Christian Renaud, Chief Architect of Networked Virtual Environments for the Cisco Technology Center will be launching a new virtual worlds tracking organization called the Metaverse Market Index (MMI) at Virtual Worlds Fall next week.
MMI will be established as a not-for-profit organization that will define and track metrics across multiple virtual worlds.
The problem is simple; there is little or no reliable and comparable data about user engagement and economic activity in virtual worlds, which makes it difficult for businesses to select the worlds in which they should invest their time and money.
MMI is founded by Nick Wilson of Metaversed.com and Robert Bloomfield of Cornell University, and will team with advisors and sponsors from industry and academia to construct a database of statistics on each platform’s user activity, economy, and technology. MMI is hoping that their work will foster informed investment in online worlds, “while also allowing developers to identify opportunities for rapid convergence toward common technologies that will allow portability across platforms.”
Data gathering will focus on three main areas: User Engagement will measure key indicators of how successfully businesses can reach customers and business partners, Economy will be a set of measures intended to capture economic activity in each world, and Technology will be focused on developing a list of key technological attributes of virtual worlds with an eye towards allowing developers to identify standards and interoperability opportunities.
MMI first step is to incorporate and work with advisors towards identifying key statistics. January 2008 will see tools development based on the initial key statistical areas, followed later by publication of the standards and data gathering.
It’s an interesting setup, and an inclusive way of building a solid foundation of usable data for businesses currently dealing with/ in virtual worlds, of those looking to do so. The setup is not unlike the Motion Picture Association of America, in that the enabling body is virtual world and corporate agnostic and in theory will undertake its activities for the good of all involved. I’ll be watching with interest the data they start publishing next year.
This concept is completely necessary for evolving businesses. And awesome I might add.
the platform developers have done it again…..very impressive
http://vidsonly.blogspot.com
What are they talkin’ bout?
A promising development – if true independence can be established this will be a success…
As someone involved in V-metrics for almost two years, and as the creator of the first commercial V-metrics tracking package, V-Tracker, I’ve given some thought to the topic of what defines a Virtual World Metric. There is so much activity that goes on inside a virtual world to consider, and then there are the after-effects of virtual worlds that end up as You Tube videos, media write-ups, or blog posts. In the end, I decided to go to the root and here’s what I’ve come up with:
Visitor Tracking – Determining who is where, over what period of time.
Visitor tracking is the heart of a 3D based virtual world metrics, as it will tell you where people go. There are many ways to get this information, and a wide spread of options as to the actual time scales used. Personally, I find that one minute intervals are most appropriate for tracking visit activity, as it takes that long for most environments to ‘rez-up’ and be recognizable. Also, there is the issue of granularity, or how much level of detail you need to know about specific short scale movements. Is it important to know that the avatar ’shifted his feet’ and moved a single meter to the left? Or is it more important to know if the avatar stood by a kiosk that played videos for the length of time it takes the video to play? Scale and granularity are important factors and as there are no standards yet, I’m thinking each V-metrics provider will come up with their own solutions, and each installation will have its own requirements. Something important to consider is how much bandwidth will be consumed and how much data will be generated by a V-metrics installation. Let’s consider a single island, and let’s make it popular, with 50 concurrent avatars visiting on average over 12 hours of every day. If we use a ‘Visit Minute’ as the root of our tracking scale, we’ll come up with the following:
1 Visit Minute x 50 avatars x 60 minutes per hour x 12 hours per day = 36,000 visit minutes/day
Run for one month = 1,080,000 visit minutes/month
To recap, one popular sim, full half the time, creates one million visit minute records in a database, using one minute intervals. Some people may read this and their eyes will gloss over, however anyone who has ever worked closely with database technology will immediately see what an issue this is. It can quickly become a staggering amount of difficult to manage data when you start to track even dozens of sims simultaneously. For this reason, trade-offs in scale, scan distances and granularity, and processing performed during each scan need to me made. Good architecture and data warehousing technology is absolutely necessary.
There are a few other areas of V-metrics that I feel deserve discussion, but the concept of a visit tracking metric root, or the visit minute, should give a nice point to start the discussion.
To Be Continued . . .