617: An Incredibly Dangerous App
01:59:03 Share

I just barely made it here. I had to rush, pair all the socks and the laundry before I came in here. I thought you were going to say you had to rush home from Manhattan, you're living the fancy lad life or something like that. Nope, just matching socks. On the bed was a whole bunch of unsorted socks. And usually, as I've discussed in the show, my roles in the household are kitchen dog, wires, and puke.

And Tiff is soft things, plants, school, and fun. Now, soft things would typically be her purview. So she does all the laundry. However, there are occasional exceptions that poke holes in those boundaries because one person really hates something or doesn't mind something. And one of the things is that Tiff does all the laundry but really hates pairing socks. So that is my job. And so I wanted to hurry up. You can't do that in a hurry, though. If you do it in a hurry, you're not going to match them up right. You underestimate my skill, John. Yeah.

No, I think you underestimate the amount of stuff I think is involved in matching socks. I think you both underestimate the sneaky, sir. And that's a reference that I bet John didn't even get. No. What now? What now? All right, we'll see if ATP References gets it. Yeah, no. So the thing about matching socks is...

It's not so much just finding the sock that is the same sock as that one. They go in pairs in terms of wear and age, you know what I mean? And right and left. So you really want to pair the ones that have always been together to be back together. Even if you have, you know, 16 of the same socks that came in a pack when you bought them all, you can't just match them up willy-nilly with their other socks when they come out. They have to go back with their partner.

See, I solved this in a different way. You only buy two of a sock ever. Yeah, well, so that's okay. Yeah. So one of them, yes, one of them is my favorite socks are from former sponsor Marine Layer. And they're one of those things where like, you know, they'll have like five pairs of socks in stock at any given time. And they just rotate out all the time. So you kind of have to buy them if you see them, if you like them. And so as a result, I have all sorts of different socks from Marine Layer. And I try never to buy two of the same pair. So I actually don't have two pairs.

identical sets of the same socks, except for the summer short socks, which is just like some big Adidas bulk pack of ankle socks I bought. So far, those are just so interchangeable that it doesn't really matter. They're just all a bunch of gray socks. That's what I'm saying. You're not as discerning. You're like, ah, you've seen one Adidas athletic sock. You've seen them all. They can go with any of them, but no, there are pairs. There aren't. Not for those. But the other thing is what really dramatically shrinks the time for, uh,

any kind of laundry folding, especially with socks is, um,

I used to do the service to my family of turning them all the right way as part of this process. Is that a service? That's just called folding laundry. Are you putting socks in your drawer inside out? Well, if you ball the socks up, it kind of doesn't matter whether they're inside or not. And so I've decided that when I remove my clothing, not to get TMI listeners, but when I remove my clothing, I leave it in the laundry, the right side in or out or whatever.

So my laundry does not have to be turned back because I am conscientious on the way into the machine. So all my laundry comes out, for the most part, the right way. So my socks are all the right way. And the rest of my family's socks, I just fold them exactly the way they left them.

A, that's spiteful. And B, if you put them into the washing machine inside out and they have crap on them, the crap is going to stay stuck inside the socks. Like if you get a bunch of little bits of leave or wood chips or whatever stuck to the bottom of your sock and you take it off inside out and it goes in the washing machine, that stuff's not coming out. It's going to be right in there when you turn it the other way. So you're not even getting your stuff clean. Hey, my stuff goes in the correct way. Well, that's, yeah. My stuff gets cleaned. Everyone should turn their clothes right side out before they put them in the hamper.

I agree, but not everyone does. Well, see, one of the services you provide by doing laundry is turn them right side out before you put them in the washing machine. Yes, that's what. Well, yeah, when I'm when I take my own clothes off and put them in laundry, that's what I do. But the rest of my family, hey, that they can do what they want to do. They're people. All right. It's America, man.

All right, let's do some follow-up. I know everyone is waiting with bated breath to know how football went this past Sunday. The Giants lost, as they always do, but I'm happy to report that the system worked, and I was able to watch a game on Fox that was not in my local area, and it was pretty much flawless, and I'm very happy about that. So I know the two of you, especially Marco, the newfound football fan, were deeply concerned, and I'm happy to put your mind at rest. All worked according to plan.

And to be clear, this is the VPN method or is this some kind of weird cable home run thing? This is a combination of actually today's sponsor, Tailscale. It legitimately is using that. This isn't a clever plug for a sponsor. It really, truly is a legitimate, honest to goodness thing. But it's Tailscale, the channels app and TV Everywhere. A combination of those three things basically got me where I needed to be. So happy to report it did work.

Now, moving on, it is the holiday season, and so we are going to talk more about Festivitas, which is Simon Stavring's, I think, hopefully, holiday light app. And one of us, I think it was John, I'm not 100% sure, probably John, was wondering, how does Simon figure out how to hang the lights on the dock? And I'm sorry, I should back up a half step. This is a thing, it's just for fun, and it's really cute, where you can put holiday lights on your menu bar on your dock and whatnot. And John, I believe it was you wondering about how Simon figured out...

where to hang the lights, especially during like magnification and stuff. And I understand that you have answers for us. Yep. Simon wrote in to say festivious festivitas. Yeah. Festivitas. I think festivitas. Okay. Anyway, his app is not using a private API. It's using the accessibility API to read the frames of every tile in the dock. And based on those frames, create a CG path, uh,

This requires sandboxing to be disabled, so the Mac App Store is still a no-go. So this is one of the other unfortunate situations that I've talked about on the Mac before where there is an API to do something. It's a public API, right?

but it's old. Apple has deprecated it. And if you want to submit your app to the Mac App Store, you can't use this API. And believe it or not, the accessibility APIs on macOS fall into that category. So no, they're not private as in, you know, you're not supposed to use them, period. Although you could. But no, this is a public API. But Apple says, yeah, you can't be on the Mac App Store if you use the accessibility APIs. And it's basically because they're old and they're like,

their permission level is not granular enough i mean they're old and creaky apis to begin with you don't really want to use them but they do what you need them to do but they're too powerful essentially like if you get accessibility permission for your app in the mac app store you could do all sorts of nefarious things and apple just says flat out okay well unless you have one of these super secret private exceptions quote-unquote temporary exceptions you can't do that um

And I understand all that, except I feel like, OK, but how many years into Mac OS X or whatever you want to call it do we need to be before a modernized version of the accessibility API comes along that you can use and ship in the Mac App Store? So anyway, there you go. Not technically a private API, but still something that keeps the app out of the Mac App Store. The sad story on the Mac continues.

Truer words never spoken. So I have not tried this yet, but a handful of people have written in with regard to my forthcoming adventure in trying to figure out when my washing machine has finished washing. And a lot of people said, oh, you should use, you know, a power outlet thing that'll read how much power the machines are using. And that's reasonable, although for

Like, I don't love the idea of a big, you know, appliance going through one of these little dinguses in the wall. And so I was thinking about using a vibration sensor to figure out when the washing machine or the dryer is completed. And a few people wrote in to say, you know, another thing you can do potentially is use an accessibility feature called recognize sounds using iPhone.

And what this does is your iPhone can continually listen for certain sounds, such as a doorbell siren, a crying baby, and notify you when it recognizes these sounds. And there's a knowledge-based article that we will put in the show notes. This all sounds well and good, and it very well may work. I had intended to try it before recording, and I apologize. I just didn't have the chance. But the real crux of the issue is that our laundry room is actually upstairs, which is

Sounds kooky at first, but it's really convenient because most of your laundry is generated, so to speak, or put in the hamper anyway, upstairs. Oh yeah, that's the way to go if you can. Yeah, exactly. So it's one of those things where at first when we were looking at the house many, many years ago, we were like, what? And then once we had it, we were like, oh, this is excellent. This or a laundry chute. Laundry chutes are the best. But anyways, so the problem is often that the

the little siren-y doodad if it isn't set loud enough, which is a user error, or if we're just downstairs on the other side of the house. Again, as I've said many times, our house is not that big, but it's big enough and there's enough activity, especially with small kids, that oftentimes we just can't hear it. And I'm skeptical that my phone would be able to hear it, but I am definitely going to give this a shot and see what happens. But we'll see.

We have some interesting feedback from Ricky Haas. Ricky writes, well, it's true that the Apple store only sells SSD modules for the Mac Pro. Apple also has a first-party parts store, the self-service repair store for Apple products, which somehow is the domain selfservicerepair.com. I don't know how that happened. You'd think Apple would be in there somewhere. We talked about this back when we first rolled this out and we were talking about the giant expensive set of tools that you get that you have to return and everything. But yeah, that same thing, it still exists. They continue to expand it.

No, I mean, I think honestly, I think that was intentional. Like when Apple launched this, they did so extremely quietly and kind of like holding their arm out all the way, like at arm's length. Like here, you guys just like take this. Like I don't think they want a ton of people to know about this and they certainly don't want their brand associated with it. Yeah.

Yeah, so it's very weird. It's worth you loading the web page or website just to see how not Apple-y it is. But either way. So Ricky writes, it typically takes some time after a product launch for parts to show up, so they don't have components for the new Mac minis yet. However...

if they sell the same components as those available for the Mac studio, they should eventually offer storage upgrades. Unfortunately, Apple doesn't allow you to simply browse all available parts. You need to enter a serial number. And so I saw this and I immediately said, okay, well, I've got a serial number for my MacBook pro. This is a year ago, MacBook pro. It's a M3 max MacBook pro. And so I entered my serial number and I was, you know, just goofing off looking at different options. And unfortunately,

For my MacBook Pro, it would appear that the SSD is soldered into the logic board or is certainly not considered a separately removable and serviceable part. And so I looked, okay, well, how much is a logic board for my MacBook Pro? A logic board for my MacBook Pro, they bill you

$4,751.12. This is 64 gigs and 8 terabytes. $4,700. Now, wait, hold on. You did just gloss over those two numbers pretty quickly. Yeah, the 8 terabyte part. Well, it's fair. But still, I mean, the whole machine, I think, was like $5,5500 or something like that. Well, that 8 terabytes thing is made of tiny diamonds. I'm not sure if you know this. Apparently. Apparently. But,

In the defense of Apple, I then continued reading down a little bit, and it says replaced part return credit is $4,312. So actually, according to Apple, if I understand this right, the price is, and they list it right there, price after credit, $439.12, which for 8 terabytes and 64 gigs and an M3 Max is...

that's actually not that bad but oh baby you got to shell out five grand to get there and then they'll eventually give you a whole pile back not to mention the the process of replacing it would probably not be simple right exactly i wonder if you could get a cheap eight terabyte uh nand that way like if something goes wrong because you don't does it say that you have to have like a broken one you know what i mean i guess i guess you have to i think well you have to return the the board but like yeah i i

i don't know how you i mean you probably shouldn't be defrauding their repair system by stealing the nand off the motherboard somehow like that's i don't think that's going to work no you'd return your intact one but you'd get a fresh one like you basically get a fresh uh logic board oh i see what you mean you know what i mean like just because it's it's so cheap obviously you don't pay for your own labor and probably you're screwing up your computer every time you open it up and close it back up but

That's what you get for having a laptop. Right. So anyway, so I went on a little side quest there. Let me get back to what Ricky was writing. So Ricky writes, and let me repeat, unfortunately, Apple doesn't allow you to simply browse all available parts and need to enter a serial number. So Ricky continues, I managed to find one from a Mac studio via a YouTube unboxing video, which makes me so happy. But anyway, and voila, NAND modules are available directly from Apple.

And so Ricky was kind enough to send some screenshots and various different modules are available. 512 gigs, one, two, four, and eight terabytes. The cheapest at 512 gigs is $395 and 12 cents. And then you get $44 back if you replace the return part or, and then you get, so that means the sum total is $351 and 12 cents for 512 gigs. On the flip side, eight terabytes like yours truly, $2,000.

$1,464. Then you get a measly $369.60 back. So your price after credit for an 8TB unit for Mac Studio, $2,094.40.

Yep. This is a profit center. I mean, look, this is like when you buy parts for an expensive car from the manufacturer. Like if you're going to have some kind of high-end sports car, if you want to go to the manufacturer to buy their brake pads, they're going to cost a heck of a lot more than everyone else's brake pads. But that's kind of their privilege. And some of it is like, well, we have very high specs for our brake pads, but a lot of it is just their profit, right? That's the business model, and that's what this is too.

And actually in the realm of expensive cars, the parts, when you buy them aftermarket from the manufacturer, like the original actual genuine parts,

way more, way more than you can possibly imagine. Like if you add up the price of all the parts that you would have to buy to build the car that you just bought, it would cost just many, many times more. Whereas these prices are merely pretty much exactly the same as the insane prices that Apple charges when you buy them. You know what I mean? Like I think the prices are very close. Like maybe they're actually a little bit cheaper, but

But as we saw before with the Bugatti headlights, it just doesn't make any sense if you look at the prices. And it's because, especially with very fancy, expensive cars, they don't want to keep a lot of parts in stock. And the few that they do keep in stock, to make it worth their while to keep any in stock, they charge...

huge amounts of money for one little piece of plastic one you know forget about like panels and body panels in the car just any little thing in the car there's like seven of these in the world and if you want one you have to pay a huge amount of money so be glad apple merely charges normal apple prices for its parts

One of my favorite kinds of product is one where there's something that's just generally trash and someone comes in and says, no, no, no, we're going to do this right. We saw this with the iPod as a great example. You know what another example of this is? And I really mean it. They didn't ask me to say this. It's

Aura Frames. That's Aura, A-U-R-A Frames. What these are, are a really great, well-designed version of those really chintzy photo frames we got like five, 10 years ago that would play, that would show digital photos. And we're all trash back then, right? Aura has done it and done it right. They're so incredibly good. And there's several different options to choose from. And they've done everything so right that one of the things you can do is like crack open a little cover on the

box for a frame and there's a QR code or something like that that's in there that lets you pre-hydrate the frame so that the moment it is opened by the person you're giving it to, it will automatically connect to their Wi-Fi network and start downloading photos. They have thought of darn near everything. It really truly is extremely well done. We have one of these in the house. I freaking love it.

Every time I go into our living room where it's mounted on the wall, I look up at it and I see just a delightful picture of Aaron or the kids. And it makes me so genuinely happy. We sent one to my parents. And you think I'm happy about these things. Oh, boy. My parents love...

seeing pictures of their grandkids and their friends, and I guess their kids too, scrolling through their living room as well. It really is incredible, incredible stuff. They have a bespoke iPhone and Android app where you can beam things to the frames from afar. It really, really, truly is excellent. So you can save on the perfect gift anywhere.

by going to oraframes.com to get up to $35 off Aura's best-selling Carver matte frames by using the promo code, guess what, ATP at checkout. That's A-U-R-A-F-R-A-M-E-S.com, promo code ATP. This deal is exclusive to listeners, so get yours now in time for the holidays. If you act real quick, terms and conditions apply. Thank you to Aura for sponsoring the show.

All right. More from our friend of the show, Jonathan Dietz Jr., with regard to SSD performance. Here again, there's a lot here, and I'm going to read pretty much all of it. So bear with me, but it's really fascinating. So Jonathan writes, the state of SSD benchmarking on the Mac is pretty sad. Nobody should be making buying decisions or even be particularly concerned about the sequential read and write speeds of Mac SSDs. Let me pause that for a second, because thank you, Jonathan, for saying what I always try to say whenever we talk about benchmarks and what used to be...

One of the main activities that I participated in online, which was technical people complaining about how benchmarks are not representative. We'd argue about Mac versus PC and CPU benchmarks and Altevec versus MMX, right? And it was all about like, well, you know, byte marks versus spec int versus this. And it was people arguing that benchmark is not representative.

a good one it can be gamed it can be cheated it's not representative of real performance or okay well i'm playing a game i get better frame rates than this well that that it has a good port on this platform but a bad port on that platform so that's not good a benchmark either all we did was argue about how benchmarks didn't mean anything and now in in the modern age go to any youtube video listen to any podcast and you're just like oh well this is what geekbench says and this is what the benchmarks say and nobody argues about it anymore they're just like yeah the benchmark says this is faster so i guess it must be faster

There's so much to be argued about these benchmarks, especially with things like Geekbench, which we really don't have. I mean, you have some knowledge of it. You can get kind of into technically, but it's nothing like the open source spec benchmarks were back in the day or like running games that are open source that you can compile for any platform and then it becomes like a compiler benchmark. So kudos to Jonathan for doing what was once a great pastime and has now fallen by the wayside, which is complaining about benchmarks. They're no good. Yeah.

Gracious. The results of the two most popular benchmarking tools, Blackmagic Disk Speed Test and AJA, AJA? System Test? AJA, maybe? AJA, are akin to comparing quarter-mile times for a car. Amorphous disk mark is better, but still miles from the type of in-depth testing that Nantech, rest in peace, used to conduct.

All of them appear to report optimistic write speeds, probably due to the effects of write caching by the system. Perhaps the biggest problem with these numbers, though, is the inability to identify the actual hardware being tested. Knowing the specific Mac model and disk size being tested is the bare minimum for making these results comparable.

Ideally, you'd also want the IO reg entry for the controller characteristics property of the Apple ANS3 NVMe controller or Apple ANS3 CGV2 controller object. Man, I don't really miss Objective-C naming. That is not something I miss in my life. Anyway...

The property provides many details about the underlying architecture of the SSD. In particular, the dies per bus key reveals the total number of NAND dies in channels from which we can infer the number of NAND packages. And the NAND marketing name key tells us the type of NAND used, number of planes, and density.

All you have to do is pop open the terminal and run the following command. No teardown necessary for M1 through M3. IO reg. Well, I'll just put in the show notes. It's a bunch of stuff. We don't need to read commands and strings online. This is an IO reg command with some flags with those two names that were read earlier. This is like reading a URL out loud. Yeah. I caught myself. I caught myself. Don't worry. Um,

So anyways, Jonathan continues, the only thing that people should be concerned about storage-wise when buying a Mac is selecting the right SSD, right size, excuse me, SSD for their needs. In most cases, selecting the next higher capacity drive will double the endurance and improve performance, but there's no underlying problem with the speed or endurance of any of the SSDs found in any of the M series Macs at any capacity. However, if you fill the drive, performance can be severely impacted, so you really do want to avoid that.

For those planning on using their Mac for eight or more years, as I generally do, 256 gigs may become a pain point even for casual users. Fortunately, Apple will likely transition to one terabit NAND dies as soon as next year, and the 256 gig capacity will go the way of the Dodo, just like the 8 gig as the entry-level RAM level did as well.

Hallelujah. We can't wait for that to happen. Although I still feel like a 512 might be small. But I think this rings true to me in all of our, you know, over a decade now doing this show or whatever. You don't hear a lot about people's SSDs going bad. I mean, they do. It does happen. Everything goes bad. Everything fails. But Apple has not historically had...

as far as I know, any Macs, especially in the ARM Mac era, any Apple Silicon Macs where the SSD was really bad. I remember there was a big kerfuffle that we talked about on the show back when I think the M1s were first out, when they were like, oh, you buy it with too little RAM and it's swapping to it, and there was something that tried to show how much life is left, how much endurance is left on the flash memory based on how many writes it's getting, and some tools reporting really huge numbers, and like, at this rate, your SSD will be dead in a year and a half, and

That turned out mostly, I think, to be a problem with the tool reporting bad numbers and a problem with interpreting the numbers that did return. And in the end, people with M1 Macs didn't all have their SSDs fail after the first year, right? I mean, we're still using one right now, and it's fine. But I also agree that one of the things...

Well, macOS doesn't like this, and add to the pile now, NAND flash drives don't like this. Don't get close to the disk storage limit. Bad things happen. Bad, upsetting, terrible things happen. The operating system doesn't like it, the flash memory doesn't like it, it's just, you're just asking for trouble, which is why it really is important to get as big an SSD as you can, or be very studious about getting stuff off of that SSD, because unfortunately...

Mac OS is not great about helping you when you start pressing up against your disk limit. It will just sort of like complain feebly for a few seconds until it's too late and then just keel over and it'll be a sad situation.

So we got some repeated news, as it turns out, over the last 24-ish hours. M4 Extreme's chip, or the M4 Extreme chip, is unlikely after Apple quote-unquote cancels the high-performance chip. Reading from MacRumors, who is in turn quoting from the information, says,

Apple is working on AI chip with Broadcom is the title, the headline of the post on the information. Apple Silicon design team in Israel is leading development of the AI chip. That team was instrumental in designing the process as Apple introduced in 2020 to replace Intel chips and Macs.

Apple this past summer canceled the development of a high-performance chip for Macs consisting of four smaller chips stitched together to free up some of its engineers in Israel to work on the AI chip, one of the people said, highlighting the company's shifting priorities. To make the chip, Apple is planning to use one of TSMC's most advanced manufacturing processes known as N3P. Apple plans to manufacture at least one of its iPhone chips next year using N3P. The other upcoming AI chips designed by OpenAI and NVIDIA are expected to use the process as well to improve performance.

For its AI chip, Apple plans to utilize a chiplet design that AMD pioneered more than a decade ago. Rather than build a single chip with sections for different functions, Apple will break up the chip and functions into smaller pieces or chiplets and then stitch them back together as one. The design reduces the chip's manufacturing complexity and potential for defects. So this was a...

Kind of I don't know if this is a separate story or just confirmation of the vague rumor we got I think multiple years ago, which was I think this was back when the m1 Extreme the the big m1 the quad m1 chip was cancelled And that was just an anonymous tip from I don't remember where the sourcing but it wasn't any particularly reliable source But just something we were told and it was Hey, if you're waiting for something on that size the earliest it could possibly appear is the m7 and

And that was back when I think the M2 was just on the horizon or coming out. So M7, no chance of any chip like that until the M7, given the lead times. To jump in really quickly, that I believe the original source of that was actually Upgrade, if I'm not mistaken, because someone had anonymously written into Upgrade that

And because this was covered in Gruber's coverage of all this, he reminded me of in his post, it reminded me of the feedback that Upgrade got, wherein someone allegedly with inside knowledge said, hey, this isn't going to happen until the M7 at the earliest. And I'm thinking maybe this team in Israel was the team working on the M7 thing.

And then they just got pulled off. And here's the thing. Many things are thwarting this chip. The fact that it's very expensive, the fact that Apple doesn't want to do it for such low volumes, you know, like this, you really need to believe in something like this to make it happen. It's, you know, well,

As my blog post said, the only way cars like the LFA happen in a company like Lexus is a bunch of people really, really want to make it happen because it just does not make economic sense. But it makes sense for the company, and you need to have enough people who believe in that. That's the whole point of the Mac Pro Believe shirt, which will be coming back someday because its job is still not done.

back pro belief shirt will be back next year wwdc well anyway uh once again apple has found a reason not to make this chip uh which it's pretty easy to do and in this case it makes some sense like obviously apple you know ai is a big thing apple doesn't want to just pay amazon to use all of its chips although apple recently i think someone from apple spoke at like the aws reinvent to talk about how they're using all of amazon's ai training chips and they're loving it and blah blah blah right so

Apple's doing what it has to do, but we also know from past stories we've discussed in the show that Apple has its own Apple Silicon and its own data centers running its own AI workloads with the private cloud computing thing that they made. Supposedly a bunch of M2 Ultras, so on and so forth. Makes perfect sense that Apple would decide to use its own Silicon in its data center to run AI workloads. The M2 Ultra is not a chip that was really designed for that. It's designed to be in Macs.

uh making a purpose-built ai training inference whatever purpose-built uh apple silicon for the data center for ai workloads makes sense and i think apple will do a good job at it the rumors are that apple is going to make that chip broad comma self-limit yada yada they're not as usual apple will not sell this chip to other people will just be a thing that apple makes for its own data center so no customer will ever see this

They're not going to sell access to it like Amazon and Google and stuff do. It's just going to be for them. It'll probably be big and complicated and expensive. But what it won't be is a Mac Pro type chip, because obviously a lot of things that the Mac Pro needs to do, a chip inside a data center doing AI workloads does not need to do.

And if you want people to make a big monster chip that can do this, probably the team that was working on the Mac Pro chip has the people that you need. And it's not like you can just leave them there and hire a bunch of new people. So I understand from a business perspective why they're doing this. And yet, if you're ever going to make a chip that makes sense in that gigantic case, you got to actually make one someday. Or is it just going to be a series of Mac studios in a giant case for the next five years? Stay tuned and find out.

I mean, they're also, in all fairness to them, this might not be purely economic. This might also be, you know, maybe they made these prototypes and they weren't as good as they thought. Because if you look at the Ultra chip, which is the two Maxes, you know, talking to each other over the silicon interposer, it does not scale perfectly in all useful ways. Right.

There is some degree of overhead. The scaling is not complete. You don't get 2x of everything when you do that. You get close to 2x in certain ways, but sometimes more like 1.5x in other ways. So obviously they're not actually achieving perfect scaling. There's other bottlenecks somewhere or the limitations. Maybe when they try to 4x design...

it just wasn't worth it because maybe a 4X design, rather than reaching 4X performance, maybe it only reached 2.5X performance or something like that. Like maybe it just wasn't technically compelling and, you know, would you pay for,

four times as much or whatever. I'm sure the cost would be absurdly more expensive. Would you pay that much for, quote, only two or three times the performance? Maybe not. And maybe they decided the better way to go was maybe to wait for something like a chiplet design where they could do things like the parts that scale well offer more of those, like the GPU cores maybe, and maybe not...

duplicate the entire chip design and all the other ways in order to achieve more GPU cores. It was like, when you look at CPU performance, we're pretty like the CPU performance, I think is probably what scales the worst as you add more cores here because of various limits and stuff.

What the Mac Pro needs to be more competitive with its previous market as well as the current market, what it needs is more neural engine cores and or more GPU cores for certain configurations. So if they're moving to a system of chiplets, which sounds very promising, where they're able to

isolate these components a bit more and to be able to say make a chip that keeps the same cpu core count as an ultra but gives you twice the gpu core count for instance and to make that somewhat economically you know it's all it's all relative of course but you know somewhat economically that would be more compelling than if you doubled everything and didn't quite get you know all of that performance so anyway the

I think I'm guessing the reason they canned the first quad was not purely economic. I'm guessing the reason it hasn't come back is also not purely economic. And if they are going to bring something like that to market, I bet it's going to take a very different strategy than just give us two ultras glued together. I think it'll be more customized.

I was talking to Gruber about this. I'm way behind on Twitter, or Twitter, sorry. I'm way behind on Mastodon for reasons we'll get into in a little bit. But I was talking to Gruber about this, but I didn't read his article about it. But anyway, one of the things I was saying was that, like we've discussed in our past episodes –

I wish I could see what the M4 Max Die looks like because if it doesn't have an interposer, that means they're not going to stick two of them together to make an Ultra. And yeah, the non-linear scaling of the Ultras versus the Maxes has been a problem. I think the M1 Ultra was the one that had really bad GPU scaling. Like the GPU was the one thing that should scale close to linearly and it wasn't because of interconnect issues.

So what I've always been hoping was that the M3 or M4 Ultra will not be two Maxis stuck together, but instead will be a dedicated custom chip that just has more stuff in the right things. Single die, you know, single massive super expensive die, but not two Maxis stuck together. Because the two Maxis stuck together approach made some sense economically and it was worth a try, but I think with two tries at it with the M1 and M2 Ultra...

It's a little bit wasteful. It doesn't have the greatest performance benefits. And that's not always the worst thing, though, by the way, because...

When you're shopping for high-end stuff, very often you are paying much, much more for just a little bit more performance. That's why it's so important to essentially make the fastest whatever that you can make. Because you can say, well, look, it is five times more expensive for only two times the speed, but it's literally the fastest thing that you can get. So if you want to spend money and get the fastest thing, if that makes a difference to your business or to your job or whatever, but you just want the fastest, you don't care that the price doesn't scale linearly with the performance. You just want the fastest. But you do want something that...

you know, performs as good as possible. And part of that is not wasting performance and money on overhead. That's not giving you better performance. So, uh, you know, one of the things I was hoping before reading the story was that the M4 ultra would appear and it would not be two M4 max is stuck together, but it would be its own dedicated chip and it would be really powerful. And then the extreme would be two of those stuck together.

You know what I mean? And so that they could design them in that way so that that gives you a more economical extreme because you're just taking two ultras, but the two ultras have their parts arranged such that it makes a little bit more sense to put them together. Yeah, and the chiplet design, once you get to a certain size, like

The Ultra with the interposer, that's basically pushing the limit. We talked about this before, the reticle limit. Like, how big can you even make things? Not just economically, but practically speaking at all. Physically. Yeah, so everyone's doing a chiplet-type design. Yeah, we keep referring to it at the Quad because we're basically trying to say something twice as big as an Ultra, but it doesn't mean four of anything. It just means count up all the stuff. Count up all the transistors, all the die area. Something that is twice as big as an Ultra, right?

That's when we're doing all the back-of-the-envelope math, like how many GPU cores and what it would be competitive with, assuming the GPU scale is close to linearly. So I'm not holding out for four of anything stuck together. I just want something that is roughly sized like four of those things with a much better arrangement of parts, less used on overhead. I don't need four copies of the Thunderbolt controllers. I don't need four different media encoders and only one of which is used at a time. You know, like, they can do better. But anyway, it just seems like this is not going to be in the cards for a long time. So now...

The only thing I have left to hope for is that the Ultra will be better than the previous Ultras, hopefully by not being too Maxis stuck together, but just being a dedicated Ultra that is its own thing. Or if it is too Maxis stuck together, that they've resolved the issues. Again, my memory's fuzzy on this, but I seem to recall the M1 Ultra, the GPU scaling was really terrible because of some thing that they didn't do right. Some buffer or something was under-resourced somewhere, and it really killed the performance. And that's just a shame because...

The GPU transistors were there. They had enough of them. If they were all in the same die, they would have worked great. But because they were separated and because there was a problem with the interconnect, you lost a lot of performance. So we'll see. But yeah, this is, you know, not that it was ever likely that I was going to get the next Mac Pro, but it's definitely looking more like Mac Studio for me. Hopefully the M4 Ultra will be good. It'll be weird for you not to have a Mac Pro. I mean, I don't think it's unreasonable for the record, but it'll be weird. I mean, I'll just, I probably will just keep it here.

Because the studio, like it can continue to be here like a piece of furniture and the studio will just be on the desk. It's so small. It doesn't take up any room. I mean, if you get a second one, you can make them into a desk. Just take, you know, those are the two legs. Get a tabletop or a door and just stick it across the top. And there you go. That's what people do with the old cheese graters, but they're a little low and I would never do that. This beautiful computer. Those are better for benches. Yeah. We are sponsored this episode by Delete Me.

Now, it's really fast and easy to a creepy and scary level to just do a quick web search and find your personal information out there on the web. Because there's all these data broker sites that gather your information often without your consent. They trade it, they sell it, and they publish it.

And so anybody can just do a quick web search and find your home address, your phone number, your relatives' names and addresses and phone numbers and emails and all this stuff. And this is both creepy and even could be used to scam people and do other bad stuff. So...

This is not great. DeleteMe is a service that removes your personal information from hundreds of data broker websites. So it gets much harder to find your info out there. They go in there and get your details removed. They do the hard work for you, ensuring that your family's information is much more private.

So this holiday season, you actually might want to give this to your relatives as a gift because their information is out there too. That can be out there sold online without you even knowing it. So again, home address, phone numbers, email addresses, those are all floating around for you and your relatives. And so you can actually give Delete Me as a gift of

of privacy and peace of mind with delete me, of course. So now this is actually available as a special discount for our listeners today. Get 20% off your delete me plan. When you go to join delete me.com slash ATP and use promo code ATP at checkout.

The only way to get 20% off is to go to joindeleteme.com slash ATP and enter ATP at checkout. That's joindeleteme.com slash ATP, code ATP. Thank you so much to Delete Me for sponsoring our show. All right, John. So potentially in preparation for you getting a Mac studio, you've been doing some house cleaning. What's going on?

Yeah. So we talked on the show. You're so pained by this. Yeah, I know. Seriously, I'm scared now. I can't remember. I think we were talking about like, maybe we were talking about our photos workflows or something on the member special. But anyway, to refresh everyone's memory, I have a shared iCloud photo library thing. I love it. My wife and I are both in the same shared library. Essentially, all of our photos are in the shared library.

I think pretty much all of her photos are in there. I keep some photos out of the shared library. They're mostly screenshots of videos about destiny stuff. My wife doesn't need to see that. So it's like, yeah. So those are my, but I have a very small, and also the pictures of the backs of appliances and stuff like that. Those are in my personal library, but for basically everything else is in a massive 200,000 plus photo shared library. Right.

And because Apple hasn't really finished the fight in Halo parlance on the shared photo library, some things are still not shared, like albums and people. And so I still need to switch to my wife's account on my Mac sometimes to do photo stuff, because if I were to do them in my account, it wouldn't have any effect because they're not shared in the shared photo library.

So anyway, she has an account on my Mac. I switch to it. Her account is set to download originals. So our full photo library is on my Mac through her account because it's set to download originals. It gets backed up with the whole rest of my Mac, right?

but also on my account is my photo library. And when I launch photos, I'm essentially looking at the same photo library that she's looking at, the shared library, right? She's set to download originals. My account is set to the optimized storage thing. But practically speaking, there's 200,000 photos. And when you go to her account, you can see those 200,000 photos. When you go to my account, you can see those 200,000 photos. And somebody wrote in a while back, and I think we talked about it on SKTP or something, asking, does macOS do, or does photos do anything smart?

with the fact that it's the same shared library and two accounts on the same Mac? Or if I set my account to download originals, would it download two complete copies of my library? And I told this person, I think it will download two complete copies of your library. And it's incredibly wasteful that it does that. But, you know, that's a pretty tricky scenario, right? Anyway, we had that discussion a while back, but it had been sticking in my mind going, you know, I wonder, especially as my disk fills and I get close to my storage limit and I'm always like fighting against that.

How much space am I wasting on my Mac by having the same photo libraries, hers downloading originals and mine not? But like, how much stuff is it downloaded?

So, you know, I've got Pearl. I can find the answer to this question. Oh, my word. Of course you turn to Pearl. Why would you, why wouldn't you just use Finder? But no, Pearl exists. I'll use Pearl. Well, I can't use Finder because it's in her account. Why not use PHP? I mean, you now know that. Gouge my eyes out. Using PHP for an extended amount has made me hate it more if that's even possible to imagine, but it's true. But it has happened. Anyway, so I wrote a little Pearl script and it ran at his root. Yeah.

He said, what could possibly go wrong? Right. I've been running Pearl as Root since before you were out of high school. Anyway, ran my script and it crawled over her photo library, my photo library, and found all the files that are identical. And I was actually, well, first, here's the file count. Any guesses?

Oh, if you include like, are we including all the little previews it generates and everything like that? I'm not excluding any files. Oh my god. There's like a directory called whatever.photolibrary.photolibrary is the extension, but it's a directory. And in there is a bunch of stuff that Photos uses to keep track of your files. So obviously the originals are in there in my wife's account, but also all the other crap of the thumbnails and previews and stuff like that. I'm going to say 2 million. No, identical files, not files total. Oh, okay. Yeah.

Oh, let's see. I think hundreds of thousands. Yeah, I mean, I'll say like 200K. 73,000 files were identical. Okay. That's actually fewer than I thought. But now keep in mind, my photo library is, how big is it? Maybe like two and a half terabytes, three terabytes, 200,000 photos. That's how big it is. So it's like, oh, I got to see some major savings here. Turns out there's only 8.6 gigs duplicated.

Now, I don't know if that's because I kind of stay away from my photo library and don't like scroll back in time and like load the full resolution images and stuff or whatever. I mostly just look at the pictures that have just come in. You know, I'm not really going back in time and doing stuff, but only 8.6 gigs. But I was like, you know what? I want those 8.6 gigs back. Oh, here we go. You couldn't just leave well enough alone. No, because that's the whole point of this. The person was asking like, does the OS do something smart about this? Like, does it share them and only have one copy or whatever? And I was like,

I can get those 8.6 gigs back. Can I guess how? You should know how because we've talked about it on the show before. But yeah, go ahead.

Are you going to use the special CP copy command that does the APFS deduplication thing for the same file? Yeah, I'm not going to use a CP command like an animal. But yeah, APFS, the file system that macOS and iOS and all the other devices use, Apple's file system, has a feature that lets you make a quote clone of a file where you get a complete independent copy of that file essentially for free.

Because it shares the storage with the other file. And not like a hard link or a symlink or an alias. Not like that. Not like where when you edit one file, the other one gets edited. They are two completely distinct files. They do not affect each other in any way. But while they are identical, they share the same storage.

If you duplicate a file in the Finder, go to your Finder now, find a 10-gig file, and make a thousand duplicates. You will not take up any more space except for the space required to store the file names of those files. That's why duplicates in the Finder are instant. Go to find a 300-gig file, hit duplicate, and be like, why was there no progress bar? You know why? Because it's just cloning the file. It's saying...

Those blocks on disk now. There's another way to get at them and again It doesn't tie them together like a sim link or an alias or a hard link They're not linked in a way when you edit one file. It will edit all the other copies. They're completely independent They will diverge as you edit them. It's the magic of copy on write It's one of the advantages of having a modern file system and most people don't even think about it But the CP command from the command line will do this Although there is a flag to tell it to do it if you don't even pass the flag It does it all the time anyway

The finder will do when you duplicate. That's why your file copies are instant. I do it all the time for podcast recordings. When I record a podcast, whatever folder ended up in audio hijack for this show or whatever, I copy it into a like separate folder, which is just like all the shows I've recorded, like from going back in time. That doesn't take up any more space. I just copy it. I also copy it into Dropbox. I'm not sure how Dropbox handles that, but it gets deleted at Dropbox pretty soon. But anyway, I can get the space back.

Take those 73,000 files and every one of them that I find where there's identical content in my photo library, clone the one from her photo library into my photo library and throw out the one that was in my photo library.

Because they have identical content. And you've got to do it carefully because just because they have identical content, they're still different files. They have different creation dates, different modification dates, different owners, different permissions, different finder labels, different comments, different access control lists, different extended attributes. Tons of stuff about these files is different, but not the data. Different resource forks, maybe, but not the data. I love that you're going through all of this on a computer. This isn't on her computer, right? This is on your computer? Yep.

why you could put another frigging hard drive in your machine. Why are you going through all this pain? Well, I mean, it's not paying that the first Perl script to find out how many files were the same real quick, like, you know, 10, 10 minutes. Right. I mean, I think it's overkill, but I I'm behind it in principle, but doing this, like, you know, swapping in place within the photos library, mega, mega air quotes database. Oh, no. So that's, that's one of the things is like, okay, well, how can I do this in a way where I feel like I'm not hosing myself? Right. Yeah.

aside from just being a good careful programmer with experience with Pearl. One of the ways I decided to tackle this was I will choose her library as the source and mine is the target. So I will only read from hers and only write to mine because hers is like the canonical good one where it's downloaded originals, blah, blah, blah. And mine is not. Worst case scenario, I just delete my photo library and re-iCloud it because there's so little is actually in my photo library proper. It's all in the shared library, right? Yeah.

So that's what I did. I used hers as a source, mine as a destination to copy the files. I did it in a couple of different steps. One is I needed to be able to find out if the files were already clones of each other. I forgot to mention this, but I did that as part of the Perl script. I couldn't find a good way to do that, except someone eventually, I eventually found someone on GitHub who had a tool that

did a best effort at it and as far as I could tell it worked. They did a bunch of test cases and always got the right answer. It essentially walks the entire extents tree for each of the files and sees if they're identical which is, it's not great but it's the only way that you can do it. There's a bunch of little flags and stuff about this even one called like

uh may share content or may share blocks like may well that's not really helpful to me i need to know if they do share blocks right so you basically just have to walk the whole extents tree so i use that was written in c i have my thing written in pearl there's a command on the mac called ditto that will copy a file with all of its metadata and stuff and then i wrote uh in swift a thing to do the extended attributes creation dates blah blah blah some things you can't restore because i think there's a thing called like a access uh

access time or attribute modification time that you can't actually update. Like it gets updated by the file system and you have no way to, there's no APIs for updating it, but I don't think anything really uses that. So anyway, I was comfortable that it would be safe. I did a test on one or two files. I looked at before and after as I examined them with every tool I had and said the before and the after file really do look identical in every way that I can determine that I care about. And so I just ran it. I got 8.6 gigs back. I'm so happy for you. And then,

I thought, you know, free disk space. Oh, no. Oh, no. Oh, no.

free disk space oh no john john john john i did my my computer has not lost anything obviously there's there are fewer redundant copies of bits on the ssd but i'm not relying on the redundancy of the bits on my ssd to save those things that's what backups are for oh god raid is not a backup neither is having multiple guys but anyway free disk space this is this is like that's like free opium what could possibly go wrong real estate i'm not doing free opium free real estate free disk real estate um

I thought to myself, most people don't know about this APFS cloning thing. It's too nerdy. But everyone likes free disk space, right? Like literally free. If you go to the Mac App Store, you'll find tons of like file deduplicators that will find duplicate files, but it wants you to delete the other ones. Like they all offer like, and I will delete them. I'll save the newest one. We'll save this, you know, but like, no, cloning doesn't make you make that choice. You don't lose any files. All the files stay exactly the way they were.

John, are you about to launch an app? Don't do this. Don't do this to yourself. The support burden will be terrible. You can have all sorts of creepy marketing about it. You can get free resources from your computer. Just give me $40. But it's a technical ability that people aren't leveraging on their own because it's too onerous to do it. I wouldn't find 73,000 files by myself. So anyway, I thought this would be a good idea for a Mac app. Oh, no, John. So I wrote it.

Oh my God. Are you launching it like right now? No. Well, here. So I, because I, I've written the Perl scripts. I'm like, this couldn't be that hard to make as a Mac app. Don't use the Perl scripts in the app for God's sake. Oh, come on. The whole point of making is an app is no more Perl scripts. No, you know, like real, real honest to goodness app. I made, you know, cause I like doing Mac development and my apps are essentially feature complete. Although I do have a couple ideas for one of them, but anyway, I,

It was exciting to start a new app. I made a SwiftUI lifecycle app, you know, started from scratch, wrote it up. I did the programmer's UI that you're all familiar with, which is a giant window with a bunch of text and buttons on it. No UI and no styling, just text buttons and rapidly changing numbers, right? Nice. Love it. Love the programmer UI. And yeah, no, it worked. And while I was doing this, I was like, you know,

I probably someone already wrote this program and I should just like stop what I'm doing and use theirs. Eventually, I remembered that.

Someone did write the program. We talked about it on the show. I feel like Casey now. We talked about it on the show. I already had it installed in my applications folder. I literally, it's a pay app that I bought years ago. It already exists. So I will put a link in the show notes. It's called Disk the Dupe by Frank Schroeder. I'm not sure how many years old it is, but it does exactly what my app is trying to do.

It's not the most polished app in the world, but it's been developed for a little while. It's on version 1.3 or something. It's available on the Mac App Store. You can get it right now if you feel like it. But I'm still making my app. Other people made podcast players too, but it doesn't stop Marco. Yeah, that's true. He still made Overcast. I'm making the app just because I like doing Mac development. It's fun.

And I think there is a place in the market for a slightly more polished version of this. But I'm doing it for a couple other reasons as well. One is after I got like, you know, a thousand lines into this app and it was basically working, but with an ugly UI, I thought, you know, let me not make the same mistake that everyone else always makes. I should turn on Swift 6 language mode.

because i hadn't at that point and so that you know burned a day but still you do it when there's a thousand lines it's better than doing it when there's a hundred thousand right there and there's a lot i don't know about concurrency but i figured it out all right uh the other thing i'm concerned about is this is an incredibly dangerous app like if i'm not careful right

It essentially deletes and then reincarnates arbitrary files that it knows nothing about. What could possibly go wrong? Because that's what it does. If you point it at a directory, it doesn't know what those files are. It just knows these two files have the same content, and I'm going to magically do some magic with cloning, and when I'm done, you won't notice anything has changed. But while I'm doing it, I hope no one's looking at those files.

Unfortunately, Mac OS being a Unix derivative does not have a really great way to tell if someone else is messing with a file. You want to run LSOF on every file. It will take a year and a day and the kernel will get angry at you. And even that is not going to tell you anything because there's a million race conditions that you only have advisory locking. You don't have mandatory locking like on Windows. Yada, yada, yada. The point is it's a potentially dangerous app.

But then on the other hand, that's why a polished version of this that is very cautious and careful maybe has a place in the market because, no, people are never going to do this on their own. They're not going to run a Perl script. They're not going to run a Perl script as root like I did that calls out to a Swift program that they wrote in a C program that they downloaded from GitHub. Like, no one's going to do that. They just want a polished single Mac app that does all the stuff. So anyway...

I am going to eventually make that app. It is essentially working now. It just has half a UI. I'm noodling with the UI to try to make it not heinously ugly. But I'm having fun. And the other thing I'm going to do with it is... So the existing app that does this is a paid upfront app. I'm going to make mine free but within that purchase because it's so ideal. You get the app for free. You try it. It tells you how much space you can save. If you can't save any space, you're fine. Whatever. But if it turns out you can save 10 gigs...

then you get the in-app purchase. And that gives me an opportunity to play with in-app purchase, which I've literally never written in any app before. So I want to do in-app purchase in a Mac app this time to see what that experience is like, because it could be fun.

Yeah, I mean. Here's where you tell me a Storkit testing is never fun. I understand, but you know what I mean. Oh, God, it's not. It's gotten better, but it's real rough. It's Storkit 2, Casey. It's not just regular Storkit. No, no, that's what I'm using, and it's still trash. It's still trash. Like, it's gotten a lot better, but it's still trash. My end up purchase will be fairly simple. But anyway. Yeah, you say that. So the final thing to say here is I have a Casey-level naming dilemma going on.

on oh yeah good luck good because here's the thing with naming i uh i have uh very particular opinions about app naming everything on the mac but when you make a new project in xcode they make you pick a name for the project and if you don't know what your app is going to be named i'm

You got to call it something. This is how Casey ends up with his stuff I mentioned as well, right? Yep, it is exactly right. That is one of the reasons why I typically cannot start a new project until I have the name. Yeah. Like I just, I can't make myself do it with a placeholder name. Yeah, also here's the other thing about names. Depending on which naming theme I'm going with,

I very often have spaces between words, but if you put spaces between words in your Xcode project, it puts underscores and files and it's just like, Oh yeah, it messes it all up. And then if you go, okay, we'll put underscores and files, we'll just remove those underscores. But then your whole project is screwed up because you haven't find every reference to that file. And it's just like, you gotta, so obviously I made a project and I gave it a name. It's not going to be the name of the app. Yeah.

Is it Flisk space? It's not as bad as a Casey name, but it might as well be my thinking on the name. I talked about this briefly with Merlin on the rectifs that will come out in a little bit. Um,

My other apps are a little bit more fanciful in names, but this one I actually want people to find when they're searching for something that does what it does. Yeah, this is like an SEO name. I can't just call it like Lemonade or something that's just like, you know, just nothing that has anything to do with anything, even though it might be good branding to come up with like a very unique, interesting name. No one will ever find it. I don't trust the app store search and my keywords and my description or whatever. So I want the name to at least give you a hint that it might have something to do with disks or storage or...

or duplicate files or something like that. And that really puts you squarely into the namespaces of the million other apps that delete duplicate files that are on the Mac App Store. So I'm not quite sure what it's going to be called.

As soon as I know, I will let you know. But the app's not quite done yet. Can you think about words that are about finding resources out in the world, like mining, miner, stuff like that? No, you can't do mining. Yeah, that's like a Bitcoin thing. Any word like that, I would want to put like disk. But people don't have disks anymore. They don't even think of it as disks. It's certainly not hard drives. Yeah, I would use it. If you're going to refer to the medium, I would call it storage or space or something like that. But ideally, you shouldn't have to.

But I don't know. Yeah, that's a tough one. Any guesses at what my working title is? What the name of the Xcode project is? Oh, no. I don't want to guess. But I think it's got to have been overdone a trillion times. But the obvious answer for this is Highlander because there can be only one. Yeah, there already is an app called Highlander, obviously. One of the names I came up with, obviously it's not a real name that I would use for reasons that you'll understand in a second, SuperDDuper.

wow that's pretty good actually because dave nanyan would come to my house and beat me up yes but it's still pretty good though that's called trade dress if i try to name my app after our well-known beloved multi-decade old like yeah which is also about storage yeah which is also a super duper if he has people don't know is a mac cloning app that i in fact use to clone my hard drive so super duper there's also carbon copy cloner but anyway super duper would be a great name if we were the same company but we're not how about canonical

It's a terrible name, but it's kind of... Isn't that a whole software company? Yeah, people aren't going to guess that that's a disk. I might have to give up on the people can guess that it's a disk thing. Yeah, I don't know. That's...

That's tough. I mean, naming anything these days, you know, anything in tech is a very crowded market these days. And naming something that solves a common problem that has existed for a very long time, like freeing up disk space in your computer, that's going to be extra difficult because there's going to be so many. Space Doubler. Oh, wait. You guys don't remember Space Doubler, but it was the name of a classic macOS utility. Didn't do this, but it was another thing. I think it was like a compression, on-the-fly compression thing.

You know, Stacker is one of those, too. And the funny thing is that name actually kind of makes sense here. Yeah. Real-time follow-up. I knew I recognized Canonical from somewhere. They're the people who do Ubuntu. I can't believe I didn't realize that immediately. Oh, okay. Well, that's out then. That's why I knew it. So anyway, a few more things. Like, now that I have this program up and running, it's obviously a two-phase thing. One where you scan where it tells you how many duplicates it's found, and two is where you consolidate them. And so I've been doing a lot of scanning, especially when developing the app. I pointed at my Documents folder.

Believe it or not, only 14,000 documents, but 12 gigs. More duplicates in my own, just in my documents folder, my own documents folder only, than there was when I was cross-comparing the two photo libraries that ostensibly are both looking at the shared library. 12 gigs in my documents library. Dropbox...

I'm probably just going to literally forbid in my app, you ever pointing it at Google drive Dropbox or anything, because I have no freaking idea what Dropbox will do when I yank the file out from under it and replace it with a clone. Yep. I'm just going to be like, Nope, not anything. That's anything that's running on a file provider type API thing like Google drive, one drive. My app is just going to flat out refuse to do just for safety reasons probably. But anyway, I got a gig 626 files in Dropbox. Uh,

My home slash library folder, which I will also forbid for what I hope are obvious reasons, stuff that's in your library is like stuff that other apps are using that you have no idea when they're touching it, right? Things that are being touched by the CF prefs demon, containers, sandbox containers, whatever. 15 gigs in my library folder. 15 gigs of duplications. So that's the type of situation where you shouldn't run it, but I might run my Perl script on it. Yeah.

John, I am in full support of you doing this for yourself. I am at best half-hearted support in you selling this as a product because I feel like the support burden would just be awful. It will be ruinous. And then you're going to not do a bunch of work for ATP that I've been happy for you to have been doing recently. That's the whole point of this program is like the one I said, the disk thing or whatever that has been on the Mac App Store.

It doesn't really have any guardrails. It will just do what you tell it to do for the most part, right? And I don't think it's a very popular program. I don't really hear about it, despite the fact that I apparently bought it years ago. Even I didn't use it that often.

I want to make something that's more polished, that has enough guardrails, that it will prevent you from really hosing yourself, right? Like, really cautious in what it does, just totally refusing to do anything that it knows is potentially dangerous, you know? And in terms of support burden, like...

We'll see. It could be like this is one of the things that was preventing me from doing is like, oh, this is you're just making an app for people to host themselves. And now they're going to be angry and you're not going to get one star reviews or whatever. But this app is already in the Mac App Store. It's got great reviews. It has no guardrails. People love it. So maybe like people find those apps find their audience. So we'll see. I mean, look, I am an expert in bad ideas. And I don't think this is not jumping out at me as a bad idea.

So I think that you might be onto something here. This might be a good thing for you to do. It is a risk. I think, you know, like any new app, it will be way more time consuming than you think it will. The support burden will be larger than you expect and predict, and it will be difficult to market it, even though you have...

one of the best marketing channels in the indie Apple scene by your existing presence in the internet and on this podcast, on your blog and everything else. But I think you could do it. And I think you could do a good job of it. And some people would buy it.

I don't know how the heck you're going to price it. I don't know anything about this market, but I think you can make it happen. Here's the thing. Any competent, experienced Mac developer could bang this out between the time this podcast is released. Anyone who's good at Mac development could do this so much faster than I will. Part of the reason I'm doing it is because I enjoy Mac development, which is a thing I discovered from my other two apps. I just enjoy doing it, period.

uh and i'll learn a lot during i want you know one of the things i enjoy is learning stuff but like i am by no means an experienced mac developer i am stumbling my way through this so it's going to take me a long time anybody good like one of the one of the many good mac developers that are out there that make these things like gee rambo probably made this app already while we've been while we've been discussing you know i mean and they'll do a better job than me too but

But maybe they're not interested in that market. Maybe they know that it's this is a dangerous tool for it to be. I just don't understand why more people haven't made. This is probably not the only one. The one I found that there's probably other apps that also do this when we discuss it on the show. I think I probably said someone should totally make an app like that. And then someone said, here's this app. And then I bought it. Right. Like, so this is not a new idea.

I just think this is an idea that more people should be implementing. And if people aren't going to do it, then I will do it. And again, even if they do do it, I'm still going to do it anyway because I'm having fun. And if it turns a big disaster, I just pull the app from the store. To be absolutely clear, as your friend,

I'm nervous for you, and I don't love it. As your co-host, oh, hell yes, because we will get so much good stuff out of this. You say that until I want both of you to run it for the show. Okay, maybe not. I'll give you the source code. You can look at it. You can step through it in the debugger to see that it's not really doing anything terrible. Oh, my gosh. I'm certainly stepping through in the debugger. I mean, I think there is...

Honestly, there is a lot to be said for taking something that, yeah, any programmer can write a script to do this in a couple hours at most. And there is something to be said for wrapping that idea and functionality in a really nice app. And there is a market for that. And honestly, there's a lot. When you look at iOS, iOS, because it is so big,

it's it's the entire world is ios and as a result it is hyper competitive it every every dollar is squeezed out of everything there you are like you can't get an iota of attention on ios without paying for it and you're going to be competing with everyone else who's paying for it with all the money they're making from their scam subscriptions and all that other crap that goes on in ios like it's a brutal market on ios it

big but brutal and you can't do apps like this on ios at all anyway true but but just like just kind of like you know the dynamics of the market and ios are just cutthroat and brutal and it's hard to to do something like nice there as an indie it honestly it really is hard

On the Mac, though, I think part of the beauty of the Mac software ecosystem that I think really peaked a lot in the mid-2000s and maybe 2010s, and we've kind of been going down since then, is people who were really just making games.

versions of simple tasks like apps that did something that yeah anybody could write something to do that but here's a really nice version of it and it's 20 bucks or whatever 40 bucks like that was a huge part of the Mac indie software scene and that's what what that's part of what drew a lot of us to the Mac in that time span and a lot of that has been has either faded or has been lost since then and

And so I love the idea of you going into this and doing something like this, even though like,

If you look at like, okay, how should you spend your money or how should you spend your time to maximize the amount of money you can make with Apple platform software? This probably is not the answer, but that's not your goal. Your goal is to make something nice on the Mac. And I think that we need more of that. And I really miss the era of that being like the thing we were all really paying attention to and doing and using. So this could be great. And selfishly, it's easy for me to say this because I don't have to handle the support load that you do. But-

I think this could be a very good idea. Or it could just fizzle out and do nothing and make you another $10. But I think odds are, if you finish this and release it and let me pick your price, I think you'll make more money from this than your other two apps combined.

Yeah, this definitely has more potential customers, but also it's much more potentially dangerous. The thing about being an indie developer is... Oh, price it per gig saved. I'm not going to be that. You're thinking of iOS. This is the Mac. We don't do that. Yeah, it's a terrible idea. Don't do that. You can compete with Apple's storage prices. Yeah.

Right? Right? That's kind of what led me to this. I was talking about this in search. But when you're running a company, you can think about what kind of product should our company make in terms of product market fit and total addressable market and so on and so forth. But when you're an indie developer, you're in the situation where essentially you can't pick what you're interested in building. And if you're lucky, you're interested in building an app that has wide appeal. But if you're unlucky like me, every app you want to make...

does not have wide appeal, but that's what you want to make, right? Like even when I was thinking of doing a Mastodon client, which is a much more difficult task, and I started it and it turns out I didn't like iOS development that much, right? Because I like doing Mac development. The apps that I want to, the apps that I'm interested in making are weird, right?

Right? And so it's like, I mean, let's just throw it out of my alley. I love file system stuff. I love like this actually surprisingly large overlap between this app and the stuff I did in my web dev career because I did a lot of, you know, high scale backend things. One of the biggest things I made was like an S3 like service that had to deal with all sorts of file system things in addition to being a web service type thing. Right? So I do actually have relevant experience here and you just don't get to pick what you're interested in when you're an indie. You're interested in what you're interested in. So yeah,

you end up making what you want to make. And if you're lucky, you get something that overlaps with a lot of customers. But if you're unlucky, like you can't force yourself to be like, I should make a to-do app. I don't want to make a to-do app, right? Nevermind that market is saturated anyway, but whatever, like anything with wide appeal, right? So-

Or if you're a very skilled developer, you can just do the underscore method and you have so many interests in so many apps and you can churn them out so quickly that you make lots of bets and eventually some of them pay off. But you have to be, A, a really good experienced developer to do that, and B, I think Underscore actually is legitimately interested in every app that he makes. He's got a wide-ranging interest. My interests are narrower and more obscure, and this is one of them, unfortunately for me and potentially unfortunately for the world. Yeah.

Well, again, I am in full support of this as a co-host. I am sort of in support of it as a friend. The support burden is what worries me so deeply. Like everything else about it, I think, heck yeah. But I am deeply concerned about the support burden. But if you say you're comfortable with shouldering that, then power to you. Let's do it. You're definitely going to want the LLC protection on this one. It's not a very complicated –

app in terms of, you know, there's not a lot of complicated UI that people would have trouble with. It's just only going to be a question of, oh, you broke my computer and deleted all my files, but it's pretty, not as easy, but like if you're super cautious in your app, you can severely limit the damage. I mean, my evidence is the DiskDedooper app that's on the Mac App Store now is not bombarded with terrible reviews. People like it. People like getting free disk space back and the operation it's performing is not

complicated it's officially supported it's you can put it in the mac app store it's not using private apis it's pretty straightforward you know you think will um full disk access be an annoying limitation uh turns out you don't need it uh because it's very confusing on mac os one of the things i discovered is it does not work the way you think it does what

I was in one of the Mac development channels discussing like how does full disk access interact with sandboxing interact with user expressed intent through the open save dialog box I don't need full disk access for this app that's bananas I can't believe that yeah you and me both I couldn't believe it either but I literally don't need it because everything that it does is user initiated like and once you get a user initiated permission through the open save dialog box I

I essentially have full access to the thing that they chose to open in the app. So would you just like have them open like the root of the drive basically? And then you can... No, they just pick whatever folder they want to dedupe. So they pick their documents folder and that's it. But what if they want to dedupe everything, their whole system? They shouldn't do that and my app will probably stop them.

Because that's bad. Now, one of the things I'm also looking into for a future version of this goes well is a privileged helper tool where you ask for admin permissions. You can do cross account, yada, yada, yada. But just version one is just files that you own in your home directory. There's I mean, there's there's data. There's space to be saved there.

No, I, of all the things I thought we were talking about tonight, this was not on the list. So saving space, saving space. Yeah. I mean, literally the show notes are internal show notes. Read John is saving space photos. I cloud shared library, multiple accounts on the same Mac. That's all we had. That's all we had to go. When I wrote that in the show notes, I hadn't, I hadn't done anything except for muse about it in between the time I wrote that. And now I wrote a Perl script and wrote a Mac app. Very good stuff. This is why we love you.

All right. Well, I am looking forward to this corner of the podcast going forward. This is very exciting. Just send in your names. We are sponsored this week by tail scale. Tail scale is legitimately one of my favorite discoveries. If not my favorite discovery of 2024. Now they've been around a lot longer than, than just this year, but I only really truly embraced tail scale earlier this year and I'm

Really, truly, genuinely, it has made my computing life that much better. They can't pay me to say that, but it is absolutely true. So what is Tailscale? Imagine if you didn't have to care what network each of your devices was on, your home network, a work network, perhaps at Linode or DigitalOcean or something like that. You just wanted to make sure that you could always talk to all your other devices, irrespective of what network they're on, irrespective of what firewalls are between you and your other devices.

If that's something that sounds good, and it should, then you need Tailscale. Tailscale is a super easy to deploy, zero configuration, no fuss VPN. VPN, you say, whoa, I don't know about that. Don't hang your hat on the VPN thing. It is a wire guard VPN under the hood, but Tailscale sprinkles all this incredible magic on top so that it just works.

works. I cannot tell you how amazing it is to watch Tailscale negotiate direct connections between devices, regardless of the fact that I'm behind a hotel or work firewall. It's really incredible. You should check it out. And the best part is their personal plan is and always will be free. So you can try Tailscale for free today. You get up to 100 devices, and you can share with up to three users total for free. That's F-R-E-E-F-R-E-E.

at tailscale.com. So what are you going to do? You're going to do yourself a favor and you're going to do me a favor and you're going to go to tailscale.com slash ATP, T-A-I-L-S-C-A-L-E dot com slash ATP and you're going to check it out and you're going to try it and you're going to install it on at least two devices and then you're going to send me a tweet or whatever and say, holy cow, Casey, you were so right. This stuff is amazing. Tailscale.com slash ATP. Thank you to Tailscale for sponsoring the show.

All right. We have semi-breaking news. I think it was yesterday afternoon. iOS 18.2, Mac OS 15.2, Vision OS 2.2 all launched. And I was not on the beta train on anything for the .2 releases.

And I have been very excited to try the .2 releases, the 15.2 and VisionOS 2.2 releases, macOS 15.2 and VisionOS 2.2, because this gives us the new fancy Mac virtual display. So if you are not caught up with

The way it worked up until literally yesterday as we record this is that if you wanted to kind of suck your Mac into the VisionOS world, you could use either the VisionOS control center or if you have a laptop, you can just look at your laptop and a little button will appear over it. This is when you have your Vision Pro on. A little button will appear over your laptop that says Connect.

And then it'll, whatever, you know, laptop or even desktop, if you go in through VisionOS's control center, whatever computer you're bringing into VisionOS, you get a 4K display in VisionOS. And you can resize that display insofar as you can make it, you know, the equivalent of 10 feet in front of your face, but it's still only ever 4K.

And I've used this a handful of times. I've used it kind of a lot when traveling, but I don't travel that often. So I've only used it a little bit here and there. And I've used it on the train and on the plane. And it's very, very cool and really nice that not only do you have a large display in a train or in a plane, but nobody can see what's happening on your Mac. It's not like, you know, screen sharing where a lot of times your Mac will be displaying what's on screen and then some other computers also displaying it.

When you do this Mac virtual display thing, the Mac screen or screens go dark. And so no passerby or bystander or whatever, looky-loo, can see what's going on. And I always thought this was good. It wasn't perfect, it wasn't stellar, but it was good. And I typically do this with the developer strap on my Vision Pro. If you recall, you can replace the right-hand side or...

AudioPod, I think they call it, which in the strap that it's a part of. And you can get one that has a little USB-C dongle on it. Then you can plug USB-C from the Vision Pro into your computer and it will actually do Mac virtual display over that rather than like local Wi-Fi or whatever. And I always thought it was good

Maybe even great. I don't know, somewhere between good and great. Marco, up until yesterday, if I recall correctly, you found the Mac virtual display to be a little bit blurry and you didn't really care for it very much. Is that accurate?

I would say that's an understatement, but yeah. Okay, fair enough. So yeah, so yesterday we got the new Mac virtual display, which does a few things. I believe the standard display is now 5K instead of 4K. Additionally, you have two other modes that you can optionally choose between. You can choose between...

What do they call it? Wide and ultra wide. Thank you. Wide and ultra wide. Now, I should also note that up until yesterday, even the standard display was well, there was that was the only option you had and it was flat. Whereas now all of the displays are curved a little bit, which generally is not something I particularly care for. But in this context, I think does make a fair bit of sense.

Um, God help me if I end up buying like one of these 40 inch curved computer displays from my desk. In all fairness, I would say like based on the virtualized size of them, the wide and ultra wide, I think basically have to be curved, especially the ultra wide. The, it, it takes up like, I mean, I'm just estimating here, maybe like 90 degrees of your field of view. So if that was just a big flat pane window in the, in the view, um,

the left and right edges would be substantially further away from your eyes than the center would be. It would be a little weird.

Yeah. So, yeah. So they added wide and ultra wide and wide is let's see. So the standard, I think what I forget the resolutions. Oh, here we go. So when I was in ultra wide mode, it offered 51, 20 by 1440, which is the default and up to 10 to 40 by 28, 80, which is bananas. But suffice it to say, I have tried this out wholeheartedly.

Holy freaking crap, this is amazing. I am so in love with this. Now, if you recall, I'm the kind of lunatic, and this is partially Marco's fault for having sent me one, but I'm the kind of lunatic that runs three 5K displays at home. I have my studio display directly in front of my face.

Then I have at, you know, like a 45 degree angle to either side. I have, uh, I have, uh, LG ultra fine five Ks. So I'm used to having an overabundance of real estate and I'm used to having to move my head in order to use my computer. I don't mind that. I, I, I think the, the juice is worth the squeeze for me personally.

The ultra wide is so freaking big in the Vision Pro that I actually think it might even be too much. And I am so happy that that is the problem I have, that this might be too much. It is unreal. I have many more thoughts about this, but I'd like to pause myself and hear Marco. You were assigned yet more homework. And listeners, I have to tell you, Marco has been a very good sport about all the homework I've assigned. And you have done more homework. And I appreciate you. And I appreciate that.

rain on my parade, and then I'll go back to telling you about how amazing it is. Honestly, if you try it, and you like it, more power to you. This...

Where I am with the Vision Pro is I have my feelings about it. I can tell you how well it works for me, but a lot of other people, it works better for them. And so if it works better for you, who cares if it doesn't work very well for me? That's my loss and your gain. So anyway, so I did try this. It is substantially better than the original version in terms of screen space and therefore the usability of that screen space.

I found the ultra-wide, the biggest setting, to actually be way too much space. I had to do so much head turning. I'm like, this is ridiculous. So I actually stuck with the middle setting, the wide. It was a lot of screen space, plenty of screen space on the middle setting for me without too much head turning. So that part was fine. The scale, the ability to use the space, that was all very good.

The main problems I had with the original version were, number one, I could notice the foveated rendering, which is bad. Because that's like you see kind of out of the corner of your eyes like which parts of the screen are blurry that you're not quite looking at. And maybe it was a little slow to update when you would move your eye to it. That, I did not notice that this time.

So I don't know if they've actually improved that or if I've stopped noticing. I think they might have actually improved it. So you don't notice the foveated rendering blurriness on the stuff you're not looking at anymore. At least I didn't. I did still notice the two other major problems I had with it the first time, which are a lack of sharpness and latency. One of the first things I tried when I first got Division Pro, I tried editing a podcast in it with Logic.

The screen space was delightful, but the problem was there's a little bit of latency. And when you're like, if you're just typing in a document, you won't see it. But when you start moving stuff around, you do see it. You do feel that latency and it feels really bad and really wrong. And it makes certain types of work very difficult to do in that environment without like driving yourself nuts.

So, latency is still there. I still notice it. And that alone would be enough for me to not really want to use this very often, if ever. Now, quick interruption. Are you using this via Wi-Fi or are you plugging in via the developer strap? The laptop it was connected to was connected to Ethernet.

But the Vision Pro was on Wi-Fi. Gotcha. That being said, the Wi-Fi router is like 10 feet away. And nothing else ever says this is a bad Wi-Fi connection right here in my seat at my desk. So anyway, and I would argue that's the way to actually judge it anyway is when using it on Wi-Fi because that's how it's really made.

Because the developer strap gives you a second cable down the other side of the Vision Pro. By the way, I actually did solve my Vision Pro standby issues. Quick aside here, I've decided when I'm not using the Vision Pro, I just disconnect the battery cable from it.

Which is, you know, that's obviously like it's like unplugging the desktop. It powers off immediately and it does not do a clean shutdown. But I don't care. And when I disconnect the battery cable for storage, I did that last time I put it away. That was probably, what, two weeks ago? I got it out this time.

Perfect charge. It booted up just fine. The battery still showed green. Like everything was totally fine. So if you actually unplug it from its battery, it does last substantially longer in standby mode. Anyway, so going back to the screen, the Mac virtual display update, the latency is still a problem for me. The sharpness is still really poor.

When you look directly at something and you kind of stabilize yourself, I'll get to that in a second, you can see what it is. It's not like the pixels are so blurred that you can't read lettering or you can't see lines. It's not that bad, but it is blurred. And when you look at the optical pipeline of what is being shown on a Vision Pro, it's

It has to be blurred. We are nowhere near the optimal resolution or screen resolution required

to actually replicate the pixel density of a computer monitor at a regular distance inside the Vision Pro. We're nowhere near it. We are many generations away from that being realistically possible if it ever comes. So it is substantially blurrier than looking at a real monitor. It has substantial latency. And then one of the big killers for me

is the Vision Pro has significant motion blur when you move your head or eyes very much. When you look at the design of Vision Pro native apps,

Everything is huge. The control, the windows are huge. The buttons are huge. The text is pretty big. Everything about Vision Pro or Vision Native apps, everything is big and it's simulated to be somewhat far away from you too. And so everything like you don't notice the motion blur as much. And then

The best thing to do, in my opinion, with the Vision Pro is watch 3D video content or immersive content. And with that, you don't really see it at all because of the type of viewing you're doing in there. The small amount of motion blur when you move your head is kind of just blurred away and compensated for with everything else you're seeing. But when you're looking at a computer screen, which has a very high pixel density, trying to make all these nice sharp windows and nice sharp text in this window...

The motion blur of the Vision Pro, just moving your own head, just even very slight movements as you operate your computer, you notice it constantly. At least I did. And so I found it very unpleasant to use. If you have already been using the Vision Pro virtual display mode before this update and it was fine for you, then this is even better for you. So enjoy it.

If you had the problems with it that I had of latency, blurriness, and motion blur, those are not improved. And I don't think it's possible to meaningfully improve them with the current hardware. So basically, if you already have a Vision Pro, this is a useful thing to try. And this might get you to use it more. If you don't already have a Vision Pro, this probably is not a good enough reason to go out and buy one.

I was with all of your summary until that very last line. So let me reiterate that for me, and maybe it's because my eyes are not great. And I know I've said this a thousand times, but just one more time to set the stage. I have a weird eye disease called keratoconus. That means I wear hard contact lenses.

Um, it's actually been said on Apple's documentation that you shouldn't wear hard contact lenses with vision pro probably won't work. I've never had a problem with that. Don't tell my, my, my vision is actually pretty good with, with my contact lenses in. I I'm like 20, 30 or something like that. I forget what I am, but it's actually not bad, but without my contacts, I'm even more blind than John. And that's saying something. So, um,

I think it is possible, it is certainly possible, if not probable, that maybe some visual wonkiness that is really a shortcoming of the Vision Pro is actually my own eyes. I don't know. This is the only eyes I've got, so I can't tell you one way or the other. But whatever the case may be, be it because of the Vision Pro, be it because of my eyes or whatever, I don't find it to be that off-putting.

I don't really get the motion blur. Occasionally, I find it's a little difficult to focus, but it usually goes away very, very quickly. That might have to do with maybe my eyes settling down and the foveated rendering calming down. I'm not sure. But generally speaking, I find this works really well. And...

So this morning I had the occasion to go to a local library and I needed to be there to return something. And I figured, well, I'll just get in the chill and work for a little bit. And so I am enough of that guy to use the Vision Pro at the library, but I'm not enough of that guy to do it in the middle of like all the other people working at the library. So I want to say rented. I reserved a study room at one of the local libraries and

And I sent this picture to Marco, and I forget exactly what you said, but you said something along the lines of, holy crap, could you not have found a bigger place to work because it is effectively a closet. And that's true. But I sat myself in this little study room.

And I put on my vision pro and I looked like a damned fool, but thankfully I wasn't in the middle of everyone. And I was using the vision pro and I was using Mac virtual display. And I agree that the ultra wide, while incredibly novel is generally speaking just too much. And,

especially since like the menu bar runs the entire width of the display. You know, if you think about the way it is on my desktop, on my desktop at home or my laptop desktop at home, I have three different displays. So they each get a menu bar and I can use the dock on any of them. Right. And it'll be centered. The dock will be centered at the bottom of each individual display and the menu bars repeated across or well effectively repeated across each of these three displays on the ultra wide. It's just one freaking huge display that,

that like you said earlier, something like 90 or 120 degrees of your vision. So if you need to find the menu bar, you got to crane your, turn your head all the way to the left to go find it. And I don't mean to imply that it's a burden, but it's, it's striking, right? Um, I think the ultra wide is a bit too much, but the wide is incredible.

and I got a lot of work done with this on. Now, granted, the comfort, I think I've actually, while I was sitting there, I ordered one of the $50 Belkin headstrap things that I should have ordered a while ago, and apparently my local store doesn't have in stock, which was a bummer. I was literally going to drive home by the Apple store to grab one of them, but it was not in stock. So leaving the comfort aside for a minute,

it was so incredible to be able to have all of this real estate and it's

It's all right there. I'm in this little closet of a room and I have a, what, like effectively a pro display XDR, maybe not literally, but effectively a pro display XDR right in front of me. It was phenomenal. And not only that, but I could put a music, the music app like above the display. So I'm using the native vision OS music app. That's in 3d space above my ultra wide display, excuse me, my wide display or whatever. It was phenomenal.

so amazing. So over the last few years, what I've really wanted to do, and this started especially during COVID, is I want to be able to work wherever I want to work, right? And leave aside the Vision Pro for a second. Just in general, I want to work wherever I want to work. And when I entered the pandemic, when we all entered the pandemic in 2020, if I'm not mistaken, I still had an iMac Pro and a very serviceable 13-inch MacBook Pro. And

Over time, I feel like we are answering more and more, or Apple is answering more and more questions. So I was thinking about this. The first question is, can I take my computer with me? And that's a solved problem, right? If you want to take your computer with you, you can just get a laptop. That is not up for grabs. We've solved this problem for 30, 40 years now or something like that. Not a big deal. The second problem is, can I power it? And...

That isn't much of a problem, generally speaking. If you go to a library or something like that, typically you can plug into power. But even if you can't, you can always bring like a USB-C battery or something like that to give you another hour or two. There's plenty of ways that you can solve this problem. So neither of these are really new problems or really that terribly interesting.

But where things got interesting, I think, was starting in around 2020 with Apple Silicon. And when Apple answered the question, can I get my work done on it? And that started in 2020 with Apple Silicon. Because now the computer that is portable, that I can pick up and walk away with, is just as powerful, if not more powerful, than that iMac Pro that was sitting on my desk. And

And part of the reason why I returned to the full-time laptop lifestyle was because my MacBook Pro, my M1 Max MacBook Pro at the time, was considerably quicker than my iMac Pro. And my M3 Max MacBook Pro is considerably quicker than a lot of Apple desktops. And

And I don't have to really make any compromises. Like it was such an incredibly freeing feeling. If you're still on Intel, it is such an incredibly freeing feeling to know that you can take this incredibly powerful computer and just walk away with it. You can just buy a laptop and walk away with it. And you don't need a MacBook Pro to do that. Like, as we've said many times, the MacBook Airs are also incredibly powerful computers that you can easily walk away with. It's unreal.

And then, so can I get my work done on it? Hell yeah, I can because they are so darn powerful. And then the other thing, and this is where the Vision Pro comes in, is do I have the space to get my work done on it? And I mean screen real estate. A 13-inch laptop is not a fun way to get work done unless you're doing very, very basic stuff. But typically in the last couple of years, and I've talked about this on the show, I bought a like,

2k portable display a bus powered portable display it is a piece of trash but i love it because that gives me a second display that i can set next to my macbook pro it's super thin super light it doesn't need a power supply like i said it's bus powered it's great and that would get me enough that i could do development work in a pinch on the go wasn't fun but i could do it with the vision pro

I can legitimately do my development work and it might be a little less convenient than having my full array of three 5k displays, but,

but not that much. The compromise, even with that spare physical 2K display, the compromise was still quite large. But now, if you can leave aside the fact that you look like a frigging idiot, the compromise is really not that much at all. And in fact, for most people that don't have 15Ks sitting in front of them, it's actually arguably better

better in the ultra wide display. Now I agree with Marco. I think ultra wide is actually too much, but that's neither here nor there. My point is just that I am now able to set up an entire workstation by way of a MacBook pro and a vision pro.

And I just think that is so incredibly cool. That is so cool that one of the things I can do, which I haven't done yet, but one of the things I could do is during the workday, if I feel like I'm just a little uncomfortable in my desk chair or what have you, I can go down to my couch and I can sit on my couch with my nerd goggles on my head and I can be doing work unnoticed.

on my couch. There's nothing wrong with it. It's almost as good as on my desk. And I just think that is such an incredibly powerful, cool thing. And again, there's so many things I could say to poo-poo this. I'm talking about what, $9,000 worth of hardware or something like that. It is an absolutely asinine amount of money that I'm talking about. I said earlier, the laptop was like five grand. The Vision Pro is 3,500 plus the $400 developer strap. We're now at like nine grand. That is bananas. But

It is so incredibly cool to have that future, I was going to say in my hands, but also on my face today. And I think that that's so incredibly powerful. But here's where I'm going to drag Marco back in. Oh, no. You know what else is missing? You know what the thing is that would really make this truly and incredibly, just perfectly portable? I can figure out power. I can figure out my display.

But Marco, what do I not have yet? What is not internal to my MacBook Pro? Internet. Internet. I cannot imagine how cool it will be when Apple finally one day, when I'm 95 years old, decides to put a cellular modem in this MacBook Pro because then...

Then this literally could be used anywhere, anywhere there's civilization. You know, I'm sure there's an asterisk there, but you know what I'm saying? I could go to my beloved park bench slash picnic table. I could go wherever the hell I could be in the car in the passenger seat. I could be wherever and I can have my MacBook pro and my vision pro and I can get work done damn near as well as I can do it on my desk.

And whether or not you think that's a wise use of my money, whether or not you think it's a smart idea to put this nerd goggles on your face, it is so powerful. I would, I would say empowering, but I think that might be a little dramatic, but it's powerful that I can get my work done in the same fidelity for lack of a better word as my desk. And I can do that almost anywhere. I can do it anywhere that I have internet. And that to me is so cool. And again,

I still mostly agree with Marco that, well, I 100% agree with Marco that if this is something that didn't work for you before, like the Mac virtual display thing didn't work for you before, you're probably not going to like it now. I concur. It is better, as Marco had said, but it's not, you know, night and day better. But if you're like me, and this is something that worked before...

But now you want to see what it's like now? Let me tell you. Oh, it's so good. It's so good. And this is the sort of thing that if you have the money to spend on this sort of thing and the desire to be able to work wherever, maybe it's only that you travel a lot. You want to be able to work really efficiently in a hotel room or whatever the case may be.

And suddenly this is becoming, and many other people have said this over the last few days, this is starting to become a killer feature for me. And I genuinely think that this will make me want to use my Vision Pro more. Not in the house necessarily. I still think all the problems we've talked about many times about not being enough content is still true in the house. But it's going to make me use it out of the house, especially on the occasions that I travel much, much more. And I cannot say enough good things about it.

If Apple really cared about this use case with the, you know, the M5 powered successor to the Vision Pro or whatever, they should just let you plug a Thunderbolt cable from the headset into your Mac to get rid of the latency thing that Marco was complaining about. Like, just fully embrace the idea of it as an external monitor. And yeah, it'll still have to do all the 3D and the tracking and the blah, blah, blah, but the latency should be gone thanks to whatever clever thing they do over Thunderbolt to, you know, essentially reestablish

into a virtual display somewhere inside the Vision Pro that it then puts into 3D space in front of you without having to do any Wi-Fi or compression shenanigans or stuff like that. But I'm not sure Apple cares that much about this use case. I know, Casey, you love this use case, but I think...

I think Apple probably thinks the appropriate level of technological and cost, technological innovation and cost is what they're doing now. Pretty good, pretty good resolution, very big display, wireless, convenient, no cables. You don't have to put all the extra hardware and memory inside the Vision Pro to support it and stuff like that. But yeah, you can see how if the Vision Pro was emphasizing this use case much more and de-emphasizing many other ones,

you first of all would have to make the vision probably even more expensive just for the memory stuff but you can do that and if you want to go really expensive you know the vision pro with an m5 in it you know what that could do i could just freaking run mac os in addition to vision os you know what i mean like with virtualization but like you could just run your mac like why do you need to have this laptop that you bring with what is the lap what is the purpose of the laptop it's just sitting there over in the corner right

again they would make it even more expensive heavier hotter you'd have to have an external compute thing the battery would have to be bigger but like that's not the use case that apple is going for so i don't think they're going to do any of that but technologically speaking because apple like taco bell is using all the same different ingredients to make lots of different things slightly different results yeah but you look at the vision pro and you're like you know all the

pieces are there like if you want if you could if you could arrange these lego pieces to make vision pro like the casey's dream like on the go mac development thing all the pieces are there except for apple cell modem that's going to put cellular into mac someday which we'll have a story on in a future episode if we remember to do it but yeah we're still waiting if they remember to do it

It's a race between a worthy Mac Pro chip and cellular and Macs, and it looks like Casey might win this one. You never know. I'm glad, actually, you brought up the plugging it in via Thunderbolt thing, because I should say that my experience, I believe I've only exclusively used it with the developer strap with the thing plugged in, so...

That's another reason why Marco's experience might have been slightly worse than mine. And it does appear that by default, it will try to hop on Wi-Fi between the computer and the Vision Pro. And I think it negotiates this via Bluetooth. But I definitely noticed once I plugged it in, and I think I actually killed Bluetooth on my Mac, if I recall correctly, it clearly renegotiated the connection over the...

cord over the cable and then I pulled it out because I realized like a ding dong I had taken the cable that's coming out my right ear and connected it to the left side of my computer I don't know what I was thinking so I unplugged it moved it to the right side of my computer and the screen immediately went black and it once I plugged it back in it took a few seconds before it renegotiated but it did come back on its own but I think that that certainly helped like latency and stuff like that some of the legitimate complaints that Marco had but yeah it either way it is so cool and I

And there's no easy way for Apple to demo this when they do like the Vision Pro demos or whatever, because it all is relying on like Apple, your Apple ID or whatever. It's Apple account now. And it would be a lot to get this to work. But, oh, man, if you have a friend where you could just strap their face computer on your face for a minute, it's worth trying because it is so freaking cool.

All right, thank you to our sponsors this week, Tailscale, Delete Me, and Aura Frames. And thank you to our members who support us directly. You can join us at atp.fm slash join. One of the many perks of ATP membership is ATP Overtime. This is our bonus topic every week. This week on Overtime, we're going to be talking about the story about there's a Mac emulator called Mini VMac that was being rejected by Apple's notarization company.

even from third-party app stores in the EU, which is quite a story. So we're going to be talking about the mini VMAC notarization rejection from the EU app stores. Interesting story. Anyway, you can hear our coverage of that in overtime this week by joining atbjfm.com. Thank you, everybody, for listening, and we'll talk to you next week.

Now the show is over. They didn't even mean to begin. Because it was accidental. Oh, it was accidental. John didn't do any research. Marco and Casey wouldn't let him. Because it was accidental. Oh, it was accidental. And you can find the show notes at atp.fm.

And if you're into mastodon, you can follow them.

C-A-S-E-Y-L-I-S-S, so that's Casey Liss. M-A-R-C-O-A-R-M, Auntie Marco Arment. S-I-R-A-C-U-S-A, Syracuse. It's accidental. They didn't mean to. Accidental. Tech podcast so long.

Before we get to Casey's thing on the after show, can I do a quick, this is not an ad discussion of one of our sponsors this week? Yeah. Yeah, go for it. This is not legit, not an ad. If you're listening to the non-member version, you by now probably have already heard the ad for our frames. This is not the ad. This is just me talking about our frames, which they did send us one of them for free. So take that with a grain of salt. I got mine for free.

I have never bought a digital picture frame. Honestly, I thought they were like, why would anyone want that? It's just like a really low quality screen in your house. I have plenty of actual pictures up on the wall or whatever. So I, you know, whatever they sent it to me, it's free. I guess I'll set it up to see what it's like.

Love the thing. I don't understand why. Like, again, they're inexpensive. It's not like it's not it's not a pro display XDR. Right. It's not like because it can't be. It's not thousands of dollars. Right. They're inexpensive. It's an inexpensive, small color screen, kind of like Casey's like 2K monitor thing that he loves, even though it's not super high quality.

I didn't like who would care that you have? Like I see, I see my pictures. They're on my phone. They're on my Mac. Like why do I want to have a frame of my house that rotates through those pictures? Turns out I do want to have that. I can't like it's so much enjoyment of looking at which picture it's showing right now. I pointed it at my favorites library, which has like 30,000 photos and something. It was like, no problem. I love the thing. Like, Oh, what was that picture? You just swipe on the top to go back to the freeze. I know this sounds like an ad. Honestly, like I,

And probably this is like, you know, a gift. Like I would not have bought this for myself. I would not have ever bought a digital frame as evidenced by the fact that I've literally never bought one. I never bought one for someone else. I always thought they were dumb. Having been gifted this one essentially as part of a sponsorship made me realize I like digital picture frames. That's it. It's not an ad that I just wanted to say that I've discovered that I like digital picture frames.

Yeah, the Aura stuff is legitimately really, really good. And you will hear me or some of you have heard me do the sponsor read. But for real, and I'm now kind of regurgitating the sponsor read, but if you've had one of these in the past, this is not like that. Because the ones in the past, you have to put an SD card in and it's just trash. And I'm not going to repeat the ad, but suffice to say this is all that but done right. It's really quite good. I took like two minutes to set it up ages ago and have not touched it since.

Yep. No, it's genuinely good stuff. All right. So I have a conundrum and I'm probably making a mountain out of nothing because, hi, have we met? But I am, I think at the point that I'm ready to release the briefly previously mentioned, or I guess not so briefly, actually previously mentioned update for call sheet where you can have multiple lists of pins. So effectively multiple batches of favorites, if you will.

So I think I'm ready to release. But then it occurred to me, you know what? Apple has said to developers a handful of times recently, you know, you can tell us when you're going to be doing a new app or a new feature or whatever. And maybe you'll get a feature on the app store. Maybe you'll get an editorial on the app store. It'll be great.

And I know full well that the likelihood of my little dinky app getting a feature on the app store is not great. But I thought, what's it going to hurt to do what they ask you to do and report in and say, yeah, I've got this big new feature. It's the thing that almost all my users have asked for at some point or another. Why can't, why wouldn't I just tell them? And so I went down the path of telling Apple, hey, I'm going to be doing this probably in the next week or two.

And completely fairly, especially given the time of year, very early on in Little Wizard, they have you go through to get the information that they want. They say, hey, we really prefer for you to give us three weeks before your launch, which in and of itself is fine. But that means I'm not hitting my self-imposed deadline of before the end of the year.

And my question to the two of you, and I think I'd like to start with Marco, if that's okay, is knowing that there's no real urgency to get this out other than a completely self-imposed deadline of wanting to be done with it before the end of the year.

Should I hold on to like the first full week of January? So like two or three weeks from where I guess three ish weeks from where we are now and give Apple the chance to feature me, even though I intellectually know that there's almost no chance of it happening. Like what's it going to hurt to delay and try, or should I just get this out the door now? What do you have a question for you? Oh, sorry. Yes. Please. Is this, do you know if Apple, uh,

not in cysts, but if Apple wants you, won't feature you if you launch without them, like is the, is the featuring contingent on you launching coordinated in time with their feature or are those two separate things?

That is an excellent question, and I don't know the answer, but I thought that I think they want you to release as they feature. I do not know if that's accurate, though. If only there's someone on the show who has experience with iOS. Okay, so this is why I came to the show. So tell me, what am I doing? What should I be doing? What am I doing wrong? What should I be doing? Help me. Okay, so first of all, I'll answer the second question first. The...

They want you to coordinate your other marketing with them. They don't want to be the only place promoting an app. Other marketing? They want to jump on a bandwagon.

And they want you to be doing a big push, and they want to be a part of the big push. That's not always the case. That's not the only way to promote something. But that's why they want to coordinate with your release. Casey, you should email ATP and tell them you have a new app that you're launching. And maybe you should offer the CEO of your company to be a guest on our show. Yes, I should do that. Oh, God. Oh, God. That's such an inside joke, but yes. Anyway, yeah, we get a lot of PR email. Anyway, that being said,

when, when you are trying to coordinate something like that, like by submitting it to them and it like in advance with the number of weeks of notice and everything, that's kind of what you're going for. That is not guaranteed to actually happen. Well,

What also happens is if they see another app that like, hey, the blogs are posting about this app. Like, you know, hey, this thing got reviewed on Mac Stories 9 to 5 and everything. If they see that, the App Store editorial team, like they have people who like look for that kind of stuff. If they see that, they will come to you and send you a message in App Store Connect saying there's a promotional opportunity if you just submit artwork of these specs by this date.

And what that means is we might feature you in the app store somewhere. Now, so that's the other part of this. If you actually succeed in being featured, that's what they will go to you with. They'll go to you with a promo art request.

And so they'll say, all right, the link link to some document somewhere on an Apple site that says we're currently looking for this type of promote. And it'll say the request like a home screen header or whatever. There's different names for the different positions in the app store, different formats and everything. And they'll send you will there's a link to like a Photoshop template.

And it's like, you know, you have to have design your artwork within these safe guides and have it, you know, feature these things, but not these things. And don't include screenshots. Don't include words. You know, it'll go through all these requirements with you. So then that's then placed on you. And they'll say, give us the promo artwork by, you know, so-and-so date. Usually give you like, you know, a week or something. If you're lucky. Sometimes it's like 48 hours. Anyway.

as developers we are given that chance and we say okay well we can either rush to put this promo art together ourselves if we even can um and if it will look any good or we can hire a designer or you know tell like because they they basically assume you have a designer ready to go like they assume you either are a designer or have one on staff because they assume everyone's a big company um so anyway so as an indie though then you're stuck with like all right i have to like

find the designer and pay them to do this for me. And then you have to start weighing like, is this going to be worth it? Or should I try to do it myself, like do my own custom promo artwork, but I'm a programmer, not a designer, and it'll look like crap. And the worse it looks, the less likely they are to actually use it and actually feature your app. Because when you submit that promo artwork to them, they basically say, thanks, we might feature you.

That's it. So you're kind of throwing it into a black hole and hoping someday they maybe use it. They won't tell you where they're featuring you. They won't tell you when they're featuring you. They won't tell you if they choose to or not to. You just kind of have to watch for it. Although I think they've recently added a notification that if they do feature you now, it finally tells you. That was like two weeks ago, very recent. Anyway, the process of being featured in the App Store...

requires something of you. It requires you to, if you're going to go through what you were asking about, about notifying them ahead of time, that's one level. If it's going to be like, you know, you having to make promo artwork, which is usually required, that's a different level. There is a certain amount of work and cost that is going to be put on you. And so the question of whether all this is worth it or not is worth asking. Sometimes it is.

I think usually for indies who already have marketing channels, usually I think it's not worth it to bend over backwards. Now, if you can address what they're asking for without too much sweat off your back, without too much work, without too much cost, then sure, do it. You might as well do it if it's going to be easy. But if what they're asking for is going to cost you like...

thousands of dollars in designer time or heck, even hundreds, it probably won't be worth it. Because the thing is, an app store feature used to be worth a lot more than it's worth now. Because not a lot of people, it seems, are just casually browsing the app store, looking around for new apps to install. It used to be that way. When iOS was younger and less mature and less established,

When people were, you know, were actually like getting their first iOS devices still, there was a lot of just browsing around that happened. Like I remember back in my Instapaper days, Christmas Day used to be a huge sales day for me. Like it was like one of my biggest sales days of the year because so many people – Instapaper had pretty good business on the iPad. And so many people would get iPads for Christmas.

And that was often their first iOS device or at least their first iPad. And so they would then immediately go browsing around, hey, what can I do on this? And let me install a bunch of apps.

Now, most people with iOS devices, this is not their first iOS device. So how many new apps they need to browse around and install goes down over time. And now there's so much out there. There's so many apps that I don't think a lot of people are just kind of browsing the editorial sections of the iOS app store. So I occasionally do get featured, which is very nice. I'm very thankful for it. But I see like when you look at the sales that happen as a result of that,

It's hard to see a bump anymore. I'm sure it helps. I'm sure, you know, any bit helps. And if you don't already have a ton of exposure in any other way...

you got to take whatever you can get. But the effect that it has has gone down over time to the point where now I think it's more of a vanity and honor thing than an actual thing that translates to sales. And so you should account for that in deciding how much to either spend accommodating their plans and or how much to change your plans in response to that. So my advice after all of that elaboration would be

don't change your plans. Release it when you want to release it. And if they want to feature it, they'll come to you. Yeah. And I guess what I'm wrestling with is,

What's it going to hurt to wait and see? Like, what's it going to hurt to throw this over the wall, give myself the remainder of the year to breathe and perhaps, you know, just make sure everything is good to go. And yeah, I have, I guess, failed in that. I haven't released the thing that I've been working on, but it's done. Like it's ready. It's released in test flight. And the only person that's really going to be disappointed in the, in it not being out is me because,

And I don't feel like that's enough of a reason to rush it. Right. And especially if, you know, God forbid, uh,

the rollout ends up being a disaster. Well, that's what I was going to say. Like, what does it hurt? I would want to get it out and make sure it actually works before I get featured. The worst thing you want to have happen is get featured dresses you launch and there's some fatal bug that causes problems with your app. And now you've like drawn lots of attention to your mistake. Like that's why I want to get it out. And for,

The wisdom of releasing it just before the holidays aside, I would want to have my – be sure the app that was going to be featured was stable and working. Which is fair, but if I'm also coordinating the big release with Apple, I can't have both, right? Unless I'm missing something. Well, that's why I was asking, do they care if you coordinate the release or do they just want to feature you? Because if they don't care if you coordinate the release, I would want to have my new version out for three weeks, two weeks.

you know, maybe get a one or two bug fix updates out and then get featured if they don't insist that you not release your app until, and Marco's thing about they want to be coordinating with your marketing push. If you don't have a marketing push, I wonder how they'll feel about that. You are the marketing push, Apple. Right, right.

Yeah, I don't know. I got a noodle on it. And certainly if App Store editorial wants to quietly whisper in my ear, not necessarily whether or not I'd be featured, but just, you know, what is the expectation? You'll never hear that. They'll never tell you that. Trust me. I know.

Ask any iOS developer you know. Ask Underscore. Ask Ben McCarthy. Ask everyone you know who has apps in the store. They're all going to tell you the same thing. I even submitted PR artwork for some of my apps at one point. And they did actually, if not write about them, then at least list them somewhere at some point in the past. But no, they did not tell me when they were doing this. I found out because other people told me, hey, I saw your app on this thing. I'm like, oh, okay.

Yeah, they did recently launch notifications for when you're feature-aware. And you can also... I've been notified through app figures. They have a thing. If you use app figures, they'll tell you. But it is mostly just like you kind of just rely on people telling you on social networks, hey, congrats, that's why you're apping this thing. And again, it's nice, but...

It's really hard to see much of a bump from those anymore just because, again, I think the number of people who are just casually browsing looking for new apps to install, I think that number is very small these days. The reality is because, as I was saying earlier, because the iOS market is so full and so cutthroat and so dramatically over-served with just huge amounts of apps available,

You usually have to pay to get people to install your app. Like, you have to buy search ads, you have to buy Facebook ads, whatever. And paying for app installs is not cheap because there's so much competition from everyone else trying to get people to install their apps. So the bids all get bid up. So, like, it's really hard to get installs of your app because no one is going out there and just looking organically unless they hear about it from your other marketing channels, from your podcast, from your blog. Like, that's...

That's where to promote your app. And just talking about it on here is worth way more than most types of app store features that any of us would be likely to actually get. Now, if you happen to be like the app of the day on the front page of the app store, then you're talking about different numbers. That is more substantial. And that is worth it, of course, if you can get it. But

This kind of smaller feature is like, oh, top apps to be cozy this winter or whatever, like that kind of stuff. You don't you're not going to see much of that. It is a nice honor. Again, like it's it's an honor to be featured and it's it's nice to have your work recognized by Apple and it feels good. But in terms of translating to actual like really meaningful sales and downloads, it's not much. Your plans are more important. So launch your app when you want to launch your app.

And you can tell them afterwards, like once it's launched and once it's settled, you can email that address and say, hey, I got this new update out and it's being well received. And here's, you know, positive reviews or whatever. You can tell them afterwards, but don't change your plans for them because first of all, they wouldn't do it for you. And second of all, you're better off serving your customers directly in the best way you see fit, not bending over backwards, trying to accommodate something that you might get from Apple that probably wouldn't have a ton of value.

Yeah. Well, I appreciate the conversation. I'll noodle on it and see what I decide to do. I don't know. I don't know. I got, I got to think on it, but this is very helpful and I appreciate it.