A new Look at the Numbers
Linden Lab put out the Q4 stats for 2010 and several have blogged about them. I suppose we probably knew whatever it is they tell us. The numbers just help us get sense of degree. We aren’t deciding if things or good or bad, better or worse, but how much better or worse. So, what do the numbers mean and what will we see in 2011?
Where the Numbers Are
First, the Linden Numbers are here: The Second Life Economy in Q4 2010 posted by Nelson Linden. I have to wonder if choosing the name ‘Nelson’ was a play on the Nielsen Ratings. One can find the quarterly reports from the Lab in the SL blog: Tags: quarterly_economic_report. The Linden$ exchange data: LindeX™ Exchange: Market Data.
Tateru Nino’s blog Dwell On It provides Second Life Statistical Charts in an interesting perspective. Just as politicians try to spin things to look good, so does the Lab. Tateru shows what information the Lab is publishing as well as the types of information they have stopped publishing. That gives one an idea of how the Lab is spinning.
Obijan Technologies provides some statistics too. See: Second Life Statistics
Another presentation of Second Life Statistics is: Second Life Grid Survey
Second Life statistics are presented from the view point of a competitor along with the stats from OpenSim grids at Hypergrid Business. Latest report: OpenSim grids reach new peak (Not really, but it looks that way, more later.)
What the Numbers Mean
The raw numbers are objective and are provided by Linden Lab, no matter who reports them or generates the graphs. The meaning of the numbers is mostly debatable. So, one can say someone is 6’ in height (1.83m). That is just a statement of fact. Whether this means the person is tall depends on many things, gender and nationality being two typical considerations. Time is another. In America 6’ used to be tall and now is pretty much average for a male. So, to one must consider the user of stats to keep them in some perspective.
So Where Are We?
Second Life continues to have a high rate of signups. The data shows it continues to bounce around between 10,000 and 14,000 per day. The Lab’s problem is poor user retention. This is shown by the number of concurrent users and unique logins per day remaining steady in the face of 10K+ new signups per day. Depending on the time of day, concurrent logins range from 30k+ to 70k+.
There is some variation over time and use of SL is seasonal. Peak usage is typically in February. August sees minimum use. So, it appears related to weather for American and European users. Still the average concurrent logins seems to be slowly decreasing. Below is unadjusted raw data for concurrent users. The adjusted data with trends revealed is here: Second Life Grid: Economic Metrics
What this tells us is open to interruption and debate. But, I think it is clear the SL Viewer 2.0 has not solved the problem of poor retention of new users. The introduction to Second Life has changed for new users too, and that is not making a visible impact on the stats. I suppose it is safe to say we have not reached Fun and Easy stage yet.
I also do not see Third Party Viewers (TPV) helping with new user retention.
There is also the Second Life real estate, goods market, and money market at which we can look. As Hypergrid Business points out, SL lost 132 sims [I think they mean regions]. One has to look closely to see how Hypergrid Business came by that loss. The number is accurate, but SL increased the number of regions online for the 2010 year. However, growth was flat for Q3 to Q4.
On the competitive note, the OpenSim regions count is strongly influenced by OSGrid. Hypergrid shows OSGrid as having 9,000+ regions online. I know that is inaccurate. Last week it was 4k+ and is currently 5k+. The bounce in numbers is because of how OSGrid culls dead regions that have dropped of the OSGrid. Two weeks ago the garbage collector was repaired and started running. Now regions that are offline for more than 2 or 3 days are removed from the list. That initially resulted in about 5,000 regions being removed. With total OpenSim regions reported as 15,623 a loss of 5k is about a 30% loss. I think all this means is that OpenSim region counting is flaky.
The Second Life virtual goods and money markets are growing. I know I am seeing an improvement in sales. The numbers show sales almost doubling since Q4 2009. Yet, we hear massive complaints from merchants saying the new Market Place is not working and sales are down. One can figure out how to resolve those two stories.
The money supply and trading volume (or money velocity) are slightly up over 2009. Through most of 2010 the numbers have been decreasing. In Q4 they almost recovered.
The value of Linden money is shown as increasing, going from 261.7/US$1 in Q4 2009 to 258.8/US$1 in Q4 2010. Basically of you are buying Lindens with US dollars the Lindens are more expensive, or spun the other way the US dollar got cheaper. With the US Treasury monetizing US debit by creating dollars (called printing money but increasing the money supply does not require printing – after all, when did you last see a paper Linden$) we can expect that trend to continue. However, the Lab manages the currency just as China controls the Yuan, so it is impossible to know what will happen.
What Does It All Mean?
Meaning of all these numbers to the residents of Second Life varies. There are some absolutes. As to growth of the SL world, it is so small most of us can’t see it in our daily SL life. The economy seems to be improving, or at least has for Q4 2010. Some of us are seeing that in a better SL income some of us aren’t.
Beyond that, things are pretty much stagnant or stable, depending on your mood.
What Is To Come?
If one tries to project the future from stats, it is very much like trying to predict the traffic in front of your car by looking out the back window. One can make general predictions but they are unlikely to predict a stalled car in your lane. One has to look to other factors to decide which lane they want to be driving in. The same is true of Second Life stats. They just aren’t good predictors.
There are, however, a number of factors that suggest what may happen.
IBM and Microsoft have been and are doing virtual world development. 3D movies and television are moving to the mainstream. Numbers of people, teams, and businesses are working on OpenSim products. There is significant interest in 3D virtual worlds and moving them forward for business and entertainment use.
Unity 3D and similar companies are developing browser based cloud served virtual world platforms. These will allow casual use of virtual worlds without downloads and installs. It will also allow use of CPU intense virtual worlds on light-weight clients, think smart phones.
Everywhere one looks 3D is a major interest. Second Life is positioned to take advantage of the 3D popularity. Whether or not the Lab can figure how to make it the popular choice is another matter. But, consider the following.
Each day Second Life is changing. The servers and viewers are getting faster and providing a better environment. Region crossings are faster. Image decompression is faster. There are fewer server crashes. Fewer user sessions end with a viewer crash.
The new SL Viewer 2.5 has hundreds of fixes and is getting more and more user suggested changes to the user interface. The SL forum is seeing some old timers talking about changing from the series 1 viewers to the series 2 viewers. The backlog of needed bug fixes is significantly reduced and some man power is transferring away from bugs to providing new features.
Mesh objects are coming.
As bad as search is right now, it is being fixed. Also it is being made Internet search friendly. So, many of the results you see in world will start to be visible from outside SL.
Display Names, the new Profiles, and the new market place are steps toward better social networking and better Internet presence. Effort is going into a browser based viewer, which is a step toward social networking and providing availability of SL to smart phone and other light weight clients.
Mesh is going to bring some professional modelers to SL. From office hours meetings we know some businesses have things planned that depend on mesh. New mesh builds will attract more people.
While I suspect most people are unaware of the Ray Cast feature being developed, it has the potential to revolutionize combat games in SL by eliminating bullets and their associated lag.
Social networking is going to expose SL to more people. We are likely to see that social networking exposure drive up the signup numbers.
Whether we will see a significant upturn in SL concurrent users and user retention is likely dependent on whether the SL Viewer can be made more new user friendly. Also, can other things like buying, opening packages, and wearing clothes be made easy and fun. The inventory outfits are making changing clothes easier.
I doubt the viewer will change much in regard to fun and easy for several months. It will likely be viewer 3.0 that has the user interface and render engine completely separated thus allowing a user to customize the viewer to their needs. Also, it will likely be viewer 3.0 that has the ‘plugin’ feature that allow people to move from building complete viewers to building just specific features and viewer modifications.
Also, Europe is turning around their economies and reducing their spending. America is in political turmoil as spending and socialistic entitlements are debated. If America controls their spending and holds the line refusing tax rate increases we may see the American economy start to improve.
So, it appears Second Life is in a good position to see an upturn in users. I am expecting considerable improvement by Q4 2011.
Great write up 🙂
Nelson’s name is actually Nelson irl. I had the chance to meet with him in San Fran the bought me fish tacos). He is director of monitization for the lab and used to be the Chief Learning Officer for Ninth House
As to the future of SL? Well if the new CEO has the coooperation of the Lab then maybe these numbers can improve