i don't think i disagree with you and (i think) neither does the author. argument (as i understand it) is that rather than piling on features that don't really

i don't think i disagree with you and (i think) neither does the author. argument (as i understand it) is that rather than piling on features that don't really fit and have complex interference you could use a better tool.