# Don't Tell Me What I Already Know
# Book: Language
# Chapter URL: https://programmer.ie/books/language/16-chapter/
# Prose fingerprint (sha256, normalized): a645cccfe630bc016e8ac1666c311e83ed58c4b8f45da89f31553f2be588763f
# Snapshot generated at build time by Hugo. This is the complete rendered prose
# of the chapter; it is not a transcript of the source file.

Iris asks about the constructed consensus-protocol talk, and the system has some evidence of her expertise. She has worked on consensus systems for two years and given talks on the subject. It therefore removes the speaker’s definition of partial synchrony from her digest: surely she knows it. But the speaker is using a particular timing assumption for the demonstration: a delay bound available to the protocol, rather than merely a bound that exists but is not known in advance. Iris knows the distinction. What she does not know is which reading this speaker chose. She sees the later performance claim without its local definition and thinks the talk is making a broader guarantee than it established. The system has omitted what she needed in order to interpret the evidence.
This is a constructed failure, not a measured one. The classic literature describes more than one partial-synchrony model; a known bound is not a universal definition of the term. Nor does the timing condition, without the protocol’s proof, tell us whether safety fails when delays exceed it. The narrower and more defensible lesson is enough: expertise in a subject does not mean familiarity with a particular source’s assumptions.
The term was familiar; its intended use in this source was not yet established for the reader. The previous chapter fixed the task and showed why different tasks need different evidence. Hold both source and task constant now. Two people can still need different amounts of explanation because their prior knowledge differs. This chapter asks a narrower question:
Given the same source and the same task, what would actually add something for this particular person?
It is a narrow question, deliberately. The proposal is not to build a portrait of Iris. It is to make one testable decision about what can be shortened without making the remaining material harder to understand or less faithful.
A model of what you know, and nothing else Answering it may benefit from a model of the reader’s knowledge for this task. That is a significant step beyond adapting a view to the current question, so the proposed model should be as small as the decision permits. Three different kinds of model are easily confused:
MODEL OF WHAT YOU KNOW ≠ MODEL OF WHAT YOU LIKE ≠ MODEL OF WHO YOU ARE This chapter needs only the first, and even that only in a limited form. The system need not infer personality, identity, demographic traits, politics, preferences or learning style. It needs evidence about whether this person can use a particular concept, in the sense and context needed by the current source. Being experienced in consensus research does not answer whether Iris has seen this speaker’s definition; disliking diagrams answers neither question. The model should say what evidence exists for a specific capability, not generalise from expertise to a kind of person.
The transformation here is:
same source same task different prior knowledge ↓ different useful representation Call this adaptation to prior knowledge rather than broad personalisation. The proposed change is local: expand, collapse or shorten material on the basis of a narrow knowledge estimate, while retaining whatever the selected claim requires. It is not an instruction to optimise engagement, build a permanent profile or infer new interests. Even a narrow knowledge estimate is information about a person, however, and must not be retained or shared merely because it is useful to the renderer.
Novelty is relational too The previous chapter defined relevance in relation to a unit and a declared purpose. Novelty requires a further relation: the unit’s specific claim or sense, the reader’s evidenced knowledge and the current context. It is not a free-standing score stored on a document.
ESTIMATED NOVELTY(unit, evidenced knowledge, sense, context) Consider the talk’s definition of a fault as a crash failure. Iris may already know that category of fault and still need to know that this talk excludes other fault models. Tom may be familiar with the historical terminology but need the local definition to compare papers. Mina may need both an explanation and an example. To isolate prior knowledge, imagine all three trying to interpret the same claim under the same declared purpose; their needs may still differ, but none can safely lose the clause that limits that claim. A familiar concept and a source-specific restriction are not interchangeable.
new in the source ≠ needed for the task ≠ new to the person A unit may introduce nothing new to an expert yet still be necessary to interpret the next claim. Conversely, a novel aside may be irrelevant to the declared task. Novelty does not override relevance, and neither overrides a preservation obligation. The variables are distinct because a single model-generated importance score could conceal which one justified an omission.
What the system may say it knows The system does not directly observe knowledge. It observes evidence: answers, demonstrations, corrections, self-reports and interactions, collected within whatever scope the person authorised. Each observation supports a limited inference. The relevant question is not “does Iris know consensus?” but “what evidence suggests she can interpret this concept in the sense the present source requires?” That keeps the model tied to a decision it can be audited for.
evidence about knowledge ≠ knowledge itself A binary known/not-known flag hides too much uncertainty. For a simple prototype, use three operational estimates, not diagnoses of a person’s mind:
KNOWN sufficient recent evidence for this concept and sense, under a stated policy PARTIAL evidence of some understanding, or evidence that does not cover this use UNKNOWN insufficient or conflicting evidence to justify a knowledge-based omission The grades are not a calibrated probability and should not be presented as one. Their meaning depends on a declared threshold and the kind of evidence available. A recent mistake can conflict with an earlier demonstration, and an apparent success can depend on hints. The safest operational result for omission is then UNKNOWN, accompanied by the conflict rather than silently replacing the older observation. A changed source definition is a scope mismatch, not a fall in the person’s expertise.
Several smaller separations protect the model from pretending to more than it has:
not observed ≠ not known mentioned before ≠ mastered familiar term ≠ known meaning The first guards against inferring ignorance from missing records: Iris’s experience may be invisible to the system. The second guards against treating exposure as mastery. The third is the opening failure, and there is a fourth boundary: knowing a standard term is not knowing the particular convention a document assigns to it. A useful knowledge record therefore identifies the sense and source context as well as the concept label.
What evidence would justify even a narrow estimate? A person’s explicit statement that they know a concept is strong evidence about how they want to be treated; it is not by itself a test of mastery. A correct explanation or successful application is more direct evidence of capability for that task, though it may be assisted or context-specific. A corrected error can reveal both understanding and a prior gap. A question asked earlier may show curiosity, not ignorance. Repeated exposure, clicks, watched videos or the mere appearance of a term in conversation show little about retention. There is no universal ordering independent of the construct being assessed, the conditions of the observation and the person’s consent.
Age matters too, but a fixed decay rate for all topics is a policy, not a psychological finding. The model should retain observation type, scope, source and date so the system can say why it believes something, and should prefer uncertainty when an old observation is the only one available.
The minimum useful record is already more than a concept and a grade. It might contain the concept, its intended sense, the task or source scope, evidence type, evidence location if retention is authorised, observation date, uncertainty or conflict, and the person’s corrections. A grade without the evidence and sense that produced it becomes a disguised dossier entry. A record should remain inspectable and reversible: the system believes you can explain quorum in this setting, based on this dated answer; you can correct that estimate or decline to use it.
This is still personal information, and data minimisation is an obligation now, not something the book may defer merely because its broader privacy discussion comes later. A session-only estimate is a reasonable default for the experiment. Any cross-session persistence, access by another system or transfer between projects needs an explicit purpose and control. Deleting a knowledge estimate should not require the reader to prove the estimate was wrong. Later chapters examine the broader personal mediator; this chapter does not assume one already exists.
Known is not safe to omit The opening failure supplies a rule: never omit a source-specific definition, exception, assumption or prerequisite merely because the reader appears to know the underlying topic. Familiarity may justify shortening an explanation of the general concept; it does not establish that the reader knows what this author chose to mean by it.
Consider three units in a chain, where each depends on the one before:
Constructed dependency example. Knowledge evidence cannot waive a source-preservation obligation. Open full-size diagram. Suppose the system has evidence that the reader understands A and B in their usual senses. But C depends on B as defined in this source, and that version of B has a narrower condition. Removing B because its label is familiar breaks the dependency needed to interpret C. The missing information is not remedied by the fact that the reader once answered a general question about B correctly.
Omission therefore needs more than a knowledge estimate. It needs a source-specific dependency check, the reader’s permitted degree of compression, and a decision about whether the claim remains recoverable if the explanation is hidden. A useful conservative rule is:
evidence supports this sense as known AND omission does not remove an obligation or necessary dependency AND the reader permits this degree of compression ↓ CANDIDATE FOR OMISSION (not automatically safe) Presentation should be more graduated than omission or display. Evidence of strong knowledge may justify a shorter reminder when no required dependency is at risk. Partial or stale evidence favours a collapsed explanation, with its existence and provenance visible. Unknown or conflicting evidence favours keeping the explanation accessible without assuming ignorance. A source-specific qualifier needed for the displayed conclusion belongs with that conclusion, not behind an undisclosed click. The decision has to pass the preservation obligations from Part I after the novelty policy has acted.
known ≠ safe to omit There are two different errors, and their costs depend on the task. Showing a definition someone knows can waste attention and sometimes obscure the decisive finding. Hiding a prerequisite or qualification can silently change the apparent claim. The latter is especially costly when the reader is making a consequential decision, but there is no universal numerical exchange rate between the two. A system should report unnecessary repetition and harmful omissions separately, with higher protection for critical dependencies. When evidence is weak, shortening or leaving material available is safer than silently deleting it.
What this is not A few boundaries keep the narrow model narrow, and distinguish observing behaviour from inferring a reader’s understanding.
I know this ≠ I like this I have seen this ≠ I know this I often open this ≠ I need more of this I rarely open this ≠ I already understand it Interest and knowledge are separate variables, and neither can be read directly from clicks. A novice may open five articles without understanding them; an expert may ignore a familiar field because the current task lies elsewhere. An engagement recommender optimises for behaviour such as opening or watching, not for whether a reader can reconstruct a particular claim. That makes click history a particularly weak basis for removing definitions. The proposed renderer should not import a recommendation profile and relabel it as knowledge.
There is relevant research, but it supports a limited inference. Kalyuga, Ayres, Chandler and Sweller’s review of the expertise reversal effect (2003) reports that instructional support helpful to novices can be less useful or even detrimental to experts under the conditions studied. That establishes a reason to test adapting explanation to demonstrated expertise, not that every expert should receive less information or that a model can identify what this particular reader knows. Knowledge-tracing research, including Corbett and colleagues’ work on student models (1994), likewise shows how a system can maintain estimates from task observations under stated modelling assumptions. An estimate that predicts responses is still not direct observation of knowledge, and neither literature validates the chapter’s proposed omission policy.
How we would test a knowledge-aware digest The experiment should hold the source, declared task and available attention budget constant while varying knowledge evidence and the presentation policy. Prepare a set of constructed and real, permissioned technical sources with independently checked definitions, dependencies and qualifications. Include traps like the consensus talk: a familiar term used in a different sense, a prerequisite far from the claim, a source revision that changes an assumption, and a novel passage that is not relevant to the task. Reviewers should identify preservation obligations and acceptable answers without seeing the system’s outputs. Disagreements become part of the evaluation record rather than being flattened into a single authoritative answer.
Compare at least four conditions: an unadapted but well-designed digest; a simple reader-controlled expand/collapse view; an adaptation based on broad expertise labels or self-report; and the proposed evidence-scoped, sense-aware policy with a dependency gate. Keep model, source access and generation budget comparable; record whether the knowledge observations were authorised, current and genuinely available at decision time. Test across readers with different demonstrated knowledge, and include an option to decline adaptation. Do not manufacture a knowledge record from their browsing history after the fact.
The primary safety measures are harmful omissions, missing source-specific qualifications, wrong answers to task questions and false confidence. Measure unnecessary repetition, task completion time, effort and reader preference separately. Check whether readers can recover a hidden prerequisite and whether corrections to the knowledge record change subsequent decisions as intended. Evaluate model-state accuracy and calibration only against an independently collected, appropriately scoped set of tasks; do not call a grade “known” merely because the system assigned it. Report results by task and expertise, including cases where a simpler collapse control matches or beats the knowledge model. This experiment is proposed; the reference example below does not run it.
Running it Running it. The existing program holds dated, hand-authored evidence about two readers. It treats a knowledge fact as scoped to a concept and a sense, rather than a global claim about a person’s expertise. The grade—known, partial or unknown—is computed from the supplied evidence and a declared age policy. The absence of an observation yields UNKNOWN rather than a judgement of ignorance. Being shown material is not evidence of mastering it, and a later mistake can conflict with an earlier demonstration.
In the constructed program, Iris’s record treats one sense of partial synchrony as known: a delay bound exists without its value being known in advance. The speaker’s local assumption is stronger, so the definition is kept. This shows the desired sense-mismatch rule, not the correctness of a model’s reading of technical literature. Her quorum definition is marked known and has no declared dependency in the selected output, so it may be omitted by that constructed policy. Mina merely opened a page mentioning quorum, so her definition stays. Tests also retain a known prerequisite needed by a displayed claim and collapse a partially known explanation rather than delete it. The person can inspect the record, correct it and ask the example to forget it. Those are implemented behaviours of the example, not evidence of how readers use them.
What it does not show: that the evidence accurately measures anyone’s knowledge, that the sense labels have been extracted reliably, or that the omission rule improves comprehension. The knowledge state is supplied by hand, not learned from real readers; the example has no durable cross-session store, and its decay periods are declared policy rather than empirical estimates. Its dependency gate only protects the dependencies it was given. It does not test privacy controls, reader outcomes or the proposed comparison above. The broader questions of storage and access come later, but no such capability may be inferred from this scripted example.
Where this leaves us Personal novelty depends on a unit’s claim or sense, the reader’s evidenced knowledge and the context. That makes it a useful hypothesis about what may add value, not permission to erase what looks familiar. The knowledge estimate is neither direct observation nor a proxy for identity or interest. It can justify a shorter explanation when the evidence warrants it, but source-specific meanings, qualifications and necessary dependencies must survive, and the reader must retain control over whether adaptation is used. The system may repeat something an expert knows; the more serious failure is to hide the very condition needed to understand the rest.
Even a correct knowledge estimate is not enough to decide how information should be presented. Two people with the same knowledge may need different forms because one is scanning between meetings and the other is studying carefully, or because one is preparing to act and the other to remember. Those are features of the present task and circumstances, not proof of a stable personal identity. The book returns to situational adaptation later, when it can also address what the system is permitted to infer. Before that, Part II must confront a simpler change: so far someone has asked the question. A person’s stated interests can persist between questions, and the next useful thing may be something they have not yet thought to request.
References Dwork, C., Lynch, N. & Stockmeyer, L. (1988). Consensus in the presence of partial synchrony. Journal of the ACM, 35(2), 288–323. DOI 10.1145/42282.42283. (Distinguishes alternative timing assumptions; not evidence of the fictional talk’s actual guarantees.) Kalyuga, S., Ayres, P., Chandler, P. & Sweller, J. (2003). The expertise reversal effect. Educational Psychologist, 38(1), 23–31. DOI 10.1207/S15326985EP3801_4. (Background on expertise-conditioned instruction, not this system’s performance.) Corbett, A.T., Anderson, J.R., Carver, V.H. & Brancolini, S.A. (1994). Individual differences and predictive validity in student modeling. In A. Ram & K. Eiselt (eds.), Proceedings of the Sixteenth Annual Conference of the Cognitive Science Society, pp. 213–218. (Knowledge-tracing precedent; not a validation of this chapter’s policy.)