Main idea

A close theory establishes that an idea had been thought. A built system establishes that it had been held, with all the trade-offs that implies — and it is the second thing that counts when you are trying to find out whether a method has already existed.

The difference is verifiable on the antecedents encountered. HWMSE did not stay a diagram: the work produced prototypes and experimental applications, notably on macroeconomic decision-making. The Experience Factory is not an organizational principle: NASA's Software Engineering Laboratory was described as early as 1992 as an operational example after fifteen years of running and several hundred studies. SWAN is not only an ontology: its authors describe a Workbench and a Browser used to build and navigate the base.

Each of these systems shrinks the claimable territory more than a conceptual article from the same family, because it paid the price of being operated.

Why it matters

This gives a priority order in the search: look for what has run before looking for what has been written. Implementations are fewer, harder to find — they do not always publish — and far more decisive.

And it protects against too quick a conclusion in the other direction: finding ten neighbouring theoretical articles and no built system is information in itself, which says the mechanism may never have been workable.

Nuances and limits

A built system may have failed, or never got past the laboratory prototype: its mere existence does not prove the mechanism works at scale.

And industrial implementations often remain unpublished, which biases the search towards academic systems — the most visible, not necessarily the closest.

Open questions

• How do you reach the systems that were built and never published, which are precisely the ones the search will not see?