Where does information collide?

The amount by which a message reduces uncertainty about which of several possible outcomes has occurred, measured in bits and computed from the probabilities of those outcomes. Shannon's 1948 A Mathematical Theory of Communication defined this quantity through entropy, the average surprise of a source, and deliberately set aside what a message means: two messages that are equally improbable carry the same number of bits whether one is profound and the other is gibberish. The referent is a statistical property of a source and channel, not the truthful content a philosopher weighs, the trade secret an economist prices, or the coded sequence a biologist reads in DNA.[1]

Unit

One bit is the uncertainty removed by learning the answer to one equally likely yes or no question

Recovery means

Not applicable

Measured by

Computed as entropy from the probability distribution of a source, in bits per symbol or bits per second

Knowledge about the quality, price, or intentions behind a transaction that is costly to obtain and often held unevenly by the parties to a trade. Stigler's 1961 The Economics of Information treated search for a better price as a costly activity worth modelling, and Akerlof's 1970 The Market for "Lemons" showed that when a seller knows more than a buyer, this asymmetry can drive good goods out of the market. Here information is a scarce economic good whose distribution shapes prices and behaviour, unlike the meaning-free bit count of information theory, the truthful content of philosophy, or the chemical instructions in a gene.[2][3]

Unit

A parcel of decision-relevant knowledge, valued by what it saves or earns and by who holds it

Recovery means

Not applicable

Measured by

Inferred from search costs, price dispersion, and the market outcomes that asymmetry produces, such as adverse selection

The precise order of bases along a strand of DNA or RNA that specifies the order of amino acids in a protein. Crick's 1970 Central Dogma of Molecular Biology framed this as directional: sequence information flows from nucleic acid to protein and cannot flow back out of protein into nucleic acid. The referent is a physical, chemical ordering that a cell copies and translates, not an abstract bit count, a scarce market good, or a truthful proposition, even though a biologist may borrow the language of coding and readout to describe it.[4]

Unit

A base sequence, read in three-letter codons that each specify one amino acid

Recovery means

Copied by replication and expressed by transcription and translation; per the central dogma it does not flow back from protein to sequence

Measured by

Determined by DNA and RNA sequencing that reads the order of bases directly

Content that is well-formed, meaningful, and true, so that a false statement does not count as information at all but as misinformation. Floridi's 2004 Outline of a Theory of Strongly Semantic Information argued that Shannon's account measures data without regard to meaning or truth, and proposed a stronger notion in which truthfulness is a requirement rather than an optional extra. Here information is a property of meaningful, accurate content, in direct contrast with the meaning-free bit count of information theory, the priced knowledge of economics, and the chemical sequence of biology.[5]

Unit

A well-formed, meaningful item of content that additionally must be true to qualify

Recovery means

Not applicable

Measured by

Assessed by whether content is well-formed, carries meaning, and is true, not by a numerical scale