US Army working on browser-based viewer for OpenSim

The Army’s Research Laboratory Simulation and Training Technology Center has begun work on a browser-based viewer for OpenSim that uses WebGL, HTML 5 and Javascript technology.

Douglas Maxwell

Douglas Maxwell

“The initial work was done using our MOSES Grid Manager that laid down the interfaces to the Open Simulator needed for this task,” said Douglas Maxwell, the science and technology manager for virtual world strategic applications at the center.

The MOSES Grid manager is an open source project, with code available on GitHub.

“In the future, grid management tasks and server control functions will be performed in the same client as the graphical interface,” said Maxwell.

He declined to provide additional details about the viewer.

“That is all we are prepared to talk about for now,” he said.

What is WebGL?

WebGL and HTML 5 are the latest developments in Web technology, and allows virtual world environments to run right inside a browser.

Cloud Party was a browser-based virtual world that required no plugins, and even had in-world building tools. It shut down in early February when the entire company was bought up by Yahoo.

(Image courtesy Beloola.)

(Image courtesy Beloola.)

More recently, Beloola launched a browser-based virtual world platform. Since the platform runs in the browser, and both mobile and desktop browsers are starting to support virtual reality, Beloola plans to have support for Oculus Rift, Gear VR and Google Cardboard headsets early next year.

Once a virtual world is running in a browser, it makes it possible to start looking at virtual reality support since Mozilla is working on adding full support to Firefox, and Google to Chrome.

For example, the MozVR project not only switches a display to the side-by-side stereoscopic view needed for virtual reality, but also allows users to change the direction in which they are looking by putting on the Oculus Rift headset and turning their heads. Similarly, the mobile version switches to the side-by-side view, and uses the smartphone’s built-in sensors to track where the user is looking.

Google VR Chrome Experiments's roller coaster application runs in the browser and is already fully compatible with Google Cardboard.

Google VR Chrome Experiments’s roller coaster application runs in the browser and is already fully compatible with Google Cardboard.

Google, meanwhile, has its Chrome Experiments for Virtual Reality, and I’ve tried it out on my Android phone and it works, though the selection of experiences is pretty weak right now.

On the plus side, it’s already built into the Chrome browser on Android phones. Just open the browser, and go to vr.chromeexperiments.com and you can check it out.

Why is this important for OpenSim?

The social component is going to be huge for virtual reality. Roleplaying games, obviously, will have to have multi-user support for their virtual reality versions.

But outside of gaming, there are an infinite number of social applications for virtual reality. Socializing, of course, but also meetings and events, education and training, group tours and group shopping, museum exhibit openings, concerts and other live performances of all kinds.

At first, there will be stand-alone apps in all of these areas. But, as we’ve found out in OpenSim, creating an outfitting a new avatar for each new applications is a real pain the butt. You might want to have a personal avatar for family and friends, a school or business avatar, an anonymous avatar, and a couple of avatars for your favorite games, but after a while you’re going to get avatared-out.

AltSpaceVR is one of several new social virtual reality startups.

AltSpaceVR is one of several new social virtual reality startups.

And once you have an avatar, you have clothing and other content associated with that avatar you might want to have access to. And the friends or contacts list. And you want to be able to take that avatar from one application to another.

At this point, it starts to look a lot like the OpenSim hypergrid.

And, in fact, the OpenSim hypergrid is currently best positioned to become that platform. It’s fully open source, completely scalable, with a big ecosystem, both professional and user-created content, lots of hosting providers and builders and developers and other vendors, and a broad base of users and developers working on the platform.

The graphics aren’t great compared to high-end video games. But if you look at the Web itself, the graphics don’t compare to high-end games either. Games can download most of their content ahead of time, so users get a fast interaction when they go online. Web sites have to be able to load all their content very quickly, since there’s no guarantee that users have already been there and had a chance to pre-cache some content. Similarly, OpenSim regions have to be able to load quickly for new visitors, and even with the graphics we have now, the loading time is rather reminiscent of the loading time of early Web pages.

If OpenSim drops the ball, and is slow to add full support for virtual reality, then it opens the way for other companies to step in with their own alternatives. And those alternatives might be proprietary. And who wants to live in a Facebook- or Microsoft-owned metaverse?

 

Related Posts

maria@hypergridbusiness.com'

Maria Korolov

Maria Korolov is editor and publisher of Hypergrid Business. She has been a journalist for more than twenty years and has worked for the Chicago Tribune, Reuters, and Computerworld and has reported from over a dozen countries, including Russia and China.

  • Chic Aeon

    Very interesting. My only historical comment is that Cloud Party ran in Chrome and so Chrome (despite its disregard for privacy) has been in the forefront. Cloud Party had different rules but it worked well within those rules. One of my favorites was that when you added say a TREE to your build, it only counted against your limited resources once. So you could have a forest for the same “cost” as one tree. That was pretty fantastic.

    Now with cycles rendering there are so many better ways to texture and keep the downloads “light”. It all helps and I look forward to seeing where this goes.

  • In your conversation with Doug, did he mention anything about what kind of *licensing* their WebGL viewer might have? I forgot to ask about that in our meeting with him, and it’s not particularly obvious that this would be of general benefit to the generic OpenSim community as a whole, after the lack of cooperation his team received, leading to them forking and going their own way. Of course I’d love to see an *open-source* WebGL viewer, written from the ground up in JS and HTML5, but I wouldn’t start counting on that just yet without a direct confirmation.

    • The MGM that they’re using as the base for this is distributed under the MIT license. This is similar to the BSD license used for OpenSim, and allows derivative proprietary commercial works.

      It is very different from the GPL license used by the standard Second Life-compatible viewers, which require that all derivative works also be distributed under the open source license.

      The license issue has, in the past, been a significant obstacle for some viewer developers.

      • But I don’t have confirmation from them that this is, in fact, the license they will use for their viewer.

      • From what I can see, they aren’t using MGM as the basis for the web viewer. MGM is written in Go. This web viewer would be developed in JavaScript, HTML5 and WebGL. So if anything, it’s the other way around, and from the other quote, it seems clear they hope to provide (as in _move_ or rewrite) MGM functionality to be within that new web viewer, e.g. if they have a web viewer, it would probably provide any new MGM3 implementation.

        I don’t really think there is a strong connection between MGM, and that’s confusing the article a bit, at least to me. Or more simply put: I believe any web viewer would be a complete separate project from the MGM admin web interface, but that they hope to eventually integrate the admin interface into the web viewer. In other words, I think this is mostly a story about their admin interface, not a web viewer. I could be wrong on all this, but they aren’t even written in the same language.

  • Dharma Galaxy

    Given that 3d movies’ audience share has been flat, and even gone down slightly over the last 4 years I might suggest that the hype over stereoscopic is just that — hype. I have no use for it myself.

    • Many people confuse virtual reality with 3D. Although virtual reality includes a 3D component, these are actually not the same thing.

      When watching a 3D movie, you’re sitting in a movie theater, surrounded by other patrons, or on the couch in your living room, with a screen in front of you, and something seems to jump off the screen at you. Except for a rare few movies — Avatar, How to Train Your Dragon — I agree, this is mostly a gimmick and I don’t usually pay extra for 3D movies.

      When experiencing something in virtual reality, you feel as though you are no longer in your living room, but that you are someplace else. It’s more akin to the Star Trek holodeck than a 3D movie. It is a dramatically different experience, a new medium, and it will transform human civilization even more than the Internet has. This is why when Zuckerberg tried it, he immediately reached into his pockets and threw all the money he had at it. (Yes, he walks around with $2 billion in his pockets. LOL)

      If you get a chance, the best experience right now is with the HTV Vive headset. The graphics are the best of any headset I’ve tried so far, and it’s super immersive.

      Yes, the technology is still clunky. The user interface still sucks. The headsets are too heavy and awkward. But remember that this is day zero for the technology.

      Again, this isn’t an existing medium with a new bell or whistle added to it.

      It is a completely new medium. It is very rare in the history of human civilization that we see the creation of a new medium.

      The internet just took existing types of content — text, images, music, and videos — and made them easier to distribute.

      Immersive virtual reality goes way past that. It allows people — programmers, artists, filmmakers, storytellers — to come together and create full-fledged experiences. The experience of riding a roller coaster. The experience of visiting a museum. The experience of traveling back in time, or to Mars, or to the inside of virus cell.

      OpenSim expands those creative possibilities to non-professionals. Anyone can use it to create, say, a virtual reality cafe where musicians can perform live as their avatars in front of an audience of other avatars from around the planet. Or create a virtual business office or classroom. Or a virtual art gallery. Or a virtual beach.

      I personally believe that combining the creative potential of OpenSim, with the immersive experience of virtual reality, offers great social and economic benefits. And I anxiously look forward to seeing it happen.

      • Dharma Galaxy

        You don’t seem to get my point. I’ll start using mobile when I can type with my hands in my pockets head up. Sitting here at home the major aspect of Opensim is chat, and the graphics — while important — are completely secondary. Having to lock myself into a headset in order to dance and chat just won’t cut it. The words are what counts, and realism is simply not desirable.

        • Ah, I see what you’re saying. Yes, the typing interface is going to be a big issue for virtual reality. Most of apps that are popular right now don’t require any typing.

          And typing is problematic for a second reason, besides just the fact that’s hard to type when you’re wearing a headset (especially a mobile-based headset without a keyboard!) — it’s also hard to read the words in virtual reality.

          For folks who prefer to use voice in-world, this is a non-issue, but I know a lot of people prefer to type, for a variety of reasons.

          I think you’re right, and the screen-based, text-chat approach will stick around for a long time as an alternative to immersive and voice-based.

          • Dharma Galaxy

            I kind of doubt the “as an alternative” part. Immersive will be fine when I can become a head in a box with toasters having more computing power than the NSA currently has, and forget this silly human existence. Until then it is not going to fly. (And I’ll still be using the “typing” part of my mind then so I can leave the vocal part listening to music.)

          • I used the phrase “as an alternative” because there are already more users of mobile-based headsets than of Second Life and OpenSim combined.

          • Da Hayward

            i think i have to agree with Dharma on some of this, although this really great and amazing technology there will always be a core of people who prefer the current set up within Open Sim and indeed Second Life.
            This is by no means that virtual reality is a no goer. There is a lot of people who will benefit by it, but also a lot will benefit and are by the present set up.
            Im always wary of headsets because of some risks whether real or not to physical health ie triggering epilepsy in some individuals, granted it is very early days and no doubt these concerns will be adressed in the future but for now, I cant see myself moving to headsets.

  • I’ve wanted web based worlds for a decade now and this is exciting to see – minus the part the military is the one working on it, which will probably become a nightmare for people at the end of the day what with the spying and etc.

    But I am losing much faith in open sim that it’s not user friendly at all. Not sure how the browser/os platform would be implemented but if it’s not intuitive and quick without having to join 13 forums to ask questions on simple basic use (and never get answered because some content creators think what they do is top secret or over the top valuable), there isn’t going to be any real future for open sim. As I pointed out awhile back elsewhere, Unity, UE, and other 3d mesh building programs – many free now – make it easier to create worlds/levels and games, there’s a huge active community that puts up current, helpful tutorials, tons of material on all aspects – and a huge collection of content both free and paid – where content creators have a realistic perspective about reality and people….none of which open sim has or ever has had.

    If the devs don’t step it up and make it user friendly out of the box, intuitive, and with some quality content so the avis can look awesome without people investing a lot out of the gate, and all the tutorials are updated, open sim isn’t going to be able to compete.

    Unity has the ability to get an open sim/second life group of people/avis/npcs and eventually those devs will work out a solution that does what open sim does with live collaboration and building…and they’ll provide proper instruction. That’s the real difference. There’s a hella learning curve with these game engines and development software but the people using them make the learning curve way easier by keeping things fresh and updated, and current.

    Case in point. Open Sim’s website is pitiful, boring, and has way too much technical stuff nobody but techhies will even bother reading. Second Life’s website shows what SL can do, what it has, it’s eye catching and elegant and professional with quality imagery. Even though people have to pay for stuff, they see the end result and SL wins. Open Sim’s website looks like whoever put it up borrowed something out of 1995 and hasn’t bothered to update anything, even the mistakes.

    I get it that the technical sorts will not care but the point here is mainstream public – they will NOT embrace open sim at all in its current presentation and format, they don’t care at all whether there’s a browser version or not. They’ll end up with SL or some other thing that’s commercially appealing.

    I hope they bridge the gap but it still seems the devs are not interested in making it user friendly, commercially appealing, intuitive, and with clear, easy tutorials created at least in 2014 at the oldest…instead of 2005.

    I check HB regularly and it makes me sad that VR is leaving OS behind and whoever’s devving OS can’t get on this train. 🙁

  • lord

    eh, not sure what the big deal is anymore…RealXtend has already done WebGL viewer for their group of Tundra based ‘Experiences”…shouldn’t be that hard to take their viewer and cannibalize it to work with OpenSimulator (not to mention that someone has already shown how to use a SL-based viewer on WebGL: http://nwn.blogs.com/nwn/2015/10/second-life-webgl-viewer.html ).

  • Nick Zwart

    So, any progress on this? Any news?

    • There’s this, but I don’t know what it means:

      https://redmine.militarymetaverse.org/projects/opensim-webgl/wiki

      https://redmine.militarymetaverse.org/projects/opensim-webgl/wiki/Halcyon_Grid_Overview

      I was hoping I could get someone to explain to me, but haven’t had the time yet to start looking.

      • This is their 3 stage plan. They are basically at the Crawl stage still:

        Crawl
        ——-
        The crawl stage is used for the development of control streams and to flesh out apis. It is a low bar, with discovery and integration with Opensim being the difficult component. The features are:

        Authenticate users and allow for some sort of presence in-world

        Present list of in-world users to the client

        Participate in text-based chat (In-world and IM)

        (bonus) Participate in In-World voice communications

        Walk
        ——
        The walk stage is a means of fleshing out and testing the primary functionality of the system. The features are:

        3-d diplay of the terrain

        Display facsimile of In-World objects, with real-time updates

        Display facsimile of In-World users, with real-time updates

        User control of In-world Presence

        Basic User-World interactions (i.e. On-Click)

        Inventory Access (No Rez/take)

        Run
        —–
        The run stage is the realization of a functional application. The features are:

        Full 3-d display of in-world assets and animations

        Full 3-d display of Avatars

        Full Inventory Access with take and Rez

        Scene Editing Capabilities

  • Nick Zwart

    it does not include a web based viewer as far as I can see. Pity.

    • It says “Sentry is a new application written to enable webgl client access to Opensim… Sentry is a plugin-less zero-install client for MOSES/Opensim based virtual worlds that targets modern html5/webgl enabled browsers. It is written targeting Google Chrome, but Firefox, Safari, and Edge support may be added …”

      Isn’t that the same as “web based viewer”?