"I can navigate to places that I'm not supposed to go"
Elan Ullendorff wants to know how you found him
Embedded is your essential guide to what’s good on the internet, written by Kate Lindsay and edited by Nick Catucci.
I found him through
!—Kate
In many ways, the internet today is so successful because we’re all shackled by the same constraints. We’re on the same platforms, posting the same algorithmically-pleasing content, and being unconsciously steered back on track by metrics if we ever stray off the path. These guardrails are what make social media feel so centralized. They also…don’t actually exist. You can do whatever you want online, and Elan Ullendorff is proof.
The director of product at The Marshall Project doesn’t look at the internet head on. Through his web projects and a class he teaches at The University of Pennsylvania, he shows how to find the side doors into the platforms we take for granted. Substack, for instance, provides its newsletter writers data on where their subscribers come from. Ullendorff, who writes
, knew there was more to the story. He reached out to recent subscribers individually to find out, in their words, how they discovered his letter. This inspired his latest project: How did you find me?The site is consists entirely of people sharing how they found it. The concept is simple, but the answers are all over the place: “someone sent a link on one of the queer furry programming discord servers i’m in”; “Once upon a time, the Lord of the Rings bonus features existed. From there it is a simple and obvious chain: I became engrossed in making constructed languages, met some likeminded people online, and 20-odd years later, one of them linked me here”; “I happened to be shitting, and an article … popped up in my RSS Reader. I opened it. And here we are.”
Ullendorff and I started our conversation focused on his latest creation but ended up talking more about the internet as a whole, and why you should keep turning over its stones to see what’s underneath.
I actually found you through “How did you find me?” What was the inspiration for the project?
A question that I basically always ask, but I think is really important to understanding how the web works in general and individual platforms work in general, is what context is being left out and who benefits from that context being left out? And as someone who makes web projects and publishes on Substack, I have the opportunity to quote unquote “learn” things about my audience. So I can attach a snippet of code to a web project and see on Google Analytics where people are coming from. Every single time someone subscribes to my newsletter, Substack emails me and it tells me where allegedly people are coming from. And I've written about Substack analytics in particular, but it's just a really reductive way of thinking about where people come from. Substack was basically telling me that a lot of people who found my newsletter were coming from the Substack network, and I ended up emailing my few hundred most recent subscribers being like, “Where did you come from?” And I found that that was a little bit misleading, what Substack was saying, but I also just found the stories that people told me really compelling. They were a lot deeper than just the last thing that the person clicked. So I became really obsessed with that idea of like, what is this longer story that you would tell if I asked you?
Do you feel like you've learned anything new about how people use the internet from hearing these stories?
There are plenty of people who are like, this platform sent me to you, which is what most analytics tell you. And it's a lot more about the person and the people that they met along the way. Or even if they're publishers, like Links I Would GChat You, that's an actual person who saw my project and liked it and shared it and had a relationship with their reader who trusted them to click it. That's a way more compelling story to me than, “this came from a newsletter.”
How did you get started making these kinds of web projects?
Not many subscribers to Escape The Algorithm know that originally it started as a newsletter called Straight To Spam, and the shtick was that it was a newsletter delivered directly to your spam folder. I liked the idea of doing the opposite of what everything else in your inbox was trying to do, which is to get your attention, and also elevating this part of your inbox that you're not supposed to pay attention to, your spam folder. The whole point of me starting a newsletter, I wasn't trying to sustain a business. I have a day job. It was just—I find it hard to write or create something just for myself. I really enjoy writing, which for me is just a tool for doing deep thinking. And it didn't really matter that the audience was big or real. It just needed to have this vague shape of being real. So it was being sent to people's spam folders, maybe one day they would look at it and they would see it. Turns out, it's pretty cumbersome to get emails reliably delivered to spam folders.
I was gonna ask—I wasn't sure if there was some fancy code you were using.
There was definitely coding involved. There's a whole industry of email deliverability experts that teach you how to avoid the spam folder, but there's very little information about how to do the complete opposite. And when I tried to ask on email deliverability forums for advice, people were like personally offended. So I was basically just breaking every rule that I could. You end up having to do everything yourself. Because if you rely on something like Substack or MailChimp, then they already have a lot of trust with email providers. So anyways, it ended up getting in the way of the thing that I was actually there to do, which was to write. So I pivoted to a regular newsletter.
Subscribers have asked if they can pay to support my work, and right now I like that I'm not really beholden to anyone except for my own obsessions. I write and I build something when I can't contain myself. But otherwise, sometimes I don't publish for a month and I don't need to worry about people paying me. And then as for the subject matter, I came to realize that escaping algorithms is something that I've always been fixated on. I might not have had that exact language for it, but when I think back to projects I did in college, I built a Twitter account that would monitor tweets and then repost any tweets the second they were deleted. I also built a Chrome extension that people could install and it would monitor whenever they Googled something that returned no results, log it on the webpage. So a lot of ideas on the same themes. And more recently, I've just started cultivating a language around it and a set of practices or algorithms for my students to replicate the steps they can take to explore these ideas. And then all of it, the teaching and the writing and the creating, all speaks to each other.
I interviewed Soren Iverson, and his tweets will go viral a lot because he'll do these made up app features. How he approaches it is basically looking at a part of an app and being like, “How can I make this worse?” The tagline of Escape The Algorithm is “finding a more human side of the internet.” Is that a throughline to your projects?
I don't always follow it, but there are certain things I'm seeking out. I'm interested in basically anything that the platform wasn't designed for. Not necessarily things that are completely hidden, but it could be things that were deleted, accounts that not a lot of people follow or not a lot of people engage with. Or invisible labor is also really interesting to me. The labor that goes into these platforms, that's not something that certain platforms like to expose. But then I also mentioned, I have a variety of techniques. I have my students each pick a platform at the beginning of the semester, research everything that they can find about how the platform works and scandals about the platform's algorithms, then they basically make a map of every possible way to navigate the platform—every button you click, every search box you can enter inputs into, every weird settings page in the background. And then I have them make their own algorithms, which are ways to navigate a platform in a surprising way. So an algorithm can be like, go to a random word generator and then go to Instagram, enter hashtag “random word” into the search box, and then have friends try that out and see what they find. And usually that's enough for them to find something really interesting.
The end goal is to learn something either about how the algorithm works or to critique the algorithm, sometimes it's to make some part of the world or the culture that's usually invisible to you, more visible, like this really fascinating subculture of the platform that isn't popular, but is still very human and interesting. The second half of the semester is, we use what we learned in the first half of the semester to imagine alternative platforms.
This just made me think, in college, I remember I had this idea, and it's kind of dated now, but at the time I wanted to crowdsource all the little video snippets that we all have from when you thought a picture was being taken, and then you realize it's a video. Because it's always you're doing a weird pose and then someone goes, "Oh, it's a video. "And I was like, it'd be kind of funny to watch all of those back to back.
I like that. It's something that no one would ever usually post because it's seen as an accident. I really like things that are seen as accidents, and trying to find the meaning in them. I imagine that if you put a lot of those together, something would feel very human about that, right? How we like to portray ourselves on social media is very overly designed, and this would be a documentation of a mistake.
When I see what you make, these fun projects without being paid or any mechanism that shows how popular it is, it feels very old-internet nostalgic. Do you have nostalgia for the old internet? Or do you not think of it in those terms?
I think there are qualities of certain aspects of the old internet. I don't necessarily think of it as nostalgia. I think there's a centralization right now that strips a lot of the expressive power out of the internet. If you're designing a platform like TikTok or Twitter, there's a lot of incentive to make all content commodifiable, not just even in a technical sense, like there needs to be a unit of information that is programmed into the interface. It could be a joke, it could be a speech, it could be a livestream video, and it's all interchangeable and it can be interchanged on the feed. And that is a very efficient way to build a company. It's very scalable.
The downside of that is that it's very inexpressive, both as a platform—it's just a bunch of boxes—and as the user, there's only so much I can do. I have very little control over how my content looks visually or how people can interact with it. So there is something to be said for being able to own something yourself and design it from the ground up. But the fact of how easy it is for anyone without any experience to just publish something online, you have to acknowledge that there are positives to that as well.
You like seeing the mistakes or the things that are hidden. Did you grow up really online? How did you get to that interest?
I would say the biggest throughline is just, I think I've always been interested in subversiveness. Even in fourth grade, if I was assigned an assignment in school, I was always thinking about, what are the ways that I could break the rules of this assignment to make it more interesting to me? And also to test the limits of what it can be. The whole joke of this project of escaping algorithms is that it's not actually possible to escape algorithms. Every way of being or experiencing the world is an algorithm or has algorithmic qualities. And the concept of algorithms is not inherently bad. Right now you might get your news from TikTok and that might give you a filtered view of the world, but before the internet, you were maybe getting the New York Times delivered to your doorstep every morning. And that was also a filtered view of the world. And I think of that as algorithmic also. Either of those deserves to be examined and questioned and subverted.
I think there are some different ways that web algorithms influence the world, but I think even more so, we have more tools at our disposal to poke back at the algorithms to better understand them and imagine alternatives. So if I wanted to understand the filtered view of the New York Times, all I could do back when it was delivered to my doorstep was read the New York Times and then read it tomorrow and then read it the next day. But if I'm on Instagram, I can use it in different ways. I can make a new account and pay attention to different things and see how that influences the algorithm. I can type words into the search box, as we talked about before. I can navigate to places that I'm not supposed to go or misuse the service in some way.
How do you feel about the current state of the internet? Are you optimistic, pessimistic?
I mean, it's both all the time. I think it depends who you're looking at. I'm pessimistic when I look at the centers of power, and I'm optimistic when I look at the margins of power, which I think is often the way that I look at the world. I think there are always a lot of really interesting things happening on the internet. I think of this project as optimistic in a way. I mean, very, very critical and also optimistic because at least for me, and I find for my students, it's really good trying to design what a more human platform would be.
Are there existing platforms or spaces online where you like to spend time?
I am helplessly addicted just like anyone else, and I spend time on the same platforms that make everyone depressed and they also make me depressed. There's a lot of discourse now about platforms making our kids anxious, and I think there's nothing inherently anxiety-inducing about the internet. Like when I'm doing my Escape The Algorithm work, and I'm doing that very active work of trying to see all these possibilities on this platform, and finding surprising, interesting things, that is really invigorating to me. That feels like a completely separate thing from mindless scrolling down my Twitter feed. I definitely pay attention to movements to this app or abstain from using this app, but as far as my Escape The Algorithm project, I'm less interested in abstinence. I don't think it's ultimately possible, even if I delete my TikTok account, I'm still living in a world that's supremely influenced by TikTok. Yeah. So I'm more interested in algorithmic agency and intentionality.
I think Arena is a platform that does that very well and when you look closely at the platforms that do it very well, it also becomes obvious why: The incentives are not there. You basically need to decide, “Making money is not the most important thing to me.” And that's the most important thing for most of these companies.
Is there a lingering subject or phenomenon or that you'd be interested in exploring next?
I mean, I have an endless backlog of ideas of things I wanna build, some things I wanna write about. I need to follow whatever's the most interesting to me at the moment. Something that I'm interested in right now because of where I am, I'm in between my first and second semester teaching, is documenting some of what I just told you. What my method is and how other people can do that, and I think that's something I'll probably focus on soon.