That makes a lot of sense: most WYSIWYG tools end up presenting a UI that cares more about appearance than semantics, so if you actually care about building good semantic HTML and complementary presentational CSS, most WYSIWYG tools utterly fail. So if a designer cares about building fluid layouts and semantic content rather than pixel-accurate recreations of Photoshop images, it makes sense that they'd feel very restricted in a WYSIWYG tool.
True, and the even harder part is building a system that treats that as a feature rather than a bug.