724: Ask Dan Abramov Anything About AT Proto

The podcast discusses decentralized social media platforms, focusing on BlueSky and its underlying AT Protocol. Unlike Mastodon, which relies on interconnected instances that complicate data sharing, AT Protocol enables decentralized networking by allowing content to be directly pulled - similar to how RSS works - making it more scalable and simpler. The protocol supports user-controlled data through Personal Data Servers (PDS), where individuals store their own content and interactions, enabling ownership and portability across apps.

The conversation explores the design of decentralized identity and data ownership, advocating for a people-first data layer on the web. Users can authenticate via domain-based identities and control their data using cryptographic keys, reducing reliance on platform-controlled "walled gardens." Apps build on this infrastructure by interpreting standardized data formats (lexicons) and creating optimized views, but the true source of truth remains in users' personal repositories. The discussion also highlights efforts toward standardization, such as the Standard Site initiative for blog interoperability, and experimental uses like decentralized scrobbling and media tracking, all promoting a vision of user-owned, interoperable digital ecosystems.

• What if you turned your personal blog into a decentralized, interoperable data source using Standard Site?

  • Move: Convert your static blog (e.g., Jekyll, 11ty) into a Standard Site-compliant feed by adopting the shared schema (publication, post, subscription) and hosting it with a plugin or tool like Adapult. Publish JSON-based records of your posts to your Personal Data Server (PDS) or public GitHub repo.

  • Why Now?: The Standard Site format is gaining traction across indie platforms and readers; early adopters can position their content for cross-platform discovery (e.g., in "Standard Reader" aggregators) before lock-in intensifies elsewhere.

  • Expected Upside: Gain portable, user-owned content that apps can ingest natively - increasing reach without surrendering control. Enables POSSE (Publish on your Own Site, Syndicate Elsewhere) at scale with minimal maintenance.

• What if you built a niche social app using AT Protocol as your backend data layer instead of a traditional database?

  • Move: Use the AT Protocol's lexicon system to define a custom schema (e.g., app.bsky.media.review) for a vertical like book or movie reviews. Connect it to a PDS so users own their data while your app acts as a client that reads and renders it.

  • Why Now?: AT Protocol tooling (e.g., TypeScript lexicon generators, BDSLS browsers) now supports rapid prototyping. The ecosystem is ripe for early entrants before dominant apps capture attention.

  • Expected Upside: Reduce backend complexity (no custom auth, storage, or sync logic); leverage existing identity and hosting infrastructure (e.g., domain-based DIDs). Launch faster with zero server-side data persistence.

• What if you proto-fied an existing SaaS product by offloading user content to their PDS?

  • Move: Modify one core feature of your app (e.g., user-generated notes, activity logs) to write data in a standardized JSON format (via a public lexicon) directly to users' Personal Data Servers instead of your database.

  • Why Now?: Tools like WebSockets and HTTP APIs for PDS access are stable, and interest in data ownership is rising post-Twitter exodus. Solo devs can experiment with real user demand for portability.

  • Expected Upside: Differentiate your product with true data ownership - users can export, audit, and reuse their data freely. Attracts privacy-focused early adopters and sets foundation for interoperability with future decentralized apps.

• Register a domain-based identity and configure DNS records to establish ownership for decentralized web protocols like AT Protocol.
• Design application data models using lexicons with reverse-domain naming (e.g., app.bsky.feed.post) to ensure compatibility and interoperability in decentralized systems.
• Implement a local materialized view database in your app to efficiently query follower relationships and feed data pulled from decentralized sources, accepting that it's a derived cache, not the source of truth.
• Adopt the Standard Site schema for blog posts and publications to enable cross-platform syndication and improve content discoverability in decentralized networks.
• Build or integrate tools that pull and process public JSON-formatted data from Personal Data Servers (PDS) via HTTP/WebSocket to create aggregated views like feeds or analytics dashboards.