Skip to content

docs: make healthcare-default framing industry-neutral#49

Merged
abrichr merged 1 commit into
mainfrom
docs/industry-neutral-terminology
Jul 21, 2026
Merged

docs: make healthcare-default framing industry-neutral#49
abrichr merged 1 commit into
mainfrom
docs/industry-neutral-terminology

Conversation

@abrichr

@abrichr abrichr commented Jul 21, 2026

Copy link
Copy Markdown
Member

What

OpenAdapt serves multiple industries (healthcare clinics and mortgage/lending, plus others). This sweep rewrites healthcare-specific terms that were used as the general/default case so the product reads as industry-general, while keeping healthcare as one concrete example (never erasing the domain).

The trigger was guides/deploy-on-prem.md heading "What runs on the clinic machine" -- "clinic" is wrong as the general term for the on-prem host.

Changes (old -> new)

guides/deploy-on-prem.md

  • Heading "What runs on the clinic machine" -> "What runs on the on-prem machine"
  • "The clinic never auto-updates" -> "The on-prem host never auto-updates"
  • Citrix section: "the clinic host inside the clinic network" -> "the on-prem host inside your network"; "a clinician's session" -> "a user's session"; "the EMR" -> "the target application (for example an EMR)"; "the local EMR's system-of-record API" -> "the local system-of-record API"
  • "wrong-patient guarantees" / "wrong-patient identity ladder" -> "wrong-record ..."
  • "never patient data" -> "never the underlying record data"

concepts/identity-gate.md, guides/troubleshooting.md, concepts/capability-ladder.md (general-mechanism prose)

  • "wrong-patient" -> "wrong-record"; "two different patients" -> "two different records"; "a medical record number" -> "an identifier"

concepts/capability-ladder.md, guides/parameters-and-secrets.md, guides/record-your-app.md, guides/induce-a-program.md

  • Sole inline "patient" examples -> "record"

concepts/regulated-execution.md

  • "a consequential clinical write" -> "a consequential write" (the clinical-write policy name is unchanged)

The docs already used "wrong-record" as the neutral term in several places (security page, substrate-model), so this makes usage consistent.

Healthcare deliberately kept (example vs evidence)

  • Multi-vertical lists that already balance domains: index.md ("500th patient referral ... claims batch ... mortgage file"), security-and-data-handling.md ("clinics, lenders, and back offices")
  • Pinned qualification evidence: OpenEMR / openIMIS / Frappe Lending rows, the "WinForms Patient Notes" named test workflow, MRN glyph illustrations -- these are real and healthcare-flavored by nature
  • The clinical-write policy identifier (a literal CLI value)
  • The multi-trace-induction worked example (a concrete illustrative scenario)
  • PHI/PII/BAA/HIPAA/PHIPA and legal-attestation lines left untouched (owned by the concurrent compliance-copy agent)

Validation

  • pytest tests/ -> 64 passed (incl. test_substrate_evidence_copy.py)
  • scripts/validate_docs.py -> Validation passed
  • mkdocs build --strict -> clean (only pre-existing INFO notes)

No em dashes added. Coordinated with the concurrent --url/browser-framing agent (distinct lines in deploy-on-prem.md/run-a-deployment.md).

🤖 Generated with Claude Code

https://claude.ai/code/session_01NyCHrzA1psrKMFfroYbzaM

OpenAdapt serves multiple industries (healthcare clinics AND mortgage/
lending, plus others), so healthcare-specific terms used as the GENERAL
default case are reframed as industry-neutral, keeping healthcare as one
concrete example rather than the implied only vertical.

- deploy-on-prem.md: "the clinic machine" -> "the on-prem machine";
  "The clinic never auto-updates" -> "The on-prem host never auto-updates";
  Citrix section: "the clinic host inside the clinic network" -> "the
  on-prem host inside your network", "a clinician's session" -> "a user's
  session", "the EMR" -> "the target application (for example an EMR)",
  "the local EMR's system-of-record API" -> "the local system-of-record
  API"; "wrong-patient" -> "wrong-record" (guarantees + identity ladder);
  "never patient data" -> "never the underlying record data".
- identity-gate, troubleshooting, capability-ladder: general-mechanism
  "wrong-patient" -> "wrong-record" and "two different patients" ->
  "two different records" (matches the neutral "wrong-record" term the
  docs already use elsewhere); "medical record number" -> "identifier".
- capability-ladder / parameters-and-secrets / record-your-app /
  induce-a-program: neutralize sole inline "patient" examples to "record".
- regulated-execution: "a consequential clinical write" -> "a
  consequential write" (the clinical-write policy NAME is unchanged).

Kept healthcare where it is a legitimate example or real evidence:
multi-vertical lists (index.md, security page), the OpenEMR/openIMIS/
Lending qualification evidence, the "WinForms Patient Notes" named test
workflow, MRN glyph illustrations, the multi-trace induction worked
example, and the clinical-write policy identifier. Left PHI/PII/BAA/HIPAA/
PHIPA lines untouched for the compliance-copy agent.

mkdocs build --strict, pytest tests/ (64), and validate_docs.py all green.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NyCHrzA1psrKMFfroYbzaM
@abrichr
abrichr merged commit 54c11fd into main Jul 21, 2026
5 checks passed
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