Advertise here with Carbon Ads

This site is made possible by member support. ❤️

Big thanks to Arcustech for hosting the site and offering amazing tech support.

When you buy through links on, I may earn an affiliate commission. Thanks for supporting the site! home of fine hypertext products since 1998.

🍔  💀  📸  😭  🕳️  🤠  🎬  🥔

The Future of RSS is Textcasting

RSS widescreen.jpeg

Earlier this week, Dave Winer (the inventor of RSS and lots of other notable software, and one of the earliest bloggers) made a podcast for me. He posted it publicly, so it wasn’t sealed like a letter, but it was addressed to me and responding to something I wrote on Threads. I don’t think I’ve ever been the personal addressee of a podcast before: like an @-reply, but in audio.

For some reason, it reminded me of the poet Frank O’Hara’s “Personism” manifesto:

[Personism] was founded by me after lunch with LeRoi Jones on August 27, 1959, a day in which I was in love with someone (not Roi, by the way, a blond). I went back to work and wrote a poem for this person. While I was writing it I was realizing that if I wanted to I could use the telephone instead of writing the poem, and so Personism was born. It’s a very exciting movement which will undoubtedly have lots of adherents. It puts the poem squarely between the poet and the person, Lucky Pierre style, and the poem is correspondingly gratified. The poem is at last between two persons instead of two pages. In all modesty, I confess that it may be the death of literature as we know it.

But this is a digression. The ideas Dave is talking about in this podcast are serious (even if he is laughing a lot), and he spells them out in text at a site called Here’s the philosophy:

  • The goal is interop between social media apps and the features writers need.
  • What we’re doing: Moving documents between networked apps. We need a set of common features in order for it to work.
  • The features are motivated by the needs of writers. Not by programmers or social media company execs.

It’s a proposal to build, using technologies we already have and understand very well, a very simple social media protocol that is completely agnostic about what editor you use to write your posts and what viewer you choose to read it. Writer/authors would have more control over styling, links, media enclosures, etc., and readers would have more control over how and where they consume it. It’s decentralized social media, but without the need to peer through ActivityPub or anybody else’s API and squeeze our toothpaste through its tubes.

Dave asked me to respond to his podcast, and while I thought about making an audio response, a blog post is more my metier.

If this is going to work, I think there are at least four problems we’d have to solve:

  • Everyday users need a default writer and reader, preferably in the same place. It would be wonderful to be able to bring your own tools to bear and plug into the endpoints and have it just work. But nobody should need to fish around for that stuff. The defaults should be just opinionated enough (for instance, I like the Mastodon client Ivory, built on the ashes of TweetBot) that people feel like they’re getting a slick, finished UI and UX experience. That’s why people are gravitating to Threads: Meta knows how to put one together. But that’s not rebuilding the electrical grid from scratch. We know how to do that too.

  • We need user and content discovery. We have to be able to find each other. And that’s what a lot of the heavy lifting of these other platforms is devoted to: to find people and things you didn’t know you were looking for. Again, this feels like a solvable problem.

  • We need metrics. Most writers work for publishers, and publishers want to know whether their content is reaching people. It’s tempting to load this up with a lot of cruft, but impressions, clicks, reposts, quote posts, etc. is doable without a ton of advertising nonsense. Social media companies figured this out, but bloggers did too, a lot earlier.

  • We need moderation. A wide-open social network becomes a highly abusive social network fast in 2023. And nobody who isn’t a Nazi wants to hang out at a Nazi bar. So we need tools to block and flag and maybe even ban people who use the platform to spam each other, impersonate real people, and abuse other people on the platform. In the now-seminal “Welcome to Hell, Elon,” Nilay Patel correctly identified content moderation as Twitter’s actual product: it’s what Twitter-the-company added to Twitter-the-protocol. We don’t necessarily need a Twitter-sized overhead, or a complex system of federated moderators, but it’s going to be a problem, so we’d need tools to address it.

But in terms of philosophy and vision, I’m all for it. RSS remains incredibly vital to what I do, and its potential as both a reading AND a writing platform remains untapped. You can write using your own tool and broadcast it everywhere. And Dave’s right: this worked for podcasts (the phrase “anywhere you get your podcasts!” is a great advertisement for interoperability breaking any single platform’s dominance), it worked for blogs, and it can work for this strange multimodal thing we’ve created called social media. It worked for the world wide web! And I will be ride or die for the open web until my life comes to an end.

Now we just need to work together to make it happen. And I confess: my tools are limited here. I can’t (really) code, I can’t (really) design, I can’t build a moderation feature. I can evangelize, I can strategize, and I can write. But I can do those things (in all modesty) very, very well.

So let’s do this thing. Why not? Twitter is dying, and Facebook is fading. None of the replacements have eaten their lunch yet. Why not make a swing for the open web? Why not try?

Discussion  6 comments

Anil Dash

In what substantive ways would this be different than, say, starting a WordPress blog?

Anthony Sorace

Listening to Dave’s podcast I kept thinking “he’s describing blogging”. Which is great! I’m 100% down with the philosophy and we need more people writing blogs, but we have blogs. I think Tim’s right about discovery, in particular, feeling like something we need to work out.

Tim CarmodyMOD

Wordpress does a whole lot of other stuff. It’s a web page/web site builder. It’s a monolithic CMS. This would be a lot more lightweight, and (to use the current term of art) headless. It would be optimized for social and mobile, not for designing and laying out webpages.

But to a certain extent you’re not wrong: as I understand it, it’s much closer to “what if an blog-first organization like Wordpress built a social media network?” than what we’ve inherited from Friendster onward.

Matt Thompson

(Hi, Anil! What a treat to again encounter sharp questions from you on a comment thread under a blog post! Where are we, 2009? This is awesome!)

I wonder if there's a link between what's being puzzled out here and something Tim just linked to from Robin, several years back, about the wish for more purpose-built tools, built not for scale or standardization. There's a whole strand of Robin's writing and creative work in the years since that expands on his love of protocols, which I think relates to this.

Wordpress (not to mention Ghost and, heck, Movable frickin' Type!) is immensely powerful, modular and flexible, but the defaults baked into it imagine variations on a theme of creating written or visual posts. It is a great tool for posting to the RSS protocol.

RSS is generic enough that it basically works for all sorts of posts. Technically, early Twitter could probably have been built on Wordpress, right? In fact, I would not be surprised if Wordpress was already advanced enough by 2007 for most of the posting functionality of Twitter to be loosely replicable by a clone. But Twitter started with a surprising feature — a tight limit on how much you could write, and soon enough, other features began to develop around that constraint.

But after years of managing all sorts of content, I've come to believe that expression might be more specific than a generic protocol can fully do justice to. And while I think RSS is fantastic for a wide range of posts, including podcasts, I don't think it's necessarily right for the full spectrum of speech and expression.

I think Robin and the network of protocol fans he's attached to might be right when they say protocols deserve widespread attention and creativity. Protocols so complex we follow them subconsciously govern how we communicate with each other in offline life. They're encoded in our languages, our mealtime rituals, our queuing practices, our willingness to interrupt one another in speech, our gift-giving traditions. You'd think our protocols for exchanging information online might begin to reflect that complexity. But instead, we've defaulted to protocols that are built for widespread scale and flexibility, and suitable for producing all sorts of content.

I'll confess, Robin's essay last November did sort of make me want to write a protocol. His closing salvo:

You feel it, don’t you? They’re all crumbling, the platforms of the last decade. It’s unsettling, but/and also undeniably exciting. Tall trees fall in the forest, and light streams down, nourishing places it hasn’t reached in ages.

But we, as users of the global internet, cannot just ride the same rollercoaster again. It’s too embarrassing to be trapped inside these hungry corporate gambits, these dumb proper nouns. The nouns and verbs of our online relationships should be lowercase, the way “magazine” is lowercase, the way “movie” is lowercase. Anybody can make a movie. Anybody can try.

New protocols abound, and more are on their way. In the months and years ahead, as you explore and evaluate them, I encourage you to ask this question:

Does this protocol recreate something that already exists?

The opportunity before us, as investigators and experimenters in the 2020s, isn’t to make Twitter or Tumblr or Instagram again, just “in a better way” this time. Repeating myself from above: a decentralized or federated timeline is still a timeline, and for me, the timeline is the problem.

This digital medium remains liquid, protean, full of potential. Even after a decade of stasis, these pixels, and the ways of relating behind them, will eagerly become whatever you imagine.

So: imagine!

[I need a part 2. I am ashamed. It's been a long while since I blogged.]

Matt Thompson

On a more fully social internet, built around the thoughtful sharing and stewardship of information, I imagine our protocols would aim to respect the complexity of human interactions in an offline world. I'd be able to express information about myself in forms that made sense to me and those in the network I wanted to share information with. People would be able to plug that information into their own applications in ways that make sense to them and those in their networks.

I find myself thinking most about birthdays. Birthdays are the last feature I found valuable on Facebook. For a while, until going to Facebook began to cause active pain, it was so sweet to be able to see people's birthdays and be reminded of them, and get to send them a message. And to make that possible, someone once upon a time in Facebook had to have created a birthday object they could begin attaching information to, like who to share the birthday with.

If we had a birthday protocol with a robust client, I could share useful information about birthdays with any other client willing to respect my rules around it. There are all sorts of information about birthdays I would want to see across platforms and applications. I would want my birthday client to be accessible to a range of people in my life.

Birthdays are especially notable to me, of course, because we have a host of rituals around them. We associate flavors and songs with birthdays. We build memories around them. We pour meaning onto them. How and when one mentions and celebrates a person's birthday has mysterious protocols around it. But nothing is more mysterious than why my experience of a birthday on the internet is locked into a particular platform.

Matt Thompson

When I imagine birthdays on a different internet, I don't imagine a cute meme. I imagine a document that grows richer with time, with recipes as well as photos. Something far more like a scrapbook than a card. The time and effort it would take to assemble this for myself is significant. But I know what it's like to do this in the tangible world, and I would love to imagine a protocol for doing this in the digital one.

The last most meaningful interaction I had on Twitter was almost five years ago, when I played a little game on the platform called "Break the Ratio." I asked my followers to only reply to tweets they found meaningful, rather than retweet them. I wanted to remember for a moment what it felt like for Twitter to be conversational, so I wanted to see only replies. I did not mind likes, but I considered them akin to "this is ok." By a not-strict definition, this was a protocol.

Under those rules, I asked people to share compliments they thought of with particular people in mind. If the compliment resonated with me, and made me think of someone I know, I'd figure out a way to share it with them. People posted beautiful compliments about others in their lives, and they did make me think of people in my own life, and following the exercise led to the last good tears a Twitter thread ever made me cry.

I relate this, because I think to invent a different internet, we have to imagine beyond a sea of generic content created by authors for wide audiences. We have the ability to create real human dialogue across space and time. For a brief shining moment, stitches and duets on TikTok were vehicles of glorious creativity and collaboration. I would love to replicate some of those features in an environment that wasn't desperately hungry for scale. 1

I find Robin's cautions both cryptic and meaningful. I am both entranced with my birthday internet and fearful that it would become a timeline, which is, I agree, the problem." Facebook is also a reminder that someone I like a lot has a birthday nearly every day (Tim, yours is still going here in California!) and a daily celebration is literally quotidien. I do not want my birthday internet to feel quotidien. I want it to feel special and precious. It is a place I want to go, not a place I require notifications to nudge me toward.

I am losing my interest in atomized pieces of content algorithmized into a novelty-rewarding timeline. I would like to make molecules of information with people on the internet. I want to be able to control how this information is exchanged and with whom. I want vehicles for real and meaningful dialogue with chosen networks of people. And I do want platforms to help smooth the process of creation and exchange, but I want them to operate by rules that feel clear and humane.

This comment is already so long, it's almost not Tim's birthday in California. But I Googled "social protocol" and I quickly came upon two links that seemed interesting in contrast to one another. One looked like a path into a nightmare, and the other looked like a fascinating rabbit hole. I will leave it to you to determine which is which.

  1. I would 100% gather with fellow college a cappella alumni on BirthdayTok to duet the birthday song every year, only for fellow college a cappella alumni and fans of same.

This thread is closed for new comments & replies. Thanks to everyone for participating!