I agree. For images, I use a definition list dd = image dt = description (in that order, since I want a bottom "caption")
FWIW, this is totally wrong. (For one, they won't be associated with each other.)
HTML5 has <figure> for images and similar documents which takes an embedded element and <legend> for a description.
-- Anne van Kesteren <http://annevankesteren.nl/> <http://www.opera.com/>
