On 11/13/21, Ralph Corderoy <ra...@inputplus.co.uk> wrote: >> What would an ideal model of output from the above input look like? > > What's the argument against CSTR 54's section 11.1 definitions being the > ideal?
Two are: - General typographic convention is to measure vertical motions in v's, not m's. CSTR #54's definitions of \u, \d, and \r do not align with this common expectation. - CSTR #54 is not even internally consistent in this regard: most vertical motions are specified in v's. \u, \d, and \r are outliers even in troff-land. Determining what historical troff did is tougher unless you have a Graphic Systems C/A/T lying around: a code analysis in http://savannah.gnu.org/bugs/?60602#comment1 concluded CSTR #54 was incorrect about troff's behavior on this point. But Heirloom troff's behavior, if a legitimate proxy for AT&T troff, calls that conclusion into question.
