Information Degradation
Reads how information loses resolution, context, source, and force as it travels through distance, time, compression, and mediation.
Expansion · Knowledge · Checking Your Map Against Reality
Mechanism
Information Degradation reads what happens to information as it travels.
A thing happens. Someone sees it. They tell another person. That person writes a summary. The summary becomes a meeting note. The meeting note becomes a slide. The slide becomes a decision. By the end, everyone may be acting on "what happened," but the object in their hands is no longer the event. It is a transmission chain.
Every link in that chain can lose something. Detail disappears. Context drops away. Source proximity gets forgotten. Uncertainty becomes clean prose. A conditional claim becomes a statement. A warning becomes a caveat. A dissenting observation becomes "mixed feedback." No one has to lie. The information becomes easier to carry by becoming less like the territory it came from.
Information degrades when each transmission makes it more portable and less faithful.This is not the same failure as Signal vs Noise. Signal vs Noise asks which observations should affect the map. Information Degradation asks what happens to an observation after it enters the channel. The original signal may have been good. The later version may no longer carry what made it signal in the first place.
The most common degradation forms are predictable.
Resolution loss. Specific observations become general claims.
Context loss. The conditions that made the observation meaningful fall away.
Source loss. The reader no longer knows who saw what, how close they were, or what incentives shaped the account.
Uncertainty loss. Confidence markers disappear as summaries travel.
Compression loss. The information becomes shorter and easier to use, but the missing pieces are exactly what the later decision needed.
Force loss. A warning keeps its words while losing its practical weight.
Control degrades information by managing it. It selects, compresses, sanitizes, and routes information until the official story remains intact. Decay degrades information by letting the channel dissolve: no source trail, no confidence markers, no version history, no distinction between direct evidence and retelling. The Range form preserves enough chain-of-custody that later action can still know what the information is, where it came from, what changed, and what it can support.
Practice
The diagnostic question is: "How far is this information from the thing it claims to represent, and what was lost on the way?"
Use this before relying on summaries, briefings, secondhand claims, dashboards, meeting notes, incident accounts, media reports, model answers, or any conclusion built from material you did not inspect directly.
Return to the nearest source. If the decision is load-bearing, do not stop at the summary. Find the field note, transcript, original data, incident log, primary document, or person closest to the event.
Preserve metadata. Keep source, date, method, proximity, confidence, uncertainty, and known pressure attached to the claim. A claim without metadata is already partly degraded.
Mark each transformation. Observation to note. Note to summary. Summary to interpretation. Interpretation to decision. Name where the information changed form.
Avoid summary-of-summary reasoning. If a claim has passed through several summaries, treat it as a degraded object until the chain is inspected.
Compare independent chains. Degradation often has a direction. Independent paths that preserve the same detail deserve more attention than one path repeated loudly.
The discipline is not source purism. You cannot inspect primary material for everything. The discipline is proportional: the heavier the decision, the closer to source the information should be, and the more visible the transformation chain needs to become.
In the Wild
An incident review begins with logs, support tickets, and interviews. A month later the executive summary says the root cause was "process adherence." The original material showed a broken handoff, unclear ownership, and repeated warnings. The summary has not necessarily fabricated anything. It has degraded the information into a cleaner object the organization already knows how to act on.
A public controversy moves through social media. The first account contains uncertainty, sequence, and context. The tenth retelling has a villain, a lesson, and a screenshot. The story may still contain a true core. Information Degradation asks what survived transmission and what was selected out because it made the story less portable.
An AI system answers from a web summary of a report, not the report itself. The answer may be fluent and mostly right, but it is now two transformations away from the source: report to summary, summary to generated answer. If the answer is being used for a serious decision, the next move is not to polish the answer. It is to inspect the source chain.
When a claim reaches you after travel, treat it as a thing with a history. Ask where it began, who handled it, what was compressed, what was made cleaner, and what uncertainty fell off before it arrived.
You do not need original sources for every small claim. You do need source discipline for the claims that will carry decisions. The map is only as good as the channel that brings reality into it.
Lineage
Information Degradation is not one inherited doctrine with one canonical source. It is a Codex translation of several lineages that study what happens when information moves.
Shannon's communication theory supplies the technical background for messages, channels, coding, and noise. The Codex uses that inheritance carefully: Shannon's model explains technical transmission, not the full meaning of human or institutional communication.
Frederic Bartlett's work on reconstructive memory and serial reproduction gives the memory lineage. Remembering is not simple playback. People reconstruct, select, normalize, and reshape material as they recall and retell it. Allport and Postman's rumor work adds the classic pattern of transmitted stories becoming shorter, sharper, and assimilated to what the reteller expects.
Organizational communication research adds the institutional pressure layer. Upward information can be softened, filtered, delayed, or distorted because of hierarchy, trust, status, and fear of consequence. That territory overlaps with Report Fidelity, but Information Degradation remains the broader transmission tool.
Source criticism, intelligence analysis, audit trails, archival practice, and journalistic verification add the working discipline: get closer to source, preserve provenance, mark transformations, and keep uncertainty attached to the claim.
The tool has limits. Not every summary is a failure. Summaries make action possible. Degradation becomes dangerous when the reader forgets that the summary is a transformed object, or when a decision demands more fidelity than the channel preserved.
Cross-references
Within the category. Signal vs Noise asks which observations should move the map. Information Degradation asks what happens after an observation begins to travel. Report Fidelity asks whether a report still warrants the interpretation and use attached to it. Rectification of Names asks whether the names that travel with the information still fit the reality.
Across the Workshop. Chilling Effects and future Bond tools such as Preference Falsification explain why some signal may never enter the chain. Rules-in-Use asks what rule actually governs the channel. Legibility explains why institutions compress reality into forms they can see.
Limitation. This tool does not say "never summarize." It says summaries are transformed objects. The heavier the use, the more the transformation chain has to be visible.