Required content: Metadata schema to help agencies fill out enterprise code inventory (7.2)

[mattbailey0]

Section 7.2 of the policy states that:

Within 120 days of the publication date of this policy, each agency must update—and thereafter keep up to date—its inventory of agency information resources to include an enterprise code inventory that lists custom-developed code for or by the agency after the publication of this policy. [..] Agencies shall fill out this information based on a metadata schema that OMB will provide on Code.gov.

[rafael5]

Code.gov metadata model recommendation

https://schema.org/SoftwareSourceCode

1. JSON-LD. This is the W3C Linked Data standard for structured data on the web, also known as the Resource Description Framework (RDF). The JSON-LD serialization of RDF is the most web-programmer friendly form of RDF, and supplements the JSON standard with the "@ context" attribute.
2. Schema.org Using a schema.org compatible model will maximize the search and discoverability of software metadata published on the web. The Schema.org metadata model used by the largest web search engines to index and search all structured data on the web, and will thus maximize discoverablity of all software artifacts published on the web. This metadata (model) is published here:

Background

The U.S. Department of Veterans Affairs, U.S. National Library of Congress, U.S. National Library of Medicine, and other federal agencies and knowledge organizations have adopted the W3C Linked Data standard (Resource Description Framework) as their metadata standard to provide master data managment for all their data resources, making them fully compliant to all W3C web search engines and technologies. VA, for example, is projecting all of its data assets to an RDF model using JSON-LD to surface all 35 years of its data:

http://vistadataproject.info

The world's largest search engines (Google, Yahoo, Microsoft, Yandex) have collaborated on creating a RDF schema such that all search engines can index and structure all data on the web using a single common schema. This is published here:

https://schema.org

[mattbailey0]

See discussion on this issue here: WhiteHouse/source-code-policy#117

[mattbailey0]

In addition to the schema definition itself, it's important that agencies have easy to understand instructions on how to work with it in building their code inventory. We've created a draft here..

• Is it clear and comprehensive?
• Can it be better organized?
• What additional questions are agencies likely to have?

[mattbailey0]

Consolidating related discussions at #41