AWSF TAC Meeting - October 23, 2019
Voting member attendance
- Daniel Heckenberg - Chairperson, Animal Logic Pty Ltd
- Gordon Bradley, Autodesk
- Pilar Molina Lopez, Blue Sky Studios, Inc.
- Michael O’Gorman, Cisco Systems Inc.
- Henry Vera, Double Negative
- Bill Ballew, DreamWorks Animation
- Matt Kuhlenschmidt, Epic Games, Inc.
- Brian Cipriano, Google & OpenCue Representative
- Sean 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
- Ken Museth, OpenVDB Representative
- Michael Dolan, OpenColorIO Representative
- Cary Phillips, OpenEXR Representative
- Joshua Minor, OpenTimelineIO Representative
Other Attendees
- Andrew Grimberg (Linux Foundation Release Engineering)
- John Mertic (Linux Foundation)
- Greg Gewickey (Warner Brothers)
Apologies
Agenda
-
Goals for TAC: Year 2 [0:00-0:05]
-
Governing Board update
-
David Morin shared rough goals for the Foundation, will be shared soon. Along similar lines as first year: expand membership base, financial stability, Premier Members committed to 2 year tenure, need to make sure ASWF delivers good value.
-
TAC goals: appetite for expanding project base, TAC is expected to contribute to this process.
-
-
Survey process (Gordon)
-
https://docs.google.com/document/d/1oYs8RABGS7qFJHvICzj5Jln83MRAZ855WC0qMWq8Vqw/edit
-
Last call for input from TAC: end of week to finalize input. At which point send to Governing Board for their additions, then send out to membership.
-
Any feedback is welcome, but need to make sure we gather actionable data.
-
Form version from John Mertic: https://docs.google.com/forms/d/e/1FAIpQLSe5MBSTs4SZBABSMQiTu-G_WSnACKGpxVOhPXZgXQUNGnjCSg/formrestricted but not currently accessible
-
Daniel: would seem “strange” to not have a list of proposed projects. Larry: if we ask anything about projects they would want “in”, it’s not useful unless they also specify “why” the project should be part of ASWF. Gordon: nice benefit of projects is that their name is unambiguous. So open question is better than a list, doesn’t imply endorsement or arm twisting. Henry: advantage of an actual list is that you can identify trends over time. Likes Larry’s suggestion that there should be a justification as to why someone would want to see a project join ASWF. Eric: isn’t the pitch for ASWF that we make projects more stable / reliable / longer term viable? Cary: agrees with Larry, there must be a realizable benefit to taking the project out of its current state and putting it into ASWF. John: related to this it would be a good idea to communicate how ASWF membership has improved the projects that have joined.
-
Kimball: also ties in to dependency management, and how many of our dependencies live within the ASWF.
-
Michael: is there any movement towards publishing the intent and results of the survey. Should have a statement of what will be done with the data, will it be posted publicly? Gordon: will we get less candid feedback if the survey results are posted? Daniel: it should be important to state how the results will be posted. Had discussion with Emily, could be good basis for annual report.
-
Gordon: added placeholder for the survey intent, and for open list of projects.
-
-
-
Goals for TAC to end 2019 [0:05-0:15]
-
Almost at November, so we have only a few weeks before end of 2019.
-
Project graduations
-
Diversity & Inclusion (wave)
-
Wave has been reaching out to a number of organizations, Daniel has been setting up calls
-
No specific info to report back for now
-
Looking for organizations to partner with, looking for specific information that could be applied to improving our own processes
-
A lot of interest in this aspect in the Outreach Committee and Governing Board
-
-
Documentation
-
Mission statements TAC list
-
Proposal from OpenEXR with discussion
-
Cary: was present at board meeting where Rob put out this request.
-
Discussed at TSC, draft by Rob Bogard and Cary, sent draft to Rob Bredow. He gave feedback that it might concentrate more on current state, feedback was incorporated.
-
What is the real purpose behind the projects, what is the scope (“guardrails”).
-
Tried to add to the home page, but hard to discuss in isolation of the rest of the information. At the top of the page there will be a description of what the project does, mission statement could be right below that.
-
Cary: not completely satisfied with the end result, OpenEXR web site itself is up for a major refresh.
-
Asking for thoughts about how this information should be presented on project page
-
Eric: would consider putting the Mission Statement as a separate page linked to by the main menu. Guides what you do and don’t do, different purpose of what you would put on the home page. Cary: also considered putting on a separate page, but feels that Rob thinks that this should be information that is “hard to miss”, and moving that to a separate page might not achieve this goal.
-
Cary: ideally a Mission Statement should be one or two sentences, but what OpenEXR currently has is a lot more verbose than that. So probably need to go back and try to prune it down.
-
Daniel: “should be tweetable”, also if we stick to the constraint of a short mission statement it makes it easier to put it in different places.
-
Cary: 1-2 sentence mission statement summary could link to more detailed page.
-
Larry: the two paragraphs proposed are a great summary of what the project is about, and do belong at the “top of the page”. Everybody’s README.md should start off with a 1-2 paragraph of what the project is about. Boiling down to the ‘140 character’ version could be a separate task.
-
Cary: write the mission statement, be thorough and comprehensive, but make sure that if you only read the beginning you still get the gist of it, and you can lift the first 1-2 sentences to provide a summary.
-
Michael: that can be what you add to other ASWF literature that points to the project.
-
-
Template project (JF)
-
Should the more detailed info go into the Wiki section?
-
John: could work on a better organization, offering to help
-
-
TAC Project
-
Some outstanding PRs that have gone “quiet”
-
Create Project Success metrics
-
Commit numbers?
-
Diversity of contributions?
-
Frequency / quality of releases?
-
Project usage?
-
-
Daniel: are we looking for feedback from TAC?
-
Should we have a “mission statement” for the tac repo? Who owns it, who is responsible for it?
-
John: typically on other foundations those repos are owned by multiple members of the TAC.
-
John: PRs against the tac repo could generate emails to the tac mailing list? Daniel: this is the first time that PRs have gone idle, so we may not really need more process. For now we can use GitHub mechanisms for drawing attention to PRs.
-
-
-
VFXPlatform 2020
- Python 3
-
CI Progress
-
-
Vote: Renew annual term of VES Technical Committee Industry Representative, JF Panisset
-
One of 3 Industry Representatives on the TAC
-
Informal approval from VES Tech Committee
-
Daniel moves to renew JF Panisset as Industry Representative
-
Seconded by Kimball
-
All ayes, no against, no abstain
-
Unanimous renewal
-
-
Technical Project updates [0:15-0:30]
-
OpenTimelineIO
-
Apache license discussion ongoing.
- No update from last meeting
-
Some meetings with Avid coming soon.
-
Meeting with 2 separate Avid people next week, and later discussion in November.
-
Approval to share some of the technical documentation they provided
-
-
Working with TSC on proposed schema improvements.
-
-
OpenVDB
-
OpenCue
- Dave Fellows: no specific updates from 2 weeks ago. Some security-related items have been noted requiring attention.
-
OpenColorIO
-
Michael: working on Python 3, discussion on GPU CI infrastructure, should have more updates soon.
-
Waiting for Andy to work with Sean to work out arrangement with AWS, and LF RelEng.
-
Wave doing investigation at Apple as to how we could do CI on GPU-enabled Mac platform.
-
Andy: recovering (email!) from leave, will be available to start working on these tasks starting tomorrow.
-
-
OpenEXR
-
Security feedback
-
Had call with Dan Hutchison from Foundry, would be a good idea to invite him to other / wider forums.
-
He seemed to agree that the steps taken in OpenEXR are the right way to go. Requirements around reporting and updating public information on reported security vulnerabilities. You post a fix, then make sure the CVE report adds the link back to the fix.
-
Kimball: make sure you have two factor auth on your GitHub account
-
Daniel: we should try to capture this in a way that other projects can use (start with email).
-
-
Cary: adoption proposal, outlines stats about project activity, CII badge process
-
License scan approved by board yesterday, so should be ready for graduation.
-
Finally set up with EasyCLA system, ready to move repo to ASWF GitHub organization. It seems to be a bit backward to move the project into the GitHub organization before full acceptance?
-
Larry had posted message about getting into cadence of patch releases, so discussion of branching strategy to support that.
-
Constant thorn in their side is Boost / Boost-Python, so looking at replacing Boost-Python with PyBind11 Not a simple project at all.
-
Getting steady stream of build issue reports, especially on Windows. Doing a reasonable job of keeping up with those reports.
-
Should be more pro-active to reaching out to distro maintainers, pushing to update to more recent versions. Some distros still include version 1.7, important to update given security-related fixes.
-
VFX Reference Platform looking at moving to 2.4.0 for Reference Platform 2020.
-
Going with 2.4.0 instead of 2.3.1 is due to the additional of an API function.
-
-
-
CI updates [0:30-0:40]
-
Access / Roles / Workflow for updates
-
Some frustration from projects
-
Possibly best handled in an email discussion
-
-
Azure best practices
-
GPU agents
-
-
Update on candidate projects [0:40-0:50]
- Media player
-
Next meeting
- 6 November 2019