Infrastructure · Primitives
ChannelCapacity
The maximum rate at which information can be reliably transmitted through a channel, given its noise
Definition
Mechanism
A communication channel (a phone line, a network link, a cell-signalling pathway) has a maximum rate at which information can pass through it reliably. Noise lowers that rate. Each message has an entropy, the minimum information needed to encode it. If that entropy exceeds the channel's capacity over a time period, the message cannot be sent reliably in that period, whatever code is used. Different codes approach capacity for different kinds of content, so efficient transmission depends on matching the code to the content. A noiseless binary on/off channel carries exactly 1 bit per use; a noisy one carries less, as the apoptosis signalling channel of human cells (measured at about 0.92 bits) does.
Invariants
- Reliable transmission is impossible if message entropy exceeds channel capacity over the available time
- Noise reduces capacity below the channel's nominal rate
- No single code is optimal for every kind of content
Risks
Failure modes
Reading every shortfall in a channel's capacity as meaningless noise, when it may reflect deliberate smoothing or computation
Treating information flow as uniform like fluid flow, ignoring that different content needs different codes
Relationships
Root pattern: builds on nothing in this library.
Used by
References
Society
Natural work overlaps: doing more of one task always means doing a little less of another, and imperfect alignment between them packs more function into the same agents
Society
Redundancy in communication is inefficient for one message or interaction but efficient for a space in which many interactions keep taking place
Bits submitted per bit accepted in a channel that computes, where rejection is meaningful work rather than noise
Mind
Model natural risk with three mathematics used together: game theory, information theory and computer theory
Reference