There's a detail in how Sign's wallet works that i keep thinking about. when someone needs to present a credential — a license, an ID, a diploma — they don't need an internet connection to do it. the wallet supports offline presentation through QR codes or NFC. the whitepaper specifically says this is critical for rural and underserved populations, and that line is doing a lot of work.
But the offline presentation piece only tells half the story. the other half is what happens when the verifier receives that credential. because the obvious question is — how does the verifier know its real? how do they confirm the institution that issued it actually issued it?
Normally that requires contacting someone. a call, an API, some kind of live check against the issuing authority's systems. which means both sides need connectivity, and the issuing authority needs to be reachable, and none of that is guaranteed everywhere.
The trust registry changes this. government agencies and institutions register their decentralized identifiers and public keys on-chain. standardized credential schemas are published on-chain. and revocation lists — for expired or fraudulent credentials — are maintained on-chain in real time. so when a verifier checks a credential, everything they need is already there. they're not calling the university. they're not pinging a government server. the verification happens against what's already on the blockchain, and it happens whether or not anyone is reachable in that moment.
What this actually means is that the two pieces — offline presentation and on-chain trust registry — are built to work together in exactly the situations where other systems would fail. no signal on the presentation side. no server dependency on the verification side. the credential still gets checked, correctly, completely.
The whitepaper frames this as infrastructure for rural and underserved populations. from what i understand, that's not just a use case — its the design constraint the whole thing is built around.
So the question worth asking is — how many identity systems right now actually work when the infrastructure they assume isn't there?
