I’m not writing about the election today. I’m sure you’ve got plenty of that already. This is a newsletter about human-computer interfaces, as always. Take care of yourselves.
I wrote about digital gardens: spaces on the internet where content is cultivated in context, rather than streaming by in an endless flow of disparate thoughts.
I think this is the the beginning of a shift in the way we engage with each other through the internet. Instead of experimenting with formats, we’ll try new structures of communication.
We spent the 2010s trying different formats for social media. We explored posts, tweets, photos; articles, listicles, chumbuckets; Stories, videos, podcasts, TikToks, memes, mashups, bookmarks, pinboards; and ads, above all, the many formats and locations for sharing ads in this new many-to-many medium.
Every form of art and creativity was commoditized. Every Instagram model became an influencer, every hobbyist got an Etsy store, every viral tweet was followed by “wow, this blew up, check out my SoundCloud”. Everyone else started a brewery. We had to figure out how to make money doing something, because having a job doesn’t pay anymore.
But we’ve pretty much run out of new formats. The big innovations right now are in licensing.
TikTok blew up because it’s easy to (legally) use a pop song in your viral video. This is in contrast to YouTube, which uses AI to detect copyrighted music and regulates preemptively, redirecting ad revenue to music labels or simply rejecting the video. In the same vein, Spotify’s podcast side is about to explode, because of the new “Shows with Music” feature that allows creators to insert songs directly into the flow of their podcasts.
Making it easy for people to recycle each other’s content is a good business move, for now, but it’s not a fundamental advance. It’s building staircases over the fences of copyright law. It won’t last.
🧗🏻♂️ What’s the obstacle to real advances in social media?
All the current social media, whatever their preferred format, rely on the same vocabulary of social gestures:
Sometimes only a subset of these tools are available, but users typically find a workaround. On Twitter, for example, hashtags, @mentions, retweets and quote-tweets were all originally hacked together by users and backported into official features of the website.
These primitive gestures are combined and repurposed as necessary. The street finds its own uses for things. And since there are incentives to use these tools for ill, there are people who have found the ways to do so. Sharing misinformation, posting privately, creating networks of bots that all follow each other. But they’re still working within the system at hand.
A fundamental advance in social media would require new social gestures that add context and control to our shared information universe. Down with one-dimensional streams and mysterious algorithmic feeds. Up with choices, commentary, citations.
🙏🏻 Which gestures will create this change in social media?
One feature that’s finally percolating into the mainstream consciousness is the bi-directional link, or backlink. This was one of the founding ideas of hypertext, neglected in the early days of the web: if page A links to page B, page B should mirror that link automatically. This changes the dynamic of commenting on someone else’s work: instead of being in their “space” giving unwanted feedback, you go home and comment from a distance. The difference between a heckler and a critic.
Lots of social media sites survive by mimicking this function in a particular silo. Reddit is a comment section for every website. Yelp allows you to comment on real businesses, creating backlinks for Main Street. Twitter is the comment section unchained, people commenting on every post on the internet, commenting on each other’s comments, like a seething mass of mealworms, devouring and birthing each other endlessly. But by siloing the commentary, they reduce context, rather than increasing it.
The backlink renaissance has been led by Roam Research, a note-taking app with a revolutionary roadmap but a rudimentary interface. Other gestures that Roam is bringing back (though by no means pioneering): transclusion, where a piece of external content is framed directly into the local page; versioning, so text can be edited non-destructively; and universal search, which allows the user to think outside of the “temporal stream”, “hierarchical outline” or “folder of documents” metaphors. They’re currently focused on a single-user model, but in a social milieu these gestures could bring nuance back to collective knowledge-building.
Venkatash Rao puts it well in his analysis from February, A Text Renaissance:
Conspiracy theories and extended universes, in the best senses of those terms — escaped reality construction might be the general category — is what Roam wants to be about.
Conspiracies are all about context.
This NAME keeps coming up, over and over again! Some of these features can be seen creeping into the legacy platforms. Twitter has built a protocol for bi-directional linking, so that each tweet can “point” at the tweet it’s replying to, and be pointed at by tweets that reply to it. This has created a renaissance in threaded essays, fed bite-by-bite into the 280 character limit.
It’s also got a simple form of transclusion in the quote-tweet — though with no edit functionality, of course. Twitter has never had edit functionality. That’s its saving grace.
🌄 What changes are on the horizon?
Did you know books didn’t always have page numbers? They didn’t always have indexes, either; nor bibliographies, nor tables of content. These affordances evolved over centuries by the active use and development of the book as medium.
The internet will evolve its own affordances as we discover what tools we need. The convocation (which I wrote about in Zones of the mind)), where users and developers mingle in backstage familiarity, is the place to watch.
Ecologically, the edge between two zones will support creatures from both zones, as well as creatures that live only on the edge. The interface is more diverse than the sum of its parts. Look for the place where users are hacking their platform, finding their own uses for things. That’s where new gestures will be developed.
Look, too, to the emerging interactions of human and machine intelligence. I know, it’s unsettling to have a robot in your house, always listening, always snitching to Amazon or Google or who knows who else. But to whatever extent you feel comfortable, let the computers process information for you. Let your phone remember your dentist appointments. Let your podcast be transcribed by AI. The advance of natural language processing will make computer interactions much more intuitive, unlocking the vast repository of social gestures developed over all human history.
Rather than packaging each of your creations into a marketable form, create a giant tangle of self-referential data. The tools to search and parse this external mind will only get better.
We’re all already being parsed, by governments and corporations, satellites and doorbells. Might as well make use of it for ourselves.
Thanks for reading,
Robot Face is a free newsletter about the interface between human and machine. It takes a lot of time and energy to write, but hopefully I make it look easy. ⛹
If you want to help me out, boost it to your social media or forward it to your friends.
And of course, I always love to hear back from you! Just reply to this email.