The converter supports document segmentation, automatically adds a navigation bar, a (hyperlinked) table of contents and index as well as cross-references within the document. Formulas, and images that can't be converted into HTML are automatically converted into GIF images. Tables can be included in HTML-style or as images.
Minor probles such as only small images as anchors in links can be fixed with not too much effort by an external script, see [7].
However, when the converter needs to deal with many images e.g. when converting documents with mathematic contents, it starts mixing up these images and doesn't do it's job properly any more. Fixing this wasn't possible when attempting to, too, due to the chaotic internal structure of the converter (a set of perl scripts).