Content negotiation tracker

[bact]

Background

Content negotiation is a set of rules in HTTP that allows the same URL to provide different versions of a resource.

In the context of SPDX 3.1, this allows the URL https://spdx.org/rdf/3.1/terms/Core/Agent to act differently depending on who is asking:

• For Humans: When accessed via a web browser, the server delivers an HTML page (.html) providing a readable description of the Agent class, with examples - intended for human consumption.
• For Machines: When accessed by a reasoner or metadata harvester, the server delivers a Turtle (.ttl) file containing the machine-readable definitions.

Two work strands

This issue proposes us to have two strands of work regarding the content negotiation:

1. Spec - Documentation about how SPDX does content negotiation (the URL, the media type, the file extension, ect) should be kept in this spdx-spec repo. They could find a place in one of the informative Annexes.

2. Implementation - Technical work items to implement the content negotiation as described in (1), including server settings, will be carried in the infrastructure repo.

This issue will serve as a tracker for all things content negotiation.

See also

• Relevant trackers
  • IANA MIME media type registration and file extension tracker #1359
  • Meta Issue: RDF Specific Issues spdx-3-model#468
• HTTP Content Negotiation and RDF https://www.w3.org/wiki/ContentNegotiation
• Examples of Apache RewriteRule for content negotiation of ontologies at https://github.com/perma-id/w3id.org/
  • Brief explanation: https://github.com/perma-id/w3id.org/tree/master/examples#example-2-supporting-multiple-media-types-mime-types