Kitely: open source virtual worlds simplified

Like a lot of virtual worlds observers, I\’ve written repeatedly on the need for virtual worlds like OpenSim and Second Life to be simpler to use – ideally web browser based. Kitely, a project underway since 2008, takes a big step toward achieving that by making the establishment of an OpenSim grid nearly as simple as it gets.

It took me under ten minutes to get set up in Kitely. Here\’s how:

1. Log in via Facebook Connect.

2. Install the Kitely plugin (Mac users note: Safari or Opera aren\’t currently supported by the Kitely plugin at present, you\’ll need to use Firefox or Chrome).

3. Create a world and choose if you want to invite anyone from Facebook groups you are part of:


4. Type a name, optional description and type of world you want to start with.

5. Click on \’Enter World\’ and your SL browser will launch (Mac users again – there\’s a known bug whereby your username and password are all entered in the Name field of the SL browser – you just need to type in that password and delete it from the end of your name)

5. Voila – you\’re now on your own island / collaborative space:


6. Three minutes later and I had my venerable log cabin rezzed on my island:


Kitely is currently in beta, and the currency used is called a KC. As part of the beta you get 50 KCs currency to start with and it costs 1KC per day to keep each world you create. On the proposed maximum discount structure that works out at US ten cents per month. It\’s an attractive proposition for someone not wanting the hassle of creating their own grid from scratch and its more than competitive with other providers. The support functionality is fairly well set up and responsive from what I\’m seeing.

There\’s still plenty of kinks to iron out but Kitely is a superb snapshot of what is going to be required for wider adoption of virtual environments: simplicity and integration with other platforms. I\’d be interested to hear from anyone who has experience in creating content in OpenSim as to your thoughts on comparability to other offerings. It\’s also worth having a read through the Kitely FAQ, which covers a lot of stuff including the approach to intellectual property (essentially the same as Second Life) and the Terms of Service.

Thanks to a number of Metaverse Journal twitter followers for the heads-up.

Update: With thanks to reader Psx_kai, who pointed out a key fact I\’d missed in the story. The pricing I described was correct but didn\’t include the extra charge of US$0.20 per minute for each visitor to your world. That\’s certainly going to get pricey after a while although it seems it\’s going to be an option to earn a fair whack of free KCs. On the upside, the \’pay for what you use\’ model is something that can work well for those wanting intermittent events without the ongoing higher monthky costs on say Second Life.

Second Life in a browser on the near horizon

\"\"\”Project Skylight\” is its name apparently, and it\’s the project that will hopefully deliver Second Life in a web browser. Tateru Nino got a confirmation from Linden Lab that testing will be occurring, so hopefully we\’ll see some concrete examples in the short-term.

It\’s a well overdue development, and one that is not so much innovative as essential for Second Life\’s longer term survival. I know that for me my time in SL use will increase with a browser based version, as I can do it alongside my other browser activities. It\’s a no-brainer really and kudos to the developers making it a reality.

The ins and outs of Intel’s OpenSim scaling

By now, you’ve probably read about Intel’s experiments in boosting the performance of open-source Second Life workalike OpenSim to very large numbers of users – or at least very large numbers of users compared to a traditional Second Life simulator.

You may have seen the video, if not, it’s here:

All of this, ultimately, is apparently going to become an open part of the OpenSim codebase.

Unfortunately, the potential utility of this is a bit limited. It works fine for ScienceSim, at present (albeit it is considered more of a demonstration than a practical system right now), but the possibilities of deriving large benefits from it if you’re not already a well-heeled organisation are actually a wee bit limited.

The system uses a ‘distributed scene-graph’ technology in a form of computing sometimes referred to as distributed- or cluster-computing. The distributed scene-graph slices the simulation-space up into optimal chunks, based on workload, and parcels out the workload to other servers, while keeping processing in lockstep so that no part of the simulation races ahead or falls behind. Here’s Intel’s Dan Lake’s slides on how it works.

The very first barrier of this solution then is hardware. You need a number of capable servers, and the simulation could wind up limited by the ability of the slowest server to cope with the load.

On the other hand, the same cluster can deal with a number of simulators concurrently, so long as things don’t get so busy as to overwhelm the hardware cluster.

The biggest issue, really, is bandwidth. The servers need to shovel a quite astonishing amount of data between them, and the cluster as a whole also needs to be able to deliver bandwidth to every client with a viewer.

If each viewer has its bandwidth slider set to no more than 500, then we’re looking at up to 500Kbps of data for one user. Ten users is up to 5Mbps, the 500 users shown in the video potentially runs up to 250Mbps. Many Second life users will tell you that 500Kbps for the viewer doesn’t exactly yield a snappy response when things get busy, so the peak bandwidth loads back to individual viewers could potentially be much higher.

So, what we’ve got here is a great technology, and a solid step forward in virtual environment simulation, but for practical uses it is limited to very-high-speed local networks, or to companies for whom the costs of hardware and high-capacity network connections are not really much of a consideration.

Interview – Kyle Gomboy, CEO of ReactionGrid

\"reactiongrid2-sml\" ReactionGrid is one of the OpenSim-based grids that\’s been making an impact during the year. The more formalised partnership with Microsoft has only increased the momentum for ReactionGrid, as has the increase in educators moving or diversifying to OpenSim grids.

I took the opportunity this week to catch up with ReactionGrid\’s CEO, Kyle Gomboy, to have a detailed chat about the company, its plans and the wider challenges for OpenSim grids.

Lowell: Can you give a potted history of how ReactionGrid initially was created?

Kyle: ReactionGrid was an experiment in controlling & developing our own virtual world experience. While managing events for Microsoft we found too many issues with X-Rated content & problems like a rolling restart in a well planned Microsoft event of over 300 attendees. We are thrilled to say these preventable situations no longer occur during our events on Opensimulator & our PG TOS has helped prevent any mature incidents. We now have Microsoft moving here & an educational and business community forming fast because of these decisions to focus on reliability and security.

Lowell: While we\’re talking reliability – why are you able to avoid issues like the restarts? Is it because of ReactionGrid\’s smaller size or do you have a different approach from an infrastructure viewpoint?

Kyle: Both actually. Our use of Hyper-V virtualization software has allowed us to give clients the option of a dedicated server which they can lock down to users registered to their world. When you now know exactly how many users will attend, exactly what the build is and exactly what resources you have, you can begin to plan for reliability for any platform.

Lowell: So the obvious question is why don\’t grids like SL have the same approach? What\’s the downside, if any?

Kyle: The downside is these smaller worlds are just that, small. The upside is with Hypergrid world hopping technology we can allow shoocls and business to go private then link up \”Stargate\” style anytime to any other world. SL doesn\’t have virtualization because their plan is a huge monolithic grid to serve all. Our plan is a galaxy of smaller, niche interconnected worlds. For this reason we rely heavily on virtualization. It also lets us migrate or copy your public world locally behind firewalls for the ultimate in uptime and security.

Lowell: To step back for a minute, can you talk a little about your own experience in virtual worlds pre-ReactionGrid?

Kyle: My experience started in 1995 when I was working as an aerospace test engineer. I started mixing 3D visualizations with data to help engineers learn how to improve part performance. I ended up using 3D to train workers worldwide, to display our parts at tradeshows and to educate our sales people by flying them in from around the world and showing them 3D animations of new product. Now there\’s the rub, we flew them in! I thought wow we could have a Quake or Unreal engine and show these CAD parts and movies in 3D. I had to wait awhile when we launched our company to prove that but now is the time.

Lowell: So what was the catalyst for you starting ReactionGrid?

Kyle: Simply running into too many development and enterprise acceptance roadblocks. I hate being slowed down, or if there\’s trouble being able to hands on fix it. After hosting web applications for almost 20 years now, we do not like our hands tied when serving our clients. Plus, how can we do such amazing events like when we taught C# inworld with Microsoft and recreate the \’39 World\’s Fair without kids? We badly wanted teachers and students to come inworld with us.

Lowell: So what platform were you using before ReactionGrid?

Kyle: We started in 97 with Macromedia Shockwave with Havok1. We got slammed by business clients for the choice. They weren\’t ready at that time for ANY downloads or plugins. So we waited for a true platform unitl we found SecondLife. We loved their concept of inclusion with user created content. So we started there really but now we have Opensimulator and recently Unity3D for our own scratch built world.

Lowell: For the non-techie, can you explain the difference between OpenSim and Unity3D?

Kyle: Sure. Open Simulator with user created content, for the near future will require a client download to fully utilize. We need something to compliment OpenSim that can operate in browser & mobile. Unity3D is perfect for this. It allows you to build your own world concept from scratch. We plan to use its .mesh import abilities to provide richer experiences for those not needing to build their own content. We will in fact be offering templates and training soon for others to learn with us.

Lowell: So are the two platforms seperate or will RG users who\’ve created content in OpenSIm be able to have others view it in a browser?

Kyle: That\’s a great question. The plan is a merging of the platforms abilities over time. Like a web content management system we store all our data for content in a database. This means we could share that info with any other systems that can query it. A little mathematical transform in between to put OpenSIm\’s data in another world is definitely a part of the roadmap R&D.

Lowell: Which leads to the related issue of interoperability: at the broadest level, do you see much progress occurring in ensuring interoperability between environments?

\"reactiongrid3\" Kyle: We had hoped the AWG group in SecondLife would mature as a basis for interop. Why not start with your most compatible world after all? But they are not keeping up with the speed of the Opensim dev\’s with items like Hypergrid now linking any Opensim world. So since we cannot get hands on to help there on the SecondLife side we have decided to experiment with our own interop focusing on the access we now haveto the worlds core database store. We can get to the data now so anything becomes possible.

Lowell: So to some extent Linden Lab is lagging the OpenSim efforts?

Kyle: Yes, recently a Linden began participating in the IRC channel with the developers so there is hope. The problem I think is the Lab deciding whether connecting to us would benefit or hurt them. We feel that kind of openess will propogate virtual worlds faster in the mainstream but I can understand all sides of the issue. I just know the web would not be the web it is if it worked like virtual worlds do now.

Lowell: Are the OpenSIm dev community still fairly tight-knit? Do you get the feeling everyone\’s still working pretty much to a common purpose or are you seeing commercial considerations increasingly creeping in?

Kyle: I think the Opensim community is incedibly tight knit. What is interesting is it is much like a company with stress, deadlines, support always testing your nerves. But in this case the company\’s struggles are fully exposed to the world. So it can seem at times there\’s infighting, but it is usually just normal wrangling over how things should be done best. An illustration of our working together is Hypergrid itself where many competeing worlds now link together and have regular cross world meetings and events.

Lowell: On that note, who do you see as ReactionGrid\’s main competitors?

Kyle: Second Life & OSGrid are 2 worlds that are run very well & offer options we do not. On our side, we also offer many optins they do not as well. For this reason in terms of attracting users, developers and more, we feel those two worthy worlds are a source of concern. At the same time, we reach out to both for partnerships & have had great success with OSgrid in this way. If we all row in the same direction we\’ll get to our destinations a lot faster.

Lowell: Which leads to the partnership with Microsoft. Can you talk a little about how that came about and what opportunities it offers you and the wider community?

Kyle: I have been a Microsoft developer since day 1 of programming for me. One thing this company does right is cater to developers. When I arrived in SecondLife I simply hoped to move my skills from Visual Basic to C#. I instantly was helped by Chris Hart who now is 1/3 owner in ReactionGrid. Over time I began to return the favor with ineractive games, not slideshows, to teach C# to new users. Over time because we used the medium properly they began to invest more in virtual worlds. Soon we spread to dozens of internal departments and to this day they all continue to try various inworld events. With the move to ReactionGrid, the plan is simply to do the same thing but now include teachers, students and families in our educational events with Microsoft. They are even developing and donating spaces for free for low budget schools and innovators to use anytime. What\’s also important on that note is you can script inworld on opensim in C# or Visual Basic or even JScript like you would with LSL2 and in fact can mix the code launguages together! This, as you can imagine, is loved by Microsoft who has suggested a 3D world toolkit with us for Visual Studio 2010.

Lowell: Really? So is that a confirmed feature at this stage?

Kyle: No, just part of the ideas for next year being tossed around. Step one is to engage the Microsoft business, teacher and student communities and see how they can help foster fun learning here. We may find they want something else other than our suggestions – teachers are amazing at innovation and this spreads to the students once they get the hang of things. So we like to listen first to our users, then generate and produce our own ideas after that.

Lowell: You mentioned earlier about ownership – who does own ReactionGrid and can you disclose any future plans around growth?

Kyle: Myself and my wife Robin Gomboy are in an equal partnership with Christine Hart in the UK. We recently incorporated as a \’for profit\’ in Florida, USA. We plan to grow initially with hosted, turnkey worlds with great support ( www.metaverseheroes.com free service) and have ideas for including our clients next year with ideas like sim ownership equals a small stock investment in the company, and other ideas to share guidance of how we operate with our end users. We believe our single file download of worlds behind the firewall recently deployed will eclipse our hosted solutions one day. We also feel by delivering superior support we will surpass most other systems in terms of customer loyalty.

On growth, since launch in January we\’re up to almost 5,000 users platform wide, with over 60 independent servers and almost 500 sims platform wide, 150+ here on our world. So extrapolating to next year, we could be pretty huge hehe….

Lowell: One of the biggest challenges for any virtual platform provider is governance. As you move beyond the current 5K users on 60 servers to five or ten times that, how do you see ReactionGrid playing its governance card?

Kyle: Keep in mind ReactionGrid is a platform for virtual world hosting first and foremost. We have our own world of course named the same which is an illustration of the power you get with Opensim. So while this world does grow itis far eclipsed by our resellers and private world owners already. We govern here simply with a PG kind of ethos. Our client and partner worlds handle ther own TOS and other governance. This is exactly like web hosting where we play no part in your worlds rules if hosted with us. If it is legal it is ok with us. As far as growth in this world we\’re focused on education and business and have laid down rules similar to those environments and have created a culture here that accepts that in order to be able to bring managers, school administrators and others inworld, they need to experience the medium safely. So we\’re hiring former teachers, architects, estate managers and more to help us as we grow on thisparticular world.

Lowell: So the obvious challenge for any business is to make money – what is ReactionGrid\’s primary commercial model now and is it likely to change into the future?

Kyle: The business model is three-fold. 3D world development is primarily our team integrating with other systems and applications. On our http://outpost.reactiongrid.com store you will find tools for business & schools to connect to 3rd party systems for single registration or remote user editing. We also are now building an ecosystem with other content creators. Our hosting services of course provide turnkey virtual worlds and web based administration for mainstream users. So development, ecosystem, hosting are the 3 tentpoles.

Lowell: So if you were able to sum up ReactionGrid\’s value offering versus others, what would it be?

Kyle: ReactionGrid is focused on the \”launchpad\” of your ideas. Our belief is when we deliver users their part of a world the process doesn\’t end there. We follow up with you. We help with sim challenges and how to improve performance. We build tools & create fixes based purely on what our users ask for. We want to get the basics right of good customer support before all while Opensim is alpha and far after it matures. Support is key. ReactionGrid sees customer service before and after the sale as critical to launch-padding our client\’s ideas.

Lowell: And you\’re confident of being able to deliver that personalised service ongoing as you scale in size?

Kyle: I am fully confident. With almost 20 years experience in such care and the fact that since Jan 1 2009 we have doubled in size every couple of months, we are already on track there and simply cannot drop the ball on this.

Open source virtual environments living server-free


Most people familiar with the Second Life grid are also aware of the existence of OpenSim technology, commonly thought of as the Open Source alternative to Second Life.

With OpenSim, you can create your own virtual environment grid without needing to pay for licensing. The grid can be made open to the public, or be kept private, only available to those on your side of the firewall.

What is the difference, then, between the OpenSim concept, and that of Open Cobalt and Solipsis? Essentially, OpenSim grids are designed to be served from a common point. Open Cobalt and Solipsis implementations are designed to be served from many points – they are both peer-to-peer technologies.

Open Cobalt: specific market niche

Open Cobalt consists of two parts: a browser and a toolkit. The browser is used to view the 3D virtual workspaces created with the toolkit. Each workspace can live on a separate personal computer. Workspaces are real time and computationally dynamic, and each can host multiple participants. Additionally, individual workspaces can be interlinked into a private and secure network of workspaces.

Open Cobalt is based on Croquet technology. Squeak is an open source software development environment for Smalltalk-80 programming purposes; the Croquet system is derived from Squeak. The Croquet system features a peer-based messaging protocol that eliminates the dependence of a virtual environment upon a single server or server cluster, and that fosters the creation of highly collaborative workspaces. The Croquet software developer\’s kit (SDK) was released in 2007, after which development under the Croquet umbrella ceased. Further development of Croquet has continued under the Open Cobalt banner.

Open Cobalt has a number of very attractive features, particularly for researchers, educators and students:

  • Open source licensing (MIT).
  • Deeply malleable, collaborative space.
  • Runs on Windows, Mac, and Linux.
  • Internet access is not required; it can run over LANs and Intranets.
  • Private environments can be created. This eliminates the incidence of griefing by outsiders.
  • Public environment can be created. This brings richness and diversity to learning environments.
  • In-world text, voice and video chat, web browsing (VNC allows access to browsers like Firefox) and annotations.
  • Access to remote applications via VNC.
  • Navigation between virtual workspaces is possible using 3D hyperlinks.
  • Workspaces can be easily saved and restored.
  • Mesh, texture, media, and whole avatar imports are possible.

Open Cobalt was started in January of 2008 by Julian Lombardi and Mark P. McCahill of Duke University.  The pre-alpha release of Cobalt (downloadable here) was announced in June 2008. Since then, Open Cobalt has progressed in leaps and bounds, featuring more functionality and more extensibility. The beta release is due this year, and a full implementation is expected to be released in 2010.

Solipsis: our market niche includes everyone.

Solipsis is also open source, and also features de-centralisation of computational work and data storage. Nonetheless, its background, implementation and philosophies are of course quite different from those of Open Cobalt.

Solipsis has been developed by French R&D partners Orange R&D, Artefacto, Archivideo, IRISA and the Université de Rennes II. The product, which has been available for download for some time, is currently in beta testing, though that is slated to be completed soon.

The Solipsis 3D project grew out a prior 2D project; the 2D browser also featured a peer-to-peer facility, and thus allowed users to engage in chat sessions without the use of centralised servers.

Beginning in 2006, with a time-line of 30 months to completion, the Solipsis 3D universe and the advanced modelling tools should now be available.

The Solipsis team has a rather grand notion of the position it will hold in the future: they desire it to replace and greatly extend the Web as it exists today. Far more than just creating a metaverse in which to communicate and collaborate with other people, they also see Solipsis as a potential way to store and present data in a more meaningful way than the conventional Web does now. Additionally, they hope that Solipsis will conquer scalability issues, promote usage and creation of high-bandwidth services, and that it will be self-organising – any part that is cut off from the rest of the metaverse will be self-sustaining.

The Solipsis GUI presents as both a stand-alone navigator, and as a Firefox plug-in and ActiveX component.

New Zealand OpenSim grid on the way

I stumbled across a communication from a consortium of New Zealand Universities, who are making a public call for support for their proposal (which is reproduced in full below):

I’ve been asked as a contributor to OpenSim to pass along the following.

There is a bid going on for a new public project in NZ, through the University of Otago, and the principles (researchers at the university) are making a general call for support for their bid.

If you could sign a letter of support for this bid within the end of the week, please forward it to me, and I will pass it on.

“Support” can be any thing of your choosing, and is non-binding. Samples are appended to this email.

I have been authorized to pass along the following excerpt from the application:

Executive Summary

Project Title: National Virtual World Grid

Project overview

Rationale for the project

The National Virtual World Grid (NVWG) project will develop an open access national virtual world grid based and bring together a community of researchers and educators with an interest in playing an innovative role in the development and usage of the next wave of the Internet: Web3.D.

Continual advances in telecommunications bandwidth and computer processing power are already giving users a more multimedia-intensive experience on the Web. Now with the advent of online virtual worlds (e.g. Google Lively, Second Life) we have the harbinger of the next wave of the
Internet: Web3.D, where users inhabit and participate in a 3D Virtual World (VW) by means of their
personal in-world avatars. An IT industry research and advisory company, Gartner, has predicted that by 2011, 80% of all Internet participants will be regular users of Web3.D technology. Already over 500 US universities are offering courses of instruction in Second Life, and major companies are
getting involved too (e.g. Vodafone, IBM, TelstraClear, Sun). The commercial nature of the most widely used Virtual World implementations, such as Second Life, places high financial costs on users who are charged for developing in-world content and for the traffic associated with accessing the remotely located servers.

This proposal describes how through the development of an national virtual world grid based on open
source software deployed on KAREN, and activities to promote the development of an active user
community, NZ tertiary educators and researchers can be given an accelerated opportunity to play an
innovative role in Web3.D development and usage.

Project outline

The National Virtual World Grid (NVWG) project will operate NZ-based computer servers running the OpenSim software (an open-source clone of the core functionality of Second Life) in “grid mode”. This will encourage uptake of Web3.D technology across New Zealand by enabling organisations connected to the Kiwi Advanced Research and Education Network (KAREN) to obtain free and high speed access to virtual world “regions” hosted in our servers, or to interconnect their own “region servers” with those of other institutions in NZ.

By developing, managing, and operating this National Virtual World Grid, and Web-based community
support tools for users, there will be an opportunity for all institutions that are connected to
the KAREN network to:

1) undertake both experimental and routine use of virtual worlds in teaching and research;

2) develop engaging, interactive in-world content customised for New Zealand use; and

3) develop new context-specific plugins enabling interaction with external applications that are
connected to the real (non-virtual) world and external services.

Expected outcome

The project will:

* Promote the establishment, growth and maintenance of a community of NZ users of Web3.D technology for tertiary education and research
* Provide the environment for New Zealand to become an innovator in Web3.D technology at the
international level.
* Facilitate the tertiary education sector to produce graduates ready to exploit Web3.D technology for the economic benefit of New Zealand

Project partners

University of Otago (Department of Information Science): Project leadership and management; grid
development and management of central grid services and Web-based community support tools.

University of Canterbury: deployment and management of Canterbury grid node (SIM); exemplar VW learning environments for two learning application areas.

University of Auckland: deployment and management of Auckland grid node (SIM).

It\’s great to see such advocacy coming from New Zealand universities. I\’m more than happy to be corrected on this but I\’m unaware of such a project underway in Australia. There\’s an enormous amount of development going on in the Australian university sector but is there anything as potentially unifying as the New Zealand model?

Thanks to Mo Hax (via Mal Burns) for the heads-up.

Second Life on your iPhone

…well sort of. Dynamo teen developer Katharine Berry has created an iPhone (or iPod Touch) version of her AjaxLife application.


Using it is as simple as pointing your iPhone\’s web browser to http://ajaxlife.net/?iphone. Once logged in you can send and receive IM\’s as well as see which of your friends are in-world.

Amazing work from a solo operator.