Hello All, Please allow me to introduce myself
I am full-time unpaid programmer and I will be fully committing myself to habilitating Collada off an on. I want to make my presence known here in the Khronos forums, because it feels like the right and obvious thing to do. I hope someone contacts me in some form or other.
When I set my mind to something it is done. It’s that simple. I am not a fickle fly by night programmer. It’s do or die.
Collada is unique. No one seems to want anything to do with it, meanwhile there is nothing on the horizon to take its place. So I reckon it’s now time to fully realize its objectives. I am not business friendly. I’d like to see the 3D field fully leveled. I don’t have time or money for business excursions.
What I do have is users. Who need to have a user-friendly means of producing and archiving artworks and other digital artifacts. It’s a personal design challenge with real world applications in other words. Failure is not an option.
I maintain a VR platform that I think is poised to take the world by storm. I am trying to get it to public Beta. The platform is an extended version of a product made by a Japanese company called From Software. They are known today for a game called Dark Souls. This product the company released in 2000 in Japan. It’s a first-person full polygon 3D video game design software suite. I envision this as being a near vertical platform for 3D media in the future. But for this it will require a 3D content-creation workflow. From Software originally used X as an exchange format for creating new components. But it’s run out of gas, and there is no alternative but to find a new horse.
Enter Daedalus. This is my bid to give users free-opensource, user-friendly modeling software. It is comprised of two parts:
A fork of the Open Asset Import Library (Assimp) which I’ve developed over the course of the last 3mos or so. This is because Collada while it may be an archival format, it is completely unsuited for transforming existing art files into Collada files. I’ve stripped Assimp down to its core, and radically streamlined it to create a modular art-preservation preprocessing stage that is used for staging files for conversion to Collada, and also for visually browsing files, where there is no good reason to convert them to Collada before displaying them. During the course of this work I developed an STL like container class that brings a modern C++ feel to the Assimp workflow, and makes the code very tidy, and programming loading modules straightforward. And I’ve developed a level-2 universal representation of mesh files. This affords a universal representation of 3D meshes, so file loading modules do not have to perform any gymnastics, and can be themselves ultra tight. The DAPP or “Digital Arts Preservation Protocol” is a portable layer that allows modules to be written using neutral identifiers independently of the underlying implementation (which could be Assimp or the fork I’ve made, which is loosely referred to as PreServe by Daedalus’ codebase, or any other implementation) with the ultimate goal of collecting modules for preserving every 3D data computer file in existence to Collada (or that’s what Daedalus primarily does. Assimp is primarily an in-memory C++ data structure representation.)
I am using the Collada-DOM project to build Daedalus. I believe eventually it will completely subsume the Collada-DOM codebase, essentially giving it a new lease on life. There are not really any alternatives. No software actually implements Collada as nearly as I can tell. Part of my work is providing these media platforms, and content-creation means. But I do this all opensource, and so the quality of the sourcecode is also paramount. Collada-DOM isn’t up to the level I require source code to be. So inevitably it will be commandeered. I’d like to think that when this is done it will be a new-lease-on-life for this work. The specification mentions the “Collada-API”. I think this might be a good name for this project. Since Collada has been so shunned, it does not seem like there is going to be a bevy of implementors. Therefore it appears that it will come down to whoever does the job first. And that will be myself. So I hope I have full support (not that it will matter.)
I hope someone here will see this thread and talk to me. I will make some other threads highlighting what I consider serious defects in Collada’s specification (1.5) and also Collada-DOM in another thread.
The From Software made product is called Sword of Moonlight: King’s Field Making Tool. Another aspect of Daedalus is it will include a modular interface for filters, which are intended to be used to browse art catalogs. For example a PlayStation disc can be thought of as an art catalog. If only it can be opened like book and pored over. So there will be filter modules organized according to the practices of companies and down to individual catalogs. The first model for this will be the PlayStation game King’s Field II (1995) and it will be ported to Sword of Moonlight as part of a larger initiative to raise awareness about art preservation. This may not be welcomed by commercial interests. However it will be associated with Collada, as part of archiving the history of digital art, and also securing the future. The filter modules will open up discs and things like open books that’ve been locked for 20 years. I also believe King’s Field II will be a killer app for VR. It’s the archetypal VR video game only seen in movies of the 90s. I personally feel it is the best modern 3D personal narrative artwork that exists (and probably among the most obscure and under appreciated as well.)
I hope all of this will be a tour de force, and I want to bring Collada along with it. Daedalus is named for the DAE files, and also the mythic artist Daedalus. It’s separate from Sword of Moonlight. The heart of modern Sword of Moonlight is to establish a streaming media model for video game like media. Streaming in the same sense that webpages are streamed–you do not download the entire website–and that you do not run a new sketchy executable file merely to sample a new piece of media. It is also very much timed to coincide with VR. It’s a general purpose platform, but more tailored to embodied games, that feature control of “avatars” from various vantages. It is not like a game engine, but is aimed instead at all peoples who can be potential 3D media artists. In fact creating 3D models is not part of Sword of Moonlight. It aims instead to organize a kind of cyberspace of digital resources, so the discipline of telling stories and staging plays can be unconcerned with the making 3D models for example, just as a director need know nothing about how to build an actor with also a strong focus on asynchronous development, so a person’s creativity is not limited by their income or wealth or ability to coordinate and keep satisfied complicated groups of people.
Website for SOM:
I see this becoming like Minecraft 2.0. And also many young people who have been prefigured by Minecraft will be eager to do something a little more sophisticated than its big fat blocks play. This will be there for them. They will take to it like fish to water.
Daedalus is part of the opensource code at svn.swordofmoonlight.net as this is the ends to which it is principally being developed. When I begin transforming Collada-DOM I’d prefer to include that in the Daedalus SDK, but I’d like to keep that effort separated out as much as can be. I don’t want to overextend myself. But I believe Collada must make a next step. I don’t see anything opensource related on Collada’s horizon. And I believe Collada-DOM basically gets it right. It just needs at least one truly dedicated programmer. I’d say I am a necessitous programmer, which is a degree even better than dedicated
Thank You All, and Please Be In Touch