Army to release browser-based viewer for OpenSim

An OpenSim viewer that runs in a regular web browser has long been at the top of most-wanted lists for OpenSim users. And the U.S. Army is stepping up.

This fall, the U.S. Army’s Simulation & Training Technology Center will release an HTML5 viewer for its MOSES project.

Douglas Maxwell

Douglas Maxwell

“The first version of the web client will be view only,” said Douglas Maxwell, the center’s science and technology manager for virtual world strategic applications. “Understanding the plumbing for what it would take to create a web client was a tremendous undertaking.”

The new web-based viewer should be accessible on any device that supports HTML5 and WebGL, he added.

“We have not focused on optimization yet, so I do not recommend the early users try to connect using an older iPad or smartphone,” he said.

The new viewer uses the Babylon JavaScript framework, used for building 3D games, and which has complete support for “full-immersion stereoscopic viewing and virtual reality head gear” according to the framework documentation.

For an example of what’s possible in the Babylon framework, check out this Assassin Creed Pirates demo.

Still from Assassin's Creed Pirates demo by Ubisoft.

Still from Assassin’s Creed Pirates demo by Ubisoft.

And the graphics capabilities are ahead of those of existing OpenSim viewers, Maxwell said.

“We believe it will be easy to convince the users to switch given that the new system will look so much better and the interface will be familiar,” he said.

“In our investigation, we were able to create stunning graphics with low overhead on the client hardware,” he said in a comment on a Google Plus post by Talla Adams about moving beyond the Second Life paradigm. “Further, the MOSES web client will be tunable, meaning we can detect what kind of client hardware you are using — mobile, laptop, desktop, GPU, etc… — and make sure you don’t have to draw a scene too heavy for your gear.”

The user interface will be customizeable, as well, he said.

“Since this is all JavaScript, anyone can come along and change the way the UI looks for their grid’s needs,” he said. “It can be as simple or complicated as you like and even have different functionality based on the role of the user.”

Building tools will be added later, he said, and scheduled to be released in the second phase of the rollout next spring. Until then, creators will still need to have a third-party viewer such as Firestorm or Singularity.

Since OpenSim isn’t designed to communicate via the web, it requires a separate piece of software, what Maxwell calls a “data arbiter.”

It sits between the web viewer and the Halcyon server, the version of OpenSim used by the MOSES grid.

“We had to create a way to efficiently shuttle data between the Halcyon server and the arbiter using protocol buffers,” Maxwell told Hypergrid Business. “You can think of the arbiter as a new kind of alternate client manager.  This heavy responsibility will eventually be removed from the OpenSim server altogether, enabling scalability.”

The existing client manager, which works with Firestorm and other third-party viewers, will remain in Halcyon for the time being.

However, it will eventually no longer be needed as the build functionality moves to the web viewer. The way that building is handled will change as well, Maxwell said.

“We plan to leverage more of the PhysX engine’s capabilities,” he said, referring to the physics engine used on the MOSES grid. “The most noticeable impact to the build tools will be in a change to how linking objects works.  We will introduce masses, offset linkages, joint limits, spring constants, and more.”

The new viewer will be released as open source.

“It will work in Halcyon, but anyone can take it and modify it to work with their flavor of OpenSim,” Maxwell said.

Related Posts'

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. Follow me on Twitter @MariaKorolov.

100 Responses

  1. And here’s a Babylon demo in VR — — on my smartphone.

  2.' Carlos Loff says:

    I could write about this for hours but I will sum it in few words – YES YES YESSSSSSSSSSSSSSSSSSSSSSSSSSSS

  3.' SkyLifeGrid says:

    I really would love to get my hands on a headset. Unfortunately I use Iphone and am not sure of a good buy. Maria any recommendations for a decent headset for Iphone ? I really really want the Rift but am worried to buy as they keep updating and, I don’t want to buy the Beta products

  4.' Nick Zwart says:

    Very good news. Can’t wait for fall to come. Anything about the voice plugin?

    •' Douglas Maxwell says:

      I apologize for missing this question, Nick. Voice will be treated as a separate service on a separate server than the web client work. We are still in the architecture and testing phase of candidate voice technologies. Our technical requirements are such that we need to be able to run with full functionality on a stand-alone network, so a service such as Vivox is not appropriate. I also wish to keep the project 100% open source and not attach dependencies that require licensing and royalties. Our functional requirements include the standard voice operations such as local broadcasting with distance fall-off, avatar-avatar telephony, and an additional requirement to mimic radio comms with simulated channels …and it all needs to be encrypted.

      We have a lot of experience with Mumble/Whisper and we currently use freeswitch. However, the most promising encapsulated solution looks like webRTC with hardware compression. Node.js is our friend! Keep in mind that all of this is currently planned as separate from the data arbiter and will likely even run on different hardware on the backend.

      Right now, we have a “distributed is better” mentality and are trying to spread the load of this architecture across as many virtual machines as possible to promote scalability and stability.

      •' Nick Zwart says:

        Hi Douglas, you really do not have to apologize, I know you are very busy so a double thanks for this reply. I do a lot of educational work on language learning and talking in virtual worlds so interacting with voice is always my main interest. Thanks for all the effort.

      •' ManBoy says:

        Hello Douglas. Will the MOSES client be built to work across all mobile platforms?

        •' Douglas Maxwell says:

          Any mobile platform that has a browser that is capable of running the Babylon.js & Node.js java scripts should be ok. A modern HTML5/WebGL compliant platform is required.

          I do need to caution you on initial performance. Lower powered machines are not going to do well. There is no free lunch, so even though this is running in a browser, you still need a GPU for best results. We haven’t performed benchmarks yet, but as soon as we get some numbers I’ll toss them over to Maria.

          Lastly, scene optimization will likely be something we tackle later in this development. All we ask is that early adopters keep an open mind and keep any performance criticism constructive.

          •' ManBoy says:

            Be careful with that. I’ve known of webGL virtual world projects that have halted because they tested on mobile devices too late in the stage of development. Mobile devices have much different architecture and a lot of webGL will simply crash their browser or produce funny side-effects. There’s no webRTC in webkit (though Apple are planning on developing it). That means you won’t get VOIP within a browser happening anytime soon. Also when iOS GPU renders images it decompresses them to their original size which can fill memory easily and crash the browser (iPhones don’t have much memory to begin with!) Also, there is a significant limit on the number of GLSL uniforms with iOS which limits the number of bones an avatar can have. You either have to remove finger bones (like the avatars in my virtual world) or use the CPU for animation. There are also shaders that can have strange side-effects on iOS that gives you a flickering screen. I haven’t even touched on areas like 4G lag, lack of keyboard and mouse, limited screen-size, responsive design, dealing with limited quotas via caching, etc… Mobile device compatibility may not be a priority for MOSES, but if it is, make sure you test early

      •' Susannah Avonside says:

        This is interesting, as I am looking for a voice implementation for one of my projects for which Vivox is completely unsuitable, as it’s commercial, and could easily eavesdrop on sensitive information. I know there was the Whisper plugin that used the Mumble server, and that a couple of years ago Timothy Rogers of Zetamex attempted to crowdfund an update, but seemingly failed, as most use Vivox, and were seemingly quite happy with that solution.

        I will watch this space!

  5.' Mac McMacintosh says:

    Question: Will this work with Second Life, too? Or limited to OpenSim only?

    •' Douglas Maxwell says:

      We made the deliberate decision after months of consultation with the community and developers that deviating from SL compatibility was the best way to go. In fact, this viewer will not work with “vanilla” Open Simulator out of the box. We are modernizing the way the clients and the server communicate with each other. What I mean by “modernizing” is that we are no longer sending clear text messages back and forth, but rather changing the protocols such that they are both binary and encrypted.

      •' Mac McMacintosh says:

        Thanks for the info 🙂

      •' Talla Adam says:

        Thanks for the extra info, Doug but I want to confirm what you are saying, that core Opensim (vanilla) wont work with the web viewer your team is building so am I to understand my existing Opensim grid will need to be run on a Halcyon version?

        If so, will the standard viewers also need updated to run the Halcyon version due to the change in protocols?

        Apart from that you have said at Opensim Virtual that Hypergrid is not being worked on as part of the Halcyon project but others could develop that separately. Can you confirm that too? Thank you.

        •' Douglas Maxwell says:

          The special sauce missing in core Opensim is our data arbiter. This is required to interface with a javascript based client. To make the web client work, the data protocols *must* be altered.

          Standard viewers will still connect to the Halcyon servers with web browser enabled code. This is what I meant by keeping the client manager code in the open sim server for the time being until we get build tools into the later versions of the web client.

          I don’t have the resources to eat the whole elephant at one meal. The high priority work is being tackled first and Hypergrid-esque functionality will come later. Unless someone steps up and takes this task on, we wouldn’t even consider working on that until next summer.

          •' Minethereé says:

            “Halcyon servers” what does this mean Doug? People can still run their own if they wish to right?

          •' XMIR Grid says:

            Halcyon servers for now means a different scripting engine, a different physics engine, a different database setup, no hypergrid support, most likely no standalone as we understand it in OpenSim core, and now also a different viewer protocol in offing. (It is essentially the Inworldz setup).

            I don’t say it cannot be consolidated, but it is not trivial. There is also the question who would do the integration?

          •' Minethereé says:

            yes, thank you xmir

          •' Douglas Maxwell says:

            Anyone may run their own Halcyon servers if they wish. More information can be found here:

            The Halcyon GitHub page is here:

            We run standalone versions of the Halcyon code base at multiple locations in our labs and in the field at training sites.

          •' Minethereé says:

            Ok, thx Doug. So for my own (and many others) purposes, it appears there will be two major opensimulator types…core as it is now with the Avination merge still being tested…which includes the hypergrid functionality as well as varregions…and the halcyon/MOSES one which has no hypergrid functionality and no varregions.

            The latter you are working on to “military harden” as you mentioned someplace or other, and some kind of stronger perms that commercial grids may like.

            The former will be moving on as it always has and has a large base of users, and is growing.

            Whether or not, for the basic user, one type or other will interest them, remains to be seen of course, time will tell…but obviously the browser code is a win-win regardless, it just needs to be optimized for core opensimulator, as others are saying also.

            I think that puts it succinctly.

  6.' Carlos Loff says:

    View Only means that we can not interact with objects, not event just touch ? As a builder Im ok in go on using the classic viewers but it would be nice to allow visitors to touch and interact with stuff just by using HTML, guess we are not there yet, snif snif

    •' Douglas Maxwell says:

      Pls keep in mind that this is a phased roll-out and there is intense interest in the capability. We want to make a good impression and balance that with getting the community something functional as early as reasonably possible. From an engineering standpoint, this is a tremendous undertaking. Managing the draw calls alone is an engineering feat. The value proposition from releasing this to the community is your assistance in testing and feedback. It won’t be perfect, but I welcome your constructive criticism and you will find that all earnest inquiries will have responses from either me or one of my engineers. Thank you all for your encouragement and support.

      •' Carlos Loff says:

        Sure, you can count on me for all testings, even if you want to use my huge lands and builds for testing, I think you are pioneering something that will change all our lives, democratising access to Virtual Worlds and bringing many folks that otherwise would not bother to come – Cloud Party, around some 5 or more years ago was using a fluid HTML access, including interactions with objects, is it some way you can gather data from that experience, I had an account and used it, it was so ahead of it´s time

  7. The best news from opensim this year for me! Thanks Doug, this is just awesome.

  8.' Anon Y Mous says:

    Seriously folks this is huge. Even without the encryption part this is huge. Since the beginning Opensimulator has been forced to play catch-up to Second Life; however, in Second Life the viewer and server are part of the same project so functionality can be shifted easily back and forth between viewer and server, leaving Opensimulator always trying to catch up and never quite getting there. This totally levels the playing field and shows a way for Opensimulator to move forward on its own. Yay and stuff! And about time for the encryption part.

  9.' John Simmons says:

    Divorcing from SL in order to bring Opensim up to speed is overdue. Hopefully the improvements in Halcyon can become merged in main Opensim. MOAP needs to use chromium to be able to use modern web apps. Updated framerates and stereoscopic views when possible are great, but “web client will be tunable, meaning we can detect what kind of client hardware you are using” is essential if Opensim is ever going to get wide adoption, especially in the third world. This is injecting a new level of hope into Opensim.

    • I’ve got to second that “new level of hope” thought. This is definitely cause for optimism! And, I, for one, would be willing to chip in so that the MOSES viewer can be adapted to work with mainline OpenSim — or towards adding hypergrid capability to Halcyon. Either way!

      •' Dharma Little says:

        Are you sure? without setting up a non-profit foundation that’s accountable and able to take in large donations to pay developers while offering a road map then Halcyon will continue to be what it is ‘A project for Moses’ did you read what XMIRE Grid wrote? If no one wants to create an accountable system but then again that would fail as well without setting up as legitimate transparent non-profit foundation with a road map and able to direct the needed changes.
        developers are not going to work for free this time around and even with crowdfunding your not going to get the kind of business/government class grants a legit recognized organization would get.
        It is very sad to see the same mistakes made 10 yrs ago repeat themselves instead of the LL/IBM Code now its Halcyon.

        • There’s been a long history in OpenSim of successful bounty-based projects. You need something that’s well defined, with clear goals, and a trusted person or organization to provide the bounty. Once the MOSES web viewer is released, and if the process of porting it over to mainline OpenSim is a reasonable one, then I think a lot of folks (including me!) will get behind it. And even if we don’t, all that means is that we’ll just wait until Halcyon gets hypergrid support. I know it’s down the line, and they have other priorities, but the MOSES team gets stuff done.

          There are already several versions of OpenSim that grids are running. Some versions of OpenSim have better hypergrid support than others, sure, and that’s an issue if you’re running a hypergrid-enabled grid. But there are lots of different takes on OpenSim out there. Kitely runs all their regions up in a cloud and has their own unique system for doing a lot of stuff, but they’re still on the hypergrid.

          And since Halcyon works fine with current third-party viewers, I’m hopeful that it’s not so different from standard OpenSim, and that a port of the web viewer will be straightforward.

          I’m with you on the foundation, though. I would like to see OpenSim get more organized and set up a legal structure, and do project planning and roadmaps and all that other stuff. But I figure that I can’t complain TOO much because am I helping do that? No. Am I putting my time into getting things organized? No. Am I helping write documentation, or helping out with testing and debugging, or any of the other million tasks I could be doing even without going out and learning C# and JavaScript and whatever? No, I’m not.

          Do I feel guilty about that? Yes, yes I do. Do I feel guilty enough to start going out and helping set up project teams and organizing meetings and creating agendas and writing documentation? No, no I don’t. Do I feel guilty about that, too? Yes, yes I do. Especially because I LOVE running meetings and creating agendas. I love setting up processes and doing the paperwork and all that bureaucratic stuff. I think I better stop writing before I talk myself into something that involves giving up my addiction to reading science fiction books and watching funny animal videos on YouTube…

          •' Minethereé says:

            I just read the last 3 books from “The Wheel of Time”…I found them online in pdf form for free. It was quite disheartening when the final sentence was something to the effect “this is the final book of The Wheel of Time”.

            I had been reading them for so long it was strange.

            There will never be more as Robert Jordan died and his notes formed the foundation of the last three books written by Brian Sanderson.


  10.' Serene Jewell says:

    “The high priority work is being tackled first and Hypergrid-esque
    functionality will come later. Unless someone steps up and takes this
    task on, we wouldn’t even consider working on that until next summer.” – Doug Maxwell

    Perhaps this is where the community can step in and make a difference. Hypergrid functionality seems to be seen as an insecure, customer-distracting, annoyance by many commercial grids. And a low priority for the army. But without hypergridding, there is no metaverse, just a bunch of standalone silos. Is this possibly where we can crowdfund some development (and documentation?)

    •' Talla Adam says:

      I think it is a good candidate for funding Serene but we will have to wait see what the web viewer looks and acts like I guess. I certainly see huge benefit in a “First Look” web viewer with Hypergrid connectivity in as much as it could bring in new people from the greater web rather than always looking to Second Life for recruitment – I really don’t think SL is cultivating that many new users anyway now and, before long, a lot of existing users may well move to Sansar or High Fidelity even. But Sansar and HF I think wont be too much like the SL/OS Metaverse anyway. Their aim is hands-free virtual reality. HF even dropped it’s chat service on screen on the premise keyboards wont be used (not so sure of that). But Opensim should have a great future if a Lite web viewer reduces the learning curve – Building, etc, can still be the preserve of a full viewer while the web viewer is all about showcase and easing newbs into the space. What I especially like is that the MOSES viewer can be viewed on multiple devices and will be open sourced. Grid owners can adapt a version for their own needs too because of that. So yes, Hypergrid is essential for our community and I have in mind to put aside at least $1000 for crowd funding in the coming year and I would really like to see others join in like we have done in the past. It would be great if some coders stepped up to the challenge as well and start talking about the technical aspects.

      More on this topic at G+ Opensim Virtual

      •' Dharma Little says:

        Sorry to be the protagonist here but any fork of Halcyon in the name of vanilla OpenSim is going to fail and here is why!
        1.OpenSim needs a road map!
        2.OpenSim is not set up as a non profit business to take in the kind of donations needed to pay developers
        3.OpenSim has no One place for documentation,learning,support like high fidelity or SL
        4.OpenSim developers are thin at best while only doing projects they like not whats needed done.

        Create a non-profit business to replace OpenSim!! paying for development,regulations,tax’s,research,documenting the code,advertisement/promotion,website,24/7 support,GitHub,SOAS 2.0, integrated viewer support and team all headed by verified real identity’s that can be trusted to carry out the road map at hand.

        Use crowdfunding not for forking Halcyon but for creating the non profit business to replace it as OpenSim 2.0 and its the only way or you will be back to stage one in no time flat.

        Going to start with fresh code you should have fresh paid developers as well in a transparent easy to access and communicate with public program.

        you have one last real chance working with a clean robust code base so please do not blow it with crowdfunding attempts with the heart in the right place but the lack of business sense to carry it thru this time around.
        some people are going to have to step in, real life ready to do the above plan
        when OpenSim was created they had the best of intentions too and dreamed of something beyond second life but look how that dream was neglected or sidetracked almost a decade later.

        •' Talla Adam says:

          I’m listening to people like Douglas Maxwell and I am not trying to arrange any forked Halcyon. We have a lot of simulator code one way or another anyway and quite a few forks of Opensim. The only real question coming out of these discussions for me is that MOSES will open source their web viewer code but they can’t be working on Hypergrid connectivity before a year or more. We don’t need any organization to pay a developer to work on that particular part of the MOSES application or Halcyon once it’s open sourced. One or more developers just needs to step up to the task after they’ve seen the code and talked to us about what might be done. In the past we got things done simply by saying clearly what we wanted and a coder did the job. Each of those who pledged money personally sent payment of their share after the code proved working and sound. What gave confidence to the developer they would get paid is that a personal guarantee was given. And since the last thing we pledged on through G+ Opensim Virtual got paid without hitch (well over a $1000) I think our word is pretty good.

          Of course, if someone or group came forward with a proposal to organize crowdfunding that too would be worth considering but I’m looking for a one-off fix here and a promise of maintenance up to a point. We had that much already to base this on.

          As for the whole Opensim project, yes, it has been messy at times and there have been long delays but it didn’t stop the community growing around it or businesses like InWorldz and Kitely making money from their version of it as, indeed, many others have too. And all the while a thriving sharing community came to exist as well. Opensim has been a remarkable development for all it’s short comings and here we are still talking about how to take it forward.

          I’m confident we can get something done if it can be done.

          • Talla, you’re totally right. There’s no point in doing work on Halcyon if MOSES is going to do it eventually anyway.

            And if we can just adapt the new Web viewer code to work with mainline OpenSim, it would probably be a lot easier and cheaper anyway, and everyone can keep their grids the way they are. You did the last bounty, with the llLookAt fix — — and it worked out great.

            Do you want to work on the new one — once we know what we’re working with, of course? I guess it would take a developer to look at the Web viewer code once its released, maybe talk with Douglas Maxwell’s team, and figure out what it would take to port it over to mainline OpenSim. Then I’d be happy to pitch in, and do what I can to help promote and publicize the project.

            I think a lot of the commercial grid owners would want to get behind this, since an easy, web-based viewer should help attract new users, especially people outside the normal Second Life-OpenSim crowd.

            And count me in to pitch with some money, as well. This has been a huge thing I’ve been waiting for, for several years, and have had my hopes raised then dashed with Tipodean ( and 3Di ( and realXtend. It would be great to have something working.

          •' XMIR Grid says:

            The problem Maria, is that you somehow assume there is a) an interested to port code from Halcyon to OpenSim core and b) someone capable of doing it.

            The reality is that, despite the offering of the MOSES project to contribute to this, they have for all practical purposes been shouted out of the building with the door slammed in their face.

            Currently, there is close to one single person making contributions to OpenSim core through the avination merge. Despite his close to heroic in effort, as it is, some aspects of current development branch is worse off than the 0.8.2 release.

            The only thing that can save the situation is a fork of OpenSim backed by a roadmap, funding and professional developers. Only then is there any point in throwing additional effort into a viewer development. What MOSES have done is exactly this.

          •' Talla Adam says:

            Yes Maria, I’ve have waited for years as well for a web viewer for Opensim and hopes have been dashed time and again with all the projects you mention. I would mention pixie viewer too and that was probably the best candidate to date. Its virtual world was an Opensim instance and it functioned on the web very well. Thomas Buchaur (Sunny Salamander in world) showed brilliantly how it can be done ( and I recall visiting the online app, playing with simple building tools and looking in a mirror – I mean, like a real mirror!

            Unfortunately, Thomas just seemed to disappear and even after promising to open source the code at an Opensim Conference where he did a demo presentation nothing came of it. I have no idea what happened so maybe he just ran into problems – we don’t know – or, like Cloud Party, Thomas may have sold up to unknown parties. As it happens I looked in on Marcus Milne, see his post below, last night on his web viewer which works quite well, not as polished as Pixieviewer I might say, but I was able to move about and chat with Marcus just fine. Marcus is looking for a buyer so he said but the price is high at $250k, which he justifies saying it has been a lot of work to get this far and he will be selling all the rights. Needless to say, I didn’t make an offer, lol. But I did suggest licensing which he wasn’t keen on. Anyway, it is another good example of what can be done with OpenGL/HTLM5. So, those dismissing any attempt to get a web viewer working for Opensim are flying in the face of the evidence.

            Referring back to your question about crowd funding, Maria yes, I am more than willing to run something under the Opensim Virtual name to get a version of MOSES web viewer adapted for simple “First Look” Opensim use in the public domain and possibly with Hypergrid connectivity, although, as you said, that may have to wait for Doug’s team to work on it. Nevertheless, once the code is open sourced and a practical application becomes available I see no reason why grids couldn’t start using the code to place a viewer window on the web site that gives visitors easy access to a welcome region for the grid in question and without any download. The visitor can be helped and advised that a full viewer could give them better performance, building tools and access the the many worlds of the Hypergrid during their visit (that should get lazy grid owners out of bed to man the welcome regions!). Seriously though, I think we could look at what MOSES produces and if work is need to polish the application enough for our basic needs then that is something I would be willing to support with a crowdfunder.

            I’m looking forward to what MOSES produces later in the year.

        •' Douglas Maxwell says:

          @Dharma Little: I’d say you were more of a realist than a protagonist. You seem to have a good handle on the frustrations we’ve had with trying to work with Open Simulator from a business standpoint. Here is a response I wrote to one of the Google+ conversations last week:

          Let’s start with the original question: “Just a hypothetical here, if a public or private entity decided to take initiative with opensim development and really fix things, what would you want fixed first?”

          It has been fascinating to read through the comments here and read the inputs from such a diverse set of perspectives. Personally, I tend to agree with almost all of you!

          To really “fix things”, you have to start at the top with the organizational leadership and structure. The ad-hoc loose management structure works fine for small startups, but as the project gains success and grows, then the management has to evolve as well.

          In the case of Open Simulator, entrepreneurs are attempting to make a living with the project but are beholden to a core group of developers who only make decisions when unanimous. Given that some of the core developers are competing grid owners, this sets up an obvious conflict of interest. This sets up a terrible decision proposition for non-core developer users – do they fork or do they accept the code as-is? It is our experience that the core developers are seized and can’t come to consensus on acceptance of even basic fixes from outsiders.

          Back when Overte existed, it was listed as a non-profit entity. I can work with that. When I contacted the core developers to initiate talks that would have led to grant funding from my program, they turned me down flat. I was told they didn’t want to deal with the finances or set up the accounting on their side needed to accept the funding. This is the first time in my 15 year research career that anyone has turned down funding from me because they didn’t want to do the paperwork.

          In reading the comments, I see funding amounts discussed in the $1k range. What is needed to advance the platform is much, much more funding than that. Bounties are nice for constrained targets, but a dedicated funding line for a goal-driven engineering effort is better.

          This is why I’m working with the Halcyon developers. They can 1) operate in a professional engineering environment and 2) accept my project funding. In short, they have their act together.

          •' XMIR Grid says:

            It has been a painful experience witnessing how your contributions to OpenSim was handled by the so-called core developers. Talk about being sabered down…

            Currently the only organization of OpenSim is a loose collection of individuals and OSGrid tugging each other in a non-coordinated fashion. It is like the scrum process from hell where no overall goal or structure exist.

          •' Minethereé says:

            In my desire to try and understand, I have read and re-read your comment, and still don’t quite understand, so I decided to just go ahead and do a comment on it, not all of it as I don’t want to get into some “long winded multi-persons” joining in thing.

            So while I am waiting for my face mask to dry (equal parts of plain oats ground finely –I use my coffee grinder–with honey and plain yogurt is the recipe for anyone interested)…and seeing so clearly how you default to non-drama, full disclosure and transparency and complete answers…which is why I would comment at all.

            And thank you so much for that Doug.

            I am in a quandary as to this statement, “In the case of Open Simulator, entrepreneurs are attempting to make a living with the project but are beholden to a core group of developers who only make decisions when unanimous. Given that some of the core developers are competing grid owners, this sets up an obvious conflict of interest.”

            Firstly, a conflict of interest doesn’t seem to be anything that should even be discussed. Since halcyon is an inworldz creation, if there is to be conflict of interest comments, they are also, aren’t they? Afterall this all gives their for-profit grid publicity and the owners (I assume them all) signed the 3 year contract you mentioned.

            Melanie of Avination (which I guess is still down) is helping with her and ubits merge of their code, which nobody has even seen happen yet, but people have no issue discussing something that is not even done and which it is proposed it be .9 version. She also has her hosting business.

            Others I don’t know if they propose code and make any money with it…I don’t think they do but I suppose it would be healthy to have some input from one or more of them, but which I doubt as they typically don’t involved themselves in this manner, preferring another method for discussions.

            It seems to me if some people want to make money with this, then more power to them…but I am seeing a bad trend from several people to show them in a bad light when in fact the work of core is what has gotten “everyone” to this point.

            The majority of opensim use runs on core.

            inworldz forked at .6.5 (I guess is how the decimals would fit).

            Absolutely everyone is beholden to the core code developers, especially the ones who try to make money with it. It hardly matters about one fork or another given that basic truth.

            So I, for one, will not forget and am very appreciate of their work that gives me something to do that I can enjoy…tho my own use is focused on the free Hyperverse mixed in with the growing amount of commercial grids who have opened their doors and enabled hypergating also.

            So any core devs and people who support them…Thank YOU!

          • Minethere — In theory, when it comes to OpenSim, where the conflict of interest comes in is when a developer creates a nice piece of code. Do they donate it to the OpenSim project? Or do they use it for their own grid?

            If they use it for their own grid and never share it, then they can profit from the new feature. BUT if other devs create a different version of the same feature and open source it, you’ll wind up with a difference between the grid’s own code, and the mainline code. That makes it more difficult to use the mainline code, its updates, security patches, etc…

            My recommendation is to create the feature, market it, profit from it, then when the marketing is over, or when you are about to release an update to the feature, open source your version and get into the main line code. That way, you continue to get the marketing benefits from your updates, while still being able to get the benefit of all the other work that other developers are contributing to OpenSim. Plus, OpenSim benefits because they’re getting cool new features.

            If you wait too long, then you wind up with a situation like InWorldz has, where their version is so far removed that they can’t just grab new OpenSim advances (like hypergrid or varregions) without significant additional work.

            I think InWorldz’ decision to open source their code (and Avination’s decision, too) is a great move by both companies. They’ve already gotten their sales benefit from the code, and now they get the PR benefit of making a big public donation PLUS they get outside developers to work on their code base.

            It has obviously paid off big time for InWorldz, because of all the work MOSES has been doing on the code base. And the general OpenSim community benefits as well, because if they’ve got a cool way to do something or solve some problem, everyone can see it and try it out and add to their own versions of OpenSim if it works well.

            Kitely has been doing this, too. They’ve got a lot of proprietary code, but when they make a fix to the heart of OpenSim, they donate it back to the community. They seem to be one of the fastest at donating the code, and even though it goes open source right away, their residents still benefit from it before everyone else because they roll it out first, and they get a reputation as a tech-savvy, public-spirited grid.

            So, overall, I think the commercial aspects of OpenSim actually help the open source project overall, especially in the long run. Developers get a chance to get paid for the work they do AND the community benefits from the improvements. Win, win.

          •' Minethereé says:

            I agree, and Kitely even added some nice hypergrid code…of course first for themselves, and then the rest of us using core….as you say, it is a win-win.

            I was just speaking to one thing is all.

            I dislike the growing propensity of people to dismiss the core developers and development.

            Of course the conversation is among a small group in relation to the much more vast, and mostly uncounted, user base using core.

          •' Kara Kelly says:

            So what is pretty clear is OS core devs don’t give a rat’s … what happens with OS. I say fork the code, come up with a current, relevant platform and get funding for it by whatever means is necessary and let’s wave goodbye to OS in its current form. The devs aren’t receptive to growth, which is WHY OS is in the place it is. It’s unfortunate but these evolutiions are showing the forks are doing what needs to be done.

            I’ll use OS and tinker around but more and more I’m moving away from it for game engines and such and while there’s drama and nonsense there, their core devs actually care about their projects and keep up.

            OS is wonderful and deserves better…so if it’s not doable to just take it out from under them and fix it, then let it die and let the devs squabble amongst themselves and release all of the community to move forward.

      •' Kara Kelly says:

        I may be misunderstanding this so feel free to reset it for me – with regard to the hypergrid/metaverse. If there’s a web based viewer at all, the world is, for lack of better phrasing, “in the web page”…so why wouldn’t it be functional as a metaverse the way it already is in 2d web? Our worlds are viewed in the browers/”web page/site” and people can load it up and explore it with the viewer and ultimately VR.

        Am I not understanding this?

  11.' Talla Adam says:

    That looks pretty good and it shows what can be done.

  12.' Minethereé says:

    Interesting. People should look into this more.

  13. “Army to release browser-based viewer for OpenSim” awesome, more time for me to work on other stuff.

  14.' ManBoy says:

    I’ve had quite a few lovely people drop by and chat to me in my virtual world recently. I assume most are from the openSIM community. I actually built this prototype to show off my 3D scanning and development skills since that’s what my startup specialises in. I used Babylon.js because I’ve been close to the team. I’ve contributed to that framework and they’ve also helped me customise it to have finer control over areas such as joystick and avatar animation. If you decide to drop by and see someone called Marcus Milne (that’s me!) feel free to say HI

  15.' XMIR Grid says:

    In my native language we have an expression for what transpires: “De glade amatørers vanvidd”.
    Not sure if it translates well.

  16.' XMIR Grid says:

    I would, but I am going to have major surgery at the beginning of next month, so I’ll have to get through that before I can put effort into it.

  17.' Talla Adam says:

    I should take it as a compliment that you, Dharma Little – who ever you are? – considers that Maria and myself are holding the fate of Opensim in our hands but I know no such compliment is intended when, in your own words, you are dismissing our views and intentions as amateurish and wishful thinking. Yes, a “tech web site editor” and a “private small g+ group” – one with 1800 members together with many other g+ groups that both reflects and promotes much of the Opensim Hypergrid community – but I wont dwell on the efforts either I, Maria or many others have put into supporting community affairs.

    You are right though about core Opensimulator seeming to be lost after the departure of people like Justin Clark-Casey but then it always was a loose-knit affair with no one claiming to lead. Heck, even OSgrid, claimed to be the test bed for Opensim, never had charitable status until recently. The community voice was mostly dismissed too, like you are doing, as a load of amateurs. But all those amateurs have grown into quite a successful community of Opensim and Hypergrid users. Of course not as big as Second Life but enough to finance and build over 200 active grids, and, for the most part, while completely independent of each other, they share a common Highway over the Internet and engage each other both as Hypergrid travelers, a common market, and through the social media of Google+ in particular.

    I have already said above that if someone or group came forward with a proposal to organize crowd funding that too would be worth considering – it doesn’t have to be a loose arrangement like I might do under Opensim Virtual for one-off fixes & features. It would be equally worth considering backing a properly registered organization that might go after venture capital though I would be skeptical of any particular grid taking such a lead. We have that now, up to a point, with Melanie Thinkler and the Avination code merge so I would suspect there is always vested interest at play. If a bunch of coders forked OS 8.2 that might be more to my liking. But what MOSES is doing with of Halcyon to provide a web viewer is something else altogether, something to wait and see.

    In fact I think “Wait and see” is the bottom line in this discussion now. 2nd gen platforms are on the horizon. Soon Sansar and High Fidelity will be competing for both the Opensim and Second Life users. Linden Lab are bleeding the SL users dry for every last penny as the old cash cow shrinks so they can finance what they hope will be platform that saves their corporate arse. High Fidelity needs a community and they’re knocking on the door of both SL and OS to give them that all important startup bunch of pioneer users until they can fully pitch for the more younger, more numerous, VR brigade. However, as has been seen, SL users don’t budge easy. OS has gained a small share over the years but many remain both in OS and SL like it was still one big community. Hypergrid is Opensim’s best feature and continues to attract new users and startup grids that see the advantage of the expanding market for virtual goods HG offers. Kitely understood this from the word go and worked hard to build their own export system and market on the Hypergrid. I would love to see a determined effort to build an Opensim 2 with voxels and more but I seriously doubt it will ever happen so I am content to stick with the rather good OS 8.2+ flavors, or good forks there of, and wait for a “first look” web viewer to make it to the public domain.

    •' Jozy says:

      I rejoined SL about a month ago for the first time in a couple of years and I really have to question whether either Hi Fidelity or Sansar is a viable alternative for most Second Lifers. These people are totally in love with their hyper-realistic mesh avis and the clothes to dress them in. Just rendering these mesh-y scenes is a strain for a lot of folks with average video cards (myself included). They are not likely to gladly embrace anything that requires even more computer horsepower for vr, and presents cartoon avis floating in vr space.

      I have also been promoting Opensim and OSGrid in the most low-key way possible: a single line in my profile, and actually did get one person to register with OSGrid. Most have never even heard of Opensimulator, Hi Fidelity, or Sansar and are blissfully convinced their virtual world will just go on forever with sexier avis. Meanwhile Linden keeps driving up the virtual land prices both to squeeze people out and to milk the user-base for more cash. The concurrent user numbers look like they are staying consistent at around 40K, so committed second lifers are not going anywhere as Talia pointed out. It’s quite likely Linden is planning on a play for the younger, gaming crowd which is legion. Second Life never got elevated much past niche status; I think at its peak the concurrency was around 100K, but the business model never really worked as expected.

      However, when Linden does finally turn the Second Life servers off there will be a huge influx of people flooding Opensimulator. Huge for Opensimulator anyway; tens of thousands seeking the only remaining platform capable of delivering the Second Life experience they are used to getting. Potentially this represents a tremendous boost for the Opensimulator ecosystem and could be what we need to finally launch this platform in the direction of the next internet. Think of a 40K concurrency spread over the metaverse! The possibilities are limitless.

      •' Dharma Little says:

        ‘However, when Linden does finally turn the Second Life servers off there
        will be a huge influx of people flooding Opensimulator.’

        Second Life is a very valuable IP Property & Branding worth Hundreds of millions of dollars and they have made that over the years as well. they will just migrate everyone at some point to new software while it might break many things it will fix many things allowing them to offer new features.

        For OpenSim to win then Second Life has to fail is that right?

        •' Jozy says:

          Second Life is a dead whale on the beach. You can’t go over it; you can’t go around it, and it smells bad. I don’t think the brand is particularly valuable. Although the concurrency has stabilized at around 40K, the product has acquired an unsavory reputation over the years, rightly or wrongly. It is steadfastly ignored by most of the new generation of serious young gamers: the millions who play WOW and EVE on a daily basis and who watch gaming tourneys on Twitch. It never proved to be the money maker Linden had in mind when it rolled out and initially took off big time in the early 2000’s. IMO Linden will turn the lights off to force migration to Sansar, and this will cause most of the population to run screaming for anything that looks even vaguely familiar. The 30 or 40K erstwhile Second Lifers may be a drop in the gaming ocean, but they are a population with a decided interest in virtual worlds. Second Life in its current form is keeping those 30 or 40K people from even bothering to find out about Opensim.

          •' dharma little says:

            Wow great .you care to share were you got all your analytical data from i would love to pour over the facts you have found maybe others as well. its very rare to get the kind of statistics your referring to you must have a great source besides just being another resident with an opinion.

          •' dharma little says:

            ‘IMO Linden will turn the lights off to force migration to Sansar, and
            this will cause most of the population to run screaming for anything
            that looks even vaguely familiar’

            Jimmie crickets i had no idea could you tell us more about Sansar maybe some of the other upcoming Linden Lab business plans some creators might find that a competitive edge worthy of being informed over. not sure about everyone else but I’m all ears at this point.

          •' Susannah Avonside says:

            Most people that have been resident in Second Life for any appreciable period (5 yrs +) are very concerned about the future. Mostly it seems they worry about what will become of their inventories, as many have spent many hours, and much money in building up that inventory.

            When they hear about OpenSimulator virtual worlds, and that they can use Third Party Viewers to access these worlds, one of the first questions they ask is, ‘Can I take my SL inventory with me?’ Of course, the answer is in the negative, and unless LL relents and allows residents to export their inventories so that they can use them in OpenSimulator worlds I would guess that, for many, the resultant bad feeling towards LL would mean a potentially huge loss of potential customers for Sansar, and instead of moving to OpenSimulator, these people would just give up on virtual worlds altogether.

            Of course, were LL to allow residents to download their inventories in a way that is compatible with importing to OpenSimulator, then we’d probably hear screams of protest from the myriad hordes of Second Life creators hoping to make a second killing selling the same content to former Second Life residents who wish to maintain similar appearance to that which they had in Second Life, which, anyone who has ever been to InWorldz or Avination knows, is a very expensive process.

          •' Minethereé says:

            This is really an issue individual freedom versus commercial for profit grids limiting people by not offering them the tools to easily leave those grids.

            The fact is, however, people can, once they deal with the reality they are being screwed, can do something about this for their futures.

            Rather than trade one limiting commercial grid for a smaller limiting commercial grid, there are things one can do to mitigate such issues by teaching themselves that there are, in fact, great options.

            It does mean that such people stop believing what they are told, and accepting it as reality when it is not.

            For example, for a commercial grid that is also hypergrid enabled, Kitely offers very good way, and easy to do ways, via their website interface, for people to not only save copies of their region but to in fact download the OARs to one’s own computer. So everything you created yourself can be saved, even to using it in other grids.

            Where SL and similar for profit grids rely on keeping a percentage of people in their grid via their accumulated inventories, there are very good options, one needs to only point these out to people who are unaware or have been blinded with disinformation.

            Those who enjoy the commercial aspects of opensim would do well the check into

            This is the actual information about OARs

            Kitely uses the existing OAR code to allow for protections to creators by removing certain perms items from the OAR save.

            If a person wants to go even further to save what they create, to their own computers, the they can join a grid such as and save both your complete region as a full OAR but also save your entire inventory as an IAR

            If you go even further you can get with a simulator hosting company and have them run up your own grid or just host you inside other grids such as OSGrid and Metropolis…you then have all the controls that hosting company will give you.

            Even more so, a person can learn how to run up their own simulators and run it themselves in such grids I mentioned, or even run it from their own computers (where their own computer power will limit or expand usage), enable hypergating, and then join in with the Hyperverse.

            So while it is true SL people will have to initially reupload and recreate their things, it does not have to be with yet another SL mini where the same thing will happen, but in an environment where once they do that, they have full control and can save it all to their own computers.

          •' Kara Kelly says:

            Here’s what’s going to happen – they’re going to lose every bit of it and discover how much money they lost and they will bail SL and it won’t be for a newer shinier version. They’ll be livid – like the rest of us – and will either look for a free open source alternative or they’ll be so bitter they’ll reject VW across the board while they “heal.”

            And then there will be a huge resurrection of the new and improved copybots…

    •' Dharma Little says:

      I do agree with you on many points including LL & HF your completely spot on in that regard.
      we have some differences in ideas and viewpoints that’s totally understandable but can agree we both would like to see a better OpenSim and if my comments offended you it was not my intention with admission of some remorse being unable to edit my posts to amend that.
      ‘I should take it as a compliment that you, Dharma Little – who ever you are?’
      who i am it wouldn’t change anything in the bigger scheme of things I’ll be forgotten an rightfully so.

    •' Kara Kelly says:

      Here’s a proposal. How about getting all 1800 members to donate $100 each. Then you’ll have $180 grand to put in, and it’s a community effort showing support for it at all. SL/HF has financial muscle and if they solve the problem finally, OS is dead in the water…

  18.' Douglas Maxwell says:

    Since we are talking about roadmaps and future direction for OpenSimulator, I’ll take the opportunity to discuss our motivations to work with the Halcyon code. There seems to be a misconception that since the creators of InWorldz forked a long time ago, the technology must be out of date or deficient. This is *not* the case. I’d like to state, for the record, that based on my team’s review of the Halcyon code: it is in many ways superior to the current core Open Simulator code. They identified and fixed some major deficiencies that we found with the core OS code – and they did it years before our efforts. Since there was a disagreement on how to address these deficiencies with the core OS developers, the Halcyon team made the decision to fork and innovate in parallel.

    Why isn’t varegion support in Halcyon? They don’t need it. They did a brilliant job of handling inter-simulator communications and they support inter-simulator script handling. Region crossing is smooth and seamless, even with a squad of 9 soldiers and 2 course managers crossing simultaneously. We’ve tested this using real training material, at a real training site, using real soldiers (see my postings at LinkedIn for more information: ). Aggregation of large numbers of sims to create a large terrain space spreads the computational and memory loads across multiple processors, thus promoting horizontal scalability. The only way to get more performance out of a varegion is to get a better computer, aka vertical scalability. Varegions often exhibit unpredictable and unstable behaviors, causing us to abandon them in field use. In fact, after working with the Halcyon code, we wrote software that slices up varegions and turns them back into regular regions!

    Why isn’t hypergrid technology supported in Halcyon? What Crista did was amazing and I applaud her effort to make a “proof of concept” technology that links disparate grids. Now what? We need to make it safe and secure to use. The Halcyon developers know that InWorldz economy is dependent on their ability to keep it secured. Whatever the solution ends up being in a practical application of hypergrids, it won’t look much like the current specification. I fear that we are in a VHS versus Betamax situation here.

    Some of you are calling for a “wait and see” stance for our work so you can then crowdfund some kind of backport of our work to core OpenSimulator. I don’t believe this will work out for two key reasons. 1) History has shown that the core developers have a difficult time with large code contributions from outside contributors and 2) what we are producing may be so foreign to the original code that porting may not be technically feasible.

    Currently the core OS code is over 1 million lines of code. The Halcyon code base has the same basic functionality, a better physics engine, support for Cassandra databases, a better scripting engine and online grid tools – with roughly 1/10th as many lines of code. What the InWorldz team did was impressive. It is tight and efficient.

    Alan Kay – “The best way to predict the future is to invent it.”

    Folks, don’t wait. You are influencers, either through your voices or direct contributions. We have already had people contact us and begin helping. OpenSim is dangerously close to turning into the AOL of virtual worlds. Embrace innovation and change.

    •' XMIR Grid says:

      I think the holdback for many is that HyperGrid “IS” the metaverse.

      Otherwise I am totally with you.

    •' Minethereé says:

      Have you come to any thoughts as to why things went pretty much silent after this post?

      For my part I wonder why you feel the need to try for the hard sell.

      You like halcyon, and so you do…why does it seem so necessary to change people’s minds?

      •' XMIR Grid says:

        Maybe, because at some time, reality strikes? 😉

        •' Minethereé says:

          maybe, I have for many years considered the fact that one person’s reality can be quite different than another’s.

          Part of the reason, I think, is the lack of ability to properly and clearly understand an issue…often by only seeing one aspect of a thing, rather than attempting to search out others.

          otoh, I am not a Psychologist so whadda I know-)

          •' XMIR Grid says:

            Every person’s views (including mine) will exist along some subjective axis, including blocking out alternatives for multiple reasons.

            In this case the (subjective) reality is that OpenSim core has more or less maneuvered itself into what might become a dead end. Rather than try and fight themselves through this quagmire, MOSES made an informed decision to base their effort on a track that has been (largely) more successful, and that early on tackled core technical obstacles that exist in OpenSim core. For an organization that does not need Hypergrid, this make perfect sense.

            In fact, if Hypergrid could be bolted on Halcyon, you’d probably see a significant move to that alternative, even by grids that have invested heavily in customization of OpenSim core. The exception is standalone, as the Halcyon database setup (per my current understanding) does not easily lend itself for installation by a casual user.

          •' Minethereé says:

            yes, subjective reality. I suppose as time goes on, we will see.

          •' XMIR Grid says:

            We will, but OpenSim, as it currently stands, is a project where not even a target audience is no longer defined.

            Originally it was SecondLife users, and to create an open alternative to SecondLife, but now that has been largely abandoned. The only person who from time to time kicks this can is Melanie, but even she is getting tired of it I think.

          •' Jim Tarber says:

            There’s nothing inherently preventing home/hobbyist use of Halcyon, but to date there have still been enough other common issues to solve that it hasn’t really been a priority to make it easy for others to set up private Halcyon grids. There are a couple of grids that are now running Halcyon software, but it does require some technical skills and a lot of persistence and dedication to getting it set up.

            I’d like to see that change in the next year, with an easy setup similar to sim-on-a-stick, a kind of Halcyon-on-a-stick. It’s really a matter of having an installer of some kind that sets up the prerequisites, then another launcher that loads the various components on demand. We’d love to get all that packaged up and polished for the novice user, and in fact I even grabbed a domain in preparation for that some time ago, but we have some pretty big challenges still queued up in front of something like that.

            In the meantime, other non-InWorldz contributors to Halcyon have been making improvements to the process, and isolating more InWorldz-specific things from the core Halcyon code to make them configurable, all in preparation for others to run the Halcyon software. It’s slow but it’s making progress and I hope to see a “stick” installer eventually. Maybe even called Homeworldz, if I’m connected to that project.

          •' XMIR Grid says:

            Thanks for the update. I think completing the installation and configuration documentation would give you a lot more traction, and make it possible for people to try Halcyon with relative ease.

          •' Jim Tarber says:

            Yes, agreed. It’s definitely something I’d love to see. But to make it a really smooth installation for hobbyists and novices, we’d probably also need to convert most of the storage to SQLite or something of that nature, to avoid the headaches of server/daemon setups, etc. None of this is a small effort, but I do think providing some form of easy Halcyon install for home users, combined with active and regular updates (e.g. every Halcyon release is accompanied by a “stick” update) could lead to a lot of traction and adoption.

          •' XMIR Grid says:

            Even getting more meat on the “expert” installation and configuration would be helpful. What I could find was rather sketchy in many sections.

          •' Jim Tarber says:

            That’s a good point; incremental improvements are helpful.

            I’d ask you to file an issue on the GitHub site, to guide us where you see gaps, but I don’t even think there’s a good summary of the *overall* steps needed.

            There’s a *tiny* bit of info on the wiki under the Configuration section, but there I don’t even see a mention that it’s only Windows currently. (The main problem with Linux is a lack of a good C# wrapper around PhysX. We’re using a modified PhysX.NET on Windows, and almost all work is there. That said, the MOSES project is Linux-based and advancing the work there to support physics, and some of the Halcyon contributors want to do most of their work on Linux.)

            Feel free to add me in Skype if you have an account there, jimtarber, and as questions about Halcyon installation. Perhaps I (or one of the Halcyon contributors) can use those questions to update the wiki and readme doc.

          •' XMIR Grid says:

            I am not in a position to pursue it till after I have recovered from major surgery I will have at the beginning of next month, taking us to after Christmas or so. Meanwhile progress may have been made on a *nix version which is what I would be interested in.

          •' Butch Arnold says:

            I agree, I have a test grid running the Halcyon code and find it to be very nice. I believe InWorldz has done a fine job in my opinion. Hats off to the InWorldz team for the work they’ve done and their decision to turn the code loose.

            It is also my opinion, based on what I’ve heard and what has been said by some others I’ve chatted with, the future is not the 9.0 OpenSim code.

            Too many people feel that the 9.0 code was “Force Fed” to the community and I have heard accusations there were “other” reasons for this, but that is not for me to decide. The fact remains, while there are some nice improvements in the 9.0 code, it is far too buggy compared with what I’m used to with 8.2 code and I don’t have a team of developers to bring it to a point where I am comfortable running it. It would be a support nightmare for me and I simply don’t have the resources to handle that.

            If not for the “Masses” expecting a HyperGrid capable grid, I would choose the Halcyon code if I were running a closed grid.

            In many ways I feel the advancements made by the InWorldz team with Halcyon, combined with the efforts from the MOSES team will provide a very solid platform indeed. If the Halcyon code was “HyperGrid” capable in some form, I’d switch to the Halcyon fork tomorrow.

            Until we become a “United” community working towards the same goals there will always be a “Split”.
            Some people feel that the current core developers of OpenSim are self serving, unfocused, not willing to change, and unhelpful.
            Many feel there are “Egos” which often get in the way of improvements and advancements and feel that we are truely at a dead end with OpenSim.

            Some people want an exact copy of the current implementation of SL, while others want something completely different. I believe the aim of the current trunk has to been to align itself in many ways to SL in hopes of attracting SL users, while at the same time being modular enough to accept customizations which will allow us to move beyond the SL implementation and ideas.

            Sure, there are a lot of SL users to attract, but there’s a whole lot more of non SL users who have yet to experience OpenSim/SL/Other virtual worlds. I don’t feel the future is with SL users at all. I feel we need to move beyond that.

            Back in the early days, Ford made a model T which was considered to be a very solid product which sold at a fair price which enabled the masses to own a car. Sure there were better cars in one form or another, but most of these were out of reach for the masses.

            Ford stayed with the Model T for many years (19 years to be exact) as they didn’t want to change what had been working for them for fear of losing the popularity they had gained. Instead, their sales number fell in the latter years causing them to lose their place in the market as many moved past the model t into other alternatives.
            Ford has never regained the popularity they once held because the did not innovate, they sat back and relied on what was working… why fix something that isn’t broken, right?

            Well here we are.
            I’ve been around OpenSim since late 2007 when I started creating 3rd Rock Grid. I’ve seen much progress over the years, but in recent years the development has slowed, the excitement has slowed, the interest has slowed, the bickering has increased, the “Split” has become greater and the future has become very cloudy.
            Look at the commits.. most are done by Ubit. We’ve lost too many developers who were once very excited to be part of OpenSim.

            I personally don’t have deep pockets but I wish that I had the resources to take the Halcyon code and “Run” with it as I believe it is far more stable and advanced than what most of us are using. Kinda weird hearing a competing grid talk about a competitor that way isn’t it?
            Well, I don’t look at any grid as my competitor, we are all part of the same community, and I share the same goal as many grid operators…. bring more people to virtual worlds.

            I don’t really care which platform that is so long as it is stable, can be expanded/improved and is easy for the end user to use. Is that Halcyon? Is that OpenSim in some form? or is it an entirely different platform?

            Guess we’ll all have to wait to find out.
            I think that we as a community should embrace “All” efforts to improve the code, whether it be Halcyon, OpenSim 8.2, 9.0, WhiteCore, or another fork and it is my opinion that this project should be managed in some form or another by the community. If the current developers want to do that then we run with them, if they don’t then we run with Halcyon, Whitecore, or another fork as it is clear to me that this “Once Darling” project is slipping beyond the waves and is need of some help.

            I think there is enough interest in our community that we can get this project back on track and make it a viable future platform which holds promise. I remember when the ambitions were to make OpenSim the “Apache” of virtual worlds.
            What happened to that? Where did it all go?
            We as a community should reclaim this project, clean it up, put it on a track where we can clearly see where it is going and go.
            Bickering, finger pointing, day dreaming, etc isn’t going to solve anything.

            United we stand! Divided we fall!

          •' Kara Kelly says:

            We as a community want the same thing. We as a community aren’t devs and it’s the devs holding up OS from its potential so we’re all kind of held hostage until a group of devs can resolve it and if they show they care and keep things updated and easy for the end user, that’s where the community will go. At this point something needs to give and it’s looking like Halcyon/MOSES might be the solution at the end of the day.

            PS I downloaded whitecore but the site seems outdated. Is this a viable alternative to OS/SOAS and the like or is it just another stagnant fork?

      •' Douglas Maxwell says:

        My motivations for responding to Maria’s queries and by extension answering the communities questions are less about changing minds and more about correcting misconceptions and rumor control. We don’t respond to trolling but I do engage earnest queries and provide as accurate a statement as possible, as long as I stay within the boundaries of what is open source. For example, I will answer any question you have related to the simulator but I won’t discuss details about how we use it in the field.

        I read comments in other forums and have fielded direct questions with concerns about the Halcyon code. This is a convenient way to set the record straight. The Inworldz folks read some of the comments directed at us last summer when our code contributions were being rejected and they contacted me directly to discuss and compare notes. The rest is history.

        The MOSES project is using the technology as a means to a research end. I am particularly sensitive to licensing as there are potentially severe consequences for me violating software licenses, even open sources ones. As part of my job, I must make an effort to ensure you are at least aware that the code is available and where to get it. I cannot make you use it.

        The MOSES project is finite and it will end someday. I’d like to see the Open Simulator in a better place when we leave it than when we found it. While you have our attention and considerable technical resources available to you, it may be a good idea to at least consider what we have to contribute and do an objective assessment of its value.

        As for why the conversation went silent, that is a question you’ll have to ask the contributors. I believe the community intuitively knows the platform must evolve or it will dry up and become an obscure historical footnote.

        •' Minethereé says:

          thank you for your response.

        • The community does know that. In our latest survey about the evolution of OpenSim (and need for VR support), only 10 percent said that OpenSim has what it takes to survive as a standalone project in its current form.

          The MOSES announcement of the coming web viewer has, I believe, breathed new life and a spirit of optimism into the community — trolls aside! LOL! — and I, for one, am feeling that OpenSim is back in the game. I can’t wait for the viewer to come out.

          •' Dharma Little says:

            Not only does the community know this but has been down this road once before.long after IBM,SunMicro an the rest of the giants pulled out another fork tried to rescue OpenSimulator but was turned away due to a mix of infighting between core and selfish agendas by those 7yrs ago to only push development to met limited needs they had. not the community’s overall needs.

            Two realXtend companies, Admino & Ludocraft, have been dedicating 20 personnel (programmers, designers and content creators)‘ to extending’ the function of the OpenSimulator Project, both server side and client side respective. = It failed the community pushed them away indirectly thru certain selfish individuals

            realXtend completely removed OpenSim from all roadmaps or considerations in the future.(ending any remaining compatibility between what had started as a fork of OpenSim .

            realXtend receives grants from the different organizations in the European union totaling over 55 million euros (They are set up as a transparent tax & donation approachable foundation)

            My bets on History repeating itself but would find myself thrilled and overjoyed if proven wrong in the next few years after all High Fidelity is not the future Metaverse just because Philip Rosedale declares it to be, nope!. only if every lets it be. from the same man who turned his back on OpenSim in the very Beginning now ‘TELLS YOU!’ he is replacing it with his own dreams! but what about yours? the shared dreamed you all have had? to me that’s enough motivation to fight an uninvited future declared on everyone.

    •' Kara Kelly says:

      I said the same thing about OS and HB mods deleted it because they didn’t like somebody pointing to the elephant in the room. I support Halcyon/MOSES and want to see what it can do. I see the OS core being updated every couple years and all the core issues from an end user POV being ignored. It’s like crickets. I hope beyond hope that a successor to OS-AOL will bring it all up to date for the web and vr tech but please, from a non coder civilian, please make it easy to deploy, use, and create with. Thanks!

      Okay mod, delete it now so nobody hears the truth…OS is becoming obsolete. Even LL had the good sense to move to an evolutionary product with Sansar and update SL to current trends. OS…isn’t doing anything.

  19.' Minethereé says:

    what’s that sound I hear?? ah, yes, that’s the sound of teeth chattering after a bucket of crystal clear bitingly cold water has been pored over them…..

  20.' Minethereé says:

    Ok. thank you, just watched it-)

  21.' Jim Tarber says:

    We’re all definitely influenced by them, but in a good way. US Army projects have produced most of what we look for in virtual worlds, from the computer itself, to the Internet, to virtual reality. Not to mention digital photography, GPS and the microwave oven. (See for some examples.)

    Now I for one am thrilled to see taxpayer money being spent on projects that try to keep people alive, rather than kill people (even if those people being kept alive are soldiers coming home as safely as possible). But beyond that, Halcyon (like OpenSim) is open-source software, meaning anyone can benefit from it. It’s not like the software teams can choose who downloads it. However if elements of the US government, or others, or school boards, or large corporations, want to use the software *and* contribute back to the community project, that might just be exactly what a project needs to focus as a professionally-run project, to help fund the work, and to reinforce the legitimacy of it was a serious project.

    Taking the government contributions and efforts seriously and responding professionally is exactly what the OpenSim team rejected. Dr. Maxwell made that very clear in his comments below. This isn’t a small crowd-funded project. I think if an organization doesn’t have at least six-digit amounts to spend on development over a period of years, they won’t be taken seriously here. Really, I think it requires a commercial company with full-time paid management and developers to pick up the ball when serious players pass it to them. Halcyon definitely isn’t going to watch the ball just roll off the court.

    •' Dharma Little says:

      Some point the Moses Project mission will be completed but Opensimulator will still continue on while i do understand your long term of moving OpenSim away from SecondLife to Inworldz that would allow you to bring a network of trusted grids into your own controlled version of HG.
      Perhaps you might attend the OpenSim conference calls for proposals providing your thoughts about the metaverse future. they may have new plans themselves and it would be a huge mistake for the OpenSim community to attempt to bypass and dismiss all the efforts past & present core developers have done.
      while I do not speak for anyone else but all efforts should be made to keep OpenSim Non-Profit (that does not mean unable to take grants or donations or unable to provide commercial features) and in the hands of the community.

      ” I think it requires a commercial company with full-time paid management and developers to pick up the ball when serious players pass it to them.”

      I completely disagree OpenSim needs a better way of operating itself. it should not be profit driven but driven for sustainability/accountability, more community driven,community powered but most of all it just needs TLC not turned over to private interests with a profit motivation even if that’s not the intent .

      Opensimulator should remain in the hands of trustees while empowering the community further as the project revives and reinvents itself.

      •' Jim Tarber says:

        I believe InWorldz has always supported a long-term readiness to connect with other secure grids, in a secure way. Unfortunately that is not what HyperGrid is (at all), so there are some overall plans to provide inter-grid connectivity through a separate gateway process that isolates such activity and does not cause invasive disruptive changes to the core region functionality. Representatives of InWorldz have attended the OSCC and will again, and in fact InWorldz founder David Daeschler has presented at OSCC recently. InWorldz has been doing everything possible to provide both code and guidance to the community, for anyone who wants it.

        I don’t really understand the comment about “long term of moving OpenSim away from SecondLife to Inworldz”, mostly because I don’t understand what moving OpenSim away from Second Life means.

        In terms of my comment about “a commercial company will fill-time paid management and developers”, that is not necessarily the same at all as a for-profit company: InWorldz is not really a for-profit company, because profits, if any, are reinvested in the future of the grid in one way or another. The key point in my comment was full-time developers and management. Without those, the problems that Dr. Maxwell described with continue to trouble and limit the OpenSim project for years to come.

        OpenSim really shouldn’t be in the position of having to turn down funding for useful projects because there’s nobody full-time on the project to manage and coordinate those efforts. It can’t be part-time, hobbyist work, otherwise “RL” will always be bumping into the way. It has to be someone’s primary employment responsibility, not something done “on the side”. Any foundation or trustees assigned to manage it should accept that with real responsibility, representing the community, probably as paid employees or at least paid contractors, rather than seeing if proposals fit into a busy RL schedule, or with their own needs or desires for the future.

        There are many changes and enhancements that contributed to the Halcyon project that are of no value or even may cause temporary problems for InWorldz that are accepted and applied because they benefit the Halcyon project itself. The conflicts of interest mentioned earlier don’t exist in Halcyon; in fact, the open-source community tends to get the changes and improvements even before the InWorldz grid does. Nor are their conflicts of interest where gatekeepers refuse updates because their own grid has a competing bit of code, even though that project has been withheld (for years) and not been made public, so someone in the OpenSim community has developed an alternative. That is dysfunction that needs to be addressed before OpenSim can grow, and Dr Maxwell identified this as a key factor in their original plan to fork OpenSim (which eventually lead to working with Halcyon instead). Trustees are not particularly effective without trust in their willingness to represent the community. These are the very reasons InWorldz forked the code forked so many years ago in the first place, and Halcyon was born. MOSES made a similar decision 6 years later, so things have really not improved in that regard.

        OpenSim should certainly be in the hands of the community, under management that represents their interests. However I think that community should realize that is not how it is, currently. There is a lesson in this story, a big one. A wake-up call. While grids themselves (like Kitely) have tried to get serious with it, OpenSim itself (as a software project) is currently what I’d call “a basement project”. A hobby project. That needs to change if it is going to be taken seriously and attract growth factors.

        •' Dharma Little says:

          “I don’t really understand the comment” i think you understand everything and your own comments made recent and over the years googled & from your own websites and even this one speaks for itself how you truly see OpenSim without me saying a word. its a good thing Halcyon is Open Source because sooner or later a MOSES will be gone once completed.
          it is not that Maxwell is untrustworthy. no. he is the credibility inworldz lacks in OpenSim and his presence helps soften perceptions the public has regarding Inworldz that’s commendable.

          it would be a mistake to have Inworldz or its management attempt to lead OpenSim or another Fork /or Replacement of it.
          a future OpenSim should be in the hands the community trusts and respects that also does the same in return.

          While most points i do agree with you it’s not the sales pitch it’s the salesman. while others might go along with it. that’s the right they have.myself i will wait for the other shoe to drop while not fully knowing what it is but certain it is there hanging over our heads and at some point will drop very very hard.

          Halcyon might be OpenSim future!!.. but a “Post-Dr Maxwell(Moses) World” run by Inworldz or its best interests is not anything i want to be part of. but then again i only speak for myself.

          •' Jim Tarber says:

            I think I’ve been pretty transparent about OpenSim. I just didn’t understand what “moving OpenSim away from SecondLife” meant. But I may have just got it. Do you mean moving OpenSim away from supporting SL protocols and behaviors? And moving it towards supporting InWorldz-specific ones instead?

            If so, let me assure you that currently those are pretty much the same, and the difference Dr. Maxwell was referring to I think is mostly the login from the web viewer needing to use JSON web tokens (JWT). That’s not really InWorldz-specific, it’s basically just support for a browser-based client and InWorldz is the first to provide it. OpenSim could also provide that, and it wouldn’t be moving OpenSim towards InWorldz; it would be moving OpenSim towards web technologies. Halcyon just (publicly) committed those changes for all to use, it’s not really a large project. Definitely something that could be crowdfunded or done by an OpenSim developer volunteer. It’s not InWorldz-specific or MOSES-specific so it doesn’t affect where OpenSim is “moving towards” for OpenSim to also pick that up.

            >> “it would be a mistake to have Inworldz or its management attempt to lead OpenSim or another Fork /or Replacement of it.”

            Absolutely. I completely agree. But okay, I think it would also be a mistake to believe that anyone associated with the Halcyon project or InWorldz would want or even consider that. That’s what Halcyon *already* is. It is already a fork of OpenSim, from 6 years ago, and since the fork, the focus has been on moving the code towards more production-ready use, replacing weaker components with stronger ones, expanding functionality to support more of what SL does, and to run InWorldz on reliable enterprise-class servers.

            The Halcyon project has continued on, for years, going on down a different path, but offering back what it can. This includes components and subprojects such as the asset servers written to replace the OpenSim asset storage, the inventory system which interfaces to Apache Cassandra, the Phlox script compiler and virtual machine runtime environment, the permission system used at another OpenSim grid, a new physics implementation, various critical security fixes, and many other open-source components and projects you can find at

            But as a result, the Halcyon code is different enough in most of the vital areas that the exchange of code is almost entirely one-way. There aren’t really any major areas of OpenSim that haven’t been rewritten in Halcyon. The Halcyon priorities are different from those of OpenSim, and the project already has plenty of work to do. I don’t think anyone at Halcyon or InWorldz is interested in trying to lead OpenSim, and especially not to create a new fork of the current code.

            So I think if OpenSim is to go on and evolve, it needs to be lead by people from its own community of users (preferably full-time, and preferably not someone associated with a specific grid). But I don’t think that’s possible anymore; the community has already lost control of the OpenSim project to one grid, and the current OpenSim 0.8 code is being replaced with a another fork from that grid. It’s okay to use words like “foundation” and “trustees” but from my perspective, those are the other two shells being moved around to distract the community while the action was going on under the third.

            There is still a chance for someone with funding to come along and adopt OpenSim, but I think the best chance for it now is fork of the 0.8 tree, and probably by someone completely external who the community had a bit of faith in: e.g. if a Microsoft or a Google or an Amazon or an individual with an interest and deep pockets and could actually come along and fund some development on it, and a team of marketeers with a promotions budget. But I don’t really think that’s going to happen, and if there was someone like that, Halcyon would probably look a whole lot more attractive anyway.

            And that’s a real shame. I’d love to see OpenSim continue on and pick up new developers to champion it in a grid-independent way. The whole community needs to grow more, and create an influx of new people, not just recycle the same current “alt grid” user base between grids, and being counted more than once. But more than that, I think it needs a rallying cry, and repeated updates on progress from the development team. I miss Justin’s “This week in OpenSim” summaries, because they brought all the grids together for coordinated excitement about new fixes or enhancements. I think those days are gone.

          •' Han Held says:

            “So I think if OpenSim is to go on and evolve, it needs to be lead by
            people from its own community of users (preferably full-time, and
            preferably not someone associated with a specific grid). But I don’t
            think that’s possible anymore; the community has already lost control of
            the OpenSim project to one grid, and the current OpenSim 0.8 code is
            being replaced with a another fork from that grid. It’s okay to use
            words like “foundation” and “trustees” but from my perspective, those
            are the other two shells being moved around to distract the community
            while the action was going on under the third.There is still a
            chance for someone with funding to come along and adopt OpenSim, but I
            think the best chance for it now is fork of the 0.8 tree, and probably
            by someone completely external who the community had a bit of faith in:
            e.g. if a Microsoft or a Google or an Amazon ”

            …preferably a RedHat, or another company with an open-source background. Also a pony…since I’m wishing for the unlikely.

            Other than that, I agree with your analysis, pretty much completely. In particular the optimizations which JCC made for the opensim conference has been thrown out the window, making 0.9 a far slower, more resource demanding proposition.

            But, as you pointed out; there’s no-one at the wheel. Particularly with Avination’s disappearance.

  22.' Jim Tarber says:

    Thanks, I appreciate that response. I definitely agree that it is better to show Halcyon for its strengths and debates centered around negativity between grids or towards Halcyon or OpenSim just do the whole community a disservice. I do try to avoid them, because I do think that if we can get *either* Halcyon or OpenSim to grow, there’s a spill-over and the whole community is likely to grow, and each of the projects will benefit if they are in a position to do so. In this case I had to be a little more blunt than usual when it seemed there was an implication that InWorldz or Halcyon was trying to make some kind of “move” towards control of OpenSim or something like that. No… no. But I’d still like to see OpenSim grow beyond its current state. I like to think of InWorldz/Halcyon/OpenSim/Aurora/WhiteCore/etc as friendly neighbors, all trying to improve the neighborhood, or different hotels in the same city, who could work together to grow the tourism or business activities in the city in order to grow the whole market there.

  23.' Jim Tarber says:

    I don’t know if a fork would be the answer; all of these things depend on the people involved. My point was that from my perspective, Halcyon already was a fork and that Halcyon folks wouldn’t be interested in leading an OpenSim fork (or OpenSim itself), as you had suggested. I just wanted to explain that wasn’t true. I am also not here to “trash” other developers, and I definitely don’t want to do that. There are many trade-offs and priorities and everyone has their own agenda (and I don’t mean that in a bad way). We’ve followed a different path. I think that is paying off now, but it’s all very subjective.

    For example, the HyperGrid seems to be a high priority for OpenSim, while it is a deferred (2.0-ish) feature at most for InWorldz. But the interesting thing to me is that this may not be true for Halcyon. I’ve heard more talk of inter-grid travels in the last 6 months or so than I did in the last 6 years. 😉 That’s the beauty of open-sourcing it and listening to the third-party developers, contributors, grid owners and users. There is a separation of agendas that happens when a product becomes multi-purpose. Now it still has to be done in a way that fits better, and passes architecture reviews, but if the industry survives the next couple of years (during which time, I think it will undergo some very large disruptive changes), then I think there’s a good chance we’ll see connected Halcyon servers. Thus is the nature of open-source and community involvement.

  24.' Minethereé says:

    I usually don’t comment to people who have no disqus to look into (i.e. one off type commenters don’t usually interest me) preferring to at least get somewhat of a better idea of where they come from in their thinking.

    Some opt to keep their disqus private for similar reasons.

    I have never been interested in investing myself in such. I do understand, of course, that some folks know that if they used some more commonly known name (which itself is usually another fake name) it would cause them to get more drama. And it might cause them to reflect more on their comments prior to posting which defeats the underlying reason for doing it in the first place.

    For myself, it never mattered…I use this name primarily as a kind of lightening rod account…and inworld I use others…I don’t hide the connections necessarily but I don’t make it public either. I prefer a more “like me” peaceful inworld experience, ya know.

    Anyway, I have been reading your comments and whatever someone else may wish people to believe, it is obvious you have a pretty good understanding of things.

    Makes you feel a bit lonely doesn’t it? I know.

    I recently decided to eschew certain people and block/mute/etc (depending on the social site rules) as just irrelevant people. Only because of their ongoing nonsense that I just no longer even wish to see.

    This goes against how I am where I prefer to know things rather than not, even if they annoy me. Some things just aren’t worth the hassle anymore to me.

    Of course this is not for all people and leaves conversational gaps which doesn’t bother me at all.

    I think I know another name you more commonly use but there is nothing served by my mention of it…and of course I could be wrong.

    I just wanted to really say to know I hope you keep it up, while some may try to show you in some kind of negative light, take that as that they know what you mean but will never say so.

  25.' Bristle Chesnokov says:

    well I am going a different way with Android. Since Android are not the same as desktop or laptop, they are more restriction on graphics. Even a Nvidia TX1 has only 256 cores of GPU and how many have a ARM TX1 with Android? The other way, is what Nvidia is doing — send out 60 frames a second with interactions both ways. Unfortunately Nvidia is closed source.
    The way things work is a server and Android clients. I did the same thing a few years ago with VNC. It was not smooth and I had a hard time doing anything but it work on my Shield. But with 60 frames a second and ability to play games or whatever, would be ideal.
    Opensimulator, I think its dead and that would be unfortunate. For both SL and Opensimulator, collada is not the way to go. FBX is support and most games now uses FBX. It is an Autodesk thing, so I am a little concern about the future. And no animation for NPC other that fake avatars– no monsters or deer. And finally, no pathfinding and there is only one article with C# instead of LSL. Really you need Opensimulator 2. And no, I am not a fan of HiFi.
    Now if you send out 60 frames a second, have interaction, have Java for UI, you may have something because if Opensimulator goes away, you can still deal with other things that has intense graphics. Or just wait for a few years for the Android that has 4 G memory or more and with 1000 cores or more.
    Now Halcyon is interesting but I don’t have enough experience to judge it.

  26.' Kara Kelly says:

    This sounds like heaven, but will it offer the same capabilities of in world building, terraforming, and all that we’re used to with downloadable viewers? SL/OS users (and game dev world builders) are used to being able to be immersed in the environments and create things. Even game dev is catching on with new toys to allow in world creation finally. If they work out an easier approach to live social aspects like SL/OS without the drama, they’ll win the VR game. Will we be able to pull up a browser and be in our VW space with the same abilities to open the tool box and create on the fly, with and without other people collaborating? Otherwise it’s walking around in another static chat space. 🙁