Topics

Open source applications as a reference platform

Jason van Gumster
 

Hi there!

After attending the BoF presentation at SIGGRAPH, it seems like a lead starting point for project inclusion is mostly on the library side of things (as opposed to the application side). That makes sense. But maybe it would be wise to coordinate with some of the larger open source creative applications (Blender, Krita, Inkscape, GIMP) to establish them as an effective reference platform for these library projects. It seems to me that the primary need to pull that off would be to have someone in the ASWF side to coordinate communication with these project teams.

So what do folks think about this? Any gotchas or impediments that would make this unfeasible?

jordanbrandes@...
 

Seems like a good idea to me. They are there so might as well use them.

Bryce Harrington
 

On Tue, Aug 14, 2018 at 11:10:17AM -0700, Jason van Gumster wrote:
Hi there!

After attending the BoF presentation at SIGGRAPH, it seems like a lead starting point for project inclusion is mostly on the library side of things (as opposed to the application side). That makes sense. But maybe it would be wise to coordinate with some of the larger open source creative applications (Blender, Krita, Inkscape, GIMP) to establish them as an effective reference platform for these library projects. It seems to me that the primary need to pull that off would be to have someone in the ASWF side to coordinate communication with these project teams.

So what do folks think about this? Any gotchas or impediments that would make this unfeasible?
Hi, are the slides or video of the presentation available by chance?

(Sorry for jumping in on the thread, I only learned of this group
yesterday.)

Bryce

rob@...
 

I've attached the slides here as presented in the BOF as a PDF.

Enjoy,

Rob

Andrew Paxson
 

I definitely think that is a good idea, I imagine there is a ton we can learn from those projects. 

On Aug 15, 2018, at 1:18 PM, rob@... wrote:

I've attached the slides here as presented in the BOF as a PDF.

Enjoy,

Rob

<aswf_bof_081318.pdf>

Deke Kincaid
 

Many of the previous mentioned projects have had very little interest in being given any guidance or direction from the vfx/film industry.  Some actively went out of their way to dissuade any contributions from us and many do not like that our projects use BSD like licenses for better inclusion of commercial software and can be quite hard core on using gpl.  I question whether software not born from vfx/film industry are actually interested in this?

Michael Hall
 

When contributing to an Open Source project you have to build trust with the community around it before they will start to consider guidance and direction from you. It's an unfortunate fact that any sizable project will have a multitude of people who want to tell them how to build it, but who won't help them build it. To insulate themselves from this noise the maintainers most often require contributors to make some significant contributions based on their existing direction before giving them a voice in changing that direction.

If you want projects like Gimp, Inkscape, Krita or Blender to make big changes to meet your needs, start by contributing small changes that meet your (and their) needs. Fix bugs reported by others, implement some features already on their TODO list, add new file format support, etc. Once they know that you are committed to the future of the project, they will let you help guide and direct that future.

But if you come into an open source project with plans to up-end something they've dedicated years of their personal life to, you're not going to get a good reaction. It doesn't matter if you know what you're talking about, of if you have years of industry experience, or even if you're an expert in the thing you're proposing. If they don't trust that you're as committed to their project as they are, they're not going to let you steer.

Michael Hall
mhall119@...
On 08/16/2018 02:40 PM, Deke Kincaid wrote:

Many of the previous mentioned projects have had very little interest in being given any guidance or direction from the vfx/film industry.  Some actively went out of their way to dissuade any contributions from us and many do not like that our projects use BSD like licenses for better inclusion of commercial software and can be quite hard core on using gpl.  I question whether software not born from vfx/film industry are actually interested in this?

Scott Petrovic
 

The goals for the ASWF will eventually need to be more concrete to help clear some of the confusion going on now. I am having a difficult time trying to understand who ASWF is trying to assist with opening up to open source projects.

I am sure there are certain projects and libraries the ASWF has in mind at first. I imagine they are also open to other projects or ideas as this idea matures. If I was starting a new foundation and wanted to see the VFX open source community grow, I would take an inventory on where some open source projects are with questions like:
  • how many people are in their core team
  • how much funding do they currently receive to pay for things like hosting and developers
  • how do they manage their project (repositories, CI, web hosting, bandwidth for downloads)
  • how much time do they spend a month improving the their library/applciation
  • what are the biggest issues they see from keeping their project going or expanding it
I am sure many of the applications that are focused on VFX users would be open to discuss and potentially collaborate. Most of them have mailing lists and use IRC for real-time communication.

I am one of the developers for Krita and would be open for any discussion or brainstorming that might need to happen
Scott Petrovic 

Alex
 

I agree with Scott, I feel like if we are looking to grow vfx open source we should be thinking about the points he listed especially since we have some of that structure already, the CI suite that the Linux foundation setup for example. Smaller OSS projects, or larger core projects with limited developer time would benefit from the support of the ASWF I think. 

Larry Gritz
 

I'm not speaking officially for ASWF here, though I was involved in the planning so I think I have a good feel for what it's trying to do.

The org's priority is about helping the projects that are developed in, by, and for the film industry specifically. The three "test projects" for the CI infrastructure (OpenEXR, OpenVDB, OpenColorIO) are exemplary of the scope of what the org is trying to do -- all were developed inside VFX studios, are critical technologies used ubiquitously in VFX (both at the studios and embedded in all the major commercial software vendors making apps for VFX), have limited use outside the industry, and have very small and precarious development communities (who are also totally within the VFX industry and have day jobs that sometimes keep their attention away from the software). The industry *needs* to take proactive steps to make sure those packages don't starve for development resources or proper governance, because those projects failing will really, directly, hurt us. The top priority for the first few years and batches of projects, is to directly address issues in those kinds of projects.

A package with a large, active development community, existing momentum, and a center-of-gravity of users and mission that is outside professional film production, isn't really within the primary mission of the new org, and doesn't particularly need our help. Projects like Gimp or Blender (excellent though they may be) aren't the reason why ASWF was created. Not that they couldn't be involved, depending on how the projects and the org evolve over time, but definitely not among the critical priorities that need to be addressed right away.


On Aug 16, 2018, at 2:57 PM, Scott Petrovic <scottpetrovic@...> wrote:

The goals for the ASWF will eventually need to be more concrete to help clear some of the confusion going on now. I am having a difficult time trying to understand who ASWF is trying to assist with opening up to open source projects.

I am sure there are certain projects and libraries the ASWF has in mind at first. I imagine they are also open to other projects or ideas as this idea matures. If I was starting a new foundation and wanted to see the VFX open source community grow, I would take an inventory on where some open source projects are with questions like:
  • how many people are in their core team
  • how much funding do they currently receive to pay for things like hosting and developers
  • how do they manage their project (repositories, CI, web hosting, bandwidth for downloads)
  • how much time do they spend a month improving the their library/applciation
  • what are the biggest issues they see from keeping their project going or expanding it
I am sure many of the applications that are focused on VFX users would be open to discuss and potentially collaborate. Most of them have mailing lists and use IRC for real-time communication.

I am one of the developers for Krita and would be open for any discussion or brainstorming that might need to happen
Scott Petrovic 

--
Larry Gritz




Bryce Harrington
 

Thanks Rob!

Have you looked at gitlab's CI service? Curious what your thoughts
are on it, and how the AWSF CI service would compare.

Bryce

On Wed, Aug 15, 2018 at 01:18:56PM -0700, rob@... wrote:
I've attached the slides here as presented in the BOF as a PDF.

Enjoy,

Rob


Bryce Harrington
 

On Thu, Aug 16, 2018 at 02:57:59PM -0700, Scott Petrovic wrote:
The goals for the ASWF will eventually need to be more concrete to help clear some of the confusion going on now. I am having a difficult time trying to understand who ASWF is trying to assist with opening up to open source projects.

I am sure there are certain projects and libraries the ASWF has in mind at first. I imagine they are also open to other projects or ideas as this idea matures. If I was starting a new foundation and wanted to see the VFX open source community grow, I would take an inventory on where some open source projects are with questions like:

* how many people are in their core team
* how much funding do they currently receive to pay for things like hosting and developers
* how do they manage their project (repositories, CI, web hosting, bandwidth for downloads)
* how much time do they spend a month improving the their library/applciation
* what are the biggest issues they see from keeping their project going or expanding it
That last question is probably the most pertinent, and would have the
most interesting and useful answers from project to project, IMO. The
other questions in that list can sometimes be hard to answer
meaningfully due to the fluidity of volunteer community work.

I am sure many of the applications that are focused on VFX users would be open to discuss and potentially collaborate. Most of them have mailing lists and use IRC for real-time communication.

I am one of the developers for Krita and would be open for any discussion or brainstorming that might need to happen
Scott Petrovic
Good points raised in this email, I'm curious too. I'm involved in
a variety of FOSS graphics projects and would also be open to
discussions and brainstorming. I'd really love to see if this becomes a
big success for the marketplace and wider community.

Bryce


Andrew Paxson
 

Larry have you some thoughts on how we are going handle measure project resources?

I think I still have some confusion on what is actionable for the next year or so? Is our goal for the next year or so to use these great test projects to establish and verify CI pipes and infrastructure?

If if that is the case where are we at within those three projects?

Michael Dolan
 

Thanks Larry - you covered it very well. I'd like to add just one clarification around what stage this project is at in this point in the journey and what the initial next steps look like as things get started.

At this point we have a number of interested parties thinking about projects and we'll be working with the initial Technical Advisory Council to identify a lifecycle for projects and an intake process for new projects. This includes identifying projects that are interesting, having a template for a project proposal to be made, stages for projects that are part of ASWF and potentially even archiving of projects that are no longer maintained down the road. There are decisions to be made about accepting projects and we have a governance model designed to facilitate that process.

The process is the key next step. If I could plan a teaser out there, the very first TAC meeting will be next Wednesday (the details are below). The TAC is initially composed of one representative from the Premier members, but it's open for anyone to listen in. We will be crawl/walking for a while until everyone's ready to run, but please feel free to listen in while we're getting everything up and running. 

Thanks for your interest, 

Mike

---
Mike Dolan
VP of Strategic Programs
The Linux Foundation
mdolan@...
---

TAC Call Info - Wednesday August 23 at 10am Pacific Time

Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/620996733

Or iPhone one-tap :
    US: +16699006833,,620996733#  or +16465588656,,620996733# 
Or Telephone:
    Dial(for higher quality, dial a number based on your current location): 
        US: +1 669 900 6833  or +1 646 558 8656  or +1 877 369 0926 (Toll Free) or +1 855 880 1246 (Toll Free)
    Meeting ID: 620 996 733
    International numbers available: https://zoom.us/u/b4Xs727sD




On Fri, Aug 17, 2018 at 1:28 PM Andrew Paxson <andrew@...> wrote:
Larry have you some thoughts on how we are going handle measure project resources?

I think I still have some confusion on what is actionable for the next year or so? Is our goal for the next year or so to use these great test projects to establish and verify CI pipes and infrastructure?

If if that is the case where are we at within those three projects?

Scott Petrovic
 

Thanks for all the clarifications. That makes sense with the way you are describing it.  It would probably be a good idea to eventually update the marketing website aswf.io to do a better job communicating these points. Looking at the website now, I can see why a lot of people are confused. A few ideas for improvements are the following:

1. Communicate strongly the words VFX. I think just saying open source software is not specific enough.
2. This organization is focused on libraries and technologies that are very specific and internal to the VFX industry.  
3. Many of Larry's points are great and this spirit needs to be carried through to the website and who the audience is:
  •  Technology that is:
    • developed inside VFX studios
    • are critical technologies found everywhere in VFX applications
    • have limited use outside the VFX industry
    • have very small and precarious development communities

Michael Dolan
 

Sorry, my notice on the TAC meeting should have said Wednesday August 22. (I had mistakenly referenced the 23rd)

---
Mike Dolan
VP of Strategic Programs
The Linux Foundation
mdolan@...
---



On Fri, Aug 17, 2018 at 3:34 PM Michael Dolan <mdolan@...> wrote:
Thanks Larry - you covered it very well. I'd like to add just one clarification around what stage this project is at in this point in the journey and what the initial next steps look like as things get started.

At this point we have a number of interested parties thinking about projects and we'll be working with the initial Technical Advisory Council to identify a lifecycle for projects and an intake process for new projects. This includes identifying projects that are interesting, having a template for a project proposal to be made, stages for projects that are part of ASWF and potentially even archiving of projects that are no longer maintained down the road. There are decisions to be made about accepting projects and we have a governance model designed to facilitate that process.

The process is the key next step. If I could plan a teaser out there, the very first TAC meeting will be next Wednesday (the details are below). The TAC is initially composed of one representative from the Premier members, but it's open for anyone to listen in. We will be crawl/walking for a while until everyone's ready to run, but please feel free to listen in while we're getting everything up and running. 

Thanks for your interest, 

Mike

---
Mike Dolan
VP of Strategic Programs
The Linux Foundation
mdolan@...
---

TAC Call Info - Wednesday August 23 at 10am Pacific Time

Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/620996733

Or iPhone one-tap :
    US: +16699006833,,620996733#  or +16465588656,,620996733# 
Or Telephone:
    Dial(for higher quality, dial a number based on your current location): 
        US: +1 669 900 6833  or +1 646 558 8656  or +1 877 369 0926 (Toll Free) or +1 855 880 1246 (Toll Free)
    Meeting ID: 620 996 733
    International numbers available: https://zoom.us/u/b4Xs727sD




On Fri, Aug 17, 2018 at 1:28 PM Andrew Paxson <andrew@...> wrote:
Larry have you some thoughts on how we are going handle measure project resources?

I think I still have some confusion on what is actionable for the next year or so? Is our goal for the next year or so to use these great test projects to establish and verify CI pipes and infrastructure?

If if that is the case where are we at within those three projects?

Sean Wallitsch
 

While the initial projects are all VFX centric, I don't believe limiting to just the VFX industry is intended.


On Fri, Aug 17, 2018 at 2:09 PM Scott Petrovic <scottpetrovic@...> wrote:
Thanks for all the clarifications. That makes sense with the way you are describing it.  It would probably be a good idea to eventually update the marketing website aswf.io to do a better job communicating these points. Looking at the website now, I can see why a lot of people are confused. A few ideas for improvements are the following:

1. Communicate strongly the words VFX. I think just saying open source software is not specific enough.
2. This organization is focused on libraries and technologies that are very specific and internal to the VFX industry.  
3. Many of Larry's points are great and this spirit needs to be carried through to the website and who the audience is:
  •  Technology that is:
    • developed inside VFX studios
    • are critical technologies found everywhere in VFX applications
    • have limited use outside the VFX industry
    • have very small and precarious development communities

Larry Gritz
 

On Aug 17, 2018, at 2:09 PM, Scott Petrovic <scottpetrovic@...> wrote:

 Looking at the website now, I can see why a lot of people are confused. A few ideas for improvements are the following:

1. Communicate strongly the words VFX. I think just saying open source software is not specific enough.


I want to clarify that ASWF is not specific to VFX per se. But it is specific to the film industry.

I was sloppy by saying "VFX" so emphatically. It (including 3D feature animation) is the part of the film industry that is the most software-heavy and has been very active in open source. VFX/animation studios have created several particular open source projects that the entire section of the industry is utterly dependent upon. I'm not sure if there are any analogous cases in other film industry disciplines, but they certainly are not meant to be excluded if such cases come to light.

--
Larry Gritz