Context
Asset organisations need to prove, decades after the fact, what configuration the asset was in at any specific moment. Regulators ask. Lawyers ask. Root-cause investigators ask. Most CDEs answer this with revision control: 'the latest state, with a manual history of past revisions attached.' That is the wrong shape for serious work on long-lived assets. What complex assets need is closer to what Git brought to source code: a full version-control discipline for data.
Most CDEs offer revision control. Weaver offers Git-style version control for asset data.
Explanation
Four capabilities distinguish a Git-style version history from a revision log. Weaver provides all four, as properties of the data model itself rather than features bolted on:
Every change is a commit
Every edit to every information container is captured as an immutable RDF transaction: who, when, what, before-state, after-state. There is no 'delete' operation that removes history; there is only 'supersede'. The history is a directed acyclic graph (DAG), not a flat log. Earlier versions remain queryable forever.
Branching and tagging
Engineers can fork the data into a branch to explore alternatives (a candidate cooling-system specification, a proposed revision to the safety case, a what-if for a regulator's question) without disturbing the main line of development. Commits can be tagged as formal baselines: as-built records, regulatory submissions, design freezes. The tags are queryable and persistent.
Diff and merge with conflict resolution
Any two commits or branches can be compared exactly, by query: 'what changed between submission 7 and submission 8'. When two teams have changed the same container in parallel, the conflict is surfaced explicitly and resolved deliberately, not silently dropped. The semantic structure (ISO 15926-11) gives merges first-class awareness of which fields are independent and which are not.
Time-travel queries are first-class
'Show me the configuration of System X as it existed on 14 March 2031' is a single SPARQL query. So is 'Show me every requirement that has changed since the last regulatory inspection.' So is 'Trace the lineage of this design decision back to the originating requirement.' Time becomes another query dimension, not a journey through file archives.
When an inspector asks 'what did this look like in 2034?', the answer is a query. When an engineer asks 'can we explore this alternative without breaking the baseline?', the answer is a branch. When two teams have edited the same thing, the answer is a merge with explicit conflict resolution. The same discipline Git brought to source code, applied to the information that describes your asset.