube creators to use our new version of Final Cut or whatever. Like it's that type of thing. So it would totally make sense now for this one, especially given the timing of the embargoes dropping today for the new Creative Suite or whatever it's called. Creator. That's right. It's totally separate from Adobe's Creative Suite. It's a studio. It's definitely, you know, ACS. It's a different ACS. It's not Adobe Creative Suite. It's Apple Creator Studio. Totally, totally different. Academic computing system. Yeah. Anyway, so, yeah, it makes sense that this was not related. And, look, we might be getting M5 MacBook Pros tomorrow, for all we know, but it's not related to this event. All right. Marco, you and I can, I think, buzz off for probably 30 to 45 minutes, because, John, you have some stuff to tell us about. We promised this last week, but Sony's TV business is being taken over by TCL. I'm sorry for your loss, John. All these years, I figured it was pronounced tickle. What a waste. I mean, you can pronounce it that way, I guess. I'll set you up, John, by reading from Jess Weatherbed at The Verge. Sony has announced plans to spin off its TV hardware business, shifting it to a new joint venture with TCL. The two companies have signed a non-binding agreement for Sony's home entertainment business, with TCL set to hold a 51% stake in the new venture and Sony holding 49%. Sony and TCL are aiming to finalize binding agreements by the end of March and start operating the new joint company in April 2027, subject to regulatory approvals and other partnership conditions. The new company is expected to retain Sony and Bravia branding for its future products and will handle global operations from product development and design to manufacturing sales and logistics for TVs and home audio equipment. Sony says that the partnership will leverage Sony's picture and audio tech, brand values, supply chain management, and other operational expertise. This will combine with TCL's own display technology, vertical supply chain strength, global market presence, and end-to-end cost efficiency. So based on the headline, everyone's like, oh, no, if you love Sony TVs, now they're dead. TCL is taking them over, and it'll just be TCL TVs. But if you read, you know, two paragraphs into any of the announcements, you realize, oh, Sony's still going to be selling Sony TVs. It's just that they're having this partnership with TCL, and yes, TCL is 51%, and they'll have 49%. So TCL will technically be in control, but there'll still be Sony products with Sony's all existing Sony proprietary stuff going into them. And the reasons behind why they might want this partnership are complex and have to do with manufacturing and stuff. And we'll get to some of TCL's announcements from CES in a little bit. But I'm actually not too broken up about this. It just seems like a sort of a business reality for Sony, and it's probably a good deal for them. This DCL has lots of manufacturing capacity and some interesting new technology they're coming out with, and that's what Sony needs, apparently. I mean, Sony has been contracting out the assembly of its televisions for a long time now. They're just changing the arrangement. I believe Sony's also part of this deal, like Sony's receivers and stuff are going to be part of this as well. And the final thing I'll say before I do an aside on TV tech is that this is like a memorandum of understanding or whatever. Like many things can go wrong that will make this deal not happen sometime between now and 2027 when they're supposed to start doing business. So maybe you won't get regulatory approval. Maybe the deal will fall apart. So stay tuned. This is not a thing that has happened yet, but this is just a thing that these two companies say that they want to happen. And I think this is a good time to take a couple steps back and look at where we are in the television technology market to see how this deal might fit into it. And this is from the perspective of, like, I don't know, technology enthusiasts, right? The TV market is huge. Most people buy inexpensive televisions. And like the sort of enthusiast level of the market does trickle down to that lower end of the market where the action is happening on the high end. Right. So that's what I'm interested in. That's what most TV nerds are interested in. And on the high end, for the past many, many years, the market has been focused on, you know, sort of competition between two different ways of making good TVs divided by whether or not you can turn on individual pixels. Every time someone talks to me about TV online, I always say that I demand per-pixel lighting control. That is obviously the technically best way to form a nice picture, which is that you can turn individual pixels on and off. And you're thinking, doesn't every TV do that? Doesn't every monitor do that? What are you even talking about? Well, lots of televisions and lots of monitors to make a single pixel turn on have to turn on a huge amount of light behind that single pixel. Studio display. Turns on the entire backlight. If you just want one pixel to be turned on. All the other pixels are trying mightily to prevent the backlight from going through. But there is just one giant backlight. And if you make a black screen with a white pixel in the middle of it, the entire backlight is on. And the screen is trying to not let you see any of it except for that one white pixel. That's very difficult. and various technologies for stopping you from seeing that entire backlight have been tried over the years, but they can't stop all the light, which is why if you take the studio display, put it in a pitch black room and make the studio display show a black screen, it will light up your room because that's just the way it is. And then as you go down the spectrum, okay, let's break up the backlight into a bunch of little pieces, and then we'll just turn on the little pieces that are behind things. So if we have a single white pixel in the middle of a black screen, will just turn on like a one inch by one inch square behind that pixel. But then you get a little glowing region around the one inch pixel. People refer to that as bloom. It's very difficult, again, to keep the light. The other parts of the monitor don't have any backlight turned on, so they're totally black. But the little pixel has a little bit of bloom around it, so you try to turn the backlight down a little bit and combat bloom. And anyway, that's one whole section of televisions, which is we cannot turn on individual pixels and make just the pixel emit its own light. We always have to turn on some portion of the backlight behind the pixels that we want to show up. And those regions have been getting smaller and smaller. It used to be like, oh, we divided it up into like five regions, 16. Now there are thousands of backlight regions on like modern displays, not the studio display. But, you know, Apple's Pro Display XDR has what, 570-something sticks backlight regions? And it's like a six-year-old thing. The monitors we were talking about from CES are like 2,000 zones. But, of course, how many pixels are there on a 5K display? Way more than 2,000. So that's one side of things. And the other side of things are the technologies that allow per pixel lighting control. And the one that is most commonly used in television these days is OLED. But just like on your phone or on the new iPad Pro, OLED, you can turn on a single pixel because the pixels themselves emit their own light. There is no light behind them that's shining through them. And this big chunky thing, the individual pixels make their own light. and there are traeoffs between them because OLEDs can't get as bright as those other ones because when you've got the big light behind everything you can crank up the light real high. More recently the battle between these things, these two different kinds of technologies has advanced to the point where it's not just brightness anymore. It used to be, okay, well OLED has the best picture but LCD TVs with backlights can get brighter and the competition is now in something called color volume, which is like, okay, how many colors can you show? And it used to be that the fanciest OLEDs were winning that as well, because the QD OLED, the quantum dot OLED that I've got on my TV, and that's in a lot of monitors now, had R, G, and B subpixels, and you could turn the R, the G, and the B on to maximum brightness, and you could get a really pure white, or you could turn on the R to maximum brightness, get a really pure red, so on and so forth, whereas those other monitors that had a backlight, they had a backlight that they would shine through, and then they had to have something that would turn the backlight into other colors, some kind of color filter or quantum dots or whatever. And on the OLED side, there were some OLEDs that had to add a white subpixel to increase the brightness. And of course, that would wash out all their colors. Anyway, QD OLED was the champion in color volume as well, because it didn't have a white subpixel. You had perfect lighting control. You could turn up the red pixel really, really high. It wasn't as bright as the best LCDs, but that was sort of like the enthusiast thing. All the Sony A95L, Sony A95K, all the Sony monitors and the Samsung monitors with Samsung's QD OLED panels, they were the champions because they had perfect lighting control, they could get really bright, and they had huge color volume. This is sort of an Empire Strikes Back-like situation where the televisions with the backlights are saying, okay, we've got a new idea. It's not really that new because Sony did it ages ago. But anyway, in the past few years, it's been like, instead of having a backlight that is just a white or blue light, most of them are blue because you can change the blue until the other colors because it's the shortest wavelength, how about we have a backlight broken up into little regions but we'll make the backlight itself like the little lights that are you know behind there the little backlight regions they will be rgb this is like a gamer's dream rgb lights they're behind my tv right and it's really complicated to do that because if you think about it i'm going to show you an image and you have you know 2 000 backlight regions and you know what color every pixel is going to supposed to be but for like the one inch by one inch region or the one centimeter by one centimeter region that's behind this particular set of pixels in this like you know drawing of a landscape or something what color should that rgb backlight be because you just got one r one g and one b for that little tiny region of the backlight what color should we make it well if most of the pixels are blue you can just make the backlight blue and then you get super duper blue and it'll be really bright with lots of color volume because you got a blue backlight going through a blue filter it'll great but what if there's like tons of different colors in that little one centimeter region on your screen what color should the backlight be now do you average them like anyway that's what that's why it's computationally tricky to do that but what it gives these screens is more color volume instead of just having a blue backlight with color filters that are imperfect now we can crank up the color volume with these rgb backlights in fact some of them don't just have rgb behind there. They have RGB backlights and some of them add a yellow RGBY or a cyan RGBC backlight regions. Very complicated. The war of words at CES was very hot because some people are like, well, if you do that, like then you have your RGB backlight region and you decide this backlight is going to be orange, but now you have some pixels that didn't need any orange and then we're getting a little bit of orange bleed through. So your colors are all muddied because you're not able to control the colors, the individual pixels you're just putting. Anyway, Anyway, it's very complicated. Sony is out there with its own sort of not particularly consistent idea of how to make a good TV, because for a while they were selling essentially the best TV in the world for years and years. But at a certain point, somebody decided this is not going to be our flagship TV, even though everyone keeps saying it's the best TV in the world. How about we make this other TV our best TV? because we think it's better to have brighter TVs that are less expensive than the less bright TV with the OLED stuff. And so they would say, yeah, we still sell those OLEDs, and they're still pretty good. It's the reason the A95L, which was their flagship, became the Braviate 2, which is not their flagship, even though it's still their best TV, because they wanted to sell LCD TVs with backlight regions on them because they could get brighter and they were cheaper. And this is complicated by the fact that the only company in the world, I believe, that makes cutie OLED screens is Samsung. and the other big company in the world that makes OLED screens that are not QD OLED is LG, and now they make the tandem OLED screens that don't have white subpixels in them. And there's probably other OLED makers. I'm not sure if they make them at TV size, but the point is it's not like Silicon Chips where it's just TSMC and like a few other companies or whatever, but the companies that can make TV size panels in any kind of volume are small in number. Everybody else uses those panels. Sony uses them, Panasonic uses them, obviously Samsung uses them, LG uses them. And TCL is one of those companies. TCL makes panels. They make LCD panels, and they're building factories to make OLED ones. So Sony partnering with them makes some sense because, like, look, we weren't making our own panels anyway. And we're not quite sure how to go forward in this business because just being at the high end is not giving us high enough volume. So we don't really have a low-end story. So TCL, you have huge amounts of manufacturing. You take stuff over for us. So that's kind of the shape of things here. It's still per-pixel lighting versus not per-pixel lighting. It's still companies taking panels made by other companies who are probably their competitors and packaging them into TVs with their own processing and everything. And, you know, Sony's been doing well there. Sony has good processing. But still, in areas like why do Sony televisions still only have like two HDMI 2.1 ports, whereas LG and Samsung have four? Because Sony outsources the chip for handling that, and the company that makes the chip hasn't made one that can do four. And it's just, it's a weird and complicated landscape. And I don't really blame Sony for doing what it's done, which is we need a manufacturing partner who can make huge amounts of TVs cheaply and who looks like they have a good technological roadmap. And speaking of roadmap, it's time to wake Casey up and talk about this next week. Huh? What? Hi. All right. So TCL has super quantum dot technology that is landed or been announced, I guess, at CES. there is a video that we will put in the show notes about this. Then from TCL's official Reddit, of all places, TCL's Super Quantum Dot technology, or SQD, represents a major advance in color performance, delivering up to a 33% increase in color gamut and a 69% improvement in quantum dot accuracy compared to previous generation TVs. RGB mini-LED systems can also produce vibrant color and cover a lot of BT-2020, which is the color space, I think. Is that right? It's like P3, but bigger. Right. But because they use separate red, green, and blue LEDs, the blending of light can sometimes cause color crosstalk or color bleed, where colors overlap and reduce accuracy and fine details. We wanted to create a TV that could get this level of color without compromise, and TCL claims that their X11L TV covers 100% of BT 2020 color. Super Quantum Dot technology avoids this issue entirely. By generating color through quantum dot conversion rather than colored LEDs, SQD delivers cleaner color separation and higher color precision, even in complex or high contrast themes, using a single chip pure white light source and SQD using more refined 5 nanometer filter particles down from the standard 60 nanometers to ensure each pixel renders color with extreme precision and no blooming or interference. So this is the next salvo here, and this is an announcement from TCL. They're saying, okay, we don't have per pixel lighting control, but everyone else is putting colored backlights behind their pixels. That's bad, and let me tell you why it's bad. And we don't have to do that. We're going to stick with a single color backlight, like we always had, broken up into little regions. But now our quantum dots are better. Quantum dots are the things that take, like, the blue backlight and change it into a red, green, and blue. Their quantum dots do that conversion better, allowing more of the color through, which is how I think this is the first I've ever seen of a quantum dot LCD TV that can do 100% of BT 2020. like historically i think the only tvs that have become close to that are the um qd oleds uh because they don't have the white sub pixel and now maybe the tandem oleds from lg that don't have the white sub pixel diluting your colors they're trying to go for you know color volume can we cover this entire volume of this color space and the color space is already shown as like a 3d thing because you've got rg and b in the three different dimensions so it's like this 3d shape filled with color and bt220 is very big and inside that is p3 and way inside that is srgb or no rec uh rec 707 or whatever the t the plain standard ftb one anyway they're saying we got these super quantum dots they're better than your quantum dots we don't have the weird colored backlight thing we don't have to do the process thing to figure out what the hell color to make our rgb backlights we don't have the color bleed through we still do have the blooming thing because you We don't have perfect light control, but whatever. And part of the reason this is a big announcement is that TCL is, I think, the only company in the industry that has decided to do this. Not because it's like, I don't think it's like exclusive tech to them, but like to do this, they basically had to either redo their existing factory or build a new one. I'm not sure which they did, but like you have to change your entire LCD manufacturing line to do this. And everyone else is like, our current stuff is fine. We don't want to bother with that. But TCL decided to make the investment. and now I believe they're the only company for now that is going to be putting out these super quantum out things and these are the TVs most people buy most people buy don't buy OLEDs most people don't buy certainly don't buy QD OLEDs those TVs are very expensive people buy plain old LCD TVs because they're cheaper to manufacture and cheaper to buy now this X11L TV with 100% of BT 2020 not a cheap TV but you know in theory as volumes increase this technology will trickle down It is, you know, in theory, it is as cheap to manufacture as LCD screens that they're putting on RAM. There are tiny OLED screens that are putting on RAM. It's inexpensive to use this technology. And so Sony is partnering with them. In theory, they could have first dibs access to a LCD non-perpixel lighting control technology that nobody else has. Meanwhile, just the year or two before this, Sony was like, we've got our own RGB backlight. Look at our backlight. We have all different colors here, and we have this cool processing to figure out what color to make all the little backlight regions. And they did the thing I think I talked about on the show where they stripped off the layers of the TV so you would just see the backlight and say, look, you can kind of see what the picture is even just from the backlight. It's like, that's great, Sony. I just look at it and roll my eyes, and I say, per pixel lighting control or bust. Anyway, that's Sony and TCL. I'm not too broken up about it. I don't buy these TVs anyway. These are most of the TVs that Sony wants to sell. They should partner with TCL to make these because they look like really good LCD TVs. All I care about is the high end, and the high end is LG Tandem OLED with no white subpixel and Samsung QD OLED with no white subpixel. That's all I care about. I will be sad if Sony stops making an OLED TV, but the rumors are not that they're going to do that. And in fact, TCL is building an OLED factory. The details of the OLEDs that they will build in that factory are as yet undetermined, but my fingers are crossed that they will not have a white sub-pixel, and Sony can finally stop buying all their panels from Samsung. How close are we, do you think, to micro-LED? Like, what are you going to call the car? It's always five years in the future? Yeah. Like, they show those TVs all the time. They're amazing. They cost $50,000. They're the size of your wall because making them small is also expensive. Like, they've been showing those at CES for, I don't know, like a decade now. They are amazing. What Marco was talking about is, hey, you want per-pixel lighting control? How about every pixel has a tiny red, green, and blue LED in it? Isn't that sort of what OLED is, right? Not really, because OLED is kind of like a sandwich with a light-emitting layer and stuff. It is per-pixel, but they're not, like, literally LEDs. Think of, like, an LED that you could buy for a little kit and shrink it real small. That's not what OLED is. It's the organic part of it that makes it susceptible to burn-in. It's kind of a sandwich, right? actual micro led where every individual pixel has a red and green and blue light up led those exist you see them in stadiums because it's really easy to make them big because if it's in a stadium the red green and blue uh blue uh um leds are huge i can see just it's like a light bright right from close up it looks terrible but like across the stadium it looks great trying to get that down to a 4k 55 inch tv do the math on how small those leds have to be and the problem is look the manufacturing. Like, you know, do you have some way to manufacture that? Like I can't do the math on how many pixels there are in a 4K TV, but there's a lot. Uh, and if you need to like place them individually and connect wires to them, that's not the way you do things. You need to be able to essentially print them. Um, that's why they're like Japanese companies have these printable OLEDs where you kind of use like an inkjet printer, kind of the same way they print the LCD stuff. Um, so far there's been no manufacturing breakthrough. So that technology continues to be a curiosity for extremely rich people. Uh, it looks real cool. It's the yes, but no, there's a bunch of other technologies too. I like, I don't remember the acronyms off the top of my head, but there's a bunch of other ways to produce pixels that make light besides OLEDs. They're just like in the experimental stage where they'll show like a postage stamp size, little experimental thing in a back room. It's the yes. And so maybe someday we can make TVs out of this, but, um, if someone actually eventually does it and it becomes manufacturable, you I sure you hear about it here from me but for now it QD OLED and Tandem OLED And honestly Tandem OLED is very exciting That what Apple M6 MacBook Pros are going to use sometime later this year as rumors have it. And that's what the iPad uses now. I'm excited about that because it's a really good technology, and those screens look great. So I'm happy with what we have. But, yeah, micro LED. And, boy, the naming, RGB mini LED LCD. It's ridiculous. It's so confusing. SQD, Super Quantum. All right, what's worse for naming TV display technology or USB standards? TV display, because in TV displays, every single manufacturer has their own branded version of what the actual generic technology is called, and USB didn't do that. We are sponsored by Delete.me. Delete.me makes it easy, quick, and safe to remove your personal data from hundreds of data broker sites online. at a time when surveillance and data breachers are common enough to make everyone vulnerable. So it's easy to find people's information online. You just search any search engine for somebody's name, and you get all these different sites advertising their home address, their phone number, their family members' names, all this stuff. It's kind of a gross business, if you ask me. And what these are are data broker sites. And data brokers buy and sell and trade and accumulate as much data about people as possible so they can then sell it to whoever comes by and wants to give them some money to get it. It's a creepy business. And what Delete.me does is they automate the opt-out and removal takedown process for all of the data brokers they can find. So what happens is, you know, you can go as a person and you can go opt-out at some of these sites, but you're going to have to figure out the process for each one, upload documents, all this stuff. No one's ever going to really do it, and they know that. Delete.me does it all for you. So Delete.me is an ongoing subscription service. You sign up, and then what they do is they go through all the data brokers they already know about, and as new ones come up, they opt you out of those, and they will file mass opt-outs and mass takedowns on your behalf to take your personal info out of hundreds of data brokers online. So it's much, much easier than doing it all yourself. Nobody ever could do it themselves. And so this is why services like this exist. And the New York Times Wirecutter has named Delete Me their top pick for such data removal services. I personally, before they were a sponsor, went looking for one of these, and I chose DeleteMe too because it's a really good site. Take control of your data and keep your private life private by signing up for DeleteMe. Now there's a special discount for our listeners. Get 20% off your DeleteMe plan when you go to joindeliteme.com slash ATP and use promo code ATP at checkout. The only way to get that 20% off is to go to joindeliteme.com slash ATP and enter code ATP at checkout. That's joindeleteme.com slash ATP, code ATP. Thanks to Delete Me for sponsoring our show. John Ternus is taking over design at Apple. Reading from Mark Gurman at Bloomberg Bloomberg. Apple has expanded the job of hardware chief John Ternus to include design work, solidifying his status as a leading contender to eventually succeed chief executive officer Tim Cook. Cook tapped Ternus to manage the company's design teams at the end of last year. People with knowledge of the move said that Cook himself was trying to expose Ternus to more parts of the company's operations. In this case, one that Cook doesn't know anything about. Hey, the role was held by John Ives. How about the ethics department? Oh, stick burn. Until his departure in 2019, Cook oversaw design from 2015 to 2017 when Ives temporarily stepped back from the position. Jeff Williams most recently held the job up until his retirement at the end of 2025. Ternus is now billed internally as the executive sponsor for all design on Cook's management team. that entails being a bridge between design staff and Apple's top brass. He represents the design organization in executive team gatherings and manages the group's leaders. The heads of Apple's design teams continue to report directly to Cook in both internal organizational charts and the company's public disclosures. Having Ternus oversee the design teams while they still technically report to Cook is a strange arrangement, according to Apple employees, but it's a sensitive situation. Changing the reporting structure would affirm Ternus' status as a rising star at a time when the company is still keeping its succession planning under wraps. they're doing a really good job keeping under wraps. Well, like, he's taking over design, but he's not really taking over design. Like, no org chart change. It was just like that Cook is essentially parachuting him in and saying, just check out this design stuff. And, like, not changing any of the reporting arrangements is so weird. Like, this is the government saying, like, you know, oh, well, it would affirm his rising star status. Like, that's just government guessing what the reasoning might be. But if it's true that the reporting structure has not changed, but instead he's sort of like being an executive sponsor or whatever. It does really seem like just kind of, again, I'm not sure if I believe that it's like Tim Cook thinks Ternus needs some experience dealing with design, but maybe he does. But either way, as we'll see with some news later in this section, it does seem like that if Ternus is the leading contender for, to be Cook's successor, that, maybe he's wishful thinking, that perhaps design is an area that he may want to turn his attention to. and that even Tim Cook can possibly see that at this point? I don't know. Just speculating. Yeah, I think, I mean, look, first of all, like, yeah, this, you know, Apple is not doing a great job hiding the idea that John Ternus is seemingly most likely the next CEO candidate. But I think this is great. The reality is that I think Apple's design leadership under Tim Cook has been just kind of weak overall because, as discussed a million times, Tim Cook seems to equate all design together. And so he's like, oh, Johnny Ive, you are designer. You are a good designer. Here, software, design this. It's all the same. And then you put Jeff Williams in charge, who is not a good designer. Right. And all the reports we've heard so far about Jeff Williams' term, you know, kind of serving that role, basically make it seem like he was pretty hands-off. But, you know, he was there. He was at the meetings, but it sounds like he was pretty hands-off. and nobody seemed to really have any problem with him. Yeah, I mean, he's not expected to do that, but at a certain point, as you go down the org chart, you have to get to someone who knows design. It's fine for the CEO not to know design. Okay, maybe the next person not to know design, but eventually it's like, okay, but who in the company is in charge and also is a designer? At least Johnny Ive was that and Alan Dye was even that, but Jeff Williams was not that. Tim Cook is definitely not that. Yeah, but so there's a couple of things that I think are promising. First, just in general, I think reinforcing the idea that John Turnis is the most likely CEO successor and that that process is starting, that I think overall is very promising for Apple's future. Again, even though we don't know what kind of CEO Turnis would be, he seemed to do an extremely good job at his current role as being hardware chief, and he seems very well liked, and he seems like he has pretty good personality traits as far as we know. We don't know that much about him, but what we do know, it seems like he'd be a pretty good person for this role. And it seems like he is a little bit more product-focused than Cook. Cook is much more kind of high-level operations-focused. Cook doesn't really have a good product sense at all, and he's shown that many times over the years. So he kind of outsources that to people below him, but he's not very good at choosing those people sometimes. So I think Chernus will be better at a lot of those things. So overall, that's a promising part of this. But also, I think this is not unrelated to Alan Dye's departure. I suspect this has kind of been in progress for a while. This kind of thing does not just happen and surprise everyone around it. Like, all the top executives and top people, including Alan Dye, probably have known about this transition for longer than we have. and my guess is that part of why Di left is that some part of Ternus and Di wasn't going to work together and one or both of them knew it. I'm guessing these are actually related to each other and that once Di saw the tide shifting towards Ternus, somehow they realized this isn't going to work and he started looking around. That's my best guess because again, the people up at that level, they know what's going on. They're not idiots. Like, we only hear a drop in the bucket usually after everything has happened. So odds are this is all related. But also, having Ternus be the executive sponsor, and we have to kind of figure out what that means. We'll talk about it in a second. But, like, that I think is promising because I'm guessing Ternus was involved in the selection of Stephen. I forgot his last name again. LeMay. Oh, my God. Sorry, Stephen. I've forgotten your last name like four times in four consecutive episodes. I'm guessing Stephen LeMay being elevated to this position was also not unrelated to Ternus' taking over. I'm guessing this all kind of happened in concert with each other. These were all related. And if Ternus was involved in the selection of Stephen LeMay, and by the way, which I guess we'll jump to the end slightly. Sorry, Casey. Sebastian DeWitt just announced that he joined the team at Apple. He was the lead designer for Halide, and he's been around the scene forever. He's a very, very good UI designer. So it looks like the tide has shifted substantially with the design leadership at Apple on the software side, and it looks like Turnus was probably involved at some level with that. So that's all very promising to me. I think this is a really good sign of things to come. Yeah, even if he's not going up to CEO, this kind of dynamic happens frequently in big organizations and is one of the skills that is actually important to have when you're a higher up or like a CEO, which is recognizing, maybe too late, but whatever, better late than never, recognizing what parts of your company are having problems or potentially having problems. Where are we weak? What are we not doing well? What dangers lurk out there? What are the risks, right? and obviously if you're slow in the draw it takes you a while to realize this and sometimes you know there are problems but there's like you know i think for example with the with the keyboards and laptops with no ports or whatever i think the company more or less knew there was a problem there but solving it you're not going to tell johnny ivy can't have his thin keyboard and then you're not going to tell the engineer organization that they need to stop trying to make the thin keyboard because they're not doing it's like you're kind of stuck where you're Like, I don't want to make these people angry because I actually want Johnny to stay because I need him to build Apple Park or whatever the hell. Like, it's a difficult situation. But, like, knowing that you have some weak area, what you do as a CEO or vice president or whatever is you look down the org chart and you find the person that you think, this person knows what they're doing. This is a competent person. I need someone to come in here to this other part of the org and fix this crap. And so they find somebody who's like, you, you are in charge of X. and X is doing great and you did a great job with X. Can you also look at Y? Because Y sucks. And they'll tell that person, in addition to your current responsibilities, take some time and go over there and fix that part of the org. And, you know, sometimes that works and sometimes it doesn't. Sometimes you pick the wrong person or whatever. But again, even if Turnus wasn't going to become CEO, just simply having him say like, you were in charge of this part of our company that's done great. Over here, I'm hearing, years too late, that we're not doing great on design, broadly speaking. I'm not going to put you in charge of design because you're not a designer, but can you go over there and, like, just fix stuff and, like, just figure it out and just, like, make it better? And, again, we don't have the privy to, you know, we don't know what's going inside the company. We don't know if he's just started doing this this year or whatever, but, like, as Marco said, from the outside it sure looks like we're having a lot of issues in design, All of a sudden, there's a bunch of turnover. All of a sudden, new things start happening in design. We start hearing new names. And the things we know about those names make us think that Apple will be making different decisions related to design in the future because we know the people that left. And we know the people that are coming in. And those people have very different opinions about how to do design. Now, does John Ternus know anything about design? I don't know. But it seems like things are changing in a direction that makes us out here who are dissatisfied with design happier. and if that's the result of John Turnitz parachuting in, even though the org chart hasn't changed and just being like, oh, I'm your executive sponsor or whatever, like representing them in meetings and blah, blah, blah. Yeah, okay, but like as Marco points out, was he part of the decision to pick John LeMay? Was he the person who said, hey, see, you keep forgetting his first name. Yeah, everything should be John, not a Steven's team. Was he the one who said, there's a bunch of great designers out there who love our platform. Can we go hire some of them? maybe like throw money at them. Like it's like the AI problem, except for with what was supposed to be Apple's core competency, right? Let's go get some good people. Everyone in Johnny, I've seen left to go to open AI, right? Everyone, you know, who's good. Like there's all those rumors of like just brain drain from Apple. People either get rich and leave Apple and go do something else, or they build their skills at Apple and decide to go elsewhere. Or they get more money somewhere. Like if ever Apple is going to overpay people to come back to the company to fix things, It would be we're having problems in design. Let's hire some good designers. And, you know, I mean, we have we have Steve LeMay, whose name we can't remember, and Sebastian DeWitt, whose name I can't pronounce. So it's two people. It's two people that we know about. But like things seem to be moving in the right direction. Obviously, we won't see the result of any of this stuff for a long time. So jury's still out. But this is encouraging news. And, yes, Marco did hoist up the Halide developer hiring up to the top, but there's a little bit more to the story. So then in a different post, Mark Gurman writes, even with hardware chief John Ternus now overseeing the operation, there's no single design decision maker at Apple. During the Steve Jobs era, the company co-founder was firmly in charge, but after his passing, the ultimate design arbiter was Johnny Ive. But it's more recently been a committee. Besides Ternus, software head Greg Federighi and marketing chief Greg Josuiak, and even services boss, Eddie Q, to some extent, all have sway. Ternus is the top voice when it comes to hardware design. Federici is the chief influence on look and feel of software, and Josviak is an important voice across the board. Of course, the heads of Apple's design teams, Molly Anderson and now Steve LeMay, have their recommendations as well. Under the new arrangement, they're now managed by Ternus, but continue to officially report to CEO Tim Cook. Yeah, so that's an interesting breakdown of, like, who that is not technically in charge of design has influence, And it makes sense that, like, you know, Ternus would, because he's the hardware guy, would be able to give his two cents on any hardware design stuff. And Fede Rigi, being charged with software, would have his opinions about software. And then Josue Act, because he's marketing, would say, you know, I want everything to be teddy bears because people love teddy bears, right? It doesn't seem like, you know. That doesn't sound like him. Well, you know what I mean. But anyway, it was like sort of cross-cutting concerns. So like the old way this was where the CEO was, you know, the entire company was run off the taste of the CEO. There's pluses and minuses to that system. Obviously, the minuses where you get stiffs leather over the presses was that Steve Jobs had a pretty good batting average. And there's one thing Steve Jobs did not lack, which was opinions. He knew what he wanted. Whether what he wanted was good or not, there was no dilly-dallying and no real committees. I mean, I'm sure there was committees below him, but he got presented with stuff and he said yes or no. And I think Johnny Abbott was like that as well. The stories when he was like, had one foot out the door for seemingly years, was like he'd just be hanging out of this place in San Francisco and, you know, come to a meeting and he would say, show me all the crap that you made. And he would say, yes, no, no, yes, no, change this, change that, you know. But this new system they're describing doesn't really reassure me, which is like a whole bunch of people who are not designers offering their opinions on designs offered to them by designers. And the designers get to give their opinions as well. But like if it's Molly Anderson and Steve LeMay arguing with Eddie Q about design, I'm like, Eddie, just like, I mean, in some respects, like the designers should have the most sway here. You can give your opinions and whatever. But like in the end, but unfortunately, org chart wise, everybody on that list is above Steve LeMay and Molly Anderson, who are the only actual designers. And honestly, I don't want Greg Joswiak, Eddie Q, and Craig Federighi and Ternus overriding Steve LeMay and Molly Anderson. On the other hand, when they delegated to Alan Dine and let him do what he wanted, he made bad decisions. So I'm not sure what to do about this org chart, but if German's reporting here is true, it is a very interesting sort of stew of people that makes it very difficult to know who to blame for things that Apple does that you don't like design-wise. Much easier when it was just Ive. I think it's, you know, standard leadership is, like, the CEO and the top execs are not going to be specialists in all areas. And so you have to be able to identify and hire good people at every level below you. Like, set things up in place that, like, you hire good execs, and they hire good VPs, and they hire good managers. You know, you hire good people because you're outsourcing that decision-making to them, and you can't do everything. Yeah, but all those people shouldn't be in the meeting. Well, right. But then also, you know, part of the process is, like, every part of that stack as you go up the chain, you need to be able to trust the people who are experts. That's why you hire them. So, like, you know, like the CEO needs to be able to trust some of the decision-making to the level below them, and then they need to trust some of the decision-making to the level below them until you get to the actual experts that know what they're talking about. But then the people above, they're responsible for more of an editorial role. So, like, Steve Jobs didn't know how to, you know, grind the corners down on a computer case to make it round. But he knew, like, he, you know, somebody at some point brought him the idea of a round a computer case. And he's like, yes, that's great. We're going to do that. We're going to do it with, you know, my ideals and qualities and whatever. That's the role of leadership. Like, you don't have to need to know. You don't need to know how to do everything. You need to have people who can do it and you need to trust them and serve as, like, an editor and a guiding voice. And so as long as the people at the top can do that, and as long as the people below them can work with that dynamic, that can produce great outcomes. It's just a question of, you know, I think we've seen times where that dynamic wasn't working very well. One of the things that we've heard a lot about Tim Cook's leadership style is this repeated thing we've heard over and over again, which is don't bring me problems. That's apparently a big thing for Tim Cook. Don't bring me problems. And he seems, from what we've heard, to lead by kind of like sheer force of will, just kind of like an ice cold, this is what we are doing, what are you still doing here, kind of like that kind of like very like cold, stern, like strict and unwavering way. Well, does the combination of that and don't bring me problems, does that sound like a kind of environment that creates good collaboration and decision making when things aren't obvious what to do or in areas that he doesn't care that much about. I don't think that is great for that. And so one of the things that we have also heard is that by pushing problems down in the stack, you create more dysfunction in other ways. So by bringing problems up into the leadership roles, by letting leaders better engage with that kind of stuff and by maybe having a new CEO coming that might have a different style of dealing with that kind of thing. Could be better. It could result in better work coming out. It could result in less weird infighting and divisional problems. Maybe the whole thing with Apple intelligence and John Gianandrea's group versus Craig Federighi's group, maybe that wouldn't have happened that way if the CEO was less about don't bring me problems and more about, hey, let's look at this thing holistically and figure it out. So I'm actually very excited about a lot of these changes coming up probably, including stuff like that that we will never even hear about at least not for years of just very basic differences in like how Ternus operates almost anything versus how Tim Cook operates almost anything. They're going to be very different people very different leaders and that's again this is part of why I'm excited about Cook's hopefully soon departure and in particular it sounds like Ternus is a pretty good candidate to replace him but we'll see how this all goes. This could go a very different way because we don't yet know what kind of leader Ternus will be given that kind of escalation and responsibilities. But I think we can look at a lot of the dynamics that we've seen in Apple during the Cook tenure, some of which have been great. You know, he's made a lot of money, good for him. And some of which have been problematic or inefficient or have other non-ideal outcomes. And some of that's due to Cook's style, and that's going to change. And I'm really looking forward to that. It's difficult to tell from the outside whether Steve LeMay, I was not stupid. Ternus was deployed. You finally got his name. We finally got it. We weren't even talking about it. I was just reading it from the screen. That's why. Whether Ternus was deployed by Cook to solve a problem or whether Ternus was lobbying, was lobbying up the management chain and saying, we've got a design problem. Let me go fix it. Right. Because, you know, yes, part of being leadership is knowing how to hire and delegate or whatever, but also part of leadership is setting direction and inspiring people. And a lot of the times, especially even in Steve Jobs case, like you said, it'll be like, People are presenting things to higher-ups all the time. Steve Jobs had a good ability to, when presented with the 75 things that people in his organization were doing, he would see one of them and Darth Vader style say, that's it, the rebels are there. Like he would say, this is it. This is what we're doing. He sees the gumdrop computer in Johnny Ives' lab along with 8,000 other things that Johnny Ives showed him and said, this is it. We're doing this. We're making this computer. This is the whole company's focus on this. and John and I were like, okay, like I showed you 25 things, but you figured this was like, that's it. We're doing this now. And because, you know, being able to detect whether it's like, this is the thing that I think people will like, it will make you a good product, it's technically feasible, like the whole big formula of like identifying that versus Cook, where I think Cook was like, glasses, it's going to be the future. But like it didn't have the right combination of can we do this now? Are you sure this will be a good product? Are you right it will be a good product? How much will it like all the things have to come together? The iMac was possible. The iPhone was barely possible, but possible. The glasses that seem to be one of Tim Cook's babies, not currently possible in the way that Tim Cook wants to make them. And he's going to be gone before probably anything comes to fruition in that. Right. So with Ternus, it's like, OK, well, Cook just sick Ternus on this to say we've got a design problem. Go solve it. And Ternus is competent in doing it. That's great. I would love it even more if Ternus was telling everybody who listened to him. We've got a design problem. We need to fix it. And then the people said to him, oh, okay, well, if you think we have a design problem, you go fix it. And he's like, fine, I will. And that's how you move up the chain. Of course, every executive doesn't want their underlings to bring them problems. It's not a unique thing to Tim Cook. That's just part of being management. The whole point is the people below you should be solving their problems. Anyway. No, but I see. Okay. I will disagree with that. Your job as the CEO is to solve the difficult problems. That is like the most difficult problems come to you. And that's your job. Yeah, but he doesn't want to be a referee for a bunch of infighting below them. Like there's a balance. And when they say, don't bring me problems, it's kind of like, look, it's your job to work this out. Like, don't don't involve don't constantly run to daddy when you have any kind of conflict, because it's exhausting for me. And it's not you people doing your job. Obviously, there's a balance. Obviously, things do bubble up and there's decision making atop. But like that attitude is trying to discourage the anti-pattern, which is anytime there's any kind of issue and people can't agree, they immediately run up the org chart and say, well, fine, our boss will decide for both of us. And it's like that's that's not a scalable system either. I don't know what you've heard about the don't bring me problems thing, but I feel like that is that is mostly a healthy dynamic within companies. And it's not saying never bring me problems. It's not saying I'll never make any opinion. It is just saying part of being at this level of management is taking responsibility and working things out amongst yourself, which is, you know, why, like, the forestall thing and everything. It's like we can't have this thing where every time anything comes up, you two fight tooth and nail and immediately run to me and say, decide which one of us do you love more, like, with every decision. That's not tenable, and maybe that's the era where that rumor came from. But, yeah, I hope Ternus is able to recognize problems before Tim Cook, which is not hard, But still, and I hope he has been lobbying to like, you know, I know I'm not in that part of the org, but like, like, like, you know, with Tim Cook reading the room or having his finger on the pulse, like, I really hope for the one thing I hope for Turner is that he is better able to accurately assess at any given time. How is Apple doing? That's really hard to do when you're high up in the org because it's kind of part of your job is to be like rah-rah Apple or whatever. But like, you have to have a clear eyed view. Steve Jobs is really good at that probably because he hated everything. Right. So whatever Apple is doing is like, oh, this is crap. We can do better, right? Tim Cook at various times seems to be distracted by his various money-making schemes and other things and not really keeping his eye on the ball. Like, how are we actually doing? What about our, like, why do people buy our products, and are we selling products to fulfill that promise? And, you know, like, take his eye off the ball with the Mac for such a long time, take his eye off the ball on design, like, just not understanding how developers' attitude towards Apple. I think to his grave, he will go to his grave, not understanding how much developers dislike Apple because of the decisions he's made for most of his tenure. He's got blind spots. So I'm hoping Ternus wasn't simply a fixer called in Pulp Fiction. Is it Pulp Fiction? Yeah, Pulp Fiction style, or is it Reservoir Dogs? Yep. Or is the fixer? I thought it was Pulp Fiction. What is it called? What's that guy's name? Mr. White? The short guy? Oh, Harvey Keitel? Yeah, Harvey Keitel, the wolf. That's Pulp Fiction. Oh, wolf. I'm sorry. I'm sorry. You're right. I hope he's not just like we called in the wolf to fix things. I hope instead he has been telling everyone who would listen to him that we need to do something about design. And by the way, we should get rid of Alan Dye. Yeah, I mean, I feel like this is slightly ham-fisted insofar as everyone knows what's going on here. So why don't we just embrace it? And I know it's not quite that simple because, you know, once they start really embracing it and really properly announcing who's next, that gets the SEC involved. There's a lot more to it than even I'm aware of. But it is kind of funny, this, like, cloak and dagger thing that they're trying to pull off, even though everyone seems to know what direction the wind is blowing. This still can go any way, you know. Like, things change all the time. And like I said, just because he's being tapped to do this, it may just be, oh, he's the most competent lieutenant being asked to do this. But, of course, we're going to ask this other person to be CEO, which seems unlikely, but you never know. It's like the Sony TCL deal. Like, it'll probably happen, but nothing's signed yet. That's true. And then with regard to Cook's departure, CEO, Apple told shareholders this month, writes Mark Gurman, that its current chairman, Art Levinson, would remain in his role past the company's February shareholder meeting, despite the fact that he's now 75, the usual retirement age for directors. That implies a chairman transition won't happen until at least 2027. Sorry, Marco. Hey, look, I'm just happy this is, you know, that there's obvious signs that this is in motion. Now, whether or not Art Levinson is there for another year or another six months, who knows? I don't think this really says either way. All this says is that he is not being forced to retire by their original rules, because, again, Apple makes the rules, they can change them whenever they want. So they're kind of holding Art Levinson over a little bit longer. I still think the most likely outcome here is sometime soon, Tim Cook replaces Art Levinson as chairman of the board, and John Turner's become CEO. That seems like still the obvious path. But as I said a few weeks ago, Tim Cook and Apple are very patient. They are happy to wait what seems like an eternity to us to move on something or to resolve an issue, to issue an update to a product, you know, whatever it is. They do things on their timeline, and they don't like having anyone else influence their timeline. So whatever Cook's plan is for the succession, he's going to take his time, and he's going to do it the way he wants to do it. And given the gravity of that situation, I think that's the right move. You know I will not say good things about Tim Cook when they're not warranted. But in this particular way, he is a very efficient leader or a very effective leader in terms of like when he has a plan on his timeline, he will stick to it. He has almost certainly planned this very carefully in a very considered way. And I'm sure it seems like from everything so far with the various executive retirements and different transitions and everything, it seems like he is executing on a plan that has been carefully considered and it's going to be on his timeline. And it didn't seem like he was going to leave in January, the way the original rumors said about because of the chairman of the board age requirement they had. That seemed a little fast. I still think it could happen in this calendar year or it could happen next year. The board is totally under Apple's control. They can do whatever they want. They don't need to wait until a one-year anniversary of this extension for Art Levinson or anything like that. Like, they can do whatever they want. So whatever their plan is, they're going to keep doing it. I still think it's, you know, this month was aggressive, but it could still be this year or next year. It doesn't matter. I think one of Tim Cook's strengths is the sort of bird's-eye view of the sort of macroeconomic picture and where Apple fits into it, which is why he's been so successful at making money and all that other stuff. And it's also part of his blind spots with China because it just makes so much sense in an economic sense. And it's like, well, the downsides, he was not seeing where this could lead. It's just that the money was too good, as they say. But, like, this transition is all about, like, I'm sure he has in his own mind, you know, ideally this is when the transition would happen in terms of stuff I probably would understand in terms of, like, you know, commodity deals and the way the market is moving and how Apple is positioned and like minimizing the impact to the company's economics. This is always going to be some hit during the transition because there's uncertainty and everything like that. He's got this idea in his head of here's the best time to make this move. And it's something that he's probably one of the only people, maybe with Jeff Williams in the world, who even understands, who holds the entirety of Apple's economic engine in their mind and the whole world market and all the political stuff or whatever and deciding when can we do this switcheroo. But in the end, it's like, you know, there's no perfect time. Like he's going to do the best he can. He's patient, you know, whatever. But like there's always going to be uncertainty during transition. He's certainly not going to rush it. But like kind of like the blind spot with China, it's like, yeah, everything's great. You're right about a lot of the things there. But if you don't pay attention to the downsides, one day you're going to wake up and it's going to be a big deal. And it's like if he takes too long and it's like, I just want it to be just right or whatever. It's like at a certain point, various crises the company might be having should take precedence over his desire to make a smooth transition. Because obviously, if there are any other crises that come up, as if, you know, Alan Dye leaving in the design thing and Apple Intel, as if we don't have enough already. But like any time something like that happens, it's like, OK, I guess, well, the transition will have to wait a little bit because we have to sort this out. It's like there'll never be a perfect time. Right. And at a certain point, which I think we're long past, you staying is the biggest problem. So you do have to eventually do it. And I hope he's a sensible person. I hope he's probably some kind of deadline. He's like whatever his deadline in his head is like, I'll try to find the perfect time. But honestly, it's definitely going to happen within the next, you know, X number of years or months. And fingers crossed that he finds a time where he feels like it's the right time and he makes it happen because he needs to go. Yeah, I mean, if you would have asked me maybe six months ago before any of the rumors started, when do I think Tim Cook might retire if it's going to be sometime soon? I would say he would wait until the Trump administration is over. And honestly, that would make a lot of sense. But I think he sees now that that's not the right move. I'm a little curious why he sees that's not the right move. I mean, that and that maybe he just doesn't want to do it that long because that's, you know, unfortunately, we still have probably three more years of that. So. So, you know, that's that's a while in terms of, you know, this timeline. So I don't think that's the plan now because I don't think we'd be hearing all this all this shuffling around and all these moves if it was that far off. I don't think it's three years off. But certainly if the Apple defenders and the Tim Cook defenders with all this Trump stuff, they point to some kind of role that Tim Cook is like heroically taking the heat so that it shields the rest of the company or maybe he takes the heat and then he leaves at the right time or whatever. If that is true, which I honestly don't think that's the whole story, if it's even any of the story. But if that is true, that timeline is too long for this. That doesn't fit what seems to be happening here. And I do wonder, like, what's going to happen with that side of the job? That's why I said I don't want him to be chairman of the board because how can you, the day you sit in the CEO seat and say, okay, Tim, I'm going to do everything the opposite of you did. I know you're still chairman of the board. What do you think of that? Well, but, you know, in terms of politics, I think Tim staying as chairman of the board for a few years, that could, maybe the plan is he stays on as chairman of the board, he keeps politicking with Trump, and that kind of leaves, turns out of it. Yeah, Trump will still think he's CEO, so it's fine. That's probably true. He'll still be Tim Apple to Trump for forever. So it's like the fact that he's no longer CEO is immaterial because Trump won't know or care about that. That's probably true. And if Cook wanted to isolate Apple from more of that heat than it needs, maybe that's what he's thinking. Because again, Cook is a very considered person. He good at playing the politics game when he needs to And so it would not surprise me if that was his plan But I think the CEO transition seems to be happening way sooner than that. So if his plan is to be like the hate sink, so to speak, it's going to be, it's going to have to at least include some of his time as chairman of the board. Lately, I can't wait for it to happen. Because, well... Alright, thanks to our sponsors this episode. Delete me and Squarespace. And thanks to our members who support us directly. You can join us at ATP.fm slash join. One of the many purposes of membership is ATP Overtime, our weekly bonus topic, which tends to be approximately 21 minutes of bonus topics every week. This week in Overtime, it's going to be, we'll be talking about the new ownership of TikTok in the U.S. and all of that whole thing. So if you want to hear that, join us at ATP.fm slash join. Thanks for listening, everybody, and we'll talk to you next week. Now the show is over They didn't even mean to begin Cause it was accidental Oh, it was accidental John didn't do any research Marco and Casey wouldn't let him Cause it was accidental Oh, it was accidental And you can find the show notes at ATP.fm And if you're into mastodons, you can follow them at P-A-S-E-Y-L-I-S-S That's Casey List M-A-R-C-O-A-R-M-E-N-G Marco Arment S-I-R-A-C-U-S-H-T-R-A-Q-U-S-A It's accidental They didn't mean to Accidental Check podcast no one All right. I don't want to belabor this, but it would be remiss of us not to at least acknowledge that what's going on in Minnesota and what's going on with ICE, and not the ICE I was dealing with, but the national ICE, it's terrible. It's disgusting. We, all three of us, absolutely stand in unison with the people of Minnesota, that we are disgusted by what our government has become. You know, Marco was tweeting a day or two ago about how, you know, it seems bananas that anyone would travel to our country these days, and I echo what Marco says. It's awful. It's really, really bad. And Tim Cook is not making things a whole lot better. Now, granted, he's not actually affecting things, as far as we can tell, But it's certainly really gross to go from the second or third shooting in as many weeks to, hey, let's watch this movie about Melania. It's just gross and disgusting. And again, I don't want to belabor it too much. It seems like, you know, there's a lot of Apple employees don't like it either. We'll put a link about that in the show notes. Tim wrote an extremely kneeling mouth memo about it. We'll put a link to that in the show notes. It's just, it's gross, especially when you compare to his letter after George Floyd was killed, which was much, much, much better. It's all gross top to bottom. I'm going to stop talking and let Marco and John take a turn. But, again, we're going to try to make this quick, not because it's not important, but because we're trying to not be total Debbie Downers over here. We will have an actual after show on a cheerier topic, but it's hard not to talk about this, like U.S. citizens being murdered by the government and then the government refusing to even have an investigation, have an independent investigation of that, and smearing the victims thing. It's just terrible things are happening here, and it's not a good situation, and it's affecting everybody even if you're not where this is happening because it could happen to you, and apparently there's no recourse if it does, and so we're all feeling kind of crappy about it. And the Timbuk-Lawagna thing, like, I'm sure he's loving this because it's like, you know, how long has this movie premiere thing been scheduled And then, like, the government's got to murder people just, like, exactly around. And not that it matters. It's bad either way. But, like, it looks even worse. You know, Tim Cook's strategy about this has not changed. And it continues to be a terrible strategy and continues to make him and Apple look terrible. And, you know, this story from The Intercept about Apple employees being cranky about it. Like, you know, it's not just people outside Apple that can't handle this. People inside Apple feel that way, too, which is another reason that the sooner Tim Cook can leave, the better chance the next CEO has to, you know, keep these employees from leaving the company, right? You don't want your employees being this angry at your CEO because they'll leave and you need them because they do the work. Yeah. And again, yeah, I have very strong feelings about this. I will point you to two posts by John Gruber over the last day or so that were excellent about Tim Cook's ridiculous statement about being heartbroken, prayers and deepest sympathies. this is a time for de-escalation like what a bunch of bulls**t from tim cook like such bulls**t yeah make sure you don't say anything to make trump angry non-committal not say it's both sides of this oh both sides have good people that's that's what this is it's so despicable and i know why he's doing it i know look part of it is he's trying to protect apple in the trump era and this, and you kind of have to just suck up a lot to do that. I get that. Also, Apple gets tariff exceptions. Apple has the government do things like block, you know, competition from like Huawei and stuff from being used. Like Apple has the Justice Department now kind of just disappearing things from being looked at. Like Apple benefits a lot from government stuff that is giving them advantages, unfair advantages or illegal advantages sometimes, but it's giving them advantages. So it isn't just about defending Apple's business from having retribution. Tim Cook is getting a lot of favors from the government that help Apple continue doing things that are kind of despicable. It's not all roses. He's not this wonderful leader who's throwing himself under the bus out of the grace of his heart. No, he's a bullshitter. He is a CEO who extracts advantages from the government by trading favors and gold trinkets and BS like that. We know exactly who Tim Cook is. He's going to keep showing us that over and over again. I can't wait for the Tim Cook era to end. And that's all I got for tonight. John, anything to add about that specifically? No, but I have a much cheerier actual tech-related after show that we should move on to. Yes, freaking please. So we don't have to think about ICE anymore. At least until we get off the air and see whatever horror has happened. Anyway, let's go back to me and my friend, my new friend, Quad Code. I had a past-keeled ATP that a bunch of stuff talked about last time. I've still been hanging out with Quad Code. Claude Code is a little bit of a needy friend and he did get money out of me. Oh! You are... I think you and I wrestle for who is more frugal and I think you usually win, but it's a toss-up. I am stunned. But in terms of value for the money, there's a few things in the tech business that many of us agree are great values. I think one of those is paying for the no ads thing on YouTube, whatever they call it this year. Like, that is obviously a great buy. And I think number two thing is whatever your AI of choice is, buy whatever their, like, 20-ish bucks a month plan is that gets you access to the good models and, like, removes restrictions and stuff. Oh, my God. Because, like, I haven't, honestly, I haven't actually used Cloud Code yet. I've used Cloud, but I haven't used Cloud Code. But I'm trying Gemini for certain things. I'm still a very heavy ChatGPT user. Like earlier today, I got a statement for a retirement account that I have. And I wanted to understand better some of the acronyms and numbers that were on it. And so I just, I scanned it and I took a screenshot without any of like my account numbers, just like the middle of the paper that shows like, you know, this table of values and headings and stuff. And I pasted it into ChatGPT and I said, can you explain this to me and tell me what this means for this year and what this number means over here? And it gave me this entire breakdown of what every single term means and how these numbers work out over time and what the different aspects, pros and cons of different things are. And I'm like, this is amazing. As far as you know. But, again, I know this is not about my experience with AI this week, but my experiences with AI continue to be extremely positive. And I cannot tell you the difference it is making in my life. The more the more I like the the walls are coming down in terms of figuring information out. And even though it isn't always 100 percent right, it's right a lot. And it's amazing a lot. But you don't know. It's like the ads. We have the money we spend on ads is wasted, but we don't know which half. Well, some percentage of your answers are wrong, but you don't know which ones. But it's right so often. I mean, look, no matter who I ask for information, like I actually I asked a financial advisor recently, a couple of questions. And it turned out later that part of what he said was wrong. Is that error rate different from... I wouldn't say chat TV has a worse error rate than a financial advisor, yes. You'd be surprised. Cutting aside, the real trick with all these things is obviously you need a way to tell whether the answer is right. I think I talked about this, maybe it was on Rectips, kind of like in the cryptography thing where it's really hard to derive the answer but it should be really fast to check it. That's the ideal. But for everything in life, you need that. How can I tell whether this answer is right? You don't, you know, you have to have some way to tell. And that's why I always say, whether you're asking a person or asking a financial advisor or asking anybody, any kind of professional, like, or hiring somebody or whatever. If you hire someone to fix your fridge and they fix your fridge, you know, real fast, whether you fix it. Does it get cold or does it not get cold? Did they fix it or did they not fix it? Right. Now, there's varying degrees, like maybe it breaks next year because they did something badly or whatever. But at the very least, if you ask someone a question or hire someone to do a thing, you can tell whether they did it if it's the type of thing where it's like, this is broken and I need you to fix it. And same thing with ChatGPT. This is broken and I need you to fix it. Did you fix it? Yes, no? Good. With other stuff that's more abstract, it gets a little bit loosey-goosey because you're like, well, if I just believe everything you say and continue along that path for a year, maybe it will lead me down a path where we went astray two months ago and I didn't realize it because it's totally an area that I don't understand. This is a sidetrack. This is a sidetrack that I'm trying to talk about. Cloud Code. I'm glad that you're continuing to have your friend that you talk with that gives you plausible answers, Marco. And a lot of the time you can just immediately check whether it's right and that it's helping you, and I get that. And that's exactly the same thing with Cloud Code. As I'm using it to do stuff, the reason I paid more, I already had the $20 a month thing, but we were getting it for free as part of a past Cloud Code sponsorship thing. But I stopped. That trial was going to end anyway, and so I paid for the $100 thing. Oh! Oh, my. The $20 thing, I was running out of tokens too much for the $20 thing, and I tell myself I'm going to do the $100 thing but just for one month. That's what I tell myself. Sure you are. Next month. But anyway, I did it. I wondered if this is a $200 plan too, and the $100 gives you five times what you get for the $20, and the $200 gives you ten times, I think. But anyway, it turns out five times is plenty for me. Now, what am I doing with this, right? Because, like, I did the pass keys thing. I did my status board thing that I talked about. I had an idea for another project. But before I had that idea, like I guess it was kind of simultaneous. My status board thing that I talked about, I'm not going to go into it again, CPath episode where I talked about this. I made a web version of Panic Status Board. That's just a website for me. And I don't want it to be accessible to the public. So I just had HTTP Basic Off on it. Do you guys remember HTTP Basic Off? Yeah. Yeah, yeah. Username, at password, colon, rest of it. Yeah, it brings up a little dialogue. Browsers still support it. Yeah. It brings up a little dialogue. You put in a username and password, and if it's not HTTPS, it sends them in clear text. It's great. Yeah. You know what also still uses that? Podcast feed sometimes. Yeah. Yeah. I mean, anyway, it's very primitive. But I said, like, and that's fine, but, like, it just bothered me. It's like, it was the 1990s. I'm using HTTP Basic Auth in 2025. I just felt wrong. And I had just added passkeys to ATP.fm. Honestly, basic auth over HTTPS is not that bad. I know, I know, I know. I'm just saying. Anyway, why don't I add a passkey-based account system to my status board? Cloud Code has already shown me that it can do it in PHP, no less. It should be easy to do it in Node because that's like an actual implementation with like real libraries and stuff. It should be a snap. like a million examples of this, like, this is the thing Cloud Code should knock out of the park. And then I had to decide, like, okay, well, how should an account system work on, like, a personal tool website that's only supposed to be accessible to me? And I had to go through, like, it's, like, really weird. So, first of all, you know, status board, its persistence layer is essentially, it's R2, essentially S3. Everything is in JSON files in S3. There's no database. It's just, you know, because the data volumes are minuscule, nothing ever, like, that's it. That's its data storage. Same thing with the account system. Like, okay, well, that's fine. There's going to be like one or two accounts on this thing ever. I'll just put everything in JSON files. And pass keys are great because you don't have to store anything. The server stores nothing. There's no hashed password. You store a public key. That's the great thing about pass keys. One of the many great things about pass keys is if there's like a data breach or whatever, all they get is public keys, information that is literally public. There's nothing in there. There's no password hashed. There's no crack. There's certainly no plain text anything that's the beauty of passkey so i'm like i have no problem putting you know public key passkey stuff in a json file and r2 not that anyone cares um well then i was like okay but how like how do i make the first account you know what i mean like if it's a website that doesn't let anybody in but there's no accounts on it how do i make the first accounts like well i guess i need like a bootstrap mode where like when there's zero accounts, you get an account creation screen, but as soon as there's non-zero number of accounts, you get a login screen? Like PHP BB. Yeah, exactly. It's old technology. And then I was like, well, what if I wanted, like, what if other people in my family want to make an account on this thing? They can't make one, because once I had the first account, they would just get a login screen. There's no, like, sign-up process, because why would there be? It's not a public website. Because if there was a sign-up thing, then anybody could go there and sign up for an account. But I don't want that. But I just want the people who I want to allow to sign up. And I could create the account for them, but I can't create a passkey for them. Because if I create an account for them, the passkey will go into my iCloud keychain and not in theirs. Right? That's the thing you could do with passwords. You could make them an account with a temporary password and tell them what the password is. And then you'd have to change it when they logged in in the old ways. But with passkeys, that doesn't work. And I was like, there's going to be no passwords. It's just going to be passkeys only. It's just all modern. It's my website. so then I came up with like token based system where you can create a token with a fixed number of accounts that can be created from the token it's just basically like a secure you know a secret url security through obscurity go to this big long url once you go there it burns the token you can create an account right and then I would just give that url to whoever wanted an account then they could create an account or whatever if I wanted to give you two accounts I would make a token with two things give the url to both of you you'd both go in create your account put your passkeys in your thing. Anyway, I spent a while doing this passkey based account system. And then I did eventually have the ability to add a password, right? Because I'm like, well, maybe if you want a password, you can add one if you really want to. And yes, there would be a hash password in there, whatever. I spent a while doing this. It's like a little miniature account system that supports passkeys that's in front of my status port site to let you in to do the status port stuff. And round the ball, I was doing this. I'm like, I have another idea for a different project. But I'm in the middle of doing this PASCII thing. I'm like, you know what? For this other project, it's the same type of deal. It's like a little personal web app just for me, but I don't want anyone else to have access to it. I'm probably going to want to use this whole PASCII account system that I just made for the status board thing. I'm probably going to want to use that in this new thing too. So I should have looked up the SKCD comic with the passing assaults. Please, Casey, find this for me. So I'm like, you know what I need? I need to factor out this whole password account system into a framework that can be used by the status port site and also the new site oh my god because if you're a programmer once you have two things that use the same thing it's like well now it's got to be factored out oh yeah it's just got to be by the way I wish that's a thing Claude Code could understand I've heard from a lot of people who are like here is the the tutelage that I put Claude Code under and it's like giant, giant essays that I write it about how to be a good programmer. But I feel like in the training day, there should be something that says, you know what? I don't want to open up a source code file and see that the same seven lines of code are repeated a hundred times. Have you heard of functions, Cloud Code? Factor that stuff out. You can't do that. Like machines, especially when you're a human, you're like, I would never do that. It's too much typing. I would never update it in all the places. Well, guess what? Apparently, Cloud Code has no problem updating the same seven lines of code in a thousand places because it's a machine. and it's like I'm never going to miss it. I'm not going to make a copy and paste type. I'm never going to make a mistake. It will. But anyway, like I don't know what's making it do this, but the code it writes just makes me pull my hair out, and I just like have to spend some quality time refactoring and saying, you're doing this same thing in 100 places. Make it a function. Put it into a library. Look, see how much. Eventually, I just went and did a bunch of stuff with hand because it wasn't getting and there's no sense arguing with the machine when I can just factor it out myself. But anyway, got to factor out this functionality into a library. I'm kind of playing it like a video game where it's like, yeah, I could just do this myself. But the video game is don't do it yourself. The video game is type in little essays to make the machine do it for you. And so I'm trying to get it to understand. It's like, no, here's the thing. Here's here's the guidelines. I bull pointed it out for it. I made this big essay in a cloud that MD. And it's like when I change things in the framework, those changes should be reflected immediately in everything that uses the framework. Like there's no like, you know, I don't know if you remember MFC where like generate source code. for you and then you would modify the source code well guess what if a new version of nrc comes out and it generates different source code that doesn't help you if you just generated it for your giant project three years ago you don't get to benefit from those changes that's what a library is you know and with node packages and stuff it was just like oh just generate this and blah blah now you're fine like no i'm not fine because if i go back into the framework and i change this character here that's not reflected in the other products that's not how a framework works i spent so long arguing for the student machine i'm like nothing can be like It has to all stay in the framework. You can't copy the files into the other. Anyway, I made a framework. I fought Claude Code. I made it make a framework for the account system. And the second little app that I wanted to make, it was actually something that someone suggested to me. After I had already made it, they said, hey, you know what you should make? You should make this thing. And I felt like replying, but I didn't want to spoil this episode. I already did. Can you guess what it is? I have no idea. Single website just for me, not public. Something to do with, like, membership tracking, but that's the status board. That's the status board thing. You're right. It's a tier list website. Oh! Of course, of course. I hate tier maker dot com. Every time we do a tier list, I have to go, like, with the Chrome dev tools and, like, hide and delete nodes. I turn on my ad blocker, and then I go in there and manually edit the DOM to get all of the ads and crap out, and you can't delete all the notes because javascript will just bring them back so you have to hide some of them and then i have to arrange it so so like as we do the tier list thing like the like the stupid buttons and copyright notice don't scroll up into view so i have to hide like it takes so long to arrange this window just so and i always do like a few days ahead of time and then i come back and see chrome has taken that tab out of memory and it reloads the page and everything is gone and i'm so sick of that and that the tier oh and by the way when we make the video of it i have to arrange my window and i use xscope to make the window the right proportions and then I use screen capture within the, you know, carefully drag out the outline where I'm going to be doing QuickTime screen capturing and then make sure that after I drag all the little tiles up that the board looks clean and no weird stuff is coming into it, right? But make sure you can see all the, you know, all the, one, if I fill the tiers, everything is still visible. I'm like, I need to make a tier list out myself. And I made it look pretty much exactly like TierMaker because TierMaker is sort of the canonical thing and for historical continuity purposes of ATP. But anyway, now I have a tierless website that has configurable aspect ratios, auto-zooming to make sure that all the tiles are always visible at all times and that you can always see at least some of the things you're dragging into the thing. But also when you were done, the thing will be the maximum size it can be. I have things to hide the scroll bars so they don't show up in the little scrolly region. I even made it work on the phone because, of course, I would. I made the status board thing work on the phone as well. and the iPad, which is a pain in the butt for Claude to do because, yeah, that's another thing with, like, it saved me a lot of time on this too. Because, like, to make a status board, it takes you two seconds. Like, the person who sent in the thing, I made the status board, you can make it in two seconds. But the thing you make in two seconds is garbage. Like, there are so many things that can go wrong. It's like, I was thinking when you program something yourself, like say you're doing drag and drop of images in, like, a tier list type thing. If you haven't done that recently, maybe you're not familiar with the modern APIs for doing drag and drop and, like, HTML5 and stuff like that. So if you were a human programmer doing it, you're like, okay, I'm going to start simple. I'm going to make a thing. I'm going to make just one thing, and I'm going to have the ability to use a dragon. Okay, now I have one thing. I have the ability to drag it. I'm going to drop it somewhere. Like, you'd build up in a piece at a time. That's how you do programming, right? Like, when you're a human, you're like, one thing at a time, piece by piece, each step of the way. I'm like, I've got something draggable. I've got a place where I can drag it. And each time you work out the things, okay, now I need multiple ones. Now they need to move out of the way. Like, you would build it up a piece at a time. Cloud Code, because the machining goes real fast, is like, I've done everything. It's all done. And then you try to use it, and, like, nothing works. Oh, you can't drag this one anywhere. You can't drag this from the top row to the bottom row. When you drag this, the image doubles. Now it triples. Now it leaves a ghost image over here. It doesn't work at all on the phone. It's just no human would ever say, here you go, I'm done, and, like, nothing works. And so you go back and forth and be like, this didn't work. And it's like you're filing tickets into your, you've got to make this work. This doesn't do anything like this. why don't we try a different strategy? Then you peek in the source code and you're like, what are you doing? They're like, reconsider this. And then you're thinking, would it be faster if I did it myself? That's the game you're playing. So I spent a long time with CodeCode working on drag and drop and scrolling and hover effects and double-touch thing and scrolling in touch. And apparently the phone doesn't support HTML5 drag and drop stuff. You have to use an entirely different API. But anyway, it's done. I have a tier list app. I needed a framework because once you've got two apps, you need to factor it out into a framework. And I did. And now it's real easy for me to make future apps. I don't currently have any future app ideas, but I'll probably polish up my tier list. And the next time we do a tier list as an ATP member special, which won't be next month because we just did one, but at some point in the future, we will do another tier list. We will get to try out my fancy new tier list website. And I will say that the one thing I stopped myself from doing was, and I may be able to revisit this, but I probably shouldn't, was, I thought, you know what? what if I did this as a multi-user thing where all three of you could be, you know, all three of us could be on at the same time and we could all be dragging things at the same time and stuff like that. Why don't you have us vote on what the rank should be, John? Well, that's what we do verbally. It's like, it's an audio podcast. Like we talk through the things, but I was like, I don't want to do that mostly because I don't want to upgrade to a different thing, a storage thing and trying to do it with JSON files. You could do it, but there'd be all sorts of things where it's like, Oh, that person started their drag first. So that person dropped it first so that invalidates yours, and then you drop it, and it goes back to its original position or goes to a different position. It would be confusing. So I didn't do that. Yeah, that's a lot of scope creep on that. Yeah, I held off. Because honestly, I feel like the Cloud Code would do a good job of that because there's so many examples of, like, the conflict-free data, whatever thing that are out there. CRDTs? Yeah, but with JSON backing, it's a little bit tricky. I mean, I did basic mutual exclusion with the JSON thing because you can do conditional puts that basically say, don't accept this put unless the content is exactly how it was when I did the get, which is the world's dumbest way to not corrupt your data. But I did at the very least I did that, but this is a single user system with a single account. But it looks really good. I've been using the 91 car logos as my test case for it, and it's shaping up pretty nice. So I'm having a good time with Cloud Code. but it is a, I wouldn't call it productive. Like, it has helped me a little bit, but honestly, I'm thinking mostly like playing Ark Raiders. Like, this is essentially a thing I am doing for entertainment. I mean, I guess it's for the show where I talk about it as well, but it is very much like playing a video game, and the video game is, can you make this other thing do a thing? That's the game, and it's kind of fun, especially if you're a programmer and you know how you would do it yourself, trying to make something else do it that's not a human. It is a pretty fun video game, and I recommend everyone give it a try. Well, this is like the power of AI coding. Like, it lowers the bar significantly for new programmers and for existing programmers looking stuff up. But also, by lowering the bar so far in all this, like, you know, kind of like grunt work for a lot of these things, what it does is it makes it worthwhile to create things that previously were just not worth the time to create. Or wouldn't have been fun. Like, I didn't do pass keys and ATP.fm because it wouldn't have been fun. I hate doing authentication and web stuff. I did it my whole career. It would not have been fun, but using Cloud Code made it fun. Yeah. That's extremely cool. So, sitting here now, do you plan to fall back to either the $20 a month or free plan after the end of your $100 month? Yeah, I think so. Like, I think I've learned how to use it more efficiently, and I don't have any other big projects. And honestly, like, it's been a big learning curve for me of, like, when to say uncle and dive in. Like, for example, I spent literally hours today. I believe it was three hours fixing this Claude Code CSS because the CSS it made, oh, my God. It's just like, like, it's like brute forces CSS. It's like, whatever. I'll just, I know how to make a CSS rule and I know how to make it apply to a thing. that's all i need i was like no no there is there is a design philosophy of not making a bazillion randomly named styles connected to a bazillion randomly named stuff especially when you have like a base framework i consolidated i don't know i should have counted how many selectors it was but like just pages and pages and pages of selectors and i spent three hours manually consolidating them down into one sane css file that like it's like 1 90th the length it's like i yeah you can write a new selector for everything you want to style and yes you can keep the same color up to date in 500 places because you're a computer but i can't stand it and so i spent three hours fixing it and now i'm like am i actually saving time with cloud code because i spent three hours editing consolidating three css files into one css file on the framework and carefully doing it and changing the markup so that the appearance stays exactly the same because it made it look the way I would say, make this look like this, make this look like that, and it would do it. And then you look under the covers and you're like, oh, my God, what are you doing? I know I said to make that look like that, and it does, but this is not – it's madness. And I have tried Gemini as well, and Claude Code is kicking Gemini's butt. I haven't tried Codex yet, but, yeah. Yeah, by all accounts, it seems like the one for Code Generation is Claude by far. Like, it seems like they're way ahead on that. Some good things on Claude. Oh, and by the way, one, remember I said that Claude erased my entire directory, or I suspect Claude erased my entire directory last time? I can't conclusively blame that, but I have something that I can conclusively blame on Claude, which is we were working together on the tier list thing, and at a certain point, despite me literally never having said anything that even remotely suggests that it should do this, it deployed my code to Claude. I'm like, I put a sternly worded instruction in all the cloud.md files said never deploy I will do that manually but it wasn't a big deal it was a private website I'm like are you kidding me because I saw it running an NPM run deploy like what in the hell it was too late it deployed it so you got to watch it you got to keep your eye on these things even if they're sandbox if they have network access don't deploy your code they don't care bananas that's extremely cool Yeah, I've been fiddling with Cloud Code some. I had it check over my entire Call Sheet code base, and it found some minor stuff that I was happy for it to find. I've had it code a couple of minor things, and what I've decided to do, this is probably, I'm surely not the first person that's done this, but instead of working on the folder or the directory that I typically do my work in, you know, when I open Xcode, I'm pointed to such and such a directory, I've created a second clone of the repo somewhere else so that it's off in its own little world, and it's doing its own thing over there, and then I have it make whatever modifications it wants to make to the files in that folder, and then I'll typically go and do a manual diff deliberately, you know, as a check, to make sure, okay, yeah, I see what it's doing here, and that makes sense. And that's worked out pretty well for me, but it's all been minor, minor, minor stuff. I haven't really had it do anything larger yet, and I'm sure with time, as I get used to Claude Code and get used to, you know, working with my new little buddy, I'll get better at doing larger and offering it larger and bigger efforts. But so far, I've been pretty impressed. And I think I've been impressed enough that I would probably pony up for, like Marco was saying, the $20 a month plan. I can't imagine ponying up for the $100 a month plan unless I'm doing something, you know, like a one-off, like you were describing, John. But you never know. I'll just put a screenshot of it if you want to see what it looks like. It should look very familiar because I was basically saying this has to look pixel over pixel like tier maker dot com because that is the canonical tier list website. And I made some slight changes, but it's essentially the same. I don't love the way it looks, but it does look like the tier list. I did. I did that, by the way, like I've had varying success with various things where you send them a screenshot and say, make a thing that looks like this. It utterly failed. Like I gave it a screenshot. I carefully composed these screenshots and what it made look absolutely nothing like this. I was like, well, the other work cut out for us. Let's dive in and fix this. But, yeah, it looks like the thing is it's all I can control the aspect ratio. I can turn on an outline that shows me exactly where I need to do the screenshot capture. It auto scales everything to make sure everything fits in the thing. It's glorious. I never have to right click and inspect on the stupid tier maker website ever again. That's very cool. That's very, very cool. And I'm glad you did it. And I hope it works. I don't mean that in a snarky way. I mean that in a, you know, hey, this is something that we're all three of us going to use. And I hope it works out the way you expect. Certainly sounds like you've tested it. So this is really, really cool. And I know, I don't think the listeners know, but Marco and I know the amount of pain. And I know John went through this a minute ago. But truly, he goes through an immense amount of pain to get these tier lists to work. And I love him for it because I do love the tier list member specials. I think they're becoming my favorites. But it is such a pain for John, and I'm so glad that hopefully, knock on wood, it will be a lot better now. I mean, it's fun, too. I enjoy making them and finding their things and coming up with the ideas. And now it'll be even more fun because I'll get to use my own app to do it and not that terrible app that's there. I mean, this says – I can also, like – I've got to add an image export. That's my last feature I need to add. But, like, so I don't have to do a screenshot of it. I can just export an image of the final tier list, which is what we want. And it looks beautiful because I don't have to worry about composing the screenshot. Everything will be in it. But, yeah, like this is going to make my life a heck of a lot easier. You can do individual play sessions with saved sessions, so I can name the sessions after the episode, and we'll always have the final state of the tier list saved in the thing. Oh, that's cool. Yeah, that's right. Actually, I had one more feature, which I want to make publicly accessible ones. So if I wanted to share a tier list, so these are all – it's all private data. You can't get this unless you're signed in. But I do probably want to make a way to have publicly accessible URLs so someone can look at the tier list. It would be cool to deploy this under the ATP domain some way, somehow, be that some subdomain or some path that's referred to this. I don't know. I'm not trying to make a product here. I know, but it relates to ATP, and it would be cool to list it there. Just the image of the final thing is fine. True. That's cool. I really dig it. I can also see it being fun, actually. I'm not trying to go creepy here, despite what it sounds, But I could see it fun leaving it possible to generate your own tier list. So, like, you know, if I'm a listener, I could take the same source material, generate my own, and then generate an image from that that I could share, you know, on social media or something like that. If I made it successful and people could make accounts on it, anyone could make their own tier list, then it would work fine. But I'm not doing that. Although it does, like, the TierMaker website is so junked up with crap, and I did this in, like, two days. Like, obviously, their website is more than just this. But, like, if I replace these JSON files with a real database, like, it's not hard to compete with a lot of these popular websites. They're just popular because they got a good domain and everybody knows about how to get there. But they're so junked up with ads and the functionality is so janky. And, like, you know, in two days, I can make a thing that has more features than theirs, substantially more features than theirs, including ones that are specifically aimed towards a weird use case of doing screen recordings for a podcast. But no one's ever going to make that product, so you kind of have to make it yourself. But even setting this aside, like even the people who just made their little toy things and sent me them or whatever, those are better than the TearMaker website, too, just because the TearMaker website is so terrible. Yep, I totally buy it. That's super cool, John, and definitely a much happier thing to talk about. So I appreciate it.