wardley-maps.sgit.ai / Method / You cannot map a gap
You cannot map a gap
Original contribution 28 July 2026 W12 · glossary
you cannot map a gap, because a gap has no evolution.
The claim
A Wardley map positions components on an axis that measures how far each has travelled from novel to ubiquitous. That measurement presupposes something being measured: a thing that exists, that people use, that has a history of getting more certain and more common.
An air gap is not a thing. It is the absence of a connection — a place where two systems that would otherwise be joined are deliberately not joined, and where the joining is done by a person carrying something. There is no component to position, because there is no component. It has not evolved from anything and it is not on its way to anything.
So the interesting question stops being where does the gap sit and becomes what does a value chain look like when there is a gap in the middle of it. And the answer turns out to be a distinctive and reusable map shape.
The shape
this will be a very interesting shape where you can have the top working quite efficiently, and then the middle bit those lines all the way to the left because they're manual… the ends are solved. the middle is people.
Both ends of the chain sit well to the right. The systems on either side of the gap are mature — often commodity — because they have had years of pressure and investment. Everything that crosses the gap sits at the far left, because it is a human being with a USB stick, a clipboard, a phone call, or a form that gets retyped. And the components hanging off that manual step inherit its position, because you cannot be more evolved than the least evolved thing you depend on.
Drawn out, the value chain sags in the middle. That sag is the finding. It is what makes the map an argument rather than an inventory: it says the investment case is not at either end, where everything already works, but in the trough — and it says so visually, before anybody has written a business case.
The contradiction this site has not resolved
The reading this page finds most defensible: the map does not position the gap; it positions the manual work that exists because of the gap. A person walking a drive between two networks is a component — it is an activity, somebody does it, it has a cost, and it is emphatically at the left-hand end of the axis. The gap itself never appears on the map at all. What appears is its shadow.
That is a reading, not a resolution, and the corpus does not state it. Published unresolved as Q3.
What would falsify it
- A defensible placement of a gap itself. If somebody can position an absence on the evolution axis and have the climatic patterns behave — have it move right under pressure, have it commoditise — the claim is wrong and the interesting question is what the axis was actually measuring.
- The shape failing to recur. The generalisation above is an assertion from a handful of cases. A survey of value chains containing a manual step that does not produce the sag would kill it.
- The trough not being where the value is. The shape implies an investment conclusion. If in practice the returns from removing the manual middle are consistently worse than from improving the already-good ends — plausible, since the ends are where the volume is — then the map is a true description with a false implication, which is worse than being wrong.
Related
- The custom-axis verdict — same brief, 28 July 2026. The
axis
air gap → file → API → event-drivenis the one the verdict sanctions, and it is the axis this map would need. - /shipped/ — the air-gap map, the translation map and the evidence-chain map were all commissioned on 28 July 2026 and all three are still listed as PROPOSED.