It's a story first, and a work of art second.
I suppose some (lots?) of people like art that makes no sense, but not me.
And I think Ebert never mentions Arthur C. Clarke because Ebert didn't read the book either, and like everyone else has no clue what's happening in the film.
There is an actual story here. A great one, and Kubrick was unable (or unwilling) to actually include it in the movie.
For anyone who hasn't seen it yet, do yourself a favor: Read the book first, then watch the movie.
In the book I read Arthur included a preface that explained how the book and movie were created together.
You are free to dislike someone's work, but to say it makes no sense is an objective statement about communication. There is an abundance of sense in 2001.
Feynman
I find it hard to believe that Feynman would apply that argument to works like Kubrick's 2001. In all of Feynman's speaking, he never took the stance that something must be explained entirely in order to be interesting. Quite the opposite; Feynman was most interested in the things that hadn't yet been explained. The things that no one else could explain.
The reason he was interested in things that hadn't been explained yet is because he wanted to find explanations - not because there is an inherent advantage to their remaining opaque.
Kubrick is considered a great director because he forces the audience to confront film as something other, and perhaps more, than simply a narrative vehicle. A Kubrick film is a visual experience, more like a photograph or a painting than a novel, and almost any given still frame from 2001 is beautiful enough to frame it and put it on your wall. (It seems important here to note that Kubrick started as a photographer before becoming a filmmaker.) In most of his films, but especially 2001, Kubrick composes these striking images together with just enough of a narrative to form an emotional impression on the audience without making it too overly familiar.
Consider some of the themes of 2001. Do you really think it makes sense to convey those ideas in a way that isn't somewhat opaque? How do you convey the mystery and terror of genuinely mysterious and terrifying themes by turning them into commonplace intellectual concepts and making the audience fully comfortable with them?
2001 conveys more than you can simply put into words. That's why it's a film, not a novel. The novel offers a more literal interpretation of the events depicted, but I think Clarke himself would be the first to tell you they are separate works of art and should be evaluated as such.
I watched the movie, thought it was horrible and incomprehensible, read the book, then watched it again and enjoyed it - somewhat, I kept getting annoyed at how unwatchable it was without the book knowledge. But except for the warp scene the visuals were great.
And no, you absolutely do not need to read the book to have a clue as to what is happening in the film. As Ebert points out, you just need to sit back and watch, and enough will be revealed.
or are you speaking of something different?
For those of you who haven't read Clarke's book 2001, you should consider putting it on your list. As garyrob has implied, it's very different from the movie - very concrete - and yet, possibly as a result, almost more hallucinogenic than the film could ever be.
http://en.wikipedia.org/wiki/The_Lost_Worlds_of_2001
I found one at a used bookstore ages ago and it was a fascinating look at the partnership and conflict between Kubrick and Clarke as well as a look at the different ideas and plots they discarded.
The Clarke novel is far more literal in spirit than Ebert's interpretation of the film. The most important difference is probably the bedroom. In the novel, it's a place where the aliens try to make Bowman comfortable while they study him. The novel describes how Bowman spits out the water he finds there, because it doesn't taste right: turns out it's too pure. The aliens knew he was supposed to have water, but they didn't understand about all the other trace chemicals we're used to. It's all very concrete.
That being said, there was a book about the making of 2001, and it talked about how there was some conflict between Kubrick and Clarke over the literalness issue. For instance, when the Star Child appears at the end of the movie, Clarke needed there to be concrete explanation of what he was doing there near Earth and why he was doing it, and suggested reasons to Kubrick, who was (in Clarke's description of the scene) "unimpressed." Clarke was also upset when he went to the first screening of the movie and saw that some of the explanatory plot was not there.
My takeaway is that Kubrick was more oriented toward 2001 being a visual poem, intended to convey his vision in a deep, evocative, non-literal way. And in that sense, it seems that Ebert's take on it is consistent with Kubrick's vision, if perhaps not consistent with what Clarke had expected/intended the movie to be.