random apps
Some reflections on non-determinism as a resource for design

Table of Contents

Lately I've been enjoying the 'random' feature on special.fish (a lovely, noncommercial social network). The 'random' button will take you to a random person's profile. People leave fun links in their profiles from forgotten corners of the Internet. From my friend Luke's profile, I encountered "Fatum theory" and the "Randonauts."

I can't recommend a deep dive into the theory (it involves some parapsychology around random number generators and other crankery), but one core idea—in my own words—is that our ability to perform new behaviors, and therefore experience new experiences, is limited somewhat by hidden determinism in our lives. There may be particular spots in our own neighborhoods we have "seen without seeing," passed by or passed over by force of habit. A prompt to notice something new, something randomly chosen, can (and, anecdotally, does) introduce new thoughts, considerations and feelings into life.

Breaking out of life's hidden determinism through randomness. I played with this idea through two phone apps, where? and who?, which I'll describe here. I'll conclude with some crankery of my own.


where? picks a random point on a map near you. Run where? on iOS or Android here. The radius for random points is the visible screen: pan, rotate and zoom to shape or constrain the randomness.

This app was directly inspired by Randonautica, but differs in a few ways. First, it removes any philosophical or parapsychological considerations around random number generation. I actually appreciate Randonautica's explicit insistence on and explanation of its philosophy. I wish more apps reflected as deeply on their ideological underpinnings. Imagine if Uber laid out its Randian labor philosophy? Nevertheless, I have seen no convincing evidence that human intent influences random number generators.1

Second, this app does not distinguish between "water points" and "land points." The idea that you somehow can't get to a water point is a social (rather than brute) fact. Likewise, the idea that you can get to land points is itself an assumption, one that frequently collides with social facts. What kinds of bodies get to move where? Physical properties of space intersect with property rights and relational characteristics like race and gender to produce differential patterns of possible mobilities. If this app helps you encounter new impediments or reflect differently on the privileges of your mobility, it's successful.

In many ways, this app (like Randonautica and others) revisits the Situationist dérive. However, the dérive fails to ask the question most interesting to me: what kinds of bodies get to move freely through space? Nevertheless, its "psychogeography" does acknowledge the effect of physical space on thought and behavior. This app doesn't engage with the Situationists here per se. Instead, it more echoes Lucy Suchman in raising the situatedness of the "user," too often imagined as an autonomous, reasoning individual, here reconfigured as a helpless one, trapped in unconscious, looping patterns.


who? picks a random contact from your contact list. Run who? on iOS or Android here.

Before "apps" existed as a social or cultural category, contacts where the original phone app.2 who? returns to the contacts app with a different lens. The original "contacts app" embeds the idea of computer-as-prosthetic-memory, a core concept in the classic human-computer interaction of the 1980s. who? reimagines the contact app as a prompt for reflection, and introduction of randomness. As we "go to" (pass by, pass through, seek out) certain physical places, we also "go to" (pass by, interact with, think about, plan around) certain people. The physical proximity of where? (places near you) is mirrored here, bounding randomness to the people already 'near' via your phone's contact list.

What one does when a contact is drawn is intentionally underspecified. Again, encountering obstacles to actually contacting certain people is, in itself, a prompt for reflection.

Personally, I find this app a bit scary. There are some people in my contact list I'd rather not think about. That fright tells me there's something deeply good about the design. When's the last time a phone app really made me nervous?


There's a lovely symmetry to my (re)discovering these ideas by clicking the 'random' button on a website.

HCI loves mushrooms—probably a fascination we inherited from Anna Tsing on our postpositivist side. I wouldn't be surprised if this fascination owes also to that side of HCI that still looks toward those transhumanist and techno-utopian impulses which themselves owe much to Terrance McKenna, Robert Anton Wilson and the like.

Anyway, I'm no exception. I was reading recently about Psilocybin (psychedelic) mushrooms.3 If you've ever used any of these mushrooms you probably had P. cubensis, which grows in the wild on feces, especially cow feces.

Why did these mushrooms evolve a psychedelic effect? One speculative answer: the compounds reduced the likelihood of the fungi being eaten by insects. However, this explanation doesn't quite answer why this insect repellent would also take the form of serotonin drugs, which just happen to have a profound effect on the behavior of mammals, in whose feces this mushroom just happens to grow. I'm not an expert on evolution, but I'm intuitively disinclined to believe in evolutionary coincidences.

As long as we're all speculating, here's another evolutionary explanation for these substituted tryptamines: they produce behavioral changes in mammals that are adaptive for the fungus. Perhaps by introducing a degree of randomness or non-determinism into cows' behavior, cows were able to sometimes find new grazing pasture, thus pooping in new places, thus expanding the mushrooms' territory.

Now, if I were forced at gunpoint to elevate this evolutionary speculation to full quackery, I'd say that hidden determinism in life is a source of epistemic risk. We don't know what we don't know in part because we're usually doing the same sorts of things. A degree of randomness adds danger and new input to life: a quantum superposition of risk and reward.

That these risks and rewards are unequally distributed based on relational characteristics—race, age, class, gender—is very much in our lane as designers and as security researchers. These apps could do a better job of explicitly considering the individual users and communities of practice, eliciting, sharing and engaging with their experiences of space and people. Something for me to think about in the future.


[2020-05-22 Fri] One more question for future consideration from Richmond Y Wong: for whom should we be adding randomness? Developers? Researchers? Users? Security practitioners? What are the implications of adding randomness for these different stakeholders?

[2020-05-22 Fri] Harrison Atkins alerted me to Max Hawkins' work—he's taken the randomized living to an interesting extreme, and is certainly the reference point for randomness in design in my book. I would love to see a little bit more about his experiences, and also a little bit of reflexivity (who gets to live in random places for two years?).

[2020-05-22 Fri] I stumbled upon cyberia.is, a round-robin game of IRC servers.



Yes, I'm familiar with the Princeton Engineering Anomalies Research Lab.


Before Nokia brickphones ran Snake, phones had contact lists. I've traced the "contacts app" to at least the 1996 Motorola StarTAC, but I'm sure the idea of a computerized rolodex goes back to the French Minitel at least. If you have any information on this lineage, get in touch.


These mushrooms are called "psilocybin" mushrooms mostly because Albert Hoffman first isolated that particular chemical. They also contain psilocin, baeocystin, maybe more.

Date: 2020-05-21 Thu 00:00

Author: ffff@berkeley.edu

Created: 2020-06-06 Sat 11:12