Link Search Menu Expand Document

ASWF TAC Meeting - December 16, 2020

Video Conference Link

Voting member attendance

  • Daniel Heckenberg - Chairperson, Animal Logic Pty Ltd
  • Gordon Bradley, Autodesk
  • Michael O’Gorman, Cisco Systems Inc.
  • Henry Vera, DNEG
  • Bill Ballew, DreamWorks Animation
  • Matt Kuhlenschmidt, Epic Games, Inc.
  • Brian Cipriano, Google & OpenCue Representative
  • Sean C McDuffee, Intel Corporation
  • Larry Gritz, Sony Pictures Imageworks
  • Jean-Francois Panisset, VES Technology Committee
  • Cory Omand, The Walt Disney Studios
  • Kimball Thurston, Weta Digital Limited
  • Eric Enderton, NVIDIA
  • Sean Looper, Amazon Web Services
  • Michael Min, Netflix
  • Michael B. Johnson, Apple
  • Dave Fellows, Microsoft
  • Sean O’Connell, Advanced Micro Devices
  • Mark Visser, Unity Technologies
  • Ken Museth, OpenVDB Representative
  • Michael Dolan, OpenColorIO Representative
  • Cary Phillips, OpenEXR Representative
  • Joshua Minor, OpenTimelineIO Representative
  • Chris Kulla, Open Shading Language Representative

Other Attendees

  • John Mertic, Linux Foundation
  • Emily Olin, Linux Foundation
  • David Morin, ASWF
  • Carol Payne, Netflix, ASWF D&I WG
  • Rachel Rose, ILM, ASWF D&I WG
  • JT Nelson, Pasadena Open Source consortium / SoCal Blender group
  • Nick Porcino, Pixar
  • Alex Meddick, Rising Sun Pictures
  • Mathieu Mazerolle, Foundry
  • Deke Kincaid, Digital Domain
  • Ashley Whetter, Python 3 WG
  • Michael Kessler, Epic Games
  • Robin Rowe, Cinepaint

Apologies

  • Joshua Minor, OpenTimelineIO Representative

Agenda

  • Goals for TAC: Year 2

    • Good success on the CI front, more work to do on the TAC side

    • Lots of activity in the WGs

    • Progress on the D&I front

  • Governing Board Update (David)

    • Last Governing Board meeting yesterday, where we review the goals for the year ahead (2021), get alignment for our goals

    • Alignment with Calendar Year (2021), including memberships

    • Timeline of ASWF history. 2 main phases: original goal of 6 projects in the first year (took a bit longer), reached $1M financing. Phase 2 was FY2020, got disrupted by COVID-19, spent most of phase 2 dealing that in various ways. Extending memberships for certain members hit hard by the situation. TAC continued working on projects and delivering on the mission. Current events highlighted need for D&I efforts.

    • Entering phase 3:

      • Funding

      • Projects

      • Outreach

    • Funding: started year at $1M run rate, but at time of renewal, Animal Logic downgrade to General Member, Blue Sky and Cisco cancellations. Back down to just below $1M.

    • COVID set us back a year in our growth plan, resuming goal of $1.3M by end of 2021. Target approved by the board.

    • 2021 member company targets: 10 companies targeted as potential premier members, 30 companies targeted as general members. More focus on getting companies to join as general members.

    • Develop Value Proposition for general members based on the experience and feedback of our existing general members:

      • Why join

      • Specifics on benefits: how does this work, testimonials

      • Cost structure

      • Membership application

      • Jordan Soles from Rodeo working on this

    • Use this to seek general membership from smaller animation / vfx companies, also software / hardware vendors that might want to join

    • 2 events: ASWF Forum in March, Open Source Days in August

    • Working with the TAC, aim to bring 3 new projects in the foundation in FY 2021, we have a harmonious set of projects, but don’t want to stick with the status quo (3 new projects showed on the timeline)

    • Looking at potential new projects: ACES, Meshroom, Rez, USD… (not an exhaustive list)

    • Develop value proposition for potential projects:

      • Specifics on benefits

      • Why transfer your project

      • How would this work

      • How to get alignment internally

      • How to deal with typical pushback

      • Help to fill the proposal template

    • Invest in D&I for the long term, and promote software engineers in our industry

    • Q: how are the funding targets assessed? Initial $1M was an “ambitious goal” what we managed to reach, new target of $1.3M is a reasonable target we can achieve, underlying goal is to ensure the foundation is well funded, fund our current activities, and funding for new initiatives. There are lots of committees that put together projects and initiatives but have no money to work with, so they aren’t as effective as they would want to be. A chief goal of creating the ASWF was to have the capabilities to reach our goals. Get to a level which gives us a solid base that we can act on and take action. Michael Min: is there an assessment of what these funds would be used for? A: put the money at the feet of the TAC, the “heart” of the foundation. Some funds are there to cover our operations (for instance services from LF), surplus is there for projects such as D&I (budget was provided for activities). Also any projects require funds (for instance for-pay services for CI infrastructure). The funds are there for the TAC. Michael: we need to tell potential members what the funds would be used for. Daniel: we’ve also looked at grants, or our own equivalent of GSoC.

    • Q: we aren’t significantly spending our existing funds, so why are we growing the fiscal target? David: the ASWF wants to grow the software engineering capabilities and projects common across the industry, want to create a strong platform across the industry, which exists in other industries where open source is playing a major role in developing platforms. We have the potential to do this in our industry: funds are needed for this, but mostly talent. Growing and supporting the TAC, giving the tools to grow the platform, need to have enough money to do so. Push to increase available funds, want to make sure we have all the funds to support any initiatives that come in the future. Want to be funded in the open source world, bringing in new members is how we grow this. Henry: not sure that having more money is the goal to have, we should identify strategic goals, and figure out how much money we need to achieve those goals. We haven’t clearly established yet what we’re using the funds. We need to be clear about what we’re trying to establish. David: the mission of ASWF is well understood, grow the use of OSS in the industry. The presentation to the board implied that context. Approval of projects comes from the TAC, the board is there to support the TAC. Henry: still not very clear. David: it’s up to the TAC to decide. For instance if we wanted to bring USD / ACES / … into the ASWF, and looking at the amount of work, may need to bring in for pay work for instance?

    • Larry: we had good success with GSoC, 3 projects who got students got a lot of value out of it, so we may want to fund our expansion of these types of projects. But yes, no one has ever told the TAC “here’s how much money you have to spend”? Would want to see concrete examples of budget items. Michael Min: could make sense to go through a capex cycle for each project to figure out what the monetary needs could be. Agree with Henry, would help to have clarity. Would want to be part of the conversation, and know what we should be budgeting for. David: these are good points, we need to get there, but should know that the money is being gather for TAC and project activities, and that there are funds available for projects (primary) and WGs (secondary, although D&I is a priority). How much money? David to provide as an action item. If there is something that costs money and would be good to do, it is possible.

    • Henry: what is the value of the foundation to premiere members vs general members? Would it make more money to focus on making the entry point for members lower to keep more premier members, would this not have more value to the foundation overall. It’s a stretch for some companies to keep their premier level. David: the value for premier members is pretty clear: a seat on the Governing Board. We did lose some premier members in this COVID year, but we didn’t too poorly (AMD, Unity), there’s been a lot of expectation of general members would be part of a network of companies. We haven’t delivered that for them, some of the companies have voiced their dissatisfaction with that. Would rather move the value for general members up, especially for smaller / medium sized studios. It’s important to diversify our membership, go beyond the smaller number of large studios we currently have, so this is why we’re focussing on general membership.

    • Henry: What is the value to the Foundation of the premier vs general membership? David: it’s always evolving, we’ve set the rules and the levels of engagement, not sure we could have lowered the premier commitment level to keep AL at premier level, they are at the general level and have a mechanism to keep valuable commitments (Daniel). We are in good shape in terms of our structure, would be a huge effort to revise membership rules and membership regulations. There was also some reservations as to the current makeup of ASWF projects. There will be downgrades and upgrades over time as to where companies stand wrt ASWF. Questions are very valid, can continue discussion, either in TAC forum or one on one. We are a young organization, and we can adapt over time.

  • TAC chair election (John)

    • Opened nominations for TAC chair, but haven’t received any nominations. So looking for someone interested in taking on this role, either for the full year basis, or on a temporary role that could be re-evaluated during the year. Chair role can seem daunting, and it is not uncommon to have trouble filling this role. Chair role is important, both organization TAC activities and interfacing with the board. LF tries to keep the time commitment to a minimum, but crucial role nonetheless.

    • Robin: willing to help. John: TAC chairperson role is require to be a voting member of the TAC, so needs to be a premier member or project representative.

    • Daniel: it has been a very rewarding activity. Commitment depends on the level of activity, generally speaking it has trended down as projects and WGs have been self managing. Currently about 5 hours a week commitment.

    • Kimball: nominate Larry? Larry: nominate Kimball? Larry: currently running 2 widely used open source projects, so not sure has enough bandwidth, this is a big year for OSL, getting GPU support going. Might be interested in another year. But would support Kimball. Daniel: second Kimball.

    • John: governing board representation has to be a single person, but for the TAC, could have a co-chair, but one of these would be the representation to the board (could swap back and forth).

    • Does TAC chair count against FTE commitment of a premier member? David: will soon start to keep track of FTE requirement, in first couple of years there was no strict tracking, but will start looking more closely. TAC chair not currently considered, but could be. David looks for a partner in the TAC chair, discussions on developing the strategy, there has to be a dialog between TAC and Governing Board. Role is collaborative.

    • Do we need to formalize this with an endorsement vote? Yes, we have quorum.

    • Daniel: nominate Kimball. Larry seconds. Any opposes or abstains? No, unanimous. Congratulations and thanks to Kimball!

  • Project proposal and discussion: rawtoaces (Kimball)

    • Bring raw2aces under the ASWF, revival a tool that people are continuing to ask for, don’t want to duplicate effort at WETA. Path towards what to do with ACES, software + data warehouse of camera spectral sensitivity characterization. How to host ACES test images.

    • Effort on how to store lens / camera metadata in OTIO, so collaboration.

    • Kimball: some other data would make sense to store in there.

    • We had set a couple of weeks to think of questions, why should we do or not do this. Should we move this to the first meeting of next year?

    • Henry: very interested at doing this at DNeg

    • Where else could this happen: Kimball: could happen on the WETA GitHub

    • Daniel: it looks like a strong project proposal. Safe harbor for contribution of data that is otherwise proprietary / sensitive / hard to share, and good place to collaborate with ACES.

    • Larry: if this project existed, it would be of great use to the members of ASWF

    • Larry: move to adopt. (seconded). Daniel: call for vote for adoption of raw2aces as an incubation stage project. No against, no abstain. Adopted! Raw2aces will be an incubation stage ASWF project.

  • ASWF Project and Community Survey (John / Emily)

    • Emily: putting into Survey Monkey, should have draft by EOD / early tomorrow for review. There are a few sections that could still use drop-down selections, OpenVDB still need to be pared down (Ken to look at it).

    • Daniel: want to make sure this doesn’t get buried in the holidays, should be in a form ready to release this week, can then decide when to push it out. Any guidance as to when we should release it? Emily: it can be tricky with the holidays, we don’t necessarily want to wait for the new year, but may want to keep submissions open until end of January so we can do a second push after the holidays. Daniel: we can keep the window open.

  • Follow up: Rust working group? (Daniel)

    • Reached out to Rust Slack channel on the creation of a WG (today), some positive support. Should be possible to find the right person to set bridges (Nick attending this call).
  • Working Groups

    • CI

      • Sony Package Manager SPK presentation: recording

      • Larry: SPI is looking if there is enough interest from a user and co-developer perspective to turn it into a ASWF project. If there is demand, will try to push for it.

    • Diversity & Inclusion

      • Carol: 2 things coming up, webinar series, partnership between ASWF D&I and universities, aimed at college age students, introduction to open source, the industry and careers. Information in Slack channel and calendars. Also will start contacting project chairs for the D&I Ambassador program for project, with a connection back into the D&I WG.

      • Presented a proposed budget to the board based on planned activities. Difference between internship vs mentorship.

    • Python 3

      • Ashley: working on creating a transition guide based on the WDA transition guide, making sure it is applicable globally across studios large and small. Also work coming in from SideFX around C++ bindings, documentation and guidelines around transitioning to Python 3 when working in raw C.
    • USD

      • Still pursuing use case examples from industry partners

      • Looking at a standard camera rig schema that could be transported between platforms, game engines. Not the same as the implied camera on the USD stage, would be a functional camera. Wave has a working example.

      • Interest as using USD as an asset standard, the camera as a “character”.

      • Daniel: using schema mechanism, but making choices as to how to encode things? Michael Min: yes, using variants scheme, want to avoid having to reinvent the wheel in each DCC, avoiding having to remap all the operations on the camera rig.

    • Review & Approval

      • Larry: attended meeting 2 weeks ago, 2 in depth demos of representative demos, 2 more this week. When we resume in January, there will be more demos. Trying to define problem domain, a deliverable has not been decided on yet.
    • Asset Repository

      • Looking to server the needs of projects first, OpenEXR is the candidate project that already has a data set but seeks to extend / modernize it. Eric: is it for at-scale testing of our existing projects, and other vendors/open source projects, or other purposes? Some discussion about this on the Slack channel. Will discuss further when Wave is back at next meeting.
  • Technical Project updates

  • Update on candidate projects

  • Dependency/version guidance beyond VFX Platform?

  • Next meeting: December 30, 2020

Action Items (AIs)

Notes

Chat