Findings / UX-024
All three rating scales start on 3
OpenHighFailurenewsroom.sgit.ai
What we saw
The usefulness, more-like-this and level-of-detail scales at the foot of every article are preselected on the middle value.
Why it matters
Neither the reader nor the author can tell "not rated" from "useful". Every rating collected so far is ambiguous.
The fix
Start unselected; record only real taps.
The record
| Status | Open |
| Severity | High |
| Kind | Failure: works against the reader |
| Product | newsroom.sgit.ai, site |
| Where | https://newsroom.sgit.ai/articles/ |
| Problems | Preselected input |
| Visibility | Subtle: felt more than seen: it costs attention or trust, but is not noticed as a bug |
| Recommendation | Start empty: inputs start unselected; record only what the reader did |
| Area | Feedback |
| Breaks | A default is not an answer |
| Observed | 2026-10-11, v0.6.4 |
| Last checked | 2026-10-11, v0.7.1 |
| Source | Review: article page and feedback |
| Prototype | Article & reader primitives, route feedback, in three directions |
Previous: UX-023 There is no page that says why the publication exists