Transcript Dublin Core Metadata – the Key?

Unlocking Access to
Information Resources
Roger A. Longhorn
Director, IDG (UK) Ltd
[email protected]
R A Longhorn, TechServ Support
Team
1
Dublin Core Metadata – the Key?
 Define consistent Dublin Core (DC)
metadata use on I*M Europe Web sites
 Investigate ways to create Dublin Core
metadata as painlessly as possible!
 Investigate the sensitivity of search
engines to DC metadata (underway)
 Examine XML applicability (pilot)
R A Longhorn, TechServ Support
Team
2
Creating DC Metadata
 UKOLN DC.dot generator
– Creates DC, RDF, XML, others
 Nordic Metadata Project DC Templates
 TagGen DC and similar commercial tools
 Comparisons?
FOR MORE INFO...
Visit … http://www.ukoln.ac.uk/metadata/dcdot/
http://www.lub.lu.se/cgi-bin/nmdc.pl
http://www.hisoftware.com/fact_sheetcc.htm
R A Longhorn, TechServ Support
Team
3
Search Engines
 How good are they are using metadata?
 Are any specifically configured to use DC
metadata?
 Will using DC metadata improve your
chances of being “discovered”?
FOR MORE INFO...
Visit http://searchenginewatch.internet.com/
R A Longhorn, TechServ Support
Team
4
Main DC Issues Examined
 Controlled lists
– Keywords (DC.Subject) – how many and
where?
– DC.Type (is “Fact Sheet” valid?)
 “Depth” of DC metadata to use
– Verbose versus brief
 HTML DC vs. XML vs. RDF vs. .???
R A Longhorn, TechServ Support
Team
5
Methodology
 Create DC metadata for 49 FP4 Libraries
project Fact Sheets (in first instance)
 Use different DC generation tools and
record time/effort required.
 Submit both old and updated Web pages to
search engines and monitor results.
 Compare results and report findings.
R A Longhorn, TechServ Support
Team
6
Current Status
 DC metadata created for all Project Fact Sheets
– UKOLN DC.dot
– Nordic Metadata Project
– TagGen DC
 Pages being published to Web now
 Search engines investigated
 Search engine results to be completed in 2
weeks, with final report/findings.
R A Longhorn, TechServ Support
Team
7