Between 1995 and 2001, Microsoft developed and operated it's own social virtual world. Named V-Chat, the service allowed users to create their own text-based, 2D, or 3D environments in which they could chat and interact with other users. In it's six years of active service, V-Chat saw two major iterations, an active end-user community numbering in the 'thousands', and gave rise to an open-source spin off in the form of Microsoft's Virtual World Platform. In 2002, several Microsoft technicians (Cheng, Farnham, and Stone, 2002) collaborated on a book chapter which outlined nine design guidelines that they learnt as a result of their virtual world experience. In relation to Second Life, most of these have been adhered to, however quite a few remain elusive. It's these elusive few, that I'm going to talk about in this post. Headings come from the opening summary given by Cheng and colleagues.
"Provide persistent identity to encourage responsible behaviour, individual accountability, and the development of lasting relationships"
For Second Life, persistant identity both is and isn't a problem. Although by and large, most people have a primary user account and a small handful of alts without causing any problems at all - a lot of people take pleasure in creating account after account to perform illegal, or simply nuisance activities. Many papers make the point that much of this relates to an inability to penalise an individual, or even an invested identity, rather than just an account. As Cheng and colleagues pointed out, creating real-world links between an individual's avatar and real-life identity seemed to be very effective at reducing the amount of in-world griefing. Although Second Life sometimes ban accounts, or IP addresses, it's difficult to ban an actual person. Perhaps if rumoured Facebook integration goes far enough, this might change to some extent. Account verification for all could be another possibility.
"Support custom profile information that addresses the privacy concerns of individuals."
As V-Chat developed it became clear to the MS team that users were not filling in their profiles with accurate information. Some fields contained fake information, others were left blank. When developing their Virtual World Platform, they instead decided to implement, a permissions feature, not unlike the one that can be seen on Facebook. Specific profile items could be assigned "Private", "Friends Only", or "Public". What the team discovered was that not only did this result in both a greater amount of fields being filled and with accurate real life information, they also discovered that users were taking the time to read other people's profiles more often. Prior to this, profiles were largely ignored.
"Help people coordinate finding and meeting those they care about to increase the likelihood of positive interactions."
One of the major problems with Second Life is networking. You only need to look at how many SL users have signed up to Twitter, Plurk, and Facebook in order to socialise with other people. This had been identified early on by Cheng and colleagues who write "being unable to coordinate finding similar people or meeting friends in a virtual environment is a barrier to entry." Perhaps then, we can reflect on some of Second Life's recent bad press as a networking failure by the journalists involved.
"Different communities have different needs and require different user interfaces."
By integrating active scripting and DHTML, Microsoft enabled users to customise aspects of their user interface in order to better suit different purposes. Typically these were customised by the world builder themselves, so much like a Sim builder would create the virtual environment, so too would they have control over the user's client. The examples they give include custom user profiles, and side-bar bulletin boards. So popular was this functionality that the Cheng and colleagues report that almost all virtual builds included user interface customisation to some extent.
Although we see this to some extent with the inclusion of HUDs, Second Life builders or sim owners do not have such a major control over the user interface. Just imagine how different Sims would be if they did! You'd visit a shop and find a sidebar load up with their special offers, you'd visit a roleplaying sim to discover that you've got a dozen new fields on your profile to fill in character information, or an enterprise sim might give you access to minutes or documents. It would certainly be a different experience.
Summary
So there you have it, four design guidelines courtesy of Microsoft's Virtual Worlds Research team. As for the other five, well they were largely adhered to. Still it's strange to think how different Second Life would be if these extra few features were incorporated, particularly that last one - giving builders and sim owners control over the client could essentially be game changing - making virtual worlds customisable on every level. I live in hope!
References
- Cheng, L., Farnham, S., & Stone, L. (2002). Lessons learned: Building and deploying shared virtual environments. In R. Schroeder (Ed.), The social life of avatars (pp. 90-111). London: Springer-Verlag. Amazon UK | US
(Hopefully) Related posts:
- Second Life and Poor User Centered Design Yesterday I wrote a post about how copybotting can essentially be seen as the consequence of bad design, I spoke about how the design...
- Virtual World Research Up until now, most of my blog posts have been fairly opinionated errr… reflections on various things, events, etc. Don’t get me wrong, I...
- Do you Collaborate in Second Life? Interviewees Needed As part of my ongoing PhD project, I’m looking to interview around a dozen or so people about how they use Second Life to...
- Second Life User Activity in 2009 Given that we’ve just passed into a new year, I thought it might be interesting to take a look at some Second Life stats....
- Copybotting as ‘intentional human error’ I’ve been reading quite a lot of blog posts on the topic of copybotting recently. For those who don’t know, copybotting is the blanket...
Lessons for Virtual World Design Courtesy of #Microsoft http://a-res.info/?p=185 — quite a few interesting tips #SecondLife
Lessons for Virtual World Design Courtesy of #Microsoft http://a-res.info/?p=185 — quite a few interesting tips #SecondLife
Hello, I enjoyed your post. I am not familiar with Microsoft’s work on virtual worlds but I am curious about the following recommendation: “Provide persistent identity to encourage responsible behaviour, individual accountability, and the development of lasting relationships”
This is one of the first instances where I have come across researchers lamenting the opportunity for multiple, anonymous, accounts. I understand how linking alts to a global account might decrease bad behavior, but I am not convinced that such steps are called for.
Might this impose an unfair limitation on those who use alts primarily as a means for the exploration of alternate self identities? Or to enjoy a virtual world without interaction from other users? Research by Tom Boellstorff and T.L. Taylor suggest that these less objectionable behaviors are a major draw to virtual worlds. Is the trade-off worth it?
Hi Richard, thanks for your comment. Glad you liked the post!
You make a good point, as someone who has alts for both of those purposes I’d prefer that they weren’t linked — at least not in a way that was public. However, I do think bad behaviour is becoming more and more of an issue, particularly in relation to copyright infringement.
With regards to anonymity, V-Chat was one of the earlier public 3D virtual worlds and was well integrated with MSN. Although I’m not aware of the specifics, the article they wrote did mention that each account was tied to a hotmail email address and MSN profile. Assuming that these tended to be the user’s primary personal address, divulging at least some personal information to other users — and MSN itself — must have been expected by most people. Further still, one of the major uses of their subsequent virtual world platform was to create support communities for individuals with particular illnesses — so the human, empathic, connection would most likely have been a major draw. As a consequence of this, I feel the Microsoft team may have come to some slightly skewed conclusions as to the degree to which people were prepared to link their virtual and real-life identities.
Ultimately, the issue comes down to how you effectively punish an avatar. I think MS were right to say that the best way to do this relates to taking action against the user rather than just the avatar, but I don’t think it’s achievable in reality. Like you indicate, even if it was possible, it would most likely come at a greater cost to the everyday user than to those who were engaging in objectionable activities.