Practical Digital Preservation: A How-To Guide for Organizations of Any Size

by Adrian Brown

This very practical guide, offering a comprehensive overview of best practice, is aimed at the non-specialist, assuming only a basic understanding of IT and offering guidance as to how to implement strategies with minimal time and resources. Digital preservation has become a critical issue for institutions of all sizes but until recently has mostly been the preserve of national archives and libraries with the resources, time and specialist knowledge available to experiment. However, as the discipline matures and practical tools and information are increasingly available the barriers to entry are falling for smaller organizations which can realistically start to take active steps towards a preservation strategy. However, the sheer volume of technical information now available on the subject is becoming a significant obstacle and a straightforward guide is required to offer clear and practical solutions. Each chapter covers the essential building blocks of digital preservation strategy and implementation including: making the case for digital preservation; understanding your requirements; models for implementing a digital preservation service; selecting and acquiring digital content; accessioning and ingesting digital objects; describing digital objects; preserving digital objects; providing access to users; future trends. This is an essential handbook for anyone involved in digital preservation in medium or smaller sized organizations and those wanting to get a better understanding of the process. It's also a useful guide to digital preservation basics for students studying library and information science, archives and records management courses and academics getting to grips with practical issues.

Genres: Technology, Nonfiction

336 pages, Paperback

First published January 1, 2013

Reviews

[Rob Tarling · Rating 4 out of 5]

Excellent read - very pragmatic and practical to get one started.

Here are my crib notes of the book as a whole:

Purpose and scope
• Keep digital materials authentic, usable, and discoverable for as long as required.
• Treat preservation as an organizational capability (policy + people + workflows + technology), not a one-off software buy.

Minimum technical baseline
• Storage & copies: Maintain ≥3 independent copies, including one offsite or on a different medium/provider.
• Integrity: Establish checksums at receipt; schedule fixity checks with alerts and auto-repair paths.
• Security: Enforce access controls, roles, and audit trails; quarantine and malware-scan before ingest.
• Characterization: Identify and validate formats from bitstreams; extract technical metadata.
• Preservation actions: Plan migration/normalization for riskier formats; record every action.
• Metadata & IDs: Capture minimal descriptive/admin/technical/preservation metadata with stable identifiers.
• Packaging: Use consistent, machine-readable packaging (e.g., BagIt/METS) to bind content and metadata.
• Access: Provide basic discovery and delivery while protecting preservation masters.
• Ops: Monitor capacity and health; document standard operating procedures (SOPs).

Policy essentials
• Purpose & scope: What you preserve and why.
• Governance & roles: Responsibilities, escalation paths, change control.
• Authenticity & integrity: Principles for checksums, versioning, and evidence trails.
• Technical approach: Technology watch, migration/emulation criteria, risk management.
• Standards: Metadata, identifiers, packaging, and exchange protocols.
• Security & DR: Storage profiles, backup/offsite strategy, disaster recovery.
• Review: Scheduled policy and practice reviews; continuous improvement.

Business case (to secure mandate and funding)
• Drivers: Compliance, institutional memory, user demand, research/reuse value, reputation/risk reduction.
• Inventory (DAR): Size, location, owners, formats, sensitivity, risk scores, cost exposure.
• Options analysis: Compare target architectures (including “do nothing”); show total cost of ownership and risk.
• Phasing: Start with a minimal viable service; expand capabilities as value and demand grow.
• Resourcing: Roles, training, support; realistic operating budget.

Roles and skills (can be combined in small teams)
• Ingester: Accepts transfers, prepares submission packages, enforces deposit rules.
• Metadata specialist: Curates descriptive/admin/technical metadata; quality assurance.
• Repository manager: Oversees ingest, preservation actions, access, compliance, reporting.
• System support: First-line help for workflow tools and depositors.
• System administrator: Storage, identity/access management, monitoring, security, capacity planning.
• All roles: Basic toolchain literacy; documentation culture; incident response awareness.

Ingest pipeline (SIP → AIP → DIP)
• Quarantine & scan: Isolate incoming content; run AV/malware checks; log tools/versions/outcomes.
• Characterize: Identify/validate formats; extract technical/structural metadata; flag blockers (encryption, DRM).
• Validate SIP: Check completeness, accuracy, conformity to deposit profile, and integrity (checksums).
• Enhance metadata: Add/confirm description, rights, agents; record preservation events with timestamps.
• Ingest to storage: Transform SIP into AIP; record custody transfer; ensure persistent links.
• Derivatives: Create access copies (DIPs) and/or normalized masters as policy dictates.

Fixity and integrity
• Establish once, verify often: Generate checksums at receipt; verify on a cadence tied to risk and storage class.
• End-to-end coverage: Validate during transfers, replications, and storage migrations.
• Remediation: Keep a clean copy; automate repair, but always log and review changes.
• Protect evidence: Secure the checksum store and audit logs; separate duties where possible.

Metadata that keeps things usable
• Descriptive: Title, creator(s), dates, subjects/keywords, summary, relationships.
• Administrative: Ownership, depositor, access conditions, sensitivity, retention, embargoes.
• Technical: File format/version, size, fixity, key properties (e.g., color space, bit depth).
• Preservation events: Virus check, characterization, validation, normalization/migration, with tools/versions/outcomes.
• Identifiers & links: Stable local IDs and/or PIDs (Handle/DOI); durable links between objects, metadata, and packages.
• Interoperability: Keep exports/imports feasible (e.g., using standard schemas and packaging).
• Pragmatism: Define a mandatory minimum; automate extraction where possible to reduce burden.

Access strategy (serve users while protecting masters)
• User actions: Decide what users can view, download, analyze, or cite.
• Appropriate delivery: Provide formats fit for purpose (e.g., PDF/A for reading; CSV/JSON/TIFF for reuse).
• Tiering: Isolate preservation masters; expose an access tier with throttles and logging.
• Persistence: Offer stable, citable links and clear rights statements.

Trusted repositories and assurance
• Frameworks & audits: Use recognized self-assessments and certifications to demonstrate trustworthiness.
• Controls portfolio: Show evidence across governance, organizational support, technology, security, and sustainability.
• Transparency: Publish policies, retention rules, preservation actions, and change logs.

Practical guardrails (avoid common pitfalls)
• Start small, iterate: Launch a minimal service now; mature via documented improvements.
• Policy ↔ procedure link: Back every principle with deposit guidance and SOPs.
• Quarantine discipline: Never ingest unscanned or unvalidated content.
• Checksum discipline: Verify all replicas; guard the checksum store; alert on failures.
• Right-sized metadata: Collect only what’s necessary at ingest; automate the rest.
• Risk visibility: Always compare against “do nothing” to highlight exposure.

[Saima · Rating 3 out of 5]

A good overview of the entire digital preservation process. Mostly examples of UK libraries. Mentions PraaS as an option.