Tools for Online Speech logo
by Jacob O'Bryant

TikTok for reading

In case you missed it: You can help unbundle social media

Recently I've exchanged a few DMs with someone about which of Yakread's value props to emphasize: should I structure and describe it as a "smart newsletter reader" or as "TikTok for reading?"

Right now, the landing page and the app's UI reflect the former. Yakread is an app for managing the subscriptions and bookmarks you already have. Yakread's timeline is algorithmically sorted, but it's mostly just re-arranging the content that you've already subscribed to/bookmarked.

However, there is another important feature: if you don't add any of your own subscriptions or bookmarks, then Yakread will use its algorithm to populate your timeline with content that was liked by other users. I added this feature so that brand new users would be able to try out Yakread right away, with the expectation that sooner or later they would add their own subscriptions.

But what if we did the inverse? In the "TikTok for reading" model, there's really no need to subscribe to anything unless you want to. Yakread shows you a scrolling timeline, you click on the first article that looks interesting, repeat. The more you use Yakread, the better it gets at finding articles you'll like.

And then as an additional benefit for power users, Yakread gives you lots of customization options for your timeline. If you add your own subscriptions and bookmarks, then those things will get mixed into your timeline. Same for your ebooks and Mastodon/Twitter accounts. You could switch between the algorithmic timeline and a pure chronological timeline. And so on.

Of course, Yakread already does all that. The question is just whether those timeline-customization features should continue to be the whole point of Yakread, or should they be more of a oh btw you can do this too thing.

What exactly are the changes I'd make if I went all-in on "TikTok for reading"? For starters, I might simplify the navigation bar on the left:

Instead of having "Subscriptions," "Read later," etc, maybe I'd just throw in a "Customize your timeline" button somewhere:

That button would let you add your own subscriptions etc, and perhaps you could create timeline presets that would show up as tabs or pills:

So the customization options would be within arm's reach for those who need them, but they wouldn't distract or confuse everyone else. (Maybe I should make the "Share" and "Advertise" navigation links less prominent too...)

Next I'd redo the landing page copy:

Instead of "Add your content" and "Read more, triage less," I'd say... something. "Read interesting stuff"? "Read cool blog posts that our algorithm finds for you"?

Or maybe I should get rid of the landing page altogether. I took a look at what TikTok does, and they didn't show me one—instead they immediately dropped me into the app and started playing a video. Same type of thing for Reddit. Maybe I should do that with Yakread! Just show people a timeline of posts right away even if they're not logged in. The algorithm can begin learning their preferences and give personalized reading recommendations whenever they visit from the same device. If you click the "Customize your timeline" button, then you'd be prompted to create an account first.

(Discuss this on Discord | Join Discord)

How can I make this newsletter helpful for you? Although I have so far mainly written about stuff I'm working on, I want this newsletter to become more generally a valuable community resource for people interested in tools for online speech. I want to include both developers and early adopters—we don't all have to be nerds here 🙂.

So, what would you like to get out of reading this newsletter?

  • Advice for contributing, like this and this? I could spend more time writing essays.
  • Curated links to blog posts and projects from other people in the tools-for-online-speech space? I could do in-depth reviews of projects other people have built. Maybe do some interviews or guest posts. I could put more work into the recommended links section (e.g. start adding commentary again).
  • Opportunities to meet, discuss, and collaborate with others in the community, e.g. via the Discord server? I could post on Discord more myself and encourage others to share their work and ideas there.
  • Something else?

I dedicate Mondays to this newsletter/community (even if much of the time I get sucked into just tweaking the theme on my website), so I have some time to work with. Right now I'm thinking of doing something like this going forward:

  • My Tuesday posts (like this one) will be off-the-cuff "lab notes," i.e. whatever I happen to be thinking about at the moment.
  • For each lab note, I'll make a separate post on the Discord forum, so if you want to discuss it, there's a nice place already set up to do so. I'll end each lab note with a link to its associated post (as I've done here).
  • After I finish writing the newsletter for Tuesday, I'll use any remaining writing time to work on an essay, or a product review, or interview, or something. Whenever I have one of these more finished posts ready to go, I'll send it out separately on Thursday. (So I'll always send something on Tuesday, and sometimes I'll send something on Thursday as well.)
  • I'll continue sharing recommended links at the end of the newsletter. If there's a link I want to give extra attention to, I'll write a lab note about it and put it higher in the newsletter. That will replace the "highlights" section I've included a couple times previously. Also, instead of my "send me a link" form, I think I'll just throw in a prompt that says "If you have something you'd like to share, post it on Discord." And of course, I'll see anything that gets posted on Discord, and thus it'll have a chance at getting into the newsletter.

(Discuss this on Discord | Join Discord)

Why social media is entering the "post-viral" age (MidRange)

Intentional Computing: How to Use Tech, Without Being Used By It (The Autodidacts)

i think my son hates me (Both Are True)

The downsides of open protocols (Moxie Marlinspike, 2016). See also Scaling Mastodon is impossible from Armin Ronacher.

For your next side project, make a browser extension (Geoffrey Litt)

Social/UGC products are uniquely difficult marketplaces (Jackson Mohsenin)

Newsletter financial infrastructure (Bits about Money)

Unbundling Tools for Thought (Fernando Borretti)

I don’t like any of the funnel or cohort analysis tools available to me (Vikram Oberoi)

For 2023, no predictions, but a wish (Heather Burns)

What does it look like for the web to lose? (Chris Coyier)

Banning noncompete agreements would be good for workers and the economy (Full Stack Economics)

Also check out my recommended newsletters.

(Discuss this on Discord | Join Discord

Published 17 Jan 2023

I write an occasional newsletter
about my work and ideas.

RSS feed · Archive

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.