OSVR may be “open source” but it is not “open standard,” and that is an important distinction says Neil Schneider of the ITA
“Open source” has become a bit of a buzz phrase in recent years in the technology world. In it’s most cavalier and casual meaning, an open source project is one that opens up the code to everyone, allowing for collaboration and bringing in new ideas from fresh eyes.
Palmer Luckey himself believes fully in the open source movement, having opened up the firmware, schematics, and mechanics to the DK1 at Oculus Connect in October. Another project, DIYVR from DODOcase, looks to collaborate with VR pioneer Tony Parisi to create an open source web-based software development tool for VR – the WordPress of VR if you will. Another project, which has the community buzzing at the moment, is Razer’s OSVR, or Open Source Virtual Reality. The project, which we dove into last week, centers around the message that the entire system is open source and made to be that way. It is an ambitious project, for sure, and could be great for the development of VR, but Neil Schneider of the Immersive Technology Alliance, begs for caution when trying to make the distinction between the colloquial definition of open source, and it’s actual definition. Schneider says there is a distinction that people must be aware of between “open source” and “open standard.” Says Schnieder, “If OSVR becomes super popular it can be a ‘defacto standard’ or even a viable open source platform – but not an open standard. It’s not the same thing and that should be made clear.”
I sat down with Schneider to talk about VR and gather his thoughts on the open source movement, OSVR and VR in general. What follows is the transcription of that conversation:
What is your role within the Immersive Technology Alliance (ITA)?
I serve as Executive Director. I’m responsible for The ITA’s day to day operation and following-through on the approved initiatives. That said, I have no voting rights in the organization. Initiatives are approved by our Board of Directors and membership.
Tell us about what ITA is looking to do.
The alliance was originally founded in 2009 to represent the needs and interests of the stereoscopic 3D gaming industry because there was no such organization before. When VR and immersive technologies launched on our doorstep, it was renamed to The Immersive Technology Alliance with an expanded mandate in March of 2014. We cover most aspects of immersive technology including virtual reality, augmented reality, stereoscopic 3D and more. We’re also no longer limited to video games and we change with the times.
We’re primarily problem solvers. For example, when it was realized that existing 3D gaming research was limited to 3D film, we raised over $600,000 of government and industry backing to launch iGO3D – an academic research initiative.
The recent Immersed conference in Toronto was the result of the alliance identifying a need to grow our industry beyond the confines of California and the need for business building tools to be more widespread.
Based on the success of Immersed, the membership and industry realized that the alliance is capable of a great deal no matter who or who isn’t working with us. There are a lot of standards and fundamental issues that need to be resolved – these are things that only a non-partisan non-profit organization can do effectively, so we are working on it.
What are your thoughts on OSVR?
I think they are on the right track in that we should have a future where one solution works on nearly all solutions. Since OSVR is open source, I expect there will be all kinds of forks, collaborations, and learning opportunities. The discussions and discoveries alone are valuable to the industry at large. I know on MTBS, our open source Vireio Perception VR driver developers are planning support for OSVR as well as other viable platforms that cross our path.
However, “open source” is not the same thing as “open standard”, and I think there are some misconceptions between the two. If OSVR becomes super popular it can be a “defacto standard” or even a viable open source platform – but not an open standard. It’s not the same thing and that should be made clear.
“‘open source’ is not the same thing as ‘open standard’ … If OSVR becomes super popular it can be a “defacto standard” or even a viable open source platform – but not an open standard. It’s not the same thing and that should be made clear.”
There seems to be a bit of confusion between what open source and open standards really are. What is the difference?
One is a software license and one is a model or expectation for how standards are created.
There are different types of licensing agreements with open source, but the general idea is that everyone can contribute to a codebase. Everything that is contributed to this open source codebase is free to be used elsewhere. You can’t sell or resell the code – at least, not directly.
If you modify the code, you have to publish your modifications for all to see and freely use. So if someone took Vireio and forked it to a different path, they have to publicize all the changes they have done unless it’s just for personal use. This is great for collaboration and for developers to go in different directions and try new ideas out as needed.
“One is a software license and one is a model or expectation for how standards are created.”
So why would a commercial entity voluntarily open source their work? Open Source makes money through plug-ins and consulting. Plug-ins can be closed source. Going back to the Vireio Perception example, we purposely closed-source the VRBoost portion of the drivers for security reasons; we didn’t think it wise for that codebase to be in the public domain. It’s freeware, but the code isn’t for the public to see. We could legitimately charge for the plug-in, but we choose not to.
If you use a content management system like WordPress or Joomla, it’s the same model. You get the core system for free, but many of the valuable add-ons or enhancements are pay to play, and they are closed source.
Another avenue is consulting. If you need special modifications done to an open source project, you can pay for programming or even customer support. Or, there are even more creative ways to make revenue if you look hard enough.
Android is open source. This is how it’s possible to have all kinds of rooted Android versions, and it’s completely legal. The reason the smartphone and tablet makers don’t just run off with their own Android versions and get rid of Google’s revenue generating store in place of their own is because unless Google’s store is included, you can’t install Google’s mobile apps like Google Calendar, Google Maps, etc. These apps are deemed the life blood of Android! So Android is a platform; a standard through sheer numbers of devices – but not a truly open standard.
It’s a fantasy to believe that decision power is uniform with open source. Valve’s Steam OS is open source. You can do whatever you want to their OS, but unless they approve the changes, it won’t be used in conjunction with the official Steam Store. Suddenly your custom Steam OS is useless without Valve’s approval because there is no audience behind it. It’s not an open standard because just one company out of many has the decision power of what is ultimately used where it counts.
“It’s a fantasy to believe that decision power is uniform with open source… It’s not an open standard because just one company out of many has the decision power of what is ultimately used where it counts.”
In contrast, an open standard is best defined as a responsibility. The responsibility is that the participating industry bodies have an equal vote regardless of size. You can’t just throw something in or take something out; it has to be discussed and agreed upon as a group. The Khronos Group which makes OpenGL is an open source platform, but the core spec is agreed upon by the membership through votes. Unless there is that equality and equal voting share, some huge innovations could get locked out because the big guys don’t want to play.
Also remember that open source is wild west. Without consistency, a standard is meaningless. Imagine the market confusion when a group of people work together on a platform, developers have a falling out, and they divide into two groups…or three! Now you have competing standards in an open source environment. There is nothing to bind it together – no order to things.
To be truly open, everyone has to be welcome and with equal rights. Before the alliance was first founded in 2009, we thought we could have a for-profit corporation behave as a non-profit. Boy were we wrong! Unless you are building a for-profit platform that you are planning to control, you need to have a formal non-profit structure or things will fall apart fast. There is no way in heck that members can have equal power if it is backed by a single individual or company. This is one of the reasons why non-profits exist: they protect us from ourselves by forcing us to be open and transparent.
Why is it important to know the difference as pertaining to Virtual Reality?
I think standards need to be handled both openly and responsibly. If we look at the biggest players in the market, they are hard at work in their labs trying to come out with the best possible VR and immersive experiences possible – or at least as good as they know how.
A standards organization has to be willing to invest the same amount of effort and diligence through discussion, debate, debate, discussion…debate…and hopefully a consensus that works. You can’t just rush something out and call it a standard. In fact, this isn’t about a standard – it’s about standards. There are all kinds of interwoven parts in immersive technology that need to be considered. There are no quick-fixes.
What should we as a virtual reality community be looking for in an open source VR world?
This question exemplifies the problem!
This isn’t about open source or closed source. These are just licensing terms. The real question is what should we be looking for in an “open standards VR world”. Who cares if it’s open or closed source?
“The real question is what should we be looking for in an ‘open standards VR world.'”
Number one would have to be a formal non-profit. If it’s intended to be a commercial platform; ok – call it a platform. Second would have to be equal voting rights for all participants. The little guys have to be as influential as the big guys. Third, there has to be consistency. If the platform can divide into a million directions without a care in the world, it’s not a standard. Are there limits on who can use the standard or how it can be packaged? Can a commercial entity have any pronounced influence over others either directly or indirectly? These are all considerations when talking about open standards. I’m sure there are others.
Whether standards are open or closed source has nothing to do with whether or not the standard is open.
Are NDAs and Royalty Fees a good or bad thing?
They aren’t good or bad, they just are.
When members join The ITA, part of the membership agreement is an NDA. This is required so people can openly speak about the industry or technologies in development at private ITA meetings. Unless you have that safety net, people will not openly collaborate – this is required by nearly all the reputable standards groups.
As for royalty fees, that depends on the technology. Khronos Group doesn’t charge royal fees (they make OpenGL), but there is no secret sauce being exchanged. Their API makes command calls to the GPU technology without the GPU having to give away proprietary secrets. If Khronos’ standards required sharing a software codec or unique method of doing something, then royalties could be a very real and fair consideration. Royalties are also necessary to sustain the organization and compensating the inventors where necessary. No royalties plus no resources equals no standard.
Do you know of an exciting open source/open standards on the VR horizon that we should watch out for?
I can’t speak to open source because that’s immaterial to the issue. What I can say is The ITA is committed to discussing standards and what they mean and how to best build them. Neil Trevett, President of the Khronos Group as well as most of the technology making members in the alliance are planning to participate.
We’ve also had input from technology makers who aren’t yet members in the alliance; everyone is going through a discovery process right now. It’s not an overnight thing, nor should it be.
At CES, we had a dinner meeting to discuss this, and it was gratifying to see many of the staunchest competitors sitting at the same table. This is probably even more important than the standards themselves. This to me is exciting.