Description
Key Learnings
- Learn about the history behind various open standards and the production challenges they solve.
- Learn about the collaboration between tech partners and creative studios driving the development of key open standards.
- Dive into how key open standards have been integrated into industry tools, like Maya and Arnold.
- Learn how studios are using key open standards on real productions to streamline workflows.
Speakers_few
- Dawn FidrickMy career spans work in animation, VFX, live action feature, and short film as well as live theatrical production. The Hollywood Reporter called my producing debut feature Other Madnesses "a film that effectively gets under your skin". I have stage managed the legendary Blue Man Group at the Astor Place Theatre in New York City and contributed as a visual effects artist on Star Trek: Into Darkness, Pacific Rim, Iron Man 3 and Wolverine. I earned my Master of Fine Arts in Computer Art from School of Visual Arts and BFA in Production Management from DePaul University. I am thrilled to have been the Producer and VFX Producer on the new planetarium show Signs of Life at the Samuel Oschin Planetarium, Griffith Observatory. My production experience in live action, theatre, vfx and animation has made me a high performing leader. I have a deep understanding and working knowledge of physical production, animation, cg pipelines, and post production. I thrive in a collaborative space that cultivates creativity. Fundamentally, I am accepting of others, believe in the links between all things, take responsibility and have a great deal of stamina to lead teams and shows from concept to completion, on time and on budget.
ALIZA CARPIO: Hello, everyone. Good morning. Good afternoon for those of you, and even good evening, for those of you joining us from other parts of the world. I would love to welcome you to our session titled The Evolution of Open Source in Film and TV Production. My name is Aliza Carpio, and I'm Tech Evangelist at Autodesk. And just like you, I'm looking forward to hearing and learning from our panel today.
This discussion is all about the continuing and broadening impact of open source in the industry, whether it is about production gains for our teams, all the way to the innovation that it enables, but first things first. We are going to learn from our panel a little bit about them. So I'm going to start with Dawn, who is on video or on Zoom right now with us from Australia, then Eric, then Matt. And when you all introduce yourself, please share your name, your role, and then a fun fact.
What is your all time favorite movie or TV show? So Dawn, take it away.
DAWN FENTON: Hi. Good morning to everyone. I'm so glad that you're able to attend, and I'm happy to be on this panel. Thank you for having me. I am a producer of visual effects and animation in movies and theatrical special projects. The breadth of my work goes anywhere from stage managing Blue Man Group to building a production studio at Griffith Observatories, the foundation in Los Angeles, and--
- We're looking on it.
ALIZA CARPIO: OK. Dawn, we'll come right back to you. They are working on it. So, Eric, we'll go to you.
ERIC: It's back. He says it's back..
ALIZA CARPIO: OK, are you back?
DAWN FENTON: I'm back. I'm back.
ALIZA CARPIO: So, Dawn, we'll go straight to your fun fact. Your favorite all time movie or TV show.
DAWN FENTON: It's got to be Ratatouille. I love food and animation.
ALIZA CARPIO: Yay. OK, we've got some Disney fans in the room. Good. Great job OK, thanks, Dawn. We'll go to Eric.
ERIC BOURQUE: Hi, everyone. So I'm Eric Bourque. I'm VP of Engineering in our Media and Entertainment Division at Autodesk. So I'm responsible for a lot of our desktop content creation tools like Macs, Maya, Flame, Bifrost, Arnold. Yeah, and I've been at Autodesk for 15 years. I'm really excited about the industries that we serve. And my favorite movie is actually a little bit complicated. So I love science fiction. And I was kind of thinking about this. I was like, is it Contact? Is it a Rival? Is it maybe Dune?
And I feel like I can't choose Dune, because it's too recent, so it hasn't really stood a test of time. But you see there's a theme there with Denis Villeneuve. So I find that Denis Villeneuve is one of these people that makes very purposeful, slow-developing movies with the most amazing visuals. And so I'm going to narrow down and say Arrival. That's going to be my favorite.
ALIZA CARPIO: Matt?
MATT SIVERTSON: And he's Canadian, right?
ERIC BOURQUE: Yeah. He's a Montrealer.
MATT SIVERTSON: There you go.
ERIC BOURQUE: My hometown. [LAUGHS]
MATT SIVERTSON: So, hi. My name is Matt Severson I am the Chief Architect for Media and Entertainment at Autodesk. I joined recently. This is my first year, so I joined back in December, so I work closely with Eric and others in media and entertainment on our film production platform. And my favorite all-time movie, I feel like it's cheating, but Star Wars. I'm a huge Star Wars nerd, so specifically, The Empire Strikes Back has got to be my favorite of all time.
And I am just like a kid in a candy store with all the new Star Wars content that's coming out. So I feel like this question should be-- I assume everybody's favorite movie is Star Wars, so we should say, besides Star Wars, what's your favorite movie?
[LAUGHTER]
ALIZA CARPIO: You love Star Wars.
MATT SIVERTSON: Yeah. That's just me.
ERIC BOURQUE: You redeemed yourself by naming which one.
MATT SIVERTSON: Yes.
ERIC BOURQUE: That's good.
MATT SIVERTSON: And the right one.
ERIC BOURQUE: And you chose the right one.
MATT SIVERTSON: That's right.
[LAUGHTER]
ALIZA CARPIO: I want to thank you all for being part of this panel. And Dawn, thank you for being here. I know it's 10:30 PM your time. So before we actually get started on, also with our first question, I'd like to ask Eric, then Dawn, then Matt, what are you most looking forward to here at Autodesk?
ERIC BOURQUE: At AU. Yes.
ALIZA CARPIO: Oh, sorry, at Autodesk University.
ERIC BOURQUE: So first is to interact with all of you. It's to meet customers of ours that are here. It's a unique opportunity that we have, so that's definitely number one. I'm also pretty excited about seeing what's happening in other industries. Obviously, I'm in media and entertainment, but there's a lot of stuff having at AU that's outside of M&E, and so that's really exciting for me to see those kinds of things. And I would say, the third thing, if you haven't had fudge in New Orleans, you need to do that, because I'm going to be doing that.
ALIZA CARPIO: Ooh, that's a good tip. It's a good tip. Dawn, what are you most looking forward to doing or seeing at Autodesk University?
DAWN FENTON: I think the talk that stood out the most for me was the new cloud-based possibilities for media production. This is something now after the pandemic and even prior to that. It's just game changing for a producer, especially and for artists and the production as a whole to be able to produce in the cloud, to kind of be able to walk away from the hardware footprint that that takes up and utilize the cloud. I think that's Thursday at 1:30.
ALIZA CARPIO: Wow. Thank you for the shoutout out there. And Matt, what about you?
MATT SIVERTSON: Well, now it's fudge.
ALIZA CARPIO: Now it's--
[LAUGHTER]
MATT SIVERTSON: I wasn't really aware of that, but now it's on my mind. For me, much like Eric, it's getting to learn more about some of the other industries that Autodesk is in. Being new to Autodesk, I've been learning a lot about our media and entertainment business. It's already interesting learning about the commonalities and differences between our industry and the others, and so this is a great opportunity to dive in and learn more about those.
ALIZA CARPIO: Thank you for sharing, everyone. I know, for me, it's about also meeting partners that I've only met on Zoom or on Slack, or maybe meeting again someone three times. There's an internal joke here, folks, but it's always great to see folks in person. So Matt, I'm going to start with you with the first question. Let's just get everyone on the same page. Could you just explain what is open source software? What is open source in general?
MATT SIVERTSON: Sure. Yeah. So open source software is software that's sort of developed in the public space with the community, so anybody can download it, modify it, use it for what they want. Often, they'll work with the community and sort of contribute back, and so they'll actually be able to influence the future of it. And it's a really great thing, because it allows anybody to get involved, but it also allows customers-- I shouldn't even say customers. It allows the users to sort of be involved in the future of the project itself.
ALIZA CARPIO: Great answer, and I just love the focus on community there. Eric, would you add anything?
ERIC BOURQUE: Yeah. I would say that with open source, there's also kind of a philosophy, and that philosophy is often defined by the license. And while licensing is not something you really want to talk about too much, there's two styles of license with open source software. There's one that says, here's something, do what you want with it. If you want to make changes and get them back, that's great, but you don't have to. And then there's another one which says, here's a software.
Do what you want with it, but you have to give us all of your changes back and you have to cite everything you do with it, and so it's kind of very different styles. And you'll see that one of these commercial companies tend to not be able to do much with. The other style, obviously, commercial companies can do a lot with. So there is kind of that fork that happens within the licensing or the philosophy.
ALIZA CARPIO: Thank you both. Recently, Doug Walker who is a Senior Principal Engineer at Autodesk, and Tech Lead also, he wrote an article titled Open Source in the Entertainment Industry. And shameless plug here, folks, you can actually find this blog on OpenSource.Autodesk.com. And so this next set of questions-- this next question is actually for Eric and for Matt. And Dawn, please jump in if you'd like also.
And what it is, it's a little bit of a rapid fire. I'm going to ask you all to describe or just tell us what is each one of these about. So the first one is universal scene description, USD. Eric, what is that?
ERIC BOURQUE: All right, so before I answer that, show of hands, how many of you are from the M&E industry a little bit? All right, cool. So that shapes my definition. So USD is really a scene description format, but the really cool part about it is that it's all about layering. So you can essentially have all kinds of different entities, you layer them together through this composition engine, which is part of USD.
And the really interesting detail is that you can do these kind of sparse overrides of attributes within your scene that get resolved when the composition engine runs, so it's highly efficient. You can have massive scenes and you can make small changes that are non-destructive, so you can import things in various places and do really crazy things with them.
ALIZA CARPIO: USD for Maya plug-in.
ERIC BOURQUE: It's almost in the definition or in the name. It really is the thing that takes USD, the thing I just talked about, and allows you to have that as a scene description format in Maya, which is a digital content creation tool.
ALIZA CARPIO: I'm just going to keep going. Arnold USD?
ERIC BOURQUE: Almost the same. The difference being that Arnold USD is something that allows you to take a USD scene and render it. So instead kind of creating the scene, you're taking a preexisting scene and you're rendering an image or an animation.
ALIZA CARPIO: I think I lost you a little bit. So, Material x?
ERIC BOURQUE: Material x is a portable material format. I'm back? OK, cool. So it's a portable material format that allows you to sort of define a material and then have it work either in a game engine or in a content creation tool or in a renderer. And you kind of have different levels of fidelity of what you're trying to represent, but at least the material itself looks similar in all those stages.
ALIZA CARPIO: This is feeling like a spelling bee. Sorry.
[LAUGHTER]
ERIC BOURQUE: It's early in the morning too.
[LAUGHTER]
ALIZA CARPIO: I figured out, he's copying.
ERIC BOURQUE: What happens if I get one wrong?
[LAUGHTER]
ERIC BOURQUE: Do I not get my fudge later?
[LAUGHTER]
ALIZA CARPIO: Autodesk Standard Surface.
ERIC BOURQUE: Oh, cool. So Autodesk Standard Surface is actually-- is not really open source, but it's an open specification. And so we have this shader, an uber shader that was very popular in Arnold, and we had a lot of other people that were trying to reverse engineer what it did and how it worked, because they wanted to have this in other renderers. They wanted it in V-ray or in Clarice or in other things. And so, rather than have them try to figure it out, we said, well, look, we'll just publish the specification.
So we'll tell you exactly what all the parameters are, and we'll tell you what are they supposed to do, and physically how is it meant to interact. In that way, each person could often do their own implementation, and that implementation could be proprietary or it could be open source. In our case, we publish also a reference implementation, which is open source, but this is it's a little bit of subtlety.
And I think we're probably going get into this a bit later, the difference between an open specification and an open implementation. They're both kind of part of open source.
ALIZA CARPIO: OpenColorIO.
ERIC BOURQUE: Oh, that's one of my favorites. Who loves color? So color is super complicated. And it turns out that there is not many people in the world that truly understand color and color reproduction, and going from what do you get in plate in a camera when you're shooting versus what do you get on a display, versus what do you get when you print or when you print a film. And you want to make sure that the color pipeline is kind of very consistent and doing the right things and the way that you expect.
And there is this project OpenColorIO, which really is a framework for defining how all those things should work. And by having this as a common foundation, all the tools can build on top of that, so we know that all the tools that are supporting OpenColorIO, they represent color correctly through that entire pipeline.
ALIZA CARPIO: And then, finally, OpenTimelineIO.
ERIC BOURQUE: Open Timeline, so OTIO. And some of these, by the way, are projects in the Academy Software Foundation, which is probably something we'll talk about later as well.
ALIZA CARPIO: Yes.
ERIC BOURQUE: So OTIO is really about trying to figure out, how do you represent the editorials? So when you're putting together a movie or an episodic-- you basically have this cut which defines what are the shots that need to go together, what is the timing of those shots. And you need to have this way to represent that timing, the metadata, all those things associated with it, so that you can move this between applications, and OTIO is really trying to do exactly that. So they're kind of creating an API and an open standard for editorial transformation.
ALIZA CARPIO: Great job. Matt, would you add anything to what Eric mentioned to any of these?
MATT SIVERTSON: No. I think he nailed it.
ALIZA CARPIO: [LAUGHS] Great job. Woohoo.
ERIC BOURQUE: I get my fudge.
[LAUGHTER]
ALIZA CARPIO: Closer to fudge, Eric. So Eric, we'll stay on you on this next question. So I want to focus on industry tools like Autodesk Maya, so how have tools like this evolved because of open source? So that's part one, and the second part is, how has the evolution of these tools benefited the broader community? So the first question is, how have tools like Maya evolved because of open source?
ERIC BOURQUE: Yes. So I think there's two phases. So if we go back, maybe even the beginning of Maya. So Maya is an application that runs on Linux. Linux itself is open source, and Maya kind of relied on a bunch kind of small entities that made up a bigger application. And so, what I mean by that is if you think about just an image format, so like PNG, if you want to be able to save a PNG in Maya, well, we used an open source PNG library. That was kind of the right thing to do.
But I think in those days, when we were building an application, we were doing all the value added, all the complicated things, and then we would use these open source projects for kind of the formats or for basic interchange, and that was kind of the way things worked up until probably between 5 and 10 years ago, when open source and this kind of what you're saying, how does it change? When you look at something like USD, so USD came along and said, hey, I'm a new format for the entire representation of your scene.
That's a different kind of thing, because now we need to look at Maya and say, how does Maya interact with this thing? And so we actually had to teach Maya how to have a completely foreign data model, but still have all the same artist workflows. And for us, that was a really heavy lift, but it's a really important lift, because it's also forcing us to change our infrastructure, so that we can bring this application along into the future, which leads to the second part of your question of how does this simplify things overall.
It's that now, when you create work in Maya and if you're using USD, you can easily continue that work in another application, so we mentioned Arnold before. So now, you can render directly in Arnold using USD without actually translating the scene anymore. Because it used to be that to render something, you had to translate it from the host format into the format that renderer needed. With USD, that's the same format now. If you wanted to then take something and open it in Houdini, which is an application that people use for effects, you can still use the same format.
So what's happening is that the pipeline is starting to sort of get standardized, and what is the data that flows through the pipeline and the applications are now just coming and talking to that data or into that format. And so that's a really interesting evolution, because it's something that people have wanted for many years. We're not completely there yet, but that's definitely the trend that we're on.
ALIZA CARPIO: And Matt, would you add anything to that?
MATT SIVERTSON: Yeah. I mean, I think it's interesting to think about not only how it's evolved to this point, but how it actually can change the future evolution of these types of software. Because what Eric is describing is really reducing the cost of moving to different applications almost down to zero potentially. Like if you can in the old world where there's more of these walled gardens, where everything was a proprietary format, it was easier to stay within one application and just work on it and do as many things as you can within one application.
But this is really a fundamental shift, where if everything is open and everybody in the industry is speaking the same language. You can say I'm going to use this piece of software for this, but I'm going to go do this one little thing, and this other piece of software because it's actually the best tool for the job. And it will change how we actually design these applications going forward. So it's not just sort of like, hey, it's open standards and it's nice because we're all talking the same language. It will actually impact the way we build the software and design it going forward.
ALIZA CARPIO: It's pretty exciting that you all mentioned, this is just, in a way, the beginning and more is to come and more evolution, so this is awesome. I'm going to switch now to Dawn. Dawn, I hope you can hear me. What are you seeing, hearing, or experiencing as it relates to how open source has created added value to industry tools used by Creative Studios?
DAWN FENTON: Yeah. I want to comment a little bit about what you were just talking about as well, which is that it's really a game changer for production. It may change things a lot on the creation side of developing, but what these tools are doing in production as a producer, specifically, is really time and money. As a producer, I'm most concerned about schedule and budget. And these things are going to help the pipeline be more seamless and more continuity.
It's going to be faster and we'll be able to hire artists that work in any software. A real common problem when you're working on a show or staffing a show is you're looking for particular artists with particular skill sets. But now, if you're standardizing and creating a base throughout a lot of the applications, then you can switch from one to the other, and you can use those talents of people and that's not restricted to just a box there.
So I think maybe to go back to your question about the value that's really added there, that is value from a producer standpoint, for schedule, and for budget, but then value in the kind of talent that you can bring to a production, and that kind of collaboration and innovation is only going to make the projects more plentiful and likely you'll be able to create more work. Because a lot of these roadblocks that you find in productions will be cleared or lessened in that case, ideally, once the workflows are reinvented, and that's another value.
Do we need to go Linear can we go back and forth upstream and downstream in production? People can work simultaneously side by side because their formats are going to be able to be compatible with each other.
ALIZA CARPIO: Thanks, Dawn. Matt or Eric, would you add anything to what Dawn just commented on?
ERIC BOURQUE: I would just say I completely agree. And for me, the analogy would be when you think about tools, like you don't want to be using vise grips for everything. There's a time to have a torque wrench, and this is really what we're talking about. You want artists to use the things that they're familiar with, to create the art. You don't want them thinking about, oh, I have to use this because it speaks that data format or this other thing.
MATT SIVERTSON: And I just think it's so interesting because it really is. This is an industry change, so like we're software vendors, she's in production. We're all telling the same story. I think that's actually really important.
ERIC BOURQUE: That's a great point.
ALIZA CARPIO: That is-- I'm going to write that down. That is a great point.
[LAUGHTER]
So we've been talking about industry tools, community, collaboration, impact, and benefit. I'm going to switch things up, and I'd like to focus on the topic of open source community groups like ASWF which is the Academy Software Foundation. And so, here's the question, and Eric, Matt, Dawn, please jump in. So the question is, how have organizations like ASWF enabled and supported Creative Studios and tech partners in the evolution of open source within the film and TV industry?
ERIC BOURQUE: Want me to just--
ALIZA CARPIO: Yeah. Let's start with you, Eric.
ERIC BOURQUE: Yeah. So I think if we look at what was happening in the film industry, we had a lot of people that were dependent on some of these open source projects that already existed, but some of those projects were not really well maintained or they weren't staffed, and they were sort of maybe going a little bit long in the tooth. And I'll tell this in terms of maybe one project, which would be OpenColorIO, where it was something that we all said earlier, we care about color. It's super important.
The industry knew it was super important, and yet, nobody was really working on that project. The original person who had written it was no longer in the industry. Nobody had really picked up and maintained things. There were kind of bugs creeping in. It wasn't working with modern compilers, modern operating systems. So all these things where it's kind of typical maintenance that you need.
And it's kind of a shame that you had this project that was like a critical project for an industry, but the industry itself wasn't really doing anything for it, because I think everybody was kind of looking at it and saying, well, I'll wait for somebody else to do it. And so, the whole impetus for forming the Academy Software Foundation, which was part of the Motion Picture Academy, is let's identify the projects that are really important for the industry and ensure that they're healthy.
And the way that we ensure that is that we get a group of people together, where all of those people are basically donating resources, so actual humans that write code, that their mandate is to keep these projects up-to-date, to add new features as they need to, and to also work with the community, so that there's a process for how do you bring new things into these projects, how do you nominate a new feature, how do you resolve a difference where one studio wants one thing and a different studio wants something else?
How do you go through that debate in an open way where you can arrive at a conclusion that really is benefiting the industry. And so that's really why the Academy Software Foundation was born. It's something where, right now, there is-- I think there's five or six projects that are what they call full status projects. There's a few projects that are in incubation, but this is something that's kind of been growing and it's, I believe, it's five years old at this point, and growing constantly.
MATT SIVERTSON: Yeah. I mean, it's really powerful for an industry to come together into an organization like that. And I think a really good example is the open review initiative that they announced recently, where it was actually-- they were proactive in saying this is something that we care about and we want to work with industry to create a framework around this. And then multiple companies came together, so Sony and DNEG and Autodesk. We open sourced.
We're in the process of open sourcing RV, which is our playback interview component, but we're going to be working with those other companies, and it's because ASWF said, this is something we want to do. It's not like they picked a project that was already in the open source and said, hey, we want to use this and go forward. They're like, we need to create a space here for companies to move into and the industry to work together.
And that, I think, is just-- it's really, really powerful because they can give an intentionality to things and they can sort of paint a picture of where they want to go. That's actually very helpful for us as a company as well to help them get there.
ALIZA CARPIO: Thank you both. Dawn, what would you add?
DAWN FENTON: I would love to add. The audio cut it out a bit, so I'm not sure if I'm repeating something. I apologize if so. But one thing that this really means in production is it creates this atmosphere where there's neutrality. And in production, you get into a studio and you've got to sign a lot of NDAs, and you have to really be careful about what information that you are allowed to share and what you are not, and this is really taken quite seriously by everyone working in that atmosphere.
So what this group allows for people to do is to be able to come together in a place and share information freely. This, to me, is kind of game changing because no one likes to redo work. And a lot of times, there's duplicate work happening in lots of studios and lots of places, because everyone needs these solutions to complex problems. But here, we're really collaborating in one space, and it's great to see this happening.
ERIC BOURQUE: Yeah.
ALIZA CARPIO: Would you like to add anything else?
ERIC BOURQUE: That's why. I do actually, so that Dawn made me think of paperwork at 8:00 in the morning. So the paperwork is actually really important for open source software. You need to know what every line of code. Where did it come from? Was it truly open source? Was it copied from somebody else in whatever it was? And so, one of the things that's great about ASWF is that, I mentioned before, when you want to kind of resolve a dispute, well, there's a governing model, so it's very well defined.
How do you choose the direction of this project? How do you resolve things like that? How do you make sure that everybody that's contributing has signed a contributors license agreement? So there's all that kind of thing. But I think the other thing that it does is it provides a legal framework for companies to feel comfortable contributing into it. So I think it used to be that you'd have big companies that were-- they had all the best intentions.
But then what would happen is that their legal departments internally would say, ah, but the risk is this, and this, and there was kind of nobody else doing it. So now that there's legal framework and there's a bunch of big companies that are contributing things, it makes it much easier to bring new companies in, and also to get bigger and bigger contributions from those existing companies, because they see that they're protected and that everybody's kind of playing by those same rules.
ALIZA CARPIO: Anything else, Matt?
MATT SIVERTSON: That's just a really good point. No.
ALIZA CARPIO: Another really good point. I will take note of that. [LAUGHS] I want to thank you all for sharing your perspective I am a member of the diversity and inclusion working group at ASWF. And I love what all of you are saying because it's so rewarding to really hear how groups, how companies, Creative Studios can come together and really be purposeful and pursue the same goal and work together.
So I love that you all said that it's a neutral place and where we can all kind of not only lend thought leadership, but also roll up our sleeves and work together. So it's crystal ball time, which means I'm going to ask you all a question. And we'll start with Dawn then Eric then Matt. If you had a crystal ball and can imagine 10 to 15 years out, how do you see open source evolving in team and film industry?
DAWN FENTON: Well, I don't have a crystal ball, but if I did, I think it seems obvious there's going to be more projects. And with ASWF really creating this group that we just talked about, that's going to be more collaboration. And right now, there's this surge of work without enough labor really to complete that work, and so, that's putting a lot of pressure on those individuals that are.
So I would hope to think that in the longer time frame, there will be pipelines that are easier to work within, that artists can really focus on creating their best work rather than trying to jump over complex problems or do workarounds. That's very common. If something's not working and you have got a deadline, you've got to find this way to work around it. So streamlining will really be helpful for everyone and I think beneficial across all studios.
I'm not sure that we have maybe time for this story, but if you look up the history of OpenEXR, and maybe matter Eric knows this better than I. But I think the evolution of how that particular project came to be is a great example of how coming [INAUDIBLE].
ALIZA CARPIO: Thanks, Dawn. I think we lost you in that last moment. I'm sure it was brilliant. We lost you in that moment, but we'll go to Eric and then Matt.
ERIC BOURQUE: Yes. I also don't have a crystal ball, but I'll take a stab. So I think that what we're going to see in the next 5 to 10 years is a unification. So all of the things, like all these really exciting projects, I think they're going to get stronger and stronger and they're all going to kind of come together to form. What everybody agrees is the thing that we need to build stuff on top of. But then I think something new is going to happen after that, which is studios are going to want to differentiate themselves.
They're going to want to do something different than what you can do with all of those things. So you're going to have all these kind of new offshoots that are going to come out. You're going to have the new OpenEXR that don't exist today, because you're trying to solve some very specific problem. And so I think it's sort of like this trend that we're going to be on, where you sort of get unification and then diversification and then unifica-- I think we're going to have this kind of pendulum that's going to happen.
And an example of that would be if you look at just computer animation and what's happening more recently with things like Spider-Verse from Sony Pictures animation, this is an entirely new stylized look. The tools that we have today, they're not really made for that. Somebody who paints, it's the same tool. They can just take that idea. They can get it on a canvas. But when you're using software, it's a lot harder to get those kind of stylized things.
And so I think what's going to happen is it's going to get more and more customized tools that will come into those kinds of studios, where once it's no longer a competitive advantage, they're going to say, oh, let's give this to everybody, and then you'll kind of get back to this unification thing. That's a guess, but that's what I would say.
ALIZA CARPIO: Thanks, Eric. Matt?
MATT SIVERTSON: I actually do have a crystal ball.
ALIZA CARPIO: [LAUGHS] Is it the 8-ball?
MATT SIVERTSON: But it's very disappointing. [LAUGHS] You just look into it and you don't see anything. [LAUGHS] I very much agree with Eric. I think that the-- so the trend is definitely only increasing, that the industry wants to move, especially more towards open standards and open source, where it makes sense. I think that's only going to grow and sort of like move up the production chain. But I think Eric's point is really important in that open standards and open source will move in to help remove the friction from the artistic process.
That's really a lot of what it's doing right now. Because the studios, ultimately, they're not acutely interested in solving these technical problems. They're interested in telling a story, and anything that gets in the way of telling the story is something that they will work through. And so these open standards and the industry software vendors can come together to remove that friction so it's easier for them to tell their stories, but storytelling is always evolving, as Eric is pointing out and they're always going to be on the bleeding edge of that.
I don't think as software vendors we will ever be the ones that are going in and saying like, I know exactly what you need to tell the creative story that you're trying to tell. So it will evolve and it will go through these periods, but I don't think there's going back though. I don't think there is going back to these walled gardens where things are separated. It's a one way trajectory. And it's just really exciting for the industry, I think.
ALIZA CARPIO: Thank you all. Now, I have one last question before we turn it over to the audience for their Q&A. And then audience, just let's start thinking about questions you'd like to ask. And then I'm going to end up repeating it because we want to make sure that Dawn hears the questions as well. So here's my final question before Q&A, and we'll start with Dawn then Eric then Matt. If you could leave one message from our panel discussion today for folks to remember, what would it be? So Dawn, we'll start with you.
DAWN FENTON: I'm going to go with one of those things we get taught as a small person, share. Just share your information, lift each other up. We could achieve a lot more together than when we work in silos. And I think that's really the spirit of open source, is community, sharing information, building off of each other's strengths, and solving complex problems together.
ALIZA CARPIO: OK. Eric, we'll go to you.
ERIC BOURQUE: Yeah. I would say with open source influence comes through contribution not money, not the size of a company, nothing like that. So you can make a really big change even as a very small group of people or even an individual. Some of the very popular projects, it's a single individual that wrote them and yet they're used everywhere.
ALIZA CARPIO: I agree. I agree. Matt?
MATT SIVERTSON: Yeah. I'll kind of reinforce Dawn's point. I think that the power of collaboration and not just on one single open source project. I think that's very valuable, but it's really impressive, I think, what the media and entertainment industry is doing as an industry, where they have just basically set this direction. And as I said, it changes everything about how we think about the software, how we about roadmaps, how we think about collaborating, who we collaborate, with all of those things.
And so, I think there's a really good industry lesson there beyond just, hey, this is a really good way to develop software and it's useful because you can do all these things on a particular project. It's really an industry transition that I think is important. I love what you all are saying because it's really talking about the industry working as a community of practice, right where we contribute together, we learn from each other, so this is awesome.
So now, we're going to go to the Q&A portion. And when you ask a question, I will just repeat it so that everyone could hear. Got questions, anyone? Please, stand up. We'll go here, to the gentleman here, and then I'll repeat your question. But go ahead.
AUDIENCE: Yeah. I guess my question goes around the things in the future for use.
ALIZA CARPIO: The crystal ball.
AUDIENCE: The crystal ball.
ALIZA CARPIO: OK.
AUDIENCE: So I was thinking how far into the future you're looking into, is it five years down the line, 10 years down the line? The way I see it, open standards in the future is to open solve a real time collaboration environment. Everyone's who are just going to collaborate in your class, you're doing faster, things faster, thus maybe because we're going to get it cheaper, much cheaper and faster to produce. How do you see that happening in the future? Is that what's going to happen maybe through even platforms like Omniverse or something that we can do?
ALIZA CARPIO: So the question, if I could just reflect back, Dawn, the question is, as we think about crystal ball and in the future, you lost piece because there was a lot of like sentences in between. Say it again. So, in the future--
AUDIENCE: Yeah. Do you believe it would actually-- open standards and also accelerate the way we collaborate together, so real time collaboration on those topic plan for the future of productions.
ALIZA CARPIO: So in the future, how do we see the future of real time collaboration and other, maybe other benefits? I'm just trying to-- anyone?
ERIC BOURQUE: Yes. I definitely think that-- if you look at the collaboration aspect in particular, having different tools that are all talking the same data format makes collaboration a lot more realistic. And so if you think about a studio environment, you know they're collaborating, but they're collaborating on a singular pipeline that they control. If you get to the point where that pipeline is sort of a well accepted way of working, then it means you can bring a bunch of different tools to this thing. And so that pipeline might be in the cloud, for example.
Once you have it there, it makes it much easier to connect from multiple different places and multiple different clients, and that's where you can get that collaborative aspect to it. So you mentioned Omniverse, that's an approach that NVIDIA has chosen to take with USD. There are other solutions that are looking at those kinds of things. I think one of the areas that we, as an industry, need to figure out is, is the collaboration truly around individual assets or is the collaboration around scenes? Is it around the storytelling aspect?
Because I think the way that you collaborate kind of will determine the evolution of some of these projects in practice. So I don't think it's that everybody wants to grab the same vertex and move it at the same time and figure out how to resolve that conflict, but it might be more, I have an animator and a modeler and a texturing artist who are all working on the same character, and how do I allow that to happen in a way that speeds up things and allows a new change to kind of impact everybody else down in the pipeline?
So I think those aspects are exciting. And you were asking when, and I would say now. It's happening now. It's going to be happening in the future, and it's going to keep evolving.
MATT SIVERTSON: Yeah. I mean, I would agree with that. And I would say that the open source and open standards is a big part of it. It kind of allows everybody to speak the same language. So it actually enables these connections, but it's not like in itself like, just because you speak the same language, it doesn't mean you're at a cafe talking to people. So the need for that location where you can come in and collaborate is not itself solved just because people are doing open source and open standards.
So I think that's definitely in development, and the process itself is so complex that there's definitely things that need to be managed beyond just the standards and the data format. So I think you're going to see that coming soon. And certainly, there's multiple approaches, but I don't think-- well, I think the story will continue to evolve for quite a long time. I agree with Eric that it's starting now and it's going to continue to evolve over the next couple of years.
ALIZA CARPIO: Dawn, did you want to add to that?
DAWN FENTON: Yeah, sure. I'd like to add that from the artist perspective, something I hear a lot is that sometimes they feel pretty hesitant to work with something that they might say, isn't battle tested. Like artists working in a production environment have a lot of pressure to complete a lot of tasks in a short period of time, and that pressure is put on by a producer or a supervisor or a lead.
And if something isn't kind of battle tested, then they are going to shy away from it, unless there's support, unless there's a central place that they can go if there's a problem. So whether that is within a studio, a department that has that, but a lot of smaller studios aren't going to have that, and they're going to want to turn to the developers to know that there's either a forum or documentation or somewhere that they can go to find those answers, and that's kind of how you'll get the adoption buy in.
I think with without that kind of communication going and documentation, it will be a struggle to convince artists to use it, because it's kind of their-- they're on the line in that moment.
ALIZA CARPIO: Thank you. Anyone have another question? And if you do, please, if y'all could start forming a line here behind this microphone? I thought we had someone here first. So if y'all could just go to this microphone right in the middle of the room, and we'll go to this gentleman, and then there was someone behind, if you could just get in line also here? Go ahead, please.
AUDIENCE: Hello. [INAUDIBLE].
ALIZA CARPIO: Hold on a second.
AUDIENCE: OK, I'll speak loud.
[LAUGHTER]
ALIZA CARPIO: Wait. Rick is coming.
AUDIENCE: My name is [INAUDIBLE]. I'm from [INAUDIBLE], Germany. I'm on the movie business. And I would like to come back to a, let's say, small part in your discussion, monetarization. Isn't it a contradiction that you've got open source community like explicitly set not working for Autodesk, but how does Autodesk who wants to earn money can use this knowledge, or is there-- how do you solve this challenge?
ALIZA CARPIO: There's almost like a conflict between, you said, monetization and open source.
AUDIENCE: Yes.
ALIZA CARPIO: OK.
MATT SIVERTSON: Yeah, I can start. I mean, we don't monetize the open source software. But the industry isn't demanding nor do I think they actually want everything to move to just free and open all the time. So we talk about USD as an example. It's a fantastic open source format that enables a lot of things. We use it within Maya. We want Maya to be the best USD editor that's out there, and that's something that people will pay for. And so, it doesn't mean that we have to make everything free and open source.
The industry just needs to decide where are the points that need to be open to enable the collaboration that they want to do, so that they can get their job done. But I don't think it necessarily means that-- I actually don't think monetization is incompatible with open source because they're all part of a bigger system. So I don't know if you have anything to add.
ERIC BOURQUE: Well, maybe just an analogy. So I'll come back to my tools, one earlier where I was talking about vise grips and torque wrenches. So if you think about you're going to make a car, you're using a bunch of tools to make that car and you want to know that a 9 millimeter wrench is a 9 millimeter wrench, and that's the standard thing. And you don't care who built it as long as it fits that thing. What you're selling is the thing you built with it, and that's kind of the way we look at this.
ALIZA CARPIO: Good question though. Thank you.
AUDIENCE: Mine kind of tailors around something Dawn eluded to. With the ability of you all now to start hiring folks with different skill sets, different backgrounds, and methodologies for the design and the artistry that you guys use in your industry, why would there be a shortage of folks that are available to help your industry or that you're saying that there might be a shortfall in people or personnel. And so that makes me wonder what would be driving that. Is it not that they're getting out?
This is not getting out to enough folks or we're not promoting this to enough folks to develop and learn this industry or to move into this field and technology? Because to me, it sounds like with all this open collaboration that you're able to have and the things that you're doing and the ability to use more of a skill set, you'd have more people available to you. Now I come from a geology background, so this is completely out of my arena, but that was one of the key things I look for is shortfalls and why we can't produce what we do?
And why we're seeing shortfalls in people and personnel, because we see it in our industry as well.
ALIZA CARPIO: Good question. So it's about the people. So Dawn, do you want to start?
DAWN FENTON: Sure.
ALIZA CARPIO: It looks like you've got an answer.
DAWN FENTON: Yeah. I think what I meant when I said kind of a labor shortage is there's so much work. There's so much work to be done, and it may be a result of the pandemic having happened a couple of years ago. And I don't have any statistics on this per se, but there's really not enough individuals to do it. And the reason why this workflow can't be completely implemented is because there's still roadblocks that aren't really clear within the pipeline.
From my perspective, the potential is there, the projects are there, the individuals are interested to work on them, but they have to be battle tested before you can put them into production. And so, that's the kind of thing from the production standpoint that we're really looking forward to, is having that safe space where something is tested and implemented, and then we can put it into a high stakes production. Otherwise, that could be pretty disastrous, if you're trying to put something that's not quite ready and implement that into a show.
ERIC BOURQUE: Yeah. I would add, it's probably the same thing that you're seeing in your industry when you said that you're seeing the same things. It really is that the demand is really much larger than the supply at the moment. If you think about a-- take a blockbuster like a Marvel movie. So Marvel movie typically has around 1,000 to 2000 artists that are working on it for about a year, sometimes more than a year, for one movie that we all sat in our couch and watched during the pandemic, and then the next night, we wanted another one.
And so when you look at that demand for all these things, the number of people it takes to generate that, it's really quite incredible. And so, obviously, we need to have ways to leverage more artists, to have more people learning these things. And I think a lot of the things we're talking about today are going to make that easier. And one example of that is that we see much more mobility right now for artists between studios. So an artist that works in one studio can take that skill set and then go into another studio and they can get working quite quickly.
Whereas, maybe 10 years ago, they had to learn things that were very specific to each studio, so they couldn't hit the ground running as much as they can today. So at least that portability of artists is getting better, but we definitely still need more artists for the amount of content that we demand right now.
ERIC BOURQUE: Yeah. And the only thing I would add--
DAWN FENTON: I think there's one more thing to, oh, sorry, go ahead.
MATT SIVERTSON: No, no. Go ahead. I'll go after you.
DAWN FENTON: I was only going to add, you reminded me there that there are-- it's worth mentioning that in most large studios, there are programs and academies and mentorships really taking hold and becoming more robust for this reason.
MATT SIVERTSON: Yeah. And I just wanted to add that there has been this just incredible content boom. There's so much more content being created now than even just a few years ago. And we shouldn't lose sight of the fact that these are highly skilled positions that we're talking about, like the artists who do this, they train for years and years and years. So it's actually not something that you can just ramp up the number of available artists very quickly. You can ramp up the demand for content very quickly by having a pandemic.
That's one way to do it. But the pandemic does not necessarily create a bunch of artists who can fulfill that, so there is definitely a shortage.
ALIZA CARPIO: Thank you all. Any last questions? Oh, OK, great. Yey, from the side. Woo.
AUDIENCE: Good morning. My name is [INAUDIBLE]. I work for Autodesk and I worked with Aliza. This is not a planted question actually.
[LAUGHTER]
I want to go back to Dawn's issue about code being battle tested or content being battle tested. So normally, when things are written down, you write the requirements and then first thing you do is you write the testing framework. So before we contribute the code or the format, I think the first thing to do would be to donate, into the open source, the entire testing framework. And make sure the curators are the known curators, so that anything that they check in on behalf of someone is battle tested just because they are battle tested.
So I think the idea would be to any open source community, to write down the clear cut requirements and contribute to the testing framework first, and then the code follows actually. So I don't know whether this happens in the movie industry, but in the other code, I mean for enterprise software, the testing framework is part and parcel, essentially part and parcel of the system here.
ERIC BOURQUE: Yeah, absolutely. So that's definitely true of projects that are open source within M&E. And then maybe for some, it wasn't so true, so one example would be OpenColorIO, the OpenColorIO that I referred to earlier. So Autodesk made some pretty major contributions to bring that into kind of a new level of compliance with what we're trying to do and to make it work on modern hardware and have better performance and better fidelity.
The first thing we did was we wrote a test suite to make sure that we didn't break anything that was already in OpenColorIO to begin with. And so we put a lot of effort into writing that test suite, because that particular software package didn't have one to begin with. But OpenEXR is an example, which is a very well established package, does have a test suite and it kind of always has. But as security issues come up, they're writing test first and they're doing those kind of thing.
So yeah, it's very important and the projects do take that as a top consideration. I think another thing that's really important for open source software that we kind of didn't really hit on directly, but I think your point is circling it, is when you think about there's a lot of eyeballs that are looking at this code. And the more eyeballs you have looking at things, the more you're going to find some of these subtle issues.
So when you think about security issues or performance issues or things like that, having many, many people look at it is kind of guaranteeing that it gets better over time.
AUDIENCE: Thank you.
ALIZA CARPIO: Anyone else? Going once. OK, did you-- do you want to come? [LAUGHS] He's like, I don't know. Do I have one? [LAUGHS] Come on up.
AUDIENCE: It's me again. It's just that I'm very curious about again the future of open standards in general. I guess the question goes around AI. So we all have seen things that made journey working fantastically. So would you think that open standards will help accelerate the process of AI algorithms to start working on all different types of tasks like 3D modeling, texturing, et cetera? That's the question.
MATT SIVERTSON: You can go on, Eric.
ERIC BOURQUE: OK, cool. So I think, yes. I think definitely the open source philosophy will have a massive impact on AI and it already is. I think with AI and you mentioned the journey, there's certain things that you have to consider, which is that mid-journey or the style of AI it's using under the hood is something that is learned by looking at a bunch of examples. And so, is the thing that is generating, what's the legality of that? It's basically copying. It's copying a bunch of different things and producing something novel.
But if you were to go take samples of a bunch of books and put them together, that would be a copyright infringement. So I think there's-- in our industry, when we're producing final images that you have to sell, there's some questionable things there. But I mean that's obviously a details that can be worked out. I think that the notion of training on open data, where people say here's like a whole style of things that I produce in my studio, that I'm putting up for training, that would be a really interesting open source contribution.
I think the train models themselves could also be open source. So once you've done a bunch of training, you can say, hey, this particular model does this thing really well. I'm open sourcing the model. Run in however you want to run it, so that it doesn't become locked into one particular vendor or something like that. So I think we're going to start seeing a lot of those kinds of things. It's obviously more in its infancy now than, say, regular open source software, but I think we're already seeing that with-- I'm forgetting the name of it, the Fusion.
I forgot the name. But anyway, there's a new AI that's out quite recently that is entirely open source. Stable diffusion?
MATT SIVERTSON: Stable diffusion.
ERIC BOURQUE: Yeah. so this would be one example where it's something-- it's been developed completely openly to my understanding.
MATT SIVERTSON: Yeah. The only thing I'll add to that is that like the stuff that is happening now with standardizing on data formats is in itself going to be an accelerant to AI. The foundation of AI is data, so the more you have standardized data. The more you can understand how it's used, how it's connected, these are things that like the open standards really contribute to and it's going to make it easier to sort of build AI's on top of that.
If you're in this fragmented world where everybody's speaking different data, it just becomes much harder to put it all together and to analyze for AI purposes. So I certainly won't predict exactly how AI is going to evolve in this industry, but I can almost certainly predict that like this standardization of data is going to accelerate it.
ALIZA CARPIO: Dawn, anything you'd like to add?
DAWN FENTON: Only that I hope it produces more fudge, [LAUGHS] duplicates more fudge. [LAUGHS]
MATT SIVERTSON: Yes.
ALIZA CARPIO: Well, with that said, folks, thank you for being here on 8:00 session, so thank you all for being here. Have a great Autodesk University experience and thanks panelists. Thanks Dawn.
MATT SIVERTSON: Thanks, Dawn.
ALIZA CARPIO: Eric and Matt.
MATT SIVERTSON: Thanks Dawn.
ALIZA CARPIO: Thank you.
MATT SIVERTSON: Thank you all.
DAWN FENTON: Thank you.
Downloads
Tags
Industries | |
Topics |