Did you read it?
From their research the colours used on the page actually have the best readability reaction times so they are in fact practicing what they preach. My first reaction was similar to yours but read it and you might be surprised.
Also, the line length is dependent on your screen size and will look best on an 800x600 monitor. Just adjust the window size and it looks better. The writers are academics not designers.
Yes, I did, albeit quickly, as it is especially interesting to me since I am a professional usability consultant. Actually, they admit that the theoretical results seem to conflict with observed reaction, which is not unusual. This stuff is very hard - but not impossible - to get objectively right.
However, what they measured was recognition reaction time - the weakness of the paper seems to be that it strongly correlates this with readibility, which is a leap too far.
If you're interested, another view can be found here:
Also, the line length is dependent on your screen size
no, it isn't. Or at least, it doesn't have to be.