Semantic Blogging

Semantic Blogging <body> <h1 style="position:absolute;left:0;padding-left:1em;font-size:48pt;padding-bottom:3px;width:70%;border-bottom:2px solid black">Semantic Blogging</h1> <h3 style="position:absolute;left:0;padding-top:5em;padding-left:10em">Making the Web more machine-digestable</h3> <h4 style="padding-top:5in;width:90%;text-align:right;"> <div style="position:absolute;width:60%;border-top:1px solid black;right:0" /> Presented by:<br/> Jordan Sissel </h4> </body> </slide> <slide> <title> Ye Olde Web Organization:

Contains individual and community websites
Linked together using URLs (web links)
Web pages are designed for human consumption

Problems:

Content cannot describe itself (What am I about?)
Viewing designed for human consumption only
Search engines required to hackishly reprocess data
Little-to-no context surrounding data
Searching requires human input to define context
Massive amounts of information become difficult to process
Not all data comes in the form of web pages (email, images, etc)

Problem example Example of searching problem:

Search for information about the rock band: Live.
Google: "live"
Google: "live rock"
Google: "live rock group music band"

Conclusion:

Context is critical. Burden of context provision is on the user.

Web Logging Explosion Blogging:

Self-publication online has grown dramatically
Individually-maintained "Blogs" are very common
Allows users to publish with greater ease and speed
Blog information will still need to be categorized by machines

Results:

Faster propogation of memes
Constantly changing the structure of the Web
More individual-oriented websites than ever before

Enter: Semantic Web

se·man·tic: of or relating to the study of meaning

Metadata processing:

Humans can inference information pieces to form new information
"I drive a Tahoe." -- What does that mean?
What does Google say? Find Out
Machines are not as "smart" as we are.

Machines need help:

Data needs a way to describe itself in a way meaningful to a computer
Metadata: Data about data
Some protocols already exist for metadata description
Solution: Resource Description Framework (RDF)

End result:

Higher levels of human-digestable information

RDF: Metadata Defined What is RDF?

An general framework for the description of any Internet-based resource.
Subject, Predicate, and Object
Subject: Resource being described
Predicate: Trait or relationship
Object: Value of trait or object of relationship
Results: directed graph showing relationships of objects in the world

RDF Example

RDF Continued RDF supports extensible vocabularies, or metadata term sets, to aid in data description. Extends and improves languages to describe data:
"Semantic Blogging" is an article whose author is a person with the name "Jordan Sissel" and major "Computer Science"

Many vocabularies exist, for example:

RSS: describes blog entries and can specify: channels, items, images, links, and descriptions.
Dublin Core: describes published work attributes: creator, date published, subject, contributors, language, publisher, etc.
PRISM: also for describing published works, terms include: byte count, edition, createion date, industy, ISSN, modification date, rights agent, etc.

What now? What Next? Blogs have been a driving force in blogging software development. As a result, blogging software has adapted to include these RDF metadata publishing features.

Why? Bloggers want to let users read content with less effort.

The future of Semantic Web:

Greater personal information consumption satisfaction
Less time spent searching
More consumable information
Google search in the future

Questions References

Cayzer, Steve. Semantic blogging and decentralized knowledge management. Communications of the ACM Vol 47, No 12 (April 2004), 47-52
RDF Standard (http://www.w3.org/RDF)
Blogging History (http://en.wikipedia.org/wiki/Blog#Blogging_begins)
Dublin Core Metadata Terms (http://dublincore.org/documents/dcmi-terms/)
PRISM Metadata Standard