5 missing pieces in OpenSim

OpenSim development — like that of any open source project — is very much focused on what developers want to do. They are, after all, volunteers, they don’t take orders.

That leaves business users in a quandary — especially those without the technical skills or budgets to do their own development work.

Last month, Daden Limited got proactive and paid a bounty to get NPC support added to OpenSim.

Maybe it’s time for OpenSim’s business users to take that approach to the next step, using a crowdfunding platform like Fundry to get together and chip in on some must-have features — that have been lagging behind.

The following are five projects that my company would be interested in contributing to, and which, I think, could be beneficial to other enterprise users of OpenSim and to OpenSim grid owners.

1. Voice

Whisper voice is great. It sounds as good as Vivox does in Second Life, and its free. But it’s not fully integrated with OpenSim.

Voice is a must-have feature for any business grid, and very much a nice-to-have feature for a social grid.

But the way it works now is that you can either have a browser that works with Vivox in Second Life, or you can have a browser that works with Whisper — there’s a little Whisper file that overwrites the little Vivox file. If you’re going to be switching back and forth between Second Life and OpenSim, you have to keep switching these files in and out — and who wants to do that? It’s not exactly an intuitive process.

One solution is to have a viewer configured to only work with OpenSim, with Whisper pre-installed, that would automatically switch to Vivox if it found itself accessing Second Life. Not a particularly difficult programming job, but a job for viewer developers.

In addition, there’s also some server-side work that remains to be done, so that Whisper voice automatically shuts off when you leave a region. After all, you don’t want to hear the things people say about you after you’re gone. Also not a particularly thorny problem.

It’s been a year since I first tried out Whisper in OpenSim and fell in love with it. Ever since then, a functional version has always been just a couple of months away.

Is anyone else with me on this? Should we raise some money and just pay a bounty to get this done?

2. App sharing

I don’t know whether this is feasible or not, but I would like to see an OpenSim module that enables the ability to share a desktop or an application window.

Say, for example, I have a PowerPoint presentation. It would be nice to embed that window on an in-world screen to show the presentation to my audience.

Or if I wanted to teach my staff how to use a new piece of software.

Most enterprise-grade virtual world platforms already have this, and so does Open Wonderland. Of course, they’ve got an advantage in that they were designed for the corporate market to begin with — and the same team gets to work on the both the viewer and the server.

This isn’t a feature that retail users would want or need. But it’s pretty high up there for corporate users.

3. Access sharing

If I want to show a Website on a screen inside OpenSim, not everybody is looking at the same site. Instead, everyone has their own in-world browser running it.

So, for example, if I put a slide show on an in-world screen, when I click to advance the slide it will only advance for me — it will only advance for my audience if they click on it, as well. Similarly, if we’re using media-on-a-prim to watch a movie together, and I pause the movie — it only pauses for me.

And if I have to log into a site to access its functionality, the audience members each have to log in individually. It’s like we’re at home accessing the site on our separate computers.

I would like it to be more like bringing my laptop to a meeting, and having everyone be able to see my screen. I want people to be able to share my access to a particular website or web application — just as if they were in my office with me, looking over my shoulder.

But I might not want everyone in the same region to eavesdrop in, so some kind of security system would be required to control access.

For example, I might have to individually invite audience members to watch my presentation, and then access priviledges would go away when I shut down the application. Or maybe invite an entire group — or invite everybody, if the presentation wasn’t particularly confidential, like a marketing presentation or a movie we were watching together.

4. In-world documents

Today, you can’t have documents — PDFs, Word documents, spreadsheets, presentations — in your inventory.

But I spend most of my time at work passing documents back and forth. If I’m going to be working in world, I need to be able to share documents — and, ideally, collaboratively work on those documents in-world as well.

This could be in the form of integration with a free office suite like Open Office. Or could be in the form of links to Google Docs. Either one is fine with me and would work with my company, though larger firms would probably want to see Microsoft Office and SharePoint integration. But here, integration with Microsoft’s new Web-based Office product might be sufficient, at least to start with.

5. Cone of silence

Many enterprise platforms allow you to set up an area that is, in effect, surrounded by invisible (or visible) soundproofed walls.

Intel’s new DSG technology, which will be part of the regular distribution of OpenSim, will allow more than a thousand avatars on a single region. That will allow companies to have busy convention centers, busy shopping areas, popular events.

This could cause problems, however. Say you’ve reserved a hotel conference rooms for a sensitive business meeting. And a vacationing couple has reserved the room above for a meeting of a more … naked … nature. Not only would their passionate moans derail your financial presentation, but your financial presentation might totally kill their mood.

Some enterprise vendors deal with this issue by having walls be sound-proofed. So if you’re standing in front of an open door, you can hear what’s going on inside, but as soon as you take a step in either direction, you no longer hear them.

Other vendors have quick drop-offs for sound — unless you’re close to someone, you can’t hear them — and then offer a public address system for when speakers need to be heard by a wide audience.

And other vendors allow users to create zones — lines on the floor — that constrain all sound within them, no walls required. If you’re inside the lines, you hear everything. Take one step over the line, and it goes silent.

I prefer the steep drop-off with a public announcement system, myself, as it seems easier to use and easier to implement. Individual objects could be programmed to be region-wide microphones, for example, using a new script command.






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.

  • Graham Mills

    1. Agreed.
    2. Is just about achievable with use of shared media and join.me. At a tools.Jam meeting, for example, I was able to show my desktop. While the resolution is sub-optimal, it's a cheap and effective approach.
    3. That depends on what shared media service you are using. AuthorSTREAM synchs viewers, for example.
    4. Agreed. I assume that's what the LL/RRR Immersive Workspaces provided?
    5. SL now has a feature whereby you can restrict sound to a particular parcel. The option is there in most clients. That said, I've not used it nor, more to the point, voice. I assume it will show up in OpenSim shortly?

    • Graham — The voice-parcel issue sounds perfect — a great solution to the voice problem. Doesn't solve the vertical voice carrying issue as much but hey, in OpenSim, you've got as much land as you want, so you might as well build horizontally. Instead of having a vertical hotel and conference center, you can have a main gathering area surrounded by smaller meeting rooms, and private cabins — all with beach-front views, of course! 🙂

      Meanwhile, authorStream, join.me and Google Docs are work-arounds — and, with Google Docs, you can't control the pace of the presentation. It would be nice to see some closer integration, make the process a bit more transparent for users.

      Maybe there's a vendor out there who's working on a solution now — if so, call me!

      The reason I'd prefer the functionality to be included in core OpenSim, however, instead of in a proprietary add-on, is that it will make it easier for pioneers to adopt. Schools, startups, non-profits, etc… who wouldn't have the budget to pay for a proprietary platform (like IBM's $50,000-Lotus Sametime 3D). This will allow for a bottom-up approach to embracing immersive technology, based on what users actually want. We already have vendors going for the bottom-down approach, selling high-end systems to corporations with all these features in place.

      We need both approaches, I believe, and OpenSim is perfectly positioned for a bottom-up growth trajectory.

  • Lawrence Pierce

    My order for prioritizing would be: 3, 5, 1, 4, 2. 3 is first because it would (finally) synchronize and greatly multiply the power of the Web within virtual worlds, and even offer much of the functionality of 2 (which can then developed over the long run). 5 before 1 since turning off excessive voice is more valuable than making voice more convenient. 4 would be fantastic for education and corporate!

    The developers may be volunteers, but surely they recognize the potential of these features. I suspect they are all technical challenges in ways not only defined by development funding. For example, media-on-a-prim works efficiently because there is no demand for synchronization between users. It seems to me that synchronization would require the web content to travel through the server side of an OpenSim world and strongly impact sim performance.

    Maybe what we need are in-world "websites". With extensions to the scripting language, actual "websites" could be embedded in a sim. They would display on an interactive media surface. Synchronization issues would be moot since the media source is already inworld. We can call them "Simsites"…

  • sarahblogging

    Actually, for me the absence of voice is a feature, not a bug. I tend to keep away from areas in Second Life where Voice is commonly used too 😉 If I have a presenter, who uses Voive, its best done with a stream, and this si possible in OS by now.

    2 would solve 3 and 4 singlehandedly

    • I think voice is a non-issue for a lot of people. It can get in the way of role playing, for example — and of gender-bending. And it can hurt people who are deaf — or who are using Google Translate to communicate with everyone. Or who have trouble picking up voice inflections (like Aspergers sufferers). or folks who don't have headsets attached to computers. Or who log in from cafes or libraries or other public locations.

      But it's critical for a business meeting, for sales training — for many business applications.

      The chief work-around for business use is to have a simultaneous Skype call or conference call — and then you've got the question "Since we're in Skype anyway, why do we also need to be in-world?"

      You need to have voice integrated in an easy, transparent way, so that users don't need to make any extra effort to have it going, for OpenSim to make sense in a business setting — and since that's what this publication is all about, that's what i care about most.

      • sarahblogging

        Which is exactly what I said in another comment on the missing killerapp. Using all the functionality Skype and other tools provide inside an OpenSim is like taking (MS-)office there. If you add Video to the mix its even better than anything an avatar could provide.

        I have a SL-Friend who tried to use this kind of functionality, espacially meetings, with his clients, and it didn't work.

  • Whisper is great but needs an extra plugin to switch the viewer, Freeswitch works out of the box without changing the viewer's plugin. It could also be an option to improve Freeswitch.

    Sharing your desktop is possible when you have a web on a prim page where you have a Ustream.tv stream from your computer. With the Ustream producer you can switch between a part of your computerscreen and/or your webcam.
    The only thing is hat you need at least web on a prim like in the Imprudence 1.4 version. There is a dalay of about 3 seconds. So all things you say or do will be heard and seen 3 sec. later.

    The same Imprudence viewer can also show google docs on a prim that are shared. So anyone can work on the same document and even draw together. The only negative point is that loading of the docs and drawings can make the viewer look like it stopped working. It needs to load very much info, but after that one minute waiting it all works. At least, on my computers.

    In Chatterdale, the virtual language village we built, we have 16 buildings where conversations can take place. Anyone inside this building will hear what is said, outside the building no one will hear. It works great, even with Freeswitch. Regionwide messages can always be sent via text.

    Most problems have a solution, and when it is not there yet, it wil come. Keep asking.

    Nick Zwart – 3DLES

  • for number two you could do Google Docs presentation dealio – it is very similar to PowerPoint and i have used it for some webinars i did with the eLearning Guild. since it's Google in a browser, it works in-world with media on a prim

    i would LOVE in-world documents! and a way to take notes in-woprld, like some form of in-world Notepad

  • Hi Maria,

    We can't afford to pay anyone but I'm definitely with you on the need to get a working viewer/server Whisper solution (if we had the human resources to spare we would implement this ourselves). I also think it needs to have support for a steep voice level drop-off, a public announcement system and a cone of silence.

    Voice implemented correctly, i.e. your #1 & #5 items, IMO, is the number one missing feature from your list. The viewer already has an option to turn voice off for people who don't want it and there are ways to get #2, #3, and #4 using websites embedded using media on a prim. However, getting avatars to talk with mouth movements and speaker indication isn't currently possible in OpenSim unless you use Whisper which, as you correctly stated, is perpetually a few months from being ready. A bounty might help get this feature gap in OpenSim finally closed.

  • Aurora-Sim made Cone of Silence sometimes ago. Very nice feature we often use for meetings and conference.
    Docs here https://github.com/aurora-sim/Aurora-Sim/blob/mas

  • This is what I have been saying all along here. FREE leaves a lot to be desired. Low prices also. It destroys quality and proper development. Progress is stalled and only money can do the difference. Yes OPEN SIM is a free open source code. Many of us are taking advantage of it. But still not far from the truth. It is what it is. A free product and because of that yes it needs lots of work and investment.
    This works not only for the development of a product but also for a Virtual World's economy and a well balanced market approach.

  • justincc

    I think that voice is very important too, and I'm surprised the Whisper module has fallen back a bit. However, all the code is open-source (at https://github.com/vgaessler) so anybody could pick it up and start working on it.

    Also, from the Mumble website it looks like it should work just fine with Linux and Mac OSX too, so I'm not sure what the issue has been with Whisper on these operating systems.

    As for application screen-sharing, this is a difficult problem – Lawrence is right in talking about the technical challenges. To be honest, I think this is going to become less of an issue as more apps move to the web & the cloud, as they can be used via the webkit browser. Screen synchronization is possible but the app has to be built specifically to allow this. I think this will become more common as collaboration becomes an even bigger feature of the web.

    As ever, many of these things are driven by code contributions or sponsorship, so the kind of good work that has been put in by companys such as Daden and Kitely is very welcome.