Advertisement

Linden Lab publishes Second Life second quarter metrics


Linden Lab's Tom Hale has published the Quarterly report for Second Life for Q2 2009: "Economy grows 94% year to year, hits new all time high in Q2; World expands and voice usage hits all time high; Bot policy affects user hours and logins"

We beg to differ in some key respects. From the figures presented, we see quite a different picture (go on, tell us you're astonished by that). Allow us to touch on a few of the high points as we walk through the report, and we'll lay it out for you.

After considerable (and very effective) effort on the Lab's part debunking and undermining user-to-user transactions as an indicative economic metric when the figure was in decline, we see the same figure touted as an indicative economic metric when it is on the rise.

We're sure that you can see our basic dilemma here. Is it a relevant indicator or isn't it? Has the method of measurement changed in such a way that the Lab's very sound reasons for touting the metric as irrelevant no longer apply?

We'll let user-to-user transactions pass without further comment. We do not see any data in the quarterly figures that backs up a position of either economic growth or of economic decline.

Xstreet sales have reached an all-time high (L$372 million). This isn't surprising as Xstreet sales growth was continuous from launch, prior to its acquisition by Linden Lab. Admittedly, this growth took a bit of a jump in Q1 2009, when the Lab began advertising Xstreet widely in the MOTD, blog and in email mailouts.

User hours are up 32.6% from Q2 2008 – but Q2 2008 was up 55.7% over Q2 2007. There's growth, certainly, but the 2009 growth is smaller in both percentage terms and in absolute numbers than the 2008 growth. Total user hours for Q2 2009 totaled 126 million.

Voice-minutes used are up to 3.14 billion for the quarter, although the definition that the Lab gives counts listeners as well as speakers. Previously, the Lab assured us that only speakers were counted. We now believe that to have been a miscommunication, however, as the historical figures remain consistent with data we've previously seen.

Due to the way the voice component of the viewer handles connections, it appears that voice minutes are almost continually accrued whether voice communications are taking place or not, so long as it is enabled in the viewer, and the user is on any voice-enabled parcel. It might be an teleconferencing industry-standard metric, but it's a metric that generates nonsensical comparisons for the mechanics of Second Life voice. [Let's try that again: Voice minutes are accrued by speakers, that is, people with an open microphone. Thanks to Tom Hale for clearing that up for us. That's closer to what we originally understood to be the case.]

Growth in voice from Q1 to Q2 was minimal in percentage terms, however, which given the above suggests a decline in the adoption rate of voice for new users. It's up 48% from the Q2 2008, granted, but correlating voice-minutes with user-hours over the same period, we see a significant per-capita decline in voice usage, and that's even taking into account the reduction in user-hours from the new policies on bots, so the actual decline in per-capita voice-usage may be even larger than the available data indicates.

Peak concurrency looked due to tip over 90K in Q1, but the growth there suddenly stalled, and has remained relatively flat during Q2. By our own figures, median concurrency has remained quite flat since the end of 2008. Concurrency peaks are a nice sort of statistic, but the concurrency medians are where the real trends emerge. We're not seeing significant growth in median user-concurrency for 2009, either in Q1 or Q2.

As for land area, the repackaging of void simulators caused a noticeable fall in private estate land area in November 2008, a trend which did not begin to reverse until April 2009. May and June have shown growth, but realistically, private estates have been set back to levels of approximately 12-13 months ago.

We agree with Linden Lab that Q2 2009 was absolutely a solid quarter, and not at all shabby by any means. What we're not seeing in the provided figures, however, is real growth, either in the economy, or indeed in any sector. That isn't a bad thing. Second Life isn't a shark or a public company, both of which must continually move forward in order to avoid drowning.

Linden Lab remains profitable to the best of our knowledge, and the lack of figures showing significant growth isn't really a strike against it. Profit's profit, and Second Life remains healthy.


Are you a part of the most widely-known collaborative virtual environment or keeping a close eye on it? Massively's Second Life coverage keeps you in the loop.