Skip to main content

Preservation Policy

Academic Stack Preservation Policy

Ensuring the long-term preservation, integrity, and global accessibility of scholarly content.

Preservation Commitment

Academic Stack Preservation is designed to secure the scholarly record against technical failures, organizational transitions, and infrastructure disruptions.


đŸŽ¯ Purpose​

Academic Stack Preservation provides long-term digital preservation services for scholarly publications and related research assets. Our objective is to ensure that published content remains authentic, complete, and recoverable over time.

This policy establishes the operational frameworks and service commitments for publishers, editors, authors, and institutional partners regarding continuity, preservation responsibility, and long-term access planning.


đŸ“Ļ Scope​

The preservation service supports key scholarly publishing assets:

📄 Core Formats

  • Journal articles & books
  • Conference proceedings
  • PDF & HTML representations
  • Structured JATS XML

đŸˇī¸ Associated Assets

  • Figures, tables & images
  • Supplementary research files
  • Bibliographic & DOI metadata
  • Crossref deposition data
Configuration Notice

The specific preservation scope may vary depending on the publisher's platform configuration, service agreement, and metadata availability. Academic Stack may define additional validation requirements for non-standard formats.


đŸ“Ļ Preservation Package​

To prevent files from being archived in isolation, Academic Stack packages scholarly content with the full bibliographic context required for future reconstruction.

Package ElementPurpose
Primary publication filesPreserve the scholarly work in its published form (PDF, XML, HTML).
Structured metadataSupport discovery, indexing, migration, and future reconstruction.
Persistent identifiersMaintain DOI and related identifier continuity.
Version informationRecord whether the item is an original, corrected, updated, or retracted version.
Technical metadataSupport integrity checks, format identification, and recovery operations.

đŸ›Ąī¸ Preservation Principles​

Academic Stack Preservation is built upon the following core architectural principles:

🌐 Long-Term Preservation

Scholarly content is preserved with the objective of maintaining long-term usability and accessibility across generations of technology.

🔍 Integrity Verification

All preserved files undergo rigorous checksum generation at ingestion and periodic validation checks to detect any silent corruption.

🔀 Geographically Redundant

Content is duplicated across multiple independent storage regions and cloud providers to eliminate single points of failure.

🔄 Automated Recovery

Self-healing systems automatically restore corrupted or missing files using verified redundant copies without human intervention.

📁 Version History

We preserve the complete lifecycle of the publication, including corrections and retractions, maintaining an auditable scholarly record.

💾 Metadata Continuity

Bibliographic, licensing, and citation metadata are preserved side-by-side with publications to ensure discoverability.

🧩 Context & Authenticity

Every file is linked to its publisher, journal, issue, and volume metadata, preserving the original scholarly context.

đŸ“Ļ Data Portability

Preservation records can be exported in standardized structures, supporting publisher continuity and platform migrations.


🔍 Integrity Monitoring​

We perform continuous cryptographic validation to detect and correct digital decay or unauthorized file changes.

Verification Workflow
  1. Ingestion Checksum: Generated immediately upon asset submission (SHA-256).
  2. Scheduled Audit: Periodic background sweeps verifying file integrity.
  3. Automated Recovery: Instant restoration from a verified replica if a mismatch is detected.
  4. Audit Logs: Full tracking of validation events and recovery operations.

đŸ—ī¸ Storage Infrastructure​

Academic Stack Preservation distributes assets across a hybrid multi-cloud architecture.

  • Risk Separation: Content replication spans different cloud providers (e.g., AWS, Cloudflare, Tencent Cloud) and geographical regions.
  • Provider Independence: Storage layers can be updated, upgraded, or migrated without disrupting service availability.
  • Access Control: Storage buckets utilize strict access privileges and immutable write-once policies (Object Lock) where applicable.

đŸ“Ĩ Ingestion and Validation​

The ingestion pipeline enforces quality gates before content is committed to the long-term repository:

  1. Submission Receipt: Publication assets are pushed from the live publishing system.
  2. Metadata Matching: Verification of DOIs, article metadata, and file mappings.
  3. Checksum Ingestion: Generating permanent SHA-256 hashes for all files.
  4. Packaging: Creating the archival unit with structural metadata.
  5. Replication: Writing copies to geographically separated storage tiers.
  6. Logging: Committing status and timestamps to the system audit trail.
Quality Gate

If an asset is incomplete, lacks critical metadata (e.g., DOI), or fails ingestion checks, the pipeline automatically alerts system operators and pauses archiving until resolved.


🔓 Access and Retrieval​

Our preservation network acts as a dark archive and recovery layer, not as a primary content delivery network (CDN).

Retrieval is reserved for specific continuity and recovery scenarios:

  • Disaster Recovery: Restoring active journal sites after system failures.
  • Audits & Verification: Exporting packages for external validation.
  • Migration: Moving publisher libraries to alternative configurations.
  • Exit Support: Exporting complete structured datasets for publisher transition.

🔒 Security​

We employ robust defense-in-depth measures to safeguard preserved materials:

  • Transit Security: All API endpoints and replication channels utilize TLS 1.3 encryption.
  • Rest Security: Storage arrays employ AES-256 encryption at rest.
  • Operational Controls: Strict multi-factor authentication (MFA) and principal of least privilege control administrative operations.
  • Tamper Protection: Immutable configurations prevent accidental or malicious file deletions.

âš–ī¸ Roles & Responsibilities​

Preservation is a shared commitment between the publisher and Academic Stack.

đŸ‘Ĩ Publisher Responsibilities

  • Ensure submitted assets are complete and suitable for long-term archiving.
  • Verify copyright, licensing permissions, and appropriate authorizations.
  • Promptly communicate retraction requests, legal mandates, or journal transfers.
  • Provide accurate metadata and timely corrections.

đŸĸ Academic Stack Responsibilities

  • Maintain and monitor global preservation infrastructure.
  • Run periodic cryptographic checksum validation programs.
  • Guarantee redundant copy replication and storage health.
  • Support timely retrieval and recovery workflows.
  • Evolve preservation procedures alongside industry standards.

âŗ Preservation Period​

Accepted content remains preserved indefinitely:

  • Subscription Termination: Canceling an active publishing subscription does not purge archived scholarly records. They remain preserved to safeguard the scholarly record.
  • Takedowns: Legal, ethical, or copyright-enforced removals are handled through formal administrative procedures in coordination with the publisher.

🔄 Changes, Corrections, and Retractions​

Scholarly publications are living records. To maintain an honest and auditable scientific timeline, Academic Stack handles updates as follows:

  • Version History: Corrected or updated versions are appended as new entries rather than overwriting existing records.
  • Retraction Manifests: Retracted publications retain their physical files in the archive but are updated with clear retraction metadata and status markers.

âš ī¸ Limitations​

While Academic Stack is designed for maximum reliability, users should note:

  • Third-Party Registries: Platform-level preservation does not substitute for national library legal deposits or specific institutional registries unless explicitly configured.
  • Ingestion Failures: We cannot recover or guarantee the safety of materials that fail the ingestion pipeline due to publisher-side errors.

📈 Service Evolution​

We regularly refine our preservation architectures to leverage emerging technologies, support new media formats, and integrate with external networks (e.g., CLOCKSS, Portico).


âœ‰ī¸ Contact​

For inquiries regarding preservation policies, continuity planning, or custom compliance audits, please contact us:

Academic Stack Support
Email: [email protected]


Last updated: July 2026