Second Life @ the University of Michigan

Observations

My initial thought was that given access to a shared virtual space and time frame, the Sakai community would spontaneously self-organize BOF gatherings much like the activity that occurs during RL conferences. There are several possible reasons why that didn’t turn out to be the case however:

1. Not enough people knew about it: I don’t think this is the case. I sent several emails to multiple Sakai maillists so I think the word got out to the community. It could be that there was insufficient lead time for people to respond, but given that the time commitment was not fixed (people could show up whenever they wanted) and on demand, scheduling didn’t seem to be much of a barrier.

2. There was insufficient program planning: The “build it an they will come” model didn’t work in this instance possibly because there was not enough clear reason for participants to show up. Unfortunately I was unable to participate much during the actual event, so despite orchestrating access and getting the word out, there was no one actively planning events nor was I able to “be there” much during the conference. I was hoping the opportunities to engage in BOFs would provide enough impetus, but based on feedback I heard from at least one Sakai member there wasn’t enough concrete reason to show up (“hey, this is kind of cool” was not compelling).

3. It’s still too early to form a critical mass around this tech: Setting up an SL account, creating an avatar, and getting acquainted with the basics of the environment may have been too big a hurdle for moderately interested participants. I have not seen a huge amount of SL interest in the Sakai community to date, and indeed this conference was intended to try to spark some activity in this area. Other LMS/CMSs are undertaking integration projects with virtual worlds (Angel and Moodle for example) and I was hoping to drum up similar interest among the Sakai community. However if the Sakai community doesn’t already have a number of folks individually participating in SL, then entering the virtual world for the first time may be a significant barrier.

4. Competition from the RL conference: Because the initial expression of interest was in “putting the Sakai conference online,” I thought it would make sense to run the in-world activities in parallel with the Paris conference. The Sakai conferences are high-energy, 24x7 events with lots of ad hoc discussions and numerous valuable hallway conversations. I was hoping the people who were not able to travel to Paris due to cost or other considerations would want to be part of the conference (if only virtually) and would extend that sense of community, perhaps bridging the virtual and real world activities. To that end, links were provided to the streaming video from a couple of the Paris venues so that in-world participants could watch and discuss the events in real time (more on this below). As it turned out however I think that many of the most likely SL participants were attending the RL conference and were not inclined to go online while richer face-to-face interactions were available. Similarly, I’m not sure that the typical tech professional is yet ready to put himself in the mental space that “conference time” requires (e.g. shutting out other distractions, blocking out dedicated participation time, opening the mind to new ideas and personal interactions, etc.) for “merely” virtual events. When we travel to a conference I believe we enter into a “conference mode” that is different from our day-to-day work mode, and virtual conferences may not yet have the necessary imprimatur of validity in order for many people to make that transition.

Other Points of Interest

Being a relative SL newbie myself (I’ve probably logged a total of 15 hours in-world), several things struck me as interesting during the time I was in-world (in no particular order):

1. The dominant communication mode was chat, even though the voice feature was available. This may be because those particular participants were just more familiar with chat or that the technical issues associated with setting up a mic, getting levels right, etc. were too much for early stage users. A small group of us did use voice chat for an in-world planning get together and I found that modality to be highly usable. It should be noted that the chat vs. speech issue is widespread in SL so noting this tension is not a particularly novel observation, but it is something that conference planners need to keep in mind.

2. My overriding perception of the SL experience is of vast expanses of interesting, but ultimately lifeless spaces. Most of the time I’ve spent in SL has been spent alone. I suppose that may be more a function of my navigational choices (there are certainly any number popular attractions that have lots of avatars in attendance), but the places I frequented during this experiment (Wolverine Island, SLoodle and a couple of other educationally spots) were invariably empty. Again, this probably isn’t a major revelation, but it’s worth noting that people don’t seem to hang out in SL in the same way that they might indicate telepresence via a Buddy List in a chat client. Likewise newbies who just want to drop in to check things out may be especially disappointed when the place is empty. Once the eye candy effect wears off, the perception may be that it can’t be of that much value or there’d be a lot of other people there. Or maybe I just don’t have enough in-world friends (I have to admit to skipping the Welcome Island orientation primarily out of an effort to get away from other strangers in-world) ☺

3. Prior to the opening of the conference, I scouted Wolverine Island with an eye for suitable gathering places for small groups (remember I was assuming we might have a few BOFs going simultaneously at that point). I settled on the red pod of the “Bouquet”, the campfire, and the first floor of the Library (where the freebies are). I was given to understand that these spot had amenities for hosting small groups (seating arrangements, etc.). When we actually met in-world however, people just sort of clumped and began talking. Since this was an experiment I took a hands-off approach and didn't try to herd people to the “right” meeting place. With larger number of participants however ad hoc gathering could become a management problem for organizers. Again, outlining more programmatic structure at the outset is probably a good idea.

4. I was interested to observe how social distance and arrangement played itself out in our small group discussion. We self-organized into a roughly proximal space that seemed more distanced than it would have been had we been chatting in RL:

I suspect this is due to uncertainty about navigation (you don’t want to walk into someone with whom you’re trying to have a conversation) and issues with breadth of vision (it is possible to change your point of world, and even to detach it from your avatar, but the default view has a narrow field of vision than RL so you have stand farther back to get everyone in your view). More sophisticated participants may form tighter physical social circles.

5. One common issue with chat as the primary mode of communication is the difficulty of creating side conversations and multiple conversation threads in a single group chat window. We saw this issue a couple of times, but for the most part were able to focus on a fairly narrow topic. Again, more sophisticated users may be able to use private chat/IM to facilitate private conversations while participating in small groups. As it was, using a shared chat window for everything gave rise to some interesting etiquette and social norming situations.

6. The notion of identity, especially as it relates to the appearance of your avatar is surprisingly important. Customizing your avatar to reflect attributes you desire to project about yourself seems to be the primary activity of many new users (in addition to exploring large, empty spaces). The choices people make regarding youth, attractiveness, height, etc. have been discussed at length in other places so I will focus on the relevance of this issue to future conference organizers. In short, no one wants to show up in one of the default avatars (that’s kind of like showing up to the skate park with shiny new board – the more ground-up the better). The act of finding your true virtual persona may be very time intensive and tire kickers may be turned off by the burden of having to spend that time lest they perceived as newbies. In other words, having a respectable appearance takes a larger commitment than it does in other types of online collaboration vehicles which require as little as a clever .sig file quote or a decent Buddy List mugshot to be perceived as with it. I suppose newbies can simply buy premade avatars, but newbie cheapskates (like me) probably are not going to reach for their credit card as soon as they log in. A default set of avatars specific to the conference might bridge this gap somewhat. Indeed a few of us sported the SLakai Paris t-shirts that were generously produced by Patricia, perhaps in an effort to appear more native while reinforcing our Sakai affiliation. A custom logo and branded swag is a definite necessity for future in-world activities!

Modeling the SLakai Paris t-shirt in one of the sandboxes:

7. Because I was personally acquainted with almost all of the folks who appeared in-world, it felt somewhat awkward to address them by their avatar names. I think this was mostly my problem however as I didn’t see many other examples of people struggling with how to address each other. It was pointed out to me by Marc that because not every else knew each other in RL, using RL names would be confusing. Duh. Still I think it might be a little odd for professionals to begin addressing other colleagues with whom they have worked closely as “Bigby Hines” or “Lava Cherrybomb.” Again, I’m sure there are multiple dissertations out there on virtual identity choices and implications, but for professionals just starting to use these spaces for sober, “important” conversations there may be some growing pains.

8. Getting access to shared, but private, space in SL may be difficult for similar events in the future. The owners of Wolverine Island were generous enough to allow access to non-U-M affiliated participants for a short period of time, but the signup mechanism of putting contact info on a Confluence page and then being manually added to an access was a bit clunky. I think we lost some participants as a result. This issue bears more consideration as I don’t think we’d want to meet in a public SL space to try to conduct conference business. But managing access to private land is overly tedious at present. The obvious solution is a Sakai Island, but that may be overkill and cost-prohibitive. There may also be other ways of managing access (e.g. sending invitations to a space that expire after a certain amount of time) of which I’m unaware.

9. I observed three kinds of group interactions while in-world:

a. General socializing: this was of the “Isn’t this cool? How are things going? What are you working on these days?” variety that one might observe as people arrive at the conference hotel and meet in the lobby. Everything very light, no attempt to talk business.

b. Discussion of a shared object of focus (in this case a RealMedia stream coming from a live talk at the Paris conference). This was exciting in that it came closest to the idea of “taking the RL Paris conference online”. We watched the slides and talking head and had our own discussion about what we were hearing and seeing. There were some technical glitches due to bandwidth limits and the fact that we had to use the external RealMedia client (I understand SL will only stream QuickTime natively to in-world displays). But the mode of interaction was very promising with regard to extending the reach of a live presentation and providing a rich context for social construction of the event. Another participant (Harriet Truscott, who was in the UK) summed up the interaction in an email to the Sakai community like this:

“We weren't able to project the conference live feed onto our wall within Second Life as it needed RealPlayer rather than QuickTime. However, we were able to watch it in our browsers and chat in SL simultaneously. The UK people found the conference feed excellent quality - our US friends seemed to find it slower and with rather dodgy audio, which was a real shame, as Laura and I felt we were getting a great taste of the Paris conference. We really appreciated being able to watch and discuss the presentation with each other - much more enjoyable and engaging than watching the feed on your own.”

A small group of us watching and discussing the Sakai conf feed:

c. A more traditional presentation (one person has the floor and others interact with that lead): Unfortunately I had to leave by the time Harriet (as Hattie Heliosense) began her presentation of a new tool design by showing mockups, discussing some of the project issues and taking feedback. Since I wasn’t there I’ll let one of the participants (Dick Ellis) describe his impressions (lifted from an email he sent to me shortly after the event):

“Hattie asked everyone to give you feedback on the demo in SL. Although she
 wasn't able to get voice working properly and had to rely on chat, her
 introduction followed by live use of the tool by several of us was for me as
 effective as a live presentation. Since the tool was meant to be used by a
small group of students, having a small group of us working with it seemed 
quite natural. I remember [being asked once] how a pedagogically-designed
 tool would be different from any other tool.... Hattie's demo would have provided an answer to 
that question. 

I wasn't able to get in early enough to see the video from Paris, so I can't
 comment on how that went. When Hattie was away getting things set up the
 conversation lagged - as it did initially when you left before Hattie picked
 up the moderator role. There was a natural conversation among participants
throughout the demo, though, which provided the point of focus for
 discussion and sharing of thoughts. I imagine having voice would only have
enhanced that. I think further use of SL for Sakai does make sense.”

Harriet described her experience as a presenter thusly:

“I then gave a short presentation and demonstration of a new teaching tool that
we have developed at Cambridge. You'll have to ask my 'audience' how they found the presentation. John Norman will be pleased to hear that rather than giving my presentation in an Second Life simulation of an auditorium, the audience all stood round in a small huddle, and I accidentally started flying a few times during the demo. It grew very dark during the presentation, but luckily one audience member had brought along a flaming torch…If there's much interest, I might even try and give another demo - this time giving a taste of what we've been working on at Cambridge for our implementation of Sakai. Expect widgets, Facebook integration, javascript and much much more.”

So although I was not there to observe first hand, the other feedback I got was pretty positive. A couple things I did take away from the notion of ad hoc presentations were the neat tricks Marc showed us. The first was creating an object (like a large flat board) and then dropping a series of images onto it for display purposes. Instant “projector”. There are some drawbacks to this technique (many newbie presenters may not know how to create an object to project on to, there may be constraints as where such objects may be created, and there is a small cost associated with uploading images to SL), but it is a nice technique for quick and dirty mockup sharing. The second was the trick of posting a URL into the shared chat window, which when clicked, becomes a personal in-world browser for each participant. It’s not shared browsing but again is an easy way to reference data and images that may not be in-world. What’s more, I think these techniques could probably be taught to new presenters in an hour or less.

Lessons Learned

Based on the observations above, I think I would do some things differently next time around:

1. Have recognized leadership with time to organize the event: This is probably a no-brainer to anyone who has ever organized even a simple conference. There are a zillion little details that need to be worked out and most of the real work occurs before and after the event. I’m not sure the virtual world changes that dynamic, indeed in some ways it adds overhead (e.g. learning how to navigate in the virtual world needs to be learned whereas most real world conference goers already know how to find a room). Having even a small group of people with known responsibilities would go a long way to expediting decisions and making things happen. Enough lead time to get the word out is also a good idea.

2. Provide a program and avoid the notion of spontaneous uptake: My original notion of a self organizing mob of participants who would show up just because they could was not realistic. Everyone is busy and in order to create the perception of value, it’s important for the organizers to clearly state what the participants can expect to occur and more importantly, what value they can expect to derive from spending their time at an event. I don’t think the program needs to be complicated (a few discrete events on one day might be best for the next go around), but it does need to be clear. Participants who are new to the environment don’t want to take on the cognitive load of defining the in-world activities in addition to figure out how to navigate, etc.

3. Consider providing training wheel sessions: This might be in the form of a pre-conference session for participants and presenters where someone is available in-world to help newbies figure out how to chat, navigate, etc. At the very least it is probably necessary to have a dedicated greeter at the teleport landing pad during official conference hours to orient new arrivals and answer basic program questions.

4. Don’t run it in parallel with RL events: People go to real world conferences because they want face to face interactions which remain the highest bandwidth communication available. Virtual conferences won’t change that (duh!) but they might fill a different niche (e.g. bringing together small groups where travel costs or logistics are a barrier). Running in parallel with a real world events asks people to choose between communicating at the bar with a colleague they haven’t seen since last year vs. logging in to a computer system to communicate virtually (something they all day every day at home). There may be overlap in the long run between real world and SL events (e.g. projecting an in-world POV to real world participants to encourage “hallway conversations” between worlds) but for now it’s probably better to keep it simple and encourage people to development the in-world conference mindset.

Next Steps

As you might have surmised by now, I felt our initial experiment with SL was very successful as a learning experience despite the small number of participants. I also believe that integration of systems like Sakai/Moodle/Angel/Blackboard with virtual worlds is going to be a requirement for many institutions soon so it would be a smart competitive move to start getting acquainted now. Once the academic year is underway at U-M, and my personal life settles a bit more in a few months, I intend to start the conversation again amongst the Sakai community with an eye toward applying the lessons learned to the first full-fledged virtual Sakai conference. I’m thinking of a one-day event with some preconference training time, a relevant keynote, maybe a dozen presentations in a couple different gathering spots, and some dedicated timeslots for BOFs and/or general exploration and sandboxing. Of course this effort doesn’t have to wait on my availability or be limited by my vision – I’d be more than happy to have someone else pick up the ball and run with it…

Acknowledgements

Of behalf of the Sakai community I’d like to thank Ted Hans, Patricia Anderson, and Marc Stephens, the U-M folks who allowed us to throw a party in their house, and even did most of the setup, on very little notice. A nod also goes out to Harriet Truscott who braved the unknown to become the first and best Sakai presenter in the metaverse (that I know of anyway). Cheers.