Andrew Ferrier’s Blog

I went to a summit on Second Life and virtual worlds in Hursley today, hosted by Kevin Aires, Jack Mason, and Roo Reynolds - it’s becoming obvious that there’s a big buzz about Second Life both inside and outside IBM - a primary bit of evidence being IBM’s recent announcement of a $10m investment in virtual worlds such as Second Life. For obvious reasons, I can’t relate everything that was discussed. However, the discussion did get me thinking about Second Life in a slightly different way from before: as a basis for social networking applications, rather than as a basis for any application: an example of the former being The Greater IBM Connection, an IBM alumni community. In this respect, it seems to be more analogous to, say, LinkedIn, than another application platform (Windows, the web, etc.). This scope seems realistic - as I’ve stated before, meetings and conferences seem to be the best application of Second Life that’s been demonstrated so far. The consumer impact of Second Life is still TBD.

Second Life is an idea I want to like. It’s not a game, and it’s not just for playing around either (despite the slightly frivolous avatars and other trivialities imported from actual games). Some of my colleagues from IBM in various R&D labs around the world - such as the Emerging Technologies Lab here in Hursley - have been doing an admirable job of promoting Second Life as a genuine business tool (articles on Slashdot, the BBC), and I think it’s great that IBM is looking at using something so bright and fresh.

Nevertheless, Second Life is in a dangerous period. The early adopters are familiar with it now, and there some important challenges to be overcome if it is to break into the mainstream:

• It needs to shake off the geeky image, as Google, Wikipedia, and others have done before. This is essential for mass sign-up. There is a huge group who will equate it with MMORPGs such as the famous World of Warcraft, and an even larger one who simply aren’t aware of its existence.
• It needs to become more robust. The demands Second Life makes on one’s machine are significant, which currently denies a huge potential market, and the client and world in general are not particularly stable - frequent updates and glitches will not be tolerated by most folks. The only way this seems likely to happen in practice is to slow the world’s technology changes for a while so they are further behind the hardware development curve.
• Second Life is a lot like the house that Wikipedia built; great from 30,000 feet, but rough round the edges when you zoom in. Left-overs from experiments are everywhere, and the world itself lacks coherency. I’m not convinced this is necessarily harmful by itself, but it is a problem that needs addressing to avoid frustrating impatient later adopters. Wikipedia has done a good job of starting to fix this with voluntary task forces attacking accuracy problems and polishing the edges. Second Life probably needs the same.
• Most of all, Second Life needs to decide what it is for. This isn’t to say that Linden Lab needs to do this (although they’d do well to encourage it if they want to stay in business), but somebody sure does. I’ve seen Second Life used for meetings, lectures, classes and similar functions, but other use cases seem to have so far evaded it. I’m sure there must be some more innovative stuff to come; Second Life is, after all, a brand new interface to manipulate data with. Yes, OK, it’s not the first 3D interface by a long shot, but it is the first with such a vast user base and such a high degree of customisability.

Many folks propose that Second Life is merely a ‘taster’ - the Yahoo of virtual worlds - and that the Google is still to arrive. Others foresee a gloomier future. I hope that isn’t the case - I want to like virtual universes, I really do. But like most people, I want to use it for something constructive - a half-way house between a communication tool and a game is no fun. Please, Second Lifers, invent a value proposition.

Update 2006-12-13: Clay Shirky’s analysis of Second Life is spot on.

James has just given me a brief tour of Second Life. Once again I had problems - cracking graphics and a crashing client - but I was a bit more impressed than last time - having a guide to show me the good stuff and get the hang of the controls helped. I also met one or two of the other IBMers using Second Life. It’s clear that there’s still a lot more to explore. I’ve decided that I should attend one of the virtual IBM events - this seems to be one of the best uses of Second Life I’ve heard of so far.

I’ve finally given in and taken my first tentative steps into Second Life (it’s legit once it has an Economist article; or something…). My first impressions are so-so; it’s obvious that there’s a lot to explore and do that I’ve only just scratched the surface of (see Andy Piper’s blog for a lot more posts on Second Life; including how IBM are getting involved). But the system obviously needs some work - the client crashed twice within the space of two hours, once bringing down Windows. The movement is pretty jerky, despite turning the detail down to minimum, and the graphics flickery, almost unusably so. I’ve already been forced to upgrade the client once (~25MB download). It’s possible my T43 just isn’t up to the job (dodgy graphics drivers?), but I don’t have anything more powerful or different to try it on.

What I’m more interested in, though, is what’s available in this virtual world and how easy it is for me to create and trade. The Economist article above touches on some of the theoretical economic differences between Second Life and ‘First Life’, but it’s obviously going to have to grow a lot bigger before the implications become obvious. I’m still doubtful, particularly given the technical problems I’ve seen above, that this particular technology can cross the chasm, but it would become an interesting lab if I were proved wrong. I suspect there’s a future for something like Second Life, even if Second Life itself is not it.

I shall continue to explore.

Update 2006-10-15: Link to Andy’s blog fixed.