Editing Link rot

'''Link rot''' (also '''linkrot''') is the process by which hyperlinks cease to point to their intended content because pages are deleted, sites shut down, URLs are reorganised, or domains expire. Closely related problems include [[Dead link]]s (immediate failure) and [[Content drift]] / [[Reference rot]] (URL still resolves but the substance has changed).

==Evidence==
A 2024 [[Pew Research Center]] data-lab study found that about '''25%''' of webpages sampled from Common Crawl snapshots spanning 2013–2023 were no longer accessible by October 2023, including '''38%''' of pages from the 2013 snapshot. Even among 2021 captures, roughly one in five were gone within two years. Separate analyses of news and government pages found substantial shares of pages containing at least one broken outbound link.

Follow-on research that re-checked Pew-style samples against the [[Wayback Machine]] has argued that many "dead" live URLs still survive in web archives—reframing ephemerality as a problem the live web has and memory institutions partly solve.

==Mitigations==
* Prefer [[Persistent identifier]]s (DOI, Handle, ARK) for scholarly objects
* Cite archived snapshots ([[Wayback Machine]], [[Archive.today]]) alongside live URLs
* Institutional [[Web archiving]] and link-repair bots (e.g. partnerships between Wikimedia and the Internet Archive)
* Cool URI design that avoids unnecessary churn ([[Cool URI]])

==See also==
* [[Dead link]]
* [[Reference rot]]
* [[Content drift]]
* [[Digital preservation]]

==Sources==
* Pew Research Center, "When Online Content Disappears" (17 May 2024).
* Academic and WADL workshop papers revisiting Pew samples with Wayback coverage (2025).
* Classic discussions by Zittrain and others on citation rot in journalism.

[[Category:Web archiving]]
[[Category:Concepts]]
[[Category:Internet culture]]