Re: [Groff] inconsistency between .R and \*R in man.tmac

[Tadziu Hoffmann]

> Quite frankly, I cannot conceive of a reasonable situation where
> ".R" for Roman would be needed.  If your base font is R (font position 1),
> all macros should process their text in their own font, then restore font
> 1 upon completion.  Therefore there should be no need to "force" an R font.
> I'm thinking in terms of ".B", ".BR", ".IB", etc.

I was speaking in terms of consistency.
The manpage macros employ a one-input-line trap for switching back
to the roman font after a font-change macro without arguments:

  roman text
  .B
  bold text
  roman text again

Other macro packages, however, require you to explicitly switch
back to the original font (you have to "know" that the default
font is roman):

  roman text
  .B
  bold text
  more bold text
  even more bold text
  .R
  roman text again

[snip]

> It is important in such instances to specify font in terms of font position
> rather than using "R".  That way you can change the fonts in an entire
> production by changing a few lines in the macro file instead of searching
> an entire book.  Using "\fR" instead of "\f1" would give Times Roman
> instead of returning to New Century Schoolbook Roman.  That'll really
> screw up a production manual.

I thought the correct way to handle this was

  .fp 1 R NewCenturySchlbk-Roman

and from then on both \f1 and \fR would refer to the same font.

Anyway, this should mostly be hidden from the user:
macros use low-level requests like ".ft 2", and the
user uses high-level macros like ".I".