3pointD.com

Sir Bruce Sterling Woodock has new population figures up on his excellent MMOGChart site, which tracks the population of massively multiplayer online games as well as a few virtual worlds like Second Life. Sir Bruce’s new data only covers up to about January of this year, but it’s worth slappping an eye on his charts, as his site is about the only one tracking numbers like these (plus it’s pretty). That said, there’s not much that’s terribly notable to report this time around (a new release is coming soon, according to the site). Asheron’s Call and Ultima Online continue to bleed customers, RuneScape continues to climb and, interestingly, A Tale in the Desert and World War II Online, though both small, seem to have leveled off without dropping customers for some time. Sir Bruce still has World of Warcraft at 4.5 million, though it’s since passed 6 million, I believe. Then there’s the issue of Second Life.

Sir Bruce puts Second Life’s “active subscribers” at about 45,000 as of last summer. This might be more or less accurate, but the numbers have definitely moved on since then. According to a recent release of some (fairly slim) economic data by Linden Lab, March saw 60,900 different accounts log into Second Life, of a total membership approaching 190,000. That unique log-in figure almost certainly counts multiple accounts owned by the same user as multiple accounts, so the number of truly active users is probably somewhat smaller.

The broader question is of how to calculate the “population” of a virtual world. This has never been easy, mostly due to the fact that game and VW companies are unwilling to release the kind of data that would be needed to produce accurate figures. But even if such numbers could be gotten, several issues remain.

In the real world, population is more or less the same as what we call concurrency in online spaces: the number of people present at any given time. But average concurrency rates are always smaller than “active” subscribers (you can’t be logged in 24 hours a day, after all, while you are present all the time in an offline space). EVE Online (which I wish I had time to play these days), has about as many subscribers as Second Life, but sees concurrency rates of between 10,000 and 25,000, while Second Life’s hover down around 5,000.

Unique log-ins over the last 30 days is probably a better measure (though it should ideally filter out multiple accounts held by the same user). But this number fails to track usage. If half those log-ins were one-shots, lasting no more than 5 minutes, should they be counted toward population?

It may be that a different kind of metric is needed, one that is based on usage and which better describes how people inhabit their virtual worlds. (Again, this presumes that companies would be willing to release such information, which doesn’t look likely. But we can dream.) It might be that some combination of weighted figures would best describe what’s going on, including, say, unique log-ins (discounting multiple accounts), average time spent per log-in, and average number of log-ins per account per month. In any case, it doesn’t look like we’ll get this kind of information anytime soon. But the problem itself is interesting to think about. What kind of census do we need?