LL should report number of 'Client Downloads' not new accounts

Some recent discussions have made me think about the reasons for the removal of CC verification. I believe that LL have two distinct ambitions (within the overall business objective of staying in business). One is to generate a vibrant online 3D world used and contributed to by as many people as possible. The other is to have SL as the most widely adopted 3D client, a kind of RealPlayer3D. The pointers to this are the championing of the relatively small client size and the moves to reduce sign up friction such as not verifying email addresses.
The tensions have arisen because these two ambitions are in conflict. LL want to show the world how successful their product is, to go past the tipping point where SL becomes the defacto standard way of collaborating in 3D or viewing 3D spaces. So why not create the concept of anonymous visitors that don't count toward the account numbers, and have limited scope, geographically and physically, but report the number of client downloads to the media just like Macromedia might have done in the early days of flash?

They should report active users for last 30/60 days.

There is a serious flaw in LL's methodology for counting subscribers, but they're not alone.

Measuring MMOs

Brief summary of Koster's analysis: The Lindens are not concerned with actual numbers of subscribers. They want artificially inflated numbers of avatars, paying or not, for appearances sake and to be seen as Major Players in the MMO world.

The numbers were semi-bogus before the CC verification change. Now they're completely invalid, but it'll look great on someone's color Powerpoint graph at an investors' meeting.

The problem from LLs perspective is that any of the much more accurate suggestions for measuring subscriber numbers will be significantly lower. I am trying to suggest a figure that would actually be quite high whilst allowing the reinstatement of CC verification, and/or a more detailed signup for someone to become a fully fledged member of the SL world.

Might as well report the number of fries McDonalds sells as "subscribers"; it would be as meaningful and look more impressive.

As far as I am concerned, there are only a few metrics which give MMOs (or any business, for that matter) any real credence.

1) Number of PAYING subscribers.
2) Number of PLAYING subscribers, measured in logged-in hours.
3) Avergage age of the above (not RL age, but how old and how active the accounts are).

That boils down to a handful of what I consder are the most important metrics:

1) How much money are you making?
2) How many people are actively participating in your world.
3) How well are you retaining people for the long-term.

Any other metrics, like numbers of accounts created, client downloads, etc, are totally and utterly meaningless because they are full of correlation errors.

But my main point is that I don't think that LL want SL to be just an MMO, so those metrics alone won't tell the whole story. Of course downloads is a duff metric, but it might satisfy LLs need to keep the media tongues wagging.

Client downloads would be over inflated numbers too. I have 3 pc's, and its as quick to redownload it from the site to each as it is to copy it across the network. And I can't be the only person who does this

Absolutely. So even better for generating buzz Again, I'm not really interested in getting a 'valid' metric in any scientific sense, only to think of ways to get the numbers up whilst allowing the CC verification for subscription to be reintroduced.

Free sex? Maybe LL could employ Escort Lindens to offer free rides at the WA?

I think it is actually a pretty good idea, Richard. I'm not sure of these number would be more or less accurate, but they would be none the less interresting.

https://secondlife.com/currency/economy.php



Of course that doesn't help you with how many of them are alts but I don't think even LL has an accurate idea of that, especially after the new registration system.

Whatever in the world makes you say that? Their "1 million subscriber" nonsense is a direct imitation of Will Wright (The Sims Online) who said exactly the same thing about TSO. It's a copycat goal patterned after Asian games like Lineage and FF or Blizzard's World of Warcraft.

Interestingly, TSO peaked at about 105,000 subscribers but has really fallen off the cliff. The last available graph at MMOGchart.com (late 2005) shows them hovering at 35,000-ish while SecondLife is going in the opposite direction.

It's big business. If LL wants this to be a profitable enterprise, they're going to have to go beyond the "open development platform" approach and make SL appeal to the masses. I understand that. What I don't understand is their tendency to make snap decisions without first putting tools into place to handle the consequences. It's a great way to NOT get where they want to go.

bad Idea I usual uninstall and redownload after each update, otherwise I get bugs

It's similar to tryign to count webpage hits.

LL could count unique IP addresses for a given period, but how many people play from multiple networks (not many I'ds surmise, but would still skew the numbers) Or households where both people play but are on the same connection.

Even counting paid accounts doesn't account for people who may have paid for more than one account.

there really is no good way to judge.

The only way to accuratly count would be if every other mmo used excactly the same methadology.

I don't know about anybody else, but I download SL onto 3 or 4 machines everytime an update comes out, on nearly as many networks. Sometimes more, I might be at a friend or family members place and hook up to their network or download onto their computer. I've got my laptop, desktop and work PC for starters, and I've got my home network, my work network, and the network at the university cafe. And on my laptop I download both the Windows and Linux versions. I'm sure I'm far from the only one with similar habits. It'd be pretty hard for LL to get any meaningful metric out of IPs or downloads.