821: Developer Roundtable 2025

Welcome to the Mac Power Users. I'm David Sparks and joined as always by your friend of mine, Mr. Stephen Hackett. Hello, Stephen. Hey, David. How are you today? I am fantastic. And today we are recording one of my favorite shows we do every year. It is time gang for the developer round table, 2025 air horn, air horn. We've got a great, we've got a great panel this year. We like to do this every year or two.

where we have some of our friends from the development community come in to talk to the listeners about how things are going for developers and Apple overall. This year, we've got a new panel. So I guess let's just go through them. First, we'll welcome Casey Liss. Thank you for joining us, Casey. Hello. Thank you for having me. It is always a pleasure to talk to all of you. Casey is...

got an app called call sheet. That's your big one. And he has also been on a Mac power user several times in the past. And he's on a little podcast called, what is that? Was that something about technology? What's that? And I think you're thinking of neutral, which is, which is.

Clearly my most popular show. No, you're probably thinking of either the Accidental Tech Podcast, which I do with our friends Marco Arment and John Syracuse, or perhaps Analog, which is here on Relay with our friend Mike Hurley. Yeah. And so Casey, hopefully he's known to some of the listeners, but also an active developer. We're looking forward to talking to you. Yep. Thank you for having me. We're also joined by Charlie Chapman.

Charlie and I are kindred spirits because we both live in the central time zone that everybody forgets. Best time zone. Charlie is the senior developer advocate at RevenueCat. Also the host of the launch podcast where, Charlie, you talk to developers about their software and how they got started. And then you have a little portfolio of apps as well, too.

Two apps that I use all the time, Dark Noise and Framus. Thank you for joining us today. Thank you for having me. I'm really excited about this. And rounding out the panel is our friend, James Thompson. James has been on the show before as well. James, can I call you the elder statesman of our developers? I was going to say, I am.

Happy to play the role of grumpy old man for the purposes of this podcast. Yeah, James famously worked at Apple, helped bring the doc to life, but also has an app called Peacalc that is... loved and beloved by many of our listeners. It's just, you know, it's the calculator app that you can run on anything Apple makes. And it's a calculator app that, that Apple never made. So thanks for joining us today.

James, you also have Dice, which is a dice rolling game. And you're frequently showing up on relay shows. So it's great to have you. You have to read the credit where I've written down that I'm the host of Connected, just for the bit. Wait. All right. You got me. Gang, we have an announcement. New host of Connected. No!

Yeah. Thank you all for joining us. We're going to get into your expertise and a bunch of topics. But before we get there, David, do you want to give everybody a quick reminder about your new field guide? Yeah, Apple Productivity Suite Field Guide. There's two days left on the intro discount. MPU launches a code. You get 10% off, but just two days left. So get in there and get it if you want it.

Hearing great things from people that have bought it. Thanks, everybody, for the support. And as always, we're going to have a more power user segment today. That's the ad-free extended version of the show. We're going to do a developer edition of State of the Platforms on that since we've got all these smart folks here. And looking forward to that as well. You can sign up for more power users at relay.fm.mpu.

So I'd like to start by talking about the annual yearly cycle of a developer on Apple's platforms. I think a lot of us feel that... The first or second week of June at WBDC is sort of the beginning of the year. We then have a summer and early autumn of betas. We have OS releases and new iPhones in the fall. And then the rest of the year is sort of this bug fix feature release cycle that many of us go through. And then we do it all over again. And James, I'd like to start with you.

We're 17 years into the iOS App Store, which is just unbelievable to me. Do you think that annual pace still makes sense? I mean, I think it makes sense for Apple product marketing. who want to have shiny new features to go with a shiny new phone for selling it in September. But I think it makes no sense for anybody else. Certainly not developers.

Inside and outside of Apple. I mean, like, everybody is crunching in this. I mean, Apple at least gets the, they know what's, well, some of them know what's happening before. But we get basically June to September to, no matter what Apple has done, turn around all our apps, or at least some of our apps. And it is a... remarkably stressful period, I would say, particularly if you're, you know, still under the delusion of trying to hit day one.

which we will come back to as a thing. Because I think what used to make sense is you get your app out on day one, you pick up all the Apple new features and Apple features you prominently in the store and you make lots of money. Problem is, it just doesn't work anymore. Casey, one of the things I think about with this is the annual release cycle is now universally applied to every Apple platform.

I've often wondered, like, I understand why some of them need annual updates, but the Mac, for instance, do you think all of these Apple platforms need a yearly update? I mean, honestly, I don't think. Anything really does with the exception perhaps of hardware, because oftentimes Apple will tie like.

I don't want to call them drivers lest Steven, you know, start wigging out on me, but like driver updates, if you will, basically low level hardware stuff or the software that powers that interacts with low level hardware. Oftentimes that'll be tied to an OS release. And that's why.

You know, when you get a new iPhone in the summertime or I guess fall, really, that'll always, always, always have the most recent version of iOS because it's only that most recent version of iOS that has support for whatever the current iPhone is. As we sit here recording today. ios 26 is required for an iphone 17 pro for example with mac os you know with max happening less frequently and with with mac os arguably being easier to update uh in turn from a user perspective anyway um i i don't

I don't think it's really necessary for the Mac. I don't think it's... Bad, although I do heartily agree with James that I don't know that it's serving anyone except perhaps Apple. And if I were to have my druthers, if I were king of Apple software development, I think what I would rather do is just roll things out, you know.

piecemeal and ad hoc is perhaps a better way of phrasing it as they're ready and if you have a big giant feature ready at such and such time then do it and i think part of the problem isn't and james was alluding to this is that they want to make a big splash in june at wwdc and if you're

If you're allowing big major features to trickle out throughout the year at any time willy-nilly, it's hard to have a big fancy feature, a tentpole feature to display and discuss at WWDC. And I think that's the crux of the issue. It also kind of seems like Apple learned a bit of a lesson last year with the Apple intelligence rollout. It seems like this year they're a little more subdued about making big promises for the rest of the year. Do you think that's a good thing or a bad thing?

I mean, I definitely think it's good. I really, really dislike over-promising and under-delivering. And Apple's general MO is the opposite, is to under-promise and over-deliver, or at least that's Apple at its best. I'll let you decide whether or not that's still the case. But I think that Apple, yeah.

Definitely got bit by the Apple intelligence thing where they did something unusual, which is they pre-announced and they said, oh, it'll be fine. It'll be coming soon. And then soon became later. Then later became not at all. See also air power or whatever it was. So, yeah, I think this is a retreat to form or return to form. It's perhaps a more complimentary way of looking at it. And I think Apple is definitely playing things close to the vest, which.

is I think it certainly serves Apple much, much better that way because they're not putting themselves in a position where they've put up a, you know, uh, uh,

What's the thing when you run and you jump over it? Oh, God, I'm drawing a blank. When you're in the track and you're running it. The hurdle. Hurdle. Thank you. Gosh, I was drawing a blank. They're putting a hurdle in front of themselves by saying, oh, we will have Apple intelligence out at such and such a time at such and such a date. And by not. putting those hurdles in front of themselves, there's no barrier in the way, right? And if they...

figure out a way to release something amazing and awesome, they'll tell us about it when it's ready. But to tell us about it in advance, I think it's a fool's errand. And the only reason they did it this time, I believe, is because they felt like the market was going to be very upset if Apple didn't show some amount of...

forward motion with regard to AI. When we're thinking about WDC, there's all these announcements. And Charlie, I'd love to get your input here as someone who is... updating multiple apps, but you also work with a lot of developers through your job at RevenueCat. How do you go about deciding what to prioritize out of, you know, sometimes a slate of dozens of announcements every June?

Yeah, I mean, I think it depends on the type of app. I think the type of apps that people that listen to MPU care about are the type of apps where being a good platform citizen is important. And for that... You know, obviously step one is get the app working. Step two is get the app working correctly within the new way of doing things. And then step three to me is like figure out what.

both Apple editorial, Apple marketing is caring a lot about and figure out what the press cares about. And then... I should say also what users are caring about but if your goal is to get out on day one and get attention as James kind of alluded to there isn't that sort of high likelihood of massive returns.

just being there on day one you have to be there on day one with something that apple and the press are picking up on and telling stories about and so for this year uh you know it was kind of looking around and being like is it going to be something with the widgets on vision os probably not is it going to be something with uh using the new apple intelligence uh foundation models and integrating that into your app and that was kind of the bet i made this year in terms of where i put

put my effort but that's kind of how i think about it is uh this is a unique opportunity once a year to get more attention on a feature and if i can like kind of latch on to that That's a way to get potentially new users that are outside of my normal sphere of influence. Casey, what about you? How do you prioritize what to tackle first? Yeah, you know, I learned from our mutual friend, quote unquote, underscore David Smith, that oftentimes a really solid choice is, if possible.

Find a feature that Apple is really, really jazzed about that may or may not. be something that your average developer would be jazzed about. So I think the canonical example of this was the genesis of Pedometer++, which was that in one of the early iPhones, I'm sure, Stephen, you remember which one, but in like iPhone 4 or 5 or something like that, they added... a step counter, the ability to count steps and whatnot. And so underscore thought, oh.

I could make an app that at the time, if memory serves, basically just displayed how many steps you've taken that day. And it was very little else. And Stephen, you know better than anyone on this panel that it's quite a bit different now than it was then. But Underscore's point was, let me just do the bare minimum.

here and see if i can get apple to notice me and to the best i recall they did and they featured pedometer plus plus and talked about it and so on and so forth and so If the stars align and if there's something that that I'm excited about and more importantly, that Apple.

is excited about then i'll try to do some feature around this and i think the foundation models that charlie was just talking about is a great example of this i haven't really done anything with this yet in any of my apps but that is a very good example but beyond that if there's nothing If there's nothing that I'm like, oh, this is the one, then then I'll probably at least start by thinking, OK, what is the bare minimum I need to get by? So in as we record this.

iOS 26 came out just a month ago or so. And so the bare minimum is, OK, a visual refresh. Keep everything roughly the same, but do a visual refresh to make it fit in with liquid glass and whatnot. If I can get that squared away, the sooner the better. Excellent. And then the reach goal is, okay, what else can I reasonably ship during the summer?

You know, for me, I still have young kids who are in elementary school. And so summer, unfortunately, is a very busy time for for kid related and family related things, which leaves less time for work related things, which is very unfortunate. But, you know, what can I reasonably ship? in the course of these two, two and a half months or three months or thereabouts. And that's kind of how I look at it. What is Apple like? What do I need to do? And what can I possibly do if I reach a little bit?

So once we are sort of past the beta season, when do you decide to ship something? James, let's start with you. I think when you ship... I used to get every single app revved and ready and out for day one. And I burnt out really badly doing that. I mean, it was... You're trying to basically hit the same targets that a $3 trillion, whatever, $4 trillion, $5 trillion company are trying to do. And you're one person.

And it's pretty difficult to do. So, you know, this year I decided, okay, my goal is going to be I am going to ship PCALC on all the sort of iOS families. day one mac comes a month later and then dice and various other things will get updates along the way And I also had the problem that due to a variety of family circumstances and stuff, I didn't get to ship iOS 18 updates.

I was doing iOS 18 and iOS 26 in one go for the apps, which was quite a lot of work. So I'm glad that I kind of limited it down. And then, yeah, once you've missed like day one, it's a case of, you know, when stuff is ready, when you think something is in a decent enough state.

Charlie, what about you? When is it time to hit submit in App Store Connect? I'm a very marketing-oriented mind, so I kind of usually work backwards. I'll set a target date, and then... try to get everything i can done within that date so like i'll de-scope uh the work that i'm doing before i'll move the date i'm aiming for like for example the original release of dark noise

sometime in like march i was like okay i need to give myself a date for when i'm going to release it and i just picked august because I always heard podcasters complaining about the sort of lull period where the betas had sort of calmed down, but it hadn't actually come out to the public yet. And I was like, I'm just going to thread that needle and see if I can get somebody to pay attention to me. And then everything worked.

backwards from there. And I tend to try and do that mostly because I'm not a very well-motivated person. And if I don't have a date, it'll just continually be a couple weeks from now for eternity.

You know, Casey, when you're talking about this period, like during the beta season, getting the updates going and ultimately the release in the fall, I think another. thing that listeners would probably be interested in is is dealing with customer support i i remember and i've told this story on the show before i was at wwdc one year and i was sitting next to a developer friend and during the State of the Union keynote, he got an email from a customer complaining that the beta broke his app.

Jeez. Jeez. Incredible. He showed it to me. I couldn't believe it. Like, you know, the guy hadn't even been back. He hadn't even had a sandwich yet, right? And apparently. you know they were upset that his thing didn't work now that's not mac power users listeners our listeners are much nicer than that but there is a load right for dealing with customer support getting this update out how do you balance all that

Yeah. You know, it's tough and it's very frustrating. It's also kind of demoralizing if I'm honest. Now I get a customer's perspective, you know, they're trying the new hotness and they want their stuff to work with the new hotness. But it's incredibly frustrating to receive an email or piece of feedback or whatever saying, hey, man, your stuff doesn't work in the beta that's been out to your point. Forty five whole seconds. What's the story?

You know, and it's like, well, I'm doing my best here. I'm one person. it's hard not to get cross with the person that writes that email and you shouldn't and really can't because this is someone who's enthusiastic about your app. Clearly that's one of the first things they've tried, but. golly is it not it takes the wind right out of your sails because you know during that first week of wwdc which is often the first or second week of june

All of us, or at least for me anyway, I should say that my, my head is spinning and is often spinning for the better part of a month when I'm trying to understand, okay, what is the lay of the land right now? And then coming back to what we were just talking about a minute ago, what direction or directions am I? And it's very hard to support beta software when you're not entirely sure what your approach is for that beta software.

Now, sometimes it can be an easy fix where, you know, maybe something with SwiftUI has changed and it's, you know, an additional thing here or there to fix it. And that's that. But oftentimes it's a much more involved fix. You're living through this whole beta period wherein you know that some of your most loyal and enthusiastic customers are getting more and more and more upset because you haven't fixed the beta.

Honestly, from a completely selfish perspective, it is not my responsibility to make my software work with Apple's pre-release software. Let me say that again. Pre- release software. So if you're a Mac Power Users listener and you're running a beta, and I know that David and Steven have talked about this many times, but you're doing that at your own risk.

That is not designed for anyone but the most powerful of power users. So tread carefully. I think you're a better person than me, not losing your temper with customers. I got one the other day who was like, I'd ask you a while ago. a while back about adding this feature you know i really like pcalc i think it would be useful um to other people if you added it last line

And if you do, I will edit the one-star review I have left. And I have to say, I wrote back and I said, this is not the way to motivate. me, or indeed any other developer to do something. And, you know, I'm not going to say that I deleted...

off my, you know, roadmap or whatever of doing this thing. Because that, A, I don't really have that much of a fixed roadmap. But I sure thought about, well, that feature's never getting ever... added and that's not the right approach but there's just that crossness that you get yeah just very quickly to build on what james said if you are withholding stars i understand the motivation for that because from a user's perspective that's about all the leverage

they have, but I can speak for myself and apparently James, and I think I speak for almost any developer I know. I am hashtag triggered the moment I see that because it's just like, oh, well, you're just a turd. You're just a turd who's not looking to work together on this. You just want what you want. You want it yesterday. And I'm sorry, but I'm not going to do anything for you, as James said.

This episode of Mac Power Users is brought to you by Squarespace, the all-in-one website platform designed to help you stand out and succeed online. So whether you have a new project or you're scaling your business, Squarespace gives you everything you need to claim a domain name, showcase your offerings with a professional website, grow your brand and get paid all in one place.

I've been using Squarespace since basically the beginning of time, really old versions of Squarespace. And from back then all the way till today, I just love working on it. It makes it really easy to build something that looks great and works well. One of the things I'm not an expert in is SEO, but that's okay with Squarespace. Every Squarespace website is optimized to be indexed with meta descriptions, an auto-generated sitemap, and more.

So more people will find your site through search engine results. And you can see how that's working with their analytics. Keep track of the stats that matter with intuitive built-in analytics tools. It makes it super easy for you to review website traffic, learn where to focus engagement, and it's perfect for online courses, exclusive tutorials, tracking revenue.

seeing where people are coming from, invoices, all of this stuff is just springs to life. Your decision-making springs to life with analytics. So head on over to squarespace.com slash MPU for a free trial. And when you're ready to launch, use the offer code MPU. You're going to save 10% off your first purchase of a website or a domain. That's squarespace.com slash MPU.

Offer code MPU to get 10% off your first purchase. Our thanks to Squarespace for the support of this show and all of Relay. Well, guys, the...

The 800-pound gorilla on the sofa through the whole last segment was liquid glass. Apple made a big change in the way the operating system looks and acts.

which often leads to a lot of work for developers. So I thought, let's take a minute and talk about that. What do you think of Liquid Glass and how's it going? James, why don't you start? I mean, I have a... wide variety of problems with liquid glass in terms of readability as the main thing and it's you know Clearly, Apple thinks the same because they are adding in the .1 releases this frosted glass setting, which is like liquid glass, but you can actually read the text.

That is on top of it. If your touchstone for your user interface is the sci-fi movies where somebody is looking at a screen that is entirely see-through, I feel you have made a mistake. But... I think some of it is fine. And I think you can still use it in a tasteful manner, but you have to be really careful about the level of contrast and accessibility that you have.

in your app and you know saying oh well people can just switch on accessibility options i think if you have to switch on an accessibility option for your stuff to be usable by let's say 75% of the population, I think you've failed. So I think there's good stuff in there. And I kind of like the animation and I like the...

I like the liquid. I don't like the glass, I guess. That's a good way to put it. You know, it really is. Now, as someone, I believe you're the only one of our developers that was developing an iOS 7 days. Yep. So iOS 7, people are still triggered by that. They did a big visual overhaul and they made under the cover changes that led to a very difficult year for developers and frankly, users.

How does this visual redesign compare? When you said iOS 7, my brain translated that to System 7, and I still said yes because it was still true. Yeah, it was. I was around for iOS 7. think i had anything in the app store at that point but i was around for i bet charlie was as well like we were we were there we were probably an android user at that time actually oh god i take everything back well how are you on the windows or windows phone uh i did that as well i mean i i think like

The iOS 7 was less traumatic because... You know, you could just strip out any personality from your app, then you fit it in with it. Whereas with this, you have to really kind of, I think you have to do some more. And I think... If you're using standard system stuff, you get a lot of this for free. And say your settings interface and stuff will probably, barring some stuff getting a bit bigger.

will generally just work and will look appropriate to the platform. And that stuff's fine. It's what you take into what I would call your interface, like the bits that you have designed. I think that is, it's not a thing where you have to take everything. You don't have to match Apple step for step, but I think you can do it in a thoughtful manner.

With pCalc, I don't have very much glass in the main interface at all. And in some of the other apps, I am looking at adapting that and bringing in stuff like... I did a... a test with the most important app about by Peacock. And I did a full glass redesign on that to kind of see what would work. And what really works is...

A burning banana moving behind a glass button looks cool. I mean, I will say. And then in Dice, I think I can do a lot of similar things, and I think it actually makes sense if you can see through some of the things, and you might see a... a dice that's like slightly under control or something. I think that could be useful. pCalc, you know, I'm going to do probably a silly complete liquid glass theme for it where everything is sort of smooshing around, but...

I'm not going to make that the default. Now, Charlie, for you, this is the first time you've gone through an Apple visual overhaul as a developer. What was your experience and how are you feeling? Yeah, I... I'm intrigued talking to people who've gone through the other platform shifts because I feel like there's been really three major design overhauls. There was iOS 7, which felt extremely dramatic.

immediately your app looks very old and you need to change everything about how it looks like across the whole board inside of your content everything right and then the next big one and the one i remember going through as a user more is when the iPhone 10 came out and we had the notch and we had, we lost the home button there, that was very disruptive to a lot of apps, right? A lot of apps were totally broken after that. This one felt more like we got a new material to play with.

There was some new ideas about button sizes and some ideas about hiding more buttons that are in your toolbars. Although those are more just choices that you can make. But for the most part... For me, it wasn't really that big of a development effort. A lot of my app is custom, but the parts that would use Liquid Glass are the elements that kind of sit on top, and those are kind of standard.

either standard controls or they're made to look like standard controls. And for that, it was pretty easy for me to just swap those out for liquid glass elements. And to be honest, I feel like it... looks and feels a lot nicer more or less just because you know there's that like a glow effect whenever you tap on a button so you get a little bit more instant visual feedback and then i i do have a big kind of

like what I think Casey was talking about, like a redesign that I was going to do. And I'm like halfway through and I decided not to try to get that out on day one, mostly because they were changing things constantly. And so... I'm glad I didn't do that. I'm kind of still waiting to see how things settle as far as how they ultimately render those so that some of my more extreme changes can fit with everything. But for the most part...

It didn't feel horribly disruptive for me. You know, as you say that, I think I use your app, Dark Noise, constantly, but I'm almost never in the user interface. Yeah, yeah. And I imagine that gives you a little bit of leeway here, right? Yeah, the biggest thing is, so the app is essentially an audio playing app. So there's a...

similar to Apple music. There's like a now playing screen that can be full screen or you can shrink it down. Uh, and I have like a full kind of liquid classified version of when you shrink it down and it's got all the jelly kind of effects and it wobbles around and everything, but.

uh that is something where it kind of needs to blend into the app and if especially earlier in the beta cycle apple was constantly changing how that's rendered and sometimes it would start looking bad in a way that was a problem for me and i was like i'm just gonna

wait until this kind of settles down a little more. Casey, I think anyone who has followed you on Mastodon or hearing you talk about this on ATP, you are still working on your redesign for a call sheet at this point. How is that going and where are some of the pain points that you've encountered? The pain points are mostly in the mirror.

Uh, what I started down the path of, so let me back up. So call sheet, uh, the 10 second version is it's IMDB by somebody who actually respects their users. And so it lets you look up, you know, like movies, films, movie shows. and cast and crew. And it is.

fairly vanilla from a user interface perspective insofar as I'm not using a bunch of esoteric, like wild custom stuff. It's for the most part, I'm mixing and matching Apple controls. And so on the surface, you would think, oh, this is a very easy thing to do with liquid glass. But because of some esoteric choices that I'm not even that esoteric because of choices I made, some things are actually a little harder than you would expect. And so I've been working on it for.

very hard for the last few months. And what ended up happening was I allowed it to spiral out of control and turn into a medium-sized refactoring slash redesign think of what marco did with overcast but not quite that dramatic um But it's allowed me to fix some longstanding foibles and things that annoyed my users about the app that weren't deal breakers, but were frustrating them.

And as I sit here today, you know, we're recording this toward the very end of October, and I really do think it'll be out in the next couple of weeks. Famous S words. But I think I'm close. But the problem was is that when you have a visual overhaul, from the system side that all but requires a visual overhaul, particularly if you're using vanilla controls on my side. And then once I'm doing that, it's like, well, I'm already in here.

i fix this thing well i'm already fixing that thing why don't i fix the other thing and it's the traditional you know classic scope creep but the problem is as much as i was snarking earlier the only person i can blame is the guy in the mirror because i'm allowing the scope creep to happen all that said i'm really happy with the place

call sheets in right now. Like I said, I'm on the precipice of shipping, but it's been a long summer and beginning of fall as well. So I'm not proud of the fact that I haven't shipped yet, but I am proud of what will be shipped when the time does come.

Charlie, have you gotten any feedback from users about iOS 26 or Liquid Glass in general? Do you have any sense of how people are receiving it out there in the world?

I was really bracing for that. And I don't know that I've gotten anything specifically about the liquid glass part of it. I've gotten people very upset that I added... ai and any sort into the app because i'm using the foundation models uh and i've gotten you know airplay issues and bugs that are probably ios related that i haven't been able to fully hunt down but

they all spiked up you know right after the release came out but yeah on the liquid glass side uh relatively chill most of my users are pretty like They're usually on the betas and they jump on things quickly. And so maybe that's part of why. But yeah, it hasn't been super disruptive, at least through my support channels.

Yeah, I'm going to answer my own question. That's been the case for our apps as well, like Panamera++ and Widget Smith and Sleep++. Really got no meaningful feedback directed at... liquid glass or ios 26 right we've gotten some feedback about some other things but about sort of that core you know what the redesign brings uh it's been really quiet for us too and

Like you, I'm a little surprised by it. I mean, you know, the community reaction was so strong this summer and parts of our community is still pretty strong. But it seems like now that it's out in the world, like...

I don't have friends coming up to me complaining about it the way I did the summer of iOS 7 or other changes that have happened over time. I'm sure you guys get this, right? Like, why did Apple do this? Because they know that I... vaguely know about apple um so i think it's gone down smoother than than a lot of us expected which is uh which is good that has been my experience like anecdotally talking to people normal people in my life as well

generally people do notice it and they're like oh this one thing is kind of weird but the overall reaction is like oh that's pretty cool and then that's that's it Yeah, I think for most normal users, it's just not that big of a deal. I mean, I think we've all seen that talking to wives and children and friends.

This is a big move, though, for Apple. And one of the questions I was curious about is where do you see this going? Apple often evolves after a big move like this. We saw iOS 7 evolve for several years. You think it's going to get more or less extreme? Casey, why don't you go first? What do you think this is going to lead to?

So again, we're recording at the end of October and in the latest 26.1 beta, I think it was like beta two or something like that. Apple now has a toggle buried somewhere. One of you I'm sure remembers where, but basically you can toggle the translucency kind of. either super translucent or maybe bring that about.

50% back, you know? And I think that they're probably going to look at that, particularly if and when it reaches wide release and see how many people are ticking that checkbox. And if most of their users are saying, this was too much, bring it back, please, then... then I think that's a hint to them that they need to adjust. And certainly you were talking about iOS 7 earlier, and iOS 7 was striking.

in the beginning and i liked it for the most part but like you know where we were used to buttons being these very large touchable or lickable if you will objects on screen they were just different colored text and that took that i mean even as a very experienced computer users i mean i grew up on the computer uh i still found that frustrating and troublesome sometimes and so they eventually walked that back and made things you know made buttons a little more obvious

made things have a little bit of depth and texture to them to a degree. And I suspect that we're going to see the same thing with liquid glass, that they will evolve, adapt, adjust as they always do. And as any good organization will, I don't think they're going to absolutely.

can it and throw it in the bin or anything like that but i think they will absolutely adjust it into to plus one what james said earlier i like but don't love the glass but i really really do love the liquid and some of those animations and affordances that they have for the liquid pieces

are very clever and very, very well done. And so I hope that they kind of dial in the glassiness of it, the translucency of it over the next month or two or maybe more. And I hope they don't abandon the liquid pieces because I do think that's very novel.

So let's switch gears a little bit and talk about some underlying technology. I want to talk about SwiftUI. SwiftUI, for those in the audience who are not developers, is a way, starting in 2019, Apple introduced this. a new way to build user interfaces across Apple's platforms. Before this, the Mac had AppKit, which routes stretch back all the way to the next step days.

iOS and iPadOS used UI kit. There was some weird watch stuff that changed over the years. And in the years since, SwiftUI has been added to and has evolved. just like the Swift programming language has in the background as well. And Charlie, I want to start with you. How much are you using Swift UI in your apps? And if you do have... in a UI kit in your iOS apps? Like, do you see moving past that over time? Yeah, so dark noise was built entirely in UI kit originally.

And it's kind of at a point now where if I build anything new with a new UI piece at all, I will almost definitely do it entirely in SwiftUI if possible. There's still a couple areas that... I want to convert things over from UIKit to SwiftUI and I can't quite get it there. But for the most, or I can't quite get it there in a way that's not incredibly risky for breaking. But for the most part, it's...

It's at a point where that is definitely my preferred way to work. And then Framus is my Mac app, and that came out last year. I had never done AppKit development at all. So I just started with SwiftUI there. And that app is a little different than Dark Noise because it's very much a utility app. So I just went totally stock, you know, standard Mac OS style app.

And in that environment, it was, I thought it was amazing. Like it was way faster to work with than even my UIKit stuff. Because I am using Mac Catalyst for dark noise on the Mac. So that one's all in UIKit as well. But for Framus... everything swift ui and i i genuinely really enjoy it um at at the bigger companies i've worked at in the past we've ran into some performance issues but those are like at more of the fringes but for uh

Kind of a standard productivity app, indie app kind of thing. I feel like it's been really, really stable at this point. What about you, James? So I'm, as I said, alluded to previously. grumpy old man side of this um the uh pcalc and dice are both uh predominantly ui kit and written in objective c so they're kind of like dinosaurs of a sort, and the Mac version of Peacock is kind of written in UIKit in that I basically...

ported everything over. So some of the classes at the bottom are different for drawing my buttons and stuff, but it's mostly the same code. Newer stuff that I've done, like the... Vision Pro PCALC About screen is entirely SwiftUI. And I use SwiftUI for anything where you're touching widgets or stuff like that. basically there is no option i don't enjoy swift ui as much to write in and i think that's because my brain you know sort of solidified at the age of whatever and uh

It's harder for me to move my mental model over. I can do it. That's why I did the about screen like that. It was to kind of prove to myself that I can actually work out.

how all this stuff works and do it and but it do i enjoy it as much no i feel i've got more control when i'm using something like ui kit and you know i can kind of guarantee what the performance is going to be like which is the main problem i have with swift ui is like i will design something badly and it will be refreshing the entire ui and i don't quite understand why and things like that and it's it's

Hard to debug those kind of problems, I've found. But again, this could just be a matter of age. You know, to jump in just very briefly, I think that there's nothing wrong, and I don't particularly disagree with anything James said, except the whole age thing. You're selling yourself short. You're an excellent developer. You absolutely could.

dominate SwiftUI if you wanted to. From my perspective, you know, I came up on UIKit and what I like to joke about with regard to SwiftUI and UIKit is the SwiftUI is the answer to, you know, UIKit's perspective of why write 10 lines of code when 100 will do? Because UIKit, for all of its many perks, of which there are many perks, it is

incredibly verbose, particularly as compared to SwiftUI. And SwiftUI is much more succinct and has a much more modern feel to it. In some ways, that's good. In some ways, that's bad. I think what I like about SwiftUI, or one of the things I like about SwiftUI, is that it is so much faster and...

to my eyes easier, although perhaps James would disagree, and I think that this is probably from what you're familiar with. But either way, it's much easier and definitely faster to spin up a whole new portion of an app, or the user interface thereof anyway, with SwiftUI because it's It's so, so little code and so...

It's quick to write. There's very little boilerplate. There's very little compulsory, you know, like state management and stuff like that. And I'm grossly oversimplifying. So if you happen to be a developer listening to this, I understand I'm oversimplifying, but it's.

much faster in so much less code and that's something that i really really really like about it it seems like one of the reasons swift ui exists is that apple no longer is just the mac company they have all these platforms and they want you know you guys making software for all of them and the one of the underpinnings of swift ui seems to be multi-platform right you figure out how to write in swift ui then you can easily adapt your apps

to everything from the watch to the Vision Pro. Is that sounding true for you? I guess, Charlie, why don't you go first? You've got your app on a few different platforms. Are you finding it easy to jump between them? Yeah, it's funny the way you phrase that, because from my perspective, you know, I mentioned I'd been doing Windows phone development and a lot of enterprise software back in the day in different programming languages.

And so when I first came over to starting to do iOS development, UIKit felt like I'd traveled back in time. I hadn't done a non-declarative UI framework in who knows how long. And so to me, the first thing, my very first WWDC, I was like, man, it'd be so cool if they gave us a declarative UI framework. And then that was the year that they announced SwiftUI.

So I almost feel, I don't know the internals. I've heard that the original story was building a framework that'll work better on the watch. But yeah, I felt like it was more like modernizing. UI development on Apple platforms. But I think what you said is how I feel in terms of... the reality of what SwiftUI has enabled outside of what Casey said, like being able to build things and mock things up really quickly, which is really nice. The portability of the UI that you declare.

in swift ui is extremely nice and that's both for like building on the mac and on ios and on watch os but it's also like the way widgets work and the fact that those can just i mean they just keep popping up everywhere

every WWDC widgets are just going to show up somewhere else. And that is kind of enabled by this framework as well. And that to me is one of the really powerful things of... of swiss ui in terms of enabling smaller development teams to to push the same stuff all over the place Yeah, I could go down a rabbit hole with App Intents and widgets here, but I've already been told the outline is too long. Casey, what about you? You have call sheet on iOS, iPadOS, and VisionOS, which is...

Exciting. We're talking more about VisionOS in the member chapter at the end of the show. But what about you? How has it been to jump between these platforms? Ah, good asterisk. I think for the most part, it's pretty good. So I am on iOS, on iPadOS, on VisionOS, and I'm also on the Mac, but in the iPadOS on Mac mode. So it's basically running the iPad.

app within the Mac. It's mostly good. I think part of the reason that I haven't spent a whole lot of time looking at iPad in particular and making a truly bespoke version for the iPad is that the... The sales pitch of SwiftUI is, oh, you can run the same stuff everywhere, which on the surface is true.

And I mean, to be fair, the iPad and Mac apps have had almost no customization done on my part. Vision OS is a bit more aggressive in the customization because it's such a wildly different paradigm. But one of the things I'll say is that... If you're trying to reuse literally the same component across different platforms, and maybe I'm just holding it wrong, but in my experience, it gets kind of gross and there's a...

term that developers use called code smell. So things just get kind of smelly. Obviously, that's not in a literal sense, but it just has this kind of like ick to it. As you start really embracing the idiosyncrasies of each platform, which a good platform citizen will, I feel my code at least.

gets a little bit more of a code smell. Now, one of the ways you can avoid this is by really embracing one of the core tenets of SwiftUI, which is composition. And what that means is, rather than having one piece of code that does an entire screen, You have a piece of code that does the entire screen, sure, but it does that by putting together this bit and that bit and that bit and that bit. And then each of those bits has subcomponents that you're amassing and putting together.

composing and and if you do a good job of that which all all the developers i know aim to do and very few of us actually do then i don't think it's quite so bad but a lot of times you know i'll speak for myself for better and for worse I meant when I'm writing call sheet, it's an iPhone app. Yes, it works on the, on vision OS. Yes, it works on iPad. Yes, it works on Mac, but to me it's an iPhone app and that's where I started from. And so I write everything.

from the perspective of the iPhone. And then I have to go back. And again, this is kind of my own fault. I have to go back and then kind of. uncouple, decouple, whatever the right word is, some of these things, decouple, I guess, decouple some of these things and make their own components out of them so I can mix and match a little bit differently and in a little bit better way for each platform. So, yes, on the surface, it does make multiplayer.

platform development easier and better, but it's not the panacea or whatever that I think Apple kind of claimed it was. Surprise, surprise. This episode of the Mac Power Users is brought to you by Indeed. Go to indeed.com slash MPU today and join more than 3.5 million businesses worldwide using Indeed to hire great talent fast.

Right now, there's a talented person out there who could take your company to the next level. Do you want to hope they see your job post before your competitors do? Or do you want to match them with Indeed Sponsored Jobs? If you're hiring, Indeed is all you need. Stop struggling to get your job post even seen on other sites. Give your job the best chance to be seen with Indeed's sponsored jobs.

They help you stand out and hire quality candidates who can drive the results you need. Sponsored jobs boost your posts for quality candidates. So you can reach the exact people you want faster. And it makes a big difference. According to Indeed data, sponsored jobs posted directly on Indeed are 90% more likely to report a hire.

than non-sponsored jobs because you reach a bigger pool of quality candidates. Join the 1.6 million companies that sponsor their jobs with Indeed. I do a lot of production around here. Let's say I needed to hire a new editor. In that case, I would go to Indeed and I would be super specific. Not just can edit audio, I'd say I need someone who's edited conversational podcasts for at least three years and has specific knowledge of specific software.

By making it detailed, I'm going to narrow that list down to just the right people, and I'm going to make the right hire. Plus, with Indeed Sponsored Jobs, you only pay for results. No monthly subscriptions, no long-term contracts. Just a boost whenever you need to find quality talent fast. People are finding quality hires on Indeed right now. In the minute I've been talking to you, companies like yours made 27 hires on Indeed.

according to Indeed data worldwide. So spend more time interviewing candidates who check all of your boxes. Less stress, less time, more results now with Indeed sponsored jobs. And listeners to the Mac Power users will get a $75 sponsored job credit to help you get your job to the premium status it deserves at indeed.com slash MPU. Just go to indeed.com slash MPU right now and support the Mac Power users by saying you heard about Indeed on this show. Indeed.com slash MPU is the URL you'll need.

Terms and conditions apply. Hiring? Do it the right way with Indeed. And our thanks to Indeed for sponsoring the Mac Power Users and Olive Relay.

Up next, we have a bit of a hodgepodge of questions. I want to start with documentation. So each year we have these changes. Apple has an extensive... developer website. And part of those web pages are supposed to be information about new APIs, new frameworks, changes in the OS. Apple has been criticized.

by developers over the years about the quality, and I would say quantity, of its documentation as well. And James, I want to start with you because you've been on Apple's platforms a little bit longer than the other guys.

Where do you think the state of Apple's documentation is now? And could you put that in context of sort of historically where they've been? I suspect the problem is... that the scale of Apple's APIs is 10 times, 100 times what it was, let's say, in the 90s or the 2000s or whatever. There's just so much more, but there's still probably, you know, like, I don't want to give a number, but you know, like, let's say 10 people in an office somewhere that are doing this stuff.

and they have to do the best that they can. And I don't know if it's 10 people, but it sure feels like 10 people. And it's not getting better. You get whole new things that appear in iOS 26, and you're sitting in Xcode, and you select it. And it's like, no, I don't have any information about what this does. Well, let's look at the headers. No, there's actually nothing in there.

Maybe there's a WWDC session where something was mentioned, which you have to go and track down and see what you're supposed to do, except they've changed the API since WWDC. And it's like, you know... I feel, you know, like with cryptic crosswords, you kind of like, you can dial into it. So you get a sense of like, I know how these things are done. I know how they're written. So I probably get a sense of...

How this is supposed to work, let's try it. That seems to work. Okay, let's go with that. But it is not getting better. And I feel that Apple should spend some of its three trillion whatever money. on getting a large team of people to support developers and write documentation. You know, it's like, give me like 100 million a year, something like that. That's nothing.

to apple and i will build like the best team of uh i won't because i'm not gonna do it but you get the idea like i i feel it really feels like apple is being cheap on the developer side. And given that we give 15% to 30% of everything we make to Apple, I would like to kind of see that spent on infrastructure and stuff. that would benefit us. Yeah, I think David Smith had the statement earlier this year that developers are some of Apple's biggest customers.

And I'd never thought about it that way, but it like turned a switch in my head because you guys do pay a lot of money to Apple through those commissions. Yeah. Yep. Casey, how do you feel about Apple's documentation? This is a mixed company podcast, right? I'm supposed to be using kind words. My mommy said, if you have nothing nice to say. No, actually, it's getting better. Back in 2020, I got a real burr up my hindquarters about this.

I wrote a blog post, which I'll leave it to you two to decide whether or not you want to link it. But I was very cranky. And I wrote a blog post, which I stand by, about Apple's piss-poor documentation. And at that point... It seemed like every new API Apple released, again, this is late 2020, every new API Apple released, you would see the words, no overview available, which was Apple's way of saying, I don't know, we'll get to it eventually.

And that is incredibly frustrating because of the one breath and, you know, talking out of one side of their mouth, Apple will say. developers, you folks really need to use new hotness. Look at all this new stuff. It's great. You should try it. We really want you to do it. It'll be great. I promise. And then the other side of their mouth is saying, what do you mean new stuff? I don't know anything about that. You figure it out. You're smart. You do it.

And it's incredibly, incredibly frustrating. Now, over the years, it's gotten better, but... And it's easy for me to armchair quarterback. But if I were, you know, like James was kind of alluding to a moment ago, if I was the the boss of Apple's developer teams, I think what I would require is you don't get to ship.

new API until there's at least reasonable documentation associated with it, which is a big ask, especially towards summertime when everything is on fire and you have to move fast, fast, fast, go, go, go. But ultimately it's whether or not do you, do you value your. third-party developers or do you not? And again, to Apple's credit, since 2020, in the last five years, almost to the day, we're a couple of weeks shy of five years from that blog post.

It's gotten a lot better. It's not perfect, but it's a lot better. And so I'm not nearly as feisty about it as I once was. But every time I open a bit of documentation and I see no overview available, just. And so it never quite goes away, unfortunately. I'll take up that feisty baton for you. You sound like that you're in a better place, but I will take the anger and just use it.

Charlie, what's been your experience here? Yeah, this is an interesting one because my role at my day job is a developer advocate, which is sort of... living in this space and so i think about this a lot from my company revenue cats perspective uh and steven you probably can say this better than me as as one of the actual users uh but

We're often praised for our documentation in particular. I think structurally one of the reasons that our docs work is I like to think of documentation as having like three layers. there's the like every single method that you release having a description for what that is written down and that's what casey's talking about where there's there's often Apple docs that literally say nothing. Or if they do, they just repeat the name of the function. Yeah, exactly. Which there's a lot of that still.

And then there's like tutorials where it's like a full-blown let's walk you through everything. And then at the very high level is... This is a framework that we're releasing. And here's an explanation like high minded why we want you to use it and how you should be thinking about it. And those are the WWDC videos. I think Apple excels at those.

They're very good at them, but those are the most expensive to produce. They're the least frequently updated. And in Apple's case, they come out once a year in huge batches. The tutorials are really light because... you know they again they only come out kind of once a year and there's not really a lot of them they do do code samples but by about halfway through the uh the os season most of those code samples don't

like run anymore i don't know if that's still the case i often don't even download them a lot of times because i'll go find external blog writers and stuff to find the docs on these things really um but at that low level that At RevenueCat, at least, that is where we have a requirement that the team that builds the product has to write those. That can be the engineers themselves, or it can be the sort of product manager that's over them.

But the idea is they're literally the only people in the world that actually understand what flipping a bit or setting true here, setting false here really means. And so they're the ones who ultimately should write that. And my understanding is that is absolutely not the case at Apple. And it's very like against their sort of worldview of like everything should be running through marketing and PR essentially. But I think unless they get to that point, we're...

Even if you, you know, James headed up the $10 million team of amazing documentation writers. I think it was $100 million I wanted. Yeah, I can't do it. $100? Oh, well, okay, maybe $100. No, but like... Even that mythical group, they would spend all of their time badgering the engineering team to be like, so what does this actually do? What does it actually do? And maybe they would get there, but it would take them 10 times longer than if the team that wrote the thing.

just wrote down a little explanation of what why they did what they did but the problem there is apple being apple are they going to leak either leak future things there accidentally or say something in a not PR friendly way. I think that I don't actually know. I don't know the people doing this, but my guess is that that's where a lot of the struggle and strife internally at Apple is and why we kind of get this result out of them as far as docs go.

They have a whole team, you know, dev pubs or developer publications. I don't know. I don't have a good view as to where the delineation is between. the pure developers, if you will, and dev pubs. But my gut is that Charlie's exactly right, that if I'm sitting there writing a new feature, a new API, I'm throwing it over the wall to Charlie to work on or James to work on or whatever. And then it's up to them to do the...

like sleuthing to figure out, okay, what did Casey just write? How does it work? And to Charlie's point, occasionally needle me about, well, can you help me with this? Can you help me with that? Meanwhile, my hair is on fire because I'm trying to ship before WWDC and so on. So I couldn't agree more. with really what everyone's said so far, but what Charlie was saying a minute ago, that I suspect part of the problem is this delineation between

you know, those who write code and those who write document documentation. And I get on the surface, you know, some people are better and more talented at writing documentation. Some are better, more talented at writing code, but. Ultimately, I feel like you need to start from somewhere and the person or persons who wrote the APIs, who wrote the code should at least take a crack at explaining it and then perhaps have someone who really knows how to do technical writing come in after the.

fact and make it sound good yeah i mean i just want to comment next to some of the like the some new um you know property has turned up and it's like what does this do You know, like, just give me like the three lines, you know, three sentences or one sentence that says, you know, set this if you're doing so and so. But so often you'll come, you will find something that's new.

You know, it's new in Mac OS 26, and it's like, a lot of it is just trial and error. Like, you know, how do I do this? Well, let's set this to true. What happens? Oh, it breaks everything. Okay, let's not do that. It's interesting to me because... I've never thought about it in this way before, but one of the things I'm hearing from you is that this is maybe an offshoot of the small teams at Apple. Like Apple has these small teams that make this software.

They're not huge teams building these apps that we use all the time in the operating system. And because of that, the people with their hands in the gears are not doing the documentation. There's just not enough of them. Yeah, I mean, I think there's a lot of teams in Apple that are much smaller than people will assume that they are, given that you're deploying code to however many billion devices. It's not like...

What was it? There was something that was like the Facebook app, 600 people working on it. I think that was a figure. And I just cannot comprehend that.

All right, let's move on, though. I think we've hit that one. The app environment has changed over the years. I remember when I first got apps I could buy for my phone, they were all a few bucks. I paid for it, then I owned it. Well, that model's not here much anymore. Now it's all about subscriptions. What do you guys think? Charlie, why don't you go first? I think you've been subscription from the beginning, right, with your app?

No, I actually switched in 2023. I was paid up front until then. Okay. Yeah, I forgot that. You're right. Well, that makes you even better for this question. What was it that led you to switch over and how's that going? Yeah, like ultimately it was, I mean, the classic tale of it. I had a decent sized user base that was continuing to use it and there were new people coming in, but it was that number was getting.

uh that ratio is changing quite a bit you know and i i i wanted something where i could keep working on this with ultimately the idea is i would love to grow this to an app that could be my full-time job at some point And subscriptions are, again, all this has kind of been beat to death, but it's a nice pairing of, you know, if you're a user that wants the software to continue to exist and you keep paying for it to continue to exist and continue to be updated.

uh then you can do that and the developer can can feed their family and and keep working on it um and for me for the most part i think it's gone well it hasn't been like this runaway train where now i'm making a crazy amount more money um it did enable me to add like a free free tier so i have quite a few users now who i probably wouldn't have had who are just using the free version and you know i'm more than happy for that to exist but uh

Yeah, I think that's kind of where I'm at with it. It's weird. I work at a company that's all about monetization and subscriptions. And I can go into it from a business perspective all day long. But as a developer and a user... It's just kind of the model that makes sense to me. And whenever I have a piece of software that integrates well into my life, and I only pay them once, I honestly start getting worried. And I sometimes even start looking for a different piece of software just because.

I want something that is going to be here for the long run. Yeah, this work is hard and you can't work for free and stay at it. Well, so I've got a paid upfront app. All my apps are in fact. And if you bought pCalc in 2008, then you've paid me $9.99 once. Oh, gracious. Don't get me wrong, I've done extremely well out of that. And it continues to do pretty well. But at that point, you're counting on growth.

to kind of keep the money coming in. Sort of a pyramid scheme. And it has done well. As a developer, I can see the logic in wanting to have the regular income. Something that you can rely on so you know roughly how well you're going to be doing. And the motivation that you want to keep those people happy. As a user, I absolutely hate subscriptions. I really, you know, I pay for call sheet. Call sheet is priced at a level where I'm like...

I'm quite happy giving Casey this money. And it's because it's also a service-based app. It's not like a utility like pCalc. I find it... I would feel bad charging a subscription for something that basically doesn't connect to the internet, doesn't do anything. It just sits there and, you know, adds up your numbers. And with something like big apps, I really resent paying Photoshop. I still pay it. But I liked the model.

where maybe I don't update every year. You know, this thing is working fine. I will use this for two to three years. Then... It's like, oh, well, it's kind of not working great on this system. I will update to this new version. And that gave me the control of saying, well, I'm going to pay for this and I'm going to use it. And then I'm...

At some point down the line, I'm probably, because I'm a happy customer, I'm going to give them more money. And that's the model that I like. And I don't like the, we're going to take your money. And yes, you're... The deal is that you get the upgrades, but it just removes the control. And I feel there's a lot of subscription burnout at this point where people are like, why have I got like...

you know, 30 subscriptions on the go. I added up every subscription I had across apps and services and things. And it's just... I get why Apple is wanting to move to that model because it gets Apple regular income as well. You know, it's not just the developers that get that nice predictable monthly. income it's apple does and that goes into their services revenue so they are financially motivated to promote that model

And that's why I think we never had anything like, you know, the famous developers asking for upgrade pricing, which never happened. So, yeah, I don't like them. I understand the developer. I mean, I would like more money. Let's be honest. I would like more money. But I just, I can't. They come back to me in a month when I've announced that Picalk is a subscription. Right now, James of 28th of October doesn't like them.

Yeah, I'm probably one of those guys that paid you $10 like 20 years ago. I know. And it's great so long as everybody is telling their friends and those friends are buying it. And, you know, there's enough people in the world that... you know, $10 from everybody is, you know, perfectly good. But yeah, it's hard to do that. That's the thing is to build on what James was saying that.

It was kind of a pyramid scheme early on, and it worked. It worked for everyone involved because the iPhone was growing and growing and growing and growing. And you could always find new users for whatever ridiculous app you had. Good apps, bad apps, didn't matter. You could always find new users. After a few years, it was quickly, quickly obvious that that wasn't going to be the case anymore and you needed to rethink things. And I think that for users, if they want to have their app.

continue to work and that might mean to james's point because it's associated with an online service like call sheet is where i'm i have to pay the movie database in order to get access to their data Or alternatively, if they just want their app to be regularly updated for the latest versions of iOS.

That work needs to happen. And if you can't find new users all the time in order to fund that work, then guess what? You're going to have to go to your existing users. And while I echo what James said, like I don't particularly care for subscriptions. I didn't want to do CallSheet as a...

subscription but as you do the math it is quickly extremely obvious that that's the only way the call sheet would be solvent and i don't think there would be a call sheet anymore if it wasn't subscription and it's only been out for two three years something like that so it's it's a very slippery slope not to go that path these days unfortunately even though i don't disagree with anything james said i mean i think i 100 get it when you have to pay somebody else you know it's like

Then it's just financially, you know, you have to do that. But I just, I don't know that it's a fit for every app. Although I'll say though, James, I remember. kind of before the subscription swing. A lot of times I would find an app I like and it would have a nice set of features, but then... development would cease on it because the developer at the other end realized that, you know, there's no bang for his buck to add the latest feature. So every couple of years.

somebody would come with a new take on the exact same app, but they would include the latest and greatest features from Apple. And if I wanted, I had to switch to a different app just because the other app... would get abandoned based on the the one and done purchase model and that's frustrating as a user right you know every couple years you got to go find the new good version of this app because the old one was good

But the guy couldn't make any more money on it, so he stopped working on it. Part of it is a limitation of the App Store business model, that you can't do paid upgrades, right? Yeah. Some people have figured out things kind of like it to do, but clearly Apple didn't want to continue that business model into the App Store era.

And there's things like, I would love if I could merge the Mac version of Peacock with the iOS version, but there's no way of doing it without doing a new version and basically orphaning all the previous purchases. And it's those things where, you know, it's their business goals, I think, get in the way of sometimes what developers would like to do.

The hot word of the day is AI. And we mentioned earlier Apple's struggles in that area. With iOS 26, some of their foundation models have come. to the OS and can be run locally through shortcuts. You can also talk to Apple's private cloud compute. And I would love to know if any of y'all are investigating this, if any of you think that these tools are a good fit for what you're building. Charlie, let's start with you. Yeah, I added it.

almost immediately and it took like the api for apple's foundation models are is genuinely one of the like nicest apis they've come out with in a while it's extremely well done and uh I have a feature in my app where you can take multiple sounds and build a mix, you know, with different volumes and everything. And in like, I think it took me like 20 minutes, maybe.

maybe less than that i think i was on a plane on the way back from from cupertino and i was able to to add a feature in where you could just type a prompt uh and apple intelligence offline because i'm on a plane was able to build a mix based on all the mixes or all the sounds that i had in my app and then and then dump you into the interface where you could then customize it from there um

I think for stuff like that, where you're not asking it to do some crazy advanced thing like the really big models that are running online, but just little stuff that happens locally. and maybe even interacts with something that you have a UI to do in your app already. That kind of stuff I think it's really good at, and it's surprisingly fast at. So I've been pretty impressed for features like that. Casey, what about you? I don't see a place for it in call sheet, but maybe you'll surprise me.

Yeah, I haven't had the time to properly kick the tires on it. I have heard from many people what Charlie said is exactly right, that the API is excellent, which is great because a lot of times APIs that Apple makes are at least... good, but I wouldn't necessarily often describe them as excellent. And this one from everything I've heard is excellent. But anyway, to answer your actual question, I don't really have a lot of place for it that I've figured out so far. I think this is also...

My enthusiasm is attenuated by the fact that I think AI has some very clever and interesting and useful uses, but I don't think that everything under the sun is one of those uses. I do think that AI can be great. And I think that what Charlie described makes a lot of sense and is a really clever way of using it. But for CallSheet, I think having a natural language way of describing what you're looking for could be helpful, but I'm not sure how I can really make that work with the...

movie databases API. And I mean that in every sense of the word, both in that literally, I'm not sure how that could work. And figuratively, is that even possible? You know, I'm not sure if that could work at all. It's definitely on the list of things to explore as soon as I get this iOS 26 thing out the door. What about on the development side? James, are you vibe coding the next version of Peacock? No.

I'd written down a comment, which is in our notes, which just in the general AI sphere is, I hate it all and I wish ill on everybody involved. Which I feel needs some nuance because... You know, if you leave aside the ethical, environmental, moral, whatever problems with AI, which is a big ask, but, you know.

Some of these systems are very good for what I think of like sort of repetitive brainless tasks, like rewrite this routine from Objective-C to Swift or convert this gradient code from an SVG gradient to a SwiftUI gradient. And it can do that stuff really well, and it can do it in a, you know, it's saving you time. I don't think...

Like when I've had problems, like I'm doing some 3D graphics thing and it's like quaternion, like whatever, complicated math. And I'm like, I ask GPT or one of these systems. And it confidently gives me an answer which doesn't work at all. And my main problem that I have with AI is I feel the learning. process is the whole point of this job. And it's like, the reason that I do this stuff is to get better at doing this stuff so that I can make better things in the future.

So I'm like learning and I'm training. And I find the more you lean on this stuff, that the... the less you've actually learned about what's really happening and some people will say well what does it matter if i get the thing done quicker and you know whatever and and and i say well i like doing it my way so james what you're saying is while flaming bananas are cool self-aware flaming bananas are not cool i mean like look

Don't get me ranting about the whole LLMs and will they lead to AGI nonsense and how this whole thing seems to be, since we were talking about pyramid schemes earlier. parts of the market propping up other parts of the market in very suspiciously circular ways of investment I would like the bubble to burst completely and then I don't know if you have wreckage left of a bubble when it's burst, but there is useful stuff here. And I think, so my law of AI...

is I think anything that turns a complicated thing into a simple thing is good, and anything that turns a simple thing into a complicated thing is bad. So like... Stuff where you're doing speech recognition or vision recognition that's got like an accessibility aspect to it. You know, even just the being able to select text in photos thing.

It's so good. Or the real-time subtitles for anything. All of this stuff, great technology. I think that's good. I wish to write a novel based on one sentence. Bad, very bad. Do you guys think this is overblown? I mean, as developers, you guys see this in ways that typical users don't. I guess Casey, why don't you go first? Yeah, my...

I could talk about this for hours. I think my short, short version is in limited use cases with strong guardrails, I think AI can be extremely useful. I mean, the website for CallSheet, which is CallSheetApp.com, that was 90% vibe coded. it was mostly vibe coded. I am capable of writing everything that's there, but it's not something that I enjoy and web development moves so fast that I don't.

keep up with it very well. And so I figured it would probably be a good use case for it. And it was, there were things I had to fix and change and whatnot, but by and large, most of the marketing website for call sheet is vibe coded. I think instances like that, it's a pretty good fit. But I also echo what James said that.

I don't think it does anyone involved. I don't think it does well for anyone involved when I'm asking it to do my job for me. And I think that for us, our unique talents are making decisions.

and implementing those decisions. And in certain cases, in certain limited cases, maybe the implementation is less important, something very simple, like a setting screen or something like that. But by and large, I think what makes us unique is our ability to... conceive of a thing and build the thing and build the thing where I think that so many people are looking to AI to just do their jobs for them.

And I don't think that's healthy for anybody involved. If you wanted a longer take on this, it's not very long, but Anil Dash, like a week ago or something, put up a really incredible post about his opinion with regard to AI, which I think I pretty much... echo entirely insofar as it's it's not the solution to all things for all people and That's okay, but we should really stop trying to ram it down everyone's throats as the solution to all things for all people.

And I'll point out that Amazon just said today that they're laying off, was it 1,500 people or something, so they can focus on AI? 15,000. Let me get that right. That was 15,000 or 1,500. Job loss and disruption from this is real. I totally agree. I have my doubts as to the upper limits of this as an LLM, but... It's certainly got some use and is changing the way people work and who does the work. Charlie, you got any big picture ideas on AI at this point?

I mean, like everybody, I could talk about this for hours. To your original question of is it overhyped, I think the answer can be yes and no at the same time. Are we in a bubble? Yes. Is this maybe the most disruptive technology in the last, I don't know, 20 years or something? Or since the sort of cell phone revolution? I also think that might be yes. Certainly in the way it's...

It's impacting all the people around me in St. Louis, Missouri, who do not know that Sam Altman, who he is, or that he's also from St. Louis, Missouri. But it's impacting their lives in ways that are both... disruptive uh i know a lot of teachers and like you can probably already in your head you think of 10 reasons that ai is really bad for teachers but then they're also really excited about all these new capabilities and ways that it helps them out um

with teaching a large class and creating material and all this stuff and so it's everywhere in the way honestly what it makes me think of is the the internet bubble yeah um which to your to your question james earlier uh are there remnants of a bubble bursting i mean i think yes i think it'll be painful uh when it inevitably happens but there is a there there um and so at this point i think it's like both recognizing the negatives and not being like an annoying hype person but also

finding where those opportunities are and where it can be used for good, like you're talking about. The accessibility stories are amazing. And focusing on those and then trying to survive the roller coaster and inevitable crash that will hit at some point. All right, so follow-up question on that for you, Charlie. How is Apple doing in the midst of all this? I've heard people argue that, hey,

They didn't spend the billions on servers, server farms. They're getting the bang for the buck without buying it. If the bubble bursts, they're going to be sitting pretty. And other people saying they've lost the future of the company because they've been checked out on this. I mean, and everything in between. Where do you see Apple? How would you grade their AI involvement so far?

It's such a difficult question. If you're just saying how have they handled the AI revolution or whatever you want to call this, it's hard to say that they've... knocked it out of the park in terms of building AI features. At the same time, they are the platform still. They were not exactly this massively successful worldwide net company.

They built the platform with with the iMac and like leaned into the Internet in a way that let them ride that wave, even though they weren't necessarily an Internet company. Right. And I think that that side of it, they have been succeeding on. everybody i know decent amount of people that work at open ai and all these companies they all use max they all like max uh it's not the thing powering their servers but like

It is the actual piece of metal or glass that you're touching whenever you're interacting with AI for a large amount of people. Okay. One follow-up question for you, Charlie. You're in charge of the AI program at Apple. What's the one thing you would do today? Oh, interesting. I know everybody wants to just say, like, buy an AI company, but...

I don't know how well they would really do with that. I don't know. I feel like I personally am more excited about them continuing to lean on making the devices work. well with all of these ai platforms and i guess continuing to push on local on-device um versions of this even if they're not as good uh because i think that's where they they have a bigger advantage kind of than everybody else

Okay. That's not a good answer. Casey, same question. How's Apple doing and what would you do if you were in charge of Apple's AI initiative? You know, if you look at Apple today, I would say compared to the rest of the market, they're not doing well. But I think that's a relatively simple way of looking at it or a rudimentary way of looking at it. I think.

So depending on how things shake out over the next year or two, maybe Apple is actually doing incredibly well. And what I mean is if this really is a bubble and if it's about to burst, I think that Apple's much more tempered approach. be that by accident or on purpose, might end up serving them. And I want to echo what Charlie just said a moment ago, like their on-device models are a differentiator because pretty much everything else LLM related has to happen on some server somewhere.

That is sometimes true of Apple stuff, but not always true. And as Charlie said, you know, their on-device models aren't as good. They're not as big, et cetera, et cetera. But they're there, and that's better than nothing. So if I were Apple, I would probably just stay the course and try to manage, you know, the market, you know, giving you stick about it as best you can. But stay the course and let that bubble burst if that's what's.

going to happen which i think it is uh if then then let the bubble burst and then you can be there when everything is you know when all the pieces have fallen down and pick up the ones you like and just keep on keeping on and say look we didn't get bought into this hype cycle. You know, we stayed the course and we did what we thought was right. And it turns out it was. Okay. That brings us to the lightning round.

We're going to try to do this quickly. I'm going to start with you, Casey. What is one tip you have for people looking to get into development? To get into development, you know, I get this question all the time and I'm going to try to make this quick. There's a very short blog post I wrote a long, long time ago back in 2014 about this, but it's basically just.

Figure out something you want to exist in the world or your own take on something that already does exist in the world and build it. Figure out how to build it. That will force you to learn how to develop stuff. If you're in...

If you're already a developer and you want to level up, or if you're perhaps an established developer and want to get noticed or whatever, the one piece of advice, I'm really digging in on this over the last several months, that the best piece of advice I think I can give anyone. and it's not just for development, is really properly give a crap. Just really give a crap. And if you actually care about the work you produce, I think that that is...

Very much a differentiator. And that probably comes from a place of privilege. But if you really and truly give a crap and really care. I truly believe that that will get you noticed. Maybe not the way you want, maybe not in the timeline you want, but your work will stand out simply by you really properly caring. You are ahead of so many people if you care. Yep, it really is true.

What about you, Charlie? Yeah, I think what I was going to say is a very similar thing. It's like, yeah, get to the cycle of... build fix build fix build fix as fast as you can i think the easiest way to do that is to make something that you want or make something that somebody you know wants or like

Make a video game or make a whatever. Honestly, if your goal is just getting into development, the language itself is less important. But if you like iPhone apps, make an iPhone app. But the faster you can get into that. feedback loop, the more likely it is you're going to actually push through whenever it gets extraordinarily painful. Because I've been told, I used to say development's easy because I felt like what I did wasn't really that hard.

But I had sort of forgotten that initial pain point and people would start trying to learn and then they would hit that and go, oh, I guess I'm not good at this and back out. And what I've kind of learned since then is it's like that initial...

it's going to feel like your brain is not meant for this at some point. And you have to push through that. And the best way to get through that is to be trying to solve some sort of problem to where you stay up all night and do it because you're excited about getting to the other side of that. James, anything to add? Yeah, I mean, build the thing you want to use yourself is the sort of canonical good advice. I would say...

As a specific, Paul Hudson's Hacking with Swift website has a number of courses on it, and they're really good. And Paul's written a lot of books. I mean, there is a lot of good third-party... guides to doing things that will really help you. The other point I think is worth making is that writing the code is almost the smallest part of the job.

particularly, you know, if you're like us and you're doing your own apps and stuff, the whole, like... marketing side of it and coming up with what features you want and doing the technical support. Casey said making the website. You know, there are so many other parts to it than the, I'm going to sit, it's not, I'm going to sit down and write code for eight hours a day and that is my job. I mean, some people do that and some people are lucky to do that, but.

For us, you need to have all these different skills beyond just the writing the code. And I think the only way... I was going to say the only way to begin is by beginning, but the only way to do it is to do it, you know, just keep at it and build that stuff up. And it's...

difficult and you can build the best app in the world and it goes absolutely nowhere all right next lightning round question uh let's start with you charlie what feature is your white whale

You've actually heard me say this before, I think, whenever I was on this podcast last way too long ago, which shows how far I've gotten on it. But I really want to rebuild my audio engine using... av audio engine uh by doing that that will enable a whole bunch of different features but the first big one that i really really want to do is basically an entire feature aimed at people with uh tinnitus

because that's a big part of my user base. David, I think you're one of those people. I have been off and on as well. Yeah, exactly. I think I learned recently I might have tinnitus and I never realized it because I was talking to somebody and they were just... If you have it, you know, trust me. Well, right. The version would be way lower, but I was like, isn't there literally isn't don't silent rooms have a ringing? And they're like, no, it's not supposed to. Yeah. Wait, what?

yeah yeah but it's yeah mine is not like uh disruptive to my life in any way at all i'm not saying that point being though is so many of my users use it that way um It just feels like that's something that I should really build explicitly a bunch of tools to make that even better. So that's kind of the big thing that I want. That's why your app is running all the time for me, frankly. It's because I don't want to listen to the ringing.

Okay, well, enough about that. Casey, Whitewell feature. What is it? You know, getting my darned iOS 26 filled out the door, I feel like it's never going to happen. No, I'm mostly kidding. I think I'd really, really love to have more robust iPad and macOS support, which is... 100% my own fault. It's just, I keep having other things get prioritized in front of it. And I would love to be able to do that. And as a bonus, there exist ways.

that you can query the Apple TV directly for what it's playing. And it's not really in a terribly useful way. And the code to do this is incredibly complicated. But... It would be extremely cool if I could ask the Apple TV, hey. what you're looking at right now and this already does work in call sheet for plex kind of uh and for the wonderful app channels but to be able to do it for anything on your apple tv would be fantastic and if i if i ever get ahead of the game

and find myself with a little free time, then that's what I'm going to be working on. Well, I apologize to all of you for asking this question because I know now you're going to get email about it. I know, I know. James, you're up. What's your white whale? So it's completely rewriting the underlying math engine in pcalc. And I built...

It's built that there's an internal API that I call for everything and it's really cleanly separated. And that little engine, that's how I can put pCalc on pretty much anything because I just move that over and then write a UI around it. But that... maths engine the point of doing that was so that i could replace that with something else and you know like it would allow me to change a whole load of stuff and and uh

enable me to do things that I can't do at the moment. But it's just, it's not a small thing. And it is kind of the, you know... chaining the foundations under an existing building and trying not to have everything collapse. But that's what I'd like to do, again, with the copious free time thing, which... I feel that there's between January and beginning of June is the time for third-party developers to actually do things, and everything else is like Apple time.

Well, I have a headcanon that Tim Cook listens to our show at 5 a.m. while he's working out. So, Tim, please take it easy next year. We all got our glass UI now. Just take it easy. Let these guys do their engine work next year, okay? So now you're set. It's all good. All right. Last lightning question. What are you most excited for next year? Let's do lightning here, guys, quick. James first.

I'm not excited about anything for next year because I haven't finished this year. I need to get this stuff out. I need to rev the apps. I need to get, at the very least, Dice redone with Glass. And then I can start thinking about next year. I don't have the space in my brain to imagine what's going to happen because do I think it is going to be good for me? Probably not.

You're most excited about catching up. I get it. I've been there. I've been there. Charlie, what are you most excited about? I think we're due for a system redesign. No, I think I'm really interested in where they'll continue to push the Apple Foundation model stuff.

I think that there's a bunch of interesting features that I could quickly build if they expand that out to audio or... Well, let's not talk about them doing video AI stuff, but... uh things like that i think i think that's that's an area that i'm interested to see what they'll do in at least yeah i would love to see developers get private cloud compute access you know beyond the the internal foundation model so you could do more with that i think that would be cool

Casey. I'm calling it now. Next year is the year of cellular max. And you can remind me of this at the end of next year when it is once again, not the year of cellular max, but. I don't do it as often as I used to, particularly during COVID, but I used to go out to like parks or the botanical garden and work from a place that doesn't have Wi-Fi and, you know, and, and to be able to do that without tethering. I know so many people are like, what's wrong with tethering? Tethering works great.

I would rather not. I don't have the time to explain. Let's just say I'd rather not. And so one day, one day I tells you Apple is going to release a Mac with like the C1 modem in it or what have you. And I will be the very first person, maybe second.

after marco but one of the first two people in line to get one of those all right well gang thanks for coming in this year i really appreciate the three of you showing up uh real quick uh where do people go to find you uh why don't you go first casey

Sure. So you can find all of my stuff at my website, which is caseylist.com, C-A-S-E-Y-L-I-S-S. You can find me on Relay at Analog, which is with our good friend, Mike Hurley. You can find me at the Accidental Tech Podcast at ATP.FM. And you can find my wonderful... app call sheet in the app store or at callsheetapp.com. How about you, Charlie? Where do folks go to find what you're up to?

I'm everywhere. It comes with a job. So I'm on basically every social network. But you can go to charliemchapman.com slash me and there'll be links to all of the different places. And we will have links, by the way, for all these three gentlemen in the show notes. But James, where do people go to catch up with you? I mean, you'll find most of my apps on pcalc.com. James Thompson, that's T-H-O-M-S-O-N on Mastodon, on mastodon.social. Those are the main places. I am not on Twitter.

So if you see me there, it's not me. And I really am not trying to sell you crypto or whatever I'm doing. And you can find me apparently as the most frequent guest on RelayFM. And yes, occasional host of Connected. Yeah. New host, right? New host of Connected, isn't it? Yeah. All right. Well, we are the Mac Power Users. You can find us at relay.fm slash MPU. You can go back there for feedback and membership and more power users.

Thanks to our sponsors today, Squarespace and Indeed. Stick around for more power users. We're going to be talking about these platforms and have a great week.