Greg,
No, I didn't intend my comment as a dig (no offense intended at Mr. Digory Hadoake (sp) either). If it sounded otherwise, my appologies. I agree with your elaboration. Mine was more from the perspective that "knowledge and experience" was used in terms of previous experience and education applied to the current subject through association and extrapolation. When an engineering issue has numerous factors that influence the outcome, actual testing of the article with all the factors included sometimes is the best way available to assess the issue. (Example: when we build a new airplane, we do hundreds of tests even though we have much knowledge and experience with very similar designs from the past. This is because the interaction of many variables in the design create complexity in analyzing as opposed to testing the actual article) Hence my comment; "Or even better, getting real data."