Monday, August 22, 2005

Linden Lab: All-Knowing?

"He's making a list,
And checking it twice;
Gonna find out who's naughty and nice.
Santa Claus is coming to town

He sees you when you're sleeping
He knows when you're awake
He knows if you've been bad or good
So be good for goodness sake!"

-- Santa Claus Is Coming To Town

During a casual conversation in Second Life recently, a friend told me something which strikes me as, well, a pretty big deal.

I consider this person to be a source you can rely on. Their claim is something that I have never actually heard another resident suggest. If they're right, the implications seem to change the face of what law, privacy and government mean in Second Life. Will privacy advocates wring their hands in mass hysteria and reach for their tinfoil? Will griefers and criminals across the grids tremble in fear?

The claim is this:

Every word you speak in Second Life via chat or instant messages, is logged by Linden Lab. Not 30-minute logs. Not day logs. Logs that go back at least as far as last year. Possibly further.

Now, it may or may not be true. Although it's incredible, it does seem logistically possible.

So I posted a Hotline to Linden question (note: the resident I mentioned in the question is not the friend who made the logging claim). Robin Linden replied: 'Logs for chat and IM aren't permanent, although I can't say how long we keep them'.

You can't be certain whether Robin physically doesn't know how long they keep them or whether their policy prevents her from revealing it. I think it's safe to assume the latter. Policy-wise, there's obvious reasons why they won't be specific.

Less firmly, my friend also suggests that Linden Lab may log even more detailed information than just chat. Possibly everything from object rezzes to gesture triggers. The works.

To my mind, the first claim seems plausible but the second beggars belief. In any case, it presents an interesting opportunity to think about what these could mean for our second lives and the future of the metaverse itself.


Section 8.2 of the Second Life Terms of Service:

"You acknowledge and agree that Linden , in its sole discretion, may track, record, observe or follow any and all of your interactions within the Service."

Whether we're pottering along at Tringo, slinging arguments at a Thinker's discussion or trying to find a pair of sunglasses that will go well with our dick, we're always vaguely aware of that privacy clause. We know that if we were somehow to stray into CS/TOS-violation territory, that a Linden might materialize behind us with their x-ray vision or that an employee at their desk in San Francisco might receive an IM and invisibly bring up our account details.

In the privacy debate, the naysayers argue, "If you're not doing anything wrong, you have nothing to worry about". But I wonder if other residents would be at least vaguely disconcerted at the thought that every single utterance they have made in the past 8 months for example, is nestled away on a storage medium somewhere, in an office, like a fly frozen in amber? All trivial fond records, an indelicate joke about nuns, a whispered aside about how god-awful your best friend's new dress is. But the serious stuff too: Personal confessions, double-lives, revelations at an in-world alcoholics anonymous meeting, a RL affair, a tearful dislosure of childhood sexual abuse, embarassing medical problems, intimate exchanges, shared Real Life phone numbers, work and financial details. All frozen in time.

Is SL any different than the rest of cyberspace in this regard? How so?

Whatever the answer, as a hypothetical it's interesting to ask yourself if you would be willing to sacrifice your privacy if it meant that every act of harassment, every Terms of Service violation, every act of fraud and real-world crime could be traced and examined with almost God-like precision. Griefers, copyright-thiefs, con-artists, paedophiles - the whole spectrum under complete scrutiny. Would you make this sacrifice in Second Life? Are you already making it?

Crimes and Evidence-Gathering

Suppose that my friend's second claim is true - that everything from object rezzes to gesture triggers are logged. Imagine that Linden Lab were to investigate the virtual-world equivalent of the Kennedy Assassination. In their God-mode recreation of the event, they can determine the timings, they can tell you what the cloud coverage was like on that day, they can tell you where the sun was, the wind velocity and direction, the ground height and slope. They can tell you the exact color of the bullet, and can give you a complete account of the physics that took place - velocity, mass, gravity, energy, the list goes on. And that's just a tiny sample of all the relevant information. They could close the case on the chat-logs alone.

That's just a hypothetical example. There's a slew of real ones available. The really virtual and the really real. Everything from allegations of in-world Nazi death-camp re-creations, to TSO's virtual child prostitutes and RL physical abuse, virtual child porn in SL, the recent client-side hack and 2 alleged land-ownership group-dispute/scams (1 and 2).

With its limited knowledge, the public (myself included) has never shied away from making its own rulings on these controversies. But would it change things if we were to learn that Linden Lab was infinitely better-informed about abuse and TOS violations than we had previously thought? That perhaps they have ruled on a hundred decisions already, the ones above for example, with 20/20 hindsight, never once revealing the extent of their evidence? Would residents actually welcome our new elephant-memory overlords?

But what about the blindspots? Private emails and forums, IRC and instant messenger sessions. Are these significant enough to be worried about? To what extent is griefing possible outside of the Second Life client itself? And if so, is it even a legitimate area for LL concern? It seems apparent that if extensive in-world logs do exist, the Lindens would not trawl through them to investigate every single abuse-report. The limited time available, and the sheer volume of information, would make this prohibitive. But will the knowledge that it is technically possible, strike fear into the hearts of griefers and criminals in Second Life? If so, will they shift their tongue-forkery off the grid and into the spaces where LL has no jurisdiction - instant-messenger programs and IRC? And will they meet there the paranoid, refugee champions of privacy and free speech? Or just all the residents taking advantage of SL-denied communication features like temporary-group chats and stealth settings (offline status)?

Government Law and Order

One of the most hotly-debated topics among Second Life residents has been in-world government. Residents ask whether Linden Lab can accurately be described as 'the government' within Second Life. They discuss whether residents might play a role in a global, in-world government or whether they should they look to ways of forming their own. And they wonder whether these resident-based governments could actually have any clout.

A few of them also wonder about 'resident review panels'.

'Resident review panels' are an aspect of the Second Life legal system that have been announced publicly but are shrouded in secrecy. You can read a bit about them here. In brief, LL chooses 25 active residents at random and sends them the anonymous case-history of an alleged serial offender. The review panel lets each resident vote yes or no on whether the offender should be banned, and provides them with a blank line in which to write a comment. LL reserves the right to overrule the majority decision and panel participants are asked not to share the details of the case with others.

One of the concerns I have often heard from the few people who actually know about these resident review panels, is that the alleged offender has no chance to present a defense. LL's response seems to be that the facts of the violations themselves are not open to question. All they're interested in finding out is whether the residents think that the violations warrant a permanent ban. But some residents clearly see in these review panels the seed of a complete Second Life in-world legal framework - one that involves greater resident participation.

But this new claim regarding Linden Labs' evidence-gathering powers, casts all these law and order questions in a new light. Is a complete legal framework with judges and juries, prosecutors and defense attorneys, any longer necessary (if it was before)? On the other hand, isn't it true that court trials are not always about disputing the facts but disputing the interpretations of those facts, and the relevant laws? If so, how often might these trickier ones actually occur in Second Life?

And in any case, is a global, resident-involved legal-system an example of 'taking the whole metaverse thing too far' or will they be a commonplace feature of the virtual-worlds of 2015?

Robin Linden recently posted:

We believe that police are not the preferred solution to the elimination of bad behavior and intolerance in Second Life. The mechanisms are in place for dealing with people who are infringing on others' rights. Be sure that we are continually refining them and looking to find ways to make them even better, but adding police isn't the means we are considering. First, it isn't scalable, and second, we think that top-down solutions from Linden Lab are not in the best interest of Second Life.

Give us suggestions for ways to improve the abuse management system, or to improve land control tools which will ensure griefing is either unprofitable or no fun.

OK, you have my vote but...

/me edges towards Yahoo Messenger.