How to Future-proof Your Bibliographies Against Link Rot

Bibliographies shouldn't come with expiration dates

Two women holding a hyperlink symbol, vigilant about protecting their bibliographies against link rot

We librarians know all about link rot: how URLs listed in a document are likely to become inaccessible over time, like lights in a building turning off one by one. Link rot happens for a variety of reasons, from simple URL changes to content takedowns to shifts in domain name ownership.

While a broken link may be a small irritant in day-to-day life, link rot as a whole negatively affects the scholarly record. If an academic article cites resources that can no longer be found, it becomes impossible for the reader to check the sources to ascertain whether the author correctly interpreted them. Worse, if a study’s link to its dataset breaks, the study can no longer be reproduced. Referencing evidence and previous scholarly work is a core tenet of academic research, and link rot interrupts the continuity and trustworthiness of the scholarly conversation.

This is a serious issue that many academics have studied in depth. One recent study by Niveditha, Kumbar, and Kumar (2022) examined over 50,000 URLs cited over a period of nine years in some of the top LIS journals. They found that, on average, 23% of the links were broken. For more recent articles, that percentage was around 10%; for older articles, it was up to 50%. We can extrapolate that, over time, more and more links in that set of articles will go rotten.

So what’s an academic author to do if they want to protect their work from link rot?

For scholarly sources, use the DOI in your citation

This is probably an unsurprising suggestion, given that it’s very much the norm these days. Reference styles like MLA and APA strongly recommend using the DOI (digital object identifier) rather than a URL in a citation. Even if the journal changes its URL, its DOI will remain the same and provide access to the new URL, meaning that readers are much less likely to encounter broken links when clicking a DOI.

Example: the study I cited previously is accessible from both a URL and a DOI.

(Note: articles like the one above may be behind a paywall, which is a whole different access issue from broken links.)


🌟 Subscribe to the LibTech Insights newsletter for weekly roundups and bonus content, including: 


For non-scholarly sources, capture the URL in a web archive

If the source you’re citing doesn’t have a DOI, then I highly recommend submitting the URL you cited to a web archive as a preservation measure. Do this around the time you’re writing or publishing your work.

How do you do this? My preferred web archive is the Internet Archive, which, among other things, takes reliable snapshots of URLs with their tool, the Wayback Machine. In its current interface, there’s a “Save page now” option on the page where you can paste a URL. Once you hit the “Save page” button (you can skip signing in), it takes a small amount of time for the URL to be saved, and then you’re presented with the new URL of the archived snapshot.

Example:

Note that not all URLs can be saved this way. Some website owners prohibit tools like the Wayback Machine from saving their content. And not all snapshots of a web page may look or behave exactly the same as the live web page.

There are other options for preserving web content, like Perma.cc. Your institution might run its own web archiving service, too.

Consider citing the archived URL in your bibliography

Now that you’ve saved the URL in a web archive, what next? You can just leave it at that, and hope that if the original URL breaks, your reader will guess that they can find it saved in the web archive you chose.

Another option is to cite the archived URL in your bibliography alongside the live URL.

Example citation:

This doesn’t strictly adhere to common reference styles, but anecdotally, I’ve had no issue with formatting my bibliographies this way in academic journals. This approach provides your reader with an easy way to access the URL snapshot. As a bonus, they’ll be able to see the page as you saw it when you cited it. This is particularly useful for web pages that may change frequently.

Consider saving your own work in a web archive

Finally, if you’re publishing something that won’t get a DOI, I strongly encourage you to save an archival snapshot of your work once it’s published. I do this with the LibTech Insights articles I write (including this one!). That way, even if the original publication location is no longer accessible, you’ll be able to access your own copy of it.

Bibliography

  • Niveditha, B., Mallinath Kumbar, and B. T. Sampath Kumar. “Rotten Web Citations Cited in Scholarly Journals: Use of Time Travel for Retrieval.” Aslib Journal of Information Management 74, no. 2 (2022): 225-243. https://doi.org/10.1108/AJIM-05-2021-0139

🔥 Sign up for LibTech Insights (LTI) new post notifications and updates.

✍️ Interested in contributing to LTI? Send an email to Deb V. at Choice with your topic idea.