With the right belts I could see it working alright, probably not worth the extra $ but who knows. I like the idea of the padded block and ramp but again, that might not work so well in action.
Here's the unit I'm working on - works great but is a PITA to shim the belts. Proof of concept but I don't know how much further I'll take it. In practice is a modest improvement over freehand on the belt itself though does offer a larger margin of error via some positional reference you don't get on a plain belt, more controlled grind rate and cooler working temps.
[video=youtube_share;fO4ZDnloxiA]http://youtu.be/fO4ZDnloxiA[/video]
Ultimately I might try to make something more along the lines of the Worksharp WS3000 that uses a spinning horizontal platter and avoid belts entirely, would give a lot more flexibility to apply different approaches. Then again, it really isn't that difficult to shim the belts and I'd only use it for initial heavy repair reprofiling anyway...