PodSearch

The Talk Show

452: ‘A Scam Held Together With Patriotism and Golden Paint’, With Quinn Nelson

 

00:00:00   Quinn, I don't know what I'm going to do in September because usually I go out to Cupertino and go to see the new iPhones that are coming out, but there's no point this year because it's been obviated. The Trump T1 phone came out.

00:00:11   That's right. I mean, what else do you need, right?

00:00:15   Oh, boy. What a disaster that thing is.

00:00:20   I don't know. I want to talk about it with you, but without discouraging listeners from going to your channel to watch your video about it. Somehow you got 30 minutes out of it, and I wanted so much more.

00:00:34   Well, good. That's high praise. I was worried that we wouldn't be able to make anything super entertaining about it because ultimately it's just a rebranded HTC phone.

00:00:45   But the process of trying to get it, and then the way that it was packaged, and even as it came, everything was so absurd that, yeah, it just ended up being a content goldmine.

00:00:57   I should have written it down, and it's so great. It's early in the video, and you say that it is a phone that is held together with patriotism, and I forget the other thing.

00:01:09   I don't remember.

00:01:10   It's not held together.

00:01:11   It's in hopes and dreams or something.

00:01:12   Yeah, hopes and dreams. And if you have an opinion of the Trump family's actual patriotism that even vaguely matches mine, you'll realize that the phone isn't held together at all.

00:01:24   No, not well. Not well, that's for sure.

00:01:28   Yeah.

00:01:29   So it is not a good phone?

00:01:33   That's the thing, is no, but it's not as bad as it was rumored to be.

00:01:39   And who the hell knew what it was going to be anyway? Because they released like nine different renderings. Some of them were clearly done by AI. It wasn't even a device that existed.

00:01:49   Once they took an iPhone and made it gold, and they're like, yep, this is the Trump phone. Okay, yeah, good luck with that one.

00:01:55   And so by the time it finally shipped, there had been so much, I think, I don't want to say misreporting.

00:02:02   I think it was, there had been so little information that people were just surmising what the phone would be.

00:02:08   And I saw an overwhelming number of outlets report that it was some kind of rebranded $120 Chinese OEM device.

00:02:17   And I was looking at this phone compared to the Trump phone, or at least the renders that they had released.

00:02:22   And I was thinking to myself, those are not the same phones.

00:02:24   It is quite literally, it looks exactly the same as the HTC U24 Pro.

00:02:29   And so that's probably what the phone is. And sure enough, I got it. And it was, that's what it is.

00:02:33   And that phone is also a bad value, but it's around the same-ish price.

00:02:39   And so if you really wanted an HTC U24 Pro, and you wanted it in yellow, because the phone's not gold,

00:02:46   then the Trump T1, maybe not such a bad buy.

00:02:50   So I think the U24, if I recall, it's like $400. And the Trump T1 phone is $500.

00:02:57   And again, it's a couple years, yeah. And it's a couple years old.

00:03:00   So like I said, it's not that it's an outright scam. It's probably a $50 premium over the HTC phone.

00:03:07   But don't, you don't want the HTC phone anyways.

00:03:09   It wasn't a good deal two years ago when it came out, and now it's two years old.

00:03:12   So, yeah.

00:03:13   I think it might even be the 24 in U24 is that it came out in 2024.

00:03:20   So it's older, but not like ridiculously old.

00:03:25   It's not ancient, and it's not using totally terrible internal components.

00:03:31   Like it's built, it's a respectable device built by a real company.

00:03:35   It's just, yeah.

00:03:37   But what do you think really happened?

00:03:39   Nobody that cared about specs was going to buy this anyway.

00:03:42   What do you think really happened?

00:03:44   I have no idea.

00:03:47   Part of the lashback that I got, of course, from the kind of MAGA side of things was this

00:03:52   is not Donald Trump.

00:03:54   It was Eric and Don Jr.

00:03:55   And I'm like, yeah, that's the same family.

00:04:00   They share the same ventures.

00:04:01   One enriches himself off the other.

00:04:04   And I even got people that were saying, yeah, well, you know, Eric and Don Jr., they were

00:04:08   at the unveil event a year ago, but they're clearly not involved in the day-to-day.

00:04:12   Otherwise, stuff would have been more well fleshed out.

00:04:15   And I was like, really?

00:04:16   You actually think that?

00:04:17   But anyway.

00:04:17   That is something.

00:04:18   So I did a little bit of investigation.

00:04:21   I didn't end up fully reporting on it because I couldn't really with full confidence say

00:04:25   that this is what was going on behind the scenes.

00:04:27   But my very heavy suspicions, and The Verge has kind of corroborated the same, is that Trump

00:04:33   Mobile is a licensed kind of sub-provider of another small carrier called Liberty Mobile.

00:04:40   And Liberty Mobile is an NBNO that mostly licensed T-Mobile's spectrum, I think.

00:04:46   And so you're running on T-Mobile's network, but I think Trump Mobile is almost a de facto

00:04:51   kind of adjacent subsidiary of Liberty Mobile.

00:04:54   And if you go to Liberty Mobile's website, you should.

00:04:56   It is hilarious.

00:04:58   I mean, if you think the Trump Mobile website looks bad, go look at Liberty Mobile.

00:05:02   Everything about this looks like an outright scam.

00:05:04   But the customer support representative that I talked to and the director of sales who called

00:05:12   me to see if I was going to buy the phone, I looked up his name and he appears to be the

00:05:16   son of the CEO of Liberty Mobile.

00:05:19   And so I think it's probably something like Liberty Mobile and their whole value proposition

00:05:25   is patriotism, network, freedom first kind of thing.

00:05:27   Exactly what the MAGA side of things wants to believe.

00:05:30   So I'm betting they're just savvy businessmen that went to the Trump org and was like, hey,

00:05:34   you guys don't have a mobile network and we could get you one really, really easily.

00:05:39   And we could even have a phone for you.

00:05:40   And then it all just spiraled out of control and it became the built in America bullcrap

00:05:46   that was spouted for months and months and months, even though everybody surely knew that

00:05:51   that was not going to be possible.

00:05:52   And here we are a year later with a thing that kind of exists in an execution that was

00:05:58   done pretty poorly.

00:05:59   And is anybody surprised?

00:06:01   No.

00:06:02   But on the flip side, how many of these damn things do you think they had to make?

00:06:06   I mean, how many pre-orders did they sell?

00:06:08   There was some speculation about it.

00:06:10   Well, yeah.

00:06:10   Well, there had been original reporting that it was a quarter of a million, but then there

00:06:15   was actually, unsurprisingly, a huge data leak.

00:06:19   The entire customer order database got hacked.

00:06:22   I forgot about that.

00:06:23   And it turned out that there had been about 30,000 people that had put down $100 to buy the

00:06:28   phone.

00:06:28   But what percentage of those actually converted?

00:06:32   I would presume they are very, very few.

00:06:34   And my question is, I mean, it is a real phone.

00:06:38   It's got a custom back cover.

00:06:40   They moved the flash around.

00:06:41   I mean, they anodized the aluminum differently.

00:06:43   So it's not like they just stuck a sticker on a phone that already exists.

00:06:46   It's not like they went up and built a phone from the ground up either.

00:06:49   But you would have to imagine that the minimum order quantities to do this amount of work to

00:06:53   a phone has got to be pretty high.

00:06:54   And I see no scenario in which they break even on this.

00:06:59   Because they have to order, I would presume, low tens of thousands of units.

00:07:03   And by all reporting I've seen, they've sold maybe a few hundred.

00:07:08   Yeah.

00:07:09   Not great.

00:07:10   I wonder what the long tail of Android really is.

00:07:14   Like the, hey, I just want whatever free phone you get.

00:07:18   All right, fine.

00:07:19   I'll sign a contract with my carrier and I'll just get it.

00:07:24   I don't really know people who have such phones, though.

00:07:26   The people, I'll admit, most of my...

00:07:28   They're alleged to exist.

00:07:29   All of my friends are, I think of one exception, who's an Android user.

00:07:35   But in my extended family, and again, I often say this, I have, throughout this course of

00:07:41   my career, very successfully stayed out of becoming the Apple IT guy for my extended family.

00:07:49   I've taken care of my...

00:07:50   That is pretty good.

00:07:51   I took care of my parents.

00:07:52   I take care of my wife's mom.

00:07:54   And in fact, I just talked to her today about something with her Apple Watch.

00:07:58   And she's so...

00:08:00   I have a great relationship with my mother-in-law.

00:08:02   Fantastic relationship.

00:08:03   It really is.

00:08:04   It's incomprehensible.

00:08:05   And she's also...

00:08:06   But she's very, very sensitive about my time.

00:08:08   And I'm like, you can take all the time you want.

00:08:10   I have nothing to do.

00:08:10   It's just lunch.

00:08:11   Ask me your questions.

00:08:12   But very few other people.

00:08:14   I have not steered my extended family towards Apple stuff because I'm John Gruber and this

00:08:19   is what I write about.

00:08:19   It's just happened that most of my extended family is all on iPhones now.

00:08:26   I don't know.

00:08:27   So I don't know, like, how many new $300 Android phones, how many SKUs are there?

00:08:35   If there's no celebrity endorsement like Trump, and I can't...

00:08:40   It doesn't seem like this is really a thing.

00:08:42   That's the thing where I think they ran into trouble.

00:08:45   Celebrities having their own phones is not really a thing.

00:08:50   You can't go out and buy a Taylor Swift phone.

00:08:52   She could sell them if she wanted to.

00:08:54   I mean, if she did the same thing Trump did, she'd sell a lot more of these goddamn phones.

00:08:58   She'd do way more numbers.

00:08:58   Yeah, absolutely.

00:08:59   Could even be the T1 phone, right?

00:09:02   It could still be the T.

00:09:03   Hey, there you go.

00:09:04   Yep.

00:09:04   Right.

00:09:05   But she's not a scammer.

00:09:07   So she doesn't try to do it.

00:09:10   Well, there's one distinguishing...

00:09:12   I don't know.

00:09:12   What do they need to do to, you know, just to do the...

00:09:16   Like you said, they didn't just put a sticker on it.

00:09:18   They actually did anodize a yellow aluminum case and move the cameras around.

00:09:24   I'll just add this, too, before I forget, is while they were in that period at the beginning

00:09:30   of this calendar year, right?

00:09:31   They started selling this thing last year.

00:09:33   They said it was going to come out by the end of the year, and it didn't.

00:09:36   And that's when everybody started getting suspicious.

00:09:38   The Verge was running a lovely one week, once a week, did the Trump phone come out column?

00:09:44   Yep.

00:09:45   And guess what?

00:09:46   You didn't need AI.

00:09:47   The answer was no.

00:09:48   Mm-hmm.

00:09:52   But when they started getting real and they started saying, well, here's a render of it.

00:09:57   And, like, there was one that was clearly AI.

00:09:59   There was one that was...

00:10:01   I don't know.

00:10:01   And the one that was just clearly an iPhone that they colored.

00:10:05   When the one that is the actual T1 came out, I thought it might have been AI, too, because

00:10:11   the lenses, the camera lenses were, like, asymmetrically arranged on the back.

00:10:15   The unveil video was 100% done with AI, because the camera buttons move, the sheen on the back

00:10:22   the device is different.

00:10:23   On one shot, there's 11 stripes on the flag.

00:10:26   And on another, there's 14.

00:10:27   It's just totally done by AI.

00:10:29   I think they fed it.

00:10:30   The probably HTC-provided, like, CAD files.

00:10:34   And then they're like, okay, make it a thing.

00:10:36   So that's what was even crazier, is they come out, you know, a year after they've been taking

00:10:41   pre-orders.

00:10:41   And they go, okay, here, it finally is.

00:10:43   We know we've shown five different versions before.

00:10:46   One of them, this is my favorite.

00:10:47   They even released a render of it being a Galaxy S25 Ultra last year, but with a gold

00:10:53   Spigen case on the phone.

00:10:55   So they actually took the phone with the case, and there's a Spigen logo at the very bottom,

00:11:00   which is how it became gold.

00:11:02   And they didn't remove that, of course, from their crappy Photoshop job and just put Tron

00:11:05   Mobile right over the top.

00:11:06   And Spigen, I think, at replied them on Twitter and was like, bro?

00:11:10   And so the whole thing was a disaster from top to bottom.

00:11:15   Again, where I'm kind of confused, they're clearly targeting a very specific demographic

00:11:20   because I think anybody that's buying a kind of budget bargain Android phones, they're just

00:11:26   going to standard carriers anyway.

00:11:27   That's the irony about this is you've got to find it.

00:11:30   You've got to seek out Trump Mobile.

00:11:32   And by the way, their pricing is not compelling at all.

00:11:35   MVNOs are usually fairly popular.

00:11:37   I'm on one, U.S. Mobile, because I'm able to be multi-network, and they have a plan that's

00:11:42   unlimited everything for like $40, and you have pretty high network priority.

00:11:46   And so it's a good compromise from the big three.

00:11:49   But Liberty Mobile does not have compelling plans, and Trump Mobile's is even worse.

00:11:53   So the Trump plan, there's one plan.

00:11:55   They call it the 47 plan.

00:11:57   And rather than the price being $45.47, they need more money.

00:12:01   So it's $47.45 is the plan price.

00:12:06   It's $47.45 per month.

00:12:07   Why leave?

00:12:08   And I think you get like 10 gigs of priority data with the bunch being like limited to edge

00:12:13   speeds.

00:12:14   So the plan is not a good plan at all.

00:12:17   It's worse than a Mint Mobile or a U.S. Mobile plan for $35, $40.

00:12:22   And you have to bring your own device or buy a refurbished iPhone, which they also offer

00:12:28   customers, or the Trump phone.

00:12:29   I think what happened is—and it doesn't disagree with yours at all.

00:12:35   Mine's just bigger picture, which is that Trump world is—and again, putting the left-right

00:12:42   politics of it aside, which I know is very difficult for some people.

00:12:46   But just put it—put the actual traditional political divisions aside.

00:12:53   I linked to a piece that Kotke found—I think it was on Reddit last year—where Trump is

00:12:58   to stupid Americans what JFK was to Catholic Americans and what Obama was to African Americans.

00:13:05   It's like, we've never been represented at the highest levels, and now we are.

00:13:10   And he's our guy.

00:13:11   And it's—I'll put a link in the show notes to the original rant.

00:13:15   But I think the Dunning-Kruger effect comes into this at some point.

00:13:19   But a lot of people just assume that people who are stupid know that they're stupid and lack

00:13:26   self-confidence because of their self-awareness of their stupidity.

00:13:29   And that's just not true.

00:13:33   Self-confidence is, I suspect, largely unrelated to intelligence and capabilities.

00:13:40   You know, and if you think about—well, I don't know if that's true in general, but

00:13:44   I sure know a lot of really smart, confident—or really smart, competent people who are—I'm

00:13:49   always telling them, you're too insecure.

00:13:51   Why are you always doubting yourself?

00:13:52   You shouldn't be doubting yourself.

00:13:53   Everybody knows people like that.

00:13:54   And everybody knows the opposite, like their idiot uncle who thinks he knows everything.

00:13:59   And if you really realize, it's not your personal life and your circle of friends and family.

00:14:05   It's the whole world.

00:14:06   They're just disconnected.

00:14:07   But that these really stupid people with no competence, competence, have all sort of found

00:14:16   themselves.

00:14:17   And they just—part of their competence is thinking that everything in the world is easy.

00:14:23   Everything is a scam.

00:14:24   Everybody else is pulling one over on other people.

00:14:27   How hard can it be, right?

00:14:29   Like, all of them, if they were capable of actually writing a book, could title their

00:14:33   book about their time in Trump world as, how hard could it be?

00:14:36   And somebody said, you know what?

00:14:39   You know where a lot of these companies are making a lot of money is cell phones.

00:14:43   Everybody's got a cell phone.

00:14:44   We should get some money on cell phones, too.

00:14:47   And that's it.

00:14:49   A lot of people were making money on crypto.

00:14:52   A lot of people are making money.

00:14:54   How about we make money on crypto, too?

00:14:55   A lot of people—

00:14:56   Including the Trumps, by the way.

00:14:58   And unlike this, where I don't think that their Trumps are making much money at all.

00:15:03   The New York Times had an investigation where the crypto coin, they made like $1.8 billion

00:15:09   or something.

00:15:10   Or he, Trump, the president of the United States, made.

00:15:13   And the Times did the work and figured out that American citizens who bought his crypto, which

00:15:20   is now, and that what they bought it at is now underwater, have lost like a collective $2.8 billion.

00:15:26   So he got like $1.8 billion of it.

00:15:29   Another billion went to whoever else organized the crypto.

00:15:33   And some unbelievable number of gullible, well-intentioned, believed in their president, Americans, lost $2.8 billion.

00:15:53   I don't think they're going to make billions from the phones.

00:15:56   But how hard can it be, right?

00:15:58   And it works out sometimes.

00:16:00   And I think that's the same gambit.

00:16:02   They found these Liberty Mobile guys are probably Trump donors.

00:16:06   And they said, hey, we've got an opportunity for you guys.

00:16:09   Eric and Don Jr. said, oh, there's a chance to make money.

00:16:12   Yeah, sure, let's do it.

00:16:13   You can parade us around, put us on the stage.

00:16:15   We'll say it's going to be the most American phone.

00:16:16   We'll pull a bunch of headlines, which they did.

00:16:18   And they got news coverage for months for free, effectively.

00:16:21   And they knew they were peddling a scam.

00:16:23   But that's what they do.

00:16:24   And then they ship a piece of crap that's kind of technically, I guess, what's partially what they promised it would be.

00:16:31   And they have enough of a product.

00:16:33   And it's real enough that they maybe shield themselves from most of the liability.

00:16:37   And who's going to sue them anyway?

00:16:40   And it might not work, but it might.

00:16:41   And so let's try it.

00:16:42   And if we throw 50 darts at the dartboard, one of them ought to stick.

00:16:45   And I don't think this is the multi-billion dollar crypto scam.

00:16:48   But it's just one of the many things the Trump org has always kind of done and put their name in front of to see if it sticks.

00:16:53   And the stuff that doesn't, like this, will probably just disappear in a couple of years.

00:16:58   It'll remain probably some weirdo kind of subsidiary of Liberty Mobile that is managed and operated by them.

00:17:05   And the couple dozen or hundred customers that continue to pay for the Trump plan.

00:17:09   We'll just have their old legacy dashboard available to them.

00:17:12   There's basically no overhead.

00:17:14   So what's the risk?

00:17:15   Right.

00:17:15   And how hard can it be is a reasonable question to ask in life.

00:17:21   And there are a lot of things in life in broad areas where maybe it's not as hard as everybody thinks it is or times change and technology changes and things that weren't hard.

00:17:33   I mean, effectively, what you and I do for a living is an answer to how hard can it be from the perspective of being a full-time media professional from, like, a generation ago.

00:17:44   Sure.

00:17:45   You couldn't do it on your own a generation ago.

00:17:47   You certainly couldn't do the video aspect that you do.

00:17:50   You'd have to get on television, and getting on television meant somehow, depending on how far back in time, getting on a very limited number of networks, right?

00:18:00   Even at the biggest that cable ever got, compared to YouTube, the opportunities for people to make a living on television were infinitesimal.

00:18:12   Yeah, compared to now.

00:18:14   And if you go back a little bit further to my childhood, they were incredibly rare because there were really only three real networks and maybe four if you count PBS.

00:18:22   How hard can it be?

00:18:23   It is still hard, but it's, like, a good question to ask.

00:18:26   But there's a big difference between crypto and cell phones because crypto, without going on a whole side rant about it, I mean, and it is kind of unique.

00:18:36   Like, maybe in the history, there's really very hard to say, oh, crypto is like, cryptocurrency is like X.

00:18:44   No, there really isn't anything in the past that was like it.

00:18:48   And there are cryptocurrencies, Bitcoin in particular, that has lasting value, apparently.

00:18:54   You know, it does not seem like it's going to crash.

00:18:56   It is real.

00:18:58   Whether it's a net win for the world, I don't know.

00:19:02   I don't want to debate it, but there is some cryptocurrency that has real value.

00:19:07   There are people who have, if you got in early enough on Bitcoin, probably there are people listening to this show who have made an awful lot of money by putting a relatively small amount into Bitcoin 10, 15 years ago or whatever.

00:19:20   There is real value, but also, because there is real value, it is the perfect, and you can make scam ones and rug pull type ones.

00:19:33   It is the, there's never been a way to make computers work better for running a scam than cryptocurrency.

00:19:40   Because there is a real there there, for some of them, and you can just make a new one, and like Trump currency, whatever the hell, Trump coin, whatever his was called, take people's money, take your end and convert it to actual dollars right away, and then see what happens.

00:19:58   And nobody would have been happier than if Trump coin had held its value or gone up in value.

00:20:04   Trump might have actually been angry if it did, because he'd be like, well, why don't, why aren't I making more money from this now that it's held its value?

00:20:10   Sure.

00:20:11   Right?

00:20:12   But it's all just, it, what did somebody long ago, somebody's great description was like explaining crypto to their mom, and it was like, imagine if you could burn fossil fuels to solve Sudokos.

00:20:26   And it's like, yeah, yeah, yeah, that's, that's basically it.

00:20:30   You're converting energy, and in the middle, there's a lot of GPUs, and at the end comes out, it's solved Sudokus.

00:20:38   And that brings us to AI.

00:20:41   No, I'm just kidding.

00:20:42   Well, but cell phones are real.

00:20:45   The difference is a cell phone is real.

00:20:47   A cell phone network is a real thing.

00:20:49   People get a phone, and they expect it to work, right?

00:20:53   And it either does or it doesn't.

00:20:54   And I think at the outset of this whole endeavor, there were people who were like, ah, how hard can it be?

00:20:59   We'll make a phone.

00:20:59   And somebody bought into the whole, why can't we just make one in the United States?

00:21:04   How hard could it be?

00:21:05   And they, turns out they were wrong.

00:21:07   And then I think what happened is shit got real around the new year, because the publicity around, hey, they took all this, these $100 deposits.

00:21:15   If you read the fine print, if you read the fine print, it actually says that they don't have to give these deposits back, right?

00:21:23   And so people really, and I thought, ah, I think they're going to ship some kind of phone eventually.

00:21:28   But I was like, would I be shocked if they just kept the $100 deposits in forever and said, whatever?

00:21:35   I wouldn't, with these people, I wouldn't be surprised.

00:21:38   But I think somebody realized, and somewhere down the legal chain, there are people whose last names aren't Trump who were on the hook for this.

00:21:47   And that's what it reads to me.

00:21:50   If you look at the box, the packaging, this was clearly something put together quickly.

00:21:54   This is not the phone that they were making a year ago.

00:21:57   This happened within the last few months, right?

00:22:00   But it is a real, tangible thing.

00:22:03   Somebody who knows their shit about the phone industry got called in as a fixer.

00:22:09   And they're like, here's the situation.

00:22:10   We took these number of pre-orders.

00:22:11   We told them they'd get this.

00:22:13   And this person said, well, we can find, like, an HTC phone.

00:22:17   We'll white label it.

00:22:18   I know somebody.

00:22:19   We can make a custom gold-toned or yellow-toned case.

00:22:24   We can do this for $500.

00:22:26   I'll bet they're not.

00:22:27   I'll bet they're, like, breaking even at $500 overall, especially in the quantities they're talking about.

00:22:32   But presumably, everybody who is buying one is going to stay on this $47, $45 plan.

00:22:37   And they'll make money.

00:22:40   That's where the money is, theoretically.

00:22:40   Yeah, that's where the money is anyway, right?

00:22:42   And so, you know, sell what's costing them probably $20 or $15 a month in service and make $30 a month from them on this.

00:22:52   And that's actually pretty good money.

00:22:53   And that's also why I think they are probably going with the HTC phone as opposed to one of these sketchier Chinese kind of reskins that they were rumored to is I honestly think it comes down to competence.

00:23:06   If they could have done it, they probably would have, but they needed something that was going to be pretty well put together and has a decent track record and they could get in the order quantities they needed.

00:23:17   And so they went to a real company that was accessible and has people that's worked with the West and they don't have to start from scratch.

00:23:26   They can be pretty highly confident that the phones that they are going to ship aren't going to have spyware on them.

00:23:31   You know what I mean?

00:23:31   Like, I kind of made that joke in the video.

00:23:33   There were all these people suggesting, oh, this thing's going to be totally loaded with spyware.

00:23:38   And after I unboxed it, I said, no, this is just an HTC phone.

00:23:41   But because I wanted to see the bootloader came unlocked on this device and I ended up installing some software on it to do some packet sniffing and to see what network activity it was doing.

00:23:50   Everything was totally fine.

00:23:51   All kosher.

00:23:52   No big deal at all.

00:23:53   Which I suspected, but I also kind of suggested, had they gone with a cheaper Chinese device, that actually might have been a problem.

00:24:01   Not because the Trump organization or Trump Mobile wanted to spy on their customers, but as a consequence of incompetence.

00:24:07   Like, I think that a bad foreign actor might have actually tried to do it, but the Trump Mobile people, they wouldn't have even known.

00:24:16   And so they probably just said, yeah, let's go with someone who's reputable, who has a track record.

00:24:20   We can get this thing out the door and, yeah, we're not going to make money off of it, but we will on the backside on the plan.

00:24:24   And then if you believe their website, this is a temporary promotional pricing and that the phone price of the T1 won't go up over time.

00:24:32   I don't think so, but yeah.

00:24:34   I think they would really just like for people to forget about it.

00:24:38   And probably if it remains a going concern that the MVNO on an MVNO, they're just going to get away from their custom phones and just give people whatever phones, used iPhones or whatever.

00:24:51   I mean, they're already doing that where you can just buy an iPhone.

00:24:53   It's actually one of the things that they have found out, and that I'm sure the person or persons who came in to fix this and make the real phone from HTC were like, you dumbasses, there's no money in phones.

00:25:05   Only Apple and maybe Samsung make money in phones.

00:25:08   That's it.

00:25:08   And nobody else makes money.

00:25:10   There is money in people owning cell phones, but it all comes on the carrier side.

00:25:15   And I think they might have known this.

00:25:16   I think that the phone was the prop to prop up the network.

00:25:20   I mean, that was the thing that was going to grab the headlines, the gold phone, made in America.

00:25:24   I mean, if they had come out and said, hey, the Trump family, we've got an MVNO, nobody would have cared.

00:25:29   The phone was what catapulted them into the front of the world, and I think they knew that, and that's why they did that.

00:25:36   And now that Trump Mobile is a thing, well, the phone exists as a byproduct of that kind of promotion, but it's not the focus, and it never has been.

00:25:45   It never will be, and frankly, it's probably, based on the people that I've talked to at Trump Mobile, in order to get the darn thing, I would suspect they have fewer than five people working there anyway.

00:25:55   And I think they all already work for Liberty Mobile, and so their overhead is effectively zero.

00:26:00   And so I think the Liberty Mobile guys, this would be my suspicion.

00:26:04   This is all speculation.

00:26:05   I have no evidence to back this up.

00:26:07   I would suggest that it's most likely the Liberty Mobile guys that are getting the sweetest end of the deal, and I think they might even be a little disappointed that it's not panicking.

00:26:15   They're not panning out quite like they had hoped or expected either.

00:26:17   Right, that they had maybe put competent people in charge of the T1 phone earlier to make it more of a thing.

00:26:23   You know, and it is maybe to his credit that his scams are sort of a shotgun approach, see what sticks.

00:26:30   I don't think they made a lot of money selling those mechanical watches.

00:26:36   Nor do I.

00:26:37   I don't know.

00:26:38   Maybe they made more money on the Trump coins that they've sold, but by trying everything and seeing what sticks, they found the cryptocurrency one, which actually made them billions of dollars.

00:26:49   Right.

00:26:50   And if you get a couple successes through dozens of failures, well, that's fine.

00:26:54   Just add it to the Trump stakes list.

00:26:56   You know what I mean?

00:26:57   Yeah.

00:26:57   There was, I forget, there was somebody who bought the Trump mechanical watch that, you know, which is sort of like, of course, also gold toned and looks very Rolex.

00:27:08   It really looks just, you can just smell it, that it smells like the type of Rolex you buy from a guy with a card table in lower Manhattan.

00:27:17   Yep.

00:27:18   It's got like a bargain bin Miota movement in it.

00:27:21   It's like very, they probably built the thing for less than a hundred bucks.

00:27:25   It was some local, I'll make a note and see if I can find it for the show notes, but there was some local news somewhere who, and I love, sometimes local TV news can make the, get the best stories.

00:27:37   It's like, but some guy bought his wife a Trump phone or Trump watch for her birthday, and it was late, of course.

00:27:47   And then when it did arrive, the Trump logo was with applied, applied to the dial.

00:27:54   You know what that means, but they're like little, a T and an R and a U that are like glued onto the dial.

00:28:00   But I forget, I swear to God, it might've been the T, I swear to God, fell off.

00:28:06   And it was rattling around in there, leaving the word rump, but it, maybe it was the P, I don't know.

00:28:11   But one of the letters fell off and was rattling around in there and it didn't, guess what?

00:28:16   It didn't keep good time.

00:28:17   Canal street quality.

00:28:19   You get what you pay for or not, I guess.

00:28:21   Yeah.

00:28:22   But anyway, the video is just magnificent.

00:28:27   And it's, it's, I think it's much higher viewers than your typical videos.

00:28:32   All of your videos are good.

00:28:33   Oh yeah.

00:28:33   But this one struck gold, no pun intended.

00:28:37   We thought it would do well, but I, even I was surprised by how well it did.

00:28:41   It was a, it was a lot of fun to make.

00:28:43   And you know, we were late to the, that's kind of why I was a little worried is we were late

00:28:46   to it.

00:28:47   I had had the phone for almost three weeks before we finally published the video, but I wanted

00:28:51   to make sure it was all really well, nicely done.

00:28:53   I fix it had already had a tear down by the time we put our video up.

00:28:56   It was already all shot.

00:28:57   But so when we put it up, I was like, oh man, this is not going to do as well as I hoped.

00:29:01   And then it got a couple of million views in two days, I think.

00:29:05   I mean this with no hyperbole.

00:29:08   I really do.

00:29:09   But I really do think that you put more work and time and a combined effort into your video

00:29:14   than all of what went into making the Trump T1 phone in the first place.

00:29:19   Even, even when they hit the reset button at the beginning of the year and were like, let's

00:29:24   just go to HTC.

00:29:25   I still don't think it was as much effort or attention to detail as, as went into your

00:29:30   video, which is just, certainly supported more American jobs and salaries.

00:29:34   That's for sure.

00:29:35   All right.

00:29:37   Let me take a break here and thank our first sponsor.

00:29:38   It's our friends at Notion AI.

00:29:40   Notion AI is the all in one AI powered by your work and it's all in one place.

00:29:46   It automatically captures meeting notes, instantly finds the exact content you need, drafts, detailed

00:29:52   docs for you and lets you chat with the best AI models.

00:29:56   Notion AI just became twice as powerful for teams, making it the best AI tool for work.

00:30:02   AI meeting notes.

00:30:04   Now, I'm lucky.

00:30:05   I live a life where I don't really have a lot of meetings.

00:30:08   But I know that everybody out in the real world with real jobs, that meetings are a real thing

00:30:14   and meeting notes are a real thing.

00:30:16   AI meeting, and there are a zillion AI meeting note solutions out there.

00:30:21   But Notion's AI meeting notes is just perfect for collaboration and for ease of use.

00:30:29   It captures every idea, every decision, every next step that is proposed in the meeting.

00:30:34   It's all captured in Notion, which is what your team is already using to collaborate and

00:30:40   share shared information.

00:30:43   So the shared meeting notes are right there in Notion, ready for your whole team to share,

00:30:47   where they can be put to work right away.

00:30:49   And even the AI will even follow up at the end of the meeting with suggested action items

00:30:54   immediately after the meeting, which you can take or leave or adjust as you want.

00:30:57   You don't have to just take AI's word for it.

00:30:59   But here's, hey, that's actually a good list of action items.

00:31:03   They got it.

00:31:03   And then you don't have to think about it.

00:31:05   You can capture anything and turn it into organized, searchable notes, including regular

00:31:10   group meetings, one-to-one conversations, lectures, even podcast episodes.

00:31:15   You can capture this episode of the talk show and get it summarized right there in Notion

00:31:20   AI.

00:31:20   It works on any meeting platform, Zoom, Google Meet, et cetera, or just with your computer

00:31:26   system audio.

00:31:27   Like if you're in a room together, the old-fashioned way, having an actual meeting face-to-face,

00:31:31   just capture it right on your device.

00:31:33   But if you're doing it remotely, it integrates with any tools that you might have.

00:31:37   Just type slash meet on any Notion page or on desktop or mobile and get started instantly.

00:31:43   Check out Notion, the best AI tool for work right now at Notion.com slash talk show.

00:31:50   And they want me to emphasize that all of those letters have to be lowercase.

00:31:55   Notion.com slash talk show.

00:31:58   All lowercase letters to try the powerful all-in-one Notion AI today.

00:32:01   And when you use that link, you are supporting the show.

00:32:06   Well, that's enough talk about Trump, I guess.

00:32:09   Yeah.

00:32:10   One more thing.

00:32:11   And about Notion, I suppose.

00:32:14   Notion's not paying me, but I do use Notion every day.

00:32:17   And I just asked Notion AI, what did I say the Trump phone was held together with?

00:32:21   Because you mentioned that earlier.

00:32:22   And it found it in my script.

00:32:23   You said it was, quote, a scam held together with patriotism and golden paint.

00:32:27   There you go.

00:32:28   Thank you, Notion AI.

00:32:30   It did it.

00:32:31   It did it.

00:32:31   It's real.

00:32:32   Isn't that amazing?

00:32:35   It's pretty cool.

00:32:37   I forget his exact quote, and I can't be bothered to look it up.

00:32:41   But my friend Daniel Joukett of Red Sweater Software, the makers of Mars, had a good long-term

00:32:45   personal friend, but he had a tweet a couple of weeks ago, or maybe right before WWDC, that

00:32:51   was basically like, the AI enthusiasts are way too into AI, and the AI skeptics are way too

00:32:58   skeptical of AI, and that the truth is somewhere in the middle.

00:33:01   And it's just turned into such a polarizing thing.

00:33:04   But there is a there there that is amazing at times, right?

00:33:10   And that's a perfect example.

00:33:11   Oh, yeah.

00:33:11   Right?

00:33:12   Yeah.

00:33:14   Yeah, no, I started out as a heavy skeptic, and I still am not a believer that it will

00:33:21   change every aspect and facet of our lives.

00:33:23   But to think that it's completely worthless is naive.

00:33:27   It's a thing.

00:33:29   Yeah, I'm exactly on board with that.

00:33:31   Me, you, and Daniel Joukett, apparently the only three people.

00:33:34   Well, speaking of AI, I guess we can roll into on the more recently news topics is,

00:33:44   I've taken a great interest in the native Mac apps, I don't know if you've noticed, on Daring

00:33:52   Fireball for ChatGPT and Anthropics, Claude.

00:33:55   And while I'm in the, and I knew this was coming, right?

00:34:00   There was the whole pre-announcement of the quote unquote super app, which is such a dumb

00:34:05   name and such a red flag.

00:34:07   And I felt like, my God, is this predicting in advance that this is going to be a disastrous

00:34:13   user experience is the easiest pickings of my prognosticating professional life.

00:34:20   So I kind of, and of course it was going to, you know, with a native app kit, Mac app for

00:34:27   ChatGPT and an Electron Codex app and the whole Atlas web browser that was a Chromium thing that

00:34:36   they had wasted so much time and money on, of course those things were going to serve as

00:34:42   the foundation of the super app, not the native app kit, ChatGPT app.

00:34:47   But as I'm writing about it is when they unveiled it.

00:34:50   And boy, oh boy, did they, I expected it to make me unhappy personally, but I did not expect

00:34:58   it to be maybe the least competent launch of a mass market app in the entire history of

00:35:08   the PC industry.

00:35:09   I'm trying to think of an example.

00:35:12   I'm trying to think of an example that was worse.

00:35:15   And they have, I know the Mac app didn't have a billion users, but ChatGPT has a billion

00:35:20   users and they, they really botched it.

00:35:23   I don't know what you think, but.

00:35:25   It was bad top to bottom.

00:35:27   I mean, their whole entire product is ChatGPT, but you wouldn't know that based on the fact

00:35:33   that the new app basically shoved chat in a corner and you're like, that's what people

00:35:38   do.

00:35:38   And the whole thing about these AI models anyways, is that you often don't have to provide their

00:35:45   context switching capable.

00:35:46   And so the fact that I have to go into a specific mode to do co-work or codex or any

00:35:53   other is so weird because with the previous app, you could just go into a chat session and

00:35:58   say, hey, I want to do this thing.

00:35:59   And I would say, hey, we really should, if we're working on files on our, on our Mac, we

00:36:03   should move to this mode.

00:36:03   And you go, okay, it just does it.

00:36:05   And now it's just flabbergasting.

00:36:07   Even just the, what was it called?

00:36:10   The blocking and tackling, just the basics of, okay, we would like to rename an app.

00:36:19   And that's, that's a tough thing.

00:36:21   I know it's a rose by any other name would smell just as sweet, but it is hard to rename

00:36:27   an app.

00:36:27   There's just a lot of little things.

00:36:30   And the list of things that you have to address are at a technical level are, it's long and

00:36:37   you want to make sure software updates go right.

00:36:40   And then there's the whole marketing angle of communicating with your users.

00:36:44   We've changed the name of this app, but I can't recall anybody ever wanting to change

00:36:51   the name of an app that does X into the existing name of an app that does something almost entirely

00:36:59   different.

00:36:59   Yes.

00:37:00   Yes.

00:37:01   And to make matters worse, to, to let you choose, hey, do you want the old icon or do

00:37:06   you want to keep this icon or what do you want to do?

00:37:08   Yeah.

00:37:08   It's chat GPT now, but it's actually kind of codex under the covers, but you use the new

00:37:12   chat GPT logo, but you can also use the old codex logo and you're like, what are we doing

00:37:16   here?

00:37:16   Yeah.

00:37:18   And with your bit, the bit you wrote about that caught me is that when you download the

00:37:22   new version of the app and mount it as an image on, you open the thing up and rather than

00:37:27   be prompted to drag it to the applications folder, it just says double click this app here.

00:37:31   But all that does is open another, another instance of this separate app.

00:37:36   And at least as they shipped the 1.0 of this, I don't know what the version numbers are, but

00:37:43   the 1.0 of the new chat GPT that's the renamed codex.

00:37:46   And it has the had chat GPT itself, the actual chat tucked into a corner and they shipped

00:37:53   this Mac app that said double click this.

00:37:55   As best I can understand it.

00:37:58   If you already were a user of chat GPT and codex apps on your Mac and you just double clicked

00:38:06   it, then it would update your codex to be chat GPT the name and it would replace your old

00:38:15   chat GPT app with chat GPT classic.

00:38:19   But that's if and only if you already had both of those installed.

00:38:23   And of course they didn't say anything like that.

00:38:26   All they said is double click this.

00:38:27   Now I personally had never installed codex.

00:38:30   I just, I've never gotten into that whole area of AI yet, even though I'm dithering.

00:38:36   Ben Thompson's always telling me I should.

00:38:38   And I do want to, I really do.

00:38:41   Cause I mean, I even have a computer science degree.

00:38:43   I, you know, I can use this to, to make software and I've been meaning to try it.

00:38:49   And I have been a chat GPT man, not a Claude man for chat driven stuff.

00:38:55   So I probably was going to try codex first.

00:38:57   I still might even after all of this, but I'd never installed codex.

00:39:01   So all I did was double click it.

00:39:03   And all it, all it did was launch a second version.

00:39:05   This is my experience as a user.

00:39:07   It launched a second version of chat GPT in my doc.

00:39:12   And I could see, and I still had the old one running and then the new one just bounced

00:39:17   for a bit and then it disappeared and I went and looked in my applications folder and there's

00:39:22   one app and it wasn't called, it was just called chat GPT, but it was the old app and

00:39:27   there was no new app and no app named classic.

00:39:29   And I launched it and it wasn't even updated.

00:39:32   It was the one from like a week before.

00:39:34   So then I quit that app.

00:39:36   I, the reason I saw them both in my doc is I'd left chat GPT running when I launched the

00:39:40   new one, which I didn't know what to do.

00:39:42   I, if I had had to drag it to the applications folder, I would have known I need.

00:39:45   So I quit it, then launched it again.

00:39:47   And I, it updated my chat GPT app, but didn't rename it to classic.

00:39:53   And so at least if I went to software updates, but it didn't rename it to classic.

00:39:58   And I'm like, well, how do I get this?

00:40:00   How do it?

00:40:00   And the answer, I even wrote a post, like, I think I even put the question mark in the

00:40:04   title.

00:40:04   How do you get the chat GPT classic app if you didn't get it automatically?

00:40:07   And there's no document.

00:40:09   I even quoted it.

00:40:09   There was a help document on their website that was like, after you upgrade, you may

00:40:14   have both chat GPT classic and chat GPT.

00:40:17   And they didn't say why you may or may not, because I'm not sure they knew.

00:40:21   I've never seen any.

00:40:23   And again, this is a company with, it's privately held at the moment, but it's, their self-proclaimed

00:40:31   valuation is around a trillion dollars with a billion users.

00:40:36   And this is how they launched an update to their flagship product.

00:40:40   It is bananas.

00:40:41   It's honestly flabbergasting.

00:40:42   Yep.

00:40:43   And on your point, the fact that it was designed for people that had both chat GPT and Codex

00:40:50   installed, they probably didn't even QA it with, I bet there's not a single person that works

00:40:54   at OpenAI that doesn't have Codex.

00:40:56   So they probably never even ran it on a machine without Codex installed.

00:40:59   So they had no idea it was going to happen anyway.

00:41:00   And they've.

00:41:02   And chat GPT and Codex didn't catch it.

00:41:04   Yeah, a week later, they kind of fixed this.

00:41:07   And it seems like now you can get both.

00:41:10   They've kind of cleaned it up and they've done some work, some obvious work in the sidebar

00:41:15   so that there's actually a top level chat.

00:41:17   Chats are in the sidebar now in the bottom.

00:41:19   They push it below all of your Codex projects, which is honestly pretty annoying, but they're

00:41:24   there.

00:41:24   And yeah, that's the thing that's the most frustrating to me is that Codex and what do they call it?

00:41:31   Do they call it work or is that clock?

00:41:32   Yeah, work.

00:41:33   No, it is work.

00:41:33   Yeah, yeah.

00:41:34   They still appear in line.

00:41:36   But again, like you mentioned, they're not the same function.

00:41:39   I mean, I have a chat GPT worker setup that goes through my email inbox using the Spark CLI

00:41:46   every morning and cleans up my email inbox and preps it for me with automatic tagging.

00:41:50   It's great.

00:41:51   Love it.

00:41:51   But it is now kind of an equal priority to other apps that have been developed using Codex

00:41:58   on my machine.

00:41:59   And you're like, this is not even they're not even the same ballpark because one is coding

00:42:03   brain and one is using kind of agentic pre-existing MCPs.

00:42:08   It's just bizarro that they're all thrown into this.

00:42:11   And then they're like, oh, yeah.

00:42:12   And then if you want to chat, it's just at the bottom.

00:42:14   So I have been doing, and it's probably what they want.

00:42:16   I've been going to chat.com to freaking use ChatGPT as a chat bot.

00:42:22   And I'm just devastated because you go back six months ago, and they had an amazing, native,

00:42:27   excellent Mac app.

00:42:29   And now it's gone.

00:42:29   Why not just do what I do and just stick with ChatGPT Classic?

00:42:32   I could.

00:42:35   I just hate having two apps that are the same but different.

00:42:39   And I use Spotlight to launch everything.

00:42:41   And it's, yeah, you want ChatGPT.

00:42:43   I'm like, no, no, no.

00:42:43   I want a classic.

00:42:44   So I have to retrain my brain.

00:42:46   Type classic when you want that.

00:42:47   And the one that's now called ChatGPT Classic.

00:42:52   It's much lighter on system resources and memory.

00:42:55   But if you are going to be using the other one anyway because you were using Codex already,

00:42:59   it does feel like, well, the last thing I need to do is give more of my precious RAM to running processes.

00:43:07   And I already have a browser open.

00:43:09   It is funny.

00:43:11   That is sort of the – it's not my biggest complaint about Electron, honestly, the RAM consumption.

00:43:17   It's really the lack of native UI idioms and support for them.

00:43:22   Yeah.

00:43:22   But it's a list.

00:43:24   It's a long list.

00:43:26   And on there is the memory consumption.

00:43:28   And it does seem – again, I can't think of any comparison in the history of the computer world where they're responsible on both ends.

00:43:39   On the one end, it is the AI mania building out data centers that has consumed – they're outbidding everybody else for the limited supply of actual RAM,

00:43:52   which has created this unprecedented RAM shortage that has driven up prices.

00:43:57   And even with the driven up prices has created product shortages, right?

00:44:02   There are still configurations of, like, Mac Minis that you cannot get.

00:44:06   I don't think you can get a Mac Mini with more than 96 gigabytes of RAM or something.

00:44:11   That's right, yep.

00:44:11   And it used to be a lot more, and now you go to eBay, and people are trying to sell loaded-out Mac Studios with more RAM than you can get today for, like, the price of a BMW.

00:44:23   It's crazy.

00:44:25   I don't know if people are buying them, but if you really need 512 gigabytes of RAM, you cannot get it from Apple.

00:44:32   It's crazy.

00:44:32   It's all from AI mania.

00:44:35   Certainly, you can't blame them entirely for the mania.

00:44:39   It's the people who've become maniacs bear some responsibility.

00:44:43   But then on the user side, they are pushing apps that use way more RAM on your personal computer than they should for the functionality they provide locally.

00:44:55   So they're driving up demand for RAM on your personal device while they're the ones to blame for the price increase and shortage.

00:45:04   Correct.

00:45:04   And I understand going with Electron as a framework, it's easy to use, it's more scalable, it's multi-platform agnostic.

00:45:13   However, they already went to the effort of building an excellent native Mac app.

00:45:17   Like, who's already done?

00:45:19   I do think.

00:45:21   Well, I keep coming back to it, and it's a bigger, broader question.

00:45:27   And it gets to a point that you raised about, or that we talked about, about where on the spectrum of, oh, the whole world is different now because of AI, or AI is entirely a scam.

00:45:38   There's nothing to it.

00:45:39   It's just a ripoff engine.

00:45:40   There's nothing.

00:45:41   I'm 100% convinced that both of those polarized views are wrong.

00:45:49   The truth is in the middle.

00:45:50   But to me, it is really, really interesting with both of these companies at this point, but at the Anthropic in particular, which is the one I wrote about first with their use of Electron, where if they're so good at coding, and one of the, you even mentioned it, one of the main reasons, and it is understandable that why organizations, big or small, for different reasons, would want to go with a write-once, run-everywhere solution.

00:46:18   This has been a recurring thing in computing all the way back to when writing programs was first.

00:46:28   We didn't even call them apps.

00:46:29   They were programs.

00:46:29   But the idea of having the ANSI standard for C so that you could write your code to the ANSI standard and then move the same C program to a different version of Unix or a different platform than Unix and still have it compile because you wrote it to the standard, right?

00:46:50   There have been efforts like this, and it makes sense.

00:46:52   Why in the world would you want to rewrite the program for every computer it runs on?

00:46:55   But if AI is so good at writing programs, why in the world can't they make it existence proof of how good they are by using Claude to make a native Claude Mac app, right?

00:47:08   And if it's...

00:47:08   Yes.

00:47:09   The argument is we don't want to spend all this time and all these resources on a platform just for the Mac, but it shouldn't, isn't the answer that your AI can do it?

00:47:17   And I think the evidence is no, it can't, right?

00:47:21   They are good at generating code, and I know people.

00:47:24   I've written scripts, and I've written things with ChatGPT that are functional scripts.

00:47:30   I can see it, and I have close friends who are using it to help them make apps.

00:47:34   But I don't know anybody who's made an entire app this way.

00:47:38   Or if they have, it is really limited in scale.

00:47:43   And it might be useful and clever, but it's not, hey, here's an entire replacement for Claude for the Mac, and it was entirely written by Claude.

00:47:52   I don't think that can happen at this time.

00:47:55   And if they want to prove me wrong, they should.

00:47:58   They probably could, but with more oversight than they're willing to allocate.

00:48:00   I think that's the main issue is they just don't care.

00:48:03   Right.

00:48:04   And I think that's a pretty sad condemnation, but I think that's probably the reality is even though it's possible, it's just not a priority,

00:48:12   and they know that people are going to use the product anyway, and so there's just whatever's easier for them to scale more quickly and rapidly.

00:48:18   And if they have to do less QA, and if you don't have multiple binaries, and you have to test multiple platforms,

00:48:23   and you can just do it once everywhere, then that's what they're going to do.

00:48:25   But it is a bit ironic that one of the selling points of these tools is how easy it makes at taking code in one language

00:48:35   and translating it to another and from one platform to the next, but they don't.

00:48:40   And part of it, it really is, I think, A, you need the expertise still, right?

00:48:45   The people who are best able to make a great Mac app, whether it's an all-new app or just rewrite an existing app

00:48:52   or add new features to an app using AI, are expert Mac developers.

00:48:57   To do it on iOS, to already be a talented iOS developer.

00:49:01   In their hands, AI can accelerate greatly what they're capable of producing.

00:49:07   And there are people who are not developers at all who are doing this, but they care, right?

00:49:12   I think Jason Snell, I forget the name of the, I think it's called Double Ender, is the name of the app.

00:49:17   It's a podcasting solution app where you and I right now are actually recording a Double Ender,

00:49:23   where you're recording audio locally and I'm recording audio locally,

00:49:26   and then it's going to go to the hands of my editor.

00:49:30   And he would like to align those two tracks, and it's a Mac app that does this in a very Mac-esque way,

00:49:37   does this one thing and does it really well.

00:49:39   And Jason is not a Mac developer, but he's a, he knows the, he cares about the idioms of being a Mac-esque Mac app,

00:49:49   and therefore was able to direct it to do this.

00:49:53   And it probably would have been easier for him to make it a web app and just drag the files into a web interface,

00:49:58   but he wanted a native Mac app because he cares.

00:50:00   And Glenn Fleischman wrote an app called Mr. Plimsoll, which I think that's the name of it,

00:50:05   but it's a super clever little app.

00:50:07   And you could just set it up to say, hey, pay attention to this drive on my Mac,

00:50:11   and when it gets to this percentage full, tell me.

00:50:15   Because maybe you have a drive where you store, like you might have a drive where you store video,

00:50:20   and it can fill up very quickly.

00:50:23   And you would like to know before the wrong time that you've reached 85% capacity.

00:50:29   It's a very Mac-esque app.

00:50:31   And Glenn is, again, he's a programmer, and he's written programs over time, but not a Mac developer.

00:50:37   But he cares, and he cares about the Mac-esque-ness of it, and therefore is able to do it.

00:50:43   And obviously, if anybody at Anthropic really cared about great native Mac apps,

00:50:47   they wouldn't have had to wait for Claude to be capable of making one.

00:50:51   They would have made one in the first place.

00:50:52   And somebody at OpenAI cared enough a couple of years ago that they put a team together at OpenAI

00:50:59   to make a native pretty good.

00:51:03   I mean, and it's like I sing the praises of it because I'm comparing it to Claude and everything else that only runs in a browser.

00:51:09   It's actually not a great Mac app.

00:51:11   It never has been.

00:51:12   Like, one of the things that, to me, just defines a Mac app is being able to open whatever the element of the app is.

00:51:20   You should be able to open that into a window on its own.

00:51:24   So Apple Notes.

00:51:25   You can go into Apple Notes, find a note, and double-click on the note, and it opens into a window on its own.

00:51:31   You can go to Apple Mail and find any mail message or a thread of them and double-click on it in the list of messages,

00:51:38   and guess what?

00:51:38   It opens into a window on its own.

00:51:40   And even messages from Apple, which they rewrote using Catalyst a couple of years ago,

00:51:46   and when they announced that they were going to do that, I was like, uh, in other words,

00:51:50   bring sort of the iPad version of messages to the Mac,

00:51:53   I got worried that it was going to lose a lot of its Mac acidness.

00:51:58   And in actuality, the version that they shipped was actually a better Mac app than the one that it replaced,

00:52:05   including support for AppleScript.

00:52:08   I don't use AppleScripts against messages often, but I have a couple of things I do.

00:52:14   I have a couple of scripts that run for a long period of time, and I want to know about the progress.

00:52:18   And then years ago, I just wrote a thing that just sends a text message to myself in messages while it's running,

00:52:25   and then I can walk away, and then my phone, like, every half hour gets a progress message from myself with the thing,

00:52:32   because AppleScript says send a message to my iMessage ID with this text and this status update.

00:52:39   All of that continued to work, and guess what?

00:52:41   When you double-click on a chat in messages, guess what happens?

00:52:44   It opens up into a window on its own.

00:52:46   Guess what you can never have done in ChatGPT Classic?

00:52:49   It cannot open into a window on their own.

00:52:52   And it still drives me crazy, but I don't want to complain too much about it,

00:52:56   because it's such a better app than anything else that does chat-style interaction with AI.

00:53:03   Oh, yeah.

00:53:04   It's honestly kind of impressive how bad Cloud is.

00:53:07   Cloud Desktop is a disaster on the Mac.

00:53:10   And again, I think it just comes down to caring.

00:53:14   I'm not a programmer.

00:53:15   I have no experience programming.

00:53:17   Well, I have incredibly limited experience programming.

00:53:19   I took one class about C Sharp in high school, and that didn't go anywhere.

00:53:25   I very quickly realized that I was never going to be a programmer.

00:53:28   However, I used Codex and Cloud Code to create, you can't say write, to supervise the creation of-

00:53:39   I think direct, you direct it.

00:53:41   You direct an app.

00:53:41   Okay.

00:53:42   To direct the creation of an app that writes our old data on our server to LTO tape.

00:53:48   And it's a fairly complex process.

00:53:51   There's a lot of moving parts.

00:53:53   It's working with real hardware that has drivers that need to be accessed.

00:53:56   And it took a little bit of troubleshooting and looking over, but it works really well.

00:54:02   We've been using it for months.

00:54:03   It's effectively, as far as I'm concerned, bug-free.

00:54:06   There are bugs somewhere, I'm sure, but not that are user-facing and not that we've had issues with.

00:54:10   We've had no issues with data loss.

00:54:12   Data recovery has been awesome.

00:54:13   It works great.

00:54:14   And it's a native Swift UI app.

00:54:16   And I'm some bozo that doesn't know what he's doing at all.

00:54:21   So, yeah.

00:54:22   And that's the really cool thing about these tools is they're capable of it.

00:54:26   But, boy, I'll tell you, it took a lot of iteration from the initial output where I would say, are you-

00:54:31   That is not what a Mac app should look like.

00:54:34   And me having to be pretty explicit in kind of wireframing out the appearance and how the interaction behavior should be and how shortcuts and hotkeys should work and all of this stuff.

00:54:45   But I did get it there.

00:54:46   It just took time and care, which it seems they're unwilling to give, which is a real bummer.

00:54:51   Because it does make the experience much worse on a Mac.

00:54:54   And there's like a whole thing.

00:54:56   And it's all, to me, evidence that a class of executive, and I think at OpenAI, we've sort of watched it happen almost publicly with the way that their executive shufflings have happened with public announcements.

00:55:14   But that the product-minded people have been replaced by people who just don't think product quality matters.

00:55:21   They don't really think that's-

00:55:23   It's not just that they don't understand it themselves or have a bar, a thermometer of taste themselves where they can taste their own product and say, oh, yeah, that's good enough or that's right or that's wrong or, boy, that sure looks confusing.

00:55:40   It's that they don't even think it matters.

00:55:42   And you can succeed with a confusing product.

00:55:46   There is evidence of it.

00:55:48   That doesn't mean it's the best way to succeed.

00:55:51   And it doesn't mean it's the—I would argue that it is always the wrong thing to do, right?

00:55:57   And it certainly never hurts a product to be well-designed as opposed to poorly designed.

00:56:03   But it's just there's so many signs of it where it's just there's no way that this shipped with product-minded people in charge when they've got, like, at this point, Claude has in their settings, like, plugins, extensions, connectors, and something else that is, like, in the thesaurus as synonyms for those words.

00:56:25   And they all do different things.

00:56:28   And somebody else—I wrote this on Daring Fireball, and somebody wrote to me, hey, you don't even know the half of it because if you're, like, in an IT organization and you're the administrator, you get connectors for something else, and the word means a different thing.

00:56:43   And it's like, these are problems—and it's like, yeah, but look at our valuation.

00:56:49   Number goes up, so obviously it doesn't matter, right?

00:56:51   But it is the sort of thing that when ChatGPT or Claude itself is making these things and nobody who knows better is there to say, no, that's not good enough, they're just like, ship it.

00:57:03   That's the other—on the valuation thing, that's one of the excuses that I've seen from people to say, oh, well, they're building quickly, and they're shipping all the time, and these are still new fledgling companies.

00:57:14   But if they have alleged effective market caps of a trillion dollars, that's their private valuation, that's bigger than Walmart.

00:57:21   That's bigger than AMD.

00:57:22   That's as big as Samsung.

00:57:24   So, oh, we didn't have time to make a native app doesn't really fly with me anymore at that point.

00:57:31   Right, these are not—they are, in some sense, because they're pre-IPO startups, but they're not startups like anything that we've ever seen before.

00:57:39   There's not five people working there.

00:57:41   Yeah.

00:57:41   Right, and they're the ones declaring and accepting these trillion-dollar valuations.

00:57:46   This establishes them as in the range, like a trillion dollars, give or take.

00:57:51   I think that puts them above Berkshire Hathaway, which is bananas.

00:57:55   Yes, I think Berkshire's nine-something.

00:57:56   Yes, which is crazy.

00:57:59   It's, like, arguably the best company that's ever existed.

00:58:04   It is quite literally crazy.

00:58:05   Right.

00:58:06   I would happily—like, my wife explained to me long ago, and it's like, oh, yeah, I see why it's like that.

00:58:14   But in law school, you had to be prepared to argue any side of the issue, whatever.

00:58:20   It's contracts law.

00:58:21   And you come in, you have to be able to argue on the spot, because you might get called on in class, the client side of the contract, the vendor side of the contract, or some other side of the contract, like somebody who was caught in the middle.

00:58:35   And you need to be ready to argue that side of the case.

00:58:38   And if you—we had, like, a debate, like, who's the best company that's ever existed in the 20—from the 20th century onward?

00:58:44   And I got dealt Berkshire Hathaway, I'd be like, oh, I might win this, because I can argue that this might be the best company that's ever existed.

00:58:51   And you've got numbers to back it up, right?

00:58:56   They put them at the beginning of their annual report every year, where they compare their returns against the S&P 500.

00:59:03   Yeah, you could just have put money into Berkshire Hathaway in, like, 1978, and no matter how much you put in, you're pretty rich right now.

00:59:11   Right.

00:59:12   And these companies are more valuable than Berkshire Hathaway, and Berkshire Hathaway hasn't, like, suffered or declined.

00:59:17   Like, they're thriving.

00:59:19   It is kind of bananas.

00:59:21   So don't sing me the, oh, they've got to move fast, they're startups.

00:59:26   It's like, forget about it.

00:59:27   But it is true, that's the other thing that is very evident, is in their list of priorities, moving fast is number one.

00:59:34   And I think to the point where it's, at this point for them, irresponsible.

00:59:40   And I'm not an AI doomer, but I do think it's irresponsible to move fast.

00:59:45   And to me, we can laugh and complain about the sloppiness of this rollout of the Mac ChatGPT codex renaming and rebranding or whatever.

00:59:54   But to me, the red flag is, what else are they doing slapdash?

00:59:58   And there are people who are putting very important decisions in the hands of their LLMs.

01:00:06   And it's, to me, that sort of hair on the back of the neck gets raised.

01:00:11   Like, hey, this, you know, and they're the ones who keep saying, this is transformative of all of society and culture.

01:00:18   And it's as dangerous as nuclear weapons.

01:00:21   And they can't even release a software update in a comprehensible, cohesive faction.

01:00:26   Because they're so hell-bent on moving as fast as they can.

01:00:30   It's very strange.

01:00:31   Yeah.

01:00:31   All right, let me take a break here and thank our good friends at Factor.

01:00:36   Hey, hunger strikes and you're exhausted.

01:00:39   Oh, what are you going to do?

01:00:41   But guess what?

01:00:41   If you use Factor, you've got something healthy in your fridge or your pantry and you can make it right now.

01:00:49   Factor is not a meal kit type thing where you get a whole bunch of ingredients in a box and then you have a recipe card and you put it all together quickly.

01:00:57   Those things are great.

01:00:58   I've used them and I really like them and I think they're really interesting.

01:01:01   But Factor is entire meals that are already prepared, but they are fresh.

01:01:07   So just like meal kits, they come in a box that's got dry ice in it to keep the cold stuff or the refrigerated stuff cold.

01:01:12   It's fresh.

01:01:14   It is fresh food that you need to keep refrigerated.

01:01:17   But then when you're ready to eat it, you can just pop it in the microwave and heat it up and it is all made with really fresh ingredients.

01:01:24   Really, really just heat it up and eat it.

01:01:27   It is fantastic for lunch.

01:01:28   Super duper fantastic for breakfast.

01:01:31   You can have it for dinner.

01:01:32   It is absolutely great, but it's really, really good.

01:01:36   And amongst the things that are really, really good about it, fresh, never frozen.

01:01:41   They have over 100 rotating weekly meals, including globally inspired flavors like Mediterranean, Asian.

01:01:47   So there's always something new to look forward to.

01:01:50   Over 100 a week and including other specialties like kid-friendly stuff.

01:01:55   If you've got kids who do not want to eat quote-unquote weird food or something like that, there's some really kid-friendly stuff that is obviously kid-friendly.

01:02:02   They've got protein-heavy stuff if you're like into maximizing the amount of protein you're intaking.

01:02:08   Whatever – if you have specific dietary needs or vegetarian or anything like that, you can definitely find it.

01:02:14   They have plus selections every week, but it's not overwhelming because they're broken down into obvious categories that you can pick from.

01:02:21   So within that category, there's not too much.

01:02:24   You're definitely not overwhelmed by choice.

01:02:26   But you're also never – every time I've ordered from them, I've never felt like, ugh, that's not enough to choose from.

01:02:33   It's a really, really good mix of choices, always fresh ingredients, and it really does come on time, and it's super easy to eat, and I really do like it.

01:02:42   I especially love the breakfast stuff with the eggs because that's when I am at my absolute laziest.

01:02:47   And I'll just admit it, I love having a little tiny steak for lunch, like a filet or mignon or something.

01:02:53   I feel like I feel like a king because ordinarily I never, ever eat something like a steak for lunch.

01:02:57   Really good.

01:02:58   Where do you go to find out more?

01:03:00   Go to factormeals.com slash talkshow50off.

01:03:07   That's factormeals.com slash talkshow50off.

01:03:14   And guess what?

01:03:15   You can guess it from the code talkshow50off.

01:03:18   That gets you 50% off and free daily greens per box with a new subscription only while supplies last until September 27, 2026.

01:03:29   See website for more details.

01:03:31   Do not miss those extra details.

01:03:33   See their website for more details at factormeals.com slash talkshow.

01:03:37   50 off.

01:03:38   All right, parlaying from ChatGPT and the Mac, how about Siri AI?

01:03:42   Have you been running iOS 27 or macOS 27 anywhere?

01:03:46   I have been, yeah, everywhere pretty much.

01:03:49   I'm running it on my iPad, on my iPhone, on one of the Macs that I'm using, not my main production Mac, but my laptop, it's on there.

01:03:56   And yeah, I've been impressed.

01:03:59   It's really good.

01:04:01   I actually ran into one.

01:04:03   And, you know, you asked that question of notion earlier in the show, like, what was it that you described the Trump phone as?

01:04:09   And I, those are the sort of things that have just been blowing me away.

01:04:12   Oh, this does, this is amazing.

01:04:15   I've mentioned, I think on a previous episode, like a door code for a doctor, you've got to punch in a certain door code.

01:04:21   I don't memorize it.

01:04:22   What's the door code for Dr. What's-his-name?

01:04:25   And then it just spins for a bit and it's like, one, two, three, four.

01:04:28   And it's like, damn.

01:04:30   And I've done it with a couple of places with door codes where some, I've tried it both ways, where I've had it in an Apple note and I've had it in the contact, in the contacts app or Cardhop, which is what I use.

01:04:41   But Cardhop shares the system contacts database.

01:04:46   You can put it in like something in the notes field for a contact or something that's in a note in Apple notes.

01:04:53   You just ask these questions or the actual questions about like recent iMessages or mail messages.

01:04:59   It just works.

01:05:01   But I ran into one the other day that it didn't work and it was actually like, I've gotten so used to it working that it didn't.

01:05:09   My son needed his social security number and I knew that I'd, and I had shared it with the, it's a shared Apple note with my wife and son with all of our social security numbers and our driver's license cards.

01:05:23   And I said, oh, what's Jonas's social security number?

01:05:25   And it spun and it said, I don't have that information or something like that.

01:05:29   I was like, hmm.

01:05:30   Oh, I was like, oh, maybe I locked that one or something because it's private.

01:05:33   But then I realized you can't lock a shared note and it wasn't locked.

01:05:37   I don't, I don't worry about us having a shared note with our Apple, our driver's license numbers and social security numbers.

01:05:43   They're not, social security numbers aren't that secret.

01:05:45   It really is just a shared note and Siri AI failed to find it.

01:05:50   And then I think I could look it up, but it's like, I asked it, like, are you sure it's not an Apple notes?

01:05:55   And it's like, I don't see it.

01:05:56   And I was like, what about an Apple note that has SSN in the title?

01:05:59   And it was, then it found it.

01:06:01   It did.

01:06:01   It actually was like coaxing a little kid.

01:06:04   What's, what's three times seven, 20.

01:06:07   No, no.

01:06:08   Think about it.

01:06:09   It's gotta be an odd number, right?

01:06:11   Cause what's seven times two, 14.

01:06:13   Well then what's seven times three.

01:06:15   Oh, 21, like you can kind of coax it, but I was actually, I'd gotten so used to it finding

01:06:21   little things like that, that when it got that one wrong, I was, I was actually surprised.

01:06:26   Yeah.

01:06:27   Me too.

01:06:27   I've had a few small instances like that.

01:06:30   We have a calendar that's an iCloud shared calendar called family.

01:06:34   And that's where we put all our kind of family stuff.

01:06:37   Yeah.

01:06:37   And for whatever reason, sometimes it doesn't properly identify the calendar.

01:06:43   The main thing that's more concerning though, is that it doesn't say, Hey, I don't know what

01:06:47   calendar you're talking about.

01:06:48   Or can you say that again?

01:06:49   I've, I've had instances where I've got event information.

01:06:53   And this is one of my favorite things about Siri, by the way.

01:06:54   And it's, it's going to be a huge thing that people will have to reframe if you're not running

01:06:58   the beta because it is a revelation once you do it, but it's so counter to how input has

01:07:03   always been based on iOS, but because it has on-screen context.

01:07:07   And I think Apple in some ways did themselves a disservice by pseudo kind of launching Apple

01:07:12   intelligence a couple of years ago.

01:07:14   Everybody knows if you take a screenshot, you can have the kind of ask or, you know, kind

01:07:19   of like Siri based prompts that have always been able to come up and they're still there.

01:07:23   But what you can do is just ask about stuff on screen.

01:07:25   So you can have, and I've had this happen before.

01:07:28   Where we'll get a text message or I'll be on a website for, I did it last night with

01:07:32   a concert.

01:07:32   I was on a website, there was a concert, they didn't have an add to calendar button.

01:07:36   And so I just held the side button on my phone and Siri popped up and I said, add this to

01:07:41   the family calendar.

01:07:41   And the first time it said, great, add it.

01:07:44   And it showed all the correct information and whatever, but it didn't look like the card

01:07:49   that I've normally seen where it kind of appears like a little calendar card.

01:07:52   So I go, oh, it was probably there.

01:07:54   And then I noticed on the little family calendar we have running in our kitchen, it wasn't there

01:07:59   five, 10 minutes later.

01:08:00   And it's always pretty quick to appear.

01:08:01   So I thought, huh, I go back in there and I try to find the conversation in the Siri app,

01:08:07   which it doesn't tend to track.

01:08:09   I've also noticed this.

01:08:10   It doesn't tend to log stuff that is action-based.

01:08:13   And so if you ask, for example, what's the pin code to this door, that doesn't appear

01:08:18   in the Siri app.

01:08:18   It's mostly stuff that's chat-based, which on one hand makes sense because you don't want

01:08:24   your Siri app to be filled with like, hey, dingus, turn off the lights.

01:08:27   Right.

01:08:27   Right.

01:08:28   But on the flip side, it is nice to kind of go back and say, okay, what happened here?

01:08:33   Why didn't that work?

01:08:34   So I tried again and said, hey, add this to the family calendar.

01:08:37   And it gave me, rather than a, hey, that worked out great, it just spun indefinitely.

01:08:43   And then it said, can't do it.

01:08:45   So I try one more time and it says, yep, I added it.

01:08:48   No problem.

01:08:48   It doesn't appear.

01:08:49   So then I tried to rephrase it.

01:08:51   And I said, add this to the calendar family, because that's technically what the calendar

01:08:55   is called.

01:08:56   It's not a family calendar.

01:08:57   It's a calendar called family.

01:08:58   And sure enough, worked perfectly.

01:09:01   And so there's just some little weird stuff because it's an LLM or at least powered by LLM

01:09:06   tech, it very much is desirous to provide some sort of successful or satisfactory output.

01:09:14   And so whereas with the old Siri, it would just say, well, I can't do that.

01:09:17   Ask again from your iPhone.

01:09:19   Search the web, which was frustrating because it never did anything.

01:09:23   Now it'll say, oh, yeah, no problem.

01:09:25   I did that.

01:09:26   And then you look and it didn't actually do that.

01:09:28   And so I think that's, and luckily that's stuff that you can kind of guardrail.

01:09:32   And I should mention that this is not the norm.

01:09:34   It's an anomaly.

01:09:35   It doesn't happen very often, but it's happened a few times where it said, yeah, sure.

01:09:40   Perfect.

01:09:40   Did that.

01:09:41   Another example is because it supports app intents and app intents have been around forever.

01:09:45   It adds stuff fairly dependably to my to-do list, which is my reminders app of choice.

01:09:50   So I can say, add, add these items.

01:09:52   And I've been doing this all the time.

01:09:53   It's amazing.

01:09:53   You have a recipe pulled up on Safari and you say, add all of these ingredients to to-do list.

01:09:58   And by the way, I already have sugar, salt, and pepper.

01:10:00   So don't include those.

01:10:01   And it doesn't.

01:10:01   It's great.

01:10:02   However, I think my microphone misunderstood me saying, add this to my grocery to-do list to add this to my grocery to-do list.

01:10:11   And it went into reminders and goes, oh, he doesn't have a list called groceries, so I better start one.

01:10:16   And so it created a new list called groceries and put all this crap in there.

01:10:20   And I went to the store and opened up my to-do list app, and I'm like, where the heck is everything?

01:10:24   And then I realized, oh, it's in reminders in a new list that it created because I didn't see one there.

01:10:28   And so that's one of those things where I understand why I did it.

01:10:32   But on the flip side, if I don't ever use reminders and every single time I tell you to do something with a reminder involving to-do list, you probably ought to just assume I mean to-do list.

01:10:40   But I think that stuff will massage itself out over time.

01:10:43   Yeah, because it's close enough.

01:10:46   You can see how that ought to be technically possible already, right?

01:10:49   Oh, totally.

01:10:50   Sure.

01:10:51   That you can kind of, they can just kind of push it there from where they already are, right?

01:10:57   And conversely, I see why it messed up.

01:10:59   It's a totally reasonable mistake.

01:11:01   I probably didn't enunciate well, but it's one of those things where when it does the thing that it's wrong or it's just to kind of crap your experience, and that's the risk I think they have.

01:11:11   And the good news is it's way better than old Siri.

01:11:13   But I ran a poll amongst my own audience saying, hey, how often do you use Siri?

01:11:17   And the vast majority of people said never.

01:11:19   I don't ever use it for anything because it doesn't do anything right.

01:11:22   I use Siri often, but it's because I know exactly what it can and can't do.

01:11:28   And that's been learned behavior over years of trying stuff and going, okay, I can't do that, so I'm not going to try that again.

01:11:34   And I think retraining people to actually give it a shot is going to be hard for Apple to do.

01:11:40   And then on the flip side, I also think that they have to yield proper results so that people don't go, I tell it to do this, and it never really does it quite right.

01:11:47   So I'm just going to stop trying altogether.

01:11:49   That's the risk they run, I think.

01:11:50   It's such an interesting point, Quinn, because now that they're calling the new version, quote, unquote, Siri AI, it makes sense as the name.

01:11:58   And it's kind of a good, there's some cleverness built in where not calling it new Siri is good because eventually it's not going to be new, right?

01:12:10   This is, I don't know, at some point three years from now, it's just Siri.

01:12:15   And Siri AI, but there is a word to use for, hey, if you're still on iOS 26, you've got old Siri, which is just called Siri.

01:12:25   And if you're on iOS 27, you can opt into Siri AI, which is the new thing.

01:12:30   But I've been thinking ever since, certainly last year, but even before they announced Apple Intelligence two years ago at WWDC, I had been wondering whether they had burned the Siri brand to the ground.

01:12:45   Because so many people that I know in my personal life have just, they just, you say Siri, they hear a stupid thing that doesn't work, you know?

01:12:56   And it's to the point where some of the things that Apple had gotten right or much improved, like I think they're, it varies.

01:13:08   And some of the people out there listening are going to be like, are you out of your mind, John?

01:13:11   It always gets it wrong.

01:13:13   But I use, I've been using the dictate my voice to text and using the Siri, they branded under Siri, the little microphone button and messages for years now.

01:13:26   And yes, some, there are certain things that it's like, oh God, why did you capitalize that word or something?

01:13:31   But it got a lot better than it was when they first announced the feature.

01:13:36   When the feature first came out, it was really unusable.

01:13:39   I mean, and I, I speak well enough and dictate well enough that I have a semi-popular podcast, right?

01:13:46   My voice is pretty standard white American male, right?

01:13:53   I mean, it is, I'm not, I don't have an accent or at least much of one.

01:13:57   I should not have had any problems with it.

01:14:00   It was unusable for a while.

01:14:01   And I know a lot of people who like tried using it and they're like, well, screw it.

01:14:04   I'm never going to try this again.

01:14:05   And it got better that they should have tried it, but I can't blame them.

01:14:09   There is a sort of first impression.

01:14:11   That's why first there's so many adages about you never get a second chance to make a, well, I guess it's one adage, but it's so often cited that you only get one chance to make a first impression.

01:14:21   It goes back to Steve Jobs' obsession with product packaging in an industry where product packaging is notoriously bad because the people's first impression with a product is when they open the package, right?

01:14:34   And you never get a second chance at that, right?

01:14:36   And if you already, if the packaging is bad, if it's one of those hard to open bubble boxes that you kind of have to almost cut yourself to open, you've put a bad seed in the person's head and they haven't even used your thing yet.

01:14:51   Now you're starting from a hole in the ground to try to make a good impression.

01:14:55   And if you shipped a bad product for years, well, Siri overall has been out since 2010.

01:15:01   And most of that time, certainly the last 10 years, it has been something that people think is terrible.

01:15:07   So I think, I honestly think with no hyperbole that getting people to actually try the new Siri AI is the single biggest product marketing challenge Apple has had in a very long time, maybe ever.

01:15:22   And I'm someone who understands the changes that have come to Siri.

01:15:27   I'm excited about them and even I will forget to ask Siri to do things because my brain, after years of training, just goes, I can't do that.

01:15:34   So I start doing something manually only to realize, wait a minute, I do it all the time.

01:15:39   Or I'll use, I'll go to ChatGPT because it's what I've been using for the last two years because of course, why would I ask?

01:15:44   Siri can't do that, but Siri can and it actually yields pretty excellent results.

01:15:48   And I think that's not just a hurdle because of all of the tech debt and the negative experience that people have had.

01:15:56   But I think the other problem, in my opinion, is it's kind of 3D touch coded.

01:16:00   3D touch was an amazing technology.

01:16:03   And all of the ways that 3D touch enabled you to use the iPhone more quickly and at a greater level of depth were, in my opinion, super impressive.

01:16:14   But it was one of those things where unless you knew something could be 3D touched, you wouldn't try to do it.

01:16:20   Because, and there was no visual context.

01:16:22   There were no clues that told you, hey, this button, you can push this a little physically harder and something will happen.

01:16:27   Now, haptic touch with being long hold, it's a learned enough behavior that I think now a lot of times you can prove.

01:16:33   But still, there are times where I'll hold something accidentally and go, oh, I didn't know this had a submenu there because I've never thought to try it.

01:16:40   And I think Siri is the same way where many of these new features are things that there's no visual prompt on screen that tells you, hey, did you know that these are ingredients and you can ask me to add these to your grocery list?

01:16:50   So people won't even try it because they don't even think it's possible.

01:16:53   And that's where I think so much of what this new Siri could do isn't well documented.

01:16:59   And maybe by the time they ship iOS 27, they'll have you go through some sort of walkthrough that tells you kind of what it can do.

01:17:06   But they can only give two or three examples as part of this walkthrough.

01:17:09   And the capability set in the new Siri is so broad and so wide and so far reaching.

01:17:15   But if there's no way to know or to be taught that that's how a thing works, I can imagine a lot of people just straight up won't use it because they won't even think to ask.

01:17:25   Yeah.

01:17:26   And that's a bummer.

01:17:28   Yeah.

01:17:29   And so I really do think they need – I don't think they can solve it with a first run experience.

01:17:34   Yeah.

01:17:35   They should put something in there.

01:17:36   But I don't think it's enough.

01:17:37   I think they really do need some kind of ad campaign.

01:17:41   And it's sort of – I didn't mention it, but it is sort of why I had a couple of linked items in the last few weeks where I was linking to brands that have had sort of, hey, we used to suck and now we don't suck anymore ad campaigns.

01:17:56   Like Domino's Pizza had one like 10 years ago.

01:17:59   And after I wrote about it, a bunch of people were like, you know, every once in a while, every time I go to my mother-in-law, she always gets Domino's.

01:18:04   And I always kind of dread it because it was so freaking terrible.

01:18:07   And you know what?

01:18:08   They actually – she still gets it, and it actually is better now.

01:18:11   It actually is pretty decent.

01:18:12   And I think it was Carlsberg Beer or something where their slogan for like 100 years was like the world's best beer.

01:18:18   And they came out with a slogan that said, not the world's best beer.

01:18:21   And they came out and said, you know what?

01:18:23   We kind of let our ingredients slip.

01:18:25   And I don't know.

01:18:26   And it's certainly not recent Apple style to run self-deprecating ads, so it would be a challenge for JAWS and team.

01:18:35   But there's a way – there's certainly got to be a way to do it, right?

01:18:39   And probably the best – one of the best ad campaigns any company has ever run, and it speaks to how much Apple liked it and how successful it was, was the – it's actually called Get a Mac, but everybody remembers it as I'm a Mac, you're a PC.

01:18:56   But it was the Get a Mac campaign with John Hodgman and Justin Long.

01:19:00   And it was – it was self – it was deprecating PCs and Windows, so that's a different challenge than a sort of, hey, we know Siri used to suck, but Siri is actually really awesome now.

01:19:12   But it's the fact that those ads were legit funny that really made it stick.

01:19:20   And they could have – there's an opportunity for fun here to say, yes, Siri used to be stupid, and now it's really useful and smart, and you should try it again.

01:19:30   And it's hard – you know, the other challenge is they tend to – for obvious reasons, they tend to make ads for the phones, not for iOS in general, right?

01:19:40   They're not selling Siri.

01:19:42   If you have a recent iPhone, you get the new Siri when you get iOS 27.

01:19:48   But I don't know.

01:19:50   I think that they should just run some Siri ads that are really like, hey, you should upgrade to the new iOS, or just say get a new iPhone.

01:19:57   Say that that's how – the easiest way to get it is to get a new iPhone and then leave it to everybody else to figure out that you just have it on your phone if you have iOS 27.

01:20:05   But just run ad campaigns in major sporting events and popular TV shows to tell people, seriously, try it.

01:20:13   It is actually good now.

01:20:14   We mean it.

01:20:14   And, yes, we can laugh at the old Siri now because we finally fixed it.

01:20:19   They're never going to crack a joke at old Siri before they fixed it, but now they fixed it.

01:20:24   And however good you end up thinking out there that the new Siri AI is, it is not a joke.

01:20:30   Yeah.

01:20:31   No, I think – and I think that's part of the reason why the new Siri app exists, why it is an app, why it will be on the home screen is to signal, hey, this isn't just your average kind of thing.

01:20:42   But what I do think is a little lackluster is the app itself, respectively, a card tile and or list view of your past conversations.

01:20:50   But I think it should be better at helping you prompt what you want or giving you ideas of what to search for.

01:20:57   The photos app has done a fairly decent job at this in the last couple of years where it kind of says, hey, you can ask questions or say, hey, generate a video of my pictures from our San Francisco trip or whatever.

01:21:09   But that is still siloed inside of the photos app, which I've always found a bit confusing, still inside of messages where I think Siri AI is the most powerful because it has, again, contextual knowledge of your entire conversation history in iMessage, years worth of iCloud synced messages.

01:21:26   You go to the search field and it's still just doing a dumb bog standard keyword search and there's nothing that suggests to you, hey, you can ask Siri how to do this or this is a text field.

01:21:37   Why don't you enter a complete sentence and we can run a search query using an LLM rather than a basic keyword index lookup?

01:21:45   Because that's what I think people are doing anyway, what they've historically done.

01:21:50   As you well know, very few people actually know how to search the web.

01:21:54   Everybody goes, dear Google, I'm looking for a blah, blah, blah.

01:21:58   And Google search has actually been made worse by trying to accommodate sentence or conversation input rather than kind of keyword search for people that actually know how to search the web.

01:22:09   And it's gotten even worse, in my opinion, given the advent of LLMs where people just go to ChatGPT and say, my tummy feels funny.

01:22:16   This is what it was.

01:22:17   And so let the system inhabit that same kind of thing.

01:22:21   When you pull down to Search Spotlight, when you click the search field in Messages, just run a rich LLM-based search and make that part of Siri AI and Apple Intelligence because that will help encourage people to kind of try to use it, even if it's not by holding the side button and saying, hey, Siri, do this thing.

01:22:39   I think it still seems kind of siloed inside of the Siri AI or UI, UX rather, where it need not be and can apply system-wide.

01:22:48   Yeah.

01:22:49   And it's funny because I can't speak to it.

01:22:52   Maybe you can.

01:22:53   But I am testing it on – actually, I could probably test it on my phone.

01:22:58   I'm running iOS 27 on my main phone, have been since beta 2, and no looking back.

01:23:03   I mean, there's a couple of issues I've run into.

01:23:05   Beta 3 seemed pretty hard on the battery, but beta 4 seems to have returned that.

01:23:10   And, you know, that's what you –

01:23:12   It's beta.

01:23:12   It would be shocking if I went through the whole summer without one of the betas being, oh, yeah, we kind of botched the battery consumption.

01:23:19   Something spiraled out of control on that one.

01:23:22   But it's really been fine.

01:23:24   So anybody out there who's at least on the public beta, it's probably – it's got to be better.

01:23:29   And I – at this point, unless something horrible happens in August, I would say jump on – you could jump on iOS 27.0 when it comes out in September, and you're probably going to have a very good experience.

01:23:42   Mail got – and they mentioned it in WWDC – Mail got, hey, search actually works now.

01:23:48   And they kind of said that in so many words in the keynote.

01:23:51   We kind of know searching for email and mail, Apple Mail, has sucked forever, like, bizarrely.

01:23:58   And now it doesn't.

01:23:59   And I've played around with it.

01:24:01   But the whole preface was that on my Mac where I've got Mac OS 27 installed, I don't even have all of my email.

01:24:06   I only have my iCloud email, which isn't my main one.

01:24:09   But on the phone, it does seem better.

01:24:12   But I asked Apple at WWDC, hey, you guys said – you've got Siri connected to the index of everything, the semantic index.

01:24:21   And you said you've fixed search and mail.

01:24:24   But what about messages?

01:24:25   And it was just one of those awkward pause.

01:24:28   And like, yeah, we might – you know, that'll be a thing to look at in the future.

01:24:32   Which is their way of saying, no, search in the messages app isn't different at all.

01:24:38   And again, it is – I don't know which one's worse, search and mail or search and messages.

01:24:43   If you're searching for a message from the last week, it's fine.

01:24:47   If you – if I'm searching for the last time, Quinn, you and I, you had you on the talk show, which might be like a year ago, forget about it.

01:24:56   It might come up.

01:24:56   It might not.

01:24:57   I don't know.

01:24:57   But if you ask Siri now, it'll find it.

01:25:00   It's there.

01:25:00   There is code on the device.

01:25:03   That's what I think your rant was about.

01:25:06   You can do this now, right now, on the device, running the beta – summer betas of these 27 OSs.

01:25:13   But if you just search in messages, which is what people do, you don't even get a hint that, hey – like a button, hey, do you want to run this same search and ask Siri to find them?

01:25:22   Yeah.

01:25:23   Which is kind of stupid.

01:25:24   Or just run it automatically and just –

01:25:25   Right.

01:25:25   Yeah.

01:25:26   Why do that at all?

01:25:27   And that's the other thing that – I mean, you can do – and there are multimodal searches where you can still do a string search.

01:25:33   If I search the word door code, that's going to be instant.

01:25:35   It's going to be fast.

01:25:36   It's going to be used.

01:25:37   No battery.

01:25:38   But recognize context.

01:25:39   If there's no results or if the user starts to enter code to door or trying to kind of play around with the word salad to get stuff through, then just run it as an extended search or prompt the user.

01:25:52   Or if the user goes in and types a question, which, again, so many people do.

01:25:56   I watched – I love my wife, and she's amazing and a genius and way smarter than me.

01:26:02   But I saw her type – this was years ago, and I teased her about it, and now she's – we've learned.

01:26:08   I've improved much more as a person ever since having met her than vice versa.

01:26:12   But the first time I saw her ever do a Google search, I was like, what are you doing?

01:26:17   Because she said, what will be the weather tomorrow in San Francisco, and should I bring a coat?

01:26:21   I'm like, what are you expecting the output to be?

01:26:25   Just search weather SF.

01:26:27   There you go.

01:26:27   And like, that's all you need to do.

01:26:30   It's going to yield you the same thing, and it's not a person.

01:26:33   It's not conversational.

01:26:34   But now, LLMs are.

01:26:36   And so every search field that isn't conversational in return feels weird and feels dated and feels bad.

01:26:42   And I am excited to see the changes in photos and in mail.

01:26:49   I'm excited to see how much further that goes because I still feel like typing provides you lesser, worse results than asking.

01:26:57   And I don't feel that those two should be materially different.

01:27:01   When I go into Spark, which is my email client that I use, and I'm not singing Spark's praises.

01:27:06   It is an electron pile of garbo that uses so much memory, it's unbelievable.

01:27:10   But they do have a function called Spark AI.

01:27:13   And when it shipped, it was pretty basic kind of garbage LLM shenanigans.

01:27:21   But now they've done really, really great indexing.

01:27:23   And I am shocked where I'm actually going into my email client and saying, hey, what was the promo code from that sponsor?

01:27:29   And it finds the answer almost immediately because it's not searching through my entire thing.

01:27:34   It's already pre-indexed, very a la Spotlight.

01:27:37   And it links me directly with evidence to the email thread so that I can double check to make sure it didn't hallucinate some garbage, which it's become increasingly less prone to do.

01:27:48   And it's made it to the point where that's my default entry and input.

01:27:52   And I think that in this era of LLM and AI technology, just make anywhere where you can interact at all, just run that.

01:28:02   How are you going to do it?

01:28:02   And I understand they have to accommodate battery limitations and certain devices will not be able to do it as extensively and or as capably as other devices.

01:28:10   And you don't want to have to go out to private cloud computer if you don't have to and do it on device.

01:28:14   And that stuff, as memory increases and gets better, that will become better.

01:28:18   But that's one of those areas where I'm like, I will say, have you noticed, I guess, are you using an iPhone 17 Pro?

01:28:25   Yep.

01:28:26   Yeah, I'm on the air.

01:28:28   I'm one of the few, I think, that has stuck with it since launch day.

01:28:31   It's my daily driver.

01:28:32   I love the iPhone Air.

01:28:33   When you run Siri for more than two or three messages back and forth, I think you could almost melt the surface of the sun with the back of the iPhone Air.

01:28:43   It is so unbelievably hot.

01:28:45   It is unreal to the point where the screen dims, the phone starts to lag, and it's only in Siri.

01:28:51   And it's only, and I do have all the new fancy pants stuff enabled.

01:28:55   I've got the enhanced voice and all the crap that's limited to the Pro-Skew variants of the phone.

01:29:01   But, yeah, wow.

01:29:02   It's rough.

01:29:03   It's an odd phone.

01:29:05   I still have my review unit.

01:29:08   And I still actually mean to write about it.

01:29:10   And I did, at first, put, like, for Beta 1, I used it on the Air.

01:29:15   And I didn't notice it getting that hot, but I chalked it more up to it being the first developer beta because I hadn't used it side-by-side with it on the iPhone 17 Pro.

01:29:29   But on the 17, and I don't know how, I think that the Air is in an unusual spot where it's got most of the top-tier specs, right?

01:29:40   The photography doesn't, right?

01:29:42   Like, you can't shoot Apple log video on the Air.

01:29:47   And there's other things, too.

01:29:49   You can't do the pro, quote-unquote, well, actual pro video stuff on the Air.

01:29:53   But for the most part, though, you can.

01:29:56   And so it has the new voices for Siri.

01:29:59   So in more ways than not.

01:30:02   It's got the 19 Pro and all the memory.

01:30:04   Yeah.

01:30:04   In more ways than not, and I think all of the knots are related to photography in particular, it is a 17 Pro-class phone.

01:30:13   But it doesn't have 17 Pro-class thermals.

01:30:16   I mean, I think that's basically it.

01:30:19   It is, A, it's so much thinner that it's the entire premise of the device, and B, doesn't have the-

01:30:26   Vapor chamber.

01:30:27   The vapor chamber.

01:30:28   Yeah.

01:30:29   Thank you.

01:30:29   So it's almost like two entirely different things, right?

01:30:34   The space, the physical space itself, and the vapor chamber.

01:30:38   So it's sort of taking it on the chin.

01:30:41   Yeah.

01:30:42   And to be clear, it's only when running the AI stuff within Siri.

01:30:46   Everywhere else, totally fine.

01:30:48   So I don't even know that it's a beta issue in as much as it is it's really making that SoC work when it's really trying to find stuff.

01:30:56   And I know the NPU is going wild.

01:30:58   And so maybe that's why they're not doing it in as many areas of the OS yet is because it would just completely nerf the experience on a bunch of devices.

01:31:07   But if that's not a problem on the 17 Pro, then maybe as we see vapor chambers make their way down the line and they end up in the base iPhone, because they will eventually, then maybe it ends up not being such a big deal in a year or two for them to flip the switch on.

01:31:22   Right.

01:31:22   Within a couple of years, it'll be the base phone that runs whatever the base system is for most people will be much more capable.

01:31:29   But they're pushing the limits on this right now.

01:31:32   And it's also easy to overlook based on the fact, and especially people in your audience and my audience, who know that nothing that Siri AI is doing is anywhere near the cutting edge of just chatbots and putting aside all the Igentic stuff.

01:31:52   But this is any kind of actual structured test that they run to benchmark new models or just an ad hoc tech test that we could run on the show or you could put in a video where, hey, let's just come up with 10 interesting questions to ask various chatbots.

01:32:12   Siri AI is not going to win that contest.

01:32:16   There's just no question about it.

01:32:18   But it's going to do credibly.

01:32:20   It is a, you know, and I admire the attitude that it takes.

01:32:24   It is just very plain spoken.

01:32:27   It's not too dry, but it's not overly cheerful.

01:32:30   It doesn't sprinkle emojis everywhere.

01:32:32   I was a little worried at first when it didn't have a setting for personality at all.

01:32:37   I was like, oh, God, don't let this be too saccharine sweet.

01:32:40   But it's not.

01:32:41   But it's not going to win.

01:32:43   And so it's easy for technically minded people who know that this is not cutting edge LLM frontier technology to just sort of say, well, it's not impressive.

01:32:52   But it really is impressive in terms of how many people out there have never used anything like this and that it's just right there, a side button press away from you asking a question or it evaluating what's on the screen right now at any moment.

01:33:08   And that it will be the baseline, right?

01:33:10   This is the worst version of Siri AI there's ever going to be.

01:33:13   And it's going to run on the least number of devices and have the most adverse effect on the iPhone air in particular or whatever other devices like smaller iPads or whatever that this might be pressing the limits of what it can do.

01:33:29   All of that's going to go away in the near future, right?

01:33:32   And that they have been they've known now they're they're now shipping products where the SOCs are meant to be running local models.

01:33:42   And maybe no, nobody ever has enough RAM and nobody's none of us are going to have enough RAM for years to come at this point.

01:33:50   For real.

01:33:51   But all of this stuff has been happening with SOCs that weren't really meant for local execution for a while now.

01:33:57   And now we're getting to the point where we're getting devices that are.

01:34:01   So it's very exciting.

01:34:02   It is.

01:34:04   It really is.

01:34:04   One of the other reasons that I'm kind of glad they did ultimately relent and decide to make in a way a chat bot.

01:34:11   I mean Siri's not a chat bot, but it is.

01:34:14   You can chat with it and you can ask it questions that it sources information from either its trained index and or even from the web.

01:34:21   And I think it was wise of them to do that, even though I know they resisted for quite a long time, because ultimately that is what people do.

01:34:27   And if you can provide a field for people to type or speak text, it's only a matter of time before they start to ask the device and by consequence Siri to do things that it can do that they would never think to prompt on their own unless they had this conversational kind of established relationship.

01:34:48   And so if they hadn't changed anything else and they just said, yeah, you can search everything's indexed now and you can ask it to provide your photos and do all this stuff, people wouldn't really try it because it's still Siri.

01:34:59   But if it can be LLM powered and act and walk and talk like a chat bot, which it does, then I think a lot of people that have been using the free version of ChatGPT will be compelled to use the version of Siri.

01:35:13   And even me, I understand, like you mentioned, the limitations of the fact that it's not cutting edge LLM tech.

01:35:19   But for stuff that I know basic LLMs can do, I'm going to ask Siri because it's free.

01:35:25   It's in my pocket.

01:35:26   It's ready to go.

01:35:27   I can prompt it via voice and it'll give me the right output.

01:35:31   I don't need the latest version of Claude Fable to know when JFK died.

01:35:37   You know what I mean?

01:35:38   Right, right.

01:35:39   It's just that same kind of thing.

01:35:41   And hopefully that extends beyond kind of basic interaction for most users.

01:35:46   And I think it will.

01:35:48   The more that I use it, the more that I see it, I think people will use it once they discover that it exists and it can do what it can do.

01:35:55   The problem is training users to understand what it's capable of with the preconceived notion that it can't do anything.

01:36:03   So I had a funny one over the last week.

01:36:08   I was at a restaurant with a French menu and I was pretty sure I was going to get it.

01:36:14   And I only asked, I'm not trying to, I'm going to butcher some French very shortly.

01:36:19   So me saying I don't speak French and don't know much about French, I'm going to prove it very shortly.

01:36:23   I was almost certain that a croquet monsieur is a ham and cheese sandwich where it's heated up and melted.

01:36:30   But I was like, well, that seems like a question, people, this put me into reviewer mindset.

01:36:35   So I asked Siri AI, what is croquet monsieur?

01:36:38   And the answer, a croquet monsieur is a classic French hot sandwich made with ham and cheese, typically gruyere or a menthol between slices of bread is often topped with a creamy becamel.

01:36:51   I don't know what that is, but it was good.

01:36:53   Or mornay sauce and baked or fried until golden brown.

01:36:57   Good answer.

01:36:58   Very good answer.

01:37:00   But then here's, I've got two issues here.

01:37:02   First is then it added a second paragraph, which is, oh, and just a quick note, always verify important details.

01:37:08   As an AI, I may make mistakes.

01:37:10   I get it.

01:37:13   And I know why I'm sure it's in the system prompt, but it doesn't do this on every one.

01:37:18   And it's got, it's like a, it's like maybe in Dungeons and Dragons terms, it rolls a D6.

01:37:25   And it's one out of every six times.

01:37:27   It'll just tack that on at the end of your chat.

01:37:30   But this is not a dangerous question, right?

01:37:33   What is a croquet monsieur, right?

01:37:35   It didn't need, I wasn't asking for life advice.

01:37:38   You know what I mean?

01:37:39   My doctor told me to take a statin.

01:37:41   Is that important?

01:37:42   I wasn't, I didn't need the O in just a quick note.

01:37:46   But anyway, on the menu at the restaurant, what they had said, it was made with Comte, C-O-M-T-E cheese, which is not Gruyere or Amantel.

01:37:56   So that, oh, so I just wrote Comte, question mark.

01:38:02   And it turns out in real life, Comte is just, it's a Gruyere de Comte.

01:38:07   So it is a type of Gruyere cheese.

01:38:10   So Siri was right.

01:38:11   It is Gruyere.

01:38:11   They just, the restaurant, because they use this specific kind of Gruyere cheese, that's what they put on the menu.

01:38:17   But all my question was, was just C-O-M-T-E, question mark.

01:38:23   And it tricked Siri AI into completely responding in French.

01:38:28   It just switched from English to French.

01:38:31   Here, I'll put a, I will, I will put a screenshot in and I'll text it to you.

01:38:38   And I thought that was pretty funny.

01:38:40   I thought that was really, really funny.

01:38:43   So it's obviously wrong, but I thought wrong in a charming and fun way.

01:38:47   Pretty charming.

01:38:48   You could, you could probably follow up and say, no, I need that in English.

01:38:52   Yeah, I'm for sure I could, but I didn't want to, I just wanted to leave it as is.

01:38:56   I was like, oh, this is a good one.

01:38:57   And so I still don't know what its answer is because I did not translate it.

01:39:02   I guess I should ask it.

01:39:03   If only you use the translation feature powered by the new Apple Intelligent.

01:39:08   No, I'm just kidding.

01:39:09   What do you think, the last one, what do you think of the new voices for Siri so far?

01:39:13   I think they're good.

01:39:15   They do need some tuning.

01:39:18   There's been, it did get better with Beta 4, I feel.

01:39:22   But sometimes the enunciation is totally wacky where it's said or finished some things where it will say like, you need to go to this door.

01:39:31   And you're like, the what?

01:39:33   This door.

01:39:33   And like, it'll, it'll cut stuff off.

01:39:36   It's very strange.

01:39:37   And it's almost always the last word.

01:39:38   Like it doesn't know how to finish, roll off a sentence or ask a question or whatever.

01:39:42   But in general, they've sounded way more natural.

01:39:46   And every time I use with disdain my HomePod to trigger HomeKit automations, I am reminded how terrible the old Siri voice was that I had just kind of accepted as being Siri.

01:39:59   Not it wasn't even I didn't even think it was that bad.

01:40:01   It was like, no, yeah, computer voice that's been around forever.

01:40:03   But the new ones are so good that, boy, they make that old one sound very robotic and terrible.

01:40:09   Yeah, it's like when retina screens came out.

01:40:12   And at first, you're like, oh, my God, this is amazing.

01:40:13   I can't really see the pixels unless I get them real close.

01:40:15   I can't believe it.

01:40:16   This is really what a leap forward.

01:40:18   This is great.

01:40:19   And it's like, yeah, it's got a great screen.

01:40:21   And then you go after a week or two, you go pick up an iPhone 3G without the retina screen.

01:40:25   And you're like, what is this hot garbage?

01:40:28   These pixels are the size of marbles.

01:40:31   This is absurd.

01:40:32   The language is like that.

01:40:34   But the old Siri had weird mispronunciations, too, for whatever reason, at least for us.

01:40:38   It does.

01:40:38   It drives my wife crazy where the old Siri always pronounces the word turnpike, tornpike.

01:40:45   There's an O in there, tornpike.

01:40:48   And we've got a Pennsylvania turnpike and a New Jersey turnpike in our area.

01:40:53   So we've got turnpikes.

01:40:54   And often those are the ones where we're using directions because it's not just a crosstown errand where we don't even need directions.

01:41:03   But it just drives my wife crazy when Siri says tornpike.

01:41:07   It's like the number one reason I could get her to upgrade is if I could tell her Siri no longer pronounces turnpike, she'll be like, okay, I'll take it.

01:41:15   And you could probably even tell it to change the pronunciation.

01:41:19   That's one of the other things I noticed that was pretty cool is we named – I've got a – my first little girl, I've got a daughter who's 17 months old.

01:41:29   And we named her a very – the name sounds exactly the same as one of my very good friends who's in my phone as a contact.

01:41:36   But the spelling is different.

01:41:37   And so every single time I would talk about my daughter, the name was misspelled because she doesn't have a contact card in my phone because she's a baby.

01:41:45   And what I –

01:41:47   You can solve that problem.

01:41:49   Well, not yet.

01:41:50   Not yet.

01:41:51   What I ended up telling Siri was, hey, when I'm messaging these people, I'm always talking about my daughter.

01:42:00   I'm not talking about my friend unless I'm probably talking with my friend.

01:42:04   And to my surprise, it said, okay, I get it.

01:42:08   And I at first thought, oh, this is just an LLM thing where it's like, yeah, sure, okay.

01:42:12   And then it would never actually do anything with it.

01:42:14   No, it actionably changed the output where almost every single time I did dictation or sent messages or wherever else, it started using the daughter's spelling correctly.

01:42:23   And it somehow, surprisingly, even maintained the correct spelling of my friend when I was talking in a group chat with him and other people.

01:42:33   And so that was one of the things that I did not expect, and it has been literally flawless ever since beta 2 where I told it, hey, stop misspelling this name.

01:42:43   It's pissing me off because I'd have to go through and correct it every single time.

01:42:45   The other thing you can do that I've tried is when you know that the device is likely to spell something wrong, you can just spell it out.

01:42:54   So you can say, oh, Comte, C-O-M-T-E, and it'll spell it out exactly as you dictate it rather than writing out the letters

01:43:02   as you would suspect it to do in both dictation and in Siri.

01:43:05   So that's an area where it's gotten way, way, way better.

01:43:08   One thing I just remembered that it did end up screwing me a little bit, we had planned to go to lunch with family,

01:43:16   and there's a restaurant nearby that opens at a certain time on weekdays.

01:43:23   It opens earlier on weekdays than it does on the weekend.

01:43:25   And so I ended up asking Siri, hey, is this restaurant open right now?

01:43:31   Or it might have been the night before.

01:43:32   What time does it open tomorrow morning?

01:43:33   And it said, oh, this time.

01:43:35   And I just was like, okay.

01:43:36   Well, we all got there.

01:43:37   The restaurant was closed.

01:43:40   And they didn't open for another 45 minutes.

01:43:43   And so I was like, oh, crap.

01:43:44   And it's sort of on you because you're the one who gets it.

01:43:47   Yes, of course it is.

01:43:48   Yeah, and it's by our house.

01:43:50   And what I realized was it wasn't that Siri had just hallucinated or made something up.

01:43:55   It's that it had depended on Yelp to determine when this – and Yelp is never right relative to the Google Maps information.

01:44:03   And so I do think that it is still going to struggle in as much as where is the data being sourced.

01:44:11   I think that it's pretty good, better than most LLMs at not hallucinating or not making stuff up if it doesn't know.

01:44:18   But it is still limited to whatever it determines its data source is going to be.

01:44:23   And I think that Siri's data sources have historically not been very great, and that's still going to be a problem.

01:44:29   It's still going to use Yelp data over Google Maps data, which – yeah, which is a bummer.

01:44:35   You know, I ran into a problem, and it all worked out well.

01:44:38   A couple nights ago, it was really late, and we hadn't eaten, and I said, I'll make mac and cheese.

01:44:45   And my wife was like, that's fine.

01:44:47   And I boiled the – just boxed mac and cheese, just basic bitch shit.

01:44:54   And it turned out, though – and my dad had – we often don't have milk in the house.

01:44:59   My son has a dairy allergy, and I don't drink milk, and my wife doesn't even drink coffee.

01:45:04   I don't take – so we often don't have milk.

01:45:06   And my dad was here for a couple days last week, and he takes milk in his coffee sometimes, but not always.

01:45:14   And I thought I bought milk, but it turned out I didn't because he was like, I'll just drink black coffee.

01:45:18   And I was like, oh, but I thought I had it.

01:45:20   I'd already boiled the pasta, and I needed to go out to get milk, and it was 9.30 at night.

01:45:25   And there's a place that's like right around the – I live right in the city.

01:45:29   I don't have to get in a car.

01:45:30   It's walkable.

01:45:31   But spot A is only a block away.

01:45:35   But I know for a while they were closing at 9 o'clock.

01:45:39   And spot B is more like six blocks away, six quick blocks, but – and in opposite directions.

01:45:46   So I couldn't pass the one on the way to the other.

01:45:49   And I looked – I asked Siri, and it said it was open until midnight.

01:45:54   And I was like, that doesn't sound right.

01:45:56   And I looked in Apple Maps, and it said it's open to midnight every night.

01:46:00   And I thought, oh, it's one block out of the way out gamble.

01:46:03   And I went, and guess what?

01:46:05   They were still open, but they closed at 10.

01:46:08   So I was lucky because it was only 9.30, but they were definitely not open as late as Siri thought.

01:46:16   And it's just because it got the map – they got it from Apple Maps.

01:46:18   And I don't know if Apple Maps gets it from Siri or from Yelp or what.

01:46:22   But, you know, and it wasn't a catastrophe.

01:46:24   But I had that same thought.

01:46:25   You know what?

01:46:26   This isn't really Siri's fault.

01:46:28   It is the data's fault.

01:46:29   You know what I mean?

01:46:30   If I made an Apple note with my son's wrong social security number, and I ask what's my son's social security number, and it gives me the wrong answer that I put in the notes, well, it's not Siri's fault, right?

01:46:42   I mean, how is Siri supposed to know that?

01:46:46   But I do think – I think with Maps in particular, that is going to – it's been – every time I sing the praises of Apple Maps, I get two responses.

01:46:56   One from people who live somewhere, often outside the U.S. or somewhere non-metro in the U.S., where they get shit directions from Apple Maps.

01:47:06   That seems to be happening less and less in the U.S., and it is more of an outside-the-U.S. thing.

01:47:11   I get great turn-by-turn directions here in the greater Philadelphia area from Apple Maps.

01:47:17   But the data about businesses problem, to me, is as bad as Apple Maps has ever been.

01:47:26   It is, to me, one of the least trustworthy sources of data that I ever, ever even look at.

01:47:34   It makes me – I never trust it.

01:47:37   I left the house thinking I'm almost certainly going to have to walk an entire two blocks out of my way, one block to see that it's closed, and then another block back where I'm going to be cursing it.

01:47:47   But I was – and it was wrong.

01:47:49   It just was wrong in a way that I was within half an hour of it closing.

01:47:53   Yeah.

01:47:54   But it's going to – as bad as it's already been that Apple Maps has bad data about businesses.

01:48:00   I forget there was one, two, where it just did not know that a restaurant had closed.

01:48:05   And I knew that it had closed because I had seen the sign on the restaurant.

01:48:08   Yeah.

01:48:08   And it's like, why are you telling me I can go there?

01:48:10   I know that they closed like two months ago.

01:48:11   It's been out of business for three months ago.

01:48:12   Yeah.

01:48:13   Yeah.

01:48:13   Yeah.

01:48:14   And it's like you don't get that from Google Maps.

01:48:16   And I feel like they've fixed their actual map data.

01:48:20   They have excellent, beautiful maps now in most places.

01:48:23   I think their directions have gotten good enough.

01:48:25   But the data about businesses, it's already a problem for the way people were using Apple Maps.

01:48:32   But they're going to use this information so much more with Siri AI.

01:48:35   And it's where Siri AI is going to go.

01:48:38   And I don't think they're going to support – hey, by the way, Siri, the way you told it how to pronounce your daughter's name, you're not going to be able to say, get your business information from Google.

01:48:49   So actually – I mean, it didn't do that.

01:48:52   But what I figured – let me just try this again.

01:48:54   So I searched the same restaurant and said, what time does the restaurant open on Saturday?

01:48:58   And it says, it opens at 10 o'clock.

01:48:59   That's wrong.

01:49:00   It opens at 11.

01:49:01   So then I responded and said, that is incorrect.

01:49:04   Please identify the correct time.

01:49:06   I didn't tell it what to do.

01:49:07   And then it said, searching the web, which it didn't do before.

01:49:10   And then it came back with the right time.

01:49:12   So it probably went out to the restaurant's website and found the right time or whatever.

01:49:16   But it was only after I said, no, that's wrong.

01:49:19   You've got to try again.

01:49:20   And so, again, I think probably for speed because they've got access.

01:49:24   It's built into Apple Maps.

01:49:25   That's what they pulled by default.

01:49:26   But that's not helpful if that's wrong.

01:49:28   And so that's the kind of stuff that's going to – it's not that big of a deal.

01:49:32   But it happens more than once or twice.

01:49:35   And people go, you know, screw Siri.

01:49:36   I'm not using this anymore.

01:49:37   So that's the best thing.

01:49:38   You know, yeah.

01:49:40   And that scenario you painted where it's a real-life scenario.

01:49:43   You've got other people.

01:49:44   You're physically – my face got flushed hearing you tell the story because I've been there where it's my fault.

01:49:49   And now you've got 45 minutes to sit around with your hands in your pockets.

01:49:53   And everybody's like, oh, we'll just wait and whatever.

01:49:55   But it's – that is the sort of real-life repercussion.

01:50:01   Nobody got hurt, but that's the sort of thing that will turn people away from using it again.

01:50:06   It really is.

01:50:07   So they kind of got to be careful.

01:50:09   All right.

01:50:09   Let me take a break here and thank our third sponsor.

01:50:11   It is our good friends at Squarespace.

01:50:13   Squarespace is the all-in-one platform for building a website to help you stand out and succeed online.

01:50:22   Whether you are just starting out or scaling your business, Squarespace gives you everything you need to claim your domain,

01:50:28   showcase your offerings with a professional-looking website, grow your brand, and get paid all in one place.

01:50:37   I just heard – and I know this sounds like I'm making it up, but I can't reveal the person.

01:50:42   But a reader of the website mentioned that I often say friends come to you to ask for help with a website because you're the nerd in their friend circle.

01:50:54   And rather than building it for them or something complicated, just send them to Squarespace.

01:50:59   Somebody who listens to the show said, I actually did that.

01:51:03   It actually worked out great.

01:51:04   The person, the friend of theirs who does it, has no idea who I am, doesn't know what the talk show is, but built a brand-new website on the Squarespace platform.

01:51:14   Loves it.

01:51:15   Couldn't be happier.

01:51:17   Can't believe that something like Squarespace existed.

01:51:20   They didn't even know it was possible.

01:51:22   So you out there listening to me tell you about Squarespace – I talk about Squarespace all the time because they sponsor the show all the time.

01:51:30   But there are so many people out there who do not know that Squarespace existed.

01:51:35   And you could be the one who tells them that they do and that it solves the exact problem they have, which is that they need a website or a new website.

01:51:46   New features or existing features that have gotten better.

01:51:49   Squarespace domains, everything, register domain, find an available domain if your first choice isn't available.

01:51:56   They've got that built in.

01:51:58   Online stores for selling goods, they've got that built in.

01:52:02   Analytics, so you can see who's coming to your website, where they're coming from, it's built in.

01:52:06   And it has a really good, comprehensible interface.

01:52:09   Just one of the best analytics interfaces I've ever seen.

01:52:13   And even stuff that's not really exactly a website but is tangential because it's online, like an email campaign, like sending out newsletters, Squarespace has that too.

01:52:23   It's all there.

01:52:24   I tell everybody on the show, try Squarespace if you need a website.

01:52:29   But it's you guys who listen to the show who can tell other people the reach is so much vaster through the audience.

01:52:36   Like when people come to you as their nerdy friend, send them to Squarespace.

01:52:41   And if you want to help out this show, give them the URL, the code TALKSHOW.

01:52:46   Send them to squarespace.com slash TALKSHOW and tell them to remember that code TALKSHOW.

01:52:52   And when they do sign up after a 30-day, no features held back, free trial, that code TALKSHOW gets them 10% off whatever they purchase the first time, including up to a year of service.

01:53:04   So my thanks to Squarespace for sponsoring this show.

01:53:07   All right, Quinn, last subject.

01:53:10   You had a post recently about Mac Gaming and the Game Porting Toolkit and stuff like that.

01:53:17   Did you hear ATP where John Syracuse said?

01:53:21   I did, yeah.

01:53:22   It's pretty much – I've never heard John Syracuse – honest to God, I've been friends with the guy for 20-some years.

01:53:29   I have never once – and I've listened to collectively – I was going to say hundreds, but it really might be thousands, a thousand hours of podcasts with him.

01:53:40   500 ATP episodes, two hours each, maybe.

01:53:45   I have never heard him say, this person's argument is right and I agree with it.

01:53:50   And I don't have any corrections or notes to add.

01:53:53   I'm so pleased to have heard it myself, too.

01:53:56   When he started talking about it, I thought, oh, no.

01:53:58   And then he said, Quinn is like my spirit animal on this.

01:54:01   He goes, I really care about Mac Gaming and I know nobody else does, but Quinn does, too.

01:54:05   And I'm like, yes, us, too.

01:54:07   Yeah.

01:54:09   But I think you're both wrong, in a way.

01:54:12   Oh, man.

01:54:15   I need John to back me up.

01:54:16   All right.

01:54:17   Well, I'm afraid of him.

01:54:18   What's your beef?

01:54:19   What's your beef?

01:54:20   Well, it's not the whole thing.

01:54:23   I think that your points and calling out how Apple is doing really – I mean, you can jump in and take it from here.

01:54:32   That the Mac Gaming Toolkit is a way to port PC games to run on the Mac.

01:54:39   And it's always been sort of technically impressive, but often practically ineffective because it results in three- or four-year-old games coming to the Mac in semi-playable shape.

01:54:53   And it's gotten better.

01:54:55   Each generation, when Apple bumps the major version number, it is major improvements in frame rate and quality.

01:55:02   And it's gotten really kind of sleeper hit good.

01:55:07   Is that a fair summary?

01:55:09   No, this year in particular, it's massively improved over version 3, which was improved from 2 and 1.

01:55:16   Yeah.

01:55:17   No, it's remarkable how much the performance uplift has come around this year.

01:55:21   Games that were unplayable last year where you were getting maybe 5 to 10 frames per second will now run at the same resolution, 40 to 60 FPS, no problem.

01:55:30   It's not every game, but it's a lot of games.

01:55:34   And I've known for years that there's at least some reasonable team of passionate, talented people at Apple who really do want to see this happen.

01:55:45   And that you're singing the praises of it is existence proof.

01:55:50   They're doing something.

01:55:50   But I don't even know if the person who initially reached out is even still there.

01:55:54   But I got like a private email from somebody at Apple and, you know, nothing, just the sort of conversation they can't officially have with me if Apple knew.

01:56:03   But just basically saying, hey, it was a couple of years ago, but I was talking about the fact that my son had gotten a gaming PC is now on like gaming PC three, I think, or at least two.

01:56:14   But somebody said that, like, in a way that things I post on Daring Fireball sometimes catch the attention of people higher up within Apple, it was caught that, hey, even Gruber's son has a gaming PC.

01:56:28   And I didn't even think, and again, I'm often surprised at what catches the attention.

01:56:32   I didn't say it.

01:56:33   I never even posted it.

01:56:34   It wasn't like, oh, I have to hold my nose and buy my son a gaming PC.

01:56:38   I don't want a crappy PC in my house.

01:56:40   I didn't know what I was doing.

01:56:41   I wasn't as much help helping him buy what to get as I usually am when he buys Apple products.

01:56:48   But I totally understood it, and I was of the perspective of, oh, when I was his age, I would have wanted a gaming PC, too.

01:56:55   There's no question.

01:56:56   I loved video games at the time.

01:56:58   So I didn't.

01:56:59   I was like, what kind of an idiot would think it's odd that a teenage boy in 2018 or 2017, whatever year it was, would want a gaming PC?

01:57:08   I mean, I think you have a hard time finding a high schooler in 2017 or today who doesn't want a gaming PC of some sort.

01:57:16   So I know that that's a real thing, but I do think they're making progress, right?

01:57:20   I mean, it's...

01:57:21   Yeah.

01:57:22   But I don't think people...

01:57:24   I don't think it's registering amongst the...

01:57:27   It's like the Siri problem, right?

01:57:29   Yeah.

01:57:29   Well, then I think it's because if you talk to Apple officially on the record, it's not for end consumers.

01:57:35   The idea is that it's designed for video game developers to take their existing titles and port them over to the Mac and or to take titles that are in development

01:57:43   and to do all the translations such that they can run on the Mac and it's low amount of work.

01:57:49   But developers are not doing this en masse because, as you've stated, there is no demand for gaming on the Mac because you don't game on the Mac.

01:57:57   And so developers are very wary.

01:58:00   I've talked to developers that have fairly popular games that have told me it's just not worth porting to the Mac because you inherit a support burden

01:58:09   and tax liabilities and all of this infrastructure that you're willing to accept if there's some level of expected sales volume, but nobody buys games on the Mac.

01:58:21   So it's not even worth doing if it takes five hours to do because that's just not worth it.

01:58:26   There's not that many people that will play games on the Mac because the idea that you're buying, that gamers are buying games through the Mac app store, it's just not a thing.

01:58:35   And you'll even see, and I mentioned this in the video, that Apple themselves are so desperate, in my opinion, to have games appear on the Mac

01:58:43   that they've walked out every major dev that has even considered putting games on the Mac in the last few years at WWDC and other keynotes.

01:58:53   And they haven't even required or asked them.

01:58:55   I mean, maybe they've asked, but some of these games haven't even shipped on the Mac app store.

01:59:00   So they end up shipping on Steam.

01:59:01   Apple doesn't get the 30% cut anyway.

01:59:03   Valve does.

01:59:04   And Apple still says, wow, look, there's games happening on the Mac.

01:59:07   So I really think they want games to be happening on the Mac and will take them in any flavor that they can.

01:59:13   But they're not really happening because nobody plays games on the Mac because there is no infrastructure to support Mac gaming.

01:59:20   And so, so far, so good.

01:59:24   I completely agree.

01:59:25   But where I feel like you and Syracuse go off the rails is suggesting that Apple make a product with this in mind.

01:59:32   I guess, again, I don't want to put words in your mouth, but to make something like a Switch or a Steam.

01:59:40   What's the, what's the Steam one called?

01:59:42   Steam Deck.

01:59:42   Yeah.

01:59:43   Put an M5 chip in there and then use the game porting toolkit.

01:59:49   Don't, don't wait for them.

01:59:50   Just download PC games and do the transliteration on the fly.

01:59:54   I don't know.

01:59:56   Pretty much.

01:59:57   And run PC games on an Apple branded handheld player that you could connect like a Switch or if you want, you play it in a handheld thing.

02:00:06   It would have actual buttons and then you could connect it by HDMI to your TV and.

02:00:11   Yeah.

02:00:12   Or just create a separate, you could do a box to throw an Apple TV, basically.

02:00:16   Same form factor, throw an M series chip in there, sell it for 500 bucks.

02:00:19   There's a little Apple game console and hell, make it an Apple TV too.

02:00:23   You can call it an Apple TV pro.

02:00:25   I don't know.

02:00:26   What do you, would you be satisfied if they only, if that was only the next generation Apple TV and it was, so it had to be plugged into a TV.

02:00:33   It wasn't a handheld thing.

02:00:34   I mean, I would love a handheld, but yeah, I'd be fine with that.

02:00:38   And I think that the opportunity is ripe because if you look at the console market, they're, they do incredible volume, but they're just getting absolutely blasted as is everybody, but absolutely blasted by all the stuff that's going on right now with memory shortages.

02:00:53   And Sony and Microsoft have both historically sold their consoles in the first couple of years at a loss in anticipation that, hey, the people buying on day one or year one are the most, you know, devout gamers that are going to buy dozens of games over the course of the life of the console.

02:01:11   And we'll make up that amount of money on the back end, and the people who will only ever buy one or two games ever, they're going to wait until year three or year four, by which point we'll be making money on the hardware anyway.

02:01:22   So it doesn't matter, but we're in this weirdo situation where it costs more money to buy a six-year-old PlayStation five today than it did back in 2020 when it came out and not by a small margin either by a lot.

02:01:35   It's hundreds of more dollars.

02:01:37   And Apple is in this really unique opportunity where I think they could sell a pretty comparable game console.

02:01:44   I mean, if you look at the really well-optimized ports that exist, their performance is slightly worse than a PlayStation five while consuming a fraction of the wattage.

02:01:55   And do I think that Apple could make an Apple TV style box with an M5 inside of it for less than $700?

02:02:02   Yeah, absolutely I do.

02:02:04   Because they, yeah.

02:02:06   So that's where I think the opportunity is ripe is there is this weird fraction in the market that won't exist forever.

02:02:12   And I'm not saying they would have to do it today or if the memory crisis suddenly disappeared that Apple would no longer have a leg to stand on.

02:02:20   But I think there is a unique opportunity to support gaming on a platform that is truly cross-platform because you buy a game on a PlayStation, you're stuck on the PlayStation.

02:02:31   You buy a game on Xbox, you're only playing it on an Xbox on a TV, kind of sometimes with Windows here and there.

02:02:37   But that's the compelling aspect of the Steam Deck is it is just a little PC.

02:02:42   And so you can play a game on the handheld and then when you want to go play it on your big boy PC and you're attached to a monitor, you can.

02:02:50   And Apple has that same ecosystem where they could run games.

02:02:55   You can run a game on a Mac and play it on an M5 Max and get really excellent performance.

02:03:00   But then run it on a dinky little $400 handheld that does a pretty decent job.

02:03:04   As good as an iPhone, maybe better.

02:03:06   And it's not an iPhone that's bugging you while you're playing games about messages and running down your battery and all that stuff.

02:03:13   So I don't disagree with you as much as I thought.

02:03:16   And I'm glad I had you on the show instead of Syracuse.

02:03:19   Because I think Syracuse on ATP was caught up on the Steam Deck idea, which is when you're saying, sure, they could do that or they could do it as a thing you hook up to your TV.

02:03:28   And that I think they could do.

02:03:30   I think they could turn Apple TV as a no-screen device that you have to connect to a TV that they might try to do this.

02:03:37   I think that's feasible.

02:03:38   They're never going to do it.

02:03:41   The idea that they would do a Steam Deck-like thing that plays anything other than iPhone games is just incomprehensible in the Apple mindset.

02:03:51   And you and you in your video where you talk about how Apple has really maybe technically fixed Mac gaming without actually fixing it because they can't actually get the games there.

02:04:04   Because they, you know, it all makes technical sense, right, that they could do it.

02:04:09   But it doesn't make Apple sense in terms of the way Apple views what they do in their platforms.

02:04:14   Whereas if it has a screen and it's roughly in between the size of a screen of an iPhone and an iPad, then it's an iPhone or an iPad and they already make those devices.

02:04:25   Yeah.

02:04:25   And that just ends the discussion right there.

02:04:28   And you can go, but, but, but, but, but, but there's these awesome games.

02:04:31   There's this, have you ever heard of PC gaming?

02:04:33   Do you realize, as Syracuse pointed out, that PC games are a big, and console games in particular, but, you know, AAA games, whatever you want to call them, are a much bigger market than all of Hollywood movies combined.

02:04:46   In terms of the dollars people spend and the time, the time they spend playing them, which makes sense that, that it makes more money, people spend more money and the time they spend on them is commensurate.

02:04:57   And all of those games, we're talking about playing those games on an Apple device and that you could have a services story by having a game store just for these type of games from Apple.

02:05:08   And you could take your precious 30% cut of them or whatever, all of that, but, but, but, but, but, but it's, it's just, forget it.

02:05:15   It just is incomprehensible.

02:05:16   If it has a small screen, then it is an iPhone or an iPad and that's it.

02:05:21   So, so I do actually agree with you.

02:05:23   That's unfortunately what I think their issue has been for so many years, however, because for as much as I think we, I, I don't want to include you in this.

02:05:33   A lot of people think, well, Apple's not going to do that because there's just not the volume.

02:05:38   There's not the, if, if it's somewhat similar to something that they already sell, they're not going to do it.

02:05:44   If they can't sell a hundred million of these, they're not even going to consider it.

02:05:47   But I think from a volume aspect, that's not necessarily true because look at the Vision Pro, look at the HomePod, look at the other home devices.

02:05:55   They're alleged to be working on, they're alleged to be working on a frigging doorbell, which to me is just insane.

02:06:01   Because sure, Apple could make a really nice doorbell that's better and more well-designed and runs better than a ring camera, but there's so many doorbells.

02:06:08   And what is Apple going to do?

02:06:09   That's going to be like, oh, well, revelation.

02:06:10   Nobody's thought of this before.

02:06:12   No, it's a doorbell.

02:06:12   Yeah, sure.

02:06:13   They could have two cameras on us.

02:06:14   You could have a package view that already exists.

02:06:16   Sure.

02:06:17   It could support.

02:06:18   And maybe we'll see this new fancy doorbell that apparently they're working on.

02:06:23   I don't know.

02:06:23   And go, oh, wow, who would have expected us to see that?

02:06:28   But I do think that they're always working on a bunch of stuff and not all of them see the light of day.

02:06:33   And I think that's Apple's strength.

02:06:34   They're not just releasing every single thing that they start working on.

02:06:37   Look at Project Titan.

02:06:38   Look at any number of other things that never saw the light of day.

02:06:42   But I do think that this is not a doorbell.

02:06:45   It's not Vision Pro either.

02:06:47   It is a device that could have mass market success.

02:06:50   The Nintendo Switch sold 20 million units in less than a year.

02:06:54   The Nintendo Switch 2.

02:06:56   And the original Nintendo Switch has sold over 150 million units.

02:06:59   Now, that's not the iPhone.

02:07:01   That's not even the iPad.

02:07:02   But it's pretty close to the Mac.

02:07:04   Right.

02:07:05   And it's a big, enormous market that continues to grow more and more and more every single year.

02:07:12   And the opportunity for Apple here, and this is where I was trying to appeal to Apple, the entity, is there, forget about the hardware.

02:07:22   There are enormous upsides to potential services revenue.

02:07:26   Because I think Apple's historic problem with gaming is, I think they're very complacent with where they are in the gaming market.

02:07:34   They think, oh, well, we make 30% off of every App Store transaction that involves gaming, which is almost 50%.

02:07:42   50% of the entire industry's gaming revenue anyway.

02:07:45   And so we're a massive player just by consequence of having the iPhone.

02:07:50   But I'm not confident that that money is going to exist forever or in the same volume that it has historically.

02:07:57   We're already seeing regions in which that's no longer the case.

02:08:00   And I think they have an opportunity to protect themselves from a services, I don't want to say rent-seeking revenue, but kind of in this historic app store services industry, if they were to release a gaming platform that by default supported translation.

02:08:18   You could log in with your Steam account, and you can download the games that you've already bought and purchased on PC.

02:08:24   And that sounds so anti-Apple.

02:08:25   And I understand it does.

02:08:26   But I think they would need to do that to get people to join.

02:08:30   Because it's not like they've invented this new thing where people are going to go, oh, I'm going to buy my first game console.

02:08:34   No, people have spent thousands of dollars on their gaming libraries elsewhere.

02:08:38   And so you need to support the people where they are.

02:08:40   But if they allowed you to ingest your Steam library and you could bring your existing purchases, but then Apple had additional features that were provided to gamers and users by virtue of a new store that they supported.

02:08:54   And it wasn't the Mac App Store built on the old bones of iTunes, but was actually designed with community and DLC functionality and marketplace stuff like Steam does so well.

02:09:06   And if it supported cross-play and cross-saving and you could auto-resume, like one of the best features of the Steam Deck is even though it's a PC, you turn it on and it's in the exact same place it was when you turned it off a week ago.

02:09:18   That's not a thing with PC games, but it is on the Steam Deck.

02:09:21   And now the Steam Machine, which is the TV version that Valve sells of the Steam Deck.

02:09:26   If Apple did that, where you could play a game on your TV and then press pause, turn your TV off, and you open it up on your iPhone or on your Mac or on your newfangled game console, and it's in the exact same spot as where you left off, oh my goodness.

02:09:41   Gamers would absolutely thrive to that.

02:09:45   And one of the things that Marco said on the podcast was, I think gamers are just incompatible with Apple.

02:09:51   They hate Apple, and there's so much bad blood over so many decades that gamers wouldn't gravitate towards that.

02:09:58   And while I understand that sentiment, gamers hate everyone.

02:10:01   They hate NVIDIA.

02:10:02   They hate Intel.

02:10:04   They hate AMD.

02:10:05   They hate everybody responsible for the memory crisis.

02:10:07   And at the end of the day, I think most people who play video games want the best hardware for the least amount of money possible and with the lowest resistance.

02:10:15   And I think that Apple, from a technical hardware positioning standpoint, is at a better position than almost anybody to enter this market.

02:10:23   And the revenue upsides are enormous.

02:10:26   I just think there's a culture mismatch that will make it never happen.

02:10:30   And that's a bummer because I think it could, and I think it could be amazing.

02:10:35   Yeah, that's a great way to put it.

02:10:37   I'll say this because I'm glad you brought up that comment from Marco, which is right.

02:10:41   But I think where Marco overlooks is how a lot can change in five years if you ship something and stick with it.

02:10:50   And the best example, I just thought of it immediately.

02:10:55   And it was like circa 2007 or 2008 maybe.

02:11:02   I don't know.

02:11:03   But it's not that long after Marco became a Mac user.

02:11:06   And maybe that's why he doesn't remember it.

02:11:08   But it was like a Ruby on Rails conference.

02:11:10   And there was a picture of the audience in a theater setting.

02:11:14   So you could, it wasn't flat.

02:11:16   It was raised.

02:11:17   And they were all MacBooks.

02:11:19   And you could see because the Apple logos used to light up.

02:11:23   And it's an audience of a certain personality type that is largely overlaps with gamers as sort of, and that was historically, quote unquote, hated Apple.

02:11:35   That these are the sort of exact, I mean, literally maybe, and the same sort of people who maybe bought a MacBook or still use a MacBook now 20 years later for their work and don't use it to bring it full circle.

02:11:51   Don't really use their Macs in a Mac-like way from my perspective.

02:11:56   I'd like to be a Mac Catholic.

02:12:01   You know, you can laugh, but you know what I mean, you know, and spend their time.

02:12:08   And the rise of TextMate as a text editor on the Mac, super, still popular.

02:12:13   It was sort of fallen by the wayside development-wise.

02:12:17   But it was so weird to me as a BB edit user because so many aspects of TextMate weren't Mac-like at all.

02:12:23   It was so weird.

02:12:24   It was like there was a menu command in the first version that was like option P.

02:12:28   That's how you, I don't know if it was print or what the hell the command was.

02:12:32   But I was like, you can't use option P as a menu command.

02:12:37   How in the world would you type a pi character?

02:12:39   And I said that to somebody, like, what do you mean?

02:12:41   I was like, if you type option P, that's how you type pi.

02:12:44   And they're like, you can just type the pi character, you know, like the Greek letter?

02:12:48   I'm like, yeah.

02:12:49   And if you option shift pi is an uppercase pi.

02:12:51   And they're like, what?

02:12:53   And it's like, the reason that it didn't occur to the guy who made TextMate is I'm sure he didn't know you could, that's what the option key is for.

02:13:01   It's not Mac-like.

02:13:01   But anyway, but it was, but it served the needs of Unix-minded developers.

02:13:06   And they, it wasn't right away, but within a few years after Mac OS X shipped, it was like, give it half a decade.

02:13:15   And then a conference full of Ruby on Rails developers who weren't traditional Mac users and had very strong opinions about Apple and would have sworn up and down five years prior that they would never buy a Mac, let alone a Mac lap or Apple anything, let alone a Mac laptop for work.

02:13:30   We're all using Mac laptops for work.

02:13:33   It can change.

02:13:33   Yep.

02:13:34   You know, but you have to ship it first.

02:13:37   You look at where development is.

02:13:39   And even in the last five to seven years, I don't know any professional developer who develops on a PC.

02:13:47   I mean, there are people that use, they run VS Code on a Mac.

02:13:50   Right.

02:13:51   I mean, it's, it's, it's where stuff happens.

02:13:53   And I think that could happen if Apple determined that they were serious about it.

02:13:58   And the problem with where they're at now is it's a mature market where I think they could innovate.

02:14:05   I think they could offer something really hyper compelling, but there is already loyalty out of convenience to specific platforms.

02:14:15   And so Apple either has to work with bringing compatibility there or offer something so hyper compelling that you would forfeit your gaming PC with hundreds of games that you've purchased over the last 15 years.

02:14:26   And I just don't think they can do that.

02:14:28   The good news is I don't think they would have to.

02:14:31   There's a famous quote from Gabe Newell, who's the CEO of Valve and gamers in general adore Valve.

02:14:37   I think falsely, not to suggest that I think Valve is bad, but they are a business that is designed to make money and they are effectively a monopoly on PC gaming.

02:14:51   I mean, they're not really, but I think they own 80% market share.

02:14:54   So they may as well be there's the Epic game store and others.

02:14:57   But one of the reasons Valve has had such incredible staying power is that their software is really good and that because they make 30% on every game transaction, they can do a bunch of kind of there's a meme good guy Valve that does all this stuff.

02:15:12   And they open source their crap and they give it away for free.

02:15:15   And Proton is multi-platform and they've made Linux.

02:15:18   They're one of the biggest contributors to Linux.

02:15:20   And they've made so many things so much better by virtue of it being altruistic, but it's not really altruistic.

02:15:26   It benefits them just on the back end.

02:15:28   And I think there is some, I don't want to say that it's all self-serving because I think there are reasons that they do it, but they can do it because they, just like Apple, are rent seekers.

02:15:39   They charge 30% of every game cell on the PC, which is a lot of money.

02:15:42   But that's where the gamers are.

02:15:44   They have no interest in leaving and they have famously worked with Apple a bunch over the last many, many years.

02:15:51   One of the, you might recall in 2017, when VR came to the Mac, if only through Final Cut Pro for 360 degree video editing, that was done with Valve.

02:16:02   They supported SteamVR.

02:16:03   SteamVR was ported to the Mac.

02:16:05   You could play VR video games on a Mac.

02:16:09   And I did on an iMac Pro in an era where that was, I mean, at the peak of the Intel era where literally nobody was gaming on a Mac unless you were booting into Windows, which if you had any interest in gaming, you were because there was seriously zero games for a Mac OS, much less the five or 10 that we get now that are big games every year.

02:16:29   But Newell has a quote about Apple where he says, we have this pattern with Apple where we meet with them.

02:16:34   We want to work with them, but then we never see them again.

02:16:37   And a year later, a new group of people show up who apparently have no idea that the last group of people were even there.

02:16:42   I just don't think that they've ever taken gaming seriously and they probably never will.

02:16:46   And that's a long time ago.

02:16:48   That was 20 years ago.

02:16:49   A lot can happen in 20 years.

02:16:51   Certainly the turnover that's happened.

02:16:53   But I think that's generally been the case where Apple is interested in gaming, but not enough to make an organizational commitment to it.

02:17:01   And that's what they would need to do to succeed.

02:17:04   And so for as much as I think, and they are in gaming.

02:17:09   I mean, the App Store is enormous, but they could further cement their position as a major player in gaming.

02:17:15   If they did more, I just, I don't think, I think it's an organizational and a cultural mismatch that they'll probably never get over.

02:17:23   But it's a bummer because they have the best silicon in the industry and they now have a really amazing, technically incredible translation platform that allows you to run DirectX Windows games on Mac OS remarkably well.

02:17:39   Better than Valve's own Steam hardware can do because Apple has Apple Silicon and Valve is a low volume kind of producer of PC hardware that has to work with AMD on getting the best they can get for a low price.

02:17:52   And they just, they can't get the performance that Apple could.

02:17:54   So Apple could dominate if only they wanted to.

02:17:57   But I just don't think there's the commitment there.

02:17:59   And I think there never will be.

02:18:00   So that's why I made the video.

02:18:02   I was like, oh, I wish they would do it.

02:18:04   But I think hoping, hoping that that's where they take Apple TV.

02:18:09   I think is a reasonable hope.

02:18:11   Hoping they make a handheld device with a screen that does it, they're never going to do.

02:18:15   The problem is I think that they need to do both in order to bring people over in any big numbers.

02:18:22   I mean, I think if they did it on the Mac and if they did it on the Apple TV, it'd be like, oh, yeah, cool.

02:18:26   You can do that.

02:18:28   But they would need to do two things to make sure that they dominated their position.

02:18:32   Number one is support non-Apple gaming stores, which I think is already a non-starter.

02:18:39   I don't think Apple would ever do that.

02:18:40   But they would have to in order to bring people over.

02:18:43   And then the second thing is make the hardware truly kind of multimodal.

02:18:48   And I think slapping a jawbone controller to the back of your iPhone is not going to cut it.

02:18:54   It's not the same experience as a dedicated game handheld with an M-Series chip and a fan and everything else that PC gaming handhelds just have.

02:19:02   Hell, the Switch has a fan.

02:19:03   But they're never going to put a fan in the iPad.

02:19:05   They're never going to put a fan in the iPhone.

02:19:08   But you need one.

02:19:08   I am so glad you mentioned that.

02:19:10   Oh, my God.

02:19:11   That was about to be the I'm going to kick myself for not bringing it up when Quinn was on the show.

02:19:16   That aspect of it.

02:19:18   Because Syracuse was like, you know, they invented these things called fans a long time ago.

02:19:22   And they really help with cooling.

02:19:24   Apple's approach to fans.

02:19:26   I'm not even defending it.

02:19:27   I'm just translating from Cupertinoese.

02:19:30   It is religious.

02:19:31   They hate fans.

02:19:32   It is in their DNA.

02:19:36   Jobs hated fans.

02:19:37   And they just let's I almost want to shut up about it.

02:19:43   And they're definitely if they were to make such a device, it would not have a fan.

02:19:48   There's no way if it was handheld and had a screen.

02:19:50   Which I think would be a mistake.

02:19:52   Because you look at it.

02:19:53   I agree.

02:19:53   But they're not going to.

02:19:55   Luckily, they're not going to make the handheld device with the screen anyway.

02:19:58   That's true.

02:19:58   But they wouldn't.

02:19:59   But I almost want to shut up about this whole thing because I don't want to jinx the fact that they make any devices with fans at all.

02:20:06   That the MacBook Pro still has.

02:20:07   They don't even call it a fan.

02:20:09   What do they call it?

02:20:10   Active cooling.

02:20:11   Oh, there you go.

02:20:12   I know it's not technically a fan.

02:20:14   But they even their antipathy towards fans is so much that they even invented a new terminology.

02:20:20   Active cooling system.

02:20:22   And thank God they I'm glad they do.

02:20:25   I only buy Mac.

02:20:26   The MacBook Pro is my machine.

02:20:27   And every once in a while, the active cooling system does activate.

02:20:32   And I can hear it.

02:20:33   And it does sound remarkably like a fan.

02:20:36   It's also one of the areas where you can get more performance for less.

02:20:41   If you could overclock an A-series chip and slap a fan on there, you could probably get an extra 10, 15, 20% out of that than if you didn't.

02:20:50   And that's why everybody does it.

02:20:51   That's why Nintendo does it.

02:20:53   And Nintendo is not someone famous for having loud.

02:20:56   But you play Switch 2, you never freaking hear the fan.

02:21:00   It's on there.

02:21:01   But I would never know unless you hover your nose over the top of the vent and go, yeah, I think I feel something.

02:21:06   I would bet that if you could talk to Nintendo's hardware engineering teams that they have a very practical.

02:21:12   And they probably do lean towards not having a fan and that their devices, their consoles always, or at least, I don't know, you'd have to go back to like N64 to say that they had one that was a breakthrough, leading edge.

02:21:28   Certainly since in the PlayStation era, they have never been the one with the technical advantages on frame rate or resolution.

02:21:35   The GameCube was the last one that was competitive, yeah.

02:21:37   Yeah, so that came after N64.

02:21:39   But, I mean, the Wii was, the original Wii, which really was the one that came that sort of, shoo, I was really worried Nintendo would still be around.

02:21:48   And it was only 480, right?

02:21:50   It wasn't even 720 resolution.

02:21:51   480 in the world where all the competition had HD.

02:21:56   But the Wii also had a fan.

02:21:58   Right, had a fan, but only supported 480 resolution and was a blast, was probably the most fun I've ever had playing video games in my life at home.

02:22:09   But here's, I just want to mention this, but it does, and I'm like, ooh, and I remember thinking, when the price increases came out last, earlier this month, I guess, it's been a long month, or was it at the end of June, whatever.

02:22:22   But when the Apple's, all right, here they are, price increases across the board, except for the iPhone, which I think we're going to hear about in September.

02:22:30   The one product that took it on the chin the hardest was the Apple TV.

02:22:36   Almost doubled in price.

02:22:38   I mean, I think, I don't know if you were on Twitter, or where, or I just, somewhere, you were just like, like, just cackling with laughter.

02:22:49   It's, the 64 gig one went from 130 to 200, and the 28 gigabyte one went from 150 to 250, a 67% increase.

02:22:59   And it's not just the storage, it's the fact that the bigger one is the one with thread networking.

02:23:04   Yep, has Ethernet.

02:23:06   There's Ethernet, and there's, I mean, the Ethernet, come on, just did not, nobody's talking about an Ethernet port shortage in today's world.

02:23:14   It's RAM, and it's not the Ethernet port that made the price go up.

02:23:18   It is overdue for an update.

02:23:20   I think this one came out in 2022, and the A-series chip in it is two years older than that, or at least one year older than that, so it's like a 2020 or 2021 era A chip.

02:23:31   And I thought, well, this makes sense in some ways, because they're pricing, they're not pricing this Apple TV, which will be preposterously priced for June, July, August, and maybe September.

02:23:44   I mean, I'm hoping, given that they already raised the price, that they're not saving the new Apple TV for late October or November, that this will be at the September event, because if it's not, then that means September and October are also months where if you do need to buy an Apple TV, you're getting ripped off.

02:24:00   And I thought, well, maybe this is all about the next one, and maybe the next one's actually worth more money, because here's where I'm the sucker, and somebody should take me aside and say, calm down.

02:24:11   That maybe this is the one that actually is going to be much more of an actual useful computer that could play games, too, right?

02:24:20   This could be.

02:24:21   But at this point, maybe $250, that's not really in the price range of what you're talking about, feasibly, right?

02:24:27   No.

02:24:28   And I also made the argument in my video, too.

02:24:30   I mean, the cost differences between an M-series chip and an A-series chip definitely exist, but they're not as great as you think.

02:24:38   An M-series chip is estimated to be anywhere from about 50% to 70% more expensive to produce, which is non-zero.

02:24:47   I mean, that's a lot.

02:24:48   It's 50% to 70%.

02:24:49   But it's not enormous.

02:24:51   It doesn't double the cost of a piece of hardware that you're selling.

02:24:54   And so if you're going to be selling the next Apple TV with a year-old A-series chip for $250, hell, just sell it with an M-series chip for $400.

02:25:02   What's the difference?

02:25:02   I mean, it's $150, but the market for $250 streaming boxes is so tiny anyways that you might as well just go whole hog, and maybe you'll find greater adoption.

02:25:13   I don't know.

02:25:13   Especially with these price increases, it really does seem that Apple has put Apple TV into a no-man's land, that they were already sort of in with the pre-price range prices, where they are so much more expensive than the competing products that are just sticks from Amazon or Roku or the Roku box, which are all under $100 and maybe a lot under $100 if it's just a stick.

02:25:38   And they're not expensive enough to actually be interesting.

02:25:41   They're in this no-man's land.

02:25:43   And $200 to $250 is more squarely in that land, where it's an even harder sell for anybody who's not already convinced that Apple TV is the best streaming device to watch on their TV, which I am.

02:25:57   But I also, again, I don't proselytize it to my family members because I don't want to be the same way that I don't want to be on the hook for their tech support.

02:26:05   It's not my job to sell somebody $150, but now $200 box to hook up to their TV, when their TV might have only cost $200.

02:26:14   And it's really big, and they're very impressed and blown away by what they got for $200.

02:26:19   And then to tell them that the thing that I'm telling them they should buy to connect to their TV to watch Netflix, which their TV says it already does, is $200?

02:26:28   And for as much as the TV software sucks, by virtue of the Apple TV existing, it has gotten better.

02:26:35   I mean, if you run Google TV built onto a bunch of TVs, it's not bad.

02:26:40   It's not great.

02:26:41   It's not the Apple TV.

02:26:43   But is it bad enough that I'm like, oh, my gosh, I have to spend $250 to get an Apple TV?

02:26:48   No, I might not even do that.

02:26:49   And I'm the Apple guy.

02:26:50   Yeah.

02:26:52   I think that, too, we bought an outdoor Samsung TV for our deck, and I'm going to connect an Apple TV to it, but I'm waiting for a weatherproof box.

02:27:02   It's a long story.

02:27:03   But in the meantime, it's just running the Samsung software, and I don't love it, but it's like, I don't really watch a lot of TV up there, so I don't really care.

02:27:12   And it's not bad.

02:27:13   And it's actually a lot better, in my personal opinion, to a Roku.

02:27:18   Yeah, Roku is by far the worst, in my opinion, and I think Samsung's right above it.

02:27:22   I think they're both pretty bad.

02:27:24   But that's a good illustration that even the stuff that sucks is like, it's fine.

02:27:29   What are you doing?

02:27:29   You're just opening the streaming app of your choice.

02:27:31   Once you're in Netflix, once you're in, heck, the Apple TV app that's everywhere, it's kind of pretty much the same.

02:27:38   It's not technically, if you have a home theater set up, and there's slight codec differences, but for most people that are just using the internal speakers on their $500 TV they got on Black Friday, eh, who cares?

02:27:50   It's like going to a restaurant where the foyer has a bad smell, and the person who checks you in is rude and competent, and it's always a hassle, even though you made a reservation.

02:28:03   And it's like, ugh, you can hold your breath through the foyer, and it's like they do eventually give you a table, and then that was really only 90 seconds of a bad smell and a bad check-in experience.

02:28:15   And then the rest of the time you're eating the meal, that's streaming, right?

02:28:18   It kind of sucks to futz around with the Roku.

02:28:21   But then once you get the show on that you wanted to watch, you watch the show, and it's the same.

02:28:26   Who cares?

02:28:27   And it's fine.

02:28:27   And so I get it.

02:28:29   I do think Roku stinks in particular.

02:28:32   I mean, I just had to use one a couple months ago for the first time in a while.

02:28:35   I was like, I can't believe this.

02:28:37   I can't believe how they made this worse than just the dumbass just list of channels in a grid.

02:28:43   They've added this weird interface that is trying to be suggested.

02:28:49   It's like, no, just show me a grid of icons, you idiots.

02:28:51   But fine, it just works.

02:28:53   But it's like the built-in thing just works.

02:28:55   So I don't know.

02:28:56   I don't know what Apple is doing.

02:28:58   But the opportunity for them to do it is Apple TV.

02:29:00   And I do think that's the other thing that's going to kill the idea that you have and that Syracuse endorses is how in the world can they sell a $700 thing that competes with the $200 thing from Apple TV?

02:29:15   I mean, maybe.

02:29:15   If it's a PlayStation, that's how I think they have to do it is if they sell it as a console that also does all the TV stuff better than, I mean, because the consoles kind of technically do it, but not well.

02:29:26   If you could say, hey, this is the one box that does everything.

02:29:28   It's the best streaming box, and it's the best game console, and everything's, oh, it could be so good.

02:29:34   They just won't do it.

02:29:35   Right, and yeah, and then maybe instead of having two Apple TVs where the difference is sort of you have to be a real nerd to, oh, it's got thread networking and Ethernet port.

02:29:45   And, like, try, I mean, how many people out there really know what an Ethernet port is?

02:29:48   It's like, what?

02:29:49   You get Internet through a cable?

02:29:50   And you're like, yeah, you used to always get it through a cable.

02:29:52   And they're like, no, you didn't.

02:29:54   It's always been in the air, and it's like, oh, but what if they had two Apple TVs, and it was regular Apple TV, and it's 200 bucks, and it's good, and there's only one size, and it has the thread networking and the Ethernet port, and there it is.

02:30:07   It's the new Apple TV, and it's 200 bucks, and then there's Apple TV Ultra, and it's a slightly bigger box, and it has an active cooling system, and it's 500 bucks or something, or 700, and it does, and then you could have games.

02:30:23   There's games that run on all Apple TV, and then there's games that are only for Apple TV Ultra.

02:30:28   Yeah, it could be cool.

02:30:30   I think the problem is it comes back to this idea of even if they did that, it's dead in the water if they don't have the store support ready to go from the beginning.

02:30:39   And they won't, because they're just going to have the app store, and they're going to say, we have this amazing hardware.

02:30:44   Developers, come bring your game.

02:30:46   And nobody will.

02:30:46   And so they're going to have to do it on their own.

02:30:49   And that's what Valve did with Linux, is they're like, look, no matter how hard we try, nobody's going to freaking port their game to Linux.

02:30:54   So we've just got to support translation of these games such that they can run on Linux.

02:30:59   And now that there's millions of people playing video games on Linux, guess what?

02:31:04   Game developers actually do target for Linux now.

02:31:06   Yeah.

02:31:07   Because they know it's going to happen.

02:31:09   A reasonable person, and who knows?

02:31:11   I mean, again, I do not think this is going to happen.

02:31:14   But there is a new CEO coming in, and we know that the old CEO, Tim Cook, had a religious devotion to the services revenue and the sort of, like, the Phil Schiller memo from 2011 that was like, when the app store first started, this came out, I think, in the Epic trial.

02:31:32   But it was discovered in one of the trials, probably the Epic one, but it was like, hey, the run rate on the app store is so exceeding what we, even our wildest dreams.

02:31:41   We're going to adjust that percentage.

02:31:43   What if we cap it at a billion dollars, and once we start making a billion dollars, we just keep lowering the commission rate to keep our side at a billion, which is good money, and then we can just lower the commission.

02:31:57   And we don't, there was no email entered with Tim Cook's answer, but we know the answer, because we know what they did.

02:32:03   And his answer was, no, I like 30.

02:32:05   How about 30?

02:32:06   There's a great New Yorker cartoon where there's, like, a business guy on the phone, and he's making an appointment, and he's like, the caption is, how about never?

02:32:16   Does never work for you?

02:32:17   And I think, how should we adjust the rate?

02:32:20   I think Tim Cook's answer is 30.

02:32:22   How about 30?

02:32:23   Does 30 work for you guys?

02:32:25   Let's go around the table.

02:32:27   What do you guys say we keep it at 30?

02:32:28   And what if we say we keep it at the exclusive way to load software, games, otherwise, whatever, right?

02:32:34   Because the obvious answer of, hey, you could just put the App Store on this new device and work out a deal with Valve and just do Valve.

02:32:42   And maybe Valve would throw 5% Apple's way, right?

02:32:45   We'll build you into the thing, and anybody who buys the thing here, you get 25.

02:32:50   Keep it at 30, and you get 25, and Apple gets 5.

02:32:54   I don't know.

02:32:55   I bet Valve would do that.

02:32:57   No, they for sure would.

02:32:58   And it would be there alongside the App Store, and all of the App Store games for Apple TV that you know and love would be there.

02:33:06   I laugh because I don't think there's many Apple TV games that people know and love.

02:33:12   Crossy Road, yeah.

02:33:13   But they'd be there.

02:33:14   And if that's the type of gaming, I don't know, maybe more people play those games on Apple TV than I think.

02:33:18   They'd all be there.

02:33:19   And all of the streaming stuff would all come from the App Store, right?

02:33:23   Your Netflix and your Hulu and your Disney Plus would all come from the App Store, not the Steam Store.

02:33:28   But then you could have a second source of games, and no one would be confused.

02:33:31   It would just be Steam.

02:33:33   And if you believe that your App Store monopoly is in jeopardy, which I think in a lot of ways they do, hey, this is a new way to re-up the service's revenue volume while still being able to say, no, we're competing honestly and fairly.

02:33:50   This device supports multiple game place marketplaces, and we're selling games via the App Store on Merit, and we think that buying games on the App Store is the best place to do it.

02:34:02   But if you have an existing library, if you want to buy it somewhere else, we're not going to put a buy the game on Steam button on the Apple TV.

02:34:09   But if you go do it on your computer, yeah, it'll show up on the Apple TV, too.

02:34:13   And I think they would kill it.

02:34:14   I think it'd be amazing.

02:34:15   But they won't.

02:34:16   No.

02:34:18   Yeah.

02:34:19   And it's not because I think Tim Cook only picked John Ternus because he agrees with everything Tim Cook does.

02:34:25   I think we're going to – I don't know what they are because I don't think we know Ternus.

02:34:28   No, we don't.

02:34:28   I think we're going to see noticeable changes, and we'll be like, oh, John Ternus has different opinions on X, Y, and Z than Tim Cook did.

02:34:35   I think we will start seeing them fairly quickly.

02:34:38   I do, too.

02:34:38   And I think that's part of – I think that is why Tim Cook picked him, that he is looking for an independent-minded CEO with different ideas.

02:34:46   I don't think that's going to be one of them.

02:34:48   No.

02:34:48   I don't think Ternus is a gamer, unfortunately.

02:34:50   No.

02:34:51   So that's not going to change.

02:34:53   He's been at Apple for 30 years.

02:34:54   Right, and I think that's ultimately what you need.

02:34:56   And if secretly he is a gamer, that could be the change.

02:35:01   That would be the change.

02:35:02   Yeah.

02:35:03   But let's see.

02:35:03   All right.

02:35:04   Good discussion.

02:35:06   Thank you, Quinn.

02:35:06   Thank you.

02:35:09   Your channel on YouTube, YouTube.com slash snazzy.

02:35:13   When did they put the ad symbols in?

02:35:14   I don't even know.

02:35:16   I've had an account for so long that you don't need to use the ad symbol on my channel.

02:35:20   I don't know why.

02:35:21   But I've got the real YouTube slug.

02:35:23   So YouTube.com slash snazzy is where you can find me.

02:35:28   Right, and I will have links to the videos we just talked about, the Trump T1 phone, the Mac Gaming one.

02:35:34   But it's a great channel, and you're continuing to do great work there.

02:35:38   And I thank our various sponsors.

02:35:41   I really mean it.

02:35:42   It's great.

02:35:43   I laughed so many times at the goddamn T1 video.

02:35:45   It is.

02:35:48   And again, of course I did, because it made me laugh.

02:35:52   Technically, it made me laugh.

02:35:53   Politically, it made me laugh in terms of the gold.

02:35:57   Anyway, I want to thank our sponsors for this episode.

02:36:00   Let's go backwards in time.

02:36:02   Squarespace, where you can build a website.

02:36:04   Factor, where you can get Meals to Eat.

02:36:08   And Notion, the all-in-one AI platform for teams.

02:36:11   Thanks.