I'm not sure how a source from 2002 proves false a citation of a source from 2007.
Edit: Look, whoever keeps downvoting this, would you PLEASE read the sources? Is that too much to ask? See my other comment below, explaining how the article arrived at its number, and why it doesn't sound right.
It's really disappointing to see HN devolve into such sloppy thinking.
http://answers.google.com/answers/threadview/id/43888.html
2002 was the last year they had full data for the series ID given there (I checked at the Bureau of Labor Statisics), but it had been stable for a few years.
I wouldn't be surprised if blogging crosses over within the next couple years. It's a very labor-intensive profession, and as the article notes, many people will pay for a friendly opinion.