The limited number of cells, the use of up/down increments to change values instead of typing, the lack of row/column coordinate system, the lack of any kind of an apparent way to operate on multiple cells at once, the fact that the cells don't actually look like cells... it's a lot easier to come up with ways it doesn't resemble a spreadsheet than ways that it does.
Sure it is a proof of concept and is about the spreadsheet computational model not making a full fledged spreadsheet application like Microsoft excel.
I thought you meant you saw some significant barriers to a full implementation or something fundamentally different in the underlying model that was not spreadsheet like.
