The Chronicle reports that
An analysis of citations to Web addresses in scholarly articles on communications studies found that a third of the links were no longer active. The two scholars who conducted the analysis are preparing their findings for publication and coming up with a list of recommendations to stop what they refer to as the decay of online citations.
. . .
Mr. Bugeja and Ms. Dimitrova focused on links used by scholars in footnotes that cite Web materials. After analyzing more than 1,126 such citations, taken from online versions of five prestigious communication-studies journals, they found that 373 of the links, or 33 percent, were dead. Of the 753 links that worked, only 424 pointed to information pertinent to the citation.
Read the article (this link only active for five days).
Decay
This is a rather serious problem, and I'm curious what solutions they will propose. I've noticed that even in the course of writing and publishing an article, some of the vital links will die. This wouldn't be such a big problem if the articles and links I needed were released under a CC license; then I could make my own backup and include links to them. Of course, if something happened to me, then the problem would persist.
This is just showing us the difficulties of applying a citation method based on the print paradigm to the new scholarship. The persistent archive that's available to the print world just doesn't exist online. It's clear that no easy solution exists, but it would help drastically if more scholarly materials were CC'ed and thus easier to archive and mirror.
thumbs up
Definitely a good point for CC licensing of texts since many scholars would provide a mirror of the works they cite. It's another keypoint:
Open access = right to create archives.
Meanwhile--and Dave B started me on this--I now use Adobe Acrobat to capture webtexts that I use in my research. At least then I have a copy of what I am referencing.
decay as de cause
I certainly agree with Matt and Charlie that we should use this kind of research to forward the cause of archiving information for research needs.
But I'm not buying that it's as big a deal as the research is suggesting. A hyperlink is not the only part of a citation; just because the hyperlink is dead does not mean that the article is no longer available.
Only a couple of times in the last few years have I not been able to find an online citation after a bit of Googleing.
But hey, if *they* want to say it's a big problem, let's agree -- maybe it'll get us somewhere that the music pimps don't want us to go...
Pertinence
"Of the 753 links that worked, only 424 pointed to information pertinent to the citation."
Anyone know how the study measures "pertinence"? I've seen any number of print journal footnotes that would likely see similar characterization.