Behind the lectern at Second Life

Last week, I attended a talk about the future of computer software... where the presenter did it from Second Life.

You’ve heard of Second Life, of course, yes? You join, you create an avatar, and you wander around — your avatar wanders around — visiting places and encountering and interacting with other people’s avatars. You can talk with people, including groups of people, and, so, you can have meetings, or “hang out” with your friends. You can go to stores and buy things, and real companies (including mine) have stores, offices, or other sorts of presence in the Second Life system. It’s a world.

A virtual world. And, to paraphrase what someone once said about virtual computers, a virtual world is just like a real world, only it’s virtual.

So the presenter gave his talk from home, not having had to travel to our meeting, and everyone at the meeting watched his avatar speak, move around, and gesture. What’s more, others who couldn’t travel to the meeting, or didn’t want to, were also able to “attend”, which was not the case with the live talks.

And how did it work out?

Not well, from where I sat, and the folks I talked with agreed. Some of the problems were just artifacts of how this particular presentation was done. The slides, for example, weren’t on Second Life, so we had to page through the slides manually, while the speaker told us when to go to the next slide. That certainly could have been done differently. And the avatar’s movements were strange and distracting, most likely because they were set up randomly. The avatar sometimes stood arms akimbo at odd times, and it shook its head from side to side at inappropriate times.

Clearly, there are a number of options for controlling the avatar, ranging from the random-ish motions that we saw, to having an array of cameras on the live speaker that allow his real motions to be reflected in the avatar. Or the speaker could simply have manually triggered certain gestures and movements, selecting appropriate ones for what he was saying. (It strikes me that doing that would cause some odd cognitive dissonance for the speaker, but one could probably get used to it and learn to operate it that way.)

But in the end, was it worth it?

That question was also the first one asked in the Q&A segment. What, someone asked, was the advantage of giving the talk from Second Life? The speaker pointed out that there were half a dozen other avatars in the “room” with him, and that he’d be hanging out and chatting with them after the talk. He couldn’t do that with just an audio or video feed. And he was reducing his carbon footprint by not travelling to the meeting. In comparison with two-way systems like Cisco TelePresence, he pointed out that the requirements for bandwidth and other infrastructure are much lower with Second Life.

I’m still not convinced. First, we’re not really comparing it with travel; we’re comparing it with other non-travel mechanisms. Presenting from Second Life is competing, here, with presentation by audio-only feed or by audio-video feed. So, while it was important to point out that he didn’t have to travel, that’s not really relevant to the comparison.

Second, it’s not true that he couldn’t have the after-talk interaction, the hang-out period, without Second Life. He had an audio feed from the conference center (that’s how questions were asked), and he could have just chatted that way afterward as well, audio to audio. I didn’t think the avatar added anything to that, and, in fact, it was worse than having only audio. Others I talked with agreed with that, and it seems clear that it was a distraction: the questions were about the form of the talk, not about the content.

We could also have used instant messaging as a communication channel during and after the talk, something that’s often done on teleconferences I participate in.

The real winner would be a video feed, where we could hear and see the real speaker, with his real motions, gestures, and facial expressions. Even low-quality video helps a great deal, and, again, it’s something I’ve used many times in meetings with remote participants. One can also envision doing some work on novel visualization mechanisms, wherein many video feeds from many participants could be seen together, highlighting who’s talking and allowing everyone to see a conversation among two or three of the participants. The network resources needed for that can be kept down by limiting the size and resolution of the video, while still making for a better “remote presentation experience.”

So it seems that doing it on Second Life is more a gimmick than anything else, done because we can. Maybe I’d think of it differently if the avatar seemed more like the real speaker than it did. But how many people will equip their home offices with the gear needed to do that properly? Meanwhile, my Macbook Pro has a video camera built right into the laptop’s lid, ready for use in an instant.

So, yes, it was interesting to check it out, but I’m not looking forward to the next presentation like that. I’d rather go for moderate-bandwidth video.

And if that doesn’t work....