Nice idea, badly implemented. The interface teaches me how to use it very badly. The most obvious interface clues are the arrows to either side of the page, bu

Nice idea, badly implemented. The interface teaches me how to use it very badly. The most obvious interface clues are the arrows to either side of the page, but these work in the usual manner (hopping from page to page), so they distract me from the heart of your interface. When I do happen to mousewheel, I can't immediately understand what that does. Am I moving myself or a divider? Why is the divider coming from the top? What's above the divider? Why is it snapping like that, instead of scrolling smoothly? The snapping erects a level of indirection between my (smooth) wheel movement and the movement of the divider, which obscures my ability to understand what's happening. I can't decide whether to scroll upwards or downwards—which is "forwards?" Why is there a white page "behind" the text if I scroll the divider upwards? Is that what's on the next page or the previous page? Why are you showing me a white page? The sense that I'm reading downward yet new text comes from above feels very strange. I see how you've ended up here, but it's highly unintuitive. Alas, it's also fundamental to your concept, so whether the concept lives or dies (once all other problems are removed) is a question of whether people can adapt to this spatial "warp." The "snappy" movement of the page divider line feels clunky. Move per pixel, in immediate response to mouse wheel (or touch drag?) movement—preferably with inertia—definitely not per line. This is not the way I want the world to scroll. But I see where you're going: leaving the text in place makes sense, and if you're going to do that then using a divider from the top follows inevitably. So the idea is worthwhile, but this implementation is its own enemy.