Interview to Shun Iwasawa, OpenToonz team member /Assistant professor of the University of Tokyo

We interviewed Shun Iwasawa, OpenToonz team member and Assistant professor of the University of Tokyo, about his working experience in the OpenToonz development, the reaction of the animation world to the release of the product and the next steps of OpenToonz.

Can you tell us something about you and your past working experience?

From 2006 to March 2014, I worked at Studio Ghibli as a programmer. I was in charge of development of in-house VFX tools and customizing Toonz. After quitting Ghibli, I became an assistant professor of the University of Tokyo and have been engaged in research on computer graphics.

​​You were in Ghibli before joining Dwango … do you believe the studio will be back on production sometime in the future or do you think that time is over forever?

​​In the first place, I’m not an employee of Dwango. I’m still an assistant professor of the University of Tokyo. For OpenToonz project I’m collaborating with Dwango as a researcher. One thing I have to mention is that Studio Ghibli is continuing their production. They are now making a short animated film directed by Hayao Miyazaki.

During the launch of OpenToonz there was a clear mention to a collaboration with the Tokyo university .. can you tell us more about that?

​​Collaboration between Dwango and the University of Tokyo has taken place as a result of choosing the best way for us to maximally focus on the development of OpenToonz. This collaboration was realized under generous understanding and support from both parties. I’m much grateful to be able to concentrate on this project.

​​Speaking more specifically of OpenToonz, what are the basic ideas that have guided the development of it?

One major purpose of OpenToonz project is to develop a new platform for connecting the academic research into frontline animation production. To achieve such purpose, Dwango engineers newly developed a plug-in effect SDK for image processing for OpenToonz. My workbefore the initial release was to merge UIs and features of Toonz Ghibli Edition into Toonz7.1 Harlequin source code, as a starting point for developing the software suitable for Japanese animation production workflow.

How was the outcome from the Anime launch and how was the reaction (to the release of OpenToonz) from the Japanese Animation Industry ?

We received huge public reaction upon release of OpenToonz. The most unexpected thing was that there were really a lot of feedbacks from abroad. Actually, the membership of “English version” of OpenToonz user forum is 18 times larger than Japanese user forum’s. Of course it may be because of a “shy” personality trait of Japanese, but I think the principal reason is that there were a lot of original Toonz users all over the world. We also received a good and expectant feedback from the Japanese Animation Industry. However we understand that the software lacks some features and usability so that OpenToonz can act as the alternative to the existing workflow. We have the intention to update the software based on their feedbacks.

Can you tell us something about you and your past working experience?

From 2006 to March 2014, I worked at Studio Ghibli as a programmer. I was in charge of development of in-house VFX tools and customizing Toonz. After quitting Ghibli, I became an assistant professor of the University of Tokyo and have been engaged in research on computer graphics.

How many are you in Dwango looking after Opentoonz? How many of them are developers ?

Dwango does not disclose about the number of engineers involved in OpenToonz. One thing I can say is that many staffs contributed to this project with not onlydevelopment work, but also management, design work, public relations etc.

Have you already established how external developers might contribute to OpenToonz?

We introduced a continuous integration tool into OpenToonz project on github, so that the test-building of each pull request – which is necessary in the first place – can be done efficiently. Also we are releasing nightly build every time the source code is changed in order to get quick feedback from users about new features and bug fixes. Actually it works well as some issues were found and reported from the nightly build versions.

We noticed there are some Toonz features you have removed from the first versions of OpenToonz; can you tell us why?

​​​​OpenToonz is based on Toonz Ghibli Edition, which had been customized for production work in Studio Ghibli. The customization was done on the basis of situations as follows:

Each production workflow ( ink&paint, color design and composition ) is done by different staff in individual division.

Animation was originally drawn on papers and converted to Toonz raster level in Toonz. (Ghibli did not draw vector-based animation, did not make cut-out animation neither.)Under the above situations, we decided to hide/remove some UI and functions from Toonz Harlequin in order to increase efficiency by limiting commands only related to each workflow.

However we noticed that there are many users who want to make cut-out animation by exploiting missing Toonz features. So we have reverted some features in response to their requests.

We see that your to-do is list is pretty big: working on stability, porting on Linux, reintegrating features from Harlequin, adding new features, supporting the studio that have switched to Opentoonz are just some points   other important activities we did not mention?

​​​​First I need to clarify that we (I and other Dwango engineers) will NOT respond to every request posted to OpenToonz website. OpenToonz is OSS, which means that any developer around the world can join the project and add any improvements to the software. We hope more users and developers to join OpenToonz community and improve the software together.

On the other hand, we think it is important not only to support existing users, but also to encourage animation productions not yet using OpenToonz to consider transition of their production tool. In an effort to achieve it, we would like to enhance consistency of the software in order to reduce the gap between existing and our proposing workflows.

Most of Japanese animation productions are unfamiliar with the way to post their requests via user forum. So we would like to go for an interview to gather feedbacks on a face-to-face basis.

​​Do you have a priority list ? If not, according to which criteria do you schedule the activities ?

​​​​The first month after the release we simply concentrate on modifying critical (i.e. crash) bugs, so we still have not any priority list for development.

As we mentioned in the previous answer, the major purpose of OpenToonz project is to develop a new platform for connecting the academic research into frontline animation production. It is necessary to customize the software to be accepted by the productions (especially in Japanese animation industry) in order to achieve such purpose.

On the other hand, it is true that not every user is producing his work with the same way as Japanese productions. I think it is also essential to respond community’s demands and keep the software available for various work styles in order to keep the community active.

In my opinion, the above two points will be the criteria for putting priority to our future development. These two ways (customizing the tool for specific workflow and enhancing the versatility of the software for accepting various work styles) will sometimes conflict. I think it would be important to keep in our mind while improving OpenToonz, to maintain a balance of both specialty and versatility.

Toonz goes Open source