Skip to main content

Layout Drift

The condition in which a Steward's surface changes the position, form or meaning of its elements between reads — by generation, adaptation or personalisation — so that each verification becomes a search rather than a glance, and the verification route is downgraded from the Steward's fastest input to their slowest.

Extended Definition​

The Transcoding Layer puts the Steward's verification route on vision because vision is the human's fastest input. That ranking carries a condition the corpus did not need to state until the industry began generating interfaces per prompt: vision is fastest only when the viewer knows where to look. A glance is a movement of the eyes to a known place and a comparison of what is there against memory. A search is the same movement repeated until something is found, and its duration is set by the layout rather than by the viewer. Layout Drift is the condition in which the surface makes the second the default — because the model designed it for this session, because it adapted to context, because it was personalised — and the Steward, who was promised their fastest channel, is handed reading with pictures.

The term names a failure of form, not of content. The Audit Surface specifies what a Steward must verify, in what time budget, at what level of abstraction; a surface can satisfy that specification on every read and still drift, if the same content is rendered in a different place or shape each time. Verification Blindness is what the drift produces — a Steward whose picture of the system lags the system — and Memo #127 located that condition in modality, state delivered as text or audio when it should be rendered. Layout Drift produces the same condition through a rendered surface. Nominal MTTI is what follows, by the mechanism the corpus already knows: a Steward who searches the surface on every check verifies less of it, verifies it less frequently, and stops verifying the parts that take longest to find, and the metric reads the withdrawal as autonomy.

The distinction the term depends on is between the value and the place. The state a surface shows changes constantly; that is its function. What must not change between reads is where each state lives, what form it takes and what each signal means. Adaptation of content is the surface doing its job; adaptation of layout is Layout Drift, and it is drift in the corpus's usual sense — a change the Steward did not commit and cannot see, because the surface that would show it is the thing that changed. A committed redesign is not Layout Drift: it is made once, at design time, with the Steward retrained against it, after which the surface is still again. Drift is uncommitted change.

The condition has two current sources and the corpus keeps them apart. A generated surface — the model composes an entire interface per prompt — drifts in every dimension at once. A component catalog — the agent selects from fixed, pre-approved elements and composes them per conversation — fixes function and meaning and leaves position and arrangement open, which is still drift, because a form built from the same three parts in a different order each time is a form the Steward has to read. Neither approach is wrong for what it was built for; single-read consumption loses nothing when the next interface differs. The Steward's surface is read thousands of times by one person whose picture of the system has to keep pace with the system, and it is the one surface on which the property that serves the first case breaks the second. The remedy is the Fixed-Function Surface.

  • Verification Blindness — The condition Layout Drift produces: a Steward whose picture of the system lags the system because each check has become a search.
  • Transcoding Layer — Puts the Steward's verification route on vision as the fastest input, a ranking that holds only while the viewer knows where to look.
  • Nominal MTTI — The metric that reads a Steward's withdrawal from slow, search-based verification as autonomy.
  • Audit Surface — Specifies what a Steward must verify, in what time budget and at what abstraction, a specification Layout Drift can break even when content is correct.
  • Steward Experience — The experience of the person reading the surface repeatedly, which Layout Drift degrades by turning each glance into a search.
  • Governance Tempo Gap — The gap between system pace and governance pace that widens when verification slows from a glance to a search.

Articles​

References​

  • Lexicon — canonical definition
  • Wiki — extended entry

Metadata​

First used: 2026-10-05
Pillar: How We Think

Part of the Arco Lexicon Ecosystem — maintained by Arco Venture Studio