PodSearch

The Talk Show

421: ‘The Ratchet of Flippancy’, With Craig Mod

 

00:00:00   What are you drinking there?

00:00:00   I've got so many coffees.

00:00:02   I've got normal water, sparkling water.

00:00:04   I've got fancy coffee, hot coffee.

00:00:07   Well, that was hot.

00:00:08   And then I've got fancy cold brew coffee.

00:00:09   I believe, am I correct here, that you have not been on the show in four years?

00:00:15   At least, yeah, four or five.

00:00:17   This is me being a derelict host of a show.

00:00:22   Because I think of Craig Mod as somebody, ah, Craig's on the show once a year.

00:00:27   Yeah.

00:00:28   Like clockwork.

00:00:30   Yeah.

00:00:30   And then I'm like, well, I should double check.

00:00:32   When was the last time he was on?

00:00:33   And then I look, and it says 2021.

00:00:35   And I'm like, did I miss an episode?

00:00:37   Is there something not there?

00:00:39   No, I think that's it, 2021.

00:00:41   Well, let's pretend that I'm not a rude host who didn't invite you back in four years.

00:00:48   Because I would like to have you on all the time.

00:00:50   How are things going?

00:00:52   You are in the middle of a book tour.

00:00:55   I'm in the middle of a book launch.

00:00:57   Yeah, the tour is, yeah, I guess this is like the virtual part of the tour.

00:01:01   Yeah, no, the book, the book launches May 6.

00:01:03   And it's funny, because this is a book, basically, right after we talked last time, I went off on a walk.

00:01:10   That was like 700 kilometers around this peninsula in the middle of COVID.

00:01:13   And during that walk, I wrote a ton, photographed a bunch, did a pop-up newsletter.

00:01:18   We can talk about that stuff if you want later.

00:01:20   And basically, for the last four years, I've been working on this one book, and it's coming out now in two weeks.

00:01:26   So that's sort of what I'm prepping for.

00:01:29   Yeah.

00:01:30   All right, teaser, we'll come back to it.

00:01:32   Think about it.

00:01:33   Do you have any, do you have an agenda of things you want to talk about?

00:01:36   I would like to talk, because you and I chat.

00:01:38   This is what throws me off, is that you and I chat over iMessage all the time.

00:01:44   And that makes me think you're on my show all the time.

00:01:49   But one of the things we often talk about is tools, right?

00:01:54   Yeah.

00:01:54   And sort of tools for the mind sort of things.

00:01:58   And so I'm curious, A, where you stand right now on AI stuff overall.

00:02:05   Overall, I think the generative stuff, like the photo and video and text generation stuff, is weird and fraught.

00:02:16   But I have a hard time parsing out what is moral or amoral about how this stuff is being used, or how the information's been gleaned, or whatever it's been sucked up from.

00:02:25   All that stuff.

00:02:26   It's like obvious, I guess it's been proven that Meta has used a bunch of copyrighted books to train a bunch of their models, right?

00:02:31   So anyway, all that stuff to me feels-

00:02:33   An enormous, an enormous trove of pirated books.

00:02:38   Yeah.

00:02:39   It's like-

00:02:40   And the emails came out that they were more or less, well, we could pay a fortune to license.

00:02:46   Some of these books, but we wouldn't even get as many books as in this pirated trove, or we could just use this pirated trove and deal with the problem later.

00:02:56   And guess which one Meta went with?

00:02:57   Yeah.

00:02:58   Well, and that's what's insane is that I don't even think in the grand scheme of tech it would cost that much money to even license all these books to use.

00:03:05   Like, it'd be whatever, a billion dollars, a couple billion dollars, whatever, just poop that out.

00:03:09   It just comes out, that's in the cushions of the couch.

00:03:12   Just pay it up.

00:03:13   Anyway, it's really weird to me that none of these organizations have been like, hey, what if we made like a really, really above the board thing where everyone was getting compensated properly and yada, yada, all that stuff.

00:03:23   Anyway, so put that aside.

00:03:25   I don't use any of the photo generation crap or whatever.

00:03:28   What I do think is profound, and I feel like it's way more morally defensible, is the code generation stuff, just because of all the open source stuff, yada, yada.

00:03:38   And for me, I've been using little drips and drabs of ChatGPT to do a few things, and then I moved over to Anthropics stuff, and I've been using Cloud.

00:03:46   And then Cloud Code came out a couple months ago.

00:03:49   And for me, Cloud Code has been probably one of the most profound computing experiences of my life, using that thing and building software with it.

00:03:57   I have been just astounded and delighted.

00:04:03   There's a part of me that's like, what am I giving up by not working through some of these problems on my own or understanding the API interface is better on my own?

00:04:10   But at the same time, I'm building software that, for like my membership program, for example, I built a Twitter clone exactly how I wanted Twitter to be like, because I have all these really strong opinions, like these stupid, these stupid verse-cron text things should act in a certain way.

00:04:26   And I was able to build it in the course of a week or two, and I've been adding features onto it, and it's become this nook of the internet for me.

00:04:34   It's a members-only chat space that, to me, because Discord is insane.

00:04:38   Discord is just like this monstrous, it's like the empire.

00:04:42   It's like the death star of software for me.

00:04:45   I think Discord, to me, here's my analogy.

00:04:48   I think you'll like it.

00:04:49   It's not just like entering a casino.

00:04:52   It's like entering a cheesy casino.

00:04:54   Yeah.

00:04:55   And I like a cheesy casino, but you got to be in the right mood to go in there.

00:05:00   And there's certain times, like first thing in the morning, you don't want to go into a bling, bling, bling, bling, bling, bling, bling, bling, do, do, do, you know, casino.

00:05:09   I can't understand how anybody uses it for anything thoughtful.

00:05:15   I find it too distracting.

00:05:17   And I get it.

00:05:17   I think some people have a mind that can tune out distractions.

00:05:22   I think there are people who could have like a serious meeting in the middle of a noisy cheeseball casino with all the slot machines pinging and ponging and beeping and popping and lights going off all around.

00:05:37   And they could just tune it out and have a discussion.

00:05:39   I can't.

00:05:40   Right.

00:05:41   I can't either.

00:05:41   One single distraction in my field of view and my mind is gone.

00:05:47   No.

00:05:47   So I've tried to do some discords.

00:05:49   I'm just like, I bounce off.

00:05:50   It's so hard.

00:05:51   And also, I think there's too much stuff that's synchronous.

00:05:54   We need more async stuff.

00:05:55   We need more stuff that's not real time.

00:05:57   Like the fact that everyone's replying instantly and all that.

00:06:00   Whatever.

00:06:00   We got enough of that in our life.

00:06:02   Don't need any more of that.

00:06:04   The other thing I don't like about Discord, and I think you, again, I think you'll agree with me.

00:06:09   And I don't blame Discord for this.

00:06:12   Discord's trying to build their own thing.

00:06:13   Yeah.

00:06:14   But to me, every time I like some, I don't want to throw anybody under the bus, but I'm a member of a couple of things that have discords you get into if you're a member.

00:06:25   And the problem is, even more so than Slack, it feels like when I'm in Discord, I'm in Discord.

00:06:34   I'm not in Craig's Discord, or if I had one, like the Daring Fireball Discord, or whoever else might have one.

00:06:41   I'm in Discord.

00:06:42   And the highest level room, it all feels too little differentiated.

00:06:49   I just feel like I'm in Discord.

00:06:51   And sometimes I just want to go to a certain place.

00:06:54   Like, oh, I want to go to Craig's little thing on the internet.

00:06:57   And I want to feel like I'm there.

00:07:00   Yeah.

00:07:01   So that's what I built.

00:07:02   I built a place called The Good Place.

00:07:04   And if you add it to your home screen, Ted Danson is the icon.

00:07:07   I don't know if you've seen the show, but it's an amazing show.

00:07:10   And it's just this quiet space that you can post two times a day.

00:07:15   So here are the parameters.

00:07:16   This is what I think the ideal social network looks like.

00:07:18   You can only post twice a day.

00:07:20   Everything disappears after a week.

00:07:22   If things get more, if older things get replies, the newest reply dictates the expiration date.

00:07:28   So you can kind of keep threads alive more than a week if they're active.

00:07:31   But if they aren't active, everything disappears after a week.

00:07:33   And so you get rid of all this stupid archive stuff.

00:07:36   You get rid of, I don't know, there's a kind of a, there's sort of a paranoia or like, there's a pathology I think that a lot of us of our generation had around that everything needs to be archived forever on the internet.

00:07:48   And I kind of don't believe that.

00:07:50   And I think Robin Sloan is also a pretty fervent believer in deleting things.

00:07:54   And he was running delete scripts on Twitter.

00:07:56   I think he was one of the first people I saw doing that.

00:07:58   And then I started doing it.

00:08:00   And it feels really good to not have those weird archives.

00:08:02   And even in the good place, which has been really active, we've got, let's whatever, a few hundred people that post a couple dozen posts a day, maybe a hundred replies a day are going in there.

00:08:11   And it just feels like the right scale and it's super quiet and it's all RSS driven.

00:08:16   That's how you subscribe to it.

00:08:18   So it's all, it's just, it's just nice.

00:08:20   It's just really, it's just nice.

00:08:22   And I couldn't have made that without the LLMs.

00:08:24   I couldn't have made it without CloudCode.

00:08:26   Actually building it with CloudCode has taught me so much about Flask, which is, I'm doing it on Python, using Flask as kind of the framework.

00:08:34   It's taught me a bunch about Flask, about using SQLite, about all these weird little things, these little quirks.

00:08:39   And the thing is with these, with something like CloudCode, if you don't know what you're doing, it's going to make a bunch of garbage.

00:08:44   So you do have to, you have to know, you have to understand the technical underpinnings of things to guide it.

00:08:51   For me, it just feels like I'm project managing and art directing this thing and all of the kind of hyper detailed API related syntax work I'm offloading to this other party.

00:09:02   And that, that feels really good for where I'm at right now.

00:09:04   Anyway, it's been exciting.

00:09:06   Yeah.

00:09:07   There's a lot to unwind there, but I think one of them, and I'm definitely guilty of the, oh, I don't want to delete anything that I've ever put on the internet.

00:09:15   And I've got email that goes back to the nineties, although I, some of it is on hard drives that I don't have plugged in right now.

00:09:24   But I feel like I, I feel like in my mind, I could get that email if I wanted to.

00:09:28   Right.

00:09:29   Right.

00:09:30   And I don't know what informed that to me, it's a very Gen X original internet youth mindset.

00:09:39   I definitely don't speak for my generation, but this particular statement probably doesn't either.

00:09:45   But it is something, something all pre-digital media were so precious or, or, or there's so many stories about old films where they were shot on the silver nitrate negatives and they started dissolving and had to be rescaned.

00:10:05   We've lost a bunch of the, the early major Hollywood motion pictures that the actual negatives evaporated or, or disintegrated newspapers were printed on the, or still are printed on cheap paper.

00:10:21   Right.

00:10:22   And it's so bananas that I grew up at a time where literally they call it fish wrap, right?

00:10:28   You'd buy the paper today.

00:10:29   Here's today's paper.

00:10:30   Here's all the news.

00:10:31   And then you would throw it away.

00:10:34   And if you needed to refer back to last week's newspaper, you'd have to like go to the library and they might have, yeah, well maybe like for a couple of weeks or months, they'd still have physical copies.

00:10:48   But for the most part, once you went back a couple of weeks, it was microfilm, right?

00:10:53   Yep.

00:10:53   And I remember doing that for certain like high school reports and certainly in college looking up stuff where, you know, and some dedicated librarian or thousands of them around the world, but at your local library, somebody's job was to shoot the newspaper onto microfilm every, every day or once a week, take a week of newspapers and shoot them onto microfilm.

00:11:20   And you couldn't search, right?

00:11:22   The search was with your eyeballs.

00:11:24   Yeah, yeah.

00:11:25   And that scarcity of historical data where you'd want to be able to look up where so-and-so had done something in 1958 or whatever, and you wish that you could just search for it the way we expect to be able to search for everything today.

00:11:47   With the internet, it obviously became technically possible to do all of that.

00:11:51   Yeah.

00:11:52   And so therefore, we should do that.

00:11:54   Right.

00:11:55   That's my mindset is it was so hard to do stuff like that up until the point where everything was digital and could be stored forever in a searchable format that once it became obviously technically possible to do it, therefore, we should do it.

00:12:14   End of story.

00:12:15   And that's my mindset.

00:12:18   And I think what you're saying and what you've been thinking about for longer than I have is maybe there's some downsides to that mindset.

00:12:27   Yeah.

00:12:29   Well, I think for some things.

00:12:30   For some things, I think that's the important distinction to make is there's stuff that should be archived and probably stuff that shouldn't.

00:12:35   Again, in real time, how do you tell what should or shouldn't be archived?

00:12:38   But my take is that most social media stuff in aggregate probably doesn't necessarily need to be archived.

00:12:45   So like all of Twitter or whatever.

00:12:46   There are certain elements of Twitter, like moments in Twitter.

00:12:48   Like I remember when Japan had the big earthquake on 311, Twitter was like the repository.

00:12:55   That was where you could kind of re-experience the earthquake in real time for a while.

00:12:59   And I gave a talk where I would show these tweets in real time about how the quake was unfolding.

00:13:04   And I think those sorts of things are pretty special.

00:13:06   But again, like how do you tell like in real time what you should be keeping or what you shouldn't be keeping?

00:13:10   Another thing I think our generation grew up with, which maybe the younger generation doesn't experience as much, is we lost data.

00:13:16   I lost hard drives.

00:13:17   Like I had hard drives crash on me.

00:13:18   I had some of the most traumatic moments of my teenage years.

00:13:23   I mean, there were many traumatic moments of my teenage years.

00:13:25   But some of the most like potently traumatic were like a hard drive failing and then realizing I lost all my NC art or my BBS settings or whatever, something like that.

00:13:34   So I think there's that element, too, where I mean, I don't know about your backup solutions.

00:13:38   I've got hard drives and things all over the fucking world, man.

00:13:42   I've got I've got I've got like Syracuse has talked about this, too, where where we definitely have the digital equivalent of what I thought with my grandparents was the food scarcity mindset.

00:13:55   I was going to say, yeah, it's like the Dust Bowl generation of data hoarding or whatever.

00:14:00   Yeah, I have so many backups.

00:14:03   It's ridiculous, which I see that is different than like keeping everything online.

00:14:08   For me, my big M.O. is make books.

00:14:11   Right.

00:14:11   And so whenever I'm thinking about should these newsletters be archived or should this stuff online be archived, my archive is to take whatever was the best of that set of things.

00:14:22   So if I do a big walk and I'd make a newsletter out of that, that doesn't need to be archived online because ideally I'm going to take that.

00:14:29   I'm going to collimate it.

00:14:30   I'm going to edit it.

00:14:31   I'm going to squeeze it.

00:14:32   I'm going to take the best of it.

00:14:33   I'm going to put the best version of whatever that was into a book and I'm going to distribute thousands of books around the world.

00:14:38   And then that's the archive.

00:14:39   I kind of see that almost as like my forcing function to get me to really look at what I've made digitally and go, okay, what deserves to kind of live on a little bit longer than the bits might live.

00:14:48   There is, again, not to make it political, but it comes up again and again in politics and government affairs like preservation of communications.

00:14:59   And the current affairs political aspect is this goofy use of signal in the Trump administration for defense stuff.

00:15:10   But there is, and God bless his soul, my pal and dithering co-host, Ben Thompson, when this signal thing first came up, he really threw me off.

00:15:22   And we don't, you know, pick a topic and we're like, okay, we'll talk about this.

00:15:26   And then we hit record.

00:15:27   And when he took this angle, it really threw me off mid-episode and it wrong-footed me.

00:15:34   I was like, I can't believe Ben's kind of defending this.

00:15:37   But it was an interesting angle and I think, you know, not to put words in his mouth, but basically grown adults right now expect to be able to text each other conveniently on a device they have with them all the time.

00:15:50   And so shouldn't there be a way that they do this?

00:15:52   And the argument against using signal is, oh, if you're the secretary of defense or you're the secretary of state or any of the people who were in this, they travel with an entourage of professional people.

00:16:07   And if they need to have a secure conversation with other people who are cleared for this stuff, they can set up a, like, I don't know, some kind of temporary secure room with secure communication to other people.

00:16:19   And for some communications, that's obviously what they should be doing.

00:16:22   This stuff should not be on the internet.

00:16:24   But on the other hand, if you're just a grown-ass adult today, don't you expect to be able to text people?

00:16:30   And again, I'm not defending what they texted and the details of it or adding Jeffrey Goldberg, the editor of The Atlantic or whatever.

00:16:39   But there is sort of a, hey, shouldn't we just be able to talk even if we're not in the same room?

00:16:47   We should be able to have a temporary conversation that's not on the record forever, you know?

00:16:54   Yep.

00:16:54   You know, what's the argument against that?

00:16:57   That the only way that people should be able to have a conversation that's not recorded forever is to be in the same physical space?

00:17:07   Right, right.

00:17:09   That doesn't make sense to me.

00:17:10   I mean, I think the weirdest thing about that, what was happening there and, like, putting aside whether they should be doing it on Signal or not, is the kind of, because of the leaks, you saw, I guess, the informality of the language around killing was so, that to me is the most depressing part of it.

00:17:31   And it also just speaks to the fallibility of humans in general.

00:17:34   Like, no one, none of us are in such a position that, like, we should be able to send a really dumb text and be like, YOLO, America first or whatever.

00:17:43   And then suddenly there's a thousand people dead on the other side of the world.

00:17:46   That just seems, that just seems like that's not how the process should unfold.

00:17:49   And that there should be, I love, there's this, I forget, listeners can Google it, but there's this thing where someone proposed that the only way a president should be able to send a nuke is that the nuke codes are embedded in a capsule in the chest of someone.

00:18:06   And they have to cut open the chest and dig them out with their hand to get the code so that they have to experience murder viscerally in order to be able to push a button to kill a million people on the other side of the world.

00:18:17   And I think there is some truth to that.

00:18:18   And I think maybe that's part of what feels so uncomfortable about using a chat app and being, using chat language, too, around war.

00:18:27   And emojis.

00:18:27   Yeah, yeah.

00:18:28   I'm laughing here, but it's, I'm laughing at the absurdity.

00:18:33   But it is, there is, there is a certain, and there, there has been for a very long time.

00:18:38   And I, when everybody thought Nixon was the worst president we ever had, there were all sorts of, once you start abstracting human lives and, and making these policy decisions where you're not really thinking, hey, if we send all these bombs to, and drop all these bombs in this country,

00:19:00   Those are actual people who are being incinerated alive.

00:19:05   Men, women, children.

00:19:06   And, and you just start thinking about it more like you're playing a game of risk, a board game.

00:19:14   I get the argument that from a strategic viewpoint, that is sort of how you plan a war, but you can lose yourself in the abstraction and lose the humanity of what you're doing.

00:19:25   And just communicating in these terms through a commercial app like Signal on your phone just sort of takes it up a level, right?

00:19:38   It's like the ratchet has moved up a level of flippancy on, on, on a subject matter that no matter who the target of the action is, or how deserving or justified the military action might be.

00:19:56   The tone and the tenor and the nature of the medium itself just feels completely inappropriate.

00:20:03   And when people complain about government inefficiency, I mean, the inefficiency is there for a purpose, right?

00:20:10   Which is that we as humans are so fallible.

00:20:13   We're so, so all individually broken.

00:20:16   There is no one who rises above the brokenness of the human condition.

00:20:20   And so you want things to kind of go slowly when decisions like this are being made, you know?

00:20:25   It's like there's a, I would say the existence of people with the amounts of money that folks like Musk or Bezos have is a bug in the system, right?

00:20:34   There's an argument to be made that was unforeseeable that people could accumulate that much wealth.

00:20:39   And the reason why you sort of don't want people to have that much wealth is because we are all so fallible.

00:20:43   There's not a single person out there who you can give essentially the ring of power to or whatever, and then they will definitely consistently operate from a good place, right?

00:20:53   That's not how humans work.

00:20:54   And so anyway, it just feels like the government thing too, when you, when you start to get to these kind of really low stakes abstractions, seemingly low stakes on the point, from the point of view of the people doing the texting or whatever, it just, it just starts to veer into a dangerous zone where you're like,

00:21:09   okay, what, what other things are being decided in such a cavalier way?

00:21:12   Do we want that to be the norm?

00:21:14   Cavalier, haphazard.

00:21:16   Yeah.

00:21:16   I don't know what other adjectives we can throw at it.

00:21:19   Yeah.

00:21:19   Yeah.

00:21:20   All right.

00:21:20   Let me take a break here.

00:21:21   Thank our first sponsor.

00:21:22   It is our good friends at WorkOS.

00:21:25   If you are building a B2B software as a service app, at some point, your customers are going to start asking for enterprise features like SAML, authentication, SCIM,

00:21:38   which is S-C-I-M, but I, it, I always mention it here.

00:21:43   They, they put a note for me who mispronounces everything that they just, if you know what it means, you know, it's called SCIM, SCIM provisioning, role-based access controls, audit trails, et cetera.

00:21:54   WorkOS provides easy to use and flexible APIs that help you ship enterprise features on day one without slowing down your core product development.

00:22:06   It's used by some of the biggest startups in the world today, such as Perplexity, Vercel, Plaid, and Webflow.

00:22:12   WorkOS also provides a generous free tier of up to 1 million monthly active users for its user management solution.

00:22:22   It comes standard with rich features like bot protection, MFA, roles and permissions, and more.

00:22:27   And you do not pay a cent until you hit over a million active customers.

00:22:33   Just a phenomenal deal.

00:22:34   And that's, they're, they're just looking to build companies that are over a million customers.

00:22:40   If you're under a million, give them a try.

00:22:42   If you are currently looking to build out your SSO for your first enterprise customer, you should consider using WorkOS.

00:22:50   Integrate in minutes, start shipping enterprise plans today, and just go to WorkOS.com.

00:22:58   No special slug or anything like that.

00:23:00   No extra thing on the URL.

00:23:02   Just go to WorkOS.com and maybe at some point in the signup process, they'll say, where'd you hear about it?

00:23:09   Tell them the talk show.

00:23:10   Boom.

00:23:11   Boom.

00:23:12   So the coding thing, I want to go back to this, where you've been using, and I've been sticking with chat GPT just because it's like, I don't know where else to go.

00:23:23   And not that I don't know where else to go, but that there's such a plethora of possible solutions.

00:23:29   And I get it that some people think Claude is better at coding, and it might be, but all of it, I feel like there are, well, like everything, there are zealots on both ends.

00:23:41   There are people who are people who are like, this is the future, the whole vibe coding thing.

00:23:45   We don't need, nobody needs to learn programming anymore.

00:23:48   The AIs will code everything for the future.

00:23:53   And there's people on the other side who don't want to say, especially with the programming aspect, I am continuously, each step of the way, and I'm way behind some people, like the people who are just throwing it all in.

00:24:11   I mean, I mean, you've built a whole thing on it, so I'm behind you, too.

00:24:14   But I've been using it more and more for the little script-type things that I write and program, and I'm continuously blown away by it, but I'm also continuously uncomfortable.

00:24:26   Because I've sort of got it in my head from the youngest possible age I can remember, when I first fell in love with computers, that knowing how to program computers is a skill you should be proud of.

00:24:42   And it's something, not that I'm not best known for being a programmer, but I have a computer science degree.

00:24:49   And it's part of my idea of myself, that this is a skill and a mindset that I have, an aptitude that I have, that most people don't have.

00:25:01   I've written about it many, many times on Daring Fireball, and in a particular context of AppleScript in particular, which was this attempt to make, and, you know, HyperCard, which was sort of used a scripting language called HyperTalk that was very similar to AppleScript.

00:25:17   But the idea was to use a more English-like language for computer programming, that that was the problem that was keeping more people.

00:25:27   Like, there was this mindset in the 80s, even into the 90s, more people should be programming.

00:25:33   Programming is something more people should be doing with computers.

00:25:36   And it turns out, I think, most people don't really want to program computers or didn't have the aptitude.

00:25:42   And the ability to do it, again, it sounds, I hate to say it, but it made me feel special.

00:25:50   Yeah, sure.

00:25:52   And now it is.

00:25:53   It turns out that these AI things are way better programmers than I have ever been or could be.

00:26:01   I could practice programming for the rest of my life, or I could have devoted my life to it instead of writing, and I'd never be as fast or as good as any of these things are at computer programming.

00:26:12   And it's very strange, because it used to seem like programming was something way fewer people could do, and it was a better sign of intelligence than verbal language.

00:26:24   But it turns out, no, it's the other way around, right?

00:26:28   Well, yeah, there's a lot to unpack there, too.

00:26:32   I mean, I think there's some analogies to photography and digital photography, right?

00:26:35   So I think a lot of people, when the digital photography shift was happening, felt this as well.

00:26:41   It's like, I've built up all these darkroom skills.

00:26:43   I look at Ansel Adams' exposure techniques and metering and all this stuff and kind of digital photography.

00:26:49   Took that all away because it made the feedback loop instantaneous, right?

00:26:53   So before digital photography, you'd go with your plates, you'd try to meter, you'd try to set things up, try to figure out, calculate, you'd have to do math, basically.

00:27:00   It was pretty onerous to take a good 4x5 or 8x10 photo.

00:27:05   And then it could be weeks later, it could be hours later, it could be whatever, it could be months later that you actually then do the developing, and then you see the result, and you've got a notebook full of little details about what did I do when I was looking at Yosemite's, this thing or that.

00:27:18   And so I think for people who went through that, when digital came around, it was like, oh my God, you're missing out.

00:27:24   You need to kind of struggle through the math of photography, of thinking about exposures like this, yada, yada, yada.

00:27:31   And in the end, did digital photography, well, first of all, it didn't kill film photography.

00:27:36   In fact, film photography now is, it tried to kill film photography, and now film photography is now weirdly stronger than I think it's been in the last 25 years.

00:27:44   Yeah, it's clear.

00:27:46   There's more cameras coming out, you know.

00:27:50   Yeah, yeah, there's new film cameras coming out, but if you talk to anyone under 30 who is a quote-unquote serious photographer, nobody shoots digital.

00:27:57   It's very weird.

00:27:58   It's very weird, but I kind of get it.

00:28:00   Anyway, so I think that's an interesting just tech example of something that sort of made a lot of skills obsolete and yet didn't kill the medium.

00:28:09   Well, don't you think we should take a step further back, though, before photography existed and consider illustration and painting and what film photography did to being a painter or illustrator?

00:28:23   Right, it's all over.

00:28:25   Yeah, I mean, and it used to be that there were very few people on the face of the earth who had a permanent likeness of them that they could refer to.

00:28:37   Just one, just one, just one painting or portrait of them at any one age, at age 18 or age 30 or age whatever, just one.

00:28:49   And here's a picture of me.

00:28:52   Very few people, you know, up until the advent of photography could say that there was a likeness of them captured on canvas or any other medium.

00:29:01   Right, and everyone thought, oh, painting's dead, and then painting evolves.

00:29:05   Again, painting today, I think paintings sell for way more than photos sell.

00:29:10   It's like the power of painting is still present and seen as a special, even more of a special, fine art activity.

00:29:16   So that's all really interesting.

00:29:19   I think another example is that despite the fact that our iPhones and the video capabilities of these iPhones are rivaling, to a certain degree, what you would have gotten out of a Hollywood camera 40 years ago or 30 years ago or whatever, like certainly digitally, right?

00:29:36   Like 20 years ago, 25 years ago.

00:29:37   And the fact that we are not seeing a massive glut of incredible iPhone films suddenly and all these, there's not a thousand Tarantinos out there.

00:29:47   So I think there's just something, I think this speaks less to the availability of tools and making things easier to produce or like lowering the bar to things and more to, I think we have a fundamental issue with education in general.

00:29:59   And that people who want to be creative don't know how to be creative.

00:30:02   There's just, I think there's some other block that's happening.

00:30:05   It's not a tool block.

00:30:06   And so the AI programming stuff to me also doesn't feel like suddenly, okay, now everyone's going to be making their own software.

00:30:12   I actually think it's people like you and me who have this long history.

00:30:15   I also have a CS degree who have this long history of programming.

00:30:18   I've grown up programming, doing technical stuff has always been a part of my identity.

00:30:22   Being able to drop into terminal and know what I'm doing was, yeah, that's like a point of pride.

00:30:25   But I would argue that it's folks like us who can take the most advantage of these tools.

00:30:30   And I use Claude code just because the interface to me is better.

00:30:34   I can, I just run it in a directory with a code base and it sucks it all in.

00:30:39   I don't have to like feed it things.

00:30:40   And then I just talk to it in the terminal and it does the changes.

00:30:43   It's quite magical.

00:30:45   And you kind of like, okay, things as it goes through, and then you can edit in whatever using VS code or NeoVim or whatever.

00:30:50   And then you kind of go back and forth.

00:30:52   You're in kind of conversation with this thing in a way that feels really interface wise intuitive.

00:30:56   I think this is just sort of self-evident, but human brains don't work like computers exactly.

00:31:03   And for some of us, our brains work a little bit more like computers than most people.

00:31:11   And I often say to people, like, I don't know what I would have done for a living 100 years ago or more.

00:31:20   And I've often said there's a part of me for, at some point in the last few hundred years, I might have been a watchmaker.

00:31:27   That there is a sort of algorithmic mindset to the way clocks and watches work that appeals to me very greatly.

00:31:37   I think even though they are not digital devices, and it's part of what I like about wristwatches in today's world, they do sort of work like that, but in a mechanical sense, right?

00:31:48   But it's like-

00:31:49   Oh, they're so beautiful.

00:31:50   Right.

00:31:50   But it's like if you have a spring that has this much energy and it expands at this rate and this gear turns at this rate, it'll turn this other gear and it'll make this happen and this will happen.

00:32:03   And it's all these sort of logic gates that are exactly like computers.

00:32:07   Yeah.

00:32:08   And because most humans' brains don't work that way, computer programming languages just don't seem natural or understandable to most people.

00:32:19   Yeah.

00:32:20   But they are.

00:32:21   It's a way of computer languages, everyone, even the most friendly ones like Python, I'll just cite as probably the sort of most syntactically pleasing.

00:32:34   Ruby.

00:32:36   Yeah.

00:32:36   Everyone quotes kind of cites Ruby, right?

00:32:38   Yeah.

00:32:38   As the most human-friendly.

00:32:40   What if we took Perl and made all the cruel stuff not-

00:32:44   What if we made it a lot more fun and took out some of that stuff?

00:32:49   Right.

00:32:49   But like Ruby or Python or something like that.

00:32:51   But even those languages, you could show the source code to a well-written Python or Ruby program to most people and with good variable names and good class names and subroutine names.

00:33:05   They could kind of follow along in their language-ish, but it's still to be able to write it, you kind of have to put your mind in this mindset.

00:33:14   And these AI systems, because they are computers, it's like, oh, that rigid structure of these languages is so much easier for them than to produce actual just English.

00:33:28   It's so much easier and so much more natural.

00:33:31   It comes naturally to them.

00:33:34   It's just an inversion of what it means to be intelligent that it's very hard for me to come to grips with.

00:33:41   And I still find myself thinking that I should, you can see me here on video, I'm putting dick quotes around should, that I should be writing code myself because it's like good posture and brushing your teeth every morning.

00:33:57   Before you go to bed and flossing once a day and do, you know, you should write your own code.

00:34:03   And it's like, well, wait, why?

00:34:05   Yeah.

00:34:06   Yeah.

00:34:07   I feel that in my bones as well.

00:34:10   But at the same time, I'm like, if I can produce this thing.

00:34:13   So making like that Twitter clone for my members, I wanted that.

00:34:18   I wanted that for a while.

00:34:19   I wanted a space that was like kind of a community space that I wouldn't have to manage.

00:34:23   But that was also implicitly kind of gatekept by rules and blah, blah, blah, blah.

00:34:28   It was just and then I could just delete it if I wanted to.

00:34:30   There's no world in which I had the time to sit down and properly.

00:34:35   Yeah.

00:34:36   You would have never done it.

00:34:37   No.

00:34:37   10 years ago.

00:34:38   Right.

00:34:38   It's like, it's like 10,000, maybe not 10,000, but it's like 5,000 lines of code that I just

00:34:43   wouldn't have coded.

00:34:43   I wouldn't have done it.

00:34:44   Maybe 10 years ago.

00:34:46   But no.

00:34:46   And but at the same time, there is no normal person who has never dropped into terminal before

00:34:52   that could have ever made this thing.

00:34:53   So it sits in this really interesting kind of liminal space of you have to know tech,

00:35:00   but at the same time, exactly like you're saying, what am I losing?

00:35:03   What's atrophying in my brain by not doing the kind of grunt work of sitting in the functions,

00:35:08   writing the functions, making sure that I know exactly how everything's moving beneath the

00:35:11   surface.

00:35:12   I'm looking at everything Claude gives me and I'm under, and I understand everything

00:35:16   it's doing, but you know, you are losing something.

00:35:20   I feel like there's a non insignificant part of my brain that is dedicated to CSS.

00:35:24   And I've just sort of like, why, why, why do I, I know CSS so well.

00:35:29   And I, whenever I go to have to edit CSS, I just feel like I'm dropping into my happy place.

00:35:33   And actually I have this aura ring and the biggest thing I found with the aura ring, aside

00:35:37   from the sleep tracking stuff is great, but looking at the stress graphs throughout the

00:35:42   day.

00:35:43   And when you drop into, there's a, there's a range called restorative, which is when your

00:35:47   heart rate drops, your breathing drops, you're not asleep, but you're in this, you're in this

00:35:52   flow place.

00:35:53   And I drop into restorative whenever I'm doing CSS work.

00:35:57   It's just so weird.

00:35:59   Anyway, it's complicated.

00:36:01   Yeah, it is.

00:36:02   I think you should do a little project with Claude Code.

00:36:05   I think it's that, that for me, that's why I think it's that.

00:36:07   And that's why I built the Twitter thing.

00:36:08   Cause I was like, I want to be touching these things.

00:36:10   I want to kind of understand what's happening with them so I can comment on them with some

00:36:15   authority.

00:36:16   Right.

00:36:16   So I can say, this is how it really feels to use these things.

00:36:18   And Claude Code is interesting enough, I think from an interface point of view and how it kind

00:36:23   of, it's way, way, way better than open AI stuff.

00:36:26   And also, I don't know.

00:36:28   It just, I want competition in this space.

00:36:30   I don't want open AI to win.

00:36:31   Yeah.

00:36:32   Yeah.

00:36:33   Well, I don't want any one system to win.

00:36:36   Right.

00:36:36   Exactly.

00:36:37   It feels like this is one of these things where maybe there won't be an infinite number of

00:36:42   them, but it would be bad if, it's just bad for everybody or for society in general if one

00:36:49   of them becomes dominant.

00:36:51   I don't know that it's not going to happen.

00:36:54   I don't feel like anybody does, but I know that the theory behind all the money that's

00:36:59   being pumped into it is that one of these systems will get to quote unquote AGI first.

00:37:05   Right.

00:37:06   And for the definition of AGI that investors care about or that these people who think it

00:37:12   might happen care about, which is what happens if one of these systems becomes smart enough

00:37:17   to be able to make the next version of itself and can do it a thousand or 10,000 times faster

00:37:24   than humans can do it.

00:37:26   Yeah.

00:37:26   All of a sudden that one system will clearly jump out ahead of everybody else and nobody

00:37:31   will be able to catch up because it'll already be a thousand times further ahead.

00:37:37   Yeah.

00:37:38   I don't think that's going to happen.

00:37:40   It doesn't seem like that's where we're heading.

00:37:43   It feels to me like we're already like as exciting as it is when open AI releases its

00:37:51   new ridiculously named oath for many, whatever.

00:37:56   Is it exciting?

00:37:56   I don't even know what any of these new things do.

00:37:59   No.

00:37:59   Like I honestly, I never know which one to use.

00:38:01   I only, here's the thing is I, I said to you, like, I don't even know which one of these

00:38:07   apps to open to use to, Hey, I want to write a script that I have this idea for a program

00:38:12   to do a thing.

00:38:14   Which one should I use?

00:38:15   Even if I just say, I'm going to use chat GPT because I like the Mac app, which is really

00:38:21   the main reason I use chat GPT is that, but I don't know which one of the models to pick

00:38:27   in chat GPT.

00:38:29   I don't know, but I really don't.

00:38:30   Nobody knows.

00:38:30   No one knows.

00:38:31   Sam Altman doesn't even know.

00:38:33   And I find that I'll switch.

00:38:36   I'll think, ah, that wasn't that great an answer.

00:38:38   Maybe I should use this one.

00:38:39   That's more supposedly better reasoning and I switch and then I forget that I switched

00:38:44   it and I come back and I'm like, Oh, am I using the other one?

00:38:47   I can't even tell.

00:38:47   No, you can't.

00:38:49   It's so ridiculous.

00:38:50   But there is though a return to Apple script as the mindset to direct these things is

00:38:59   sort of like the dream of Apple script, right?

00:39:02   where like somebody like you or me is writing very precise English to direct Claude to do

00:39:11   the thing, right?

00:39:12   Like when you're, when you're doing this, you're not just like looking off to the side and you're

00:39:21   being very precise in your instructions of what you want out of it to make, to write the program.

00:39:26   But you're, that, that was the dream of Apple script, right?

00:39:30   That you could just sort of describe what you want in regular English, which is native to

00:39:37   our brains and get computer code out of it.

00:39:41   Yeah.

00:39:41   And this is, I use this to build tools all the time and I use, I used to build Alford

00:39:47   workflows.

00:39:47   So I use Alford as my launcher and you know, you can tie scripts into Alford launch things

00:39:52   and whatever you can have the, the output be fed back into Alford and I use it all the

00:39:57   time.

00:39:57   I have, I have one that I just built that was Google drive.

00:40:01   So one of the, I love Google docs and I love Google sheets.

00:40:04   I just, I think they're fantastic.

00:40:07   Like if you have to collaborate with anyone on anything with a, about a text and need edits

00:40:11   and stuff like that, Google docs is sort of unbeatable.

00:40:13   I've found.

00:40:13   And the worst thing about Google docs is having to touch Google drive to use it.

00:40:18   It just sucks.

00:40:19   There's just something about Google drive.

00:40:20   That's terrible.

00:40:21   I hate, I really hate Google drive.

00:40:23   And I, I personally do not like Google docs, but I know, I know what you mean.

00:40:29   Yeah.

00:40:30   Yeah.

00:40:30   There are certain things that are annoying, but I find the benefits are pretty, pretty big.

00:40:34   So anyway, so the other day I was like, what can't I just use the API and in Alford, just

00:40:41   type in GD space.

00:40:42   And then the first few characters of a title of one of my Google docs that I know I want

00:40:46   to open and then just have it, pull it up.

00:40:48   And then I just hit enter and it opens it.

00:40:49   And yeah, in, in three minutes, I had that script built with chat CPT and, but I could see

00:40:54   it.

00:40:54   Oh, it's this bash thing.

00:40:55   It does this thing and authenticates with the Google it ball.

00:40:57   It's, it's a great, like those, the malleability of Mac OS is now, I think.

00:41:04   Giving it in the age of these LLMs, giving it like another breath of like superpower life

00:41:10   that iOS and iPadOS simply don't have access to.

00:41:13   And to me, that feels really exciting.

00:41:15   It's funny you bring that up because I had this on my list of things to maybe bring up

00:41:21   with you.

00:41:21   And I thought, ah, we're going to talk about so much stuff and I always go long, so I'm

00:41:25   not going to bring it up, but you brought it up, but I'm a launch bar guy, not an Alfred

00:41:30   guy.

00:41:31   But I feel, I feel like way back when there was Quicksilver.

00:41:36   Did you use Quicksilver?

00:41:37   Quicksilver.

00:41:37   Yep.

00:41:37   All right.

00:41:38   So Quicksilver was sort of the first like command space, type something, tab into an action.

00:41:44   And I feel like the misconception that so many people who don't use any of these tools use

00:41:52   is they're familiar now with Spotlight, that Spotlight, and again, it's close to 20 years

00:41:57   where it's been built into Mac OS, where you type command space and the name of text and text

00:42:04   edit comes up and you hit return and it launches text edit.

00:42:08   And they think of them as launchers and some people call the app category launchers, but

00:42:13   I feel like the, you're underselling what these things can do, right?

00:42:18   Because you can, it's more than just launching an app.

00:42:21   It's, it's that you can have these actions that come built in with Alfred and launch bar and

00:42:27   then the up and comer is Raycast, but you can, and if there's three things that you really

00:42:33   care about, you can have this custom workflow that makes this thing that you do.

00:42:38   do all the time so much faster, 10 times faster than it would be before.

00:42:44   And it's one of those things, again, it comes back to why I hope open AI doesn't dominate

00:42:52   the way Google search has dominated search, where I feel like one of the things that makes

00:42:57   the Mac great is we have three or four of these launcher command space utilities that are all

00:43:04   great. And if one of them went away of launch bar, just if they announced a, we're dropping

00:43:09   support for the next version of Mac OS, you have to move on. I know that I can move to Alfred

00:43:14   or Raycast and I'd be disrupted in some, some way, but overall it's like, how great is it that we have

00:43:23   three or four of these apps that are absolutely terrific and nothing like that is technically

00:43:30   even possible on iPad or iPhone. Yeah. But I, I wrote a, cause I've been, I had him on my show a couple

00:43:38   months ago, Vlad, uh, Prilovic, the founder of Kagi, the search engine. Oh yeah. Yeah. Yeah. I use that.

00:43:46   I think you got me onto that. That's been, so you're using it. Yeah. Because of you kind of

00:43:51   saying, Hey, I use it. I started using it a couple of years ago. I've got a Kagi t-shirt that I wear to

00:43:56   bed sometimes. Well, but I, but I launched most of my web searches through launch bar, not Safari,

00:44:04   because I'm, I may not be in Safari at the time. I just want to hit command space K and I'm in Kagi

00:44:10   and start typing my search and launch bar makes it super easy. The way that every browser other than

00:44:18   Safari lets you add a custom search engine with just a sort of a URL with an asterisk as the placeholder

00:44:25   for the search terms. Launch bar does that too. And that's what I've been using for Kagi for years.

00:44:30   And I thought, wow, but I would love to have like auto suggest. So like if I start typing

00:44:35   Craig space M and it fills in Craig mod and maybe the kissa by kissa or something like that.

00:44:43   And I thought, Oh, well I, I, I know launch, I've never written an action and I should have,

00:44:49   I've been using it for 20 years and I love to write scripts and stuff like that. And I was like,

00:44:54   I bet I could use AI to help me. I'll use like open AI or something like that. And it turns out to make

00:45:00   the action. I only needed two lines of JavaScript. So I didn't need AI. It was literally two lines.

00:45:06   And I, then I thought to myself, what I really learned how to make these actions for launch bar.

00:45:11   I found this, the exact same thing. It's like a lot of these little scripts I'm building are so

00:45:16   tiny and easy. I made one the other day. I keep a text file called you're not a piece of shit.md.

00:45:21   Anytime someone sends me a nice message or a nice email, I copy it. And then I add it to the bottom of

00:45:28   that you're not a piece of shit.md. And the file is like 60, 70,000 words long now,

00:45:34   which is like nice to see that number grow. But it was always a pain in the ass to open it,

00:45:38   paste it in the bottom, put a couple of spaces, add four dashes. And so I was like, I wonder if I could

00:45:44   just do a quick little Alfred thing. And now I can, I open Alfred, I just type not shit, hit enter,

00:45:50   and it takes whatever's in my clipboard and it pastes it onto the bottom of the text file. And it was

00:45:55   like two of the stupidest, simplest Unix commands ever. It's like PB copy, whatever, less than sign,

00:46:01   blah, blah, blah. Anyway. Yeah. These things aren't complicated, but the AI removes that stupid wall

00:46:07   of friction and just gets you to do the thing, even if the solution's simple.

00:46:10   Yeah. There's like, I don't know, some kind of Unix punctuation command line thing to prepend to a text

00:46:17   file. And then there's a different sequence of characters to append to a text file. And every

00:46:23   time you forget, it's like, you don't need it for another five years. And so, but why,

00:46:29   why even worry about the fact that you forget it? You can just ask the AI and it'll, it'll come up with

00:46:35   it. And, and, and, and it really is this massive relief. It is, Oh, it's like, here's the memory I

00:46:42   wish I always had for the command line incantation to just append to a text file.

00:46:49   Yeah. Yeah. Yeah. Yeah. It's interesting. I mean, going back to your point about like whether one

00:46:55   company will win or not, and we don't want that. I think to me, one of the hopeful lights on the

00:47:00   horizon is the fact that these open models are actually so good that you can, and I have a maxed

00:47:05   out MacBook pro so I can run a lot of these big open models. And I do every now and then I'll,

00:47:11   you know, the old llama thing, get that going, download a model, see how it works on my, on my

00:47:15   local laptop. To me, I think we're getting pretty close to where almost all this programming stuff

00:47:20   that I'm doing now could be done with one of these local models. Certainly all the stuff we're talking

00:47:26   about right now in terms of like shell scripts and little JavaScript things and all that, but also

00:47:31   translation. I mean, all this is language. And the other thing that these models shine with is,

00:47:35   is translation. So I'm constantly going between Japanese and English. And I find that I use it a lot.

00:47:41   Like when I'm writing formal emails in Japanese, I write the formal email and then I

00:47:45   dump it into one of these things and I say, Hey, can you make this, make me sound a little

00:47:49   less nudnicky? And then it corrects it. And it reminds me, Oh, that's right. We can use these

00:47:53   verbs and we're doing these formal things. Oh, this is too formal. So I'll dial that back a little,

00:47:57   but anything language related, these things, these models are so powerful with, and I think we're,

00:48:02   we're very, very close. I mean, like a year or two away from most of this stuff running locally

00:48:07   on a maxed out apple chip. And that, that gives me hope that there isn't going to be one dominating

00:48:13   force in the universe. Yeah. All right. Hold that thought. Let me take a break here. Thank our next

00:48:18   sponsor. It is our good friends. First time sponsor on the podcast, but they sponsored the website last

00:48:23   week. Decaf coffee roasters, decaf with a K D E K A F and decaf believes those who drink coffee purely

00:48:32   for its flavor are the true connoisseurs while other roasters. And there are many great coffee roasters

00:48:40   out there and the founders of decaf love them all too. But you know, most of them treat decaf,

00:48:46   decaffeinated blends as a side project. They have one or two or something like that. Decaf has made it

00:48:53   their entire mission. I'm going to tell you, Craig, you know, this, I drink a lot of coffee and I drink

00:48:59   coffee. It's, it's probably the single most consistent thing in my entire life, which is that,

00:49:05   and I was trying to think the other day when I wrote my thank you to them, when was the last day

00:49:11   that I did not drink coffee? And even 14 years ago, I had surgery on my hand because I cut a tendon in

00:49:20   a finger, but I had to go to the doctor. I went to the ER the night before and they said, go to this

00:49:26   doctor at Jefferson hospital. And I went there and they got me an appointment first thing in the morning

00:49:31   and it was great. And I went there with coffee in my hand and my surgeon who turned out to be my

00:49:37   surgeon took, he was like, Hey, he's like, what's, what's in this coffee? Do you drink it black? And

00:49:41   I was like, yeah, I just drink black. And he was like, okay, good. Cause we could do cert, you know,

00:49:44   if you haven't had any milk, we could do surgery on you at noon. So even the day I had surgery on my

00:49:49   hand, I had coffee at like eight in the morning. I drink coffee every day. I love caffeinated coffee.

00:49:57   But the other thing is I used to always make coffee. I make a second, like a half size batch in the

00:50:02   afternoons. And I got into my forties and it's sort of started screwing up my sleep and I gave it up.

00:50:10   I really, really enjoy decaf blends. They sent me a couple. They also, in addition to their fully

00:50:17   decaffeinated blends, they have these half to 50% decaffeinated blends, which is what I've been

00:50:23   enjoying in my afternoons. And if you like tea, God bless you. Me personally, I do not like tea. I like

00:50:31   hot black coffee or cold, cold brew coffee. And that's the other thing that they've got. They have

00:50:38   a brand new, this, and maybe it's a familiar word. Maybe you can correct my pronunciation,

00:50:43   Mizudashi Cold Brew Concentrate. It's a Japanese style of cold brew coffee. I do, in hot weather,

00:50:51   I do love cold brew coffee. And also that's sort of like my afternoon jam, like even in the summer,

00:50:58   hot coffee in the morning, cold brew in the afternoon, but I don't want all that caffeine anymore.

00:51:03   They make it with a slow drip of cold water over the beans. Instead of steeping them for 12 or 24

00:51:10   hours, it results in a smooth, low acid, and very, very flavorful cold brew style. Very, very good.

00:51:17   They've got nine single origin blends or single origin decaffeinated coffees. Six blends. The blends are

00:51:25   more of my jam, but maybe you're a single origin person, but you know, that's 15 styles of regular

00:51:31   coffee. They've got the Mizudashi Cold Brew. All of their coffee ships within 24 hours of roasting.

00:51:38   So go to decaf.com with a K, D-E-K-A-F.com and use the code DF, just DF like daring fireball. You get

00:51:49   30% off your first order by thanks to decaf. Holy crap. That's, I, of all your sponsors, honestly, I think that's the most

00:52:00   exciting sponsor I've ever seen you, you have. Not to, not to like, not to rank your sponsors, but like, I, I'm also, I'm the

00:52:08   same. I, I, during COVID, I realized I was having panic attacks and I thought it was because of like COVID and the end of the

00:52:14   world, but it was just because I was brewing way more coffee than I normally did. Cause I was stuck at

00:52:18   home. Yeah. And I was bored. Yeah. I was definitely doing it. I was making like a second pot of coffee.

00:52:22   Yeah. Yeah. So I had to, I, yeah, yeah. I had to module. I had to learn. I was like, Oh my God,

00:52:29   I have a limit now. And again, maybe it's this over 40 thing, but I have a limit of how much caffeine I can

00:52:33   take in a day. But like you, I really love drinking coffee. So actually decaf, I don't think they shipped to

00:52:39   Japan, which is very depressing, but I'm going to try to pick some up. We'll have to look at, we'll

00:52:44   have to look into this world tour, this world tour I'm about to head off on. But I, in general, I love

00:52:48   that decaf coffee is kind of be becoming a, a real thing. People are kind of looking at it again. It's

00:52:54   great. I, I, I, again, the sponsor reads over, but I have to say, I just, I I'm 52 now, but I, I've, I think

00:53:01   I've had before they sponsored, I think I had three cups of decaffeinated coffee in my life.

00:53:05   I mean, it's disgusting. It was so gross. So gross.

00:53:09   And I think at least two of them were by mistake where I was like, Hey, is this like decaf? And

00:53:15   they're like, Oh yeah, sorry. And yeah. Whereas decaf's coffee actually tastes really good. So

00:53:22   yeah, I'm excited. I'm excited to try it. I got to get some of those beans.

00:53:26   All right. So to return to what we were talking about before we started talking about decaf,

00:53:31   part of this whole, I don't want to go totally sideways talking about Siri and Apple, but part

00:53:37   of the whole contrempt, whatever we wanted to call the controversy over the last six weeks

00:53:43   of, Hey, some of the stuff Apple promised last year, isn't going to ship this year. And Hey,

00:53:50   Apple's sort of having their, Hey, we, we actually have to do something about the fact that Siri sucks

00:53:57   and everybody can kind of see it now because these other things don't suck. Right. You can

00:54:06   give these plain English commands to these other systems and they understand you and they do their

00:54:15   best. And again, people, they can hallucinate or have you, have you had this with coding where every

00:54:21   once in a while, like I was coding programming something and maybe it's because I personally

00:54:27   lean towards Pearl, which is the language I'm most familiar with. But I was writing a script like a

00:54:33   week ago and chat GPT effectively hallucinated a module, like an open source module that I could use

00:54:42   to do a thing. And I was like, I don't think that exists. Cause I went to CPAN, the comprehensive Pearl

00:54:49   archive, the thing where you download all the open source things. And there was no module with

00:54:55   that name, but the code that it gave me would have been the perfect API for the thing I wanted to do.

00:55:02   And I said, I don't think that exists. And it was like, Oh, you're right. It doesn't exist. Sorry

00:55:06   about that. And it moves on. But again, if you kind of know what you're doing, you'll catch it. Like

00:55:12   you don't just blindly copy and paste, you copy paste, use double check and you figure it out,

00:55:18   you know? Well, and, and with the code, it's instantly verifiable, right? It's like, you try

00:55:22   to run it. It doesn't work. You're like, okay. Right. That's, that's the other thing that's always

00:55:27   been so beautiful to me about computers. Right. And it works or it doesn't. Right. It's, and it's why

00:55:35   using these systems in these early days to write computer programs sort of really works because it

00:55:44   gives you code, then you copy and paste, you make the code. And then does it actually do the thing you

00:55:49   wanted it to do or not? Yeah. Well, look, I think, I think with the series stuff, they are just way

00:55:56   overthinking this and you just create the most bounded, simple set of, of actions within which

00:56:03   you can, you can do things with, with Siri, like all the home kit related stuff. So turning on lights,

00:56:09   turning off lights, setting timers, asking dates, stuff like that. All there's, there's a, there's

00:56:12   like probably a hundred commands for the most basic commands. And you just force the LLM to stay within

00:56:17   those bounds and that's it. And just make those work well. Right. That would, that would 99% of the

00:56:22   way get Siri to a sane place. Right. Right. So that, that to me is really weird. And I was thinking

00:56:28   about it the other day. I was like, I don't think there's a single piece of technology that has created

00:56:33   more aggregate frustration than Siri. Just if you think about the number of people who've sworn at a

00:56:39   device, because like that Larry David thing in the, in whatever the last. Curb your enthusiasm. Yeah.

00:56:45   Last curb season. Where he was trying to get driving directions. It's like, I don't think there's

00:56:50   ever been a thing ever in the history of humanity that has been sworn at more than

00:56:54   Apple Siri devices. But we just ran into it here because Amy made a whole bunch of stuff

00:57:00   for our family Easter feast over the weekend. And she was setting timers, multiple timers for

00:57:09   multiple things that she was baking. And at one point she had a cake in the oven and said,

00:57:17   Hey, dingus, how much time is left on the timer? And it said, there is, there are no timers.

00:57:22   Yeah. Or, or my favorite is my favorite is when you ask her how much time is left and then it

00:57:27   cancels the timer. Yes. Yeah. Yeah. That might've been what happened. I don't know. But the worst part

00:57:34   is, well, for me, the worst part is her anger goes towards me because I'm not that I made Siri, but I'm

00:57:43   the nerd who brought Siri into our kitchen. Right. And so therefore it's my fault. And the cake turned

00:57:49   out perfect because, but she had to kind of eyeball it from that point forward because she didn't know how

00:57:55   long it had been in. Yeah. And just kind of had to open the oven and sort of stick. Like, I don't know what she was

00:58:01   putting in there like a fork or something to sort of test how done it was, but she was so mad. And I know she tried

00:58:08   to set the timer, right? Well, I mean, there's no doubt that she wouldn't have remembered the moment she put the

00:58:14   raw cake batter into the oven. She definitely set a timer, but it telling her there is no timer. There are no open

00:58:21   timers. And my heart sunk because I was right there. And I was like, oh, I'm in for it now. Yeah. Yeah.

00:58:28   Yeah. Well, in the Wayne Ma information report on Apple putting Siri under Mike Rockwell's leadership,

00:58:36   moving Rockwell from Vision to Siri and having him report directly to Craig Federighi and seemingly

00:58:43   giving him carte blanche to do whatever it takes to make Siri good is one of the nuggets was that

00:58:51   Federighi has apparently instructed his staff that if they need to use open source LLM systems to do what

00:59:00   they want to do it, do whatever it takes. If it includes using open source LLMs, don't worry. Whereas

00:59:08   before that the directive was you can only use Apple's own systems to make Apple's products. Yeah. And

00:59:16   it's sort of a throwaway line in a non-technical article. So I understand why it isn't technical, but

00:59:25   my mind races to, well, what exactly does that mean? And which open source LLMs are you talking about?

00:59:32   And to me, one of the things is sort of based on something you mentioned like half an hour ago,

00:59:38   where it's like the moral or ethical quandaries over how these things were trained, right? So it's not

00:59:45   just like, oh, are we going to include this open source LLM as an executable thing on people's devices,

00:59:53   or is it just something they're using to train their own system on their end? And then the output comes out

01:00:02   of it. And because one of the ways to train new LLMs is to use existing LLMs to provide the training

01:00:09   data. And well, what if Apple can't vouch for the training data?

01:00:16   Yeah, that feels like a weird. Again, I think they're just overthinking this stuff. They don't

01:00:20   need Siri to answer every possible question in the world. It just needs to do basic fundamental stuff

01:00:26   on the device, within the boundaries, within the domain that they know they control. Like this Siri

01:00:32   LLM should be the literal easiest LLM in the world to produce and get working right. And they shouldn't

01:00:38   have to use an open source LLM. And they should be able to train their own and know exactly what's in

01:00:41   the black box. Because that's the other thing is like, these things are such black boxes. You can't

01:00:45   an open source LLM. It's not like you can open up and see where it all came from. So you get into the

01:00:50   ethical issues again there. It's very strange. This mainly to me just makes me feel like there's some

01:00:54   fundamental management issues at Apple now. And maybe we've been feeling it in the cracks of software

01:00:59   for a while. And I think iPadOS is another great example of feels to be like someone is not steering the

01:01:04   ship quite as strongly as it used to be. And I don't want to say like, oh, Steve Jobs is the only

01:01:09   person who can ever do any of this stuff right. But there does seem to be a bit of a mess in taste

01:01:15   around software. Like Apple Music on macOS is one of the worst pieces of software I've ever used in my

01:01:22   entire life. I opened it up because I'm an Apple Music subscriber because Spotify didn't come to Japan

01:01:27   until late. And so Apple Music got me first. And nothing brings me less joy than using that app on my

01:01:34   Mac. It just there's so many broken pieces to it. And it drives me nuts. And it just makes me feel

01:01:41   like no one inside of Apple ever uses it or cares about it. Because if anyone did, if anyone who is

01:01:46   an executive there used Apple Music and really, really cared about it, there's no way they could

01:01:50   let this thing ship or continue shipping as it is. It just feels broken.

01:01:53   It's a good example. And it is sort of an interesting, slow boiling frog example where there was never a,

01:02:04   oh, that's the year 2013 where they threw out all the old code and started a new thing. No, it's been this

01:02:13   continuous evolution from iTunes, which used to be Sound Jam before Apple bought it. And you, the original

01:02:21   version of iTunes, you could kind of see some of the Sound Jam roots in there. And it was a great app

01:02:27   that we loved and deservedly so, right? Like when iTunes 1.0 came out, it was such a great app. And

01:02:35   we could quibble about the brushed metal interface as an aesthetic thing. But in terms of...

01:02:43   Even that's cool.

01:02:44   Even that's cool.

01:02:45   Yeah, it was kind of cool at the time.

01:02:46   Yeah, right. Yeah. But it was just a great app. And it was like, oh, but I have 10,000 songs or I have

01:02:54   20,000 songs. It's like, no, just throw, put them all in your music folder. iTunes will be able to deal

01:03:00   with it. I've never heard of anybody back in the day, 25 years ago, who had too many MP3 files for

01:03:08   iTunes to handle, right? It was great at enormous libraries. It was great at building playlists. It was

01:03:15   great for the common scenario of, I really only listened to the same 100 songs over and over again.

01:03:21   It had the cool visualizations. It had so many cool things and people... And it was scriptable. So nerds

01:03:28   like us could write scripts or LaunchBar or Alfred, which wasn't around at the time, but something like

01:03:34   Alfred or whatever. You could write little commands that would skip to the next song or something like

01:03:40   that. Everything was great. And then it just slowly got worse and worse and worse. And somehow we wound

01:03:49   up here where even just resizing the window feels janky, right? It's like, even when you just resize the

01:03:57   Apple Music app on macOS now, it's like, or just resize the column from the sidebar to the main window.

01:04:04   It's just the way the selection moves around. It's like, all of this feels janky. How did we get

01:04:08   here? Yeah.

01:04:10   Shouldn't this be a beautiful app?

01:04:11   It should.

01:04:12   And I do. I feel like it's another example, though, of what we were talking about with open AI and a fear

01:04:20   of, well, what if only one LLM type system takes over and everything else is sort of the equivalent of

01:04:27   what Google is to search. And I feel like Apple is that too tasteful UI. And I kind of feel like an awful

01:04:38   lot of what smart critics of Apple's current software problems, what it boils down to is which

01:04:50   perspective you're looking at. Is this, are these critiques accurate? Is this app janky? Is this

01:04:58   buggy? Yes, yes, yes. This is all correct. But from the other perspective, which is where I think

01:05:05   Apple's leadership looks at it from, is it better than anything else out there? And their answer is

01:05:11   yes. And so therefore, what's the problem? Right? None of this stuff makes me want to switch to Android

01:05:17   at all. And I have an Android phone and I look at it and none of it's better there. The problem is

01:05:23   that there's only one company making really good computers and it's Apple.

01:05:27   Yeah. Recently in the last maybe year or so, I've thought more and more about grabbing a framework

01:05:33   laptop and just seeing how that feels to use and throwing Ubuntu on there and just seeing where it is.

01:05:40   Again, to feel like the pain of polish of Mac versus not the polish of Linux or whatever.

01:05:44   But it's the first time in a long time I've felt like maybe, you know, having a system where

01:05:49   everything, especially with LLMs being able to do a lot of the heavy lifting around configuration or

01:05:54   whatever, maybe we are at this moment where Linux and being able to like just super dial in this thing

01:05:59   like a glove made perfectly for your hand as a system might be a closer reality. But yeah,

01:06:05   Apple Music, I mean, when we say like something feels janky, to me it just means that the underpinning

01:06:11   code, like there's a foundational, there's like a bedrock of code that these apps live on top of and

01:06:15   the bedrock is just really wonky. And it feels like there could be sinkholes at any moment. It's like

01:06:21   just moving around the app. Things don't respond reliably or consistently. Play buttons, there's like

01:06:27   three or four play buttons on each screen. It's sort of, it's just so weird. What am I supposed to

01:06:31   click? How do I start this thing? And then if I hit the space bar, is it going to scroll the screen

01:06:35   or is it going to activate pause or play? It's just, there's a certain amount of brittleness to

01:06:39   the software that I don't think existed. Like you were talking about when iTunes was the thing back

01:06:44   in the day. It feels thin in a way. I don't know how else to describe it, but it should feel thick

01:06:51   because it's old and it's been there for a long time, but somehow, somehow it, it now just presents

01:06:58   itself as a thin veneer. Yeah. On top of a web view. Yeah. Basically. And the, the polar opposite

01:07:05   example of software to this is for me, things I still use. I've been using things for, I think,

01:07:10   almost 20 years now, 18 years, 17 years. I love things so much. And I love it because it's like

01:07:16   picking up a beautiful hammer. It's just like the perfectly weighted hammer with the beautifully

01:07:21   worn handle made of some like teak or something, like whatever the perfect hammer handle wood would

01:07:27   be. I don't know what that is, but every time I'm in things, I just feel so swaddled in thickness of

01:07:33   good software. I love it. I just love it. I love cultured code. I wish they made other pieces of

01:07:39   software besides things. Well, I also feel like things is a great example though, of an app that

01:07:46   has been around for a long time and has kept moving forward visually, right? Like things today doesn't

01:07:53   look like things pre iOS seven, but manages the, the right balance between trendy enough, but not going

01:08:03   for fads and staying true to, Oh, we have this concept for how you can manage things that you want

01:08:12   to do. And we'll have projects and folders and a couple of metadata items on a task. You could put a due date

01:08:22   on it and there's a notes field and then that's it. And if you need more than that, go to another, there's a zillion

01:08:30   other task managers to do systems. Or if you want to use ours, here's the way we want to use it. And

01:08:38   20 years later, they've kind of stuck true to that without, Oh, here's a major new version of things.

01:08:46   And you have to forget everything you knew how to use things before. No, it's very familiar. So somebody

01:08:52   who had been in a coma for the last 12 years and they'd be like, is things still around? And you'd be

01:08:58   like great news. Here it is. They would be like, Oh yeah. Yeah. Oh, it does look nicer. Or well,

01:09:05   and weirdly you could say, and it's better. Yeah, it is. It's everything. The feature creep has been

01:09:10   so admirably sort of maintained at a same level. Like there's nothing unneeded in things. Yeah. I think

01:09:18   the biggest controversy in the history of that piece of software is the synchronization code they

01:09:22   built. I remember that maybe a year longer than they thought it was going to take, but it works

01:09:27   perfectly. I don't know. I love it. I love touching it. I want cultured code to remake every Apple app

01:09:32   at this point. Keynote still feels good. Keynote feels amazing. I think it probably feels amazing

01:09:37   because it's been forgotten that it exists there. I don't think there's a, there's many people working

01:09:41   on it, but Keynote is to me still one of the best pieces of software. I think that I've ever used.

01:09:47   I totally agree. I know that I turned you onto Acorn fairly recently.

01:09:52   Yeah. Yeah. Yeah. Right. Which is great. I mean that all the time now,

01:09:55   but there's the whole Pixelmator is now an Apple product and we don't know what's going to happen.

01:10:02   We don't know for sure. They're going to do something with Pixelmator. I mean, I don't know this.

01:10:07   Nobody's at Apple has told me, but it would be insane if there's not, I mean, maybe they'll change

01:10:12   the name because it's, it's, it is a weird name that I have. Pixelmator. Pixelmator. I have famously

01:10:20   mispronounced it. That talk about a 20 year old gag, but there's also their companion app,

01:10:26   Photomator, Photomator, which is more like a pro version of the photos app that uses your

01:10:36   Apple photos library, but like a dark is actually good. Well, yeah. But is Apple regretting the fact

01:10:45   that they moved away from aperture and are they going to go back to having, okay, there's the photos app

01:10:51   for most people. And here's our pro photos app for people who are more serious and shoot raw photography,

01:11:00   et cetera. Pixelmator though is going to stay around, but in some way, but I've heard that it's

01:11:07   going to be more like a peer or it's in the group with keynote and pages and numbers, not with final

01:11:14   cut, but I'm not quite sure inside Apple where that line is. I'm not sure that line is as demarcated as

01:11:21   we might think. Yeah. I think, I think one of Apple's biggest missteps in software of the last

01:11:29   15 years or whatever is, is that, that breaking away from here's the true pro tools and here's the less

01:11:35   pro tools. And I think the photos thing and getting rid of aperture, trying to make photos be a place

01:11:41   that you edit photos and do complicated photo editing in feels like a huge failure. And the way the photo

01:11:46   editing tools work in photos too, to me feels really unintuitive. It's a strange interface. It

01:11:51   feels overly complicated. They're like, a lot of times I just want to rotate a photo. Like that's

01:11:56   like a strange thing that the iPhone doesn't get right. Still. It's oftentimes you'll do an overhead

01:12:01   landscape and it'll be in the wrong orientation. And to rotate a photo, you have to click like three or

01:12:06   four buttons to just imagine rotating a photo is probably one of the most used acts out of photo.

01:12:12   They did make a change though, two or three iOS versions ago where the default rotation angle

01:12:19   switched from clockwise to counterclockwise. Right. And by switching, it actually corresponds more to the

01:12:28   mistake you make when you hold your phone. And at least for me, you put the volume buttons up top.

01:12:36   And so for me, for years, it was like three taps to get to rotate. And once I got to the rotate tool,

01:12:43   I had to tap it three times because I had to go one, two, three. So they did fix that, but I do agree.

01:12:50   And again, that feels like the sort of thing that a modern computer should be able to recognize. Like

01:12:56   they should, they should be able to tell when you shot a photo sideways. Right. Absolutely.

01:13:02   Absolutely. Because you know it, right? Like, I feel like we're getting more and more to the point

01:13:07   where anything where a human being can say, oh, that, that photo is sideways. It needs to be rotated

01:13:12   90 degrees. The computer should know that too. Yeah. Yeah. Yeah. Yeah. Yeah. That's it. And that's

01:13:18   what I mean by there, there's a, like a weird brittleness to the underpinnings of some of these things

01:13:22   where it's like, why is it not doing that automatically? And then also it's just, I think from an

01:13:26   interface design perspective, as soon as you try to meld pro and not pro activities into the same

01:13:32   space, it just, it doesn't work. And I think separating those out, it makes it easier for

01:13:37   software developers. It makes it easier for users. I mean, it just simplifies in a weird way. Every,

01:13:42   by maintaining my, by maintaining two code bases, it actually makes it easier for everyone because you

01:13:48   know where to go to do the thing you want to do, as opposed to trying to like muddle through

01:13:51   layers and layers of weird complexity. Yeah. Anyway, I'm being overly negative about some

01:13:57   of this Apple stuff, but for the most part, it's great for the most part. It's great, but we love

01:14:01   it. And we are criticizing with, from a loving place of love. Yeah. But I also criticize it more

01:14:08   and more. I feel like I criticize it from a place of fear, which is what if everything at Apple continues

01:14:16   to go south? What are we left with that if we're going to use as tools? Right. And with physical

01:14:22   tools, if you have like a great old set of knives, well, you can just always just take good care of

01:14:29   them, get them sharpened every once in a while. Yep. And even if the knife maker isn't around anymore,

01:14:35   well, you've already got your set of knives, you're good. And computers aren't like that. Like we kind of

01:14:41   need Apple to keep being the Apple we love and keep having that standard. And if their only

01:14:49   perspective inside is, are we better than every other competing platform out there? That's not good

01:14:58   enough, right? That's to me is the thing that Jobs, Steve Jobs always had was Steve Jobs could envision

01:15:05   what this could be. What level could we be at? And he was competing with that sort of platonic ideal of,

01:15:14   hey, we could make a system that works this good. That's who we're competing with, not Microsoft or

01:15:22   Google or whoever else is out there. It's what's the best potential thing I can envision or I could feel

01:15:29   that we could make. And that's what I worried that Apple's sort of losing, that they're losing the,

01:15:34   hey, we should be shooting for the best possible thing we can possibly make. And instead they're just

01:15:40   shooting for, ah, we're still way better than whatever crap Samsung's shipping on their phones. So who

01:15:46   cares? These, screw them. 100%. Yep. All right. Let me take a break here. Thank our next sponsor.

01:15:53   It is our good friends at Notion. Notion combines your notes, documents, and projects into one space

01:16:01   that's simple and beautifully designed. You can leverage the power of AI right inside Notion across

01:16:09   all your notes and docs without jumping between your work and a separate AI powered tool. Just right there

01:16:16   within Notion, you can just start putting AI commands right in there. And it uses the context of

01:16:22   all the notes and docs that you have stored in Notion to sort of inform and instruct the

01:16:29   instructions that you've had. So rather than say, hey, take a look at all of these documents to learn

01:16:34   everything about this project that I'm working on. It's all right there. And everything is right there

01:16:41   for you ready to query in completely natural language to AI right within Notion itself. It's your one place

01:16:48   to connect teams tools and knowledge so that you're empowered to do your most meaningful work. The fully

01:16:54   integrated Notion AI helps you work faster, write better, and think bigger doing tasks that normally

01:17:01   take you hours in just seconds. You can try Notion for free when you go to notion.com/talkshow. All lowercase letters, you just notion.com/talkshow

01:17:14   Go to try the powerful, easy to use Notion AI today. And when you use that link, they'll know you came from here. Notion is used by over half of Fortune 500 companies. But even if you're just like a one person team or a small multi-person team, it is also a great system for that too. Big corporate environments, small individual teams. Notion works great for you. Go to notion.com/talkshow today.

01:17:42   Boom. And if you work at notion, if you're a notion employee listening, it turns out I'm giving a talk at notion in about two weeks in a couple weeks on May. Uh, what is it? Seven to eight. I think on the five, six, on May 8th in New York City, in the notion offices, I will be giving a talk. So you can tune in virtually or in person. If you're an employee serendipity. I did not know that Craig. How would I?

01:18:10   Uh, let's talk about your book. Impossible. Let's do it. Speaking of May, what day does the book come out? May 6th.

01:18:18   All right. The new book is called Things Become Other Things. Kissa by Kissa was your previous book. And I think that might be when you were on the show the last time.

01:18:28   Yeah. But yeah, we talked about that. Yeah. But that was more or was self published. You did all the work and yeah.

01:18:38   And this is, I don't know how else to say it. And I'm curious what you're thinking is behind going this way. But I mean, this is random house. I mean, this is, yeah, yeah. I got my copy right here.

01:18:49   I don't want to say that a self published book isn't a real book. It is. And it's probably the way I would do it if I were going to put out a book. And I get that that's the way you published your previous stuff. But it is sort of like a thrill for me that one of my pals has a big new random house book coming out. Right. How did this happen?

01:19:10   Yeah. It was a little bit complicated. I did the walk four years ago, like I mentioned at the beginning of the show in the middle of COVID. Did this walk, did a pop up newsletter about the walk, wrote 30, 40, 50,000 words on the walk. And then from that, that was the grist that became this book. I was kind of feeling like I was getting close to finishing it about two years ago. And I ended up going the themes of this book. So Kissa by Kissa, my previous book was about me walking from Tokyo to Kyoto and eating a bunch of pizza toast along the way.

01:19:38   And we kind of talk about Japanese cafe culture, Kissa 10 and blah, blah, blah, blah, blah, blah. And it was sort of lighthearted, fairly short book, maybe 20,000 words with photos. Very proud of it. We've just gone into our sixth printing, which is insane because the book costs 100 bucks a copy. And so we've, I've sold something like 7,000 copies of that book now, which is just bananas. You don't, that normally doesn't happen with a hundred dollar book. And so that book was its own little thing and it's done really, really well.

01:20:06   This one is a little different. It's still about a big walk in this peninsula of Japan. And you, in reading this, you're getting access to this part of Japan that you'll never be able to visit as a normal person. You just won't be able to talk to these people. Like I can talk to them and you won't be able to go to these villages because you just won't know about them. And they just aren't on the radar.

01:20:25   And so in some ways, this is a book about really the hidden parts of Japan that if you're interested in Japanese culture or what's happening in contemporary Japan today in the kind of the countryside, this book kind of really illuminates that.

01:20:37   But at the same time too, as I was doing the big walk, I was reflecting back on this friendship I had with my best friend from elementary school. And this is someone I hadn't thought about in 25, 30 years.

01:20:47   And so this book is also a letter to this old friend, Brian. And there was something about the themes in this book that to me felt like I wanted to give them a bigger platform than I could do just on my own.

01:21:02   And as successful as a self-published book can be, and you know, I'm doing fine art edition. So again, like, you know, maxes out at, if it does really, really well, 10,000 copies maybe.

01:21:12   And I just thought this story and the themes and honoring the memory of this friendship, this deserves a little bit of a bigger platform.

01:21:20   So I went to New York and I flogged myself beneath the feet of all the big publishers and agents and all this crap.

01:21:27   And everyone kind of rejected me. And finally, then I had this amazing meeting with a random house and that turned into the book deal that this book came out of.

01:21:37   I've got a bookmark. I've got a couple of bookmarks in my copy here, but one of the chapters, it just speaks to me.

01:21:43   Shitholes and not shitholes.

01:21:46   And there is, it is so you, because I know you and it's like, well, that sounds like inflammatory or you're shit stirring to make a chapter like that.

01:21:57   But I know how thoughtful you are.

01:21:59   Right. And so much of you comes through in the book. I don't want to embarrass you here, but it is a, you know, what's a walking memoir?

01:22:08   It is a memoir, but by making it more about central Japan and the people and the areas that you are documenting with a sort of, you are there, this is what it's like description.

01:22:25   It's really you that's coming through. It's filtered through Craig Mott in a way that, I don't know, it's hard to describe it. It's a good book, I guess is what I'm trying to say.

01:22:36   Thanks.

01:22:38   It's different though. Are all of the photographs yours?

01:22:41   Yes.

01:22:42   So there's, there's two film stills that were such a pain in the ass to get the rights to, but we got them. There's a film still from Akira and a film still from Ikiru. And so those are in there. These are a Kurosawa film and, and the classic anime Akira.

01:22:59   It's an interesting type of book though. Fundamentally, if you could only see the pictures or only read it, you'd, you would take the words version. At least I would. It is, it is a written work, not a photo book.

01:23:14   It's not full color. It's black and white offset printing, but the photos come across very well. I, I, I've never, I, I'm, I'm trying to think, and I was looking before you came on the show, like flipping through the, I was going to say hundreds, but maybe, I don't know, however many hundreds of books I have in my office.

01:23:37   I can't think of another book I own. That's quite like it in terms of the mix of photos and prose, but that it's just printed black and white offset printing like this, including full bleed photos, which I couldn't find another book on my shelf that has that.

01:23:54   Yeah. Well, in Random House, one of the reasons why I agreed to do this edition with Random House is because they got it. They really got it. And I, I, I've been looking at PDFs for months or a year almost. And then I finally got the printed final, final, final printed edition about a week ago. And I'm, I'm really delighted by it. I was thinking about the printing quality is going to be lower. I know what the paper type is going to be. It's going to be that rough kind of standard mass market hardcover paper.

01:24:22   And so I redeveloped all of the photos specifically for black and white and for that paper. I think the end result is it is a really special mass market hardcover book. I don't think I also haven't, I haven't seen a book that, that, that works like this either.

01:24:38   And I didn't realize until I got the final copy, like how special it was going to feel. The long and short of it is I'm really proud of how it turned out. And I'm really happy with how the team worked together. The designers over there worked together to kind of put this thing to, to, to slap it together. And the full bleed stuff I think is really unique and it works well. And the photos come across great.

01:24:58   In my opinion, in my opinion, as a person who, who is very, very persnickety about printing and all this stuff, I think for what it is, I think it looks fantastic.

01:25:06   I think that the photos to me are like, uh, when you're eating wings or spicy, like Buffalo wings and you have glass of milk with you, like every couple of wings, you need a sip of milk just to calm your tongue.

01:25:24   The pros is the fire and you turn the page and there's a photo that goes along with it. And it just sort of goes down like a nice sip of cold milk. Like it's not the main thing, but it's very soothing and it wouldn't be the same without it.

01:25:42   But the other thing, I mean, it's partly where the name dithering comes from. I've always had this affinity for black and white printing of photographs. And we were talking earlier about how crappy newspaper printing was by, by necessity, but there was something great in some ways about a great newspaper photo just printed in black and white.

01:26:07   And I remember as in particular, the Philadelphia Inquirer went color before I think the New York Times did. But when the New York Times print edition started printing in color, it kind of broke my heart because, and I get why they did it, that there was this interstitial period of years between the heyday of black and white ink stained fingers, newspapers.

01:26:35   And, oh, I don't read a printed newspaper anymore. I read it on my phone or on my, on my computer. There was a brief period of years where to maintain relevance, they kind of needed to be printing in color, certainly for the advertising. And if you're already paying for the color for everything else, the photos should be color. But to me, they just, I don't know. And again, I guess I'm old and I know like my son, he's 21.

01:27:03   He really does not like black and white movies. He just doesn't. Right. And I guess I get it. But I remember being a very little kid thinking black and white TV shows and movies suck because they're old. Color is cool.

01:27:17   And then I got to be like a teenager and I appreciate, oh, black and white could be really, really cool. Kurosawa. To me, well, he has some color ones too that are great too. But to me, my favorite Kurosawa movies are the black and white ones.

01:27:32   And they're just beautiful. And yet to me, I completely forget their, their black and white by the time I'm two or three minutes into them and black and white printing on kind of crappy paper, where if you really, you know, you don't have to have great eyes or look all that closely.

01:27:49   You can kind of see the half tone dithering of the dots. You can see how it's made, but it speaks to me in a certain way. And there is that sort of newspaper pulpy vibe to the book.

01:28:04   And I don't know. It's quite unlike any, I, again, I can't think of another book that's quite like it.

01:28:09   Yeah. I think it works with the story as well, because, you know, when photos are printed like this, there's no pretentiousness to it. Right.

01:28:17   It's just, this is the bare bones version of printing that we can do for this. And where I come from in the story of me and Brian is one, again, that we come from a place that had absolutely no pretensions, had no sense of elevated culture or anything like that.

01:28:31   So in a weird way, like being printed on paper like this, in this way, feels like the most natural compliment to the story of the book itself. So I think it works. I think it works.

01:28:41   I definitely do. And when does it come out?

01:28:43   May 6.

01:28:45   All right. And is there, what's the best place where people can go to, to find out more about it?

01:28:51   I mean, just, if you just search for things become other things, it'll throw you to a...

01:28:56   Craig Mod, things become things.

01:28:57   Things become other things. If you just get, I mean, my website,

01:29:00   Well, no, I'm not saying that's the title of the book. I'm saying that if people just search for that, they'll find it.

01:29:05   Oh, they'll find it. Yeah, yeah, yeah. It'll pop up.

01:29:08   If you just go to my website, it's going to be the main, big, giant, mega link at the top of the page.

01:29:13   That has... So I'm doing this world tour. I don't know. When is this episode coming out, do you think?

01:29:18   Sometime before the end of the month.

01:29:21   Okay. So if you're listening to this, it's recently come out, then I'm doing a American tour.

01:29:28   I'm going to be doing four events in New York, D.C., Chicago, Seattle, San Francisco, Los Angeles, possibly Boston.

01:29:34   We're trying to snap that on at the end.

01:29:36   I would love, love to see readers come out. We've already sold out most of the events. I just heard a launch event at Rizzoli in Midtown Manhattan. We already have over 200 people signed up for that, which is insane for a book launch event. And I'm just a schmuck. I'm not like Stephen King or something like that.

01:29:55   Everyone's kind of freaking out. So if you want to come, on my homepage, there's the Things Become Other Things page. It has the tour list. There are RSVP links for all the events. Try them out. If it's sold out, just come anyway. We'll squeeze you in. We'll just squeeze you in.

01:30:09   Tell them you know John Gruber.

01:30:11   Yeah, yeah, yeah. Tell them you heard it on the talk show.

01:30:14   Yeah, and they'll fit you in.

01:30:15   And they'll have a hair shirt for you, and they'll squeeze you in the back.

01:30:18   All right. I've got one last sponsor to thank. It's our good friends at Squarespace. And Squarespace is the all-in-one platform where you can build your own presence online. More or less, if you need a website, you can go to Squarespace, and Squarespace will let you create a website.

01:30:35   I could end it right there. That's sort of all you need to know about Squarespace. That's sort of the Squarespace brand. But Squarespace keeps moving things forward. Again, there's an AI angle to everything, and Squarespace definitely has an AI angle.

01:30:52   They call it design intelligence. But they've always had WYSIWYG, what you see is what you get editing for the visual nature of your Squarespace website, where you log in as the admin, and the website you see in the admin interface is the website everyone else sees, except you get the admin tools where you can drag things from the left to the right, or change the font, or change the style, or change the hero image, or something like that.

01:31:19   But now they've added this new thing, design intelligence. It's really, really cool. All the old stuff is there. So you can design your website as traditionally as you want. If you're familiar with Squarespace already, nothing changed. You can keep using it the way you want to. But if you want to use AI, and just sort of say, make this feel more fun, or make it look like this is more for kids, or something like that.

01:31:47   Whatever your art, more or less just becoming an art director, and giving art direction to the design tools, as opposed to doing it yourself, and being the designer. You can just sort of give it art direction, and it actually works. And it's really kind of cool and clever.

01:32:04   They've got all sorts of other stuff built in Squarespace payments is the easiest way to manage payments in one place, and they support everything you'd want to support in a modern payment system, like ACH Direct Debit in the US, Apple Pay, After Pay, Clear Pay, which is sort of like the After Pay in the UK.

01:32:23   All of that stuff is all there. Design intelligence is the new AI stuff. And again, if you don't want to use it, it stays out of your way. If you do want to use it, it's really fun, very clever. Totally give it a try.

01:32:37   Analytics built in, in my opinion, the best analytics dashboard interface I've ever seen for seeing where people are going, how much traffic you're getting, what the sources are for a website.

01:32:48   I wish every analytics interface was as cleverly and clearly designed as Squarespace's.

01:32:54   It's so good.

01:32:55   Go to squarespace.com.

01:32:57   You get a free trial, 30 days, no watermarks, just 30 days to use it.

01:33:03   Make your own website.

01:33:04   It's, you know, everything you want.

01:33:06   And only at the end do you need to pay.

01:33:08   And at that point, go to squarespace.com slash talk show, squarespace.com slash talk show.

01:33:15   Remember that code talk show that saves you 10% off your first purchase, including you can prepay for a whole year.

01:33:24   You can register the domain name, all of that stuff.

01:33:27   You save 10% with the code talk show.

01:33:29   And that's how they know that you came from the show.

01:33:33   All right.

01:33:35   Every time I have somebody on who's written a book, I try to get to, because I've never written a book, and I try to get to, how do you write a book?

01:33:43   Like, what tool do you use to write a book?

01:33:46   Oh, you mean like literally?

01:33:48   Yeah, like literally.

01:33:49   And when did you know that this was a book and not just like a series of articles for your website?

01:33:56   Like where, because that's to me is the bigger, I guess there's two questions.

01:34:00   It's the tooling and where do you actually do the writing and where did you realize this could be a couple of hundred pages that you cut it off neatly and it's a book as opposed to your website, which is an ongoing concern.

01:34:16   You have members.

01:34:17   You're always, you never stopped writing for your website while you're writing the book.

01:34:21   Yeah, yeah, yeah, yeah.

01:34:22   So for a lot of things, I do what are called pop-up newsletters, right?

01:34:27   So it's a newsletter that is time box.

01:34:29   It starts on this day.

01:34:30   It ends on this day.

01:34:31   And usually I'm doing a big walk.

01:34:33   So I've done three walks from Tokyo to Kyoto or Kyoto to Tokyo.

01:34:36   I've done a bunch of 500 miles, four or 500 miles, something like that.

01:34:41   It's a big walk.

01:34:41   Yeah, yeah, big walk, big walk.

01:34:43   And I've done thousands of miles down on this peninsula.

01:34:46   I've done other walks in other parts of the country.

01:34:48   I've done other things where I've, let me go to 10 weird mid-sized cities around Japan and then I'm going to walk 50K in each city and then I'm going to write about that.

01:34:56   But I do these things, these pop-up newsletters, because I think people are exhausted by newsletters in general.

01:35:01   There's just so many of them now.

01:35:03   And they're exhausted by things that don't end.

01:35:05   And so with a pop-up newsletter, I think the real benefit is saying this is only going to last for two weeks or three weeks or four weeks.

01:35:12   That's it.

01:35:13   You sign up.

01:35:13   And if you don't, if you miss it, there's no archive.

01:35:16   So you got to get in before it starts.

01:35:19   And then it's going to end.

01:35:20   And what I've found in doing this is, first of all, the open rates are bananas.

01:35:24   Because it turns out when people know something's going to end, they engage with it.

01:35:27   They're like, okay, I can handle that.

01:35:29   There's only going to be 20 of these or 30 of these.

01:35:31   And then for me, what's great about it is it just motivates me to write way more than I would in the middle of these things.

01:35:36   Like I said, four years ago when I did my big walk that became Things Become Other Things, I was writing two, three, 4,000 words a day and sending that out with a pop-up newsletter.

01:35:45   That's a lot of words, Craig.

01:35:47   Yeah, no.

01:35:49   I would do eight hours of walking.

01:35:50   I would walk 30, 40K.

01:35:52   I'm carrying this huge pack, whatever.

01:35:54   I'm interviewing people, photographing people.

01:35:55   And then I'd spend four or five hours every night editing photos.

01:35:59   It was all digital when I did that walk.

01:36:01   And editing photos, writing two, three, 4,000 words, and then putting together a newsletter with photos edited and words.

01:36:08   And then a lot of times, too, I was also doing binaural audio recording.

01:36:12   So I'll import the audio, clean it up, publish it to a podcast that night.

01:36:16   And so I'm doing all this kind of real-time creative work as I'm doing the walk.

01:36:21   It's sort of almost like this insane, masochistic, ascetic practice that I do.

01:36:25   And then the thing is, you pop out the other end, and you've got essentially a book written, what could be a book.

01:36:30   And I've done a bunch of these.

01:36:32   And I actually know the next five books I want to write.

01:36:34   And it's because I've done the basis for it.

01:36:38   I have the next 15, 20 years of work just sitting there.

01:36:42   You just made a face.

01:36:44   You saw me.

01:36:45   I made a face.

01:36:47   Because if you said to me, I know the next book I want to write, I think, ah, I would just raise my eyebrows.

01:36:51   I'd be like, ah, Craig's a pretty savvy guy.

01:36:53   Five books, though?

01:36:54   Yeah.

01:36:55   Yeah, yeah, yeah.

01:36:56   I know the next five books I want to do.

01:36:57   Yeah.

01:36:57   And, you know, in doing them, you'll find other things, and you'll change course or whatever.

01:37:02   But I can see them.

01:37:03   All right.

01:37:04   So, where are the notes, and where are those books right now, technically?

01:37:11   So, I collect, as I'm walking, I dictate into notes.app, just the notes app, transcriptions.

01:37:19   And I'll dictate thousands of words a day as I'm walking, little notes, snippets of conversation, thoughts, whatever.

01:37:25   And then I'll get to the hotel at the end of the day, and then I'll use that as my jumping-off point to write 2,000, 3,000, 4,000 words.

01:37:31   And usually, by the time I sit down to write, I already know the hook of the day.

01:37:37   I know the conversation I want to build off of, and it just kind of flows.

01:37:40   I've not had any knock-on-wood issues of writer's block doing that.

01:37:44   So, notes.app kind of collects the random thoughts throughout the day, and so I have a bunch of those for each walk.

01:37:49   And then, Ulysses is the app I use to write the newsletters.

01:37:53   That's where I write the essays.

01:37:55   And Ulysses is also the app I use to write the drafts for a book.

01:37:59   And I use Ulysses because it lets me just have a Dropbox folder with markdown files.

01:38:06   It doesn't do anything too fancy.

01:38:08   And it lets me order the files in the side panel arbitrarily.

01:38:14   And that's really important to me because I want to be able to move chapters around.

01:38:18   So, each file is a chapter.

01:38:19   And what's great about Ulysses is you can, in the sidebar, you can select, say, five chapters in the sidebar.

01:38:25   And then, in your writing space, it's all turned into one editing view.

01:38:31   So, you can focus in on one chapter if you want and have that be your editing space.

01:38:36   You could select something in the sidebar, hit Command-A, select all the files, and you've got the entire book in your editor right there.

01:38:43   You can move around really quickly.

01:38:44   The search in Ulysses is quite good.

01:38:47   It now does interlinking and link backs and stuff like that.

01:38:49   It's just software that is simple enough.

01:38:53   It gets out of the way.

01:38:54   And it does what I need it to do to be able to write in the way that I like to write.

01:38:58   I have a file called Dead Darlings.

01:39:01   And so, I will cut stuff, kill my darlings, right?

01:39:05   That's like the phrase that they use, right?

01:39:07   So, sentences, paragraphs that I love, but just don't work.

01:39:10   And as long as I can put them somewhere, I don't feel like I've lost it.

01:39:14   So, I put it in the Dead Darlings file.

01:39:16   And it's like I can have that pinned to the top of the folder and it's easy to dump stuff in there.

01:39:20   Yeah, it just works for me.

01:39:21   Ulysses works really, really well for me.

01:39:23   And then, like with Random House, we had to move into this mode of editing, going back and forth with an editor.

01:39:28   And thankfully, we were able to use Google Docs for that, which to me is a thousand times better than what normal publishers use.

01:39:37   And by the way, all my friends in publishing were like, I can't believe they let you use Google Docs because they all want to use Word files.

01:39:43   And not like Word files and pages, DocX files or whatever.

01:39:47   You have to buy Word and use Word and it is terrible.

01:39:51   It is terrible.

01:39:52   It sucks.

01:39:53   There's a part of me that was like, hey, this sounds pretty good.

01:39:56   And I've never been a big Ulysses user, but I could be.

01:40:02   Like, Ulysses is one of those apps where I'm like, oh, I could see how I could get into this if I had more than a file to edit.

01:40:10   Google Docs, that would be a stretch.

01:40:12   Microsoft Word, that would actually, I don't mean to be precious, but it is very me.

01:40:19   I would just be like, we have to cancel the book project.

01:40:22   I'm not doing it.

01:40:24   Yeah, I mean, so Microsoft Word didn't enter the equation until the copy editing process.

01:40:31   But I have heard that from so many people who have written books.

01:40:34   The other thing is, I don't know anybody who's like, yeah, it was fine.

01:40:37   I liked it.

01:40:37   No, I don't.

01:40:38   I have friends who've written books who are very opinionated about software, who make software.

01:40:45   I have friends who have written books who aren't very technically minded at all.

01:40:49   Nobody has good things to say about using Microsoft Word for writing books.

01:40:54   But it is somehow so entrenched in the industry.

01:40:58   I can't think of any piece of software that's more entrenched in an industry than Word is in the book industry versus how anybody feels about it at a qualitative level.

01:41:12   Right.

01:41:12   Like the fact that today we're throwing Word files back and forth still is just psychotic.

01:41:18   It's just insane.

01:41:19   And thankfully, I mean, my editor, Molly, who was amazing.

01:41:22   Thank God she was OK with Google Docs because that made the process so much simpler.

01:41:28   Commenting and going through comments and approving and rejecting stuff.

01:41:32   It's so smooth in Google Docs.

01:41:33   It really just works well.

01:41:34   To me, the fundamental problem with Google Docs, all of them, not just the actual like word processing docs, but their slide thing.

01:41:43   What's their spreadsheet called?

01:41:45   Sheets.

01:41:45   Sheets.

01:41:46   Yeah.

01:41:47   And I know people like it.

01:41:48   What people love about it is that you can share it and that it's just a URL you send to somebody and you send them the URL and then they're in.

01:41:56   They can see it or you can grant them access.

01:41:59   And I do appreciate that that is a huge leg up versus a local .docx file.

01:42:06   But the problem to me is that the concept of all of Google's apps like that is all based on Word or the Microsoft Office suite.

01:42:15   What if we took the Office apps and put them in a browser and made it shareable?

01:42:19   And that is an improvement, but it's also an improvement on terrible software to begin with.

01:42:28   Yeah, and I think the way to approach Google Docs and the way I do it is there's a set of settings you can set and you can establish a kind of typographic baseline.

01:42:38   And once you've done that, you can ignore everything else in it.

01:42:42   And it's just a nice online editor.

01:42:45   And actually the spelling correct, being able to do it all from the keyboard.

01:42:49   I remember we had a bunch of messages back and forth.

01:42:51   We were trying to get macOS to do accepting spell correct suggestions all entirely from the keyboard.

01:43:02   Yeah.

01:43:02   We even wanted to invoke it, though.

01:43:04   Like while you were typing and you got the red underlines, what if you didn't need to reach for the mouse and you could just hit a keyboard shortcut and bring up the spell check menu?

01:43:14   Yeah.

01:43:15   So Google Docs does that really well.

01:43:16   And it was because of Google Docs that got me thinking about that because when you, the cursor's on the misspelt word, it just says a pop-up and you can tab into it and you can tab to get rid of it and tab to accept.

01:43:26   And it works really well.

01:43:27   Yeah.

01:43:29   All right.

01:43:30   I have one last question.

01:43:32   Sure.

01:43:33   How did you get a blurb from William Gibson?

01:43:37   Email me.

01:43:38   And I don't want to throw any of your other blurbers under the bus, but to me, getting a blurb on the back cover from William Gibson to me was like, I'm so happy for you.

01:43:51   I'm like, holy shit, Craig got a blurb from William Gibson.

01:43:55   Oh, dude, I've, honestly, that was one of the coolest moments of doing this whole book was getting the email from Molly, who was like, holy shit.

01:44:05   William just sent the blurb over.

01:44:08   Because, you know, it's like, we knew he had the book and he's like, yeah, yeah, I'll take a look.

01:44:12   And you just go, all right, whatever.

01:44:14   This guy's got a million people pitching him books.

01:44:16   And you just, you kind of write it off.

01:44:17   You assume, okay, that's a lost cause for good reason.

01:44:21   He's old.

01:44:22   You know, what, how old is Gibson now?

01:44:24   He's in his seventies.

01:44:25   Whatever.

01:44:26   He's busy.

01:44:26   He's got stuff that he's doing.

01:44:28   Everyone's pitching him crap.

01:44:30   And, uh, you know, this random ding dong sends him a book about Japan and to not only get a blurb, but to get such a generous blurb that maybe, maybe you could read the blurb for, for listeners.

01:44:44   I will in one second.

01:44:45   Hold on.

01:44:45   I want to do the math on how old he is.

01:44:47   He is 77.

01:44:50   Yeah.

01:44:51   Yeah.

01:44:52   All right.

01:44:53   Here's the, here's the blurb from William Gibson, which actually is unsurprisingly to me, the perfect synopsis of the book.

01:45:00   Craig Mod takes the reader along on an epic, exquisitely detailed journey on foot through a rural Japan.

01:45:09   Few of us are likely to experience uniquely unforgettable.

01:45:14   I feel like we could leave off the uniquely unforgettable, but I feel like that first sentence, that's the book, right?

01:45:21   It is a unique journey, comma on foot, comma that few of us are likely to experience.

01:45:29   Perfect.

01:45:30   Right?

01:45:30   Yeah.

01:45:31   He's a pro.

01:45:32   He knows what he's doing.

01:45:33   He knows what he's doing.

01:45:34   No, that was, that was really exciting.

01:45:36   And again, yeah, because obviously Gibson's work and in particular books like Pattern Recognition.

01:45:43   Yeah.

01:45:44   I love that word.

01:45:45   It's such a good, I keep coming back to it every couple of years.

01:45:47   To have someone whose work you respect like that, not just read what you've done, which in and of itself is like a big honor to just have someone take the time to read what you made.

01:45:57   But then to be so generous with a blurb like that, I don't know.

01:46:02   To me, there is, it's, you know, novels do it and memoirs can do it, but it's just the written word and it doesn't make that big a difference whether it's fiction or nonfiction or a blur between the two like Hunter S. Thompson.

01:46:18   It is a sense of you are there and somehow, even though it's just prose, it can put in your mind the closest relay of this is what it's like.

01:46:36   This is what the people are like.

01:46:38   This is what the place smells like.

01:46:40   This is what it looks like.

01:46:42   It's old or it's new or in a way that far richer multimedia experiences can't capture.

01:46:50   The pictures don't.

01:46:52   And your book has pictures.

01:46:53   We've mentioned it.

01:46:54   But like a movie doesn't capture it.

01:46:56   Or in my opinion, immersive VR wouldn't capture it in the way that truly well-written prose at a moment in time from a certain perspective, that perspective is relayed to the reader to conveys.

01:47:16   And that's what your book has.

01:47:17   And that's why I keep coming back to books.

01:47:20   That's why I like books.

01:47:21   I just, they just feel like one of the few mediums I can really do that.

01:47:25   I can really handle that.

01:47:26   And the other thing is, and it kind of comes full circle to something we talked about earlier, where if social media can be ephemeral and you can just say, hey, this stuff, I toss out my blue sky.

01:47:39   Every six months I delete it or whatever.

01:47:42   Or I've, in your case, you've made a little Twitter-like thing for your membership.

01:47:47   And every seven days, anything that hasn't been commented on, any thread that's seven days old without any mentions, just gets expunged.

01:47:56   And everybody knows that up front, so nobody's losing anything they didn't expect to lose.

01:48:02   But like a book is here forever, right?

01:48:06   Yeah.

01:48:06   Yeah.

01:48:07   Well, and especially, you know, you make thousands of copies, shoot them around the world.

01:48:11   It's highly unlikely that it'll ever be completely erased from the ledger of humanity, too, which is sort of like a weird, there's still nothing in the digital universe that kind of maps to that, right?

01:48:24   Yeah.

01:48:24   Oh, one last thing.

01:48:25   I'm so glad I thought about it before I was just about to say goodbye.

01:48:28   But the other weird thing, and it's just such a funny little coincidence, previous episode of the show, I had Glenn Fleischman on, and we were talking about his book, Six Centuries of Text and Printing, or whatever.

01:48:40   Six Centuries of something.

01:48:42   But such a good book.

01:48:44   Hold on.

01:48:46   I got to get the title right.

01:48:47   That's not right.

01:48:47   But his book set in Bembo.

01:48:53   Oh, yeah.

01:48:54   And your book set, Six Centuries of Type and Printing.

01:48:57   I was correct that that is the title.

01:48:59   Your book set in Bembo.

01:49:02   But to me, a very different version of Bembo.

01:49:05   How much input did you have on the design of the book?

01:49:07   Good question.

01:49:10   Awkward question?

01:49:12   Well, not an awkward question, but I assumed because, look, I went into this Random House thing going, okay, I'm going to have no control.

01:49:21   And everyone told me you're not going to have any control.

01:49:23   So I went in expecting that.

01:49:25   And I actually produced, there's a much longer story to the genesis of this book, but I actually produced a different,

01:49:32   edition of this.

01:49:33   My fine art edition of this book came out 18 months ago.

01:49:36   And this was also what was bizarre about the Random House contract is they gave me a provision for a fine art edition.

01:49:41   And it was a very unique contract.

01:49:43   So I felt like I had produced my own design canonical edition.

01:49:47   So I was able to come into this with less preciousness.

01:49:50   But that said, in the end, the cover, I art directed extremely heavily.

01:49:56   That's my photo on the cover.

01:49:57   And I sort of art directed a lot of the type and stuff.

01:50:00   But there was a designer who kind of like did the finishing mechanics of it.

01:50:04   And then the internal sequencing of images, obviously, I had control over because the images sequencing was important.

01:50:10   And I also had control over the sizing of them and the placement.

01:50:13   So like if it was in line, some of the images are kind of in line with the text.

01:50:16   Others are full bleed.

01:50:18   Others have borders around them.

01:50:19   So I had sort of control over that.

01:50:21   But I had no control over the typography, over the type selection, over the way chapter headers are set, or how you have the running headers on the sides of the book with the title and page numbers.

01:50:32   I didn't have any input on that at all.

01:50:35   But I liked the designs that were sent to me so much I didn't feel like I needed to have input on it.

01:50:40   And we were just lucky.

01:50:41   I mean, they did a great job.

01:50:43   I really do think it reads well.

01:50:44   It feels good.

01:50:45   The pages feel balanced.

01:50:47   I'm as happy as I could have.

01:50:49   I couldn't be happier if I did it all looking over the shoulder of the designer.

01:50:52   There is, I don't know how to describe it, but it's like a certain sense of how much ink is on the page.

01:50:59   And it's like you, when you can read the language that a book is printed in, it's kind of hard to do it.

01:51:10   And I remember when I was learning print design that a lot of people said, you should turn it upside down.

01:51:17   Turn it upside down and flip through it that way so that you can't read it and just sort of look for the balance of that.

01:51:23   That's sort of a gimmick.

01:51:26   But it's like I sort of have that sense.

01:51:29   And the type is a little light on the page, not too light.

01:51:34   I mean, but in an appropriate way that to me is balanced by the number of photographs.

01:51:40   Right.

01:51:41   So in Glenn's book, which is at least my edition, is letterpress printed.

01:51:47   The type is much heavier and there's much more ink on the page from the letter forms.

01:51:52   And in yours, Bembo is a little lightweight in a very appropriate way because to me the ink is going, is balanced by the number of photos that are there throughout the book.

01:52:06   And it's such an interesting two books that are based on the same family of type from, again, 600 years ago.

01:52:14   And it was probably still all, you know, knock on wood, the humanity still around will still be used 600 years from now.

01:52:22   There will be somebody printing a book in a Bembo derived typeface.

01:52:27   Probably, yeah.

01:52:28   It's beautiful, though.

01:52:29   It really is.

01:52:30   And to me is, is, I don't know, it just has the right taste.

01:52:35   I don't know, or smell, right?

01:52:37   There's like a smell to the right font for a book.

01:52:40   And I would not have picked that.

01:52:43   If you had asked me what you should use, I would not have suggested Bembo.

01:52:48   But once I see it, I'm like, oh, yeah, this is right.

01:52:50   This is very nice.

01:52:51   It's a beautiful book.

01:52:52   Beautiful book.

01:52:53   Well written.

01:52:54   Thank you.

01:52:54   Craig Mod, thank you for coming on the show.

01:52:56   Thanks for having me.

01:52:57   And everybody out there, go look for where Craig is going to show up and go see him in person.

01:53:05   Please do.

01:53:06   Say hello.

01:53:07   Give me a hug.

01:53:08   I'm going to need hugs on this tour.

01:53:10   It's going to be exhausting.