OK, first some business: I'm going to start sending this newsletter regularly again, and I'm renaming it from "Future of Discovery" back to simply "Jacob O'Bryant's Newsletter." I'm also merging it with my little "Software Invention" newsletter (a bit more about this below) which I set up a few months ago, got several subscribers, and never sent a single issue of. (To the six of you who weren't already subscribed to Future of Discovery, hi!)
I'm on a quest to help people share and discover meaningful information on the Internet. That started with recommender systems, but I've since expanded my scope to include everything in one's "information workflow." Besides working on The Sample, I think a lot about individual parts of that workflow, how to unbundle them from social media, and how to rebundle them in ways that serve individuals.
Personal newsletters are one of those parts. I think just about anyone who lives on the internet should have their own newsletter. What if having a newsletter was as common as having a Twitter account? The way you'd write it would be different, for sure. Right now, having a newsletter is usually at least a small commitment, as evidenced by the fact that they usually have titles. No one gives a title to their Twitter account.
One of the problems with social media is that, thanks to all that growth, there are too many posts, and it's hard to know which ones are worth showing you. And as a poster, your posts will only ever be seen by a small fraction of the people who follow you. Benedict Evans talks about this in The death of the newsfeed. He describes how Stories are one attempt to address that:
Meanwhile, you could propose that the Stories format that Snap invented, and Facebook... continued to invent, is also a way to address overload [...] maybe Stories mean you share more things, but by bundling them into one thing you place less load on your friends and reduce the need for a filter.
Newsletters can serve the same function in a platform-agnostic way: instead of trying to compete with the rest of people's Twitter/Facebook feeds, you could bundle a few good posts into a weekly newsletter. I think the format of a newsletter lends itself well to less volume and more quality.
I think it would be extremely interesting for a newsletter platform to go after this use case and make it go mainstream. Imagine an ambitious version of TinyLetter that didn't get killed by Mailchimp, in contrast to most existing newsletter software which is built for people who are trying to make money from their lists in one way or another.
Maybe I'll build this some time after The Sample makes enough money for me to start taking a salary, but in the mean time I'm trying it out myself as a guinea pig. Especially since I miss writing regularly. I'd like to develop a workflow which allows me to send out a quality email weekly without having to spend much time on it. I'd like the newsletter to be a light curation of my work, thoughts and discoveries from the past week; not an additional piece of work.
I won't stick strictly to any particular topic, in the same way that I don't do that on Twitter or Facebook, which is why I'm removing the title and referring to this just as "my newsletter"/"Jacob O'Bryant's Newsletter". I'll no doubt talk a lot about information workflow stuff since I'm interested in that, but I'll also talk more about software invention (hence merging in the other newsletter) and whatever else is on my mind. I'll try to format the newsletter in a way that you can easily skip over any topics that aren't relevant to you.
And finally: as part of my workflow experimentation, I've made a "Software Inventors" Discord server to go along with this newsletter. There's a #jacob-obryant channel where I'm happy to chat about anything. Here's an invite. The idea is that anyone who identifies as a software inventor can have their own channel and get some discovery by cross-pollinating with other channels in the server. But for now, it's just me.
Web3 without blockchain. Twice recently I've seen people say that what they find compelling about web3 is the possibility of having lots of different apps operating on the same shared data without being controlled by a commercial platform. The funny thing is, I'm totally on board for that vision! I'd just rather get there via plain, boring protocols (like HTTP, RSS, SMTP, and maybe some new ones) rather than trying to do it all on blockchain, which for most use cases would be over-engineering.
The most interesting pro-web3 argument I've heard is that regardless of technical aspects, there are a lot of people interested in the movement, and that could drive change. Also, "web3" is a much catchier name than my "user-centric data". I wonder if web3 could be redefined to be primarily about the top-level goals rather than the implementation, in which case I'd love to be part of it.
Nadia Eghbal on meaning in the creator economy.
The Melancholy of Subculture Society, from HN a few weeks ago. This made me feel melancholy. It feels like an incomplete model, though—status isn't our only source of satisfaction.
Interoperable serendipity. Can you have two applications that work on the same data without coordination from at least one of the authors? Personally I think the answer is "no" and thus we should look for other ways to get good software (interoperability is just a means to that end IMO), but I'm glad people are thinking about it.
Not so safe, is it?
We can be friends until the revolution.
I'm sure 2022 will be different.