363: The Floor is Lava


00:00:00   You've had that car for several years now, right? Yeah, it's a 2014 car with 30,000 miles.

00:00:04   Yeah. My kids complain that it still smells like a new car. They complain? That's the best smell in the world!

00:00:09   That's what I say, where they're like, "I don't like it." What does it smell?

00:00:13   I mean, does it actually smell like a new car, or does it smell like shampoo? Like, that scents it like a new car?

00:00:16   No, it's a new car smell. The inside has never been shampooed. Hmm. It's just mild new car smell.

00:00:21   Because I don't let the kids destroy it, so it doesn't smell like they're spilled chocolate milk and other things

00:00:28   they've wedged into the seat and other gross stuff. Like, it's a car mostly used by an adult who doesn't make a mess.

00:00:33   You? I'm surprised. Mm-hmm. I will say, so we successfully made it through,

00:00:40   I think, the majority of the snacking in the car age that our child went through,

00:00:47   without ever allowing him to snack in the car. And so none of our cars ever smelled like Cheerios or, you know,

00:00:53   things that like little kid/baby cars usually smell like. Because we just said, "No,

00:00:58   we're not allowed to eat in the car, and like, we'll eat when we get there."

00:01:00   And we stuck to that, like, even on long trips, and it was surprisingly easy. It was just like, "No, you can't eat."

00:01:08   "Okay." And then we ate when we got there. And as a result, my car never smelled like Cheerios.

00:01:15   Yeah, Aaron's car, it is relatively straightforward to clean out all the detritus that is everywhere,

00:01:20   but it is chock full of detritus. And, you know, periodically I will clean out the inside and get it all out of there, but

00:01:26   it's real bad. And she's better about it than me, because, you know,

00:01:30   she is fully aware that that is the family car, and that that's what happens to the family car, and so on and so forth.

00:01:35   But I look at it, and I'm just like, "Oh, God, what has happened in here?"

00:01:39   Well, let me assure you, it is optional. You can say, "This is not going to be a snack car."

00:01:44   And it doesn't have to be a snack car.

00:01:46   Well, the food comes back later, because my kids now want to grab an on-the-go breakfast on their way to school or something.

00:01:55   So they have teenager food of cans of sparkling water and random—

00:02:01   I mean, even just fruit stuff or yogurt. Alex is always bringing yogurt into the car on the way to school.

00:02:06   He wants to grab a yogurt and eat it on the way, and so he's got a yogurty spoon sitting in like a big—

00:02:11   There's a middle period where there's not much food in the car, and then later they get old enough to just be like,

00:02:16   "I'm just bringing food in the car. I need a snack on the way to school, because I didn't have time to eat breakfast."

00:02:22   And you're not getting in the car, or you're not getting a breakfast, because that's—

00:02:26   Like, yogurt is such a joke. Yogurt is just melted ice cream.

00:02:29   It's not melted ice cream.

00:02:31   Yeah, it's dairy and sugar.

00:02:33   Margot also doesn't know where food comes from.

00:02:35   No, no, yogurt is basically—it's like, you know, dairy plus sugar.

00:02:38   It's easily spoiled. It's full of sugar, so it will be sticky and be covered in bees pretty soon.

00:02:44   It's just—that's what happens. Yogurt is—it's fine when you're eating it, but it's not something you want ever going near your car.

00:02:51   No. Just let him bring in an ice cream cone. It's the same thing.

00:02:55   If you would allow yogurt, bring in an ice cream cone.

00:02:59   Most yogurt has slightly less sugar than ice cream.

00:03:02   You'd be surprised how little less. It's really quite similar.

00:03:06   That's why the kids like it so much. Because it's ice cream, basically.

00:03:12   [Music]

00:03:14   I'd like to start the show, if I may, with Casey's complaints about his co-host corner.

00:03:18   Yes.

00:03:19   First of all, I would like to begrudgingly congratulate you. Game respect game.

00:03:24   I was—I was talking to somebody—shoot, I can't remember who it was now.

00:03:28   But I wanted to refer back to the BMW that I had built for the post-show neutral last week.

00:03:34   And BMW has this wonderful feature where whatever build that you configure on their configurator gets a unique URL.

00:03:42   Right.

00:03:42   So that you can—you can then copy and paste that URL, send it around, save it, refer back to it later, and it just has this, like, preset configuration of whatever you set it to.

00:03:50   Exactly right. And so I knew that I had put my configuration—quote-unquote "my" configuration in the show notes.

00:03:56   And I wanted to refer back to it. And so I clicked on it in the—you know, I went to ATP.fm and I clicked on it.

00:04:02   And I saw a white BMW. And I thought, "Oh, I must have clicked on the wrong thing." And I clicked it again.

00:04:10   And I saw the same white BMW. But it looked at this—oh. Oh. Oh. Well, okay, fine.

00:04:20   Well done, you insufferable pain in my hindquarters. Well done indeed.

00:04:25   I—it took me a minute to realize the joke. And then I realized the joke and I realized that Marco had swapped my Tanzanite Blue 7 Series for a white one.

00:04:34   I was quite annoyed. And I was actually more annoyed at the quality of the joke than I was the fact that you had replaced my blue car with a white one.

00:04:43   In fact, I respected the joke enough to—I went to actually replace the link with the correct one because I had the correct one still.

00:04:50   And then I thought, "No, no. He earned this one. I'll let it sit." And so it stole the white one on ATP.fm.

00:04:57   But rest assured, listeners, I had built a blue one, dammit.

00:05:01   This one just so happened to be white.

00:05:03   It happened to be white, Marco. It can just happen.

00:05:07   Also, I'd like to complain about Mr. John Siracusa. You had mentioned offhandedly on Reconcileable Differences #121, entitled "Impotent Hydra," that you don't like Wegmans.

00:05:18   And I would like you to defend yourself, sir, because this is blasphemy.

00:05:21   We've talked about this on this very show, I think.

00:05:24   The one reason is that the Wegmans near me is in a very inconvenient location.

00:05:28   It's on an annoying, divided highway that's difficult to get to and always clogged.

00:05:33   And then once you're on the side you want to get on, it's hard to get back on the other side.

00:05:37   It's a mess. So that's a big reason why we don't ever go there.

00:05:41   Because even though it's not that far away as the crow flies, it is terrible to get to.

00:05:47   And the parking's not great and just everything about it.

00:05:50   The second thing is, even if it wasn't terrible to get to, there's another one that's also even worse to get to that's farther away that's in a mall.

00:05:58   It's kind of like Whole Foods, where we go all the time because we have Whole Foods close by us.

00:06:03   But Whole Foods, like Wegmans, is... I mean, you know what they're like.

00:06:08   They're not like regular grocery stores.

00:06:12   Whole Foods is worse than Wegmans in this respect.

00:06:16   Where a regular supermarket that you're accustomed to, you can picture in your head of here are what the different aisles are and here's what's in them and here are the kind of products they have.

00:06:27   Just a generic, straight up the middle American supermarket.

00:06:32   Wegmans and Whole Foods have fancier stuff, different stuff, different brands of things, lots of prepared foods, all sorts of stuff like that.

00:06:40   The environment is different. It's not just aisles of stuff with carts going back and forth. It's all these weird stations and areas and stuff like that.

00:06:49   And when I want to go grocery shopping, most of the time, unless there's something special that I want to get that I know I can't get at a plain grocery store, I just want the grocery store experience of going up and down the aisles and getting the stuff and everything being where I expected.

00:07:02   Being able to find all the brands that I'm accustomed with. But Wegmans doesn't fill that. So maybe if it was close, I would go there more than I go to Whole Foods.

00:07:12   But I always feel like I'd have to go to multiple places like Wegmans/Whole Foods for weird specialty items or things that we know are better there or something we want to get that's fancy.

00:07:23   And then a regular supermarket for regular supermarket stuff.

00:07:27   I actually am sort of playing a shtick here. I do enjoy Wegmans. They just, in the last couple of years, came to my area.

00:07:36   I do like Wegmans actually most as a place to work. They have a really nice patio outside, one of the local Wegmans near me.

00:07:45   And additionally, they have quite a large food court kind of sitting area inside as well. Large enough that if you just go and sit and do some work, that's not really unusual.

00:07:55   And it's not...nobody's looking at you like, "Why are you here? Get out."

00:08:01   For those of you who don't know, Wegmans is a supermarket chain here in the States. I would say it's, like John had said, more similar to Whole Foods than a traditional supermarket, but not quite all the way to bespoke, artisanal everything like Whole Foods is.

00:08:17   But anyways, Wegmans is very expensive. My favorite vice, well not my favorite, one of the things I enjoy to do for lunch occasionally is go to Wegmans hot bar where you get a styrofoam or whatever container.

00:08:31   It's probably not styrofoam, it's probably cardboard. But anyway, you get a cardboard to-go box and you fill it with all sorts of different things.

00:08:37   And of course, if you're like me, you want the really, really heavy things like mac and cheese. And then you go and you go to the checkout counter and they weigh your box.

00:08:47   And it's so much per half pound or pound or what have you. And it is preposterous how much a single human being can spend at the Wegmans hot bar.

00:08:57   I think I have spent like $15, $20 on my meal alone. And this is just like cafeteria food for all intents and purposes. It's very tasty, cafeteria food, but consider your source, I have low standards.

00:09:07   And so it is ridiculous how much you can spend at the Wegmans hot bar. And Whole Foods is just as bad if not worse.

00:09:13   But yeah, a lot of people are devout Wegmans fans. I definitely like it. It is very expensive even for groceries as far as I'm concerned.

00:09:21   But it's a very nice store and I enjoy it. And I just wanted to hear you defend yourself and I can't really argue with anything you've said, unfortunately.

00:09:28   I can't believe that you go to the hot bar in a grocery store and get macaroni and cheese. That's basically the worst thing you can get there because it is always priced by weight.

00:09:37   And what you should really do, I mean there's all sorts of hacks of like you get all sunflower seeds or whatever. It's cheaper than buying sunflower seeds per pound or whatever.

00:09:47   But within the realm of normal meals, I feel like a salad with a bunch of stuff added on top like sunflower seeds or other nuts, things like that, that's almost always a way better buy than anything that is inherently very dense and heavy like baked potatoes or pasta.

00:10:04   Oh no, I completely agree. There's a stretch of time when Wegmans was new that we were doing this once every month or two. The family would go for lunch or something like that. And over time I realized I'm being an idiot.

00:10:17   I need to get like salad with maybe a little kiss of macaroni and cheese on the side rather than an entire platter. Not an entire platter, but you know what I mean.

00:10:25   Like a large portion of mac and cheese and then some other heavy thing next to it. So yeah, I completely agree with you.

00:10:32   I don't partake of any of the parrot foods. I don't begrudge people them. Occasionally if you're in a crunch and you don't have time to do anything, you can get them, but they're so expensive and everybody has sneezed on them and you have no idea how old they are.

00:10:44   It's just everything is aligned against them. The main reason we go into our local Whole Foods aside from it being very close by is they do milk as a loss leader. So they actually have the cheapest milk in the area.

00:10:56   So every time we go to Milk we end up in Whole Foods and then if our kids are with us they make us buy like the $2 mochi which are good granted, but A) not that good and B) they're $2.

00:11:08   I have never found anything in the grocery store hot bar to be worth getting a second time. And I will forget this about maybe once every two years and I'll go get another big lunch or a couple lunches there. And every time I'm like, "Why did I do that?"

00:11:28   It's actually very similar to the WVDC box lunch side dishes where you get something and you're like, "Here's a side of macaroni salad or something." And you taste it and you're like, "This is not anything like what I expected."

00:11:45   And also somehow manages to taste like nothing and yet is also not good.

00:11:50   And it's $8.00. And it costs you $8.00 if you're lucky. In reality I'm usually closer to $12.00 because I think I'm underweight and then I get to the checkout and it's like, "Oh yeah, it's $12."

00:12:02   And afterwards I'm like, "I could have walked down the street to any casual fast food place like a Chipotle or something and gotten a way better meal for less money than what I just spent at the stupid hot bar at the grocery store."

00:12:15   I actually think that Wegmans and Whole Foods both are tasty and decent food, but I completely concur that it is not worth the money that one would spend on it.

00:12:24   We've gotten way off the rails already and it's my fault so let's just move right along. Why don't you tell us about what could have caused your iMac Pro's fan problems, Marco?

00:12:33   Actually I have a small update on that. Alright, so first let me do the feedback item here. So we got feedback from Chris Harper who says, "I started my career as a technician in the entertainment department at a large theme park. One of my jobs was repairing broken equipment.

00:12:50   Early in my training I was told to never use the air compressor for blowing out electronics because air compressor tanks and even the hoses often get condensation in them, possibly due to the heat of the motor. Essentially you'd be blowing moisture on your stuff.

00:13:03   So you had to use canned air instead or vacuum them out using special electronics vacuums."

00:13:07   I think I heard from a couple of people on Twitter who also said that apparently air compressors can spray tiny drops of oil from inside themselves into electronics, which is obviously very bad or little metal shards. I don't think any of that happened and my air compressor is a very small consumer unit that probably doesn't have either.

00:13:27   I can tell you I've never oiled it and probably doesn't get hot enough to ever make condensation inside. So I don't think that was an issue here. However, after a few days of operation my fan problem is back.

00:13:43   Now I also heard, I forget who, I'm sorry, I have to look up.

00:14:08   I heard from Tom Bridge who said, "A second possible explanation for Marco's iMac Pro's reduced fan RPMs post-vacuum would be a reset of the system management controller, the SMC reset, which was accomplished when I unplugged my iMac Pro for more than 15 seconds."

00:14:22   So this is a long-standing Mac spin around and fix the problem kind of thing that you try. You do a rain dance, you zap the PRAM and you reset the SMC. That's what this is. So the way you reset the SMC on any modern Mac is you cut off power for 15 seconds and you plug it back in and the SMC resets.

00:14:40   And that is the component that is responsible for things like thermal management, fan speed, stuff like that. So that is much, because I said last episode, I'm pretty sure that not a lot came out of it dust-wise.

00:14:55   And I wasn't sure why it fixed it because it seemed like enough stuff didn't come out of it for that to be the problem. So if that indeed wasn't the problem and if the SMC is just kind of getting buggy over time, which does happen with Mac sometimes, that's a much scarier outcome for me.

00:15:13   Because now the problem is back. I didn't have time to try this yet because I used my computer so I didn't have time to unplug it for a while and play around today. But I am a little concerned that this just might be my life now.

00:15:26   And if that's the case, I'm going to be very sad because, again, one of the biggest reasons I love the iMac Pro is no matter what I do to it, I never hear the fans. And now I'm hearing the fans spin up and down on a pretty regular basis as I do Xcode builds and stuff. And that's really not great.

00:15:42   Let it be known, listeners, that as we record, it is the evening of the 29th of January and the year 2020.

00:15:49   I have not yet ordered a Mac Pro. In fact, go to the next follow-up item and I'll tell you why that's relevant to me.

00:15:56   Uh-huh. Well, let me just tell you, listeners, that I am putting it on the record that before WWDC, Marco will have a Mac Pro in his house. You've heard it here first. All right, tell me about why you're supposedly not getting a Mac Pro, Marco.

00:16:10   This is my item. He'll chime in after I explain what's actually in here. And I don't think it's actually relevant to his decision-making process.

00:16:17   This is just a follow-up on the Pro Display XDR reference modes. We talked about them last time, like what they're used for and the P3 1600 nits versus 500 nits and all that stuff.

00:16:27   Anyway, there is an official Apple, whatever they're called now, Tech Notes support article that explains all the different reference modes. There's many more than the ones we mentioned.

00:16:35   So if you want to see them all and understand how they work and what they're supposed to be used for, you can check out the link in the show notes.

00:16:40   The one interesting thing that I pulled from it is there's a section towards the bottom that's like, if you see this scary icon showing like a, you know, a caution triangle exclamation point symbol next to a monitor that appears in the AirPlay menu, your Pro Display XDR might be in a low-power mode and using limited brightness.

00:16:57   This can occur if the ambient temperature of the room is 77 degrees Fahrenheit or higher and your display has been at 500 nits or higher for a long time. Unplug your display from your computer, wait 5 to 10 minutes, then connect your display and try again.

00:17:08   If the issue continues and the ambient temperature of the room is less than 77 degrees, contact Apple.

00:17:13   The reason I think this is irrelevant to Marco is because I really don't think he ever keeps his house higher than 77 degrees.

00:17:19   And he's also probably never going to have the monitor over 500 nits because he's not doing HDR video editing/mastering, so it's mostly irrelevant.

00:17:27   And I can tell, like I said when I first talked about this monitor, you cannot hear or I cannot hear the fans in the monitor.

00:17:34   I'm assuming they're spinning right now, but I have never heard them other than when I shoved my ear against the back of the monitor when I was setting it up and even then it was difficult.

00:17:42   People can hear the tower computer over there off in the distance a little bit, which is why I think still with your iMac Pro it may be noisier under load than this.

00:17:50   It is definitely noisier under load than this thing is because this thing never changes sound whatsoever, but its constant noise level is higher than the idle noise level of an iMac Pro, which I basically can't hear.

00:18:02   This I can always almost hear and it just literally never changes volume at all.

00:18:07   Unfortunately, my iMac Pro was like that for a long time and now isn't, so just wait John, wait until about year 2 or year 3 of your Mac Pro.

00:18:16   We have a little bit more thermal overhead than you do considering how vast my cooling system is and how little is actually in the box.

00:18:23   I'm not on the ragged edge of anything. I really need to get, I think the iStat Menyoo guy said they were updating their stuff to support the new temperature.

00:18:32   They were going to send me a beta and I don't think they sent me one.

00:18:35   Look at that, they're usually pretty fast.

00:18:37   I do want to see, do the fans ever go any faster no matter what I do?

00:18:41   Because like I said, I can't hear them change volume at all even when I do disk benchmarks, GPU benchmarks, CPU benchmarks.

00:18:52   Even playing games in Windows, which I assume doesn't really understand the fan control on this or whatever.

00:18:58   Like playing Destiny, I don't know what I have to do to this computer to make it make any more noise, it just doesn't.

00:19:05   Because I've got the wimpiest video card and a fairly wimpy CPU and three gigantic fans plus a blower fan and almost nothing in that box.

00:19:14   Well, wait two years and then you'll see.

00:19:16   The reason why this Pro Display XDR, I would call it an overheating issue maybe in a warm environment and running at full brightness, it sure sounds like that.

00:19:27   Apple would never use that term but that's kind of what this is.

00:19:30   You basically could have problems with the XDR retaining full brightness if you're running it in a room that's hotter than 77 degrees Fahrenheit.

00:19:38   Well, I do that all summer long.

00:19:42   When I'm at the beach, one of the reasons I like being there is that while it is warm, it is not too hot.

00:19:51   Like it almost never hits above like 85 degrees and there's a nice breeze.

00:19:56   And whether or not there's a nice breeze, I can augment it with a big ceiling fan.

00:20:00   So I'm able to work very comfortably with breeze and/or fan blowing on me up to the low 80 degrees.

00:20:09   Like up to like 80 to 85 degrees Fahrenheit.

00:20:12   That's the temperature outside. Check the temperature in the room you're sitting in.

00:20:15   It's not going to be lower than outside.

00:20:18   I think if the room you were sitting in was 77 degrees or higher, you would do something about it.

00:20:24   You don't know summer mode, Marco.

00:20:26   You got the breeze going.

00:20:28   Anyway, if you have the windows open and fans going in, it would be fine.

00:20:33   Because you're not mastering HDR stuff. You're not going to have it over 500 nits.

00:20:36   Put your monitor up to max brightness right now. That's not even 500 nits.

00:20:40   You're never going to have it that bright.

00:20:42   How bright does the iMac Pro monitor go?

00:20:44   Not even up to 500, I don't think.

00:20:46   I think it might. And I keep it most of the time at about 75 to 80 percent brightness.

00:20:50   In the middle of the day in the summertime, I might crank it up because the room will be brighter.

00:20:54   It's not a factor.

00:20:56   Alright, you should try it.

00:20:59   But I will be shocked.

00:21:01   Because you're not doing HDR stuff.

00:21:03   You don't realize how bright 500 nits is.

00:21:06   And if you have the windows open and the salt air is blowing in and corroding your monitor.

00:21:10   This is only if you're in a stifling room with no airflow that's over 77 degrees.

00:21:17   And you're doing things for a long period of time.

00:21:19   We'll see because I don't actually have whole house air conditioning or anything that's going to save this thing.

00:21:26   So when summer comes in my tiny little stifling room that I'm actually in, we'll see if I ever have any issue.

00:21:32   I can imagine that not only will it not have any problem, but I also will continue to not hear the fans in this monitor.

00:21:39   By the way, real time follow up, the iMac Pro does apparently have up to 500 nits brightness as do the most recent MacBook Pros.

00:21:45   So what do you have yours set to now?

00:21:47   About 80%. And this is at night in the winter.

00:21:49   So is the room over 77 degrees? And has it been at 500 nits or higher for a long time?

00:21:55   It's never been at 500 nits because you never crank it to full brightness probably.

00:21:58   And the room is never over 77 because who keeps their house that hot if they can help it.

00:22:02   And you have central air so you can help it.

00:22:04   Yeah, but all summer long. I'm telling you.

00:22:07   So basically what this tells me is I probably shouldn't buy a Pro Display XDR for the beach.

00:22:12   Now granted I probably wasn't going to anyway because it's so damn expensive.

00:22:15   But again that puts a pretty big thorn in the side of any possible Mac Pro plans.

00:22:21   Because one plan would be get a Mac Pro and haul it back and forth and have a monitor at both places.

00:22:26   But that now seems like an even worse idea than it did last summer when we were first talking about this.

00:22:32   I think it's very likely you're going to buy a pizza for the table. You're going to buy a computer for the house.

00:22:37   I don't know what it's going to be yet. It could be a Mac Pro. It could be an iMac Pro.

00:22:40   But this is going to be a computer for the house.

00:22:42   I don't think so.

00:22:43   Absolutely. You're going to get a computer for the house.

00:22:44   A monitor is very likely.

00:22:46   Everybody loves it when you get a computer for the house.

00:22:48   I will almost certainly, if I don't haul an iMac Pro back and forth, which I think is still probably the best option.

00:22:55   But if I don't do that I would probably keep an LG 5K out there and then whatever the heck my home solution was.

00:23:01   Computer for the house.

00:23:02   Imagine if you had ever had experience with the LG 5K at the beach house.

00:23:07   You would know exactly what you're getting into.

00:23:09   I have.

00:23:10   That's my joke, Marco. Good God.

00:23:13   I spent a whole summer with it. It was really mediocre.

00:23:16   I know. But if you had it hooked up to a preposterously overkill Mac Pro, I think you would get over the mediocreness.

00:23:23   Anyways, I think you're getting a Mac Pro.

00:23:26   I think you will start by having an LG 5K at the beach.

00:23:30   Don't you already have one at the beach right now at somebody else's house?

00:23:33   He buried it in the sand. It's marked with an X.

00:23:35   Secret treasure.

00:23:38   I stuck a kid's plastic shovel right above it so I could see where it was.

00:23:43   It would be better if your name was Casey because then it would be the secret of Casey's gold.

00:23:45   But the secret of Marco's gold doesn't have the same ring to it.

00:23:47   No.

00:23:48   In any case, I think here's your future.

00:23:51   Your future is in the next six months. You're going to be getting a Mac Pro.

00:23:54   And I think you'll get a Pro Display XDR for your home in New York.

00:23:59   For your home that you're sitting in right now.

00:24:01   And then I think you will go to the beach.

00:24:04   You will try to convince yourself that you can live with the LG.

00:24:08   And then before you depart the beach from the end of the summer, you will have ended up ordering a second Pro Display XDR and put it at the beach house.

00:24:16   That's my hope.

00:24:17   Just wait until the newly redesigned iMacs with built-in 6K displays come in colors.

00:24:22   And then we'll get whatever color matches whatever room that it goes into in the house.

00:24:27   Now we'd be talking new, like, 6K iMacs? Now I'm very interested.

00:24:31   In colors. It's going to come in teal.

00:24:33   Yeah. But see, I'd be very interested not because I would want two of them, but because I would just replace my iMac Pro with the new 6K iMac Pro.

00:24:40   You're getting two of them. I don't understand why you're resisting this. You're not going to keep plugging that thing back and forth. You're going to get two.

00:24:45   I honestly don't think you would get two.

00:24:48   Keeps it from the table. That's going to happen.

00:24:50   I honestly think that Marco will end up with two Pro Display XDRs, but I think there will only be one machine.

00:24:57   He's rapidly getting too old to be lugging a computer on his shoulder back and forth. Just wait.

00:25:01   But now I wheel it. It's easy.

00:25:02   Right.

00:25:03   Yeah, I'll see.

00:25:04   The funny thing is, it's probably an easier solution to just get a second iMac Pro.

00:25:09   That's what I'm saying.

00:25:10   That's both cheaper and easier than bringing a Mac Pro back and forth.

00:25:14   I didn't say what kind of computer it would be. I would just say there's going to be a computer for the house.

00:25:19   I can't handle this conversation. Can somebody please save me and tell me about True Tone?

00:25:23   No, because why are we having some problems with the notes today? You go from top to bottom. In this country, we read from top to bottom.

00:25:30   Oh, I'm sorry. I saw the Chris Harper and I went below it. God, we are disastrous, especially me.

00:25:34   We? Well, some of us.

00:25:35   Is that especially me? Oh my gosh.

00:25:38   You and Marco both have done technical fouls in the follow-up section.

00:25:42   All right. So I don't want to hear about True Tone. Never, ever. What I'd love to hear...

00:25:49   Oh, God. See, I skipped it. My subconscious skipped it because I don't give a s*** about a rackmount Mac Pro.

00:25:55   Did you watch the videos?

00:25:56   I watched like four seconds of it and I realized I don't give a s*** about this.

00:25:59   Four seconds. That's your YouTube attention span. All right. I'll handle this one so you don't have to. It's a quickie.

00:26:04   We've been talking about the rackmount Mac Pro, which is available for sale. Someone put up a YouTube video of unboxing it and comparing it to his old setup.

00:26:15   It was Neil Parfit. They're really good videos. If you want to know what the deal is with the rackmount Mac Pro, you'll see it from every angle and learn all the features and learn how one person's actually using it.

00:26:28   The interesting thing about it, I think, is when you see it, especially if you're familiar with the plain old tower Mac Pro, it is so clear that unlike the Xserve, this is not a purpose-built rackmount computer.

00:26:42   This is a tower computer turned sideways and put into a weird case that kind of goes in a rack, which is fine.

00:26:50   There's nothing terrible about that, but when you look at the Xserve, which was absolutely purpose-built, shared nothing with anything, like all new industrial design just for being in racks, it's so clear that Apple at one brief time in the semi-distant past was serious about making rackmount computers.

00:27:10   I think with the Mac Pro, they're just realizing some pros, the situation in which they want to use a Mac Pro, they have a rack available. They don't want it to be freestanding. So can we make this tower computer rackmountable?

00:27:22   Yes, and it'll look fine, but it's not particularly space efficient. It is just a Mac Pro turned on its side. That's why there's weird gaps and stuff inside the case. They did some interesting solutions for you to be able to get at the different stuff because the RAM is on one side and most of the other cards are on the other side, which makes sense when you have the tower Mac Pro where you pull off the whole case and you have access to it from all sides.

00:27:43   But the rackmount one is not like that. So it has a door on the top and then it has a second door on the bottom. It's like, why is that stuff on the bottom? Why didn't they just make everything accessible from the top? Because this is the tower Mac Pro turned on its side and that's the way the tower Mac Pro works.

00:27:55   The interesting comparison, so there was the unboxing and everything, there was the comparison of fan noise and again, he was very impressed that it was very quiet.

00:28:02   Neil was filling his thing with almost every card slot filled because he does audio stuff and he's got all these weird audio interfaces. So he stuffed it to the gills and had all these power cables going everywhere.

00:28:13   He was really happy with it because at the end of the second video he shows, this is what this is replacing. I have this big giant whatever it is 5U rackmount thing, which seems like it's really big, but it replaced and he does a little cut in the video showing.

00:28:26   Two trashcan Mac Pros and a billion peripherals and like Thunderbolt docks and cables and everything. Just this giant mess of hardware all now fits inside this one box.

00:28:37   So I think he was happy with it even though I think that as a rackmount computer it is a little ungainly and ugly, but it's clear that some people want exactly this. They do not want a tower computer. They want it to fit in a rack.

00:28:48   So I'm glad Apple's making it and I'm glad at least one person seems happy with it.

00:28:52   I feel better for having known this piece of information. Thank you, Jon.

00:28:56   You should watch the video. You can hear the cool little snick that the door makes when you take it off the side and click it back in. It's cool.

00:29:02   Just like Wolverine. Would you tell me about True Tone, please? Because I'm oh so excited to hear about it.

00:29:08   This is Chris Harper chiming in on, oh no, that's not, that's the compressed air. Now all three of us have done Fall Affairs.

00:29:15   Here's how the show notes work, Jon. You read from top to bottom.

00:29:18   The instruction pointer is friggin, no one knows where the instruction pointer is anymore. We're just, we're lost. We're lost in follow up.

00:29:26   Oh, this is amazing.

00:29:28   Someone issued a jump statement and now we have no idea where we are.

00:29:32   You are such a nerd, I love you. Oh, God.

00:29:36   Christopher Bowers writes, regarding True Tone, as someone who works professionally in printing photos, the ideal situation is to have the room lighting and display white points be equal. That's why having True Tone on is the better, more accurate setting.

00:29:50   So there's another person with some professional experience chiming in saying, True Tone, yes, go for it, even if you're doing something that's important to color.

00:29:57   And he also gives a link to ISO 12646. I looked at that and I didn't, I think it's just, you know, well here's the text from the thing.

00:30:07   Specifies the requirements for two conformance levels for the characteristics of displays to be used for soft proofing of color images.

00:30:14   Included are requirements for color uniformity and variations in electro optical properties when viewing, with viewing direction for different driving signals. That's a lot of jargon.

00:30:23   I think the standard is just basically saying how you should set things up to be judging photography on a computer device. It's hard for me to tell. That's a lot of jargon. Plus they spell color with a U so it's all very confusing.

00:30:35   True Tone, endorsed by one person who claims to be a professional. It's good enough for me.

00:30:42   Go team. I was using it already anyway.

00:30:47   We are sponsored this week by Yes Please Coffee. This, I gotta say, this is not in the script. This is one of the only coffee things that I would ever recommend. I've used it myself for a long time and it's fantastic.

00:30:58   So there's a lot of productivity hacks, but there's one productivity tool that performs wonders for almost everyone. It's coffee.

00:31:06   Yes Please is a coffee subscription service. It's Yes PLZ. They want to bring the very best roasted beans into your kitchen.

00:31:15   So, you know, do you really need a coffee subscription? How often do you run into coffee at home?

00:31:19   True coffee fiends are pretty good about never running out, but you also know that scoring truly superb beans, especially reliably, can be hard.

00:31:27   So, Yes Please, this is at YesPLZ.Coffee, I love the domain, sources and roasts some of the finest stuff from all over the world.

00:31:35   They also change it up every week and then they deliver it to your door. So it says whole bean, very freshly roasted coffee, delivered weekly, fortnightly, monthly or just whenever you need it.

00:31:46   You can pause and cancel anytime, no hassle. And you also get this cool monthly print magazine that they make featuring culture, food, music and more.

00:31:54   It's kind of fun. It's like a nice analog moment for your mornings in an increasingly digitally distracted world.

00:31:59   So the founders of this, they have quite a strong background behind them. So it's from Tony Knesney and Sumi Ali.

00:32:05   They are veterans of coffee's third wave scene. You actually might know their previous coffee subscription business called Tonks.

00:32:11   That's the name of Tony Knesney from back in the day. And these guys believe great coffee should not require much fuss.

00:32:17   Making a perfect cup should be as easy as making a box of macaroni and cheese if you start with the best beans.

00:32:22   And I can personally say I have used this service. I used it all summer long when I couldn't be home. Usually I'm a home roaster.

00:32:29   But there's a lot of times when I'm too busy or I'm not home for a while. And whenever that's the case, I go to Yes Please.

00:32:34   This is actually the coffee service I use when I want to buy beans. Go to Yes Please. That's Y-E-S-P-L-Z dot coffee.

00:32:41   Use offer code Marko sent me, all one word, to get 25% off your first shipment.

00:32:45   Upgrade your morning coffee game and you'll upgrade your whole day.

00:32:48   Thanks a lot to Yes Please for making amazing coffee and sponsoring our show.

00:32:52   Why don't you tell me why you don't understand Xcode, please?

00:32:58   This is about the rant last episode about me trying to do GUI layout in Xcode using the various systems.

00:33:06   A lot of people offered advice. A lot of people offered commiserations, saying that a lot of things I said were familiar to them.

00:33:13   Even people with lots of experience that, you know, some of these problems are just sort of long-standing problems.

00:33:18   And yes, you do sort of get better at dealing with them and come up with techniques for them.

00:33:22   I want to call out Alex Kent in particular, who gave a video demonstration of him basically laying out my dialog box in Xcode.

00:33:32   And I think that was important because lots of people sent like, "Here, look, I made your dialog box with SwiftUI.

00:33:37   I made it with Xcode. I made it with this. I made it with that. See how nice it is? Look, it looks just like yours."

00:33:41   First of all, a lot of people who sent those in, I applaud the effort, but many of them didn't actually look like mine.

00:33:45   It was kind of the point of like, "Yeah, you can get 90% of the way there, but then if you want to get it pixel accurate, it's a little bit trickier than you might think."

00:33:51   But whatever. But Alex's video let me see how he uses the tool to build it.

00:33:59   First, it was refreshing to see him run into many of the things I complained about in real time when he was doing it and then deal with them.

00:34:08   And then I said, "Okay, well, this is going to happen when you try to use this tool. Here's how you deal with it."

00:34:12   So that was very helpful. Just kind of the mindset of what things you should bother doing and what things you should let the tool do for you.

00:34:21   It was helpful to see that. Anyway, we'll put a link to the video so you can check it out.

00:34:27   It's kind of long, but if you've never used Xcode to do layout before, there are a lot of different tools available to you, a lot of different systems, a lot of different ways to use the tools.

00:34:37   And he demonstrates one particular technique. You don't necessarily have to do what he does, but I think it was eye-opening to see someone else's workflow.

00:34:45   I think the most useful advice he gave me, and this is one thing that I didn't mention last show and I should have, and a lot of people writing with advice were also not taking account for.

00:34:53   I was, maybe foolishly, maybe not, aiming for a situation where I have no warnings, because Xcode does these little yellow warning symbols in the sidebar if it thinks you have any layout issues.

00:35:09   And always my goal was no warnings, because my whole project was built, no warnings, no warnings or errors. It's got to be clean, right?

00:35:16   Same thing with the layout. So if I got a layout, it was very easy for me to get a layout that I think was mostly correct and mostly looked correct, but it had a thousand warnings on it.

00:35:24   So that was part of what I was banging my head against. Even when I thought I went over the finish line, I'm like, "There, finally, everything's lined up. Let me run the app, let me look at it, let me measure with Xcode, boom, everything's lined up."

00:35:32   And then I would look in the sidebar and there's 800 warnings.

00:35:35   I'm like, "What do I have to do to make you happy, Xcode? What are you complaining about?"

00:35:39   And I had all these complaints, "This doesn't have an X position." I'm like, "It doesn't have an X position? Are you kidding me? It has an exact distance from the thing that it's next to, which has an exact distance to the edge of the superview."

00:35:51   That's, like, it can't be anyplace else, X-wise. But it was like, "No, I need an X position."

00:35:57   I was like, "What do you want from me?" And I have all these things like, "Add missing constraints."

00:36:01   And if you try to do that, it just throws constraints all over the place and it makes this big mess. So that was part of my frustrations. I'm trying to get to Xaro Boogs. I'm trying to get to zero warnings in Xcode.

00:36:13   And Alex had this one particular piece of advice, which I had not figured out on my own, which was very helpful.

00:36:19   For the localization warnings, it'll warn you, "Oh, you have a field that's like a fixed width. If you try to localize that text, the German word's going to be longer and it's not going to fit."

00:36:29   One way you can deal with that is you can just somehow, I don't know how to do this, but somehow I'm told that you can convince Xcode that you're not going to do localization and it'll shut up about that.

00:36:37   But I don't want to close the door to localization. Who knows? Maybe it will localize someday. But I do want Xcode to shut up. I don't want those warnings to go to the sidebar.

00:36:48   So Alex says, "For localization fixed with warnings, they can be fixed by making the width constraint priority lower than the content compression resistance horizontal priority." So the auto layout constraints have priorities, and the maximum is 1,000.

00:37:03   You can do it in one number increments. So a lot of the defaults are like, "This has a priority of 1,000. This has a priority of 750." If you make the width constraint have lower priority than that compression resistance, what that means is that, yeah, the width will try to keep it at this width.

00:37:18   But if the contents are too big and the compression resistance, like the resistance of the contents to be squished or be clipped, is a higher priority than the width one, it will win. And even though the width is supposed to be 10, because the content is too wide, it will bust itself out and make itself 11 or 12 or whatever.

00:37:37   And just by setting -- it doesn't actually change the layout, because again, I'm not localizing. But just by changing those priorities to be just one different from each other in either direction, you can get it to get rid of those warnings.

00:37:49   And that, I think, was the final thing that I really needed to get over the hump. So when I was -- I'm working on my other app that has a Swift UI thing. It also has a regular storyboard, like I said.

00:38:00   And after watching his video and learning about that trick, I went back to that dialogue and relayed out the whole thing with auto layout. And it was -- I'm not going to say it was painless, but I feel like now I have a system.

00:38:12   Now I feel like I can look at a setup and figure out, okay, I can defeat auto layout and make a layout that looks the way I want, uses auto layout everywhere, and has no warnings, now that I know these various tricks.

00:38:24   So thank you, Alex, and I recommend everybody check out the video just to see how one person gets it done.

00:38:30   That is just -- it's so ridiculous that you have to jump through so many weird hoops like that. And the tools are not making that easy or intuitive.

00:38:42   There are some features in the tool that I was avoiding. Like Xcode will offer to fix a lot of stuff for you. It's super helpful. It's like, do you want me to fix this? Do you want me to fix that? And it does -- I can imagine the people doing these UIs. What more do we want to do for you, the people who offer this in Xcode?

00:38:56   We give you -- not only do we tell you what's wrong, but we give you every conceivable reasonable option on how to fix it and explain what they each mean and let you pick from it in a popover.

00:39:05   And they do, but the problem with that is you look at that and you're like, well, which one of these do I want to do? I mean, these all sound like they might fix it, but then you pick one and it has this cascading series of other consequences for your layout that you don't quite understand.

00:39:22   And then you're like, did I pick the wrong one two steps ago? I'm not sure what to do.

00:39:25   So I think the key is knowing which one of those tools to partake in. So if you watch Alex, what he does with the video, which I think he's maybe doing intentionally just to show how the tools can help you, is he chucks a bunch of controls onto the canvas willy-nilly and then just makes a bunch of layout constraints using the right-click-drag shortcut thing instead of going the long way.

00:39:48   And then just makes all the constraints. And then the feature he uses a lot of is misplaced items. The sidebar will say a bunch of items are misplaced, which means that you put some things on the canvas and you put a constraint that said this should be 10 pixels from that. But it's not 10 pixels from that. It's not even close to 10 pixels from that. That's a misplacement.

00:40:05   And then you hit the fix button, right?

00:40:07   Yeah, it will fix the misplacement for you. And that one is one of the least destructive ways you can accept its health. Because otherwise, like, do you want me to add missing constraints? I don't think I ever do want you to add missing constraints.

00:40:19   Trust me, you never want that.

00:40:21   The constraints you think are missing, they're not actually missing. You're just, you know, whatever. But the fixed-mix placement will essentially move things on the canvas for you based on your constraints. So if you put a big jumble of items and you just do a bunch of constraints and then you just fix misplacements, they all spring into where, more or less, where they're supposed to be and then you can fine-tune.

00:40:39   So that feature seems like a safe feature to use. And I found myself using it more. Like, once you're in Auto Layout mode, you don't have to do that single-pixel alignment stuff. My problem was always that, like, when I did Auto Layout, I could never get the layout to look right when it actually ran the window.

00:40:53   And I could never get all the little warning stuffs to go away until I started, you know, I was chasing my own tail. I was like, well, if I do this and fix that and this warning goes away, but this other one comes, I finally get all the warnings away. And I had this ridiculous series of constraints that made no sense and didn't put things where I wanted them.

00:41:07   But I'm better, I'm doing better now.

00:41:09   So are you a fan of Auto Layout at this juncture? Are you anti-Auto Layout? Where do you come down on that?

00:41:16   See, I've been watching years, literally years, of Auto Layout videos at WWDC at presentations, like in-person and on video. So I'm not unfamiliar with Auto Layout. I've more or less sold on the idea that Auto Layout is a cool thing.

00:41:29   And I heard all the complaints when Auto Layout first came out and people's frustrations with it or whatever, but I hadn't actually used it myself. So going to use it myself, I was already bought in of the idea of, like, I understand springs and struts, I understand the system.

00:41:41   People complain about the system, but I know there were bugs about the constraint solver or whatever, but it's better now and I mostly buy into it.

00:41:47   But I was so frustrated by my own desire to get rid of all those little warnings. A lot of the warnings I didn't agree with. I'm like, there must be a way to get down to zero warnings. And I was just chasing my tail.

00:42:01   Now that I, A, have that one fixed with localization thing in my back pocket to address those without screwing up my, without adding more constraints, and B, sort of know how to use the minimum number of constraints to get what I want done and how to address the complaints if it says, "Oh, I need this."

00:42:20   Sometimes it still complains that it doesn't have an exposition, and I'm like, "You do have an exposition." But I know how to give it now a redundant constraint to shut it up. I think I mostly prefer it to springs and struts, even for static layouts and especially for resizable layouts.

00:42:37   So I don't say I like it, but I think I've made friends with it. And I think if I have to make a new view, I'm going to try to use Auto Layout first. That's my first choice now.

00:42:49   We are sponsored this week by Linode, my favorite web host. Whether you're working on a personal project or managing an entire enterprise's infrastructure, Linode has the pricing, support, and scale you need to take your project to the next level.

00:43:03   They have 11 data centers worldwide now, including their newest one in Sydney, Australia. All of these feature enterprise-grade hardware. They now have an S3-compatible storage option and their next-generation network.

00:43:13   So Linode delivers the performance you expect, but at a price that you don't. Get started on Linode today with a $20 credit for our show, and you can get access to their native SSD storage, a 40-gigabit network, industry-leading processors, their revamped cloud management built with an open-source single-page app, and all their servers give you root access to your server along with their new API version 4 to automate things.

00:43:37   They have a Python CLI if you want. I actually used that myself to set up my servers really quickly. They now have these Nanode plans that start at just $5 a month that currently gets you 1 gig of RAM on that server.

00:43:49   And they have all sorts of plans above that no matter what your needs might be, they have something for you.

00:43:53   So they have specialties like dedicated CPU plans, GPU compute plans for machine learning, video processing, AI, block storage, object storage, one-click installs of the most popular apps including WordPress, the LAMP stack, and things like Minecraft servers.

00:44:09   So you can get all this at Linode. I have all of my servers hosted there. All of Overcast and Market.org are hosted at Linode.

00:44:15   I'm just so happy. I've been a customer for theirs for more than I think 8 or 9 years now and it's just been wonderful the entire time. I have used a lot of web hosts in my career and Linode is by far my favorite one.

00:44:27   See for yourself by going to linode.com/atp and using promo code ATP2020 to get a $20 credit towards your next project.

00:44:37   And also they're hiring. Linode.com/careers if you want to find out more about that. Otherwise, once again, linode.com/atp. Promo code ATP2020 for $20 in credit.

00:44:47   Thank you so much to Linode for hosting all my stuff and sponsoring our show.

00:44:51   [Music]

00:45:20   What's been going on, man?

00:45:22   I've been trying to write an audio app in iOS 13.

00:45:25   That's a mistake. Why would you do that?

00:45:29   I've had a lot of challenges with the latest version of Overcast and Beta.

00:45:35   I spent the last week or so basically issuing 1-2 new builds a day trying to chase down this crazy problem that I was having with my AV audio engine based playback engine.

00:45:50   Sometimes, for some people, audio would not play when the screen was off.

00:45:57   You could send the app to the background. All of that would be fine.

00:46:01   It would do background playback, but if you turned off the screen to your phone, it would pause.

00:46:07   And the second you turned the screen back on, it would unpause.

00:46:11   So it's not that the app was crashing, because it wouldn't unpause if that was the case.

00:46:16   And it's not that the app didn't have background audio capability.

00:46:19   If that was the case, it would not play in the background at all. It even went to screen on.

00:46:24   So that wasn't the issue. The issue was if the screen was off, this would be the problem.

00:46:29   Now, I tried a whole bunch of different stuff.

00:46:31   First of all, I suspected maybe my AV audio session is not getting correctly the fact that I have an active session.

00:46:40   Or that it's a long form session or whatever.

00:46:42   So I did all sorts of tricks with the AV audio session, trying not deactivating it ever.

00:46:48   Trying to activate it every time I hit play and then checking it, polling it to see if it was getting deactivated or whatever.

00:46:54   I tried so many things with the audio session manager and I wasn't getting anywhere with that.

00:46:58   Then I realized, or I remembered rather, when I wrote Overcast's original audio engine,

00:47:04   which is still what you're hearing this theory if you're playing it in Overcast if my new update hasn't been approved yet,

00:47:10   there was this issue that if you run an AU graph, which is Apple's old, old audio framework for audio units,

00:47:16   if you run an AU graph and you want it to be backgroundable, there's a certain property that you have to set on all the audio units in it

00:47:24   that is documented only by this one tech note. It's like one Q&A 1606, something like that.

00:47:31   And there's this maximum frames per slice setting.

00:47:35   Because the idea is that when you're pumping audio through an audio graph,

00:47:40   there's this real-time audio thread that requests a certain number of samples every time it calls back your function.

00:47:46   Your function has to give it that number of samples.

00:47:48   And usually that number was either 256 or 512.

00:47:52   And so it was calling your function. Every 512 samples, and keep in mind there's 44,000 of these per second most of the time,

00:47:58   so it's calling this function a lot. When the screen turns off, the phone switches to a lower power version of this,

00:48:06   where instead of requesting smaller blocks more frequently, it starts to request much bigger blocks.

00:48:12   So it doesn't have to poll you as often, so the CPU can be in a lower power state more of the time.

00:48:18   And so it starts, instead of asking for 512 or 256, it starts asking for, I believe, 4,096.

00:48:25   And if you didn't set this property on all your audio units that was barely documented, welcome to Core Audio,

00:48:31   if you didn't set this property to at least that 4,096 number to say, like, what's the maximum number of audio samples I can provide you at once,

00:48:41   this exact problem would happen. If you turn the screen off, your audio would just pause.

00:48:47   And then you turn it back on, it would unpause. The challenge of this to me is that once you have an AV audio engine type API,

00:48:55   you can't set that property anywhere. There's nowhere where you can do that.

00:48:59   What was extra weird is that if you do that wrong with an AU graph, the old way, it fails every time.

00:49:05   You can reproduce it 100% of the time in development, just turn your phone screen off and it pauses, you're like,

00:49:09   "Oh, I got a problem." You can fix it, and you can see if your fix works. But this is an intermittent problem I was having.

00:49:14   It never happened to me once. I could not get it to happen in development at all,

00:49:20   but every beta build I would send out, I would get a large number of people saying,

00:49:24   "Within a few hours of it going out, this is happening, this is still happening, this is still happening."

00:49:29   So the only way I had to test whether my fix was working was to issue a new build to all the beta testers,

00:49:35   send it out to, like, thousands of people, hope many of them install it, and then look at feedback 8 to 12 hours later

00:49:41   and see how many people were saying this and then instill this feedback.

00:49:44   And so I was just, like, blindly shooting in the dark here, and eventually I found,

00:49:50   so based on that hunch of maybe it was related to that old tech note, I eventually found there's a property on AB Audio session

00:49:59   where you can tell it what latency duration to use, basically, like you can set your preferred IO buffer,

00:50:06   and if you set it to something, it effectively disables the automatic switching between more blocks, less blocks.

00:50:14   So I just set it to exactly 81.92 samples, which is twice the old 49 to 6.

00:50:22   I set it to exactly that number, and the problem vanished.

00:50:26   So I think I have solved this problem. I have not gotten any reports of it, or at least many.

00:50:34   It's gone from, like, a ton to, like, two or three, so it seems like I think I have fixed it.

00:50:40   And, oh my goodness, it's been quite a ride.

00:50:44   So anyway, that's a long, long, long story to say. There's this crazy audio bug in iOS 13 that happens sometimes to some people.

00:50:51   I think I have fixed it in this version of Evercast that might be out by the time you listen to this.

00:50:56   And, oh god, this is the kind of thing that just crushes my morale, because it's so frustrating.

00:51:04   I'm sure all developers go through this. When you have a bug, not only can you not reproduce it,

00:51:09   it doesn't even make sense that it could happen, and then you go through all these crazy difficult tricks and tests and logs,

00:51:17   trying to figure out how the heck can this possibly be happening, and then eventually you try some really obscure thing.

00:51:23   And I think this worked, but I'm like, first of all, it's frustrating that I can't be sure that it worked.

00:51:30   And it's extra frustrating that, like, there's this weird undocumented thing that, or barely documented thing.

00:51:37   What hope does anyone else have of ever figuring this stuff out?

00:51:41   Like, I happen to know, from having written this old audio engine five years ago,

00:51:46   I happen to know that there was this little weird thing in the API that you had to do to make it background properly.

00:51:52   Had I not known that, I never would have figured this out.

00:51:56   There's nothing on Stack Overflow, there's nothing anywhere that's useful.

00:51:59   There's one discussion thread on Apple forums that references this Q&A, which is how I eventually found it.

00:52:05   I can't overstate how important it is that Apple document their damn APIs.

00:52:13   Like, so much of Apple development, you're shooting in the dark here,

00:52:17   and the documentation is either wrong or incomplete or just totally missing.

00:52:25   Audio is definitely one of those areas where there's almost no documentation for audio,

00:52:30   and what's there is usually incredibly incomplete.

00:52:33   And the errors that you get are usually incomprehensible to the nonexistent.

00:52:38   Weird stuff happens, like, the main thread will get blocked for like five seconds by a random audio call to start playing audio.

00:52:45   Like, it's just weird stuff happens with the audio subsystem,

00:52:48   and then whenever there's a new OS version or a new beta season, audio gets really messed up.

00:52:54   I don't know why, like, it doesn't seem like the kind of thing they would be changing often, but they do change it often.

00:53:00   And some of it makes sense, like, you know, when they launch new things like AirPods,

00:53:03   and you know, you gotta have all the support in place for that, and stuff like the audio sharing and AirPlay 2,

00:53:07   like, they do a lot of stuff in the audio realm,

00:53:10   but the audio subsystem is so buggy and so undocumented,

00:53:15   and so just, you know, bad at reporting errors, and the tooling around it is nonexistent,

00:53:21   so it's hard to even test most of these error cases.

00:53:24   Like, it's so hard to make an app that just plays audio reliably.

00:53:30   The audio APIs on iOS are so incredibly difficult to work with for unnecessary reasons,

00:53:37   you know, things like bugs and documentation problems, not the APIs themselves.

00:53:41   They have some problems, but they don't have major problems.

00:53:44   I would never have guessed when I set out to run an audio app that I would be fighting the platform so often.

00:53:50   I would imagine that the common cases that they use, whatever it is, AV Audio Player,

00:53:54   like the super high-level API, which is like, "Don't you worry about audio.

00:53:58   You just tell us what you want to play, and we'll handle all the details,

00:54:01   and no, you can't control anything about it or have limited control, but don't worry, we'll get it done."

00:54:05   Like, that I feel like is probably the most, you know, the paved road, as they say, for audio.

00:54:13   Obviously, you are going off the beaten path.

00:54:15   You want to get at the guts, you want to get right in there, you want to make your own engine,

00:54:19   you want to have total control over the whole stack, and then you are essentially working with the guts of their system,

00:54:25   which, you know, it doesn't surprise me that they're totally undocumented and buggy and weird,

00:54:29   because they're like, "Oh, well, no one's going to ever..."

00:54:31   And I know they're public APIs. I know you're not using private APIs,

00:54:33   but they're like, "Well, no one's going to ever use these things."

00:54:35   I mean, who knows if they even have tests for them or how they would test it.

00:54:38   They're like, "As long as our top-level front door just please play an audio file for me,

00:54:44   as long as that works fine, we'll test that and make sure that mostly works."

00:54:47   But anyone mucking around on the internals, like, and you don't have the source code to it, right?

00:54:52   So if they're like, "Oh, I could write a good audio engine,

00:54:54   because here next to me is the source code to the audio system,

00:54:57   and I can see how it's all working, and I can step through it in the debugger and say,

00:55:00   "Oh, I see what the problem is. I see right here there's this conditional thing.

00:55:03   Oh, you've just gone into the background and changed the buffer size."

00:55:07   You'd see that if you were working on the guts.

00:55:10   You're working with the guts, but you don't have access to the guts,

00:55:13   so you're just fumbling around in the dark saying, "Does this work? Does this work?

00:55:16   This thing's not documented, but it seems like it has these functions,

00:55:20   and I guess you're supposed to use them this way, right?"

00:55:23   If you could just see the comments above that function to be like,

00:55:26   "Oh, here's what you've got to do with this, and keep in mind X, Y, and..."

00:55:28   But you can't see any of that. It's just a bunch of binary libraries.

00:55:31   So I feel your pain.

00:55:34   But yeah, and it doesn't, you know...

00:55:37   Lobbying for them to make that better is probably much more difficult

00:55:40   than if you said like, "Oh, I just tried to play an audio file

00:55:42   with your top-level audio API, and it didn't work."

00:55:45   Yeah, and it goes beyond just that I'm using special APIs.

00:55:49   It's also, in a lot of cases, it's really hard to tell if it's even my bug,

00:55:54   because iOS 13 in general has had a lot of audio bugs.

00:55:57   I find even just switching in and out of AirPods,

00:56:02   or I find bugs in the music app, trying to play music between itself,

00:56:08   or HomePods, or again, AirPods, like switching devices, CarPlay.

00:56:12   There's so many problems that are audio-related in iOS 13,

00:56:16   not to mention all of iOS 13's other issues.

00:56:18   For instance, I still have to go back to the All Inboxes list in Mail,

00:56:22   and then back into the Inbox to refresh messages sometimes.

00:56:25   Again, I don't know why Apple thinks it's okay to break the Mail app so often,

00:56:32   but they do, but they also seem to break audio very often.

00:56:36   So I don't even know if this is my bug or not sometimes.

00:56:41   A lot of times, if the audio engine just hangs,

00:56:44   and I've given it the exact same conditions that I've always given it,

00:56:48   and this time it hangs and last time it didn't,

00:56:50   how am I supposed to know, is this my issue, or is this,

00:56:55   that that person's phone had a weird little glitch with iOS 13.1 point whatever,

00:57:00   and that's actually an OS issue?

00:57:02   Again, it adds to the massive rippling effects of iOS 13 being such a buggy release.

00:57:10   It puts a huge tax on developers all year long,

00:57:15   because we have to also fight around the bugs with our own apps

00:57:20   and try to figure out, first of all, is this our bug, or is it not our bug?

00:57:25   And if it's not our bug, do we have to work around it somehow anyway,

00:57:29   because Apple's not fixing it?

00:57:31   - Which framework is all of this stuff in?

00:57:33   Is this Audio Toolbox or AV Foundation or none of the above?

00:57:37   - There's a lot of overlap there.

00:57:40   Most of this that I've been telling you about is AV Foundation,

00:57:44   but under the hood it uses a bunch of Audio Toolbox stuff.

00:57:47   - Well, but Marco, AV Foundation is 88.2% documented

00:57:50   as per no overview available.

00:57:52   Surely there's almost nothing missing.

00:57:54   - Well, and the great thing is, a lot of times,

00:57:57   this won't be measured correctly,

00:57:59   because most of these functions actually do have official sentences

00:58:05   to describe what they do in the documentation.

00:58:07   They're just incredibly incomplete and sometimes wrong.

00:58:11   - Oh, yeah, no argument.

00:58:12   I'm snarking, but no overview available was done,

00:58:15   I believe by Matt from NS Hipster, if I'm not mistaken.

00:58:19   But anyways, we've talked about it in the past,

00:58:21   but it just visually shows you how godawful Apple documentation

00:58:25   has become over the last few years.

00:58:27   And I'm assuming it was better in the past, maybe it wasn't,

00:58:30   but there is a lot of stuff here

00:58:34   that is just really, really poorly documented,

00:58:37   and I have been whining and complaining and moaning

00:58:40   and fetching to anyone who I know inside Apple

00:58:43   who will listen to me, which is basically nobody.

00:58:46   I know a lot of people who don't listen to me,

00:58:48   saying, "This is bold. This is awful.

00:58:50   "I can't do my job as effectively as I want,

00:58:54   "because you guys, for political reasons, for, I don't know,

00:58:58   "some sort of..."

00:58:59   I don't know what it would be other than political reasons.

00:59:02   I mean, they have enough people, they have enough money.

00:59:04   I know that there are teams working on documentation,

00:59:06   but somehow it never makes it out.

00:59:08   And just pointing to a WWDC video and saying,

00:59:10   "There's your documentation," is obnoxious at best

00:59:12   and just plain wrong at worst.

00:59:15   I don't know, this is a bugbear

00:59:17   that's really been driving me nuts recently,

00:59:19   and I ran up against this as recently as earlier today,

00:59:22   and so I've got a real bee in my bonnet about it.

00:59:24   But it's just so frustrating.

00:59:27   I guess maybe ultimately the problem is that

00:59:31   I think of myself as more important to Apple,

00:59:33   not me, Casey, but developers, third-party developers,

00:59:35   as more important to Apple than we really are.

00:59:38   But you would think for a company that wants to provide

00:59:41   really robust, well-done APIs, and often does,

00:59:45   yet they don't spend the time to actually document anything,

00:59:48   and to your point, Marco, half the time it's documented,

00:59:50   it's incompletely documented or wrongly documented.

00:59:53   It's just unfair.

00:59:55   Like, I know I'm kind of whining right now,

00:59:57   but it's just plain unfair.

00:59:59   Well, and I think it speaks to a bigger problem of, like,

01:00:02   the technical foundations of this company are rotting.

01:00:06   This is a company that is doing way too much

01:00:09   and is pushing way too hard on the new stuff

01:00:12   and is not allocating its resources properly

01:00:15   to keep things at a high-quality level.

01:00:18   And this goes all levels of the company.

01:00:21   This goes from, you know, from, like, the front-end apps,

01:00:24   like I was complaining last week about how they haven't

01:00:26   really designed good UIs in a long time.

01:00:29   Their design is all over the place,

01:00:30   and they seem to not have time or the care

01:00:32   to lead that correctly.

01:00:34   And then this also goes for things like software quality

01:00:37   and API stability and documentation

01:00:40   and, like, and the released OSs.

01:00:42   This is all related.

01:00:44   This is all the company not being able

01:00:47   to keep up with itself.

01:00:49   And this is an entirely self-created problem.

01:00:53   It's management problems.

01:00:55   It's, like, how you allocate resources,

01:00:57   what kind of timelines you set,

01:00:59   how you decide when to release something versus when not to.

01:01:03   This is all within their control.

01:01:06   And it's not being balanced well right now at all,

01:01:10   and it hasn't been for years.

01:01:12   And I don't know how to fix that,

01:01:13   'cause I don't know enough about, you know,

01:01:14   their internal workings or their people or whatever.

01:01:17   I don't think this is just a purely, like,

01:01:20   "Oh, fire Tim Cook" kind of problem.

01:01:21   Like, I think it's probably more complicated than that,

01:01:23   but this is ultimately on management and leadership to fix.

01:01:28   Like, it's about allocating resources

01:01:30   and setting deadlines and choosing what to focus on

01:01:32   and how to, and what's important to you.

01:01:35   And it seems like software quality

01:01:37   is not important enough to them at all,

01:01:39   and that goes to everything from documentation

01:01:42   to the APIs to the OSs to the software itself.

01:01:46   - I think one thing, I don't know if we talked about this

01:01:49   in our last, previous conversations

01:01:51   about Apple's software strategy and stability problems

01:01:54   and their release cadence and all the other stuff,

01:01:57   but it occurs to me, partly part of the conversation

01:02:01   that's been happening this past week

01:02:02   about the 10th anniversary of the iPad,

01:02:03   which we probably won't get to talk about today, but--

01:02:06   - Oh, we should.

01:02:07   - Has been cycling around,

01:02:09   that different products, you know,

01:02:14   are at different points in their lifetime,

01:02:16   different Apple products.

01:02:18   And even if they start at the same time,

01:02:19   they might progress at a different time.

01:02:21   And during the various life stages

01:02:25   of any product or product category,

01:02:27   different strategies are appropriate.

01:02:30   And at this point, with the iPhone in particular,

01:02:33   even though it is still, you know,

01:02:35   the majority of Apple's business

01:02:37   and by far their biggest product

01:02:39   and super important and everything like that,

01:02:41   the category of smartphone has slowed down

01:02:46   progress and innovation-wise as compared to, you know,

01:02:51   a decade ago or whatever.

01:02:52   Like, in the beginning, you know,

01:02:54   if you look at the first iPhone up to the iPhone 4,

01:02:57   it's like, whoa, like, there's some big changes

01:02:59   hardware and software-wise.

01:03:00   And at this point, software-wise,

01:03:03   if you look across the entire market,

01:03:05   there are important new software features

01:03:07   that you have to have.

01:03:08   Photography, Face ID was important or whatever.

01:03:11   But in general, the pace of change

01:03:14   of smartphone operating system software

01:03:16   and sort of platform features has slowed significantly.

01:03:20   And so I feel like it is an appropriate time

01:03:23   for Apple to consider leaning more towards

01:03:29   the iOS 12 strategy of, like,

01:03:32   let's just keep making what we have faster

01:03:35   and pick one or two or three important new features

01:03:37   that are required to stay ahead in the industry

01:03:40   or keep pace in the industry,

01:03:42   but not feel like, you know,

01:03:44   iOS N+1 has to advance on all fronts every time,

01:03:48   because that's just not the nature

01:03:49   of the smartphone market anymore.

01:03:51   There is not--I don't think there's pressure to do that,

01:03:53   and I think the internal pressure,

01:03:55   the sort of self-imposed internal pressure

01:03:57   of, like, the next iOS has got to be better in every way

01:04:00   and have all new features and subsystems,

01:04:03   that's not what users want,

01:04:05   that's not what the market demands,

01:04:06   and it's not required to stay ahead,

01:04:08   and it is actually detrimental to staying ahead,

01:04:10   because what's required to stay ahead

01:04:12   is one or two or three really important new features

01:04:15   that advance probably with some hardware component.

01:04:18   That's important to keep doing.

01:04:19   Like, don't just say, okay, well, it's done,

01:04:20   and we never need to add anything.

01:04:21   You'll never get Face ID that way,

01:04:22   and Face ID was important, right?

01:04:24   So pick the way you're going to advance,

01:04:26   but then all the other stuff, you're like,

01:04:28   do we need to change this?

01:04:30   Is this one of the tentpole features of the new OS?

01:04:33   If not, let's dedicate all those resources

01:04:36   to an iOS 12 strategy of refinement, fix bugs,

01:04:40   make it faster, you know, just document,

01:04:44   do all the things that you supposedly never have time to do,

01:04:47   and that should sort of be the default,

01:04:49   and then only a few things are blessed

01:04:51   in each major release as the, you know,

01:04:54   this is going to be where we're making some big advances.

01:04:57   These are going to be our big

01:04:58   customer-facing marketing features.

01:05:00   I haven't been keeping up with the rumors for iOS 14.

01:05:02   I did hear briefly in some story

01:05:04   that there was some rumor that it was going to support

01:05:06   all the same phones as 13,

01:05:08   and it immediately made me think/hope, like,

01:05:11   oh, maybe that means 14 is going to be more like 12 again,

01:05:14   where they'll try to sort of regroup

01:05:17   and make everything less buggy and faster.

01:05:19   Again, there was nothing in the one headline

01:05:21   that I saw that indicated that,

01:05:22   but the fact that I was hoping that that was the case

01:05:24   makes you feel like that's at least what I want as a consumer,

01:05:27   is when iOS 14 comes out,

01:05:30   I want it to be a no-brainer update

01:05:32   that is as stable as iOS 12 was

01:05:34   and is as pleasing as iOS 12 was.

01:05:36   Oh, everything's a little bit faster,

01:05:38   and there aren't any weird bugs,

01:05:39   and maybe there's one or two new cool features I want.

01:05:42   That, I think, is appropriate for the smartphone market,

01:05:44   and it came up in the context of the iPad

01:05:46   because the iPad is in a very different place.

01:05:48   The iPad's pace of innovation has been slow,

01:05:51   and there is tons of room

01:05:53   for major new advancements in that area,

01:05:55   so I think the strategy for iPadOS,

01:05:57   which now has its own name, but unfortunately,

01:05:59   under the covers is very much iOS but slightly different,

01:06:03   I don't know how they square that,

01:06:05   but, like, iOS needs to make some major advances,

01:06:08   maybe not competitive pressure-wise,

01:06:10   but just product-wise,

01:06:12   because the tablet market is different

01:06:14   than the phone market, because the tablet market

01:06:16   has not blossomed and matured

01:06:18   and had this sort of, you know, competitive pressure

01:06:21   to evolve through competition

01:06:23   to what the smartphone market has,

01:06:25   and since those are basically both the same OSes

01:06:27   under the cover, it's difficult to strike that balance,

01:06:30   but, hey, they did give it a separate name,

01:06:32   and I'm perfectly fine if they fork off from each other

01:06:34   and continue advancing that way,

01:06:36   and also because most of the features on the iPad

01:06:38   that the iPad needs to deal with

01:06:40   shouldn't require major guts redesign,

01:06:43   I hope, and I don't actually know.

01:06:45   Like, maybe there is something that needs to be done

01:06:47   to the guts to mess with that,

01:06:49   but that's my thinking on the software quality,

01:06:53   and in particular on the phone and iOS,

01:06:55   that it is -- I'm not going to say it's time for Apple

01:06:58   to slow down or the time to Apple to give up or whatever,

01:07:01   anything like that.

01:07:02   I'm just saying that, like,

01:07:04   they may not be choosing the right strategy

01:07:07   for each of their products' particular stage.

01:07:10   It's like picking dog food for your dog.

01:07:12   Like, what life stage is your dog in?

01:07:14   You know, you get the puppy chow, you get the --

01:07:16   what is it called? Like, the geriatric dog chow,

01:07:19   you get the adult dog chow.

01:07:21   I think it's time for the iPhone to have the adult dog chow.

01:07:26   [ Laughter ]

01:07:27   -There seems to be this disconnect

01:07:29   between the pace of change and feature evolution

01:07:34   and new stuff that Apple keeps doing

01:07:38   and what customers actually want and need.

01:07:41   And, you know, iOS --

01:07:43   So another thing that I'm doing in this Overcast release

01:07:47   is here now, like, you know,

01:07:49   four or five months after iOS 13 launched,

01:07:53   I'm restoring support for iOS 12 in this release

01:07:57   because not enough people have updated to iOS 13,

01:08:00   and adoption seems to have flattened out pretty significantly.

01:08:03   So we're standing at something like 75% adoption,

01:08:06   depending on how you measure it.

01:08:08   And that's good, but it's not good in the sense

01:08:12   that a quarter of the user base seems to not want iOS 13 at all.

01:08:16   That's bad in the sense that if you want your app

01:08:19   to run on 25% of devices, which is not a small number,

01:08:23   you can't use all this new stuff yet

01:08:26   because iOS 13 has been such a botched release

01:08:29   with bugs and everything like that that, like, you know,

01:08:31   people are avoiding it and, like, holding on to 12.

01:08:34   The same way I'm still not on Catalina

01:08:36   for all the same reasons, 'cause, you know,

01:08:38   I'm holding on to whatever the heck this is,

01:08:40   Hi Mojave, whatever I'm running.

01:08:42   And, like, and so what Apple's customers want

01:08:46   is not, you know, new ways to pose animoji with faces

01:08:51   and glue them together with iPad drag and drop.

01:08:53   Like, no, what Apple's customers want is reliable phones

01:08:56   and good battery life and good performance

01:08:58   on their older stuff.

01:08:59   When you introduce something like iOS 13,

01:09:01   like, yeah, there's some fun new stuff in it here and there,

01:09:03   some nice features, but there's a big price to pay

01:09:06   in the much more important stuff of

01:09:09   everything needs to work well and, again, to a lot of people,

01:09:12   it has to work quickly on old phones

01:09:14   and it has to support their old phone to begin with

01:09:16   and stuff like that.

01:09:17   And if they're falling down on those fronts,

01:09:19   nobody cares about the new features.

01:09:21   And we have this series of releases now

01:09:24   where it seems like the new features

01:09:25   are less compelling than ever

01:09:28   at the same time that the basics are getting screwed up

01:09:32   more often than ever.

01:09:34   And that's dangerous.

01:09:35   That shows that, like, things are not being managed correctly.

01:09:38   Like, this is a problem.

01:09:40   I don't know whose problem it is or how to solve it,

01:09:44   but this is ultimately a management and resource allocation

01:09:48   and release planning type problem.

01:09:49   And this doesn't seem,

01:09:52   it doesn't seem like Apple is taking this seriously.

01:09:55   It doesn't seem like they're actually making changes.

01:09:57   Like, we hear rumors about, like, you know,

01:09:59   Federighi having memos or meetings where they say

01:10:02   they're gonna, like, you know, get better,

01:10:03   but what if they actually change to make that happen?

01:10:07   Are they not gonna release every year anymore?

01:10:10   Are they gonna cut back significantly on new features?

01:10:13   Are they going to, you know,

01:10:14   make the old OS run on new hardware?

01:10:16   Who knows?

01:10:17   It doesn't seem like they've actually changed anything.

01:10:20   It seems like all they've done is

01:10:23   say they're gonna get better.

01:10:25   But we haven't heard anything about actual changes

01:10:27   that have actually happened

01:10:28   that would be material to that.

01:10:30   So that I'm still very much concerned about.

01:10:33   You know, anybody can say I'm gonna get better,

01:10:35   but without actually changing significant things

01:10:38   that will enable that change,

01:10:40   I don't know how that's ever gonna happen.

01:10:42   - Well, they did iOS 12,

01:10:43   which shows that they can do a release like that.

01:10:47   Of course, we don't have any transparency,

01:10:48   so we don't know, like I was saying,

01:10:50   is iOS 14 on an iOS 12-like path, or is it not?

01:10:53   We're not gonna find that out

01:10:54   until basically they announce iOS 14,

01:10:56   and then they'll either brag about the fact

01:10:58   that they concentrated on performance

01:11:00   and stability enhancements, or they won't,

01:11:01   and we'll have to surmise from what they say publicly

01:11:03   whether it was like that.

01:11:05   - They're gonna say that anyway

01:11:06   because 13 had a bad reputation,

01:11:08   the same way like iOS 11 had a bad reputation,

01:11:10   which is why iOS 12's marketing message

01:11:12   was basically we're focused on stability and performance

01:11:15   because 11 was a mess.

01:11:16   - But they did, though.

01:11:17   But they did, and 12, I feel like they succeeded

01:11:20   in that strategy of giving a stable OS

01:11:23   that has fewer features

01:11:24   that was far less buggy than 13 for sure,

01:11:28   and also than 11,

01:11:29   and it was generally a well-received OS.

01:11:32   Customers liked it.

01:11:33   People upgraded to it.

01:11:34   It was stable.

01:11:35   It did what they said.

01:11:36   It wasn't just marketing BS, right?

01:11:38   But we didn't know that until 12 basically was announced,

01:11:41   so if Apple is doing that now,

01:11:43   we're not gonna know about it

01:11:44   because they're not gonna tell us

01:11:45   about what they're doing internally,

01:11:46   but I think when iOS 14 is announced, they'll tell us,

01:11:50   and 13, they didn't make the same pitch as 12.

01:11:53   13, it was like, "Look at all these new features."

01:11:55   The pitch was not, "We really regrouped them,"

01:11:57   or, "We're mostly concentrating on performance."

01:11:58   That wasn't the 13 pitch at all,

01:11:59   so I think it is totally conceivable

01:12:01   they could be doing that,

01:12:02   but yes, it's frustrating that we don't actually know

01:12:04   because that's sort of their own internal business,

01:12:06   and you don't know their internal business.

01:12:08   We'll find out when they announce iOS 14,

01:12:11   but here's hoping.

01:12:13   - Well, we don't know except when we do.

01:12:15   There was a "Gurman" article from late November,

01:12:18   "Inside Apple iPhone Software Shake-Up

01:12:21   "After Buggy iOS 13 Debut,"

01:12:23   and this article talks about how apparently Federighi

01:12:26   got everyone, all the troops in the same room,

01:12:29   and basically said, "Hey, we're gonna do

01:12:30   "a whole bunch of things totally different,"

01:12:32   and I don't remember the exact details,

01:12:34   and I haven't read this article since it came out,

01:12:36   but the short short of it is that they're planning,

01:12:39   or they were planning on changing

01:12:41   kind of the way they develop stuff leading up to release,

01:12:46   and I had heard from a couple of different people

01:12:49   that this article, while the details were wrong,

01:12:52   surprise, surprise, but the general gist of it was accurate,

01:12:55   that they are going to rework the way they do their work,

01:12:59   and I don't know any more specifics than that,

01:13:01   but that is indicative to me that there is at least

01:13:04   some amount of acknowledgement and understanding

01:13:06   that things are broken.

01:13:07   Now, with that said, if it's anything like

01:13:10   any of the places I've worked,

01:13:12   the answer to a broken process is what, John?

01:13:15   - More process.

01:13:16   - Oh, yeah, more process,

01:13:18   and so it very well could be that this is just gonna

01:13:20   make things worse and not better,

01:13:21   but it's at least a small step in the right direction,

01:13:26   and I think that we should acknowledge that,

01:13:28   even though I do agree with everything both of you guys said.

01:13:31   - I think that's what Marco was getting at, though.

01:13:32   Like, when we hear stories like that,

01:13:33   which obviously are not coming directly from Apple,

01:13:35   so it's always, you know, a game of telephone or whatever,

01:13:37   that that always sounds like the stuff you hear

01:13:40   in any company when something bad goes wrong.

01:13:42   They're always going to say,

01:13:43   "We're gonna do things differently,"

01:13:44   or "We're gonna try harder," or whatever,

01:13:45   and I feel like because we're not privy to what's actually going on,

01:13:49   rumors like that could go either way,

01:13:51   'cause they're gonna say that internally no matter what,

01:13:53   and it could be they're just like,

01:13:54   "Don't actually make any changes that effectively change things,"

01:13:57   but I feel like when they announce the OS,

01:13:59   Apple knows by that point, "Have we pulled it off?"

01:14:01   Like, when they were announcing iOS 12,

01:14:02   it wasn't just like iOS 12 was gonna be a stability release.

01:14:05   They had been looking at iOS 12.

01:14:06   They know, "Guess what?

01:14:07   "It's pretty--the number of bugs is low.

01:14:10   "We did what we set out to do."

01:14:12   By the time they're announcing it to us,

01:14:14   they know whether the thing they said

01:14:16   nine months earlier turned out to just be a bunch of BS

01:14:19   or whether it actually changed things for the better,

01:14:22   whether their process changes actually, you know,

01:14:24   so that's why I feel like

01:14:25   because we don't have a view of the internals,

01:14:28   I think it's--we just have to assume

01:14:30   anything they say to themselves internally is meaningless

01:14:33   and could go either way,

01:14:34   but when they say it to us and we get the bits,

01:14:38   you know, and when they say it to us, they're honest.

01:14:40   When they say it to us,

01:14:41   if they emphasize that it's stable and performant,

01:14:44   it probably is, right, 'cause they already know.

01:14:47   Like, you know, the WWDC build is a good way to judge things.

01:14:51   Sometimes the WW--we always know the WWDC build is early,

01:14:54   but if the WWDC build is actually kind of okay,

01:14:56   that is a very good sign, right, and if it's not okay,

01:15:00   it could still go either way, but it's not a good sign,

01:15:03   so, you know, I'm waiting for June.

01:15:06   [laughter]

01:15:08   Like, I'm also not entirely sure that their alleged process changes

01:15:13   from the article are going to be enough,

01:15:16   even if they do it, I mean, 'cause, you know, it did--

01:15:18   I mean, honestly, it did kind of sound like,

01:15:20   "We're gonna fix this by working harder,"

01:15:22   which is not ever the solution to any real problem,

01:15:26   but right now, iOS 13 is still full of tiny little paper-cut bugs,

01:15:32   and it's been, like, five months.

01:15:35   If they make it so that the initial releases are better,

01:15:39   what does that put them at?

01:15:40   About where we are now?

01:15:41   This still isn't good enough,

01:15:43   and from what I'm guessing, or from what I'm hearing,

01:15:45   Catalina, which is only slightly younger,

01:15:48   is not doing that much better.

01:15:51   It's similar to the issue of what I said last week

01:15:55   that I honestly don't think Apple is capable

01:15:57   of making great UIs anymore.

01:15:59   It's almost seeming like they're not capable

01:16:02   of shipping high-quality software anymore,

01:16:04   and that's really scary,

01:16:05   because if you look at what they've actually shipped,

01:16:07   you know, over the last, I don't know, five years,

01:16:10   there's not a lot of bright spots in quality.

01:16:13   There's not a lot of releases you can point to

01:16:15   and say, "That one was really good."

01:16:17   Even iOS 12 had some issues.

01:16:18   Like, it wasn't as bad as 13, but, like,

01:16:20   iOS 12 is pretty good, but I think that's been the best one.

01:16:23   Like, even whatever the Mac OS version I'm running now

01:16:26   was not great.

01:16:27   Like, Mac OS seems to have even more issues than iOS

01:16:29   a lot of times, but, like, I don't know.

01:16:31   I'm concerned.

01:16:32   Like, what they need to change are, like,

01:16:35   fundamental priorities and resource allocations

01:16:39   and process things, and I don't know

01:16:41   that they've changed those, and I don't know that,

01:16:43   I mean, a lot of that might even be above

01:16:45   Craig Federighi's head, so we don't even know

01:16:48   if he can change those, but regardless,

01:16:51   like, it's unclear whether that level of change has happened,

01:16:55   and anything less than that, if they're just saying

01:16:57   we're gonna work harder and we're gonna, you know,

01:16:59   feature flag disable bad stuff until we know it's ready,

01:17:02   that isn't necessarily gonna be enough.

01:17:05   - I'm glad at least I have a sort of canary for OS quality,

01:17:10   kind of like the app modal preference dialog box

01:17:13   in iTunes was a good canary for whether

01:17:15   they've actually changed the app.

01:17:16   Now I have the, you know, the undead toolbars

01:17:20   that keep popping up in my Finder window.

01:17:22   I'll know that someone has finally paid attention

01:17:24   to Mac OS when that bug gets fixed.

01:17:27   I figure it will probably last another three or four

01:17:28   releases before anyone even acknowledges its existence,

01:17:31   and then in five or six years it'll get fixed,

01:17:33   'cause nobody cares about these bugs,

01:17:35   'cause everyone just leaves the toolbar

01:17:36   visible all the time, so why should anyone care?

01:17:39   But yeah, it's good to have some way to measure,

01:17:45   like, is my personal experience of Mac OS stability

01:17:49   or iOS stability representative,

01:17:51   is there a systemic problem?

01:17:52   Obviously, part of that is you just look at the press

01:17:53   and see if other people are complaining about it,

01:17:55   whatever, but paper cut bugs, like Margot was talking about,

01:17:58   sometimes those last for a really long time,

01:18:00   'cause no one cares about that paper cut,

01:18:02   or no one is staffed to work on that application at all,

01:18:05   so it is not being changed at all,

01:18:07   or if it's being changed, this was a thing

01:18:09   back in the early days of Mac OS that was an epidemic,

01:18:11   and people on Apple used to complain about it.

01:18:13   Like, you know, in 10.2 or 10.3 or whatever,

01:18:16   you'd be responsible for the whatever, the mail app Finder,

01:18:19   and the only thing you'd get to do in a release

01:18:22   is you would be told that what you have to do

01:18:25   is implement support for and certain new headlining feature

01:18:28   of Mac OS here, whether it's Spotlight

01:18:30   or some new graphic system

01:18:32   or make a dashboard widget or whatever,

01:18:34   it's like, but I wanted to do feature X with my application,

01:18:38   or I wanted to refactor something

01:18:39   or work on stability and performance.

01:18:41   It's like, nope, you're doing Spotlight support,

01:18:43   because that's the thing that we all, you know,

01:18:45   there were so many major new features to the OS

01:18:47   that it was important for Apple to be able to say,

01:18:49   look, here's a major new technology or feature,

01:18:51   and all our applications use it to great benefit

01:18:53   to show you how good it is,

01:18:55   and that would mean that with small teams,

01:18:57   that's all they could do during the release,

01:18:58   so they wouldn't have time to address just basic bugs

01:19:01   with the boring functionality,

01:19:02   because they would be worried about just making their app

01:19:05   work with the new feature,

01:19:06   integrating it in a good way

01:19:07   and making sure that their thing doesn't crash,

01:19:09   because now they're linking to some library that, you know,

01:19:11   is being released for the very first time.

01:19:13   That I don't think is happening anymore,

01:19:15   but I can definitely see signs that applications

01:19:18   that need some love and attention

01:19:20   are getting very little of it,

01:19:21   probably a fraction of some single person's time,

01:19:24   just to make sure they still build

01:19:26   and function more or less correctly in the new OS,

01:19:29   but beyond that, not much.

01:19:30   Hopefully, like, mail on iOS does not fall into that category.

01:19:33   Hopefully, mail on iOS does not have a fraction

01:19:35   of a single person's time working on it,

01:19:37   but the Finder, I look at the Finder, and I say,

01:19:39   well, in this release, it's clear that they mostly worked

01:19:41   on the thing where it replaces iTunes for your device management,

01:19:44   and that must have been a super pain in the butt

01:19:46   filled with tons of old, creaky code pulled out of iTunes,

01:19:49   so I feel for that team.

01:19:51   How toolbars are suddenly spawning,

01:19:53   it's probably unrelated to that work,

01:19:56   but I'm not surprised that they didn't have time to fix it

01:20:00   and won't in the near future,

01:20:01   because the iTunes integration stuff is probably buggy,

01:20:04   and they'll probably have to fix that in the next release.

01:20:06   - On the other side of the coin, though,

01:20:08   how quickly all three of us forget

01:20:10   that they changed out the file system

01:20:13   on, like, what, two billion devices,

01:20:16   and I have not heard a single problem with any of them,

01:20:19   not to say that I'm the clearinghouse of it, but--

01:20:21   - That was a project that took, depending on how you measure,

01:20:24   like, a decade, so certain internal projects

01:20:27   have slightly longer timelines.

01:20:29   Nobody was saying, we need a new file system,

01:20:31   we need it by this release, or maybe they said that

01:20:34   a couple times, like when they announced, pre-announced ZFS,

01:20:36   but it's like, I don't know the political things

01:20:40   that made that happen, but tech-wise, practically speaking,

01:20:45   that was worked on for so long at a seemingly leisurely pace,

01:20:50   which is the right thing to do for the file system,

01:20:52   and that, you know, I'm glad they pulled it off.

01:20:55   And with many sort of false starts of, like,

01:20:57   what kind of file system do we want to have,

01:20:58   but what about one like this, what about one like that?

01:21:00   I heard a story recently, I think it was some ex-Apple person,

01:21:04   like, spilling some beans from long dead stuff,

01:21:07   but, like, Spotlight, the, I don't know if this is true,

01:21:10   I don't even remember the source,

01:21:11   'cause I see a lot of things on the internet,

01:21:12   but the idea was that Spotlight, search engine we know and love.

01:21:16   Well, at least the search engine we know.

01:21:18   Yeah. The one of, it began its life with one of the many,

01:21:21   many efforts to figure out what to do about the file system,

01:21:25   because, you know, HFS and HFS Plus,

01:21:27   it had been creaky for a long time,

01:21:29   and the idea was, like, let's just do sort of,

01:21:33   instead of doing everything with file hierarchy,

01:21:35   let's make a file system that does everything,

01:21:37   sort of just by indexing metadata, right,

01:21:40   and just treat file path as another piece of metadata,

01:21:42   which semantically is what every file system does,

01:21:45   but, like, let's do it, like, physically,

01:21:47   like, let's not privilege paths and anything else in the file system

01:21:50   that have other metadata hanging off,

01:21:51   let's just make everything arbitrary metadata

01:21:54   in this big sea of things,

01:21:55   and you can view it as a file system if you view it

01:21:57   in terms of that one piece of metadata,

01:21:59   but it is not privileged over any other,

01:22:01   and apparently that didn't work out,

01:22:03   but by the time they'd done that enough to experiment with it,

01:22:06   they said, well, this is not gonna be our new file system

01:22:09   'cause it doesn't fulfill that purpose very well,

01:22:11   but it's a pretty good search engine,

01:22:13   so can we reuse it for that?

01:22:15   And that's what they did.

01:22:16   They made it a search index system

01:22:18   that lives on top of HFS+

01:22:20   but is not itself a replacement for the file systems.

01:22:23   That's, according to the story anyway,

01:22:25   one of their experiments about what to do about the file system

01:22:27   actually did ship but not as a file system,

01:22:29   so eventually we got APFS,

01:22:31   but there were many other things, including ZFS,

01:22:33   in between now and then,

01:22:35   and that is a project and an effort

01:22:38   that probably didn't have that many people on it,

01:22:41   like a handful of engineers,

01:22:43   but they worked on it for like 7 to 10 years

01:22:46   and had many blind alleys and restarts or whatever,

01:22:49   and obviously not every project,

01:22:51   especially a project that's like for competitive purposes,

01:22:53   like, oh, to be competitive,

01:22:55   we need to have, like, a Face ID-like feature.

01:22:57   You can't spend 7 years on that

01:22:59   because the world will pass you by,

01:23:01   but for file systems, they dragged it out

01:23:03   pretty much as long as they can, and they did pull it off,

01:23:05   so props to them.

01:23:07   We are sponsored this week by BlueVine.

01:23:10   Owning a small business is no easy task,

01:23:12   and sometimes your cash flow just isn't moving

01:23:14   at the speed of your business.

01:23:16   Through BlueVine, you can secure a business line of credit

01:23:18   in as little as 5 minutes.

01:23:20   BlueVine is an easy, fast way to help support your business growth

01:23:23   with a line of credit of up to $250,000.

01:23:26   Whether you need money to offset upfront costs,

01:23:28   secure inventory, or pay an unexpected expense,

01:23:30   through BlueVine, you can help yourself

01:23:32   and your business stay secure for any reason.

01:23:35   There's no fee to set up your line of credit,

01:23:37   and BlueVine never charges maintenance or prepayment fees.

01:23:40   Applying is easy. Just go online to getbluevine.com/atp.

01:23:44   Fill out a few simple details,

01:23:46   and you are done with your application within minutes.

01:23:48   And seeing an offer will not affect your credit score.

01:23:51   Once you get approved, funds can be received

01:23:53   in as little as 24 hours.

01:23:55   Have peace of mind knowing that funds can be drawn

01:23:58   with the click of a button for any business expense.

01:24:01   BlueVine has helped more than 20,000 customers

01:24:03   and has delivered over $2.5 billion in funds to businesses.

01:24:07   BlueVine also has advisors available by phone

01:24:10   to answer any questions and help meet your business needs.

01:24:13   With an A+ rating from the Better Business Bureau

01:24:16   and nearly 5-star reviews on Trustpilot,

01:24:18   see why thousands of satisfied business owners

01:24:21   have chosen BlueVine as their go-to source for financing.

01:24:24   For listeners of the show, BlueVine is offering

01:24:26   a special, limited-time promotion of a $100 gift card

01:24:30   when you take out a loan or open a line of credit with BlueVine.

01:24:32   Go to getbluevine.com/atp for more details.

01:24:37   All you have to do is go to getbluevine.com/atp and apply.

01:24:41   That's quick, easy, and meaningful help to your business

01:24:44   in as little as 24 hours.

01:24:46   This promotion will offer a subject to terms and conditions

01:24:48   that can be found at getbluevine.com/atp.

01:24:52   Thank you so much to BlueVine for sponsoring our show.

01:24:55   [music]

01:24:58   - Marco, are you using your iPad these days?

01:25:00   - Every day.

01:25:02   - For anything other than playing overcast in the kitchen?

01:25:04   - Well, I use it for, like, you know, email

01:25:08   and stuff like that sometimes, like light web browsing,

01:25:11   occasional, like, you know, Amazon ordering and stuff.

01:25:14   But I don't know, it's funny, like, literally just today,

01:25:17   I was booking a trip where I just had to book, like,

01:25:20   you know, a flight in a hotel, standard stuff,

01:25:22   and I started doing it on my iPad to, like,

01:25:25   look up flights and everything, and then I went to my computer

01:25:27   to actually book it because I knew I'd have to be on the

01:25:29   iPad, I knew I'd have to be, like, juggling multiple tabs

01:25:32   and pulling stuff out of one password and getting, like,

01:25:35   my TSA pre-number, stuff like that, like,

01:25:37   and I knew that would just be, I knew I could do it all

01:25:39   on the iPad, but I also knew it would be a lot easier

01:25:42   to just do it on my Mac, so I went to the office

01:25:45   and did the whole thing here instead.

01:25:47   - Don't blame you, don't blame me at all.

01:25:49   I don't know, I have such a, I'm not love/hate,

01:25:54   'cause I like my iPad, so I have, and I presume you do

01:25:57   as well, the, as we are recording this,

01:26:00   the most recent iPad Pro, I have an 11-inch,

01:26:04   so this is the first one with Face ID,

01:26:07   and, you know, I was, I was at Declan's swim lesson

01:26:12   this afternoon, and I was trying to draft a blog post,

01:26:16   and I was using the excellent Working Copy app,

01:26:19   which is a Git client that's actually a pretty decent

01:26:21   text editor, especially for Markdown,

01:26:24   and I was able to, like, set the iPad wherever Michaela

01:26:28   happened to be standing at that moment

01:26:30   as she was running around playing with things,

01:26:32   and write a bit, and then if she ran somewhere else,

01:26:34   it was so easy to just grab it, fold it up,

01:26:36   and go somewhere else with it, and it was light,

01:26:38   and I didn't even really have to think about it

01:26:40   when I was carrying it, and in so many ways,

01:26:42   I love it so much, but then I was trying to book

01:26:45   our summer vacation just a few days ago,

01:26:48   and I was doing a lot of, like, preliminary research

01:26:51   on the iPad and realizing that it's just frustrating,

01:26:55   like, even though I have about the same amount

01:26:56   of real estate as my MacBook Adorable,

01:26:58   I just can't use it the way I want to,

01:27:00   because the multitasking on iPad,

01:27:02   as I'm sure we'll discuss, and Gruber has talked about

01:27:05   a lot recently, is difficult to unintelligible,

01:27:08   to incoherent, to incomprehensible,

01:27:11   depending on where or what you're trying to do,

01:27:13   and it's just, it ends up that all too often,

01:27:17   if I'm doing anything even remotely serious,

01:27:20   even something like just trying to plan a trip,

01:27:23   I'm reaching for a computer, and oftentimes,

01:27:25   I'm going to my office and going to the 27-inch screen,

01:27:28   because that's the easiest one to get things done on,

01:27:30   and so I completely echo what you've said.

01:27:32   - What's a computer?

01:27:33   - What is a computer, even?

01:27:35   It's a thing where you can actually get things done.

01:27:37   That's what it is.

01:27:38   - Oh, no, oh, God, please don't.

01:27:39   Oh, God, all the emails.

01:27:40   - Please email Marco.

01:27:41   - I can feel the emails from here.

01:27:42   - Please email Marco.

01:27:44   So anyway, so it's just, it's tough,

01:27:46   because when my iPad does what I want it to do,

01:27:50   it feels like the future, man,

01:27:53   but all too often, it feels like a future

01:27:56   that hasn't quite arrived yet.

01:27:59   So, Jon, I think of the three of us,

01:28:01   you're probably the most consistent iPad fan,

01:28:04   and probably use yours the most.

01:28:07   What are you on these days,

01:28:08   and what are you doing with it?

01:28:10   - I'm still just using a regular iPad Pro.

01:28:12   I need to upgrade.

01:28:13   I also have, this is a personal issue,

01:28:15   as my eyes get worse,

01:28:17   I'm constantly adjusting my glasses prescription,

01:28:21   because I'm having more trouble seeing close up,

01:28:23   and so at a certain point, you go to bifocals,

01:28:25   and my eye doctor keeps telling me

01:28:26   I'm not quite there yet for bifocals,

01:28:28   but it is still, with my glasses on,

01:28:31   especially when my eyes are tired,

01:28:32   it gets harder and harder for me to focus on

01:28:35   something held in my hand close to my face, right?

01:28:39   So when I use my iPad now, there's a distance.

01:28:43   I can't bring it, I don't have to hold it at arm's length,

01:28:47   but I used to apparently hold it

01:28:48   much closer to myself than I do now,

01:28:50   so now I have to put it back like one or two inches

01:28:52   to be able to focus on it when my eyes are tired at night,

01:28:54   and that is actually cutting into my iPad time.

01:28:56   A lot of times, what I will switch to instead

01:28:58   is using my phone super close to my face with my glasses off,

01:29:02   just as a more comfortable thing to do

01:29:04   at night when my eyes are tired,

01:29:06   but that's nothing to do with the iPad.

01:29:08   That's just me personally.

01:29:09   All that is to say that I'm using my iPad

01:29:11   slightly less than I used to be, but no, I do use it,

01:29:15   and I do, what Marco described is absolutely true

01:29:17   of my experience as well in that

01:29:19   when I want to do anything complicated,

01:29:22   I want to do it on the Mac, and complicated means

01:29:24   I'm going to be juggling multiple things.

01:29:26   I'm going to be like, a good example is a thing

01:29:29   that I often start on my iPad and switch to the Mac is

01:29:32   like if I'm looking for something,

01:29:34   to give an example, all the crap I bought

01:29:36   when I was preparing the way for my Mac Pro,

01:29:38   things like the little end table,

01:29:40   and cables, and cable guides, and the UPS,

01:29:45   and power plugs that go into them,

01:29:49   and adapters, and pigtail things,

01:29:51   and just all that stuff.

01:29:53   A lot of that I started out, I'd be like in bed on my iPad,

01:29:56   and I'd be like, "Let's look at this. Let's look and see."

01:29:58   I'd do a search for UPS things and watch some reviews,

01:30:01   and especially I'd watch some YouTube videos or reviews,

01:30:03   but at a certain point I'd say,

01:30:05   "Okay, now I'm in serious research mode.

01:30:07   I'm going to open 500 tabs about UPSs."

01:30:10   Or, "I'm going to open every furniture site

01:30:13   in the entire world, every store, Target, Walmart,

01:30:16   Wayfair, Ethan Allen, Crate & Barrel.

01:30:20   They're all going to be open in 100 windows and 100 tabs,

01:30:23   and I'm going to be taking screenshots,

01:30:25   and copying images into a notes document,

01:30:27   and saving URLs, and typing down measurements."

01:30:31   That is time for the Mac,

01:30:33   because I could do that on the iPad,

01:30:36   but I would drive myself insane with the text selection alone.

01:30:39   Right? So I will go to --

01:30:42   And I think that's not a condemnation of the iPad,

01:30:45   because the whole point of the Mac is it's supposed to give you

01:30:48   more space, more flexibility, more precise input,

01:30:50   than a thing that you can carry around with you.

01:30:52   Like, it's a desktop -- I'm using, anyway, desktop computer,

01:30:55   and my screens are huge.

01:30:57   Now, to Casey's point, that's not always true.

01:30:58   There are Macs that have smaller screens,

01:31:00   and they also have a lot of these benefits.

01:31:02   But anyway, I do feel that tension in, like,

01:31:07   at a certain point, like, "Uh, let me just go to the Mac."

01:31:10   And a lot of that is also because the Mac

01:31:12   is what I grew up with.

01:31:13   It's what I'm most comfortable with.

01:31:15   And when I was reading Gruber's thing,

01:31:17   where he was sort of slamming the iPad's crappy multitasking

01:31:20   and how undiscoverable and awkward it is,

01:31:22   I mostly agree with him, but there is an angle on that,

01:31:25   which is like, "Well, if you grew up using this,

01:31:27   it wouldn't be so weird to you." Right?

01:31:29   It's not quite the same as the -- God, I forgot the name of the thing.

01:31:33   What the hell is the app that had an intentionally obscure UI?

01:31:36   Snapchat.

01:31:37   To make the -- you know, it's not quite Snapchat,

01:31:39   but it's similar in that if you grew up --

01:31:41   you know, anything you grew up using,

01:31:43   A, seems, you know, seems natural to you

01:31:46   because it's all you know,

01:31:47   and B, there is a certain kind of cache to, like,

01:31:52   you're young, you have a lot of time to invest in learning a weird thing,

01:31:55   and once you know it, it's part of your,

01:31:58   you know, personality and self-image that, of course,

01:32:00   you know how to use this thing,

01:32:02   and anyone who doesn't is a square, using a term from the '50s.

01:32:06   [ Laughter ]

01:32:08   Reading Gruber's thing, you can look at it and say,

01:32:10   you know, "Well, Grandpa,

01:32:13   you may feel comfortable with the Mac,

01:32:14   but if you grew up using the iPad

01:32:17   or if you adopted it early enough in your life,

01:32:20   it is totally natural, and none of this stuff is weird,

01:32:22   and you learn it once, and it's no big deal,

01:32:23   and you just sound like an old fuddy-duddy." Right?

01:32:27   But I, you know, I think that angle is there,

01:32:30   and there is a tiny bit of truth to that,

01:32:32   but there is much more truth to the idea that, like, look,

01:32:35   there's ways you can measure how good a UI is,

01:32:38   and Gruber touches on it a lot in the article that we'll put in the show.

01:32:41   One of the ways is just across the entire customer base for this product,

01:32:46   all the people who own it.

01:32:48   How many people use this feature?

01:32:50   How many people know that it exists?

01:32:51   How many people enjoy it?

01:32:52   How many people find it indispensable and useful to use? Right?

01:32:56   How many -- and on the other side,

01:32:58   how many people, if you describe the problem

01:33:00   that this feature is supposed to solve, would say,

01:33:02   "Oh, I have that problem, too," and then when you say to them,

01:33:04   "Oh, well, then why don't you use feature X to solve that problem?"

01:33:07   They'll be like, "I didn't know how to use feature X.

01:33:09   I didn't know it existed,

01:33:10   and once you tell me it's too awkward to use,"

01:33:12   like, there are ways to measure the success of any feature,

01:33:14   and that way is not find the person who is the most expert

01:33:18   in using this product and see if they like it.

01:33:21   Like, you can actually look at the customer base, you know.

01:33:24   How many people -- and this is true of weird features in the Mac, too,

01:33:27   but you can say, like,

01:33:28   "How many people who own the Mac use the dock?"

01:33:30   I think it is widely used.

01:33:32   I think it's very widely used.

01:33:34   Whatever problem the dock is solving

01:33:36   is a problem that people have because they use the dock,

01:33:39   and I think most people use it, right?

01:33:43   Or how many people ever have more than one window open on their Mac?

01:33:47   I think that number is also high.

01:33:48   People discover how to open a second window

01:33:51   because it solves a problem that they have

01:33:52   that they want to see more than one thing at the same time,

01:33:54   and they're able to successfully do that with a Mac.

01:33:57   It is a thing that happens all the time.

01:33:59   On the iPad, very similar fundamental things,

01:34:02   like how many people have more than one thing open on the iPad at once?

01:34:05   If it's a fraction of the people who do it on the Mac,

01:34:08   how many people know how to do it on the iPad?

01:34:11   Also a fraction of the people on the Mac.

01:34:12   How many people know but still don't do it, right?

01:34:15   Those things, I think, are mostly, you know, uncontested.

01:34:20   It doesn't mean the iPad is necessarily bad,

01:34:24   but in a weird way it frames the iPad as most useful to deep experts.

01:34:29   Then there's a big gulf,

01:34:31   and then useful to people who only need simple functionality,

01:34:34   and it's that gulf that the iPad needs to span.

01:34:37   I was thinking about this.

01:34:38   I had what I thought was a coherent thought,

01:34:39   and I put a note to myself in the show notes,

01:34:41   and then I looked at my note,

01:34:42   and I realized either my thought was not as coherent as I thought it was,

01:34:46   or I forgot what I thought.

01:34:47   Oh, this happens to me all the time.

01:34:49   They're trying to write down a dream.

01:34:50   Yeah, exactly, but it was basically this.

01:34:52   Like, the group wrote this thing about how the multitasking on the iPad is weird and all this stuff,

01:34:55   and it's like the iPad, the things that are great about the iPad,

01:35:00   and we talk about this many times,

01:35:01   and the reason people are enthusiastic about it

01:35:04   and the reason I'm still enthusiastic about it is that it provides,

01:35:08   it gets rid of a lot of crap from the world of PCs

01:35:12   that you don't have to deal with anymore.

01:35:14   Some would call it a simplification,

01:35:16   but I think it's mostly just dropping stuff that only exists for weird historical reasons.

01:35:22   So, like smartphones, the iPad is more comfortable for people to use.

01:35:28   They're not worried that they're going to screw it up

01:35:30   or that they don't know how to get stuff done.

01:35:33   It's generally tractable and low maintenance.

01:35:37   That is the big thing that makes people keep saying,

01:35:39   "Why do people keep saying the iPad is the future of computering,

01:35:41   and why do people like phones instead of their PCs?"

01:35:44   And writ large, the answer is that they're less of a pain in the butt to deal with, right?

01:35:48   And that is the greatest feature of all these things.

01:35:52   And that's why it's so weird that where the iPad falls down,

01:35:56   like in terms of like, "Oh, let's try to also make it possible to do more powerful things,"

01:36:01   it falls down in like just stepping out of that and expanding its power

01:36:06   by including a multitasking system that's the opposite of that.

01:36:09   It's more weird and complicated than the PC.

01:36:12   It brings more obscure rules and strange physically difficult gestures.

01:36:17   It doesn't make as much sense, and it is more obscure.

01:36:21   It's not like they took--

01:36:24   The problem is not that they said, "Okay, well, we want to make the iPad more powerful,

01:36:28   so let's give it some of the power of the Mac but in a simplified form."

01:36:32   Maybe they thought that we were doing that,

01:36:34   but what they ended up doing was giving it the power of the Mac

01:36:38   in a more complicated form that's harder to use.

01:36:41   And it's so weird because you don't need--

01:36:46   I think what I was trying to get is like you don't--

01:36:49   This is not a problem that the iPad had.

01:36:52   The problem was that we need to come up with a whole new way

01:36:57   of working with multiple things that is simpler.

01:37:00   Or maybe that was a problem that it had, but they didn't do it.

01:37:02   They come up with a whole new way of working with multiple things

01:37:04   that's harder and more obscure and more complicated and more difficult to use

01:37:08   and, at the bottom line, less used.

01:37:11   It is less successful as a feature because fewer people actually use it.

01:37:15   So are you really helping people if they say,

01:37:17   "Oh, well, I don't do that because it's just too weird and too complicated."

01:37:20   And again, you can't look at the outliers and say,

01:37:22   "Look at how this maestro of the iPad is able to get tons of work done."

01:37:27   That's good that it's able to do that,

01:37:29   but part of the reason it's so impressive is--

01:37:32   What is it like to bear riding a unicycle?

01:37:34   It's like impressive that they can do it at all

01:37:36   because it's generally acknowledged to be difficult and complicated,

01:37:39   and that's not how it should be.

01:37:40   No one sees someone on the Mac with two windows open and goes,

01:37:43   "Wow, you must be a Mac expert."

01:37:45   But you see someone with three things in split screen on an iPad and a slide over,

01:37:51   and you assume they're an iPad expert,

01:37:53   and that is probably the biggest condemnation of the iPad's advancement.

01:37:58   And Gruber is pretty harsh on the article saying,

01:38:00   "Look, they need to just start over.

01:38:03   Clean slate--" It's not unprecedented.

01:38:06   It's been done plenty of times before.

01:38:08   "Rethink this.

01:38:09   Whatever paradigm you thought you were going for is not resonating and not working.

01:38:13   Come up with some sort of new consistent paradigm,

01:38:16   or if you can't do that, use the old paradigm of a bunch of windows

01:38:21   that you can move around,"

01:38:22   which I don't think anyone is asking for

01:38:24   because that's the nightmare of an iPad enthusiast.

01:38:27   But the current system is not succeeding.

01:38:30   It is failing in all the same ways that the whole rest of the iPad succeeds.

01:38:35   It's too complicated and too obscure.

01:38:37   So I'm not demanding that Apple wipe the slate clean,

01:38:42   but I also don't quite know how they can salvage what they have

01:38:46   without at least one or two fundamental adjustments

01:38:48   to how they think about doing more complicated things on the iPad

01:38:53   because it's got the CPU power.

01:38:55   A lot of them have the screen space now,

01:38:57   and the input methods are there as well.

01:38:58   You've got the pencil and the finger

01:39:00   and all these different sensors and everything.

01:39:02   It's the software that's falling down.

01:39:06   It's not bugs or whatever.

01:39:09   It is software design.

01:39:11   There's bugs too.

01:39:13   Well, I suppose.

01:39:14   Try using keyboard shortcuts in mail.

01:39:16   Yeah.

01:39:17   The iPad has evolved to this point through a series of small steps,

01:39:23   none of which really seem like they ever had the big think

01:39:26   of like -- some of it did the big -- you know, Xerox,

01:39:29   and Xerox and the Mac team did various big thinks of like,

01:39:33   what does it mean to have a grass-full user interface?

01:39:36   How does it work?

01:39:37   What are the building blocks?

01:39:38   And then riffed on those and built on that fairly solid foundation

01:39:43   of the WIMP windows -- what was it?

01:39:45   Windows icons menus and pointer.

01:39:47   That was a very solid -- it continues to be a very solid, flexible foundation.

01:39:51   If there had been no foundation but it had just been like,

01:39:54   well, we've got kind of a GUI, but we can probably --

01:39:56   we have some kind of graphical interface.

01:39:58   Let's add this feature.

01:39:59   Let's add that feature.

01:40:00   Let's let people have more than one thing.

01:40:02   Let's do this thing.

01:40:03   If you grow at piecemeal like that and never have sort of a solid,

01:40:08   extensible underlying philosophy,

01:40:10   you will end up in something that's a complicated mess.

01:40:13   So that's, I think, what they need to do with the iPad is.

01:40:16   Step back and rethink.

01:40:18   They did that for the iPhone.

01:40:20   Let's rethink.

01:40:21   Oh, the application takes over the whole screen.

01:40:23   It's edge-to-edge, blah, blah, blah.

01:40:24   They came up with a paradigm for phone applications

01:40:27   and for simple iPad applications that totally works

01:40:31   and people love and use.

01:40:32   It's the next step where they're falling down,

01:40:35   where what about when you want to do something more complicated

01:40:39   than you could do on a phone because we have this big, powerful iPad?

01:40:42   How do we deal with that?

01:40:43   I think the answer is not what they've currently done.

01:40:46   I think they need to regroup.

01:40:50   And I'm not saying they need to totally redesign iOS.

01:40:52   I think the next thing after we got through,

01:40:53   we're saying they're supposed to do refinement releases

01:40:55   and fix stability, but this is a design thing

01:40:57   that I totally give them a few years to work on.

01:41:00   I just hope their plan is not,

01:41:02   we'll continue to refine what we have

01:41:04   because I just don't think it's working.

01:41:06   - The biggest characterization of the iPad that I've felt

01:41:10   and I think you actually just basically agreed with

01:41:13   implicitly is like, if you think about,

01:41:16   and I should blog this, but I'm not going to

01:41:18   because I don't want the response.

01:41:21   So if you think about a graph,

01:41:24   a standard XY, two-dimensional graph,

01:41:27   on one axis you have the complexity of what you're trying to do

01:41:30   and on the other axis you have how easy it is to do.

01:41:34   I think for most standard computer, Mac PC kind of things,

01:41:40   that line is kind of a gentle slope.

01:41:43   The complexity of what you want to do

01:41:46   versus how easy it is to do it

01:41:48   kind of ramps smoothly for most things.

01:41:50   Yeah, there are some bumps along the way.

01:41:52   That being said, where that starts,

01:41:56   like the zero point of that graph, is high.

01:41:59   There's a certain baseline of knowledge

01:42:01   required to use a computer at all and do tasks at all

01:42:05   and there's a certain baseline level of complexity

01:42:07   you have to deal with even for easy things.

01:42:09   However, as you scale up, that line climbs,

01:42:15   but there's not a lot of huge spikes in it.

01:42:18   Whereas iPad usage, the ease of use of easy things

01:42:25   is way lower. That line starts way lower.

01:42:28   However, as you progress towards trying to do more complicated things,

01:42:32   it just shoots way up.

01:42:34   Easy stuff is easier, but hard stuff is way harder.

01:42:39   And what is even considered a hard thing on iOS

01:42:43   is oftentimes not a very hard or uncommon thing

01:42:48   in the PC/Mac world.

01:42:50   I don't think iPads, or phones for that matter,

01:42:54   but I think it's more of a problem than an iPad,

01:42:56   I don't think they have ever really found a good way around this.

01:43:00   The whole reason the iPad is easier to use for simple things

01:43:06   are fundamental aspects of its design.

01:43:09   Things like that all apps, at least start out full screen,

01:43:13   and that all apps only have at least normally one window,

01:43:16   and that all apps, you know, you tap the icon and that just shows,

01:43:20   and then you hit the home button if you still have one,

01:43:22   and then you see all your apps and you hit the next one,

01:43:26   and they take away the concept of files,

01:43:28   and that's all great when you're doing really simple things.

01:43:32   But as soon as you have to do something that's slightly outside of that

01:43:37   golden path of easiness, the barrier of complexity and difficulty goes way up.

01:43:44   As John was saying, you really have to be an iPad expert

01:43:48   to do a lot of things that on a computer are considered pretty trivial things.

01:43:54   Stuff like shuffling files in different places between apps,

01:43:57   and dealing with multiple windows.

01:44:00   I still can't, for the life of me, drag a tab to the right side of Safari

01:44:05   without accidentally opening a new window.

01:44:07   And then God knows how to close it. I have never figured this out.

01:44:11   I eventually have to start kicking stuff out of the force quit thing,

01:44:15   and eventually, and I never can recover that tab.

01:44:18   I just got to start a new one. That tab's gone.

01:44:20   It's not just expertise. That's the important point of when I was saying

01:44:24   you can actually measure this. It's not just, you know,

01:44:26   it's like, well, you have to be an expert to know all these secrets.

01:44:29   It's not knowing. It's like physical capabilities.

01:44:31   How hard is it physically to do this?

01:44:34   Part of the beauty of the Mac user, and again,

01:44:37   Gruber touched on this with double clicking, is you can say,

01:44:40   like, this is how a feature works. Can people do it?

01:44:44   Is it actually physically hard?

01:44:45   And I'm not talking about people who have disabilities.

01:44:48   Like, just all of us, everybody, like, everyone is disabled to some degree.

01:44:52   It's just a question of what that degree is.

01:44:53   There is no discontinuity. It's just a smooth spectrum

01:44:56   from supposedly able to supposedly disabled.

01:44:59   It's difficult to precisely control things as you drag.

01:45:05   Like, you have to hold pressure on the screen,

01:45:07   because if you pick your finger up on the iPad or any touch device,

01:45:10   it cancels the drag. And you know the edges are hot,

01:45:13   like a game of Operation, right?

01:45:15   If you accidentally go to the edge, something different's going to happen.

01:45:18   I had a big rant a couple, you know, many shows ago

01:45:21   about trying to rearrange icons on Springboard on the phone.

01:45:24   And, you know, to add to that, like, if you're going to pick up multiple items

01:45:28   and do all those multi-finger things,

01:45:30   it is actually physically difficult to do those,

01:45:33   even if you are a "expert" and know exactly what you're supposed to be doing.

01:45:37   And that's the sign of a feature that is measurably bad for most users,

01:45:42   even very dexterous, very careful ones.

01:45:46   It's frustrating, because you don't know where the live, hot areas are.

01:45:49   There's no visual indication, even if you have incredibly precise motor control

01:45:53   and you're in a situation where you can do that,

01:45:55   because the iPad is well supported and you're not on the go.

01:45:57   Like, that's not hard because it's an expert feature.

01:46:01   That's hard because it's a bad feature.

01:46:03   It is hard for human beings to use.

01:46:05   And fundamentally, because it is a thing that you carry

01:46:08   and because there's no pointing device and because, you know, whatever,

01:46:12   like, there are some -- that's the problem they have to solve.

01:46:15   But there are other parts of it that they've chosen to do themselves,

01:46:17   the fact that there is no Chrome in any of the windows.

01:46:19   Again, a simplification, but it also means there's no affordances,

01:46:23   visually and touch-wise, in the UI

01:46:25   for you to have a safe area to grab or move things to or anything like that.

01:46:30   So that's the challenge they face,

01:46:32   and I feel like their solutions to it are measurably not as successful

01:46:37   as they should be just because they're so hard that lots of people can't do it.

01:46:41   Like, Marco, you know what you're trying to do with that stupid tab,

01:46:44   but you just physically can't -- it's like a video game that's hard to play,

01:46:47   and that's not a good UI.

01:46:49   Yeah, it's like, I just have a very high error rate for this common task.

01:46:52   And I think a lot of that, by the way --

01:46:54   I know you're hearing from a few people in the chat

01:46:56   that apparently you can long press on the tabs icon,

01:46:59   and there's a "merge all windows" button to solve my tabs problem.

01:47:04   What? That's the undiscoverability,

01:47:06   because there's no Chrome or menu bar or anything like that.

01:47:08   I would never have found that.

01:47:10   I had no chance of ever knowing that.

01:47:12   So, yeah.

01:47:14   And I think this is part of the thing.

01:47:16   What makes it so simple to do the simple tasks

01:47:19   is a large degree of both physical and of design --

01:47:25   both hardware and software simplicity visually.

01:47:29   Everything -- yeah, everything happens full screen,

01:47:33   and you said there's no window Chrome and everything,

01:47:35   and then that's great.

01:47:37   It keeps simple things simple.

01:47:39   But then where do you put the advanced stuff?

01:47:41   You start doing things like overriding what edges mean,

01:47:45   and you start hiding things behind long press.

01:47:47   And one of the greatest challenges of UI design

01:47:50   for touch devices since the beginning

01:47:52   is how do you make things discoverable

01:47:54   that aren't immediately visually on the screen?

01:47:57   Like, if you're not going to just have an icon

01:47:59   or a button on screen that can do something,

01:48:01   how do you show people that they can do it?

01:48:04   That is a largely unsolved design problem among touch devices.

01:48:09   Like, we have a few bad solutions.

01:48:12   We have, like, the first launch overlay or arrow saying,

01:48:16   "Ooh, click here to see all these options."

01:48:18   Those are terrible.

01:48:19   We have the junk drawer design.

01:48:21   First it was the hamburger menu.

01:48:22   Then it was the action menu.

01:48:24   Now it's the dot, dot, dot menu.

01:48:25   They're all the same thing.

01:48:26   It's all just like "more."

01:48:28   Shove a whole bunch of stuff into the "more" button,

01:48:31   and people will push it to find "more."

01:48:34   You know, like, that's --

01:48:35   All of these solutions are, you know, mediocre at best.

01:48:38   They just don't have the kind of discoverability

01:48:41   that you have in a typical PC interface

01:48:44   where you typically have way more real estate on screen,

01:48:48   and you have, you know, smaller buttons

01:48:50   because you're controlling it

01:48:51   with a more precise pointing device,

01:48:52   so usually you have, like, you know,

01:48:54   you can fit more buttons or icons in a tool bar.

01:48:57   Then you have the whole menu system,

01:48:59   and you can pull down the series of menus

01:49:00   and see what things are,

01:49:02   or you can right-click on things to see what you can do.

01:49:05   - Right-click is a great example

01:49:06   because that is an example of a thing

01:49:07   with no visual affordance

01:49:08   that seems to be totally non-discoverable,

01:49:10   but because it is so unfamily consistent

01:49:14   in the PC/Mac interface,

01:49:16   it becomes culturally successful.

01:49:18   Like, there is no visual indication

01:49:20   that you can right-click on things,

01:49:21   but, again, look out into the world.

01:49:23   How many people who use a personal computer of any kind

01:49:26   know that right-clicking exists?

01:49:28   Now, it is less than 100%,

01:49:29   but the percentage is way higher than the people,

01:49:32   like, who right-clicked stuff on their PC.

01:49:34   It's way higher than the number of people

01:49:36   who use split-screen and multitasking --

01:49:38   sophisticated multitasking on the iPad

01:49:40   because it is one additional thing to learn,

01:49:43   which is, you know, one thing

01:49:45   that you wouldn't have to learn for simple things,

01:49:47   but it is just -- it's everywhere.

01:49:49   If you see something

01:49:50   and you don't see a visual way to interact with it,

01:49:53   one of the things that you can try --

01:49:55   and usually your first instinct is,

01:49:56   "Let me try right-clicking it,"

01:49:57   because people have been trained

01:49:59   through very consistent use of that,

01:50:00   that even though there is no indication

01:50:01   this is gonna be successful,

01:50:03   it's a good first guess.

01:50:04   "Can I right-click it?

01:50:05   Is there some menu to right --

01:50:06   Can I right-click on this?"

01:50:08   And then there's a menu that says "Delete."

01:50:09   "Can I right-click it to get info about it?"

01:50:11   And right-clicking is incredibly successful,

01:50:14   so successful that Apple,

01:50:15   which insisted on single-button mouse for a long time,

01:50:17   had to come around to basically enable right-click,

01:50:20   you know, by control-click in the beginning

01:50:23   and then eventually by making us a single-button mouse

01:50:25   that lets you do right-clicking.

01:50:26   There are solutions to these problems.

01:50:28   They're tricky.

01:50:29   And getting back to the double-clicking thing

01:50:31   that Gruber pointed out,

01:50:32   just to clarify what we were talking about,

01:50:33   if you don't bother to read the article,

01:50:35   double-clicking on the Mac is a tricky situation

01:50:38   where they added this thing,

01:50:40   which is like, well, single-click to select

01:50:41   and double-click to open, right?

01:50:43   And it's a very simple,

01:50:44   very consistent paradigm on the Mac,

01:50:47   but it has the problem that double-clicking

01:50:50   is a physically tricky thing to do.

01:50:52   Not that tricky.

01:50:53   You're like, "Ah, double-clicking.

01:50:54   It's not like I'm doing

01:50:55   a multi-finger, five-finger drag on the iPad, right?

01:50:58   Double-clicking. How hard is it to double-click, right?"

01:51:00   It is actually physically difficult

01:51:02   because what determines if something is double-click

01:51:04   is the space between the clicks,

01:51:05   and that's why, since the beginning,

01:51:07   the Mac has had a feature

01:51:08   where you can adjust the tolerance

01:51:10   for what the OS, you know, considers a double-click.

01:51:13   There's a setting, I believe, to this day

01:51:15   in System Preferences and on the original Macintosh

01:51:17   that says, "Should I give you lots of time

01:51:19   if you go click, click, click?

01:51:21   Is that a double-click,

01:51:22   or do you have to go click, click, click, right?"

01:51:25   Because it is difficult for people to do that.

01:51:28   In particular, I saw this early on with the Mac

01:51:30   when mice were new,

01:51:31   it's difficult for people to double-click

01:51:33   without accidentally moving the mouse

01:51:35   between the two clicks,

01:51:36   which makes it kind of not count as much as a double-click

01:51:38   depending on the heuristics of the operating system, right?

01:51:41   That is, you know, compared to multi-finger drag,

01:51:44   that's nothing, but that is an epidemic problem

01:51:46   that has never really gotten better.

01:51:48   Some people still have trouble with double-clicking.

01:51:50   The second thing is, unlike culturally with right-click,

01:51:52   where everybody knows,

01:51:53   "I can't figure out how to use this thing.

01:51:55   Let me try right-clicking it,"

01:51:56   which is generally a safe thing to do.

01:51:58   It's not going to do anything harmful,

01:52:00   like, culturally, it's understood or whatever.

01:52:02   Single-click versus double-click, despite being,

01:52:05   I think, a very clear, easily explained paradigm

01:52:08   from day one on graphical user interfaces,

01:52:11   still is a source of confusion for people.

01:52:14   You see people double-click links in web browsers.

01:52:17   You see people double-click literally everything

01:52:19   because it's basically like,

01:52:21   "I clicked it and it didn't work,

01:52:22   so maybe I should double-click it,"

01:52:23   and then people very often short-circuit that

01:52:25   and say, "If I see anything on the screen

01:52:27   and I really want it to work,

01:52:28   I should double-click it instead of single-click it."

01:52:30   And in general, that will work.

01:52:33   Like, if it requires a single-clicking,

01:52:35   you double-click, most of the time,

01:52:37   it will do the thing,

01:52:38   so maybe that's why people are trained to do it,

01:52:40   but even that one minor complexity,

01:52:43   through decades of use,

01:52:45   like, human nature is such that single-click versus double-click,

01:52:47   if you had to go back and say,

01:52:48   "What is the paradigm that you might want to rethink

01:52:51   about the WIMP, you know, aggressively user interface?"

01:52:54   Single-click versus double-click

01:52:55   turned out to be surprisingly tricky for people to master,

01:52:59   and if they haven't done it by now,

01:53:00   they probably never will,

01:53:01   so it's always going to be slightly less successful

01:53:03   than you'd want it to be,

01:53:05   versus right-click, which wasn't thought of back then,

01:53:07   even though the Xerox mouse had, like, five buttons on it

01:53:09   or three buttons or whatever and was even more complicated.

01:53:12   Right-click, I think, has been much more successful,

01:53:14   so I do have some hope that there are solutions

01:53:17   that don't require visual affordances

01:53:19   that can be culturally successful

01:53:21   and physically easy to do.

01:53:23   You know, you could argue, "Oh, that's long-press,"

01:53:25   or, "That's forced-press," or, like, you know,

01:53:27   "Apple has tried a whole bunch of things in that area,"

01:53:28   but none of them have been consistent

01:53:31   or popular enough to get by.

01:53:32   I think maybe long-press is getting there.

01:53:35   People do kind of understand

01:53:36   that you can long-press things in touch interfaces,

01:53:38   and you might get, you know, options or whatever,

01:53:41   but anyway, like... -Do they?

01:53:43   -I think more than people know how to do split-screen

01:53:45   on the iPad, that's for sure.

01:53:47   Like, I've seen people in my family do long-press on things,

01:53:50   but I've never seen anyone in my family

01:53:52   do split-screen on an iPad, so...

01:53:54   -I'm not even sure how I'm supposed to be doing split-screen.

01:53:56   Every time I try to do it, I occasionally will use it,

01:53:59   and every time I try to do it,

01:54:01   it takes me a very long time to figure out

01:54:03   the exact combination of where I press and hold,

01:54:06   where I drag it, where I let it go.

01:54:08   Do I, like, stick it down or stick it up or stick it?

01:54:11   Like, it takes me a very long time,

01:54:13   and I, like, officially, like, I officially know it.

01:54:16   Like, I ostensibly know how to do this,

01:54:18   and, yeah, there's a lot of waiting.

01:54:20   There's a lot of, like, "Oh, I accidentally didn't,

01:54:22   you know, stick this down in the right spot,

01:54:24   so it wiggles back to somewhere else,"

01:54:25   and, like, I'm good at computers.

01:54:28   I do this for a living.

01:54:31   I use an iPad every day,

01:54:33   and I have multitasked on every version of iOS

01:54:36   that has supported multitasking,

01:54:38   and I'm still bad at doing it on the iPad because,

01:54:42   I mean, I guess I don't do it enough,

01:54:45   but I think the bigger problem is it's just so fiddly,

01:54:49   and a lot of this is down to, you know,

01:54:52   factors like we're saying about, you know,

01:54:53   undiscoverability of touch gestures,

01:54:55   and a lot of it is down to because there are basically

01:54:59   no buttons in the hardware,

01:55:00   and because the software UI paradigm

01:55:04   is to make everything full screen

01:55:06   and to not have things like window, chrome, or menus,

01:55:10   that inherently makes it really hard to design these things

01:55:14   in a way that is intuitive and discoverable.

01:55:17   Like, it'd be one thing if Apple redesigned iPad OS

01:55:21   so that every app by default had a title bar

01:55:24   that had some buttons in the corner

01:55:26   that would do things like maximize it to the full screen

01:55:30   or shrink it to half screen.

01:55:32   - That would have to be big chrome.

01:55:34   Like, that's the reason they're resisting

01:55:35   is 'cause your pointing device is so imprecise

01:55:37   that it would have to be comically large

01:55:38   and now you're eating into your limited screen space.

01:55:40   Like, I understand why they don't, but--

01:55:42   - Yeah, but they just made all the screens bigger.

01:55:44   - Yeah, I know.

01:55:45   - If they really wanna make multitasking

01:55:47   on the iPad intuitive

01:55:50   and something that everyone is able to do reliably

01:55:53   and able to figure out, they have to put it on screen.

01:55:56   They have to have on-screen controls

01:55:59   for multitasking to be really easy.

01:56:01   And that just goes completely against

01:56:04   the software design paradigm of iPad OS

01:56:07   and that may make certain things worse.

01:56:09   That might make that floor of how simple things can be,

01:56:13   it might raise it very slightly.

01:56:15   - 'Cause you could accidentally hit the window controls

01:56:17   when you didn't mean to and then you're like,

01:56:18   "Oh, where'd my window go?"

01:56:19   - Right, but I would argue that's actually better overall

01:56:22   than where we are now, which is you have these

01:56:24   kind of secret gestures that come up

01:56:26   when you move near edges or do weird things

01:56:28   that it would be not at all surprising to me

01:56:31   if the vast majority of invocations

01:56:34   for iPad multitasking were accidental.

01:56:36   - Yeah, everything's hot on your screen.

01:56:39   - Yes, which raises that floor.

01:56:41   - Yeah, I use Slide Over a lot, surprisingly.

01:56:44   I more or less know how it works

01:56:45   and I like the thing where you can switch

01:56:46   between the Slide Over stuff.

01:56:48   I've worked that into my workflow and it mostly is okay,

01:56:51   but because the floor is lava,

01:56:54   because everything is hot in iPad OS,

01:56:56   when I'm watching, for example, Netflix,

01:56:59   and sometimes I'll be watching Netflix

01:57:00   and I'll bring over Twitter to just look at something

01:57:02   'cause I'm watching something cruddy or whatever,

01:57:04   like I'm multitasking on my iPad, look at me go.

01:57:06   Sometimes when I bring over Slide Over,

01:57:10   that gesture gets registered by the underlying

01:57:12   Netflix or Hulu player and the Slide Over comes on,

01:57:16   but the video I'm watching goes back 30 seconds

01:57:18   because that app has a gesture for a swipe backwards

01:57:21   and depending on the speed that you do it,

01:57:24   'cause you can, to Slide Over,

01:57:26   you have to start in a certain position and bring it over,

01:57:28   but if you go a little bit too fast,

01:57:30   the underlying app registers at a gesture.

01:57:31   Arguably, that's a bug in the app below it,

01:57:34   but you can't blame the app author.

01:57:35   They're like, look, if someone swipes their finger

01:57:37   from right to left, I should go back 30 seconds

01:57:40   and as far as they're concerned,

01:57:41   that's exactly what's happening,

01:57:42   but as far as the OS is concerned,

01:57:43   they're like, oh no, they were activating Slide Over,

01:57:45   but both of the things get it.

01:57:46   There is no safe area and I think the experts,

01:57:50   talking about doing split screen

01:57:52   and multitasking and dragging things,

01:57:54   the experts learn mentally the map of the safe,

01:57:58   hot regions and how long you have to hover

01:58:01   and how long you have to press and when it's safe to move

01:58:05   and what areas you should avoid.

01:58:07   That's part of the expertise is learning

01:58:10   this invisible heat map and this invisible

01:58:13   four dimensional timing map of,

01:58:15   it's not intellectually, but through repetition,

01:58:18   I know I've gotten better about putting things into Slide Over

01:58:22   and going into multitasking just by learning the routes

01:58:25   and everything and that can have a feeling

01:58:28   of mastering a video game and being good at it

01:58:30   and can give positive feelings towards the experience

01:58:34   because they're like, I invested time into this

01:58:36   and now I'm good at it and now I feel like a real maestro,

01:58:38   but again, that's not a characteristic

01:58:41   of a good user interface.

01:58:43   It's fun to master a game.

01:58:45   It's not fun to have to master something

01:58:48   that you're just trying to shop around for,

01:58:50   new UPS for your fancy computer, whatever it is you're doing.

01:58:54   You want more people to be successful.

01:58:57   You don't want the people who master it

01:58:59   to feel an incredible amount of satisfaction.

01:59:02   You want everyone to feel that satisfaction

01:59:04   because everybody is able to successfully

01:59:06   have more than one window on the screen or whatever.

01:59:09   So to that end, this is my favorite thing in the world,

01:59:12   and by favorite, I mean least favorite thing in the world.

01:59:14   So let's suppose that you have your iPad Pro or iPad or whatever

01:59:19   and you have your smart keyboard attached and open

01:59:21   and so on and so forth, and let's suppose you're looking

01:59:24   at Safari full screen, right,

01:59:26   and you want to get a second app in split view.

01:59:30   The most direct way to do this, at least on the iPad Pro

01:59:33   that I have, is you swipe up just a little bit

01:59:37   from the bottom, because if you swipe up too much,

01:59:39   you're going to go home. You don't want to go home.

01:59:41   You just want to swipe up a little bit.

01:59:42   You swipe up a little bit,

01:59:43   and you grab one of the things in your dock,

01:59:45   and then you just drag that icon over to the side,

01:59:47   and suddenly you have split view.

01:59:49   Yeah, split view, right?

01:59:51   I don't even know the damn terminology for this stuff.

01:59:53   Anyway, so that's all well and good,

01:59:55   but what if you don't have what you want in your dock?

01:59:59   So let's suppose that I don't have notes in my dock,

02:00:02   even though I actually do.

02:00:04   Let's suppose notes is on one of my home screens,

02:00:06   and I'm looking at Safari full screen,

02:00:08   and I want to get notes off to the side.

02:00:10   Well, I can't go to the dock because it's not there.

02:00:13   So when I have the keyboard attached,

02:00:15   it's actually slightly straightforward.

02:00:16   I just Command + Space and type in "notes,"

02:00:19   and then I get the notes icon,

02:00:20   and I drag it from the spotlight window over to the side,

02:00:25   and now all is well. So far, so good.

02:00:29   How do I do that if I can't hit Command + Space?

02:00:32   I have an answer to this question,

02:00:34   but can you guys tell me how you do this?

02:00:37   I read an answer earlier, so I'm spoiled, but it's ridiculous.

02:00:40   Right. So what I've figured out,

02:00:43   and maybe this isn't the best way,

02:00:45   but what I've figured out is you start by going home.

02:00:48   So I'm now looking at my home screen.

02:00:50   I find notes on the home screen.

02:00:52   I pick up the notes icon, and I drag it,

02:00:54   which of course is starting to make me rearrange the home screen.

02:00:57   Right, or delete notes.

02:00:59   Or delete notes.

02:01:00   Then I swipe up from the bottom with a different hand,

02:01:04   just a bit, not too much now, just a bit.

02:01:08   Now I've got multitasking up.

02:01:10   I'm still holding with the other hand.

02:01:12   I select Safari for multitasking.

02:01:14   Now the icon becomes a window,

02:01:16   and I drag the window off to the side.

02:01:19   Once I've done this 10 times, it vaguely --

02:01:24   I wouldn't even say it makes sense,

02:01:26   but it's a repeatable series of actions.

02:01:28   Let's put it that way.

02:01:29   But how preposterous is this?

02:01:32   How does any human being figure this out

02:01:35   without being instructed directly?

02:01:37   How do you figure this out?

02:01:38   Even if you're instructed.

02:01:39   Can you imagine teaching someone to do that?

02:01:41   Oh, don't go too far.

02:01:42   Oh, wait, no, hold that, but don't hold it.

02:01:43   There are so many hot regions.

02:01:45   If you hold it over this for too long, it does this thing,

02:01:47   but you want to not let it go.

02:01:48   Remember, don't take your finger off the glass,

02:01:50   but pull up just a little bit, but go to the left.

02:01:52   Now use the second finger to do this.

02:01:54   Right, exactly.

02:01:55   It's like trying to teach someone to dance,

02:01:57   but what most people will say is,

02:02:00   "It's already too complicated.

02:02:02   I'll just do it the other way."

02:02:03   They give up on trying to have two applications there

02:02:07   at the same time.

02:02:08   They're like, "I know how to switch between them

02:02:09   the old-fashioned way.

02:02:10   I'll just do that even though it's less convenient."

02:02:12   It's not worth it to go over that hurdle,

02:02:15   and that's not a successful feature.

02:02:17   If what you're trying to do is enable people

02:02:19   to use two apps at once, and they're bailing on it

02:02:22   because it's too weird, rethink.

02:02:25   My other favorite part of this is,

02:02:27   let's assume I'm trying to do this all over again.

02:02:29   I was in Safari.

02:02:30   I go home.

02:02:31   I'm looking at my home screen.

02:02:32   I pick up the Notes icon.

02:02:33   I'm moving it around, and then I see,

02:02:35   "Oh, Safari is also on the screen,

02:02:37   perhaps on the dock, perhaps just on the home screen.

02:02:39   I want to bring that back."

02:02:40   So what do I do?

02:02:41   I tap the Safari icon.

02:02:43   Oh, shit, I'm now collecting two icons.

02:02:45   That's not what I wanted at all.

02:02:47   It's like the floor is lava.

02:02:50   But it's so easy to drop that second icon, Casey.

02:02:52   You know how to do that, don't you?

02:02:53   Yeah, right.

02:02:54   Yeah, right, uh-huh, exactly.

02:02:55   Where's the safe place to drop it?

02:02:58   Yeah, it beats me.

02:02:59   It's just so frustrating because it could be so good.

02:03:03   It could be so, so good.

02:03:05   And when I do get that moment, and it happens,

02:03:08   and it doesn't happen infrequently,

02:03:10   when I get that moment, a great example of this is,

02:03:13   I've taken a screenshot of something,

02:03:14   and I want to mark it up.

02:03:15   Well, on the Mac, that's kind of a pain in the hindquarters.

02:03:19   I have to figure out some way of doing this on the Mac.

02:03:21   On the iPad, I grab the pointing device connected to the iPad.

02:03:27   That's amazing.

02:03:28   That is so much better than anything I would have to do on the Mac.

02:03:31   And it works so much more nicely.

02:03:33   But so much of using the iPad is an exercise in frustration.

02:03:37   And I don't really want to open this can of worms, but here I go.

02:03:44   If I need to write a series of shortcuts

02:03:48   in order to do the things that I can do so much easier on the Mac,

02:03:52   then I'm just going to do them on the Mac.

02:03:54   And yes, I understand that there are human beings that prefer iOS.

02:03:56   I get that.

02:03:57   I really, truly do.

02:03:59   But for me, even if I don't have a particular preference

02:04:02   between Mac OS and iOS,

02:04:04   I can tell you that because I'm fluent in both,

02:04:07   it is so much faster for me and easier for me to do stuff on the Mac

02:04:13   that all too often that's what I'm choosing.

02:04:16   And it's so frustrating because the physical hardware of the iPad Pro,

02:04:20   especially with Face ID, it's beautiful, it's light, it's powerful, it's fast.

02:04:26   Cellular.

02:04:27   It's cellular.

02:04:28   It's so much what I want.

02:04:30   It's so much what I want that the software just falls on its face

02:04:34   and it makes me so sad.

02:04:35   And I would even say there are lots of things that the software is better at too.

02:04:39   It isn't that iPad OS is just bad and Mac OS is good.

02:04:42   There's a lot of differences in the OSs and especially in the apps on those OSs.

02:04:47   There's no clear way.

02:04:48   I can't just say everything on Mac OS is better or everything on iOS is better.

02:04:52   But it just seems like there is so much about iPad multitasking

02:04:58   and iPad UI use that relies on modes or timing or hidden gestures.

02:05:07   And ask anybody who has ever tried to design computer interfaces,

02:05:13   modes, timing tricks, and hidden gestures are recipes for total disasters of usability.

02:05:21   And what makes the iPad great, and especially what made it great initially,

02:05:26   is none of that.

02:05:28   What makes it great is the opposite.

02:05:30   It's the simplicity.

02:05:31   It's that when you're on that golden simple path of doing everything full screen

02:05:35   and just doing simple things where you never have to deal with files or multitasking or anything.

02:05:40   When you're just doing that, you don't have any of those problems.

02:05:43   And it's glorious.

02:05:45   But as you try to do more stuff on your iPad, you slam into those walls pretty quickly.

02:05:51   And to get over those walls is so cumbersome.

02:05:54   And it's all because you have this device that Apple really wants to keep simple

02:05:59   without losing what's good about that low barrier to entry for it.

02:06:03   And I think the only solution to this is to put goddamn controls on the screen.

02:06:08   Like put them in the UI.

02:06:10   You have buttons that you can push to do things.

02:06:13   That's how you solve this.

02:06:15   And it's not always going to look pretty.

02:06:17   And it's not always going to be conceptually design purity maximum.

02:06:22   But design purity maximum is how you get pretty things, not usable things.

02:06:27   And it is possible to have both.

02:06:29   If you have a nicely designed UI, it has its own form of beauty.

02:06:33   Nice, easy to use UI is itself beautiful.

02:06:37   It's a different version of beauty than clean minimalism, Johnny White room, junk drawer design.

02:06:43   Look what they do now.

02:06:45   It's a different style.

02:06:47   But it's not worse. It's not ugly. It's just different.

02:06:50   And I don't see how you add the complexity that power users want, or that all users want, honestly.

02:06:58   I don't know how you add that complexity and power to iOS without sacrificing some of that visual purity.

02:07:06   Because again, having a title bar with windows on the screen, even if you can't freeform shape those windows,

02:07:15   even if you don't drag the corner to resize it, there's different choices you can make there.

02:07:21   But if there was just a basic amount of window chrome, think of the functionality you get out of window chrome on the Mac.

02:07:27   Not only do you have sizing controls, you also have hiding and showing that window,

02:07:33   and you have closing that window for good, like getting rid of it, minimizing it, big, small.

02:07:38   Then you also have the title bar, which tells you what app you're in or what thing you are viewing.

02:07:44   Then you have the proxy icon, which is a shortcut to tons of power.

02:07:49   And then of course, if it's visually combined, you have the whole toolbar, which itself is a massive amount of potential power.

02:07:55   Buttons, and you have navigation back forward, you might have a search box.

02:08:00   There's so much that is in standard window chrome.

02:08:03   And that's not even including if you want to drag the corners to resize it, because that may or may not be a good idea in iPad OS, who knows.

02:08:10   But if you just had a small amount of window chrome on every window for apps that supported this,

02:08:15   that would expose a ton of basic multitasking functionality in a really easy way that wouldn't require all these weird gestures that are too easy to do accidentally and too hard to do intentionally.

02:08:28   There's a reason why people do this. It's like when I made the magazine and I tried to make no setting screen,

02:08:33   and I had such incredible complexity because I didn't want a setting screen, and that the result ended up being worse and harder to use,

02:08:41   and then I had to eventually just put a damn setting screen in it because that was actually better.

02:08:44   I feel like, think about all the hoops Apple has jumped through and made its users jump through to avoid having basic window controls on iPad OS.

02:08:54   Is it really worth it?

02:08:56   They're trying to preserve the simplicity. That seems to be their job one, which you can kind of understand because the simplicity is the big selling point,

02:09:01   but they haven't found, the solutions we're always going to, I think many, many shows ago when we discussed this,

02:09:06   we were just talking about having menu bars on iPad apps. Do you remember that discussion?

02:09:10   We keep going to solutions that we know work in the Mac world because it's like at a certain point,

02:09:16   and we've had this discussion over many years, at a certain point it's like, look, you didn't come up with something better.

02:09:20   We're not saying you've got to do everything the same way the Mac did it. Come up with a new way, show us.

02:09:25   You tried a bunch of ways and they all suck. So we get back to this point where it's like, well, how about just window chrome?

02:09:31   Because you haven't come up with something better. I'm not saying window chrome is a solution.

02:09:35   And in fact, when you're discussing window chrome, it made me think about some of the regressions they've had on the Mac.

02:09:40   As you know, if you're staring at a Mac right now, there isn't window chrome to speak of.

02:09:45   And I have seen many people have difficulty resizing Mac windows because to do it,

02:09:52   there is a very small region of pixels near the edge of a window where you will get the appropriate arrow cursor.

02:09:58   And there's different arrow cursors because Mac windows are resizable from all sides, which power users love.

02:10:03   But regular users might not understand that if you reach for the corner and you end up grabbing a side and you get the little arrow,

02:10:09   you can only resize in one direction now. And that's frustrating.

02:10:12   And in general, just getting to the point where you get the appropriate resize cursor because there's no visual affordance for it.

02:10:18   And because the windows are basically borderless, which looks very clean, there is a very small region where you have to do that.

02:10:24   Compared to Mac OS, you know, the old Mac OS with the platinum interface that had a big textured square with little grippy things,

02:10:31   which wasn't a huge target, but it was absolutely visual and it was big enough where you could get it.

02:10:37   And, you know, when windows weren't resizable from the edges, you didn't have that problem.

02:10:40   So even in systems that are tried and true, it's possible to, through visual refinement, end up making interfaces harder to use.

02:10:49   Even if people aren't confused about how to resize windows, I think resizing Mac windows is now harder than it used to be for that reason.

02:10:56   And the purity argument of like, oh, the interface isn't as pure, I think the, I keep calling it WIMP, I hate that acronym, but anyway, it's my root, you know.

02:11:05   The Mac user interface, I know it's not just the Mac, but anyway, has incredible purity.

02:11:10   Menu bar at the top of the screen, windows with scroll bars that are resizable with closed widgets and, you know, with window widgets.

02:11:18   And even the dock, like, there is incredible purity to that paradigm and incredible flexibility.

02:11:24   It has stood the test of time, you were able to do amazing things with it.

02:11:28   It is more pure than the iPad OS interface, which has many different paradigms all fighting with each other and various things tacked onto the end, and none of them are sort of composable and flexible enough to get the job done on their own.

02:11:42   All so they can preserve the simple interface, which, you know, I'm all for preserving the simplicity, it's just that you need to find a way to enable more sophisticated actions without killing that simplicity, but also by making a consistent system, and they haven't done it yet.

02:11:59   And so we say, well, have you tried, you know, have you tried just sticking a menu bar?

02:12:03   Think of how much flexibility putting just a plain old menu bar at the top of the screen on an iPad would do.

02:12:08   You'd have to make it like a finger capable menu bar or whatever, but we, you know, and we said, oh, that's old thinking, you just want it to be the way the Mac is.

02:12:15   I don't actually, I want something better, but if you can't think of something better, I'll take the compromise of, we couldn't come up with any better ideas, so here is a finger accessible menu bar, right?

02:12:24   We couldn't come up with any ideas, so here is some kind of window chrome in iPad OS, which necessarily has to be larger and, you know, and hurts the complexity, hurts the simplicity or whatever.

02:12:35   But you do, that's the challenge, that's the challenge Apple faces, come up with something that preserves the best of both, but provides the power and is consistent, and they have not risen to that challenge.

02:12:45   And honestly, I don't think we have risen to that challenge either, because everything we're suggesting, I think, would be mostly an abomination and terrible.

02:12:51   But on the high end, it would make the complicated things that we're describing easier and better to do.

02:12:59   It's just that on the low end, they would make people, you know, confused and frustrated, and I think in general, the public would think an iPad that works that way is worse than an iPad that works the way it does now, because one of the, you know, I was saying how we never see people using multitasking on the iPad.

02:13:13   Also, for all the iPad users that are intimately familiar in my life who aren't tech nerds, they're mostly satisfied with their iPad experience of not using those expert features.

02:13:23   They don't care that the iPad has a very inconsistent, inflexible, difficult to use interface for doing more complex things.

02:13:31   And if you try to show them, they mostly bail, but their satisfaction with the product remains.

02:13:38   So I'm not saying that the current compromise is successful, because clearly it's not, but it does have the beneficial characteristic of most of our users would still be satisfied, even as we screw up our ability to do this high end.

02:13:52   So it's kind of like the super high end experts are satisfied because it's a video game they can master.

02:13:56   Most people are fine because they never have to do that, but then there is like, you're leaving money on the table and that like these devices, we know these devices could enable people to do more complicated things.

02:14:06   And it's just that they're not good at bridging that gap.

02:14:10   Like it's one of the things that the Mac was always good at.

02:14:13   Hey, you're not a computer expert, but if you sit down in front of this computer and learn some basics, you can be successful at doing things you never thought you could do.

02:14:21   And the iPad does that for the low end.

02:14:24   Hey, you can browse the web, you can send people email, you can chat with people, you will be successful.

02:14:29   If you buy this product, it's not too complicated, you'll figure it out.

02:14:32   But that stops. And it's like, you want to do anything more than that?

02:14:35   You're going to have serious problems.

02:14:37   And unless you want to be a super duper expert, just retreat to the simple things.

02:14:40   I would also, you know, just one final closing thought on like the purity angle here. The history of the iPad is filled with, you know, starting out doing very literal, having things very restricted.

02:14:52   And then somebody or some third party starts proposing like, hey, you know, what if we, you know, add a keyboard?

02:14:59   Although, by the way, Apple shipped a keyboard on day one, but nobody bought it.

02:15:02   But like, hey, what if we, what if we had a keyboard?

02:15:04   What about, what if we make these capacitive stylus things?

02:15:08   And, you know, yeah, Steve dropped me a ton of styluses a long time ago, but it turns out they're kind of nice.

02:15:13   And then eventually, you know, someone does, you know, Microsoft does it or third parties do it.

02:15:17   And then eventually, Apple adds their own option.

02:15:20   And before Apple does it, everyone's like, no, you should never have a stylus. That would ruin everything.

02:15:26   You should never have, you don't need a keyboard. That would make it a laptop. Just buy a laptop. Blah, blah, blah.

02:15:30   And then Apple does it. They release a new thing. They release their own stylus and it's good.

02:15:35   They release their own keyboard and it's good. They, you know, they had multitasking systems, didn't have multitasking before, and that, you know, is okay.

02:15:42   But like, eventually, the market keeps wanting this device to get more capable.

02:15:49   Well, again, although, Jon, you're right, not all of its market and possibly not even most of its market.

02:15:54   But a portion of this market, especially, you know, its high end for sure, wants these devices to get more capable.

02:16:00   And in most cases, the way that has happened is to make them more like laptops.

02:16:07   And everyone b******s and moans and complains beforehand.

02:16:12   And then somebody allows them, usually Apple, to make it more like a laptop and everyone's like, oh, oh yeah, that actually is nice.

02:16:19   Like, and that's, for me, that's what changed. Like, what changed the iPad for me from a cool toy I kept buying and never using to something I use every day is the keyboard.

02:16:29   Like, Apple made a really good keyboard for it and it turns out the difficulty of text input was holding me back quite a lot and discouraging me from using it very often.

02:16:38   And once I had better text input via a physical keyboard, I now use it way more.

02:16:44   You know, for a lot of people, the pencil did that, you know, where like drawing things on it or, you know, precise input in other styles was difficult or impossible or limited.

02:16:56   And the Apple pencil fixed that and opened up whole new worlds to that kind of use.

02:17:02   And, you know, then everyone's like, oh, it should never have a way to browse files.

02:17:07   Well, then Apple adds the files app, which mostly works.

02:17:09   Ah, I don't know about that, but I don't want to derail your point.

02:17:12   And then Apple adds the files app, which occasionally works and kind of does what people want.

02:17:15   And, you know, and the conceptual purists, when they saw that, were like, why do you need that? It's an iPad.

02:17:21   You know, just keep data in apps.

02:17:22   But no, they added it and it actually solved a lot of problems and opened up a lot of doors for people.

02:17:26   Not well, but, you know, it started knocking down some of those walls.

02:17:29   And so there's this whole history throughout the entire iPad of people wanting to defend its purity for, you know, ideological sake or for fear that it'll ruin it, somehow make it more complicated.

02:17:43   But then eventually Apple does make it more complicated because that's what certain use cases need.

02:17:48   And then it's fine. And people, you know, with the exception of a lot of this multitasking stuff that's hard to use because it's weirdly designed and the gestures are all weird.

02:17:56   Like, most of this stuff doesn't have a negative impact on people who want to keep it simple at all.

02:18:01   Most of it is simply additive.

02:18:03   So if making it a little bit more like laptops here and there has made it so much better for power users and so much more capable of a platform,

02:18:13   we shouldn't be afraid of things that are in that direction that sound scary that we haven't tried yet that might have a big impact.

02:18:20   Things like, as I said, basic window chrome.

02:18:22   That sounds really weird right now because we don't have it and we haven't seen it.

02:18:27   But in three years, all iPad apps might have window chrome by default.

02:18:31   They might have a title bar with simple multitasking buttons on the left or whatever.

02:18:35   And that might just be fine.

02:18:37   And we might look back on the time before that, the same way we look back now on the time before the keyboard and the pencil.

02:18:43   Or copy and paste.

02:18:45   Yeah, or copy and paste. Yeah.

02:18:47   Making the iPad a little more laptop-like when that is the best solution, which is not never.

02:18:53   Like, it isn't always, but it's not never.

02:18:56   When the best solution is to borrow something from the laptop world, the iPad has done it a lot of the time.

02:19:03   And there's lots of ways it hasn't.

02:19:07   And I'm putting off for now a lot of discussion about the apps themselves and what they can do and what they can't do.

02:19:14   And various, like, you know, sandboxing issues and file dealing with issues and stuff like the app store issues.

02:19:19   There's a lot of other issues that we could talk about.

02:19:21   But just talking about, like, the basics of usability of the device, getting around apps, you know, getting in and out of apps, multitasking, physical interface.

02:19:29   Borrowing from the laptop world so far has succeeded very well on iPads in the few times they've done it.

02:19:38   And so doing it one or two more times for something as rudimentary as, like, the way the multitasking system works, which they've already heavily borrowed from.

02:19:47   They already have a dock and an alt tab switcher or command tab switcher.

02:19:51   They already have a lot of the same things.

02:19:54   You know, the search box, like, there's a lot of overlap already and it's worked mostly okay.

02:19:59   To borrow a couple more things to make the multitasking system way more easy for anybody who's not Federico to use, I think would go a long way.

02:20:07   And we shouldn't rule it out simply because, oh, why don't you just use a Mac?

02:20:10   No, it turns out there's good ideas and we've already borrowed a lot of them.

02:20:14   It wouldn't harm the world to borrow a couple more.

02:20:17   So what we're saying, as usual, is that Microsoft Surface got it right and their only problem is they started from a crappy desktop operating system and it's only a matter of time before Apple figures it out and ends up, like, I saw a good graphic recently.

02:20:29   It was, like, show the Microsoft Surface line and someone was like, this is what I expected the iPad line to look like by now.

02:20:34   It had the giant Surface Studio 27-inch touchscreen thing and it had the small ones with, they looked like laptops with removable screens.

02:20:41   And Microsoft has the unified OS strategy of they have Windows with Window Chrome on their tablet, sort of, kind of, like, you know, obviously, you know, starting with Windows was a serious detriment and I'm not saying Apple should have just put Mac OS on their tablets because that would have been the wrong thing too.

02:20:57   But what you're describing, Marco, is essentially where Microsoft is now, but with the details different, right?

02:21:04   So, you know, having something that works and looks and borrows a lot from a laptop but is also a touch device and works in that mode as well, just better, right?

02:21:15   That's what we're looking at.

02:21:16   The details matter.

02:21:17   Yeah, the details definitely do matter.

02:21:19   There's a reason we're not using Microsoft Surface, but, you know, the overall strategy, if you didn't look at the details and just described it in broad strokes, that's more or less what you're talking about.

02:21:26   And I still contend that, like, ideally you come up with a solution that is better, but if you try for a long time and can't think of anything, it's time to just consider the menu bar.

02:21:36   Like, I would consider the menu bar before Window Chrome, actually, because I just feel like it's a simpler element and you could put one of them up there and they already have a status bar.

02:21:42   It's almost like a menu bar anyway and so on and so forth.

02:21:44   But that's the type of stuff you can't just add piecemeal.

02:21:48   You have to have a good rethink, right?

02:21:50   Just, you know, step back and say, "Next-gen iPad OS. Describe the paradigm."

02:21:57   And then you can choose which things from the Mac world are we going to pull in.

02:22:01   And, you know, if someone says, "Oh, we don't want to pull that in," it's like, "Well, then what's your better idea?"

02:22:04   Right?

02:22:05   Because we tried these things and they're worse.

02:22:07   If you don't have a better idea, it's time to consider does the menu bar fit in, right?

02:22:11   Does a menu bar fit in?

02:22:13   Does Window Chrome fit in in some fashion?

02:22:15   And, you know, so I hope that happens eventually.

02:22:18   If it doesn't, I don't think actually it's the end of the world because I still think that people are generally –

02:22:24   most people are satisfied with their iPads using them in simple mode, right?

02:22:28   But it also means that the iPad is never going to threaten the Mac for complicated tasks if they can't figure out how to make it, you know, easy to do –

02:22:41   you know, make it comfortable and fun and easy for even experts to do on the iPad.

02:22:47   It's not.

02:22:48   It's too difficult.

02:22:49   It's too complicated.

02:22:50   And only if you really, really love the iPad will you go through that effort and everyone else will just run through the thing they're more comfortable with.

02:22:58   And we are continuing to make generations of people who are comfortable to some degree doing complicated tasks on desktop computers.

02:23:07   So it's not like those people are dying out either.

02:23:09   New generations of people know how to use Macs and PCs because they have to use them to do all the things that they generally can't do on an iPad.

02:23:18   So that divide will remain and iPad customers will be happy and Mac customers will be happy.

02:23:25   But the future where iPad eventually expands to absorb the problem space of the Mac will never happen if they don't figure this out.

02:23:33   Thanks to our sponsors this week.

02:23:34   Linode, Bluevine, and Yes, Please.

02:23:37   And we will see you next week.

02:23:40   Now the show is over.

02:23:44   They didn't even mean to begin.

02:23:46   Cause it was accidental.

02:23:48   Oh it was accidental.

02:23:51   John didn't do any research.

02:23:54   Marco and Casey wouldn't let him.

02:23:57   Cause it was accidental.

02:23:59   Oh it was accidental.

02:24:02   And you can find the show notes at ATP.FM.

02:24:07   And if you're into Twitter, you can follow them.

02:24:12   @C-A-S-E-Y-L-I-S-S

02:24:17   So that's Casey List M-A-R-C-O-A-R-M

02:24:21   N-T Marco Arman S-I-R-A-C

02:24:26   U-S-A, Syracuse

02:24:29   It's accidental.

02:24:31   They didn't mean to.

02:24:34   It's accidental.

02:24:37   Tech podcast so long.

02:24:41   I didn't get the time to shove it in because we were going long anyway.

02:24:46   But I have the opposite of Marco's bug.

02:24:48   I had a beginner bug that is not the result of, but is adjacent to the fact that there is tons of documentation for AppKit.

02:24:59   AppKit, if you go look at the documentation for some common class in AppKit, there are so many attributes, there are so many methods.

02:25:08   One failure mode is, oh you didn't see this method or this attribute?

02:25:14   Cause if you had read it, it's obviously what you want, but it was on a page with 500 methods and you didn't happen to see it.

02:25:20   So this bug was a total dead beginner's bug.

02:25:24   But it baffled me for a while.

02:25:26   I was just trying to make a window appear and then let you close it.

02:25:32   So I had a menu command and you pick the menu command and a window appears.

02:25:35   And then when you're done with that window, you close it.

02:25:37   And then my bug was, alright so I did it.

02:25:41   You select the menu command, I would make a new window, I would say window please show yourself.

02:25:46   It would show itself and then you would close the window and then I would invoke that same menu command right after that.

02:25:50   And it would hit the same code and say I'm about to show a window.

02:25:53   And I tried it both the naive and the quote unquote fancy way.

02:25:57   The fancy way was I don't need to make the window every time, I'll just keep that window around.

02:26:00   The next time they take the menu command I'll just show it cause they already made it.

02:26:03   Simple, straight forward.

02:26:05   The problem was that every time I did the command the second time, the app would crash.

02:26:10   Nice.

02:26:12   And I was in, I'm using Swift here, so I was in my great Perl motto, when it's segfaults it's not your fault.

02:26:20   Because that's the promise of Perl, it's a high level language.

02:26:23   And if it's segfaults, that's not on you.

02:26:25   You should, that's, you know, Perl's broken, that's a, you know, it's a compiler problem.

02:26:29   It's not your fault, you shouldn't even be able to do that, right?

02:26:32   John Syracuse says, Perl's broken.

02:26:34   Yeah, well if it's segfaults, that's, you know, that's not your fault.

02:26:37   Unless you're writing C code to interace with Perl what you can do.

02:26:40   But in general if you're writing Perl code.

02:26:42   So in Swift, I know that's not actually true, but I was having that attitude.

02:26:45   And up to this point, I'd never had a crashing bug, because I'm doing simple things and Swift is, you know, handles the allocation and everything for me.

02:26:53   I'm sure I was making memory leaks here, my apps are so dinky, it doesn't matter anyway.

02:26:56   But yeah, it was, I'm like, what, I'm just making, so I said alright, stop trying to be fancy.

02:27:03   And just every time they do the command, make the window and ask it to show it.

02:27:06   And for some reason that wasn't working either.

02:27:09   And I couldn't for the life of me figure out, like, why does it work the first time, but then the second time it doesn't.

02:27:16   And like, AppKit, the thing about AppKit is there's a long legacy of the way it used to work back in the next days, in the early Mac days and so on and so forth, and there's a million different attributes.

02:27:25   And so I, you know, I've been heavily leaning on one of the slacks I'm in that's filled with, you know, a vast wealth of experience of Mac development and next development going back ages.

02:27:36   And I'm like, you know, I come in there only with my tail between my legs after I've spent an hour doing this, and I'm like, look, I'm just trying to make a window appear and then close it and be able to do it a second time and my app is crashing.

02:27:46   What super obvious thing am I missing? And someone with experience said, oh, what you're missing is that unbeknownst to you, every window, every NS window has an attribute called is released after close or something.

02:28:01   And it's set to true by default, so if you show the window and then close it, the memory for that window is released behind your back in the system, and that is the correct default for this particular mode working with nibs back in the day, right?

02:28:15   But it lives on as an attribute in every single window, and why didn't you know about that? Because if you go to the NS window documentation, which is incredibly extensive, there's a million things there, and if you're not looking for an attribute called is released after close, you'd be like, why is my app crashing?

02:28:28   So it would show the window, I would close it, it would release that memory, and the place where it used to be is now just initialized garbage memory, and the next time you go and try to do something with that, it crashes.

02:28:38   Cool. Isn't this a failure of Arc? Like, shouldn't Arc be holding onto that?

02:28:43   Set is released after close. Well, the is released thing just ditches the memory for it, and that just calling, causing the bug.

02:28:53   Well, no, dalloc ditches the memory for it, release just loses a ref count.

02:28:57   I don't know what it actually does. I don't know what the attribute is called. I don't know if I got the name right. It's release after close or something similar, but for all we know, it is dalloc-ing. I don't know what it's doing behind the scenes. All I know is it's an attribute among the many million attributes.

02:29:09   I mean, the lesson is don't try to hide and show your own windows. Use NS window controller instead, because it will handle all these things for you, and if you use nibs, this also would be handled for you.

02:29:17   Like, this is all the machinery that's behind the scenes that you don't know about, but it is a brain dead simple thing that I'm assuming. I didn't even look up the docs, but I'm assuming totally is documented like everything else, but if you don't know where to look or you're not familiar with the weird history of how AppKit works and how it works with nibs and how you're supposed to use it, and you try to be like, "I know how to make my own window. I'll construct it and ask it to show itself, and it will work great."

02:29:39   It does the first time, but then your window is gone.

02:29:42   on.

02:29:44   (beeping)