PodSearch

The Talk Show

439: ‘A Mitigated Disaster’, With Daniel Jalkut

 

00:00:00   Daniel, it's good to see you.

00:00:01   You too.

00:00:01   We're both snowed in right now, right?

00:00:04   Well, the whole eastern half of the United States, I guess.

00:00:07   I don't know.

00:00:08   It's one of those times where everybody I know on the East Coast has the exact same weather.

00:00:13   Yes, it's true.

00:00:15   Sub-freezing and about a foot of snow.

00:00:17   Exactly.

00:00:18   Yeah.

00:00:18   I was just out before we started recording.

00:00:21   I went out to fill up a portable gas tank thing for my snowblower, and it's kind of surreal out there.

00:00:27   I don't know if you've been out at all, but it's just almost...

00:00:30   No, Amy and I went out for a bit.

00:00:31   Okay, yeah.

00:00:31   Just walked like four blocks.

00:00:32   Yeah.

00:00:33   Took some pictures.

00:00:34   It's kind of insane out there.

00:00:36   But yeah, it's fun.

00:00:37   I like it when everything kind of slows down.

00:00:39   Everything is forced to be slow, so...

00:00:42   Yeah.

00:00:42   We get very little...

00:00:44   Well, relatively speaking, we get little snow in Philadelphia, and we haven't had more than six inches of snow, and I forget how many years they said, but it doesn't happen very often.

00:00:56   But I love it when it does, because the city...

00:00:59   I mean, because we live right in the city, and it gets so quiet from when it snows, because there's no traffic, or almost no traffic, and the traffic that is there is just creeping by on a bed of snow.

00:01:11   And then I thought to myself, well, if I like it so much, why the hell do I live in a noisy city?

00:01:15   Oh, yeah.

00:01:16   Well, you probably like all the things that you can get to amidst the noise.

00:01:21   It's like when you like something once a year.

00:01:24   Yeah, right, right.

00:01:26   What else is going on?

00:01:28   Are you ready to hear all of my gripes?

00:01:29   I don't know.

00:01:30   I'm worried that everybody who reads Staring Fireball is going to get sick of me complaining about macOS 26 Tahoe.

00:01:36   I think it's kind of weird.

00:01:38   It's kind of surprising and weird.

00:01:40   Maybe not too surprising, but the way that everybody's complaining about Tahoe right now, not just Accidental Tech Podcast is going off on it, that is partly inspired by these recent blog posts, like criticizing the icons and blah, blah, blah.

00:01:55   You would think that Tahoe came out like last week.

00:01:57   Yeah, and I was thinking about it.

00:02:00   And I forget when Apple first started consistently doing, I guess it's been a long time, where the betas come out at WWDC in June for all of the operating systems.

00:02:12   And I feel like in the early years of that, there was a lot more, hey, I know it's a beta, so let's forgive glitches, but let's review it now.

00:02:24   If they've changed X, Y, and Z, let's review it.

00:02:28   And if you don't like it, call out what you like, like, hey, here's a really great thing to look forward to when this hits release later in the year, and here's some stuff that Apple should reconsider.

00:02:37   I've gotten away from that, and I feel like I got enough pushback from people at Apple that it's like, hey, it's unfair to review the beta that's in beta.

00:02:46   And I think, maybe it's just me, but I think that the betas have become more in flux in recent years.

00:02:53   Whereas what ships in the middle of June when WWDC hits isn't supposed to be what it's going to look like in September.

00:03:01   And it's all sort of, at least for me, I'm blaming myself, not Apple, not anybody else, but I'm blaming myself where it's like, well, then I don't want to give them a hard time over the summer because they're still working on it.

00:03:11   And then by the time the fall comes out, a lot of the complaints don't feel fresh anymore.

00:03:16   And so I don't make them when it hits the release either.

00:03:20   And then it's like, well, when am I supposed to complain?

00:03:23   Well, and if you think about it, if nobody complains from June, there's no feedback.

00:03:29   I mean, what's the point of having the beta if you don't take the complaints?

00:03:33   Maybe what they're saying is, hey, file the bugs privately.

00:03:37   Don't write about it publicly.

00:03:38   But it's the kind of thing we need a collective reckoning of these major releases.

00:03:43   And sometimes I think the feedback really does reach them.

00:03:48   And remember a few years ago, like they completely overhauled Safari's tabs.

00:03:53   And it was kind of like the tabs were like up at the top of the window edge.

00:03:57   It was really unusual.

00:03:59   And everybody I know pushed back.

00:04:01   And so that didn't happen.

00:04:04   Yeah.

00:04:05   And I contributed to that.

00:04:09   I don't want to, I'm not trying to take credit, but obviously if I make a good point and make it hard, it has carried some weight outside Apple.

00:04:18   And I think that's one of the ones where I got some pushback, like, hey, okay, we'll fix this.

00:04:23   But, you know, maybe take it easier on it while it's in beta.

00:04:26   And I kind of feel like with Tahoe in particular, there was a sense with Liquid Glass in general, hey, what they're showing us at June and what we could see right away, like on the iPhone, where it was the most polished and clearly was the device that Liquid Glass started on conceptually.

00:04:48   And then it's like, all right, now how can we apply this to the other devices?

00:04:51   It was so clearly, oh, this isn't finished, right?

00:04:55   It really wasn't in June.

00:04:56   And a whole bunch of the animations they showed in the keynote weren't present in the initial betas.

00:05:02   They were, whether they were simulated for the keynote or just in private builds that weren't merged with the main build, whatever.

00:05:10   It's, this still doesn't look like what they're saying it's going to look like.

00:05:14   And to their credit, especially on iPhone, by 26.1 and 26.2, it really does look like what they promised us in June.

00:05:23   But I feel like with a lot of it, on all the OSs, it was like, well, this is clearly not finished, so let's give them a break.

00:05:29   And I think with Tahoe in particular, it was like, well, a lot of this, they can't possibly intend to ship like this.

00:05:35   And that's how they shipped it.

00:05:37   You're right.

00:05:37   Right, right, right.

00:05:38   Everyone was saying that release after release.

00:05:40   Wait, where's the fix for this?

00:05:41   There's, people are using the expression paper cuts a lot.

00:05:45   And I think that is accurate.

00:05:47   I'm sitting here.

00:05:48   I'm one of those people.

00:05:50   I think I usually start using the new OS, like, in June, and then just never look back.

00:05:55   Sometimes it bites me, but for the most part, I just, I'm used to kind of like living through uncomfortable stretches of the OS.

00:06:05   And I'm kind of perplexed.

00:06:07   I was just thinking about it, before you had even asked me to come on the show, I was thinking, how is Apple living with themselves over some of these things?

00:06:17   Because they're the kinds of things, like little misalignments all over the place.

00:06:22   They're the kinds of things that Apple is supposed to care about that nobody else cares about.

00:06:27   Right.

00:06:28   And to see so many of them.

00:06:30   I'm just, I'm sitting here now, the app switcher, command tab.

00:06:33   You've probably seen this too.

00:06:34   You notice how, like, at the edges, the one at the first and the one at the last, often the name is not under the thing you've selected.

00:06:41   It's been driving me crazy.

00:06:43   That's one of the ones where I was like, come on, fix that.

00:06:45   Yeah, yeah, yeah.

00:06:47   But somebody should be ashamed.

00:06:49   I mean, that's the kind of thing where, I don't know, I'm a shame-driven developer.

00:06:53   You will not see a release come out fast enough if somebody points out something that just looks like it was reckless or, you know, inconsiderate of the quality of the product.

00:07:03   Yeah, another one, you may not have it handy in front of you right now.

00:07:07   I can see it because I'm podcasting.

00:07:09   I am, as I keep reiterating on Daring Fireball, I'm using Tahoe on my podcast Mac down here in my basement.

00:07:16   I'm not using it.

00:07:17   And my current plan is to ride out the whole year without upgrading on my work machine up in my office.

00:07:24   But I can see it right now.

00:07:26   And I think it's a problem here on Tahoe 26 that when you command tab, in addition to the thing where the name does, yeah, the first one, like now that it's back to Google Chrome for me, because it doesn't fit centered under the icon.

00:07:40   It just, it's squished in, but off to the right and like halfway under, it's halfway under the next icon.

00:07:49   Yep.

00:07:50   Right?

00:07:50   So yeah, you haven't been seeing that because you're not on it full time, but that has been there, I believe, since day one.

00:07:56   Right.

00:07:57   And it needs to be a long name.

00:07:59   So Google Chrome is a nice long name for an app.

00:08:02   And so when Google Chrome is my leftmost icon, when it goes back to Mars edit, then it's nice because it fits.

00:08:08   But Google Chrome, no, not so much.

00:08:10   Yeah.

00:08:11   The other problem is, do you agree?

00:08:13   I don't think that the highlighting, when you're R command tabbing and you're like command tab, tab, tab, because you want to go back like three apps that you most recently used.

00:08:22   The highlight around the one that you're selecting is way too subtle, in my opinion.

00:08:27   It is.

00:08:27   But on iPadOS 26, it's almost invisible.

00:08:31   It is, it's almost like an eye test.

00:08:35   Can you see a highlight around the third icon?

00:08:38   And again, that's the whole point of the command tab switcher, is to switch to another specific app.

00:08:46   The whole point of it is to pick an app.

00:08:49   And so the highlight should be extremely vivid, not subtle.

00:08:54   And whatever urge there is within Apple to have subtle UI elements and that this, clearly, I mean, this has been a decade plus, maybe longer in the making.

00:09:08   Of all sorts of complaints across the Apple's OS is, hey, why is that so subtle?

00:09:15   I remember when Safari tabs didn't have fav icons, fav icons.

00:09:21   And it was like, somebody in Apple didn't want color icons in the tabs.

00:09:25   And they were like, okay, if you pin a tab, we'll let you put an SVG icon.

00:09:31   But you have to use a custom format, because we're Apple, that is black and white.

00:09:37   So it's a monochrome SVG icon.

00:09:39   When every single other browser in the known universe just puts the fav icon in the tab.

00:09:45   And then I complained about that, and I kind of do take credit for getting that one fixed.

00:09:50   There was no good argument against putting the color icon of the website in the tab for that website.

00:09:58   Other than, oh, it looks a little cluttered, and we don't get to pick the colors, and CNN's icon may clash with Yahoo's icon or something.

00:10:06   But it's your icons.

00:10:08   These are the websites you've gone to.

00:10:10   They should be there.

00:10:12   And with the command tab switcher, any argument that like, hey, selection states should be a little subtle as opposed to being very vibrant.

00:10:20   Well, why would you want that in the command tab switcher, which is only up temporarily while you're holding down the keys for command tab, right?

00:10:29   It's not like it's in your face all the time.

00:10:31   It is the definition of a user interface element that is only up while you're using it, and then immediately goes away.

00:10:38   So it should be as vibrant as possible.

00:10:41   And yet it's more like a puzzle.

00:10:43   Did you get the right app?

00:10:44   Who knows?

00:10:45   Let go and see.

00:10:47   Well, anyway, I don't want to try to fit in the whole rant about this without taking a break.

00:10:53   So let me take a first break here and thank Returning Sponsor.

00:10:58   This is a sponsor who was here a couple months ago, and they had such a good response that they are back.

00:11:03   It is our good friends at Sentry.

00:11:06   Sentry is an all-in-one application code monitoring and fixing toolkit.

00:11:12   Sentry, S-E-N-T-R-Y, not century.

00:11:16   Sentry, like somebody looking over your app, doesn't just catch errors.

00:11:21   It catches all the stuff that makes your app feel broken.

00:11:24   The random slowdowns, the freezes you can't reproduce, the bugs that only show up once a real user hits them.

00:11:30   When something goes wrong, Sentry gives you the whole chain of events in one place.

00:11:35   Errors, traces, replays, logs, actually connected so you can see what led to the issue without digging through five different dashboards.

00:11:44   And that's what SEER, S-E-E-R, Sentry's AI debugging agent, builds on.

00:11:50   It takes that full context, explains why the issue happened, and points to the code responsible, drafts a fix, and even flags if your PR is about to introduce a new problem.

00:12:02   So the workflow stays simple.

00:12:05   Something breaks, Sentry shows you the full context, and then SEER helps you fix it and catch new issues before they ship.

00:12:13   Sentry is used by millions of developers behind some of the biggest apps like Claude, Disney Plus, Duolingo, big apps with lots of users.

00:12:24   They support over 100 languages and frameworks across front-end, back-end, and mobile.

00:12:29   S-E-N-T-R-Y.

00:12:32   Try it free at Sentry, S-E-N-T-R-Y dot I-O slash talk show.

00:12:40   That's Sentry dot I-O slash talk show.

00:12:43   And when you use that code, talk show, you get 80 bucks in Sentry credits.

00:12:49   80 bucks for free just for listening to the show.

00:12:52   What else?

00:12:54   What else about Tahoe?

00:12:56   Do you agree?

00:12:57   Am I nuts?

00:12:57   Here's the thing.

00:12:58   And I don't think – I feel like I've been hammering this point long enough that if there was widespread pushback –

00:13:04   because sometimes when I feel like I'm veering towards an extreme opinion on macOS 26 Tahoe,

00:13:09   which is that it's an unmitigated disaster and an embarrassment to Apple and that I'm really best avoiding it.

00:13:16   And I think, well, that's a harsh take for me.

00:13:18   But I'm not really getting much pushback from readers.

00:13:21   I mean, I'm getting a little from some who are like, hey, I upgraded back in September.

00:13:26   And there's a couple of the things you talk about that are annoying.

00:13:28   But overall, it's fine.

00:13:30   And of course it is, right?

00:13:32   It's not, oh, you install it and everybody's MacBooks are kernel panicking, right?

00:13:37   It's not losing data.

00:13:39   It's not corrupting your – it's not like they have file system bugs or something truly horrendous.

00:13:45   It's just annoying.

00:13:47   And I kind of feel like even – it's like the confirmation to me is that even the people who are – the readers who are politely, almost all, telling me,

00:13:56   hey, I don't really see why you're so hung up on it, seems fine to me.

00:14:00   The sentiment is exactly what you were saying earlier, Daniel, which is, yeah, fine is not the Apple way, right?

00:14:07   Right.

00:14:08   Right.

00:14:10   Like, how about the one that a couple of people wrote about the problem with – what was his name?

00:14:16   Norbert – he's the guy – he's one of the developers of LaunchBar.

00:14:18   Oh, yeah.

00:14:19   It looks like he started a new blog just to post this because it's the only post on his blog.

00:14:25   Norbert Heger – I hope I'm pronouncing that right because I believe he's German or something.

00:14:30   Why it's difficult to resize Windows on Mac OS.

00:14:33   Oh, yeah.

00:14:34   Yeah, yeah, yeah.

00:14:35   That one was interesting for me because I didn't notice it until I read the post, which is saying something.

00:14:42   In other words, it didn't affect me or it didn't affect me enough that I became aware of the problem.

00:14:49   But now that I've read the post, I'm missing the target all the time.

00:14:52   It's kind of weird, like the knowledge of it.

00:14:54   Or maybe I'm just noticing that I'm missing it.

00:14:56   But, yeah, that is – that's another little problem here.

00:14:59   I just – I guess I should stake out my position on Tahoe because you said it's an unmitigated disaster or you're leaning towards feeling that way.

00:15:07   Yeah.

00:15:08   That's how I feel.

00:15:10   You never hear people talk about it, but maybe it's a mitigated disaster.

00:15:14   Yes, yes.

00:15:15   Nobody says it's a mitigated disaster because everyone's getting by.

00:15:21   It's fine.

00:15:22   It's like the Titanic sinking is an unmitigated disaster, but the Titanic showers being cold is just like a mitigated disaster.

00:15:32   So I don't know.

00:15:34   The fact is I was actually thinking about this a little bit.

00:15:36   When you asked me to come on the show, you warned me we're going to talk about Tahoe.

00:15:40   And I was doing some errands and I was thinking, why am I not bothered more by the little things?

00:15:46   And I started thinking about it.

00:15:47   Basically, I think since I've – since my whole career working at Apple, working on my own stuff, I think I'm more dedicated to the pursuit of giving users a clean, flawless experience than I am insistent upon having that experience myself.

00:16:09   So maybe it's like a dress designer who goes around wearing blue jeans.

00:16:13   It's like they don't need to be wearing the fancy ballroom dress to appreciate the craft and the work that goes into it.

00:16:22   Yeah.

00:16:22   And I feel for the people inside Apple.

00:16:26   And I am not hearing from people currently at Apple because they know better.

00:16:30   But I'm hearing – and I think it's a sign of my age that I know a fair number of people now who are retired from Apple because a lot of people, as it's kind of common sense, that people not much older than me are able to retire early if they spent their career at Apple because they were there for a long number of years.

00:16:51   And from people who are out of Apple, I'm getting, go, keep it up.

00:17:03   This needs to be heard inside the company.

00:17:05   And I think Tahoe, because it's very specific, it is this OS for the Mac, this version that has taken such a sharp turn.

00:17:16   And again, I mean this with no hyperbole.

00:17:19   I wrote the other day on Daring Fireball that I can't think of anything that makes me want to upgrade to Tahoe.

00:17:25   And there have been years past – I've been doing this a very long time – where there have been years where I almost never upgrade.

00:17:32   I don't upgrade to my work machines in June like you do.

00:17:35   And I know that developers often have to.

00:17:37   I usually wait until like the .1 update in October or November, shake out some bugs and upgrade then.

00:17:47   And sometimes the .2, depending on when it comes out.

00:17:50   You know, some years it's a little buggier than others, but this one feels like this sharp turn.

00:17:55   But just concentrating on this, it really helps to clarify that an awful lot of things at Apple in recent years I think have been – like how can this ship?

00:18:09   How can they excuse it as this bubbles up the hierarchy to the executive level in the company?

00:18:13   And I think it really boils down to a bit of maybe almost unavoidable arrogance that comes from, well, this is still the best there is, right?

00:18:24   Like this – for whatever you want to say about some of the iPhone interface problems in liquid glass, it is still the best mobile operating system user interface, right?

00:18:36   And the problem for me with Tahoe is it's not the best desktop computer interface.

00:18:43   Last year's version to me is.

00:18:46   And I could even see – and I'm sure there's listeners out there who are itching to jump in and interrupt us and say, no, no, no.

00:18:53   You've got to go back like 10 years to like 2014, 2015.

00:18:57   And I kind of agree and it was a post where I went back and I was looking at Stephen Hackett's collection of screenshots from every version of Mac OS X from the Mac OS X public beta in 2000 up through Tahoe.

00:19:10   And it's, man, the versions from around 10 years ago, like when Alan Dye first came in, they look so good and they don't look old.

00:19:20   They don't look like, oh, it's a little bit of a retro Obama-era feel.

00:19:26   They look like they could ship that interface today and it would look right at home as the Mac version of Apple's current design language right next to iOS 26 running on an iPhone.

00:19:43   Now, it's not liquid glass 10 years ago, but, you know, it's the San Francisco font.

00:19:48   It's a similar color palette, a similar overall aesthetic, but applied to a real serious work desktop graphical user interface.

00:19:58   It's gotten worse.

00:20:00   It really has.

00:20:01   And it's not just different.

00:20:03   It's not like, oh, I don't like more transparent things.

00:20:07   Transparent things are not to my liking.

00:20:09   It's real things like, hey, once Norbert Heger posted this, everybody's like, oh, that's why I can't resize windows.

00:20:18   Right?

00:20:18   Yeah.

00:20:19   Yeah.

00:20:21   Like I said, Apple should be embarrassed about all the little problems.

00:20:24   I'm sure some people at Apple are embarrassed.

00:20:26   I had that metaphor with the ballroom gown.

00:20:30   It needs to be perfect, right?

00:20:31   So imagine like a beautiful like Oscars red carpet ballroom gown and then there's like a big long thread hanging off of the bottom or like a ripped piece of fabric that was like pasted on the bottom.

00:20:45   It's like you're looking for an elevated product.

00:20:50   It needs to be an elevated product.

00:20:51   And as soon as it starts, I guess what I'm just to reiterate, I guess what bothers me is some of these problems should be driving people at Apple crazy that they're not fixed yet.

00:21:04   And maybe they are, but I don't understand.

00:21:06   There's this other little issue I noticed where, oh, probably like about half of all of Apple's.

00:21:12   So you know how like sometimes there's an explanatory text under a radio button or a checkbox and it's usually aligned with the text of the radio button or the checkbox.

00:21:22   There's no automated way to do that.

00:21:24   You have to just kind of indent it the right amount.

00:21:26   And if you start looking around on Tahoe, you'll see in like Apple's preference panes everywhere that the text is a little bit to the left because they made the checkbox or the radio button a little wider.

00:21:40   So this is one of these things where I was like, yeah, that's annoying, but actually you could probably fix that.

00:21:47   I mean, so the problem is they have to fix it in every app, but they could have somebody just be like, I'm the fixer.

00:21:54   Give me access to all the source code and give me two days and this will not be a problem.

00:21:59   I feel like there must be systemic issues that are preventing widespread perfectionism these days.

00:22:07   And as a developer, somebody who's trying to target the latest OS in as many ways as are sensible, I'm often looking to Apple for direction as to how to embrace the design.

00:22:21   And there are very few, even Apple apps.

00:22:24   I'm hoping with this new iWork suite coming out that we're going to get some direction because right now, looking on my Mac, Safari has adopted Liquid Glass.

00:22:36   Lots of other apps either haven't adopted it or have adopted it in a very light way.

00:22:42   I was looking at TextEdit just now.

00:22:44   TextEdit.

00:22:45   That's what I was looking at.

00:22:46   I was looking at because I know TextEdit still has an old-fashioned settings dialog, not the new iOS style one.

00:22:52   And I was looking at it to look at the checkboxes.

00:22:54   And I see what you mean about the labels being closer to the buttons.

00:22:57   Yeah.

00:22:58   Actually, now I'm looking at Xcode.

00:23:00   Xcode might be the single most complex app I've seen that embraces Liquid Glass.

00:23:09   So that's kind of funny, actually.

00:23:11   But, you know, if they shipped this in, what, September?

00:23:14   And I guess it's not that long.

00:23:17   But I was expecting to see Apple, like, just coming down with – because making such a major UI change, the way to beta test that internally is to change your own apps to adopt it.

00:23:30   That's how you figure out what design changes work and don't work, is to actually try to apply them to apps.

00:23:38   I'm sure they did it to Xcode.

00:23:39   They did it to Safari.

00:23:40   But there's a lot of, like, complex UI stuff that apps like Pages or Keynote have that Safari doesn't have.

00:23:51   So it's sort of adding to the overall incoherence of the OS that, for instance, even among Apple's own apps, TextEdit is an app that everybody uses.

00:24:03   I mean, it must be 95% of all Mac users use both Safari and TextEdit to some extent.

00:24:11   And they're using this top-of-the-line OS that ships with two apps that everybody uses that don't adopt the same design language.

00:24:19   So I think that that inconsistency – here I am ranting after saying I'm fine.

00:24:24   I get by fine on the OS.

00:24:25   But I'm ranting on behalf of the other users.

00:24:29   You know what I mean?

00:24:30   I don't care as much.

00:24:31   I would like to see it coherent and pristine, but it bothers me.

00:24:37   I still have a tiny bit of mental Apple ownership.

00:24:41   Like, I identify as an Apple employee 20 years after I quit.

00:24:44   And after only working there relatively briefly.

00:24:48   Yeah, six or seven – six, seven years.

00:24:52   But I still feel proud of Apple.

00:24:54   And so when they do something that I think is shameful, it sort of makes me feel ashamed.

00:25:00   It kind of makes me feel like a sucker for still defending them.

00:25:04   But it's not so bad.

00:25:06   It is a mitigated disaster.

00:25:11   I guess – I just don't think some people get for how long.

00:25:17   It wasn't like, oh, there was a brief heyday for just a handful of years where inside Apple there was a real focus on user interface perfectionism.

00:25:30   And that doesn't mean every choice was perfect, which can't be, and that's totally subjective.

00:25:37   And probably inside there were always, should we do it this way or that way?

00:25:41   And you may not agree with the fact that they chose that way instead of this way.

00:25:45   But when they did that way, it was pixel perfect.

00:25:48   And if it was supposed to look like this, then it would look exactly like this to the pixel.

00:25:57   And I've heard, and you've heard, and some of the stories are public, that was really the way prototyping would work within Apple.

00:26:07   That if the design team was making mock-ups using Photoshop or some other just graphic tool and got approval maybe all the way up to Steve and was like, okay, this is what it should look like.

00:26:21   Then that's what it was supposed to look like to the pixel, and it was going to look like that to the pixel.

00:26:26   Or somebody, including Steve, is going to notice that this is one pixel – this whole sidebar is one pixel too low.

00:26:32   Fix it.

00:26:33   Yeah.

00:26:33   And that would rise to the level – I think that a fair criticism of Apple over the years is that sometimes fixing 50 little misaligned text boxes or divider bars, using your time to do that is time better spent than adding another user feature.

00:26:52   And that's a philosophy that many people in this world do not agree with because it leads to less features in the product.

00:27:02   But for those who do, Apple has been the place.

00:27:05   And so when that starts to deteriorate a little bit, it's like, come on.

00:27:10   You've got to do this.

00:27:11   Yeah.

00:27:12   That's a good point.

00:27:13   That if your philosophy is fewer but better features, one way to achieve that is by striving for perfectionism in their presentation and implementation.

00:27:26   That anything that could be fast, anything that could be fast, anything that is slow, the bottlenecks are identified, and it is made fast.

00:27:34   Anything that is supposed to look a certain way looks that way.

00:27:39   Anytime there's a weird edge case that is like, hey, if you turn off the toolbar, it lets you make the window a lot smaller than it does when the toolbar is visible.

00:27:50   And when you make the window really small, this, that, the other thing breaks layout-wise.

00:27:55   Like, oh, yeah, I didn't really – nobody really tested whatever the app is with the toolbar off.

00:28:01   Okay, but let's fix it, right?

00:28:03   Or let's set the minimum – one way or the other, let's fix it.

00:28:07   Let's either fix those layout bugs when you make the window too small when the toolbar is not visible, or let's make it so that the minimum window side is big enough that the – one way or the other, let's make it so that can't happen.

00:28:17   Exactly.

00:28:18   And reminding me – I've said this over the years to other developers who maybe they're asking for feedback about their UI.

00:28:25   And I often say, don't let users make your app look bad.

00:28:30   You know, and that's an example of that is don't let them – because if you don't do anything about it, you can – in a Mac app, you could resize the window to be the size of a postage stamp.

00:28:40   And it would just, like, squish all the stuff off the screen, and it looks bad.

00:28:45   So I always go by that idea that, similarly, don't let them set ugly colors on the UI, et cetera, et cetera.

00:28:52   Where was I going?

00:28:54   Don't let them make your app look bad.

00:28:57   Don't let them make your app look bad.

00:29:00   I think that's it.

00:29:01   Yeah.

00:29:02   At a certain point, that's the job of design.

00:29:06   It is making decisions for the user.

00:29:10   That is one simple argument of what design is.

00:29:14   It is making decisions and hopefully not making them arbitrarily but making them for good reason and with lots of debate and being able to defend them.

00:29:26   And there's just so many things in Tahoe that it just – it's like the – I think it was getting – the Mac interface was getting stretched thin over the – roughly the Apple Silicon era, which is sort of helped disguise the user interface decline because the hardware story suddenly got so much better where it was like, holy crap.

00:29:48   These machines, these machines, these laptops run forever, and they don't get hot, and the screens are better than ever.

00:29:54   And I have a four-year-old M1, whatever, and it still feels like brand new.

00:30:01   I don't feel – this hardware is amazing.

00:30:02   And everything – it was another, at an engineering level, extremely, almost bizarrely seamless experience for users in terms of all the Intel not yet updated for Apple Silicon apps that you still had to use just took a little bit longer to launch at first and because of Rosetta.

00:30:25   But then actually, because Apple Silicon was so fast and Rosetta was such an efficient translation layer that unless you had a really high-end Intel Mac, your Apple Silicon Mac ran your Intel compiled apps faster than the Intel Mac you switched from, which is amazing.

00:30:42   And all of that just sort of helped keep our attention away from the slight decline in user interface attention to detail.

00:30:51   And then I feel like as thin as it was stretched, Tahoe bursted open, and now there's garbage all over the interface.

00:31:00   You know what's funny?

00:31:01   I think the last time I was on this show was very shortly after we both got our M1 MacBook Pros because I remember I was just elated.

00:31:11   It was like the most positive version of me.

00:31:13   No, you were on in 2023, so I'd had mine for a while.

00:31:15   But we still might have been excited.

00:31:18   Okay, well, I was excited because I was like, everything is just fast and works.

00:31:23   And to your point about things like Rosetta working so well that it's possibly even better.

00:31:30   It's an example of its evidence that the people behind the product were asking themselves questions all along the way.

00:31:37   Well, what happens if X?

00:31:39   And then, well, this is annoying, but we're going to have to write a complete emulation layer for Intel apps.

00:31:47   Another company might say, no, we're just going to make them buy new versions of the apps.

00:31:53   Apple covered so many bases with this Apple Silicon transition.

00:31:58   And the fact that we were just elated by them represents the number of things they must have thought about, acknowledged the problem, and then fixed before it was even shipped.

00:32:11   And I'm going to switch to a little bit of hopefulness here because we're talking about how great of a hardware company Apple has shown itself to be, particularly in the last few years.

00:32:21   And I don't have to remind anybody listening to this show that we had the keyboard fiasco.

00:32:28   There was a time there where you could reasonably suspect Apple of having lost all of its Apple-ness with respect to hardware quality.

00:32:37   And so maybe we're in a software valley like the keyboard valley.

00:32:44   And especially, I'm optimistic, especially because of the change in leadership for design.

00:32:51   I can't imagine anybody at Apple being more embarrassed about the appearance of stuff in the OS than the leader of design.

00:33:02   I mean, the buck stops with them.

00:33:04   So their reputation is on the line.

00:33:06   Yeah.

00:33:08   And we don't still, you know, here we are six weeks or eight weeks after the big announcement that Alan Dye has been poached by Meta and took his inner circle with him.

00:33:19   And long-time user interface designer Steve LeMay is taking over the software design chief role at Apple.

00:33:28   And so many people still, even now that the news isn't so fresh, are still sending me notes that they're cautiously very optimistic because Steve LeMay is just exactly the right type of person.

00:33:40   The concern isn't about him personally.

00:33:43   The concern is whether anybody can write this ship and turn, maybe not turn it around 180 degrees, but it needs a significant course correction on, especially on macOS.

00:33:56   I would argue everywhere, but especially macOS.

00:34:00   And there is an institutional aversion at every institution to admit mistakes, right?

00:34:10   Nobody likes to.

00:34:11   And proud companies and teams are less likely to want to do it.

00:34:16   And the longer you go on a good, successful run, we're Apple.

00:34:21   We don't make bad mistakes.

00:34:23   And part of Apple's identity is that Apple ships terrific user interfaces, right?

00:34:30   It's foundational to the identity of the company, and it's why almost everybody I know who works there went to work there, at least obviously on the software side.

00:34:39   And on the hardware side, it's because they make great hardware that looks good and runs well and that you go there to make good stuff.

00:34:46   And it's good design, great design, the best design in the industry.

00:34:51   And so how do they get out of this without admitting mistake?

00:34:55   Is their pride going – this is the fear.

00:34:57   Is their pride going to get in the way of changing some of these things back or come up with – not necessarily exactly the way they were, but oh, now when you select a menu item or a menu in the menu bar,

00:35:11   the actual highlight of the view menu when you click view is visible as opposed to almost invisible.

00:35:18   And the shape looks like a rectangle, not an oval because why in the world is – why in the world would you break that, right?

00:35:27   Where the idea, the graphical appearance of a selected menu in the Mac menu bar is a rectangle, which is the pulled down menu, with a rectangular selected tab at the top left, which is the name of the menu that you've opened, right?

00:35:46   And now it is an iconic appearance, and there are – I can think of other apps that if you need an icon that represents a menu, that's what it looks like.

00:35:56   It looks like a little rectangle with a tab that's highlighted at the top left corner, and you could make that icon very small, and everybody would know, oh, that's a menu.

00:36:06   So you could make a little 32-point icon that any user would recognize, oh, that little icon, that's a menu.

00:36:15   But now the actual menus don't look like menus.

00:36:18   Yeah, right.

00:36:20   Well, part of my optimism about the next few years for Apple is, to your point about being willing to own the mistakes, actually, Alan Dye leaving is a perfect opportunity to sort of elide the shame.

00:36:36   It's like, well, we got a new design head in, and he just had some slight changes in mind for perfecting something that was already almost there.

00:36:47   You could say it in a way that was, like, complementary of the current team that implemented it, but also leaves room.

00:36:54   In other words, emphasizing more, like, a different take than a correction.

00:36:59   Yeah.

00:37:00   And that's my hope, too, is that they're obviously never going to mention him, right?

00:37:04   They're not going to come out.

00:37:05   Let's say come June, and all of our dreams come true, and when they introduce macOS 27 in the keynote in June, that it has – I don't know what they're going to call it.

00:37:20   They're not going to call it all new because last year was all new, but improved user interface.

00:37:25   And as they start showing right away screenshots and movies of macOS 27 in action, everybody watching the keynote live is excited as they watch.

00:37:36   Oh, they fixed the menu bar.

00:37:37   Oh, they fixed the window shapes.

00:37:38   Yes, yeah.

00:37:39   Oh, and you see these things flashing by.

00:37:41   And again, Apple did this recently.

00:37:43   They did it just last year when they talked about the Photos app, right?

00:37:46   Which a lot of people were very upset about the year before in iOS 18 when they changed the sort of – the tabs at the bottom of the Photos app to sort of steer you away from just looking at your library all the time and to show you these smart groups of things like friends and pets and places and stuff like that.

00:38:08   And people really didn't like it, and they addressed it.

00:38:12   I think it was Federighi himself or somebody who said, hey, we listened to you, we heard you, and in iOS 26, Photos has an improved layout that answers the question or solves the complaint that people had.

00:38:25   And I think it does, right?

00:38:26   I think people – for whatever other complaints they have about liquid glass, I do think the Photos app responded to the widespread criticism.

00:38:35   So there's a hundred of those things in macOS 26.

00:38:39   I don't think they're going to get them all next year, but if they start getting some of them, it would be a sign – we'll quickly be disappointed at all the ones that they don't get to.

00:38:49   But if it's clear that they're responding to some of them, we'll be like, ah, thank God.

00:38:55   They're listening.

00:38:56   Yes.

00:38:56   Sanity has prevailed.

00:38:58   And I think they can do it in a way that does, without ever mentioning him, does sort of put the blame for all of it on the guy who left.

00:39:08   Yep.

00:39:09   I mean, it's a great opportunity, actually.

00:39:11   And I think my complaint about the lack of cohesiveness, that might be a place for them to focus.

00:39:17   Because when you make things more cohesive, it inherently means leaving one of the choices aside.

00:39:25   So there's going to be some – if you talk about the release, like we have this great new UI, there's some things about the old UI that might have even been better still, and we just want to unify it all.

00:39:38   When you're unifying something, you have to choose, do you want the oval selection on the menu or do you want the rectangle?

00:39:44   And that's a time to sort of blame the fact that a choice has to be made, right?

00:39:50   A choice has to be made if you're going to unify things.

00:39:53   Maybe not with the menu bar, because obviously the menu bar is unified, but the way buttons behave from one app to another or the size of checkboxes, et cetera, et cetera.

00:40:02   Yeah.

00:40:03   All right.

00:40:04   Let me take a break here and thank our next sponsor, and it's our good friends at Notion.

00:40:09   Notion brings all your notes, docs, and projects into one connected space that just works.

00:40:16   It's seamless, flexible, powerful, and actually fun to use.

00:40:20   With AI built right in, you spend less time switching between tools and more time creating great work.

00:40:28   And now with Notion Agent, your AI doesn't just help with work.

00:40:32   It finishes it.

00:40:33   Notion Agent can do anything you can do in Notion.

00:40:38   So, like, this whole idea that we're moving towards a world where our system-level AI and Apple Intelligence or whatever they call it on Android, but you can just sort of talk to AI, and the AI can kind of take over and run your computer.

00:40:49   I know a lot of people are using Cloud Code for that across the system.

00:40:52   Notion Agent does that within Notion now.

00:40:56   And Notion is sort of – it's more than just notes.

00:40:59   It really is databases and documents and a lot more.

00:41:03   And you can do anything you can do.

00:41:05   You can direct Notion's internal agent to do for you in Notion.

00:41:10   It can tap into your Notion workspace.

00:41:12   It can go to the web.

00:41:13   It can use connected tools that you've connected to your Notion account like Slack and Google Drive to complete assigned actions that have been assigned to you.

00:41:21   You can focus on what you want to spend your time and delegate to almost another version of you that knows your style, your workflow, and your preferences.

00:41:32   It is really powerful.

00:41:33   Notion is used by over 50% of Fortune 500 companies and some of the fastest-growing companies in the world like OpenAI, Ramp, and Versil.

00:41:42   But it's also used by really small teams and individuals, too.

00:41:45   It is the type of app that scales everything from one-person companies all the way to the Fortune 500.

00:41:51   Try Notion, now with Notion Agent, at Notion, N-O-T-I-O-N, Notion.com slash talkshow.

00:42:00   That's all lowercase letters.

00:42:02   Don't put any capital letters in the URL or it's not going to work.

00:42:05   Notion.com slash talkshow to try your new AI teammate, Notion Agent, today.

00:42:11   And when you use that link, you're supporting the talkshow, Notion.com slash talkshow.

00:42:17   One of the things I wrote about recently, and I really felt like it was, if I do say so, a good point,

00:42:23   is combined with these growing rumors that John Ternus is internally to Apple, the intended successor to Tim Cook, is the CEO.

00:42:36   And it makes sense in a lot of ways, makes sense age-wise, that Ternus is only 51 or 52.

00:42:41   I forget when, I don't know when his birthday is.

00:42:43   But relatively young, and it's, you probably feel the same way, but as somebody in his early 50s.

00:42:51   I really like hearing somebody in their early 50s described as young.

00:42:55   Yeah, right?

00:42:56   Yeah.

00:42:57   This is our time.

00:42:57   Come on.

00:42:58   I'm not in my early 50s, by the way.

00:43:00   I'm in my early 50.

00:43:02   Yeah.

00:43:03   Early 50.

00:43:03   Okay.

00:43:04   I got you.

00:43:04   It's a singular still.

00:43:06   Yeah.

00:43:06   So you'd be a really young CEO for Apple.

00:43:08   That's right.

00:43:09   Give me a call.

00:43:10   But there is the, and again, and who even knows, right, like just to name somebody, just Federighi, right?

00:43:17   And I think in some scenario, Federighi's got to be on a short list of people who could be, right?

00:43:24   If there was some kind of accident and a couple of other executives were, let's just say, let's be nice about it or optimistic and say they were hospitalized for a while and incapacitated, including Tim Cook.

00:43:38   You could certainly imagine a scenario where Craig Federighi would be the interim CEO, without question.

00:43:43   He's got charisma.

00:43:44   He's got experience.

00:43:45   Maybe, though, Craig Federighi really does not want to go any higher than the head of software, that he's had conversations with Tim Cook, and Tim Cook has said, hey, what do you think?

00:43:56   And he's all, I don't want your job.

00:43:58   I want to stay right where I am.

00:44:00   I could really believe that about Craig Federighi.

00:44:02   I don't get the sense, but they're all poker players to some extent.

00:44:07   If there is some kind of elbow-throwing rivalry within their senior executive ranks for who would take over when Tim Cook does retire, we wouldn't know it.

00:44:19   But at a basic level, it makes sense that it would be Ternus, in addition to his age, in addition to the fact that he is good and charismatic on stage or in keynotes that aren't on stages anymore in public, but also just the basic truth that what part of Apple is running the best for the last five to ten years?

00:44:39   I think without question, it's their hardware, right?

00:44:42   So why not name the hardware guy as the CEO of all the areas of the company?

00:44:47   Hardware, software, services, marketing, right?

00:44:51   And marketing, JAWS' job at marketing is not just doing ads.

00:44:56   That's just a tiny part of it.

00:44:58   Product marketing at Apple is really more about the product than the marketing.

00:45:01   But the buck stops somewhere.

00:45:03   And, you know, the most talked about commercials from Apple in recent years have been ones that were problematic, right?

00:45:11   Yeah, the crushed musical instruments where they had to have an apology.

00:45:16   The Apple intelligence commercial that they had to pull from YouTube because it wasn't actually going, the feature that it advertised wasn't going to ship in the year that it came out.

00:45:27   And software quality, again, like we're talking about Rosetta and other things, engineering quality of software might well be better than ever, right?

00:45:38   Fewer bugs, fewer crashes, fewer data losses, more reliable syncing.

00:45:43   My uptime, I just looked at it because I'm writing, I'm working on my Apple report card for six colors.

00:45:49   My M1 Max MacBook Pro upstairs has an uptime right now of 73 days.

00:45:56   Like it's been 73 days since I've restarted that computer.

00:46:00   And you probably didn't restart it because of a problem.

00:46:02   You probably restarted it before.

00:46:04   When I upgraded from 15.7.1 to 15.7.2.

00:46:07   And I haven't upgraded to 15.7.3 yet because I've got a bunch of stuff open that I don't want to close to restart the computer.

00:46:14   73 days of uptime.

00:46:15   And there's nothing that's flaky or wrong.

00:46:18   I better restart over the weekend because X, Y, or Z is starting to flake out.

00:46:24   Incredible engineering, right?

00:46:26   But the one thing at the company that it's hard to even think of something to complain about is hardware across the board.

00:46:33   Yep.

00:46:34   No, it's absolutely.

00:46:36   You know, I'm thinking back to way, way, way back on the System 7, System 8 days of Mac OS, which is when I started working at Apple.

00:46:45   There was a lot of inner conflict at Apple at that time.

00:46:49   You reminded me of this by talking about Rosetta and maybe its relationship with hardware.

00:46:54   At that time, there were two competing engineering groups for what was the standard interface to the Mac.

00:47:03   In the sense that the PowerBook team, as opposed to the desktop iMac-type team, made its own software.

00:47:14   They made – I'm sure you remember, John.

00:47:16   Many of our listeners remember Control Strip.

00:47:19   I knew you were going to say that.

00:47:20   So Control Strip was like a rogue move by the PowerBook team.

00:47:25   And they were like – for a while back then, Control Strip was only available on PowerBooks.

00:47:30   Yeah, I know.

00:47:30   And it was maddening because it was so cool.

00:47:32   It was so cool, and I think there were hacks to get it to work on desktop Macs.

00:47:37   It probably didn't take much to make that happen.

00:47:39   But I remember being at Apple at the time, people were furious that the PowerBook team was exerting its own taste on the OS, essentially.

00:47:50   And adding a system-level feature.

00:47:53   I mean, how would you describe Control Strip for those who aren't as old as we are?

00:47:57   Or if they are as old, weren't using Macs at the time?

00:48:00   I mean, it's really – it's pretty easy to compare now to the status items at the upper right corner of the Mac screen.

00:48:08   But essentially like that, but positioned at the bottom left of the screen in a kind of – almost like a zipper-like expanding bar.

00:48:17   Yeah, like a little pull tab on it.

00:48:20   Yeah.

00:48:20   Yeah, and you could just click on the pull tab, and it would go away.

00:48:23   But just leave the tab in the corner of the screen, and you could click on it again, and it would snap out.

00:48:28   I think you could click and drag if you wanted to make it so instead of being full width, you could make it partial width, and then there would be arrows at the sides to kind of go through the icons.

00:48:36   Yeah, that sounds familiar.

00:48:37   Yeah, but the quickest way to describe it would be those – yeah, exactly.

00:48:41   Those menu items in the top right of your Mac menu bar were not – A, because displays were so much smaller back then.

00:48:51   There wasn't room for them to the right of the applications menus, right?

00:48:56   Because even back then, serious apps like QuarkXPress and Photoshop and FileMaker and BBEdit had a lot of menus.

00:49:03   So there wasn't a lot of space up there, and there was no good API.

00:49:07   There were always hacks, right?

00:49:09   There was the – what was it?

00:49:10   The precursor to FastScripts, OSA script from Leonard Rosenthal.

00:49:15   There were ways to put a system-wide script menu up there.

00:49:19   But the control strip was an official sanctioned way to get those type of status icons in a strip.

00:49:28   And I think it was a more elegant solution than what we've been living with in Mac OS X for 20 years where we stick those icons in the menu bar because they're not really menus, right?

00:49:39   They're something else.

00:49:40   Like the menu should be for the commands of the system and the commands of the application.

00:49:46   It shouldn't be a place for many apps, and the control strip was a better place for that.

00:49:51   And I don't know.

00:49:53   It just looked cool, but it was incredibly frustrating that it was only for PowerBooks.

00:49:58   And that's the sort of thing Apple does once in a while that's like, oh, come on.

00:50:04   Why don't you – why doesn't everybody get that?

00:50:07   Put it in perspective, I think it's possible that the reason it was only for PowerBooks was because the desktop team refused.

00:50:15   And so eventually they won.

00:50:18   The PowerBook team won.

00:50:20   Like it became a standard element of Mac OS.

00:50:23   But there's some funny examples like that of things that just weaseled their way in because of a hardware decision, not a software decision.

00:50:32   I hope – I don't even know.

00:50:34   I doubt he listens to every episode.

00:50:36   I hope the statute of limitations on off-the-record comments have expired on it.

00:50:43   But Phil Schiller told me a story once that he prototyped the first version of Control Strip using – what was the macro media tool that you could use to sort of prototype software?

00:50:54   Oh, it's not coming to me.

00:50:57   It was like a precursor to Flash because it was only on the desktop.

00:51:00   But he was on an airplane on a PowerBook and was like mad and frustrated about not being able to get – I don't know what it would have been because Wi-Fi didn't exist yet.

00:51:12   But there was something like the volume or something.

00:51:15   And so on the plane, he prototyped the first version of the Control Strip.

00:51:20   And then he got off the plane and it came from there.

00:51:22   I don't know what it looked like.

00:51:23   But I thought that was a cool story.

00:51:25   But it was specifically because he was on a PowerBook and couldn't get access to some of these system controls.

00:51:32   That's a good example of what you said earlier about marketing not just being ads.

00:51:37   Absolutely.

00:51:37   That was him marketing based on his own experience.

00:51:39   Right.

00:51:39   But it really was – there was – and you were there.

00:51:43   I was outside.

00:51:44   But, I mean, it was infamous just reading about the company that the politics inside the company were bad.

00:51:51   Yeah.

00:51:52   There were fiefdoms.

00:51:54   Totally, yeah.

00:51:54   And there wasn't – there's that famous cartoon of the org charts of Silicon Valley's biggest companies.

00:52:00   And it was drawn when Steve Jobs was still alive.

00:52:02   But Apple's org chart was just – it's just a circle around one person in the middle.

00:52:07   And everything goes to that one person.

00:52:10   You can't really recommend that structure, the Steve Jobs executive leadership style in business school because it sort of hinges upon having a Steve Jobs intellect and level of taste and assertiveness and confidence and charisma and you name it.

00:52:30   You name all the things that were great about Steve Jobs and the person who does that to organize a major company like that kind of has to be all of them.

00:52:38   But it worked and it did solve – it solved that little fiefdom thing where desktop Macs had a different interface to those things and were missing a cool feature that was PowerBook only.

00:52:50   And it's just so spiteful because everybody – and I remember being at the college newspaper where, of course, we had all desktops.

00:52:57   I think we did have one PowerBook that was like so – because they were so expensive at the time.

00:53:02   They were so much more expensive than even powerful desktop Macs.

00:53:05   So we had one PowerBook that like if a reporter was going to the student council meeting and wanted to take notes on the computer, they could sign it out and take it.

00:53:15   But all the other computers were desktops and we all wanted that feature.

00:53:18   And I think you're right.

00:53:19   I kind of remember there was some kind of hack to enable it or something.

00:53:22   And, of course, we did it.

00:53:24   Yeah, of course.

00:53:26   Yeah.

00:53:27   Well, I was thinking when you said that about the kind of like circle with one person around it, that kind of harkens back a little bit to me to the concern about attention to detail and perfectionism within Apple.

00:53:41   Because it strikes me that a company like Apple can thrive even with a relative lack of sort of top-down control management because everybody who works at the company has.

00:53:56   Inherited the taste of the company, you know what I mean?

00:54:01   If you go into Apple with Apple taste, you don't need a manager's manager's manager telling you like how to make a control strip or whatever.

00:54:10   You have it built in.

00:54:11   And I think a lot of times since I left Apple, people ask me, what are Apple interviews like?

00:54:17   And, well, there's no answer to that because every single group in every part of the company does it a little differently.

00:54:24   And by contrast, Google's interviews are like famously the same.

00:54:28   At least they used to be.

00:54:29   But it was like at Google, they have a top-down – like every interview will go like this.

00:54:35   And at Apple, I don't think they're – the only thing – granted, this was 20 years ago, 25 years ago.

00:54:42   But the only thing I had to do in preparation for interviewing a candidate for my first time was either watch or go to some kind of like seminar where they teach you how to not be discriminatory.

00:54:55   And I think it was worthwhile.

00:54:56   I probably would have made mistakes if it wasn't for that.

00:54:59   But apart from that, the only guidance about interviewing came from my manager.

00:55:04   And that's how it worked as far as I know all throughout Apple.

00:55:07   It was all – how do we hire the way that the specific manager wants to hire?

00:55:12   The other thing too – and this is where you really have to wonder – I have to wonder what the hell – how did this break down?

00:55:23   Is it just used to be that you might disagree with the choice.

00:55:31   Like I said earlier, hey, should the grip strip on the control – or should the control strip buttons look three-dimensional or should they look flat?

00:55:39   And they looked three-dimensional.

00:55:41   They were – it was in the three-dimensional buttons looked like things – actual buttons you could press and go in.

00:55:46   But if somebody said, well, maybe these should look flat because the menu bar is still flat.

00:55:51   The menu bar doesn't look three-dimensional in any way.

00:55:53   And the control strip should be more like a portable version of the menu bar.

00:55:57   It should be flat.

00:55:58   And somebody said flat.

00:55:59   And somebody said they should look 3D.

00:56:00   And 3D won.

00:56:01   But once 3D was the choice, which you might disagree with, there was an agreement of what looks properly 3D.

00:56:09   What looks pixel perfect as 3D?

00:56:12   And if there was like a gap between some of the buttons but not some of the other buttons in the control strip, everybody would see it.

00:56:19   And it would be like if you're the one working on it and you're showing it to me, you're like, I know there's a gap there.

00:56:24   I'm going to fix that.

00:56:25   Right?

00:56:26   It wouldn't be like, ah, who cares?

00:56:29   So what?

00:56:29   There's a gap between some of them.

00:56:31   Right?

00:56:32   So what?

00:56:33   Who cares if sometimes you're scrolling a sidebar in Tahoe and the search field content goes underneath it, rendering both the search field and the content underneath completely inscrutable?

00:56:45   Who cares?

00:56:46   Right?

00:56:47   Somehow something broke where you can point to this and say, well, here's a box where I can still click on it and I can type what I'm searching for, but I cannot read the letters I'm typing as I'm typing the search query because of this transparency and layering.

00:57:05   And somehow they're like, well, we'll still ship it.

00:57:09   And that is so far beyond the sort of, hey, this is misaligned by two pixels.

00:57:17   That used to be like a priority.

00:57:20   Well, we absolutely cannot ship this off by two pixels.

00:57:24   It just used to be, and that thinking permeated the whole indie Mac culture too, right?

00:57:33   All the indie Mac developers I know were, if anything, more persnickety.

00:57:37   I remember one time when I was at Bare Bones, probably towards the end, like in 2002, and we were working on Yojimbo, which is still around.

00:57:48   It's sort of a notes slash bookmarks, passwords, junk drawer app.

00:57:55   And in office, and I went down the hallway to our mutual friend, Jim Correa, who was at Bare Bones at the time, and I could tell that the sidebar was one pixel too low, one pixel.

00:58:06   And he looked at it, he's like, oh, yeah.

00:58:09   And he knew why.

00:58:10   He's like, oh, yeah.

00:58:11   You wouldn't understand the explanation completely, but it was Bare Bones' first Coco app, and Jim and its colleague Steve Kalkorf had thrown themselves and learned Coco, and they'd made this sort of use every aspect of Coco app.

00:58:24   And it was like, it's a subclass of something, and when you subclass it, it's going to go down.

00:58:29   And at the moment, and it was like, let's say it was a Wednesday afternoon, he was like, I don't think there's anything we can do about that.

00:58:35   And I was like, all right.

00:58:36   And it was one pixel, the whole sidebar was one pixel too low.

00:58:39   And then the next morning, Jim called me in, and he was like, I couldn't stop thinking about it, and here, look.

00:58:45   And he fixed it, right?

00:58:46   Of course he did, yes.

00:58:48   He lost sleep that night because it bothered him.

00:58:53   And while he couldn't think of it, and it was only literally just one pixel, and it didn't look wrong.

00:58:59   I just knew that it didn't used to look like that, and it shouldn't look like that.

00:59:02   But then once I, and Jim hadn't even noticed, but then once I pointed it out, he couldn't sleep until he fixed it.

00:59:07   That's amazing.

00:59:08   Everything was like that everywhere across macOS, right?

00:59:12   And then you'd get to an app like, oh, Microsoft came out with a new version of Word, and it's not like that at all.

00:59:19   And that's why Mac users kind of held their nose to use Microsoft Office at the time, because it's like, this was the difference between Mac-like and un-Mac-like software.

00:59:29   It's not just that some of the idioms aren't right.

00:59:33   It's like there's just no attention to detail.

00:59:36   And the attention to detail is the one thing that, no matter if you disagree, should this button be a rectangle or an oval?

00:59:45   You might think one way.

00:59:46   You might think the other way.

00:59:47   But whichever way it's decided, that rectangle or that oval should be pixel perfect.

00:59:52   Yeah.

00:59:52   You're reminding me that I've had this experience.

00:59:56   So I started at Apple as a QA engineer.

00:59:59   So I was – my job was to look for problems.

01:00:01   It was great.

01:00:03   I was already a fussy person.

01:00:04   I just look for problems and report them.

01:00:06   And the key was you were considered to be doing a good job the more things you noticed and reported.

01:00:12   And so –

01:00:15   Right.

01:00:16   It wasn't like, hey, why are you reporting all these things?

01:00:19   Right, right, right.

01:00:20   You should shut up.

01:00:20   It was the opposite, right?

01:00:22   It was the opposite.

01:00:23   You sent bug reports in like this word is one pixel off.

01:00:28   And what I've learned over the years since then is I'm pretty good about reporting bugs both to Apple and to other engineers, other developers of apps.

01:00:37   There are some people, some developers I don't report bugs to because they have demonstrated their lack of attention to detail by responding to my bug reports with like dismissive, well, who cares?

01:00:52   Or why does that matter?

01:00:54   Who cares if the sidebar is one pixel too low?

01:00:56   And other developers though, it's kind of like when you – like I know when I read Daring Fireball, I can feel good about reporting any typo or grammatical issue because I know you strive for perfection in that department.

01:01:12   And there are other people I have come to realize the last thing you want to do is report a typo to them.

01:01:19   I take it as a personal attack and it's like, okay, well, I was just trying to make you better.

01:01:24   But it's the same thing with software.

01:01:26   And I think there's some people at Apple who are saying who cares if the pixels one stays off.

01:01:31   Or that they're mad, right, that they're offended.

01:01:32   They're like, why are you picking on me?

01:01:34   They feel attacked.

01:01:36   They're taking it personally.

01:01:37   And I've brought this up multiple times.

01:01:40   But one of the most interesting classes – and there's like this one thing I learned from it.

01:01:45   But I took a screenwriting class in college at Drexel.

01:01:49   And the guy who taught it, his name was Martin Donoff.

01:01:52   And he was a great professor.

01:01:54   And he had worked on a whole bunch of shows, TV shows, over the decades.

01:01:58   He had been a – maybe not the showrunner, but he was one of the writing people for ALF, which was a big hit.

01:02:07   And then he was the showrunner on the ALF Saturday morning cartoon.

01:02:12   And he'd worked on other things like PBS, like Masterpiece Theater, like real serious adult drama.

01:02:16   He had no embarrassment about having worked on ALF.

01:02:19   It was a big hit.

01:02:20   I think he made a lot of money on it.

01:02:22   I hope he did.

01:02:22   But the lesson that he learned – we would write – every week, it was a once-a-week class.

01:02:26   And every week, you have to write like a scene for something.

01:02:29   It might be a comedy.

01:02:30   It might be a drama.

01:02:31   Whatever.

01:02:31   You had to write a one-page screenplay for a scene.

01:02:34   And then everybody before the class, there'd be copies that you could get of everybody else's.

01:02:40   And I guess we got them electronically, like by email or something.

01:02:43   And you were expected to read everybody else in the class's scene.

01:02:47   And then there'd be a group discussion where we would critique each other's scenes because you were supposed to have read them before the class started.

01:02:53   And the golden rule was you criticized the work, not the person.

01:02:58   So you say, hey, this line of dialogue doesn't sound right for the character.

01:03:05   The dialogue, the character, not you, Daniel, the person who wrote this scene.

01:03:11   You shouldn't have had the character say that.

01:03:13   You're making the same point, but the way you phrase it is – and he would interrupt.

01:03:19   And people weren't taking it personally and nobody in the class was mean.

01:03:22   But if anybody phrased it like you shouldn't have – you should have this character say more here.

01:03:28   He would interrupt and he'd say, not the you, the character.

01:03:32   The character should say more here.

01:03:33   And we all got it.

01:03:34   But it was a brilliant breakthrough because it takes that personal aspect of it.

01:03:38   But there are still people who, if you phrase your criticisms this way, right, that you focus on the work, they still take it personally, right?

01:03:47   And that's what you're talking about.

01:03:48   And I hate it.

01:03:49   I love typo reports from readers for my writing.

01:03:53   Certainly if it's like a spelling mistake or, God forbid, I get somebody's – spell somebody's name wrong, which is like the worst.

01:04:00   But even for things that aren't errors but like, hey, I don't know.

01:04:04   This sentence confused me, right?

01:04:06   I love feedback like that because nine times out of ten, if I reread the sentence, I'm like, oh, I do see that that's not – it's sort of confusing.

01:04:15   I know what I meant because I wrote it, but I can see how that sentence is ambiguous.

01:04:20   And I love feedback like that because I don't have an editor.

01:04:24   Right.

01:04:25   But – and the people like you who will frequently report typos, you don't preface it with, hey, sorry, but I think you spelled this word wrong.

01:04:34   But when I get like first-time reports from random readers, they're always very apologetic.

01:04:39   Hey, sorry, I think you spelled this word wrong or something like that or I think you're missing a punctuation mark here.

01:04:45   And it's very obvious that I'm missing a punctuation mark because the paragraph ends without a period, right?

01:04:50   Yeah.

01:04:51   It's like – but they're so cautious because they know other people are offended.

01:04:54   And I try to always write back to those emails and say, thank you.

01:04:57   No need to apologize.

01:04:58   I love getting typo reports.

01:05:00   I fixed it.

01:05:00   Thanks to you.

01:05:01   And – but people feel like that.

01:05:04   And I do think that at Apple, it was – and the indies – the best indie software was all from people.

01:05:11   They might be annoyed that there's a bug to fix, but they're not annoyed at you for reporting it, right?

01:05:17   Right, right, right.

01:05:18   It's – there's a difference.

01:05:20   Yeah, I still have to often – I mean it's funny actually.

01:05:24   Our mutual friend, Paul Kafasis, he's an active user.

01:05:27   Yeah, it's this guy from Boston.

01:05:29   He's an active user of my crossword app for iOS.

01:05:33   And he sends me lots of bug reports.

01:05:35   And it's great.

01:05:36   And even him, it just so happens today, he sent me like a litany of reports.

01:05:42   And he's like, sorry for the – whatever – for the glut of reports.

01:05:45   I'm like, no, no, no.

01:05:46   I'd rather have you report it to me than everybody in the public report it to me.

01:05:51   I – he reports a lot of typos to me as well.

01:05:56   I bet, yeah.

01:05:57   You could tell when he's –

01:05:58   He's got a good critical eye.

01:06:00   He's actually – and it's funny you mentioned Paul, but he's actually very good at pointing out ambiguous sentences to me.

01:06:06   Not just typos, which he will, but he's one of my top reporters of, hey, maybe you should rephrase this sentence just because it's not clear or could be read two different ways.

01:06:17   It's such a benefit, but you really have to have the right mindset.

01:06:20   And I feel like something has been lost within Apple.

01:06:23   And then the other – the problem is it cascades because once it kind of gets out of control, like the way you achieve near pixel perfection in software is to get near pixel perfection in software and then try to stay there.

01:06:39   Like little – something changed, you upgrade to a new version of OS, and now this button is way too close to the bottom of the dialogue.

01:06:49   Or, you know, it's like here's what happened.

01:06:51   I upgraded the OS, and now this dialogue box looks wrong.

01:06:55   It sticks out because there aren't that many mistakes.

01:07:00   But now it's like where do you start, right?

01:07:03   It's like, you know, like a wild dog ran through your house and tore up all the furniture and ripped open a trash bag.

01:07:12   That scene in Christmas Vacation where the dog goes nuts in the kitchen and tears everything apart.

01:07:18   Where do you start cleaning it up?

01:07:20   It's got to start somewhere, right?

01:07:22   Right.

01:07:23   The Journey of a Thousand Miles.

01:07:25   Well, let me take another break here and thank our next sponsor of the show.

01:07:28   And it is, once again, Shocker, our good friends at Squarespace.

01:07:32   This podcast is sponsored, once again, by Squarespace.

01:07:35   And Squarespace is the all-in-one website platform designed to help you stand out and succeed online.

01:07:43   Whether you're just starting out or scaling your business, Squarespace gives you everything you need.

01:07:48   Or someone you know who comes to you for help as, like, the resident nerd in their social circle.

01:07:54   Send them to Squarespace and send them with my code.

01:07:57   They give you everything you need to claim.

01:07:59   Everything from your domain to setting up the way your website looks.

01:08:03   To tweaking the way it looks.

01:08:05   To updating the content on the website on a regular basis.

01:08:09   Like, if it's got, like, a blog or you just have little news items or whatever.

01:08:14   You update the website through Squarespace, too.

01:08:17   Not just create it and then use something else.

01:08:20   All of it you could do in one place.

01:08:23   They have so much new stuff.

01:08:26   One of the new things they have is analytics, which they've improved greatly in recent years.

01:08:31   The built-in analytics at Squarespace websites is one of the best analytics dashboards I've ever seen.

01:08:37   Where it's not minimal.

01:08:39   It has a lot of information.

01:08:41   But it is unlike so many other analytics packages.

01:08:45   Google Analytics, I am looking in your direction.

01:08:47   It is cohesively designed in a way that you don't have to, like, read a manual to understand the analytics.

01:08:55   You just look at it and it's as clear as, hopefully, the dashboard of your car or something like that.

01:09:00   So, selling content, selling time, selling products, whatever it is you're hoping to sell, whether it's something virtual like software or whether it's something physical like physical goods or whether it is your time because you're a consultant or a trainer or something like that.

01:09:16   You can set up your Squarespace website to handle that sort of stuff.

01:09:20   And one of the other new features they've got is Squarespace invoices where it ties together where if whatever it is that you're selling needs an invoice, the invoice you can do through Squarespace, too.

01:09:31   Your customers get the invoice, pay the invoice, and it comes to you, including the payment processing, all through the Squarespace platform.

01:09:37   Where do you go to find out more?

01:09:40   Go to squarespace.com slash talk show.

01:09:45   Head there, squarespace.com slash talk show, and save 10% off your first purchase of a website or domain.

01:09:52   I want to circle back before I guess we can move on to something other than bitching about Tahoe.

01:09:58   But the other thing when I was talking about Ternus, and yeah, of course, he's the guy who they're talking about as the next CEO of the company because he runs hardware, and hardware's the best thing going on.

01:10:07   The other point I made on Daring Fireball, and I think it's really key, is that Apple's hardware designers are absolutely unafraid to keep things the same year after year if they don't have a better idea.

01:10:21   That they've gotten MacBooks to a point where this is the way a MacBook should look.

01:10:28   It should have round corners.

01:10:29   It should have a trackpad that is a very large part of the palm rest, but not the whole palm rest.

01:10:34   It should have a keyboard that looks like this.

01:10:38   And every couple of years, there's a major redesign, and even the major redesigns, like from the Intel, the last era Intel MacBooks to the first redesigned Apple Silicon era ones, they don't look altogether new, right?

01:10:54   They're just sort of thinner, flatter.

01:10:56   The MacBook Air is no longer wedge-shaped.

01:10:58   It's just flat, but thin.

01:11:01   And then the M3 MacBook Air looks like the M2, and the M4 MacBook Air looks like the M3, which looks just like the M2, and maybe they tweak the colors.

01:11:09   But they have this confidence to keep things the same if they know that they're still the best.

01:11:15   And that's where I feel like things under Alan Dye, really.

01:11:20   That's the bottom line of how did this happen.

01:11:22   I think, and I know it sounds like a harsh word, but I really do think Alan Dye is a fraud.

01:11:27   I do not think he is a good user interface designer at all.

01:11:31   I actually don't know that he's a very good graphic designer at all, which is what he came from.

01:11:36   But he's definitely not a good user interface or interaction designer.

01:11:41   And his skills lie elsewhere.

01:11:45   The politics of navigating a company like Apple and pleasing the people above him on the org chart.

01:11:51   And managing, to some degree, the enormous design talent in the ranks of the rank-and-file designers at Apple, who are still incredibly talented.

01:12:00   And that's why, overall, the design of Apple's software is still good overall, looking at the rest of the industry.

01:12:07   But there's a lack of confidence that comes from being a fraud, where you feel like you've got to keep new, even if new isn't better.

01:12:16   And I think if, like, over the last 10 years, the Mac OS interface had changed as little as the Mac hardware has changed over the same 10 years,

01:12:27   this whole conversation would not be happening.

01:12:30   And you probably wouldn't even be on the show right now.

01:12:34   Yeah.

01:12:35   Well, that's a really good point.

01:12:36   When you talk about that, it reminds me, people often compare Apple to Porsche for the way that, like, the, like, Roadster or whatever would look the same.

01:12:45   9-11, famously.

01:12:47   Oh, the 9-11.

01:12:48   Okay.

01:12:49   So, that's if it's not broken, don't fix it type of thinking.

01:12:52   And you're right.

01:12:53   The MacBook, I'm looking down at my MacBook right now.

01:12:56   My first Intel Mac practically looked just like this, just thicker.

01:13:01   It was like, things have not changed in decades in a significant way, except for they, of course, get thinner.

01:13:07   And I actually was just remembering the touch bar is actually a really good metaphor or comparison for the control strip.

01:13:13   It was the one we were trying to kind of explain.

01:13:15   So, but that's the example of where Apple took a bold step, but then recognized that it was a mistake, and they pulled back.

01:13:23   So, yeah, I don't know.

01:13:26   I feel kind of like you're right.

01:13:29   Like, we should know by now whether a button should be rounded or square, what size it should be, whether it should look 3D or not, whether it should turn to glass when you click it.

01:13:41   Some of these things we should just know as facts, and they shouldn't be kind of like taste-based, I think.

01:13:49   And they shouldn't be going for novelty every few years, right?

01:13:53   The menu bar should still look like the menu bar did 10 years ago, where it doesn't have to be white, but it should be in light mode.

01:14:00   It should be a close to white menu bar with black text, and when you click it, a rectangle around the menu name is highlighted in a very vibrant way, and the rectangular menu drops down beneath it.

01:14:14   That's the way it should look.

01:14:16   Like, and if it looked, it was a solved problem, leave it solved, and put your attention and your urge to mark your territory like a fucking dog on new stuff, right?

01:14:27   It's this, I think, a deep insecurity in Alan Dine and her team led them to want to put their mark on things that had come before them, and they've just, I really don't mean this in any hyperbolic way.

01:14:43   It's if they've vandalized it, they really have, and I really do think it's fundamentally a lack of confidence.

01:14:50   It's an unconfident designers who insist on new, because new, at least, you can say, well, you just, your complaint is just that you don't like new stuff, right?

01:15:00   You're a fuddy-duddy, and you want things to stay the same, and you don't get it.

01:15:04   The software needs to keep changing.

01:15:06   It is sort of a defense mechanism, and it's kind of hard to argue with that.

01:15:10   It's a sophistry, I don't know if that's an adjective or not, but it's a way of debating that sort of avoids debating the actual point of, put aside whether it's new or old, is this a good design for this?

01:15:23   And is there a reason to break people's expectations, right?

01:15:27   The shape of the windows, the way windows look, whether it is obvious at a glance, instantly, if you see four windows on screen at once, which one is the active window?

01:15:38   That should be obvious.

01:15:40   That is just part of using a computer, whether it's light mode or dark mode or Windows or Mac or a version of Linux or iPad with multiple windows open.

01:15:50   If there are multiple windows, and it is important because keyboard focus can only go to one place at a time, which one is my typing going to go to?

01:15:58   That should be very obvious.

01:16:00   And instead of making it more obvious, and changes to it should be in the favor of making it more obvious in a way that looks cool, which one is front most, not less.

01:16:10   And they just keep making – it's like they're bleaching the whole OS and making it ever more inscrutable, which one is active.

01:16:20   I think – getting back to my optimism a little bit, I think what you're describing is a situation where there's probably like a collective sigh of relief among many people who worked for Alan Dye, whose ideas or priorities may have been suppressed over the past several years.

01:16:40   There's got to – under any leader, there are people who are suffering with disagreements with the leadership.

01:16:47   And I kind of envision now Apple's design department as having like – if it's a garden, like somebody has come in and raked all of the leaves out, and everything can grow now.

01:17:01   And that's my really kind of like rosy, rose glasses view of hopefully the next few years.

01:17:07   Not only is a leader who has been criticized gone, it's almost poetic that he went to one of Apple's biggest competitors slash enemies.

01:17:18   I would say the biggest, right?

01:17:20   In a way.

01:17:20   Not that they compete directly, but they kind of are starting to with the glasses, right?

01:17:25   Where Meta is perhaps more – if you name just the big five or six companies, Microsoft, Google, Amazon – who else should we put in the list?

01:17:35   NVIDIA, I guess, deserves to be there.

01:17:37   But NVIDIA doesn't make their own devices, right?

01:17:40   The one company amongst all of those that is making the sort of devices that Apple makes is Meta with the glasses and with the headsets and stuff like that.

01:17:50   But I think culturally it's more than that, right?

01:17:52   Meta is the company where Tim Cook, in an interview when they asked him, hey, what would you have done if you were Mark Zuckerberg and you got into the – I think it was the Cambridge Analytica thing.

01:18:02   I don't know, whatever.

01:18:02   It was some kind of privacy-related scandal.

01:18:05   And he's – I wouldn't have been in that situation in the first place.

01:18:09   Stone cold.

01:18:10   Yeah.

01:18:12   And there is a culture within Apple, certainly once you reach senior levels, that these are people, largely men but not all men.

01:18:19   Someone like Deirdre O'Brien has been at Apple her whole career, largest part of her career, that you get to that level and Apple is your career.

01:18:28   And you retire from Apple and you bleed six colors, right?

01:18:33   That's something people in Apple talk about.

01:18:35   And it's – you could say, oh, that sounds like a cult and maybe it could, but it's really just more like a camaraderie, right?

01:18:43   And a like-mindedness that grows the longer you're there.

01:18:46   It is extremely unusual for somebody at the senior vice president level at Apple to leave for another company.

01:18:53   I mean I'm trying to think of somebody before Alan Dye who did.

01:18:56   It's very, very rare.

01:18:58   Right, and just to name one, I'll bet Craig Federighi gets a lot of emails from recruiters looking for CEOs, right?

01:19:06   Sure, yeah.

01:19:07   I'll bet if like you looked at like the last 20 years of Craig Federighi's email, how many times he's been offered serious job offers as the CEO of another tech company.

01:19:18   It's probably ridiculously large and I don't know that he ever even reads them, right?

01:19:23   That they just go right to trash because he's not looking for that.

01:19:27   But of all the companies for someone to actually leave for to be meta, to me it just proves the point that Alan Dye never belonged at Apple in the first place.

01:19:35   What I was thinking as you were describing it is they are not business competitors as much as philosophical competitors.

01:19:43   Like if you look at the Venn diagram of overlapping philosophical beliefs with Apple, you can find some that are shared with Google and are shared with Microsoft.

01:19:55   And Amazon in certain ways, right?

01:19:57   Sure, yep.

01:19:58   But there are very few that are shared with Facebook or Meta.

01:20:01   Yeah, yeah, right.

01:20:03   There was a guy, I don't know if I linked it at Daring Fireball or not.

01:20:07   I couldn't, I don't remember, but it doesn't matter.

01:20:10   But at the last AWS conference, a couple months ago, sometime like late 2025, Amazon had their big AWS conference.

01:20:18   And somebody from Apple came and spoke for like a 10 or 15 minute stint in the keynote.

01:20:25   The way that Apple, when they had onstage keynotes at WWDC, would bring in somebody from Intel or from Microsoft to talk about Office and the new APIs or whatever.

01:20:36   Here's Jerry Yang from Yahoo to talk about doing the weather app for the original iPhone or something like that.

01:20:42   And it was somebody from Apple who came to talk about how much the App Store uses AWS behind the scenes and how running Swift code on the server at AWS made something faster.

01:20:57   Like way faster and fixed a whole bunch of bugs because Swift is memory safe and all this and it just runs great on AWS.

01:21:04   So they were there to talk about some Apple technology like running Swift on the server, but it was at an AWS conference.

01:21:10   And you think, hey, when's the last time you saw someone from Apple speaking at somebody else's conference?

01:21:15   Well, there's some overlap with Amazon where they go, guess whose developer conference I don't think we're ever going to see an Apple person speak at.

01:21:23   Yeah, exactly.

01:21:23   No, you're right.

01:21:25   I might soil my pants if I found somebody said, holy shit, somebody from Apple spoke at Meta Connect or whatever the hell the name of their developer conference is.

01:21:34   Right.

01:21:34   Yeah.

01:21:35   It would be shocking.

01:21:37   Whereas with the Amazon, it was like, oh, yeah, of course, it kind of makes sense that Apple is leaning on AWS for a lot of the server infrastructure for the App Store rather than building it out on her own.

01:21:48   Before we move on, the other one I really do want to talk about is the fucking icons in the menus.

01:21:55   Oh, yeah.

01:21:56   Okay.

01:21:57   Yeah.

01:21:57   I this is one of those things.

01:22:00   The more I look at it, because the more I'm using Tahoe here while I'm complaining about it, the more angry.

01:22:07   Oh, it makes me so angry.

01:22:09   This and the fact that like old versions of the Hig specifically said, hey, don't just add icons to every menu item that just looks disorganized and confusing and makes it hard to read the menu.

01:22:20   And then they have like examples.

01:22:22   And there's one from like 1992.

01:22:24   And then there's an updated one from, I don't know, 2004 with a totally different art style to match the look of the Mac at the time.

01:22:31   But the don't do this.

01:22:33   It looks exactly like screenshots from Tahoe.

01:22:36   Yeah.

01:22:38   I heard a good amount of conversation about this.

01:22:40   Maybe you did, too, on the ATP.

01:22:42   And something I think John Syracuse pointed out that even now, apparently, the guidance says not everything needs an icon.

01:22:53   Right.

01:22:53   But I'm looking right now at Safari and damn near every menu item has an icon, even to the point where they use a generic globe for all the like sites you visited that don't have a favicon.

01:23:07   Yeah.

01:23:08   Right.

01:23:08   And then it's funny, though.

01:23:10   I think that's always been the case for the history menu, though, because they do show the five fav icons for ones that do.

01:23:15   So now you're looking at the one that always.

01:23:17   That makes sense.

01:23:19   Yep.

01:23:20   One of the best examples, and the Hig always said this, too, that if adding an icon can clarify the meaning, then, yes, add it.

01:23:31   And it really is that it's the lesson from the Incredibles movies from Pixar, where they outlaw being a superhero.

01:23:39   And it's like if everybody's equally special, then nothing is special.

01:23:43   If every single menu item has an icon, then none of the icons are special.

01:23:49   And they're just you can't come up with icons for some of these things.

01:23:53   Right.

01:23:53   What is the icon for show reader?

01:23:55   It's a rectangle with a little, you know, it looks like a tiny little text document because I guess you're reading text.

01:24:02   And there were two really good blog posts about it.

01:24:05   There was the Tonsky one that I linked to both of them.

01:24:09   Yeah.

01:24:09   It's hard to justify Tahoe icons by Nikita Prokopov.

01:24:15   Hard to justify.

01:24:15   Oh, just.

01:24:16   Okay.

01:24:17   Nikita Prokopov.

01:24:19   And I'll put links to these back in the show notes.

01:24:21   But he pointed out how if you go to two apps that have the same menu command from Apple, both from Apple, they use different icons.

01:24:28   Like it was maybe partially a rush job, but partially like, well, some of these commands, they're so abstract where it's like, how do you come up with that?

01:24:40   What would like, or I'm looking at Mars at it right now because I was looking up those blog posts.

01:24:45   So you've got in the view menu for which things you can view in a window, like title field, slug field, link field, keywords field.

01:24:55   Well, what's the difference between a title field and a link field, right?

01:24:58   Right.

01:24:59   And it's like, well, a link, you could use like the chain link link thing.

01:25:04   And the title, you can make like a capital T or something.

01:25:08   But then all of a sudden, they're both just text fields.

01:25:11   So the icons don't really look as similar as the, you know, why?

01:25:15   This was a solved problem.

01:25:17   You know what people do with menus?

01:25:19   They read them.

01:25:20   They read menus.

01:25:21   When you go to a restaurant, you read.

01:25:24   And a menu at a restaurant is a perfect example.

01:25:27   And I bet you know this firsthand is a lot of times you go to a restaurant, they will have an icon to represent vegetarian dishes.

01:25:38   And that is a very useful icon.

01:25:41   And you put it next to certain items on the menu, which have no animal products in the making of the meal.

01:25:49   And if you are a vegetarian, that is a very useful, very useful icon to have.

01:25:55   But if every menu item on the menu has its own icon to represent that dish, then having one special icon to mark some of them for a very specific reason that is very important to some people, all of a sudden you can't see it.

01:26:13   Right.

01:26:14   And you've got to, like, run your finger down the page and kind of look for the icon that you're looking for.

01:26:18   In the window menu, move and resize, and this has been true for a couple versions, you can go and you can, like, it's a system-wide feature.

01:26:26   Go to window, move and resize, and you can move a window to the left.

01:26:31   Or you can move it to the right, and it'll take up half the screen on the right.

01:26:35   Well, those commands have had icons on macOS for a couple of years now because the icons show you exactly what the window will look like after you use it.

01:26:45   And it's actually better to use the icons for what you want to achieve than to read, right?

01:26:52   Another one that didn't have icons, but I think in hindsight should have, is to rotate a photo like in Preview or the Photos app.

01:27:02   It'll say in the menu, rotate counterclockwise or rotate clockwise.

01:27:09   So I am pretty good with clockwise and counterclockwise, but I still have to think for a second.

01:27:15   But there's a lot of people out there, and I live with two of them, who always get it wrong.

01:27:20   They just, if you say clockwise, you know, it's like a piece of buttered toast that always falls on the wrong side.

01:27:25   They're always guessing the wrong way.

01:27:26   But the icon, which shows which way the rotate command works, is actually more clarifying than using the words clockwise or counterclockwise.

01:27:36   Draw the arrow showing which way the photo is going to rotate.

01:27:40   That would have been a good icon to put in macOS 20 years ago or two years ago or 10 years ago or whenever.

01:27:48   Now it's there in macOS Tahoe, but those icons are lost because all the other menu items have icons too, and they don't stand out.

01:27:58   Whereas if only rotate clockwise and rotate counterclockwise had icons, they would draw your attention, and you'd be like, oh, yeah, those are the two rotate commands.

01:28:08   One rotates this way, one rotates that way.

01:28:10   Because everything is special, nothing is special.

01:28:12   Because everything has an icon, no icons stand out.

01:28:16   I think that's a really good point, especially with the comparison to restaurant menus.

01:28:21   We don't often think of menus on computers as directly comparable to restaurant menus, but of course that's where it came from.

01:28:28   I mean, at least the idea, a menu of options.

01:28:31   And they're organized into sections, right?

01:28:34   Like here's the entrees, here's the appetizers, here's the beverages.

01:28:38   Right, and so if Apple was in charge of the restaurant menu, there would just be like a generic food icon for anything that nobody could come up with an icon for.

01:28:48   I'm looking in preview right now.

01:28:50   This is another example that sort of drives home the uselessness of it.

01:28:53   If you look in preview under the view menu, you got hide sidebar and thumbnails, which are about as similar an icon as you can get without being identical.

01:29:05   It shows a rounded rect with a visual sidebar with three dots representing the content.

01:29:13   And it's like, oh, I know.

01:29:15   Which one do I want?

01:29:16   Oh, I know.

01:29:16   I want the one with the three dots that are more centered.

01:29:19   I have to take my glasses off and get closer to the screen to see the difference.

01:29:23   I can see that the icons aren't exactly the same, but I would have to get closer to the screen to see what the difference is.

01:29:28   And it's two totally different commands, right?

01:29:31   It's madness.

01:29:33   All of the zoom icons, they're so small, I can't even tell which ones have a plus.

01:29:39   There's actual size is like a magnifying glass with a one in it.

01:29:43   Zoom to fit is a magnifying glass with, I guess, arrows in it.

01:29:48   Zoom in is a magnifying glass with a plus.

01:29:51   Zoom out is a magnifying glass with a minus sign.

01:29:54   The magnifying glasses are so small that those little plus minus or a number one, whatever the hell is in zoom to fit is they're all you can't see them.

01:30:03   They all just look like magnifying glasses.

01:30:07   So how do you, they don't help.

01:30:09   One of the things that was so great in Prokopov's, I think it was his, the other guy who wrote the really good post on this was icons and menus everywhere send help by Jim Nielsen.

01:30:18   So I forget if it was Jim Nielsen or Nikita Prokopov who wrote this, who just said, okay, take all the words out of the menus and just show the icons.

01:30:28   Can you tell what they do?

01:30:30   And he like showed an example.

01:30:31   You can't tell what they do.

01:30:33   There would be no possible way to possibly navigate the menu bar in Tahoe by the icons alone if you took the words out.

01:30:41   But other scenarios where words go with icons like toolbars, right?

01:30:47   Like the toolbar, the Cocoa toolbar in Apple Mail or in MarsEdit.

01:30:53   And you can set the toolbar to icon only, icon and text or text only, right?

01:30:59   That's been this way since Mac OS X shipped 25 years ago.

01:31:03   Because a toolbar only has enough, you know, what's the most toolbar items an app usually has?

01:31:09   A dozen is probably pretty busy, right?

01:31:11   So it's fewer than a dozen toolbar buttons.

01:31:14   Probably more like six, seven, eight of the most used commands.

01:31:19   And because they're the most used commands, you can come up with icons that represent them.

01:31:23   So you could use it icon only.

01:31:25   You could use it icon and text.

01:31:27   Or you could use it text only because either way you can kind of remember or learn what they represent.

01:31:32   But when it's 50 different menu commands, you can't possibly memorize them by icon alone.

01:31:38   And so the icons just add noise.

01:31:40   It is so frustrating that Apple of all companies shipped an OS with this as the design change.

01:31:49   All right, let me take a break here for one more sponsor read.

01:31:53   Our fourth and final sponsor of this episode, our good friends at Factor.

01:31:57   Factor's chef-prepped, dietician-approved meals make it easy to stay on track and enjoy something comforting and delicious and easy to prepare no matter how hectic your life is.

01:32:13   Factor, it's not a meal kit where you get a bunch of ingredients and then you spend 15, 20, 30 minutes putting them all together and then cooking it.

01:32:21   These are pre-made meals that come sealed up, refrigerated.

01:32:25   They come in a box with dry ice to keep everything cold and refrigerated.

01:32:30   You put them in the refrigerator.

01:32:31   But these are fresh meals that then when you're ready to eat them, you can just heat them up in the microwave.

01:32:39   It takes just a couple minutes to heat them up and you've got a full meal ready to go.

01:32:43   And it is really good food.

01:32:45   I am just a huge fan of so many of the items there, including their breakfast items in particular.

01:32:51   But they've got everything and they've got all sorts of categories, lots of meat options, lots of steak and pork chops and stuff like that, lots of vegetarian options.

01:33:00   They have a whole category of new Mediterranean diet options that are packed with protein and good for you fats from Mediterranean diet type meals.

01:33:11   And they've got Asian-inspired meals with bold flavors influenced by China, Thailand and more from better choices, from more choices to better nutrition.

01:33:21   That's why 97% of customers of Factor say that Factor helped them live a healthier life.

01:33:28   These are really good meals.

01:33:29   They really do feel healthy, but they are delicious too.

01:33:32   So you get variety.

01:33:34   You get tons of choices every time you order a new package of Factor meals.

01:33:41   So many choices to choose from, something to please everybody, a whole bunch of great breakfast items too.

01:33:46   I am a big fan of it.

01:33:48   I really do.

01:33:48   I eat them all the time.

01:33:50   And when I run out, I'm a little bit sad when I'm looking for what I'm going to have for lunch.

01:33:54   Eat smart at Factor meals, F-A-C-T-O-R-M-E-A-L-S, factor meals.com slash talk show 50 off.

01:34:05   And use that code talk show 50 off.

01:34:08   The 50 is five zero.

01:34:10   Talk show five zero O-F-F, talk show 50 off, factor meals.com slash talk show 50 off.

01:34:18   And you get 50% off your first box plus free breakfast for one year.

01:34:22   That's talk show 50 off at factor meals.com for 50% off your first box plus free breakfast for one year.

01:34:29   Get delicious ready to eat meals delivered with Factor.

01:34:32   Offer only valid for new Factor customers with code and qualifying auto renewing subscription purchase.

01:34:39   My thanks to Factor.

01:34:41   You had mentioned in a group chat that you and I are both in that you just like last week or Friday maybe added a feature to Mars edit that you said you never would have even tried if not for AI.

01:34:56   And I think cloud code in particular.

01:34:57   Can you describe it quickly?

01:34:59   Because that's a bold statement.

01:35:03   I wouldn't have even tried this without AI, but I did it with AI and it was kind of easy.

01:35:06   I will explain it because I'm delighted by it.

01:35:10   And I think I can explain it in a way that everybody, programmer or not, will be able to appreciate.

01:35:14   So the situation is Mars edit connects to WordPress through the WordPress XML RPC API.

01:35:24   One day I'll switch to their newer RPC API, but for the time being, it uses this.

01:35:28   And one of the persnickety things about the API is that it can be hard to determine when a user types their password wrong, whether the error you get, or I should rephrase that.

01:35:45   There are errors you get from WordPress that mean either the user typed their password wrong, they're not authenticated, or they're trying to do something on the blog that their user's permissions don't allow.

01:35:59   And so it kind of gives the same status code for each.

01:36:03   In some cases, it's a real mess.

01:36:06   It's like different between different API calls, et cetera.

01:36:09   So for years, I have had this little hack in there that says, if the error mess, because it gives you error message text.

01:36:16   And so if I get what looks like you need to.

01:36:20   So the code, the code doesn't change, but the error message does change based on the context.

01:36:26   And, and so long and short of it is that I have, I call it implicit password sniffing.

01:36:33   Of course, no, no, implicit authentication sniffing, because I have to determine in the app, is the error I'm getting actually something I should put up a password prompt for?

01:36:44   Or is it something I should just convey the message directly?

01:36:47   Because you don't want to show a user a message that just says your password is wrong without giving them a field to put the right password in.

01:36:57   And you don't want to give a user repeatedly a password prompt when all, when they're already logged in, they've already been able to download posts.

01:37:05   And it turns out the only thing they can't do is like edit other people's posts or something like that.

01:37:09   So I ran into a customer.

01:37:13   This is one of these situations.

01:37:14   This happens to me all the time.

01:37:15   A customer's, so getting back to the perfectionism thing, getting back to the no bug report is too trivial.

01:37:23   This person running in German, MarsEdit, got, reported a bug to me that every time they tried to edit a post, it asked for the password again.

01:37:35   And I was like, what the heck is going on?

01:37:37   And I finally realized it's this situation where I never see it myself because I have a workaround in English.

01:37:47   So then the question becomes, okay, so what I'm doing is I'm searching the strings for phrases that I think imply permission problem.

01:37:54   Like if the string says not allowed, I will, instead of putting up a password prompt, I'll just pass the string along to the user.

01:38:02   Like you're not allowed to edit posts.

01:38:04   Obviously I can't, so I see this situation.

01:38:07   There's no like programmary API level way to tell that it's a authentication thing.

01:38:13   I have to use the string.

01:38:15   And so I'm thinking, well, I can't just say like, what does not allowed in German mean?

01:38:20   And then if I do, that will only fix it for German.

01:38:23   So what I used Claude, as it turns out, and ChatGPT in tandem, what I used it to do was to derive a list of error strings that WordPress returns in every language that WordPress supports.

01:38:43   And then to map them from the other language back to the English version.

01:38:49   So that you, if you know the English ones, you can map it to whatever other language there is.

01:38:54   That's exactly right.

01:38:54   But more importantly, as it turns out.

01:38:56   For the same error.

01:38:57   Back from every other language.

01:39:00   Because you can envision how if you're on the German Mac, and MarsEdit gets an error back, and it says, you know, whatever in German, you can't edit posts.

01:39:07   Right.

01:39:08   I now, instead of looking for not allowed, I first translate it to English, and then look for not allowed.

01:39:18   And this would have been terrifyingly difficult to put together.

01:39:23   How many languages are there?

01:39:25   Oh, I think it's like 70 or something.

01:39:27   It's a lot.

01:39:27   And if I was doing it by hand, there's no way I would have supported all of them.

01:39:32   I probably would have just picked, you know, the 10 most popular languages.

01:39:35   But.

01:39:37   Back to French.

01:39:39   So just at a high level, just to kind of, if people might be interested to hear, like, how did I approach this with AI?

01:39:45   So at first I realized, luckily, the XML RPC API in WordPress is pretty much dead, like it's not being developed anymore.

01:39:54   So with high confidence, I can predict that the error strings in that section of the product are not going to change.

01:40:04   So I point Claude at the source file for the XML RPC server on WordPress, and I just say, give me a list of all the unique error strings that are passed to the WordPress localization function.

01:40:18   So it gives me this list, whatever it is, it's like, you know, I don't know, 50 something, it's all these different error messages, like you are not allowed to upload, you are not allowed to da da da.

01:40:28   So then I ask it to, like, filter out ones that obviously don't have any permission-related context.

01:40:34   So I get this list of all the English strings.

01:40:37   And then WordPress is localized online with an automatic, not automatic, WordPress.org-based custom, it's called GlotPress, I think.

01:40:48   And it's a custom, like, anybody can contribute the translations, blah, blah, blah.

01:40:52   But I now need to get a copy of these strings in all whatever 78 languages.

01:40:58   And there's no way I would just, like, type every string into every language.

01:41:04   So I figured out, with ChatGPT's help, how I could access the site programmatically, supplying an English string and getting back the localized string.

01:41:16   Oh, and somewhere in here, I also had to have it figure out for me what all of the international, like, language codes were for all of the pertinent languages.

01:41:25   And some of them, it had to, like, infer it because it wasn't, like, the language code listed wasn't the same way it was managed on the site.

01:41:33   Just to say, I got this list of language codes, 70 or 80 language codes, a bunch of English strings, and then I had it write a script that would go to the WordPress localization site with each of those strings, with each of those languages.

01:41:49   And it created a one megabyte text file that contained all the mappings of all of the strings.

01:41:57   And so then I had it write me some code in Swift that would just, like, let me say, okay, here's the English, give me the translation.

01:42:04   And then, kind of like a Jim Correa moment, I wasn't going to sleep at night if I knew that I was dedicating one megabyte of my app to supporting the rare case that somebody has a weird error in another language.

01:42:18   So, with more AI help, I managed to deduplicate their situations where, like, a lot of the phrases are translated the same in, like, Mexico and Spain and whatever.

01:42:31   Anyway, I ended up with this now, I think it's 16K or something small, something small enough that I'm, like, sold.

01:42:39   I'm in.

01:42:39   Wow.

01:42:40   And that's when I committed the feature, I tested it, testing against a local WordPress install, I can change the language at will, it's working everywhere.

01:42:49   Why wouldn't it work?

01:42:50   It's great, fantastic.

01:42:52   And there's no way in hell I would have even bitten off the first part of that challenge.

01:42:57   And, like you said, I mean, I did this in, like, a matter of, like, an hour or two.

01:43:03   It was something that would have taken days, at least, I think.

01:43:09   I mean, I would have, as a programmer, I would have automated what I could.

01:43:15   Yeah, of nonstop work.

01:43:16   I would have probably figured out how to access the translations with a URL and a script and automate it.

01:43:22   I would have written a Python script, probably.

01:43:25   But just having it do all of it for me is incredible.

01:43:28   I, that is incredible.

01:43:32   I have a much, much smaller example.

01:43:34   But it, and I don't, I feel like for programming type stuff, I don't use AI enough.

01:43:39   And I don't do a lot of programming.

01:43:40   And so, and when I do, it is sort of, like, go back to black ink and solving puzzles.

01:43:47   I kind of like doing it by hand because it feels like a puzzle to make it work.

01:43:51   But the other day, I linked to a post that my friend Om Malik wrote.

01:43:55   And there, I've just, back to reporting typos or grammatical things.

01:43:59   He repeated the same sentence in two paragraphs.

01:44:02   And it was a very good sentence.

01:44:04   But I knew that that was a mistake, that he didn't find in editing.

01:44:07   And one of them was, like, paragraph 17 and one was paragraph 23.

01:44:11   And I just, I wanted to be able to tell him which two paragraphs it was in when I told him about the typo.

01:44:17   And I thought, wouldn't it be great if I could just click a button and have numbers for all the paragraphs in Safari?

01:44:23   And then I thought, huh, you could probably do that with a bookmarklet.

01:44:26   And I was like, I could probably figure out how to do it.

01:44:28   And I was like, I'll bet ChatGPT could just make me a bookmarklet.

01:44:31   And I just asked for that.

01:44:33   Like, it was just a very simple prompt, like one sentence, like, write me a bookmarklet that would add a number to each paragraph.

01:44:42   And it did, and that's all I asked for, and it gave it to me, and it worked on the first time when I dragged the bookmarklet to Safari.

01:44:50   And that's how I know that they were paragraphs 17 and 23.

01:44:53   And the other weird thing is it just counts every P tag in HTML, and sometimes outside the article, like on Daring Fireball, the by John Gruber is a P tag.

01:45:04   I don't know why.

01:45:05   You know, what else is it going to be?

01:45:07   But then, you know, it throws the count off.

01:45:09   And so I just asked it, you know, and now do a version of it with a click handler.

01:45:14   So if I click on any paragraph, that paragraph becomes the new first paragraph.

01:45:18   And the ones that are before that become zero, negative one, negative two.

01:45:24   Yeah, it's a –

01:45:26   Just type that, gave it to me, and it just worked.

01:45:29   And I was like, oh.

01:45:30   You didn't ask it.

01:45:33   And then I spent two hours tweaking the CSS to get it to look the way I want.

01:45:39   Manually.

01:45:40   No, but that was fun.

01:45:42   But, you know, just – I never in a million years would have sat down to write that bookmarklet because I need paragraph numbers in Safari.

01:45:49   Like, it occurs to me like once a year.

01:45:52   Like, there's a situation where I want it.

01:45:54   So I never would have done it myself.

01:45:56   But it literally was as fast as –

01:45:59   I had a working bookmarklet that I was using in Safari quicker than I just described it to you here on the podcast.

01:46:07   That's how quick it was.

01:46:10   And that's the sort of thing that there are people like you, developers, who are actually using AI.

01:46:15   And you're like, yeah, duh.

01:46:16   It does a lot more than that.

01:46:18   It does really –

01:46:19   But I think my message here is for the people who are still somewhat skeptical that they could get anything out of it or that there's really something to it.

01:46:28   And I'm not even a programmer.

01:46:29   Why would I do it?

01:46:30   But you might want a bookmarklet in Safari or whatever, Chrome, whatever, that does a weird thing that you kind of want.

01:46:37   And that's the sort of place.

01:46:40   Like, you're not building a whole app.

01:46:41   You're just making a bookmark that you can select that will add blank to every paragraph on a web page.

01:46:48   Right.

01:46:49   And you did it – what you did was you gave it the chance to be easy.

01:46:54   Like, a lot of people just say, I'm sure this isn't going to be easy, so I'm not going to try it.

01:46:58   You say, well, what the heck?

01:46:59   I'll just ask, can you make this for me?

01:47:01   And then, of course, in 10 seconds it was done.

01:47:03   And that's the magic of these LLMs.

01:47:08   And I agree with you.

01:47:11   I mean, I know that there's – we're going to get feedback on this show.

01:47:15   Somebody is going to say, you know, we're ruining the planet, whatever, blah, blah, blah.

01:47:18   There's still a huge amount of animosity from some people towards anything AI.

01:47:24   It's a very charged situation.

01:47:27   But there's just no denying –

01:47:29   But like on the energy thing, it's like a chat GPT prompt on average takes the same amount of energy.

01:47:36   And I've seen this source – I mean, I can't verify the math, but it makes sense to me – that it takes about the same amount of energy as it does to watch eight seconds of video from Netflix.

01:47:46   I mean – and I've seen things like a lot of the energy was already spent to build the models, blah, blah, blah.

01:47:51   Right.

01:47:52   Are these people going to get in their car and drive to Walmart after they're done criticizing my use of electric power?

01:47:59   The fact is humanity uses a lot of energy.

01:48:01   We use a lot of energy to do a lot of things.

01:48:05   And technically speaking, we don't need to use any but maybe a little fire.

01:48:09   We're basking in the glories of excessive energy use.

01:48:14   And I'm here for it because it is, you know, the whole like bicycle of the mind comparison of like a Mac.

01:48:23   The way that like a tool like a bicycle can make you more powerful than you ever could have been without the tool, that is what AI is doing for many of us.

01:48:35   Like that was one little example.

01:48:37   I often have examples where I'm like I feel like I'm compelled to go share.

01:48:41   And in this case, I just shared it with you on the chat we're in.

01:48:44   Now I'm sharing it with everybody.

01:48:46   But it is remarkable.

01:48:49   And as somebody who is a generally like a humanist person, a lot of what I do, you know, the reason I report typos to you, the reason I make software that I think people will like is because I want other people to be more successful.

01:49:02   If I only cared about my own success, I wouldn't be bothering.

01:49:06   I wouldn't care about fixing your typos.

01:49:08   I wouldn't make software for other people unless I guess unless I made a lot of money.

01:49:12   But what I feel about AI is it's too good not to share.

01:49:16   In other words, like people out there are missing serious improvement.

01:49:22   And it is overhyped, right?

01:49:26   It is absolutely 100% overhyped.

01:49:28   But that doesn't mean that it shouldn't be significantly hyped.

01:49:31   Everything that deserves a significant amount of hype and really is transformative is going to get overhyped, right?

01:49:38   The existence of the internet was overhyped.

01:49:40   Personal computers back in the 80s were overhyped, right?

01:49:44   But it really was groundbreaking like that, you know, an accountant who is still doing all of the work and paper in a green book and say to his colleague who is using Lotus, you know, the first spreadsheet.

01:49:57   I think Lotus 1-2-3 was the first spreadsheet, right?

01:50:00   I think so.

01:50:01   Not for – my memory is not that bad.

01:50:02   Wait, yeah.

01:50:04   You know, but they're using a spreadsheet.

01:50:06   Oh, no, no, no, no, no.

01:50:06   No, it wasn't Lotus 1-2-3.

01:50:08   No, what was it?

01:50:09   It was Dan Bricklin.

01:50:11   Why am I – it's the classic.

01:50:16   I got to look it up right this minute because –

01:50:18   I know.

01:50:20   VisiCalc.

01:50:20   That's it.

01:50:21   VisiCalc.

01:50:22   VisiCalc.

01:50:22   And I think Lotus 1-2-3 was –

01:50:24   I was right to question myself.

01:50:25   Yeah, yeah.

01:50:25   I think Lotus 1-2-3 was like an early kind of clone copycat.

01:50:29   Well, but somebody in the early PC era is using a spreadsheet and they're like, I don't see the difference.

01:50:34   It doesn't take me that long.

01:50:35   I have to type in the number.

01:50:36   You have to type in all the numbers anyway.

01:50:38   I write them on paper and then type them in a calculator and I get a sum and it's like, okay, but, you know, you've got 500 transactions from a bunch of different states and I've got 500 transactions in my spreadsheet in a bunch of different states.

01:50:51   But I just realized that I had the sales tax for Maryland wrong.

01:50:56   And I update that in one square, one time, and the whole thing recalculates instantly.

01:51:03   And then the guy who's doing it on paper is like, oh, I need that, right?

01:51:08   And AI can do things like that too, like where we've all become totally used over the last 40 years to the things like that scenario of, oh, yes, you update the Maryland sales tax once and the sum of 500 transactions, only 10 of which were from Maryland, updates, you know, you get a new sum of the total instantly.

01:51:32   We're all used to that.

01:51:33   But we're at this moment where there are things like your scenario with 70 different languages and WordPress to try to just add an ease of use, a nice, wouldn't it be nice if every single language that a MarsEdit user connected to WordPress is using got a sensible description of what the problem is for any language?

01:51:57   Two hours.

01:51:57   Two hours.

01:51:58   Right?

01:51:58   That is like the, you know, it's maybe not quite instantaneous, but for that task, it feels that might as well be instantaneous.

01:52:06   And it's not just like fun and it's not just amazing.

01:52:12   It's like it's amplifying my ability to create things.

01:52:17   And that's for me, I'm a lucky person.

01:52:20   Most of my needs are already met.

01:52:22   So for me, it mostly is about fun and continuing to improve the things I like.

01:52:26   But for other people out there, this could be like a career changing, you know, like your accountant, the one who didn't get the spreadsheet, made a lot less money the next year.

01:52:38   You know what I mean?

01:52:40   Like, so I, yeah, yeah, yeah, or it was, it could handle a lot fewer clients.

01:52:46   At your own peril, ignore the technological advances.

01:52:49   I've got a couple of wrap up points here at the end.

01:52:52   I looked it up.

01:52:54   It was Macromedia Director.

01:52:56   Oh, yeah.

01:52:56   Okay.

01:52:57   Yep, yep, yep.

01:52:57   Director.

01:52:58   That was the prototype.

01:52:59   Yeah.

01:53:00   I want to mention on the menu items in Tahoe that our mutual friends, both of them, including, we keep mentioning him, Paul Kafasis at Rogue Amoeba and Brent Simmons with the latest updates, the Tahoe update to Net Newswire.

01:53:16   Brent figured out a way to not have those ugly fucking icons on every menu item and not only shipped a version of Net Newswire for Tahoe that does that and takes them out, but explained, of course, because he's Brent, exactly how.

01:53:32   And now Rogue Amoeba's apps for Tahoe do the same thing.

01:53:35   And I will put links in the show notes to their posts about that.

01:53:38   I still need to write about both of them on Daring Fireball, but there was no way that after mentioning Paul for reporting typos and complaining about the menu icons in Tahoe that I couldn't mention that Paul spearheaded Rogue Amoeba shipping software updated for the Tahoe interface without icons on every goddamn menu item.

01:53:57   And last but not least, I want to say that you, Daniel Joukett, are a liar.

01:54:04   That's true.

01:54:05   Or is it false?

01:54:08   Well, in this particular case, it's because your long-running podcast with Manton Reese, Core Intuition, ended last year.

01:54:17   And then there were 12 episodes after it ended.

01:54:20   I'm a dirty liar.

01:54:23   Yeah.

01:54:24   Well, we adopted – I know you know this, John, but we adopted the Apple versioning system.

01:54:30   So last year we had episode 26.0 and 26.1.

01:54:36   So I guess this year we'll move on to 27.

01:54:39   Wait, is that right?

01:54:42   Yeah.

01:54:42   It's actually – yeah.

01:54:46   If – you know, and that is actually kind of – I actually – I'm making fun of you.

01:54:51   But I actually think that's kind of an interesting way to semi-retire a podcast, right?

01:54:59   Where, okay, we're going to stop doing it.

01:55:02   What was the old schedule?

01:55:03   You guys once a month?

01:55:03   Once a week, yeah.

01:55:04   Once a week.

01:55:06   We were weekly.

01:55:06   Holy shit.

01:55:07   Yeah, I bet.

01:55:08   Well, I missed them.

01:55:08   I mean, we were up to 400-something, I think.

01:55:10   About where you are.

01:55:12   But you know what?

01:55:14   I don't miss them now.

01:55:15   Oh, yeah.

01:55:15   Now that you're doing them like twice a year, I don't miss episodes of Core Intuition.

01:55:19   And it's a very interesting way to say, okay, we're not going to do once a week anymore, but we still – every once in a while, we want to talk to each other and do the Core Intuition thing about what's new in the Apple developer community.

01:55:32   And I'm so glad you did it.

01:55:33   And it is kind of a neat way as a fan of the show to be like, hey, it's not over-over, right?

01:55:39   It's like, okay, Letterman doesn't do his nightly talk show anymore, but he's got that thing on Netflix and once a year there's like five special episodes.

01:55:46   I think when we said it was over, I think we said something like, who knows what's going to happen, so don't delete us from your podcast feeds.

01:55:55   And then I put out an episode and all these people were like, what?

01:55:59   I thought you were over.

01:56:00   I deleted it already.

01:56:01   That's your problem.

01:56:02   Don't delete my dinosaur feed.

01:56:04   All right.

01:56:08   Well, all right.

01:56:10   That's it for me.

01:56:11   I don't have any other follow-up points.

01:56:13   Where else can we direct?

01:56:14   So there's Core Intuition, which they can listen to in wherever they're listening to the talk show right now, whatever podcast app you're using to, just search for Core Intuition.

01:56:22   You'll see the new episode for sharing numbers that Daniel is talking about.

01:56:28   Your software is at RedSweater.com, including MarsEdit, Black Ink, which you mentioned, your crossword app.

01:56:35   That's pretty much it.

01:56:37   What else?

01:56:37   That's just kind of like my old nostalgia stuff.

01:56:40   People can enjoy your musings on social media.

01:56:44   What's your favorite?

01:56:45   What do you have?

01:56:46   Mastodon, probably?

01:56:46   Well, I start with Micro.blog because of our good friend Manton Reese.

01:56:50   Oh, yeah.

01:56:50   So what I do is I have pretty much taken a copy everywhere approach.

01:56:55   So I write at Micro.blog, which clones it over to Blue Sky for me.

01:57:01   And then I could probably automate more of this.

01:57:03   But right now, I just end up putting it on Mastodon and Threads as well.

01:57:07   I took some time off from Threads because I'm not philosophically aligned with Meta.

01:57:12   But I'm back there because, honestly, every one of these services has become the exclusive social networking choice for somebody – some of my favorite people online.

01:57:30   Right, right.

01:57:32   I keep saying it.

01:57:35   In some ways, it was better – it was ideal when Twitter was good overall and it was the one Twitter-like thing.

01:57:43   And then when Twitter got bad and it was the one Twitter-like thing, it was terrible.

01:57:48   And once Twitter went bad, there was no going back.

01:57:53   And now we've got multiple things that are Twitter-like.

01:57:57   And in some ways, that's good because it's a healthier – well, if one of them goes bad, if Threads goes bad because Meta owns it, we're not all sunk because we haven't all just thrown all of our stuff into one place.

01:58:11   But the downside is it's like people are all over the place.

01:58:16   And you can't just go to Twitter to get everybody's Twitter-like musings that you're interested in.

01:58:22   You've kind of got to bounce around.

01:58:24   But probably better overall.

01:58:27   So people can find you there on all of these networks and you've got a wonderfully searchable surname.

01:58:33   They'll probably find Jalkit first.

01:58:38   All right.

01:58:39   Let me just wrap it up by thanking our sponsors.

01:58:43   We have four amazing sponsors for this episode.

01:58:45   Notion, the AI workspace where teams and AI agents get more done together.

01:58:49   Squarespace, where you can build your new website and register domains online.

01:58:55   Sentry, a real-time error monitoring and tracing platform.

01:58:58   And Factor, healthy eating made easy.