Clay Shirky and the "Try Me" virus

For once I get to (respectfully) disagree with Clay Shirky, who has just published an analysis of the synthetic world Second Life: while he injects some welcome sanity into the current SL frenzy (see this previous post) he also got, I believe, a couple of things wrong.

Clay's piece basically says three things:

1. The SL membership figures (about two million) are meaningless. That figure refers to anyone who has ever created an SL account; Clay estimates that SL's churn (people who try it out and never come back) "is in excess of 85%" and says:  "I doubt that the number of simultaneous SL users breaks 10'000 regularly". He's right, simultaneous users are actually somewhere around 10-15'000 these days, but there is no doubt allowed: those figures are public, right up on SL's homepage for everyone to see.
2. Second Life is nothing new: Howard Rheingold wrote about MUDs (Multi-User Dungeons) in 1993 in "Virtual Communities" with the same enthusiasm used today for SL; a few years later Silicon Graphics' VRML vas touted as a 3D turning point that would transform the Internet; other virtual worlds have come and gone, as have other overhyped technologies (Clay mentions the "information push" system Pointcast and videophones).
3. The press is the culprit: the current media mania around SL is about young, unexperienced reporters who don't know of the previous failed virtual worlds and like SL because it's "a story to good to check", it's conceptually simple to understand ("you are a person, in a space"), it's supply-driven ("most of the stories are about providers' adoption", IBM virtual meetings and Reuters' SL office and so on, which can be told in press releases that the press is happy to copy/paste).

Here is the central paragraph of his post:

I suspect Second Life is largely a "Try Me" virus, where [media] reports of a strange and wonderful new thing draw the masses to log in and try it, but whose ability to retain anything but a fraction of those users is limited. The pattern of a Try Me virus is a rapid spread of first time users, most of whom drop out quickly, with most of the dropouts becoming immune to later use. Pointcast was a Try Me virus, as was LambdaMOO, the experiment that Second Life most closely resembles.

Clay is right to bring some historical perspective into this, which is often forgotten (Ethan Zuckerman, who agrees with him, adds some more, and clarifies a bit the distinction between games and non-game virtual worlds). But while his social analysis is cogent, he somehow forgets to apply that prism to technology. Which also has a history. SL is different than previous attempts at creating synthetic 3D worlds in that it is a platform that lets users generate their own spaces (through a scripting language called LSL) and incorporate various kinds of media, but also because it comes at a different time - a time of powerful computers and broadband to the masses. Skype was also around a dozen years ago - there was a system developed by Israel's Vocaltec that was basically the same thing, only without broadband nor compression algorithms nor powerful machines capable of handling voice and peer with others, and indeed only a few thousand people were simultaneously using it at any given time. Today Skype is at the forefront of a movement transforming the world of telecoms.

Second Life is still an early version of a useful 3D world; it's clunky and rough, requires powerful hardware, and walking in comes with a steep learning curve; most of the time, the experience of "being there" still trumps the content. But that's SL in its current incarnation. What really matters, and what accounts for at least part of the current press interest in SL, is its potential developments:

• SL or similar synthetic worlds may become the next Web: the next platform for accessing information. The embedding of various media formats through hyperlinking is only step one. True, turning Amazon.com into a 3D world where to virtually walk the aisles and check out book spines is not really interesting, but that's also not very imaginative.
• SL adds a social element that the Web lacks today: we may be many visiting the same webpage at the same time, but that remains an individual experience; everything one does on SL is instead eminently social: the other users are there, as avatars, visible, contactable. SL shows that there is a potential for linking what you know (the information you access) with who you know, and that's the Web's next frontier.
• This will be facilitated by the future integration of additional features: VoIP calls being top of the list.
• Right now, SL functions on a proprietary platform, controlled by Linden Labs. The users' growth and the involvement of big entities such as IBM may hint to the emergence of "external" SL islands (possibly open-source), and there will be demand to link them all.

Now, the hype is disturbing, and all of the above doesn't mean that that future platform will necessarily be Second Life itself: we could soon see another synthetic world emerge, better and more flexible and more open and more usable. The point is that, after many false starts, 3D worlds may have reached a tipping point.

UPDATE 9 January 07 - About the forth point above (the emergence of "external SL islands (possibly open-source)", well, apparently it's happening.

Comments

[Clay Shirky]

And a parallel hypothesis is that 3D Worlds may not have reached a tipping point.

I certainly am not comfortable divining between those two possibilities without knowing what the people with the most at stake, the people who have tried it, actually think.

At this point, I don't know whether I am a Second Life skeptic, but I do know I am a Linden Labs skeptic. The issue is perfectly illustrated by your quoting of their Currently Logged In numbers; I have access to those same numbers, but they are useless, as they mix real users with 'try it and bail' users. My question was and is "What is the number of simultaneous _return_ users?" Neither you nor I have access to those numbers; only Linden does, and though they report lots of numbers, none of the published numbers help answer that question.

[BG]

Hi Clay. Agreed: we are both Linden Lab skeptics (as I said in the post, the future 3D platform may well not be SL). And definitely the figures for return users would offer some elements that could help speculate about the future. I have a possible survey group for that, I will see if I can poll them.

[david spring]

after a week or so in SL to write a story for a good old print magazine, i noticed a few topics.

first, i think that we were all suprised by the gap between the 2.3 millions residents and the 10 000 or so that are "online now" given by secondlife.com. but i am happy with this, because the lag problem in SL is so bad that you cannot even access to some places, because of the crowd. ok, it's not the museums but more "amsterdam" or places like that...

then, LL knows the problem of user retention, if you read this quote on new world notes: “Although Second Life is still challenging to get used to, about 10% of newly created residents are still logging into Second Life weekly, 3 months later. 10% is pretty good given the computer requirements and steep learning curve” Philip Rosedale, CEO of LL.

then, i was a bit disappointed by reuters offer (yes, i dare). you can wander in SL with the latest news at the corner of your screen, which is cool, and very cyberpunk. but if you click on the title of the article, you'll spend hours waiting for your browser to open. yes, the full text is still on the net, this old media. so i dream of the moment when parts of the web will be really integrated in SL, and that you don't have to switch from SL to Firefox anymore. Maybe the magazine PixelPulse, available only in SL and not on the web, shows kind of a direction.

another small thing. i met very interesting and surely very busy scientists in SL. i could spend time to chat with them, which i am sure is not easy to do RL or by phone. it reminds me the good old days of email, when you could get answers from even the most known people. this way of communication was new, so busy people (like linus t.) took time to answer to nobodies. maybe this "extra time" moved to SL for the moment.

one very last thing: in 1964, daniel galouye, a sci-fi writer, has written the novel simulacron 3 (http://en.wikipedia.org/wiki/Simulacron-3). it kind of anticipates all virtual worlds... thanks to patrick gyger, director of the "maison d'ailleurs", who knew this book! so maybe the inspiration for SL is older than the famous novel "snow crash", from neal stephenson.