> Scientist should...

Have you seen the state of scientific "computing".