Skip to content

docs: names reference — how Compose and Module were chosen#67

Open
wmadden-electric wants to merge 3 commits into
mainfrom
docs/names-reference
Open

docs: names reference — how Compose and Module were chosen#67
wmadden-electric wants to merge 3 commits into
mainfrom
docs/names-reference

Conversation

@wmadden-electric

Copy link
Copy Markdown
Contributor

Adds agent-os/product/names-compose-and-module.md — the consolidated reference for the two settled names, written for anyone who wasn't in the room:

  • The names at a glance — Prisma Compose (framework), Module (unit), Prisma App (artifact), plus the supporting registers, and the one-breath sentence that uses them all.
  • The philosophy — three principles, each learned by shipping a wrong name first: products are named for value but names must refer as well as describe; vocabulary nouns are tested in speech, not on the page; identity belongs to the registry, not the unit noun.
  • The rubrics — the product value tests, the referential battery (workbench frame, artifact-collision hard rule, bare token, identity frames), and the eight unit-noun spoken tests.
  • Why Module / why Compose — the Hex → System → Module and MakerKit → Prisma App → Compose histories, what each failure taught, why the winners pass, and the runners-up with their disqualifiers.
  • Lessons worth keeping — the transferable rules (test in speech; test at the smallest grain; never name a tool after its output; the gloss test; verify "nobody says X").

Linked from the product README as the naming entry point; ADRs 0025/0026/0027 and vocabulary-tests.md remain the binding records.

🤖 Generated with Claude Code

wmadden-electric and others added 3 commits July 13, 2026 08:56
The consolidated story of the two settled names: the register-based
philosophy (value words describe; milieu names must refer; vocabulary
nouns are tested in speech; identity belongs to the registry), all
three rubrics, the Hex/System/MakerKit/Prisma App failure history, why
Module and Compose won, the candidates that lost, and the transferable
lessons. Linked as the naming entry point from the product README.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: willbot <w.a.madden+machine@gmail.com>
Signed-off-by: Will Madden <madden@prisma.io>
Naming a tool after its output is a red flag that predicts referential
failure, not a hard rule that decides by itself — the referential
frames (workbench, bare token, identity) are what decide. Drops the
"every mention parses as the artifact / no prefix can rescue it"
over-claim from the names reference and the rubric.

Co-Authored-By: Will Madden <madden@prisma.io>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: willbot <w.a.madden+machine@gmail.com>
Signed-off-by: Will Madden <madden@prisma.io>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: willbot <w.a.madden+machine@gmail.com>
Signed-off-by: Will Madden <madden@prisma.io>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant