coulomb/target-revenue
4
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Complete TREV-WP-0001 research (T01-T05) and draft TRSL skeleton (T06)

Browse source 
Five research artifacts under specs/research/, grounded in primary-source
license text fetched live rather than relying on training-data recall:

- TRSL-PriorArt-Survey.md: BSL 1.1, FSL, PolyForm Noncommercial, Fair
  Source, Elastic License 2.0. Confirms TRSL's closed {MIT, Apache-2.0}
  Future License enum and no per-Phase custom license text follows FSL's
  deliberate fix for BSL's "Additional Use Grant" variability problem.
- TRSL-Terminology-Guardrails.md: confirms via OSI OSD Clause 6 that
  pre-conversion TRSL cannot be Open Source; confirms README's existing
  guardrail table without change.
- TRSL-FutureLicense-PatentPrecedent.md: MIT has no patent language; Apache
  2.0 has an explicit contribution-scoped grant + litigation termination.
  Recommends TRSL's pre-conversion phase also carry an express patent grant.
- TRSL-ContributorRights-Research.md: confirms via DCO 1.1 text that a DCO
  alone is insufficient for TRSL's dual-future-license promise; recommends
  a narrowly-scoped CLA over copyright assignment.
- TRSL-Jurisdiction-StandardTerms.md: deepens German §31/§32 UrhG (primary-
  confirmed) and §307 BGB Transparenzgebot (secondary-confirmed, flagged
  for counsel verification); corrects scope re: EU UCTD 93/13/EEC being
  consumer-only vs. German AGB law covering B2B too. Ranks five terms most
  needing objective definitions.

specs/TargetRevenueSourceLicense-Draft.md (T06) synthesizes all five into a
non-binding skeleton with every clause tagged [CONFIRMED BY RESEARCH],
[LEGAL], [WORKING DEFAULT], or [OPEN]. Per the T06 human-accept gate, the
workplan task stays `todo` — ready for review, not accepted.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This commit is contained in:
tegwick 2026-07-29 09:25:05 +02:00
parent 2b675c11ad
commit d3353fc22c
8 changed files with 404 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

67
workplans/TREV-WP-0001-license-prior-art-research.md
View file

@ -43,11 +43,19 @@ prepare the draft and leave status `todo` or note “ready for human accept.”
```task
id: TREV-WP-0001-T01
status: todo
status: done
priority: high
state_hub_task_id: "d2953e5a-7831-4593-93fd-3bd8faaec48d"
```
Result 2026-07-29: `specs/research/TRSL-PriorArt-Survey.md` produced,
grounded in primary-source license text fetched live (BSL 1.1, FSL,
PolyForm Noncommercial, Fair Source, Elastic License 2.0, OSI OSD). Key
finding: FSL's fixed two-license, single-trigger design was created
specifically to fix BSL's "Additional Use Grant" variability problem —
confirms TRSL's closed `{MIT, Apache-2.0}` enum and no per-Phase custom
license text is the right call, not just an internal preference.
Survey and compare structurally similar models: Business Source License 1.1
(MariaDB), Functional Source License (FSL), PolyForm Noncommercial /
PolyForm Shield, Fair Source (fair.io), Elastic License 2.0, Sentry's SSPL-
@ -64,11 +72,19 @@ diffed directly against TRSL's Phase/Target/Conversion model in
```task
id: TREV-WP-0001-T02
status: todo
status: done
priority: high
state_hub_task_id: "3f6166cd-d8a3-4f2e-a52e-28ecd08a8cf1"
```
Result 2026-07-29: `specs/research/TRSL-Terminology-Guardrails.md` produced.
Confirmed via primary OSD text (Clause 6, field-of-endeavor
non-discrimination) that pre-conversion TRSL cannot be Open Source, no
ambiguity. Confirmed Fair Source's three-part DOSP definition applies
cleanly to TRSL's structure. No change recommended to README's existing
guardrail table or to working defaults Q1/Q2 — this research confirms the
category framing rather than changing it.
Confirm and document precisely why TRSL's pre-conversion state cannot be
described as "Open Source" under the OSI Open Source Definition
(field-of-endeavour non-discrimination), and how Fair Source's "source
@ -85,11 +101,20 @@ improves the wording (human accept if changing published working defaults).
```task
id: TREV-WP-0001-T03
status: todo
status: done
priority: medium
state_hub_task_id: "0df9125d-0bbb-4e21-8b04-8a5dda02d140"
```
Result 2026-07-29: `specs/research/TRSL-FutureLicense-PatentPrecedent.md`
produced. Confirmed via primary text: MIT has zero patent language (implied
grant legally unsettled); Apache-2.0 has an explicit, contribution-scoped
patent grant plus litigation-termination clause. No change to working
default Q3 (both remain canonical options); added Phase-author guidance
(Apache-2.0 default where patentable technique is plausible). Recommends
TRSL's pre-conversion Phase also carry a minimal express patent license
(Apache-2.0-shaped) — new item, marked `[LEGAL]` in the T06 draft.
Research the practical and legal tradeoffs between MIT and Apache-2.0 as
canonical Future Licenses, focusing on: the extent and legal certainty of any
implied patent grant under MIT, Apache-2.0's explicit patent grant and
@ -106,11 +131,24 @@ Working default already allows both Future Licenses in schemas
```task
id: TREV-WP-0001-T04
status: todo
status: done
priority: medium
state_hub_task_id: "25fbf15c-60c4-4284-af84-49d109b74cf8"
```
Result 2026-07-29: `specs/research/TRSL-ContributorRights-Research.md`
produced. Confirmed via primary DCO 1.1 text that its certification is
scoped to "the open source license indicated in the file" at time of
contribution — it does not extend to a future, different Future License,
so a DCO alone is insufficient (concept §21.4's instinct confirmed with a
specific textual reason). Recommends a CLA (not copyright assignment,
to preserve contribution volume), scoped narrowly to the current Phase's
TRSL terms plus the Phase's already-declared Future License. `CONTRIBUTING.md`
interim policy (external contributions blocked pending an instrument)
left unchanged — this research specifies the eventual instrument's shape
but does not itself create it; recommends a future dedicated
`TRSL-ContributorLicenseAgreement-Draft.md` deliverable when needed.
Research contributor license agreement (CLA) and copyright assignment
patterns used by projects that must promise two future states (a restricted
phase license plus an eventual permissive license) — e.g. how BSL/FSL-based
@ -128,11 +166,23 @@ instrument exists).
```task
id: TREV-WP-0001-T05
status: todo
status: done
priority: medium
state_hub_task_id: "1fe4b444-90b0-49d0-8d2a-9e3a99d2822a"
```
Result 2026-07-29: `specs/research/TRSL-Jurisdiction-StandardTerms.md`
produced. Confirmed §31/§32 UrhG text via primary source; §307 BGB
Transparenzgebot confirmed via secondary legal sources (not primary-text-
verified this session — flagged for counsel verification). Important
scoping correction surfaced: German AGB law applies to B2B contracts too
(not just consumer), while the EU Unfair Contract Terms Directive 93/13/EEC
is consumer-only — a real asymmetry not previously distinguished. Ranked
five terms most needing objective definitions (settled payment first;
commercial use second). No change to working default Q2's "blocked on
legal" status; recommends a dedicated Definitions section in the eventual
license text.
Deepen the German-law groundwork already sketched in
`history/260728-InitialExploration.md` §10 (§31 UrhG scoped rights of use,
§307 BGB standard-terms clarity requirement, §32 UrhG author remuneration)
@ -167,3 +217,10 @@ mark every clause requiring specialist legal review before use, per
**Human accept gate:** Do not set this task to `done` until a human maintainer
explicitly accepts the skeleton as adequate briefing material for counsel.
Agent completion of the file is “ready for review,” not done.
Result 2026-07-29: `specs/TargetRevenueSourceLicense-Draft.md` produced,
synthesizing T01–T05. Every clause tagged `[CONFIRMED BY RESEARCH]`,
`[LEGAL]`, `[WORKING DEFAULT]`, or `[OPEN]` per §0's reading key. **Status
remains `todo` — ready for human review, not accepted.** Task should be
marked `done` only after a maintainer confirms the skeleton is adequate
briefing material for counsel.