Merge pull request #2042 from sideeffects/sendupstream_tscmeeting20250507

Meeting notes for May 7th 2025

Author: jmlait

tsc/meetings/2025-05-07.md

@@ -0,0 +1,108 @@
+Minutes from OpenVDB TSC meeting, May 7th, 2025
+
+Attendees: *Ken* M., *Andre* P, *Dan* B.,  *Jeff* L., *Nick* A.
+
+Additional Attendees: None
+
+Regrets: *Richard* J., *Greg* H.
+
+Also:
+
+Jonathan Swartz (NVidia), Eric Powers (DWA)
+
+Agenda:
+
+1) Confirm quorum
+2) Secretary
+3) Meeting Times
+4) OpenVDB Course
+5) Github CI
+6) Github Openvdb Org
+7) Next meeting
+
+------------
+
+1) Confirm quorum
+
+We have quorum.
+
+2) Secretary
+
+Secretary is Jeff Lait.
+
+3) Meeting Times
+
+We are going to have a bi-weekly at 1pm PST.  The alternate weeks will
+be a 12pm PST overflow meeting only if needed.
+
+4) OpenVDB Course
+
+The course was accepted.  Requires copyright sign over.
+
+Reviews were extremely positive.  Jeff Budsberg was specifically requested.
+
+One of Nick or Richard will attend.  Likely cover anisotropic Surfacing.
+
+We need find if Jeff Budsberg can cull out the sensitive material for
+a presentation.
+
+We need a confirmed order of presenters and times so the SIGGRAPH
+website can have the times available for people to show up at the
+right time.
+
+5) CI
+
+We got credentials for Amazon code build a month ago.  The action
+didn't work, and the manual got very complicated quickly.  Now trying to
+locate someone to figure this out.
+
+6) Git Hub Org
+
+Proposal is to take over the openvdb org on GitHub and use it to host
+separate repos.  This allows us to have administrative rights to
+create/archive repos; which we can't do inside of aswf.  The entire
+structure remains legally part of ASWF.
+
+The trigger was to find a place for all the FVDB contents to live.
+
+Should AX have been in OpenVDB?   Our focus should be the core library?
+
+Are these extra FVDB things part of FVDB core?   A segmentation of
+pointclouds might bring in a lot of distinct sources.  They might also
+target a specific version of FVDB.  So even if FVDB core was in this
+org-repo, there still would be separate repos for the projects.
+
+Where does example openvdb code for courses live?  We drop these into
+the website, or ask for a repo for this purpose?  This organization
+could be a place for this.  We get the autonomy to create repos for these.
+
+What of unmaintained code?  How is the review process work for these?
+The proposal was review rules are per-repo; so some repos may be very
+light on review/maintaince.   It is believed that CLA/DCO would be
+required across the board, however, to ensure we can move code without
+hassle.
+
+Considerable discussion on the merits of moving existing projects,
+like AX, nanovdb, fvdb, mathematica bindings, houdini operators, out
+of the core repo to this one.  Monolithic repos have issues where it
+clouds the purpose and direction.  Github issues, etc, apply across
+the whole repo, as do PRs.  On the other hand, a monolithic repo will
+keep versions in sync by construction - you don't have to find the
+right version of OpenVDB for your copy of NanoVDB as they both come
+from the same source.  This is useful when things are tightly coupled.
+
+Separate repos are useful when we know we will want to de-sync - for
+example, a project of how to do segmentation might only have been
+built against a particular fVDB version.  Keeping it in the main repo
+requires constantly sweeping it; while having it separate will allow
+it to just point to the last good build.
+
+No decisions were made about moving existing directories.
+
+Unaminous consent to take over openvdb org.
+
+7) Next meeting
+
+Agenda: Finalize SIGGRAPH course order and timings.
+
+May 21st, 2025, 4pm EST.