Skip to content

DayBack v0.1.0 — reviewed prototype

Pre-release
Pre-release

Choose a tag to compare

@sharonbasovich sharonbasovich released this 10 Nov 21:50

DayBack v0.1.0 — reviewed prototype

A local-first capacity check for an invented student week: lose a study day, see the first deadline that no longer fits, preview an explicit earlier-time or effort repair, then Apply or Undo. Protected commitments and deadlines stay put.

This release is pinned to commit 7fe87224a95b919aaf0ff39ee6f5d1fbcd9d6e07. There are no new runtime changes beyond that reviewed source.

Live demo

Open DayBack on free static GitHub Pages hosting. Baseline, missed day, late/early repair preview, Apply and Undo were checked on the hosted app. No account or API key is needed. Edits are ephemeral and lost on reload.

The tag, source ZIP and HTML asset remain the original reviewed snapshot; their embedded documentation predates this hosting update. This update changes documentation, not runtime code or asset bytes.

Downloads and running

  • DayBack.html: self-contained HTML with the stylesheet and four modules embedded.
  • dayback-v0.1.0-source.zip: the exact 16 published source, test, documentation and portable-artifact files.

The documented run route is a local HTTP server. Extract the source ZIP, enter its folder and, with Python 3 installed, run:

python3 -m http.server 8000 --bind 127.0.0.1

Open http://127.0.0.1:8000. The single-file artifact was also checked over local HTTP. Direct file:// opening and disconnected-device/offline execution remain unverified. No app dependency downloads, API keys, accounts or paid services are required. The live demo above is an additional run route; these downloads remain unchanged.

Verification and limits

23 local automated tests passed. Independent checks compared the embedded module/HTML graph with source and exercised packaged handlers. Bounded native browser checks of the packaged HTML over HTTP covered baseline, missed day, late/early preview, Apply and Undo. These are not comprehensive accessibility or device certification; screen-reader and OS reduced-motion runtime acceptance remain unverified.

The model assumes one person, divisible half-hour work available at horizon start, explicit deadlines and non-overlapping study time with protected overlap excluded. It does not model fatigue, prerequisites or uncertain effort. Synthetic data only; no user study or measured productivity/grade benefit. All entered data is ephemeral page memory: reloading loses it. No calendar integration, telemetry or persistent storage.

Design, implementation, tests and review were produced with OpenAI AI-assistant assistance. No third-party runtime dependencies or external media assets are included. No open-source license grant is made; applicable copyright is retained. This release is not an event submission or award claim.

SHA-256

80b76482b43d3675c79e9cae9db8d39718214bf7f0766730d5bdd2713c79b876  DayBack.html
d79a925c7aad32aa7c93eb2a4c5cc93f6141ea56602b90a55a1eaa5be8311161  dayback-v0.1.0-source.zip

Screenshot walkthrough

DayBack-walkthrough.mp4 is a 96-second silent screenshot walkthrough of the synthetic example, not a live recording. Still screenshots are editorially arranged; captions explain baseline, lost day, late/early repair, Apply, Undo and limits. DayBack-walkthrough.srt and TRANSCRIPT.md provide the caption text. No narration, music, user study or real student data is included.