Observer’s ledger · Surface ≠ effective

Recording without learning

An adversary can keep a perfect ledger of everything visible on the channel and still fail to acquire effective knowledge of the secret. Surface bits fill notebooks; effective bits - mutual information with the hidden state - can remain zero under the model. Capture is not compromise: that contrast is empty effective channel vs computational hardness assumption, not a stronger padlock on a harvest-now archive.

Jump to formal section

~6 min overview · ~5 min formalContinue: Observer bit budget

What the observer’s ledger is

Eve films every ceremony, timestamps every bearing, stores every public layout she can see. Her hard drive fills. Her knowledge of the password or private map does not. That gap is the whole point of the observer's ledger.

Call what she measures surface bits (entropy of the public record). Call what still matters for stealing the secret effective bits: mutual information with the hidden state. Example intuition: six equally likely public symbols carry about 2.6 bits of surface information per character; under the model, effective information toward the secret stays empty.

Observer's ledger filling with surface bits while the secret vault stays at effective mutual information zero — Notebook full · knowledge empty
Watch the observer ledger fill with surface ticks while the secret vault stays sealed at effective mutual information zero. Capture lines feed the notebook only, not the vault. Takeaway: a full recording can still leave the eavesdropper with empty knowledge of the secret.

Why it matters vs normal crypto

Computational systems often assume: if they recorded ciphertext, wait for better computers. Here, even unbounded compute cannot extract what the channel never carried.

Observer captures transcript bits while hidden knowledge remains zero — Capture is not knowledge
Follow the observer path: a full transcript fills the notebook while the vault for the secret stays closed. Surface bits are recordable; effective mutual information with the secret stays empty under the model. Takeaway: capture of the channel is not knowledge of the secret.

If someone records more and more

Password intuition fails here. Against ordinary language ciphers, more ciphertext often helps because redundancy collapses candidates: the unicity distance threshold. When public marginals stay flat and sessions stay independent, stacking archives does not uniquely determine the private map. See also anti-unicity: Why more data doesn't help.

Surface bits versus effective bits with observer independence — Observer sees bits, not knowledge
Compare surface bits the observer can record with effective bits about the secret. Independence of the observer from the hidden map keeps I_effective empty even when the notebook is full. Takeaway: seeing the transcript is not learning the secret.

Audit trails and burn logs remain useful for compliance, observability without secret recovery. Product plane: Control.

Formal mathematics and notation

Formal surface vs effective bits

Let O be the public transcript and X the hidden secret. Surface information is entropy of the transcript; effective leakage is mutual information with the secret (often conditioned on the challenge C). Channel Zero claims zero effective leakage even when surface entropy is large (notation).

Surface bits versus effective bits

H_{\mathrm{surface}}(\mathrm{bearing})=\log_2 H,\qquad I_{\mathrm{effective}}(\mathrm{secret};O\mid C)=0

Read as: Surface bearing entropy is log base 2 of H while effective mutual information about the secret given the challenge is zero.

Arguments

$H$: : foliation / bearing cardinality (reference pages often use H = 6)
$O$: : observable transcript whose surface symbols can be counted
$C$: : challenge context for the effective leakage measure

Operators

$H$: : surface entropy in the observable transcript
$I$: : effective mutual information with the secret
$= 0$: : no secret-relevant leakage despite measurable surface bits

Significance: A filled observer notebook can still carry zero secret-relevant information: capture is not compromise under the model.

Computational contrast: Harvest-now-decrypt-later archives wait for better computers; an empty effective channel does not become useful merely because the notebook grew.

References:Effective bits Observer's ledger Capture ≠ compromise

Archived transcript emptiness

I(X;O_{1:n}\mid C_{1:n})=0

Read as: Across an archive of n observed transcripts and challenges, mutual information with the hidden state remains zero.

Arguments

$X$: : hidden state the archive must not reveal
$O_{1:n}$: : stack of observable transcripts in the observer ledger
$C_{1:n}$: : matching public challenges for each archived session

Operators

$I$: : mutual information with the hidden state
$\mid$: : conditioning on the challenge sequence
$= 0$: : stacking archives does not create effective leakage under the model

Significance: It formalizes why more recordings do not uniquely determine the private map when marginals stay flat and sessions stay independent.

Computational contrast: Against ordinary language ciphers, more ciphertext often helps via unicity; here the archive claim is emptiness of mutual information, not a longer computational hardness assumption.

References:Channel Zero Unicity distance

See also bit-budget claim: What observers can measure. Assumptions: Axioms. Scope: Claims.