# Abstraction Layers Convert Into Coordination Cost ## Claim When a tool invents its own vocabulary for a construct that already has a standard name, it transfers the translation burden onto the humans at the handoff boundary, and that cost recurs on every exchange. <!-- atomicity: 1 idea ✓ --> ## Scope & conditions - Holds where both disciplines must reason about the *same* underlying construct — layout, spacing, type scale. - Does not hold where abstraction genuinely reduces complexity for one side without the other needing the detail (a designer choosing "elevation 2" needn't know the box-shadow values, because nobody hands elevation back across the boundary). - The test is directionality: if the abstraction has to be reversed by someone downstream, it's a cost; if it terminates, it's a genuine simplification. ## Evidence Penpot's developer-facing product tour argues that most design tools invent a proprietary layout model and translate it to CSS on export, producing a lossy "the design doesn't match the build" tax. Penpot's counter-move is to use real CSS terminology as its native model — flex is flexbox, grid is grid, `1fr` means `1fr` — so the inspect panel emits code rather than approximating it, removing the translation step entirely. ## Links - [[3. Resources/Permanent Notes/Developer-Tool Adoption Tracks Cognitive-Load Reduction Not Feature Count]] — because extends: vocabulary alignment is one specific mechanism of cognitive-load reduction - [[3. Resources/Permanent Notes/Progressive Disclosure Keeps A Plugin's Core Trustworthy]] — because contrasts: that note is about *when* to reveal complexity; this one is about *whose words* describe it — different axes of the same interface-design problem - [[3. Resources/MOCs/Visual Thinking MOC]] — because applies-in: how tooling choices shape design-to-engineering communication ## Tests / falsifiers - Count handoff clarification requests per design spec across a team using CSS-native tooling vs. proprietary-layout tooling; the claim predicts fewer for the former. - Find a counterexample: a case where a shared low-level vocabulary *increased* coordination cost by forcing designers to reason about implementation detail they didn't need. If common, the claim needs narrowing. ## Maintenance - Last tended: 2026-08-16 - Next review: 2026-11-16 ## Source - [[3. Resources/Literature Notes/A Developer's Guide to Penpot (Design Tokens, Grid, Flex & Code)]] — the shared-vocabulary argument and the CodePen parity demonstration.