Distribution and versioning

Gapwise separates canonical ownership, public distribution, and runtime consumption.

Multi-university dataset discovery

To discover available universities and campus models programmatically, use the public API discovery endpoints:

# Discover 13 supported universities
curl https://api.gapwise.ca/v1/universities

# Discover 15 campus models and metadata
curl https://api.gapwise.ca/v1/campuses

In the canonical GapwiseHQ/data repository, each supported university maintains its campus-wide dataset at universities/<id>/campus.json (e.g. universities/carleton/campus.json), containing canonical building identities, coordinates, and navigation metadata.

Raw first-party distribution

The current canonical raw distribution channel for detailed UTM exterior entrance geometry, routing graphs, and access audits is published under:

https://data.gapwise.ca/datasets/utm/latest/

A production build generates:

https://data.gapwise.ca/datasets/utm/latest/manifest.json

The manifest lists each distributed artifact with its byte size, SHA-256 digest, and first-party URL. This makes raw consumers independent of GitHub’s raw-content URLs and provides an integrity boundary for mirrors and caches.

Example:

const campus = await fetch(
  'https://data.gapwise.ca/datasets/utm/latest/buildings.geojson'
).then((response) => response.json());

Why Gapwise itself does not live-fetch it

The student web app, routing engine, and public API must keep functioning if the Data portal is unavailable. Core therefore pins and validates a snapshot during development/build rather than fetching data.gapwise.ca for every student request.

This gives us both:

• one official source of campus facts; and
• no new production single point of failure.

latest versus immutable releases

latest is the human-friendly current channel. Consumers that require strict reproducibility should pin a checksum or an immutable dataset release once one is published. Do not assume that latest will remain byte-identical forever.

The public API uses its own versioned compatibility contract. A data artifact changing does not automatically imply an API breaking change.

API versus raw data

Prefer the API/SDK when you want Gapwise semantics such as building resolution, route calculation, warnings, confidence, or gap feasibility. Prefer raw distribution when you are doing research, visualization, provenance inspection, validation, or building your own derivation pipeline.