2 days ago · 7 min read1321 words · Tech · hide · 0 comments

I recently ventured down an atproto rabbit hole, poring over an unhealthy number of RFCs and specifications. This post unwinds the stack on some of that information, attempting to lay out some of the talking points I find most interesting. Today I'm focused on DIDs, or decentralized identifiers. The atproto docs explain that the protocol primarily uses two forms of identifiers: handles and DIDs. Handles are your user-facing, friendly identifiers. Typically a domain name, e.g. mgmarlow.com. DIDs are stable, long-term identifiers that are not easy to remember or read, e.g. did:plc:bxgklbscdfspxtn5l3x4fq4g. Think of your handle as a pointer to your DID. Today you might identify as me.bsky.social, but tomorrow you might rather be me@mydomain.com. Since your handle is just a pointer to a globally unique DID, you can freely migrate handles without affecting your global identity. So long as the DID stays the same, you're in the clear. There are two ways to configure handles in atproto so…

No comments yet. Log in to reply on the Fediverse. Comments will appear here.