linkrot

[Baker, John]

        Anyone who's paying attention has seen plenty of examples of how
unreliable Googlecounts are.  A large or small number of Google hits is
a quick and dirty way to get some idea (not always accurate) of a term's
popularity, but it's a mistake to see it as more than that.  At the end
of the day, there's no alternative to actually looking at the uses.

        "Link rot" has been around longer than I would have guessed,
considering that I don't recall previously running into this useful
term.  From the newsgroup bit.listserv.qualrs-l, Google Groups,
11/1/1996, from a list of "new definitions in the high tech world,"
"Link Rot" is defined as "The process by which links on a web page
became as obsolete as the sites they're connected to change location or
die."


John Baker


-----Original Message-----
From: American Dialect Society [mailto:ADS-L at LISTSERV.UGA.EDU] On Behalf
Of Barnhart
Sent: Friday, June 13, 2008 8:49 AM
To: ADS-L at LISTSERV.UGA.EDU
Subject: Re: linkrot

American Dialect Society <ADS-L at LISTSERV.UGA.EDU> writes:
>linkrot. The one-word version gets 26,000 hits; the two-word version
>(link rot) gets 461,000 hits. A real winner of a term, I'd say -- I'm
>sure to use it with a certain frequency.

This computer term is interesting from several linguistic viewpoints.
However, I wonder at the number 461,000.  How many are totally
irrelevant?
 Knowing the penchant for Google to produce duplicates, how many are the
same?  Don't get me wrong.  Google is an important resource.  But, for
me it is way down the scale for reliability and usefulness from more
traditional sources.

I have come to the conclusion that the fascination with Google numbers
for suggesting that a new word is "established" or at least
"establishing"
derives from the ease with which impressive (i.e. large) numbers can be
generated.

Far more impressive is the status of the term in such traditional
resources as Nexis (in LexisNexis), where some conceptually manageable
(i.e. comprehendible) numbers may be generated.

word form  ....  major world papers  ...... all news (including blogs)

linkrot ...........7  ........................................... 37
link rot ..........5  ..........................................  88
link-rot ..........0  ..........................................    0

So, _link rot_ and its variant _linkrot_ are a significant new term, but
hardly by the weight suggested in the unmanageable numbers presented by
Google searching.

The newly-revealed resource at Brigham Young (americancorpus.org) shows
3 hits for _link rot_ but 0 for _linkrot_ and _link-rot_.

(See the "Working Knowledge" in _Dictionaries--Journal of the Dictionary
Society of North America_ [No. 28, 2007], p 131-162).  Here you will
find comments by myself, Orin Hargraves, Ian Brooks, John Simpson, Ieda
Maria Alves, Victoria Neufeldt, and Allan Metcalf.

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org