About
Straight answers to the questions people ask first: how trust works without a CA, what happens if you lose a key or the network is down, how it compares to Sigstore, PGP and SPIFFE, and how far along it is.
About › FAQ
The protocol. This page describes the BetterSign protocol and its open-source core. For what the BetterSign app and products do today, see Products.
No. There is no global ledger, no consensus, no tokens and no mining. Each identity has its own small, signed, append-only log that only its followers care about. Trust comes from replaying that one log, not from a network-wide chain.
No. Point your tools at a VLAD and BetterSign keeps the keys current for you. The cryptography stays under the hood. Day to day you deal with a stable name and the occasional signed update.
Verifiable Long-lived Address. It’s a permanent, self-certifying name: anyone can verify it, it does not change over time, and it points at you.
The math, not a company. Anyone can replay a VLAD’s provenance log and check every change for themselves. The network (VLADemlia) only helps peers find each other. It never decides what is true.
Records are content-addressed and verified locally, so a lying or malicious peer cannot forge anything. Bad data simply fails verification. And if the network is unreachable, you can still verify any log you already hold. Discovery needs the network, trust does not.
BetterSign has a built-in recovery authority: a stronger, better-protected recovery key that outranks your everyday key and can authorize a rotation if the normal key is lost or compromised. It is set up when the identity is created.
Yes. Post-quantum signatures and key encapsulation are first-class, so identities and messages created today still verify and stay private after quantum computers arrive.
Honestly: it is an early, actively developed system, not a turnkey product yet. The cryptographic core and provenance-log verification are implemented and tested, and integrations for SSH, TLS/SVID, WireGuard and git signing work today. Evaluate it against your own risk tolerance. Because it interoperates with existing standards, you can pilot it on a single surface before committing.
No. BetterSign speaks X.509, SPIFFE/SVID, OpenSSH and PGP, and ships a drop-in gpg replacement and a zero-trust mTLS proxy, so you can slide it under existing tools and adopt it one surface at a time.
They pull; you do not push. You publish the change once, and anyone following your VLAD looks it up, verifies it themselves, and switches to the new key automatically. Nothing is hand-delivered to individual machines.
Sigstore centers on keyless signing with short-lived certificates and a transparency log tied to an identity provider (OIDC). BetterSign gives you a long-lived, self-owned identity whose keys you rotate freely, verified by replaying its own log. It does not depend on an external identity provider being online.
PGP ties your identity to a specific key, so rotating keys makes you look like a stranger, and its web of trust is famously hard to use. BetterSign separates identity from keys and replaces social-graph trust with a signed log anyone can verify. The comparison page lays it out side by side.
SPIFFE and SPIRE issue workload identities from a central server and its database. BetterSign issues the same kinds of identities but roots trust in a provenance log instead of a server, so rotating a key, or even re-keying the authority, is a routine signed update. See the SPIFFE demo.
| Trust | Math you can check, not a CA |
| If keys are lost | Built-in recovery authority |
| Maturity | Early, actively developed; core implemented |