Post

🦇 DEADvin 🎃 GHASTney 👻
@devingaffney.com

@danabra.mov has been on a bit of a tear with a really good mantra: "we can just do things". It feels like there's something in the air as of late, too - tooling, community, and network effects feel like they're finally hitting an inflection point in ATProto. Can we get at it with data? 🧵

17 reposts130 likes

🦇 DEADvin 🎃 GHASTney 👻
@devingaffney.com

There's a ton of ways to cut it, but I think the simplest way is asking something like "how many projects have been abandoned vs are actively being worked on". If the scene is dying, we'd expect collections strewn across PDSes, not having any new records for weeks or months.

🦇 DEADvin 🎃 GHASTney 👻
@devingaffney.com

Someone gets excited about ATProto. Creates app.coins.coin, creates app.coins.sleeve to hold them in a digital sleeve. Gives up after a few users try it, its just not the right storage, ecosystem, what have you. We'd expect to see *lots* of these if this was a dead ecosystem. If we don't... its live

🦇 DEADvin 🎃 GHASTney 👻
@devingaffney.com

In fact, 80% of non-BSky collections are under active development with records written in the last week - 90% in the last month. Almost *no* abandoned projects where more than a small handful of uptake. The ecosystem is ~sticky~. Why?

🦇 DEADvin 🎃 GHASTney 👻
@devingaffney.com

This is the same thing @danabra.mov is getting at - "we can just build things". In ATProto, the cold start problem is easy. Getting some users is easy, we all have a passport to each other's apps. Booting with some content, some social graph, etc, is native. Also, no postgres - just an app, *now*.