It is easy to make digital preservation sound like a shopping exercise: choose an archiving tool, install a validator, buy more storage and assume the collection is protected. In practice, useful preservation begins with a question about the material and the risk. A tool is appropriate only when you can explain what problem it solves, what evidence it produces and what happens when it fails.
This guide connects historical discussions of preservation tool registries with a practical method for selecting tools today. It is an independent TwoSquares resource, not an official mirror of the former Digital Curation Exchange tool list.
Start with the preservation action, not the software name
A preservation workflow usually contains several distinct jobs:
| Task | What you need to know | Typical output | |---|---|---| | Discovery and inventory | What exists, where it is stored and who is responsible | File list, URL inventory, collection register | | Web capture | Which pages and assets were accessible at a point in time | Web archive or captured HTML and assets | | File identification | What formats and versions are present | Format identification report | | Fixity checking | Whether the bytes of a file have changed | Checksum manifest and verification log | | Characterisation | Which technical properties affect future use | Metadata about formats, encodings and dependencies | | Packaging and transfer | How files, metadata and relationships travel together | Transfer package and manifest | | Access and reuse | How a future user can find and use the material | Stable records, access copies and documentation |
A single product rarely handles all these jobs well. Larger systems may coordinate several utilities, but the underlying decisions still need to be documented.
COPTR: a community-owned tool registry
The Community-Owned digital Preservation Tool Registry (COPTR) is a community-maintained resource for exploring tools and workflows by preservation function and content type. It is an appropriate starting point when you know the job but are unsure which software may fit.
The registry emerged from concern that preservation knowledge had been scattered across overlapping lists and project websites. In a January 2013 account, Paul Wheatley argued for pooling information rather than duplicating tool registries. Today, COPTR remains a living community resource. See our source record for the historical proposal.
A registry is a discovery aid, not a certification of suitability. Confirm that a tool is maintained, that its licensing is compatible with the planned use, and that its outputs can be validated independently.
Categories worth understanding
Website capture and web archiving
For a public website, the critical task is to capture pages together with enough of their dependencies to make them understandable. The result may use a WARC-based workflow or another well-documented format. Dynamic interfaces, external scripts, personalisation and authenticated content create gaps that must be acknowledged.
Before capturing, identify the pages that actually matter: research outputs, support documentation, historical announcements, downloadable files, policy pages and frequently cited URLs. A crawler that retrieves thousands of navigation duplicates but misses a single important PDF is not a successful preservation strategy.
Format identification
A filename extension is a clue, not definitive evidence. Tools and reference databases can help identify actual content types. The UK National Archives maintains PRONOM, a technical registry of file formats. Format identification should preserve the original file, record the method used and flag uncertain results.
Fixity and integrity
A cryptographic checksum such as SHA-256 creates a fingerprint of a particular file's bytes. Matching checksums offer evidence that a copy has not changed between checks. They do not prove who originally created the file or whether its contents are truthful. Record the time, algorithm, location and outcome of each verification. See our file formats and fixity guide.
Packaging and metadata
Preservation packages should describe what has been included, how it was acquired, which checks were applied and what the relationships are between original files, metadata and derived copies. The exact structure depends on organisational policy and available infrastructure, but a manifest and readable acquisition note are useful almost everywhere.
Emulation and migration
Sometimes a format cannot be used with modern software. Migration creates a different representation, ideally with documented testing. Emulation attempts to reproduce the environment required by the original object. Neither is universally superior: both introduce questions about authenticity, dependencies, usability and future maintenance.
How to evaluate a tool before introducing it
Test it on representative material. A utility that works on a modern PDF may behave differently with a damaged disk image or a legacy proprietary format. Keep samples that expose unusual cases.
Check reproducibility. Can someone else repeat the process? Can they obtain the same version, understand the command options and interpret the result?
Review security and privacy. Sensitive files should not be uploaded to a third-party service without an appropriate legal and security assessment. Identify where data travels and how long it remains there.
Avoid single-vendor dependency. Exportable records, documented interfaces and open formats make later migration easier. Closed systems can still be suitable, but their exit conditions need to be understood.
Maintain an evidence trail. Preserve logs and human decisions, not just successful outputs. A warning about a missing file is part of the collection's history.
A lightweight starting workflow for a small organisation
Begin with a complete inventory, a rights register and reliable copies stored in more than one place. Create checksums for important files and maintain a basic verification schedule. Capture the public website and retain its URL inventory. Document contact details, access requirements, system versions and restoration steps. Test a sample restore and record what did not work.
This is a better foundation than deploying an elaborate preservation platform without clear ownership. More advanced systems become valuable as collection size, risk and access requirements justify them.
For businesses protecting websites and online documentation, TwoSquares website archiving focuses on the practical capture and recovery side. For sites that have already disappeared, see website restoration.
Questions people ask
Is COPTR software I can install?
No. COPTR is a searchable registry of tools and workflows, not a single preservation application.
Can a checksum prove a file is authentic?
It can help demonstrate that the bytes match a previously recorded copy. Provenance and authenticity require additional evidence about acquisition, source and handling.
Should I preserve every page of an old website?
Preserve what is possible and lawful, then prioritise pages with documentary, legal, research or user value. Record gaps rather than presenting a partial capture as complete.
Primary resources
- COPTR registry
- About COPTR and its scope
- Open Preservation Foundation: Creating a Community-Owned Tool Registry
- The National Archives: PRONOM
Explore next: Digital forensics workflows | Research library | Digital preservation hub