Posted by: Peter Quirk | August 1, 2008

Reasons to use Opensim/realXtend for team meetings inside the enterprise

I’ve been really busy putting together a couple of submissions for our Innovation Showcase, helping my son get established in his new apartment, and doing my day job which has hit a peak workload. That, and general slowdown on the realXtend front due to the July vacations overe there, has meant that I have haven’t had the usual time and slew of topics to write about.

I thought you might be interested in the reasons I put forward for holding team meetings in Opensim or realXtend rather than using conventional audio, web or video conferences. The following points are extracted straight from one of my submissions, with slight edits.

Thesis: Current collaborative technologies used in the workplace are difficult to use and distract us from the act of human communication


1. The problems with Audio Conferences


For everyone

  • Poor audio quality, lots of background noise
  • Complex dial-in menus

For presenter

  • Ensuring that each user has correct dial-in info
  • Managing roll-call
  • Gauging participant engagement
  • Capturing audience questions, sharing answers in written form
  • Keeping audience involved

For participants

  • Listening without talking for up to an hour
  • Holding questions until the appropriate time
  • Knowing when to talk to avoid a collision with someone else
  • Side conversations not supported
  • Catching up if late to a meeting
  • Leaving early without interrupting speaker to explain early departure
  • Finding the presentation content, which may be separate from the calendar entry
  • Tough to attend a meeting beyond an hour in length

2. The problems with Web Conferences


For everyone

  • Lots of context switching: Outlook, browser, phone, LiveMeeting console

For  presenters

  • Complex interface on top of separate audio interface
  • Need two screens to monitor user feedback while presenting
  • Can’t present video or audio when using phones for audio
  • Sharing application with collaborators is complex

For participants

  • Collaboration is difficult
  • Side conversations not supported
  • May see same slide for 10 minutes if presenter doesn’t pace presentation well
  • If presenter & content is engaging, half-day meetings are feasible, otherwise 1 hour max.

3. The problems with Video Conferencing


  • Limited support for group video conferencing (only certain rooms & locations)
  • Polycom video is expensive per user (>$100 per user) (Note: assumes you don’t want video to be relayed by an external service like Skype, Adobe Air Connect, etc. for security reasons)
  • Switching between video and slides requires switch gear and an operator
  • Playing video from presenter’s PC doesn’t work
  • Need a separate device to present audience feedback or questions to presenter
  • Often need a producer at the primary site and interlocutors at remote site
  • Need guaranteed bandwidth to ensure satisfactory viewing experience
  • Presenters often restricted in what they can do by fixed cameras
  • Can’t present the very small or the very large without pre-recording material and mixing with real-time video

4. Advantages of collaboration in virtual worlds


Meeting experiences are more engaging

  • Eyes, ears, voice, hands and brain all involved
  • Participants less likely to drift off to email and other distractions

Meeting experiences are more social, more spontaneous

  • Real team spirit develops when teams share a unique moment that sets them apart from others

More natural conversation

  • No dialing, no voice menus – just talk whenever you want, mute when needed
  • High fidelity sound even with cheap (<$15) headsets
  • Hear voices coming from the direction of each avatar
  • Start a break-out session by moving some distance away from the crowd and conversing with your colleague(s)
  • TXT  when necessary to share a URL, to clarify pronunciation, or to create a log of the conversation
    • No need to switch to email or IM.
  • Start a private conversation with someone by right-clicking on their avatar and selecting Call/IM
  • Leave a message for an absent colleague
    • Message will be delivered when they next login
  • All-in-one interface – reduced cognitive load

More interactive

  • Virtual worlds enable activities not available in other meeting technologies
  • Creating ad-hoc break-out sessions
    • Break into teams about 30 feet apart
    • Facilitator can move from one to another, joining in the conversations
  • Exchanging notes, gifts, photos and objects
  • Photographing and video recording scenes for later use
  • Reviewing detailed profile info of participants
  • Instantaneous creation of groups
  • Co-designing “real” and abstract objects
  • Shared browsing of web sites and VNC sessions
  • With little work, an application screen can be linked to a surface to demonstrate new or proposed functionality
  • Integrating external data. With some code development, real-time data from Enterprise systems can be instantiated as geometric objects (e.g. bar charts, thermometers, gauges) for review and manipulation by participants

Rich media support

  • Stream video onto any surface
    • Video is played at user’s discretion, asynchronously from any other user’s video stream
    • Typically displayed on a projector screen or TV set in the virtual world, but can be mapped onto anything
  • Create a gallery of video samples for your team to review at their leisure
  • Integrate pre-recorded company quarterly meetings and discuss as a group
  • Integrate MIT Open Courseware videos, TED Conference videos
    • Brainstorm as a group after watching one of these videos
  • Integrate live streaming media via Veodia service
  • Watch company all-hands video feed or a remote video conference participant at work or at home while enjoying the experience of being present with others watching the video
  • Add background music or podcasts to any parcel of land
    • Users can play them at their discretion


  1. Great article!

  2. Lots of pros. Any cons?

  3. Fair question!

    1) Hardware requirements, especially ATI or nVidia graphics controllers: I expect this to change with realXtend as it uses the Ogre graphics pipeline which supports Direct X as well as OpenGL. They still have to strip out the Second Life OpenGL stuff.

    2) Learning curve: People who have never played in 3D sometimes find it a bit daunting at first.

    3) Stability: OpenSim and realXtend are still Alpha quality code.

    3.1) Functional completeness: To protect newbies from themselves, you need a working permissions system. Bits and pieces work.

    4) Bandwidth: remote workers at home need DSL or Cable

    5) Network Latency: I haven’t had a chance to test this yet, but I surmise that someone in India accessing my platform in the US might find network delays annoying, both for voice and manipulating their avatar.

  4. Peter, I totally agree on your evaluation of communcation methods. I also find virtual worlds a better alternative. Sadly, the Opensim project is not reliable enough for us, and Second Life meets all requirements easily.
    So why promoting Opemsim instead?
    I have not yet found a single advantage except that people can develop on the code and behind their company firewalls. But then again, what is desirable about it if it closes out the rest of the world? ;-)

  5. Thanks for your feedback!

    I’m lumping Opensim & realXtend together as solutions for meetings inside the firewall, even though Opensim doesn’t have a fully fleshed out voice solution yet. (Dr. Schofield has demonstrated integration with the Asterisk PBX, though that might not be acceptable to most IT organizations either.) RealXtend does have a working voice solution, and also offers a Skype alternative. With the conversion of OpenLife Grid to realXtend and with the merging of realXtend functionality into OpenSim, we should have good alternatives to Second Life soon that allow us to talk inside or outside the firewall. Not to be outdone Linden Lab is currently testing a version of Second Life for hosting inside company firewalls. We shouldn’t ignore Croquet, Cobalt and Qwaq either, since they also support voice.

    Also emerging are the cellphone based solutions. The idea of being able to talk to anyone with a cellphone from a virtual world, and to call an avatar from a cellphone is really compelling.

    I think the real issues for voice across the firewall are compliance and trust. A lot of electronic communication is subject to regulation and e-discovery, so we need functionality for recording, archiving and indexing voice inside the firewall. Secondly, spoofing of voice needs to be prevented. You saw what an utterly simple spoofing of a cell phone call did to Sarah Palin. We know how many people sucked into scams by spoof emails. We need to ensure that we don’t create yet another worldwide architecture for conning people.

  6. Great summary and discussion. Thanks for adding the link to

    I’ve been going around “selling” virtual meetings around TED and Open Courseware so was very happy to see you suggesting it as well.

    My problem with virtual world meetings is the barriers (your “cons” above) plus the lack of non-verbal communications (except through “artificial” menu selections.)

    I’m looking forward to a blend of video and virtual, an advance as profound as voice coming to SL.

    Thanks again for your blog. I always enjoy your perspectives.

    Best wishes,

  7. I’d have to totally disagree with B. Koenig on this one. I’d bet I could count at least 25 reasons why I’d choose OpenSim over Secondlife, and the number one? Uhh control. There’s also maintain, modify, host and the fact that you own it, not Linden labs. I’ve been able to get a local standalone modrex build running along with external grids, connected regions to OSGrid and ScienceSIM but this external region of modrex still eludes me.
    I can’t tell you how many nights I’ve been lost in amazement with what’s possible with this OGRE-mesh rendering, oh yea there’s another one SL does not do. I can’t wait to see it stabilize a bit.

    Peter, do you think we may be able to talk you into a quick tutorial on configuring an “external” or publicly visible, modrex/rexserver standalone or standalone grid? I have yet to truly find a good one, so many variants especially in the last .31 to .5 releases. I’m getting a little confused in exactly what’s needed to authenticate an avatar with the addition or cable beach, worldserver, openID, etc.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s


%d bloggers like this: