Try out the Yakread pre-release

Welcome to my weekly newsletter about practical ways to make the Internet better, focused on my own work in that space. I'm Jacob O'Bryant.

I'm deep into coding mode, finishing up the last batch of essential features and bug fixes before I release the Yakread rewrite I began several weeks ago. You can use it right now if you head on over to yakread.com/home. You don't even need to be signed in.

Besides the general redesign and heightened amount of polish—intended to give an accurate first impression of the app's maturity—there are a number of conceptual differences in the upcoming release which are worth highlighting.

Emphasis on algorithmic recommendations

As discussed previously, this was the main reason for doing another Yakread overhaul. I wanted to make Yakread more frictionless so that people can get value out of it without needing to add their own newsletter subscriptions first. As such, the home page is now populated immediately with algorithmic recommendations, the quality of which—if I may say so myself—is not bad. This turns Yakread into "TikTok for blogs," basically. You can still add your own subscriptions and bookmarks, and they'll get mixed into your feed.

Better support for manual curation

As a complement to the algorithmic feed, Yakread has something that's even better than a chronological list of articles: a chronological list of your subscriptions.

The algorithmic feed is great when you have only a few minutes to spare and you want to jump into reading something right away. If you have a larger block of time, you can scan your subscriptions for individual publications that you'd like to catch up on. Clicking on one will present a chronological list of posts from that subscription. After you pick a post to read and scroll to the end, the list of remaining posts will be displayed again, so you can repeat until you're ready to move on to a different subscription.

I'm going to also add a "pin" button so you can choose a handful of your favorite publications to keep at the top of the subscriptions page.

Sharing and comments

This bit is only half-completed. Currently there is a "share" button which triggers a popup menu like so:

I'm going to rename that button to "comments." When you press it, Yakread will search Reddit, Hacker News, and anywhere else possible (Bluesky?) to see if there are any existing discussions for the current article. Yakread won't host any comments itself, but it will aggregate discussions from elsewhere on the web. If there aren't any discussions yet, then you can always use one of the share buttons to start a new discussion.

I'm interested in this for two reasons. On a personal level, it's an approach to social media that might actually click with me. I've made several feeble attempts in the past to get into posting regularly on Twitter and (more recently) Mastodon, and kinda-sorta Substack Notes, but I'm not sure if it'll ever stick. A few weeks ago I culled my follow lists and made a point to do some scrolling each morning and look for things to reply to. But the signal-to-noise ratio was just too low to justify the time it took.

The Yakread feed, on the other hand, is marvelous. It always has meaningful things for me to read. If I can effortlessly look up comments for articles I've read, maybe then I'll be able to talk more to other people instead of only being absorbed in my own work (the latter being quite a temptation for me as an introvert). I also like the idea of centering discussion around articles I've actually read.

Second, this discussion aggregation will fulfill my vision for an unbundled web where reading, publishing, and discussion are all separate and interchangeable. Like any other RSS reader or email client, Yakread doesn't care where the articles you read are published. Writers have freedom to post their content anywhere, which makes publishing apps more competitive—they have to compete on features and pricing instead of network effects.

So Yakread will soon do the same thing for discussion: if reading apps aggregate discussion from anywhere, then there's less need for all discussions to happen in the same place. More competition, better consumer choice. And with publication and discussion being outsourced to and aggregated from other locations, Yakread itself is subject to competition. Yakread has no privileged access to anyone's content. Anyone is free to make a different reading app that has all the same access to content that Yakread does; no gatekeepers.

Thus all three parts of the media ecosystem—publishing, reading, and discussion—will be subjected to competitive pressure and forced to improve and evolve instead of relying on network effects for their continued survival, until the Singularity occurs and the Earth becomes as a sea of glass and...

(OK maybe getting a little carried away, but you get the idea.)

Stuff I'm removing

Just as important as the things I'm adding are the things I'm taking away. In the first announcement for Yakread, I emphasized how Yakread takes any form of written content and munges it into article-length: books get divided into sections and tweets get compiled into longer posts. So your entire Yakread feed mostly of things that take about 5-10 minutes to get through, and you have a single place you can go to for all your reading (helpful for building a habit).

While a worthy experiment, I've decided those things are nonessential: better to focus on content that's been written specifically for the article format. You can put a bunch of tweets together in a batch, but that doesn't make them any more worth reading unfortunately. It turns out that my solution for reading Twitter has ended up being "just stop reading Twitter."

As for books, I personally have dedicated Sunday to reading them. On Monday - Saturday, Yakread is my go-to reading app; on Sunday, it's my ebook app. Books aren't written to be chopped up into bits anyway—they work better with larger chunks of time.

I'm also getting rid of the recommendation profiles. Again, an interesting experiment. However since implementing them, they've just felt kind of "meh." Conceptually there's no reason they need to be built into Yakread instead of some other app (like a publishing app), and I don't see them being a meaningful part of Yakread's growth strategy.

I've even decided to nix the Pocket integration (same for Instapaper). Not many people use it; Pocket itself is buggy and in maintenance mode; and I think better to just provide a URL form + bookmarklet so people can add bookmarks directly without needing to go through a 3rd-party service. Eventually perhaps I'll hire a freelancer to make some web browser extensions/mobile apps so that bookmarking things is more convenient.

---

Stuff I've read recently: Bell Labs didn’t give researchers as much freedom as you might’ve thought. Cultural differences between Bluesky and Mastodon. The economic benefits of blogging (though I suspect there’s some availability bias here). How Twitter’s algorithm really works. A review of the Arc browser. A series of articles on supreme court cases relevant to the first amendment. Mistakes from a first-time manager. How the EARN IT act would backfire.