Comparison Table¶
Template — instantiates Cognitive Representation Externalization
A table that externalizes categories, criteria, contrasts, or alternatives.
A Comparison Table lays already-named alternatives down the rows and the criteria that distinguish them across the columns, so that the single relation the mind was juggling — how do these options differ, criterion by criterion? — becomes a grid you can read at a glance. Its defining move is that it encodes a contrast relation over a fixed axis of comparison: every option is scored against the same criteria, which forces the implicit weighing that happens silently in someone's head into explicit, cell-by-cell view. It is not an elicitation tool — the options and criteria must already be named before the table can exist — and it is not a causal or sequential model. It is the mechanism for the moment when the pieces are known but their relative standing is a fog, and the fog clears the instant the empty cells demand to be filled.
Example¶
A hardware team must choose a battery chemistry for a new outdoor sensor and keeps re-litigating it in circles because each engineer is silently optimizing a different thing. They build a comparison table in the spirit of a Pugh matrix:[1] rows are lithium-ion, LiFePO4, and nickel-metal-hydride; columns are cold-weather performance, cycle life, unit cost, and supply-chain risk. Filling the grid, they discover the argument was never really about chemistry — it was about cold-weather performance versus cost, the only two columns where the options actually diverge. The other columns are near-ties and can be dropped from the debate. The table's real output is not the winning row but the visible structure of the trade-off: it shows which criteria decide this, and the footnoted caveat under it records the one criterion they deliberately left out (recyclability, which no supplier could yet quantify) so no reader mistakes the grid for the whole story.
How it works¶
- Fix the axis of comparison. Choose the criteria (columns) that actually distinguish the alternatives; a criterion every option shares carries no information and wastes a column.
- Populate every cell against the same criterion. The discipline that makes the table honest is applying each column's standard identically across rows, so the contrast is real rather than rhetorical.
- Read the columns that diverge. Columns where options cluster are non-deciding; columns where they spread are where the real choice lives.
- State what the table omits. Because a grid looks exhaustive, the criteria and considerations deliberately left out are noted alongside it, not silently dropped.
Tuning parameters¶
- Criteria granularity — a few coarse columns versus many fine ones. Finer criteria expose subtle differences but dilute the deciding contrast in noise; coarser ones stay legible but can hide a decisive sub-factor.
- Cell resolution — checkmarks and X's, ordinal ratings, or exact measured values. Richer cells carry more but invite false precision where the underlying data is soft.
- Weighting — whether columns are treated as equal or explicitly weighted. Weights make the trade-off's priorities visible but can smuggle the author's preference into an apparently neutral grid.
- Baseline framing — scoring options absolutely versus relative to a reference row (as a Pugh matrix does). A reference row sharpens "better/worse than" but hides absolute adequacy.
When it helps, and when it misleads¶
Its strength is collapsing a tangled multi-way comparison into a surface where the deciding criteria become obvious — it is the fastest way to see which dimension the choice actually turns on and to stop an argument that was really about mismatched criteria. It excels when the alternatives are enumerable and the criteria nameable.
Its failure mode is the tyranny of the tabulated: only what fits a cell gets compared, so hard-to-score but decisive factors (a partner's trustworthiness, a design's elegance) are quietly excluded, and a filled grid radiates a false authority that a messy judgment lacks. The classic misuse is reverse-engineering the columns and weights until the pre-chosen option wins. The guarding discipline is to record the omitted criteria in a boundary note, to keep weights explicit and challengeable, and to treat the grid as a structured argument, not a verdict.
How it implements the components¶
relation_encoding— the grid encodes one specific relation, contrast across a shared criteria axis, as the intersection of every option row with every criterion column.granularity_setting— the choice of how many criteria and how finely each cell is scored sets the level of detail at which the comparison is made inspectable.representation_boundary_note— the caveat recording which criteria and considerations were left out marks what the tidy grid does not cover.
It cannot arrange options it has not been given — it does not elicit an unformed model from someone's head (implicit_model, cognitive_task_context) — that is its template twin Canvas Template; and it does not test the grid's meaning through inspection (shared_review, mismatch_capture), which is Model Walkthrough Review.
Related¶
- Instantiates: Cognitive Representation Externalization — the Comparison Table externalizes the contrast structure of a choice so it can be inspected and argued over.
- Sibling mechanisms: Canvas Template · External Memory Aid · Model Walkthrough Review · Process Flowchart · Schema Draft · Whiteboard Modeling Session · Concept Map · Causal Loop Diagram
Editorial Notes¶
Form Classification¶
Form family: Representation, Specification & Plan
Rationale: A table that externalizes categories, criteria, contrasts, or alternatives, making its operative form a non-executable information artifact that externalizes static or prospective structure.
Independent corroboration: The frozen evidence defines Comparison Table as 'A table that externalizes categories, criteria, contrasts, or alternatives', so its operative form is Representation, Specification & Plan.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Engineering & Design
Origin pattern: Single lineage
Present-day reach: Multi-domain
Rationale: Engineering selection practice cohered the Pugh-style comparison table as an explicit criterion-by-alternative representation.
Related originating lineages:
- Operations Research — Multi-attribute decision analysis contributes weights, common scales, and transparent trade-off structure.
Review resolution: The fixed criterion-by-alternative grid descends directly from engineering comparison and Pugh matrices. Multi-criteria decision analysis informs systematic scoring and weighting, but the source template does not require those additions, so a single engineering-design lineage is the better origin-mode calibration.
Review outcome: Reconciled after independent review; high confidence.
References¶
[1] The Pugh matrix (Stuart Pugh, Total Design, 1991) scores design alternatives against weighted criteria relative to a reference concept — the canonical named form of a comparison table used to make a multi-criteria trade-off explicit and reviewable. registry ↩