pki.sgit.ai / packs / registry-mvp / doctrine

Appendix C — Doctrine

Wardley's forty doctrines are the part of his model that always applies — the training, not the battle plan. This appendix explains them for readers who know the maps and have never met the doctrine, then rates this project against all forty. The shape of the result turned out to be the finding.

Read this as a self-assessment by the party being assessed, which is the weakest kind there is. It is published in that form on purpose: every rating names the artefact it rests on, so a reader who disagrees has the evidence in front of them. Assessment version .

What doctrine is, and why it is not strategy

Most people meet Wardley Mapping through the maps — components positioned against a value chain and an evolution axis. The maps are only one layer of the model, and they are the context-specific one. Doctrine is the layer underneath, and its defining property is that it is context-free: it applies whatever your industry, landscape or strategy happens to be.

LeadershipThe context-specific moves — gameplay. What you do, here, now.
DoctrineUniversal good practice. Applies to everyone, regardless of context. This page.
ClimateThe rules that govern change — evolution, inertia, the things you do not control.
LandscapeThe map. Positions relative to an anchor, usually a user need.
PurposeThe mission. What game you are playing at all.

Wardley's own analogy is military: doctrine is the training of your forces — the standard techniques any unit benefits from, drilled regardless of which battle they end up in. You would not send troops into a fight to learn how to communicate. Strategy tells you where to attack; doctrine is why you are capable of attacking anything.

Strategy and mapping are context-specific. Doctrine always applies — so an organisation's ability to handle a hard problem tends to track how well it practises doctrine, before any strategy is chosen.

One distinction matters, and Wardley is explicit about it: these are principles you choose because the evidence says they are broadly useful — not immutable laws. Several writers separate doctrine, meaning chosen and evidence-based good practice, from dogma, meaning belief held past the point of questioning. Rating yourself against forty of them, as this page does, is only useful if you are willing to argue with the rating.

How they are organised

Roughly forty principles, grouped into six categories and sequenced into four phases of increasing maturity. The phases are the useful part: they say what to fix first. There is no point distributing decision-making before anyone agrees what the words mean.

The four phases

Click any phase to filter the forty below. The bar shows practised · partly · not practised · no basis yet.

The six categories

Where this project actually stands

Assessed as if the pack were built as specified — so the ratings cover the design, the published record and the one shipped surface, not a running registry.

All forty

What the shape says

The result is not a score, and it is not flattering in an interesting way. This project practises the doctrines that a small, documentation-heavy effort can practise on its own: agreeing what words mean, publishing the reasoning, challenging assumptions including its own, knowing the details of what it measured, using standards rather than inventing them, and admitting error in public. Those are real, and they are also the cheap ones for a project of this shape.

It is weak on every doctrine that requires other people. Know your users is the sharpest: the only user is the project itself. Listen to your ecosystem has no ecosystem to listen to. Be the owner has an empty Sponsor field on its own specification. Distribute power and decision making routes every open decision to one person, and the register proves it. Five more cannot be rated at all, because rating team structure with no team would be theatre.

The two findings the pack already had — that nobody outside the project has been asked whether this is a need, and that nothing can move past Draft without an accepting authority — are not two problems. They are the same doctrinal hole, showing up in two places.

Which gives the appendix a use beyond the scoring. Phase I is nine doctrines: five practised, three partly, and one not — and the one that is not is know your users, which is the one that gates everything after it. The cheapest next action in the entire pack — ask five operators whether anybody has ever asked them to prove an agent's authority — is a Phase I doctrine fix, and it needs no registry, no REP and no code.

Honest tensions in the assessment itself

TensionNote
Self-assessedThe party being rated wrote the ratings. Mitigated only by naming the artefact behind each one, which turns disagreement into something a reader can act on rather than merely feel.
“No basis yet” is convenientFive doctrines are excluded because there is no organisation. That is true, and it is also the answer that flatters us most — a reader should treat those five as unearned rather than neutral.
The phase model implies a ladderWardley sequences the phases, and real projects practise bits of Phase III while missing bits of Phase I. This one does exactly that.
Forty is approximateThe canonical list is “roughly forty” and different sources group and name them slightly differently. The categories and phase placements here follow the briefing this appendix was built from; a few placements are ours and are arguable.
Rating design rather than a running systemThe registry is unbuilt. Doctrines about operations and flow are rated against how the project is run, which is a weaker claim than rating a service.

Sources