Understanding Data

First off, the trouble with Data is that it's ultimately nebulous. URLs (Access Addresses or Locators) and URIs (Names/Handles) are properties of Data..

What is Data?
How we express observation via collections of attribute (properties or predicates) & value (object) pairs that coalesce around an observation subject (entity) identified by name using a variety of naming/identification mechanisms called identifiers.

What is Structured Data?
Data representation constrained by a schema where clear rules are in place for representing (modeling) subject (entity), attributes, and attribute values.

What is a Document?
A timeless resource that serves as the canvas (surface), container, disseminator of data. It is endowed with attributes that cover "whole data representation" (i.e, content) and provenance (who, what, when, and where) in the form of document metadata.

What is a Web Document?
Typically, a document where the content loosely 'mentions' or makes 'reference' to many things. But, when you factor in Linked Data, you end up with a more specific kind of document with a very specific purpose i.e., description of an unambiguously named subject using an HTTP URI (typically).

The effect of the above leads to the following hierarchy:

1. Generic Representation -- typical Web pages or documents
2. Description oriented Representation -- Linked Data and RDF documents
3. Definition oriented Representation -- Linked Data and RDF documents that describe Properties and Classes that ultimately facilitate fine-grained semantic relations.

In all cases, you have documents bearing content. Actual content representation transmitted to clients (user agents) is always negotiable.

What is a Link?
A reference to data.

What is a Hyperlink?
An electric medium reference to data .

What is a URL?
A data access protocol abstraction that enables hyperlinks function as network addresses or network resource locators. Basically, data access oriented identifiers.

What is an URI?
An abstraction that enables unique entity identification via hyperlinks.

What is a de-referencable URI
A URI that resolves to data, directly (as in the case of URLs) or indirectly when used to provide indirection oriented abstraction that distinguishes a data object from its representation (or values). Basically, you end up with the ability to make explicit statements about data objects which is basically the fundamental basis of object theory.

What is Linked Data or Hyperdata?
The use of hyperlinks to enhance structured data representation in a manner that leverages the expanse and ubiquity of networks. Examples would include the World Wide Web and/or your firewall protected private network.

I've attached a series of diagrams to this post that put the above into context re. Linked Data and the dimension of the Web it seeks to expose. The very first diagram is TimBL's original WWW proposal diagram (note the describes links). The others are variants that I've drawn up to aid (hopefully) understanding of the use of URIs to enhance InterWeb scale structured data representation and publication.


1. http://t.co/3hOTNcoF -- understanding data objects
2. http://www.w3.org/DesignIssues/LinkedData.html -- original Linked Data meme
3. http://goo.gl/ovqPL -- from Open Database Connectivity (ODBC) to Open Data Connectivity
4. http://www.slideshare.net/kidehen/understanding-linked-data-via-eav-model-based-structured-descriptions -- understanding how to describe things using entity-attribute-value model based 3-tuples or triples
5. http://youtu.be/ZaJPNrf1DPY -- "there is no spoon" clip from the Matrix.

#Web #LinkedData #SemanticWeb #BigData #SmartData #BroadData #Data #Hyperdata #Hypertext #URI #URL #HttpRange14 #ODBC #JDBC
4 Photos - View album
Shared publiclyView activity