Really Specific Stories: Marco Arment Duration: 01:18:05 SPEAKERS Martin Feld, Marco Arment Martin Feld 00:21 Thank you so much for joining Really Specific Stories, Marco; it's a pleasure to have you on the show. Marco Arment 00:26 Thanks! Thanks for having me here. This is great—great honour! Martin Feld 00:28 I'm honoured too! Now, first question, as is the case with every participant: how did you first get into podcasts? Marco Arment 00:35 I think it, my, my story is the same, is a very similar story, but at a much earlier time than many people. I was bored while driving, and then later while walking to work, that's it! Like those are, and so you know, just for me, it happened pretty early on, like, we, my wife and I started listening to podcasts (before we were even married) on long road trips, and early days—like this, you know, this was like, 2005, 2006, really early days—we would listen to just the handful that were out then. You know, you'd have things like, I believe the TWiT network was operating back then, um, you know, stuff like MacBreak Weekly; I'm pretty sure that was, or at least This Week in Tech, I, I think those were there back then. I remember listening to an old show called The Word Nerds, which is all about like, you know, word origins and stuff like that. And, and back then the variety of shows, it was kind of all over the map, because they were small in number, but nobody really knew how to make podcasts yet. And so there were all these different formats people would use: some people would do it like a radio show where they would talk for a while and then just play a song, and then get back to talking for a while. And then like, it was, it, no one really knew what to do with the medium yet. Marco Arment 01:39 And then, my clearest memories when I was really getting into them, besides just long road trips, like when it became an everyday thing for me, was my early days at Tumblr, I was working in the city, and I would have, you know, a decent commute to the city, I would take the train and walk for like 20 minutes, that was 2006 to 2007. And I would listen almost every day to some kind of podcast on the way there, either You Look Nice Today, which was very, you know, very good, formative for me back then. I believe The Ricky Gervais Podcast was over by then but when that was on, I listen to that a lot, and The Stack Overflow Podcast, which is still going today. It says, you know, that, because that was when they were developing Stack Overflow. And they, Joel Spolsky and Jeff Atwood—the two main co-founders, I believe—um, they did a podcast where they were kind of just explaining their thought process of developing Stack Overflow, how they were building it, the choices they were making, why they were making it. Um and that was all, you know, super-early days. And again, I can't give enough credit to You Look Nice Today. I mean, anybody who knows me from ATP, if you didn't listen to You Look Nice Today, you might not realise how much of my editing style and how much of just my general, like, kind of podcast sense of humour, I just straight-up lifted from You Look Nice Today. Like, it was a huge influence on me and I think the whole world of podcasts around, like, my circle, now. Whenever I'm editing ATP, what I'm really doing is totally ripping off their style (in the context of a tech podcast). Martin Feld 03:02 I think it's really interesting in there that you said that you listened to podcasts with your wife. Was that correct? Marco Arment 03:08 Yeah, for long road trips, yeah, you know, we would try, we tried audiobooks here and there, but podcasts really stuck with us more. You know, we'd be driving driving across Pennsylvania, you know, takes forever! And we'd have a number of hours to spare in the car, and, and yeah, we'd listen to podcasts. And at the time, you know, the playback device was a little bit in flux at the time, oftentimes, we would be burning them to CDRs as MP3s and I had like an MP3 CD player in the car! And so that that was the early version, uh, you know, running through a cassette adapter into the, into the car's head unit! You know, that was the only version of podcasts; later on, I did eventually get an iPod—the iPod Video, so it was pretty late in the iPod age, really—and then, uh, soon after that I got an iPhone, and then that was it. Martin Feld 03:52 See that's really interesting to me because when you talk to a lot of people about listening to podcasts, it's a very (stereotypically) individual or solitary experience with their headphones in, but you were sharing it in a car with a loved one. How would you explain the kind of experience or shared memory that you would have of listening to podcasts together? Marco Arment 04:14 I think, I mean, you're right, that it is largely, I think, a solitary medium, but I think more people listen together in the car than you, than you might expect. Based on feedback I've gotten over the years as both a podcaster and as the developer of a podcast app, it seems like listening together in a car is actually pretty common. Um, now, whether both people in the relationship can agree on what to listen to, uh, or you know, whether one is being very kind to the other, that varies! And so from that angle, you know, I knew on my long car trips, I knew that, like, I would save the tech shows for when my wife was asleep or when I was driving alone, you know, because, just out of courtesy to her if she really didn't care that much about some of the, you know, some of the nerdery they were getting into. And so we would find these common shows that we would enjoy together that were, you know, things like comedy shows, storytelling shows, you know, obviously the old This American Life that was back then, Ricky Gervais again, that was, I would listen to that with her as well. Um, so there was a lot of that and largely at that time, I wasn't really needing to listen to podcasts much outside of that, because at that time, I didn't have a long commute. Marco Arment 05:21 Um, so my, my day-to-day, like, you know, going to work and back was not really listening to podcasts. And what I found, you know, as, as a person, and as a podcast, you know, business person, podcasts listening correlates pretty strongly to how people and when people are commuting. Like, for instance, like at the beginning of COVID, I saw a huge drop-off in usage. I also see drop-offs in usage on the weekends. Um, it just, you know, there's a there's a direct relationship between how much people are commuting and how much they listen to podcasts. And that isn't the only time people listen, but that is certainly a big time. And so I didn't have that back then, so I, my listening was more communal, like only during road trips, basically, the way many people listen to audiobooks today, you know, many people will will kind of save up an audiobook until they have a long drive to do and then listen to it then. Uh, that's kind of what podcasts were to me at the beginning. And frankly, there weren't that many of them, so I couldn't really, you know, it wasn't like I had this giant surplus of podcasts that I just couldn't get to all them in time and I had to listen all throughout the week; it was, it was more like the three podcasts I would listen to at any given time would release episodes every week or two, and maybe... and so I'd, like, save them up for road trips. Martin Feld 06:29 Yeah, and you mentioned the TWiT network for your personal tech podcast listening. Can you tell me more about that experience about listening to tech podcasts in that early time and what some of the shows were that you were listening to—what you got out of it at that time? Marco Arment 06:46 I know that there was, that like, there were there were radio shows, or cable TV shows about tech, you know, many times featuring like Leo Laporte and these, many of these same people, before tech podcasts were a thing, but I never had any of those, I never had access to them for whatever reason, either, you know, they weren't broadcast in my area, because I'm just from Ohio. It wasn't like a tech hub at the time, so they weren't broadcasting my area, or, in the case of cable, like, I didn't get those channels or we didn't have the right cable package or whatever. So, there really was no tech media that was available to me, except webpages, you know, like, you know, web publications, and print magazines, like that was where I got all my tech news before that. I had no tech audio or video or you know, TV, nothing like that. So podcasts at the time, and they still serve us well today to some degree, they were like one of the only ways I could get this information in like a dynamic form like that. And it was amazing to me! So I'm like, 'Oh, my God, I can be driving my car, listening to tech stuff!', like that concept was totally novel because there were no like, you know, AM talk-radio tech stations in Ohio and Pennsylvania. It just wasn't a thing. Marco Arment 07:58 So, it really served a wonderful role for me back then; it was so novel, that I could hear people talking about nerdy stuff, in what appeared, you know, in form, to basically be a radio show that was also on demand! I mean, it was it was magical that, this was one of the reasons why I think so many early podcast consumers were nerds like me, because most of us didn't have access to that kind of stuff. And it was novel for most of us and the technological barriers that were in place that made it a little bit hard for many people to access podcasts were less important and less restrictive to us, because we were already nerds, we knew how to operate nerdy stuff. Uh so, it played a huge early role in podcasting, for sure. And I think, you know, while podcasting has gotten a lot bigger today and in the intervening decades since then, tech, I think, is still a really strong core of it. And we're not the biggest by any means, you know, comedy and storytelling and true crime, those categories are way bigger than tech. And tech will never be as big as the mainstream stuff, you know, the celebrity podcasts now. Now we're back like in our corner, you know, but tech has always been like the stronghold of podcasting. It's been there since the beginning, it will hopefully always be there and nerds like me will always really appreciate it despite whatever happens in the larger podcasting landscape. Martin Feld 09:11 You've called yourself a nerd, which I love, and naturally, you have this history or passion for technology that led you to listen to and want to make podcasts for a living (that's brought you to this point). Can you tell me what that spark, that initial story, was for you with technology? What was your earliest experience that brought you into this lifelong interest? Marco Arment 09:35 You know, I don't, I'm not sure I can pinpoint it to a specific thing. I've always loved technology, both like physical and conceptual. Like, when I was a kid, my mom's friends would give us, like any, if they had like a broken VCR or something, they would give it to me so I could take it apart and just like play with all the stuff inside, like that's, that's the kind of kid I was, you know, I was like, you know, people would give me old stuff to just take apart and our garage was full of random electronic crap I had taken apart. You know, I didn't even have a computer until I was about 13, something like that. Um, 14? And before that, like I would use computers at other people's houses or in school computer labs to, you know, to the small degree those existed back then. And I was just fascinated by them, I loved computers, you know that any little bit of time I would get with one, I absolutely loved it. Marco Arment 10:21 I didn't know what I was doing on the computer, but I just loved doing it, I was one of the people who would just love to use it just for the sake of using it even though I had no real work to do on it. Like even my first computer, I didn't even have internet access for, for a year or two, after I got it; I would just play on the computer, I would just like, and not only games, I would like, open up MS Paint and just draw stuff. I would, you know, poke around the different programs on the computer and see what—what did these things do? You know, move files around whatever it was, I was just happy to use the computer. I don't think there's any particular inciting incident that made me want to be a tech person, I've just always been drawn to it. I've always loved it. And, you know, certain people are drawn to, you know, sports or different activities or different passions, and this was mine: I have no control over it and I'm kind of happy that I landed wherever I did, because turns out tech is awesome, and, and I was very fortunate that this thing that drew me in, that I was like, you know, born to do, happened to also have a career associated with it that I had access to. Martin Feld 11:17 And how did that initial interest that developed over time (and experimenting with computers at home and around family) how did that lead to developing skills in programming or software development? Marco Arment 11:29 It took a little while for me to figure out how, how I could access programming on my computer. I'd known from early on about the BASIC programming language from various sources, you know, back, this was, I was coming up in the, you know, this was like the early to mid-90s. And at the time, there were like magazines, like, there was a 3-2-1 Contact Magazine that, they would have BASIC programs printed, like on a page, as source code in the magazine, so that you could copy the code off this page, you could type it all in and run it on whatever computer you had that had BASIC. And I had, at the time I had, I had a Windows 3.1 PC and I didn't know enough about it to know how to access a lot of stuff and you'd turn it on, and it would just boot into Windows and that would be it! And Windows 3.1 didn't have any kind of obvious way for me to get to a BASIC prompt. And so I would like write down programs on paper, I just loved the concept of BASIC, I would get books out of the library on it, because I just, it just kind of seemed cool to me. I didn't really have access to it for a while, until one day, one of those magazines had printed, like, in the little corner of it: 'Here's how to access basic on...', you know, '...if you have this kind of computer, if you have this kind of computer...', and it's like, '...at the DOS prompt, type in "Qbasic"'. Marco Arment 12:40 And I tried that, and I'm like, ‘Oh my God! This is how to do it! It's here!' And so I spent the next couple of years really just playing in QBasic. You know, QBasic had, it was the DOS BASIC interpreter, and it was my first, the first programming language I had access to and that was on my home computer. I couldn't believe it! I could type in whatever I wanted! And it had built in documentation and everything because, you know, it was made, this was in the days where most things were totally offline. My whole computer was offline, I didn't have access to anything, so the developer environment was made to kind of be self-sufficient, to explain itself. It had all built-in documentation, all built-in tools. I understood almost none of the tools. You know, it had, like, debugging and building and everything. I understood, I understood none of that, but I had access to BASIC then and it just kind of went from there. I was. I was mostly self-taught until college. I would get a book out of the library sometimes; I would you know, get a magazine sometimes that would tell me certain things; but for the most part, it was banging my head against the wall in QBasic and you know, checking the documentation, figuring stuff out and very slowly becoming a self-taught programmer. Martin Feld 13:40 Wow, I love the idea that you were actually taking stuff from print. That's really intriguing. Marco Arment 13:45 Oh, it was terrible! Martin Feld 13:46 Terrible? No nostalgia involved? Marco Arment 13:49 If you made one mistake, like one typo, in when you were entering in the program, like that'd be it. Good luck finding it and you know, just try again! Martin Feld 13:56 Wow, OK, yeah, I can't imagine. But as you moved from that self-taught environment into college, as you said, in a more formal learning environment, what was that experience like, moving into a place where you were really, really focused on learning that? Marco Arment 14:11 It was great. I mean, and I'm not a good student, in most ways, like, I don't really, I'm not really motivated to do homework and stuff like that. I've always had problems with that. And there were certain parts about even computer science that kind of went over my head, or it didn't interest me. But I found enough that did, and it was amazing because, you know, I when I got to college, I had a whole middle school and high school experience of BASIC programming. I'd never done anything besides BASIC. I eventually moved from QBasic to Visual Basic, but that was that was my entire experience at that point. And college introduced me to C and the Unix stack and that kind of stuff. And it was, I mean it was it was very rudimentary, you know, what I was working on: I was working on these like old, you know, Sun SPARCstations, running Solaris and stuff like that. Eventually we moved to Linux, like in the, in the older years of college, but, but it was, you know, very rudimentary stuff. But I was writing straight C-code and getting segmentation faults and all the stuff that C-programmers do! You know, learning the basics of getting around a Unix command line, stuff like that. Marco Arment 15:15 It was very different than what I had done before with Visual Basic, but I found that I really enjoyed it. I loved working at a low level, I loved learning about this whole world that existed outside of the Windows world that I knew. This was before I was a Mac person, like, all I knew was Windows PCs, and then there was this whole world of all these cool like Linux and Unix programmers that were doing their own thing and didn't give a crap about my Windows PCs. And at first, I found it kind of off-putting, but eventually I'm like, 'Oh, wait, this is a lot of advantages!' Martin Feld 15:43 Mmm... Marco Arment 15:45 It was glorious, because that was my first actual education in programming. You know, my, at the time, most high schools and middle schools did not have coding courses. And now they're, now it's much more commonplace, thank goodness, but in the late 90s, when I was like in high school, that didn't exist, like it, at least didn't exist in Ohio, maybe other places—I'm sure California had a lot of them but Ohio sure didn't! Uh so, from high school, I was I was all self-taught. College was the first time when I had actual experts that were actually teaching me in a professional way, you know, how programming works and how to how to write in, you more, more advanced languages than BASIC and, and exposing me to the rest of the world of programming, besides just my little Visual Basic, you know, hole I was in. So, college was amazing for me, even though I was a C-student, just like any, just like the rest of my life. And I, you know, academically speaking, it was challenging as as any academic setting ever was, for me, but intellectually speaking, and especially in regards to computer science, I got a lot out of it. Martin Feld 16:47 And once that world had been opened up to you, exploring different systems, different languages, hearing experiences from different people, where did that new knowledge take you? Marco Arment 16:57 At first, it didn't take me very far because while I was good at the academic side of it, and I could code, which not everybody who who majors in computer science, or gets a degree computer science, not every not every one of the people can write code, it seems counter-intuitive, but it's true. And so, I could actually code and so that gave me an advantage in terms of like applicable skills in the job market. But what I really didn't know and didn't get, you know, a good head start on is how important connections are, and internships and stuff like that. Like, in order to get a job, you can't just graduate with a computer science degree and say, 'Alright, world! Give me a job!' You know, it takes more than that; you have to, like, get your name in places, use any kind of social connections that you or your parents or your friends might have, you gotta like, use them. 'Hey, do you know anybody, you know, looking for a programmer?' You gotta push all that. Marco Arment 17:45 And it took me a few months after college to really kind of realise, 'Oh, wait a minute, I have no leads'. I didn't do what I was supposed to do in terms of getting summer internships during college and stuff like that. I didn't do any of that. I didn't know I was supposed to. No one, like no one handed me a booklet saying, 'Here's what you have to do for, you know, to get your career started. Um, I was lucky that one of the members of my graduating class had recently gotten a job in Pittsburgh at this company called Vivísimo, it doesn't exist anymore, but they briefly made the Clusty search engine (which people might be familiar with) but their main product was, was enterprise software. And my, my classmate had gotten a job there, and she reached out to me and said, 'Hey, we're looking for somebody else, do you want to come interview?' And so, I got my car and drove to Pittsburgh. Thank God for her connection to this company, because I never would have found it otherwise. And that started my career in programming! Martin Feld 18:36 And what was that first role like for you? Marco Arment 18:40 Oh, it was amazing, because I got my butt kicked! It's 'cause, so I walk into this job... the reason I got the job was because part of the interview was a C programming test. And I had just done C a lot in college, I understood it pretty well, I wasn't making constant, you know, memory errors and stuff, the way a lot of people would if they weren't too familiar with the language, and I aced the programming test. And so they were willing to overlook my crappy academic record, and my lack of any internships and any experience, uh, because they said, so few people passed their their programming test. And it was, it was something like, you know, it's one of those standard, like, C-interview things of like, you know, write a program that, you know, counts duplicate input lines. It was something like that. And so there's a couple of opportunities to, like, allocate memory or fail to do so or do things wrong, and so they can kind of gauge, like: do you really understand much of this language or not? Marco Arment 19:28 And that job, it was incredibly formative for me because, first of all, you know, whatever academic path I took, I didn't learn about stuff like version control, and bug tracking and all those kinds of you know, I learned computer science, I learn algorithms and, and data structures and stuff like that. And so, that job taught me the basics of like, here's how you actually get work done in the real world with a team. You have the meetings that are, 'Tell me what you're doing', you assign tasks, you have this bug tracker that helps you organise things, you have version control to help you manage the codebase and everything. Um so that was great for that. And then also, I was working in C and low-level language with programmers who were just way better than me! (And of course, way more experienced as well) So I walked in there thinking I was a hotshot, because I was a 22-year-old who like, knew how to write C, I'm like, 'Yeah, I'm a programmer, this is great, I'm smart!' And I got there, and like, 'Oh my God, I got my butt kicked', because I was not nearly as smart as I thought I was. And they, in the most graceful way, formed me into a better programmer, and you know, with with much more experienced people, you know, being my bosses and my co-workers. So that, I learned a tonne from that job. And I was only there for two years, um but it was it was incredibly influential and formative. Martin Feld 20:43 And what happened at the end of those two years? Marco Arment 20:45 Um, f various reasons, uh, my wife and I wanted to move to New York. That start was in Pittsburgh, and and so we moved to New York, and I looked for a job on Craigslist, and that job became Tumblr. And then I was there for four years, and then I left that and became an indie app developer. That's, that's how I got here! Martin Feld 21:03 That's pretty amazing. That's a huge shift, and people listening naturally would know that Tumblr is a major part of your history. Can you tell me about the experience of working at Tumblr, what that meant to you and how it shaped you and led you to that indie role that you just mentioned? Marco Arment 21:21 Oh, sure! I mean, when I was first moving to the city, looking for a job, first of all, I had tried all the West Coast tech companies to see maybe I want to, maybe I want to work on the West Coast. None of them even called me back, like no, I couldn't get I couldn't even get a phone call from any of them. Except, eventually, Amazon did a phone interview: it took forever to arrange; after the phone interview, no one got back to me either way, for a few weeks; and then I heard the person I was interviewing with was fired. And then it fell on the floor and never went anywhere again! So, I didn't have any little any luck getting into the tech companies. West Coast is out then, we'll move to New York. Moved to New York, I had a couple offers: Bloomberg Technology was one of them; and what, the small consultancy called Davidville, that would become Tumblr. Marco Arment 22:04 And I chose David Ville over Bloomberg, because David was this cool, young guy who would let me work on a Mac. And I'm like, 'Well, this is great. I've been using a PC all this time, now I can work on a Mac, this is fantastic!' So, I chose that even though it was like it seemed really, like not stable or not, not safe, of a choice. You know, Bloomberg was this big company that had a big glass building and all these people inside who were really smart, way smarter than me, and it seemed like a really safe choice but the actual work I would be doing, I think, would be fairly miserable for me, because I don't care about finance. I didn't like the, the tech situation that they had going on there it was very, you know, PC-based a lot of like, super-old code, like I think like Fortran running on their, their, like main stuff. It was, it would have been a different scene for sure. Whereas Davidville was like, here, I'll buy you a Mac laptop, and we're gonna make websites with PHP. I'm like, 'Great! Sold!' Marco Arment 22:57 So I chose that instead. Um, and then, and Tumblr kind of came up from almost by chance. I mean, we were making websites for for other people as a consultancy, and we had a small gap of like, a few weeks between clients. And David was like, 'Hey, I want to make this thing; I've had this idea for a little while, let's build it in a couple of weeks, see what happens'. That was Tumblr! And we built it and eventually, we dropped, it took maybe six months before we dropped off our consulting work and focused just on that. Uh, but that's how that went. And then, because it was taking off like crazy, you know, David went out and got funding so that we wouldn't have to worry about where the bills were, how the bills were being paid. And it was, oooh, it was quite a ride. I learned so much there because here I was like, you know, a 20-something-year-old, working with a 19-year-old... Martin Feld 23:45 Mmm! Marco Arment 23:45 ...on this thing that was exploding in popularity and I didn't know how to scale stuff, I didn't know how to write, like the super-intricate, like, I learned along the way. Because it's like, you're going along riding a train, and you're like, you're building the track in front of you as the train is moving. Like that's, that's how it felt. We would face problems like, 'OK, well, we have this, they have this server and it's reaching its limits, we have to scale to multiple servers or spread this out somehow. How do you do that?' And we'd look at each other like, 'I don't know, do you know? No?' Marco Arment 24:15 Like, 'Oh, I heard this term "sharding" before—you want to look, see what that means?' Marco Arment 24:19 'Yeah, sure! I'll look it up!' But you know, it was that kind of thing. You know, we would occasionally meet with other people or you know, have, had like phone calls, like our our investors would set up like, we'd set up a meeting between me and somebody who worked at Twitter or like, tell me how they did stuff. You know, there'd be occasional help like that but for the most part, we were kind of ploughing through figuring stuff out on our own. And it was quite a ride. And it was very stressful but I learned a lot too. You know, David is David Karp, the founder of Tumblr, he's really a genius in a lot of ways, and he has incredible product sensibility. He's very, very good at making product decisions and product designs. He's also very good at marketing those designs and getting a lot of those details right. And I learned a ton from him on how to do all that stuff. It was very influential in my life, you know, both working with him and working on that product, you know, working on Tumblr. It was massively influential. It levelled me up so much that, you know, first of all, I never want to do it again, it was a lot of stress! But it, it levelled me up in a lot of ways that enabled me to do what I do now, which is, you know, make apps like that. Marco Arment 24:22 Without the Tumblr experience, I would not have been qualified to do this, not from not from a technical side, but from a, like, product-design side. I got a lot of good product sensibility from working at Tumblr with David and, and I wouldn't have gotten it otherwise. And so it was extremely formative, it was a heck of a ride, I mean, geez, the stress the, you know, the the ups and downs of that ride, it was a lot! And that's why I would never do it again. OK, and now I'm too old for that. Now, I'm like, I just, I'm almost 41, I'm like this, you know, this. I know that in in general terms it's not very old, but but in, you know, for like staying-up-all-night programming work. That is, you know, I'm out of that! Martin Feld 26:02 You've had that experience; you've addressed that. Marco Arment 26:04 Yes, exactly. That's, it's like, it's like college, you look back in college, 'Well, I would stay up all night partying in college and be able to go function the next day!' Like, that doesn't happen anymore when you're 40. And you know, similarly, like the the start-up lifestyle of that crazy roller coaster where you just you can never stop moving because if you stop at all, your site explodes. Like that, I'm done with that now but I had that, it was a fun ride. Martin Feld 26:29 Now I'm really interested in what you were saying in that story of Tumblr there: blogging in general, or a platform, like Tumblr, in essence, has some crossover or similar ideas to that of podcasting, the idea of sharing something about yourself or connecting with other people online. When you were building Tumblr together—through that stressful ride, as you described it—do you recall any ideals or aims that you had for Tumblr? What did you imagine that it would be when you were working on it? Marco Arment 27:00 The biggest, I think, shift that we had while developing it was when we were first developing it, it was initially made to just be a publishing platform, not a social tool. Martin Feld 27:11 Mmm. Marco Arment 27:11 The social features were added, like a couple years in, or like a year, a year or two in. They, originally it was basically just like a CMS that you would, it was just a free CMS; you'd go and make your own tumblelog (is what we called them), and and we didn't invent that term, like there were there were a couple of the tumble logs that existed before Tumblr. And David saw them and thought the format was cool and was like, 'I'm gonna make a tool to make this easy'. And then all the social stuff like with the dashboard and the re-blogging, all the stuff, that all came later. Uh, that was not part of the original conception of the product. And so part of it was we had to kind of grow into being a social product and this was at the time, you know, at this time, Myspace was already out and very well established, Facebook was already out and very well established. Facebook had kicked Myspace's butt by that point, but they were both still existing. LiveJournal was around, but it wasn't, neither of us really familiar with it. And there were a couple other, you know, social kinds of sites, Friendster, but it wasn't like today's environment of social networks. Twitter was around at the time, much, much, much smaller. Marco Arment 28:09 We didn't have the scale that we have now in social products. We didn't have the abuse, we didn't have the spam, we didn't have the harassment, the problems with things like racism and white supremacy; we didn't have that as much back then, to the scale it is now. Obviously, there was always some degree, any social product, you'll have some of that stuff but what you have today is this whole different ballgame. So we were able, back then, to be a much smaller team and to make much more, I think, risky, or bold product decisions, because it was a very, very small world compared to what we have today. And so, as we were growing this into a social product, we made a lot of decisions that were kind of against the norm at the time, you know, Facebook, this was a time when Facebook was really building its dominance. And Facebook was all about using your real name, your real identity, and using the product to find connections to people who you knew in real life. And David and I kind of thought that was gross, like, that there, you know, there was a place for that but that shouldn't be the only place, and that shouldn't be the only option. Marco Arment 29:12 For many of us, and this goes back to you know, being a nerd, for many of us the real world is not a comfortable place, or does not have a lot of the people that we want to hang out with, or expose ourselves emotionally and expose our ideas to or show our work to. You know, many people don't find their communities that easily in real life. And to the people who do, they don't even see that as a thing. Like, if you're the kind of person who can just like walk into any bar and make small talk with all the strangers and get along really well with them and have a totally easy time and not have any stress about that, that's great, and those kinds of people tend to think they're the only kind of people because that part of life has been easy for them. But a lot of people that's not the case for, including me and David! And so we built Tumblr, we designed everything from the start to be more of like a, like a nerd or artist or outcast kind of paradise, because that's what we wanted, because that's what we needed. We didn't want all of our, like people we knew in real life, you know, our family and our bosses, we didn't want them finding our blogs, so that's why everything on Tumblr is anonymous-first. You know, it's all about you can create your identity, it can be whatever you want and you can have multiple Tumblr blogs, and they can all have different identities. And if you don't tell people, 'This is my secret comic blog', or whatever, they can't tell, they can't, they can't figure out that to you. And we did things that we thought were socially-healthy for people compared to how things were done, like we did, it was a one-way following system, you know, like Twitter, but we didn't show follower counts in most places. We tried to make it as kind of low-key as possible, try to reduce the kind of emotional burden and emotional load, make it less about stats, and more about positive-only reinforcement. So we'll show you how many likes you get, we're not going to show you, we're not going to make a dislike mechanism, you know? You can follow people, but, you know, they can't tell that you're following them. And so you can unfollow them, that kind of thing. You can, you can use your real name or not, and most people don't, and it's fine. You can make your anonymous art blog, and no one's going to know it's you. Marco Arment 31:16 That kind of outlet at the time really didn't exist. Because you either had blogs like you know, WordPress sites, and everything, which you had some of that going on there but because it wasn't free to host them, in general, at least in most places, in most ways you had to pay, and Tumblr was free, so we got a lot of people who couldn't pay or didn't want to pay for stuff. And then we also got all the people who didn't want to use their real names, for lots of reasons, many of them very good reasons. So, it kind of made Tumblr a place for artists and outcasts and nerds; queer community really had a blast there and found it to be a safe space, which I'm very proud of, obviously, you know, not every story is that good, but I think overall, it was overall a safe space. And it was amazing to have built that for people, you know, I will occasionally hear people now talk about like the time they spent on Tumblr—I'm talking about it as if it's in the past, it's still around, but it is smaller than it used to be for the various various things that have happened—but people talk fondly about Tumblr as like, the safe space they were able to, like, hang out and be themselves when they were younger, usually. Marco Arment 32:21 And it's kind of funny, because most of the time period that they talk about is after I left. I was there from from its start in 2006 until late 2010. And then I left. And when you ask people what years they use Tumblr, usually they started after 2010. So, the version of it that I saw was very different than than what it became for most people, but even the version I saw, I, we were very careful to make it that kind of safe space for, you know, cool people who like didn't quite fit the mould of the buttoned-up Facebook of the world, where you had to have your real name and your parents would see everything you posted, and everybody, and your boss could see it and say, you know, you could get in trouble for stuff you posted on Facebook like—no, if you if you wanted to have like, you know, whatever, your cool anime blog on Tumblr, no one would know it's you, and that's totally fine, and you could hang and you could find other people like you, and this could kind of be your hang-out place. That's what it was. We were incredibly proud of that, and I don't think there was a lot else out there at the time that did that. Martin Feld 33:20 That's awesome. And I totally appreciate what you're saying about giving a safe place to people of the queer community or different minorities or even just not wanting to share your name, that's... Marco Arment 33:29 Mm-hmmm. Martin Feld 33:29 ...a very big deal and speaks to the kind of ideals of the Web or being able to communicate how you wish safely and comfortably. Something (or a few things you said in there)—words like 'nerd', 'artist', 'outcast'—it kind of lit up this light bulb in my head about 'the crazy ones' and the idea of using a Mac. And you said that you wanted to, earlier in that story just there, that you were keen on working at Tumblr because you could use a Mac. Now I'm interested in that tech history that you shared. What was that moment of going from that Windows environment that you'd been used to, to the Mac environment? Marco Arment 34:06 So, when I, uh, left college in 2004, I got my first Mac because I knew I'd be moving around a bit as I'd got my career, you know, settled, and I wanted a laptop and Apple made the best laptops. And I had some some graduation money from various family, they'd given, and it was just enough to buy a laptop. Like, alright, great, I'm finally gonna be able to afford to buy a Mac laptop! I'd heard they were really good. And at that time, I'd been using PCs since I was a teenager and I was kind of getting tired of how much work and maintenance Windows required to stay healthy and running well. You know, I would have to like reinstall Windows every nine months or whatever, like and I was just getting tired of it. And it was fun for a while when I was younger when I was in high school and middle school, that, you know, that was like a cool tinkering kind of thing. As I got older, I realised, like, I'm wasting a lot of time on this and I kind of don't want to be doing this anymore. I'd rather you know kind of move up the stack and start you know building, you know, spend more time programming and more time building stuff. Marco Arment 35:02 So anyway, I wanted to move to the Mac, I was able to do it with that college graduation money with the laptop, but I still had my PC desktop. And I would go to work every day at the Pittsburgh job and I would use a PC desktop there. And I remember there was this one day where Windows Update decided in the middle of the night at like, on my work computer, it rebooted my computer to install an update. And it lost all the work that I, all the windows I had open, it lost a couple of unsaved, you know, documents and stuff, it lost all to, it, because it forced, it force-restarted the whole computer—without my permission, without warning me ahead of time. So I walked in one morning, and my computer was all reset and it had a little, helpful dialogue: 'Your computer was restarted to install this update', or whatever. And I said, 'That's it'. And I unplugged the monitor from the PC tower and I plugged in my laptop, I connected the cable, plugged it into the DVI port on my laptop. And I was like, 'I'm just going to use my Mac', plugged the keyboard into it, plugged the mouse into it like you know, because at that job, fortunately, that was possible because what we were mostly doing was coding in terminal windows like using Vim logged into a Linux server that was actually hosting our development environment. So, all you needed was terminal windows, and email and like a web browser. It's like, 'Well, I have all those on my Mac laptop here!', so I just started bringing my laptop every day. And I started, I started doing all my work in the laptop. My work PC slowly, you know, went into disrepair, and I pretty much never used it. Marco Arment 36:25 And so then when I was changing jobs moving to New York, I had already been accustomed to working on a Mac from that time. I had loved it so much, but you know, in 2006, which is when I was looking for these jobs, not a lot of professional programming jobs would even be able to have you use a Mac; most of them were, were jobs that, that required Windows just for the environment you'd be working in—with the exception of web development jobs. You know, this was the time this was the rise of Ruby on Rails, TextMate, the nice, early Mac web-development boom. This was all going in 2006. So, I was able to kind of choose between like, you know, the the kind of old world of programming of working on some giant, you know, mainframe using a PC terminal with a little, you know, wedge of space on a cubicle line at Bloomberg, or, I could go work for this young guy in this weird cartoon office and he would let me work on a Mac that, which I had already been, you know, been kind of addicted to at that point. So, it was kind of a no-brainer, I was like, 'Alright, well, this weird cartoon job with this young guy', it was like, it was a little bit less money, and it seemed less stable. My mom thought I was nuts for, like, turning down a job at, like, this established company, Bloomberg, you know, to go work for some guy. I was like, 'I really think I will enjoy this better'. And, and I did and it was, oh it was amazing to be able to work on a Mac full-time. And David was also a Mac nerd, so like to be able to work in a Mac full-time and have like all the good equipment, like I got the big monitors and, and that was was right when the Intel transition was was going. And so I got, I got the first white plastic MacBook and it was so much faster than my than my G4 PowerBook. It was a great time to be a Mac user and I felt, I felt like I was getting away with something, like, 'Wait, I can go work at my full-time job... on a Mac... all day?!' Like, I'm like, I, 'This has to be some kind of trick! Like surely, I'm not going to get paid to work on a Mac all day!' But that's what it was, it was amazing. And at the time that, again, that was that was very rare, except in web development, and so that's why I went that direction. Martin Feld 38:27 And as you were diving further and further into the brand, consuming more Apple stuff, using more Apple stuff, what kind of news, or publications or other things were you enjoying as an adjacent part of that fandom? Marco Arment 38:43 The Apple tech-podcast thing was still pretty young in 2006/2007. Like that, that was the time I was I was more into like, you know, You Look Nice Today, This American Life, you know, that, The Stack Overflow Podcast, like that... Martin Feld 38:54 Mmm. Marco Arment 38:55 There weren't a lot of tech shows that really grabbed my attention back then. I think most of the tech-podcast renaissance happened maybe five or six years later, like around, you know, if you look at like 2010, you have things like, um, obviously, uh, 5by5 came out. That was a, I was part of it, so I, you know, I can't, I can't really take full credit but, I but I think that that formed a pretty strong community and moved tech podcasting into new audiences and new directions that, and so I think 5by5, you can't overstate how influential that was at the time. And then also, you know, the stuff that was going on with with TWiT, you know, getting bigger over time and everything. So the podcast world was being, was great at the time. Also, just you know, the web publications were really strong for tech, and most of them I don't even think are operating anymore. That was the era of the big, like, Weblogs, Inc. blogs, you know? You had like Engadget and, you know, Gizmodo, like that whole battle between those two things, like, all those sites started around that timeframe. Lifehacker was big back then. That kind of stuff! And the world of blogs, you know, podcasts were still in their infancy for tech stuff then, but blogs were, I think, in their heyday. That was like, that was the peak of blogs! Marco Arment 40:04 We had, you know, all of us nerds were using RSS readers and we had all these amazing blogs that were get, were launching and getting popular, and it was an incredible time for blogs. That's where we got all of our news. Digg was really popular, like, you know, you'd go, this is kind of where Instapaper came from, go and find like all the cool stories of the day on Digg. Reddit was just getting started around that time. You know, Hacker News came a little bit later, I think, but it was, you know, similar relative timeframe. So you had this amazing world of blogs, and blog posts and news articles, and opinion articles and stuff like that; they were all flying around and we were building up these great tools around them and these great networks, and you'd send traffic these these big, you get these huge flows of traffic from Digg and stuff like that. It was a cool time. And it was, it was an amazing time for blogs and the RSS ecosystem, that I really miss that now. And there's a lot of reasons why, why it went away, or got got much smaller. Some of them are good reasons, some of them are bad reasons but it's been a lot it's most multivariate—lots of reasons why, why that's, that happened. But I miss that world and I think back on it very fondly. It was it was really an amazing time for blogs and RSS. Martin Feld 41:16 And during this golden age that you're talking about, what was it like to make that leap to being an independent developer and how did you use that to enter the world of podcasting? Marco Arment 41:28 I'd been talking on and off with Dan Benjamin about, he wanted me to do a show on 5by5, and I was like, 'I don't know, I'm a little nervous, like I...' And the main reason we couldn't get it together was just scheduling-wise, Dan wanted to record during the work day, and I was still at Tumblr and I couldn't, because I was just too busy and that was never gonna happen. Like it was just, it wasn't gonna fly. So, I really couldn't do a podcast until, as long as I was full-time-employed. And around late 2010, Instapaper had been going strong, I had been doing Instapaper on the side, because it didn't honestly, it didn't take much work in the early days. And so I was just doing, you know, evenings and weekends, and it was really taking off, like when, Instapaper did pretty well on the iPhone, but then when the iPad came out in early 2010, it took it to another level because we had this amazing reading device in the first iPad. And that took Instapaper's business, like, through the roof. So that was when, in early 2010, that was when it started making enough money that it could be my full-time job. Marco Arment 42:24 And then in late 2010, Tumblr was going through some significant growing pains: it was about to expand massively into a huge, new office; hire a huge, huge amount more staff; and so, my role was going to change dramatically into being much more of a people manager than a tech manager (or than a tech contributor). And I'm not good at people management. And so David, David and I sat down one day, and were like, and basically worked out, like, 'Look, this was a good time for me to leave, because my job is going to change dramatically in directions that I really don't excel in or want to do'. And timing-wise it it was like a good time, we had some people who could replace me coming up, so it was a, it was a it was a good time for me to step out, and Instapaper was finally, you know, big enough that I could. So we decided, 'Hey, you know, let's part in good terms. This is cool, it's been fun. I'll step out now'. And so I left, late 2010, and I took Instapaper full-time and, and that, because I left my job, that gave me the time to start podcasting as well! So that's why around that same time, I think it was early 2011, um that I started Build and Analyze with Dan Benjamin, and that's what got me into podcasting, uh, as a producer. And this is, you know, I wasn't making a podcast app yet, this was long before that. But it was, you know, I was doing Instapaper full-time then from that point, and uh, being a podcaster kind of on the side, and it was a tonne of fun! Martin Feld 43:44 And what did you learn in connecting with people through podcasts and sharing your knowledge with (at least at first) an invisible audience? Marco Arment 43:52 Yeah, I had been writing a blog this entire time as well. And I occasionally I had dabbled in, like, you know, putting sponsorships on it and making it more, you know, more professional, like more of a thing that I did. And blogging, I still like writing as a medium, I do it a lot less now, at least publicly, but I do it a lot less. But the problem with writing is that the audience has different perceptions of your work and different reactions, different ways to react, different ways to spread your work. And you have to be very careful with writing because the medium suggests a higher level of seriousness and rigour than you might actually be putting into it or being or intending. So, I would find with, with writing, that if I if I made a little, tiny mistake, or if I wrote something a little bit the wrong way that I didn't necessarily think I was, you know, having a certain attitude or saying a certain thing, people would take it, you know, a different way than what I intended, which, you know, it's not their fault, you know, it's my fault for writing it poorly, but they would take it a different way than I intended and I'd get a huge amount of blowback from, from that kind of stuff. And it would just destroy, like, the rest of my week. Like, I would, I would feel so bad, I would try to, you know, write corrections or try to update... and it was tough. Marco Arment 45:06 Podcasting, I, I fairly quickly realised, podcasting didn't have that dynamic, nearly as much, it was almost none. Because podcasting, you're hearing a person talking, and you're hearing their tone, you're hearing their personality, you're hearing a lot of, a lot more context around what they're saying. And I think you inherently, by having more information about the person and what they're communicating, I think you give them more the benefit of the doubt. So if I say something that sounds a little bit wrong, if you take it a certain way, you're most likely not even going to notice, because you're kind of auto-correcting in your head, as you're hearing me talk, you're hearing, oh, this is a casual, unscripted conversation. This is this person with this kind of personality, like you kind of get all that, that doesn't come across in writing nearly as much or nearly as well. So, if I would say something in writing that could be misconstrued in some terrible way. Not only would the people who, who read it on a regular basis have a pretty high chance of having that, you know, having that interpretation, but then it would spread. And then, and people would like, you know, take it link to it, quote it. And then other people who had no context of who I was, or my personality, or what I was trying to say, they would then also see that in the worst possible way, and it would spread like wildfire and create this horrible dynamic. Marco Arment 46:27 Well podcasts, you have easier interpretations in the first place, you have people giving you more benefit of the doubt in the first place, because they can get more information about context and your personality and the environment because of that. And then also, it doesn't spread like that; there is no mechanism that's really effective and podcasts for spreading, you know, small clips of a podcast in ways that spread really easily and get retweeted and reposted and go viral and get somebody in a lot of trouble. That doesn't really happen much in podcasts. And so the downside of, like, how bad things can be, if you slip up or say something a little bit wrong, is much, much lower. Now, the corollary is the upside to how many people you can reach is also much lower, because of that lack of spreading dynamic. So it's a very different environment. Like, if I'm trying to cause change to happen, say I have some like, you know, big cause that I want to fight for or that I want to advocate for, or something I want to like persuade Apple to change some behaviour in something, writing is the better mechanism for that. Martin Feld 47:27 Wow, so you had this very early experience of positivity with the medium, particularly with Build and Analyze starting on 5by5. Where did this show and this very early positive experience take you? What were the opportunities that followed? Marco Arment 47:27 Writing is more effective, it can spread more easily, it can get further, but when I'm just having casual discussions or reporting on the news for the week and discussing it, I greatly prefer podcasting. It is way less time and effort, which is another huge benefit. But the audience dynamic of not assuming the worst of me constantly, is wonderful. And then even though the the height of how how far you can reach is lower with podcasting, because of the lack of spreading. The people who are there are very loyal. And they have lots of context of who you are. And so the audience that you get, it's going to be smaller than like a YouTube audience, a blog audience, like it'll be smaller, for sure. But it'll be a better audience. And I have found over time, I greatly prefer that, compared to trying to get as many people as possible to see this blog post I wrote, and then have it blow up in my face. Marco Arment 48:37 When I was on 5by5, you know, Instapaper was still a thing. I think I, I don't think I... let's see, I launched Overcast after... yeah so. So it didn't... what it did for me, it got me into the world of this community, like this, that, you know, that was still very much forming at the time, but still, the community that that kind of became my main audience, my main professional community, it got me into that world, it got me, you know, knowing these people, you know, meeting meeting the different in a different the different hosts on 5by5, you know, this is how I got to got to know like, Merlin and John Siracusa, like, it was a very informative, influential time for that. And it kind of got me into this world of podcasting of like that group, you know, because podcasting is a big world. And there's lots of different podcasts and lots of people who listen who have, who have no idea who any of us are, but this got me into that group and started forming that group. And so that, that was massively influential and, and formative, and then it also, it made me just get better at trying to communicate my ideas. Marco Arment 49:36 You know, when I first started doing Build and Analyze, I was so nervous every single episode. It was hard, like, to do it what is effectively a form of public speaking, every single week, you know, unscripted and mostly unprepared besides maybe an outline, it was hard, but it got me more comfortable with it over time. And I did that show for a couple of years but by the end of it, I was really comfortable in front of the microphone. And I was able to talk to more people, I was able to express myself more clearly, I was way less nervous. And yeah, and I just got better at it. And that enabled me to then take the next step of, a little while after that, after that show ended, doing Neutral with John and Casey, who, Casey I knew as a kid, John, I met through 5by5 and that whole scene, and then Neutral became ATP. And then that had its own, you know, slow build into what it is now. And same thing with ATP, like, I started out being a little more nervous, and you know, not as good at it. And over time, like we've kind of, we've established a really great pattern for ourselves, and a really great format for the show and a really great audience. And now we have this amazing community there. It all started with that little 5by5 community all the way back then. And we're kind of doing the same thing now, just better. Martin Feld 50:49 And for people who are listening, I assume they know what ATP or Accidental Tech Podcast is, but for those who maybe haven't explored it enough or have stumbled upon this episode, how would you define Accidental Tech Podcast? What is it? What has it become for you? Marco Arment 51:06 if you had to categorise it, it's a show about Apple and tech news. But podcasts, at least the kind of podcasts that I make, and frankly, the kind of podcasts I listen to, like, I don't I don't listen to most of the big podcasts, like, if you if you ask a random person on the street, what podcast they listen to—first of all, it's amazing that we live in a time when they actually know what that means, and that they probably do listen to something, that's great, 'cause that wasn't always the case—uh, but if you asked, you know, random people they listen to, they're gonna give you a list of popular shows. And whatever you can imagine is on that list, chances are I listen to none of them. What I listen to is shows like what I make, by people like the people who I hang out, hang around with on my shows. So you know, for the most part, it's other nerds talking about nerdy things—not always tech, but a lot of tech. So, you know, a lot of the Relay shows, still listen to Merlin a lot, you know, John Roderick, you know, like that kind of stuff, this kind of constellation of nerds and nerdy things that are kind of near each other. That's what I listen to. And that kind of show, unlike some of the big popular shows, this kind of show is so much more about the people than the topic. You know, it's way more about the hosts, the chemistry they have with each other, the personalities they have... oftentimes the diversions they go off, are oftentimes are more interesting or more entertaining than the official topic of the show. Marco Arment 52:31 And I think ATP is that same way, where our strength is that we are, we are ostensibly a news show, and we talk about news topics, but what our listeners are there for is us first and the topic second, because we've stumbled upon this good trio of hosts that work well, you know, we work well together, we have good chemistry, we're all friends, actually. And it works really well and people like hearing that. People like hearing conversations between friends that, you know, and it's almost like you're, it's almost like you're in the room. Like, when you're listening in the car to somebody's podcast that's of this style, you almost feel like you're hanging out with your friends, you're just not talking much. But, or you're yelling back at the podcast, and they aren't hearing you. But like, that kind of feeling, that's what I like about podcasts. So the podcasts that are not that, that are more like, 'Here's a scripted thing that we wrote with our staff of 20 people with an audio bed and these clips mixed in', that's a different style of show altogether. And there's a place for that and there's a huge audience for that, but I'm not that audience. What I want is nerdy people who are friends and have good host chemistry talking to each other about stuff I kind of care about. Marco Arment 53:37 And when you have that cool dynamic, where you like, you know, a set of hosts, and you like their personalities, and whatever it is, they can talk about almost anything at that point and you're probably there for it. Because that's kind of how a conversation between friends are! When, you know, if you are talking to your friends, you don't just talk about one thing with your friends in real life, you can talk about lots of things and you enjoy those conversations. That's how a lot of people listen to podcasts. They are there for the hosts, and the dynamics and the personalities, and the topics are secondary. So, going back to your question, what ATP is, is people who get together to talk about tech and Apple stuff, but we actually talk about a lot of different things. We go off on huge diversions and derailments, and it is on one level, infuriating, if you're only there for the news, and we hear from those people occasionally. But for the most part, we've lost all those people already, so the people who are left are the ones who were there more for the conversation and the personalities. So it is ostensibly about tech news but it's really a show about the three of us BS-ing with each other mostly about tech, but not always. Martin Feld 54:40 And it's really interesting and great to hear you speak with such enthusiasm about it and that you are all friends. How has it been to make that into your career? Is it a balancing act to have that fun mixed with professionalism alongside your development work? Marco Arment 54:57 I think for most people it would be. For us, we are just damn lucky that we can make the show... I think for all three of us, I think making the show is very, it comes very easily to us, you know, because for what, you know, for whatever reason, like, you know, if you look at, like, you know, musical bands that stay together for decades, what gets bands to stay together is like, they work really well together, they like each other, they don't have major conflicts, they can just work together nicely and easily. And then, you know, you look at bands that break up, and it's like, they didn't get along well enough. They didn't, it was, it was too hard for them or things didn't work out or whatever. ATP is easy for the three of us to make because we work together really well, we're all really laid-back and chill to work with. There's no, like, diva kind of mentality, there's no friction, we just work really well together. And then making it a business is similarly easy for us. Because we happened to luck into a decent-sized audience and because tech people tend to be on the on the, you know, middle, middle-to-upper-class range (on average) we are also very attracted to advertisers. And so we're, we've been able to fairly easily I mean, you know, it goes up and down. But overall, we've been able to fairly easily sell our ad inventory over time, and it makes decent money. And that's first of all a huge tribute to our audience. We, we're extremely thankful to them, but it also is just luck that like the demographic that we happen to attract with making a tech show happens to be a good audience for advertisers as well and the medium of podcasting. Marco Arment 56:26 You know, I was saying earlier about how the listeners you get, they're, you know, they're smaller in number compared to things like YouTube, but they're more loyal, they stick around, they listen, every week, they listen more intently. That also makes them great for advertisers. Because they are more likely to hear the ad in the first place, instead of just skipping over it or you know, zoning out and having it playing in the background not paying attention, they're more likely to hear the ad, they're much more likely to hear multiple episodes of the show over time, and therefore hear repeated ads, and that builds customer loyalty for the repeat advertisers. And they're more likely, if we say during the ad, 'Hey, we use this thing and it's pretty good', that also lends credibility to the product, and that makes the ads more effective as well. So all of that makes it a pretty good business for advertisers to advertise on our show, which makes it pretty good for us as a business then. Marco Arment 57:10 Um, and then secondly, we have again, that same, the loyalty people have to the show, you know, the the love people have for their for their nerdy stuff, that also has enabled us to launch a membership program, where we now have paid memberships. And you can pay eight bucks a month and get a couple of bonus things, ad-free version of the show and stuff like that. And again, like, you can't do that if your audience isn't, you know, (A) somewhat large, and (B) likes you a lot and is really into your show. And so, that, again, is mostly luck in the sense that, you know, we we serve a nerdy audience like ourselves, and they like us, and that's mostly just luck. And so running ATP as a business and making it a business, I'm sure we've, you know, we played a role, obviously—I'm not gonna, I'm not gonna say we didn't do anything to cause this—but it's also a lot of luck, that we happen to have a good audience in a good market that is worth money to advertisers and it just worked out for us. So, it's honestly been a very easy business. It's a decent amount of work every week, but still not like a crushing amount, you know? I'm still able to do other stuff, you know? So it's, it's a decent amount of work, but it's a great business and, and we love working together, with each other and membership has really taken it to another level in terms of like smoothing out the ad up and downs. And so it's been great and I really see no end in sight. Martin Feld 58:26 And you mentioned in there that you have time to do other stuff; you mentioned the app Overcast that you develop. I recall during your story also made references to things like the medium of podcasting doesn't really allow that viral spread that text can give you. When it comes to the infrastructure of podcasting and this other half of your professional career or development tasks that you set out for yourself, what can you tell me about your passion for or interest in the underlying technology and delivery of podcasting? What does Overcast and other software mean to you? Marco Arment 59:06 So in, as as I was going through my 5by5 days and everything, I slowly went from using, you know, I mean, originally when I was listening to podcasts originally on, you know, an iPod or burning CDs or whatever, then moved into the early iPhones. In the iPhones, it was literally just a tab in the Music app for a while. Eventually they broke it into its own app. And then eventually I discovered third party podcast apps that were just better! And the one I got super-into, Downcast was the one that ended up fitting me the best. I kind of went on that journey of going from the first-party app to Downcast during Build and Analyze, during my 5by5 time. I was working on Instapaper, so that was a whole thing, but Instapaper was starting to get, you know, kind of get out of my control. Like, that, it was needing too much, and frankly, the money was going down because I had a one-time purchase, and that works great for a little while and then it doesn't work that well anymore! And there was a lot of competition in that market and that was a whole thing. People want an Android app, people want... and so it was like, everything was was kind of spreading me too thin, I decided to sell it. Marco Arment 1:00:04 And around that time, I was like, 'Wait, you know, I've been using third-party podcast apps, and they've been fine, but I would do things a little bit differently'. And of course, I'm a programmer. You think about every every app you use when you're the kind of nerd like, every app, you're like, 'Well, if I made this app, I would do things a little bit differently than what they do', you know? But unfortunately, there's only so much time in the day, you can't make every app. Uh, but you can make one or two! And, and so as Instapaper was becoming too big, and I then I chose to sell it, I had an opening in my capacity. I had had a couple of ideas for what I would do with a podcast app, with Overcast, it ended up being Smart Speed and Voice Boost, and I made a prototype and it worked pretty well. Like I, I had a prototype that was running these processes on the audio and I'm like, 'Wow, this does sound a lot better, or this does make it a lot easier to listen'. And so I had those, those sparks of ideas. I'm like, 'OK, I want to make an app around these, I want to make my own podcast app'. I was listening to more and more podcasts at the time, they were really becoming big. And you know, any app that you use a lot, you know, you want to replace. So I decided my next app will be a podcast app. Marco Arment 1:01:08 And I also, you know, the podcast landscape at the time, it was the big app of Apple Podcasts. And there were a couple other things that were coming out, like um, Stitcher was out, and Stitcher scared the crap out of me for the future of the business, in particular, because at the time, all the nerdy podcast apps—Pocket Casts, Downcast, you know, all that stuff—they were all paid apps. This was like at the time of the App Store progress, where paid apps were still very commonplace, but they were just getting destroyed in the market by free alternatives. Again, Instapaper was the same problem. And also at the time, YouTube was really getting big, really, really big, and YouTube was basically and basically locked down video for itself. And Facebook was getting really big, and Facebook had almost entirely locked down the role of text publishing. You know, at that time, if you were writing a blog, or publishing a newsletter, whatever, you really depended a lot on Facebook to be the gatekeeper of all your traffic. And if you were making video content, there were other video sites before YouTube, or alongside of YouTube in the early days. Most of them are either dead or very marginalised now. If you want to publish video, you basically have to do it on YouTube; there were no alternatives that were viable, coming into, like, you know, the the 2012/2013 time. And I think that is largely still true today. I mean, today we have, it's a little more complicated with like TikTok and Instagram, but I think for the most part, YouTube is still extremely dominant in video. Marco Arment 1:02:35 And it scared me that video and text were getting locked up by these like, you know, single large companies. It scared me that it might happen to podcasting. And Stitcher had come out, and again, this is an area where you had Apple being the free app, built into the phone, and then all these other cool apps that were all paid that nobody effectively was buying compared to the free versions. Stitcher came out and very quickly got pretty strong market share, I think it was on the order of like five or 10 per cent, which is like, in a very short time. And the reason people were using it because it was free. Like, I would ask people who were using it, you know, why do you, what is that app and you know, why do you use it? And people kept saying over and over again, 'Oh, I'm using because it's free'. Marco Arment 1:03:16 And I realised, OK, the podcast, app environment needs more competitors in it, to make it more diverse, and they have to be free. And what scare me about Stitcher was that they were not being good participants in the open, RSS-based podcast ecosystem. They were proxying files and re-hosting them themselves and building their own little walled garden. Again, that scared me from the point of view of like, you know, somebody who had been through the decline of blogs at that point and the, you know, the capturing of video by YouTube. So I decided I wanted to make a podcast app anyway just for myself, and decided it's going to be a free app and I'm going to try to get as many users as possible, I'm going to try to go mass market with it, to try to stem the tide away from these things that are that are going to come in here for free and lock it up, like Stitcher. And so that was the goal with Overcast, was like: make it free up front, find other ways to make money with in-app purchase, or unlocks, ads or whatever, get out there and get something that is simple enough that people will want to use it. Because again, again, my app of choice at the time was Downcast, and Downcast has a tonne of features. It's a very power-user kind of app, but it really, it was very intimidating for average consumers, it, because it's a very complicated, visually very complicated app. Marco Arment 1:04:30 So I was like, I want to be able to have this kind of functionality, this level of power, but have it be simpler, so people can actually use it and have it be free, and also build it around these two prototypes, Smart Speed and Voice Boost, that I had made, that I thought were pretty cool. So that was how Overcast was born, and that's why those choices were all made, that's why it was free upfront, and because I was the first kind of like nerdy podcast app that was also free, I got a huge amount of market share early on. And eventually all those other apps had to eventually follow my lead and become free as well. And I didn't do that to attack them, I did it to establish a foothold in the consumer market, like, in like the regular-people market. And it's funny, eventually I did pass Stitcher in market share and that was a glorious day. And that's not because I dominated the world, it's because they went down, but still, it was like I, it felt very good to me! And it kind of was like a win for open podcasting. And that's ultimately what I wanted to do, is like, preserve open podcasting, and make a really cool app that I thought was, uh, that that fit my needs. And that's what Overcast is! Martin Feld 1:05:28 That's fantastic, and it's very interesting and useful, the way you've characterised it alongside the, as you say, fall of more open or competitive video and text. Aside from your own business success in this area, your role as a podcaster, your interest in producing this content, if someone asks you, 'Why is open podcasting important?', how do you answer that question? What is it about the openness and that basis in RSS that is so significant? Marco Arment 1:05:56 I think if you look at the Web in general, you know, what has made the Web—and people hardly even use that term anymore—but you know, the Internet... Martin Feld 1:06:05 Mmm... Marco Arment 1:06:05 ...what has made the Internet so powerful and so liberating to so many people, and so disruptive to so many industries, is that it's (for the most part) decentralised and that anybody can go launch a blog, or website or whatever, and have it be part of this big ecosystem that everyone else can access. And you don't need to like go make deals with all these different gatekeepers to get your stuff to be visible on the Internet, like you don't have to go work with all the backbone providers, you know, 'Oh you gotta go to like Comcast, and Verizon and UUNET', or whatever, you know whatever, whatever, all these people are! You don't have to, like work with, you know, go make a business deal with some big company, sign a contract, so you can you can deliver to their customers over their fibre lines, like, there isn't any of that with the open Web. And what that does, is enable much smaller entrants to the business to come in and participate with no real disadvantages over the big, the big entrants. Most industries don't work that way. You know, like, most industries, most businesses in the rest of the world are not that easy to get into and are, and are not that open and free to access. And so it's very important for me that the web works this way. Marco Arment 1:07:11 Unfortunately, the way the Internet has developed and the way you know, the way these companies have developed over time, it also makes it possible for somebody to build up a whole lot of audience in one place, and then for them to start dictating terms to other people to access that audience. While the infrastructure underlying it all—servers, IP, networking, you know stuff like that—that's all open and decentralised, but audience attention is oftentimes focused through gatekeepers. And it scares me that like, if I want to publish something, in some form, that this wonderful, open medium that we've had for so long in the Internet, that openness is being marginalised in certain areas, because, 'Oh, well, if you actually want to access anybody watching video, it has to be on YouTube'. Well, that puts YouTube in a position of being able to dictate terms to the entire publishing world. And we do see, like from other businesses, like I know, people criticise Walmart for using this term 'monopsony', which is basically like there's, you know, instead of being like one provider to multiple buyers, it's there's one buyer for all the providers, and like Walmart is the one buyer for so many types of goods, that they can dictate terms to manufacturers and kind of have have an overreach of their control, and have negative effects on the rest of the, the rest of the ecosystem. That same thing happens with content and publishing, where, if you have too much audience attention being gated by one company's power—Facebook, YouTube, Google, whatever the case may be—they can then dictate terms to everyone else who tries to make content in that entire medium and you kind of have to play ball. And I hate that on so many levels, and it's, to some degree, some of that's going to be inevitable and in certain markets, but podcasting has has resisted that way better than any other market has, way better than text, way better than video. Marco Arment 1:09:03 By the way, largely, that's Apple's fault. Largely, you know, Apple has a very dominant position in podcasting, but they don't abuse it, at least not yet. They never have. They're kind of like sleeping giants: podcasting is kind of too small for Apple to really try to lock it down and try to extract tonnes of value out of it that for themselves and make it worse for everyone else; but their position is so big, that other people really can't, like, break them up and take market share from them. So, podcasting has greatly resisted those efforts to be locked down or to have all the attention stuck behind one gatekeeper and have to dictate terms. It's great because I can still, just like, put up a website on a server anywhere and put up some MP3 files and an RSS feed, and I have a podcast. And I take a couple of steps, like, you know, submit it to the Apple Podcasts search directory, and then it shows up in every app. That's amazing, and that's how the Internet is supposed to work. And that's how many things on the Internet used to work, and now it's harder or more complicated, or, you know, there's not enough people in that mode to support it. Podcasting still is that way. And that's why it's just, it's so important for me to keep it that way because it's so powerful as a publisher, and frankly, as a listener. Marco Arment 1:10:20 You know, when things get locked up behind single companies, not only does that hurt the publishing side of things, in the terms of, you have to then work with that company, and whatever terms they dictate, you have to play along, if they're going to be middlemen between you and your audience, you're probably going to lose control, you're probably going to lose some money, you know, so that's, that's the whole thing. And then on the consuming side of things, you know, for the, for the customers, or listeners or users, they have little choice in how they consume things, either, like, if you are watching video, you know, you're gonna have to watch it on YouTube, mostly, and you're gonna then be, first of all, you're gonna be tracked, you're gonna be served ads, you know, all the stuff there, and you're gonna have limited choices in your client experience. You're gonna have to use their app in all likelihood, and, you know, whatever choices their app makes, you're stuck with those choices. Marco Arment 1:11:05 Well, when you use the open ecosystem of the Web, as your basis for things, you can, you know, the publishers can do whatever they want, with fewer to no middlemen. And then the users, they can consume that content in whatever way they please, as long as the publishers make it available in that way. So you have options, you have different podcast apps, you have different web browsers, you have different reading apps that can parse webpages, like Instapaper, and, and you know, serve them in different ways. You can read on your Kindle, you can, you can read on your tablet, whatever, like... you have the diversity on both sides: you have, you know, publishers doing whatever they want, not having gatekeepers getting in their way; and you have users being able to consume the content in whatever way they want, without having to use somebody's, you know, mediocre ad- and tracking-heavy app. And that's, that's a great place to be. And the more, the more spaces that we can preserve that still work that way, the better. And unfortunately, there aren't many left. But podcasting is one of them. And I intend to keep it that way as long as possible. Martin Feld 1:12:02 That's a super comprehensive view of how the Web should work and I really appreciate it. It's put a lot of your own story and the motivation for how you develop your app, Overcast, and how you produce the content that you do. I suppose, with that idea of how you want the Web to continue going, are there any other elements of podcasting's future that you would like to comment on, or are there any things that you would like to see in your own future as a podcaster and developer? Marco Arment 1:12:30 You know, back when Overcast started, I mentioned that Stitcher was kind of scaring me. Martin Feld 1:12:34 Mmm. Marco Arment 1:12:35 Today it's Spotify. Spotify scares me now, in podcasting. Spotify has gotten more control in the podcasting market than Stitcher ever had. And I don't see it, I don't see them ever losing that control. Fortunately, Apple still dominates, you know, they they battle with numbers and stats back and forth, you know, PR, there's a lot of PR that suggests that Spotify is the largest podcasting app, and it really isn't by far, but you know, there's like certain words you can use, like, if you put this word over on this claim, it makes it true or, but it's kind of misleading. Spotify has a good level of control, though, they have, you know, a strong market position. And they, the way they do things is much more locked down: it is not as much of a participant in the open Web; it is not as much of a good citizen in the podcasting world; and they are more about those levels of consolidated control. You have to use their apps, you have to listen to their ads, you have to have their tracking and their experience. And, frankly, their podcasting experience is not very good. Marco Arment 1:13:37 A lot of the, the entries in the podcasting space, over the years, have been people who have been told by their bosses or some trade magazine that podcasting is a big deal. And so they're like, 'Oh, OK, uh... we're gonna make a podcast feature!' And they do a podcast feature without really understanding it, uh, without really having much love for the medium at all or having much experience with the medium. And they make a really crappy experience for people. And Spotify combines a crappy experience with locked-down control. So of course, it offends me on multiple levels. Uh, so, I hope that they don't advance any further into podcasting and into their control over the medium. I know they're trying to, of course they are, they're a big company, that's what they do. You know, you show big company a way to lock something down and insert themselves as middlemen to make money off a whole industry and they're going to try for sure, uh, it's in their nature. They're gonna keep trying to insert themselves into the entire business of podcasting. They, I think, would love it if all of us podcasters had to work with them, had to use their platform, had to distribute to their audience, had to use their ad tools and their demographic tools and things like that. And I hope that doesn't play out that way. Marco Arment 1:14:47 So far, you know, they took a big chunk of share, but then, I think they stopped. I think they, what they could easily get kind of ran out. And I hope that's where it ends, but we'll see what happens. They are a big force to be reckoned with for sure. And secondly, I hope Apple doesn't turn bad. They kind of scared us a little bit, when, uh, about a year or two ago, they launched premium paid podcasts that are only available in Apple Podcasts. That was not a great thing for the ecosystem; it still remains not a great thing for the ecosystem; I'm still not not happy about for lots of reasons, but it didn't have a huge impact. You know, it wasn't like an extinction event for podcasters, and for independent podcasts and apps. But Apple could cause such an event if they tried. And hopefully, they won't. And I think that I think the people there have their heads in the right place with this. I'm not too worried. It's funny, like, only one time, I had a chance to very briefly meet Eddy Cue. And Eddy Cue is the Apple executive, who, I believe, is the top of the organisation that Podcasts is in. (I think.) Martin Feld 1:15:50 Mmm. Marco Arment 1:15:51 I was like walking past him at an event one time, and the one thing I said was, 'Please don't ruin podcasting'. I had, I had the chance to say one thing to Eddie Cue, that's what I chose to say! I think overall, their head's in the right place. I worry, as Apple tries to seek more and more services revenue, I worry they're going to try to push harder into podcasts, because most of the low-hanging fruit that they could pick service revenue from is already picked. So, they're gonna start, you know, it's like fracking, like they're, they're gonna start like finding different ways that are more destructive, to extract a little bit more service revenue out of different places. And at some point, some executive is going to look at podcasts, and be like, 'Mmm... maybe we can get more money out of that!' And I worry, the amount of power they have in the business, if it went bad, that could be very destructive to the podcasting world. I hope it doesn't go bad. And so far, it hasn't gone bad, but that's always a future risk. So I worry about those things, but for the most part, I'm very proud of our, of our industry here, for being extremely resilient towards attempts to lock us down. Many people have tried over the years, most have completely failed in a comical way. And the few that have gotten any inroads at all, like Stitcher or Spotify, for the most part, we've kept them from becoming any kind of dominant force. I hope that dynamic continues, but we'll see. Well, I mean, it's not it's not a fight that I intend to stop fighting, because I'm not I'm not going to turn my back on it for a second. But I think, I think our odds are pretty good. Martin Feld 1:17:21 No, that's fantastic, and thinking about your narrative, everything you've said about the industry and the medium of podcasting, is there anything that I haven't asked you in this conversation that you would like to touch on before we finish? Marco Arment 1:17:33 No, I think that's good! Martin Feld 1:17:34 There we go. Well, I want to say a big 'Thank you', Marco, for joining Really Specific Stories. It's been really intriguing to hear about your own narrative and where you think podcasting has come from and where it's going. Marco Arment 1:17:46 Thank you so much for having me. It's been a lot of fun. Thank you for listening to all my rambling!