The Aboard Podcast

Paul Ford: Richard!

Rich Ziade: It’s become your signature—Paul. How are you?

Paul: I get very Brooklyn when I’m around you. How are you today?

Rich: I’m doing well. How are you doing?

Paul: I’m doing fine. It’s busy day. A lot going on.

Rich: Busy day. A lot going on.

Paul: Good stuff. We are The Aboard Podcast. I’m Paul Ford, you’re Rich Ziade. And we have a software company that builds software using which technology?

Rich: AI.

Paul: Artificial intelligence, which, you know, gives people the willies.

Rich: Yeah. And we deliver AI into clients.

Paul: That’s right. We are a transformative AI play.

Rich: Absolutely.

Paul: And so today we have a rare treat. We have a guest who is a professor, a thinker, a writer.

Rich: Whoa.

Paul: An essayist. [Rich whistles] A literary person. And why in God’s name would such a person be on this nerd capitalist podcast?

Rich: Let’s find out.

Paul: Let’s find out. We’re going to talk to Andrew Leland in a second. We’re going to play our wonderful theme song, and off we go.

Rich: Let’s do it.

[intro music]

Paul: Andrew!

Andrew Leland: Hello. Hi.

Paul: Welcome.

Andrew: Hi, Rich.

Rich: Welcome.

Paul: Welcome to The Aboard podcast. We’re so glad to have you. For people who don’t know you, I’m going to introduce you for, like, 20 seconds and then you’re going to correct me. Okay?

Andrew: Yep.

Paul: You are a professor, a visiting professor at Wesleyan University. For people who don’t know, that’s, like, one of those, like, little liberal universities kind of somewhere in New England. That’s kind of most, that’s what—yeah. Which state is it in?

Andrew: Connecticut.

Paul: Yeah, we knew. I knew that. I just, I wanted, I was testing you. [laughter] And you are also the, you’re also an author and an essayist. I actually interacted with you many years ago when you were editing The Believer. And you have a very long range of work around a whole lot of stuff. But in the last few years, because you have a degenerative eye condition, you have been writing about blindness and what it means to go blind. And your book, The Country of the Blind, is really, really good. Like, and I say that not in my normal, like, podcast, oh, my gosh, I’m really excited you’re here. But, like, boy, it’s got good sentences.

Rich: Yeah?

Paul: You would like it. Richard hasn’t read it yet. You would really like it.

Rich: Yeah.

Paul: Rich is funny for a capitalist because he also, like, he’ll come in on a Tuesday and be like, “George Saunders is amazing.” [laughter]

Rich: Oh, Lord.

Paul: It’s real, though. No, no, you’re in a safe space.

Rich: You can do both of those things. I can make money and like George Saunders.

Paul: That’s—no one has ever said that before.

Rich: That’s possibly true. That’s groundbreaking.

Andrew: That’s how George Saunders survives, is that, you know, people make money reading.

Rich: Exactly, exactly.

Paul: There’s an angle for you on this podcast that I don’t think people might expect, which is, while you are a very thoughtful and careful essayist, a family man, a person who is narrating an experience of a disability, you were kind of a huge nerd already, but this situation has turned you into an unbelievable nerd.

Andrew: Yes.

Paul: And a very, very thoughtful product person.

Andrew: Hmm.

Paul: Building tools for yourself. And I think, let me just throw that back to you. So, like, we start as an essayist, we start as a professor, we start as a thinker, but you’re also now a very specific kind of technologist. And I would love it if you could just kind of talk a little bit about that and help the audience frame it.

Andrew: Yeah. I got my start at The Believer magazine where I was interacting with you, as you said. And I was managing editor for many years, which meant, I remember I dropped out of college to take the job. And Dave Eggers, who started that magazine, told me, like, just buy a manual for QuarkXPress and, like, read it on the plane as you fly from northeast Ohio to San Francisco. And I did that because I was like, basically, you know, the reason that The Believer could exist is the overhead was so low. There was only one full-time employee, and I was dropping out of college to be that guy. It was all done in QuarkXPress, you know, later transitioned to Adobe InDesign. But that was, I just, that was such a formative experience for me, reading a software manual end to end on an airplane that was going to, like, this is going to be my life. Okay, Let me, like, learn the keyboard shortcuts that will dictate my professional career.

Paul: I want to just pause for one second.

Andrew: Yeah.

Paul: And remind everyone that you just said that Dave Eggers started your technology career. [laughter]

Andrew: Yeah, yeah. I mean, he’s an interesting guy. Like, he would write—I don’t know if he still does this. I hope he doesn’t. But he would write novels in QuarkXPress, which is not a normal thing, right? Like, the usual workflow is—

Rich: Whoa.

Andrew: You write in a word processor and then export it like several iterations later into it. But he was a design, he is a designer as much as he is a writer. And that ethos, I think, was a big part of my education as an editor and as a writer.

Paul: I mean, he came out of, what was the magazine, I think it was called Might was his…

Andrew: Mmm hmm.

Paul: Yeah. So he was this sort of very San Francisco, almost kind of—it was, like, zine transitioning to magazine, very Gen-X thing.

Rich: Yeah, yeah.

Paul: And The Believer kind of came after McSweeney’s. Anyway, it’s a Wikipedia page that people should explore. [laughter] So that’s really interesting because everything with him was low-budget and kind of lo-fi, right? So he’s like, “Andrew, you want this job, you better learn the software.”

Andrew: Yes. And you can’t learn software on an airplane without a laptop, which I did not own at that time because, you know, I was a college student and it was 2003.

Rich: Those thick…

Paul: I remember those books, right? They were heavy.

Rich: They were heavy books and they were a good time.

Andrew: It was like reading a travel guide to a country you haven’t visited yet. Right? I was like, “Oh, okay, I guess that hostel sounds good, but I’m not really going to know until I get there.” Anyway, I got there and it was a technically nerdy job. Like, I did have to really get dive into the guts of Quark and then InDesign at the same time that I was, you know, writing copy and editing pieces. And I think that I start my answer to your question with that because in some ways, like, it was this lesson that there is no real divide for me between that kind of like wonky keyboard shortcut memorization and, like, understanding how a piece of software structures your art and your thinking and your craft all the way up until now. And blindness changed that. But it wasn’t as big of a change as you might think. I think it’s a similar dynamic that I’m experiencing.

Paul: Well, you know, QuarkXPress had, you know, tags and structured data and so did InDesign. Like, like when you dig into these tools, a lot of them were very structured and organized. And a lot of that three thinking was some of the earliest thinking that got translated onto the web around navigation and structure. So, you know, it’s funny because I think people could hear this and hear like, oh, well, he was putting together a magazine with desktop publishing. But the reality is a lot of that workflow stuff is very much like programming. Like, it’s a—putting together a magazine is almost a software process in a lot of ways. It’s just that the end result is a magazine instead of a piece of code. It’s a workflow. Anyway, so you were very workflow-oriented.

Andrew: I loved watching the little fetch dog run as I uploaded gigantic PDF files to the printer in Canada.

Paul: Yeah, that’s right. This is all I want to talk about.

Andrew: Yeah. That, nothing made me feel more like a, like a coder, or—not that I was even thinking remotely like a coder, but that was like, the most adjacent I got, HTML and so on. But all to say, yes, I was making a digital artifact, even though it ended up as a print product.

Paul: Okay. So on you go in your editorial career.

Andrew: Yes.

Paul: And have a lot of success. You’re writing, you’re doing all kinds of stuff. Blindness happens.

Andrew: Well, yes—

Paul: And you find yourself—okay, go ahead. Yeah, no, I mean, please, I’m sure there was more.

Andrew: I mean. No, the only thing that I have to interrupt you about there is that “blindness happens” is complicated, and—

Rich: Paul.

Andrew: The disease I’ve got—yeah, Paul, why are you trying to simplify my life? [laughter] Just kidding.

Rich: Finally.

Paul: Really, honestly, I welcome this. Let’s get into it. Okay, go.

Andrew: You will understand momentarily why this is important to me. So, RP, the name of the degenerative retinal disease that I’ve got, Retinitis pigmentosa, it goes different speeds for different people, but I have this sort of pretty much normal version, I think, that is the most common, and it’s just unbelievably slow. And I think this is important to point out only because it informs my perspective as the kind of, like, product, what did you say? Product user? Product expert? I forget what you called me.

Paul: No, no. You’re a product developer, you’re a product strategist.

Andrew: Product-adjacent.

Rich: Take it. Take it and run, Andrew.

Paul: Yeah, yeah, trust me, the money’s good. You want it. [laughter] Keep going.

Andrew: So that’s all to say that, like, you know, when I was working at The Believer, there was already things happening here where I had night blindness. I stopped driving at night, but I still drove during the day. And, you know, it’s basically slowly encroaching tunnel vision. So these days, I have a couple of degrees of useful central vision, but I still use them a lot, all the time to see stuff, particularly when I can control the visual elements.

Paul: Now you’re looking at us right now on Riverside, which is the podcast. Like, what do you see?

Andrew: Basically, I see very little at a time, but what I do see, I can see fine. So, like, if I’m looking at you, I mean, I also have zoom. You know, I can kind of control it, So I can zoom way in and actually see your face pretty well to, like, point at you in a lineup, if you’ve committed a crime. I think I’d be able to ID you. [laughter]

Paul: I’m glad that’s the use case.

Andrew: Yeah, I mean, I’m expecting that to happen at some point. I use a white cane, though, because, if I’m like, later today when I’m going to be navigating Grand Central, the idea of like, imagine strapping a toilet paper tube to your head and trying to navigate Grand Central, right, where you could only see like a pair of shoes at a time, it’s not going to work. So…

Paul: So nothing on the periphery. Nothing—so just, like, this, and then reading mostly audio, or do you like, what do you do?

Andrew: So, like, I can spot check text, especially if it’s large and especially if it’s white, a light color on a dark background, but that is not super sustainable. So if I’m going to read anything more than a couple of words, I’ve got audio going or I’m zooming way in, controlling the visuals. And that’s been a very, very slow process.

And so it’s, it’s, I think, yeah, I interrupted you only because, you know, in the reporting I did for my book, I talked to plenty of people who were in a car accident and, or, you know, gunshot victim and lost their sight instantly, and people who were born blind. I’m in this other weird category where I’m just very, very, very slowly going blind. And I think as a writer and as a product guy, which I’m not, but I like that you think of me as that, it’s interesting because it allows me to have a foot in both worlds. Right? I’m still able to do a lot of things visually and I still can kind of relate to software visually. But also I’m increasingly reliant on the accessibility angle. And that gives me, I think, yeah, I can kind of see it both from both sides in some ways.

Rich: Andrew, I mean, most people live with the condition, obviously, research the tools. Maybe their, their doctors will say, “Hey, you should use this tool or that tool.” As a product person who’s, the idea of product creation is changing over time. Do you feel like, hey, I can go make something or I can, I can have an opinion about the tools that affect my life? I don’t have to just wait for some product to come out. Talk us through how you think about this. Is it empowering or is it empowering and frustrating? Like, yeah.

Andrew: Well, first of all, like, the doctors do not tell you what tools to use. The doctors are, like, I am going to cure your blindness. And then if they can’t, they’re, like, “Dang, here’s a support group that you should go to.”

Rich: [laughing] Okay, that’s interesting.

Paul: Call this number.

Rich: Interesting.

Andrew: Yeah, exactly. And then, I mean this is, then you call the number and you might get, like, a blind—you might get, like, a very benevolent sighted lady or guy who’s like, “Oh, yes, like, here is the solution. You’re going to want to, you buy a gigantic closed-circuit television that will take up your entire bedroom. And that will be what you use to, like, painstakingly read books.”

Rich: Whoa.

Andrew: And what I realized as I kind of entered this world is that it’s a little bit like asking somebody where you should go to college. Right? Like, there is a lot of options. Some of them will be terrible options or not right for you. And to this day, and this is sort of to answer Paul’s question about, like, why losing vision made me into a nerd. Because the deeper I went, the more complex and interesting, but also, like, totally unsatisfying it became in terms of what the tools are. It’s not just, like, oh, you know what? You need a screen reader. Here’s the good one. Go for it.

Rich: Yeah.

Andrew: I mean, it could be if I wanted to make my life simpler. But if you care about this stuff, it’s not.

Rich: Let’s be brutally honest here. Is it just a lot of bad software out there? A lot of bad tools?

Andrew: Well, yeah, that is such a good question. I mean, one thing that I’ve been wrestling with is that as this process started, and I mean, and the process really is like, okay, I need to use a screen reader, right? Like, I need to have my computer talk to me.

Rich: Yeah, yeah.

Andrew: And I resisted it for a long time because I have vision. But, like, the thing that I experienced that so many people losing vision experience is, like, the longer you cling to the visual, especially if you have a degenerative condition, it’s just diminishing returns. And, like, I’ve met people who are, like, one highway-sign-sized letter at a time, like, trying to read anything like that. It’s just a recipe for frustration and failure.

Rich: Mmm.

Andrew: And the successful blind people—and a lot of my reporting and writing was, like, you know, ostensibly to write a book, but also I was just like, how did the people I want to emulate myself after, like, how are the successful blind people doing it? And you got to do the screen reader, basically.

Paul: So, so leaning into a screen reader was your way to kind of own the condition. Like, okay, here we go. This is, I’m going to get good at this.

Andrew: Yeah. There was a moment where it was, like, scary and hard and frustrating. But I kind of got into it in the same way that, like, I think anybody who is sort of technically minded or interested in software is like, oh, okay, I’m like cracking open the computer in a way that I haven’t previously. Like, I’m, I’m engaging—because the thing that a screen reader does, which I did not understand and I still, as I’ll soon reveal, don’t totally understand, it’s not just what it sounds like. It’s not like, oh, I’m going to, here’s a bunch of screen, here’s a bunch of text on the screen and it’s going to read it. Right? There’s a thing called, and you know this, and your listeners know this better than I do, like an accessibility tree. Right? That is sort of like pulling from the guts of whatever the system is, like whether it’s a web page or, like, an API or, I don’t even know what these words mean. What’s it called? The toolkit. The webkit. It’s got kit in the suffix.

Paul: I mean, there’s many kits, Andrew.

Andrew: It pulls from the kit, Paul!

Paul: Yeah, it’s from the kit.

Andrew: In other words, like, it’s generated. So it’s, like, okay, here’s all the buttons, here’s all the lists, here’s all the headings.

Rich: It’s trying to make sense of the order of things. It’s not just rambling out all the stuff and leaving it to you to decipher it. So there is a hierarchy, so to speak.

Andrew: Yes. And I think partially because I had vision, I have vision, and so it wasn’t, I wasn’t like, when things didn’t work right, I wasn’t, like, completely screwed in the way that many blind people are when something’s inaccessible. It sort of became, like, because I was writing a book, it kind of became an interesting project that I was excited. I was like, oh, okay. I’ve been a Mac user my whole life. But, like, most blind people are on Windows and therefore the screen readers on Windows are way more robust than the ones on, than what Apple offers for the desktop, for the laptop.

Paul: This is one of the worst things that this condition could do to you. [laughter]

Andrew: Yeah. Send me back to Windows.

Rich: To Windows XP.

Andrew: Which sucked. But also, like, I was learning and I, and I basically, I had this honeymoon period where I was like, this is leveling me up as, like, a person who’s engaged in technology and thinks about it.

Rich: Yeah.

Andrew: And then the kind of rude awakening that I think this is like a cycle I’ve been in where I’ll get jazzed about something and then be, like, ah, this is always—a wonderful person named Chancey Fleet, who works for the Andrew Heiskell Talking Book Library at NYPL, at the New York Public Library, she’s like one of the like great blind technology thinkers of our time. Always going to be a second-class experience on some level if you’re working with mainstream tools, even like the companies that are just, like, you know, the Apples of the world that are like doing these great commercials that show, like, you know, a university full of disabled people using their products and loving them. And the products are great. There’s always fundamentally a feeling of like, I am going through the side entrance here and it’s not…

Rich: How come? Explain that. I mean, there’s so much time and energy and money that’s been spent on building these tools. Why aren’t they great?

Andrew: I mean, I think one reason, I don’t know, I was reading something from a blind hacker guy last night and he was talking about the problem of keyboard input on a touchscreen. And there’s something just sort of fundamentally, like, the metaphor is fundamentally visual. Like, you look—

Rich: Sure.

Andrew: Like, the way that the touchscreen keypad, keyboard is built, you look at it, you type, then you get visual feedback that you’re typing the right letters, and it’s all visual.

Rich: Yeah.

Andrew: So there are ways to make that accessible and there’s been some ingenious ones. But his argument, and I think this gets to your question, is like, there’s just fundamentally, like, a mismatch. Like, if you were going to design a text input with, to be non-visual, that’s, you wouldn’t do, you wouldn’t build it like that.

Rich: Yeah.

Andrew: And so there’s always a sense that, like, you know, carpet laying over the nice wood floor that you want.

Rich: Yeah, yeah, yeah. You’re negotiating with these experiences that are assuming full capabilities in the user. And it’s always caked on, it’s never from ground up, because they are, they’re software add-ons, oftentimes on top of the world that’s been created.

Andrew: Yeah. I mean, and there’s this, you know, I don’t want to get too nerdy with history, but if you think about the history of computing.

Paul: Oh no, this is a safe space.

Andrew: Okay, okay.

Paul: Yeah.

Andrew: So like in the DOS days, right, where everything is text, blind people were kind of in their element. Right?

Rich: Mmm hmm.

Andrew: Because you could just, once you got that talking, you know, there are still problems on the command line about like reviewing text.

Rich: Yeah.

Andrew: But fundamentally, there was, like, everything was legible in that way. And then you had the GUI, the dawn of the GUI, and it was a nightmare. Right?

Rich: Yeah.

Paul: This just feels like it’s going to, you’re going to end up in Unix, if you, like, at some point.

Andrew: Well, I have. I mean, that’s the—

Paul: There we go. [laughter]

Rich: No, wait a minute. I mean, you must be over the moon with AI.

Paul: Well, hold on, let’s get there.

Rich: No, but like—

Paul: I want to get him to the terminal.

Rich: He’s made friends. I want to hear about his new AI friends. [laughter]

Paul: All right, so wait a minute. Mac to Windows, and this feels like you’re going to end up in the terminal, is that what happened?

Andrew: That’s ultimately what happened. Yeah. I’ll spare you every agonizing move, but basically I met a guy, another blind hacker, who I just, I’m very drawn to the blind hackers. And there’s a lot of them out there.

Paul: Where do they hang out? Where do you find blind hackers?

Andrew: Mastodon is a pretty good space for it, I’d say.

Paul: Okay.

Andrew: Listservs, email lists.

Paul: Yeah.

Andrew: Clubhouse was popular for a minute during the pandemic, but then that got shut down. You know, like, there’s a really cool documentary that is just about to come out called Joy Bubbles about one of the original phone freaks.

Paul: Sure.

Andrew: Who was this blind guy who could whistle the 2600Hz tone that would unlock the phone system, and he’s like a OG blind hacker guy. Really wonderful film. But, yeah, I met this blind hacker named Patrick Smith who, I was sort of talking to him about some ideas I was thinking of, sort of after my book came out and he was. He really just shoved me down the rabbit hole and he’s like, “Look, man, if, like, this is the journey you’re on and you’re thinking about this stuff in the way that you’re thinking about it, like, forget about—like, sure, Microsoft Word is accessible with the, with JAWS or NVDA, which are the two big Windows screen readers.”

Rich: Yeah.

Andrew: “But like, you should be in Markdown, man.” And I was like, okay, I’ll try Markdown. He was like, and if you’re in Markdown, you should, like be using Pandoc on the command line. And like, plain text is more accessible.

Rich: Yeah, he sent you down a path.

Paul: This is just exploitation, is what that was. [laughter] Which, good for him.

Rich: Yeah.

Paul: Yeah.

Andrew: Try Linux!

Rich: Oh, yeah.

Paul: What a gateway, straight into all of our favorite cults. That’s wonderful. Okay.

Andrew: Yeah, yeah.

Paul: So he took this nice literary man, and… [laughter]

Rich: Sent him off.

Andrew: He’s got a PhD in English. He wrote his dissertation in org mode in Emacs. And that’s where I ended up, for better or worse.

Rich: I’m just going to leave.

Andrew: I’m back on the Mac.

Rich: I’m just going to leave you two. [laughter]

Paul: This conversation just became Richard’s nightmare. I promise we’ll talk about AI in a minute.

Rich: No, no, no. Go, go, go.

Paul: Okay. So for people who don’t know, I am an enormous advocate of, like, just classic text editing. And Emacs has been a text editor I’ve used for 20 years. And watching Andrew get seduced by it gives me happiness in the same way that any cult member coming in and being new makes me happy.

Rich: Yeah. Yeah.

Paul: So welcome. I’m glad you’re here.

Andrew: Well, and it speaks directly to this conversation, too, because for all the reasons that you for 20 years have been enjoying it, those things are good for me, even if I’m not writing code, but from an accessibility perspective, I mean, on the one hand, it’s a nightmare. Like, if you want to think about accessibility in the sense of, like, I click on a download button and then I have an app that I know how to use. Right? Emacs is like your worst nightmare because it takes 20 years to learn how to use it, right?

Rich: Yeah.

Andrew: Or so. But for instance, there is no app I have ever used where invert colors works 100% of the time. Like, it’s really, glare is horrible for me. So it’s really important that, like—dark, I mean, these days dark mode is becoming hip, so it’s getting easier. But for a long time, like, inverting colors, there would always be, like, some bright bar of nonsense that I couldn’t fix. Or it’s same thing with making text larger. Like, you can magnify the whole screen, but it’s sometimes nicer to have, like, all of the UI elements large. Right? And that would always inevitably break things or, like, things are running over each other. Until I found Emacs, where it’s just everything is like, you know, Lego blocks that fit perfectly. Like, and not only that, but I can change the typeface of everything. Right? And the real joy is that this guy, I think also probably about 30 years ago, or 20 years ago, named T. V. Raman, who is a blind developer, he was at Adobe and now he’s at Google. He built something called Emacspeak, which is essentially like a screen reader that just works for Emacs.

Paul: He’s really legendary. Like, I mean, if you know about, you know about this tool, you actually end up learning about him.

Rich: Oh, no kidding.

Paul: Yeah. Because Emacs has been a real tool—because of him, it became a tool for blind users 20 years ago. Right?

Rich: Got it.

Paul: So it’s sort of baked into it as what it’s about.

Rich: Yeah.

Paul: So, okay, so now your world, and so, you know what’s interesting here. There’s two things. One is the same thing that got you into where you were, like, all right, I’m going to learn QuarkXPress from this book, is also you coming into software and technology as a blind person, or as a person losing sight?

Andrew: Yeah, yeah, yeah.

Paul: And then what we’re learning, and I think going back to Rich’s earlier question, is sort of like, what’s not good is that you need a very malleable foundation in order to adapt a system to your specific needs. Right? And so, like, MacOS just can’t be that malleable and be a consumer product that millions of people can pick up.

Rich: Yeah. I mean, and even if it was, it’s often hard to bend it. Like, you know, actually, Apple has invested a ton in accessibility tools.

Paul: They have a good rep.

Rich: And capabilities.

Paul: I think people don’t hate them, right? Like, in general, folks are like, yeah, Apple’s pretty good about this.

Andrew: For sure. I mean, the iPhone is, like, the number-one tool for a blind person.

Rich: It’s incredible.

Andrew: The MacBook, there are very few blind people who love MacBooks.

Rich: Interesting. Yeah. I actually met, years ago, the head of accessibility for Apple, and they were just about to come out with, like, that whole suite of features. Remember, people don’t often talk about it, but it’s a wide breadth of features in iPhones for people with disabilities.

Paul: I think sighted users don’t realize, because it’s such a visual medium, how much accessibility tooling is built into iOS.

Rich: Correct. Correct. But I actually, I have no doubt in my mind that they’re so, they’re buried four layers deep in settings. Like, unless you really sit down and learn it, they’re not immediately accessible. And there’s no, they’re not as malleable because they are a set of features. Whereas Emacs, like, Emacs nerds just love the control. Right?

Andrew: Yeah.

Rich: This is about control, to a large extent.

Andrew: Yes.

Rich: So it’s not just usability. Even if you’re good with computers or phones, you’re like, ah, you almost got me there. Like, it’s that—

Andrew: Yeah.

Rich: You know, it fell just short, which is normal. Totally normal.

Andrew: Yeah. And for me, like, as a reader and a writer, like, my relationship with text is paramount, right?

Rich: Sure.

Andrew: And I sort of got to a point where I could, like, see the workflow and, like, I could do it in Word or I could, you know, it was totally fine. But then when I started to actually, like, follow this guy down this rabbit hole, that feeling of control was intoxicating, because I realized, like, oh, no, I don’t have to make these compromises. It felt a little bit like the cartoon character who gets into the mech suit, where you can like, oh, now I can reach 40 miles with my precision syringe and go drill down right here. And that was transformative for me.

Rich: Yeah.

Paul: And now, so Emacs has got this very thorny language inside of it that makes it very customizable, called Emacs Lisp. But it’s not a gentle glide path. I mean, I’ll be very frank. Like, even as a big old Emacs nerd, I don’t like using it. Except AI shows up and now you can start to not just like—

Rich: Finally. Are we done with Emacs? Are we done? [laughter] Can we move on?

Paul: We’re never done, Richard. We’re never—it’s running on my Mac right now.

Rich: We’ve shed so many listeners for the last 15 minutes.

Paul: We just attracted so many listeners.

Rich: We may have attracted a new—this podcast is sponsored by Rush, the band. [laughter]

Paul: Yeah, that is true. The vector of, like, Rush fans and Emacs users.

Rich: Yeah, yeah, it’s strong. It’s a strong band. Andrew, I gotta ask.

Andrew: Yeah.

Rich: This new command-line transcription-based experience lands like a spaceship on Earth three years ago.

Andrew: Mmm hmm.

Rich: Yes, it’s not as buttoned up as DOS. It’s more patronizing, kind of clueless sometimes, and maybe damaging long-term.

Andrew: Yeah.

Rich: But were you ecstatic? What did you think about this? It has become the dominant interface of the last few years.

Andrew: Yeah. I first saw it through the blind communities I’m a part of, and a lot more people who have a lot less vision than I do, and the way I saw it was in the multimodal stuff, which did land like an asteroid in the blind community. Because, if you think about it—

Paul: Oh, it could describe a photo.

Andrew: Exactly.

Paul: Yeah.

Rich: Unbelievable, yeah.

Andrew: Five years ago, you’re, like, asking your wife to do it, right, or, you know, have to hire somebody or you’re outsourcing it.

Rich: Yeah.

Andrew: And I was just reading something this morning from a guy who’s, a blind guy who teaches blind people how to use technology, really beating the drum for AI and saying, like, if you’re teaching blind people life skills and you’re not including AI, like, you’re not doing your job.

Rich: Yeah.

Andrew: And one thing that he emphasized, which, that was interesting, was how patient the chatbots are. Which I think for, if you think about a blind person learning to do anything, you’re doing it in a way that is, like, fundamentally counter—not counter, but it’s just utterly to the side of how the rest of the world does it. Right?

Rich: Sure.

Andrew: Like, non-visually, there are very few things that, like, the mainstream does non-visually, right? And so to have a chatbot that has all the power that you’re talking about, right, but also that is patient in that way and will be, like, let’s try it again, that’s huge.

Rich: Sure, sure.

Andrew: But, yeah, the multimodal thing, like, the independence that that gives you is tremendous.

Rich: Yeah.

Andrew: And now with wearables, like, it’s not just describe this image, right? But it could be like, I’m at the airport, like, tell me the gate numbers that you see here, and people are just like, wearing Meta Ray Bans and cruising through airports independently.

Rich: Yeah.

Andrew: And it’s a game-changer.

Rich: What do you think? Do you use them? Do you like them? Do you use Meta’s Ray Bans and all that?

Andrew: So I’m not using the wearables and I still have enough vision that I will just zoom in or take my glasses up and smash it up to my face if I want to see something that bad.

Rich: Yeah.

Andrew: But I like to know that it’s there. And, like, Chancey, who I mentioned before, she reads graphic novels, novels with multimodal AI, which I just think is wild.

Rich: Wild.

Andrew: Like, she’ll open a page of, like, an Alison Bechdel graphic novel and say, like, “Okay, hit me with page 23.” And then she’ll, you know, she tunes it.

Paul: That last Alison Bechdel book is really good, by the way, if people haven’t picked it up.

Andrew: Oh, really?

Paul: Oh, it’s all about hippies in Vermont. And it’s just brutal and hilarious. It’s really good. [laughter] Anyway, it’s called Spent, I think. Anyway, keep going, keep going.

Andrew: I was just going to say that for me, it really started to feel powerful, and Rich is going to be unhappy with this answer, but I don’t know Emacs Lisp. I know enough to just barely open the front door. But with Claude Code, I could say, “Hey, this package is broken. Fix it.” And it does it. Or blind people, the blind vibe-coding revolution is upon us for sure.

Rich: Tell us what that’s like. What is that?

Andrew: For one thing, not to just Emacs it out forever, but the open source aspect of it is pretty important. So, for instance, you know, the great thing about Apple is that VoiceOver, the Apple screen reader, just comes built-in on every Apple product you buy, whether it’s a watch or the phone. You don’t have to pay for it. In the past, like, and this is a thing in accessibility in general, like, if you want an assistive tool you have to pay thousands of dollars because the economies of scale are terrible because there’s so few people buying it and Apple makes it for free.

Rich: Expensive, yeah.

Andrew: Yeah. And there’s a, there’s, like, a proprietary screen reader for Windows that’s one of, like, the most venerable and powerful ones called JAWS. But now there’s this open source upstart called NVDA that is really taking a lot of the market share from JAWS. And it’s free, it’s open source, and it has this add-on library that’s all in Python that, and you can write your own add-ons. And so blind people are just going bananas writing NVDA add-ons.

Rich: Amazing.

Andrew: You know, Wikipedia, for instance, is a totally accessible webpage with a screen reader. But it’s also, like, not that easy to just sort of do the thing that visual users do where you, like, glance at the headnote and like glance at the TOC, and, like, you can, you can get fast with it. But this guy just, you know, had an LLM write him an add-on for his screen reader that, like, basically scrapes Wikipedia and, like, Cold-Stone-Creameries it up into, like, his particular confection that he wants, and then he uploads it to the NVDA add on store and now like, you know, lots of other blind people are using it and there’s just like. That’s just…

Rich: Amazing.

Andrew: And people are building, you know, mods for games that are inaccessible. So I think there’s two angles of it. One is, like, here’s an inaccessible tool and it’s going to, like, remediate it. And the other is, like, here’s an open-source tool that could sort of, like, slot into my other open-source assistive tech, and that’s the sort of NVDA or Emacs.

Paul: See I want to, I want to break that down for one second because something you said really sticks in my head, which… So Wikipedia is a lot of very big web pages and they are, they’re very accessible by all the, like, of course Wikipedia is a big not-for-profit. It follows all the guidelines and I’m sure there have been lots of disabled people advising over the years, right? But what this person needed and what they really wanted was just a really scannable version. Right?

Andrew: Mmm hmm.

Paul: That was not on the product roadmap for Wikipedia. I’m sure people have thought about it before, but everybody just kind of, you know, they just kind of go along, they kind of figure it out and, you know, everybody was sort of doing the best they could, but now this person can go, “Wait a minute. I just want this. I just want it. Let me see it. Let me see what it would look like.” Wrong choice of words.

Rich: It’s empowering.

Paul: But the—

Andrew: No, it’s fine.

Paul: [laughing] Okay, but he wanted to perceive this thing and use it and, and have it. And as a result, everybody gets to have it. And that, to me, that is the moment where—and I want to talk to you about a slightly different thing, too, which is sort of like me, you straddle two worlds. You are in—well, and actually you straddle two very specific worlds. I straddle technology and sort of culture. But you are of a very specific part of the culture industry, which is very thoughtful, very caring, very sort of essayistic, very sentence-focused. Those people all hate AI, top to bottom. [laughter] They want it gone and killed with fire.

Andrew: Yeah.

Paul: And on the other side, you’re in the blind community where really must feel good. It just must—

Andrew: Yeah.

Paul: I can feel the relief as you’re, because you have control over your own life. And so you have one community that is losing control because of this technology, and another one where it’s so foundational that it changes your day-to-day living.

Andrew: Mmm hmm.

Paul: And sort of—how do you, do you organize that in your brain at all? Or are you just sort of, like, I’m sorry, I’m like, I don’t have a lot of vision. I’m going to just use this shit. Like, what, how do you react to that split?

Andrew: Yeah, that’s such an interesting question. I mean, my wife is an English professor and she’s, like, I have to, like, you know, there’s a thing called Screen Curtain with a lot of assistive tech where you can, like, basically make your screen go black, but you can still hear it. And, like, I will turn on Screen Curtain if I’m using Claude and she walks by because, like, she will just like, I’ll get in trouble. You know, like, she really, she hates, she’s one of those kill-it-with-fire people.

Rich: Oh, interesting.

Paul: Sure.

Andrew: You know, I think, I think the way that I connect those two worlds, for me, it’s not—I definitely have the thing you ended with, which is, like, you know what? It’s, like, I gotta get out the door somehow. It’s, like, the same impulse that makes me stick with Amazon. Like, I feel bad ordering tiny things from Amazon, but also I’m, like, well, I don’t drive. And, like, but is that, am I just making that justification or, like, is it really a blind thing or what?

Paul: No, no, let’s just… [laughter] We’re gonna give you that pass.

Rich: We’re using Amazon on, too, Andrew, if you make sure any better.

Andrew: Yeah.

Paul: Yeah, probably—this podcast is probably being recorded to an Amazon server.

Andrew: Yeah.

Paul: Like, you just, you can’t own a structural problem. It’s okay. [laughter]

Andrew: Yeah. Okay, thank you. But I mean the thing that I wrestle with, and I’m still not sure how to articulate is the, like, the idea of a mainstream product versus a niche product. And I think, like, the thing that justifies it for me is, you know, you were, you were mentioning earlier that you liked it when I said that you can really tell when a blind person has made a piece of software.

Paul: Yeah. And I really want you to talk about that. I want to understand the difference. Go ahead.

Andrew: Well, I was just going to say that, like, so many of these, you know, like, the Wikipedia-scraping add-on, like, that’s never going to be a feature that any mainstream tech company is going to make. Right? And like you can have the handful, like, you know, the big tech giants have blind developers and disabled developers and they have accessibility initiatives, but it’s always going to be that sort of side-door thing or, like, an overlay kind of approach. I mean, not always necessarily. Those companies like Google and Apple and Meta are all doing really good things for accessibility.

But ultimately I feel, like, I saw this developer who has a mobility disability and scrolling is really hard for him and he talked about how a friend of his hacked together like a cool scrolling app that made him able to scroll more well on his Mac, but it also sort of didn’t do what he wanted. And then Apple actually has something called Dwell Control that, like, was supposed to solve the problem, but like it didn’t work on all the websites. And then Claude Code, you know, he just like in a couple hours he made it and now he can, like, scroll with the best of them.

Paul: I mean, listen, Andrew, I can give, I can give this back to you and it says you’re being very polite here. And look, I think they are, people are, have learned a lot about accessibility and they’ve learned about disability. Right? And they brought in teams and they’re, they’re making, but you’re never the main event.

Andrew: Yeah.

Paul: And what you’re saying is, boy, it’s really good to be the main event sometimes. It’s good that this product is for me.

Andrew: Yes.

Paul: And when a blind person makes something that I can use, I can just tell and I can feel it because they’re the frickin’ main event. And that to me is always the miracle about this work is that you can solve a problem for an individual and then that can scale out. And you’ve been down on the list. You’re on the list, but you’re down on the list.

Rich: I mean it’s, they’re literally—

Paul: And now you’re up.

Andrew: We’ll never be at the top of the list. Like, there’s, there’s never gonna—

Rich: It’s a numbers thing.

Andrew: Unless there’s some, like, Jose Saramago event where, like, you know, three quarters of the world population is blinded? [laughter] We’re never gonna, that’s never gonna be the case. And so, but I don’t want to, I don’t want people to hear this and think like, oh cool, like, AI solved accessibility then. So, like, for my next product, like, we don’t need to think about it. You know, I think, because I think that, like, I do have the experience, like you know, say what you will about Facebook, like, there’s a lot of blind people on Facebook and they yell and scream and stamp their feet when Facebook pushes an update. Like, I just got like a really excited email this morning from the American Council of the Blind that was like, “Great news!” And I was, like, oh, is this like some legislative thing? And it was, like, “The Facebook, the new—if you update your Facebook app, like, the back button works again. Fuck yeah! We did it guys.”

Paul: Yeah!

Andrew: And I was like, wow, that… But, but also as a user, like, Facebook is a pleasant app to use with VoiceOver. Like I, when it works, like, when they push those updates?

Rich: Yeah.

Andrew: Like, compared to Twitter, for instance, or like, a lot of other apps, like, it’s just nice. You can, like, interact with the posts. You can like them easily. You can, like, scroll easily, skim easily, all of that. So, like, that stuff is really important and real. But at the same time, like, those regressions are just, like, a fact of life. And so this, you know, this is, and this is where I get all Linuxy and, like, open-sourcey. I’m, like, maybe I just divest and, like, what if I could just like live with the weirdos where, like, the back button breaks every day but, like, I can fix it myself and I’m, like, with these other weirdos and that’s the kind of, like, back and forth.

Paul: Yeah, baby, come on in. Come on in. [laughter] We want you here. We want—the weirdos want you.

Andrew: Good.

Paul: Listen, you know, I mean, there’s another thing I’ve heard, too, which is, it’s very easy to be like, make this app accessible. You can say that to Claude Code, and it’ll, like, do an okay job, but from what I’m hearing, it’s just kind of not there. Like, it’s not really… It won’t quite get you there. You still need real users and you need real review.

Rich: I’m going to be, I want to, I want to close this podcast with a futurist sentiment. Going beyond the, this community or other, you know, communities of disability or frankly any community. Communities that have, that just don’t get attention because they do the, like, total addressable market sort of cold calculation and are, like, it’s not a big enough market.

And the truth is we’re heading towards a place where a bespoke solution to a problem doesn’t require buy-in at a large scale because you can just do it. Everything I’m hearing is born out of the community itself. Right? Not some R&D project at Microsoft. And it’s oftentimes driven by a personal frustration, an individual frustration that is, like, maybe others are frustrated like me.

And oftentimes it doesn’t hit the mark. It happened to work for that person who built the add-on, but missed the mark for other shades of a condition or whatever it may be that people carry. I think we’re heading towards a day—and this is the futurist piece I’m going to say right now, so put the music in, whoever’s producing this— [futurist music starts]

We’re coming to a day where something that is so custom tailored to a high fidelity around your individual needs can come to life for you. And others will be like, that’s weird. That’s not how I would do it. Well, go make your own because it’s going to be easy. And the empowerment here is in that ability to sort of address, it’s almost like targeted therapy. Right? Like, you have a particular set of needs. It’s not just about your condition. Andrew, you also are a writer. You’re not doing something else. You actually those, there’s aspects of your life that are meaningful to you—

Paul: They have a lot of things that they need.

Rich: Go. Yeah. [music ends]

Paul: And they never get them. [laughing]

Rich: I think it’s an exciting time for the tinkerers who are able to build stuff that empower them and then maybe share those things with the world. I think that’s a very interesting… We’re in a very interesting time.

Andrew: Yeah. And I think that your point about the sort of diversity within disability is important from a product standpoint because, you know, like, I talked to this blind guy who was, like, on the very non-technical end of the spectrum and he had the ear of his, like, local zoning board or whatever and he was like, you know, “Blind people really like this kind of asphalt and when you fill in the potholes with the other kind, it’s not good.” And I was like, “What are you talking about, man? Like, this is just like, that’s a you thing.” [laughter] But now, like, the whole, you know, the whole community council is like, “Okay, well the blind community really likes this style of asphalt.”

Rich: The blind community is the guy.

Andrew: Yeah! And like, you see this the same thing, like, there are people who never identify as blind their entire lives who lose their vision in their 80s.

Rich: Totally. Totally.

Andrew: That user is radically different than, like, the blind hacker who’s blind from birth, who’s been, like, Unixing it up since childhood. That kind of malleability. I mean, that’s one of the things that makes accessibility hard. Although I think the main thing that makes accessibility hard is just that people aren’t teaching it in schools and it’s not on people’s radar. But even when you do kind of want to make a good-faith effort, often you’re, like, doing great for the low-vision people or the colorblind people. But like, you know, the people who are using joysticks or a different screen reader are left out.

Rich: In different environments and different cultures and different professions that have different requirements that that sort of layer on. I think, look, I think generally speaking, we talk about it on this podcast, we’re heading towards a post-product world with this stuff, because the ability to sort of assess a situation and build something that’s utterly custom tailored and has a customer of one.

Paul: We just made Andrew a product manager and you’ve already—

Rich: Welcome to Aboard, Andrew. [laughter] Yeah, yeah. This was an interview process.

Paul: It’s over now.

Andrew: This was my interview.

Pul: No more roles. No, you know, here’s. You know what’s funny is, I was thinking, like, two years ago, I think I would have been like, “Hey, Andrew, I got an audience of product managers for you and you can tell them what you need, you know, and maybe they’ll listen and this will be good. It’ll be a nice educational moment for everybody,” on and on. But I think what I’m hearing now is you saying, “Hey, if you could just, like, make sure I know how to talk to your thing and then kind of get out of my way a little bit?”

Rich: “I’ll take it from here.”

Paul: “I’ll take it from here.” And I think, like, I really do want to get that into people’s brains. You’re ready to go. You’re ready to tell Claude what you need. You just need some good APIs and some nice documentation for the LLM to read and you’re kind of up to try stuff.

Andrew: But that’s me. I think if we’re talking about the community, you have to be ready for the 80 year old who lost her vision.

Rich: Yeah.

Andrew: And needs a thing that will talk to her.

Rich: Yeah, she’s not getting on Claude Code.

Paul: But Andrew, five yous can help 80,000 of them. Like, it’s really, because you could build like you’re writing, you’re busy, you’re teaching.

Andrew: How?

Paul: Okay. But like let’s say, let’s say this was your job. You could solve problems at a scale very quickly for a lot of people with low vision using these tools all day long. Like, the scaling of somebody who actually has the experience and doesn’t constantly needed to need it sort of translated to them.

Rich: I partially agree with you because the example, the Wikipedia reader add-on probably solved it for, let’s say hundreds or thousands of people beyond that individual. But I think there’s a diminishing return on it. I think that this is kind of the, the challenge with this is that how people live their lives, the state of their condition, and how it intersects with how they live their lives and their professions or whatever.

Paul: Yeah, but think about it—let me just, like, just a future state, a couple years from now.

Rich: Yeah.

Paul: Ten years from now somebody develops a condition and they’re starting to lose their eyesight and it’s been going on for a while. And then a service pairs them with a developer who hangs out with them for a day or two.

Rich: Yes, as a service, and then makes them a tool for them.

Paul: In a day or two, can build them five or six key pieces.

Rich: That’s fascinating.

Paul: Uses their smart-home tools. So I think there’s a bespoke—

Rich: The barriers are gone.

Paul: There’s a bespoke personal way to do software that’s been unimaginable.

Rich: It’s almost like a clinician, in a way. They’re sort of, like, I’m going to watch how you live for a couple days and then make you something.

Andrew: All of that is true. But I feel like there is one thing that is, like, the boring thing that people have been that, like, disabled activists have been yelling about for decades that that doesn’t totally carry over. Which is just that, like, there does have to be a baseline.

Paul: Yes.

Andrew: Like, WCAG is not made irrelevant by AI. Right? Like, there are still basic accessibility features that without—

Paul: Yes.

Andrew: Without which I can’t even get to the Claude prompt. You know what I’m saying? So, like…

Rich: Yeah, yeah.

Paul: To be fair, I’m assuming an enlightened society, which is probably a mistake

Rich: We’re still in futurist mode here.

Andrew: Okay, good. [laughter]

Paul: Just like, let me, let me, let me back that out.

Rich: Post-product. Post-war. Post…

Paul: Let’s get to the enlightened society with inclusivity first.

Rich: Everybody’s in robes. [laughter]

Paul: Yeah. And then we can build the custom adaptable hand-holding products. That’s fair. Okay, look, I actually could talk about this kind of stuff, I mean, look, we’d love to stay in touch here. I kind of want to, I don’t know, I just kind of want to vibe code. Like, it sounds really fun. Like, are you pushing stuff to GitHub? Are you, are you there yet?

Andrew: I’m not there yet, but I do own some 2026 versions of my, like, 500-page QuarkXpress manual that I have on my to-do list to, like—

Rich: Nice.

Andrew: Learn Python and…

Rich: Stuff.

Andrew: Learn more, you know, the idea that, like, if I know Python, I can tell Claude to be smarter about the Python that Claude will write for me.

Paul: You really can. Although I gotta be frank, it’s mattering less and less.

Rich: Yeah.

Paul: Okay, so look, if you want any good vibe-coding help, you know.

Rich: Paul’s available.

Paul: You got, you got friends here, as does everyone else listening to this podcast.

Rich: Reach out at aboard.com.

Paul: They need a little advisor. Let’s just work that marketing right into it. Andrew, look, people can visit your website to learn more. What’s a good way to reach you if they’d like to reach you?

Andrew: Yeah, my email address is on my website, andrewleland.org. It’s aleland@fastmail.com I appear on all of the major social media networks and I live in Western Massachusetts.

Rich: Amazing.

Paul: All right, we’re going to put that link in the show notes in case anyone is too lazy to figure out how to spell andrewleland.org. it’s a not-for-profit, it’s a community organization. [laughter]

Andrew: It’s actually .org because it’s an org file.

Rich: Oh!

Paul: Oh, no, thank you. Thank you for wrecking it.

Andrew: Sorry, Rich. Sorry, Rich.

Rich: Oh man. Oh man.

Paul: All right, so this was lovely. I hope you’ll come back on at some point.

Rich: This was great.

Paul: And yeah, if you, you know, like, let’s get you into GitHub. I’m just really excited. I get so excited when I see like an essayist become a big Python nerd.

Rich: The, like, population of three here. [laughter]

Paul: It’s very, it’s very—

Rich: It’s quite a club.

Andrew: Give me a year. Give me a year. I’ll be there.

Paul: Aw, just anything you need. Okay, good. Well, we will talk to you soon. And if anybody out there wants to learn more about Aboard, check out our website. We build and we solve problems and we transform your organization with AI, and we do it in a way that really sits next to you and we work with you really closely. So imagine if Andrew was a large corporation, like in the insurance industry. [laughter] That’s how we would approach this problem. So hello@aboard.com. Rich, did we leave anything out?

Rich: I think that’s it. Everybody take care of each other. Have a wonderful week.

Paul: Andrew, thank you so much.

Rich: Thank you. Andrew.

Andrew: Thanks, guys. This was awesome.

[outro music]