The author addresses 'blind belief' in computer results by offering a useful critique that 'important details are overlooked due to the flood of information' from the output of computer calculations. Indeed, errors in the numerical model may lead in extreme cases to structural f…