For the category of people you describe, I agree this has been the case in the past, and perhaps even the case at present. I think (or perhaps hope) the future

For the category of people you describe, I agree this has been the case in the past, and perhaps even the case at present. I think (or perhaps hope) the future may hold something different though. There are some industries such as bionformatics where the boundaries between science and scientific computing are being blurred. Whatsmore, I think this blurring of lines between science and computing will continue as a generally trend. Steven Wolfram (Mathematica's creator) articulated this, albeit poorly and hubristically in a "New Kind of Science". That work gained rightful criticism from the majority of the scientific community, but it did articulate a clear trend - scientific research and computation are becoming increasingly intertwined. We may not differentiate so sharply between the types of work you describe in future, and perhaps value scientific programming more.