Identity, Reference, and the Web

Harry Halpin, <>

Two Black Holes

Introductory Comments, Edinburgh, May 2006


We will be audio-recording some of the talks for pod-casting, and will post an RSS feed of the talks at our web-page:

The papers are also on the Web in HTML - the way conferences should be :

There may be an upcoming journal to anthologize some of these talks and papers - so give feedback!

Pat Hayes will be giving his talk over live audio, and sends his apologies.

We will have a discussion at the end of the workshop to find consensus (if any) on these issues. We'll then write final notes up to post on the web-page and possibly to the W3C

Lastly - Book Dinner will be at Verandah


TimeAuthorTitle of Paper
11:30-12:00 Harry Halpin (Introduction) Identity, Reference, and Meaning on the Web
12:00-12:30Dan ConnollyA Pragmatic Theory of Reference for the Web
12:30-1:00A. Gangemi and V. PresuttiThe bourne identity of a web resource
2:00-2:30Allen GinsbergThe Big Schema of Things
2:30-3:00David BoothURIs and the Myth of Identity
3:30-4:00Coffee Break
4:00-4:30B. Parsia and P.F. Patel-SchneiderMeaning and the Semantic Web
4:30-5:00Pat Hayes (via live audio)In Defense of Ambiguity
5:00-5:30Steve PepperThe Case for Published Subjects
5:30-5:45H. Choi, S. Kruk. S. Grzonkowski., K. Stankleicz, B. Davis, and J. BreslinTrust Models for Community-Aware Identity Management
5:45-6:00John BlackCreating a Common Ground for URI Meaning Using Socially Constructed Web Sites
6:00-6:30Henry Thompson (Chairing) Discussion: Where Next?

A Black Hole?

Questions of identity, reference, and meaning are thought of as a "black hole" that has long puzzled philosophers and well as causing never-ending discussions on W3C list-servs.

Can we shrug these questions off?

Black Hole as Metaphor for Listserv Philosophy

The Semantic Web is not a separate Web but an extension of the current one, in which information is given well-defined meaning, better enabling computers and people to work in cooperation (Berners-Lee, Hender, and Lassila, 2001).

Is the killer term meaning or people? Or the combination of the two? How can meaning transfer from computers to people and back again?

What do We Mean by Meaning?

Meaning is a thorny word:
  1. Machine-readable
  2. Has a relation to a formal model?
  3. Easily understood by humans?
  4. Somehow connected to the world in a robust manner?
  5. and more...
  6. How are questions of identity, reference, and meaning mixed together, and can they be teased apart? What luck can we have with the Web if these questions are so difficult with the wide world outside the web?

    We are not analyzing a world, we are building it. We are not experimental philosophers, we are philosophical engineers. (Tim Berners-Lee)

Broad Overview

By building on the best of past work and applying it to the context of the Web progress is more likely.

Our approach will be to look at the problems and the debates under the rubric of looking at some past theories of how identity, reference, and meaning operate.

  1. The Identity Crisis
  2. Russell and the Direct Theory of Reference
  3. Frege and Interpretation
  4. The Solution of the TAG
  5. Social Meaning and Tagging
  6. Towards a Practical Solution...

The Identity Crisis of URIs

The first step in the creation of the Semantic Web was to extend the use of a URI (Uniform Resource Identifier) to identify not just web pages, but anything

Some find this a bit weird, but if the Web is a universal information space, it seems literally everything should have a URI and be disambiguated.

Instead of just visiting Tim Berners-Lee's web page,you could use the Semantic Web to make statements about Tim himself.
Means that Tim Berners-Lee has a phone number.
Means that his home-page has a phone number! Which is nonsense.

Ambiguity and URIs

Things can get genuinely ambiguous because most things don't have URIs, and it's hard to tell what a URI is supposed to mean without further information (Ginsberg).

Seems to mean the Eiffel Tower itself was built by Gustave Eiffel.

While the second seems to mean a web page about the Eiffel Tower was made by Artifice Inc. This problem of what URIs identify is called the Identity Crisis.


URLs (Uniform Resource Locators) gave you a location of some information that you could access via a scheme.

Unlike names in natural language, a URL implied under normal conditions possession of the thing.

URNs (Uniform Resource Names) explicitly forbid any sort of access mechanism...but this hasn't been adopted.

Larry Masinter and others have tried to resurrect the distinction with schemes like tdb (Thing Denoted By).

What should a URI for thing not on the Web return? A Published Subject (Pepper)?

Redirection (Miles) allows a URI for a thing to be redirected to a URI that hosts representations, while maintaining distinct URIs.

Or, Wittgenstein-like, are URIs determined only by their use? (Booth)

URIs Identify One Thing

TimBL has stated that URIs identify one thing in a global context.


This thing is a resource, which may or may not return representations over schemes.

Who decides what the thing is a URI identifies?

TimBL seems to believe its the owner, while Masinter believes in the power of readers over writers

Leibniz's Law of Identity: If two objects have all their properties in common, then they are identical and so only one object.

Can we use this to define identity on the Web?

The Direct Theory of Reference

There is a historical parallel to TimBL's vision
Russell's Direct Theory of Reference

. Bertrand Russell

Patches of sense-data known through direct acquaintance allow one to ground the atoms of logical statements or create descriptions that can form the basis for names.

The priority of reference over meaning

We are also assuming the real world (as opposed to any logical idealization of it) is divided neatly into individuals (Strawson).

But how can we deal with referents in the past, or imaginary referents?

Kripke and the Causal Theory of Names

A name can be transmitted through time, with a name being given its original referent through a process Kripke calls baptism. The name is a historical process, not a description.

See the Web Proper Names paper for a detailed exposition of this.

Saul Kripke

Example:The identity of a URI is established by fiat by the owner, and then communicated to others in a causal chain in the form of Semantic Web metadata and maybe publishing documents at the URI (or a redirection thereof).

Yet how can things truly be communicated without ambiguity (Hayes)? What type of common ground (Black) would it take for machines to communicate, or people, or both?

Frege and Meaning

Frege, the father of logic, posits that the actual thing in the world is the referent, and a name is a symbol that identifies a referent(s). The sense is the mode of presentation, a type of public, objective(?), knowledge about that private concept among a shared community. The third party of sense (meaning) mediates the reference relationship.Gottlieb Frege

The priority of meaning over reference

Example:Hesperus has a sense ("the morning star") different from that of Phosphorus ("the evening star"), yet both have the same referent, the planet Venus.

As formalized in logic, meaning is taken to be given by interpretations that define the truth conditions of the statement.

Ambiguity is built in: The web of logical statements is the bearer of meaning, and whatever satisfies the sense of the statement could be a referent.


interpretation example

Syntax and Semantics Example: The syntax of the numeral 2 is about the number two. The interpretation of numerals to numbers is arbitrary. 11 is three in binary and eleven in decimal notation.
Interpretation in OWL Example OWL interpretation example

Thanks to Ian Horrocks for the slide..and hear more by Parsia and Patel-Schneider.

Information, Resources, and Representations

Resoource and Representation

The W3C TAG (under httpRange-14) took on: What is the range of the HTTP dereference function?

Information resources are defined as anything whose essential characteristics can be conveyed in a message where these characteristics are encoded, with varying degrees of fidelity, into a sequence of bits.

"Cars and dogs" are non-information resources because their essence is not information

Picture and Text from TAG's Architecture of the Web.

The TAG's solution

Resource and Representation If a http resource responds to a GET request (Connolly)...

1. With a 2xx response, then the resource identified by that URI is an information resource

2. With a 303 (See Other) response, then the resource identified by that URI could be any resource;

3. With a 4xx (error) response, then the nature of the resource is unknown.

Can not distinguish between the case of an information resource being redirected to another information resource and a non-information resource being redirected (Gangemi and Presutti).

Social Meaning and Tagging

XML Schemas, Ontologies, and other languages are based in social meaning.

David Lewis According Lewis, a language is defined formally but established and used according to convention. A convention is used by a community to solve some co-ordination problem, even if such a choice is arbitrary.

Example: the American versus European method of writing dates - 5/23/2006 versus 23/5/2006.

Out of a game-theoretic analysis is that people will in general use the minimum amount of convention to solve their co-ordination problem.

The uptake of tagging over Semantic Web formal semantics in the "Web 2.0" is a point-in-case.

Beyond Tagging

An overlap of tags is given as a sign of similarity or identity, and two pieces of data that share the majority of the same tags are thought to identify the same thing.

Tagging already does not scale well nor is tagging capable of expressing complex relationships even on the scale of RDF.

It's difficult to share tagged data from proprietary Web 2.0 services.

Hard to tell what a single natural language term means, might need contexts and namespaces (URIs) - and even formal semantics - to share properly.

So maybe the Semantic Web is what the Web 2.0 needs?

But how to lower the cost?

The Future: Practical Solutions

Will the Identity Crisis just go away?:

While in 1996 this may have seemed unlikely, with the development of RFID tags for everyday objects, the gulf between a thing and its representation on the Web may not be a gulf at all. Whether one makes an incorrect triple about you on the Semantic Web or manufactures a false biography on Wikipedia, representations on the Web are now part and parcel of reality. Even by just minting a URI for a thing, you have in a very real sense made that thing on the Web.

One way forward

It seems some combination of human-readable documentation that clarifies the social meaning, Semantic Web metadata, and a (RDDL-like or RDF) format for distinguishing between non-information and information resources could be the way forward.

Now on with more detailed talks on the problem...