That Twitter had more downtime last year (84 hours) than any of 15 social network sites measured by an uptime monitoring service should surprise no one: The site's "fail whale" is so famous it was just featured in the New York Times.
However, what may surprise some Twitter users and industry observers -- me, for example -- is that the site's availability performance showed dramatic improvement over the second half of 2008, according to Pingdom.
And the same cannot be said for LinkedIn, which appears to have caught a case of whatever had been knocking Twitter offline so regularly.
Those are a few of the findings from a report out last week -- "Social Network Downtime in 2008. The report covers Facebook, MySpace, LinkedIn, Twitter, Friendster, LiveJournal, Orkut, Bebo, Hi5, Windows LiveSpaces, Last.fm, Classmates.com, Reunion.com, Xanga and eight months worth of performance from Imeem.
Five from that group -- including heavyweights MySpace and Facebook -- recorded outstanding uptime records of 99.9 percent or better; the other three being Classmates.com, Xanga and Imeem.
"The survey reveals several interesting trends," says Pingdom's Peter Alguacil. "For example, a full 84 percent of Twitter's downtime came during the first half of 2008, when the service was still struggling with stability issues. July and onward has seen a significant improvement for the service. LinkedIn, on the other hand, is having the opposite problem. Each quarter showed a larger amount of downtime for LinkedIn than the previous one."
LinkedIn had 45.8 hours of downtime last year, second only to Twitter on the dubious distinction scale. Their public relations department didn't respond to my request for comment.
Third most often offline was Friendster, which logged the most lengthy downtime episode -- 23 hours over less than three days in November -- thus marring an otherwise middle-of-the-pack uptime record.
My takeaway: I'm going to lighten up on Twitter. ... And perhaps those of you who use LinkedIn can keep an eye on them for me.