I have a diagram with a bunch of boxes on it. Each box represents a part of a computer system that will evolve over time. I’m using colors to represents the different stages of evolution for each box. Each box can evolve independent of the other boxes. So I have a base diagram that shows all boxes, at the beginning of time, with no colors (gray). From that base diagram I want to build a sequence of diagrams where the colors of various boxes change sequentially over time. The trick is that we have to make frequent updates to the base diagram, where we add/subtract/move boxes as we learn more about the system.
To represent the base diagram, I’m using a shared layer, and I’m using a separate canvas for each scene in the sequence. For each box that changes in a given scene/canvas, I copy the box from the base (shared) layer, into the sequence layer so that it precisely overlays the box in the base layer, and I then change the color on that copy of the box.
The problem with this approach is that if I move or delete boxes on the base (shared) layer, I then still have to go to each canvas and delete any overlays for that box. It seems to me that the ideal solution to this problem would be if I could somehow use a layer to not introduce any new objects, but only to style objects on the layer below. Is something like this possible?