NGDA Archive Data Model

This is the NGDA archive data model and ingest API. Features:

• Representation of archival objects as directory trees of files, and literal storage as such.
• Two types of directories that distinguish between 1) decomposition of objects/components into subcomponents, and 2) grouping of equivalent, alternative representations of objects/components.
• Explicit association of objects and object components with "definitions" (format specifications and other kinds of specifications that define semantics and interpretation), which are themselves represented as archival objects.
• Inter-object relationships.
• Object- and component-level lineage tracking.
• An XML manifest for archival objects that records the above information.
• Ingest templates, and association of objects with templates.
• An HTTP-based ingest API that supports one-time, incremental, and bulk ingest, and both template-based and non-template-based ingest.
• Support for archive self-description, including representation of the manifest and template schemas as archival objects.

Each of the following schemas is available in three schema languages: RELAX NG (compact syntax and XML syntax) and XML Schema, respectively. The first is the preferred and most readable format.

Schemas

manifest		[RNC | RNG | XSD]
template		[RNC | RNG | XSD]
ingest		[RNC | RNG | XSD]
format		TBD

Additional specifications

Identifiers as history
Standard inter-object relationships
Usage guidelines

Examples and notes