Resolution

From ITUwiki

Jump to: navigation, search

What to what?

  • Identity to identity
Private to public
Asserted to Alias
Alias to public
  • Identity to location
Service to network

Resolution is the process in which an identifier is the input — a request — to a network service to receive in return a specific output of one or more pieces of current information (state data) related to the identified entity. Resolution to e.g., a location (URL) is one instance of this. Resolution to a "related identity" is another instance. General agreement that we should be looking at the higher level of abstraction (any current state data) rather than the lower (location, aliases) since solving the higher allows all the lower functions.

If resolution is to a related entity should this be simple freeform relation -which could be useful for some apps (as in a URL, or the simple use of freeform "Desc" in a Handle record); or declared in a structured manner (at its most structured, a consistent and robust definition from a contextual ontology). Digital identity attributes —or data —exist within the context of ontologies: http://en.wikipedia.org/wiki/Digital_identity

Has to allow recursion - resolution might be nested, or require several levels of indirection. Has to be contextual (who can ask may determine what is received). Contextual resolution need not necessarily mean giving control of resolution to the requestor: e.g. there may be a default resolution record controlled by A, which B has the right to modify only if B has appropriate credentials. Example quoted of Handle contextual resolution in the Defense Virtual Architecture : http://www.handle.net/apps.html,

Resolution space

What is identity? "Identity" defined by a set of identifiers (where identifiers are attributes)- unique key and other qualifying attributes. Which attributes are relevant to defining identity depends on context: what may be necessary to be distinguished in one context may not be necessary in another context. Example discussed: ISBN treats all copies of one edition of a book as identical, but if I buy a copy of the book I don't buy the whole print run. This is also an example of the distinction between qualitative and numerical identity (http://en.wikipedia.org/wiki/Numerical_identity#Qualitative_versus_numerical_identity )

The distinction between "class identifiers" and "instance identifiers" is arbitrary: any instance identifier can be made more granular and so becomes a class identifier decomposed into several instance identifers. Example: "Fred" may for some purposes need to be distinguished into "Fred on Monday" and "Fred on Friday"; for other purposes, "Fred" suffices. (Functional Granularity: it should be possible to identify an entity whenever it needs to be distinguished.)

Identity may belong to a profile; a profile is an abstraction, and any abstractions may need to be identified (if there is a functional requirement).

Where is “identity”? Taken to mean: who controls the assignment of the set of attributes to an entity? Since the attributes returned from resolution mya be influenced by context, we need to consider appropriate resolution of an identifier to yield some identity attribute. (e.g. - Location context, Geographic or Network). The higher level of abstraction (managing any form of attribute) is to be preferred to the lower (location, geographical definition) since solving the higher allows all the lower functions. There are examples on the web of contextual resolution through work-arounds of DNS (e.g. OpenURL: http://www.crossref.org/03libraries/16openurl.html, to solve the "appropriate copy problem").

  • Protocols
DNS
Handle - a distinct protocol, running on top of internet protocol only - does not use DNS (http://www.handle.net/overviews/dns.html)
[note] ENUM - not a protocol but an application of DNS (putting E.164 into DNS)
Personal tools