Re: RFC111 AI/LLM tool policy: proposal for a significant revision
Daniel Evans via gdal-dev <[email protected]> Thu, 14 May 2026 09:30:19 +0100
| Newsgroups | gmane.comp.gis.gdal.devel |
|---|---|
| Message-ID | <CANnaQaYuoY0D-MFmmjPN23i040LzWkTXLj8nYHRb0=xNKHm9YQ@mail.gmail.com> |
--===============9027233042499560336== Content-Type: multipart/alternative; boundary="000000000000c3520d0651c2e809" --000000000000c3520d0651c2e809 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable I also strongly suspect that "require an issue to describe the issue" would simply translate to "the LLM is asked to generate even more fluff than it usually puts in the PR description". Also, would it mean that one-line docs typo PRs need an approved issue? If not, what would be the threshold for "this proposed change must be approved" and who would decide it? Cheers, Daniel On Wed, 13 May 2026 at 23:10, Even Rouault via gdal-dev < [email protected]> wrote: > Andrew, > > I doubt you can enforce that through github configuration, and even if we > could, that seems to me like adding overhead to all the people acting > reasonably (some "institutional" large projects may have that kind of > policy, but none of the projects I regularly or occasionally contribute t= o > require this). That said, opening an issue to discuss a problem and expos= e > some ideas can certainly be done when needed (and going through a full RF= C > for substantial changes), but someone contributing a typo fix shouldn't > have to open a ticket first. > > Even > Le 13/05/2026 =C3=A0 19:38, Andrew Bell a =C3=A9crit : > > Would it be helpful to require an issue that describes the change to be > made and then some approval before permitting an PR to be opened? (I have > no idea if this is technically feasible.) > > On Wed, May 6, 2026 at 12:29=E2=80=AFPM Even Rouault via gdal-dev < > [email protected]> wrote: > >> Hi, >> >> based on the experience gained from the initial policy, I propose to >> significantly revise it to drastically limit their use. See >> https://github.com/OSGeo/gdal/pull/14500 >> > > -- > Andrew Bell > [email protected] > > -- http://www.spatialys.com > My software is free, but my time generally not. > > _______________________________________________ > gdal-dev mailing list > [email protected] > https://lists.osgeo.org/mailman/listinfo/gdal-dev > --000000000000c3520d0651c2e809 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"><div>I also strongly suspect that "require an issue t= o describe the issue" would simply translate to "the LLM is asked= to generate even more fluff than it usually puts in the PR description&quo= t;.</div><div><br></div><div>Also, would it mean that one-line docs typo PR= s need an approved issue? If not, what would be the threshold for "thi= s proposed change must be approved" and who would decide it?</div><div= ><br></div><div>Cheers,</div><div>Daniel</div></div><br><div class=3D"gmail= _quote gmail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">On Wed,= 13 May 2026 at 23:10, Even Rouault via gdal-dev <<a href=3D"mailto:gdal= [email protected]">[email protected]</a>> wrote:<br></div><blo= ckquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left= :1px solid rgb(204,204,204);padding-left:1ex"><u></u> =20 =20 =20 <div> <p>Andrew,</p> <p>I doubt you can enforce that through github configuration, and even if we could, that seems to me like adding overhead to all the people acting reasonably (some "institutional" large projec= ts may have that kind of policy, but none of the projects I regularly or occasionally contribute to require this). That said, opening an issue to discuss a problem and expose some ideas can certainly be done when needed (and going through a full RFC for substantial changes), but someone contributing a typo fix shouldn't have to open a ticket first.</p> <p>Even</p> <div>Le 13/05/2026 =C3=A0 19:38, Andrew Bell a =C3=A9crit=C2=A0:<br> </div> <blockquote type=3D"cite"> <div dir=3D"ltr"> <div dir=3D"ltr"> <div>Would it be helpful to require an issue that describes the change to be made and then some approval before permitting an PR to be opened? (I have no idea if this is technically feasible.)<br> <br> </div> <div class=3D"gmail_quote"> <div dir=3D"ltr" class=3D"gmail_attr">On Wed, May 6, 2026 at 12:29=E2=80=AFPM Even Rouault via gdal-dev <<a href=3D"mai= lto:[email protected]" target=3D"_blank">[email protected]</a= >> wrote:<br> </div> <blockquote class=3D"gmail_quote">Hi,<br> <br> based on the experience gained from the initial policy, I propose to <br> significantly revise it to drastically limit their use. See <br> <a href=3D"https://github.com/OSGeo/gdal/pull/14500" rel=3D"n= oreferrer" target=3D"_blank">https://github.com/OSGeo/gdal/pull/14500</a><b= r> </blockquote> </div> <div><br> </div> <span class=3D"gmail_signature_prefix">-- </span><br> <div dir=3D"ltr" class=3D"gmail_signature"> <div dir=3D"ltr"> <div>Andrew Bell</div> <a href=3D"mailto:[email protected]" target=3D"_blank"= >[email protected]</a></div> </div> </div> </div> </blockquote> <pre cols=3D"72">--=20 <a href=3D"http://www.spatialys.com" target=3D"_blank">http://www.spatialys= .com</a> My software is free, but my time generally not.</pre> </div> _______________________________________________<br> gdal-dev mailing list<br> <a href=3D"mailto:[email protected]" target=3D"_blank">gdal-dev@list= s.osgeo.org</a><br> <a href=3D"https://lists.osgeo.org/mailman/listinfo/gdal-dev" rel=3D"norefe= rrer" target=3D"_blank">https://lists.osgeo.org/mailman/listinfo/gdal-dev</= a><br> </blockquote></div> --000000000000c3520d0651c2e809-- --===============9027233042499560336== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ gdal-dev mailing list [email protected] https://lists.osgeo.org/mailman/listinfo/gdal-dev --===============9027233042499560336==--