I think what you said about 'trying to optimize it' is actually on the spot. I think what too many people try to do is to 'maximize' or 'minimize' it without really understanding the impact of designing based on metrics. But I think what we can do is use a metric to understand some of the impact of our design decisions, and help clarify why it might make sense to do something. Even the best designers will know when it makes sense to break rules (because there are often good reasons to do so).

You are right in that there is a lack of empirical evidence in how to apply the data-ink ratio to design. This is one of the reasons why I am trying to sift through the previous research that makes claims without really testing them properly, so that I can hopefully make better sense of it with better research.