the accountable net notion is peer-to-peer (I'll be accountable to you and you be accountable to me, and that only works with a persistent identity). The first commandment: Do ask, don't lie. You have every right to ask who I am and what I've done, you don't need to answer, but if you do answer, you cannot lie.

Where2.0 conference: all about maps and location and GPS and stuff like that. What the conf was really about was identity: you in the map (avatar in SL, place on GoogleMap) there are tools for representing self on the map. Do we want privacy or do we want attention? Newest tools are about attention, establishing identity. We don't realize the slime trail we're leaving. When we do see it, we may be alarmed.

(The designated bomb-thrower of the panel) Quotes from a novel about a guy who's trying to get on a ship before it sails, talking to the authorities who are demanding identity patpers from him... "I don't need any papers, I know who I am." "Yes, but others want to know as well"

You have to think about identity in the context it is used for. Identity is really an assertion claim to access resources. There are two security strategies to go about testing the assertion: accountability and authorization. Accountability keeps systems functioning better, but don't work well where there are catastrophic outcomes for a mistake. Authorization doesn't scale well.

Oracle is his day job, but he's here representing the Liberty Alliance (http://en.wikipedia.org/wiki/Liberty_Alliance). Founded in 2001, it was a backlash/reaction to MS Passport, but hopes they've moved beyond that now. The three principles or Liberty Alliance:

• there must be user consent (user has to choose, or opt into the federated relationship)

• the user should have a choice of identity provider
• there should be a decentralization of identity repositories

how do we get to convergence on the problem? We don't even know what the problem is to be solved.

• Profile Exchange: folks are looking for profile exchange (user doesn't want to have to retype info every time they join another social network).
• Authentication of identity: Log onto the internet versus log onto each site.
• Authorization: How do I prove I am who I say I am?
• Privacy management: How do you prove you've gotten to use information in a certain way.
• Recourse: What are the responsibilities of the user and how do you take away their rights if they misbehave?

Need to be able to go and manage your identity in the way you do in the real world, via different personas (me as dad, me as co-worker, me as consumer, etc)

What is user-centric? It's a consistent user experience for the user. The user picks the tool to manage their identity. If you put the user in the middle, it can scale. If you don't have the user in the middle, you've got to educate whenever any changes are made (and leave you wide open to phishing).

sees two very big challenges at this time. Was somewhat optimistic meeting them 10 years ago, but is less opt today.

1. The security compromise challenge: have to think seriously about what happens if things don't work, if your system breaks.
2. The "collapse" of the lawful access premise: It used to be we felt comfortable that there was accountability when accessing citizen's data. As recent events have shown, there's much behind-the-scenes access happening.

The American public's attention to privacy matters has been on the increase since Sept 11, especially as their privacy has been invaded more and more on behalf of "making their country safer."

Instead of the VA suggesting that the veterans (whose data was compromised by the recent laptop theft) go out and get their "free credit report" to watch to see if anything untoward has happened, the VA should be required to pay for the monitoring service to proactively watch these veterans' records. Why's the cost not borne by the agency responsible?

She's looking at human behavior and interaction through signal theory. Her presentation is "Fashion risk and identity: social signals in the real and mediated world"

Since so many things indicating what we want are "hidden" we have to rely on signals. This is a competitive landscape (competing to be what we want), so we rely on signals that can be deceptive.

Basics of signaling theory:

• competition motivates deception (signals aren't as important in a cooperative environment)
• wasteful costs ensure reliability (if there are enough resources to actually be able to waste them, the signals are more reliable)
• punishment also helps ensure the reliability of signal (impersonators are prosecuted)

In the animal world, "costly signals" include big horns and antlers in that it takes a lot of metabolism to hold up these racks. Moose can judge whether or not to engage in a fight for territory or mate or food by how big the horns are on the competition.

In designing new "vocabulary" for social spaces, we need to be sure to be aware of both intended and unintended meanings.

Hypothesis: Fashion is a signal of information access and adaptability. (remember: fashion is often considered a wasteful activity)

Blogs are an example of using link structure to see who has the fastest access to data on the web.

Signals of daring: suntanning and smoking. The fact that it's bad for you is what makes it attractive. If we want to change people's behavior, we need to look at how we message safe sex, anti-smoking, etc. This may help explain why so many young people are posting "risky" content on MySpace (pics of themselves naked, lists of drugs)... that's what's sending the fashion signal.

How do people manage social risk?

It's a matter of personal and impersonal exchange. There are parallel processes involved in how we work through things:

• organizations -> social actor-critic (who you know)-> optimal control
• institutions-> selfish actor-critic (what you know)-> optimal use

If you can make it in people's interest to cooperate, they'll usually figure it out and cooperate.

McCabe's Two Open Questions:

1. Can we design a joint decision making scenario where different people would have different criteria for what it means to cooperate?
2. Should we allow groups to define what measures go into constructing a reputation score, or only use group affiliation to provide the cue?

Things to keep in mind about virtual worlds: they are populated by citizens from around the real world. They're in the position of creating economical and social environments that cross international and geographic borders. The bad part: it's really common in these worlds for the participants to have ambiguous and anonymous and even multiple identities. (Second Lifers sometimes have as many as six different characters).

We want trust because it lowers the transaction cost of taking risk.

Some of the questions Linden Lab is struggling with: what happens when real-world governments want to engage with the virtual world of Second Life?

If you can have a tradeable preference, you can get supply and demand. Credit cards on the global system are the best identification system ever made. Baby boomers are soon going to start spending their wealth on their health. In our 11 Trillion dollar market, there's 1 Trillion dollars spent on creating demand.

The best way to set up an auction is to set it up so that the highest bidder actually pays the price submitted by the second-highest bidder plus a penny. Google adopted this method and their revenues sky-rocketed. Google knows that if they can hot-wire that 1 trillion dollars to a better back end, they've got it made.

Every single company that has to get its product to market is having to deal with a completely new way of delivering the signal in this $1T market.

Vertical market: consumer is constrained by the corporation creating the market. The new technologies break down the walls of the walled garden. A similar hypothesis made in 1938 about television about how liberating television would be from the tyranny of radio. We all know how well that played out.

There are a couple ways of looking at the evolution of the walled garden. If we give people the tools, will they take the initiative to control their own futures, or will they look for others to "rule the garden"?

As a juice-maker, started Odwalla back in 1980. When the natural food business in the early 1970s, it was a $3 Billion business (self-organized co-ops, back-to-the-land groups, etc). today is a $75 billion business. In the last month, Walmart announced that they want to include organics in their stores. On the right you have Walmart, on the left you have Whole Foods and in the middle you have the other grocers (Safeway, Giant, etc). The middle is disappearing, as they have to take sides on which approach they want to take.

What's difficult as enterprise to think of is "what do we do with this new paradigm of trust?" Shared a corporate-based proposal to be able to mine the data in customer's shopping baskets based on what's been purchased using their frequent-shopping card. The customers likely wouldn't be comfortable to know that the corporation is doing such targeted datamining based on their use of the affinity card.

My Virtual Model's (MVM's) mission is the be the standard in virtual identity by being the leader in high quality 3D visualization and innovating Personalized Shopping experiences. Users pushed the idea further to get them to go B2B, directly to the source. With user demand, the needs are being addressed. With enough users and enough critical mass, easily identifying the need will follow. Thanks to MVM's approach to proactively sharing data, lots of opportunities are opening up to them. One example is that their 3D products can neatly port into games for product placement. There's no need to convert from 2D to 3D, they've already done it, so it's easy to provide the content to games, or even to online visual community members (members of Second Life, for instance) to use to create their own detailed environments.

MVM shows what happens when you incorporate the customer early on in the design and follow them rather than try to lead only.

At the event as a software vendor and as a supporter of Higgins . He sees that you need to apply the identity technology to the area with the greatest risk (bank info, etc) to the trust. Is excited about the opportunity for Higgins. Sees the infrastructure and the plumbing as being a place where consumers and corporations can interplay, but Higgins is ultimately about the consumer having the choice of where to store their info. The infrastructure folks will likely not be the ones who also store the info.

Is IBM leaving money on the table by focusing on the infrastructure instead of on the storing of data? No, because that's not the brand identity of IBM to provide the technology, not to provide the secured data intermediary services. That would be alienating their customers (the banks) and getting IBM into a space they don't have strength in.