The Permanent Contribution: Digital Preservation and the Human Record

Full report

On this page

The internet is fundamentally ephemeral. Despite its vast scale and the illusion of permanence it projects, the digital human record is subject to constant attrition. This attrition is driven by shifting commercial business models, rapid technological obsolescence, and the inherent fragility of hypertext architecture. Valuable public knowledge disappears daily when platforms close, domain registrations expire, or server paths change. Addressing this vulnerability requires transitioning from passive digital consumption to deliberate digital stewardship. The central inquiry of modern digital archiving is how individuals and small organizations can architect public knowledge so that it remains identifiable, accessible, attributable, discoverable, and useful over generational timelines.

The Ephemerality of the Digital Record: Analytics of Decay

The loss of digital public records is not a theoretical vulnerability; it is a measurable, accelerating crisis known as digital decay. This phenomenon manifests primarily in two forms: link rot and content drift. Link rot occurs when a URL fails entirely, typically returning a 404 Not Found, a DNS resolution error, or a server timeout1. Content drift occurs when a URL still resolves, but the underlying page has been fundamentally altered, no longer reflecting the source material the original author intended to cite1. Both degrade the integrity of the web, but content drift is particularly insidious because it silently undermines the reliability of historical records without triggering automated broken-link detectors. Recent large-scale analyses reveal the staggering scope of this decay across all sectors of the digital landscape.

The centralization of online discourse into proprietary social media platforms has radically accelerated this erasure. Platform-hosted content exhibits a significantly higher disappearance rate compared to independently hosted websites5. Centralized platforms prioritize real-time engagement and advertising revenue over historical preservation. When a platform shuts down, alters its business model, or purges user accounts, entire communities and their accumulated knowledge are frequently deleted without recourse. Historical cases validate the dangers of centralized reliance. The closure of early web communities like GeoCities erased millions of early personal websites, while server migrations at MySpace resulted in the permanent loss of over a decade of independent music. In contrast, a 2024 Oxford Internet Institute study found that decentralized, independent content demonstrates far superior longevity5. University-hosted academic content boasts a 91 percent survival rate after a decade, and independent personal websites with dedicated hosting exhibit a 73 percent persistence rate over the same period5. Conversely, platform-hosted content averages just a 31 percent survival rate beyond five years5. The viral commentary that shaped a cultural moment or the community forum that crowdsourced niche technical solutions are routinely lost when the host decides the database is no longer profitable to maintain.

Institutional Preservation Frameworks

The mitigation of digital decay relies on robust methodologies developed by libraries, universities, open repositories, and public archives. These institutions utilize standardized frameworks to ensure the longevity and integrity of digital objects, shifting the paradigm from mere backup storage to active digital curation. The foundational standard for these efforts is the Open Archival Information System (OAIS), formalized as ISO 147217. The OAIS reference model defines the specific functions an archive must perform to preserve digital information over the long term, including the creation of structured Information Packages (Submission, Archival, and Dissemination)7. Under this model, an archive assumes responsibility for ingest procedures, persistent archival storage, data management, and long-term preservation planning7. Operating in parallel to OAIS are massive harvesting initiatives like the Internet Archive, which deploys automated web crawlers to capture publicly available web pages, preserving them in the Wayback Machine8. Similarly, Software Heritage focuses exclusively on harvesting and preserving the world's software source code from public Git repositories, ensuring the foundational logic of the digital age is not lost to repository deletion or corporate acquisition. Institutional repositories at universities utilize DOI-based publication workflows to ensure that academic outputs remain permanently accessible, even if researchers transition between institutions5. To evaluate and improve technical capacity in digital preservation, the National Digital Stewardship Alliance (NDSA) established the Levels of Digital Preservation11. This tiered framework provides a matrix for organizations to assess their preservation maturity across several functional areas:

1. Protect your data (Level 1): Focuses on basic bit preservation, keeping at least two complete copies in separate physical locations on non-removable disks, and establishing routine virus checks12.

2. Know your data (Level 2): Requires comprehensive metadata, the generation of cryptographic checksums (fixity checks) to detect unauthorized changes or data corruption, and verifiable file manifests12.

3. Monitor your data (Level 3): Involves routine, automated fixity checking against the established checksums, the auditing of logs, and the geographic dispersal of multiple copies12.

4. Repair your data (Level 4): Ensures the capability to automatically retrieve and replace damaged files from trusted backup nodes, conducting full disaster recovery simulations to ensure system resilience12.

Balancing Accessibility and Stewardship: FAIR, CARE, and TRUST

Institutional repositories operate under guiding philosophies that balance open access with ethical stewardship. The FAIR principles—Findable, Accessible, Interoperable, and Reusable—provide a data-centric approach to making digital assets technically shareable and machine-readable15. FAIR mandates the use of globally unique persistent identifiers (like DOIs), rich standard metadata, and clear, machine-readable usage licenses17. However, absolute openness can conflict with privacy, ethical obligations, and community rights. To address these gaps, the CARE principles for Indigenous Data Governance—Collective benefit, Authority to control, Responsibility, and Ethics—were developed16. While initially formulated for Indigenous data sovereignty, the CARE principles offer a vital framework for any digital preservation effort dealing with human subjects, sensitive communities, or marginalized knowledge. CARE ensures that data governance considers potential future harms, enforces contributor consent, and establishes cultural governance protocols regarding who is authorized to access specific records17. These principles are frequently synthesized with the TRUST principles (Transparency, Responsibility, User focus, Sustainability, and Technology) to form a holistic governance model for digital repositories15.

Responsible Archiving and the Right to be Forgotten

Digital preservation exists in constant tension with data privacy legislation, most notably Article 17 of the General Data Protection Regulation (GDPR), which enshrines the "right to erasure" or the "right to be forgotten"21. This right allows individuals to request the removal of their personal data when it is no longer necessary, when consent is withdrawn, or when it was unlawfully processed21. For archival projects, this creates a complex balancing act. Once information spreads online, achieving complete erasure is technically nearly impossible due to caching, decentralized backups, and immutable ledgers21. Furthermore, GDPR Article 17(3) contains explicit exceptions where the right to erasure does not apply, including compliance with legal obligations, freedom of expression, and "archiving purposes in the public interest, scientific or historical research, or statistical purposes"21. A responsible archiving practice requires strict protocols to evaluate withdrawal requests case-by-case. Archives must assess whether the requested data involves an overriding public interest, if the information is genuinely irrelevant, or if removing the data would fundamentally distort the historical record21. A technical infrastructure must support granular takedown requests and clear communication channels for contributors to exercise their consent rights. Archives must also be wary of the Streisand Effect, wherein the act of suppressing or deleting a record inadvertently draws massive public attention to the very information the individual sought to hide21.

Categorization of Tooling: Preservation vs. Discoverability

A robust archive requires a synthesis of disparate technologies. Some tools ensure cryptographic permanence, some ensure semantic discoverability, and others actively threaten the longevity of the record.

Formats for Longevity

Proprietary file formats are highly susceptible to software obsolescence. Durable public knowledge must rely on accessible, portable, and open formats.

• HTML, Markdown, and TXT: Plaintext and its lightweight markup derivatives are the gold standard for text preservation. They require no specialized rendering software, possess no hidden binary dependencies, and remain human-readable even if parsing tools fail24.
• PDF/A (ISO 19005): Standard PDFs rely heavily on system fonts, external links, and multimedia elements that degrade over time25. PDF/A is an ISO standard explicitly designed for long-term preservation by enforcing total self-containment25. It mandates the embedding of all fonts, device-independent color spaces, and XMP metadata while strictly prohibiting encryption, external references, and JavaScript execution25. Archives utilize open-source validation engines, such as veraPDF, to verify that generated files strictly adhere to PDF/A conformance levels (e.g., PDF/A-2b or PDF/A-3b) before ingest26.
• WARC (ISO 28500): The Web ARChive (WARC) file format is the international standard for archiving web content31. It captures complete HTTP transactions, including request headers, response headers, server status codes, and the underlying payloads (HTML, CSS, JSON, images)32. A WARC file serves as a transparent container that allows archived websites to be replayed locally exactly as they appeared at the time of harvesting, making it invaluable for offline processing and long-term repository storage9.

Identifiers, Metadata, and Citation Practices

Information is only useful if it can be found and understood. Discoverability and attribution depend on robust metadata schemas and persistent identifiers.

• DOIs and DataCite: Digital Object Identifiers (DOIs) provide a permanent, unbreakable link to a resource, mitigating link rot10. DataCite provides the global infrastructure for registering DOIs alongside a standardized metadata schema10. The DataCite Metadata Schema requires six core properties—Creator, Title, Publisher, Publication Year, Resource Type, and Identifier—ensuring that independent research outputs enter the global scholarly knowledge graph and are fed into platforms like Google Dataset Search34.
• Dublin Core (ANSI/NISO Z39.85): The Dublin Core is an internationally recognized standard (ISO 15836\) consisting of 15 foundational metadata elements (e.g., Title, Creator, Subject, Rights, Identifier)36. It is the foundational vocabulary for institutional repositories, ensuring cross-system interoperability through the "Dumb-Down Principle," which allows complex metadata to degrade gracefully into simpler formats for older systems37.
• Schema.org and JSON-LD: While Dublin Core is utilized heavily by backend library systems, Schema.org is the metadata vocabulary of the open web39. Embedding ScholarlyArticle metadata via JSON-LD (JavaScript Object Notation for Linked Data) within the HTML of a webpage ensures that the knowledge is highly structured and machine-readable41. This facilitates automated citation extraction, provenance tracking, and rich indexing by search engines39.

Decentralization, Versioning, and Networked Consensus

Relying on a single centralized server guarantees eventual failure. The preservation of the human record requires protocols that distribute trust, manage versions securely, and enable temporal navigation.

• The Memento Protocol (RFC 7089): The web is natively spatial, not temporal; a URL points to a location, returning whatever state exists at the time of the HTTP request43. Memento introduces an HTTP framework for time-based access to resource states43. It utilizes Accept-Datetime headers to negotiate access to prior versions of a webpage45. A designated TimeGate resolves the datetime to the appropriate archived snapshot (a Memento), while a TimeMap lists all available versions of the resource43. This allows users to request a resource exactly as it looked when a specific historical claim was made.
• Webmention (W3C Standard): Webmention is a decentralized protocol that allows websites to notify one another when they link to each other47. It serves as a decentralized citation and notification system47. When a writer on a personal domain links to an article on a public repository, their server sends a Webmention POST request to the repository's endpoint47. The repository server fetches the source URL, verifies that the link exists, and can then display it as a verified citation, reply, or bookmark49. This shifts the power of discovery, indexing, and conversational threading away from proprietary social silos back to independent domains.
• LOCKSS (Lots of Copies Keep Stuff Safe): LOCKSS is a peer-to-peer digital preservation network developed to mitigate operator error, internal attacks, and external threats51. Operating on a polling and consensus protocol known as LCAP, mutually distrusting network nodes regularly confer on the cryptographic hashes of their stored content51. If a node detects that its local copy has degraded or been altered, it repairs the data from the consensus of trusted peers54. This architecture fails gracefully and gradually; tampering or bit rot is detected and flagged for human review long before all copies across the network are compromised53.

The Benefits and Risks of Syndication

Syndication and maintaining multiple copies across different platforms (e.g., cross-posting content to Medium, Substack, LinkedIn, and personal blogs) increase immediate reach and create redundant off-site backups. However, syndication introduces significant risks of content fragmentation and canonical confusion. If an author issues a correction, updates a broken link, or retracts a statement, they must manually update every syndicated endpoint. Over time, these disparate versions drift apart, leaving researchers unable to identify the authoritative text. Furthermore, syndication can severely dilute link equity; if inbound citations are split across five different domains hosting the identical article, the discoverability and search ranking of the authoritative original suffers. To mitigate these risks, all syndicated copies must utilize a standard rel="canonical" HTML tag pointing directly to the durable archive, and the archive itself must act as the ultimate source of truth, utilizing DOI versioning to manage updates explicitly10.

Application to BuildToResist.com: Architecture for Persistence

BuildToResist.com intends to act as a voluntary public knowledge archive, linking individual contributions to stable pages while supporting export, citation, attribution, and decentralized discovery. Achieving this requires transitioning away from standard dynamic content management systems toward a rigorous, standards-compliant technical architecture.

1. System Architecture and Search Implementation

To eliminate database dependencies, software vulnerabilities, and high hosting costs, BuildToResist will utilize a Static Site Generator (SSG) architecture (such as Hugo or Eleventy)24. In this framework, the entire website is compiled into raw HTML, CSS, and JS files prior to deployment.

• Portability: The source files (Markdown and JSON metadata) are highly portable, completely framework-agnostic, and easily version-controlled in public Git repositories24.
• Security & Costs: Static sites cannot be compromised via SQL injection and cost mere fractions of a cent to host on standard content delivery networks24.
• Offline & Static Search: Dynamic searches typically require active server instances. Instead, BuildToResist will utilize Pagefind, an open-source static search library built in Rust24. Post-build, Pagefind crawls the static HTML, constructing a highly compressed WebAssembly index24. Search occurs entirely client-side within the user's browser, meaning the site remains fully searchable even if mirrored to an offline intranet, accessed via IPFS, or viewed from a local USB drive24.

2. Digital Preservation Risk Matrix

The following risk matrix dictates the technical countermeasures implemented by the BuildToResist architecture:

3. A 10-Year Maintenance Strategy

A durable archive requires realistic financial and operational forecasting. By utilizing static architecture, maintenance is vastly simplified compared to database-driven architectures, allowing small organizations to survive decade-long operational timelines.

4. Permanent-Link and Versioning Policy

Adhering to the architectural principle that "a cool URI is one which does not change"1, BuildToResist will enforce strict routing constraints.

• Stable URLs: URLs will follow a semantic, date-independent structure based on a unique cryptographic hash or DOI prefix (e.g., buildtoresist.com/archive/10.xxxx/yyyy). Once a page is published, its URL routing is immutable.
• Versioning: If a contributor makes a significant update to a published work, a new record is created. The DataCite Metadata Schema mandates that new DOIs be minted for major version updates to maintain citation integrity10. The original URL will remain active but will feature a prominent UI banner linking to the updated version, ensuring historical citations remain valid.
• Memento Compatibility: The server infrastructure will implement RFC 7089 Memento headers44. Requests for older versions of an article will be routed through a TimeGate, directing crawlers to the specific timestamped Memento59.

5. Downloadable-Content Strategy

To ensure true decentralized ownership and mitigate the risk of platform failure, users and readers must be able to export knowledge in robust, offline-capable formats. Every article page on BuildToResist will feature a "Preservation Asset Bundle" containing:

1. Source Markdown: The raw, human-readable text and frontmatter.

2. JSON-LD Metadata: The extracted Schema.org entity data ensuring semantic context is retained offline.

3. PDF/A Document: An automatically generated ISO 19005 compliant PDF for offline reading and institutional archiving, verified via a background veraPDF script to guarantee font embedding and color space compliance25.

4. WARC File: A localized web archive (ISO 28500\) generated via tools like warcio, capturing the article's full HTTP transaction, ensuring absolute fidelity of the page's styling, layout, and structure32.

6. Practical Backup, Restoration, and Link-Monitoring Procedures

• Backups (NDSA Level 3/4): The entire site is managed as a Git repository. Every commit automatically triggers a distributed replication to three geographically distinct storage nodes. A secondary LOCKSS-inspired daemon will randomly poll the cryptographic checksums of the static files on the edge servers against the source repository to detect silent bit rot53.
• Restoration: Because the architecture is entirely static, restoration requires zero database reconstruction. In the event of catastrophic server failure, the entire site can be redeployed to a new hosting provider directly from the Git repository in under five minutes.
• Link-Monitoring: A continuous integration script will run monthly, utilizing an open-source link checker. If an outbound link returns a 404, the system will not alter the author's original text, but will automatically append a data-archived-url attribute pointing to a verified snapshot on the Internet Archive, seamlessly intercepting the broken click for the user1.

7. Archival-Ready Article Template

Contributors submit their work utilizing a strict Markdown template with embedded YAML frontmatter. This frontmatter maps directly to Dublin Core and DataCite schemas34. During the build process, the static site generator translates this into semantic HTML with embedded Schema.org JSON-LD40.

BUILD TO RESIST - ARCHIVAL FRONTMATTER

title: "The Mechanics of Digital Decentralization" creator:

• name: "Jane Doe" orcid: "0000-0000-0000-0000" datePublished: "2026-10-12" abstract: "An analysis of peer-to-peer web archiving tools." subject: \["Digital Preservation", "Decentralization", "Webmention"\] license: "CC-BY-4.0" version: "1.0.0" relatedIdentifier:
• type: "IsSupplementTo" identifier: "https\://janedoe.com/original-post"

The Mechanics of Digital Decentralization

\[Body text written in standard Markdown\] At compilation, the build system automatically injects the corresponding JSON-LD into the HTML \<head\>, rendering the document highly machine-readable for automated citation networks41. It is vital to note the limitations of semantic discoverability regarding emergent technologies. While archiving in this format ensures structural integrity, semantic interoperability, and discoverability by human researchers and standard search indexers, it must be explicitly understood that archiving does not ensure inclusion in AI training data. Large Language Models rely on opaque, proprietary scraping schedules and filtering heuristics that operate entirely independent of open web metadata standards.

8. Contributor Guide: "Make Your Ideas Last"

The following text forms the foundational guide provided to all authors publishing on the platform, establishing behavioral norms for decentralized knowledge creation. 1\. Own Your Core Node Your knowledge should originate from domains you control. Write first on your personal website, and utilize BuildToResist as your durable, cryptographic secondary repository. Use rel="canonical" tags on your own domain pointing to the BuildToResist DOI to unify your citation metrics, or vice versa depending on your preferred source of truth. 2\. Leverage Webmention for Discovery Do not rely on algorithmic social media feeds to map the conversations around your work. BuildToResist fully supports the W3C Webmention standard47. Place your BuildToResist article URL in your email signatures, newsletters, and community profiles. When other authors link to your work on their own independent sites, their platforms will ping our Webmention endpoint. We will verify the cross-site link and display their commentary directly beneath your article. The conversation becomes a permanent, decentralized appendage of the work itself, entirely independent of commercial silos50. 3\. Cite Durably When writing your article, assume every outbound link will eventually rot4. Provide deep context in your text so the link is not a single point of failure. Whenever possible, archive the pages you cite using the Wayback Machine or local WARC tools, and provide the archival URL alongside the live URL. 4\. Opt for Open Formats Do not upload ideas trapped in proprietary word processors. Submit text in Markdown, data in CSV, and visualizations in open SVG. Information encapsulated in proprietary software is destined for obsolescence. 5\. Consent and Community Control By publishing here, you grant a perpetual, non-exclusive license for preservation. However, following CARE principles, you retain the ethical authority over your work16. Should you identify future harm associated with the publication, you may trigger a formal withdrawal request. The content will be cryptographically redacted, leaving a Dublin Core metadata "tombstone" (a 451 Unavailable For Legal Reasons response) to explain the removal to future researchers without violating your privacy rights.

Conclusion

The preservation of human knowledge in the digital era cannot be left as an afterthought or outsourced entirely to the goodwill of centralized commercial platforms. The statistics detailing digital decay represent an ongoing, silent burning of the modern library. Constructing a resilient public knowledge archive like BuildToResist.com requires deliberately blending the rigorous institutional frameworks of the OAIS, NDSA, and DataCite with the agile, decentralized technologies of the open web, such as static site generation, Webmention, and Memento. By prioritizing open formats (Markdown, PDF/A, WARC), implementing strict cryptographic fixity checks, minting persistent identifiers, and respecting the ethical boundaries of data sovereignty, individuals and small organizations can engineer spaces where knowledge survives. In this ecosystem, ideas are no longer held hostage to shifting corporate algorithms, domain expirations, or the decay of hyperlinks, but are instead embedded into a distributed, durable, and highly discoverable record of human contribution.

Works cited

1. What Is Link Rot and How Can You Stop It From Hurting Your Brand?, https\://bitly.com/blog/what-is-link-rot/

2. When Online Content Disappears \- Pew Research Center, https\://www\.pewresearch.org/data-labs/2024/05/17/when-online-content-disappears/

3. When Online Content Disappears \- Pew Research Center, https\://www\.pewresearch.org/wp-content/uploads/sites/20/2024/05/pl\_2024.05.17\_link-rot\_report.pdf

4. How long does a web page live: what the research says about link rot, https\://archivarix.com/en/blog/link-rot/

5. The Internet Is Breaking, and Nobody's Talking About It \- Medium, https\://medium.com/@shivashanker7337/the-internet-is-breaking-and-nobodys-talking-about-it-7894f50f1f3f

6. Link rot \- Wikipedia, https\://en.wikipedia.org/wiki/Link\_rot

7. OAIS Reference Model (ISO 14721\) Explained \- CASRAI, https\://casrai.org/dictionary/term/oais-reference-model-iso-14721

8. Gone but Not Forgotten: Recovering the Dead Web, https\://blog.archive.org/2026/04/23/gone-but-not-forgotten-recovering-the-dead-web/

9. WARC (file format) \- Wikipedia, https\://en.wikipedia.org/wiki/WARC\_(file\_format)

10. Best Practices for DOI Registration \- DataCite Support, https\://support.datacite.org/docs/best-practices-for-datacite-members

11. Fedora & the NDSA Levels of Digital Preservation, https\://fedorarepository.org/fedora-ndsa-levels-digital-preservation/

12. Getting Started \- Digital Preservation Handbook, https\://www\.dpconline.org/handbook/getting-started

13. The NDSA levels of digital preservation: an explanation and uses, https\://www\.digitalpreservation.gov/documents/NDSA\_Levels\_Archiving\_2013.pdf

14. NDSA LEVELS OF DIGITAL PRESERVATION: A REVIEW IN, https\://www\.digipres.org/publications/ipres/ipres-2023/papers/ndsa-levels-of-digital-preservation-a-review-in-terms-of-trustwo/

15. Geodetic Data: Applying FAIR, TRUST and CARE Principles, https\://geodesy.science/2026/geodetic-data-fair-trust-care/

16. CARE Principles for Indigenous Data Governance | ARDC, https\://ardc.edu.au/resource/the-care-principles/

17. 9 FAIR and CARE principles, https\://learning.nceas.ucsb.edu/2023-06-delta/session\_16.html

18. Data Ethics in Cultural Heritage: FAIR and CARE Principles, https\://culturedigitalskills.github.io/2024-ethics-data/fair-and-care-principles.html

19. 5.1 Introduction to FAIR \- DAG \- CESSDA Data Archiving Guide, https\://dag.cessda.eu/Chapter-5/1-Introduction-to-FAIR

20. Working with the CARE principles: operationalising Indigenous data, https\://www\.adalovelaceinstitute.org/blog/care-principles-operationalising-indigenous-data-governance/

21. Right to be Forgotten: Deleting Your Digital Past \- ComplyDog, https\://complydog.com/blog/right-to-be-forgotten

22. Elements of Erasure to Determine the Sufficiency of a GDPR Article, https\://scholarlycommons.law.case.edu/cgi/viewcontent.cgi?article=1115\&context=jolti

23. Delete Your Privilege: The Hidden Economy of Digital Ghosts, https\://medium.com/@elliotJL/delete-your-privilege-the-hidden-economy-of-digital-ghosts-9cb695602a5d

24. Building a Lightning-Fast Blog with Astro: My Complete Setup Journey, https\://www\.revibyte.blog/posts/building-lightning-fast-blog-with-astro-complete-setup/

25. PDF/A (ISO 19005): The Standard for Long-Term Electronic, https\://dev.to/derek-compdf/pdfa-iso-19005-the-standard-for-long-term-electronic-document-preservation-and-conversion-tools-3ml6

26. What is PDF/A? The complete guide to PDF/A‑1/2/3/4 \- Nutrient, https\://www\.nutrient.io/blog/what-is-pdf-a/

27. ISO 19005-1 \- PDF Association, https\://pdfa.org/resource/iso-19005-1-pdf-a-1/

28. PDF/A \- Wikipedia, https\://en.wikipedia.org/wiki/PDF/A

29. FAQ \- PDFix, https\://pdfix.net/faq/

30. PDF Format Preservation Assessment Part 2: PDF/A Profile \- org, https\://wiki.dpconline.org/images/2/22/PDFA\_Assessment\_v1.0.pdf

31. The WARC File Format (ISO 28500\) \- Information, Maintenance, Drafts, https\://bibnum.bnf.fr/WARC/

32. Scrapfly Crawler API | WARC Format Reference, https\://scrapfly.io/docs/crawler-api/warc-format

33. The stack: An introduction to the WARC file \- Archive-It, https\://archive-it.org/blog/the-stack-warc-file/

34. Working with DataCite, https\://datacite.org/wp-content/uploads/2023/08/DataCite\_Brochure.pdf

35. Towards richer metadata \- perspectives from three DataCite projects, https\://datacite.org/blog/towards-richer-metadata-perspectives-from-three-datacite-projects/

36. Information Standards | NISO website, https\://www\.niso.org/publications/standards?page=1

37. Dublin Core Metadata Standards Explained \- Sourcely, https\://www\.sourcely.net/resources/dublin-core-metadata-standards-explained

38. Dublin Core Metadata Element Set \- Standards Information System, https\://standards.clarin.eu/sis/views/view-spec.xq?id=SpecDcMes

39. Reference, Citation & Publishing Terminology \- CiteKit, https\://citekit.com/docs/terms/

40. An introduction to Schema.org and JSON-LD.pptx \- Zenodo, https\://zenodo.org/records/14920117/files/2025.02.25%20-%20An%20introduction%20to%20Schema.org%20and%20JSON-LD.pptx.pdf?download=1

41. ScholarlyArticle \- Schema.org Type, https\://schema.org/ScholarlyArticle

42. abstract Schema Field: Format and Examples \- Karpi Studio, https\://www\.karpi.studio/schema-glossary-terms/abstract

43. Memento \- Knowledge Sidekick, https\://knowledgesidekick.org/standards/mementos/

44. RFC 7089: HTTP Framework for Time-Based Access to Resource, https\://www\.rfc-editor.org/info/rfc7089/

45. GitHub \- lanl/invenio-memento-proxy: Memento protocol support for, https\://github.com/lanl/invenio-memento-proxy

46. Introduction to the Memento Protocol, https\://mementoweb.org/guide/quick-intro/

47. Social Web Protocols \- W3C, https\://www\.w3.org/TR/2016/WD-social-web-protocols-20160603/

48. Webmentions with batteries included \- Fabio Manganiello, https\://blog.fabiomanganiello.com/article/webmentions-with-batteries-included

49. Social Web Protocols \- W3C, https\://www\.w3.org/TR/2016/WD-social-web-protocols-20160112/

50. How Webmentions Work on brennan.day, https\://brennan.day/how-webmentions-work-on-brennan-day/

51. Frequently Asked Questions \- LOCKSS Program, https\://www\.lockss.org/about/frequently-asked-questions

52. How universities preserve and protect digital assets with Grafana, https\://grafana.com/blog/how-universities-preserve-and-protect-digital-assets-with-grafana-dashboards/

53. Preservation Principles | LOCKSS Program, https\://www\.lockss.org/about/preservation-principles

54. Enhancing the LOCKSS Digital Preservation Technology, https\://www\.dlib.org/dlib/september15/rosenthal/09rosenthal.html

55. LOCKSS: A Peer-to-Peer Digital Preservation System, https\://cgi.di.uoa.gr/\~mema/publications/SOSP2003-long.pdf

56. Essays \- Manton Reece, https\://www\.manton.org/categories/essays/

57. Archives Month: Web Archive (WARC) File Format | The Texas Record, https\://www\.tsl.texas.gov/slrm/blog/2021/10/archives-month-web-archive-warc-file-format/

58. People & Blogs Reader, https\://peopleandblogs.feedland.cloud/

59. Memento Framework — TimeGate 0.5.0.dev20160000 documentation, https\://timegate.readthedocs.io/en/latest/memento.html

60. 1.7.0 • pypi-warcio • tessl • Registry • Tessl, https\://tessl.io/registry/tessl/pypi-warcio/1.7.0