PodSearch

ATP

540: The Points Don’t Matter

 

00:00:00   I'm in a really good mood tonight.

00:00:01   Good!

00:00:02   You know, when you're writing Swift UI,

00:00:05   sometimes things don't go your way.

00:00:08   Sometimes you hit walls and you can't figure out

00:00:12   why something isn't updating or you can't get something

00:00:14   to look or work right or whatever the case may be.

00:00:17   Today, I had an incredibly good Swift UI day

00:00:21   where I'm like, I was like firing on all cylinders,

00:00:23   like getting everything working.

00:00:25   Like I actually would like go like, yes.

00:00:28   Like when I got something working to nobody

00:00:29   and you know, I'm alone in the house

00:00:32   in a room by myself, clapping to myself.

00:00:35   This was a really good day.

00:00:36   I feel so good right now.

00:00:38   This is like the Swift UI high.

00:00:40   Things are working, things are looking good.

00:00:43   I'm actually making progress.

00:00:46   Now I just have to rewrite the entire rest of the app.

00:00:49   But the things I've been working on have been really good.

00:00:53   - Other than that, how'd you enjoy the play, Mrs. Lincoln?

00:00:54   No, all kidding aside, that's really fantastic.

00:00:57   - I love a good Swift UI day.

00:00:59   A good Swift UI day, like I really just like the whole thing

00:01:03   about a 10X developer and I think that's mostly been put

00:01:06   in the recycling bin as things that we think are true.

00:01:09   But when I'm having a good Swift UI day,

00:01:14   I feel like one of those mythical 10X developers.

00:01:16   I get so much done in so little time.

00:01:18   Like when you and Swift UI are holding hands

00:01:21   and skipping together, it is the best.

00:01:25   It is the best.

00:01:26   (electronic beeping)

00:01:28   Five minutes before the show,

00:01:29   I finally got the Vision OS SDK downloaded and installed

00:01:32   and I got to run Overcast, the current version of Overcast,

00:01:36   not the new Swift UI rewrite.

00:01:37   Got to run that in the simulator,

00:01:38   just like in iPad window mode.

00:01:40   And hoo boy, do I have a lot of work to do.

00:01:42   (laughing)

00:01:44   That's, you know, it's funny.

00:01:46   Like earlier I was, I had tried, hey, let me see if I,

00:01:50   should I be doing a Mac Catalyst version as well?

00:01:53   I decided, last week I mentioned like I'm not gonna do

00:01:56   the native AppKit version because it just wasn't worth

00:01:59   the amount of trouble it would be.

00:02:02   And I'm looking at Catalyst, I'm like,

00:02:03   I might do Catalyst, I don't know.

00:02:05   But it seemed like even that's like a decent amount of work.

00:02:08   And then I look at Vision OS, like, oh no.

00:02:10   (laughing)

00:02:11   This is gonna be a lot of work.

00:02:13   But you know, we'll see.

00:02:16   I mean, the iPad version seems to work.

00:02:20   I wouldn't say it's good.

00:02:22   But neither is the iPad version that runs on the Mac now

00:02:25   for M1 Macs, like, that's not good either.

00:02:27   - Yeah, it's not that bad.

00:02:28   It's not that bad.

00:02:29   - It's not that good.

00:02:30   And I know that, 'cause it's just an iPad app

00:02:32   running in the window.

00:02:33   You know, there's only so good that can be.

00:02:36   And so, you know, it's fine.

00:02:38   It's better than not having anything.

00:02:40   It's certainly better than my horrendous web app.

00:02:42   But it is not nearly as good as something that has

00:02:45   like any care put into it at all

00:02:47   to customize it for the platform.

00:02:49   And I would like to just keep that running on the Mac

00:02:52   like in the way it is now and just like tweak things

00:02:54   a little bit, like, you know, give it a nice toolbar,

00:02:56   give it, you know, a couple little things.

00:02:57   But I don't think I can actually do that

00:02:59   without making a Catalyst app.

00:03:01   And that's, making a Catalyst app is harder

00:03:03   'cause it's a whole separate target,

00:03:05   a whole separate binary.

00:03:06   You gotta submit it to the Mac App Store,

00:03:07   everything separate from the iOS app.

00:03:09   Like, it's kind of a different beast

00:03:11   that's probably not worth the amount of trouble.

00:03:12   But Vision OS, I'm looking forward to.

00:03:14   I think I have a lot of work to do.

00:03:17   And I'm really glad it isn't coming out like, you know,

00:03:19   next week or next month or anything like that.

00:03:21   Thank goodness we have a lot of time

00:03:23   because we're gonna need it.

00:03:24   - Now, I actually think you're selling yourself

00:03:29   a little short.

00:03:30   Overcast for Apple Silicon on the Mac is really not that bad.

00:03:33   The only complaint I have about it,

00:03:34   which you probably can't fix, is that, you know,

00:03:36   as we've discussed many times, I'm a devout spaces person

00:03:39   and Overcast lives in one of the spaces

00:03:41   on one of my monitors.

00:03:43   And the media keys work fine as long as Overcast

00:03:47   is on a visible space.

00:03:49   But if I like hit play when Overcast is on a different space

00:03:53   that I'm not looking at, when I come back to the space

00:03:56   that Overcast is on, that's when it starts playing.

00:03:59   Which again, I don't know that there's really anything

00:04:00   you can do about that short of going the whole

00:04:02   catalyst route, but that drives me bananas.

00:04:05   Other than that, I actually think it's really not bad at all.

00:04:07   It's more than serviceable.

00:04:08   It's actually even pretty decent.

00:04:10   So you're selling yourself short.

00:04:12   - I mean, it's bad, but I mean, you know,

00:04:14   this is coming from the guy who was like,

00:04:16   "Yeah, my iMac reboots itself like, you know,

00:04:17   "three times a day, but it's not that bad."

00:04:19   - Oh, stop.

00:04:20   At least I'm being generous, right?

00:04:22   - And the media key thing, I don't think

00:04:23   I can do anything about that.

00:04:24   I can try, but like I made some change early on

00:04:28   in the M1 Mac era.

00:04:30   - Which made it much better.

00:04:31   Yeah, yeah, I remember, I don't remember

00:04:32   what it was specifically, but I remember you doing it

00:04:34   and it made it much better, but it didn't get us

00:04:36   100% of the way there.

00:04:37   - Yeah, it was some like, you know, change

00:04:39   to one of the remote control command things

00:04:41   or some audio session thing, I forget.

00:04:42   Some like little tiny detail.

00:04:43   But I mean, I've found, personally,

00:04:45   I've found Mac media key handling, in general,

00:04:48   not just in Overcast, like even just with Apple's own

00:04:51   music app, has been so buggy that I usually

00:04:54   don't use media keys anymore.

00:04:55   Like it works so infrequently, I usually just

00:04:58   switch over to the music app and hit space bar

00:05:01   because it's just so unreliable.

00:05:03   Like if I hit play/pause on my keyboard at any given time,

00:05:06   I don't know what's gonna happen, usually nothing.

00:05:08   And so I'm just like, that's not worth the hassle.

00:05:10   I'll just, like I just give up on 'em.

00:05:13   - Yeah, I think the problem for me with media keys,

00:05:15   normally I feel like they work okay,

00:05:17   but if there's a scenario where I have multiple things

00:05:20   that I could ostensibly want to play/pause,

00:05:23   so let's say I have Apple Music open,

00:05:25   I have a YouTube video open in Safari,

00:05:29   then the Mac, I don't know how it's supposed to intuit

00:05:32   which one I want, but I feel like it always guesses

00:05:36   the wrong one. (laughs)

00:05:37   And so like, you know, I like just opened up a tab

00:05:40   in Safari that happens to have a YouTube video in it,

00:05:42   but music is playing, so I wanna pause my music

00:05:45   so I can watch the YouTube video,

00:05:46   and what ends up happening inevitably

00:05:48   is it plays the YouTube video

00:05:49   while the music is still playing.

00:05:51   It's like, no, I wanted the other one.

00:05:53   So yeah, it's a hard problem to solve.

00:05:55   Like again, how can you possibly know

00:05:56   which one I want to control with this one button?

00:05:59   But it's just funny to me

00:06:00   that it seems to consistently choose the wrong one.

00:06:03   - Wait till they add eye tracking to Mac OS,

00:06:05   then they'll tell what you're looking at and play/pause that.

00:06:07   - Yeah, that's true, that's true.

00:06:09   All right, and speaking of eye tracking,

00:06:10   we have a lot of followup to do,

00:06:11   and it starts with Vision Pro,

00:06:13   and Brandon Jones has some feedback for us

00:06:16   and information for us.

00:06:17   You wanna talk about this, Jon?

00:06:19   - Sure, this is something that was in the WWDC videos,

00:06:21   but Brandon Jones did a good job

00:06:23   of summarizing sort of the implications of it all.

00:06:25   So I'll read, these are from several Mastodon posts

00:06:28   that I've sort of compiled here.

00:06:29   Apple's Vision OS significantly limits how applications

00:06:32   are allowed to interact with the user,

00:06:33   especially regarding their new gaze-based input,

00:06:36   and I think it's worth talking about.

00:06:37   The summary is, if you wanna do AR apps,

00:06:40   you must give Apple full rendering control.

00:06:42   A lot of it centers around Apple's choice

00:06:44   to both make eye tracking central to the headset's input

00:06:46   and simultaneously declare it to be too private

00:06:48   to be exposed to apps, which to be fair,

00:06:50   is pretty sensitive data.

00:06:51   We've talked about this before

00:06:52   when we were talking about the WWDC sessions,

00:06:54   like Apple was saying you wouldn't get eye tracking data,

00:06:57   you couldn't tell where the user was looking,

00:07:00   and how that might preclude certain types of games

00:07:03   or whatever because it's a privacy concern.

00:07:05   It's not the same as where your cursor is

00:07:07   in Mac OS or whatever, or gaze data is more sensitive

00:07:11   because you haven't even yet decided to do anything,

00:07:13   it's just where your eyes are wandering.

00:07:14   Anyway, that's the principle behind that, right?

00:07:16   - Yeah, and also, and not just that,

00:07:17   but also the Vision Pro is also looking at your entire room

00:07:21   and all of your surroundings,

00:07:22   and parts of your body and everything.

00:07:24   And so not only can apps not tell where you are looking

00:07:27   until you choose to click on something,

00:07:29   but also they can't see your room.

00:07:32   And they don't even know necessarily where the walls are

00:07:36   in regular app modes.

00:07:38   And so yeah, a lot of these details we'll get to.

00:07:40   But the gist of it is your app can say,

00:07:44   "Here's a button, let me know if someone taps it."

00:07:47   But it can't tell if you're hovering over it

00:07:50   by looking at it or selecting it or anything like that.

00:07:52   So there's all sorts of implications with this

00:07:54   that I think are pretty interesting.

00:07:55   - Yeah, so continuing here from Brandon's summary,

00:07:58   in the Safari session they described a whole new element

00:08:01   highlighting pattern that discards the web's built-in

00:08:03   hover APIs in favor of one handled entirely by the OS

00:08:06   in the name of not leaking gaze data.

00:08:08   But it's not just the web that gets this treatment.

00:08:10   If you want to display anything in the user's space,

00:08:12   you don't get to render and shade it yourself.

00:08:15   Instead, you hand off meshes and high-level materials

00:08:17   to the OS and it renders it for you.

00:08:19   The relevant WRC session,

00:08:20   if you wanna learn about this by the way, is 10096,

00:08:24   build great games for spatial computing.

00:08:26   Although I haven't actually watched the session,

00:08:27   but I do wonder if they talk about

00:08:28   shooting things with your eyeballs.

00:08:30   Anyway, Brandon continues,

00:08:32   "From the OS perspective, this is an attractive approach.

00:08:34   They can handle object selection and highlighting

00:08:36   without ever exposing gaze or hand data."

00:08:38   Now let me just summarize what this part

00:08:40   is talking about here.

00:08:42   So when you look at something,

00:08:43   you might want it to highlight, as the controls do,

00:08:46   the little buttons and stuff or whatever.

00:08:47   But hey, whatever you're doing.

00:08:48   Say you write an app and it's got a bunch of 3D objects

00:08:50   and you want the thing that the person's looking at

00:08:52   to highlight so they know when they make

00:08:54   the pinch gesture, it will do it.

00:08:55   If you as the application developer

00:08:58   were able to actually render the 3D objects,

00:09:00   you'd be like, okay, well now if they're looking at it,

00:09:03   I have to make it glow.

00:09:05   So you'd have code for that.

00:09:06   But you don't know when they're looking at it.

00:09:08   The OS handles all the rendering for you.

00:09:10   And because the OS has handed the meshes

00:09:13   and it says here you go, here's the raw materials OS,

00:09:16   you render it, the OS can decide when to highlight

00:09:20   your button, your 3D object or whatever

00:09:22   without your app knowing.

00:09:24   Because that's all happening on the rendering side.

00:09:25   So you don't get to choose like,

00:09:27   oh, when they look at it, I want this to look like this

00:09:31   or whatever, the OS handles that entirely,

00:09:32   which is very unlike 3D APIs in the normal sense

00:09:36   where you, yes, you construct the objects or whatever,

00:09:38   but you also choose how to render them.

00:09:40   And not so in Vision OS and that's so the OS

00:09:43   can do things like highlight without your app knowing.

00:09:45   To continue, they can light the scene

00:09:47   without exposing environment data.

00:09:49   What that means is, what Mark was just saying,

00:09:51   how they don't know what your room looks like.

00:09:53   Like if you're doing it in AR mode,

00:09:54   they take like your room, whether you know,

00:09:57   say you have like your walls are painted red

00:09:59   and you have lots of lights on.

00:10:00   That's gonna bounce a bunch of red light around.

00:10:03   That is incorporated into the rendering

00:10:05   of whatever you put in there.

00:10:06   If you have a little 3D object in your app

00:10:08   or your little windows or whatever,

00:10:09   the red light bouncing off the walls

00:10:11   influences how they're rendered.

00:10:13   But because you as the app developer

00:10:15   don't even have access to the surrounding room,

00:10:17   you couldn't do that.

00:10:18   The OS does it for you.

00:10:19   The OS sort of treats your room as kind of like,

00:10:21   I imagine kind of like a cube map or whatever

00:10:23   and lights the objects that you put in front of the user

00:10:27   so they look like they're incorporated into the room

00:10:29   without your app ever having any idea

00:10:32   what the room looks like

00:10:33   because you don't have to render that.

00:10:34   You don't have to figure out what the lighting should be.

00:10:36   You don't have to figure out how to like

00:10:38   make it look like that object is really on the coffee table

00:10:40   given the room's lighting.

00:10:41   The OS does that for you.

00:10:43   Continuing from Brandon here,

00:10:44   they can include virtual objects

00:10:45   without exposing room geometry.

00:10:47   Again, they can, you know,

00:10:48   they know what's blocked by your coffee table or whatever.

00:10:52   Brandon says, "Want to try to experiment

00:10:53   with different input models

00:10:54   or try out an advanced new rendering technique?"

00:10:56   Sounds like you simply can't,

00:10:57   at least not in any AR environment.

00:10:59   Apple is offering more flexibility

00:11:01   if you're doing a VR app,

00:11:02   but there is no camera/gaze data allowed.

00:11:05   So I thought this was really interesting

00:11:07   and like it's, there's sort of the technical side of this

00:11:10   that allows Apple to do two things.

00:11:12   One, to handle a lot of the drudgery for you.

00:11:15   Like I don't want to, you know, I just have an iOS app.

00:11:17   I don't want to figure out how my little floating window

00:11:19   should look like it's well integrated into the room.

00:11:22   But two, even if you're writing

00:11:23   a full fledged native Vision OS app

00:11:25   with a bunch of 3D objects and whatever,

00:11:27   you know, whatever you come up with,

00:11:28   a cool app that would only work in Vision OS,

00:11:31   you both don't get to render those things.

00:11:34   Also, you don't have to.

00:11:36   The OS will handle it for you.

00:11:37   And this is, you know, a great example

00:11:39   of how privacy focused Apple is,

00:11:42   that this sort of like fundamentally defines

00:11:45   the application programming interface

00:11:47   for 2D and 3D stuff on Vision Pro

00:11:49   entirely in service of privacy

00:11:52   and not letting apps get at information

00:11:54   that they don't have to have to do their job.

00:11:57   - Yeah, I mean, you can imagine if they didn't

00:11:59   put all these protections in place,

00:12:01   look at the terrifying, awful, like morally bankrupt way

00:12:06   that ad tech and surveillance tech over the years

00:12:10   have exploited even the smallest bits of information

00:12:14   they can collect from you or they can infer about you.

00:12:18   Anything that you give them, any like,

00:12:20   they'll try to sniff your browser configuration

00:12:22   and confederate, try to fingerprint you.

00:12:24   Like, you know, the obviously IP address stuff,

00:12:26   like there's so much stuff they try to derive

00:12:30   from any little bit of data they get about you.

00:12:32   Imagine if they could see your room,

00:12:35   everything around your room, everything about it.

00:12:37   They could derive where you were,

00:12:38   they could derive who you are very, very easily,

00:12:41   even more easily than they do now, you know.

00:12:43   Then imagine gaze data, imagine if they could track

00:12:46   your eyes in real time as you like browsed a webpage.

00:12:48   If you think algorithm generated content is bad now,

00:12:53   imagine if they had that much data

00:12:55   on exactly where you look and when and how.

00:12:58   Imagine what they could create, what kind of hellscape,

00:13:02   like the web would become even more horrible than it is now

00:13:06   with data they could like generate based on,

00:13:09   well, this will generate not only the most clicks,

00:13:12   but this will make people look at it more.

00:13:14   Like, it would be a terrifying hellscape

00:13:17   even more than it is now.

00:13:19   And so having this level of focus on an AR environment

00:13:23   is something that I don't think any other major tech company

00:13:27   would even consider, let alone put this much effort into

00:13:32   and have this much conviction about,

00:13:33   'cause you know this is going to, in some ways,

00:13:37   hurt app development on VisionOS.

00:13:39   And I have more to say about that later,

00:13:41   but talk about like principles and courage

00:13:44   to this heavily restrict such fundamental data

00:13:49   about your app's environment and interaction

00:13:51   from being visible to apps and to put as much effort

00:13:54   into it as they did to keep those things separate

00:13:56   and private, that really takes a lot of dedication

00:14:00   on this front.

00:14:01   And again, I think it's the right call.

00:14:02   I mean, this platform is extremely early.

00:14:06   It's really hard to say at this point

00:14:09   what will be the best decision in retrospect,

00:14:11   what will work out long term, will they have to loosen up

00:14:13   on some of these things over time?

00:14:15   We don't really know yet, it's way too early to say.

00:14:17   But I think this is gonna prove to be a good move,

00:14:21   even though it will create a bit of pain in the buttery

00:14:24   for certain development tasks and things like that.

00:14:27   Certain things will become a little bit tricky

00:14:29   that you have to work around, or certain features

00:14:32   or interactions that you'll have to either not do

00:14:34   or stick with the stock way that they're done,

00:14:36   rather than doing your own custom thing.

00:14:37   But again, I think overall this is probably the right move.

00:14:42   And only Apple would have done this.

00:14:44   - Yeah, and if they didn't do this,

00:14:45   the ad tech companies would be reading the spine

00:14:48   of every single book in your bookshelf,

00:14:50   reading the publications that are sitting

00:14:52   on your coffee table, doing face recognition

00:14:54   against everybody in your family,

00:14:56   listening to every sound that plays,

00:14:57   like determining where you live

00:15:00   based on looking out your windows at the angle of the sun

00:15:02   and cross, you know, calibrating with Google Earth

00:15:06   because you didn't give a location date.

00:15:07   - Oh, they're watching you, they'd be watching your arms,

00:15:10   your skin, your fingerprints, anything they could see

00:15:14   they would use, and they could see a lot from something

00:15:17   that has such a broad view of the world.

00:15:19   - And the flip side of that is that, you know,

00:15:21   so this is, it makes certain things in apps more difficult,

00:15:23   but it does make the experience more coherent.

00:15:27   Not having control over the rendering means

00:15:29   every object that is rendered in augmented reality,

00:15:32   everything that is supposed to be floating

00:15:33   in the middle of your room or sitting on your coffee table

00:15:36   will be rendered in a consistent way.

00:15:38   If you control that rendering yourself,

00:15:40   app developers could do whatever they wanted

00:15:42   with the rendering and the texture mapping

00:15:44   and the shading and the shadowing,

00:15:46   and would they all choose to do things in the same way

00:15:49   with the same level of skill?

00:15:50   Probably not, and that would lead to a,

00:15:53   more of a hodgepodge inconsistent experience,

00:15:56   even not in any kind of malicious way,

00:15:58   but just if, you know, app A developer makes different

00:16:01   choices about how to shade things in app developer B

00:16:03   and you have them both running at the same time,

00:16:05   it'll look, they'll look different.

00:16:06   Even if they both try to incorporate themselves

00:16:08   into the room somehow, they would look different

00:16:10   from each other due to different decisions,

00:16:11   whereas in the OS does it, it's gonna be consistent

00:16:14   across every single application, because it has to be,

00:16:16   because the apps literally don't control the rendering,

00:16:19   the OS does.

00:16:20   - Yeah, I mean, look at, on iOS,

00:16:22   you have many different visual styles

00:16:25   and many different interaction styles,

00:16:26   and even in the absence of any kind of malice,

00:16:29   what you get is a world of a lot of web apps

00:16:31   and electron apps, and the way they behave,

00:16:34   on that we all know this by using them,

00:16:36   they behave in subtly different ways

00:16:38   in lots of little implementation details,

00:16:40   lots of feels, lots of looks, lots of behaviors,

00:16:43   they're just different from the system defaults.

00:16:46   Imagine that in AR, it's easy when we see,

00:16:50   we're seeing this new platform right now,

00:16:52   we're seeing it in PR demo mode.

00:16:54   We're seeing Apple's polished apps

00:16:57   that they showed during the demo,

00:16:58   we're hearing about the press demos that everyone else got,

00:17:01   that are all these carefully walked through events

00:17:05   that were carefully polished for those demos.

00:17:07   What we're not seeing yet is what's gonna happen

00:17:10   when this gets exposed to the world,

00:17:12   and we get every crappy developer

00:17:15   trying to make crappy corporate apps for this thing.

00:17:18   It's not gonna be that perfect,

00:17:20   and the more Apple puts in place at the start

00:17:23   to try to control the basics,

00:17:26   the better this platform can be,

00:17:27   and something like how an object is rendered in your room,

00:17:31   that probably should be consistent,

00:17:33   because if you imagine different objects on screen

00:17:35   at different times from different apps,

00:17:37   or different styles, or different renderers

00:17:39   that are being used, different libraries

00:17:41   that are being used between different apps,

00:17:42   if one of them's using whatever the AR version

00:17:46   of Electron will become, like R Electron or whatever,

00:17:48   if somebody is using R Electron apps today,

00:17:51   their corporate BS app, 'cause they don't wanna write it

00:17:53   for three different AR headsets,

00:17:55   that's gonna have a different rendering style.

00:17:57   The light might hit it differently,

00:17:59   it might have different textures,

00:18:00   or different edge behaviors,

00:18:02   or different interaction behaviors.

00:18:03   There are so many ways that that could be

00:18:05   really disorienting, or just crappy looking and sloppy,

00:18:09   and so much about AR needs to be both consistently rendered,

00:18:14   and rendered with a lot of sophistication,

00:18:18   in order to prevent pretty basic fundamental problems,

00:18:21   like motion sickness, or disorientation,

00:18:24   or things like that, and so, you can imagine,

00:18:26   like if there's something, if there's some custom renderer

00:18:28   out there used by R Electron apps

00:18:30   that is a little bit crappier than Apple's built-in one,

00:18:33   and you know it would be,

00:18:34   it could cause weird problems for a lot of people.

00:18:36   This kind of environment is very sensitive,

00:18:38   it needs to be controlled a lot,

00:18:40   and to have Apple control more of that rendering pipeline,

00:18:44   from, again, from this early point of view,

00:18:46   that who knows what we'll think in the future,

00:18:48   but this early point of view,

00:18:49   that sounds like the right call.

00:18:51   - Well plus, what you're looking at

00:18:53   isn't constrained to a small, or at best,

00:18:57   medium-sized box in front of you, right?

00:18:59   Like, I'm looking at two 5K screens,

00:19:02   and even still, all of the UI wonkiness that's happening

00:19:06   is contained in those two rectangles.

00:19:08   Granted, it's not real, but when you're suddenly--

00:19:11   - You're not real, man.

00:19:12   - Yeah, right.

00:19:13   When you're busting all of these things

00:19:15   into your quote-unquote real life,

00:19:17   I mean, again, I know it's not exactly true,

00:19:19   but that can be quite a bit more jarring

00:19:21   than when it's limited to just a couple of rectangles

00:19:24   that are directly in front of your face.

00:19:25   And so, I think having that consistency makes sense.

00:19:28   And I really applaud Apple,

00:19:31   'cause I think I could imagine other companies

00:19:34   going for AR and just shrugging at all the privacy stuff.

00:19:38   What I commend Apple for is they clearly

00:19:40   have really thought it through.

00:19:42   Now, as Marco said, ad tech and tracking

00:19:46   and surveillance tech, they are as slimy as they come,

00:19:48   and I'm sure they'll find a way

00:19:49   around some of these protections,

00:19:51   but it sure sounds like Apple's put a lot of thought

00:19:55   into making sure this is safe for users,

00:19:56   which, as a user, whether or not I get one of these

00:19:59   immediately or later or whatever,

00:20:01   I really appreciate that.

00:20:02   - You're getting one immediately.

00:20:03   - Well, one of the three of us is gonna have to,

00:20:05   and I'm not sure which one of us is gonna be

00:20:07   without a chair when the music stops.

00:20:08   - I'm definitely, do you see how crappy my app looks?

00:20:10   I have to get one.

00:20:12   - Yeah, we do always know you'd find an excuse.

00:20:15   (laughing)

00:20:15   - I think, given that I'm a professional iOS developer,

00:20:18   I don't think it's an excuse.

00:20:20   - Are you a professional XROS,

00:20:22   I mean, vision OS developer?

00:20:23   I don't think so.

00:20:25   No, I'm giving you a hard time.

00:20:26   We're probably all in for one.

00:20:27   All right, John, tell me about--

00:20:29   - Actually, before we move on,

00:20:31   for the people who listened to what Casey said before

00:20:33   and think, I don't care that all my apps on my screen

00:20:35   have different UIs.

00:20:36   In fact, I think that's a strength.

00:20:37   I like the fact that each app can have its own UI.

00:20:40   It's not so much like, oh, people will be confused

00:20:42   that apps have different UIs

00:20:44   and they won't know how they look.

00:20:46   It gets to more of what Casey was saying,

00:20:47   like they're not just confined to his screens,

00:20:50   they exist in the world of his screens.

00:20:53   And there is some consistency in that world.

00:20:54   The OS controls like the shadows on the windows,

00:20:57   for example, or whatever,

00:20:58   although some of that's over-rightable.

00:20:59   But the whole point of augmented reality

00:21:03   is that everything, even the stupid little rectangles,

00:21:05   even the stupid little flat rectangles

00:21:07   that are just a phone app or an iPad app,

00:21:09   even those that are apps that have nothing to do

00:21:11   with virtual reality or 3D, it's just a plain old app,

00:21:15   the rectangle is supposed to be floating

00:21:18   in the air in your room.

00:21:19   That is not true of any of the windows on your screen.

00:21:22   They're not supposed to be in your room.

00:21:24   They are supposed to be in the room,

00:21:27   on a monitor that's in your room,

00:21:29   but they do not incorporate anything from your room in there.

00:21:33   They're not like floating above your desk.

00:21:35   They're in the world of the monitor.

00:21:37   So everything in Vision OS,

00:21:39   when you're in AR mode and not in VR mode,

00:21:41   where you're seeing your room and everything,

00:21:43   is supposed to look like it is in that room.

00:21:46   That's the consistency they're going for.

00:21:48   Not that everything on the screen has to look the same.

00:21:50   You can make wildly different apps.

00:21:51   Take all those wildly different apps

00:21:52   that Marco was talking about on iOS,

00:21:54   you can run them all in Vision OS.

00:21:56   But when those little rectangles are floating in the air,

00:21:58   they better look like they're floating

00:21:59   in the room that you're in.

00:22:01   The rectangles themselves will be wacky

00:22:03   and have stupid banner ads and whatever the hell they are,

00:22:05   but they have to look like they're in your room.

00:22:07   And of course, obviously for 3D objects,

00:22:08   for actual apps that take advantage of Vision OS

00:22:10   and aren't just little floating flat iOS and iPad apps,

00:22:13   same deal, but even for the flat ones,

00:22:15   the rectangles have to look like they're in your room.

00:22:17   And that goes all the way up to the,

00:22:20   I was playing with the Vision OS demo thing too,

00:22:22   the basically the window backgrounds.

00:22:24   Like someone was pointing out like,

00:22:26   hey, there's no light mode and dark mode in Vision OS.

00:22:28   You wanna know why?

00:22:29   The window material,

00:22:30   the thing that windows are made out of in Vision OS,

00:22:32   is this weird, magical, translucent, frosted glass,

00:22:36   watch a moozy.

00:22:37   And they have a bunch of like demo rooms that you can flip,

00:22:40   like kitchen during the day, kitchen at night,

00:22:43   living room at night, living room.

00:22:45   Try all the different lighting.

00:22:46   It's like, how can this window be legible

00:22:48   in all this different lighting?

00:22:48   What color is this window?

00:22:50   Is it a white window?

00:22:50   Is it a dark window?

00:22:52   I can't really tell.

00:22:53   It's the OS handles this for you.

00:22:55   Like, because if you tried to do it yourself and you said,

00:22:58   well, my app is always a black window with white text,

00:23:01   and then you put it in a pitch black room

00:23:02   and no one can see the window anymore,

00:23:03   just like the text is floating in midair.

00:23:05   The OS is handling so much stuff for you

00:23:07   to maintain the illusion that things really are integrated

00:23:12   into your reality.

00:23:13   And the final point I'll make of this is that

00:23:15   by Apple doing all of this,

00:23:17   when Apple gets better at doing this,

00:23:19   when the new hardware comes out,

00:23:20   when they revise their software,

00:23:22   when they get better and better at making it look

00:23:24   like it's photorealistically floating in your actual room,

00:23:27   every app will benefit from that

00:23:29   because the apps never controlled that rendering

00:23:31   to begin with.

00:23:32   It's not like if someone does an app

00:23:33   and doesn't update it for three years,

00:23:34   it'll look like a cruddy, you know, old version.

00:23:37   It doesn't look like, no,

00:23:38   every app since they control the rendering

00:23:39   will advance along with Apple doing this stuff.

00:23:42   So I think this is definitely the right choice.

00:23:44   I do definitely think they should expose gaze data

00:23:47   in full VR for games,

00:23:48   but those are two entirely separate things.

00:23:51   One has an awareness of your surroundings

00:23:52   and the other has no,

00:23:53   it doesn't mean I have any awareness of your surroundings,

00:23:55   at which point I have no problem passing my gaze data

00:23:58   to the shooting game of letting it know

00:24:00   which thing I'm looking at to shoot.

00:24:02   So hopefully Apple will figure out

00:24:04   how to separate those priorities.

00:24:06   - Yeah, and I think this is one of those things

00:24:08   where Apple needs to see how are we gonna use it?

00:24:12   You know, I'm not the first to say that the Apple Watch,

00:24:16   you know, Apple had ideas about how we were gonna use it,

00:24:18   both in terms of just regular use

00:24:20   and in terms of developers,

00:24:21   and over time they refined, you know,

00:24:23   what they thought the watch was for

00:24:25   based on what people were actually using the watch for.

00:24:28   And I suspect we're gonna see a lot of that

00:24:30   with the Vision Pro.

00:24:31   And if there's a cry for,

00:24:33   oh, we really would like to do such and such,

00:24:36   but we need eye data,

00:24:38   then presumably Apple will figure out a way to facilitate it,

00:24:41   be that, you know, only in VR mode, like you're saying,

00:24:44   or maybe there's some sort of way to do it in AR mode

00:24:47   that's privacy conscious,

00:24:48   or maybe you have some sort of like a dialogue

00:24:51   that you have to approve that says,

00:24:52   hey, you know, this app really, really, really

00:24:54   wants your location.

00:24:55   I mean, eye tracking data, and you know,

00:24:58   is that okay with you?

00:24:59   And any of those things,

00:25:00   I don't know if they're the right answer,

00:25:01   but they are answers.

00:25:03   So we shall see.

00:25:05   Speaking of gaze and things like that,

00:25:08   we have some links to some various WWDC sessions,

00:25:11   particularly 10073 Design for Spatial Input.

00:25:15   Tell me about this, please.

00:25:17   - Yeah, so this is just kind of like,

00:25:19   hey, how do I design things for Vision OS,

00:25:21   assuming I'm like making a native app for it

00:25:22   and not just like running one of my existing ones.

00:25:25   One of the things that caught my eye

00:25:26   was they showed like the minimum area for UI elements.

00:25:31   They say that it is 60 points,

00:25:34   but the elements can be smaller than 60 points with margins.

00:25:36   And so there's two little diagrams.

00:25:37   They showed like a button and a 60 point square.

00:25:41   And they said, hey,

00:25:42   so the button doesn't fill the 60 point square.

00:25:44   In fact, there's an eight point margin on either side,

00:25:46   making the actual button, dun, dun, dun, 44 points,

00:25:49   which a number that stuck out to me,

00:25:51   wasn't that exactly the same size that they used to say

00:25:54   it was like on the original iPhone,

00:25:56   the minimum size of your buttons should be 44 points.

00:25:59   So I thought that was interesting.

00:26:00   And also interesting that the eye tracking

00:26:02   is apparently less precise than our meaty fingers

00:26:05   because they want the hit area to be bigger than 44.

00:26:08   The hit area is 60, but the visual thing is 44.

00:26:11   Now you may be hearing this and saying, wait a second,

00:26:13   what the hell does 60 points mean in an environment

00:26:17   where my window is floating in midair

00:26:19   and can be pulled closer and farther away from the user

00:26:21   and the person can walk around?

00:26:22   Like, does points mean anything?

00:26:24   On a screen, it means something

00:26:26   because no matter how far away your iPhone screen is to you,

00:26:28   you have to touch your finger to it to touch it.

00:26:31   And once your finger touches it,

00:26:32   44 points given Apple's DPI and blah, blah, blah,

00:26:35   is roughly the same size for across different lines.

00:26:37   It's varied a little bit,

00:26:38   but you can see how point size for touch targets,

00:26:41   having a standard makes some sense.

00:26:42   But how does having a point size on an AR app

00:26:46   make any kind of sense?

00:26:47   Well, Apple continues in that same session to explain this.

00:26:51   We'll put a timestamp links of these two different offsets.

00:26:54   You can see that those parts,

00:26:55   if you don't wanna watch the whole video.

00:26:56   Apple says, "The system provides dynamic scale

00:26:59   for app windows.

00:27:00   You can see how the window scales larger as it moves away

00:27:03   and smaller as it moves close.

00:27:04   Dynamic scale makes your UI fill the same field of view

00:27:07   and preserve the size of the targets

00:27:08   no matter where the window is positioned."

00:27:10   So you can see this in the video, but think of it this way.

00:27:13   Normally, when objects get farther away from you,

00:27:15   they look smaller.

00:27:16   That's just perspective, right?

00:27:18   What Apple does is as you push a window away from you,

00:27:22   it makes the window bigger.

00:27:24   So no matter how far you push it away and vice versa,

00:27:27   or how close you pull it to yourself,

00:27:28   it always fills the same proportion of your field of view.

00:27:31   That is a dynamic scale mode for windows.

00:27:34   So 60 points makes sense because it's like,

00:27:35   go ahead, put that window wherever you want.

00:27:37   I am gonna make sure that in the world of this AR thing,

00:27:42   that button is always 60 points.

00:27:44   You can't make it bigger by pulling the thing towards you.

00:27:46   You can't make it smaller by pushing the thing away.

00:27:48   It's gonna be 60 points no matter what,

00:27:50   because you have to be able to target it with your eye

00:27:52   and their judgment of what is comfortably eye-trackable

00:27:55   given their current eye-tracking technology

00:27:56   in people's eyes is 60 points.

00:27:59   So that's really weird if you see it happening

00:28:01   like they show in the video.

00:28:02   From your perspective, it seems okay, right?

00:28:05   But if you were to look at it from the side,

00:28:07   like if you were to, you know,

00:28:08   they give you like a view in the 3D world of like,

00:28:10   well, what if there was another 3D camera over here

00:28:12   looking at, you can see the window getting bigger

00:28:14   as it gets pushed away and smaller

00:28:15   as it gets pushed forward and it looks really weird.

00:28:18   You can use fixed scale instead,

00:28:20   and in that case, it behaves like a regular 3D thing

00:28:23   where when you push it away from you,

00:28:23   it gets smaller, when you pull it towards you, it gets bigger.

00:28:26   But I thought that was super interesting

00:28:27   and that answers a question

00:28:29   I hadn't really thought about until, you know,

00:28:30   actually try using these apps in the simulator.

00:28:32   Like, can you make an app unusable

00:28:35   because it's too far away from you?

00:28:37   And in their default mode,

00:28:38   they try to make sure that doesn't happen.

00:28:40   - Vision OS, where everything's made up

00:28:42   and the points don't matter.

00:28:43   - Mm-hmm.

00:28:44   Well, the points do matter, but everything's still made up.

00:28:47   - Hey, hey you, you ATP listener.

00:28:50   Hi, hey, it's me.

00:28:51   We are sponsored this week by ATP Membership.

00:28:54   Now, if you're hearing this, you're probably not a member

00:28:56   because members get an ad-free version of the show.

00:28:59   So all these little promos that I'm sticking in like this,

00:29:01   and of course the regular sponsor reads

00:29:03   are not included in the member version of the show.

00:29:05   You can add that feed to any podcast player you want.

00:29:07   We don't care what you use.

00:29:08   Well, I care, but you know,

00:29:10   the rest of us don't care so much.

00:29:12   And all of that's just eight bucks a month

00:29:14   or whatever that multiple is per year.

00:29:15   Now, we also give you a bootleg version of the show.

00:29:18   If you want, you can listen to that version,

00:29:20   which has its own feed, again,

00:29:21   add it to any podcast player you want.

00:29:22   That includes everything we do on the live stream,

00:29:25   and it's unedited, raw audio, you know.

00:29:27   It doesn't have chapters, but it does have Casey swearing,

00:29:29   and it does have, you know, whatever we talk about,

00:29:32   you know, before and after the show.

00:29:34   Yeah, maybe I'll cut it from the published show

00:29:35   'cause it might not fit or whatever,

00:29:37   but we'll put it in the bootleg.

00:29:38   A lot of people love that perk.

00:29:39   You also get occasional discounts on merchandise

00:29:41   during some of our sales, but for the most part,

00:29:43   you're paying to support the show.

00:29:44   And look, we love you for listening

00:29:46   no matter how you support us.

00:29:48   You can support us by listening to the show

00:29:49   with the sponsors in it.

00:29:51   That is very great, and we thank you for that.

00:29:53   If you wanna join and become a member

00:29:54   for eight bucks a month and get those cool perks,

00:29:56   you can do that too.

00:29:57   It's a different way to support the show.

00:29:58   We love all of our listeners equally,

00:30:01   but certainly if you wanna do membership,

00:30:02   we strongly encourage that.

00:30:03   So anyway, atp.fm/join to learn more about that.

00:30:07   Thank you so much, and now back to the show.

00:30:10   - And then Christopher Masto had some information

00:30:17   on how our eyes work and will likely cope with Vision Pro.

00:30:22   And so this begins with the two techniques

00:30:26   your eyes have to focus on stuff.

00:30:28   There is, and I read up on this earlier,

00:30:30   I'm probably getting some of these details wrong,

00:30:31   but the general idea is there's vergence and accommodation.

00:30:34   So vergence is your eyes ever so slightly,

00:30:39   they're pivoting along a vertical axis.

00:30:41   So you're not going literally cross-eyed,

00:30:44   but your eyes will come closer together a little bit

00:30:47   as you're focusing on something far away,

00:30:49   I think I have that right,

00:30:50   and they'll spread out a little bit, so to speak,

00:30:52   as you're focusing on something closer,

00:30:54   and that's vergence.

00:30:55   Then accommodation is basically,

00:30:57   hey, you have a lens inside your eye

00:31:01   that is used to focus things.

00:31:03   And Christopher writes that the discrepancy between the two

00:31:06   is one of the sources of discomfort in VR

00:31:08   because everything is at the same eye squish distance,

00:31:11   the same vergence, regardless of the binocular distance.

00:31:15   And so this creates what's called

00:31:17   the vergence-accommodation conflict, or VAC,

00:31:20   which is a visual phenomenon that occurs

00:31:22   when the brain receives mismatching cues

00:31:24   between vergence and accommodation of the eye.

00:31:26   This commonly occurs in virtual reality devices,

00:31:28   augmented reality devices, 3D movies,

00:31:30   and other types of stereoscopic displays.

00:31:33   The effect can be unpleasant and can cause eye strain.

00:31:36   We'll put links to all of this stuff in the show notes.

00:31:39   - Yeah, last week I was talking about

00:31:40   the two different things that we focus,

00:31:42   and it was the moving and the squishing,

00:31:44   and these are the actual names for them, right?

00:31:45   And to be clear, the squishing, in mammals anyway,

00:31:48   the squishing is tiny little muscles in our eyes

00:31:50   are squishing the lens of our eye.

00:31:52   There's like a little lens that you can make flatter

00:31:56   to change how the light is focused on the back of your eye,

00:31:58   where the retina is.

00:32:00   And that's, as your lens gets less squishy as you get old,

00:32:04   you have trouble focusing as much.

00:32:05   Basically, there's a measurement of like,

00:32:07   how much can you squish the lens in your eye?

00:32:09   It's like really round or really flat.

00:32:11   And that distance, like how much you can squish it

00:32:13   is measured, it's like measured in diopters or whatever.

00:32:16   As you get older, you have less range of squishing.

00:32:19   So the squishing thing is the accommodation

00:32:22   and the vergence is like Casey said,

00:32:24   the cross-eyed type thing.

00:32:25   If something's real close to you,

00:32:25   your eyes kind of point towards the bridge of your nose,

00:32:28   if something's far away, your eyes point straight out.

00:32:30   Now, here's some stuff I don't actually know.

00:32:33   Everyone who has sent in information about this has said,

00:32:36   "Hey, I use VR stuff."

00:32:38   And in VR stuff, the vergence distance,

00:32:43   the vergence is the same.

00:32:44   Like there is no pointing your eyes straightforward

00:32:48   to look at things that are far away

00:32:49   or going cross-eyed to look at things that are close.

00:32:51   Every single thing that is seen on the screen

00:32:53   in most VR headsets is at the same vergence level.

00:32:58   And that's this, you know,

00:32:59   vergence-accommodation-conflict type of thing.

00:33:01   Because you think you're looking at something farther away,

00:33:04   but your eyes, I might've got it backwards

00:33:06   that it's the same accommodation.

00:33:07   Anyway, the discrepancy between the vergence

00:33:09   and accommodation, because one of those two things is fixed.

00:33:12   Now, is that how it is in the Apple Vision Pro?

00:33:16   Is everything at the same focal distance

00:33:18   to regardless of how far away it is?

00:33:20   Will the Apple Vision Pro suffer

00:33:22   from vergence-accommodation-conflict?

00:33:24   Here's what, again, session 10073,

00:33:29   designed for spatial influence,

00:33:31   had some things to say about it.

00:33:32   And I'm not sure what this means.

00:33:34   I can interpret this both ways.

00:33:36   So at four minutes, they said,

00:33:37   "We should also consider depth

00:33:38   when thinking about eye comfort.

00:33:40   Depth is a unique feature of spatial experiences.

00:33:42   Placing your content near or far away

00:33:45   creates different feelings in your projects."

00:33:46   Okay, that seems like a reasonable truth,

00:33:50   regardless of this whole vergence-accommodation thing.

00:33:53   They continue, "But our eyes focus on one distance at a time

00:33:56   and changing the focus depth frequently

00:33:58   can create eye strain."

00:33:59   Now, wait a second.

00:34:00   How would you change the focus depth

00:34:05   of anything inside the headset

00:34:07   if the focus depth is always fixed at a particular depth,

00:34:12   no matter how far away anything is?

00:34:14   Because they're trying to say this,

00:34:14   "Hey, you know, don't bring things close and far

00:34:17   and make the person focus at different distances."

00:34:19   But if the Apple Vision Pro is just like most headsets,

00:34:24   no matter how close or far the objects are from you

00:34:27   in the virtual world, your eyes never have to refocus.

00:34:30   And to be clear, like we said last week,

00:34:32   the screens are fractions of an inch from your eyeball,

00:34:34   but the focal distance due to lenses in the headset

00:34:38   is not inches from your eyeball, right?

00:34:40   The focal distance is like, I don't know,

00:34:41   like six feet in front of you or something, or nine feet,

00:34:43   whatever it is.

00:34:43   Like, in most headsets, there is a fixed focal distance

00:34:47   and that focal distance is far in front of you.

00:34:49   That's why we were saying, like,

00:34:50   you're not gonna get myopia from looking at screens

00:34:52   that are a fraction of an inch away,

00:34:53   because the focal distance is farther.

00:34:54   And how is the focal distance farther?

00:34:56   There are lenses in the Apple Vision Pro,

00:34:58   even if you don't wear glasses.

00:34:59   There are additional lenses that you add onto that

00:35:01   if you wear glasses,

00:35:02   but even if you have perfect 20/20 vision,

00:35:04   the Apple Vision Pro has lenses inside of it

00:35:07   that set the focal plane of, you know,

00:35:09   hey, if you wanna see the images on that screen,

00:35:11   focus your eyes as if they're looking at something

00:35:13   six feet in front of you,

00:35:14   or whatever the focal distance plane is.

00:35:17   So Apple continues,

00:35:18   "Look to keep interactive content at the same depth

00:35:21   to make it feel effortless to switch between UI.

00:35:23   By maintaining the same Z position,

00:35:25   your eyes don't need to adapt to the new distance."

00:35:28   Again, I am confused.

00:35:29   Z distance, by the way, is like,

00:35:30   how far is it from your face?

00:35:31   Like, is it really far in the distance,

00:35:32   or is it really close to your nose?

00:35:34   And they're saying,

00:35:35   hey, don't make things different Z distances.

00:35:37   Keep things close together

00:35:39   so your eyes don't need to adapt to a new distance.

00:35:41   They could be talking about

00:35:43   a vergence-accommodation conflict, saying,

00:35:44   hey, when you put things closer and farther away,

00:35:46   people's eyes will instinctively try to refocus on them,

00:35:50   but unfortunately in our headset,

00:35:51   everything is at the same focal distance,

00:35:53   and that will cause vergence-accommodation conflict.

00:35:55   So don't make people do that.

00:35:56   You could also interpret it as saying,

00:35:58   wait a second, it's possible to focus my eyes

00:36:00   at different distances inside this headset.

00:36:03   Now, there is technically a way they could do that.

00:36:05   Because they know where your eyes are pointing individually,

00:36:09   they could tell where, like how, you know,

00:36:12   what your vergence is.

00:36:13   Like, you know, if two laser beams

00:36:15   came out of both of your eyeballs,

00:36:16   where would the laser beams meet?

00:36:18   And that is the distance where you're looking,

00:36:20   like, you know, the vergence distance that you're looking.

00:36:22   Because they know where your eyes are looking

00:36:24   and which direction they're pointing,

00:36:26   they could calculate that,

00:36:27   and then they would have to mechanically move the lenses

00:36:30   inside the headset to refocus at the new focal distance

00:36:33   where your eyes are.

00:36:34   But I don't think they're doing that,

00:36:36   because A, they would have bragged about it a lot.

00:36:39   And B, focusing and refocusing in response

00:36:42   to where the vergence of your eyes

00:36:44   would have to be really fast, like, you know,

00:36:47   super expensive, like, you know,

00:36:49   super expensive mirrorless camera fast,

00:36:51   and would make noise and would have motors

00:36:53   and would destroy battery life or whatever.

00:36:54   So I don't think they're doing that.

00:36:56   So my interpretation of this, my best guess is,

00:36:58   there is a fixed focal distance inside the Apple Vision Pro.

00:37:02   Everything, no matter how far away it is from you,

00:37:05   is at that focal distance.

00:37:07   And that means that this headset does suffer

00:37:10   from vergence-to-combination conflict,

00:37:12   because your eyes will expect things farther away,

00:37:15   they will expect you to focus on them

00:37:17   at a different distance, but you won't have to, right?

00:37:19   And then this whole session is trying to say,

00:37:21   hey, don't do that, don't put one thing really far away

00:37:23   and then one thing close, then far away, then close,

00:37:25   because people's eyes will constantly be trying

00:37:26   to focus on them and they don't need to do that

00:37:28   in our headset.

00:37:29   If we actually had a headset or were able to try it,

00:37:31   this is a thing that we might look into,

00:37:34   but for now we're just speculating.

00:37:35   But I thought this was fascinating because it's a topic

00:37:37   that Apple pretty much entirely didn't touch on.

00:37:40   I didn't hear anybody who tried the headset

00:37:43   talk about this either.

00:37:44   And I personally don't have any experience with headsets

00:37:47   to say how bad vergence-to-combination conflict is,

00:37:50   or if it's tiring or eye-straining or whatever,

00:37:52   but that's definitely something we will follow up on

00:37:54   when Marco gets his headset in case he maybe gets his.

00:37:57   Are you not in for one, John?

00:38:00   - Maybe. - We'll see.

00:38:01   - Oh, come on, okay. - Boo.

00:38:03   - No, I think when you look at the design of VisionOS,

00:38:07   how it seems to be structured so far,

00:38:10   so first of all, I think people have,

00:38:11   I forget whether Apple said this

00:38:13   or whether people have analyzed it,

00:38:14   but it seems like the default physical,

00:38:16   like fixed focus distance of the Vision Pro

00:38:20   seems to be something like two meters in front of you,

00:38:22   so two to three meters in front of you, something like that.

00:38:24   If you look at the design of VisionOS,

00:38:26   when new windows are shown

00:38:28   or when the home screen icons are shown, et cetera,

00:38:30   they all seem to be shown at about that distance from you.

00:38:33   It seems like what VisionOS is doing

00:38:36   is kind of having a default distance from your face

00:38:39   that it renders things at,

00:38:40   and that is almost certainly

00:38:42   the actual accommodation distance

00:38:44   that the optics are designed to simulate.

00:38:45   And so if you leave things at that distance from your eyes,

00:38:50   that the default distance is placing things from you,

00:38:52   I'm guessing you don't get VAC,

00:38:54   or at least it's as minimized as it could be.

00:38:57   - Well, I mean, the problem in AR mode

00:38:59   is the whole rest of the room also exists,

00:39:00   and the wall is farther and the coffee table is closer,

00:39:03   and so if you do choose to look at them

00:39:05   and try to quote-unquote focus on them,

00:39:06   it will feel weird because--

00:39:08   - That's true, yeah. - You don't have to focus

00:39:10   on them because it's, you know,

00:39:11   that's part of the unreality of it,

00:39:13   like does it look like the room

00:39:14   or does it look like a screen?

00:39:15   Well, if you're looking at the room

00:39:16   and you're looking at the coffee table,

00:39:17   you will have vergence and accommodations

00:39:20   such that you can focus on the coffee table

00:39:21   that's a foot and a half and away.

00:39:23   In the headset, with a fixed focal distance, you won't.

00:39:26   It'll, you'll just put your eyes over there,

00:39:28   you'll point your eyes at it,

00:39:29   and it'll already be in focus,

00:39:31   except if your eyes try to adjust

00:39:33   with the expectation that it will not be, or anyway.

00:39:35   So, but yeah, you're totally right about where,

00:39:37   like even when you put the windows to the side,

00:39:40   it tilts them, it's trying to maintain that radius

00:39:42   of like this is the non-VAC region.

00:39:45   If you're anywhere in here, there is no conflict,

00:39:48   and anything that is closer or farther away

00:39:50   is potentially conflicting.

00:39:52   - Yeah, and I think, I'm not an expert in this,

00:39:54   I've just read some of the same articles

00:39:56   that you probably have, but it seems like this effect,

00:40:00   like the eye strain and discomfort that can result from VAC,

00:40:03   it seems like it's magnified more when stuff,

00:40:06   when you're holding or looking at stuff

00:40:07   that it's simulating to be too close to you,

00:40:09   rather than too far away.

00:40:10   Like it seems like the closer the thing is,

00:40:13   it magnifies the problem.

00:40:15   So it wouldn't surprise me if that's the case

00:40:18   to see VisionOS not give you lots of ways

00:40:21   to pull things that close to you,

00:40:23   and instead try to stick with stuff like more

00:40:26   at this distance and a little bit further from

00:40:28   or a little bit closer, but not a lot closer.

00:40:30   And there's all sorts of things that go into that too.

00:40:33   I kinda had this funny thought a minute ago,

00:40:34   like what if, like we keep seeing all these simulated rooms

00:40:39   in Apple's demos or simulator and everything,

00:40:41   like everyone has all these rooms

00:40:44   that have a good amount of empty space

00:40:46   a few meters ahead of where you're sitting.

00:40:48   And I was thinking like, if this actually becomes

00:40:51   like a really meaningful computing platform

00:40:54   that lots of people are using to get a lot of work done,

00:40:56   can you imagine people possibly starting to like rent out

00:40:59   offices that are just circles,

00:41:03   like the wall is just a circle

00:41:04   that is like a three meter radius around you?

00:41:07   - Just go in VR at that point.

00:41:10   Just turn on the Yosemite, you know what I mean?

00:41:12   - But then you could just,

00:41:15   if you're sitting inside your three meter donut room,

00:41:17   you could place anything on any of the walls.

00:41:20   You can even have like your coffee sitting three meters

00:41:22   away from you over here and like a picture of your family

00:41:25   sitting three meters away over there.

00:41:27   - That works with three meter long arms,

00:41:29   I don't know if that quite works.

00:41:30   You know, in the simulator,

00:41:32   I was playing with the Vision OS simulator in Xcode,

00:41:35   it just came out like, I don't know, hours before we recorded

00:41:38   so I didn't have that much time to play with it.

00:41:39   But you can fly the camera around

00:41:41   and one of the first things I did was I flew the camera

00:41:43   around to the side of a window to see if it had depth

00:41:46   and I'm pretty sure it doesn't.

00:41:47   Like I looked at it edge on and it basically disappeared.

00:41:50   The other thing I tried to do was take windows

00:41:52   and bring them right up to my nose.

00:41:54   And I think maybe I just don't know how to do that

00:41:56   in Vision OS, obviously everything in the simulator

00:41:58   is weird, have you ever tried to use an iPhone app

00:42:00   in the simulator, it's weird and this is 10 times weirder

00:42:02   because you're using like a mouse and a keyboard

00:42:04   and a scroll wheel to substitute for eyeballs and pinching,

00:42:07   it's super weird.

00:42:08   But I couldn't like, it wasn't conducive to that.

00:42:13   I could zoom in, I could zoom the camera in

00:42:16   but I was like just flying the virtual camera

00:42:18   like in the simulator, I don't think you'd be able

00:42:19   to fly the virtual camera that way in when you're wearing it

00:42:22   because when you're wearing it,

00:42:23   you'd have to push your head closer to the thing.

00:42:25   But this is another thing to try.

00:42:27   Like again, with the idea of the things,

00:42:29   the floating rectangles or whatever,

00:42:31   trying to look like they're positioned in space,

00:42:34   there's that button, whatever it is,

00:42:36   I forget which button it is,

00:42:36   whether it's the digital crown one or the other button

00:42:38   that like re-centers all your crap in front of you.

00:42:40   And I think if you were sitting on the couch

00:42:42   and had the windows in front of you at the intended,

00:42:44   you know, focal distance, and then just got up

00:42:47   and walked forward two steps and shoved your face

00:42:50   into Safari, it would let you get as close as you want,

00:42:53   but it would be useless because now you're,

00:42:55   now all you can see is three letters of the Safari window.

00:42:58   And if you press whatever button that is,

00:42:59   it would re-center things nine feet away from

00:43:01   or six feet away from where your head is now.

00:43:04   So that re-centering and re-distancing thing,

00:43:06   you know, makes sense.

00:43:07   And it's also why everyone that you see

00:43:08   in one of these demos, they're not using,

00:43:10   they're not like, again, they're not walking through

00:43:12   their house, walk and talk Aaron Sorkin style,

00:43:15   and using 17 apps while they walk and talk.

00:43:17   They're sitting on a couch, they're sitting in a chair,

00:43:19   they're stationary so that the things can be placed

00:43:21   into the real world with them,

00:43:23   but also maintaining that focal distance.

00:43:26   I don't know if there even is a mode where,

00:43:28   like, as you walk, your windows follow you, you know,

00:43:33   maintaining the six feet distance or whatever,

00:43:35   because you're probably walking to a wall

00:43:36   or do something terrible.

00:43:39   We are brought to you this week by Rocket Money,

00:43:42   a personal finance app that fines and cancels

00:43:44   your unwanted subscriptions, monitors your spending,

00:43:47   and helps you lower your bills all in one place.

00:43:49   'Cause do you know how much your subscriptions

00:43:51   really cost you?

00:43:52   Most Americans think they spend around 80 bucks a month

00:43:54   on subscriptions, but the actual total is closer to $200.

00:43:58   So if you don't know exactly how much you're spending

00:43:59   every month, you need Rocket Money.

00:44:02   Over 80% of people have subscriptions they forgot about.

00:44:05   Chances are you're one of them.

00:44:07   I was one of them, so I installed Rocket Money

00:44:09   when they sponsored our show, they said,

00:44:10   "Here, try it out."

00:44:11   I installed it, it instantly found not only one,

00:44:15   like I think four or five subscriptions

00:44:17   that I'd kind of forgotten about,

00:44:19   or something I had two copies of for me and Tiff

00:44:22   to both use it, we really only needed one now.

00:44:24   So there were a number of things where it found ways

00:44:27   for me to save money within an hour of installing it.

00:44:29   It was so fast, it was instantly paid for itself.

00:44:34   Rocket Money will quickly and easily find

00:44:36   your subscriptions for you, and for any

00:44:38   that you want to cancel, you just hit cancel,

00:44:41   and Rocket Money will cancel it for you.

00:44:43   It's that easy.

00:44:44   They also help you manage all your finances

00:44:46   all in one place, Rocket Money can automatically

00:44:48   categorize your expenses, you can easily track

00:44:50   your budget in real time, and also,

00:44:52   you get these cool email alerts if anything looks off.

00:44:54   So they've emailed me a few times since I set it up

00:44:56   to say, "Hey, this transaction looks a little unusual,"

00:44:59   or, "It's higher than it usually is,

00:45:01   "you might want to take a look at it."

00:45:02   It is a great app, over three million people

00:45:04   have used Rocket Money, saving the average person

00:45:06   up to $720 a year.

00:45:08   Stop throwing money away, cancel unwanted subscriptions,

00:45:12   and manage your expenses the easy way

00:45:14   by going to rocketmoney.com/atp.

00:45:18   That's rocketmoney.com/atp, rocketmoney.com/atp.

00:45:22   Thank you so much to Rocket Money for already saving me

00:45:27   a pretty good amount of money, and for sponsoring our show.

00:45:34   - So the Vision SDK is out, there is, at a glance,

00:45:38   some really frickin' good documentation.

00:45:42   I think hell may have frozen over, my friends,

00:45:46   because Apple's been like firing on all cylinders

00:45:49   about documentation, it's been very strange.

00:45:51   But we'll link to, dive into featured sample apps,

00:45:54   where they have a series of sample apps,

00:45:56   including Happy Beam, where you shoot hearts

00:45:59   at grumpy clouds, if I understand this properly.

00:46:01   Anyways, but it's a full-on walkthrough of the code.

00:46:05   Not only here's what we did, but why we did it.

00:46:08   Like, brav frickin' oh, no sarcasm.

00:46:12   This is excellent.

00:46:13   So I'm gonna have to dig into this at some point

00:46:15   when I'm not well over my head deep in call sheet stuff.

00:46:19   But yeah, this looks really, really good,

00:46:22   and it's worth checking out.

00:46:24   And then it's also worth noting, I think John had noticed,

00:46:27   Jonathan White, I believe former Apple employee,

00:46:30   pointed out that if you have a PlayStation 5

00:46:32   or Xbox controller hooked up to your Mac,

00:46:33   that you can use that to navigate

00:46:35   in the Vision OS simulator.

00:46:37   - Which is vastly preferable to trying to use

00:46:39   the tiny little camera controls they have

00:46:41   in a little floating pallet.

00:46:42   Again, it's very awkward to use,

00:46:43   to try to use a VR headset simulator

00:46:47   with the mouse and keyboard.

00:46:48   The main reason I put this post in here

00:46:50   is because he posted like an animated GIF

00:46:53   of like one of the little, you know,

00:46:55   they have virtual rooms for you to be in in the simulator,

00:46:57   'cause obviously there's no headset looking at your room,

00:46:59   so they have a bunch of virtual rooms for you to try.

00:47:01   And it's just a box floating in space.

00:47:03   And he takes the PlayStation controller or whatever

00:47:05   and just flies the camera in and out of the room

00:47:07   and you can see there's just a giant black void

00:47:09   with a living room sitting in it.

00:47:10   Neat.

00:47:12   - Going back a second, when Casey was talking about

00:47:13   all their good documentation of this,

00:47:15   which I'm very happy, I haven't seen it yet,

00:47:16   but I'm very happy to hear that, that's fantastic.

00:47:18   And then explaining why, you know,

00:47:20   I think what's very important here is that

00:47:23   they want us to develop apps for this platform

00:47:27   that fit and make sense and are good,

00:47:31   but we not only have never tried this platform,

00:47:33   but we also won't have a chance

00:47:35   to try this platform for a while.

00:47:38   - Well, that's half true.

00:47:39   Did you see, you might not have seen,

00:47:41   but I think it was Underscore that pointed out

00:47:43   that somewhere, I don't know where,

00:47:45   there was Apple documentation that said,

00:47:47   starting next month, you can do the lab thing

00:47:51   that they have at like five or 10 locations

00:47:53   around the world.

00:47:54   So that is, I mean, I still think your point, Marco,

00:47:58   is fair, but breaking news, apparently it is as early

00:48:01   as July, if we all read this right,

00:48:03   that one could go to Cupertino, London, et cetera, et cetera,

00:48:07   and actually try your stuff on Vision Pro.

00:48:09   - I mean, that helps, but that's a very limited,

00:48:13   like, you know-- - Oh, sure, totally.

00:48:14   - When you look, you know, when the iPhone App Store

00:48:17   came out, we had already had iPhones for a year.

00:48:20   - Yeah, yeah, yeah. - And we already,

00:48:21   and before that, we had had smartphones for a while.

00:48:23   Yeah, they sucked, but we had other, you know,

00:48:25   so like, the concept of a smartphone

00:48:26   and what smart software should be was fairly known,

00:48:30   and then we all had iPhones for a year,

00:48:32   and so we knew how iPhone software should behave

00:48:35   and how it should be designed, and yeah,

00:48:37   it was a learning process, you know,

00:48:38   we didn't get it all right right at the start,

00:48:40   but we had a huge head start.

00:48:42   When the iPad came out, we didn't have a chance

00:48:44   to use them before it launched to the world.

00:48:47   However, we kind of assumed, like,

00:48:49   this is probably like a giant phone,

00:48:51   and just kind of use that as a starting point,

00:48:53   and yeah, our initial apps sucked,

00:48:55   but they weren't that far off, you know,

00:48:58   and again, it was a learning process.

00:49:00   With this, every other major, you know,

00:49:04   VR, AR kind of headset that's had any success at all,

00:49:07   besides like very, very specialized, you know,

00:49:10   narrow markets, has been gaming focused,

00:49:13   and so we kind of have some idea,

00:49:15   like hey, if you wanna make a VR game,

00:49:16   here are some principles that you might wanna follow,

00:49:18   or here are some things that work and don't work.

00:49:20   That's fine, but the idea of a spatial computing environment

00:49:25   of like using an AR environment

00:49:27   for general purpose computing,

00:49:29   that is still really in its infancy

00:49:32   in the rest of the world, and most of us have not used that,

00:49:35   and none of us have used specifically the Vision Pro

00:49:38   for this purpose yet, and so to try to make software

00:49:40   for this platform that we have no experience with,

00:49:45   and that can't really be simulated very well,

00:49:49   and even if it was, you know,

00:49:51   even if the simulator was really good,

00:49:54   and this could get into like, you know,

00:49:55   they said they're gonna offer dev kits at some point,

00:49:57   and you know, we mentioned like we don't really know

00:50:00   what that means or what the dev kits will be able to do,

00:50:03   but if the dev kit is anything less

00:50:06   than a full-blown environment that has all of Apple's

00:50:10   built-in apps already working and everything,

00:50:11   which it probably will be, like I can't imagine

00:50:13   that those will all be ready yet, you know,

00:50:16   I expect the dev kit to be basically like the simulator,

00:50:19   in terms of what it has and what it doesn't have,

00:50:22   like in terms of other apps and everything,

00:50:23   and you're not gonna be able to like install stuff

00:50:24   really, besides your own app, so you're not gonna get

00:50:27   a good experience of like, what it's like to actually

00:50:30   really use this thing for computing,

00:50:33   and because it's so radically different

00:50:36   from what we've used before, Apple needs to really step up

00:50:41   above and beyond to, with things like documentation

00:50:45   and tooling and you know, explaining the, you know,

00:50:48   human interface guidelines, explaining why they're doing

00:50:51   things, explaining how things should be designed,

00:50:53   why things should be structured this way,

00:50:56   because we don't know, because we can't use this platform

00:50:58   yet, and the good thing is, it sounds like that's

00:51:00   what they're doing, and it's gonna take a whole lot

00:51:03   of like immersing ourself in this documentation

00:51:06   and in the simulator and in the tools and reading up

00:51:09   and experimenting with things for us to get stuff right,

00:51:11   and it's still not gonna quite be right when we get it,

00:51:14   because we will have no experience actually really using

00:51:17   the hardware and using the software and figuring out

00:51:21   how things work and how things should behave,

00:51:23   how things should feel, how things should look,

00:51:25   you know, there's so much of that.

00:51:27   When you look at the iPhone, there's so much kind of

00:51:30   built in, like, you know, here's how things should be

00:51:32   laid out on the screen, here's how navigation works,

00:51:34   here's how things should be structured,

00:51:36   here's the different conventions that we use.

00:51:39   We know none of that with VisionOS, and until we're

00:51:42   actually able to use it with real apps, with our real data,

00:51:46   with, you know, really in our actual world trying to get

00:51:48   things done in our real lives, we're not gonna have

00:51:51   those feelings kind of naturally inherent to us.

00:51:54   So it's gonna be a learning process, but in the meantime,

00:51:58   Apple really has to step in, and so I'm very glad

00:52:00   that it seems like they are doing that.

00:52:03   - Yeah, very much so.

00:52:04   All right, John, tell me about the John Turnus external

00:52:08   GPU question that was surfaced at the talk show.

00:52:11   - Yeah, we talked about it last week, I think.

00:52:14   I asked about external GPUs, and Turnus said, like,

00:52:17   he doesn't quite see how they would incorporate that

00:52:20   into their system of built-in GPUs or whatever,

00:52:22   it was kind of closing the door on external GPUs,

00:52:25   and also saying that it just didn't even seem

00:52:27   to make sense to him.

00:52:28   It occurs to me, thinking about that question and other,

00:52:32   you know, and the talk show in general, that maybe that

00:52:36   is also a question that could have been asked to Craig

00:52:38   Federighi, because Turnus is the hardware dude, right,

00:52:41   and you think, oh, external GPUs, like, that's a question

00:52:43   for him, right, because if he's like, oh, you're gonna make

00:52:45   a Mac Pro, are we gonna support external GPUs or not,

00:52:48   and how can we incorporate those,

00:52:49   that's the hardware guy, right?

00:52:51   But there is a software component to this.

00:52:55   As far as I'm aware, there's nothing inherent about any

00:52:59   of the M whatever SOCs that precludes the idea of,

00:53:03   for example, using an external GPU for compute,

00:53:07   like setting aside the video stuff, I don't know what

00:53:09   the deal with that is, but just like, just doing it

00:53:11   for compute if you're trying to do, like, you know,

00:53:13   AI model training or something, like using GPUs for a

00:53:15   compute or whatever, especially in the Mac Pro,

00:53:18   where you actually have card slots and all that other thing,

00:53:19   right, so the hardware may be capable, but you can't

00:53:23   actually use any of those GPUs to do anything unless

00:53:27   you have a driver for it, and that is a software question

00:53:31   that Turnus doesn't really control, that's on the

00:53:33   Federighi side, it's like, hey, are we going to support

00:53:37   on Apple Silicon, you know, Apple Silicon native drivers

00:53:41   for, let's say, AMD graphics cards, and it's not like,

00:53:44   you know, Craig Federighi, like his organization maybe

00:53:47   writes all those drivers or whatever, but certainly it falls

00:53:49   under the umbrella of what kind of things does Mac OS

00:53:51   support, because, you know, in my Mac right now, I've got

00:53:55   a bunch of drivers for my AMD graphics cards, and they

00:53:57   came with the operating system, and the operating system,

00:54:01   that's Craig, so it would have been interesting to pose

00:54:04   the same question to, obviously, you know, not gonna make

00:54:06   the whole talk show just to be the giant Mac Pro show,

00:54:09   that would be what it would be like if I was hosting it.

00:54:11   - I was gonna say.

00:54:12   - But, and that's why Apple does not give you executives

00:54:15   to interview.

00:54:16   - Yeah, but anyway, that's, you know, it's just an

00:54:19   interesting thing to think about, that there's more to it

00:54:22   than just like, oh, you know, when you design the Mac Pro

00:54:24   hardware, you didn't make it support this, or you're,

00:54:26   you know, because the SOC has the GPU on it, it can't do

00:54:28   that, as I think we've talked about many times in the past,

00:54:31   Intel GPUs, Intel CPUs had integrated graphics for years

00:54:35   and years, and Macs with Intel CPUs with integrated

00:54:39   graphics also supported discrete graphics, in fact,

00:54:41   Apple had this wacky thing in the OS where it would use

00:54:44   the discrete versus using the integrated and go back and

00:54:46   forth and all that stuff, and of course, in the Intel Mac

00:54:48   Pro, actually the Xeon I don't think does have an

00:54:50   integrated GPU, but anyway, you could have put, like,

00:54:53   instead of a Xeon, put like an i9 or whatever, or i7,

00:54:56   or whatever the most recent one, you know, put something

00:54:57   with an integrated GPU and still use the external GPUs.

00:55:01   Apple did all that work for Intel, but that work is a

00:55:04   software work, that is not hardware work, so that would be

00:55:08   an interesting thing to muse on as we go in the next round

00:55:12   of the Mac Pro to see if Apple changes course on this,

00:55:16   to see if it would even be technically possible to change

00:55:18   course, for example, with just a new version of macOS

00:55:21   with a bunch of drivers that suddenly let you plug in,

00:55:24   you know, GPUs, and I don't think it's gonna happen,

00:55:26   obviously, on the M2 Ultra Mac Pro because all of their

00:55:29   slots have, are banned with Star, and they would need

00:55:32   external power connectors to power a big GPU and yada, yada,

00:55:35   yada, but I was just thinking about this, that it, you know,

00:55:39   it's not just a hardware question, it's also a software

00:55:41   question.

00:55:42   - That's fair.

00:55:43   All right, Apple has added passkeys to Apple ID and iCloud

00:55:47   logins, so they are dogfooding this, or allowing us, I guess,

00:55:51   to dogfood it.

00:55:52   I haven't really messed with passkeys at all, so I don't

00:55:56   really have much to say about this, but I applaud Apple

00:55:58   for actually making it possible.

00:56:01   - They say this from a Six Color story, they said that

00:56:03   the future has been rolling out as of yesterday and can be

00:56:05   tested on devices running iOS, iPadOS 17, or Sonoma betas,

00:56:10   and some other people have said if you run, if you're running

00:56:13   Ventura and you use Chrome, because Chrome, the latest

00:56:15   versions of Chrome know about passkeys, that you could

00:56:17   actually do it on Ventura with Chrome.

00:56:20   - That's funny.

00:56:20   - But they say it's rolling out, and that must, it must not

00:56:23   have rolled out to me because I tried for a while earlier

00:56:25   today to get to the point where it would let me use a passkey

00:56:28   or prompt me, like by going to the iCloud login page, and I

00:56:31   just don't have access to that feature yet, so it's coming.

00:56:35   Eventually, but I wasn't able to try it.

00:56:37   I have used passkeys with other things, like in Chrome,

00:56:40   because again, Chrome supports them, but with other services

00:56:42   that support passkeys, and it's fun, and it's nice.

00:56:44   I use them in addition to names and passwords.

00:56:47   I haven't yet had the guts to take any of the services that

00:56:49   support them and say, you know what, forget about my

00:56:52   password and do everything with passkeys.

00:56:53   Although honestly, I shouldn't really, it shouldn't really

00:56:56   be that scary because, you know, we all know how the world

00:57:00   works, like oh, well what if passkeys break or don't work

00:57:03   or whatever, it's exactly the same thing you do when you

00:57:06   forget your password, which happens to all of us.

00:57:08   In the end, your email address is the ultimate key to your

00:57:10   stuff because people forget their passwords.

00:57:11   You'll never forget your passkey, but say there was a bug

00:57:14   and the computer screwed up and the computer, you know,

00:57:16   forgot your, quote unquote, forgot your passkey.

00:57:19   You're just gonna use the forgot password link and have

00:57:22   them email you something, yada yada, which is not great,

00:57:25   but in theory, once the passkey feature is worked out and

00:57:28   all the bugs are ironed out and it basically works reliably,

00:57:32   you don't have to worry about forgetting anymore and I think

00:57:34   humans forgetting is going to happen way more often than

00:57:37   iCloud Keychain, like deleting stuff accidentally.

00:57:40   - Tell me, with regard to passkeys and passwords and things

00:57:44   like that, tell me what's going on in Sonoma.

00:57:46   - Yeah, so there's an app on Mac OS that's been around for,

00:57:49   I don't know, a decade or two, called Keychain Access,

00:57:52   that gives you a view of your Mac OS keychain.

00:57:55   It started off looking at just your local keychain and

00:57:57   eventually iCloud Keychain was introduced and they put that

00:57:59   into the app.

00:58:00   It's a pretty nerdy app, it hasn't been updated in a really

00:58:02   long time, it's scary and confusing and in general, people

00:58:05   should not mess with it because it's very easy to screw

00:58:07   yourself up because you'll say, I don't know what this is,

00:58:10   do I need this?

00:58:11   I'm gonna delete this.

00:58:12   Sometimes you need it.

00:58:13   (laughing)

00:58:14   Don't, like, don't, it's like people used to go into a

00:58:16   library folder on the early versions of Mac OS X being like,

00:58:19   library, what the hell is this crap?

00:58:20   And just delete everything.

00:58:21   It's not a good idea.

00:58:23   So on Sonoma, when you launch Keychain Access, it throws a

00:58:26   dialogue in your face that says, manage your passwords in

00:58:29   system settings.

00:58:30   Go to passwords in system settings to manage your passwords

00:58:33   in pass keys, set up verification codes and view your

00:58:35   security recommendations.

00:58:36   And it has a big button that is the default button that says

00:58:38   open passwords and the non-default button is open

00:58:42   Keychain Access with a checkbox to say don't show this

00:58:44   message again.

00:58:45   So for users who are used to using Keychain Access because

00:58:48   they're nerdy and that's the place where they go for

00:58:50   passwords, they're trying to tell them, hey, we have a

00:58:53   significantly more friendly interface to passwords over here

00:58:56   in system settings that you might not know about because

00:58:57   it's buried.

00:58:59   Please use that one instead.

00:59:00   If only that friendlier interface to passwords could be a

00:59:05   standalone application that could replace Keychain Access.

00:59:08   Maybe not replace it, 'cause Keychain Access does much more

00:59:10   fancier stuff.

00:59:11   So I hope they keep Keychain Access around, but I do like

00:59:14   the idea that they are trying to herd users towards the

00:59:17   nicer friendlier interface that is in a totally

00:59:19   different place.

00:59:20   - Also, can I take this moment to thank the Xcode team or

00:59:26   whatever part of the Xcode team is responsible for

00:59:28   automatic certificate management and in particular,

00:59:31   whoever at Apple made that work for CarPlay apps as of a

00:59:34   few years ago or a year ago, I have not had to go into

00:59:38   Keychain Access for probably two or three years now.

00:59:42   I used to on a regular basis because I used to have to do

00:59:44   all the provisioning certificates and everything and all

00:59:46   that management and everything and inevitably something

00:59:49   would mess up and I'd have to go in there and clean up

00:59:51   some garbage and I haven't had to do that in like two years.

00:59:54   - Nice.

00:59:55   - So thank you, whoever did that, thank you.

00:59:58   - Yeah, go team.

00:59:59   John, there's good news about Notification Center,

01:00:01   we're being told anonymously.

01:00:03   - Yeah, like I said last week, two things I had to say

01:00:05   about Sonoma was that my weird bug seems to be fixed,

01:00:09   but also that the notification interface actually has

01:00:12   buttons that you can click that don't disappear when you

01:00:14   try to click them.

01:00:15   So here's some anonymous feedback on that in Sonoma.

01:00:18   Notification Center has been significantly re-engineered

01:00:21   for macOS Sonoma, probably in big part because it's

01:00:23   responsible for hosting widgets, which can now be added

01:00:25   to the desktop, so it wouldn't surprise me that the

01:00:27   notification hover bug is finally fixed.

01:00:29   Before Sonoma, the Notification Center process would be

01:00:31   responsible for all things related to widgets, including

01:00:34   when to update them, the running of extensions and

01:00:36   archiving of views.

01:00:38   With Sonoma, they brought ChronoD, the Chrono demon,

01:00:42   to macOS, so now all Notification Center has to worry about

01:00:45   is actually placing views on the screen.

01:00:47   So here's to rewriting a part of the system that had been

01:00:49   broken for three years.

01:00:51   I don't know what ChronoD, I think it's some internal thing

01:00:54   or whatever, but this matches with my experience.

01:00:58   The Notification Center, the interface looks the same.

01:01:00   It's the same stupid, like, oh, we can't show you the

01:01:02   buttons until you mouse over them because we're afraid your

01:01:04   little brain will explode.

01:01:05   But now, when you mouse over them, you can actually

01:01:07   click on them.

01:01:08   So there is some support for the idea that it's not just my

01:01:11   imagination that they actually made this particular user

01:01:15   interface work.

01:01:16   Now we can get back to complaining that this

01:01:17   particular interface is dumb, but at least it's dumb and

01:01:20   working as designed.

01:01:21   Hurray!

01:01:23   All right, there's a couple of tidbits that we wanted to

01:01:25   share with regard to iOS 17.

01:01:29   First of all, if you recall, Apple has done for the last

01:01:32   year or two a semi-private Slack for WWDC.

01:01:37   I was going to say attendees, but that's not really true.

01:01:39   People who are interested in WWDC.

01:01:41   And in there, somebody caught, and then it was eventually

01:01:45   posted to MacRumors, that somebody from Apple said, as

01:01:50   of iOS 17 and MacOS Sonoma, disabling the iCloud Drive

01:01:54   switch will no longer disable syncing in your app.

01:01:58   It'll be controlled by the individual switch for your

01:02:01   app, which means, Marco, that you can still sync with

01:02:06   CloudKit and iCloud stuff, even if iCloud Drive is turned

01:02:09   off because of some sort of work provisioning.

01:02:11   This is critical for Marco today, but not a small amount

01:02:15   of call sheet also rides on CloudKit.

01:02:17   And so this is critical for me soon.

01:02:19   So I am very excited about this.

01:02:22   This is very good news.

01:02:23   Yeah, I'm very curious to see, once I get running on 17 and

01:02:28   get some real user data out there, because right now, as a

01:02:32   quick reminder, and I just verified, if I look at my

01:02:38   CloudKit state of whether I can save data in CloudKit, I

01:02:42   track this in Analytics so I can make a decision in the

01:02:44   future about whether to switch to that for sync.

01:02:46   And right now, I got 12 and 1/2% of my user base saying no

01:02:50   account.

01:02:51   And I believe that's the result you would get in iOS 16

01:02:55   and below when iCloud Drive was just disabled, in addition

01:02:58   to not actually being logged into an iCloud account.

01:03:01   And in that case, you can't use CloudKit.

01:03:03   So right now, I'm at 12 and 1/2%.

01:03:05   As my iOS 17 adoption goes up, which is already at 2%, as

01:03:12   that goes up, I'm curious to see if this number goes down.

01:03:15   And my analytics system is not advanced enough that I could

01:03:18   say, what is this value just for iOS 17 people?

01:03:22   I'm kind of curious to see how that tracks.

01:03:24   Maybe I'll add something separate for that.

01:03:25   But we'll see.

01:03:28   Right now, this doesn't help me, because that's still way

01:03:31   too many people to not be able to use iCloud

01:03:34   with iOS 16 and below.

01:03:36   And maybe in a year from now, that'll be different.

01:03:39   Right now, I still can't use this.

01:03:42   I can't rely on iCloud Drive being universal enough, or

01:03:46   rather, CloudKit being universal enough.

01:03:48   But maybe next year, I will.

01:03:50   We'll see.

01:03:51   I hope so.

01:03:52   Either way, this is a very, very important step towards

01:03:56   that possibility that might enable this

01:03:58   possibility next year.

01:03:59   So looking forward to that.

01:04:01   And also, with regard to iOS 17, apparently there's now a

01:04:05   crossword puzzle, like MiniApp, within Apple News+.

01:04:09   I don't even remember hearing rumblings of this until today.

01:04:11   Maybe I missed it.

01:04:12   But starting with iOS 17, Apple's taking a page from the

01:04:15   New York Times and integrating crossword puzzles into the

01:04:17   News app.

01:04:18   And puzzles will be available to Apple News+ and Apple One

01:04:21   subscribers.

01:04:22   This is from 9to5Mac.

01:04:23   We'll link it in the show notes.

01:04:25   Again, news to me.

01:04:27   It makes sense to be in the News app.

01:04:28   But this is just sort of raising the bar for

01:04:30   everybody else.

01:04:31   Like, we're complaining when Apple adds functionality to be

01:04:34   built into the--

01:04:35   if they make a News app, and part of News is the New York

01:04:37   Times, and part of the New York Times is the New York

01:04:39   Times crossword, you can see how they got there.

01:04:41   I still think there's plenty of room for better third-party

01:04:43   crossword apps.

01:04:44   But it just goes to show that you never know when Apple's

01:04:47   going to come for your little section of the market.

01:04:49   That being said, if it's limited to News+ subscribers,

01:04:53   I'm not sure that's a huge threat to the market.

01:04:57   But we'll see.

01:04:59   Yeah, I mean, again, there's third-party opportunity there

01:05:02   for sure, because A, you got all the free crossword apps.

01:05:05   And B, what if people don't want Apple News or Apple One?

01:05:08   They just want crosswords.

01:05:09   Well, they just go to the App Store and find one.

01:05:12   And Marco, we have news with regard to why Apple isn't

01:05:17   allowing third-party watch faces in WatchOS 10.

01:05:19   Do we--

01:05:20   This is as per-- well, not really.

01:05:21   Does this actually say anything?

01:05:23   No.

01:05:24   It does say something, but what it says is BS.

01:05:27   As per MacRumors, in an interview with Swiss newspaper

01:05:31   something or other published today, Apple's VP of

01:05:33   Technology, Kevin Lynch, and product marketing employee

01:05:35   Deirdre Kaldbeck explained--

01:05:37   and these quotes are machine translated from German--

01:05:40   Lynch said that Apple puts, quote, "a huge amount of

01:05:42   effort," quote, "into every watch face to ensure that they

01:05:45   work," quote, "uniformly and simply," quote, and said,

01:05:48   "Apple needs to plan ahead to make sure watch faces continue

01:05:50   to work if we want to change something or add new

01:05:52   possibilities." This is the most non-answer answer I've

01:05:56   read in a little while.

01:05:57   But what are you going to do?

01:05:58   Because you know Apple would never change how someone

01:06:01   integrates into their operating system, like, say, a different

01:06:03   system for writing widgets or an entirely different system

01:06:05   for writing apps for the watch.

01:06:08   This is-- yeah, so we don't want to do this because what

01:06:12   if we change our mind about how the watch faces work?

01:06:14   You change your mind about things all the time.

01:06:16   That's what WWDC is about.

01:06:18   That's what we're all doing when new OSes come out.

01:06:20   It's like, we can't give you watch faces because what if we

01:06:22   change our mind about the API?

01:06:25   Nothing has changed more than going from--

01:06:27   what was the previous one, the crappy one?

01:06:29   Is it WatchKit to non-WatchKit?

01:06:31   I don't remember.

01:06:32   They change stuff all the time on every OS.

01:06:34   That's the nature of what they do as a platform owner.

01:06:36   So that is not the reason why watch faces don't exist.

01:06:40   We have a huge-- put a huge amount of effort.

01:06:42   We want them to work uniformly and simple.

01:06:44   You can use this as an excuse for why you don't want to have

01:06:46   the App Store as well.

01:06:48   This is-- I mean, this may be the reason why they're not

01:06:51   doing it, which is basically saying, oh, it would be more

01:06:53   work if we did it.

01:06:54   I agree.

01:06:54   It would be more work, just like it's more work to run the

01:06:56   App Store.

01:06:57   But geez, please, third party watch faces.

01:07:00   First of all--

01:07:01   yes, I know this is a machine translation.

01:07:03   But first of all, Apple's own faces don't work

01:07:06   uniformly or simply.

01:07:08   So let's rule that out right now.

01:07:10   They don't.

01:07:11   So that's not a good reason.

01:07:12   But also, everything that we've seen from the way

01:07:17   widgets are rendered and the way Swift UI works suggests

01:07:20   that they could do a really good watch face kit using the

01:07:26   widget rendering system, where your process isn't even needing

01:07:29   to run all the time, just like complications,

01:07:31   just like widgets.

01:07:33   You can kind of use pieces.

01:07:35   And they could give you like, here's clock hands for the

01:07:38   current time, or here's a digital version of the time,

01:07:40   or whatever.

01:07:41   They can give you components.

01:07:42   And you can place them in your view and lay them out however

01:07:45   you need to and style them with a few custom ways.

01:07:49   They could do that so easily with what they've already

01:07:52   built and what they're already shipping, not even just brand

01:07:56   new this year, but what they've already been shipping

01:07:57   for a few years.

01:07:58   So there really isn't any technical

01:08:01   merit to this argument.

01:08:03   The reason we don't have third party watch faces yet is

01:08:06   because they don't want to.

01:08:08   That's it.

01:08:09   There is no technical justification.

01:08:11   There is no good argument.

01:08:12   It's simply they don't want to.

01:08:15   Apple has a new system for installing betas.

01:08:18   Anyone with a free Apple developer account can do it now.

01:08:21   You don't need to pay the $99 like you used to.

01:08:24   There's a write up in Ars Technico about this.

01:08:26   Basically, you go fiddling in settings and select which beta

01:08:29   train you want to be on.

01:08:31   And then you reboot, and you're on that train, which is

01:08:33   kind of neat.

01:08:33   It used to be able to go to the developer website and

01:08:35   download an installer, basically.

01:08:37   And you could still--

01:08:38   I think you can get a restore image for Apple Silicon only

01:08:41   to do that.

01:08:42   But that's not what they want you to do.

01:08:44   They want you to go to software update, which is

01:08:47   buried in system settings.

01:08:49   And you got to know to click on the little i in a circle.

01:08:51   Everything's about the little i in a circle.

01:08:55   That lets you pick, hey, I want beta updates.

01:08:57   And importantly, it lets you pick which Apple ID you want

01:09:01   to use as the Apple ID that has a developer account that is

01:09:05   being used to download the thing.

01:09:06   It doesn't have to be a paid developer account anymore.

01:09:08   So you don't have to pay $99 to get access to the betas.

01:09:11   You can have a free developer account.

01:09:12   But the question is, if you're like us old school Mac users

01:09:16   or developers, we have multiple Apple IDs.

01:09:18   Which Apple ID is the one that has a developer account?

01:09:22   And you want to configure that.

01:09:23   Because if you don't, if you're trying to use your real

01:09:26   Apple ID with a beta, but another Apple ID has a

01:09:31   developer account, you'll end up in your other developer ID

01:09:34   in the beta.

01:09:35   You'll be like, but this isn't my Apple ID.

01:09:37   I want to be in my real Apple ID.

01:09:38   But oh, I have to stay in this Apple ID because that's the

01:09:40   one that gets the updates from the beta.

01:09:42   They're two separate things.

01:09:43   You just have to-- again, it's the little tiny i in the

01:09:45   circle, and the stupid system setting interface lets you

01:09:47   pick which Apple ID you want to be the one

01:09:50   that gets the betas.

01:09:51   And that doesn't have to be the same Apple ID as the one

01:09:53   you're actually signed into in Mac OS or whatever.

01:09:55   So I don't know how this works outside of Mac OS, because

01:09:58   I've only done the Mac OS beta.

01:09:59   But I think the new system is a little bit confusing, but

01:10:03   it's better than installing dev profiles,

01:10:05   which is one option.

01:10:06   I think it's maybe not better than downloading installers,

01:10:09   but we'll see.

01:10:10   But just FYI, I don't feel like you're trapped to actually

01:10:13   using your developer Apple ID as your user's Apple ID.

01:10:17   - Indeed.

01:10:18   And then finally, very, very good news.

01:10:21   DP Review was purchased by Gear Patrol.

01:10:23   So they are not going away into the sunset.

01:10:26   They will live on, which is great news.

01:10:29   - Well, it was kind of a shame that they will live on after

01:10:32   like half of their employees bail to get new jobs, which who

01:10:34   could blame them.

01:10:36   Obviously, the price for Gear Patrol is probably lower if

01:10:39   you wait until after the site is declared to be shut down and

01:10:43   half the employees leave.

01:10:44   But they're also getting a less worthy asset, less valuable

01:10:48   asset, because some of the good people who are at DP Review

01:10:51   are no longer there.

01:10:53   All in all, probably not handled the best, but it's

01:10:56   better than it just actually literally dying inside Amazon

01:10:59   and nothing coming of it.

01:11:00   So I'm rooting for its resurrection.

01:11:02   I do wonder if Gear Patrol is actually dedicated to it.

01:11:06   Will they try to hire back some of the people who left to

01:11:09   go elsewhere?

01:11:10   We'll see.

01:11:10   But apparently it will live on to fight another day.

01:11:14   - OK, so Safari 17 has a bunch of new features, a plethora

01:11:19   even, of new features.

01:11:20   But one in particular is very interesting, link tracking

01:11:23   protection.

01:11:24   So my executive summary of this is, you know how you get

01:11:28   those really heinously long URLs whenever you share

01:11:32   something from Instagram or Facebook or something or

01:11:35   Twitter or something like that.

01:11:36   Well, apparently Safari 17 will do its best to strip out

01:11:40   the query string entries that are clogging up those URLs and

01:11:43   leave only the ones behind that matter.

01:11:45   And I haven't tried this myself, but this sounds good.

01:11:49   I'm here for it.

01:11:50   - Like all these blocking and prevention things, it really

01:11:52   depends on how good a job it does.

01:11:54   So it's deleting stuff from your URLs.

01:11:57   So you take a URL and it's going to say like, no, some of

01:11:59   the stuff in this URL we think is being used to track you, so

01:12:01   we're just going to delete it.

01:12:02   Some websites are surely designed so that if you don't

01:12:05   have the tracking information, they don't work right.

01:12:07   And that's crappy.

01:12:08   But other times, I do worry that this thing is going to

01:12:11   delete something from the URL that it thinks is a tracking

01:12:13   parameter but is not.

01:12:15   Because the only way it has to tell is based on the names.

01:12:18   Apparently, Jeff Johnson found that there's a queryparam.wp

01:12:21   list file that contains a list of the URL query parameters

01:12:24   to be removed.

01:12:26   And it currently contains 25 of the usual suspects.

01:12:29   And he lists a bunch of query parameter names.

01:12:33   I do wonder if it can get specific, like, oh, and this

01:12:35   website, this parameter, or if it just looks for any query

01:12:38   parameter named GCLID, which is some Google

01:12:43   thing, apparently.

01:12:44   But yeah, I hate it when ad blocking, privacy, whatever

01:12:48   things cause a website not to work.

01:12:50   Because you're trying to do a thing on the web.

01:12:52   You're trying to order dinner.

01:12:54   You're trying to buy movie tickets.

01:12:55   You're trying to book a reservation.

01:12:58   You click a button to do a thing, and nothing happens.

01:13:02   You're like, is this web page broken?

01:13:04   Is one of my blockers screwing it up?

01:13:06   Same thing when someone sends you a URL and messages.

01:13:09   You click it.

01:13:09   You try to load the thing.

01:13:10   It says, I tried to load it, and it didn't work.

01:13:11   And they're like, well, it works for me.

01:13:13   And I'm like, well, it doesn't work for me.

01:13:14   And little do you know that Safari 17 stripped out a query

01:13:17   parameter that it thought it was for tracking.

01:13:19   And you can always turn off content blockers and blah,

01:13:21   blah, blah.

01:13:21   And I hate that dance.

01:13:23   But I also hate tracking stuff.

01:13:25   So I applaud the effort.

01:13:27   But this is an extremely hard problem,

01:13:28   especially since you have so little information.

01:13:31   You've got a string.

01:13:32   It's shaped as a URL.

01:13:33   It's got query parameters.

01:13:35   You know some names are commonly used for tracking.

01:13:37   Is it OK to delete this?

01:13:39   I don't know.

01:13:39   Let's try it.

01:13:40   So I'm a little bit nervous about this feature.

01:13:42   But I'm sure you'll be able to disable it

01:13:44   if you don't like it.

01:13:46   Indeed.

01:13:46   I mean, I agree with everything you said.

01:13:48   But I'm excited.

01:13:49   This hopefully will work really well.

01:13:52   DocKit is a new thing that is coming with, I guess, iOS 17.

01:13:57   There's a WWDC session about it called

01:13:59   Integrate with Motorized iPhone Stands Using DocKit.

01:14:02   And the description begins, "Discover

01:14:04   how DocKit can help your camera app create incredible photo

01:14:06   and video experiences with motorized stands.

01:14:09   We'll show you how your app can automatically track subjects

01:14:11   in live video across a 360-degree field of view,

01:14:14   take direct control of the stand, and blah, blah, blah."

01:14:16   So this sounds pretty cool.

01:14:18   I haven't looked at the session yet.

01:14:20   This is one of those things where I don't personally

01:14:22   think I have a need for this at the moment.

01:14:24   But I could totally see how you could make, like,

01:14:26   an incredible sunrise video with you panning as the sun is

01:14:30   rising or something like that.

01:14:31   I think this could be very, very, very cool.

01:14:34   The integration of this is exactly the thing

01:14:36   they showed in the keynote, which is, hey,

01:14:38   you can use continuity camera and FaceTime with Apple TV now.

01:14:42   And it'll follow you around.

01:14:43   Yeah.

01:14:43   So you're putting your camera by your TV.

01:14:46   I mean, imagine if Apple made a TV set.

01:14:49   But of course, the only tool it has to use

01:14:51   is the center stage, which is just, like, oh,

01:14:53   big, wide, fisheye lens.

01:14:54   And then we crop a portion of it.

01:14:55   But now, if you had one of these docket things integrated

01:14:58   with FaceTime, integrated with Apple TV,

01:15:00   and you had a bunch of kids in front of you

01:15:02   on the couch squirming around, it

01:15:03   could actually turn the camera and use, like, the good camera

01:15:06   and change how the picture is framed by literally changing

01:15:09   how the picture is framed.

01:15:10   Imagine if the docket didn't just rotate but also tilt it

01:15:14   and everything.

01:15:16   This is working towards the idea of just everyone

01:15:18   sits down in front of the TV.

01:15:20   And you can just do FaceTime with the whole family.

01:15:22   And it just makes sure everyone's in the frame

01:15:23   and zooms in on people when they talk.

01:15:25   And it all looks good.

01:15:26   This is a piecemeal way to do it,

01:15:29   supporting continuity camera, having an API

01:15:31   for these third-party things.

01:15:33   Could you actually find a TV that

01:15:35   supports Apple TV, that supports docket, that supports FaceTime,

01:15:38   that supports continuity camera, and put this all together

01:15:40   and in your living room with your iPhone

01:15:43   and your set of cameras so that it works?

01:15:45   Who knows?

01:15:45   It's not exactly an integrated Apple solution.

01:15:47   But it's more of the type of thing that people always

01:15:49   say they want from Apple, which is like a geeky-type endeavor,

01:15:52   where all the pieces are there to do something really cool.

01:15:55   You just got to assemble them yourself.

01:15:58   No, this sounds very, very neat.

01:16:00   And just to be clear, I don't know

01:16:02   what it sounded like when I said it.

01:16:03   But it sounded like Jon was saying docket.

01:16:06   It is dock-kit.

01:16:07   And I might have made the same mistake when I introduced this.

01:16:10   So just to be absolutely clear.

01:16:11   It's two Ks in a row.

01:16:12   Yes.

01:16:12   It's not hover.

01:16:14   It's hover, apparently.

01:16:16   Don't start with that again.

01:16:19   We are brought to you this week by Squarespace,

01:16:22   the all-in-one platform for building your brand

01:16:24   and growing your business online.

01:16:26   Stand out with a beautiful website,

01:16:28   engage with your audience, and sell anything-- your products,

01:16:31   content you create, even your time.

01:16:33   Squarespace makes it so easy to make any kind of website.

01:16:36   And they have made the most amazing business features

01:16:39   out there.

01:16:39   So Squarespace makes it easy for, for example, member areas.

01:16:42   Creators can monetize your content and expertise

01:16:44   in whatever way fits your brand.

01:16:46   You can unlock a new revenue stream for your business,

01:16:48   free up time in your schedule by selling access

01:16:50   to gated content like videos, online courses, or newsletters.

01:16:53   They support any kind of online store, physical goods,

01:16:56   digital products.

01:16:57   No matter what it is, they have all the tools

01:16:59   you need at Squarespace to start selling online.

01:17:01   You can even use analytics to help grow your business,

01:17:04   learn where your site visits and sales are coming from,

01:17:06   analyze what your marketing channels are most effective.

01:17:08   You can build a marketing strategy

01:17:10   based on your top keywords or your most popular products

01:17:12   and content.

01:17:13   Whatever you are selling or whatever

01:17:14   you're making a website for, Squarespace has you covered.

01:17:17   And it's all visual.

01:17:18   All what you see is what you get.

01:17:20   You're never writing code if you don't want to.

01:17:22   You never have to worry about patches or upgrades or security

01:17:25   updates or anything like that.

01:17:27   All that's taken care of for you at Squarespace.

01:17:29   And you can just focus on whatever it is you're doing.

01:17:31   They have email newsletter support now, too.

01:17:33   Squarespace email campaigns, so many tools

01:17:36   built into Squarespace.

01:17:37   It's amazing.

01:17:38   See for yourself by starting a free trial

01:17:41   at squarespace.com/atp.

01:17:43   There's no credit card required.

01:17:45   You can build the entire site in trial mode

01:17:47   and see if Squarespace is a good fit for you.

01:17:49   Frankly, I think it will be.

01:17:50   And when you're ready to launch, use off for code ATP

01:17:52   to save 10% off your first purchase of a website

01:17:54   or domain.

01:17:55   So once again, squarespace.com/atp

01:17:57   to start that free trial.

01:17:59   And when you're ready to purchase,

01:18:00   use off for code ATP to save 10% off your first purchase

01:18:03   of a website or domain.

01:18:04   Thank you so much to Squarespace for sponsoring our show.

01:18:11   Nick asks, do the capabilities of the Vision Pro

01:18:13   make you more bullish about Apple's potential in the car

01:18:16   space?

01:18:17   Assuming Vision Pro ships, Apple will

01:18:18   demonstrate both delivering projects

01:18:20   very long in development and a ton of visual capabilities.

01:18:24   The thing that I think is most interesting to me about Vision

01:18:29   Pro is the R1 chip, I believe it's called.

01:18:31   Is that right?

01:18:32   The one that does all the real-time processing.

01:18:35   Because if I understand it properly,

01:18:37   the R1 chip is running some sort of real-time OS on it

01:18:42   or something like that.

01:18:43   I might be making things up here,

01:18:44   so definitely check my work on this.

01:18:47   But it runs a real-time OS.

01:18:48   And that's the sort of thing that you

01:18:50   would need for a car safety computer or something

01:18:53   like that.

01:18:54   I don't necessarily think Apple is

01:18:55   going to do any actual automobile.

01:18:58   I personally feel like that ship has probably

01:19:01   sailed, to use a different vehicle metaphor.

01:19:04   But I do think that this is probably either--

01:19:08   and I think I said this either last week or the week before.

01:19:11   I think this real-time stuff was either born out

01:19:13   of the car project or could be used by the car project

01:19:17   if it wasn't born there originally.

01:19:19   So does it make it more bullish?

01:19:21   I don't know about that.

01:19:23   But it certainly expands the possibilities of things

01:19:26   that Apple could potentially conquer.

01:19:29   It's one thing to do a screen used for infotainment.

01:19:33   It's another thing to do the gauge cluster.

01:19:34   And obviously, Apple has already announced

01:19:37   they intend to do the gauge cluster.

01:19:38   But all of us were very worried.

01:19:40   I think, well, how does that work when Apple doesn't really

01:19:43   have a real-time OS that will keep

01:19:44   the speedometer on your screen and refreshing quickly always?

01:19:48   And well, now they have hardware and software to do it.

01:19:52   So it's definitely interesting, although again, I

01:19:55   don't personally think an Apple car will ever be released.

01:19:58   Or at least I don't think that's sitting here now.

01:20:00   Yeah, this doesn't give me any more confidence in the car

01:20:02   project for one very important reason.

01:20:05   With Apple Vision Pro, your life is not on the line.

01:20:09   At least I hope not.

01:20:10   I guess the battery could burst into flames in your pocket

01:20:13   or whatever.

01:20:13   But the stakes are so much higher.

01:20:16   I get what you're saying.

01:20:17   Oh, Vision Pro, real-time stuff, Vision stuff, AR.

01:20:20   The technology matches there.

01:20:22   Some related car thing, self-driving cars, car

01:20:25   visual experience.

01:20:26   You can see how this technology overlapped.

01:20:28   But there is no overlapping consequences.

01:20:31   This is like you're literally on your couch.

01:20:33   And yeah, maybe the thing crashes

01:20:35   and you can't use an app ride or whatever.

01:20:36   The stakes are just so much lower.

01:20:39   And it's not-- so look, Apple can

01:20:41   ship things that have been in development for a long time.

01:20:43   The car thing is not about the length of development time.

01:20:46   The car thing is about the consequences if it's not great.

01:20:48   The consequences of the Vision Pro is not great

01:20:50   is whatever the product that flops.

01:20:52   But people aren't going to die no matter what Apple

01:20:55   does in the car space unless it's just a better car play.

01:20:58   Even just better car play, there's higher stakes.

01:21:01   But if they're doing anything having

01:21:02   to do with controlling a car or whatever,

01:21:05   boy, that's so much harder.

01:21:06   And it's an area that Apple has never really gone into.

01:21:10   None of the things that Apple ships

01:21:12   are so directly responsible for the life of the people who

01:21:17   are using them as something that is actually controlling

01:21:20   anything about a car.

01:21:21   So I still think that Apple has a lot of hurdles

01:21:24   to overcome to figure out anything in the car space.

01:21:27   Maybe a better analogy would be between health and cars,

01:21:29   because health, like the blood glucose thing or whatever

01:21:32   is rumored, health and car, both things where people's lives are

01:21:36   on the line.

01:21:36   Vision Pro, hopefully no one's life is on the line.

01:21:40   Going back to the Realtime OS thing for a second,

01:21:43   when you look at-- we did hear rumblings back years ago

01:21:48   that Apple was working on a Realtime OS as part of the car

01:21:52   project.

01:21:53   Now, whether that was the same thing as the Realtime OS that's

01:21:56   running in Vision OS, we don't know.

01:21:59   I bet there's some overlap.

01:22:01   I bet there's at least some shared expertise there,

01:22:04   and if not some shared code-- although honestly, there

01:22:07   probably is some shared code as well.

01:22:09   One thing we learned last summer when they kind of pre-announced

01:22:13   that new carplay system with the gauge clusters,

01:22:15   as you were just saying, Casey, and RS-Technica

01:22:18   did a good article about it.

01:22:19   We talked about it on the show about how just

01:22:21   by regulatory reasons, gauge cluster OSes

01:22:25   have to be Realtime OSes.

01:22:27   And listeners wrote in to inform us that-- I didn't even

01:22:30   know this-- that Apple's modern chips, the modern Apple

01:22:34   Silicon ARM chips, are able to run in virtualization two

01:22:40   OSes at the same time.

01:22:41   One of them can be a Realtime OS, and the other one isn't.

01:22:44   And I think this is how-- I think

01:22:47   some part of Google's stack for cars, some part of it

01:22:51   does this, I think.

01:22:52   And so we were speculating at the time, hey,

01:22:54   maybe future iPhone, iOS versions

01:22:58   will be able to run this Realtime OS as some slice

01:23:03   of the resources of the chip, and then run iOS side

01:23:08   by side with it.

01:23:09   And that little Realtime OS could run the gauge cluster

01:23:12   in CarPlay, if your phone is running CarPlay,

01:23:14   without disrupting the rest of iOS.

01:23:17   Well, it sure looks like, when you

01:23:18   look at how Apple describes the architecture of Vision OS,

01:23:22   it sure looks like it's that exact same architecture,

01:23:25   with the Realtime OS running on the R1,

01:23:27   and running the AR pass-through functions of Vision Pro.

01:23:34   And then having Vision OS running inside of that,

01:23:39   but the Realtime OS is not disturbed.

01:23:40   So if some part of Vision OS hangs or crashed or something,

01:23:45   your reality is not weirdly paused or warped,

01:23:47   or some way that could make you sick, or anything like that.

01:23:50   So it seems like they're doing that kind of split architecture

01:23:54   here.

01:23:55   And maybe the car project produced the Realtime OS,

01:23:59   and gave them the foundations to do simultaneous Realtime

01:24:02   OS and other OS.

01:24:04   And maybe they didn't use that to make a car,

01:24:07   but they used it to make the AR headset, and also

01:24:09   future CarPlay stuff with the iPhone.

01:24:12   So I think the car project--

01:24:14   I still think the car project is as doomed as it ever was,

01:24:18   but I think we're also seeing stuff falling out of it that

01:24:21   is good, and that's being able to be used in other areas.

01:24:25   Now, going back to the actual car project as a product

01:24:29   possibility, in addition to everything John said,

01:24:31   which is correct, that I don't think Apple is super

01:24:35   willing to make products that could potentially kill you,

01:24:38   I think also, when you look at the general purpose computing

01:24:42   landscape, who else is going to make an AR headset that

01:24:48   has strong general computing possibilities?

01:24:52   Maybe Microsoft, maybe.

01:24:54   Maybe Google with Android.

01:24:56   I mean, Microsoft did.

01:24:57   It's called HoloLens.

01:24:59   It costs as much as the Vision Pro and is not as good.

01:25:02   Yeah, I understand it's useful in a couple of very narrow

01:25:05   markets, but it's not at all a mass market thing.

01:25:08   And I don't even think Microsoft even thought it would be.

01:25:11   But you have basically three companies--

01:25:14   Microsoft, Google, and Apple-- who even could make a general

01:25:18   purpose AR headset.

01:25:20   And we've seen what they make in other areas, and they're fine,

01:25:23   but they're not great.

01:25:24   So if anyone was going to make a great AR computing experience,

01:25:28   it was going to be Apple.

01:25:29   Look at cars, though, and I don't see that strong need.

01:25:33   Why does this have to be Apple?

01:25:35   Lots of people make really nice cars.

01:25:38   Whatever Apple thought they could add to that market that

01:25:41   would be so much better than what everyone else is making,

01:25:44   I don't see it.

01:25:45   I mean, look, maybe I don't have the imagination for it.

01:25:48   Maybe I'm not seeing something.

01:25:49   Maybe I just don't know what amazing thing that they had

01:25:53   or thought they could do or still think they could do

01:25:55   there.

01:25:56   I don't know.

01:25:57   But I just don't--

01:25:58   the amazing computing prowess they

01:26:01   have to make awesome computers that, for general purpose use,

01:26:06   I don't see them having that same amazing advantage

01:26:09   in making cars.

01:26:10   We haven't heard much about the car project recently.

01:26:13   It seems like it gets disbanded and regrouped every two years

01:26:16   or something, or even more than that.

01:26:19   Maybe it's finally now sizzling out.

01:26:22   I don't know.

01:26:23   But I sure hope we don't hear more about it, honestly,

01:26:27   because what we see now is the headset,

01:26:30   here's something that probably took a similar scope

01:26:34   and scale of resources and time investment.

01:26:37   But what they were making was what

01:26:40   is probably going to end up being a pretty great computer.

01:26:42   And they're really good at making great computers.

01:26:45   And I can't imagine them spending a similar amount

01:26:48   of time and resources making a car that would actually

01:26:51   be worth all that time and resources.

01:26:53   Even if the tech was directly transferable

01:26:55   and exactly related, which I don't think it actually is,

01:26:59   it wouldn't matter because that's just the tech.

01:27:01   In the end, the product, the car thing,

01:27:03   if you want to do anything with cars other than just show

01:27:06   the current speed and play songs from your phone,

01:27:08   if you're going to control some part of the car,

01:27:10   like all the self-driving stuff or whatever,

01:27:12   even a safety system, that's the hard part.

01:27:15   Not like, oh, implementing something

01:27:17   that you could implement self-driving on top of,

01:27:19   like a platform with a real-time OS and a chip and a control.

01:27:22   Now you have to write the software

01:27:24   to actually drive the car.

01:27:25   That's the hard part.

01:27:26   And it's the hard part both technically,

01:27:27   because no one's figured out how to do it,

01:27:30   even as well as a human at this point,

01:27:32   but also it's hard organizationally for Apple

01:27:35   because they're not accustomed to shipping products like that.

01:27:38   And that's why I'm saying success with the Vision Pro

01:27:41   does not make me more optimistic about the car project

01:27:44   because the things that are hard about the car project

01:27:46   are not the things that Vision Pro is doing well.

01:27:49   Again, even if you are willing to believe that the tech is

01:27:51   directly transferable, that's not

01:27:53   the hard part of the car project.

01:27:55   The hard part of the car project is

01:27:56   outside of the technical platform foundations.

01:27:59   It's everything else, like, OK, now drive the car.

01:28:03   Not to mention, now sell the car and service the car

01:28:06   and provide repair parts and a dealer network for--

01:28:10   there's so much about cars.

01:28:12   Apple have a self-repair program, though,

01:28:13   to send you a two-post lift in the mail

01:28:16   in a big giant Pelican case.

01:28:18   Yeah, a two-post lift in a Pelican case, yep.

01:28:21   That's definitely it.

01:28:23   Mark Blender writes, do you have any sort of app uninstaller,

01:28:28   such as AppCleaner on Mac OS?

01:28:29   Or excuse me, do you use any sort of it?

01:28:31   I know the official way to uninstall something is

01:28:33   simply to drag it to the trash, but I find myself, perhaps,

01:28:35   unreasonably worried about leftover files.

01:28:37   Leftover files are definitely a thing,

01:28:39   but I haven't run one of these in forever.

01:28:42   So I don't know, Marco, let's start with you.

01:28:45   Is there anything on your computer

01:28:47   that serves this purpose?

01:28:48   I don't use anything like this.

01:28:50   I don't like having to install accessory apps to serve

01:28:57   accessory functions that I think shouldn't need to exist.

01:29:01   I know that's kind of a broad statement, but in general,

01:29:04   I don't run a lot of different utilities on my Mac

01:29:07   that are always there or replacing some system function.

01:29:11   For the most part, I like to use the built-in stuff

01:29:13   the way it's meant to be used, because that tends to not get

01:29:16   me into trouble.

01:29:17   It tends to not break stuff.

01:29:18   And we were raising a drama saying earlier

01:29:19   how sometimes privacy-preserving stuff or ad blockers,

01:29:23   whatever, can break websites.

01:29:25   I always worry that using apps like this

01:29:28   might break something else.

01:29:30   And I don't see the need.

01:29:32   I just delete stuff.

01:29:33   If something has its own uninstaller,

01:29:36   I will run that uninstaller.

01:29:37   So certain things that install hooks in parts of the system,

01:29:41   they'll have uninstallers, so I'll use that.

01:29:43   For regular old apps, I'll just delete them.

01:29:45   This is a big focus of my early Mac OS X reviews,

01:29:48   was talking about the Next-based bundle system, where

01:29:51   you have a folder with a .app extension,

01:29:52   and inside there are a bunch of files,

01:29:54   and that's how applications are made up.

01:29:56   And so I was comparing it to the classic Mac, where

01:29:58   you had resource forks, which were

01:29:59   their own sort of structured thing, where

01:30:02   you could have different kinds of resources that

01:30:03   belong to the application, and they

01:30:05   were individually editable with the resource editor.

01:30:07   But the Next bundle system did that, but in the file system.

01:30:10   And because I'm an old-school Mac user,

01:30:12   uninstaller is just something Windows users use.

01:30:13   Mac users don't need to do that.

01:30:14   You just drag the app to the trash.

01:30:16   But the reality of Mac OS X was, especially in the early days,

01:30:20   there were some apps that came with quote, unquote,

01:30:22   "installers," which would spray files all over your drive.

01:30:25   And also, when you drag the app to the trash,

01:30:28   obviously, any file that wasn't inside that app bundle

01:30:32   is still on your computer, because all you did

01:30:34   was put the app bundle in the trash.

01:30:36   And there has never been, to my knowledge,

01:30:37   any part of the Mac operating system in the post-Mac OS X

01:30:41   days that magically knows that when you drag an app

01:30:43   to the trash, it has to go clean up a bunch of other files

01:30:46   or whatever.

01:30:47   There's no hooks for third-party developers to do that.

01:30:50   There's nothing for the first-party developers

01:30:51   to do that.

01:30:52   So this market for app cleaner things is like, hey,

01:30:55   when you drag your app to the trash,

01:30:56   do you know it's leaving some files behind?

01:30:58   Two things about that.

01:30:59   One, sometimes you want it to leave files behind.

01:31:03   This was true in classic Mac OS, and it's true of Mac OS X.

01:31:06   I want it to leave the preference file behind,

01:31:08   because if I ever reinstall that app,

01:31:10   I don't have to reset all my settings.

01:31:12   Preference files are tiny.

01:31:13   They've always been tiny, even classic Mac OS.

01:31:15   They're definitely tiny now.

01:31:17   Please, leave that on my computer

01:31:19   so that if I ever do decide to reinstall this app,

01:31:22   it'll be able to read those settings.

01:31:23   Even if it's a newer version, hopefully,

01:31:24   it'll read an old version of its settings file.

01:31:26   So that's an example of residue that I want to leave behind.

01:31:28   Second, app cleaner apps, those have no idea

01:31:33   how any individual app works.

01:31:35   They're making a best guess based on heuristics,

01:31:38   based on maybe some knowledge of some specific applications.

01:31:41   But I don't want an app from a third party

01:31:43   to try to know what things it can safely delete

01:31:47   that other apps put there.

01:31:50   Most of the time, you'll be fine, because it can say,

01:31:52   oh, well, the preferences are in the preference folder,

01:31:53   and anything in caches can be deleted,

01:31:55   and this, that, the other thing.

01:31:56   But things get complicated when you

01:31:57   get into more sophisticated applications

01:31:59   or suites of applications, where I'm not confident

01:32:01   that a third party application even

01:32:02   could know what the right thing to do is to delete this stuff.

01:32:05   Hell, I'm not even that confident that the uninstallers

01:32:07   that people write work correctly.

01:32:10   So yeah, I'd never use one of these app cleaner

01:32:12   uninstaller type things.

01:32:13   Now, the final note is, as the Mac operating system

01:32:16   and the post Mac OS X era has matured over the many, many

01:32:20   years, Apple has slowly but surely been removing

01:32:24   every single kind of thing that cannot be inside the app

01:32:28   bundle.

01:32:28   It used to be that whole swaths of common functionality

01:32:33   could not be inside the app bundle,

01:32:34   and now they've just been moving them all in there.

01:32:36   Extensions, secondary applications, helper apps,

01:32:42   login launch items, menu bar things,

01:32:45   so much stuff that used to have to be outside your application

01:32:48   that it'd have to be sprayed into your slash library

01:32:50   folder at the top level of your disk

01:32:52   or until the slash library or whatever.

01:32:54   All that stuff now can live inside the app bundle, which

01:32:58   means that when you drag the app to the trash, a really good

01:33:01   modern well-behaved app, you've deleted its login item.

01:33:04   You've deleted its menu bar thing.

01:33:06   You've deleted its embedded helper application.

01:33:09   You've deleted everything.

01:33:11   You may even have deleted its helper command line thing

01:33:13   because they could have just put a sim link in user local that

01:33:16   just sublinked into the bundle or whatever.

01:33:18   Not every app is that well-behaved.

01:33:20   That's true.

01:33:20   But the path that Apple has been paving

01:33:23   is we have a way for everything to be inside.

01:33:25   Even Safari extensions.

01:33:27   If you ship a Safari extension now, the way you do it

01:33:29   is you embed the Safari extension inside an app.

01:33:32   It's not like the Safari extension

01:33:33   goes into slash library slash Safari slash extensions

01:33:35   and you have to remember to dig it out of there

01:33:37   to get rid of it.

01:33:37   No, it's literally inside the app bundle

01:33:40   in your applications folder.

01:33:41   That's where it is.

01:33:42   And when you put the application in your applications folder,

01:33:44   the Mac OS scans it, finds the Safari extension,

01:33:47   and tells Safari about it.

01:33:49   So Apple is trying to make it so that when you drag it

01:33:51   into the trash that it's not just an illusion that really

01:33:54   you are trashing everything having to do with it.

01:33:56   Finally, there are still some apps

01:33:57   that spray things in various locations, which

01:33:59   is why I use and recommend the app launch control, which

01:34:03   lets you see the things that are outside the app bundle

01:34:06   that apps may install.

01:34:07   And that is the only thing remotely like an app cleaner

01:34:10   that I use.

01:34:10   And all it does is show me what exists

01:34:13   and then I use my knowledge and experience

01:34:15   to know which things I can safely disable or delete.

01:34:17   But app cleaner is even more dangerous than KeychainX--

01:34:20   not app cleaner-- launch control is even more dangerous

01:34:22   than KeychainX.

01:34:23   Do not muck about it if you don't know what you're doing.

01:34:26   It is very easy to break things.

01:34:28   Instead, just ignore apps in this class

01:34:30   and just hold on tight and wait until everything

01:34:32   is actually moved into the app bundle for every app

01:34:34   you care about.

01:34:36   Dave Copeland writes, in the discussion of Google Auth

01:34:38   and syncing, there was an implication

01:34:40   that Keychain can perform two-factor authentication.

01:34:43   I've seen others say that one password can as well.

01:34:45   But how is that a second factor?

01:34:47   Doesn't it turn a second factor into a first

01:34:49   since access to your computer and Keychain means passwords

01:34:51   and 2FA?

01:34:53   Shouldn't 2FA be kept as a separate factor

01:34:55   on an iPhone or other device?

01:34:58   I feel like I can take a stab at this.

01:35:01   But Jon, I think it might have been

01:35:02   you that have added some very relevant replies

01:35:06   into the show notes here.

01:35:07   Yeah, we don't have to answer it because it was answered

01:35:09   by other people I'm asking on.

01:35:10   So here's Sebastian Cohen saying,

01:35:12   this is all just two-step, not two-factor.

01:35:15   This can still be valuable in case your credential entry gets

01:35:18   intercepted, for example.

01:35:20   So I know there's confusion about what is two-factor,

01:35:23   what is two-step, whatever.

01:35:23   You have to think about what is this protecting against.

01:35:26   One of the things this protects against is if, say,

01:35:29   your password is intercepted somewhere,

01:35:31   your two-step thing, even though they're both on your phone

01:35:34   or both on your Mac or both on your whatever,

01:35:36   saves you here because the thing that was intercepted

01:35:39   or was dumped from some database or website or whatever

01:35:41   is just your password.

01:35:42   They don't have the other factor.

01:35:44   Even though both of the factors are on your phone,

01:35:46   like the password's on there and the 2FA code

01:35:49   is also on your phone, what they got

01:35:51   was a cracked password dump from some website

01:35:53   that had bad security.

01:35:54   They just have your password.

01:35:56   They don't have the other thing.

01:35:58   And Drew writes, the central threat model motivating 2FA

01:36:01   is not that someone has access to your computer.

01:36:04   It says they've captured your password from someone else's

01:36:05   insecure system.

01:36:07   Given the centrality of email to the account 2FA management

01:36:09   workflows, if someone has access to your computer,

01:36:12   they can likely reconfigure your quote, unquote, "true" 2FA to.

01:36:15   This goes back to what we were saying before.

01:36:17   Your email account in the end for most services

01:36:19   is the key to everything.

01:36:20   If they have your phone, don't worry about the fact

01:36:22   that your password and your two-factor code

01:36:25   are both in the same place in settings.

01:36:28   Worry about the fact that they have your phone.

01:36:30   That means they have access to your email.

01:36:31   That means they can reset any password

01:36:32   and all the 2FA stuff.

01:36:34   It's just, yeah.

01:36:34   So you're thinking about the threat modeling wrong

01:36:38   if you think, if only my second factor

01:36:40   was someplace separate, I would be safe.

01:36:42   There are cases where that's true,

01:36:44   but if someone has access to your computer, your phone,

01:36:47   as you, like if they have that kind of access,

01:36:49   you're probably screwed no matter what,

01:36:51   unless you're really hardcore with the security keys

01:36:54   and you never keep them in the same place as your phone,

01:36:56   because how did they get your phone?

01:36:57   They didn't get your little YubiKey thing at the same time?

01:37:01   Security is difficult.

01:37:02   But I will say that whether it's just two-step or two-factor,

01:37:08   it is better than nothing.

01:37:09   It does predict against more scenarios

01:37:11   than just a password, which is why I recommend it.

01:37:13   Same deal with pass keys.

01:37:14   They have a different set of trade-offs.

01:37:15   Pass keys, you don't have to worry about forgetting them.

01:37:17   They're impossible to fish,

01:37:19   because the human does not choose when to submit them.

01:37:21   The computer does, and the computer is not fooled

01:37:23   by fake-looking emails.

01:37:24   So yeah, security's complicated,

01:37:27   but two-factor is not pointless.

01:37:29   It is just different than you might think it is.

01:37:32   - Thanks to our sponsors this week,

01:37:34   Squarespace and Rocket Money.

01:37:36   And thank you to our members who support us directly.

01:37:38   You can join, please do, at atp.fm/join.

01:37:42   And we will talk to you next week.

01:37:45   (upbeat music)

01:37:48   ♪ Now the show is over ♪

01:37:50   ♪ They didn't even mean to begin ♪

01:37:53   ♪ 'Cause it was accidental ♪

01:37:54   ♪ Accidental ♪

01:37:55   ♪ Oh, it was accidental ♪

01:37:57   ♪ Accidental ♪

01:37:58   ♪ John didn't do any research ♪

01:38:00   ♪ Marco and Casey wouldn't let him ♪

01:38:03   ♪ 'Cause it was accidental ♪

01:38:05   ♪ Accidental ♪

01:38:06   ♪ Oh, it was accidental ♪

01:38:08   ♪ Accidental ♪

01:38:09   ♪ And you can find the show notes at atp.fm ♪

01:38:14   ♪ And if you're into Twitter ♪

01:38:17   ♪ You can follow them at C-A-S-E-Y-L-I-S-S ♪

01:38:22   ♪ So that's Casey List M-A-R-C-O-A-R-M ♪

01:38:27   ♪ Auntie Marco Arment S-I-R-A-C ♪

01:38:32   ♪ U-S-A-C-R-A-Q-za ♪

01:38:35   ♪ It's accidental ♪

01:38:36   ♪ It's accidental ♪

01:38:38   ♪ They didn't mean to accidental ♪

01:38:42   ♪ Accidental ♪

01:38:43   ♪ Tech, podcast, so long ♪

01:38:46   - So I have some updates with regard to computers and cars.

01:38:53   So which one would you like to know first?

01:38:56   - Car.

01:38:58   - I don't remember where we last left our heroes, but.

01:39:00   - You would have ruined Aaron's car.

01:39:04   - Yes.

01:39:05   - And you still don't drive an EV somehow.

01:39:06   And you consider yourself a driving enthusiast.

01:39:08   - Well, because I would like to shift for myself, sir.

01:39:11   And yes, I'm aware of the Toyota garbage.

01:39:13   - Yes, everyone said that.

01:39:15   - I can't decide if that's delightful.

01:39:17   - We were way ahead of the curve on that.

01:39:18   We talked about this almost a year ago, I think,

01:39:20   on the show.

01:39:21   Remember, when we talked about the EVs with MagneTrans,

01:39:22   which I think was a prototype Mustang at the time.

01:39:25   Anyway.

01:39:26   - Yeah, yeah, yeah.

01:39:26   Anyways, pro tip, you do not,

01:39:30   if you're an Allstate customer anyway,

01:39:31   you do not need to use Safe Flight Auto Glass

01:39:34   in order to replace your windshield.

01:39:36   And may I recommend that you do not use Safe Flight Auto Glass

01:39:39   to replace your windshield.

01:39:41   - That good, huh?

01:39:42   - Oh, it's been a nightmare.

01:39:43   So we are on our second replaced windshield

01:39:45   in as many weeks as we speak.

01:39:48   The short, short version is,

01:39:49   somebody came out to install the new windshield.

01:39:51   They said, "Oh, they sent me the wrong thing.

01:39:53   Your car has a heads-up display.

01:39:55   This windshield will not work with the heads-up display."

01:39:57   So we wait another week or two.

01:39:59   A very nice gentleman comes to install the windshield

01:40:02   at our house.

01:40:03   And he says, "Oh, this isn't an OEM windshield.

01:40:09   It isn't a Volvo-produced windshield.

01:40:11   It's a third-party windshield.

01:40:13   It should be fine."

01:40:15   So, okay, great.

01:40:16   And then the next time we get in the car,

01:40:18   everything is lit up with warning signs

01:40:20   with regard to safety stuff.

01:40:21   No lane keeping aid, no speed limit signs,

01:40:23   no pilot assist, nothing.

01:40:26   Stupendous.

01:40:27   Okay, so we call Safe Flight.

01:40:29   We go on the way there,

01:40:31   where I'm expecting them to do

01:40:33   their super fancy calibration.

01:40:34   On the way there, all the warnings go away.

01:40:37   Super.

01:40:38   So I get there.

01:40:39   This is a 15, 20-minute drive away.

01:40:41   I get there.

01:40:42   I say to them, "Look, it wasn't working.

01:40:44   Now it says it's working.

01:40:45   Just can you hook it up to whatever computers you have

01:40:47   and at least just tell me that it should be fine?"

01:40:50   And they did.

01:40:50   And they said, "Yeah, it should be fine."

01:40:53   The next day, everything lights up again.

01:40:55   Stupendous.

01:40:56   All right, so I go back to say,

01:40:58   I call Safe Flight back and say, "Hey, it's me again.

01:41:02   Can I have an OEM windshield, please?"

01:41:04   'Cause obviously everything associated with this is screwed.

01:41:07   Said, "Sure."

01:41:09   Hey, how long do I have to wait for that?

01:41:10   Like a week and a half.

01:41:12   Stupendous.

01:41:13   So we at least at this point have a windshield

01:41:16   that from the outside looks good,

01:41:18   but the heads-up display is blurry

01:41:21   and none of the automated driving sensors work.

01:41:24   It ended up that we drove an hour and a half

01:41:26   just a couple of days ago to go to the Blue Ridge Tunnel,

01:41:28   which I think I've talked about in the past, just for fun.

01:41:30   It was on Father's Day.

01:41:31   And we took the Volvo 'cause we took the dog with us.

01:41:35   And we didn't have any form of cruise control

01:41:38   'cause it wouldn't even let you do regular cruise control

01:41:41   because it wasn't able to do automated cruise,

01:41:44   the distance following cruise.

01:41:46   And it was like, "Well, if I can't do automated cruise,

01:41:47   there is no other cruise."

01:41:49   And so that was kind of a pain, but whatever.

01:41:50   So today, literally today,

01:41:53   we went back to Safe Flight in downtown Richmond

01:41:55   and they put in a Volvo windshield

01:41:59   and they recalibrated everything and knock on wood,

01:42:03   I think everything might be okay now.

01:42:05   But it has been a month and two days, I believe,

01:42:08   since I slightly shattered Aaron's windshield

01:42:11   and it is now hypothetically fixed.

01:42:14   What a nightmare.

01:42:16   And as I'm going around and around with Allstate,

01:42:19   with Safe Flight and all these other people,

01:42:21   come to find out, and this is what I said earlier,

01:42:24   despite the fact that when you fill out a claim online

01:42:26   with Allstate, they're like, "Oh, it's glass?

01:42:28   Here, go to Safe Flight.

01:42:29   They'll take care of it."

01:42:30   Come to find out, you don't have to use Safe Flight.

01:42:34   And I could have gone to Volvo from the get-go

01:42:38   and had Volvo do it, where Volvo presumably knows

01:42:41   how to replace their own windshields

01:42:43   and certainly should know how to calibrate the god darn thing.

01:42:46   - Oh, man. - So I could have done that,

01:42:49   but I had no idea.

01:42:51   And so the Volvo people were very nice,

01:42:53   but they were like, "My guy, why didn't you bring it to us?"

01:42:55   And I was like, "My guy, I didn't know I could!"

01:42:57   But sure enough, I could.

01:42:58   So anyway, so pro tip, if you have a glass issue,

01:43:01   you might want to investigate with your insurance

01:43:03   whether or not you have to go to the Safe Flight people

01:43:07   because you might not have to.

01:43:09   But that seems resolved, as far as I can tell.

01:43:11   With regard to Aaron's computer,

01:43:14   I don't think I mentioned on the show,

01:43:16   I might have mentioned it offhandedly,

01:43:18   but the MacBook Adorable randomly came back to life

01:43:21   after I let it sit for a couple of days.

01:43:23   I didn't do anything.

01:43:24   I don't know what happened.

01:43:25   It just decided to work again.

01:43:27   But that isn't really a long-term solution.

01:43:29   And I still don't love the idea

01:43:31   that with the MacBook Adorable dead,

01:43:33   and even with it alive, potentially,

01:43:35   I didn't really have any sort of backup plan

01:43:37   if my computer kicked the bucket.

01:43:39   I do have the Mac Mini that I use for Plex and channels,

01:43:42   and it is more than computationally capable

01:43:44   of doing a podcast recording if necessary,

01:43:48   but it would be kind of a pain to move it

01:43:50   and attach monitors to it and so on and so forth.

01:43:53   So I wouldn't want to have to do that.

01:43:56   So I was trying to figure out what the right answer is

01:43:58   in order to replace Aaron's computer

01:44:01   with something more modern,

01:44:02   preferably Apple Silicon and so on and so forth,

01:44:05   and I was hemming and hawing about what the right answer was,

01:44:07   and it seemed kind of wasteful

01:44:10   to replace what is effectively

01:44:13   her Kroger online shopping computer

01:44:16   if it's still puttering along, somewhat working.

01:44:19   But out of the woodwork came,

01:44:22   I don't know if this person wanted to be anonymous,

01:44:24   and I will assume anonymity just to be safe,

01:44:28   but out of the woodwork came a random person

01:44:30   who I know through the internet,

01:44:32   but not extraordinarily well,

01:44:35   who was kind enough to say to me,

01:44:37   "I have an M1 MacBook Air that's collecting dust.

01:44:39   Would you like it?"

01:44:40   To which I said, "Oh my frickin' God, yes I would."

01:44:44   - Wow. - So this exceptionally

01:44:46   kind individual who I would name generally,

01:44:49   but I didn't think to pre-clear this conversation with them,

01:44:52   they shipped, well I paid for shipping

01:44:55   because it was the least I could do,

01:44:56   but they shipped this computer to me,

01:44:59   and it has arrived, I have loaded it with Aaron's stuff.

01:45:03   Oh my word, it's so much nicer than the adorable.

01:45:05   I will always and forever love the adorable.

01:45:07   I will always love that computer, always.

01:45:10   But oh my God, it's so much nicer than the adorable.

01:45:13   It is so fast, and this particular model,

01:45:16   I genuinely, I said I'll take it.

01:45:18   I had no idea how much RAM,

01:45:20   no idea what disk it had in it, did not care.

01:45:23   - It doesn't matter, even the lowest, tiniest configuration

01:45:26   is miles better than the MacBook Adorable.

01:45:29   - Exactly, and sure enough, it arrived,

01:45:32   and because this person is a developer

01:45:33   for a very big technology-related company,

01:45:36   not Apple, but someone in that kinda neck of the woods,

01:45:40   this computer happened to have a terabyte hard drive

01:45:43   and 16 gigs of RAM.

01:45:44   I could not have asked for this to go any better for me.

01:45:49   So thank you to the, whether or not you want it to be,

01:45:51   anonymous individual who has,

01:45:54   well, I mean, I guess I paid for the shipping,

01:45:55   but they still shipped me the computer,

01:45:57   and I am so thankful for it,

01:45:59   because now I have a backup computer,

01:46:01   which now Marco is thankful,

01:46:03   because I have a backup computer

01:46:05   in case my computer has problems,

01:46:07   and it is genuinely very nice,

01:46:09   so much nicer than we need or deserve,

01:46:11   and so thank you very, very, very much for doing that.

01:46:14   It was extremely kind, and the best part of this,

01:46:18   it's a person after my own heart, right?

01:46:20   I opened the box. - Is it white?

01:46:21   - Well, first of all, well, no, it's the silver,

01:46:24   whatever, but no, I opened the box,

01:46:26   and first of all, it came in the Apple box.

01:46:27   I had no idea what was being shipped to me,

01:46:30   other than that it was an M1 MacBook Air.

01:46:31   That's all I knew, and that I understood it

01:46:33   to be in good condition.

01:46:35   So I opened the shipping box, and sure enough,

01:46:39   there's the original MacBook Air box.

01:46:41   Bonus points for this individual.

01:46:43   I opened the box, and there are probably three lunatics

01:46:48   in this world that would have done this.

01:46:51   I'm one of them.

01:46:52   I bet Jon is one of them,

01:46:54   and this just wonderful anonymous human

01:46:57   not only kept that shrink wrap, not shrink wrap,

01:47:01   but you know what I'm saying, that plastic stuff

01:47:03   that's wrapped around the computer

01:47:04   when it was shipped from Apple?

01:47:06   Not only did they keep it, they put the friggin' computer

01:47:09   back inside it the best they could.

01:47:11   I was overjoyed to see this.

01:47:13   - No, I do that too, when I sell a computer

01:47:15   and do the same thing.

01:47:15   - Oh, really, okay, fair enough.

01:47:16   I apologize, Marco, for selling you short that.

01:47:18   I thought you were a little more normal

01:47:19   than Jon and I. - No, I do that.

01:47:21   (laughing)

01:47:23   - So this extraordinarily tasteful individual,

01:47:26   they did that, they sent me the power brick.

01:47:28   I didn't expect that.

01:47:29   Not only did they send me a USB-C cable,

01:47:32   I'm pretty sure that it had been used at some point,

01:47:35   but they have the dexterity to somehow put it back

01:47:41   in the little cardboard thing that it came in.

01:47:43   Maybe they never used it, maybe I'm wrong,

01:47:45   but I think it had left--

01:47:46   - The trick is never take it out of the cardboard thing,

01:47:47   just FYI, someone who also packages products,

01:47:49   they'll just never take it out.

01:47:51   - Exactly, but either they never did, and I'm mistaken,

01:47:54   or they took the time to make it look really close

01:47:58   to the way it came from the factory.

01:48:00   Again, this is a person who gives a crap,

01:48:03   and I am forever indebted to them

01:48:05   and in love with them in a friendly way.

01:48:07   So Aaron's computer problems are solved

01:48:10   because now this incredibly kind individual

01:48:12   has sent me a M1 MacBook Air, which I don't deserve,

01:48:16   but I am so thankful for.

01:48:17   The best part, of course, of this whole story, though,

01:48:20   is that Aaron's working on, for Declan's yearbook,

01:48:25   it's very weird the way his school does it.

01:48:27   You order it online, you don't get it until the summertime,

01:48:30   you don't get to go around to your friends

01:48:31   and have them physically sign it,

01:48:32   but you can electronically sign it, it's very unusual.

01:48:35   But anyway, she's been working on this

01:48:37   for the last several days and trying to amass pictures

01:48:40   because you can pay a couple of bucks

01:48:43   to put in a few extra pages with your own pictures

01:48:46   and content in it.

01:48:46   It's like the world's crummiest photo book,

01:48:48   but it's really cute.

01:48:49   And Aaron's been working on this really, really hard,

01:48:52   and I look over at her,

01:48:54   and she's using the goddamn adorable.

01:48:56   I'm like, Aaron, what are you doing?

01:48:59   And she's, oh, I was-- - I was gonna say,

01:49:00   what did you do with that?

01:49:02   What did you do with the adorable?

01:49:02   Apparently you did nothing with it.

01:49:03   - I didn't do anything with it yet.

01:49:05   I've also been well over my eyeballs deep

01:49:09   and call sheet stuff, like everything's fine,

01:49:11   but I'm trying to get this out the door and blah, blah, blah.

01:49:12   So I didn't do the due diligence of hiding the adorable

01:49:17   or anything, or even in Aaron's credit,

01:49:19   I didn't really have a lot of conversation

01:49:20   with her about it other than saying,

01:49:21   hey, your new computer's ready.

01:49:23   But anyway, so I look at her using this damn adorable,

01:49:25   and I'm just like, oh my God,

01:49:27   how are you doing this to yourself?

01:49:28   I was like, well, why don't you use a new one?

01:49:29   I don't know, this is the one I grabbed.

01:49:31   I was like, okay, well, that's fine.

01:49:33   But anyway, so the car, hypothetically fixed,

01:49:36   thanks to me just being basically a Karen to Safe Flight

01:49:41   and finally getting them to do the job properly.

01:49:44   And then the computer's fixed,

01:49:45   thanks to this incredibly kind listener.

01:49:48   Someone in the chat is perhaps tongue in cheek,

01:49:50   perhaps not saying, am I gonna pay it forward

01:49:52   and send your adorable to someone?

01:49:53   I should, but I'm not,

01:49:55   because I fricking love that computer

01:49:56   and I will keep it forever,

01:49:57   even though it is a pile of garbage.

01:49:59   - You think of someone an unreliable, crappy computer,

01:50:01   that's not paying it forward, that's a punishment.

01:50:02   - Yeah, that's not a gift at all.

01:50:05   - Exactly, so yeah, so that's my updates.

01:50:07   - That's good.

01:50:08   Man, you've had a better tech week than I have.

01:50:10   I have had a terrible family device week.

01:50:14   - Oh no, what's going on?

01:50:15   - All right, so long story short,

01:50:18   someone in my house who is not me

01:50:20   accidentally sat on someone else's iPad

01:50:24   and caused what appears to be a crack in the screen.

01:50:27   - How bad, how big, how deep, tell me more.

01:50:31   - It's so subtle that it could be easily confused

01:50:34   for a scratch on the surface,

01:50:36   but it's across like a third of the screen.

01:50:38   - Well, I think we have developed a hierarchy here now,

01:50:41   the rock paper scissors style.

01:50:44   So an iPad can destroy a car windshield,

01:50:47   but a butt can destroy an iPad.

01:50:49   (laughing)

01:50:50   So if you touched your butt to the windshield of your car,

01:50:53   it would just shatter.

01:50:55   - Just blow it right off.

01:50:56   - Into a million pieces.

01:50:57   (laughing)

01:50:59   Well done, John, well done.

01:51:01   Anyway, so the owner of the iPad was extremely upset

01:51:06   at the butt crack situation.

01:51:09   And because the iPad is the owner's primary computing device

01:51:14   and they didn't cause the damage,

01:51:18   I thought it would be inhumane to make them suffer

01:51:22   with this damage forever.

01:51:23   But it is a crack, it is not a scratch.

01:51:27   The pencil doesn't work across it.

01:51:28   There's a whole bunch of little weirdness about it.

01:51:31   And this person uses the pencil, so I decided I have,

01:51:36   this is an M1 11-inch iPad Pro.

01:51:38   I have the exact same one as my iPad.

01:51:42   I'll just swap it.

01:51:43   Now, my iPad, because I use it so infrequently,

01:51:48   but when I do, I really need cellular usually,

01:51:52   because of that, I had gotten the T-Mobile

01:51:55   like super cheap data plan on it.

01:51:57   It was like five bucks a month or something.

01:51:59   It was super cheap, way cheap.

01:52:00   Normally we're an AT&T family.

01:52:02   It was way cheaper than AT&T's cheapest option.

01:52:05   So this was active on T-Mobile for, I don't know,

01:52:07   a year or two.

01:52:08   When I transitioned it over, that plan didn't carry over

01:52:12   'cause it was some weird prepaid thing

01:52:13   and the eSIM instantly got lost when I did the restore

01:52:17   and everything and all this stuff.

01:52:18   I'm like, "Ah, fine, who needs T-Mobile anyway?"

01:52:21   It isn't that good.

01:52:22   And the new owner of what was previously my iPad

01:52:26   actually used the cellular pretty heavily.

01:52:29   And oftentimes while we're on road trips,

01:52:31   so it's like, "All right, well,

01:52:33   "T-Mobile is not great for that, coverage-wise."

01:52:36   I want this to be on AT&T now.

01:52:38   - Can we, let me just interrupt you very quickly

01:52:40   and to go on a very brief tangent.

01:52:42   I want to be a T-Mobile customer so badly

01:52:45   because they seem to actually treat their customers,

01:52:48   as far as I can tell, like human beings

01:52:50   rather than just another cash machine to extract money from.

01:52:54   - Well, try canceling an iPad plan.

01:52:57   It's not super easy.

01:52:58   - Okay, so maybe I'm getting ahead of myself,

01:53:01   but I think I might've talked about this on the show.

01:53:03   Like a year or two ago,

01:53:04   I did the T-Mobile try-on or something like that.

01:53:07   I'll probably forget to put a link in the show notes,

01:53:08   but you can download an app on your iPhone

01:53:11   if you have an eSIM,

01:53:12   and you can literally provision your eSIM on your iPhone

01:53:15   to have a T-Mobile data plan for a month.

01:53:19   And so you can tell your iPhone,

01:53:21   "Hey, use Verizon, AT&T, whatever,

01:53:23   "for your existing, or excuse me, for phone calls,

01:53:27   "but use T-Mobile for data."

01:53:29   And you can even tell it,

01:53:30   "Don't fall back to your old carrier."

01:53:32   So you know like, "Oh, if I'm not getting data here,

01:53:35   "it's because T-Mobile sucks."

01:53:36   I tried this admittedly a year or two ago,

01:53:39   and when T-Mobile worked, it worked really damn well.

01:53:42   But let me tell you, in the Richmond, Virginia area,

01:53:45   it was very spotty, very, very spotty.

01:53:48   And I tried it thinking, "Maybe I will switch."

01:53:50   And now, mm-mm, no chance, or at least I'll,

01:53:53   maybe I'll have to try again in a few years,

01:53:54   but not right now.

01:53:56   So it bumps me out,

01:53:57   because they really do seem, cancellation issues aside,

01:54:00   they seem to treat their people, like their customers okay.

01:54:03   But golly, the service around here anyway, no good.

01:54:06   So anyway, I apologize, carry on.

01:54:08   So you need to now cancel this plan that's been orphaned.

01:54:11   - Yeah, so I can't, anyway,

01:54:13   long story short, T-Mobile makes it hard.

01:54:14   You gotta like talk to a web chat person,

01:54:16   and it was complicated.

01:54:18   Anyway, got that done.

01:54:20   So now I have this iPad that was on T-Mobile

01:54:23   that its new owner really needs it to be on AT&T.

01:54:26   I have had the hardest time ever trying to get this done.

01:54:30   Now, one could argue,

01:54:33   why don't I just fix the butt crack screen?

01:54:36   I don't know why, for some reason,

01:54:38   I didn't get AppleCare on this iPad.

01:54:39   I should have gotten AppleCare.

01:54:41   I often do for family devices.

01:54:43   I didn't on this for whatever reason.

01:54:45   The out of AppleCare repair price

01:54:48   for an M1 11 inch iPad Pro for a screen is $600.

01:54:53   Oh, gosh.

01:54:55   Now, a new one with the same configuration is $1100.

01:55:00   So, you know, that's less,

01:55:03   but it certainly doesn't feel like a good way to spend money.

01:55:06   And I thought, my iPad use is relatively minimal.

01:55:11   I don't care about this crack.

01:55:12   So I'll take it as mine, no big deal, fine.

01:55:15   Anyway, so switching the other one over, the good one,

01:55:18   trying to get this iPad that was on T-Mobile

01:55:21   to now be active on AT&T has been a nightmare.

01:55:25   I have had so many like,

01:55:28   all right, resend the eSIM, please.

01:55:30   Like, try to get carrier settings to update.

01:55:33   I have verified that T-Mobile is no,

01:55:34   it's no longer active on T-Mobile.

01:55:36   The plan I paid for has expired like weeks ago now.

01:55:39   It is no, it's definitely no longer active there.

01:55:42   Going back and forth to AT&T, both web chat,

01:55:45   and then I brought it into a store,

01:55:46   and it's like, it'll activate in just my house,

01:55:49   and then if I leave my house and go anywhere else,

01:55:51   it deactivates, says no service.

01:55:53   - What?

01:55:53   - Or it'll activate for like a day,

01:55:55   and then the next day, no service.

01:55:57   - That doesn't make any sense.

01:55:59   - I finally have it right now where it partially works

01:56:02   because I abandoned the eSIM,

01:56:04   and the person in the AT&T store gave me a physical SIM,

01:56:07   and that has made it work a little bit better

01:56:10   than the eSIM has.

01:56:11   But now, and like, there's like eight of those messages

01:56:14   stacked up in settings saying AT&T wants you

01:56:16   to install an eSIM.

01:56:18   This whole system, I've always had it work pretty well,

01:56:24   where I get a device, and it's only on AT&T

01:56:27   from the moment I get it, and I never have any problems.

01:56:30   I have activated and deactivated so many iPads and watches

01:56:34   for AT&T over the years, and it's great

01:56:36   'cause you can do them all through the web interface,

01:56:37   like no chat bots, no calling anybody.

01:56:40   I don't think it works with phones,

01:56:41   but for smartwatches and iPads,

01:56:43   you can do that with AT&T.

01:56:44   It's all web-based, so you can go on, cancel,

01:56:47   add new ones, no big deal, it's super easy.

01:56:49   This one, because it was on T-Mobile before,

01:56:52   has been a nightmare.

01:56:53   So I strongly suggest people out there,

01:56:56   the thing that Casey just told you to do

01:56:58   where you try T-Mobile, yeah, don't do that.

01:57:00   Definitely don't do that. (laughing)

01:57:01   This has cost me so many hours, literal hours,

01:57:05   of going back and forth with either doing the eSIM dance,

01:57:08   calling T-Mobile, going to AT&T.

01:57:10   I have honestly thought I should just replace this iPad.

01:57:13   It would be easier at this point to buy a new iPad

01:57:16   and to save myself hours of time than it would be,

01:57:20   or to pay the $600 to get the butt crack fix.

01:57:23   If you would've told me two weeks ago,

01:57:25   hey, look, you can pay $600 to solve this problem,

01:57:29   or you're gonna have two weeks of stress

01:57:32   and at least 10 hours of wasted time,

01:57:35   I would've paid the $600.

01:57:36   Knowing now what I know then,

01:57:38   that would've been the better idea.

01:57:40   My time is that valuable right now.

01:57:42   I'm so crunched on all angles.

01:57:43   Like, I need to recover time

01:57:45   from different places in my life.

01:57:47   This iPad is like ruined by this weird cellular weirdness

01:57:52   that I can't get either Apple or AT&T or T-Mobile

01:57:56   to figure out.

01:57:57   It's been a nightmare.

01:57:58   So yeah, word of advice, when you get an Apple device,

01:58:02   leave it on one carrier for its entire lifetime.

01:58:04   Do not change carriers, and especially in the eSIM era.

01:58:09   I would be saying the same thing as you

01:58:10   after that experience, but for what it's worth,

01:58:12   I have never, not yet anyway, had any such problems.

01:58:16   Like if you recall a year, year and a half ago,

01:58:18   something like that, we switched from AT&T to Verizon.

01:58:20   I'm pretty sure we were,

01:58:23   I don't remember if we were on eSIMs for AT&T at this point.

01:58:26   I think we were, but we went to Verizon eSIMs,

01:58:29   and that was all fine.

01:58:30   I've put different SIMs on it, or eSIMs on and off iPads,

01:58:36   if I'm not mistaken.

01:58:36   I have never run into this.

01:58:38   I used to use T-Mobile physical SIMs years ago,

01:58:40   'cause they used to have this absolutely

01:58:42   extraordinarily great pre-pay plan.

01:58:45   It was like five bucks for like five gigs

01:58:49   that you could use over the course of like

01:58:50   something like six months or something like that.

01:58:53   And that doesn't seem like that much,

01:58:54   but if it's an accessory device that you're not using

01:58:56   to like watch YouTube nonstop,

01:58:58   just a few gigs will last a real long time,

01:59:01   and that was the best, but they don't do that anymore.

01:59:03   Anyways, I haven't had these personal experiences,

01:59:06   so I don't wanna make everyone forever

01:59:09   and always think that eSIM is garbage,

01:59:11   but that experience is unquestionably garbage.

01:59:13   Like that sucks.

01:59:15   And what do you do?

01:59:16   Like if Apple can't do it,

01:59:17   and if T-Mobile can't do it, and AT&T can't do it,

01:59:19   what do you turn to?

01:59:20   Do you throw the thing in the ocean and start anew?

01:59:22   - Yeah, and like I posted the Mastodon

01:59:24   like maybe almost a week ago now.

01:59:26   I heard from a bunch of people,

01:59:28   many of whom have had similar problems,

01:59:30   and they're like, oh well, you know,

01:59:31   you have to go to like super high level support

01:59:32   with AT&T or Apple.

01:59:34   One person said they had to go through AppleCare,

01:59:36   and get the whole device replaced.

01:59:37   - Oh my gosh.

01:59:38   - Actually, most people said that.

01:59:40   Most people said, you know, the AT&T,

01:59:42   you gotta like get them to properly register

01:59:43   the IMEI in their system.

01:59:45   It's somewhere, it's not associated right or whatever else,

01:59:47   and I had the person in the store at least look at that,

01:59:49   and they couldn't figure out any problems with it.

01:59:51   Someone else said that it takes all these like levels

01:59:54   of escalation through different customer support,

01:59:55   and I'm like, how many more hours am I going to spend?

01:59:58   Like honestly, I was looking at trade-in prices.

02:00:01   I'm like, can I just trade this in?

02:00:02   Like how quickly can I solve this problem

02:00:04   without going through hours more customer service,

02:00:08   and then mailing it off and being without it for a week,

02:00:10   and like all this stuff.

02:00:11   Like, oh my God, like there's so,

02:00:13   and all this for the stupidest problem of like,

02:00:16   the cell service doesn't work.

02:00:17   And for this iPad, cell service is a must.

02:00:19   Like it's very frequently needed for this iPad.

02:00:22   So like I'm not gonna just not have it.

02:00:24   I'm not gonna tether.

02:00:25   Believe me, we've been doing that for the last,

02:00:27   like we just took a big road trip, did a lot of that.

02:00:30   It sucked.

02:00:32   Oh, tethering is the worst.

02:00:34   Yeah, anyway, I need cellular for this iPad,

02:00:36   and I'm seriously, I'm like about to replace it.

02:00:39   Like it's that bad.

02:00:41   - That sucks, I'm sorry.

02:00:43   - Anyway, I'm glad to hear your windshield

02:00:44   and computer worked out.

02:00:45   - For now, for now.

02:00:46   - See if your friend has any extra iPad Pros lying around.

02:00:49   - Yeah, right.

02:00:50   - I feel like this is your punishment

02:00:51   for not selling me that iPad, Margot.

02:00:52   - Do you want this one?

02:00:54   - Yeah.

02:00:55   - If you don't need cellular.

02:00:56   - I mean, I don't need the cellular, but it's too late now.

02:00:58   I already bought a brand new one,

02:01:00   but I was ready to buy your M1 iPad Pro off of you,

02:01:02   and you're like, yeah, I wanna keep it.

02:01:03   I might need it for something.

02:01:04   And look what happened.

02:01:05   - I can offer you the butt crack one for a good price.

02:01:07   - No.

02:01:08   - 'Cause Apple's traded and currently values it at $80.

02:01:10   - Oh, your chance to sell to me

02:01:13   was before someone cracked it in half with their butt.

02:01:15   - I'll give it to you for $81.

02:01:17   - Pass.

02:01:19   - Wow, even then.

02:01:21   - Who wants the thing with a cracked screen?

02:01:23   - If you don't use the pencil

02:01:24   and you don't look that closely, you'd never know.

02:01:27   - Yeah, yeah, the little slivers of glass

02:01:29   that are driving themself into your thumb,

02:01:30   you'll barely notice.

02:01:31   Beep. Beep. Beep.

02:01:33   .