Comparison Chart¶
Artifact — instantiates Strategic Juxtaposition
Encodes selected dimensions across items in a compact grid; useful when relation inspection requires structured rows, columns, or criteria.
A Comparison Chart encodes several items against a common set of dimensions in a compact grid — items on one axis, criteria on the other, each cell holding the item's value on that criterion. Its defining move is abstraction: it discards the items as whole objects and keeps only their attributes on the chosen dimensions, so that many items and many criteria collapse into a single scannable structure. Where a display keeps the items themselves adjacent, the chart keeps only an encoded relation; its power and its danger both come from that compression.
Example¶
A team choosing an issue-tracking tool for a 200-person organization has demos, sales decks, and opinions scattered across a dozen threads. To make the decision inspectable, they build a comparison chart: five candidate tools down the rows; the criteria that actually matter across the columns — SSO support, API completeness, per-seat cost, on-prem option, migration effort. Each cell gets a normalized entry (yes/no, a score, a dollar figure), and the criteria that carry the decision are weighted.
Rendered as a grid — a weighted decision matrix in the Pugh tradition[n1] — the choice that felt like a tie in conversation now shows a clear leader, and, just as usefully, shows which criterion is deciding it. Cell footnotes flag the two scores the team is least sure of, so a reader can see where the grid is soft rather than trusting the total blindly.
How it works¶
The chart is built by selecting dimensions (the columns — the criteria worth comparing on), fixing the item set (the rows), and choosing a cell encoding: raw value, normalized score, rank, or symbol. Weights may be applied to reflect that criteria matter unequally, and an annotation layer carries flags, footnotes, and confidence marks on individual cells. The essential move — the one that separates it from arranging the real artifacts — is that everything not captured by the chosen dimensions is thrown away.
Tuning parameters¶
- Dimension selection — which criteria become columns. This is the highest-leverage and most manipulable dial: the chart's verdict is largely set by what it chooses to measure.
- Cell encoding — raw figures, normalized scores, ranks, or symbols. Scores enable summation but invite false precision; symbols resist precision but can't be totaled.
- Weighting — whether and how criteria are weighted. Weights make priorities explicit and also hide them inside a single number.
- Density — how many rows and columns. Beyond a point the grid stops being scannable and becomes a spreadsheet.
When it helps, and when it misleads¶
Its strength is compressing an item-by-criterion space too large to hold in mind into one structure the eye can scan, making trade-offs across many options explicit at once.
Its failure modes follow from the abstraction. Dimension-selection bias lets the author decide the outcome by choosing which criteria appear. False precision creeps in when soft judgments are scored to two decimals and summed as if commensurable. And the grid inherits the general map-is-not-the-territory problem: the encoded attributes are not the tools, and a criterion nobody put in a column is invisible to the decision. The guard is to expose the criteria and weights for challenge and to mark low-confidence cells rather than letting a tidy total imply certainty the inputs don't support. A practical discipline is to run the chart twice under different weightings: if the winner flips, the decision is being made by the weights rather than by the evidence, and that should be visible to everyone reading the grid.
How it implements the components¶
contrastive_display— the grid is the materialized display that makes the item-by-dimension relation perceptible and discussable.comparison_set— the items occupying the rows.annotation_layer— the weights, cell flags, and confidence footnotes carried on the grid.
It does not keep the real artifacts adjacent under a curator's hand — adjacency_structure and interpretive_owner are Exhibit Curation's; the Comparison Chart abstracts each item into encoded cells and needs no curator, where the exhibit preserves the objects themselves.
Related¶
- Instantiates: Strategic Juxtaposition — realizes the structured, multi-criterion display: many items and dimensions in one grid.
- Sibling mechanisms: Before/After View · Side-by-Side Comparison Display · Benchmark Comparison · Exhibit Curation · Juxtaposed Narratives · Paired Case Comparison · Paired Example and Non-Example · Red-Team/Blue-Team Comparison
Editorial Notes¶
Form Classification¶
Form family: Representation, Specification & Plan
Rationale: Encodes selected dimensions across items in a compact grid; useful when relation inspection requires structured rows, columns, or criteria, making its operative form a non-executable information artifact that externalizes static or prospective structure.
Independent corroboration: The frozen evidence defines Comparison Chart as 'Encodes selected dimensions across items in a compact grid; useful when relation inspection requires structured rows, columns, or criteria', 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 design selection cohered Pugh and weighted decision matrices that place alternatives against common criteria in a compact grid.
Related originating lineages:
- Operations Research — Multi-criteria decision analysis supplies weighting, normalization, and aggregation across attributes.
Review resolution: The source is an attribute-by-alternative design comparison grid, recognizable in engineering selection practice and the Pugh tradition. Operations research contributes optional weighting and normalization, but the basic chart cohered within design practice and therefore remains single-lineage.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] The Pugh matrix (Stuart Pugh) and the broader weighted decision matrix score options against weighted criteria to make a multi-attribute choice explicit and challengeable. The method's value is the visible structure; its risk is treating the resulting total as more objective than its inputs. ↩