<div dir="ltr"><div dir="ltr"><div dir="ltr"><div dir="ltr"><span style="background-color:transparent">Greg Troxel <<a href="mailto:gdt@lexort.com" target="_blank">gdt@lexort.com</a>> ezt írta (időpont: 2026. jún. 9., K, 12:16):</span></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Regardless of how the consensus ends up</blockquote><div><br></div><div><br></div><div>Since consensus on a main-repository AI/LLM policy seems difficult right now, I propose an experimental compromise: </div><div>- create <b>a separate "postgis-agentic-lab" repository</b> !<br>It would allow AI/agentic experimentation without adding noise or review pressure to the main postgis repo.<br>It would create no obligation for maintainers, but useful results could be manually extracted by a human postgis maintainer and submitted upstream through the normal process.</div><div><br></div><div>So some postgis users may be able to contribute not only bug reports, but also AI-assisted candidate solutions. ( Fable/Mythos welcome ) </div><div>Even if these are not accepted upstream, they may still be useful to other postgis users with similar issues.</div><div><br></div><div>Imre / wearing my postgis power user hat /</div><div><br></div><div><br>=====================<br><br><b><font size="4">Proposal: ~ `postgis-agentic-lab`</font></b><br>( brainstormed with Chatgpt-5.5-High ) </div><div><br></div><div><span style="background-color:transparent">The goal is to ensure that AI-assisted work demonstrably reduces, rather than increases, the maintenance burden on PostGIS maintainers.</span></div><div><br><b>Main PostGIS repository</b><br>- Only human contributors may submit upstream issues, pull requests, review comments, or mailing list messages.<br>- <b>Existing trusted postgis commiters/contributors may use AI assistance f</b>or small, reviewable changes, preferably under a soft limit such as ~50-100 changed lines. ( new contributors with AI --> `postgis-agentic-lab` ) <br>- Substantial AI assistance must be disclosed, including the tool/model used where relevant.<br>- The human contributor remains fully responsible for understanding, testing, licensing, and maintaining the contribution.<br>- Any `AGENTS.md`, skills, prompts, or agentic instructions in the main repository should clearly say <b>that agentic experiments belong in `postgis-agentic-lab`,</b> not in the main repository. ( including postgis end-users agentic code suggestions , fix ) <br>- Autonomous agents must not act directly in official PostGIS project spaces.<br><br><b>Postgis-Agentic-lab repository:</b><br>- Create a new fork<b> `<a href="http://github.com/postgis/postgis-agentic-lab" target="_blank">github.com/postgis/postgis-agentic-lab</a></b>` as<b> an experimental AI staging/lab repository for postgis</b><br>- The lab is for AI-assisted reproduction, testing, analysis, benchmarking, patch prototyping, prompt sharing, and agentic workflow experiments.<br>- The lab is continuously synchronized from the main PostGIS repository, but nothing flows back automatically.<br>- AI agents and human+agent workflows may create issues and pull requests there under special rules.<br><b>- Work in the lab creates no obligation for PostGIS maintainers to review, respond, accept, or merge anything.</b><br>- Stale lab issues/PRs without a human owner or clear progress may be <b>automatically closed after 30 days.</b><br><b>- Agentic tools such as openclaw, hermes, or similar systems should work in this lab repository, not directly against the main PostGIS repository.</b><br><br><b>Supported use cases in the lab</b><br>- Generate a minimal, reproducible, human-friendly bug report from a real-world problem ( by the postgis-end users or agents ) <br>- Sanitize and reduce private or company data into a safe public reproducer.<br>- Create a failing new PostGIS regression test as a separate commit.<br>- Analyze the root cause and document competing hypotheses.<br>- Create a detailed fix prompt for a general AI coding agent.<br>- Produce a candidate patch as a separate commit.<br>- Document test results, benchmark methodology, and performance impact.<br>- Share useful prompts, agent instructions, negative results, and failed approaches.</div><div>- ....<br><br><b>Contribution categories in the agentic-lab:</b><br>- "Supported": reproducible bug reports, minimized test cases, benchmarks, root-cause analysis, documentation of methodology.<br>- "Accepted for human upstreaming": small, tested, human-owned patches with clear provenance and AI disclosure.<br>- "Tolerated / prototype" : speculative fixes, large refactors, experimental prompts, agent-generated ideas, incomplete prototypes.<br>- "Forbidden": autonomous / generated spam, hallucinated test results, undisclosed private data, unclear license provenance, or any change the human owner cannot explain.<br><br><b>Upstream promotion rule</b><br>- The lab is not an official upstream contribution queue.<br>- Only a human trusted postgis committer may manually prepare and submit a final PR to `postgis/postgis`.<br>- The upstream PR must be small, reviewable, tested, disclose substantial AI assistance, and link back to the relevant lab artifact if useful.<br>- Ideally, the upstream PR should contain the prompts and:<br> - one commit for the failing test,<br> - one commit for the fix,<br> - optional documentation or benchmark notes.<br></div><div><br></div><div>etc. <br>=====</div><div><br></div></div></div>
</div>
</div>