Re: [Groff] Werner's Margin Notes]

[Peter Schaffter]

On Sun, Jan 16, 2005, Werner LEMBERG wrote:
> 
> > I created 2 files, MN.tmac and werner.t and then ran,
> >   groff -ms werner.t > werner.ps
> 
> I suggest to add the command line option `-ww' to see all warnings.
> Then you'd probably seen the problem by yourself.
> 
> > Also, is there any "standard" support in groff for margin notes.
> > i/e, any standard macro package, which makes creation of margin
> > notes simple.
> 
> I don't think so.  Even the new `mom' macro package doesn't have
> support for margin notes, as far as I can tell -- maybe you can
> convince Peter to add this feature...

I foresee some problems with this, since I'm not sure how people
would want to use it.  I don't presently have need of margin
notes--never had--so I'd need some suggestions.

If I created a macro called MN (Margin Notes), I imagine it would
store the margins note and associated ev in a diversion, then, when
finished, would output the margin note *starting on the baseline
that was current when MN was called*.  Page overruns would be
handled by the bottom- and top-of-page traps.

i.e.,

    a line of text
    .MN
    <text of margin note>
    .MN OFF
    more lines of text

would output the first line of the first margin note on the same
baseline as "a line of text".  If the note ran long, only the
portion that fit on the page would be output, with the overflow
being output starting at the nominal first baseline of running text
on the next page.

Problem with that scenario is, what happens if users go margin note
crazy?  There'd be no way, in user space, to know if the text of one
margin note ran so long that a second margin note on the same page
might overprint the end of the first.

Which would mean that all margin notes for a given page would have
to be collected in a diversion, then output when the bottom-of-page
trap is reached.  This would remove any possibility of overprinting,
but could well result in margin notes that aren't located near the
text to which they refer.  In effect, they'd become like vertical
footnotes, except they wouldn't have any markers in the text itself
to identify them.

Then there's the problem of margins.  In the event a user didn't
set up left margin large enough to permit decent typsetting of the
margin notes, should a warning be issued?  Or should printing carry
on anyway (with ghastly results and possibly reams of wasted paper)?
And what would constitute a reasonable margin?  Seems to me some
correlation would have to be made between the type point size and
the available line length, in order to determine whether "decent"
typesetting was possible.  Not the easiest judgment call in the
world to make.

All of which is to say, if I get a better idea of how people would
use margin notes functionality, I'll add it to mom.  But I do need
feedback, since, as I see presently see the issue, it strikes me
as a lot of trouble for not particularly useful results.  All
suggestions welcome.

-- 
Peter Schaffter
  Author of _The Schumann Proof_ (RendezVous Press, Canada)
  http://www.golden.net/~ptpi/theschumannproof.html