I suspect it's roughly the same as a crawer -- same issues of fair use, TOS/CFAA, etc -- but likely there's no expectation that humans will read and follow robots.txt.