👀

Haven't been super on top of my flashcards — family stuff (fantastic! but time-consuming). Plus, I burned a couple of nights researching governance and social media theory, prototyping, thinking; I, ill-timedly, adjusted my caffeine intake just as our little family needed more of me; and I spent a couple recovery evenings on RuneScape: Dragonwilds too. Time to lock in, though: JLPT registration for the December test is this coming Monday (8/24)!

👀

If the problem isn't privacy, and it isn't just feeds that behave badly, but governance models built to serve the platform operator rather than the community using them, what shape do solutions start to take? 💭

👀

As Urbit collapsed, I came to realize that the fantasy of individual sovereignty on which it was built is essentially anti-political. It regards political and social ties—the fabric of cooperation—as mere bondage. It assumes that pure self-reliance is both desirable and feasible; in my experience, it is neither. Our dependence on others does not make us less human. On the contrary, it defines us as social beings.
Urbit’s anxious disavowal of dependency appealed to rugged “power-users” eager to exit MEGACORP for the freedom of the digital frontier. By offering users ownership of “galaxies,” “stars,” and “planets,” it promised a virtual liftoff from earth and into outer space. But the same fantasy that attracted users contributed to Urbit’s dysfunctional governanc. It obscured the role played by concrete human relationships in the real world in operating and maintaining this (and any) technology. As a result, Urbit has come to resemble Web 2.0: an enshittified infrastructure controlled by a small handful of actors immune to democratic accountability. If Yarvin’s political vision proved to be an unsound model for governance for a company, one can only imagine how it would play out on the scale of a country.

I'd written another LinkedIn post about CUI and software development (mostly about test harnesses I've built to improve confidence over new features), I'll post that one here too.

Testing MCP with Claude Code running locally, main benefit right now seems to be that the combo automates some parts of my development workflow, e.g. no more manually pulling named branches from browser-based sessions, then copy and pasting error messages as I debug getting new features working locally. Local access to the live app also appears useful even if I'm writing features by hand, plus read-only database access means some tasks, like checking that seed data actually landed, are easier? Though I probably still want the seeding itself handled reproducibly, as artifacts under version control in the application's source code. So feature work still happens in a regular Claude Code session, just with these tools along for the ride whenever something needs checking against the running app. Neat!

Tertulia on localhost is officially MCP-enabled!

Oof, swapped my coffee around. Bad idea, caffeine headache from what used to be my regular daily intake 😵‍💫