Media Publishing Platform With Content Library

Publishers already have an asset problem long before they need a platform. The photograph exists in three places, the version in the article is not the version in the library, and the rights expiry is written in somebody’s notebook.

The library was therefore built to be the single place a piece of media is registered, with the article referencing it rather than copying it. Every asset carries its licence, its credit line and its expiry, and the publishing workflow blocks an article that uses something whose rights have lapsed. That one rule has done more for compliance than any review process.

Editorial workflow was kept deliberately light: draft, review, scheduled, published, with an audit trail of who changed what. Anything heavier was going to be abandoned within a month by people who write for a living.

The interesting constraint was storage and delivery. Originals are large, and serving them directly made the front end slow on the connections most of the traffic arrives on. Derivatives are generated at upload and the original is kept only where the licence requires it, which cut page weight substantially.

The search and filter combination was tested against the real library rather than a sample, because relevance behaves very differently on ten thousand assets than on fifty.

Handover covered the derivative sizes, the licence-expiry rule and the export path, since those three are what a new editor will need.