Spaceship Identity

I'm reading a fascinating book by string theorist and astrophysicist Brian Greene called The Fabric of the Cosmos at the moment. It's all rather mind bending, but this passage put a wry smile on my face.

String theorists can be likened to a primitive tribe excavating a buried spacecraft onto which they've stumbled. By tinkering and fiddling, the tribe would slowly establish aspects of the spacecraft's operation, and this would nurture a sense that all the buttons and toggles work together in a coordinated and unified manner. A similar feeling prevails among string theorists. Results found over many years of research are dovetailing and converging. This has instilled a growing confidence among researchers that string theory is closing in on one powerful, coherent framework—which has yet to be unearthed fully, but ultimately will expose nature's inner workings with unsurpassed clarity and comprehensiveness.

Explorers of digital identity can take encouragement, then, that we're not the only ones poking around with sticks and flints by flickering candlelight. ; )

Authentication through triangulation

Kim Cameron and Phil Windley write about their intriguing conversation regarding "identity context and transfer of trust".

They discuss how context-specific identifiers such as a passport or a driving license are used as de facto identifiers in contexts beyond their officially-intended remit, often as a way of "bootstrapping" a new trust relationship that is thereafter independent of the original identifier. Phil writes:

Identity credentials have contexts. When I was talking to Kim Cameron, he used the example of a Government issued passport and coffee club card. The context for the passport is a border crossing. The context for the coffee club card is buying coffee. Identity credentials are often used out of context. Sometimes, out of context use doesn’t make sense—think of presenting the coffee club card during a border crossing.

Other times, however, it’s a critical part of establishing a relationship or transferring trust. As an example, you might use a credit card to pay for your purchase at the coffee shop and be asked to present some kind of identity credential. In that case, using your passport at the coffee shop would be out of context, but you’d be doing so to transfer the trust that the government has that you are a particular person to the coffee shop cashier.

Interesting indeed, and redolent of an notion for an item on my blogging "to do" list. In abstract terms (my favourite kind):

Two entities may establish an independent authenticated relationship through an already mutually-authenticated (or "trusted") third entity. This process of authentication "triangulation" allows for the modular construction of authentication relationships in an identity meta-network. The triangulation process could be utilised either:

(1) As a one-off introduction and assertion of identity attributes, or

(2) As an ongoing means of authentication.

(1) is illustrated nicely by the passport and coffee club card example above. (2) could be achieved, for example, by Service A, of a pair of mutually-authenticated online Services A and B, issuing a user with a token that asserts the user's authenticated status and (selected) identity attributes; the user could then present this token to Service B, thus establishing their possession of all or some of the identity attributes required by Service B.

The key point with authentication through triangulation is that is facilitates the bottom-up emergence of an identity meta-network.

Topic: identity

Pottering around in the Digital identity garden

I've been doing a little gardening in the Wikipedia Digital identity article—pruning and hoeing a touch here, watering and fertilizing there. I also dug over a new seed bed for the cultivating of ideas on Identifiers... seeds and t.l.c. anyone? : )

P.T. Ong's taxonomy of digital identity

P.T. Ong has written an interesting and informative piece in response to my discussion with Kim Cameron about ontologies and digital identity. In his piece, Ong maps out a provisional taxonomy of digital identity. It makes a lot of sense to me in the main, with just a few points I feel could benefit from further discussion.

I believe that the best design will include only the minimum set of concepts which is needed to model reality.

I couldn't agree more. And the simpler the concepts, the more potentially powerful the applications.

An entity is a real-world object (person, legal entity, device)

I additionally included digital objects in my definition of entities for my draft of the Wikipedia entry for Digital identity. Let us consider the rationale for such an inclusion. Starting with a counter-argument, we might suppose that the fact that a digital object (i.e. data set) can be exactly cloned would imply that it is less quintessentially unique—and hence less an "entity"—than the collections of molecules that constitute, say, a particular human being or the flower vase on my coffee table.

Yet imagine, for a moment, we were to clone Wikipedia in its entirety (application and data) at another web address—would it be an identical object as the original? No. Naturally, all the incoming links would be broken, and our Trikipedia's PageRank would be zero. But the difference from Wikipedia could be divined even without reference to entities beyond Trikipedia itself: from the moment of cloning, Trikipedia would not respond to input as Wikipedia, because the community of people participating in editing and regulating Wikipedia would be absent. An off-topic page edit by an unsuspecting "outsider" (who didn't know about Wikipedia) would not garner the same response from our inert Trikipedia as they would from the bustling Wikipedia, where people keep close watch on pages for inappropriate changes and reverse them promptly.

Wikpedia, a digital object, functions as an independent organism, and thus can be considered an entity. This independent organismic functionality can be understood as a key attribute for an entity.

Once again, though, we are thrown back on our subjective, human perceptions in this regard, as organismic functionality is hardly an objectively definable property. Just when does an embryo become an organism independent from its/his mother? And my flower vase is only an entity in my eyes because of my emotional attachment to its form and function—otherwise it would end up in the recycling bin with the empty bottles ready for melt-down and recycling into a new form. Even our perception of each other as categorically-separate individual human beings is merely a contextually (very!) useful contrivance, when examined from the point of view of particle physics (or indeed psychic phenomena). The status of being an entity—along with the nature of that entity's identity, as I wrote in a previous piece—can be understood then as a wholly subjective attribute.

An identity is a set of claims made by one entity about itself or another entity (to paraphrase Cameron). Identities are owned by their entities.

Firstly, I feel that sets of claims are the tip of the identity iceberg (see Kim's blog sidebar for illustration!) that is currently practical to work with. Our digital identities in the broadest sense surely embrace all data that are attributable to us?

Secondly, as I indicated in saying that identity is a subjective property, I believe that far from being an objective data set (such as a set of claims), identity itself emerges subjectively from a relationship between two entities. Sets of claims about an entity's identity are... sets of claims! Perceptions of identity, as all human perceptions, arise when information is processed cognitively, and that cognitive processing inevitably takes place in the context of our beliefs and emotions.

This holds true for both our perceptions of others' identity and perceptions of our own identity. It's relatively easy for us to accept that we can never objectively know another entity's identity, but perhaps harder to accept the same notion about our self-knowledge. Yet a great proportion of our minds' processes take place on an unconscious level—so in this sense our conscious selves exist in a subjective relationship with ourselves as a whole, just as we exist in subjective relationship to others!

To return to the definition, then, perhaps we could say something like: "Aspects of the identity of physical entities may be expressed in terms of a digital entity or entities (which for practical reasons may be limited to sets of claims about the physical entity). A subject's perception of an the identity of an entity (physical or digital) arises from the subject's relationship to the entity."

I use "aspects" because it seems clear that a digital representation of a physical entity will in most cases be partial. Note that our Wikipedia example still works expressed in these terms, as it is the communal identity, as perceived by a user or "subject", of the physical entities—the community of humans—that participate in Wikipedia that endow it with its entity status.

topic: identity

Meeting minds

So Meetup.com have started charging group owners a monthly fee, much to the dismay of its unsuspecting users. Metroblogging, a group blog network of which I'm an author (on the London site), is cancelling plans to collaborate with Meetup forthwith.

How about actually talking with and listening to your community, guys? When people start to build a web service into the fabric of their lives, their identity mingles with that of the service—and vice-versa. It's hardly surprising users feel angry when treated as mere units in a cashflow spreadsheet.

My one experience of dealing with Meetup.com is sending a request for a "#joiito IRC" topic (within Meetup's centrally-administered and hierachical topic directory—no tagging here!) for all the people who hang out on the #joiito freenode IRC channel. They never did reply...

Wikimedia to adopt Drupal's SSO protocol?

Could Single Sign On across web services spread in pragmatic patchworking initiatives like this?

Of course, privacy and security issues will present challenges—but it's an interesting trend to watch, nonetheless.

Via Identity Commons mailing list.

Community Fabric

I’ve just set up a wiki for a new open-source project called Community Fabric with some friends (from the joiito freenode IRC channel). The idea is to allow people to harness existing social net tools to blog, bookmark, collaborate and post photos from within an easy-to-use and integrated community environment.

I’m really hopeful that this software could empower people with little knowledge of computers to get published and engage in conversations and online community, as well as making life a lot simpler for those of us crazy enough to play around at the bleeding edge of the social net. But to make it happen, we need techies with the ability and drive to transform the vision into reality!

So if that could be you, leave me a comment, come on into the wiki, or accost me (weaverluke) in the joiito IRC channel, and let’s talk. : )

Ontological distance within the identity net

I've recently exchanged emails and blog posts with the redoubtable Kim Cameron about the concept of embodiment in digital identity. It seems to me that in order to get a clear view of this topic, and indeed of digital identity as a whole, it will be very helpful to be able to visualise the geometry of the "identity net" within which identity might be embodied.

I outlined in a previous post how a geometric representation of a network where all nodes are potentially equidistant (in terms of network logic) requires n-1 spacial dimensions (where n is the total number of nodes). Of course, that's impossible to actually visualise in our 3D brains, but it's useful to remind ourselves that 2D and even 3D representations of many-dimensional networks inevitably distort the true geometric relationships between nodes. Crucially, once we eliminate the arbitrary variation in distance between nodes introduced in lower dimensional network mappings (by mapping the network in n-1 dimensions), we are free to re-introduce distance as an expression of an intrinsic property of the network itself. But what could that property be?

From a human user's perspective, the identity net primarily mirrors our psychological, rather than physical experience of the offline world: it transcends physical and geographical limitations, as do our minds, and the data schemas it employs to represent our identity are, ideally, digital adaptations of our personal belief systems or ontologies and beyond that those of our communities. In the sense that to perceive ourselves as unique individuals, I must perceive you as not me, and vice-versa, you and I inevitably have different ontologies. And, according to our differing experiences in life, we will invariably have developed many other ontological differences such as "my friends", "my family", "my self-image", "my moral values", "the one true religion" (optional!) and so on. Given that we recreate our personal and communal ontologies within the identity net, and that those ontologies are inevitably mutually divergent, it seems to me that distance in the identity net can most usefully be used to express ontological distance between the nodes of the network. In simple terms, we can then explore identity as an expression of ontological distance between two entities.

So will we soon have the ability to construct a universal, albeit ever-evolving, map of the identity net? Well, no—for two reasons. Firstly, we currently lack the technical apparatus to interoperate, and hence measure the similarity of or difference between diverse data schemas. Secondly, even supposing we could somehow measure the similarity between diverse schemas, ontological distance appears to be an intrinsically subjective property, governed by the emotional importance a person attaches to any given aspect of any given relationship.

An example: you might meet someone with whom, on the face of it, you have loads in common with—age, location, interests—and yet something just doesn't "click". There is some aspect of the other's identity, for instance a tone of voice that reminds you of someone you dislike, that sets up a negative emotional charge in you, possibly unconsciously, or perhaps consiously, which overrides the positive matches of shared attributes. Conversely, you might feel you have almost nothing in common with certain members of your family, and yet the fact that they are blood relatives and have shared many experiences with you might well lead you to continue to socialise with them (even if you don't always enjoy it!). In each case, your emotional weighting of certain matches or discrepancies between your ontology and that of the other person influences your perception of the relationship. And it seems likely that different kinds of relationships will entail emphases of different aspects of one's ontology—physical attraction is important in assessing a potential sexual partner, for instance, whereas experience and qualifications are primary when choosing an accountant.

I suspect the identity net will function analogously: for a true measure of ontological distance between a person's digital self and another digital entity, that person's subjective feelings must be factored into the equation. A topic for another post, perhaps.

Thirdly, given that we (or more accurately, "entities") will very often only disclose selective aspects of our ontology to another entity, no measure of ontological distance can be sure to take into account the ontology of each entity as a whole—such a measure will always be provisional and dependent on the nature of relationship between ourselves and the other entity.

So let me try to bring this all back to the contemporary digital identity debate. What are the implications for contemporarily-practical applications of digital identity? Kim Cameron's focus on identity assertions as a do-able goal is, I feel, just right. Microformats—small data-structures for a specific purpose, such as expressing identity attributes, for example—seem to be the most effective way right now to enable two entities to inter-operate their ontologies within a specific remit, agreeing on the smallest-practical patch of common ontological ground for the purpose. We can presumably anticipate a Darwinian contest between rival identity microformats leading to the evolution of a few or many common microformats for digital identity. As for context-dependence for the disclosure of identity attributes, I understand this as being absolutely in tune with the notion, expressed above, of identity as a wholly subjective property that arises from within relationship.

But then again—let's not mistake for the moon the finger that points at it. It seems to me that digital identity is a wholly more complex and wonderous thing than we have yet been able to fathom, and I for one look forward to the discoveries and insights that lie waiting for us around the virtual corner...

Topic: identity

Wikipedia discussion on Digital identity

Further to an edit of the Digital identity wikipedia entry and a couple of emails from Jaco Aizenman L. I've kicked off the Discussion page for the article. I warmly encourage all you Identity bloggers to get stuck in! : )