The latest fake but fun reply to a student reminded me of an article by Jay Schalin I found linked from National Review Online's education blog, Phi Beta Cons.

[Full disclosure: I was a student at the University of Washington, and did get more information about classes using this system. However, said classes either had only one section in a given quarter or were math courses that were exceedingly similar across sections, so it did not benefit me as much as it would others.]
So let's run a cost-benefit analysis:
Professors would have to put up some information for the class when students can register for it. If the professor knows in advance that he or she has this course, the professor has taught this course in the past, and does not want to change the course materially from last time, then this cost is actually fairly low. (As a non-professor, I have no way to evaluate how likely it is that they would know enough in advance which course to teach.)
Students could use this information to register for a class that they are more interested in, think they would do better in, or both.
It would raise transparency levels as to what a class taught, and by extension what a degree is worth.
One limitation is that this seems to be more useful in humanities and social sciences rather than hard sciences. I would expect that it is difficult to personalize math, engineering, and physics courses.
I am sure that there are other costs and benefits that I am missing, and those would certainly be valid points. Mr. Schalin's full report deals with more than I've presented here.
So is this a good idea? I think so. I would love to have more detailed descriptions for the courses I am taking for my master's. And if I teach, as I hope to, I would provide that information for students who are interested in the classes I'd have if there is a place for it.