ATP

00:00:00 ◼   ► I am back.

00:00:02 ◼   ► - Where were you?

00:00:03 ◼   ► - I was using HomePod minis.

00:00:06 ◼   ► - Okay.

00:00:08 ◼   ► - And I am back.

00:00:09 ◼   ► - On big boys?

00:00:10 ◼   ► - As long as they last.

00:00:12 ◼   ► - Oh my gosh.

00:00:13 ◼   ► - With big HomePods.

00:00:14 ◼   ► - It was like the cheese graters.

00:00:15 ◼   ► - Oh my goodness, how did this happen?

00:00:17 ◼   ► - So last time I went to Westchester,

00:00:20 ◼   ► there they were two perfectly good full-size HomePods

00:00:24 ◼   ► being used in a kitchen by nobody.

00:00:26 ◼   ► - Poor house has just been gutted.

00:00:28 ◼   ► So, you know, I'm like, you know,

00:00:31 ◼   ► these are just sitting here, rotting away

00:00:33 ◼   ► with whatever electrical flaw is what eventually kills them,

00:00:36 ◼   ► that capacitor or whatever everybody says is bad

00:00:39 ◼   ► or wrong or whatever.

00:00:40 ◼   ► No one's using them.

00:00:41 ◼   ► And I'm sure they're gonna get replaced at some point

00:00:43 ◼   ► in the next decade by Apple, but until that happens,

00:00:46 ◼   ► let me take them back.

00:00:48 ◼   ► So I brought them and I have been using them

00:00:50 ◼   ► and they are exactly as glorious and annoying

00:00:55 ◼   ► as they always were and I'm so, I'm so happy

00:00:58 ◼   ► I made the switch because while they are playing music,

00:01:02 ◼   ► which you can eventually make them do most of the time,

00:01:05 ◼   ► eventually, they do sound incredible.

00:01:10 ◼   ► (laughs)

00:01:11 ◼   ► - Can you tell me something,

00:01:12 ◼   ► and I need you to really and truly be honest,

00:01:14 ◼   ► are they white or are they black?

00:01:15 ◼   ► - They're white, they happen to be white.

00:01:17 ◼   ► - So here's the thing, I'm so glad you answered

00:01:19 ◼   ► that question, I feel like Jack Ryan in "Hunford October"

00:01:22 ◼   ► about a crazy Ivan, we should do that as a movie thing,

00:01:24 ◼   ► by the way. Anyway, I feel like your white home pods are my white BMW. Hear me out. When

00:01:31 ◼   ► my white BMW was working, it was amazing. It was only working for about a week every

00:01:38 ◼   ► month, but for that one week, oh, it was good. It was real good for that one week. Your white

00:01:44 ◼   ► home pods, which by the way just happened to you, those when they're working, I guess

00:01:48 ◼   ► they're real good, aren't they?

00:01:50 ◼   ► No, and by the way, I chose the white because I know this is unpopular for a nerd like me

00:01:57 ◼   ► to say that the black option is not the best option on something, but white home pods look

00:02:01 ◼   ► better than black home pods.

00:02:02 ◼   ► They're not really white, they're gray, I mean, come on.

00:02:05 ◼   ► They're pretty close to, I mean, you know, the outside is made of something that kind

00:02:10 ◼   ► of resembles cloth, so you can't get it like super super white.

00:02:13 ◼   ► They're not white.

00:02:14 ◼   ► They're not even starlight, they're light gray.

00:02:15 ◼   ► Well, and the black is definitely dark gray.

00:02:18 ◼   ► - Yeah, the black is very dark gray.

00:02:20 ◼   ► - But yeah, anyway, so the white looks better.

00:02:22 ◼   ► And so I chose them and I stand by that.

00:02:24 ◼   ► I'm unashamed in choosing the white HomePods.

00:02:28 ◼   ► I also actually, so my phone is temporarily white again,

00:02:31 ◼   ► but I'll tell you what, some quick follow up

00:02:33 ◼   ► on last week's case discussion.

00:02:36 ◼   ► I briefly said last week I had tried the Peak Design

00:02:39 ◼   ► and the Pataka, and the Pataka, I liked the way it felt,

00:02:42 ◼   ► but I hated the way it looked.

00:02:44 ◼   ► The Peak Design was a little bit bulky

00:02:47 ◼   ► and I didn't like how little relative tackiness

00:02:50 ◼   ► that the back kind of cloth-like surface had.

00:02:53 ◼   ► Well, I have been using the Peak Design case all week

00:02:58 ◼   ► because I just like it.

00:03:00 ◼   ► It makes relatively little sense.

00:03:04 ◼   ► I can't really justify it over the other cases

00:03:07 ◼   ► in any kind of subjective, or objective, I guess,

00:03:11 ◼   ► objective means. (laughs)

00:03:13 ◼   ► It is more expensive than most of them.

00:03:15 ◼   ► It is thicker and it's less grippy.

00:03:18 ◼   ► However, it feels the nicest and it looks the nicest.

00:03:23 ◼   ► Therefore, I've been sticking with it and I kinda like it.

00:03:26 ◼   ► I kinda like the weird little square on the back.

00:03:28 ◼   ► It's kinda like a reverse popsocket.

00:03:30 ◼   ► Instead of having something that sticks out,

00:03:32 ◼   ► it's like a hole that you can put your finger

00:03:34 ◼   ► on the inside of the hole to lift it out of your pocket

00:03:36 ◼   ► or whatever, it's nice.

00:03:38 ◼   ► I even, I just now, or just yesterday,

00:03:40 ◼   ► I just ordered a couple of their mounts that go on the back.

00:03:43 ◼   ► Just see what I can do with that.

00:03:43 ◼   ► - Oh, here we go.

00:03:45 ◼   ► - You're gonna put it on your motorcycle?

00:03:46 ◼   ► - Yeah, yeah.

00:03:48 ◼   ► First I have to buy a motorcycle just for this case.

00:03:50 ◼   ► I like the case so much, I bought a motorcycle

00:03:51 ◼   ► and then I bought the mount,

00:03:53 ◼   ► and the motorcycle happens to be white,

00:03:54 ◼   ► and I bought the mount that loops around that way.

00:03:57 ◼   ► And I'm gonna start becoming a,

00:03:59 ◼   ► what are those people who make GoPro action videos

00:04:03 ◼   ► on their vehicles?

00:04:04 ◼   ► What are those, do we have a name for those?

00:04:06 ◼   ► Like cloggers or something?

00:04:08 ◼   ► - Oh my gosh, yes, let's go with cloggers.

00:04:10 ◼   ► That's perfect.

00:04:11 ◼   ► - So I'm gonna become a clogger.

00:04:14 ◼   ► And it's all thanks to this case.

00:04:15 ◼   ► - I tried that Marco, not as easy as you think.

00:04:18 ◼   ► (laughing)

00:04:19 ◼   ► - Yeah, here we go.

00:04:21 ◼   ► So anyway, turns out that's the case

00:04:23 ◼   ► I'm going with for a while.

00:04:24 ◼   ► - Yeah, update on my creaky clear case.

00:04:26 ◼   ► I've got a, this is the first clear case

00:04:28 ◼   ► I've used for any pre-show amount of time.

00:04:30 ◼   ► I've got quite a collection of crumbs and dust collecting.

00:04:34 ◼   ► Visible all around, if you just circle all around

00:04:38 ◼   ► like the top edge of the thing, it is so gross.

00:04:42 ◼   ► I cannot wait to get this thing off.

00:04:43 ◼   ► One of my cases shipped, I got a shipping notification.

00:04:45 ◼   ► I think possibly it'll be here by next week's show,

00:04:49 ◼   ► but I'll keep you updated.

00:04:49 ◼   ► I cannot wait to get this thing off my phone.

00:04:52 ◼   ► - You know, an alternative is to just not care

00:04:55 ◼   ► about having a covered bottom.

00:04:56 ◼   ► And then there's zillions of cases you can try.

00:04:58 ◼   ► - There's plenty of options too.

00:05:00 ◼   ► I mean, I picked two of them because they were expensive.

00:05:02 ◼   ► There's more than two options I could have picked.

00:05:04 ◼   ► I figure I'll like one of these two,

00:05:05 ◼   ► but I'll let you know.

00:05:06 ◼   ► - You can try the Peak Design.

00:05:08 ◼   ► It comes in grayish black and a weirdly light green color,

00:05:12 ◼   ► And that's it.

00:05:13 ◼   ► - Yeah, I mean, obviously if I needed a secure mount

00:05:16 ◼   ► or something, I would probably pick that,

00:05:17 ◼   ► but the little square that you mentioned,

00:05:19 ◼   ► like, oh, I can use it to pull my phone out of my pocket

00:05:21 ◼   ► or whatever, I just, I don't want that to be there.

00:05:23 ◼   ► I feel like my fingers would find it

00:05:25 ◼   ► and it would just be annoying.

00:05:26 ◼   ► - You're right, that is a thing.

00:05:28 ◼   ► And the little square, the very first day I used this case,

00:05:31 ◼   ► I thought it felt a little sharp.

00:05:32 ◼   ► Now it doesn't feel sharp to me anymore.

00:05:33 ◼   ► Maybe it just sanded down my fingers, I don't know.

00:05:35 ◼   ► But what I like about the Peak Design,

00:05:37 ◼   ► in addition to the fact that it looks nicest

00:05:39 ◼   ► feels nicest, the buttons also feel nicest, the mute switch toggles a little deep, like

00:05:45 ◼   ► the whole, the little recession to get to the mute switch is a little bit deep, but

00:05:48 ◼   ► the side buttons, you know, all the buttons are covered, the side buttons feel great in

00:05:53 ◼   ► this case, better than most cases, and overall, this case is the only, it's the only iPhone

00:05:58 ◼   ► case I have ever used from anybody except Apple where it felt like they went for the

00:06:03 ◼   ► the nice materials instead of the cheap materials.

00:06:06 ◼   ► And it is by far the nicest case I've ever seen,

00:06:10 ◼   ► felt, or used that was not made of leather.

00:06:13 ◼   ► So for whatever that's worth,

00:06:15 ◼   ► if you either wanna go leather-free

00:06:16 ◼   ► or if you just want something nice, it's really nice.

00:06:20 ◼   ► - All right, let's do some follow-up.

00:06:24 ◼   ► What is going on with shared photo libraries

00:06:27 ◼   ► and what is shared with them, Jon?

00:06:28 ◼   ► - Last week I said some things

00:06:31 ◼   ► about what wasn't shared in photos

00:06:32 ◼   ► and I was wrong about a few of them.

00:06:35 ◼   ► I said that keywords, favorites, and location stuff

00:06:37 ◼   ► are not shared.

00:06:38 ◼   ► They are shared.

00:06:39 ◼   ► So those are attributes of the individual photos.

00:06:42 ◼   ► So keywords, and I tested this with my little thing.

00:06:44 ◼   ► If you put a keyword on a photo,

00:06:45 ◼   ► it gets shared with the other person.

00:06:46 ◼   ► If you favorite it, it shows up.

00:06:48 ◼   ► Then we're talking about photos

00:06:49 ◼   ► that are in the shared library.

00:06:51 ◼   ► You can do that stuff to them.

00:06:52 ◼   ► You can assign keywords, you can set their favorite status,

00:06:54 ◼   ► you can add location information, and everybody sees that.

00:06:57 ◼   ► All the other stuff I talked about,

00:07:00 ◼   ► the non-photo items remain unshared.

00:07:02 ◼   ► So you don't get albums, you don't get smart albums,

00:07:04 ◼   ► you don't get slideshows, you don't get book projects,

00:07:06 ◼   ► all that stuff.

00:07:07 ◼   ► And also someone asked about this

00:07:09 ◼   ► when I was having a discussion about it on Twitter.

00:07:12 ◼   ► What about duplicates?

00:07:13 ◼   ► What if two people have the same photo,

00:07:17 ◼   ► like maybe like a person airdropped the same identical photo

00:07:20 ◼   ► to two different people at two different times,

00:07:22 ◼   ► and they both add it to the shared library, what happens?

00:07:26 ◼   ► The answer, and according to my experimentation

00:07:28 ◼   ► using the Mac version of photos only

00:07:30 ◼   ► and two different accounts that have the exact same file.

00:07:33 ◼   ► I imported the file into both of their private libraries

00:07:36 ◼   ► and they're separate at that point.

00:07:37 ◼   ► And then I had both of them add that photo

00:07:40 ◼   ► to the shared library and you end up with two copies

00:07:43 ◼   ► of that photo in the shared library.

00:07:44 ◼   ► Just two completely identical copies of the photo

00:07:46 ◼   ► in the shared library.

00:07:47 ◼   ► I didn't edit them, so it's not like one of them was edited

00:07:49 ◼   ► and one of them wasn't.

00:07:50 ◼   ► I just imported into the individual things.

00:07:51 ◼   ► Then I had one person add it to the shared library

00:07:53 ◼   ► and then had the second person add it to the shared library.

00:07:56 ◼   ► Still in beta, maybe they'll change that.

00:07:58 ◼   ► And interestingly, I was like, okay, well,

00:07:59 ◼   ► maybe they do that for safety or like just to make sure

00:08:02 ◼   ► photos don't squish other photos with edits or whatever.

00:08:04 ◼   ► But isn't there a dedupe feature

00:08:06 ◼   ► that they've been talking about in iOS 16 and stuff?

00:08:09 ◼   ► They will find your duplicates for you,

00:08:11 ◼   ► like built into the iOS 16 photos thing?

00:08:13 ◼   ► - I believe so.

00:08:14 ◼   ► - If that exists in Mac photos, I could not find it.

00:08:17 ◼   ► So it's kind of weird if that ends up being

00:08:19 ◼   ► an iOS only feature.

00:08:21 ◼   ► There are obviously there are tons and tons of ways

00:08:23 ◼   ► to dedupe photos in your photo library

00:08:25 ◼   ► with Mac applications.

00:08:26 ◼   ► Lots of third party applications

00:08:27 ◼   ► been doing that for years, but it's kind of weird that if they're bringing that as a first

00:08:30 ◼   ► party feature to the Photos app on the phone that they don't also bring it to the Mac.

00:08:34 ◼   ► So I hope that appears in a future beta or something.

00:08:39 ◼   ► We are sponsored this week by Collide.

00:08:41 ◼   ► IT admins often feel like they have to choose between their commitment to cybersecurity

00:08:45 ◼   ► and their duty to protect their employees' privacy.

00:08:47 ◼   ► Naturally, of course, you need to safeguard company data against hacks and breaches, but

00:08:51 ◼   ► you also don't want to turn your workplace into 1984.

00:08:55 ◼   ► Personal MDMs give the IT team complete access and control over company devices to a fault

00:09:00 ◼   ► because since employees are inevitably going to use their work laptops for some personal

00:09:04 ◼   ► activities, those tools can saddle you with surveillance capabilities over personal stuff

00:09:08 ◼   ► you never wanted access to like photos or browser history.

00:09:12 ◼   ► Before you know it, your end users are complaining about all the security agents slowing down

00:09:16 ◼   ► their laptops, developers are frustrated by the lack of autonomy, and people start secretly

00:09:20 ◼   ► working on their personal devices just to get things done.

00:09:24 ◼   ► In that case, everyone loses, so it's easy to fall into these traps of top-down security,

00:09:29 ◼   ► but this is not the only option.

00:09:32 ◼   ► Collide is an endpoint security solution built around honest security.

00:09:36 ◼   ► Their philosophy is that employees aren't your biggest security risk, they're your biggest

00:09:40 ◼   ► allies.

00:09:41 ◼   ► Your relationship with them should be based on transparency and informed consent.

00:09:45 ◼   ► Collide works by notifying your employees of security issues via Slack, educating them

00:09:50 ◼   ► on why they are important and giving them step-by-step instructions on how to resolve

00:09:54 ◼   ► them themselves.

00:09:55 ◼   ► For IT and security teams, Collide provides the right level of visibility for Mac, Windows

00:09:59 ◼   ► and Linux devices and addresses high-risk issues that can't be solved through brute

00:10:04 ◼   ► force or automation.

00:10:06 ◼   ► And your end users can see exactly why and how every piece of data is being collected

00:10:10 ◼   ► via Collide's User Privacy Center and their open source code base.

00:10:15 ◼   ► You can meet your security goals without compromising your values.

00:10:18 ◼   ► Visit collide.com/ATP to find out how.

00:10:22 ◼   ► If you follow that link, they'll hook you up

00:10:23 ◼   ► with a goodie bag just for activating a free trial.

00:10:25 ◼   ► That's collide, K-O-L-I-D-E, collide.com/ATP.

00:10:30 ◼   ► Thank you so much to Collide for sponsoring our show.

00:10:34 ◼   ► - All right, the European Union has mandated

00:10:42 ◼   ► that all phones in somewhere between one and 17,000 years

00:10:47 ◼   ► will have USB-C on them, as was foretold.

00:10:52 ◼   ► - Yeah, well, do we know?

00:10:54 ◼   ► I mean, the Europeologists here will tell us,

00:10:57 ◼   ► like, well, this actually, they didn't really mandate it yet.

00:11:00 ◼   ► Now it has to go to this committee,

00:11:02 ◼   ► then it has to go to this board, then it has to--

00:11:03 ◼   ► - I bet there's something in the notes

00:11:05 ◼   ► that has the answers to those questions.

00:11:06 ◼   ► - Let me read from the notes.

00:11:08 ◼   ► (laughing)

00:11:10 ◼   ► I didn't know if we were going to make

00:11:11 ◼   ► any initial opening remarks, I guess not.

00:11:13 ◼   ► So here we go.

00:11:14 ◼   ► - That's why you shouldn't have editorialized

00:11:15 ◼   ► based on the title item.

00:11:18 ◼   ► Because notice what the title item says.

00:11:19 ◼   ► Would you like to read the title item as written?

00:11:22 ◼   ► - EU USB-C mandate passes vote.

00:11:24 ◼   ► - Passes vote.

00:11:25 ◼   ► It does not, so it's mandate as a noun, right?

00:11:28 ◼   ► The USB-C mandate, what happened to it?

00:11:30 ◼   ► Did it go into effect?

00:11:32 ◼   ► Is it real?

00:11:32 ◼   ► Is it whatever?

00:11:33 ◼   ► No, but it passed the vote.

00:11:34 ◼   ► What the hell does passes the vote mean?

00:11:35 ◼   ► And here we get the text.

00:11:37 ◼   ► - I mean, look, hey, hands up, who wants us to pass?

00:11:39 ◼   ► Three of us, one, two, good.

00:11:40 ◼   ► Okay, it just passed a vote.

00:11:41 ◼   ► Like, that doesn't mean anything.

00:11:43 ◼   ► - Oh, it means something.

00:11:44 ◼   ► Anyway, go on.

00:11:45 ◼   ► On October 4th, the European Parliament voted overwhelmingly in favor of new legislation

00:11:49 ◼   ► that would eventually require all mobile phones sold in the EU to use a USB-C port for wired

00:11:55 ◼   ► charging.

00:11:56 ◼   ► The EU's new rules are yet to be formally approved.

00:11:59 ◼   ► Although they've been given the thumbs up by the bloc's parliament, the common charger

00:12:02 ◼   ► legislation still needs to be signed off by the Council of the EU and published in the

00:12:07 ◼   ► EU official journal.

00:12:09 ◼   ► It would then enter into force 20 days later.

00:12:11 ◼   ► So clear.

00:12:12 ◼   ► But even once that happens, just you wait.

00:12:16 ◼   ► But even once that happens, companies like Apple will still effectively have a two-year

00:12:20 ◼   ► grace period that's designed to ease the transition to a USB-C future.

00:12:23 ◼   ► This means the rules are likely to come into force by the end of 2024.

00:12:28 ◼   ► Devices already on the market won't need to be withdrawn, so if Apple launches a Lightning

00:12:32 ◼   ► Port iPhone ahead of the deadline, it can keep selling the phone.

00:12:36 ◼   ► Or what if they just kill the ports entirely?

00:12:38 ◼   ► They already did one, why not take the other?

00:12:40 ◼   ► - Yeah, so 2024 is a ways out if things go according

00:12:44 ◼   ► to what they seem like they're gonna go to.

00:12:46 ◼   ► That's plenty of time for Apple to transition to USB-C,

00:12:49 ◼   ► quote unquote, on its own.

00:12:51 ◼   ► You know what I mean?

00:12:52 ◼   ► - I love that it says companies like Apple too.

00:12:54 ◼   ► Like this is all about Apple.

00:12:55 ◼   ► Like what other companies are left?

00:12:57 ◼   ► - There's no company, as Tim Cook will tell you,

00:12:59 ◼   ► there are no other companies like Apple.

00:13:01 ◼   ► Only Apple would be stubborn enough

00:13:02 ◼   ► to keep lightning for this long.

00:13:04 ◼   ► - Nice.

00:13:05 ◼   ► I don't know how I feel about this.

00:13:07 ◼   ► So I'm gonna steal from a former guest,

00:13:10 ◼   ► Christina Warren who tweeted about this and I agree with her. I would like USBC on my

00:13:15 ◼   ► phone for several reasons which we can investigate if we care, but I would like USBC on my phone.

00:13:20 ◼   ► I don't love that it will arrive there if at all because of a government mandate rather

00:13:26 ◼   ► than either market forces or I don't know. I feel very like, "Oh, free market this. No

00:13:32 ◼   ► government." And that's not how I feel about most things, but in this particular case,

00:13:36 ◼   ► I don't love how we're ending up on a position

00:13:39 ◼   ► that I think I'm going to love.

00:13:40 ◼   ► - I mean, it's gonna end up there not

00:13:42 ◼   ► because of this mandate,

00:13:43 ◼   ► because if the 2024 date ends up being

00:13:45 ◼   ► anything close to correct,

00:13:46 ◼   ► Apple essentially has painted itself into a corner

00:13:50 ◼   ► to transition regardless of this law,

00:13:52 ◼   ► because they keep making phones

00:13:53 ◼   ► that can create these massive files,

00:13:56 ◼   ► like shooting the high resolution video

00:13:59 ◼   ► with high frame rate,

00:14:01 ◼   ► and trying to get them off a phone is a nightmare.

00:14:03 ◼   ► And so they either have to upgrade Lightning

00:14:05 ◼   ► to be much, much faster, which would be a big hassle,

00:14:08 ◼   ► I'm not even sure if it's possible,

00:14:09 ◼   ► or they have to go to USB-C.

00:14:10 ◼   ► So there is a technical motivator

00:14:12 ◼   ► that's going to make Apple make some kind of change,

00:14:15 ◼   ► whether it's to USB-C or something else,

00:14:17 ◼   ► just because it's plain ridiculous

00:14:19 ◼   ► how long it would take to get video off of a phone

00:14:22 ◼   ► that you can record at the highest quality.

00:14:25 ◼   ► It just takes too long, it's too slow.

00:14:27 ◼   ► So that's my guess about why Apple will be transitioning.

00:14:32 ◼   ► I'm sure this mandate doesn't hurt the schedule, right?

00:14:36 ◼   ► It can only help it,

00:14:37 ◼   ► but given how long these things take effect

00:14:40 ◼   ► and given how it's been, not weakened,

00:14:41 ◼   ► but like they've been allowing people to ease the transition,

00:14:44 ◼   ► give them a long time to do it,

00:14:46 ◼   ► not having to withdraw existing products,

00:14:47 ◼   ► which really helps them.

00:14:48 ◼   ► Like Apple could release it,

00:14:49 ◼   ► depending on the timing within 2024,

00:14:52 ◼   ► they could release their last lightning phone

00:14:54 ◼   ► to be the 2024 phone, right?

00:14:57 ◼   ► And not have to worry about it until a 2025 phone, right?

00:15:00 ◼   ► That's how much time they have to do this,

00:15:01 ◼   ► but I think they'll probably change before then.

00:15:04 ◼   ► David Schwab had some other ideas

00:15:06 ◼   ► about things Apple could do to skirt this.

00:15:07 ◼   ► I think these are less likely than the straightforward thing

00:15:10 ◼   ► which is they just go to USB-C, but here they are.

00:15:12 ◼   ► He says, "The EU law contains an exception for devices

00:15:14 ◼   ► "that only use wireless charging.

00:15:16 ◼   ► "Assuming Apple really doesn't want to switch the out front

00:15:17 ◼   ► "from Lightning to USB-C, do you think they might

00:15:19 ◼   ► "just replace the Lightning cable with MagSafe charger

00:15:21 ◼   ► "and the EU implement one of these as a legal workaround?"

00:15:24 ◼   ► These are probably, not increasingly silly,

00:15:27 ◼   ► but some of them are silly.

00:15:28 ◼   ► "Put a service only sticker or plug over the Lightning port?"

00:15:31 ◼   ► I don't see that happening.

00:15:33 ◼   ► It's kind of like the little diagnostic part on the watch.

00:15:35 ◼   ► Disable charging through the lighting port and software.

00:15:39 ◼   ► Right, you could still have the lighting port and not USB-C

00:15:41 ◼   ► as long as it doesn't charge through it.

00:15:43 ◼   ► Yeah, only for EU phones, right?

00:15:45 ◼   ► Ship an EU phone without a lighting port

00:15:46 ◼   ► just like the US phone doesn't have a SIM tray.

00:15:49 ◼   ► I guess they put a plastic spacer in there.

00:15:52 ◼   ► And he says, I think EU regulators may have underestimated

00:15:55 ◼   ► how much Apple hates being forced to modify their designs

00:15:57 ◼   ► to satisfy regional laws.

00:15:58 ◼   ► But I really don't think that's what's

00:15:59 ◼   ► going to force Apple to do it.

00:16:01 ◼   ► the march of progress of storage size and the size of video files and for that matter photo files.

00:16:08 ◼   ► Say you're shooting everything with 48 megapixel RAWs, each of those photos is 80 megabytes and

00:16:13 ◼   ► you got a one terabyte iPhone, try transferring that at lightning speeds, at the USB 2.0 speeds

00:16:18 ◼   ► or whatever it is. It's a little bit silly. Their "pro" phones need to be able to transfer

00:16:22 ◼   ► data faster and at this point trying to make a new version of lightning that is faster seems very

00:16:28 ◼   ► silly in light of how much USB-C is spread throughout the rest of Apple's lines. You

00:16:33 ◼   ► know, they did it on the iPad. They're either going to do it on the phone or they're going

00:16:37 ◼   ► to get rid of ports altogether, but I don't see them. Like, it's not the EU that's forcing

00:16:42 ◼   ► them to put out a non-lighting phone in 2025. It's just sanity.

00:16:47 ◼   ► I want them to do it not because a government is forcing them to, but because it's the right

00:16:53 ◼   ► thing to do. And I think they, you know, the rumors have been fairly consistent that starting

00:17:00 ◼   ► with next year's iPhones that we are apparently going to USB-C and that's been consistent

00:17:05 ◼   ► now for a number of years. So I kind of put some weight behind it. And I think, you know,

00:17:11 ◼   ► Apple must have decided a couple years ago, like, you know, to make this change and changing

00:17:16 ◼   ► over the iPhone in any kind of major component change is not a small deal. They have to worry

00:17:22 ◼   ► about like, first of all, can we even get or create

00:17:25 ◼   ► enough USB-C connectors to keep up with the iPhone's volume?

00:17:30 ◼   ► Like that's actually a real concern that, you know,

00:17:33 ◼   ► something that's as high volume and as,

00:17:38 ◼   ► I guess, high stakes as the iPhone is.

00:17:41 ◼   ► 'Cause every single thing that an iPhone has or has to do

00:17:45 ◼   ► has to be nearly 100% perfect, nearly 100% of the time,

00:17:49 ◼   ► because they just sell so many of them,

00:17:52 ◼   ► and it's so important to the company,

00:17:53 ◼   ► that if they have a part that has like a .001% failure rate,

00:17:58 ◼   ► that's too high, they can't have that,

00:18:00 ◼   ► 'cause that could cause a scandal

00:18:01 ◼   ► that could have a big problem for the iPhone that year.

00:18:04 ◼   ► So they have to be so careful,

00:18:06 ◼   ► and they have to make sure they can create the volume,

00:18:09 ◼   ► and have the high yields, and have the good reliability

00:18:11 ◼   ► of all these different parts.

00:18:12 ◼   ► You know, when you look at the iPhone as a product,

00:18:17 ◼   ► It's really amazing when you compare it to,

00:18:20 ◼   ► not only anything else that Apple makes,

00:18:22 ◼   ► but anything else that anybody makes.

00:18:24 ◼   ► It's really amazing how rarely anything goes wrong with them.

00:18:29 ◼   ► Like how, like manufacturing defect wise,

00:18:31 ◼   ► like how often have you opened up a new iPhone

00:18:34 ◼   ► and something's been broken and you've had to exchange it.

00:18:36 ◼   ► It's almost unheard of, like it happens so rarely

00:18:39 ◼   ► compared to the number of them that they sell.

00:18:42 ◼   ► And so, again, they have to be super cautious.

00:18:44 ◼   ► So I'm sure there were reasons like that

00:18:46 ◼   ► that led them to take this long to get here.

00:18:50 ◼   ► But I do think it looks like things are lining up

00:18:53 ◼   ► that they will be getting here.

00:18:54 ◼   ► And I think part of the reason, as John said,

00:18:57 ◼   ► the transfer rates of having these giant video captures,

00:19:00 ◼   ► they're literally, they're marketing the Pro phones

00:19:02 ◼   ► and they're making these software features

00:19:04 ◼   ► and hardware features to optimize for things like ProRes

00:19:08 ◼   ► and raw photos and everything that generate

00:19:10 ◼   ► these huge files that are comically slow

00:19:12 ◼   ► to get off the phone.

00:19:14 ◼   ► That is one side of this.

00:19:16 ◼   ► But the reality is, and Apple knows this,

00:19:18 ◼   ► most people with most iPhones will never do those things.

00:19:21 ◼   ► And so that doesn't necessarily need to be the reason.

00:19:25 ◼   ► That's a reason.

00:19:27 ◼   ► But I think the reason, the much bigger reason,

00:19:30 ◼   ► is just that it's a pain in the butt

00:19:32 ◼   ► to have two different phone chargers out there.

00:19:35 ◼   ► And when Apple went with Lightning,

00:19:38 ◼   ► the Android world was not as unified as it is now.

00:19:41 ◼   ► Now, everything is USB-C,

00:19:44 ◼   ► and it has been for a number of years now.

00:19:46 ◼   ► And it's spreading to all sorts of other devices

00:19:49 ◼   ► that aren't even phones.

00:19:50 ◼   ► The laptops now all charge via USB-C,

00:19:52 ◼   ► they can at least, they don't have to but they can.

00:19:54 ◼   ► And then you look at every, all hardware in the world,

00:19:58 ◼   ► flashlights charge via USB-C.

00:20:01 ◼   ► I saw, I got an ad on Instagram for a power drill

00:20:04 ◼   ► that charges via USB-C. (laughs)

00:20:07 ◼   ► Which by the way, well targeted ad.

00:20:09 ◼   ► Everything is USB-C now.

00:20:13 ◼   ► And the very few things that aren't are every iPhone

00:20:18 ◼   ► and like our AirPods cases or whatever.

00:20:20 ◼   ► It's like there's not much else left.

00:20:22 ◼   ► Oh, and the stupid Apple Watch.

00:20:23 ◼   ► - Don't forget my keyboard.

00:20:24 ◼   ► - Yeah, keyboards, yeah, my trackpad.

00:20:26 ◼   ► But like-- - My stupid keyboard.

00:20:28 ◼   ► - The Apple battery case that isn't a case.

00:20:30 ◼   ► - Yeah. - Whatever it's called.

00:20:31 ◼   ► - But yeah, but you look at the market

00:20:32 ◼   ► and like everything else is USB-C now.

00:20:35 ◼   ► It's a very, very different scenario now

00:20:39 ◼   ► than it was when Lightning was introduced.

00:20:41 ◼   ► You know, when they had to decide to go from the dot connector to Lightning, again, that

00:20:45 ◼   ► was a very different world.

00:20:47 ◼   ► There wasn't this consensus.

00:20:48 ◼   ► There wasn't this one amazing universal standard.

00:20:51 ◼   ► There was a bunch of miscellaneous crap and mostly micro-USB, which sucked.

00:20:56 ◼   ► And this is a totally different ballgame now.

00:20:57 ◼   ► It's a different time with different needs.

00:21:00 ◼   ► And the right thing to do now, whether the EU gets around to mandating it for them or

00:21:05 ◼   ► not, is to use USB-C for everything.

00:21:08 ◼   ► So rumors again suggest we're going there

00:21:10 ◼   ► and I hope they're right because it is such a pain

00:21:13 ◼   ► in the butt to have a family of mixed devices

00:21:17 ◼   ► of just like, you know what, we need USB-C

00:21:19 ◼   ► for pretty much everything except our iPhones.

00:21:22 ◼   ► That's so annoying.

00:21:23 ◼   ► Well and our Apple watches again, whole separate thing there

00:21:26 ◼   ► but the world would be better off

00:21:29 ◼   ► if they made the iPhone USB-C

00:21:31 ◼   ► and I hope that's the reason they're doing it

00:21:33 ◼   ► and not because of government pressure

00:21:36 ◼   ► or not because of only pros needing it.

00:21:39 ◼   ► No, everyone needs it.

00:21:40 ◼   ► It'll benefit everyone.

00:21:42 ◼   ► - Speaking of not having enough parts and stuff,

00:21:44 ◼   ► I seem to recall a story back when lightning first came out

00:21:46 ◼   ► that the limiting factor on the phones

00:21:48 ◼   ► they could manufacture was the ability

00:21:50 ◼   ► to get that little lightning connectory thing

00:21:53 ◼   ► because obviously Apple was the only company

00:21:54 ◼   ► in the entire world that needed that thing made

00:21:56 ◼   ► and they needed a lot of them

00:21:57 ◼   ► and they needed a lot of them fast

00:21:58 ◼   ► and they needed to be up to Apple standards of quality

00:22:01 ◼   ► and everything and that was a problem.

00:22:03 ◼   ► But at this point, getting quality USB-C connectors

00:22:07 ◼   ► should not be a problem for Apple.

00:22:09 ◼   ► - One would hope not.

00:22:10 ◼   ► But no, I mean, I haven't been traveling too much,

00:22:12 ◼   ► but I've been traveling more than zero,

00:22:13 ◼   ► which is more than I can say for the last couple of years.

00:22:15 ◼   ► And having everything USB-C is extremely convenient.

00:22:19 ◼   ► And yes, I did spend an absolutely hilarious amount of money

00:22:22 ◼   ► on my travel mag safe situation,

00:22:26 ◼   ► but it would still be nice to know

00:22:28 ◼   ► that if I needed to plug in,

00:22:30 ◼   ► all I need is the cable that I can use for my laptop

00:22:32 ◼   ► the switch or for any number of other things and that I don't need a bespoke cable just

00:22:37 ◼   ► for my iPhone. And yeah, I don't begrudge Apple for having gone lightning. I do kind

00:22:42 ◼   ► of begrudge them for not having gone to USBC sometime in the last year or two. And I think

00:22:46 ◼   ► that, you know, it would have been better a year or two ago. The next best time is as

00:22:50 ◼   ► soon as possible. And, and, and I just, the other day I had recorded, the kids were doing

00:22:55 ◼   ► like a little play for the two of us, for Aaron and me, and I recorded it and I recorded

00:22:59 ◼   ► it as like one 10 minute, you know, I don't remember what my settings on my phone are,

00:23:03 ◼   ► but I think it was a one 10 minute 60 frames per second 4k video. And it's something like

00:23:07 ◼   ► three gigs. I don't even remember how big it is, but it's massive. Right. And yeah,

00:23:11 ◼   ► getting that, uh, oh, I'm sorry, it's actually nine gigs, nine gigs. Getting that off of

00:23:17 ◼   ► my phone via cable was effectively impossible. Like, yes, it is. It is literally possible,

00:23:23 ◼   ► but it was effectively impossible. And so what I ended up doing was air dropping, which

00:23:28 ◼   ► took a couple of tries and was not exactly reliable, but I eventually got it from my

00:23:32 ◼   ► phone to my computer, and I'm going to eventually, you know, put it in Final Cut Pro and do things

00:23:37 ◼   ► with it, but it is not fun to get, you know, more than a minute of 4K 60 frames video off

00:23:45 ◼   ► of your phone using a lightning cable. It's just, I know you guys said this a minute ago,

00:23:48 ◼   ► but it's so true. And this is something that I ran into just in the last three days. I

00:23:54 ◼   ► cannot wait for USB-C to be a thing.

00:23:57 ◼   ► That being said, I wouldn't be entirely surprised

00:24:02 ◼   ► if Apple went no ports at all,

00:24:05 ◼   ► or perhaps no ports on non-pro phones

00:24:08 ◼   ► and ports on a pro phone.

00:24:10 ◼   ► And the reason I say that is,

00:24:11 ◼   ► what do they have to care about?

00:24:12 ◼   ► They have to care about developers,

00:24:13 ◼   ► who they don't really care about,

00:24:15 ◼   ► or people who have wired CarPlay.

00:24:17 ◼   ► And there are solutions, like I use,

00:24:19 ◼   ► they're not great, but they work,

00:24:21 ◼   ► to change wired CarPlay into wireless CarPlay.

00:24:23 ◼   ► So do either of you guys see them going to a completely wireless world?

00:24:28 ◼   ► There was that, uh, the rumor that like the mag safe puck,

00:24:31 ◼   ► like imagine a mag safe puck as they exist now,

00:24:33 ◼   ► but also with like a thing in the middle of it that lets data be transferred and

00:24:37 ◼   ► that would be their solution essentially like mag safe.

00:24:39 ◼   ► I don't know what number they're on mag safe two, three.

00:24:41 ◼   ► I thought I guess the numbers are on the laptops and I believe they reset the

00:24:44 ◼   ► timeline on that one. Yeah, it's just, it's, yeah. Yeah. As we said,

00:24:47 ◼   ► let previous episode, not mag safe, but mag safe. Anyway. Um, that was a,

00:24:51 ◼   ► They had patents filings on that and everything,

00:24:53 ◼   ► but it's hard to tell whether that's just a thing

00:24:55 ◼   ► they were considering for the MagSafe puck

00:24:57 ◼   ► and just didn't do, or if it's a thing for the future.

00:24:59 ◼   ► It doesn't really solve the carplay problem at all.

00:25:02 ◼   ► One of the problems it does solve is Apple's ability

00:25:04 ◼   ► to charge peripheral manufacturers money to sell things

00:25:08 ◼   ► that's sort of made for iPhone, whatever.

00:25:11 ◼   ► Does anybody make a MagSafe puck besides Apple?

00:25:15 ◼   ► Like the actual puck?

00:25:16 ◼   ► - Yeah, well, sort of.

00:25:18 ◼   ► My travel thing that I keep talking about

00:25:20 ◼   ► over and over again, there's three pieces.

00:25:23 ◼   ► There's a Qi charger that's about the shape

00:25:27 ◼   ► of an AirPods case, there's an Apple Watch charger,

00:25:30 ◼   ► and then there's a honest-to-goodness MagSafe,

00:25:32 ◼   ► not a puck, but there's a MagSafe mat on,

00:25:36 ◼   ► I think it's the rightmost of the three different parts

00:25:39 ◼   ► of this charger.

00:25:39 ◼   ► So yeah, it is full-on honest-to-goodness MagSafe,

00:25:42 ◼   ► but it is not a first-party MagSafe puck,

00:25:46 ◼   ► as far as I'm aware.

00:25:47 ◼   ► - Well, and to be clear, I'm sure they make

00:25:50 ◼   ► a decent amount of money with the licensing of everything.

00:25:53 ◼   ► But have you ever seen anybody in the real world

00:25:57 ◼   ► using officially licensed MFI stuff

00:25:59 ◼   ► that didn't come with their phone?

00:26:01 ◼   ► Everyone just buys the knockoff crap

00:26:02 ◼   ► from the drugstore on Amazon.

00:26:04 ◼   ► And I doubt they're getting any money from that.

00:26:06 ◼   ► So I wonder if we might be overestimating the value of that.

00:26:10 ◼   ► - I think there's a lot of it in there.

00:26:12 ◼   ► I think all of the manufacturers that you see selling stuff

00:26:16 ◼   ► on Apple Store, for example Belkin and stuff,

00:26:18 ◼   ► they're a venue manufacturer.

00:26:19 ◼   ► They send tons of stuff.

00:26:20 ◼   ► If you just do a random Google for any kind of wire peripheral

00:26:23 ◼   ► thing, the odds of you getting a Belkin match are high.

00:26:26 ◼   ► And I'm assuming everything Belkin does is on the up and up

00:26:29 ◼   ► because, again, they're an Apple store.

00:26:30 ◼   ► So they're probably--

00:26:32 ◼   ► I don't know.

00:26:32 ◼   ► It's someone's job to maximize MFI income.

00:26:36 ◼   ► And that person is not in charge of the whole company.

00:26:38 ◼   ► But that is a factor in weighing this.

00:26:41 ◼   ► And obviously, it's not going to stop them from going to USB-C.

00:26:43 ◼   ► They did it on the iPad.

00:26:45 ◼   ► No more revenue from all those lightning cables

00:26:47 ◼   ► that we were selling to iPad owners.

00:26:49 ◼   ► In the end, they'll do the right thing

00:26:50 ◼   ► from a technical perspective,

00:26:51 ◼   ► but it remains to be seen what Apple thinks

00:26:53 ◼   ► is the right thing on the phone,

00:26:55 ◼   ► because in the same way they removed the headphone port,

00:26:58 ◼   ► although they removed that from the iPad too,

00:26:59 ◼   ► but anyway, they may say, "Oh, on the phone,

00:27:01 ◼   ► "we need every ounce of space we can get.

00:27:03 ◼   ► "It's not like we have room for plastic spacers in there,

00:27:05 ◼   ► "so we gotta get rid of the port

00:27:07 ◼   ► "and everything is going to be magnetic pucks from now on."

00:27:10 ◼   ► I hope they don't do that.

00:27:11 ◼   ► It seems like a much more straightforward

00:27:13 ◼   ► and smarter thing to do to go with USB-C,

00:27:14 ◼   ► and in the end, I think the iPhone,

00:27:17 ◼   ► for all of Apple's sort of punctuated moments of daring

00:27:22 ◼   ► tends to be a conservative product.

00:27:24 ◼   ► And, you know, change happens slowly.

00:27:27 ◼   ► So the iPhone 10 was a big change,

00:27:29 ◼   ► going Retina was a big change,

00:27:31 ◼   ► the big iPhone 6 was a big change,

00:27:33 ◼   ► and going from 30 plane to lightning was a big change,

00:27:36 ◼   ► I think, but you know, but those events happen.

00:27:39 ◼   ► But in general, Apple's not keen to rock the boat

00:27:43 ◼   ► on the idea that a phone is something that's a rectangle

00:27:46 ◼   ► where you plug in a thing at the bottom to charge it.

00:27:48 ◼   ► So right now my money is still on a USB-C port

00:27:52 ◼   ► where the lightning port was.

00:27:54 ◼   ► I don't know, we'll see.

00:27:56 ◼   ► - It is tempting, you know, Apple famously,

00:27:59 ◼   ► frequently overdoes their minimalism,

00:28:01 ◼   ► especially in hardware, and so I see why we would think

00:28:06 ◼   ► that, you know, oh, this is a big risk,

00:28:08 ◼   ► and I think it's a risk that they might do this,

00:28:11 ◼   ► but I think it's a small risk, because ultimately,

00:28:15 ◼   ► Wired is in many ways, of course it's in many ways simpler,

00:28:20 ◼   ► but it's also in many ways better.

00:28:23 ◼   ► Wired charging first of all is way more efficient.

00:28:25 ◼   ► Now we've already seen, I believe there's a feature

00:28:27 ◼   ► in the 16.1 beta that came out yesterday or today,

00:28:30 ◼   ► that there's now an option to set your iPhone to charge

00:28:34 ◼   ► when it has environmentally friendly energy generation

00:28:39 ◼   ► in your area, so if there's a time of day in your area

00:28:44 ◼   ► where they use only solar or wind power or something,

00:28:47 ◼   ► it can have your phone try to only charge during those times.

00:28:50 ◼   ► And that's the kind of feature that,

00:28:52 ◼   ► the reason they do that kind of feature

00:28:53 ◼   ► is that it has a pretty massive environmental impact

00:28:57 ◼   ► when you're talking about the number of iPhones

00:28:59 ◼   ► that are out there.

00:29:00 ◼   ► If you can make them charge a little more efficiently

00:29:03 ◼   ► or using certain resources instead of others,

00:29:06 ◼   ► it's a small power draw, yeah,

00:29:07 ◼   ► but it's like millions and millions and millions

00:29:10 ◼   ► of small power draws.

00:29:12 ◼   ► And so to go to wireless only would make almost everyone

00:29:17 ◼   ► use an inefficient charging method on their iPhone

00:29:22 ◼   ► that loses a good percentage.

00:29:24 ◼   ► I mean, what is the Qi MagSafe efficiency?

00:29:27 ◼   ► It's probably something like 70 or 80%.

00:29:29 ◼   ► - I think that's very optimistic.

00:29:31 ◼   ► - Yeah, exactly.

00:29:32 ◼   ► - I would guess that it's less than 50%, much less.

00:29:35 ◼   ► - Right, and especially if you have a case on your phone,

00:29:38 ◼   ► then you're getting those coils further apart,

00:29:40 ◼   ► and I bet it makes the efficiency worse.

00:29:41 ◼   ► So if they're gonna put all the effort environmentally

00:29:44 ◼   ► to do other good things for like energy conservation

00:29:47 ◼   ► and smart energy usage and everything,

00:29:50 ◼   ► it seems like a step backwards to require

00:29:53 ◼   ► all of a sudden everyone to go Qi or wireless only.

00:29:57 ◼   ► In addition to the fact that I just hope

00:30:00 ◼   ► that this rumor is not true or that this idea

00:30:02 ◼   ► wouldn't happen, I hope because I know

00:30:05 ◼   ► from a developer's point of view,

00:30:06 ◼   ► like the Apple Watch development situation,

00:30:08 ◼   ► not being able to hardwire to it,

00:30:11 ◼   ► is just so inferior to the iPhone

00:30:13 ◼   ► where I can just hardwire in.

00:30:15 ◼   ► So I really hope they don't do that.

00:30:17 ◼   ► - Oh, agreed, agreed.

00:30:18 ◼   ► - But yeah, I think there's lots of reasons for this

00:30:21 ◼   ► why they probably wouldn't do it for any iPhone,

00:30:24 ◼   ► let alone for all iPhones.

00:30:28 ◼   ► - Yeah, I mean, I don't know.

00:30:29 ◼   ► I don't want there to be an all wireless future,

00:30:32 ◼   ► but I wouldn't put it past Apple one way or the other.

00:30:36 ◼   ► - We are sponsored this week by Linode,

00:30:39 ◼   ► my favorite place to run servers.

00:30:41 ◼   ► Visit linode.com/atp.

00:30:44 ◼   ► Look, if you need to run servers,

00:30:45 ◼   ► and if you're listening to this show,

00:30:47 ◼   ► I think you're disproportionately likely,

00:30:48 ◼   ► which is probably why they advertise with us,

00:30:50 ◼   ► you probably have needed to run a server at some point,

00:30:52 ◼   ► and need to run one now,

00:30:54 ◼   ► and Linode is by far my favorite place to do that.

00:30:56 ◼   ► I have run so many servers in my career,

00:30:59 ◼   ► and I'm actually trying to run fewer of them,

00:31:01 ◼   ► but for some reason to do that,

00:31:02 ◼   ► I need to keep adding more.

00:31:04 ◼   ► (laughs)

00:31:05 ◼   ► But I am so happy with Linode as a customer.

00:31:08 ◼   ► They have everything I need from compute instances

00:31:11 ◼   ► to specialty plans like GPU compute, high memory, dedicated

00:31:14 ◼   ► CPU, stuff like that.

00:31:16 ◼   ► And they have all sorts of managed services now as well.

00:31:18 ◼   ► So in addition to things I've used for a while,

00:31:20 ◼   ► like load balancing and managed backups,

00:31:23 ◼   ► they also now have block storage, object storage,

00:31:26 ◼   ► like S3, and managed database services now launching.

00:31:30 ◼   ► And so it's really just a huge amount of value

00:31:33 ◼   ► being created there.

00:31:34 ◼   ► And they really don't charge much for their offerings.

00:31:37 ◼   ► They are an amazing value and I have not found

00:31:40 ◼   ► a better value in the business.

00:31:41 ◼   ► And that's why I went to them in the first place

00:31:44 ◼   ► and that's one of the reasons why I've stayed there

00:31:46 ◼   ► all this time because I run a lot of servers,

00:31:48 ◼   ► the money would really add up in any other place

00:31:50 ◼   ► and Linode is by far the best value I found in the business.

00:31:55 ◼   ► So it's just, it's amazing, I got an amazing support,

00:31:57 ◼   ► amazing control panel, they have a full API if you need it,

00:32:00 ◼   ► all sorts of capabilities that I don't even understand

00:32:02 ◼   ► 'cause I don't need to but it's just, it's a fantastic host.

00:32:04 ◼   ► So see for yourself at linode.com/atp

00:32:08 ◼   ► and let them make your cloud computing fast, simple,

00:32:13 ◼   ► and affordable so you can focus on your projects,

00:32:15 ◼   ► not your infrastructure.

00:32:16 ◼   ► Once again, linode.com/atp.

00:32:19 ◼   ► Create a free account there, you get 100 bucks in credit.

00:32:21 ◼   ► Linode.com/atp.

00:32:23 ◼   ► New free accounts there, you get $100 in credit.

00:32:25 ◼   ► Thank you so much to Linode for being an awesome host

00:32:29 ◼   ► and for running all my servers

00:32:30 ◼   ► and of course, for sponsoring our show.

00:32:32 ◼   ► [Music]

00:32:36 ◼   ► So I have a couple of updates with regard to my weird Apple Watch

00:32:39 ◼   ► band return. John, not Syracuse, has a theory that I genuinely wish I could have confirmed or denied,

00:32:46 ◼   ► but all this came in after I had completed my return. John writes, "Here's why I think you

00:32:52 ◼   ► were asked to bring your watch bundle to return the band. The bundled band is not exactly the

00:32:57 ◼   ► same as the retail version, as you can see in pictures that John provided, which really you

00:33:02 ◼   ► You can just use your mind painting to figure this out. It has no barcode and no specifications about its color from the case

00:33:09 ◼   ► I can only tell that it's in my case a sports band on the other hand

00:33:13 ◼   ► The retail version has some more stickers for the model in the barcodes see another picture again

00:33:17 ◼   ► We're not gonna bother with pictures in the show notes partially for John's OPSEC. But anyway, I don't think this is the case

00:33:23 ◼   ► I thought on certainly on the exterior, you know how on the Apple watch

00:33:27 ◼   ► It's like a piece of paper that wraps the watch box in the band box

00:33:32 ◼   ► Well on that it's not a piece of paper like a thin piece of cardboard

00:33:35 ◼   ► Well anyways on that it says what band is within now

00:33:39 ◼   ► Maybe the band wasn't specific, but I could swear it had a picture of the correct band on the outside

00:33:44 ◼   ► I could swear it had like the size on it and so on and so forth

00:33:48 ◼   ► So, I don't know maybe John is right, but I'm skeptical but it's it's a plausible theory now from an anonymous Apple Watch employer

00:33:55 ◼   ► Apple Watch employee from an anonymous Apple employee, they had the following to say with

00:34:00 ◼   ► regard to the genius or retail person that I spoke to. They were wrong. We don't need the watch. All

00:34:06 ◼   ► we need is the serial number of the watch. Here's what actually happens during band swap. Systematically,

00:34:10 ◼   ► regardless of whether your watch is there or not, Apple is returning the whole thing and reselling

00:34:14 ◼   ► it to you with a new band. It's just that they've rejiggered the system to hide most of the exchange

00:34:19 ◼   ► from us, as in the retail employees, and the customer. The steps look like this. One, the specialist

00:34:24 ◼   ► chooses the item swap option on their ISAAC, which is the handheld device they carry.

00:34:28 ◼   ► Two, the system asks for a watch serial number. This can be scanned using the barcode from

00:34:33 ◼   ► the box, or it can be manually entered if the box isn't there. We can also scan the

00:34:38 ◼   ► original receipt. Since every serial number is unique, that number allows us to pull up

00:34:41 ◼   ► the original transaction. Quick aside, I later asked this anonymous genius, "Hey, what if

00:34:46 ◼   ► I had the W1234-5679 online order number?" And they weren't sure, but they said, "Yeah,

00:34:52 ◼   ► I bet that would work. All right, so back to the retail employee. Step three, after we have received,

00:34:57 ◼   ► retrieved, excuse me, the original transaction,

00:34:59 ◼   ► the system asks for us to scan in the new band. Four, if the new band is the same price as the old band, the

00:35:05 ◼   ► transaction is pretty much finished except for the receipt, which is printed or emailed. If the new band is a different price, the customer pays

00:35:11 ◼   ► the difference or receives a refund for the difference depending. Easy peasy, lickety-split. So this is

00:35:17 ◼   ► mostly what happened when I went back with the boxes and so on and so forth,

00:35:21 ◼   ► But the key here is that really all they need is a way to get the serial number and once they got that

00:35:27 ◼   ► Then they're off to the races and that makes sense and kind of stands to reason based on what I saw

00:35:33 ◼   ► But here's an anonymous Apple employee telling us exactly what the truth is

00:35:37 ◼   ► Alright, so let's talk about something that is definitely brand new and definitely hasn't been talked to death over the last two months

00:35:44 ◼   ► Let's talk about AI art

00:35:46 ◼   ► And I feel like we should start right away by saying you should really consider listening to Cortex episode 133,

00:35:54 ◼   ► "The Ethics of AI Art."

00:35:56 ◼   ► That was a real--I really love that show--

00:35:58 ◼   ► but that was a really, really great episode in which a lot of the ins and outs of all this was discussed.

00:36:03 ◼   ► How do we want to approach this? I guess we should maybe kind of do the quick summary of "What do I mean by AI art?"

00:36:10 ◼   ► John or Marco, jump in when you're ready.

00:36:13 ◼   ► But the general gist is there's been a lot of work put into various products, some open source, some not,

00:36:18 ◼   ► that allow you to do many different things.

00:36:21 ◼   ► But one of the things they allow you to do is type a prompt, type a picture of the three hosts of the Accidental Tech podcast

00:36:28 ◼   ► drawn as pixel art.

00:36:30 ◼   ► And these different products will use a whole bunch of machine learning and artificial intelligence to try to,

00:36:37 ◼   ► figuratively speaking, draw

00:36:40 ◼   ► whatever picture you've asked them to draw. Some of them are better than others, and we're going to talk about that

00:36:45 ◼   ► I think a little bit, but it's very very interesting and some of these products

00:36:49 ◼   ► I haven't played with all of them. In fact, I've only played with one of them,

00:36:52 ◼   ► but the one I played with, a lot of times it gave me straight-up garbage,

00:36:56 ◼   ► but occasionally it would come up with something reasonable, and when it did it was kind of

00:37:00 ◼   ► mind-shattering that I could ask a computer with plain text, like describe a phantom picture

00:37:05 ◼   ► I had in my mind in plain text, and have the computer

00:37:08 ◼   ► basically come up with it, it's really wild.

00:37:11 ◼   ► And so that's kind of what we're talking about here.

00:37:14 ◼   ► That, I don't know, John, I think you were most excited

00:37:17 ◼   ► to talk about this, how do you wanna proceed?

00:37:19 ◼   ► - We should list some of the ones that are out there,

00:37:20 ◼   ► I think you wanna try them, we'll have these links

00:37:22 ◼   ► in the show notes, there's DALL-E, D-A-L-L hyphen E,

00:37:25 ◼   ► that's a play on the Salvador DALL-E and WALL-E,

00:37:28 ◼   ► the robot from the Pixar movie.

00:37:30 ◼   ► There's Stable Diffusion, Midjourney, Google Imogen,

00:37:34 ◼   ► a whole bunch of other ones, right?

00:37:36 ◼   ► One of them is available as like a standalone application

00:37:39 ◼   ► that you can run on your ARM-based Mac.

00:37:41 ◼   ► What is that one called?

00:37:42 ◼   ► - Stable Diffusion, I believe that's the one.

00:37:43 ◼   ► - Right, but is it like Diffusion B or something?

00:37:45 ◼   ► - It's Diffusion B.

00:37:46 ◼   ► - Yeah, we'll put a link to that.

00:37:47 ◼   ► That's the one I play with on my Mac as well.

00:37:49 ◼   ► A lot of the other ones have web interfaces.

00:37:50 ◼   ► I think Dolly used to be invitation only for a while,

00:37:54 ◼   ► but now I think it's open to everybody.

00:37:55 ◼   ► So you can follow the links and try them out,

00:37:57 ◼   ► and people post the interesting things they come up with.

00:38:01 ◼   ► And the thing I wanted to talk about with this,

00:38:04 ◼   ► it's been a topic of conversation

00:38:05 ◼   ► because, I mean, first it went around, like, I don't know,

00:38:08 ◼   ► a year or two ago, whenever the first one of these

00:38:10 ◼   ► started coming out, it was like a technical curiosity,

00:38:13 ◼   ► and it started to get mainstream enough

00:38:14 ◼   ► that you'd see articles about it

00:38:16 ◼   ► and just regular sort of tech websites

00:38:18 ◼   ► and other stuff like that, and at a certain point,

00:38:21 ◼   ► it started to be so mainstream that people were using it,

00:38:25 ◼   ► not just as a technical curiosity,

00:38:26 ◼   ► but they were using it to make a picture

00:38:29 ◼   ► that they would then use.

00:38:29 ◼   ► I think there was some controversy

00:38:30 ◼   ► 'cause one person was actually writing,

00:38:32 ◼   ► I don't think it was an article about AI Art,

00:38:34 ◼   ► but they, or maybe it was,

00:38:35 ◼   ► They use the AI art to, they talk about this

00:38:37 ◼   ► in the Cortex episode, to make an image

00:38:40 ◼   ► that they included with their article online

00:38:43 ◼   ► rather than paying an artist and people were mad,

00:38:45 ◼   ► like why didn't you pay an artist to do this?

00:38:46 ◼   ► And that sort of gets into the,

00:38:48 ◼   ► I think the most interesting part of this debate is,

00:38:51 ◼   ► given that people are working on this tech,

00:38:53 ◼   ► what does it mean for the future of all things

00:38:57 ◼   ► related to making pictures?

00:38:59 ◼   ► And by the way, there are movie ones now as well,

00:39:01 ◼   ► where you can ask it, like someone,

00:39:02 ◼   ► You can ask it like, a painting of an ice cream cone melting in the sun, and it will do a video of,

00:39:09 ◼   ► you know, a video of a painterly style ice cream cone melting in the sun, right? So it's not just

00:39:13 ◼   ► audio, or not just still images. And this stuff is developing so quickly, it presents a lot of very

00:39:21 ◼   ► thorny questions. Obviously, if you are, if your profession is drawing pictures for money,

00:39:29 ◼   ► and there's a program that lets people draw pictures

00:39:31 ◼   ► by typing what they want to see in the picture.

00:39:34 ◼   ► That probably doesn't make you feel good

00:39:35 ◼   ► about your chosen profession.

00:39:38 ◼   ► There is the predictable sort of Luddite

00:39:41 ◼   ► versus tech enthusiast, you know, battle there

00:39:45 ◼   ► between saying, "Oh, a computer can never do

00:39:47 ◼   ► what a human does," and, you know, like that.

00:39:52 ◼   ► Anytime there's any kind of technology

00:39:53 ◼   ► that previously does something

00:39:54 ◼   ► that could only be done by humans,

00:39:55 ◼   ► there is this battle saying that the new way to do it

00:39:58 ◼   ► It is soulless and bad and evil and is going to corrupt the youth and so on and so forth.

00:40:04 ◼   ► And then the other people who are excited about the tech and just want it to go forward.

00:40:08 ◼   ► And many times throughout history there has been a technology that has caused entire professions

00:40:12 ◼   ► and entire industries to basically disappear or shrink to the point or transform in a way

00:40:17 ◼   ► that's not even recognizable.

00:40:19 ◼   ► Witness the entire industry surrounding having horses pull things with people in them and

00:40:24 ◼   ► the advent of the automobile.

00:40:25 ◼   ► It's not like we don't have horses.

00:40:26 ◼   ► It's not like people don't have jobs making saddles for horses

00:40:28 ◼   ► and shoeing horses and taking care of horses.

00:40:30 ◼   ► All those jobs still exist.

00:40:31 ◼   ► But boy, that industry looks a lot different

00:40:33 ◼   ► than it did before the advent of the automobile,

00:40:35 ◼   ► a lot different.

00:40:36 ◼   ► And so this AI art thing brings all those issues up,

00:40:39 ◼   ► and then people are going around and around that debate.

00:40:41 ◼   ► But I think one of the most interesting aspects

00:40:45 ◼   ► of this debate is how these things work.

00:40:48 ◼   ► Like, how do you make a program where you type in words

00:40:50 ◼   ► and it draws you pictures, right?

00:40:52 ◼   ► And like most sort of machine learning style models,

00:40:55 ◼   ► they are, I don't know if trained is the right word,

00:40:58 ◼   ► but they are given a set of images

00:41:02 ◼   ► and associated words and phrases and stuff

00:41:06 ◼   ► to say, to feed into the model so that they can do this.

00:41:11 ◼   ► So I don't know, so like millions and millions of images,

00:41:13 ◼   ► I'm not sure how they're tagged,

00:41:14 ◼   ► maybe each one is just tagged with a caption

00:41:16 ◼   ► or something like that,

00:41:17 ◼   ► and they grind that up into a big soup,

00:41:19 ◼   ► and that's the technical term.

00:41:21 ◼   ► And then when you say a picture of an ice cream cone,

00:41:23 ◼   ► they give you a picture of an ice cream cone

00:41:25 ◼   ► because they have enough images of enough things

00:41:27 ◼   ► with enough of the words associated with it

00:41:29 ◼   ► that they can synthesize a picture

00:41:31 ◼   ► based on everything they've ever seen

00:41:34 ◼   ► and the association of those words to the images

00:41:36 ◼   ► to get you, not just one,

00:41:38 ◼   ► but multiple pictures of ice cream cones

00:41:39 ◼   ► and trying various attempts at it.

00:41:42 ◼   ► And that raises a lot of interesting questions

00:41:46 ◼   ► in particular, what images were fed to this thing?

00:41:50 ◼   ► Most of the things that are done

00:41:53 ◼   ► for universities or whatever,

00:41:56 ◼   ► this is like sort of research type stuff.

00:41:57 ◼   ► And the first versions of these,

00:41:59 ◼   ► there are lots of image sets that are used

00:42:02 ◼   ► for research purposes that are presumably

00:42:04 ◼   ► like millions and millions of correctly annotated

00:42:09 ◼   ► royalty free images that have been used

00:42:12 ◼   ► in lots of different computer vision studies

00:42:14 ◼   ► for years and years, right?

00:42:16 ◼   ► But there are so many of these things

00:42:18 ◼   ► and they're so popular and the new ones come up every day,

00:42:21 ◼   ► it's not entirely clear what's in all the image sets

00:42:23 ◼   ► that they're using.

00:42:24 ◼   ► So there was a story about this on Arstactica,

00:42:26 ◼   ► I guess actually it's fairly recent.

00:42:28 ◼   ► Headline is artist finds private medical record photos

00:42:31 ◼   ► in popular AI training data set.

00:42:33 ◼   ► (laughing)

00:42:35 ◼   ► - What?

00:42:36 ◼   ► - Someone who found private medical record photos

00:42:38 ◼   ► taken by her doctor in 2013 referenced

00:42:40 ◼   ► in the LA ION 5B image set.

00:42:43 ◼   ► - Whoops.

00:42:45 ◼   ► - Because a lot of the ways they find images

00:42:47 ◼   ► is like, oh, I'll just scrape the web

00:42:48 ◼   ► and anything I find the web I'm sure is fine to use.

00:42:51 ◼   ► Well, no, because there might be a doctor website

00:42:53 ◼   ► that has poor security and has an image exposed to the web

00:42:55 ◼   ► and a web scraper comes along and finds it.

00:42:57 ◼   ► And lo and behold, your like medical images

00:43:00 ◼   ► end up in a dataset that is, you know,

00:43:02 ◼   ► and to be clear, it's not like--

00:43:04 ◼   ► - Isn't this also how my email address

00:43:05 ◼   ► ends up on so many people's lists?

00:43:07 ◼   ► Oh, you must have bought something from us.

00:43:09 ◼   ► Nope, I sure didn't.

00:43:11 ◼   ► - Well, I mean, it's like bad website security

00:43:14 ◼   ► in that like there's a page that is protected,

00:43:15 ◼   ► you know, security through obscurity.

00:43:17 ◼   ► Technically, if you know the URL, you can get to it,

00:43:19 ◼   ► but there's no link to it.

00:43:20 ◼   ► So how can people find it?

00:43:21 ◼   ► Well, a lot of the web scraping things can find links that are not visible because they're

00:43:24 ◼   ► hidden on a page or they do scraping techniques that allow them to find things by iterating

00:43:29 ◼   ► on IDs or stuff like that.

00:43:32 ◼   ► And that's not great.

00:43:33 ◼   ► And on top of that, as we all know, if you are an artist who does any kind of art and

00:43:39 ◼   ► puts it on the internet, you will find that art all over the place.

00:43:44 ◼   ► And you may think this is a show with three software developers.

00:43:47 ◼   ► That doesn't apply to us.

00:43:48 ◼   ► So we make a small amount of art.

00:43:51 ◼   ► You may have seen some of it and purchased some of it

00:43:52 ◼   ► on our t-shirts, and let me tell you,

00:43:54 ◼   ► that very simple art that is on our t-shirts

00:43:56 ◼   ► is all over the freaking web.

00:43:58 ◼   ► Not because we put it there, but it's everywhere, right?

00:44:01 ◼   ► If you put an image on the web,

00:44:03 ◼   ► oh, and you go to the Cotton Bureau page

00:44:04 ◼   ► and it shows you a picture of the shirt,

00:44:05 ◼   ► there's that image on the web,

00:44:07 ◼   ► and a web scraper could find it,

00:44:08 ◼   ► because that page at Cotton Bureau, totally unprotected.

00:44:11 ◼   ► So if a web scraper finds it,

00:44:12 ◼   ► that image is probably in some image set somewhere, right?

00:44:15 ◼   ► It's probably in all these image sets,

00:44:17 ◼   ► because it's in the wild west out there on the internet.

00:44:19 ◼   ► You can find the image, you can put it in.

00:44:20 ◼   ► So there are, and obviously this is just a tiny,

00:44:23 ◼   ► we're doing t-shirt graphs, whatever,

00:44:24 ◼   ► imagine you are an actual artist by profession

00:44:28 ◼   ► and you do this artwork and maybe you have it

00:44:29 ◼   ► on your website that shows your portfolio

00:44:31 ◼   ► of all your great artwork that you're gonna do

00:44:33 ◼   ► and your artwork, probably also your name

00:44:36 ◼   ► and the descriptions of that stuff,

00:44:37 ◼   ► gets shoved into one of these image sets

00:44:40 ◼   ► to the point where, let's say you're a famous artist

00:44:42 ◼   ► and you have a name that people know.

00:44:43 ◼   ► It says like, you know, picture of a toaster oven

00:44:47 ◼   ► in the style of Ralph McQuarrie, right?

00:44:50 ◼   ► That knows who McQuarrie is based on the images they pulled

00:44:54 ◼   ► as they're all over the freaking internet

00:44:55 ◼   ► and his name is attached to all of them

00:44:56 ◼   ► and it's gonna show you a toaster oven drawn in his style.

00:45:00 ◼   ► I think he's dead now, so it's not, but hey,

00:45:01 ◼   ► like tons of living artists,

00:45:04 ◼   ► like you'll type something on one of these things

00:45:05 ◼   ► and they're like, that looks a lot like something I did once

00:45:09 ◼   ► and you know why?

00:45:10 ◼   ► Because some of your artwork is probably fed

00:45:12 ◼   ► into this machine and it's popping out.

00:45:14 ◼   ► And these people are like, well, how can you do that?

00:45:17 ◼   ► Are you allowed?

00:45:18 ◼   ► And they're like, well, it's not really your art.

00:45:19 ◼   ► This was made by the AI.

00:45:20 ◼   ► It's an original work.

00:45:22 ◼   ► It's like, yeah, but it's an original work

00:45:24 ◼   ► informed by work that I did.

00:45:26 ◼   ► And that doesn't seem like,

00:45:27 ◼   ► should I get some kind of royalty for this?

00:45:29 ◼   ► Should you have my permission?

00:45:30 ◼   ► - Does that count as a derivative work

00:45:32 ◼   ► for copyright reasons?

00:45:33 ◼   ► - Yeah, or should you at least get my permission

00:45:35 ◼   ► to include any of my artwork in there?

00:45:36 ◼   ► Or should you clean your data set

00:45:38 ◼   ► to make sure that the images you have,

00:45:39 ◼   ► you actually do have the rights to?

00:45:41 ◼   ► And that's almost impossible because you need millions of images to do this, or at least

00:45:44 ◼   ► a very large number of images.

00:45:46 ◼   ► Having a human vet each one for copyright, it's just, you know, it's like everything

00:45:49 ◼   ► that's true about the internet is true of these AI image things.

00:45:52 ◼   ► Like the perfect world where you're like, oh, we have to make sure every one of the

00:45:56 ◼   ► images that contributes to this input is free and clear and we know all the rights to it,

00:46:01 ◼   ► and that's impossible at scale.

00:46:02 ◼   ► Like that's just not how the internet works.

00:46:05 ◼   ► We can't even get all the movies and television shows made in the pre-internet or on streaming

00:46:09 ◼   ► services because people can't figure out how to do the rights.

00:46:10 ◼   ► And that is a much smaller problem than millions of images and image sets.

00:46:15 ◼   ► So there's that whole debate and rathole about do artists deserve to get paid?

00:46:22 ◼   ► You know, should they be allowed to do this?

00:46:25 ◼   ► Is there some kind of royalty structure?

00:46:26 ◼   ► Should this stuff be removed?

00:46:30 ◼   ► I think the final interest, well, two more interesting things about this.

00:46:34 ◼   ► is given that that's the way these things work,

00:46:39 ◼   ► well actually before we move on to that,

00:46:41 ◼   ► I should ask you too, do you have an opinion

00:46:42 ◼   ► on the artists having their work sucked into the thing?

00:46:46 ◼   ► Like what do you think about the validity

00:46:50 ◼   ► of the artist complaining in that scenario?

00:46:53 ◼   ► - I think it's really still yet to be proven

00:46:56 ◼   ► like what our acceptable standards are for this.

00:46:58 ◼   ► So my barometer for like what is an unacceptable

00:47:04 ◼   ► level of copying, you know, just ethically.

00:47:06 ◼   ► And there's legal definitions as well,

00:47:08 ◼   ► which I think kind of comport with this, but anyway,

00:47:11 ◼   ► is it's based on like, are you, to make a new work,

00:47:16 ◼   ► are you pretty much lifting most of your stuff

00:47:19 ◼   ► from the same source or the same very small number

00:47:23 ◼   ► of sources, then that's kind of over the line.

00:47:26 ◼   ► Whereas if you are taking bits of inspiration

00:47:29 ◼   ► from a diverse set of sources so that the resulting work

00:47:34 ◼   ► doesn't look like just a straight up clone

00:47:36 ◼   ► of one other person's work, but it looks like,

00:47:38 ◼   ► okay, maybe you were inspired in this way by this person,

00:47:40 ◼   ► and this way by this work, and this way by this style,

00:47:43 ◼   ► but it all comes together into a more diverse soup

00:47:46 ◼   ► of a product, I think that's okay.

00:47:49 ◼   ► And so, you can look at these AI generators and say,

00:47:51 ◼   ► well, if you ask for something that is in one particular

00:47:56 ◼   ► person's style, that could result in something

00:48:00 ◼   ► that is over the line, whereas if you just ask

00:48:04 ◼   ► for an image of a slice of pepperoni pizza on a table,

00:48:07 ◼   ► like that's gonna be probably drawing

00:48:09 ◼   ► from so many different data points and input sources

00:48:12 ◼   ► that I don't think if the texture on the pepperoni slice

00:48:16 ◼   ► happens to look like the way you texture something

00:48:18 ◼   ► in Photoshop once, I think that's less of a concern.

00:48:21 ◼   ► But the problem is you can use these tools

00:48:24 ◼   ► the way, whatever the operator wants.

00:48:28 ◼   ► And if the operator says rip off this one person's work

00:48:31 ◼   ► or their style, you're gonna have a problem.

00:48:35 ◼   ► But I don't necessarily know that that's the fault

00:48:37 ◼   ► of the technology, that's the fault of the user.

00:48:40 ◼   ► - So assigning blame on this,

00:48:45 ◼   ► how is this different than other scenarios?

00:48:47 ◼   ► Assigning blame is always fun when it's a computer program,

00:48:50 ◼   ► quote unquote, doing it,

00:48:51 ◼   ► but then the user is prompting it to do it.

00:48:53 ◼   ► And that kind of leads to the question

00:48:56 ◼   ► that you usually end up at in these type of debates is,

00:48:58 ◼   ► Is this thing doing anything different than what people do?

00:49:02 ◼   ► If you ask a human artist to draw something,

00:49:05 ◼   ► they have a corpus of images they've seen

00:49:07 ◼   ► through their entire life that contributes to the output.

00:49:10 ◼   ► You could say, well, if I give it to a person,

00:49:15 ◼   ► they're gonna do original work,

00:49:16 ◼   ► but the original work of an artist is necessarily informed

00:49:19 ◼   ► by their entire life experience of seeing everything,

00:49:23 ◼   ► of seeing things in real life, obviously,

00:49:24 ◼   ► but also of seeing other pictures and works of art,

00:49:27 ◼   ► inevitably and

00:49:28 ◼   ► Some people would say well this AI program what it's doing is absolutely no different than what a human does

00:49:33 ◼   ► It has a series of inputs and that contributes to what it's going to make if you ask a human

00:49:39 ◼   ► To give you a logo in the style of a solid bass logo

00:49:42 ◼   ► They can probably do that because they know about those logos because they're very famous and he's a very famous logo designer

00:49:46 ◼   ► And if you do that

00:49:47 ◼   ► He's not keeping a dead people. He's not gonna rise from the grave and sue you

00:49:51 ◼   ► because you can't sort of

00:49:55 ◼   ► Trademark a style if I tell you to draw something in the style of any living artist

00:50:00 ◼   ► You can do that and they can't say oh it's illegal for you to do that because you just copied myself now

00:50:04 ◼   ► They may look down on you and say you didn't come up with your own original style

00:50:07 ◼   ► But every style is a you know

00:50:09 ◼   ► Everything's a remix every the style is a that we think of as new and novel is itself informed by all the other styles that came

00:50:15 ◼   ► Before it so in one sense. I agree that this program is

00:50:21 ◼   ► doing something that if you squint it looks very similar to what people also do.

00:50:27 ◼   ► But that leads to the second question which is, can this program, not can it make anything

00:50:34 ◼   ► new, can any of these programs make anything new?

00:50:37 ◼   ► But like if you sort of fast forward this you do the, whatever it is, it's not argument

00:50:42 ◼   ► ad absurdum, no, but if you just, it's an infinite timeline argument.

00:50:46 ◼   ► So say the artists become the horse and buggy salesman, right?

00:50:52 ◼   ► And they still exist and they're out there, but boy, there are a lot fewer of them because

00:50:55 ◼   ► these AI programs get so good that in the average working life of a person, nobody actually

00:51:00 ◼   ► pays an artist to do anything.

00:51:02 ◼   ► We just type words into a program and we get an output, right?

00:51:06 ◼   ► If all the input to these programs are images made before these programs existed, then how

00:51:12 ◼   ► does that sustain itself?

00:51:13 ◼   ► Can you feed the output of these things back in as the input?

00:51:16 ◼   ► So forget about computers.

00:51:17 ◼   ► You've just got humans making art.

00:51:19 ◼   ► Humans make art, then new humans arrive and see the art made by the previous humans, and

00:51:23 ◼   ► they in turn make "new art" that then the future humans see and it feeds back in.

00:51:27 ◼   ► So you can see how the people are kind of like programs in this scenario where they

00:51:31 ◼   ► see existing art, they make "new art," and then that cycle repeats itself.

00:51:36 ◼   ► If you took the humans out of the equation, could the machines continue to do the same

00:51:41 ◼   ► thing?

00:51:42 ◼   ► as input all the art ever made by humans,

00:51:44 ◼   ► and then going forward, taking as input

00:51:46 ◼   ► all the art made by AI programs.

00:51:49 ◼   ► Or would they stagnate and feed in on themselves

00:51:52 ◼   ► to everything that was just a giant gray mush

00:51:53 ◼   ► because like, you know,

00:51:55 ◼   ► mixing all the paint colors together?

00:51:56 ◼   ► Or would they be just as averse as human artists?

00:51:59 ◼   ► And that I think makes me personally circle back to,

00:52:02 ◼   ► are they doing what people do?

00:52:04 ◼   ► And I think fundamentally they are not doing what people do.

00:52:07 ◼   ► Big strokes it seems like they're doing,

00:52:09 ◼   ► hey, they see pictures and they make new pictures.

00:52:11 ◼   ► That's exactly what people do.

00:52:13 ◼   ► But it's not.

00:52:14 ◼   ► Like, all these AI things,

00:52:15 ◼   ► and that's why general artificial intelligence,

00:52:17 ◼   ► or whatever you wanna call it, is so far away.

00:52:20 ◼   ► Even if these programs were operating the exact same way

00:52:24 ◼   ► that the center of our brains that makes pictures do,

00:52:27 ◼   ► and they're not, but even if they were,

00:52:29 ◼   ► there's so much more to a human mind

00:52:32 ◼   ► than the part that makes pictures based on word prompts.

00:52:36 ◼   ► And what these programs don't have,

00:52:38 ◼   ► and won't have for a long, long, long time

00:52:41 ◼   ► is the life experience of a human,

00:52:44 ◼   ► all the sensory input they've ever had,

00:52:46 ◼   ► all the emotions they experience,

00:52:47 ◼   ► the way humans judge a picture,

00:52:50 ◼   ► whether it accomplishes the goal they set out from this,

00:52:52 ◼   ► the ability to set a goal for themselves,

00:52:55 ◼   ► the ability to experience art

00:52:56 ◼   ► and feel what the art is meant to feel,

00:52:59 ◼   ► thus judging whether this art has achieved

00:53:01 ◼   ► what you wanted it to achieve

00:53:02 ◼   ► or inspiring you to do something else

00:53:04 ◼   ► based on how something you saw made you feel.

00:53:06 ◼   ► None of these programs can do any of that,

00:53:08 ◼   ► and it necessarily makes the funnel through which

00:53:13 ◼   ► they have to shove all of their creative efforts so narrow.

00:53:16 ◼   ► They do not have the wealth of experiences of a human.

00:53:19 ◼   ► All they have is visual input and descriptions.

00:53:22 ◼   ► They don't have an experience of the art.

00:53:25 ◼   ► So the art that they make can only be informed

00:53:27 ◼   ► by those tiny little things

00:53:29 ◼   ► 'cause they literally can't experience anything else.

00:53:31 ◼   ► They have no memory, no life, not memory in that sense,

00:53:35 ◼   ► like no memories, no life experience,

00:53:37 ◼   ► no sensory organs, no emotions, no thoughts,

00:53:40 ◼   ► no awareness like they're not artificial intelligence

00:53:44 ◼   ► in that sense.

00:53:45 ◼   ► Do you need that to make a picture of a slice on a table?

00:53:48 ◼   ► No, but I think you need that to continue the cycle

00:53:53 ◼   ► of creation of art with the quality level

00:53:57 ◼   ► that we have come to expect from humans.

00:53:59 ◼   ► Because as we make each new generation of humans,

00:54:00 ◼   ► they have new experiences, their life experiences

00:54:03 ◼   ► and the art that they see and the things that they feel

00:54:05 ◼   ► and form the things that they create.

00:54:07 ◼   ► and it is a rich tapestry, as they say.

00:54:09 ◼   ► And it's great to be able to feed that into an AI

00:54:11 ◼   ► and have it chomp that down,

00:54:12 ◼   ► but if you take the humans out of that equation

00:54:14 ◼   ► and leave the AI as stupid as they are now,

00:54:17 ◼   ► it would basically be like they were working

00:54:19 ◼   ► from the same set of data forever

00:54:21 ◼   ► and they would just grind it to a pulp

00:54:22 ◼   ► and it would just be this incredible stagnation.

00:54:24 ◼   ► Not that I think this is gonna happen

00:54:25 ◼   ► because you can't stop humans from making things.

00:54:28 ◼   ► Unless the machines kill us all Terminator style,

00:54:31 ◼   ► we don't have to worry about that.

00:54:32 ◼   ► But just as an academic exercise,

00:54:34 ◼   ► I don't think AI art is a sustainable thing

00:54:39 ◼   ► without human creativity as an important input.

00:54:41 ◼   ► It would be sad to think that the only purpose

00:54:43 ◼   ► of human creativity in artwork would be to feed into AIs

00:54:45 ◼   ► to do most of the drudgery, and then, you know, again,

00:54:47 ◼   ► they'd be like the people who own horses now.

00:54:49 ◼   ► They're out there, there's a lot of them,

00:54:51 ◼   ► but not nearly as many as there were.

00:54:53 ◼   ► And I don't think we have any particular fear

00:54:55 ◼   ► of that in our lifetime.

00:54:57 ◼   ► But that's kind of where I come down on this.

00:54:59 ◼   ► Setting aside the legalities and everything,

00:55:01 ◼   ► These programs are so dumb and so bad at what they do,

00:55:05 ◼   ► we're impressed.

00:55:06 ◼   ► You know, it's, this one analogy,

00:55:08 ◼   ► like seeing a rhinoceros dance.

00:55:09 ◼   ► You're impressed that it could do it,

00:55:10 ◼   ► but boy, the dancing isn't that great, right?

00:55:12 ◼   ► And they're never going to be adequate

00:55:16 ◼   ► to sustain a creative timeline of works of art like humans

00:55:21 ◼   ► until their experience of life is as rich

00:55:24 ◼   ► as a human's experience of life,

00:55:25 ◼   ► in which point we have lots of other problems.

00:55:28 ◼   ► And we're not even close to that, so don't worry about it.

00:55:31 ◼   ► Don't let the people who are tell you that AI is going to take over and kill us all.

00:55:34 ◼   ► If you're listening to this now, that will not happen when you're alive, so don't worry

00:55:36 ◼   ► about it.

00:55:37 ◼   ► Kind of like self-driving cars.

00:55:38 ◼   ► Hey-oh!

00:55:39 ◼   ► Hey-oh!

00:55:40 ◼   ► Well, but I think there's actually some overlap there, because I don't think that this is

00:55:45 ◼   ► going to put artists out of business as a whole.

00:55:51 ◼   ► It's more like thinking about this is a new digital tool that can save a lot of busy work.

00:55:58 ◼   ► it's gonna make certain types of art more accessible

00:56:01 ◼   ► than they were before to more people,

00:56:04 ◼   ► and it's going to save a bunch of time

00:56:07 ◼   ► on work that previously was more manual.

00:56:10 ◼   ► So if you think about it kind of like

00:56:12 ◼   ► when digital art came around,

00:56:15 ◼   ► when Photoshop and everything came around

00:56:17 ◼   ► and digital drawing tools and things like that,

00:56:19 ◼   ► there was a whole industry before that

00:56:22 ◼   ► of people who were doing a lot of this stuff by hand.

00:56:25 ◼   ► If photo retouching by hand,

00:56:27 ◼   ► painting and illustrating by hand.

00:56:29 ◼   ► And when you move to digital,

00:56:31 ◼   ► a whole bunch of things got easier.

00:56:33 ◼   ► And things became much more easily possible

00:56:36 ◼   ► that weren't easily possible before.

00:56:39 ◼   ► And so that did inevitably put out of work

00:56:42 ◼   ► like sign painters and things like that to some degree.

00:56:46 ◼   ► But most people who were artists in some way

00:56:49 ◼   ► embraced the new tools in some form and just became,

00:56:53 ◼   ► their job just became a little bit different.

00:56:55 ◼   ► But it didn't kill the art.

00:56:57 ◼   ► it just changed what was out there and what was available

00:57:00 ◼   ► and how you had to use it and what was possible.

00:57:04 ◼   ► And again, a bunch of new people were able to do it

00:57:06 ◼   ► who weren't able to do it before

00:57:07 ◼   ► or maybe it was like a little bit too tedious before

00:57:10 ◼   ► and now people were able to do things

00:57:12 ◼   ► who like wouldn't have done the old tedious way

00:57:15 ◼   ► but were willing to do the new digital way.

00:57:17 ◼   ► And so it just, it changes things.

00:57:18 ◼   ► And not everyone comes along on those transitions.

00:57:22 ◼   ► Every time technology gets better,

00:57:24 ◼   ► certain jobs aren't necessary anymore.

00:57:27 ◼   ► of the horse analogy, and not every person

00:57:30 ◼   ► who was keeping horses became an auto mechanic.

00:57:33 ◼   ► It doesn't work that way, but a lot of people,

00:57:37 ◼   ► a lot of people do become auto mechanics

00:57:38 ◼   ► when that demand rises up.

00:57:40 ◼   ► In this case, when digital art tools came around,

00:57:43 ◼   ► a lot of people became digital artists.

00:57:46 ◼   ► Not everyone who was previously drawing stuff by hand,

00:57:48 ◼   ► which by the way still exists and is fine,

00:57:50 ◼   ► but not everyone who did that went digital,

00:57:53 ◼   ► but many people did, and many new people started on digital.

00:57:57 ◼   ► And so art is a thing that's still a major thing in the world.

00:58:01 ◼   ► It's just different than it used to be.

00:58:02 ◼   ► I think these AI tools, using them and figuring out

00:58:06 ◼   ► how to make these text prompts, how to control them,

00:58:11 ◼   ► what knobs and dials to adjust, how you word things,

00:58:14 ◼   ► what you even think to create, that's all art.

00:58:18 ◼   ► That's part of the process.

00:58:20 ◼   ► These are now just able to generate things much more

00:58:24 ◼   ► quickly than a human can.

00:58:25 ◼   ► But then humans are still directing them,

00:58:27 ◼   ► humans are still tweaking them,

00:58:29 ◼   ► humans are still deciding, okay, you know what,

00:58:31 ◼   ► generate 100 pictures of this thing,

00:58:33 ◼   ► and I'm gonna pick the one that I like out of this 100

00:58:36 ◼   ► and have you riff on that a little more.

00:58:39 ◼   ► And then go to that one, generate 100 riffs on that one.

00:58:43 ◼   ► Okay, I'm gonna pick these two,

00:58:44 ◼   ► let's follow these through and do more with these.

00:58:47 ◼   ► That's art, that's humans doing art with a different tool.

00:58:51 ◼   ► And it doesn't have to be entirely used

00:58:53 ◼   ► for entire images too.

00:58:55 ◼   ► As the tooling and as technology gets more mature

00:58:59 ◼   ► and more established, these kind of tools

00:59:01 ◼   ► can be used for things like, okay, you know what,

00:59:03 ◼   ► I'm drawing this thing in Photoshop,

00:59:05 ◼   ► I have a brick wall here, can you just put a brick texture

00:59:07 ◼   ► on this wall that looks good, that hasn't been used

00:59:09 ◼   ► a million times by everyone else

00:59:10 ◼   ► who's ever used Photoshop in their life?

00:59:12 ◼   ► And it can generate a brick texture, fine.

00:59:14 ◼   ► Or hey, this car that's in the background of this photo,

00:59:16 ◼   ► I don't want this car to be here,

00:59:18 ◼   ► can you delete that in a way that's even smarter

00:59:20 ◼   ► than constant aware fill and stuff like that?

00:59:22 ◼   ► As the AI tools get better, it adds a lot of capabilities

00:59:26 ◼   ► for artists to eliminate busy work that used to exist

00:59:31 ◼   ► or to do things in a nicer way

00:59:33 ◼   ► than they used to be able to be done.

00:59:34 ◼   ► And so I see this really as a mixed bag.

00:59:39 ◼   ► There are, yes, some downsides,

00:59:41 ◼   ► and some artists will be put out of work by this,

00:59:44 ◼   ► but it also opens up so much potential

00:59:46 ◼   ► for artists to use these tools.

00:59:49 ◼   ► the work that is going to be reduced by this

00:59:53 ◼   ► is gonna be stuff like the crappy client saying,

00:59:56 ◼   ► "Hey, can you show me 50 different versions of my logo?"

00:59:59 ◼   ► Like, maybe they can skip that step

01:00:01 ◼   ► and move on to more interesting things.

01:00:03 ◼   ► And again, that's not gonna take everyone along with them,

01:00:05 ◼   ► but I don't see these tools as like a universal bad

01:00:09 ◼   ► or like a doomsday scenario for human-created artwork.

01:00:13 ◼   ► Quite the opposite, it's just new tools for humans to use.

01:00:16 ◼   ► - I think of the crappy state these things are now,

01:00:18 ◼   ► like the relatively affirmative state.

01:00:20 ◼   ► I think now is the time that is the most rich for artists to potentially have legal action

01:00:26 ◼   ► against them because it's very difficult to tell without sort of knowing, you know, I

01:00:30 ◼   ► know this is not how they work, but imagine if you could ask one of them, "Okay, so you

01:00:34 ◼   ► made an image for me.

01:00:35 ◼   ► Can you tell me what images contributed to this image that you made?"

01:00:38 ◼   ► And again, that's not how that works.

01:00:39 ◼   ► They don't just take five images and smush them together, but like big picture wise in

01:00:43 ◼   ► the abstract, lots of millions of images in input and then output, right?

01:00:46 ◼   ► And sometimes I can imagine that these more primitive, very early versions of this produce

01:00:53 ◼   ► a work where you could overlay an actual existing thing from its corpus on a section of it and

01:00:58 ◼   ► say, "Okay, this is literally just lifted.

01:01:00 ◼   ► It's smushed and smoothed a little bit, but literally I drew this slice of pizza."

01:01:04 ◼   ► And you put it in the image and you rotated it and scaled it.

01:01:07 ◼   ► And that's a no-go.

01:01:10 ◼   ► If you did that with, if you did like a, you know, a cover of a magazine and you did it

01:01:16 ◼   ► by like stealing the cover of a different magazine and just, you know, cropping out

01:01:20 ◼   ► everything except for the slice of pizza, like draw your own slice of pizza, right?

01:01:24 ◼   ► There is a line to be drawn there like, "Oh, I was doing a collage or whatever," but that's

01:01:27 ◼   ► what these legal cases are about.

01:01:29 ◼   ► And you know, there's a whole other thing to be said about the sad state of legal cases

01:01:32 ◼   ► on songs that are identical, but for artwork, you could say, "Oh, I was doing a collage,

01:01:36 ◼   ► it's a derivative work, so on and so forth, versus I just straight-up lifted this pizza

01:01:40 ◼   ► slice from this other artist thing. It didn't change it enough for it to be legally distinct.

01:01:45 ◼   ► As these things get better, there will be less of that, less chance of that happening, that

01:01:49 ◼   ► it really will be all new, totally fresh work. But part of that relies on a big sort of leveling up

01:01:57 ◼   ► of these things in understanding literally anything. What they understand now is so limited.

01:02:03 ◼   ► You could say, "These things don't know what a slice of pizza is." Well, they kind of do,

01:02:06 ◼   ► because of all the pictures of slices of pizza and the fact that they could say

01:02:09 ◼   ► that triangle thing is probably the pizza slice because I have a hundred

01:02:12 ◼   ► thousand examples of it right? I mean in all fairness like most of America doesn't

01:02:15 ◼   ► know what a slice of pizza is. Right exactly but they don't actually know

01:02:18 ◼   ► they may be able to pick out the thing that corresponds to pizza from an image

01:02:23 ◼   ► but they have no idea what pizza is you know could they do something like draw

01:02:29 ◼   ► heat lines coming off a pizza only because they have existing artwork that

01:02:32 ◼   ► shows heat lines but they don't know pizza as hot they don't know what hot is

01:02:34 ◼   ► They don't know what pizza is. They don't want food is like they're again

01:02:37 ◼   ► They're they're uh, they're so incredibly dumb that even even if they are synthesizing new images in the same way that our brains synthesize

01:02:45 ◼   ► New images our brains have so much other stuff that informs the thing that we're making which is why

01:02:50 ◼   ► You're saying we're gonna you need a human to guide this because these things know nothing and so you can't even guide them

01:02:55 ◼   ► To do things that require them to have literally any understanding of anything that they're doing right, you know, like

01:03:01 ◼   ► Could you put more place settings at that table?

01:03:04 ◼   ► Maybe if they have images that say place setting

01:03:07 ◼   ► and there's different numbers of them,

01:03:08 ◼   ► then I could adapt, but they don't know what a table is.

01:03:09 ◼   ► They don't know what a place setting is.

01:03:10 ◼   ► They don't know what people are or that they sit at tables

01:03:13 ◼   ► and like future versions of this will be better

01:03:15 ◼   ► in that regard and then they will be much better tools

01:03:18 ◼   ► because they have to have some kind of understanding.

01:03:21 ◼   ► In fact, probably in very specialized areas,

01:03:23 ◼   ► they'll gain that understanding,

01:03:24 ◼   ► but getting computers to understand what a person is,

01:03:27 ◼   ► what a table is, what a pizza is,

01:03:28 ◼   ► how they relate to each other,

01:03:30 ◼   ► we've been working on that for decades and decades.

01:03:32 ◼   ► It's way harder than you think it is.

01:03:34 ◼   ► These things look like magic because they're like,

01:03:36 ◼   ► why are we doing that?

01:03:36 ◼   ► It's like trying to make an airplane

01:03:38 ◼   ► by making a thing that flaps its wings.

01:03:39 ◼   ► That's the wrong way to do it.

01:03:40 ◼   ► Even though that's how birds fly,

01:03:42 ◼   ► it's stupid for us to try to make a mechanical bird.

01:03:45 ◼   ► Instead, how about we make a fixed wing thing

01:03:47 ◼   ► and we put a lawnmower engine on it and a propeller

01:03:50 ◼   ► and that's a way better way to make an airplane,

01:03:52 ◼   ► even though it has nothing to do with how birds fly

01:03:55 ◼   ► or a little bit to do with it,

01:03:56 ◼   ► but it's not an ornithopter, right?

01:03:58 ◼   ► These are like that for image generation.

01:04:00 ◼   ► - Wait, what?

01:04:01 ◼   ► Because we can't make we can't make a thing that thinks yet

01:04:05 ◼   ► But we can make some incredibly dumb thing that we feed enough of our intelligence into by saying here's a bunch of images

01:04:12 ◼   ► Here's descriptions of them the words, you know what words are

01:04:14 ◼   ► Well there now do that and we've got just enough to do this magical stuff

01:04:18 ◼   ► But like as a tool

01:04:20 ◼   ► Trying to herd this towards something that you want is even harder than trying to herd an artist or what you want

01:04:25 ◼   ► Because at least you can tell the arcs to make the logo bigger and if they don't do it

01:04:28 ◼   ► It's because they think you're a jerk now because they don't understand what you mean

01:04:30 ◼   ► Ornithopter, a machine designed to achieve flight by means of flapping wings. Today I learned.

01:04:36 ◼   ► Yeah, same. I knew that I knew the word existed. I could not in a million years have told you what

01:04:41 ◼   ► exactly it is. You should know it from the 1994 movie Dune where they all talk about,

01:04:44 ◼   ► "Let's get in the ornithopter," and they get into these things that have wings that do not flap.

01:04:47 ◼   ► Come on. Marco didn't see it. Neither did I. Please don't make me, John. Please don't.

01:04:53 ◼   ► The new Dune movie, those wings flap, baby. Now, the thing that really changed my opinion

01:05:00 ◼   ► about this. Well, I didn't have a strong opinion about it, but what really kind of blew my

01:05:04 ◼   ► mind, I guess, is the three of us are in a Slack together, and another person in that

01:05:09 ◼   ► Slack was saying, oh, and I'm heavily paraphrasing here, but, oh, I was looking at designing

01:05:15 ◼   ► like an app icon or an image, I forget exactly what it was, and I knew a vague direction

01:05:20 ◼   ► of where I wanted to go with it, but I didn't really know specifically what I wanted to

01:05:24 ◼   ► do. This wasn't the case, but like let's say for the sake of discussion they were trying

01:05:27 ◼   ► to draw a settings icon and they knew they wanted a gear, but they didn't know

01:05:32 ◼   ► do I have a gear or like several concentric gears, do I have a series of

01:05:36 ◼   ► gears all touching each other on the outside, like you know what exactly

01:05:40 ◼   ► what am I looking for here? I just know I want something with gears and they

01:05:43 ◼   ► said they put basically app icon with gears in it or something along those

01:05:47 ◼   ► lines into one of these projects and it spit out you know like 15 different

01:05:51 ◼   ► And this person didn't really love any one of the options, but they said that it did a really good job of getting their creative juices flowing, saying, "Okay, now I have something to work with. Now I know where I want to go with this."

01:06:04 ◼   ► And this is what you were alluding to earlier, Marco. I just think something like that, having this kind of fascinating tool in your tool belt, is extremely cool.

01:06:13 ◼   ► And as someone who can barely draw a stick figure,

01:06:17 ◼   ► I think being able, especially as these things get better,

01:06:20 ◼   ► being able to, I don't know, like make my own app icon potentially,

01:06:24 ◼   ► or get close.

01:06:26 ◼   ► You know, not that I have any problems with the app icons I have.

01:06:28 ◼   ► I love them, and I got them from a dear friend of mine.

01:06:32 ◼   ► But nevertheless, it would be neat if I was capable

01:06:35 ◼   ► of even putting like a placeholder icon there that wasn't utter garbage.

01:06:39 ◼   ► and I just think having this tool available to more people,

01:06:44 ◼   ► especially non-artists, I think that's neat.

01:06:47 ◼   ► And the thing that gives me pause about it is,

01:06:50 ◼   ► well, what happens to artists?

01:06:52 ◼   ► In the same way that I worry about GitHub co-pilot

01:06:55 ◼   ► or whatever it is, and I worry, well, what happens to us?

01:06:58 ◼   ► What happens to developers?

01:07:00 ◼   ► - Don't worry about that.

01:07:01 ◼   ► - I know, but you get my point, though.

01:07:03 ◼   ► - Well, and in the same way,

01:07:04 ◼   ► co-pilot is basically fancy autocomplete.

01:07:07 ◼   ► And I think we can look, like, this is basically, like,

01:07:10 ◼   ► fancy bucket fill.

01:07:12 ◼   ► - Fair.

01:07:13 ◼   ► - Well, but the difference is,

01:07:14 ◼   ► not to go off on the tangent on Copilot,

01:07:16 ◼   ► bucket fill, to determine whether it has done

01:07:19 ◼   ► its job adequately, you look at it and go,

01:07:21 ◼   ► is it okay for what I want it to be?

01:07:22 ◼   ► Fine.

01:07:23 ◼   ► Code, not quite that easy.

01:07:26 ◼   ► 'Cause if we could look at code and figure

01:07:27 ◼   ► and know whether it was doing what we intended it,

01:07:29 ◼   ► we wouldn't have to use Copilot.

01:07:31 ◼   ► So Copilot will generate some code,

01:07:33 ◼   ► does the code do what you want it to do?

01:07:35 ◼   ► Why don't you look at it and tell me?

01:07:37 ◼   ► That turns out to be really, really hard to do.

01:07:40 ◼   ► So I don't think we have, because in the end,

01:07:43 ◼   ► you can use AI art things to full stop substitute

01:07:47 ◼   ► for a thing that a person could do.

01:07:49 ◼   ► But copilot, you need a human to look at that

01:07:53 ◼   ► before you check it into the air traffic control system,

01:07:56 ◼   ► let's say.

01:07:57 ◼   ► Maybe for a game, you can get away with it or something,

01:07:59 ◼   ► or non-information critical,

01:08:00 ◼   ► but a copilot has no idea what it's doing,

01:08:03 ◼   ► even more so than a person.

01:08:05 ◼   ► So we need people to look at it and to check that what,

01:08:08 ◼   ► same way when you use autocomplete.

01:08:10 ◼   ► - But John, wouldn't your unit test catch any problem?

01:08:12 ◼   ► - Yeah, if you autocomplete and instead of--

01:08:15 ◼   ► - Oh, can it write my unit test for me?

01:08:16 ◼   ► - You wish. - Maybe, it probably can.

01:08:18 ◼   ► But are the unit tests right?

01:08:20 ◼   ► Like in the same way when you do autocompleting,

01:08:21 ◼   ► you think you're autocompleting NSString

01:08:23 ◼   ► but in autocompletes, what was the thing

01:08:24 ◼   ► that used to do NSSet or whatever before Xcode?

01:08:26 ◼   ► - It was much less common, like NSScanner

01:08:30 ◼   ► or something like that.

01:08:31 ◼   ► - Yeah, whatever came first alphabetically.

01:08:32 ◼   ► If you don't notice that's what came out of autocomplete,

01:08:34 ◼   ► Guess what? Your program's not gonna work.

01:08:35 ◼   ► You always have to look at the code that is doing it.

01:08:38 ◼   ► Not like it's a help.

01:08:39 ◼   ► It's like, you know, content-aware fill

01:08:41 ◼   ► to really help you, especially on programming interviews.

01:08:43 ◼   ► You should say, "Can I use Copilot?

01:08:44 ◼   ► Great, now I'm gonna reverse this red black tree for you."

01:08:47 ◼   ► But yeah, you gotta check its work.

01:08:50 ◼   ► But for the AI things,

01:08:51 ◼   ► the checking of the work is much simpler.

01:08:53 ◼   ► You look at it and you decide,

01:08:54 ◼   ► "Am I happy with what it has made?"

01:08:57 ◼   ► - Yeah, yeah.

01:08:57 ◼   ► I just, I fear and feel for artists

01:09:01 ◼   ► that, you know, maybe wouldn't be able

01:09:03 ◼   ► to make a living as artists anymore,

01:09:05 ◼   ► but I agree with what you were saying,

01:09:07 ◼   ► that that's quite a ways in the future,

01:09:09 ◼   ► and we're nowhere near there yet.

01:09:11 ◼   ► And eventually, the only thing that's inevitable is change.

01:09:15 ◼   ► If you're a developer that's getting your job usurped

01:09:18 ◼   ► by artificial intelligence,

01:09:19 ◼   ► then you're gonna have to find a new way to make money,

01:09:21 ◼   ► and same thing with an artist.

01:09:22 ◼   ► But I don't know.

01:09:24 ◼   ► My initial reaction was, get off my lawn, this is barbaric,

01:09:28 ◼   ► we can't take from real and true artists,

01:09:31 ◼   ► But I worry that as I get older,

01:09:34 ◼   ► that's my natural reaction is to just yell,

01:09:36 ◼   ► "Get off my lawn."

01:09:36 ◼   ► And so I'm trying very desperately to fight that.

01:09:39 ◼   ► And I think having this tool available,

01:09:41 ◼   ► especially to people like me,

01:09:42 ◼   ► who I have no artistic ability whatsoever,

01:09:45 ◼   ► I think it's exciting and I think it's very fascinating.

01:09:47 ◼   ► - So the problem with having, well, first of all,

01:09:49 ◼   ► just to address what you said there,

01:09:50 ◼   ► I totally agree with Marco that this is yet another tool

01:09:53 ◼   ► that will be used by artists.

01:09:54 ◼   ► Like Content-Aware Phil is the best example.

01:09:56 ◼   ► They didn't use AI to advertise that

01:09:58 ◼   ► because it wasn't in FAD at the time,

01:09:59 ◼   ► but, and it works slightly differently,

01:10:01 ◼   ► but content-aware fill powered by this type of thing

01:10:04 ◼   ► gets even better.

01:10:06 ◼   ► It's a tool that artists will use.

01:10:07 ◼   ► And the mundane tasks that artists do,

01:10:10 ◼   ► for example, for my childhood,

01:10:13 ◼   ► painting cells in Disney animation,

01:10:15 ◼   ► like coloring in like the people's dresses

01:10:17 ◼   ► and making the sky blue and the grass green and everything,

01:10:19 ◼   ► that used to be a job where you would paint

01:10:22 ◼   ► to fill those regions,

01:10:23 ◼   ► 'cause how else are you gonna make something

01:10:25 ◼   ► filled with the color green

01:10:25 ◼   ► if you don't fill it with the color green?

01:10:27 ◼   ► Computers made that real easy.

01:10:29 ◼   ► you just click the bucket tool and look,

01:10:30 ◼   ► it just filled the whole area with green.

01:10:32 ◼   ► That put all the people who were painting

01:10:34 ◼   ► those cells out of a job.

01:10:35 ◼   ► You're like, oh, they were just doing a mundane test.

01:10:37 ◼   ► That was incredibly skilled work.

01:10:39 ◼   ► It's only mundane for the computer to do it

01:10:41 ◼   ► because very often the strengths of computers

01:10:44 ◼   ► are the exact opposite of the strengths

01:10:45 ◼   ► and weaknesses of humans.

01:10:47 ◼   ► A computer finds it really easy to fill a region,

01:10:49 ◼   ► especially if that region is correctly contained,

01:10:51 ◼   ► with a solid color, whereas a human

01:10:54 ◼   ► has to carefully control a brush.

01:10:55 ◼   ► And so you would say the person doing that work

01:10:58 ◼   ► is incredibly skilled and the computer doing that work

01:11:00 ◼   ► is as dumb as rocks and that often is the case.

01:11:03 ◼   ► But it doesn't change the fact that they're out of work

01:11:04 ◼   ► because now the computer just does the fill

01:11:06 ◼   ► on all that stuff.

01:11:07 ◼   ► Same thing for hand-drawn animation

01:11:09 ◼   ► in the age of 3D animation.

01:11:10 ◼   ► Doing 3D animation is incredibly difficult.

01:11:12 ◼   ► There are incredibly skilled artists that do that.

01:11:14 ◼   ► They have to have all the skills of traditional artists

01:11:16 ◼   ► on top of computer skills.

01:11:17 ◼   ► But if you are a 2D artist and you don't know

01:11:20 ◼   ► how to use computers and don't care to learn,

01:11:22 ◼   ► you've got a problem.

01:11:23 ◼   ► In fact, if you watch the Disney Plus,

01:11:24 ◼   ► on Disney Plus there's an ILM documentary,

01:11:26 ◼   ► what is it called?

01:11:27 ◼   ► I don't know, just go to Disney Plus and search for ILM.

01:11:31 ◼   ► And part of that documentary is seeing what the advent

01:11:34 ◼   ► of computer technology did to Industrial Light and Magic,

01:11:36 ◼   ► to the people who were there, like the model makers

01:11:38 ◼   ► and the creature shop people or whatever.

01:11:40 ◼   ► Like if technology comes slow enough, people die and retire

01:11:44 ◼   ► and then the new generation does new tech.

01:11:46 ◼   ► But if it comes fast enough, people actually end up

01:11:48 ◼   ► getting booted out of their jobs

01:11:49 ◼   ► or have to learn new skills.

01:11:50 ◼   ► And that's just part of the world.

01:11:51 ◼   ► But I think mostly this stuff,

01:11:54 ◼   ► even at the rate it's developing,

01:11:55 ◼   ► people think, oh, it's going so fast.

01:11:57 ◼   ► next week you'll be able to make a feature-length movie by just writing a phrase like no you won't

01:12:00 ◼   ► because to get to Casey's earlier or a later point um you like the fact that you don't have artistic

01:12:05 ◼   ► skill but you can just ask this thing to make uh you know a picture now i'm gonna sort of uh show

01:12:11 ◼   ► the uh counterpoint to my earlier point about it's easier to tell whether you're happy with the

01:12:17 ◼   ► picture than to tell whether the code copilot generated does what you wanted to do part of

01:12:22 ◼   ► making that decision so let's say i have no artistic taste but now i can just make it make

01:12:26 ◼   ► make the image for you.

01:12:27 ◼   ► If you have no artistic taste and no artistic skill,

01:12:30 ◼   ► your ability to judge whether what it generated is good or not

01:12:33 ◼   ► is also impaired.

01:12:34 ◼   ► [LAUGHTER]

01:12:34 ◼   ► Fair.

01:12:35 ◼   ► Fair.

01:12:36 ◼   ► Like, yes, being able to do it is one skill,

01:12:39 ◼   ► but also having the taste to know this is the good icon

01:12:42 ◼   ► with the gears versus this is the bad icon for the gears

01:12:44 ◼   ► is itself an artistic skill.

01:12:46 ◼   ► And just because a computer made you 50 of them

01:12:48 ◼   ► and you get to pick one, the picking of the one

01:12:50 ◼   ► is the skill, which is why you need

01:12:52 ◼   ► artists to use these tools.

01:12:54 ◼   ► Content-Aware Philo is available

01:12:55 ◼   ► on all our copies of Photoshop,

01:12:57 ◼   ► and yet if we use Photoshop,

01:12:58 ◼   ► we can't do what a great artist can do with Photoshop

01:13:00 ◼   ► 'cause we're not great Photoshop artists, right?

01:13:02 ◼   ► And I think picking, you know, like,

01:13:04 ◼   ► just look at anything,

01:13:05 ◼   ► anything that requires any kind of, you know, taste.

01:13:08 ◼   ► Like, even if you're presented with a thousand options,

01:13:10 ◼   ► if you don't know which one is actually better

01:13:12 ◼   ► or the one you pick is not the one

01:13:14 ◼   ► that the world thinks is good,

01:13:15 ◼   ► and, you know, for reasons that you don't understand why,

01:13:18 ◼   ► it's not pleasing, but you don't like any of them,

01:13:20 ◼   ► or the one you think is awesome everyone else thinks is ugly,

01:13:22 ◼   ► Like there's always a place for that because again,

01:13:25 ◼   ► the things that are generating this have no awareness

01:13:29 ◼   ► of anything, they're just being led by us.

01:13:32 ◼   ► And so like any tool, the result is going to be

01:13:37 ◼   ► heavily informed by the person using the tool,

01:13:40 ◼   ► even if the quote using of the tool is just pointing

01:13:43 ◼   ► to a grid of pictures and saying, I like that one.

01:13:45 ◼   ► And even if you repeat that process a thousand times,

01:13:47 ◼   ► give me 50 more, give me 50 more, give me 50 more,

01:13:49 ◼   ► and just keep pointing to the ones that you like.

01:13:51 ◼   ► If you have bad taste, you will end up with a bad icon

01:13:54 ◼   ► at the end of it no matter what.

01:13:56 ◼   ► No computer is gonna save you.

01:13:58 ◼   ► And that gets back to what is it that makes good taste?

01:14:02 ◼   ► Are all the computers doing what we do?

01:14:06 ◼   ► And I have to say that like most things in AI,

01:14:09 ◼   ► the answer is no until their experience of their existence

01:14:14 ◼   ► is something close to what our experience is,

01:14:17 ◼   ► which would allow them to learn things

01:14:20 ◼   ► and have memories and experience life the same way we do,

01:14:25 ◼   ► something that doesn't do that will never be able to create

01:14:29 ◼   ► or judge art in the same way that we do.

01:14:31 ◼   ► So we will always be a necessary ingredient in that stew.

01:14:34 ◼   ► - We are brought to you this week by Squarespace,

01:14:39 ◼   ► the all-in-one platform for building your brand

01:14:41 ◼   ► and growing your business online.

01:14:43 ◼   ► With Squarespace, you can stand out

01:14:45 ◼   ► with a beautiful website, engage with your audience,

01:14:48 ◼   ► whatever you need to, and sell whatever you want to sell.

01:14:51 ◼   ► Your products, your content, even your time.

01:14:54 ◼   ► Squarespace makes it super easy to make all kinds of websites

01:14:58 ◼   ► from very simple, basic informational sites,

01:15:01 ◼   ► all the way up to full blown storefront hosting.

01:15:04 ◼   ► They support it all.

01:15:05 ◼   ► You can sell your products in an online store,

01:15:07 ◼   ► whether you sell physical or digital products.

01:15:09 ◼   ► Squarespace has all the tools you need to sell online.

01:15:13 ◼   ► You can have email campaigns with your customer.

01:15:15 ◼   ► Click email subscribers, convert them into loyal customers.

01:15:18 ◼   ► Start with an email template and customize it

01:15:20 ◼   ► by applying your brand ingredients,

01:15:22 ◼   ► like your site colors or your logo.

01:15:23 ◼   ► You've built an analytics measuring the impact

01:15:25 ◼   ► of every email sent.

01:15:27 ◼   ► All this comes with SEO tools so that your website

01:15:30 ◼   ► or online store has a suite of integrated features

01:15:33 ◼   ► and useful guides to help maximize prominence

01:15:35 ◼   ► among search results.

01:15:36 ◼   ► Everything you need to run a business.

01:15:37 ◼   ► Analytics, member areas if you wanna sell member content.

01:15:41 ◼   ► So much is available at Squarespace

01:15:43 ◼   ► and it is all backed by amazing support,

01:15:46 ◼   ► amazing infrastructure, and just a whole bunch of stuff

01:15:48 ◼   ► that you don't have to deal with.

01:15:49 ◼   ► You get to do whatever it is you do with your business

01:15:52 ◼   ► or your hobby or whatever your pleasure is

01:15:54 ◼   ► for making your site, you get to focus on that

01:15:56 ◼   ► and let Squarespace handle all of the boring

01:15:58 ◼   ► and tedious hosting and maintenance kind of stuff.

01:16:01 ◼   ► So, see for yourself at squarespace.com/ATP

01:16:05 ◼   ► to start a free trial.

01:16:06 ◼   ► When you're ready to sign up and launch your site,

01:16:08 ◼   ► use offer code ATP to save 10% off your first purchase

01:16:12 ◼   ► of a website or domain.

01:16:13 ◼   ► Once again, squarespace.com/ATP to start that free trial

01:16:17 ◼   ► and use offer code ATP at launch

01:16:18 ◼   ► to save 10% off your first purchase.

01:16:20 ◼   ► Thank you so much to Squarespace for sponsoring our show.

01:16:24 ◼   ► - All right, let's do some Ask ATP

01:16:29 ◼   ► and let's start with Philip who writes,

01:16:32 ◼   ► I recently got a MacBook and I'm struggling

01:16:34 ◼   ► to build a solid mental model

01:16:35 ◼   ► for window management on macOS.

01:16:37 ◼   ► I've been using Linux with a tiling window manager

01:16:39 ◼   ► and things felt simpler.

01:16:40 ◼   ► I'm not trying to replicate this setup

01:16:42 ◼   ► and want to learn the quote unquote Mac way,

01:16:44 ◼   ► but I can't seem to grok it.

01:16:45 ◼   ► I'm not sure which features I should be using

01:16:48 ◼   ► between mission control spaces,

01:16:49 ◼   ► application switcher, hide, minimize,

01:16:50 ◼   ► full screen with tiling, hot corners, et cetera.

01:16:53 ◼   ► And I often end up with a lot of window clutter

01:16:55 ◼   ► where I can't even seem to find the one window I need.

01:16:57 ◼   ► Can you refer to a primer on macOS window management?

01:16:59 ◼   ► How do you think about and organize

01:17:01 ◼   ► your applications in Windows?

01:17:02 ◼   ► I've listened to the Windows of Syracuse County,

01:17:04 ◼   ► but can't tell if Jon is messing about.

01:17:07 ◼   ► No, he was not messing about,

01:17:08 ◼   ► which is why Marco and I still five, six, seven years later

01:17:11 ◼   ► are gobsmacked by that episode.

01:17:13 ◼   ► I think it's episode 96, if I'm not mistaken.

01:17:16 ◼   ► That half-hour, 45-minute segment of ATP

01:17:19 ◼   ► might, to this day, be my favorite part,

01:17:21 ◼   ► or favorite segment we've ever done on ATP,

01:17:23 ◼   ► because we didn't know it was coming.

01:17:25 ◼   ► And I don't wanna speak for Marco,

01:17:26 ◼   ► but I'm gonna speak for Marco in saying

01:17:28 ◼   ► it was flabbergasting, like just stupefying

01:17:32 ◼   ► the absolute bananas way in which John Siracus

01:17:36 ◼   ► manages his seven trillion windows.

01:17:38 ◼   ► - And yet, what have you learned since then?

01:17:39 ◼   ► - Nothing.

01:17:40 ◼   ► - Have you amended your ways?

01:17:41 ◼   ► No, you have not.

01:17:42 ◼   ► - Because I'm not a monster that keeps

01:17:44 ◼   ► the ED5 building Windows open.

01:17:45 ◼   ► - No, you're just a little baby.

01:17:46 ◼   ► - Right, exactly.

01:17:48 ◼   ► - So getting to Phillip's question,

01:17:50 ◼   ► what is the Mac way to manage your Windows?

01:17:52 ◼   ► I think you two, at this point,

01:17:53 ◼   ► are better equipped than me to answer this,

01:17:55 ◼   ► because I think the answer is just do what Windows people do.

01:17:57 ◼   ► Zoom everything to full screen,

01:17:58 ◼   ► 'cause you have no freaking idea how to deal with Windows.

01:17:59 ◼   ► - No. - No, no, no, no.

01:18:02 ◼   ► - The great thing about macOS window management

01:18:05 ◼   ► is that macOS 10 started out trying to get people

01:18:10 ◼   ► who used Mac OS 9 and earlier to like it.

01:18:13 ◼   ► - I don't know about that.

01:18:14 ◼   ► I feel like it did the opposite of that.

01:18:15 ◼   ► - I didn't say it succeeded.

01:18:17 ◼   ► - I'm not sure it really even tried,

01:18:19 ◼   ► but I get what you're saying.

01:18:20 ◼   ► There was some acknowledgement that there was something

01:18:23 ◼   ► that existed before the Mac,

01:18:24 ◼   ► but it was a grudging acknowledgement.

01:18:26 ◼   ► - Right, so anyway, as it went on,

01:18:29 ◼   ► it later took a larger role in trying

01:18:32 ◼   ► to get Windows people to like it,

01:18:34 ◼   ► and then later on, it took a larger role

01:18:37 ◼   ► in trying to get iOS people to like it.

01:18:40 ◼   ► And now it's in this weird mishmash

01:18:42 ◼   ► where they're trying to move it forward with iOS

01:18:45 ◼   ► and the iPad somehow.

01:18:47 ◼   ► And so the result of all of this

01:18:49 ◼   ► is that there are a million different ways

01:18:52 ◼   ► to manage Windows on macOS.

01:18:54 ◼   ► It basically ended up with this mishmash

01:18:56 ◼   ► where it kind of supports all of these different things

01:18:58 ◼   ► you might wanna do.

01:19:00 ◼   ► So what I would recommend is basically

01:19:04 ◼   ► play with different options and just see what works for you.

01:19:07 ◼   ► Now, I can tell you what I do,

01:19:09 ◼   ► 'cause I was a Windows person until 2004-ish,

01:19:12 ◼   ► when I began a two year transition

01:19:17 ◼   ► to Mac full-time at that point.

01:19:20 ◼   ► But anyway, what I do, first of all, hide versus minimize.

01:19:24 ◼   ► You want to hide.

01:19:26 ◼   ► Command-H is your new best friend.

01:19:28 ◼   ► As a new person on Mac, you're gonna want

01:19:31 ◼   ► all the keyboard shortcuts to make this stuff easier,

01:19:33 ◼   ► and Command-H is gonna be one of the things

01:19:35 ◼   ► you use the most, up there with Command + Q for quit.

01:19:38 ◼   ► - I almost never hide Windows, almost never.

01:19:39 ◼   ► - Oh no.

01:19:41 ◼   ► Anyway, it might be new to you, the fact that on the Mac,

01:19:44 ◼   ► an app can have no Windows but still be running.

01:19:47 ◼   ► So when you're done with an app, Command + Q.

01:19:50 ◼   ► If you wanna close a window or tab, Command + W.

01:19:53 ◼   ► And then Command + H for hide.

01:19:54 ◼   ► These are things, other, Windows has Alt + Tab,

01:19:58 ◼   ► we have Command + Tab, it's in the same position

01:20:00 ◼   ► on the keyboard even.

01:20:02 ◼   ► We also have, since, you know, Alt + Tab on Windows

01:20:04 ◼   ► goes between different windows all individually.

01:20:09 ◼   ► On the Mac, Command Tab goes through apps,

01:20:12 ◼   ► much to John's chagrin in certain cases

01:20:14 ◼   ► the way it does this, but it goes through apps

01:20:16 ◼   ► and then Command Tilde, the button right above the Tab key,

01:20:19 ◼   ► on the US keyboard at least,

01:20:20 ◼   ► goes between different windows of the app you're currently in

01:20:23 ◼   ► so again, these are kind of like training wheels

01:20:26 ◼   ► and getting into Mac window management from other systems,

01:20:29 ◼   ► most likely windows.

01:20:31 ◼   ► So those I think are the main entry points.

01:20:35 ◼   ► And then whether you like maximize to full screen

01:20:38 ◼   ► like John just accused Casey of,

01:20:40 ◼   ► which Mac OS makes kind of difficult.

01:20:42 ◼   ► Whether you even use full screen mode

01:20:44 ◼   ► or just make Windows big, that's up to you.

01:20:47 ◼   ► On those things, I make Windows big when they need to be

01:20:50 ◼   ► but not when they don't.

01:20:51 ◼   ► I don't use full screen mode on anything

01:20:53 ◼   ► because it sucks in many other ways.

01:20:55 ◼   ► And then spaces, well that's virtual desktops.

01:20:59 ◼   ► every windowing system has that.

01:21:01 ◼   ► If you use spaces in your previous systems,

01:21:03 ◼   ► you might need it here, you might not.

01:21:05 ◼   ► It's up to you.

01:21:06 ◼   ► You can try it, feel free, no one's gonna bother you.

01:21:09 ◼   ► The things like mission control,

01:21:12 ◼   ► or what used to be called expose,

01:21:13 ◼   ► where you kinda like zoom all the windows out at once

01:21:16 ◼   ► and do stuff with them.

01:21:17 ◼   ► I don't really do that.

01:21:19 ◼   ► I used to, like when it was expose back in the early days,

01:21:22 ◼   ► I don't really do that now.

01:21:24 ◼   ► The one thing that I do a lot is the F11 to show desktop,

01:21:28 ◼   ► where you hit F11 or whatever the desktop key is

01:21:30 ◼   ► on the Apple keyboard,

01:21:32 ◼   ► and it shoves all the windows out to the sides

01:21:36 ◼   ► and exposes the desktop where I keep,

01:21:37 ◼   ► yes, files I'm working on.

01:21:40 ◼   ► That's a whole thing.

01:21:41 ◼   ► I do it, everyone does it, who cares?

01:21:43 ◼   ► So I oftentimes will zoom all the windows out,

01:21:46 ◼   ► grab a file on the desktop, hit F11 again

01:21:49 ◼   ► to bring all the windows back in,

01:21:51 ◼   ► and then drop that file onto something I'm working on,

01:21:54 ◼   ► onto a window of an app or whatever.

01:21:55 ◼   ► - Marco, you're showing your keyboard preferences here,

01:21:59 ◼   ► because F11 isn't what you're describing.

01:22:02 ◼   ► It's volume down.

01:22:03 ◼   ► - It is when you're using a non-Apple keyboard.

01:22:04 ◼   ► - Yes, but for everyone else in the world

01:22:07 ◼   ► that uses a Mac, it's volume down.

01:22:09 ◼   ► I understand what you're going for,

01:22:10 ◼   ► but your preference for keyboards is coming through here.

01:22:14 ◼   ► - Yeah, well anyway, so yeah, so basically,

01:22:17 ◼   ► that's the kind of like the basics that I like to do,

01:22:20 ◼   ► but whether you use all this stuff,

01:22:22 ◼   ► like the mission control and all that stuff,

01:22:25 ◼   ► It's really personal preference, and again,

01:22:27 ◼   ► because of the history of Mac OS

01:22:30 ◼   ► and the jumbled design leadership it has had over time

01:22:33 ◼   ► and the very different targets it has tried

01:22:35 ◼   ► to attract people from over time,

01:22:37 ◼   ► it kind of offers all of these things.

01:22:40 ◼   ► And now we're even gonna have Stage Manager

01:22:42 ◼   ► if it ships in Ventura, and that's its own whole other thing

01:22:47 ◼   ► that frankly I don't think is working very well so far.

01:22:51 ◼   ► But that's its own thing.

01:22:53 ◼   ► Try it out, see what works for you.

01:22:54 ◼   ► If you ask any two Mac users,

01:22:56 ◼   ► you're gonna get two different answers

01:22:58 ◼   ► because there are so many methods

01:22:59 ◼   ► and Mac users are largely,

01:23:01 ◼   ► at least we used to be power users.

01:23:04 ◼   ► And so everyone has their own certain workflows

01:23:07 ◼   ► and quirks and habits and preferences

01:23:09 ◼   ► and they're all gonna be a little bit different

01:23:11 ◼   ► 'cause there are so many different ways to do it.

01:23:14 ◼   ► - All right, I don't think either of you

01:23:17 ◼   ► is necessarily wrong,

01:23:17 ◼   ► but I think you've gone directly into the deep end

01:23:20 ◼   ► and I think we need to back up a bit.

01:23:23 ◼   ► So coming from Windows, I have used Linux,

01:23:26 ◼   ► but it's been, oof, it's been a long time.

01:23:29 ◼   ► Coming from Windows, I think the most startling thing

01:23:32 ◼   ► about using the Mac is what you had made brief mention

01:23:37 ◼   ► of earlier, Marco, is that you can have an application

01:23:40 ◼   ► that is running even though it does not have

01:23:43 ◼   ► not a single window open.

01:23:45 ◼   ► This is very different than Microsoft Windows,

01:23:47 ◼   ► or at least the last I used it 10 years ago,

01:23:50 ◼   ► where if you close the final window of Outlook

01:23:52 ◼   ► for the sake of discussion,

01:23:54 ◼   ► you're suddenly not gonna get any new email.

01:23:56 ◼   ► Again, this may not be true anymore, it doesn't matter,

01:23:57 ◼   ► but that's the way it used to be.

01:23:59 ◼   ► If you close the last Outlook window,

01:24:00 ◼   ► you're not getting email anymore.

01:24:01 ◼   ► If you just close, not even necessarily hide,

01:24:04 ◼   ► close the last mail window,

01:24:07 ◼   ► the mail app is still open,

01:24:10 ◼   ► and this isn't universally true,

01:24:12 ◼   ► which has gotten even squishier over time,

01:24:14 ◼   ► but generally speaking, that's true.

01:24:15 ◼   ► If you close the last window,

01:24:17 ◼   ► that does not necessarily mean that the app is quit.

01:24:19 ◼   ► The app could still be running,

01:24:20 ◼   ► and mail is a quintessential example of this,

01:24:22 ◼   ► Safari if you don't have any tabs open for example, they're still running. So what I

01:24:28 ◼   ► would say, and I think what Marco got right, is get used to the keyboard because the keyboard

01:24:34 ◼   ► is your friend for doing windowing things on macOS. You don't have to do it, you don't

01:24:38 ◼   ► have to touch the keyboard at all, but it is your friend. So when you're done looking

01:24:44 ◼   ► at something, Command+W closes that thing, be it a tab or a window. Generally speaking,

01:24:50 ◼   ► You're not quitting the app, usually, even if you close the last window, you're just

01:24:55 ◼   ► closing that window.

01:24:56 ◼   ► So as an example, if I'm looking at my mail using the standard macOS Mail app, after I'm

01:25:02 ◼   ► done reading and responding to mail, I Command+W.

01:25:05 ◼   ► That closes mail, but it does not quit mail.

01:25:08 ◼   ► It leaves it running to get new mail if new mail comes in.

01:25:12 ◼   ► If I want to play ignorant and don't want to get new mail, then I Command+Q for quit,

01:25:18 ◼   ► that will quit mail such that I don't receive new mail.

01:25:22 ◼   ► So the whole close versus quit thing, close being Command-W, quit being Command-Q,

01:25:28 ◼   ► that is something to get used to. And similarly, if you look at the stoplights in the upper left,

01:25:32 ◼   ► the red stoplight is not quit, it is close.

01:25:35 ◼   ► So you are closing an entire window. That maybe that is a window with a bunch of tabs in it,

01:25:41 ◼   ► maybe it's just a single window like in mail, but you're closing it, you're not quitting it.

01:25:44 ◼   ► And there is an option buried somewhere in system preferences, I forget where, that you can have like little light bulbs that show

01:25:51 ◼   ► under the app icon in the dock to indicate what is actually running, which is the things that have a light bulb.

01:25:57 ◼   ► They aren't light bulbs anymore. They used to be.

01:25:59 ◼   ► Well, whatever they are.

01:25:59 ◼   ► Now they're just dots.

01:26:01 ◼   ► You're right, you're right. But now I'm showing my age.

01:26:03 ◼   ► Yeah, and also Mac OS lies about that. It's way more complicated than you think it is.

01:26:07 ◼   ► You're right, you're right. But I'm trying to do it. I'm trying to ease into the shallow end here.

01:26:11 ◼   ► - Yeah, the problem with what we're saying here

01:26:13 ◼   ► is everything that Casey has just said

01:26:16 ◼   ► has a bunch of asterisks on it now,

01:26:18 ◼   ► again, through the course of history

01:26:19 ◼   ► and different goals and different efforts.

01:26:21 ◼   ► - You're right, you're absolutely right.

01:26:22 ◼   ► - But you're right, overall.

01:26:23 ◼   ► Your overall theme is right, but it is,

01:26:25 ◼   ► I think it's part of the reason,

01:26:26 ◼   ► it's a little bit frustrating that the Mac

01:26:29 ◼   ► is not as simple as it once was,

01:26:32 ◼   ► because there have been all these different ideas

01:26:34 ◼   ► and directions and then band-aids over bad designs

01:26:37 ◼   ► over time. (laughs)

01:26:39 ◼   ► - I mean, but the thing is,

01:26:40 ◼   ► First of all, Casey, you're going off on a tangent here,

01:26:42 ◼   ► which is basically about process management

01:26:44 ◼   ► versus window management.

01:26:45 ◼   ► And I see how they're somewhat related.

01:26:46 ◼   ► - Well, they're interrelated.

01:26:47 ◼   ► - Kind of, but detecting the details that we know about

01:26:52 ◼   ► for like, you know, I think I had a big paragraph on it

01:26:55 ◼   ► in one of my iOS 10 reviews that basically you can have

01:26:57 ◼   ► applications with no windows that are running,

01:26:59 ◼   ► applications with windows that aren't running,

01:27:01 ◼   ► applications with a dot under it

01:27:02 ◼   ► in the doc that aren't running,

01:27:03 ◼   ► applications without appearing in the doc that are running.

01:27:05 ◼   ► Like every combination of things

01:27:07 ◼   ► that you think shouldn't be possible are possible.

01:27:09 ◼   ► But that's just what we know from a technical perspective.

01:27:12 ◼   ► What's important is the user model, the mental model.

01:27:15 ◼   ► You're not supposed to know that an application,

01:27:17 ◼   ► the doc that has a dot under it

01:27:18 ◼   ► might not actually be running

01:27:20 ◼   ► because the OS is working to provide the illusion

01:27:23 ◼   ► that if it's got a dot, it's running

01:27:25 ◼   ► and it does everything it can to make that illusion true.

01:27:28 ◼   ► So for example, if the OS has quit an app

01:27:31 ◼   ► but left the dot under it in the doc,

01:27:33 ◼   ► it did that because the app supports

01:27:35 ◼   ► whatever the hell the thing is called.

01:27:36 ◼   ► But when you click it again, behind the scenes,

01:27:39 ◼   ► it relaunches it for you and lets it auto restore the state.

01:27:42 ◼   ► So to you, it just looks like it just brought that app

01:27:44 ◼   ► to the front, which is what it would do if it was running,

01:27:46 ◼   ► but it didn't actually relaunched it, right?

01:27:49 ◼   ► And by the same token, sometimes when you quit an app,

01:27:52 ◼   ► the OS has the option to go,

01:27:53 ◼   ► yeah, I'm not actually gonna quit it.

01:27:55 ◼   ► I'm actually gonna keep it running.

01:27:56 ◼   ► So if you quit an app and the dot disappears

01:27:58 ◼   ► from running the dock, and then you click that icon

01:28:01 ◼   ► on the dock again, you're like,

01:28:02 ◼   ► wow, that launched really fast, you know why?