Week 3 Reading Notes

Introduction to Metadata: Setting the Stage

• Metadata = data about data, describes information object
• Necessary for "identification, representation, interoperability, technical management, performance, and use of data contained in an information system"
• Metadata should tell us: content (what it contains, intrinsic), context (who what where why how, extrinsic), and structure (formal set of associations with other objects, like MARC records)
• In libraries: indexes, abstracts, bibliographic records
• Automated: metadata mining, Web crawling
• Need for standardization of metadata
• Was emphasis on structure and context --> now more emphasis on content with new tech.
• Metadata is more than description and resource discovery - also object behavior, function and use, relationships, and management over time
• NEW: user-created metadata, tagging, folksonomies

Types of metadata:

• Administrative
• Descriptive
• Preservation
• Technical
• Use

Attributes and characteristics:

• Source of metadata - internal (intrinsic, original creator) vs. external (outside source)
• Method of metadata creation - automatic vs. manual
• Nature of metadata - nonexpert vs. expert
• Status - static vs. dynamic, long-term vs. short-term
• Structure - structured vs. unstructured
• Semantics - controlled (standardized) vs. uncontrolled
• Level - collection-level vs. item-level

Phases: Creation/reuse --> Organizing/describing --> Validation --> Searching/Retrieval --> Disposition --> [repeat]

• Metadata: not necessarily digital, more than description, comes from variety of sources, continues to accrue, one object's metadata can be another's data
• Important for: increased accessibility, retention of context, expanding use, learning from metadata, system development and enhancement, multiversioning, legal issues, preservation and persistence

The main question I was left with after reading this chapter was: How much metadata is too much? This chapter talked about all the advantages of having all kinds of different types of metadata for information objects, but how much is too much? We use metadata to cope with information overload of data, so what about information overload of metadata? Aren't we just repeating the same problem?

An Overview of the Dublin Core Data Model

• Dublin Core Metadata Initiative (DCMI) = "international effort designed to foster consensus across disciplines for the discovery-oriented description of diverse resources in an electronic environment" or standardized metadata architecture

DCMI Requirements:

• Internationalization
• Modularization/Extensibility - namespaces used for independent definitions of same term
• Element Identity - specific, standardized definitions of terms
• Semantic Refinement - richer semantic definitions
• Identification of encoding schemes - less ambiguity
• Specification of controlled vocabularies - "allow for additional understanding of contextual information"
• Identification of structured compound values

Need to describe properties of resources

This project, while some of it is beyond my comprehension, makes sense to me as an effort to standardize the way we use metadata. The benefits of standardization in many aspects of information management is pretty self-evident to me, so this project sounds like a great idea.

EndNote X5: Introduction

• EndNote = bibliographic software program

Can be used to:

• Import citations from saved literature searches - hyperlinks to location on Web
• Develop a personal library of references - catalog and track information, save notes
• Create and format citations for papers and publications - automatic formatting

Flexible, customizable

This program sounds extremely helpful for someone who is dealing with a lot of resources for a job or a particular project. Having a way to organize and catalog these different resources would aid in accessing them in the future.