Main idea

Looking for "the new idea" inside a method is the wrong question, and the prior-art search on Contexte shows it block by block. Provenance has a literature and a standard. Event Sourcing rebuilds a state from a history of events. Architecture Decision Records keep a decision, its rationale and its replacement. Truth Maintenance Systems already held several incompatible sets of assumptions back in the 1980s. HWMSE, Experience Factory, KCS, ScholOnto and SWAN already assemble big chunks of it.

What remains is not a component: it is a work of selection — which blocks, among all those that exist — of combination, and of normalization. A method has its vocabulary, its objects, its constraints, its procedures, its particular way of separating source, knowledge, state, history, synthesis, decision, contradiction, handoff and deliverable, and its rules for circulation between autonomous spaces.

No historical system possesses exactly that grammar. Almost every one of its words has one, and an earlier one.

Why it matters

This changes what you claim, and it makes the claim sustainable. "Here is what I borrow, here is what I adapt, here is what I combine, here is the form I give it" survives the discovery of one more antecedent; "this is new" falls at the first.

It also shifts the criterion for judging a method: not the novelty of its pieces, but the coherence and ease of use of the whole they form.

Nuances and limits

The position is unspectacular, and it sells badly: a combination cannot be told as a discovery, even when it is worth more.

And it is not a licence: selecting and normalizing can also produce an arbitrary assembly, coherent only to its author. The combination is only worth something if it makes the mechanisms usable together.

Open questions

• How does a combination get passed on to someone who knows none of the traditions it is made of?