No correlation.
No tracking.
No way to stitch interactions together.
And technically… that’s true.
With systems like $SIGN using zero-knowledge proofs and primitives like BBS+ signatures, each interaction becomes its own isolated event. Different context, different identifier, no obvious connection.
It feels like a clean break from the surveillance-heavy systems we’re used to.
But here’s the part that doesn’t get enough attention:
Unlinkability doesn’t eliminate structure — it displaces it.
When Everything Becomes Isolated
If every interaction is truly unlinkable, something unusual happens.
You lose continuity.
No persistent reputation
No accumulated trust
No shared history across contexts
Every proof stands alone. Every interaction resets the system.
That sounds privacy-preserving — but also… impractical.
Because real systems don’t just verify facts.
They rely on context over time.
So Where Does Continuity Come From?
If verifiers can’t correlate activity directly, the system still needs a way to answer questions like:
Has this credential been revoked?
Is this the same entity as before?
Can this identity build reputation?
And this is where things quietly shift.
Continuity doesn’t disappear.
It just moves behind the scenes.
The Rise of Invisible Coordination Layers
To keep systems usable, something often steps in:
Issuers maintaining identity anchors
Registries tracking revocation or status
Policy layers defining how proofs relate over time
None of these explicitly “break” unlinkability.
But together, they create a new kind of structure — one that operates outside the proofs themselves.
The Trade-Off Nobody Mentions
This creates a subtle but important tension:
If interactions are linkable → you get tracking risk
If interactions are unlinkable → you need coordination
And coordination is rarely neutral.
It introduces dependencies.
Not always obvious ones.
Not always visible ones.
But dependencies nonetheless.
Does the Problem Actually Go Away?
$SIGN solves the cryptographic side of correlation extremely well.
The math works.
The unlinkability is real.
But systems are more than math.
They still need to answer:
How does identity persist without being linkable?
And in answering that, there’s always a risk that:
Correlation isn’t eliminated — it’s just relocated.
The Real Question
So maybe the deeper question isn’t:
“Can we prevent correlation?”
But rather:
“Where does correlation reappear once we do?”
Because if it always resurfaces just in less visible layers
then the challenge isn’t removing it…
…it’s making sure it doesn’t quietly regain control.