Dead References and Their Implications

The issue of dead references is one that academic publishing has recognized and addressed. The Digital Object Identifier (DOI) system is utilized by over 5,000 publishers, providing each published work with a unique, permanent identifier. A DOI functions like a URL by directing users to a specific piece of content; however, it is guaranteed to consistently lead to the original source. If a publisher updates their systems, the DOI can be modified to ensure it always points to the correct location. Since its inception in 2000, there are now over 190 million unique items referenced through this system.

Traditionally, "publication" implied a permanent version of a work, with any updates released as separate articles. While this holds true for academic publications and books, it doesn’t apply to much of the internet. Websites frequently change, reports may be removed, and domains can expire. Research suggests that the average lifespan of a webpage ranges from 44 to 100 days. Ironically, one of the sources used in this research is no longer available online.

This transient nature of the internet is acknowledged in various referencing guidelines. For instance, the Harvard referencing style mandates that non-DOI sources include an "accessed" date, which at least indicates when the material was consulted. Yet, how can readers effectively utilize this information? What steps can we take to mitigate this issue?

Strategies for Addressing Dead References

To combat the problem of dead references, I propose a couple of solutions.

Firstly, authors should utilize the "save it" feature of the Internet Archive's Wayback Machine to capture a snapshot of the referenced page. This tool has archived over 451 billion web pages, which are regularly crawled after being saved. In the article's reference list, the displayed URL can show the original link, while the href can point to the archived version. This method allows readers to access the current version while easily referencing the original.

Secondly, I advocate for academic journals to require authors to submit copies of all referenced materials along with their articles. While I hope the Internet Archive remains accessible indefinitely, it has faced numerous legal challenges regarding copyright for the materials it archives. Supporting the Archive in various ways is essential, but relying on a single source for reference material is risky. Storing copies alongside published articles could ensure their longevity, ideally as supplementary material.

Navigating Copyright Challenges

Copyright laws complicate the storage and potential republication of reference materials. Academic publications are typically behind subscription paywalls, though open access is gaining traction. I aim to publish my work under open access licenses, but this is still not the norm and can be costly. I managed to secure funding from the Imperial College London Open Access Fund; otherwise, I would have faced processing fees of around £1,000.

Despite these challenges, I believe the principle of fair use should apply to the republishing of reference materials, and journals ought to adopt this practice routinely.

My Commitment Moving Forward

In light of the lack of action from publishers, I will take the initiative to save references to the Wayback Machine, even if they do not currently link to it. This will create accessible copies for future readers.

For every article I write, I plan to archive all reference materials and upload them to a public GitHub repository post-embargo. This repository will include the published article and copies of all references. It will be archived to become read-only, and I will upload it to my Internet Archive account, triggering a save with the Software Heritage Archive.

For those using WordPress, automation is possible. I utilize the LH Wayback Machine plugin, which automatically submits all WordPress content to the Internet Archive whenever changes are made.

Preserving important resources is crucial in an ephemeral online landscape. This doesn’t mean hoarding data (like emails); rather, it entails thoughtfully considering what reference materials may be relevant in the future.

For further insights on this topic, check out this episode of the a16z podcast featuring the founder of the Internet Archive.

George R.R. Martin discusses his references to the Grateful Dead in his works, shedding light on the connection between literature and music.