Self-Annotating Identifiers       INDEX -- SPECS -- EDITOR -- USES

Mary Poppins-11143435085969103

The above default String is a typical SAID - a SELF-ANNOTATING Identifier which not only uniquely names its intended meaning, but overtly describes it by using a controlled word (red), semantic axioms it obeys (blue), and other required metadata outlined below.

The Shape = Mary Poppins-1

The symbol to be defined, in lexicography, is sometimes called a shape. One generally can assign no unique meaning to a shape lifted out of context, but a well-chosen numeric suffix can be a work-around to that problem.

If a full suffix is known separately (a namespace file), a UI may truncate it to one digit - the relative order of creation among the SAIDs for each defined term. This yields a simpler unique shape for that term's "senses", as in ontology-1 or apple-2.

Assertion Time = 1143435085969

Declaring intended meaning for an identified subject is a deliberate act at a specific time. Recording the time plays an indirect role in SAID semantics, by clarifying the specific version of everything that was formally involved in creating that symbol. Potentially, this can help widely distributed SAIDs evolve smoothly over time when future changes occur in definitions.

On more basic levels, time tags make each SAID nearly unique independent of a namespace. Long numbers (msecs. since 1970) are awkward, however, so they are often seen in sense notation, or as the age of the SAID in hours.

The Axioms = 103

The last 3 characters in our suffix represent basic semantic predicates true of the shape's meaning. The author of the identifier deliberately encoded them by using this web form, and declared its intended meaning to be as described here:

Nature 1

Individual vrs Collective
Specific vrs Indefinite
Imaginary vrs Real

Realm 0

Natural vrs Artificial
Concrete vrs Information
Basic vrs Mediating

Type 3

Composite vrs Characteristic
Persistent vrs Incidental
Discrete vrs Continuous

The Interpretation = the imaginary physical Object

Axioms are okay for digital reasoners, but most people may relate more easily to a phrase-like summary of the merged facet meanings.

These combined approaches should semantically qualify a SAID's shape enough to indicate its intended sense. The referent of that word sense - its conceptual meaning TO ITS PUBLISHER - is what the identifier properly denotes. This is the identified "Subject".

Upper Ontology Facets = Nature, Realm, Type

The Axioms and Interpretation tap a very powerful indexing technique known as faceted classification [1], which is now making resources much easier to find in advanced libraries, web sites and data bases. Instead of one taxonomy (notoriously hard to create "by committee"), this approach classifies concepts simultaneously in several smaller, orthogonal ones called facets.

Each Axiom and Facet assertion can be separately added to range or domain constraints, and used in faceted queries of the problem space. More flexible than "class" in such usages, they often split up team disagreements into more tractible chunks as well. All these effects have led well known authorities on modeling [2], [3], [4] to advocate such techniques in the uppermost layers of ontologies.

Namespace = http://sandbox.net/to/play/in/#

To be used widely, per many tools, a SAID requires a namespace. Above is the one currently in use here. To change it, visit the INDEX page.

A random namespace could work syntactically ([5], [6]), but to be truly a "published subject identifier" (PSID) takes more than URI syntax. Subtle issues of trust, authority, accuracy and stability arise, which many feel the authority [7] of the namespace must resolve. This has led to numerous "URI-registry" plans (e.g., [8], [9]), often complex to support.

Authority Lite = OASIS Recommendations

The nicely crafted alternative standard above has emerged, which this page informally extends. It makes web-wide URI identifiers of any style easy to post as open standards, even by small businesses [10]. Yet it still offers a way to verify that such "published" URIs are backed by the authority of the namespace creator, who professionally endorses them.

Specifically, the namespace for any valid published subject identifier should resolve to decent documentation on the namespace, its publisher, its planned lifespan, and any custom namespace-specific extensions [11]. It not, users are warned by the non-compliance to beware: either the publisher has failed to follow standards, or the PSAIDs are bogus.

[1] Faceted Classification - The Wikipedia, 2006
[2] Top Level Categories - Upper Ontology Facets. Sowa. 1999
[3] Formal Ontology - see esp Individuals, Universals & Collections. Smith. 2006
[4] Simple Bio Upper Ontology - guidance. Rector, Stevens, Rogers. 2006
[5] Namespace Specs - The W3C standard. Bray, Hollander, Layman, 1999
[6] Myths on Namespaces - They ONLY name XML nodes. Bourret, XML.com, 2000
[7] URI syntax - scheme:/authority/path/query/fragment. INet Soc., 2005
[8] DOI Handbook - Digital Object Identifers are persistent, 2005
[9] Electronic Stnd. Book Number - Another software registry - Wikipedia
[10] Lexikos PSIs - Examples for CTM & WORDS constraints. Corwin. 2004
[11] Lexikon Standards - Senses, extra facets, web services. Corwin. 2006



This upper ontology registry can define words, phrases and sentences so that software "understands" what they denote. In June 2011, it won U.S. patent 7,962,328.

SAIDs can add speed and accuracy to many semantic applications - ours or yours. For more about using them, write Dan Corwin