This question wouldn't matter to me if I didn't have so much respect of what he has to say. He does his homework, is willing to question and test present practice and comes up with useful recommendations.
How he presents it seems to me to be designed to make the material look crowded, harsh and impenetrable on the screen - a problem I believe is easy to solve. See what you think when you compare these two presentations:
(I haven't done a usability test on the two and would like to know what you think - send me a note with your thoughts and comments.)
And then there way it is written. HIs tone is strident and he hectors his readers. There is no shading and little subtility, no feeling for quirks and few exceptions to the rules.
Jakob Nielsen is also empiricist. He believes in finding out how things work by trying it out on people and seeing what they do, not what he thinks they should do.
The empirical tradition, as I understand it, is never completely sure of anything. It about probabilities, not certainties.
Statements like: "How Users Read on the Web They don't." or "users only care about content" seem to me to go completely against the skepticism that is at the center of what Nielsen practices. In all fairness, the detail is there about how many subjects were involved, what the test was and the conditions it was done under. It sounds like there are plenty of reasons to go on with tests to see if what he found is true using different populations and different material.
It also sounds like the directions he gives make good rules of thumb to follow when writing for the Web.
I only wish he would follow his own advice, especially, "Users detested "marketese"; the promotional writing style with boastful subjective claims ... promotional language imposes a cognitive burden on users who have to spend resources on filtering out the hyperbole to get at the facts."
Instead he should attend to the truths Ellen Ullman recalled when she was listening to Linus Thorwalds:
We were reminded that software engineering was not about right and wrong but only better and worse, solutions that solved some problems while ignoring or exacerbating others. That the machine that all the world seems to want to see as possessing some supreme power and intelligence was indeed intelligent, but only as we humans are: full of hedge and error, brilliance and backtrack and compromise.
Truths that don't lend themselves to bulleted lists, but that do place our opinions in a better light, the light of something less than absolute certainty.
