My guest today is Alexis Lloyd. Alexis is VP of Product Design at Medium and co-founder of Ethical Futures Lab. Previously, she led design and innovation work at The New York Times, Axios, and Automattic. Alexis has been thinking about the future of media for a long time. In this conversation, we focus on the evolving ways we consume and produce media.
Listen to the full conversation
- Alexis Lloyd
- @alexislloyd on Twitter
- Research & Development at The New York Times
- John Maeda (MAEDASTUDIO)
- Ethical Futures Lab
- The future of news is not an article by Alexis Lloyd
- Ted Nelson
- Project Xanadu
- Link to Text Fragment Chrome extension
- The Informed Life Episode 6: Beck Tench on Tinderbox
- Roam Research
- Building a digital garden by Tom Citchlow
- Dropbox Paper
Some show notes may include Amazon affiliate links. I get a small commission for purchases made through these links.
Read the full transcript
Jorge: So, Alexis, welcome to the show.
Alexis: Thank you. Thanks for having me.
Jorge: I’m very excited to have you here. Can you please tell us about yourself?
Alexis: Sure! So, I’ve spent most of my career working on designing experiences for how people read and write and share information on the internet, broadly speaking. Thinking about things like, how do we create frameworks for people to learn, to communicate? How do you create clarity and understanding for people? And how do you create tools that give people superpowers to express themselves in the way that they want to. And I’ve done that both with one foot in the present, one foot in the future. I spent nearly a decade at the R&D lab at The New York Times, the last few years, co-leading that lab, which was really thinking about working within a media organization, but thinking about how emerging technology and emerging consumer behaviors… what kinds of new opportunities that might create for how people engage with media and information, looking like three to five years out. So that was doing a lot of research through making and working with a team of creative technologists, prototyping everything from screen-based interfaces to interactive environments, to data visualization, and then trying to pull that back to understand not just what’s possible to make, but what’s desirable to make. And then continuing from there, I did some more work in media organizations, was part of the kind of starting team of Axios, where I led design there and really helped take the founder’s high-level vision and build out what those products would be like both for readers and for people who are writing internally.
And then more recently I’ve done similar work around these information experiences in companies that have more of the DNA of tech companies. So, John Maeda reached out to me and invited me to come work with him at Automattic, which was a wonderful opportunity, loved working with John. I headed up design innovation there, so it was really thinking about, how do you take this kind of 15 year-old blogging platform and think about what does it mean to democratize publishing now, and how do we build upon that in new and interesting ways? And now I’m at Medium heading up product design there. So, obviously a lot of these things tie together. And then last year also started a side project with my partner called the Ethical Futures Lab, which is… came out of, we had worked together at the R&D lab at The Times and we were both missing some of that thread of looking outward, looking more speculatively, thinking about this idea of how do you build things that are good for people, good for society? How do you evaluate the possibilities and risks of emerging technologies in really tangible ways, so less big picture theoretical and more like, how do we actually design for these things? Like, what do you do to make a system that, you know… how do you build a database that works well for people? And so, we’ve been doing that for the past year and that’s kind of where I’m at now. And I am just really excited to talk to you more, about information and how we design for it and how we think about that.
Types of media
Jorge: When you say information, you’ve been talking about Automattic and The New York Times and Medium and Axios. And I think of those as primarily text-based media. Is that a fair take?
Alexis: Yes and no. I think that they lean towards text-based media. Certainly, at The Times we were thinking well beyond text, especially because we were thinking further out and doing a lot of experimentation around audio, video, and then also how does information make its way into environments that aren’t screen-based and what does that look like? But yes, I would say that I spent a lot of time thinking about text and how you work with text and what that means and how you engage with it.
Jorge: Yeah, I’m hoping that we’ll dive a little bit into those distinctions because you spoke about two sides to this equation. One was the people consuming the information and the people producing the information, right? And you talked about designing tools that give the producers superpowers. So that’s kind of one side of what I heard.
And the other side is… and I don’t know if you said this now, but I certainly have gotten this from your readings is providing the people who are, gosh, I don’t know if I like the word consuming, but the people who are interacting with the information that someone else has produced, give them novel ways to become informed? Is that fair?
Alexis: Yeah. Absolutely.
Jorge: And my expectation would be that both sides of that equation would come at the information very differently if it was text versus video versus audio.
Alexis: Absolutely. I think that’s true. And I think there are also different things you can do with each of those media. Although I think there are also some interesting things that I’m seeing in the world that are starting to blur the distinctions between the two? I’m forgetting the name of it, but there is an app that allows you to basically… it automatically creates text transcripts from video, and then you can edit the video by editing the text. And so, it’s kind of this cross-media manipulation that I think is super interesting. But historically, yes, I think that we think about those formats really differently and people are producing them in different ways and with different concerns in mind.
Alexis: So, I think actually one of the things I’m interested in is, how do you blur some of those lines? So, maybe if you’re creating something with audio that it automatically creates a text version that you can then edit and refine, or the same kind of thing with video, but thinking about ways that you can create more fluidity between media formats? Both for people who are creating them and also for people who are consuming them. So thinking about people who prefer different… some people love video, I personally am not a fan of video. Like I hate being forced into a linear format, and I like to be able to skim text and skip around. So, I think they are interesting ways that you can start to blur those lines a little and create more flexibility for both the way people share information and the way they absorb it.
Articles and particles
Jorge: Yeah, this very issue is one of the reasons why I wanted to speak with you. You posted an article on Medium about the granularity of information. I think that the distinction that you set up is between articles and particles. I was just going to ask you if you could elaborate on that, because I’m really interested in that distinction.
Alexis: For sure. So that was actually a concept that we had come up with back in the R&D lab at The Times… it was about the future of news not being an article, but coining this term “particles,” which is that there are all these pieces of information – a quote, a piece of analysis, of fact, an idea – that live within this article construct, which has been kind of the atomic unit of news, largely because of historical constraints of print and how you produce journalism. But that we don’t actually necessarily have all of those constraints — we still have some of them in terms of daily production schedules — but you can see a similar thing with the web where we’ve kind of been tied to the way that words and ideas work in print and in physical surfaces where we think of pages as these atomic units, and links being the bonds between those pages, and so really thinking about this idea of kind of particles and bonds, which is like you have ideas or concepts or quotes, and then you have bonds, which are the ways that you point to those or annotate them or contextualize them, or connect them with others. And I think that there are spaces where we can start to innovate on how that works. There’s been thinking about this happening in the margins of the web for a long time.
Jorge: I got the sense from your comments earlier that when you’re on the consumer side of this equation, you prefer information that is scannable. Like, that doesn’t force you to go through a strictly linear sequencing. And I can see that reflected in this distinction because breaking articles or breaking content down into particles that are brought together by bonds, I would expect would give you a lot of flexibility as to how you reconfigure those particles for different needs.
Alexis: Exactly. So that’s, what’s super interesting about that is that when you were talking about, for people who are reading or consuming information, what are interesting ways to be able to engage with that information? Or even for people who are producing it, there’s this whole thinking about how people annotate, how people remix, how people reassemble found ideas from elsewhere and then layer on top of them is a lot of the potential power of the internet that we’ve somewhat realized, but I think there’s probably more there. And so, I think that there are these pieces of information that are locked within structures, that if we start to unlock those structures or loosen them, that it creates more possibilities for how people can engage with ideas and with each other.
How granular should we get?
Jorge: I had an interview last year on the show… I spoke with Beck Tench who was doing her work towards her postgraduate studies and using Tinderbox to implement Zettelkasten. Have you seen this way of notetaking?
Alexis: Yeah, I think I have. This is kind of the Roam Research thing, right?
Jorge: That’s right. Yeah. And I’ve gotten into that whole way of working just because it is doing what you are talking about here, which is like breaking down information to more granular units that then are interconnected in ways that reveal perhaps new patterns, new connections.
Alexis: Yeah. Have you actually been doing it? Because I’m fascinated by this and it also hurts my brain trying to think about how to implement it or approach it. I haven’t quite gotten there yet, so I’m curious if you’ve actually worked with that in practice.
Jorge: So, not with Roam, but I am implementing something very similar using DEVONthink on the Mac. Which is, gosh, it’s the same thing as with Tinderbox… these are applications that are very hard to describe in a sentence because they are so open-ended, right? And the idea is like, I think it’d be fair to say that they’re both note-taking or note-keeping apps, but they’re designed to break you out of the kind of predetermined connections that happen when you’re thinking linearly. They’re set up to let you discover new connections between things and for that to work, you have to be granular in the way that you’re capturing content.
Alexis: Right, right. Do you find that you need to have an extra layer of process around how you capture ideas to make sure you’re annotating and tagging things so that they can fit into that system?
Jorge: Absolutely. And I’ve recently come across a wealth of writings on the subject and this is all so new to me that I can’t tell you about the actual posts off the top of my head, but I will include them in the show notes. But there is an image that I heard that I think captures it very nicely. It’s the phrase “digital garden.” And this idea that what you’re doing is gardening is, I think, very apt because a garden, you know, needs tending. And these things need tending. And the thing is, and I feel like I’m being a little rambly here, but there’s an actual question in here for you, which is, I find that this way of note-taking, this way of breaking down content is a very different way of thinking for me and a very useful, and… like I am finding value in doing this. I am also curious about how much value another person could get from my digital garden. It feels to me like for this to be of any use to anyone else, I must do the… turning the serendipitous connections between particles into something more linear that can then have a narrative. And I’m wondering if that’s something that this notion of the tension between particles and… I don’t know if to call it storytelling, but the fact that we…
Alexis: Yeah. The narrative is how we convey ideas, but then the way that we capture what’s interesting about those ideas may be much less linear or formal or necessarily clear to someone outside of our own heads, right?
Jorge: Yeah, exactly. I’m wondering in your investigations around this, if you have found the optimal balance between the granularity of content and the point where it stops conveying narrative meaning? Is there a way to gauge that?
Alexis: Yeah, I think that I wouldn’t argue for content. I wouldn’t say that content should be more granular in the way it’s created. But I think that in the way it is to be extracted, annotated, remixed, and built upon, that the structure should afford more granularity than the original output.
So, for example, to make that more concrete, I should be able to tell a story in a fairly linear way, but then — and I’m going to look at this through a machine lens and then through a human lens — so, through a machine lens, there should be ways for a machine to be able to understand underlying pieces within that structure and pull out the appropriate ones to be recombined or surfaced in other contexts. So, if you look at some of what Google search has been doing in terms of pulling out more structured data within documents or web pages that are surfaced within search results, I think that’s an example of it. That’s not how those pages are written, but you can start to train machines or start to structure that content in a way that it is more machine discoverable and understandable at a more granular level. And then for humans that I should be able to grab something, I should be able to highlight something or share it, annotate it, re-share it, and add layers on top of it. I should be able to take a bunch of different pieces and pull them together and build a narrative around that. So, I think it’s less that I’m saying everything should be more granular, but that we should be able to like, understand and leverage and make use of the more granular pieces within traditional narrative structures.
Jorge: And that certainly corresponds to the way that I understand how tools like Roam work, right? And it’s what I’m trying to do with the DEVONthink thing. And it feels to me like that reconfiguration of the particles, with its annotations and stuff like that, are meaningful to me. I’m wondering for something like a newspaper… you worked at The New York Times and The New York Times is the “paper of record,” right? Like, where the narratives are articulated where a large group of people can have broad agreement on what we think about things. And I’m wondering in this world where we can reconfigure narratives, because they’ve been made more “particular”…
Alexis: I like that. I hadn’t extended out the grammar of that.
Jorge: The one issue that I have, that is still an open question for me, is what happens to narrative and what happens to argument? Because it is very hard, I think, to put across a compelling argument when things have been deconstructed, and then you’ve been given like a kit of parts for you to make your own meaning, you know? I think that that opens up a lot of interesting questions. Like I said, I’m super excited about experimenting with it for my own note taking, but like, I would not want The New York Times to work like that, because how would I make sense of what’s going on? You know? How do I make sense of what’s going on?
Alexis: Well, and like I said, I don’t think… I wouldn’t argue for that either. I think that the thing that’s interesting about particles, about these more granular ideas is not that is how information is disseminated, but that within the more traditional narrative, you can extract those things and then do something with them, whether it is to remix them and contextualize them within a different narrative, like I’m going to pull these five different quotes that I think make an interesting argument. And then I’m going to make that argument with those quotes as references, or, I’m going to create a collage of different media elements, or I’m going to find a historical timeline of every time this idea was referenced for the past 10 years or something like that. I think that it allows you to do more with the ideas within a narrative once that narrative is created. But I wouldn’t argue that we shouldn’t be creating the narratives. I think that we still very much need those. And I think what this allows us to do is to use those building blocks of previous narratives to build new ones.
Jorge: Yeah, for sure. That’s what I was trying to say earlier is like, I use it mostly as like grist for this mill that then will put out stuff that looks traditionally consumable, but it really is made very differently. The process of making connections between concepts has been taken out of the meat computer here and put into the silicon computer in a place that has a lot more memory than this one. Right? And better search capabilities for sure.
Alexis: Yeah, that’s what I wonder about with those. Like, I’ve approached that and I always have this idea… like, when I was in college, or high school, every semester, I would like try to find the perfect notebook system that was going to keep me perfectly organized, which of course like nothing has succeeded at because it’s my brain! Like it’s not this ideal state. It is what it is. And so, I feel like now I do this with software where I get really excited about tooling and I’m like, this is the thing that will make me able to keep track of everything and connect it all together in the way that I want. But then it requires so much effort to tend the thing, the structure that I’ve set up, that I end up sort of abandoning it and doing whatever’s easiest, which might be like an array of disconnected Dropbox Paper docs, or opening up my Notes app. It’s always like finding the thing that’s the least friction in the moment, but then the thing that’s the least friction in the moment doesn’t lead to longer-term connections or understanding the way that I want.
Designing for bonds
Jorge: In this scenario where content is broken down into smaller particles that then come together through bonds, I would expect that the bonds matter a lot. And I’m wondering about the bonds and the degree to which we as designers, perhaps, if we’re not specifying the linear narrative, perhaps there is work to be done on the bond side of the equation. And I’ll be, concrete. I think that the metadata that we assign to the particles will likely have a big impact on the context in which those particles emerge, no?
Alexis: Yes. I think that’s true. And I think it’s also, what are the affordances we create for doing that work of extracting, annotating, remixing, contextualizing. So, I think that, for example, if you look at Quotebacks, that is really privileging reference and attribution. So that’s built into the system, to say that it has those ideals or values built into it. So, it has a value built into it than it is important to attribute and reference someone that you’re quoting and to make it easy to get back to the original source material and that’s built into the bond and the way that it’s designed and literally the design of the thing that you embed in the page when you create a Quoteback.
So, I agree. And I think if you look at something like Pinterest is kind of a weird example of this, right? But I can take images from anywhere and I can assemble them, and I can create, if not a narrative, at least a theme around those and share that with other people. But that’s really privileging imagery. Whereas if you look at something like Arena, which does something similar, but it’s not just privileging imagery, even though it is very visual, it’s allowing that to really extend across different kinds of sources and webpages. And so I think your point about bonds is well taken, in that it’s the tools that you create for being able to extract or remix or annotate or share and how those things get shared and with what kind of metadata, really changes the experience of what people do with it, or it leads people towards different outcomes based on what the defaults are. Designing those defaults is really important.
Jorge: Yeah, absolutely. The way I talk about it usually is, I speak of the “tyranny of screens” because I find so many designers drawn to the allure of the final thing, the final artifact that people will be interacting with, when the real power in a lot of these experiences lies deeper, at the structural layer of whatever the system is that you’re designing.
The future of media
Jorge: You are someone who is thinking a lot about these issues and the future. And I’m wondering, what has you excited about the future of media?
Alexis: I don’t know that there’s like a product out there or something like that, that I would point to and say, “this is the thing that’s gotten me really excited.” I think we’ve gone through these very distinct phases of the internet and particularly the web as a medium for creation and consumption of media. And we had this kind of early open web, open internet that was like the first decade or so of the web, which had certain advantages. Like it had high bound for creativity. You could basically make whatever you wanted, if you had the skill to do it. If you could learn some HTML and CSS and could put anything out there, it was accessible to everyone. But it wasn’t highly discoverable. Like, being able to know what was out there, how to get to it or who to connect to was not very easy. And then as a creator, it was theoretically accessible to anyone, but you needed to have some technical skills, you needed to set up your web hosting and you needed to do all this stuff to just say a thing on the internet, wasn’t super easy.
And then you’ve had the past like decade or so of platform-driven web that’s where people are largely consuming and publishing within the bounds of the Facebooks and Instagrams and TikToks and Twitters of the world, which has radically increased the accessibility of publishing. It is extremely easy to say something on the internet and it’s increased discoverability, so it’s very easy to find who you’re connected to, who you might want to be connected to, and to make sure you’re getting all the things that you want from them. But it has kind of limited creativity in that everything is… a Facebook post looks like a Facebook post looks like a Facebook post, you know, it’s all very same-y, which leads to, a) just a little bit dull, but also it leads to some context collapse where it flattens the texture of information and the way that people can say things.
And the thing that I’m starting to see is, I’m starting to see people try to go out on their own more. I mean, I think you’ve seen it in the past few years in sort of the emergence of more podcasts and newsletters and things like that, where it comes back to building this more relational aspect between the people who are creating the people who are reading or listening to your work. And it’s getting out of the algorithmic platform delivery, feeling of things. And so, I feel like there’s a sense of people wanting to recapture some more independence and then also some more creativity. One of the things that we’ve been doing at Medium is building work creator tools that allow for more expressiveness, so that it’s not all the same, so that you do start to create more context, and having more ways to kind of build relationships between readers and creators.
So, I feel like, to answer your question, the thing I’m optimistic about is that we could be heading for a space where we start to build the best of both worlds, where we have a lot of the affordances the platforms have brought us in terms of ease of use, and in terms of the kind of network effects — although there obviously have been some not so positive network effects as well — but the ability to connect and the ability to easily create, while recapturing some of the kind of individuality, the creativity, context that we’ve lost somewhat in the last several years. And so, that’s where I’m hopeful. I don’t know that that’s the world that will come to be, but that’s my thread of what I hope for is that we can start to bring back some of the web that we lost while retaining the affordances of the newer technologies and platforms that we’ve been building on.
Jorge: Well, hear, hear! And I just want to mention that, well, you spoke of, your newsletter earlier, Ethical Futures Lab. So, I wanted to thank you for that, because I think that it is an example of that kind of writing. And I want to encourage folks listening to the show to sign up for the newsletter. Where can folks follow up with you other than the newsletter, which I’ve already plugged?
Alexis: Yep. So, you can sign up for Ethical Futures Lab, at ethicalfutureslab.com. And then you can follow me on Twitter @AlexisLloyd or from my website, alexislloyd.com.
Jorge: Well, fantastic. Thank you so much for being on the show, Alexis.
Alexis: Thank you so much, Jorge!