There's interesting news that seems to have mostly skipped media attention:
MediaGoblin, a free media publishing platform, got a support for 3D models thanks to Blender. And that's just the beginning.
Before we dive into the topics like 3D models and digital asset management, let's talk about basics.
The MediaGoblin project was started in 2011 with an aim to create a self-hosted platform for publishing multimedia data — your own YouTube, your own Flickr, your own SoundCloud etc.
MediaGoblin has a flexible storage system: it's up to you if you want just local file storage or something like Rackspace Cloudfiles via OpenStack's "Swift" storage system, or even something entirely new.
It also supports many things that most social networks have: commenting, geolocation and maps (via OpenStreetMap), galleries and collections, etc.
But hey, 3D models and Blender?
Moving forward with 3D
How did 3D models support become possible? First of all, MediaGoblin is
extensible from ground up. New media types can be freely added, and that was a good start.
Few weeks ago developers of
LulzBot (a project to create a 3D printer with open specs) aproached the MediaGoblin team. They asked if 3D support could be added to make MediaGoblin a kind of free version of Thingiverse. They also donated to MediaGoblin and suggested a 3D printer as a reward for such a project.
Subsequently in just two nights
Aeva Ntsc implemented the feature. The code uses Blender to render still images from STL and Wavefront OBJ files. For the preview in browser it makes use of thingiview.js (requires WebGL support).
Sorry, this video will not work because your web browser does not support HTML5 video.
You can get a modern web browser that can play this video at
This code is still in a branch and needs to be merged into the master Git branch, but it's available: you can clone the repository and try it.
Meanwhile extensbility of MediaGoblin is growing thanks to Will Kahn-Greene who has just landed a plug-in system. According to developers, media types could become generalized plugins soon, so they could do even more than they already can.
The second part of the MediaGoblin news is that the project MediaGoblin is currently running a
fundraising campaign to secure full-time work on the project by Chris Webber, the project lead.
The deadline is quickly approaching, there are
just few days left.
What's on the list to deliver? LGW asked Chris about that, and he outlined the following features that have to be implemented and will definitely be worked on:
API for 3rd party projects;
federation, to hook together separate MediaGoblin instances;
audio and video podcasting support;
more and improved plugins;
easier installation, better documentation;
new media types;
more general polish.
Edit: if the 60K goal isn't reached, the team still gets the money. But Chris will have to find another income source, which means that he won't work on MediaGoblin full-time.
And then there is a whole interesting topic of digital asset management. Where exactly does it come in to play here?
MediaGoblin as part of a future DAM solution
In a manner of speaking, the world of free multimedia software is small. The DAM idea comes from the
Tube open movie team who are great friends with Chris Webber. In the past Chris even developed several new tools for the movie project.
At the moment the Tube team uses
Helga, a web-based collaboration tool, but there's a need for more. Thanks to its flexibility and scalability, MediaGoblin sounds like an interesting platform for a new digital asset management system.
Fateh Slavitskaya, writer and producer of Tube, shares:
I'm especially excited about MediaGoblin, because it will establish the core functionality we can use to implement a lot of cool ideas we've had during Tube production.
It's about a collaborative platform that also fills the huge need for a solid free software pipeline, a kind of Super-Helga with some interesting properties.
At the Blender Conference this year Tube and MediaGoblin developers came together again, joined by
Over to Nathan Letwory, co-producer at renderfarm.fi and CTO at Studio Lumikuu:
During Blender Conference we had a short meeting with several interested parties on digital assets management. One of the points that came forward is that different users have different needs, so
a system that is flexible needs to be created, with tools and API(s) for most if not all involved parties (from top to bottom and left to right). Essentially transcending the box of DAM into a more complex all-in production management toolkit (even cross-production).
Julius Tuomisto, Project Manager at rendefarm.fi and COB at Studio Lumikuu, adds:
Learning about MediaGoblin from Fateh and Bassam, both of whom I respect a great deal, made me think. In our talks about animation, the conversation always seems to veer towards digital asset management and how to successfully handle it in a Blender based animation project.
DAM (as for example in Tube) seems to almost always be directly linked with handling rendering passes — which in any larger scale animation project means connecting to a renderfarm.
Starting with the fun and relatively simple stuff, we would definitely be interested in offering some kind of plugin / API integration for MediaGoblin on Renderfarm.fi. This could for example enable a MediaGoblin server to automatically monitor and download frames from Renderfarm.fi for a given project.
Sounds exciting, but how much doable is that? Chris Webber replies:
Increasingly doable! But it would need community members interested in churning in that direction, and a few features to fall into place regarding plugins. It might be something worth looking for grant funding for or some specific contracting in the same way that our API was paid for by the Icelandic Government.
The project could also benefit from using Open Assets Import Library (Assimp) that supports
quite a lot of file formats. Moreover, this summer there was a successful GSoC project by Alexander Gessler to implement FBX importing in Blender via the library. VIDEO
On that Chris Webber comments:
It would be pretty great, if we could support more general 3D stuff, especially 3D files that incorporate textures etc. That might be either an extension to the existing plugin that's just arrived, or an entirely new plugin. Using Assimp might be very useful.
We're only using Blender to import things. I don't think Assimp and Blender are incompatible things in the way we're using Blender. It's entirely possible that we might continue to use Blender to render things and use Assimp in the future.
However, it's not really best to speculate.
We'd need someone to spec out and start working towards such a feature.
There's another imporant comment from Nathan to consider:
As for 3D content, I want to underline that here too it is important to champion
open standards over closed ones (COLLADA vs FBX). Granted, COLLADA is struggling with its existance, but making sure we push for good integration, possibly even helping out with implementation of proper import and export, this would be only a good thing.
If you are interested,
here is a helper project from Alexander Gessler that makes 3D models accessible to JS/WebGL via Assimp.
If you are concerned about having full control over images, videos and audio records that you put online, you have just few days left to support development of MediaGoblin — an awesome free software project that decentralizes media storage.
If you are a VFX or animation studio, or even a 3D printing company, you have even more reasons to support the project. With initial support for 3D models (STL and OBJ) MediaGoblin has a great chance to grow into a scalable digital asset management solution that is free to use and modify.
Finally, if you are a developer who's good at Python, MediaGoblin could do with your contributions.
P.S. The featured image for the article is based on an
original painting made by Justin Nichol.
Was it useful? There's more: