Neilsen tested only speed if you read only the first half of the results. You might want to read the next one as well. It's called "User Satisfaction".
He doesn't say how long he asked people to read for. My suspicion is that the longer the timeframe, the more results between print and monitor (and Kindle and iPad) will diverge. You don't get eyestrain from even light reading on a monitor.
Also, satisfaction can mean a number of things, and he didn't define it, as evidenced by comments about weight. A lot of people could have had higher reading satisfaction on an iPad because it's an iPad.
Neilsen's results have no bearing one way or the other on my original statement, which is that people don't buy iPads to read books.
User satisfaction includes all of the relevant issues for a person, and in this test they were testing reading, so eye strain would matter to people.
People absolutely buy the iPad to read books. It may not be the only thing they do with it-since it is a multi function device. However reading books is a very large component of the purchase for everyone that I know that has bought one.
My graduate program just bought 65 iPads for the incoming class to read their books and PDFs. They bought those based on user feedback that people hate reading on the tablet and don't like hauling around forty seven pounds of dead trees. The majority user request was to load the reading on an iPad.
Nielsen's tests very emphatically disagree
http://www.useit.com/alertbox/ipad-kindle-reading.html