Introduction to Document
Management
Documation ‘97
February 25, 1997
Santa Clara, California
Kurt Conrad
conrad@sagebrushgroup.com
Goal of Tutorial
To help you to understand the fundamental
changes which are occurring in the field of
document management and their relationships
to process and technology alternatives.
Fundamental Changes
P Just now learning to use computers to
improve organizational performance.
P Destabilizing the nature of work
< Organizational purpose
< How individuals contribute value
P Document management “in the cross-hairs”
< Concept of the document
< Measures of value
Hidden Importance
P 80-90% of corporate information in
documents
P Documents claim
< 40-60% of office worker’s time
< 20-45% of labor costs
< 12-15% of corporate revenues
P Emerging metaphor for organizing complex
information
Documents as Strategic Assets
P Contain information critical to complex
organizational behaviors
< Provide context
< Integrate, document, and communicate
understanding
P Critical to customer satisfaction
P Inconsistently recognized as strategic
< Real men do databases
< CALS, ATA 2000, ISO 9000, etc.
What the Tutorial Will Cover
P What is Document Management
P The History of Document Management
P Document Management Architectures
P Implementation Issues
P Workflow Automation
P Integration Points
P Impact of the World Wide Web
What is Document
Management?
Simple Definition
Systems for managing collections of
documents
Wide disparity of approaches
P Document Image Management
P Full Text Retrieval
P Compound Document Management
P Online Viewing
P Workflow
P Object-Oriented Databases
What is Management?
Actions taken today to protect the future
Protecting the Future
P Do all your documents (or the information in
them) have the same future?
< “One size fits all” solutions are a common mistake
P How much will the future cost?
< Cost = f(Legacy, Vision)
P Future value is defined in terms of human
and automated behaviors
Metadata Determines Future Value
P Metadata = data about data
P Metadata is the basis for
behavior
P Humans can create metadata
and resolve ambiguous
metadata
P Computers can’t
P Documents are often rich in
ambiguous metadata
P Are your documents “smart
enough” to meet future needs?
What is Document Management?
P Document Management processes and
technologies protect the future value of
documents.
P A wide variety of approaches have been
developed which are based on different
concepts of the document and emphasize
different definitions of document value.
History of Document
Management Systems
History Overview
P Mirrors the evolution of the concept of the
document
P Conceptual changes closely tied to
technology and metadata changes (chicken
and egg)
P Three primary concepts
< Paper documents
< Automated paper documents
< Electronic documents
Paper Documents
Focus on the dynamics of the physical artifact
P Metadata implied through visual clues
< Linear sequence
< Typography and formatting
< TOC, lists, indexes, cross references, etc.
P Human interpretation creates meaning
P Efficient use of space often more important
than retrievability and reuse
P Innovations target the independent efficiency
of production, storage, and retrieval
Automated Paper Documents
Speeds the processing of physical documents
P Paper hides a multitude of sins
P Focus on visual formatting
< Laser printers allow more control
< HW/SW tools function like fast, powerful pens
< Metadata / operator interaction based on
formatting codes
P Illusion of control
P Management of meaning and semantics
limited to relational database world
Automated Paper Documents
Solutions often focus on a subset of the document lifecycle
Authoring
Editing
Research
Formatting
Viewing
Publishing
Retreival
Delivery
Storage
Automated Paper Documents
Technologies
P Paper-based interface standards
P Graphics, Wordprocessing, and Desktop
Publishing tools
P Manage information about the documents
< File management systems
< Image management systems
< Other database-based indexing systems
Electronic Documents
Conceptual Shifts
P Increased information density
P Documents are more than their paper
representations
< Time-based media
< Hyperlinks and other navigational aides
< Formal relationships to other sets of information
P Paper becomes a portable, high-resolution
display technology
Electronic Documents
Conceptual Shifts
P Processing-neutral encodings that support
multiple representations for delivery
P Emphasis on meaning and semantics
< Richer, more descriptive metadata that serves as
a basis for integrating the entire document
lifecycle
P Tied to new organizational models that are
based on shared pools of information
Electronic Documents
Performance
P Time and quality become dominate values
< Use and reuse of knowledge
< Customer satisfaction
P Performance and value increasingly limited
by production process
P Increased importance of up-front design
< Formalized structures and validation
< Explicit metadata that supports complex human
and automated behaviors
< Software and data interfaces
Electronic Documents
Technologies
P Manage information contained in documents
P Data encodings as interface standards
P Structured authoring
P Hypermedia authoring (including links,
annotations, workflow, other relationships)
P Component management systems
P Convergence of competing concepts
What is Document Management?
Revisited
P Today’s high-performance documents are
based on relationships
P Emphasis is shifting away from
< Simple storage and retrieval
< Independent management of life cycle phases
P New emphasis on integrating interrelated
information lifecycles
P Systems often encompass competing
concepts of the document
Overview of Document
Management Architectures
Overview
P Three models
< Image-based
< WYSIWYG DTP
< Compound document management
P Components
< Data encoding standards
< Software interoperability standards
< Task-specific tools
< Communications and repository infrastructure
Image-based Architectures
P Dragging paper documents into the
electronic age
P Heavy reliance on human interpretation
P Layering of metadata to capture meaning
and understanding
P Workflow automation and annotation
innovations
WYSIWYG DTP
P Control of visual aspects
P File-based and BLOBS
P Production focus
P Short-lived documents
< Advertising
< Novelty
< Drama
P WWW
Compound Document
Management
P Control of individual information objects
P Structure and semantics
P Late binding of typography
P Customization of both form and content
P Addressing and transformation issues
P Encompasses and consolidates other
architectures
Data Encoding Standards
General Questions
P Who controls the standard?
P What classes of metadata (conceptual
models) does it support?
P What behaviors does it support?
P Portability, platform independence, ability to
support required transforms
Data Encoding Standards
Text
P Paper
P Image
P Text
P Page image
P Traditional markup
P Generalized markup
Data Encoding Standards
Graphics
P Paper
P Image
P Vector
P Semantically-rich vector graphics
Data Encoding Standards
Other
P Audio
P Video
P Voice
P Positional
P Hyperlinking
P Rendering
P Behaviors
Software Interoperability
Standards
P Programming languages
P Application Programming Interfaces
< Single vendor
< Vendor consortium
P Examples
< Shamrock, DEN, ODMA, OLE, OpenDoc, CORBA
P Stability
Task-Specific Tools
Authoring
P Traditional
< Word processing and DTP
< Graphics
P Structured authoring
< SGML/HTML
< Forms
< Graphics
P Layering
< Browsers
Task-Specific Tools
Editing
P Heavily reliant on human interpretation
P Syntax checkers and validators
< Content (spelling, grammar)
< Markup
P Batch vs real-time
Task-Specific Tools
Formatting & Publishing
P Converters
< Scanners
< OCR/vectorizers
< Programmable
P Composition tools
P Physical media and associated hardware
P Hypermedia authoring tools
P Print on demand
Task-Specific Tools
Delivery & Storage
P Dependent on published form
P Relational and object-oriented databases
< Square pegs
< Tables, hierarchies, and non-linear relationships
< Performance
< Data model designs
< Granularity
P Email, workflow, other network-based
transport mechanisms
Task-Specific Tools
Retrieval
P Database queries
P Full text
< Boolean searches
< Weighted thesauruses
< Vector searches
< Context-sensitive searches
< Natural language
P Image matching
Task-Based Tools
Viewing
P Text readers
P Native file viewers
P Raster viewers
P Page viewers
P Binary browsers
P Fixed markup language browsers
P Arbitrary DTD browsers
Infrastructure
P Repository and communications subsystems
P Scope
P Granularity
P Encodings
P Versioning and configuration control
P Target of most software interoperability
standards
Implementation Issues
Human Issues
P Difficulty of adopting enabling technologies
< Conceptualization
< Learning
< Foresight
P Perceptions
< Technology problem
< Uniqueness
P Who knows?
Organizational Issues
P Reengineering
< Complex behavior based on richer semantics
< Self-awareness
P Information politics
< Stakeholder interests
< Policy development & governance
< Allocation of decision making
P Competing interests of information owners
and technology vendors
Technical Issues
P Adequate communications infrastructure
P Cross-platform integration
P Selecting standards
P Legacy systems and data
P Addressing and granularity
P Planning for obsolescence
P Labor costs
Workflow Automation
Issues
P Often confused with document management
< Check-in and check-out
< Component-level configuration control
P Convergence with document management
< Routing and communication
P Ad hoc vs engineered workflows
Opportunities
P Basic reengineering model
< Shift from linear flow to shared pools
< “Linear” process flows still remain
P Documenting transformations provides
additional context to information objects
< Facilitates understanding
< Simplifies reuse in new contexts
P Additional “publishing vectors”
Integration Points
Organizational Integration
P Information suppliers and consumers
P Metadata requirements
P Process, policy, politics
P Values
Data Integration
P Encoding standards
P Software interoperability standards
P Transformations
P Addressing
P Synchronization
Impact of the World Wide
Web
Primary Impact
First time that a large number of individuals
and organizations have used non-proprietary,
vendor-neutral encoding and communications
standards to implement a truly heterogeneous
computing environment.
Additional Impacts
P Encoding standards
P Software design
P Focus for consolidation
Encoding Standards
P HTML hides a multitude of sins
P A application of SGML
< Conformance issues
< Volatility
< Theology
P Easy to get into
P Danger in thinking that more than a delivery
encoding
Encoding Standards
P Simplicity limits utility and drives divergent
publishing models
< Complex graphics
< Structured data at the server
P Competing/complementary efforts
< Stupid HTML export
< Proprietary encodings
< Increased visual sophistication
< Structural flexibility
P XML Initiative
Software Design
P Viewer-centric
< Customized views
< “Do everything” browsers
P Smaller apps (e.g., plug-ins, java applets)
P Platform independence
P Authoring metaphors
Focus for Consolidation
P Aim for the accident
P Change changes change
< Perceptions of value
< User needs
< Vendor desires
Conclusion
P Use encodings as primary integration
mechanism
P Choose tools that let you control metadata
structures and object granularity
P Layer new relationships and meanings as
identified
P Engage stakeholders in all phases of
document lifecycle to identify metadata
requirements
