Academy Software Foundation Technical Advisory Council (TAC) Meeting - April 29, 2026

Join the meeting at https://zoom-lfx.platform.linuxfoundation.org/meetings/aswf?view=list&projects=aswf

Voting Representative Attendees

Premier Member Representatives

Project Representatives

Carol Payne - OpenColorIO Representative
Cary Phillips - OpenEXR Representative
Chris Kulla - Open Shading Language Representative
Daniel Greenstein - OpenImageIO Representative
Diego Tavares Da Silva - OpenCue Representative
Jonathan Stone - MaterialX Representative
Karen Ruggles - Diversity & Inclusion Working Group Representative
Ken Museth - OpenVDB Representative
Nick Porcino - Universal Scene Description Working Group Representative

Industry Representatives

Jean-Francois Panisset - Visual Effects Society

Non-Voting Attendees

Non-Voting Project and Working Group Representatives

LF Staff

David Morin - Individual - No Account
Emily Olin - Academy Software Foundation
John Mertic - The Linux Foundation
Yarille Ortiz - The Linux Foundation

Other Attendees

Cottalango Leon - Sony Imageworks / Dailies Notes Assistant
Doug Walker - Autodesk / OCIO
Bill Ballew - Dreamworks
Cory Ormand - The Walt Disney Studios
Guillaume Brossard - Autodesk
Jim Helman - MovieLabs
JT Nelson - Pasadena Open Source consortium / SoCal Blender group
Olga Avramenko - Sony Imageworks / DNA
Robin Rowe
Sam Richards - ORI
Tony Micilotta - Autodesk
Lee Kerley - Apple

Antitrust Policy Notice

Linux Foundation meetings involve participation by industry competitors, and it is the intention of the Linux Foundation to conduct all of its activities in accordance with applicable antitrust and competition laws. It is therefore extremely important that attendees adhere to meeting agendas, and be aware of, and not participate in, any activities that are prohibited under applicable US state, federal or foreign antitrust and competition laws.

Examples of types of actions that are prohibited at Linux Foundation meetings and in connection with Linux Foundation activities are described in the Linux Foundation Antitrust Policy available at linuxfoundation.org/antitrust-policy. If you have questions about these matters, please contact your company counsel, or if you are a member of the Linux Foundation, feel free to contact Andrew Updegrove of the firm of Gesmer Updegrove LLP, which provides legal counsel to the Linux Foundation.

Agenda

General Updates
- AGENDA TOPIC: AI code assistant policy #1195
- Dev Days 2026 #1288
- AGENDA TOPIC SLP 2026 #1306
Annual Review: Rez #492
Annual Review: Machine Learning WG #1029
Linux Workstations in VFX - Wayland #1317

Notes

General Updates
- AGENDA TOPIC: AI code assistant policy #1195
  - Carol: have been talking about TAC level AI code contribution policy. Posted link to #tac channel, please read it and comment. Projects should also have their project level policies. Link to doc
- Dev Days 2026 #1288
  - Carol: coming up in a couple of weeks, make sure everything for your project is up to date.
- AGENDA TOPIC SLP 2026 #1306
  - Carol: If you have volunteered to review applications, please do so. You should have received an email about it. Deadline is May 2nd. Also if you signed up for any other SLP volunteer role, watch out for email. Also if you still want to apply as a mentor, please DM Carol.
- Linux Workstations in VFX - Wayland #1317
- Carol: WG approved! Next steps coming, it’s moving forward.
Annual Review: Rez #492
- Slide Set
- Stephen Mackenzie / Jean-Christophe Morin
- Used new in-development template + current one
- No Major change to overall slide, but made progress on OpenSFF Best Practice Badge, went from “nothing” to 85% on Silver, 57% on Gold
- Roadmap
  - Blank! We’ve been historically bad on roadmap. A lot of our time is spent on Python, dependencies.
  - We have some ideas… required for graduation!
- Releases
  - Going with larger releases
  - rez-3.3.0
    - Features
      - Shifted Python support from 3.7+<3.12 to 3.8+<3.14
      - wildcard support in rez-test
      - gitbash compatibility improvement
      - logging improvements in payload caching
      - rez-plugin registration using entrypoints
      - new settings for payload caching disk usage
    - Fixes
      - platform compatibilities (slashes, case sensitivities, exit codes)
      - race conditions
      - large-scale updates to vendored dependencies
    - Docs improvements
      - security policy, connected to GitHub security infrastructure, haven’t gotten security bulletins yet, even though we vendor some projects
      - rez-bind deprecation warnings
      - improved plugin and .rezconfig.py docs
- Graduation Requirements Report Card
  - Trying to get it over the finish line!
  - Mising:
    - Defined Mission & Scope
    - Roadmap
      - Not yet itemized in clean form
    - Growth Assessment
    - 85% silver, 57% gold OpenSSF badge
    - Release signing almost done, a few hours of work left
- LFX Insights
  - “Healthy” project
  - Lots of stars, lots of forks
  - Contributions outside work hours: 67%
- Contribution Dashboard (Individual)
  - Jean-Christophe Morin - 50%
  - Stephen Mackenzie - 10%
  - Lots of small / one time fixes
  - A lot from smaller studios trying Rez, experimenting, or maybe not understanding something and contributing a docs fixes.
  - People who worked on Rez at one studio, then went to another studio, then get to a studio which uses Rez, they start contributing again.
  - 2-3 years ago was mostly TSC contributors, nice to see higher variety
- Contribution Orgs
  - 3 most reliable orgs:
    - Eyeline - working on a large PR
    - Laika
    - Dreamworks
  - but we’re not the focus of many orgs
- ADOPTERS.md
  - No one adds themselves to the list, but look at the Slack channel, ping them and ask if they can be added. Needs to be updated since last year. Many new people, but not adding themselves to ADOPTERS.md. I don’t like to remove defunct studios, but maybe split into active / inactive.
- VFX Industry Build Matrix updates
  - rez is pure python so almost nothing in the sheet applies to its own operations
  - Shifted Python support from 3.7+<3.12 to 3.8+<3.14 at rez version 3.3.0
  - Found cmake-4 incompatibility to be addressed
  - Old-school rez-gui probably needs a Qt lift, many are unaware it even exists. Personally hoping to replace with a dedicated / separate project.
- Engineering Time Contributions
  - No changes : we don’t get any
  - rez asserts there are zero ASWF member orgs that dedicate 20% of their work time towards the project.
- CI / Security / Marketing Updates
  - Type annotations PR will hopefully enable mypy usage in CI
  - Signing releases / tags PR probably just needs a few more hours of work
  - JCM made rez-pip2 into the actual first project on PyPI to create / upload SLSA attestations
  - Small but consistent improvements to docs
  - We have an active security policy now, and the correct GitHub/ASWF-email things wired up to receive vuln notices
  - Seeking “dynamic analysis tool” for Python tools to fulfill SSF badge requirement
  - Had conversations with a group from NYU that found us, offering security analysis assistance that found us through CLOtributor, but it fizzled out
  - TSC Meetings have had intermittent cancellations due to TSC availability / burnout at times, but still continue
- AI / ML Code Generation Use /Reporting
  - 1 or 2 docs PRs, CoPilot generated
  - Looking for small / reviewable contributions
- Non-Code Aspirational LLM usage scenarios
  - Scrape our public Slack for…
    - docs improvement opportunities
    - common FAQs to add to FAQ docs
    - … or to write more extensive usage scenarios
    - finding “we should do…” notes to assemble improvement lists, generate GitHub issues, etc
    - develop referenceable “studio strategies” that get circulated repeatedly
      - packaging tips
      - OS-specific advice
  - Collating related GitHub issues for combined solution design
  - Generating release notes
  - Remove meeting-notes / summarization / action-item-tracking needs from maintainers
  - Analysis of project inconsistencies (e.g. logging, path handling)
  - Security analysis / threat modeling
- Areas the project could use help on
  - Replacement TSC member
  - Better code reviewer SLA
  - More member company engineering time
  - Test coverage / project graduation
  - JCM’s rez-pip2 is seemingly mature enough that it likely makes sense to migrate soon (flagged last year)
- Feedback on working with ASWF
  - Rez is given the best chance under ASWF
  - When we ask for help, we generally get it
  - But could we ask for more contribution from FTE equivalent?
  - Looking to leverage LLM developments
  - But generally working under ASWF has been great!
- TAC Open Discussion
  - Carol: around your test needs, have you create Good First Issues for Dev Days? Your project is Python, this could be doable tasks? Stephen: the quirk around Rez and testing is that the testing is around platform and shell testing, lots of weirdness around cmd.exe / powershell.exe, weirdness is around command line interaction. But there are areas of Rez that can be better tested. Of course there’s interaction between writing the test and writing the code. LLMs could help identify what needs to be better tested.
  - Gary (chat): I’ve been working on a new build tool called “pcons” inspired by SCons and CMake; there are probably ways I should be integrating with rez. Stephen: We take build systems shims! But we also moved towards supporting arbitrary build_commands and that handles 99.9999% of cases
  - Carol: feel free to use whatever slide template you prefer to use. Great progress on the OpenSSF badging, halfway to gold is great!
Annual Review: Machine Learning WG #1029
- Slide Set
- Tommy Burnette, ILM
- What is the ML WG?
  - Goals
    - Incubate concrete open source projects within ASWF
    - Establish a community to share information and expertise within the film industry
    - Lead vision, terminology, norms, and technology related to ML
  - Non goals
    - Avoid doomerism and boosterism
    - Avoid getting involved with issues or politics outside of the film industry
    - Do not take creative control or agency away from artists. Empower, not replace.
- Who are the ML WG?
  - Acting Chair: Tommy Burnette - ILM
  - Key Participants
    - James Spadfora - ILM
    - Dmitry Grankin - Vexa.AI
    - Cameron Target - Disney
    - Tommy Snyder - Autodesk
    - JT Nelson - PasOS/SCB
    - Sam Richards
    - Cottalongo Leon - Imageworks
    - Larry Gritz - Imageworks
- Agenda
  - What have we done?
  - How did we do?
  - Waht comes next?
  - What do you have to add?
- What have we done?
  - Born from conversations at OSF last year. Larry wrote initial proposal on April 28th 2025, wanted one active project by SIGGRAPH. Initial idea was Digital Notetaking Assistant (DNA). Presented to TAC on May 14th. RMTC presented on Jun 25
  - DNA was a good first project, automate some labor, some concerns from production folks. That project “sucked all the oxygen” from the larger project. Wanted to refocus on larger goals, so we split off DNA into its own group / TSC.
  - Proposal emerged to re-invigorate WG into “mutual learning sessions”
- How did we do?
  - Goals
    - Incubate concrete open source projects within ASWF
    - Establish a community to share information and expertise within the film industry
    - Lead vision, terminology, norms, and technology related to ML -> not done yet
- What comes next?
  - WG-ML Mutual Learning Sessions
  - A series of learning sessions around topics related to ML
- Accelerating ML adoption is creating siloed exploration
  - As machine learning and AI adoption accelerates across the industry…
- Connecting practitioners to build collaborative pathways
- Four guiding principles for our sessions
  - Mutual learning
  - …
- Strict boundaries on intellectual property and sales
  - No confidential information
  - No commercial product pitches
- Applying ML to content creation workflows
- Leveraging LLMs for pipeline and productivity
- Infrastructure trade-offs and the scrapheap
  - Production deployment
  - Infrastructure trade-offs
  - …
- A low-friction pipeline for speaker selection
  - Call for speakers
  - Attract submission
  - Leadership curation
- Moving from shared knowledge to open-source tools
  - Phase 1: Knowledge sharing
  - Phase 2: Networking
  - Phase 3: …
- What do you have to add?
  - Carol: WGs evolve over time, which is good.
  - Carol: should share the document that was shared with #tac to the #wg-ml group, will do that.
  - Robin: is there a call for speaker year? Tommy: not yet, but soon. Don’t have a timeline, but we will get that under way. Carol: also SIGGRAPH will be a good time for that.
David: The WG on ML is tackling what we should be tackling, but would be interesting to hear the guidance / non guidance you get inside companies. Can anyone comment on this?
- Carol: they are just tools, depends on what you are willing to pay for? Definitely as an individual. “You get what you pay for”, but you can still get a lot done with a lot of the open source models. David: recommendation from LF execs were about the cost, don’t go straight to Claude to fix all your problems, but first Google Gemini which was less costly for them. There is a lot of information in practical sessions about what models to use and how.
- Stephen: “try everything!”. I’ve experimented with different models, sometimes it refuses to answer in anything but Markdown. There are very specific use cases, and likely going to be a lot of fragmentation. I’ve seen a lot about how tools are used with LLMs, but don’t see much about the machine vision stuff, which is particularly interesting to our field. This group should have experience with that?
- David: it can be a touchy subject! Carol: not sure it’s touchy, but it’s big. Think about how does this impact the ASWF, where do we focus our time and energies. We are talking about “what tools make sense where” within projects. How we run our dev environments hasn’t bubbled up to the top. David: we talk a lot about incoming code that’s AI generated, but are there best practices on how to use the models to generate code. It can be considered internal practices that are to be kept private within a company, but it would be great to have a workshop where someone talks about how they are using those tools for code generation. Carol: would have to keep that limited to open source tools. Gary: I’m mostly retired, writing entirely open source software, I’m using Claude almost all the time, Opus 4.7. Working through my lifetime backlog of projects I wished I had time to work on while I was working, and 10x ing them all. Opus 4.7 seems to be the way to go. I occasionally max out my $100 subscription but it’s worth it. Will be different in large companies where you have restrictions. But for open source, in one project I’m identifying all commits as co-authored by Claude, but not in another project, since there are no best practices established so far.
- David: perhaps your learnings could be put together into a small workshop? Gary: I would be happy to, count me in if that happens. Carol: yes, a good perspective, a lot of us on this call may be more limited about what we can say on this topic. Gary: happy to contribute, lots of ideas / thoughts to share.
- JC (chat): We can get access to Copilot through LF, so technically we can use non-free models, if it helps for discussions. Stephen (chat): Similar experience for home/hobby projects. Had a wishlist project for 10+ years. Prototyped it last weekend. Slay that bucket list.

Next Meeting Agenda

General Updates
- 2026 Security Reviews #1137
- TAC Industry Representative Election #1291
Revamp Annual Review Structure/Template #1336
2026 TAC Priorities #1208