I like to bet people a quarter. I pick a quarter because when someone wins or loses, it's completely about the principle rather than the money. If you bet someone $100 and you win, you're focused on the money. If you bet .25c, it becomes totally about pride.
So I just bet Cory a quarter over the following proposition: Which will have more North American users two years from now: Second Life or WoW?
Not surprisingly, Cory is going with SL and I'm taking the other side.
The bet, of course, becomes a test of both SL's growth projections (quite robust of late) and WoW's ability to retain or grow its subscriber base. It could also therefore be seen as a test of the long-term growth potential of user-created content versus the more standard entertainment industry model.
What say you? Help us make our own futures market here.
Dmitri, I'd recommend that you and Cory settle on a definition of "user" as soon as possible, lest he swindle you out of your 25c. ;)
The various definitions of VW "users" has been discussed here on TN recently, though I don't believe we came to any consensus. In that thread, Mike quoted Cory as saying that SL’s “concurrency numbers are rapidly approaching 4500, about 17,000 residents were in SL in the last 24 hours, and 50,000 in the last 30 days… If you go back even 90 days you get about 90% of the accounts having logged in.” The latest numbers released from Blizzard state that WoW surpassed six million players worldwide last month, and had over one million paying customers in North America last August. (For anyone interested, Blizzard gives their definition of "customer" at the bottom of each of their press releases, be sure to check it out.)
So what's a good definition of "user" when dealing with two VWs that are so vastly different? Obviously subscribers won't work, and I don't think monthly or yearly income is a useful measurement in this case. I think registered users unfairly inflates SL's number, and would be a difficult number to come by for WoW. So what does that leave us? Peak concurrency? Average concurrency? Unique users logged in during the past 1/7/30 day(s)? Some leverage of ARPUP (Average Revenue Per User Per Month) to convert SL's numbers to something subscriber-like, and then compare that with WoW's subscribers?
Whatever measurement makes the most sense to the two of you, it might be useful to decide upon that now, rather than two years from now. Our inability to settle on one number as an industry is a frustration to me, so I'll be interested to hear what decision you come to.
As for my own personal take on the bet -- I generally think it's a bad idea to bet against Blizzard. ;)
Posted by: Samantha LeCraft | Mar 23, 2006 at 03:53
Might be good also to define WoW and SL. What if Blizzard bring out WoW2 or some variant, does that count? And what if SL go to the model of user operated servers that has been mooted, does that count as SL or not?
Posted by: ren reynolds | Mar 23, 2006 at 04:59
If Linden is releasing concurrency numbers, that's the best measurement. That might give wow a small advantage, but WOW is not heavily two boxed or botted.
Posted by: Kathy | Mar 23, 2006 at 09:09
I'd be tempted to put money on SL. I think WoW's a hot virus -- it'll burn fast and two years from now WoW may be a ghost town (*may* be). SL is more of a slow cooker, so it might just be hitting its stride in '08.
Posted by: Chip Hinshaw | Mar 23, 2006 at 10:05
A definition of 'user' is definitely in order, not mention business models. Wow has about 6 million paying subscribers worldwide; how does that equate with SL's 'free to try' model?
Posted by: Jessica Mulligan | Mar 23, 2006 at 10:26
Oh, and no offense to SL (I really like the Linden Lab crew and am intrigued by what they are doing), but WoW will definitely have more users, no matter how you describe them.
Why? Simply: because game MMOs serve a need for entertainment that is demonstrated to have at least 60 million customers worldwide, while non-game virtual worlds such as SL are intended to appeal to the one group that is most resistant to downloading a large client and needing to learn how to use it: the casual user.
That is not say that SL won't be successful in it's own right or that in some future time or that we won't find a means and method of getting enough casual users interested to start an explosion; it just won't happen in the next two years.
Posted by: Jessica Mulligan | Mar 23, 2006 at 10:34
I'd be willing to put .25c down on Second Life being unable to achieve 1/10th of WoW's current peak concurrency rates in the next 2 years.
This would roughly be 150k -- I can't seem to locate a more current number (500k peak concurrency for WoW) than when they were at 1.5 million total subscribers, but extending this proportion conservatively to their ~5 million total subscribers lends itself to suggesting 1.5M peak concurrency...
Posted by: Ralph | Mar 23, 2006 at 10:52
What's the Second Life appeal for people who have a real life full of friends?
Maybe I'm a bit "extreme", but I do think that people who spend times in this kind of virtual world need to consult.
For me, virtual world whould be place where you relax/forget about everything that is stressful in real life. Second Life needs commitment/time that I doubt the "mass" will ever have.
World Of WarCraft don't need dedication or much time to be enjoyable and fun.
Posted by: LEKO | Mar 23, 2006 at 12:22
The spirit of the bet was total registered users for SL anywhere having logged in in the last several months (shall we say 4?) vs. current paying players for WoW in North America. The numbers would most likely come from LL's reporting and Blizzard's press releases, unless there is a better source by then.
I think those terms are generous for SL but I'm OK to dangle the quarter out there even so.
Posted by: Dmitri Williams | Mar 23, 2006 at 12:33
To heck with principle; Cory, I'll put $250,000 on Dmitrii's side.
:)
Posted by: Andrew Zaffron | Mar 23, 2006 at 12:50
I've definitely got my quarter on top of Cory's.
I think most of us still don't grok Second Life's position because we're comparing it to worlds that have tried similar things under very different Web conditions. I can see the development of the Web pulling SL right along with it, connecting up with other programs, markets, social networks, etc. And remember it's not a role-playing game and doesn't require massive time investment. You can hop in when something interesting happens, meet with your real life friends as your real life self if you'd like, and hop back out to your blog or Word doc, all with a click of a SLurl. No literal "second life" required. I'd suggest you really can't think about SL's prospects for growth by looking at the VW industry at all. It's a different animal, like a new technology being adopted rather than a particular experience being had and exhausted.
And consider that there are existing companies with gigantic user bases and an eye to build in VWs who could drive millions of users within *days*. If SL or other platform worlds could handle that load right now, we may well have seen it already. WoW is awesome, but trapped in a narrative, and as a user or third-party developer you can't build anything in it. SL's content covers the whole spectrum, from charity fundraisers to pr0n to virtual office space and "3D websites" to appearances by real life celebrities to whatever, and it's pure positive feedback with more people meaning more builds, more builds meaning more people and so on. I'd stand Jessica's comment on its head and say (in a very friendly and respectful way!) there's no way WoW will have more users than SL.
As a random example of what can happen in (less than) two years, Google bought Keyhole in late 2004 and launched Google Maps in February of 2005. Google Earth and open mapping APIs weren't even on the...um...map until then :). Now it's off to the location races and it feels like we've had Google Maps fooorrreeevvveeerrr.
And so my quarter's on the table! This'll be fun to watch develop.
Along these lines I've added the "Active Worlds --> Second Life" entry to the bottom of Tim O'Reilly's What is Web 2.0 list copied below:
DoubleClick --> Google AdSense
Ofoto --> Flickr
Akamai --> BitTorrent
mp3.com --> Napster
Britannica Online --> Wikipedia
personal websites --> blogging
evite --> upcoming.org and EVDB
domain name speculation --> search engine optimization
page views --> cost per click
screen scraping --> web services
publishing --> participation
content management systems --> wikis
directories (taxonomy) --> tagging ("folksonomy")
stickiness --> syndication
Active Worlds --> Second Life
Posted by: Jerry Paffendorf | Mar 23, 2006 at 20:14
Leko> What's the Second Life appeal for people who have a real life full of friends? <
I’d guess Second Life appeals to people who don’t see it as some separate plane of existence from “Real Life”, and their friends there as somehow “unreal”. Would you consider the friends listed in your phone book as real, and the friends on your MSN list as not? Perhaps it’s a generational thing, but if MySpace can quite rapidly become just another place to hang out with your friends for a section of the population, I don’t see why SL or something like it couldn’t follow. Two years though? Hmm. I’ll risk a quarter on Cory’s side anyway.
Posted by: Hellinar | Mar 23, 2006 at 21:38
Leko> For me, virtual world whould be place where you relax/forget about everything that is stressful in real life. Second Life needs commitment/time that I doubt the "mass" will ever have.
World Of WarCraft don't need dedication or much time to be enjoyable and fun. <
Another thought. Gameplay in Second Life requires much less effort and commitment than WoW. Gameplay in WoW is D&D combat, and requires quite a complex understanding of your fantasy character’s abilities. To level up you have to successfully kill stuff, potentially quite a challenge.
Gameplay in Second Life is the familiar Consumer Game. You level up by shopping. Very simple. Very familiar. SL is still in the early adopter beta state, mostly full of people putting big time and effort into making stuff. I’d imagine a “bigger than WoW” state would be mostly full of people buying stuff. Compare the number of people who used to play D&D to relax and forget about the stresses of real life to the number of people who go to the Mall for the same purpose. “When the going gets tough, the tough go shopping”. Given long enough, I think the virtual shopping game will have the bigger numbers.
Posted by: Hellinar | Mar 23, 2006 at 21:56
This argument reminds me of the CompuServe v AOL arguments of 10 or 15 years ago: "Which of these private, monolithic systems is best?".
Give it 10 years, and put your 25c on the answer being "something else".
Richard
Posted by: Richard Bartle | Mar 24, 2006 at 03:47
I think Cory is being a bit too optimistic about what people want as a gaming experience. Unless the next immensely huge social phenominon happens in SL, it can't happen. And that can't happen until SL gets a universal portal... Login to it from cell, handheld game console, pda, home console, home pc, you name it. Only then could it command the numbers to compete with warcraft.
Posted by: geedeck | Mar 24, 2006 at 08:56
If due diligence reveals that SL is in talks with MS or Yahoo about becoming the front end for their messenging software, then I'm all over that bet like a rash: gimme some sweet quarters (cheap at the current exchange rate, anyway: I'll hedge with them).
Otherwise, if business models are relatively static: WoW it is.
Posted by: Endie | Mar 24, 2006 at 11:21
Chip Hinshaw> I think WoW's a hot virus -- it'll burn fast and two years from now WoW may be a ghost town (*may* be)
My suspicion, also... There are many WoW elements that seem to benefit short-term game, but unless they re-invest ALOT of their spectactular earnings to address these issues, these issues are going to come to the fore long before another two years go by.
But... I also don't see SL hitting the current numbers of WoW by then- I'm not entirely convinced of the scalability of the system that far...
So it's a debate of "will WoW's decline be drastic enough that SL's modest-to-ambitions incline will beat it, 2 years from now."
My $.25 is on Second Life.
Posted by: Chas York | Mar 24, 2006 at 11:42
Richard has an excellent point, but I'm with Andy Zaffron--but keeping it to 25c. (I'd be happy to lose that 25c!)
Posted by: greglas | Mar 24, 2006 at 12:29
You win. Second Life has surprised me, because I wrote it off as another Habitat/There, that is, a VW with no point (that is no goals, not a game) and therefore a dead end. They demonstrated that their creation tools allow users to produce enought stuff to create mini=environments that DO have a point. But to be an SL creator means you either need to be a Photoshop jockey or a programmer, so a limited subset of users will be creators--and it's awfully hard to explain to non-users why participating is indeed fun and interesting. I give Linden a lot of credit for making this work, but I really don't see it going mainstream.
Posted by: Greg | Mar 25, 2006 at 22:52
An interesting bet. Even if Second Life has to pass WoW's 1 million North American subscriber mark that is still quite a daunting task. I assume you're banking on Second Life repeating the webpage/blog mass audience appeal. Those media allowed a fair amount of personal customization and content. Also a fair number of resources and customization tools exist for them. If Second Life can appeal to the desire to express one's individuality, Second Life has fewer limitations than WOW in what it can do to generate interest. Still 2 years is not a lot of time. I think you'll lose, as WOW isn't exactly going to stay still. Are its current numbers the peak of what it can reach? Probably not.
Posted by: David | Mar 25, 2006 at 23:30
I tend to Agree with Dmitri on this one as well. Simply because WoW's business model suggests that they will continue to drive new (or returning) revenue, through highly advertised and hyped expansion packs and constant well timed updates (about every 60 days), that will inevitably make their customer base return. As far as concurrency and total user base, the lower these numbers are the better it is for Bliz anyways as the higher their paying customer count to active users ratio is (say 5 customers for every daily concurrent user) the cheaper their overhead becomes. What one wants to do, in my most humble opinion, is to drive a game towards those who will play the least, but provide a stable and constant user base.
I think a better bet/example would be more the average number of hours per captia played between the two games, if we are defining which is the better of the two games in any case. As a game with the financial backing of VU Games regardless of quality will outsell a smaller independently financed MMOG. However, if the driven market is identified and catered to correctly, the number of hours played should be in theory greater. Atleast given the age of the MMO being greater then 2 or so years, as the new-hype has worn off, and it comes down to directed marketing and quality of gameplay.
-J
Posted by: Jesse | Mar 28, 2006 at 17:10
I think it's interesting that we're at a point where people (other than Cory) are actually having a debate about this and putting money on both sides.
Richard: Give it 10 years, and put your 25c on the answer being "something else".
And that would have been the smart bet a few years ago too, but the "something else" most people would have been imagining at that point would probably have looked at lot like EQ or WoW or the imaginary "WoW beaters" that people spent a lot of time at AGC talking about.
If nothing else, SL's recent success gives me greater hope that the "something else" we imagine 10 years from today may not just be an incremental improvement over WoW.
Posted by: Jim Purbrick | Mar 28, 2006 at 17:15
Just to clarify, WoW hit 1.5 Million NA users several months ago, and probably approaching 2 Million by now. WoW's peak concurrency is higher than most VWs, anywhere from 20-35% of their subscriber base.
SL has maybe 20k or so real life, comparable users.
This bet is certainly comical, and if anyone would like to put some real money on the table, I'll wager World of Warcraft's total gross Revenue for 2008 vs Second Life's, any day of the week.
Posted by: Jeff S. | Mar 28, 2006 at 22:35
This reminds me of comparing the number of WoW-players (~2-3 Million in China using 'WoW Point Cards' for purchase) and the number of Yulgang-players(~19 Million in February'06) or players of MU in China. Comparing only the total subscription numbers, Yulgang or MU wins...
To compare the registrant numbers of a MMOG using a traditional subscription model with another one or, in the case of SL, with a 'Virtual World' using the "free-to-play and pay-for-virtual merchandise" business model for the most part, doesn't make sense.
Posted by: Phall | Mar 29, 2006 at 09:39
Phall> To compare the registrant numbers of a MMOG using a traditional subscription model with another one or, in the case of SL, with a 'Virtual World' using the "free-to-play and pay-for-virtual merchandise" business model for the most part, doesn't make sense.<
I think you are missing the point. The bet, as I understand it, is about which model will be introducing more people to VWs in a couple of years time. WoW or SL? Though the payment model is different, they are comparable in the sense of making VWs a common part of peoples lives. Its like comparing email and IM. Neither were part of regular peoples lives a decade ago. Now both fill the same niche of “using the Net to communicate with friends”, even though the delivery methods are rather different.
Posted by: Hellinar | Mar 29, 2006 at 16:19
An interesting thought, what happens when gameplay, graphics and marketing, stop being the focus of the game, but rather the games ability to facilitate a growning and dynamic community. Perhaps look for the future headlines, that don't talk about Stunning Graphics and Inventive Gameplay but rather about lag free chat and DIY political structures.
Posted by: Jesse | Mar 29, 2006 at 16:32
Hellinar>I think you are missing the point. The bet, as I understand it, is about which model will be introducing more people to VWs in a couple of years time.<
I agree that this is very probably the point of this bet.
But to be able to compare different concepts, those concepts need to have basics that allow to compare them. Like comparable subscription models if the comparison is based on user numbers like in this case. The question itself is very interesting but it won't be answered here.
It's a funny bet for sure but I think even if SL would have the same user numbers as WoW one day, this bet could not be interpreted as a test of "long-term growth potential of user-created content versus the more standard entertainment industry model" under the given circumstances. .)
Posted by: | Mar 29, 2006 at 18:09
Erm, that was me...
To my excuse, it's quite late here in Germany. :/
Posted by: Phall | Mar 29, 2006 at 18:15
Phall>It's a funny bet for sure but I think even if SL would have the same user numbers as WoW one day, this bet could not be interpreted as a test of "long-term growth potential of user-created content versus the more standard entertainment industry model" under the given circumstances. .)<
Are you reading that as “long term revenue growth potential”? I’m reading it as long term population growth potential. It seems to me the VW that is profitable on the smallest subscriber revenue per head is likely to be the population winner, even if behind on total revenue. SL has the edge there, as they only have to put up servers, and the users fill them with content. Blizzard have to keep paying those artists. That’s an exaggeration of course, it would be interesting to know for example how Blizzard’s CS cost per head and SL’s compare. Maybe an advertising supported VW, which appears “free” to the users, will beat them both in the population stakes.
Posted by: Hellinar | Mar 29, 2006 at 19:51
The strict test is the absolute number of North American users; you can read whatever subtext you want into it. I like Hellinar's interpretation of it, and I think I would agree that SL has a definite advantage. Most people playing WoW are already quite aware of VWs; this might not be true of SL.
On a different note...
I think it would be interesting if Linden Lab permitted payment in the form of hardware… Someone tell them to think about it. They could wipe the hardware to make sure there aren’t any viruses or anything, and then hook it in, which shouldn’t be terribly hard, I can’t imagine.
Then people would be able to help them grow as they wished. Of course, it’s kinda non-refundable. =P
Posted by: Michael Chui | Mar 30, 2006 at 06:04
Hey thanks Dimitri - between you and that idiot Cory you have just wrecked Second Life. They removed all restrictions on signup, the world is filling with under-age griefers which Linden Lab has no way of controlling, and paying customers are leaving in droves.
This must be the first time a multimillion dollar company has been destroyed by a 25 cent bet - you'll go down in history.
Posted by: Chris Light | Jun 20, 2006 at 07:21
I'll put a quarter on Second Life not existing in 2 years if LL continue down their current path.
Posted by: KR | Jun 20, 2006 at 08:13
Quarter Bet Watch: Reuben Steiger extrapolates SL's growth and gets 3.6 million SL users by July 1, 2007.
Posted by: Jerry Paffendorf | Jul 18, 2006 at 08:17
Quarter Bet Watch reality check: Reuben Steiger extrapolates SL's growth and gets 3.6 million registered accounts by July 1, 2007.
--matt
Posted by: Matt Mihaly | Jul 18, 2006 at 11:26
Contentious! lol thinking about 3.59 million alts :).
But I dunno, Matt, I work with a company that makes things in open virtual worlds and I get to see the outside projects who want to come in. If SL (and other open platforms) can support their wishes (and the wide load), they're coming in and bringing the party. And that just augments the people already in there building it from the inside out.
I'm looking back on Ted's post, "The Golden 1M: Please Welcome the Next Candidate, World of Warcraft". Some random comments pulled from that thread (not to compare apples and oranges, it's just interesting to reflect):
"I don't think WoW is going to get to a million. My best guess, once the dust settles and the initial phase is over? Somewhere around 400k to 500k subscribers."
"I don't see it hitting 1 million soon"
"Judging from the number of servers and other anecdotal evidence, I'd say that WoW hitting 350k-500k US subscribers is looking highly plausible at this point."
"One million WON'T happen, come on. If they'll break that record they'll do worldwide when both Korea and Europe will launch, but not NA alone. /The current results are still impressive and they could lead to around 400k NA subscribers after the Christmas (and directly beat EQ1 records and so every other american mmorpg to date) which is exactly what I predicted months ago."
"Wow will hit 1m in China alone and perhaps 2m globally. Concurrent users would be in in 500k range."
"Interesting to read some of these comments now that WoW has reached (and surpassed) its 1M subscriber milestone."
And I think it's obvious that the market for open-ended worlds, as they mature and weave themselves into the web, is growing much larger and more mainstream than the market for WoW (let me qualify that by saying that in open worlds or platforms you can still create defined communities of interest, games, and experiences, so it's not all let-the-chips-land-where-they-may Burning Man all the time--you can go to your own special little world within a world--even a WoW-style world, enter Multiverse--where you can be yourself and do your business, and then of course you can go somewhere else, a beautiful pollination cycle).
The story of Second Life since 2003 has been "They only have 1,000 users," "They only have 5,000 users," "They only have 10,000 users," "They only have 25,000 users," "They only have 50,000 users," "They only have 100,000 users," "Sure, they have 350,000 users, but those are all alts and the accounts are free! Cheaters!!" It's been a trip to follow it this whole time, and we'll see what happens next. I'm obviously a supporter of open gaming, virtual world, and map platforms. Go Second Life, go Multiverse, go Google Earth, go whomever's going to free us up to build and connect to the things that we want, no matter what they are. We should want them to succeed.
(Woa, looks like I jumped on a horse somewhere in there. ;)
Posted by: Jerry Paffendorf | Jul 18, 2006 at 13:51
It's too bad you didn't phrase the bet like this:
"Which will have more users in 2 years, Second Life or World of Warcraft?" Then my quarter would be stacked there easily with Cory's and Jerry's. Why? Because there's LOTS of people pouring in and they WILL have more users in 2 years, no contest.
I see from the microcosm of my rentals where I went from about 50 people nearly 2 years ago to about 600 now, with 100 of them just in the last 45 days. And these are real people, real loggers, not alts, who spend money.
However, my quarter has to go over on Dmitri's pile. Because he worded the bet like this: which will have more NORTH AMERICAN users, WoW or SL.
And hands down, WoW will have more NORTH AMERICAN users. SL will have more European, Asian, and even Latin American and African users. The Americans will retreat into WoW in keeping in general with their whole injured international psyche. The rest of the world will be interacting more and trying to become like the old America.
Again, I can only go by my hundreds of customers passing through -- I see the demographics changing. I see American 20-something males and some females leaving to go play WoW because there isn't enough shooting in SL or not enough "game" -- and it's a girls game or it's "too gay" as they often describe it.
I see the f32 demographic burrowing in-- it *is* a girl's game and many women are empowered through SL to even start home-based businesses to help through through the Catherine's Wheel of being stretched between children, husband, old parents, and part-time or full-time work.
I see lots of Europeans and Asians coming in, and they are the hungry kind who have DSL connections through big companies they work for, or through richer relatives or friends, but they want to hit the ground running, learn the skills, and sell avatar wares or land like Anshechung.com.
The unverified, non CC credit cards have made the Europeans and Asians pour in. They will keep pouring in. At some point they may realize that their customers, those skittish young North American males, are gone and not logging on to buy stuff, but that won't be in the next 2 years.
Reason? The African American and hispanic American cohort will keep growing in SL significantly but not enough to put it over the top because no one will think to market to this important demographic in SL.
Then after the 2 years I'll put my quarter back on the North American pile again because, a slower but significant demographic will help swell those numbers back to Cory's side:
o out of work Dems who will have lost the elections again and will be roaming in the Internet/blog wilderness looking for the Next Big Thing
o early retirees from government disenchanted with Republicans and replaced by the growing cohort of Hispanic youth booming demographics
o injured Iraq war veterans
o out-of-work steelworkers and coal miners in Pennsylvania
o newspaper and TV producers and reporters whose media is increasingly suffering losses due to the Internet, blogs, podcasting
o psychiatrists, health workers, and psychics
o unemployed art college students
o me and my alts
Posted by: Prokofy Neva | Jul 18, 2006 at 15:51
Jerry, I think maybe you missed a thread in preparing your reply. This ground of counting (or mis-counting) users has been fairly well covered.
The thing with Second Life isn't that people don't like it, appreciate it, or think it's highly innovative and may well point a solid direction for the future. No one I know wishes SL anything but success. It and others like Multiverse may well point the future of online world-environments (including games).
The push-back you're hearing comes from repeated claims of "exponential growth" and the like, citing registered users or long-tail trailing numbers that either don't hold up on any scrutiny or don't mesh with user-base measures that they're often compared to. So rather than accept "registered users" or other vague definitions of users that have been touted in various places, many of us have tried to find more apples-to-apples comparisons such as unique logins in the past week, weekly average or peak concurrency, or best of all, average revenue per (in this case registered) user (this being a good measure of both an environment's draw and its ability to monetize its use).
So the graph on Steiger's blog strikes me as just a bit nonsensical. If it's measuring anything useful, that would be active users (if it is just registered users, it's not that informative). But he starts by assuming about 350,000 of those, which, unless you use highly generous definitions (e.g. unique logins in the past 90+ days) I don't believe SL has hit that level (and if I'm wrong I expect Cory will be along to tell me :) ). Further, it assumes a linear rate of increase (22% per month), which seems pretty unlikely, given the profile of every other network-growth product (not just MMOs). At some point any such product reaches plateaus in market penetration and adoption, and those dropping off are a larger percentage of those coming in.
Citing old posts about WoW doesn't do much to bolster your case either. Sure, no one predicted anything like their success -- few thought they'd break a million. We were all wrong. OTOH, they had over 250,000 active, paying users in the first three days, so that should have given us a clue. Does SL have the equivalent of 250,000 paying users now? I don't know, but I am inclined to think not. Given that, I think Dmitri's quarter is pretty safe. The probability that they're somehow going to exceed WoW's numbers -- that one is going to collapse or the other explode to the degree that their ~100,000 and 6,500,000 lines cross in the next few years -- seems remote to me.
Posted by: Mike Sellers | Jul 18, 2006 at 16:04
My $0.25 is on Second Life. I go with Linden because I see them as developing an interface for future internet access rather than simply a directed virtual world.
Posted by: monkeysan | Jul 18, 2006 at 18:31
Jerry wrote:
I'm obviously a supporter of open gaming, virtual world, and map platforms. Go Second Life, go Multiverse, go Google Earth, go whomever's going to free us up to build and connect to the things that we want, no matter what they are. We should want them to succeed.
Well, sure, we want whoever is going to provide the services we want to succeed. Success is about substance though, barring flash-in-the-pan internet bubble nonsense, and there's really nothing to be gained (except for Linden) by repeating hype.
Second Life may be profitable now (I have no idea, but I know they were pretty close to it a few months back) but the idea that it will have 3.6 million actual users (as opposed to 3.6 million registrations) next year is a little unrealistic barring stretching the term 'user' beyond usefulness.
--matt
Posted by: Matt Mihaly | Jul 18, 2006 at 18:46
If Linden does get a few Web2.0-type partnership deals done, SL may be the 3D Yahoo-type social/business space that monkeysan bets that it will become.
Will it happen in the next few year? Not the way it is at the moment, but if they, or a Web 2.0 partner, hack a brower-based portal into SL then it may take off like myspace.
Which leads me to think that Cory may have insider information on what's brewing strategically and tactically in Linden. Prokofy has given some insights into that.
If properly promoted to Asia and Europe, Cory can win the bet. Phall's comments point to this. If an interesting free-to-play VW is properly promoted to Chinese audiences, the said VW can get millions of registered users that have been active over the past 4 months (per the term of the bet).
Posted by: magicback (Frank) | Jul 18, 2006 at 21:28
Really quick thoughts as I'm dragged away from the machine:
Prok > The unverified, non CC credit cards have made the Europeans and Asians pour in. They will keep pouring in.
Prok, as an aside on strong international growth, my roommate (Christian Prior in SL--bug him :) wrote some code that lets you use Google's language translator in SL. It's only as good as Google's translator, but it works. Anyone using stuff like this that you know of? Are their still problems displaying characters for some non-English languages? If an answer's too off-topic find me inworld.
Mike > Jerry, I think maybe you missed a thread in preparing your reply. This ground of counting (or mis-counting) users has been fairly well covered.
Matt > the idea that it will have 3.6 million actual users (as opposed to 3.6 million registrations) next year is a little unrealistic barring stretching the term 'user' beyond usefulness.
Mike, I'd read and enjoyed your post and I've spoken with Sir Bruce from mmogchart.com about it as well so I'm not in the dark. Sorry if I was short on that. But I do think the definition of users and how they're measured is stretching.
Some questions off the top of my head: Am I a Second Life user if I log in once a month to attend an event? How about if I'm on the web and avatars are sending me SL IMs through AIM (maybe I teleport through a few of the SLurls they send me, maybe I don't)? How about if I'm playing a game that was prototyped in SL and recreated elsewhere? Do I count in the system? How do we measure the connected pieces?
It's also hard to measure "how far away" people are from signing up. Once the right content in a free and easy to access virtual worlds gets ya, you're gotten. You have an avatar account. It's a soft science, but I think there are quite a few invisible virtual world "sleeper cells" out there waiting to be activated by the content that appeals to them. They'll probably want free access, and they'll probably be hard to measure by traditional MMO standards.
Some thoughts with no looking back. Have to hop at the moment.
Cheers,
Posted by: Jerry Paffendorf | Jul 18, 2006 at 22:06
No, I don't think someone sending you SLAIMs counts, no.
I mean, you're either on the bus, or off the bus, Jerry.
As for translations, people are not going to go outside to fetch google or babelfish. There's this idea that everybody is seemlessly going from SL to the Internet like it's a cake walk. But it means minimizing your screen often, and that makes your avatar slump. Not everyone wants to go AFK to do something. Or SL sometimes acts up, right after the patches, and you can't get the window resizing thing to happen right.
What people sometimes do is try to find an inworld translator through friends, through Live Help, or mentors or other people from that country.
I personally as a RL translator won't replace myself or others with Google or Babelfish. We make sense. They don't : )
Posted by: Prokofy Neva | Jul 19, 2006 at 04:32
To Matt's point about what counts as a user:
I agree that there are all kinds of definitions. Second Life publishes tons of data about what's going on in SL here: https://secondlife.com/currency/economy.php
You have been very clear and vocal that you don't agree with our definitions, which is fine. As I've said before, both on TN and via the backchannel, we can just wait a while and then we'll be able to clearly judge the best metrics. I don't see any particular reason why SL would correspond to MMORPG rates of concurrence, tie, or whatever. We don't have a grind or leveling, so the reasons to log in are driven by very different motivations. If you leave for a month to play WoW, you haven't lost the ability to play with your friends because you're 20 levels behind them, nor are you forced to pay us during that time.
What is clear is that every method of measuring SL's progress -- from residents, to residents online per some unit of time, to concurrency, to economic activity within SL, to real-world transactions -- have *all* been exhibiting gentle exponential growth since we changed the EULA to allow residents to retain their intellectual property. To assert otherwise is to simply ignore the data.
To compare SL's growth curve to any existing MMORPG is a little silly. WoW launched with approximately $10 trillion in advertising (I may be off by a factor of 2 one way or the other), like all other MMORPGs, and is primarily driven by retail sales. SL launched with approximately $0 in marketing and is completely based on viral, referral growth. What is great, from SL's perspective, is that there are numerous examples of other communities and online businesses that have generated extended, exponential growth from this model. I'm sure Mike was just mistyping, but any, fixed month-on-month growth is exponential, not linear, and many examples of this style of growth exist. MS Live Arcade, eBay, MySpace, FaceBook, etc.
Posted by: Cory Ondrejka | Jul 19, 2006 at 09:42
Jerry asked: Some questions off the top of my head: Am I a Second Life user if I log in once a month to attend an event? How about if I'm on the web and avatars are sending me SL IMs through AIM (maybe I teleport through a few of the SLurls they send me, maybe I don't)? How about if I'm playing a game that was prototyped in SL and recreated elsewhere? Do I count in the system? How do we measure the connected pieces?
Excellent questions. These go to the "blurring of the edges" around online world usage that is only going to increase. To define users, many current environments (and significantly, WoW, just because it's the biggest) have chosen "people who pay every month or who log in at least once a week" as "active" users. This presumes a certain usage model and may be too stringent -- to presumptive of a hardcore usage model. OTOH stretching this out beyond "those who log in once every two weeks" or maybe once a month seems kind of silly. Washington DC could try counting all the tourists who visit once a year as part of their native population, but we'd say that was nonsensical.
In addition, whether I'm checking in occasionally or via AIM or whatever will eventually skirt the issue entirely. From the POV of commercial ventures (and these are all commercial ventures now and for the foreseeable future), what matters is how much, on average, every hard-core, casual, occasional, and tourist user brings in: Average Revenue Per User. That accounts for all the matters you mentioned earlier, for different usage patterns, etc.
So on that basis, WoW has something above $15/per user per month the way they measure users, for more than 6.5M users. If SL wants to count all those who have logged in during the past 90 days as users, fine -- but then what's their 90 revenue averaged across all of those users (not just a subset of them)? On that basis, I think Dmitri's Quarter is pretty darn safe (btw, I think "Dmitri's Quarter" could be a new shorthand for comparing virtual worlds on this basis -- better than the Sword of Damocles).
It's also hard to measure "how far away" people are from signing up. Once the right content in a free and easy to access virtual worlds gets ya, you're gotten. You have an avatar account. It's a soft science, but I think there are quite a few invisible virtual world "sleeper cells" out there waiting to be activated by the content that appeals to them. They'll probably want free access, and they'll probably be hard to measure by traditional MMO standards.
Well in terms of measuring online worlds, that strikes me as pretty well nonsensical (a "soft science"?). How many people are "this close" to buying a new car or seeing a new movie? How do they affect the popularity and commercial success of those things? I think the answer is "not at all." In a commercial sense, until someone has bought, they haven't. This is why we've seen so many hugely popular online environments -- from hotmail to MySpace -- with commercial revenue models that are questionable at best. Monetizing users who are there but not spending, or who are "this close" to forking over money, turns out for many efforts to be incredibly difficult. As a result, the only number they can count as in terms of online users for a commercial effort is zero -- and even that ignores the negative aspect of their cost with no revenue benefit.
Posted by: Mike Sellers | Jul 19, 2006 at 09:44
SL launched with approximately $0 in marketing and is completely based on viral, referral growth.
OTOH, you guys have, to your immense credit, driven a free PR campaign that has left many industry heads spinning (as in "how do they do that?"). Lower cost than magazine spreads, but likely more effective too.
I'm sure Mike was just mistyping, but any, fixed month-on-month growth is exponential, not linear, and many examples of this style of growth exist. MS Live Arcade, eBay, MySpace, FaceBook, etc.
Yes - where rate of increase is fixed, growth is exponential. My point was that no network-centric effort that I know of has maintained a fixed rate of growth (e.g. 22%) over a significant period of time: plateaus and even dips are inevitable. The graph shown (not by LL) assumes that rate of growth is constant, which is unlikely except in hockey-stick marketing-land.
Posted by: Mike Sellers | Jul 19, 2006 at 09:50
Prok > As for translations, people are not going to go outside to fetch google or babelfish. There's this idea that everybody is seemlessly going from SL to the Internet like it's a cake walk.
I should have been clearer. His code makes the translator accessible inworld. Typed chat is automatically translated out loud into or out of the language you prefer. Babler(tm) ha :).
> OTOH stretching this out beyond "those who log in once every two weeks" or maybe once a month seems kind of silly. Washington DC could try counting all the tourists who visit once a year as part of their native population, but we'd say that was nonsensical.
Those tourists bring lots of bling and they certainly are measured. Also (lord knows) you don't have to live in DC to be affected by what happens there. As a US citizen, I'm a DC user :). And globalization makes us users of all sorts of places we never (or rarely) spend physical time in. Am I stretching too far? It doesn't feel like it to me. You're not always totally on the bus.
> (btw, I think "Dmitri's Quarter" could be a new shorthand for comparing virtual worlds on this basis -- better than the Sword of Damocles).
Awesome, but only if Cory let's you do that. After all, I think it will be his quarter soon :p. I do hope the Quarter Bet goes down in history. It's most definitely going to be the title of a section in the Metaverse Roadmap doc that outlines this debate.
> Well in terms of measuring online worlds, that strikes me as pretty well nonsensical (a "soft science"?). How many people are "this close" to buying a new car or seeing a new movie? How do they affect the popularity and commercial success of those things? I think the answer is "not at all."
I guess I, very respectfully, couldn't disagree more. If people want to do something and it becomes newly possible within the flow of their life, they do it ("Hey, thinking about a new car? Here's one you like and can afford." Yoink. "Thinking about seeing this new movie but going to the theater isn't part of your life pattern? That's cool, now you can order it at home." Yoink. Tipping points). MySpace is a great example. It showed how many people were "this close" to making their own webpage or putting themselves online but didn't have the know-how or the right online venue to be where other people are.
Related, I think you'll love this. Check out Ze Frank's video on design democracy and ugly MySpace pages (starts with a funny song, kicks into a killer speech 1/3 in). As Ben Linden writes and has been a topic of conversation in our house since it came out, it also applies to Second Life and user-creation in virtual worlds.
Posted by: Jerry Paffendorf | Jul 19, 2006 at 10:57
The point about "almost there" populations is that from a commercial POV, they're not there -- you can't count them. If I told you "X million people plan to travel to New York City in the next month" that's an "almost there" figure. But if I tell you this on Sept 10, 2001, your results are going to prove to be far different from your plans. Yes, you can (and in a commercial sense, have to) plan for growth and conversions of non-paying to paying customers, but you can't count non-paying customers as paying customers until they really are. So, from the POV of an online environment, you have to plan ahead in terms of available servers and infrstructure, but you can't count conversions to paying customers until they've actually happened. So in any real sense, "this close" is immeasurable and really does not count.
This brings up an important point about "tipping points" -- they're only visible in retrospect. They're fun to predict because anyone can do so and seem perceptive. Which ones actually happen is a different matter. Will new forms of online environments (SL, Multiverse, etc.) create a tipping point in how people experience being online? I certainly hope so. Will MySpace create a tipping point in how people use identity web spaces? Maybe. Despite the company's monstrous acquisition price, the jury of the online marketplace is still out on that one -- let's wait and see if MySpace retains its userbase past the fad stage and, more importantly, manages to eke out a revenue stream that can support its social popularity. Similarly, will the current standard/soon to be old-style MMOG grow alongside these new forms, or will WoW and its like turn out to be the dinosaurs of our age, dying thunderously while new mammalian forms scurry around? No way to tell. But at this point, it's too soon to start betting against the brontosaurus.
Posted by: Mike Sellers | Jul 19, 2006 at 11:25
Riffing on the bronto-mammalian front, and also the generally predictable nature of cyclical "tipping points" or major shifts (predictable in their development, but not their exact evolution--who all the players will be, exactly when, etc.). This is Raph Koster from a conversation at the Metaverse Roadmap Summit (single quotes because I slavishly transcribed hours and hours of audio and may have some "a"s and "the"s switched etc.). He's discussing a 7-10 cycle in the MMO industry (and pre-industry) and what he sees happening next.
'User-created and dynamic content will replace topdown content. We'll see more scrappy, independent content. The rise of independence is clearly a story. In 1997 there was an apocolyptic event that killed all the existing MMO providers. It was the shift to a subscription-based business model paired with game level production values. The thing was the dinosaurs that got got killed off were the heirs of an earlier apocalypse in 1989 which was the shift in earlier business models to an hourly closed service model, heirs to a shift in 1982 from academic VWs with no business models. And in all those shifts, the existing companies pretty much all died. And we're due for a shift. The last explosion was 1996, 1997 (Dark Sun, Ultima, Lineage, Asheron's Call, Active Worlds, and others). There will be a production shift married to
a business shift and the Blizzards of the world will face a new round of mammals.'
He goes a little further too, but I'll leave it there so he can either go further himself or smack me for quoting him on that :).
Posted by: Jerry Paffendorf | Jul 19, 2006 at 11:55
Jerry Pappendorf wrote:
Some questions off the top of my head: Am I a Second Life user if I log in once a month to attend an event? How about if I'm on the web and avatars are sending me SL IMs through AIM (maybe I teleport through a few of the SLurls they send me, maybe I don't)? How about if I'm playing a game that was prototyped in SL and recreated elsewhere? Do I count in the system? How do we measure the connected pieces?
Why don't we count anyone who uses a Visa or Mastercard as a user too, since Visa and Mastercard are affected by Second Life. Or, being reasonable, we might decide that simply because one is affected by something doesn't make that person a user of something in any meaningful sense. In any case, if you're going to start counting second-order relationships as users, give up the ghost now, as WoW has tons of those as well (friends, family, spouses, of players, etc. All affected.)
Cory wrote:
What is clear is that every method of measuring SL's progress -- from residents, to residents online per some unit of time, to concurrency, to economic activity within SL, to real-world transactions -- have *all* been exhibiting gentle exponential growth since we changed the EULA to allow residents to retain their intellectual property. To assert otherwise is to simply ignore the data.
Sure, I don't think anyone is disputing that SL is growing, but that's not what this discussion seems to be about.
WoW launched with approximately $10 trillion in advertising (I may be off by a factor of 2 one way or the other), like all other MMORPGs, and is primarily driven by retail sales.
With all due respect, I don't believe that's true at all. WoW's success is largely due to referrals. Yes, they run advertising, but as a recent study clearly showed, the most important factor in purchase is positive word of mouth from those you know. We run quite a bit of advertising for our company's size (small), but referrals from friends still make up about the same as what all other marketing/advertising brings in, combined. Referrals from friends act as a multiplier for all other forms of marketing, and purely anecdotally, I'm sure you are as aware as I am of the many people around you who started playing WoW because of word of mouth rather than advertising.
Cory wrote:
SL launched with approximately $0 in marketing and is completely based on viral, referral growth.
Well, ok....but advertising is not typically counted as viral or referral growth, and SL advertises. I have no idea how many players SL's advertising gets it (could be trivial), but it's not based completely on viral, referral growth. I was playing with Adwords just the other day, for instance, and came across Second Life ads in a couple of the keyword groups I was bidding on.
Mike Sellers wrote:
So on that basis, WoW has something above $15/per user per month the way they measure users, for more than 6.5M users.
To be fair, a good portion of their users don't pay anything near to $15/month (such as their Chinese users).
--matt
Posted by: Matt Mihaly | Jul 19, 2006 at 12:19
Matt Mihaly> the most important factor in purchase is positive word of mouth from those you know ... Referrals from friends act as a multiplier for all other forms of marketing
I agree, where we're talking about gamers. But what about distinguishing between those who are already gamers and those who've never played an online game?
I can see referral being an effective way to market to people who are already plugged into some network of users (such as the community of people who play online games). Is there any evidence that mass market advertising is more effective at bringing new players into a network?
--Bart
Posted by: Bart Stewart | Jul 19, 2006 at 15:50
I simply think that it's impossible in SL's current incarnation to provide the hardware that could possibly support as many users as WoW. You'd need a facility the size of freakin' Manhattan to hold all the servers and an exclusive contract with Hoover Dam to power it all. Just ask Blizz how many servers they have worldwide - I'll counterbet you $.25 that SL already has way more.
Posted by: Psyra Extraordinaire | Jul 19, 2006 at 16:46
Blizzard has spent years building a positive world of mouth around its brand. This takes nothing away from WoW or Blizzard, the point was simply that comparing the WoW launch to SL's launch is silly.
And, yes, the discussion is about growth and how we measure growth. Mike has repeatedly insisted that we use MMORPG metrics to measure SL. There is little reason to default to those metrics for SL. We aren't a game, we definitely aren't an MMORPG, so there is little reason to suspect that metrics from those products should be applied to SL.
Posted by: Cory Ondrejka | Jul 19, 2006 at 16:47
Cory, I'm not insisting on anything other than apples-to-apples comparisons. To the extent we talk about MMOG measures it's just because the business models are sufficiently similar, and the MMOGs established the model first. Without equal basis comparisons we quickly get into all sorts of nonsensical claims: I can say that my system has over a billion users with equal validity that you say SL has 350K users, even though mine isn't out yet (because I measure my user base by everyone on the planet who might someday play it -- taking Jerry's "almost" users just another step). Along these lines, saying that SL will have more users than WoW requires such an even-field comparison (and hey, it's not my bet).
Moreover, using commercially relevant measures of users in a commercial sphere is the only thing that makes sense. Talking about registered users, lumping together paying and non-paying users, or propping up actual active usage with long-trailing numbers might be good (if obscurative) marketing-speak, but doing so does not give an accurate picture of the commercial viability of a product/service that depends on consumer-based revenue to succeed.
This goes beyond trying to measure popularity or shallow "king of the hill" kinds of questions. Measuring revenue, users, and (as I've said a few times now, echoing others) revenue averaged across all users (ARPU) is the only way I know to assess likely commercial viability for commercial online environments. If SL has ten million "users" by its own measure but only a highly concentrated revenue stream that equates to a monthly ARPU of $1, that's not nearly as commercially viable as having 10% of those users with a $15 ARPU.
Or to use a different example, if MySpace has 48 million users as touted, that's great, right? They win! Or do they? If MySpace also makes only $30M this year, suddenly that number is put in some perspective, with a monthly ARPU of around 63 cents. Yes it's popular in a faddish way and people spend hours there... but as a business their revenues are about 4% of traditional MMOGs on a per user basis. Suddenly the mad rush to "be like MySpace" (see, e.g., a discussion of Walmart's ill-considered entry here) doesn't look like such a no-brainer (and IMO, the $500M+ acquistion of MySpace begins to look more like the $400M+ acquisition of hotmail several years ago).
So the point here isn't popularity or trying to hang on to the primacy of MMOGs over other online environments. But there are social and commercial reasons for wanting to compare users on an equivalent basis. If you know a better way than talking about ARPU, I'd be interested to hear it.
Posted by: Mike Sellers | Jul 19, 2006 at 19:58
Talking about registered users, lumping together paying and non-paying users, or propping up actual active usage with long-trailing numbers might be good (if obscurative) marketing-speak, but doing so does not give an accurate picture of the commercial viability of a product/service that depends on consumer-based revenue to succeed
The problem is that SL doesn't depend on consumer-based revenue. They depend on hot-air marketing and idiot venture capitalists. My understanding is that they aren't making enough money off the "residents" to pay the electricity bill, much less finance the entire operation. So in that context, why even bother to compare an online world that's making money hats for everyone involved, with one that isn't producing anything more valuable than carefully scripted "marketing-speak" sound-bites?
Posted by: | Jul 19, 2006 at 20:24
Mike,
I've never disagreed about using ARPU, however since nobody is sharing ARPU, it seems like we spend a lot of time circling around other metrics that aren't apples-to-apples.
Posted by: Cory Ondrejka | Jul 19, 2006 at 20:41
Part of the problem is that, contrary to most other online VWs which require some nontrivial payment to keep an account active, with SL's new registration system that doesn't require payment info (or even a valid email address!) to be submitted, what defines an "active user" has become exceedingly tricky.
Is someone who has paid LL money in the past N months a user? I think that is fair to say, even though the amount paid can vary by orders of magnitude (and could be as little as $0.30, compared to $15 for a traditional subscription-based game) from user to user.
The trick comes in the counting of those who provide zero revenue, as not everyone who has registered an account can be legitimately considered a "user".
This basically comes down to a statistics problem. i.e., if a free account is registered, what is the length of idle (which I'll define as no login, no payment made) time such that you can be reasonably sure (say, 95%) the user will not return and can be eliminated from accurate counts?
Obviously, we don't have access to your user stats, so we can't know things like the distribution of mean time between logins, but in my opinion, an estimate of that distribution and account counting based on such a statistical metric is a necessary first step to making a meaningful comparison. Saying that "90% of users have logged in at least once during the past 90 days" doesn't mean much if your usage stats show a person that goes 20 days without logging in is 95% likely to never log in again.
Posted by: Tom | Jul 20, 2006 at 11:29
Cory said:
"To compare SL's growth curve to any existing MMORPG is a little silly. WoW launched with approximately $10 trillion in advertising (I may be off by a factor of 2 one way or the other), like all other MMORPGs, and is primarily driven by retail sales. SL launched with approximately $0 in marketing and is completely based on viral, referral growth."
Amen. It's really interesting to watch this debate unfold but ultimately frustating because its almost intrinsically doomed because of the the apples to oranges problem. Although Second Life looks like an MMORPG and uses much of the same tech, the similarities sort of end there. Both in terms of adoption and usage patterns, Second Life's growth has been much more like that of the Internet than like a single game title.
My prediction and hope is that it will continue to do so (grow like the Net). In fact, one could argue that the accelerating success of SL is the result of a series of conscious anti-MMORPG-model decisons. Listed briefly these would be user created content, user ownership of their own content, the allowing of free trade in virtual items and finally the move to a free model where the majority of free users support the paying minority and allow them to pay much larger monthly fees.
For SL to continue to move toward an Internet-like model, it will require sustained vision and counterintuitive decision making on the part of Linden Lab. Specifically, to accomodate the scope of their vision, they will need to continue to sacrifice control and/or ownership of slices of the software stack, placing those in the hands of users or 3rd Party companies.
To me the question isn't will SL catch WOW, it's when. And while it's fine for us to spend our time arguing over definitions of users or time spent in world, the truth is that SL and WOW are radically different beasts. WOW is arguably the most polished example of MMORPG the world has seen to date -- this type of highly orchestrated and refined content is something that distributed content creation will never (IMHO) threaten. I'd love to be proven wrong and will be overjoyed to sit enthralled in the audience watching a movie like Titanic that was made by thousands of loosely connected prosumers, but I highly doubt it'll happen. And hence, there'll always be room for WOW and it's sucessors.
SL is something radically different and ultimately much larger. What I'd love to see is a shift in focus, away from the "Who's Bigger and Badder?" debate. Instead, why don't we start puzzling out the really tough problem of how Second Life (and ultimately others like it) should handle the radically difficult problems and great responsiblity that come with creating a true 3-D internet.
Posted by: reuben steiger | Jul 29, 2006 at 21:58
I think it's unfortunate and missing the point to call this discussion a "grudgematch" (as you did on your blog, Reuben) or to think that it's about who's king of the hill between the old cathedral-like MMOGs and the new bazaar-like Second Life. From my POV, it's not about that at all. The reason this discussion has legs is because it's ultimately about honesty, accuracy, and about not giving into or believing the hype of the moment.
World of Warcraft is a phenomenally successful commercial venture (stating the obvious, but stick with me for a moment). We know this because Vivendi, its parent company, releases both revenue and earnings statements drained of marketing exuberance. When WoW was first deployed, many industry watchers (e.g., venture capitalists) did not believe any online world would ever reach a million paying customers; the ones I was talking to at the time didn't believe there were a million people altogether in the worldwide MMOG market, much less that could be attracted by a single game. And of course, none of us industry insiders or crystal-ball gazers remotely predicted WoW's commercial success.
So we were wrong in much the same way that in the mid-1990s many industry watchers, venture capitalists, and old Internet hands (e.g., Cliff Stoll) predicted that the Web was a fad and would never really be useful for anything or would matter to any non-geeks. This is how things are with disruptive events and technologies; they're only visible in retrospect.
The other thing that often happens in situations like this is that a type of 'irrational exuberance' (to co-opt Alan Greenspan's Internet bubble comment) emerges: if things are going this well now, they'll surely go much better in the near future! This sort of thinking gives rise to MMOG business plans predicting "conservative" user numbers in the 5-10 million range, in much the same way that it gave rise to visions of remote-driven (or even flying) cars when the interstate highway system was new.
And it may in the same way be allowing us to indulge in visions (whether prophetic or flights of fancy is difficult to say) of a 3D web, a Metaverse, a distributed, ubiquitous set of virtual environments that pervade our daily life as email and html do now.
To have a vision of this is terrific. To act on it, to put your heart, mind, and career on the line for it as many at Linden Labs are doing is an example of the best sort of innovative spirit.
But: to cheer these efforts on with dissembling numbers, puffing the situation up beyond where it is or shows signs of being, is in my opinion a mistake and ultimately dishonest.
So when someone says Habbo Hotel, for example, has over 50 million users, huge red flags go up in my mind: what do they mean by "users"? When this is clarified to over 50 million characters have been created, that makes more sense and is somewhat less remarkable. When they report that over 7 million unique users log in to HH every month, that to me is still questionable but also more significant -- and the capper is the ultimate commercial quantification, that they report (unaudited) over $30M in revenue last year. If they do have 7M active users that's an ARPU of only $0.37 per month, but who cares: with their likely margins that's a terrifically successful venture (and here the irrational exuberance monster perks up again: just think how much it'd be if they were getting 10x to 40x that amount per user, like WoW is...).
Back to Second Life. Cory and I (among others) have sparred on how many users SL "really" has, since that (along with user-based revenue) is a key mark of of market adoption, and is a necessary precursor if SL is going to lead us into the promised land of the 3D Internet Metaverse. Not surprisingly, continued real (user/revenue) growth also appears to be a necessary component of LL's business plan. As with Habbo Hotel (and many other open/non-subscription online environments) answering the "how many" question depends what you mean by "user." And, for a commercial venture, how much revenue you're able to extract from each one.
If SL defines users as "anyone who's logged in in the past 90 days" to me that's not useful or even entirely honest: what's the probability that someone who hasn't logged in for 30, 60, or 90 days is ever going to log in (much less pay money!) again? I don't know, but I imagine LL does. And I'll bet that even with a non-core, non-grind usage model, after 30 days the probabilities fall of precipitously. At some point they're just not coming back, and calling them a user or a resident unduly inflates the number of people actually adopting the service in any meaningful way.
So: to bring this back to your original graph, Reuben, and other visions of the future. You "did the math" as you said, starting with about 350,000 users for SL. Where did this number come from? Have there been 350K unique logins in the past 7, 30, or even 90 days? (FWIW, Blizzard uses a 7-day figure for non-subscription users, as I've pointed out before, which is both conservative and fits their usage model.) You then assumed a continual 22% growth rate, compounded monthly. No leveling off, no reduction in the monthly growth rate, no saturation effects. That seems to me to be unlikely to say the least, and reminiscent of the bad old days of bubble-era hockey-stick growth curves.
But so what? Why is this a big deal at all? I believe it is a big deal because, as you say, there are truly difficult problems facing us as technology innovators working within a dynamic whitewater market. There are many difficult questions we each face: what products and services will people pay for, and how much, within an online environment? Is this a broad untapped market or a narrow niche to be mined further? Can online environments be applied to non-entertainment contexts in a commercially sustainable fashion? If so, does the technology exist to support these users? And ultimately in a lot of future-watchers' minds, is a 3D Internet an inevitability, or will it be relegated to the same page as flying cars in the history books?
We all want to know. We want the future to get here as fast as possible. And some of us -- most of us reading blogs like this one -- want to be part of making it happen. All well and good; as the saying goes, the best way to predict the future is to create it. But making it doesn't mean faking it. Outside of the most facile of marketing, it doesn't mean jumping ahead of where things really are or positing unlikely scenarios for success.
I desperately want WoW, SL, Habbo Hotel, and many other online environments to succeed. But not enough that I think we should overstate their merits or overlook their potential shortcomings. This isn't about "who wins" or who's bigger. It's about reality, and not losing sight of that as we all pull as hard as we can on bringing the future into the present.
Posted by: Mike Sellers | Jul 30, 2006 at 01:05
I was one of the people who was wrong about WoW. I thought they'd hit 450-500K, maybe 1M world-wide. I underestimated both the amount they would cannibalize from the existing games and the draw of the brand in Asia. In hindsight I can see the statistical shadow of a "Waiting for WoW" effect on the industry growth curve that should have clued me in, but hindsight is only useful for revising future analysis, not forgiving past errors.
SL has great potential, and impressive growth, but a naive projection of best-case numbers is setting things up for a fall. The biggest problem is, as Mike Sellers has stated, a lack of transparency. I can suspect marketting spin on other games, but sooner or later they have to report a "hard number", generally revenue.
Habbo looks impressive in number of players, until you see the revenue figure, then you have to assume that they are using either an extremely elastic definition of user, or they are doing really poorly at monetizing those users.
Second Life is not very transparent in their numbers, rarely qualifying the ones they provide. Which leads to them being treated with skepticism, especially when they seem to push the boundaries of a basic sanity check (SL is bigger than every US game except WoW, but almost no-one outside of these circles is talking about it much?).
For example, I could claim 20K users for VHR. That number of accounts have been created, after all. But VHR has a free trial, and in spite of a pretty high conversion rate for a free-trial game (2%), 20K in the front end doesn't add up to that many at the level that counts. Fortunately costs per user are very low and revenue per user is pretty high. And growth rate is very good (25% monthly) with no advertising. But projecting that growth forward and saying that in 4 years we'll have over 10M users and annual revenue of $2B wouldn't just be optimistic, it would be ridiculous.
Extraordinary claims require extraordinary evidence.
--Dave
Posted by: Dave Rickey | Jul 30, 2006 at 02:37
Reuben,
As you can see, I'm on the bet here figuring SL will grow a lot more. But one of the ways it will grow is by Lindens and post-Lindens not being so ideologically fastidious about insisting that "it's not a game". Of course it's a game. To most of the people using it. Long before Hamlet made this joke, I was saying "It's a game [slap]. It's a platform [slap]. It's a game AND a platform [cries] [Chinatown]."
It's a game. It plays. It has play. It's a game. This constant Newspeak around it gets tiresome. You can airily call it a "platform" and cite all the educational and business use of it, and try to front all the cancer fundraising and other good stuff but these 350,000 or million that you and others want to get here are going to be playing house, playing store, playing the land market, playing the stockmarket, playing an internal game-within-game, playing hard to get, whatever. *Playing*. So it will be a *game*. This is not about the uses of SL to perform complex remote neurosurgery by Indian doctors getting lower salaries to operate on somebody in America -- um, at least not yet.
I will stay out of the apples-and-oranges debate because I don't know enough about what the REAL numbers of any MMOPRG is. I belong to WoW and any number of games where I don't log on for 90 days or even 120 days, but somebody may be counting me.
What I find intriguing and also frankly a bit creepy is your statement here: "Finally the move to a free model where the majority of free users support the paying minority and allow them to pay much larger monthly fees."
So...let me get this straight. People like me who pay big tier bills compared to the average user, or people who are huge content creators and have at least some land to put a store on, or a workshop, those are the ones who will be "supporting the free majority".
I'm not sure what we get in exchange for this very heavy burden, a microcosm of LL's own heavy burden in absorbing and sustaining free accounts.
This is a very different model than WoW where we all have to pay $49.95 for the game, then $15.00 for the subscription or whatever the fees are and where "content creation" could only be understood as non-inventoriable content like a guild with an interesting name and history and capacity, etc.
So...you want to build SL on our backs. O-kay. I think many inworld investors would like to have more say in things given their status as back-providers, but that's a whole big discussion. For the short term, however, you really need to come clean with some more numbers so we can understand how back-provision is really working.
The number of private islands sold per month is displayed on the economic statistics page. It's not clear if it is a complete number including hidden, Linden, and cut-rate educational islands, but likely it's most of them.
The number of mainland parcels is displayed for sale, which is a rough way of following how this "let's have the tier-payers support the whole enchilada" concept is going.
But what we really need to see is the following:
1) Number of Premium Accounts -- people who subscribe, for either $9.95/month or $72/a year and get $400/week stipends now, valued at $4.80 US at current rates of 300/$1.00 so that's about a half-price off if you cash out your game loot. How many of these are there??? We were once told there were 9000 of them when the population was listed at 40,000. And now? Now that the 2000 in their packets (if they signed up before 7/21, they have 2000, or $6.00 value per month; that would pay for an annualized subscription but no longer the quarterly or monthly, and that has caused some people to dump it -- after all, if they need 2000, they can just buy it for $6.00 outright now.)
2. Of these premium accounts, how many make use of their 512 and just get the 512 tier-free first land only?
3. Of that number, how many of these get more than 512 (we were once told that this figure was 6,000 out of the 9,000, illustrating that most people bought the premium account and likely annualized it or at least put it at quarterly to get discounted Lindens in their box every week automatically).
4. How many free accounts turn to premium, and over what time?
Now that island rentals, and island billing, are done as a separate system outside of premium accounts and the mainland tier system, how many premium accounts are there with land more than 512, sustaining at least the mainland experiment?
This figure, if it is growing, might indicate world-growth and health -- it might be growing too slowly or there might be proprietary reasons why these figures aren't released.
Eyeballing the figures of land for sale, looking at the map, and looking at the figure of Linden sources released into the world each week (that includes those old $50 freebie accounts too, so it's hard to tell), I figure there are probably at least 20,000 premiums now, maybe a lot more? Estimates?
Looking at the email list of auction byers that Ryan Linden once accidentally copied to everyone on the list (bcc is your best friend btw), I'd have to say there probably aren't more than 500 big tier payers today eligible for the concierge services. But I could be very wrong about that.
I don't know what the cash burn is for LL every month with payroll and accounts payable etc but since Philip does tell us "it is close to being profitable" I'm thinking that we may be close but not REAL close to this ideal: "finally the move to a free model where the majority of free users support the paying minority and allow them to pay much larger monthly fees."
There's a lot of painful decisions coming up, Reuben, you're absoutely right. The question is, who will be feeling the pain. And I imagine it will be the people who already have something being built on their backs, eh?
One of them is how to give political autonomy and share power with these "paying minority" people who help Linden lab manage this very complex world, but who have the status of Russian serfs and quit-renters on the Governor Linden Estate. The new group tools of 1.12 might help liberate these serfs a little more, but many complex governance issues remain.
The idea of making SL a bunch of Internet pages is how the Lindens are going, but that steps on the world. The world is the product. The world and its coherence and immersiveness is what makes people show up. When they show up and find their land is lagged, littered with 16m2 sign griefing and extortion, and teenage shooters, all introduced as part of the "like-the-Internet" concept, they log off, and fall back into that quagmire of statistics called "the 350,000".
Posted by: Prokofy Neva | Jul 30, 2006 at 15:26
Cory said:
"What is clear is that every method of measuring SL's progress -- from residents, to residents online per some unit of time, to concurrency, to economic activity within SL, to real-world transactions -- have *all* been exhibiting gentle exponential growth since we changed the EULA to allow residents to retain their intellectual property. To assert otherwise is to simply ignore the data."
This is true. However the growth rate appears to be highly dependent on what is being measured, and Reuben's naive use of the "22% per month population growth" figure to predict the true userbase of SL is shaky at best.
True, the number of accounts registered may be growing at 22% per month, but given the ease with which one can register a throwaway SL account, that number isn't terribly meaningful. As an alternate measure, using semi-public stats one can show that the peak concurrency is growing at rate closer to ~5% per month.
Projecting true active userbase a year in advance from these disparate growth rates leads to a difference in population estimates of an order of magnitude. Even the "conservative" estimate of 10% per month is likely to be overly generous by a factor of 50% or more when projected a year out.
Posted by: Tom | Aug 11, 2006 at 15:33
Here's an interesting set of figures posted by Sally Linden this week on the official Linden blog, in answer to queries by residents who were puzzled over wierd fluctuations of the numbers on the home page, going up and down quite a bit:
"The number that is currently on our home page is a time-weighted average between “total number of signups ever” and “total number of logged in users over the last 60 days”. As of right now, those numbers are 493,563 and 225,028."
Total number of logged in users over the last 60 days is way more than I thought; this is the number many would like the Lindens to publish regularly.
See full discussion here:
http://blog.secondlife.com/2006/08/11/what-is-that-number-anyway/
Posted by: Prokofy Neva | Aug 13, 2006 at 01:54
Даже хроникерам трудно выжать из себя что-то вразумительное после трех нулевых ничьих подряд. Аналитические заметки же страдают не столько от нулей на табло, сколько от невразумительсти действий клубного руководства, от огромной разницы между провозглашаемыми намерениями и получающимися ( надеюсь, пока) результатами.
Мы все аплодировали Президенту Зенита Сергею Фурсенко, когда, вступая в должность, он сказал, что Зенит должен выигрывать все турниры, в которых участвует. Это было выступление, достойное команды, города и болельщиков. Естественно, мы ждали сформированного плана и вразумительных шагов. Надежды наши подкреплялись тем, что уж лучше Газпрома найти хозяина для команды в России в принципе невозможно. Богатая компания, с серьезнейшим политическим и деловым авторитетом в России и Европе, которая может сделать то, чего не было никогда во всей истории Зенита – потратить достаточно денег для серьезного усиления.
Posted by: ЦСКА | Aug 13, 2006 at 05:58
I was just wondering what everyone's take was on Entropia Universe, formerly Project Entropia. It's designed as a complete universe and is going with the free to play model. It also differs from normal MMOGs by allowing players to deposit real world money for in game money at a fixed rate at any time.
Rather than being designed as a universe by the users for the users as SL is, EU is designed to be a game. However, unlike WoW, it's impossible to do everything. To begin with, there are a wide range of skills covering the normal hunting/mining and ranging to obscure professions such as humanoid investigator or plastic surgeon. The second issue is that each of these skills goes to a maximum level of 20,000.
MindArk do realise that this is a bit restrictive for the players and have made it possible to transfer skills between different characters, at the cost of skill being slightly degraded. The question is whether the trade off is worth the time saved.
Version updates happen on a regular basis and new skills are being implemented too. The single universe for all players regardless of their location is a good idea but I am not too sure about how convinced I am that linking loot drops to the actions of all the users is a good idea.
Finally, there is already a person who is making his real world living through investments that he made in the game universe. I know that it is not directly related to the issue of which of the two VWs will have more users in 2 years but it definitely says something about how viable the universe model of MindArk is.
Posted by: Sam Jacob | Oct 03, 2006 at 20:58
Quarter Bet Watch: Second Life just hit 1 million accounts made, with almost 400,000 logging in in the last 60 days. Numbers on secondlife.com. This is faster than Reuben's graph.
No analysis, no arguments, just saying. (Full disclosure: I am a 25 cent investor ;))
Posted by: Jerry Paffendorf/SNOOPYbrown Zamboni | Oct 18, 2006 at 13:16
Second Life just hit 1 million accounts made, with almost 400,000 logging in in the last 60 days.Numbers on secondlife.com. This is faster than Reuben's graph.
No analysis, no arguments, just saying. (Full disclosure: I am a 25 cent investor ;))
In my opinion, Reuben's graph is not terribly useful. But, if you're not going to do any analysis of the available data, then please allow me... :)
(all numbers are derived from my own analysis of the unique login and total population data that Linden has made publicly available)
Accounts created in the last 60 days: 416,165
Unique accounts logged in over the past 60 days: 396,616
Difference: -19,549
So, no fewer than 5% of the accounts registered in the past two months didn't bother to log in even once.
Assuming a 10% initial-abandonment rate, we find that roughly 22,000 of the accounts that existed on August 17th (60 days ago) have logged in since then. That is, roughly 4.2% of the 524,000 accounts that existed on that day have been in-world over the past 60 days. This activity trend can be modeled with a Gamma distribution over time, with a peak login percentage around 10% at the 14-day point (i.e. 10% of accounts registered prior to Oct 3rd have logged in over the past two weeks).
Plotting the unique logins vs. time interval data, we find that approximately 53,000 unique accounts logged into SL yesterday, with 40,000 of those being previously-existing users, and the other 13,000 being new sign-ups.
If SL follows the industry-standard convention that roughly half of all active users will login over the course of a day, this amounts to a real population of around 90k-100k. This is consistent with the fact that the economic statistic of "users who bought something over the past month" is sitting around 82,000.
So, all this suggests that although there have been one million accounts registered since SL opened, a heck of a lot of them that are just taking up disk space...
This is just a sample of the information that can be gleaned from the public data available.
Posted by: Tom | Oct 18, 2006 at 14:24