I concur. It also makes it much harder for us to introduce an outside package. Labeling as a layer is just fine.
In some sense, this is analogous to the distinction between data.frame and matrix in R (there are other historical diff's as well). The point is that labeling the mathematical object is a non-mathematical operation. If we know that we use the same dictionary, then we can defer all operations to the math layer and not worry about the results being right. On Thu, Mar 18, 2010 at 7:50 AM, Jake Mannix <jake.man...@gmail.com> wrote: > I personally agree that keeping the name as part of the mathematical > vector is wrong, >
