Chairman of the Blender Foundation
Sciencemotionology talked with Ton Roosendaal, whom some people may know as the "father" of Blender
The Blender Conference in Amsterdam that Sciencemotionology attended in 2014 was surprising to anyone with a purely scientific background. The audience was extremely participative and - as befits the subject - open in their remarks. After the conference, SciMoLogy had the opportunity to visit the Blender Foundation and to talk with Ton Roosendaal, Chairman and Founder of the Blender project - whose views on open source and copyright issues are of particular interest - and to visit with the crew working on the Gooseberry Project, later released as the open project Cosmos Laundromat.
The Blender event was an entirely different experience from what is usual at scientific conferences. Unusual, in the sense that the community revealed a down-to-earth approach with people sharing, not only their work, but also, more importantly, their failures.
Ton found nothing surprising in this because their collaborators and developers love to talk frankly about the uses of technology and art in building something.
In fact, artists, developers and movie directors shared their problems, talking about issues that came up when building their new animation, video game or apps. The audience pitched in with suggestions to overcome problems raised. One leader of a team described how, when working on a video game, they had lost an entire year of work because of a wrong approach.
For Ton Roosendaal, the Blender community is amazing: the developers, the makers are very passionate. They see it as a vitally important tool. Perhaps a Linux, computer graphics, or other conference is different in that regard, less emotionally involved.
What makes Blender special is its story, its source, its origins.
In the nineties, Ton worked in a commercial studio for around eight years, but began to feel unsatisfied. He says that the clients always wanted to do the same kind of work.
In the beginning, I learned from it, and I grew, but the clients did not. They started to hold me back and they simply use you because you are good. So I didn’t learn from that.
I wanted to grow. I wanted to do new things. We had developed a certain software in our studio, so I thought: let’s try to put it online and see what happens.
It was 1998, and we only had a silicon graphics version, but that already gave very interesting dynamics. That resonates with the 3D graphics communities, because that is a way of life, something like skateboarding. It’s something special. It’s something we do full-time. It still is, but in the 90s it was even more full-on, it was magic! If you have a magician’s toolkit, and if you get many “Harry Potters” together, you can make magic happen!
The first Blender Open Movie opened to the public in 2002,
I started a new project but after two years, I felt that the users were absent and I knew that the coders could help them. This is art in an open way, you can see what goes on, you can have some kind of studio model, just as you can see it at Pixar and other places, and you work on what is needed, instead of working on what everyone is screaming about.
That model carried on through 2005, 2006 and then Ton created an Institute and started making the central software. Developers and so many others appreciated that easy approach. Ton says you can still develop Blender even if you are not a highly trained engineer.
Obviously the task of managing so many artists and developers in different countries was not easy.
Roosendaal managed the Blender Project too, with online developers, people working on the website, documentation, and mailing lists, et cetera.
He finds it hard to explain all the complicated workings of it.
We have many formal aspects to it. We have project module ownership, typical of how open source works, and we have small teams that can work autonomously on a part of Blender. We say to people “this is your corner” and we agree how we should work together and on the big picture. Within your box, we say you can do practically whatever you wish. People have to manage themselves. You do it in an accessible language, for artists and developers, with mailing lists and in other places where everyone can communicate.
This project has been so successful that it has grown to over a hundred people, resulting in problems becoming magnified.
It can get confusing. People are going in and out, sometimes a couple of hundred working on things. Perhaps, in total, we have three or four hundred people trying to work. And then we get people saying they cannot get in, and that it’s not accessible anymore...
How do you get through with this without the bureaucracy? It’s horrible… It has to stay fun! It’s hard to keep it accessible. The people who can get through any obstacles are those who are really very good and highly motivated.
Blender has achieved a certain dimension, where more people are trying to get in and this gives them the chance to be more selective.
It’s not always the people who have the most experience, who are the best. I get people from the film industry, with much experience saying: “I don’t have time for this. Why don’t you appoint me as boss?” They are not used to have a bottom-up system, where people come in with small schemes that can evolve.
It’s my way of working, and it is also a cultural issue. In France, they have a strict hierarchy and you have to listen to the boss. In the Netherlands, we are used to agree on things firstly and on getting my team on board. To motivate them we listen to them; try to adjust some things, to make sure everybody is happy. Unfortunately, that’s not always what works. Making everybody happy and making a good movie - it might not be the same thing.
We are experimenting with an open model, and open production where people can have a subscription and where they can get everything. Creative commons in a public domain. If you have the content in open access, you can copy everything and share it, of course.
Naturally, the copyright of digital property is an issue for whoever produces digital content. The trend has always been that people expect not to pay for internet content. Everything can be copied freely and shared. Ton Roosendaal believes that the only way to get a revenue stream is by providing added value.
People are thinking differently about music. At first, it was extremely valuable to have a record, or a CD. Then it became really valuable to have your mp3 player full of songs. Now it’s all disappearing. Now you get it all in streaming from music websites like Spotify. It doesn’t belong to you, but who cares? It was never yours anyway.
Of course, if the internet stops working that’s a bigger problem. Imagine if all the music that the Beatles made disappeared! Then, you start making music again…
Ton spoke to us about the ongoing problems with copyright models, which he feels should be re-assessed.
His opinion is that the value people can make in their creative expression has to be reconsidered. The value is not so much in the sharing of it, nor the fact that you can have a DVD. The value is in the moment it is created. Musicians should have been paid the moment they created music: and so they did. Nevertheless, he asks why would their grandchildren, and their great-grandchildren and so on, for the next two hundred years get paid for the same piece of work?
Copyright lasts for seventy years! It was extended because of the Beatles. The limit was fifty years, but in 2013, when it would be exactly fifty years since the Beatles made their first album, it was increased to seventy. That gave the Beatles twenty more years to make money! Otherwise, it would all be in the public domain. For me that doesn’t make sense. In the USA, they are trying to make it one hundred years after the death of an artist. It’s the ownership of an idea to the extreme.
In the case of little-known musicians and groups, Ton considers they do not benefit at all from copyright. In fact, he says they do not want to have ownership of an idea, preferring instead to have gigs to play and to get paid for it. They want a platform where they can share their music.
Imagine a journalist. Then, he or she could say: “Well, everything I write is my own and anyone who wants to read it has to pay for it.”
Then he says: “Oh, I would rather have a decent income from a magazine or a journal, or whoever pays my salary, and I get paid for what I write. But, what about next year? I will have to do something else. Then again, I will be paid for what I do.” What if you work on the streets, sweeping and cleaning, or if you worked in a restaurant or in a studio making animations? What right has the animator to say: “Well, I want to get paid for my animation, again and again”? But, the musician has the right? You should pay the musicians for their time and that’s it.
This discussion brings us to the hot topic of copyright in science, or the issue of having to pay for most scientific papers.
According to Ton Roosendaal, some scientists are now kicking against that. They are starting to open their own publishing websites, from their universities and sharing their knowledge, and sharing their failures.
It’s a new trend I read about in a Dutch newspaper. It spoke about open access, and it was about Dutch universities saying that the current model is sometimes corrupted.
There have been scandals about scientists making up their results, by manipulating statistics. If you don’t publish you don’t count and you can only publish if you have a result. But if in your research the result is “no”, there is no correlation between chocolate and cancer, let’s say, nobody will publish it.
Roosendaal makes a very pertinent point here because, generally speaking, if scientific papers are not published, finance is not forthcoming. If you do not get financed, you cannot carry out the research.
Universities should train smart people, those who can conduct experiments. Sometimes it will prove useful, sometimes not. That’s how it goes. That’s science. Then they should share. Talk about it. That’s why computer graphics became so big. Especially in the 90s, before the internet almost, in the era of CD-ROMs, the big thing was that the developers, the people, were sharing everything. Everything, all the time. The area was so new, that it made no sense to block things.
That’s why everything went so well in the 90s. And now… it’s coming back! Companies are starting to do open source. Even Microsoft is talking about making Windows an open source. That is completely different thinking. It’s also because of Apple. They’re getting their momentum back. Apple supported the Blender Project. In 2002, 2003, 2004 and then it disappeared. Apple was becoming a gated community. Recently they returned, because Pixar now likes Blender!
Apparently, studios are finally opening up. Blender depends on the good interoperability of technology. Roosendaal claims their business is not about taking other people’s jobs, they do not wish to be exclusive.
To get the best pipeline of cash ever, you have to work on it. Forever. You can’t copy Pixar. You have to build it up and it will take you ten years. So, there’s not any real reasons to not share. That’s the way to grow!
Obviously, it is hard for smaller studios to compete with larger ones like Pixar. To reduce costs, some use Blender, which is free and therefore a great resource.
There is the problem of the amount of work behind an animation. The amount of time it takes to make it. You can’t have one single person opening their studio to a (scientific) animation and at the same time be an expert on everything: modelling, animation, lighting, after effects, at least If you want some quality.
I invented what I call the “Magic Number”: that is the amount of work (in months) per minute. For Blender, it is between six and ten. In total, it was fourteen minutes, and we invested one hundred and forty months of work, with a team of between ten and twelve people.It was super low budget. Pixar is 250 to 500! 1 minute equal to 250 months of work! That’s why it’s so difficult to compete with someone like Pixar. They can invest so much of their resources. At DreamWorks, they have 2200 people. They only produce ONE film per year. You can calculate the magic number. If it were 2000 people, a film of 100 minutes per year, and in 12 months the magic number is... 240! Of course, you have to factor in administration, copyrights, sales people, and writers.
Scientific animation may not be as elaborate as film, but you there are other problems to solve. It depends on what is wanted.
If you want Blender, it is ten months of work per minute. One month of work is 5000€, quite a reasonable cost.
Therefore, 5000 times 10, that’s 50 000 per minute. Even in the 90s, in our studio, we tried to calculate on a minimum of 250 guilders. It would be 250 euros per second. If you want a minute - 15 000. If you do it for less, it’s not sound business.
Ton says they collaborated on BioBlender with Mike Pan, who made large parts of it, also making the Molecular Flipbook. People like Monica Zoppè use Blender, creating plugins for scientific purposes.
I always wonder why the universities - especially the mathematics, physics, IT departments and simulation - don’t collaborate more and say “Guys, we have a great opportunity. There’s this open source software, let’s make our version, let’s upgrade it, and let’s give it to artists. They can do crazy things with it! That’s how scientific visualization or simulation works too. Sometimes you have to see it to understand it. That doesn’t happen much. We had that for fluids. We had that collaboration for Blender. It was a scientific project. It’s ridiculous that there is no open source platform training.
Ton Roosendaal making the inaugural notes at the beginning of the Blender conference in 2014, at the Balie in Amsterdam, NL
Elephants Dream - the world’s first open movie, made entirely with open source graphics software such as Blender. Released in 2006.
"To make everybody happy and to make a good movie - it might not be the same thing."
“Guys, we have a great opportunity. There’s this open source software, let’s make our version, let’s upgrade it, and let’s give it to artists. And they can do crazy things with it!
"I invented... the Magic Number!"
"They only came back recently because Pixar now likes Blender!"
"The only thing you have is the file on your hard drive.
But nothing else, and that's the problem."
"It's the ownership of an idea to the extreme"
"It's far more interesting to know that those guys in Denmark tried it already and they failed. Awesome."
Ton is a great conversationalist, besides a designer, and entrepreneur. The discussion was much broader than just about Blender or the topic of animation.
is like a way of life. (...) it's something we do full-time."
"You have to build it up and it will take you ten years."