A news aggregator for a galaxy brain undazzled by virtue fireworks

A lot of people miss Google Reader, but several companies have tried to build applications that imitate what Google Reader offered without ever really nailing it. Maybe it's time to start somewhat fresh and build something that does what Reader did well, while also drawing on some of the lessons learned in recent years. Plus innovating a little.

And so, I propose, a new news aggregator.

The point of this aggregator would be to create a site that has just enough social for you to take advantage of the hive mind but limits it enough to undermine the performative incentive that make social a cesspool of toxicity and virtue fireworks.

The business intent would be to build something that's a public service. More a "stayup" than a startup. Something that can stay quite lean, too.

So I am going to describe a design. Not a design in terms of how it looks, but in terms of its basic user experience. I'm going to do it with words because, well, I use words.

Here's the short version: It's a news aggregator that lets you see what people with similar interests and sensibilities are reading. Comments on the service work more like a group text. Identity on the site bends more toward Wikipedia and less toward Instagram. It's a site much more about text than pictures. It's not at all about video. It will never pivot to video.

It would be a news aggregator that's social but it isn't needy. So here are the key points about it. I have sketched these out in even more detail, then I realized it was too much for this blog post. So this is the simpler version — if you're curious for more, let me know. Here we go:

First of all it has a front page, and stuff gets voted to that front page, but the stuff that gets voted there doesn't get associated with the people that posted it. The front page is open and public, like the front page of Twitter was in the very early days and the front-page of Reddit now, but without its bragging rights.
Other than the front page, every user will have their own personal page with three very clearly marked tabs: mine, social and select (let's say).
Mine: This is the most like Google Reader. It would use RSS and scraping to show you the metadata of updates, with the link, to changes on websites you are following (header image, title, summary, date, maybe keywords). If you want to read it, you gotta click through, though.
Social: This is all the stuff people you follow are reading, posted here in the order they read it. When the site's server get pinged that they looked, it goes into the social feed for everyone that follows them.
Select: This is where the people you are following are curating for you (and you for them). People can hit stuff to be shared. When they do that, they can't write any new text, but they can slap an emoji on it (just one) and they can boost it with a limited set of votes they have to give out each day. They can give all their votes to one thing or vote once for multiple things. It's up to them. It would run on some sort of quadratic voting system.
Select would only sort by shares and votes. That's it. No more hoodoo under the hood to try to read people's minds. If you follow someone, then the stuff they share and vote for gets sorted accordingly on Select. If other people you follow vote the same link, it gets pushed up further, but no mystery in the sorting scheme. Votes, by thew way, evaporate over time.
On the emoji: The idea here is that different emoji would take on certain meanings within the site. Eventually that would help determine categories that could be used to flesh out the front page from a straight firehose to something more nuanced that had sections, like a newspaper does but without completely copying newspaper sections. The emoji should make that more dynamic (like maybe one week there's a cartwheel section and the next week there is not, but apple replaces it).
Everything would default to preserving privacy, which means you can always hide stuff and the app errs on the side of deleting data after a reasonable amount of time.
Your feed stops saving stuff further back than 30 days. If you haven't gotten to something by then it just disappears. No one needs that unread posts thing growing infinitely.
There are comments in Social and Select, but only mutuals can see the comments (so if you follow someone who doesn't follow you, you see what they read and what they share, but not what they say). Comments also auto-delete after 30 days. They are end-to-end encrypted between the parties in the conversation, so that even the site's admins can't read them.
The front page is determined by shares and votes across the whole user base of the site.
This probably goes without saying, but it's not essential for a user to follow a site to share a link. They can always cut and paste a link straight into their feed.
Business model, freemium. If you don't pay, you can't share or follow folks or vote, but you can still keep track of a bunch of sites and scan the front page. Payment would be super cheap. Let's say: $1/month or $5 per year. Why? Because price elasticity of The Man.

Furthermore

We all know that a variety of other factors that have nothing to do with people's curatorial brilliance determine prominence on social sites. So a few more design elements:

The first browser always optimized for is Tor. Why? We're aiming for a niche first before the whole world. So then Thunderbird, then Opera, then Brave, then Edge, then Chrome. Chrome last. Always.
Everyone's profile has an avatar photo but it can't be a face. It can't even be a simulated face or a representational face. Machine learning and some other stuff. It will work.
No real names.
No user search. If you don't have the link to a user's profile, you can't follow them. Maybe we'd introduce some discovery early on but the goal would be to get rid of it.
People would have a public front page that would show what they were into at the moment, but of course that could be hidden.
You can 100% use the service without a login, you just won't get the share feature for stories, nor votes and you won't be able to comment. Follow-only, both news sites and people. No problem. I honestly think a lot of people would want this. It's cool. Twitter should do a version of this.

Hit me up if you want to build this. I don't know what I could do substantively but I'm not a bad hype man.

—Brady Dale
April 25, 2020