> But: A short survey showed that everyone I asked uses "
> instead of u2033.

Well, laziness shouldn't be an excuse for not doing it right
(it's not as if getting the right character would involve a
disproportionately huge effort), and reduction of ignorance
seems to be the main purpose of your paper.


> I notice that there is a \(fm (foot mark) request, but
> couldn't find a \(in (inch) or \(im (inch mark) request.

Ah, yes, this is handled somewhat inconsistently.  Usually
you use one prime for feet and two primes for inches,
and one prime for minutes and two primes for seconds.
The signgle-prime character is called \[fm] (foot mark).
and the double-prime character is called \[sd] (second),
Strange, huh?  (In the Postscript "Symbol" font, they're
called "minute" and "second".)


> How do you write inch in an english text? \(fm\(fm?

I'd say this is perfectly okay, too (but the above is easier).
I don't think the slightly larger spacing is objectionable
(you could kern them a bit if you wished), and as far as I
know, the original troff only defined the foot mark.


> I will change that into:

I'd suggest something along the lines of:

  Die Schreibmaschinen-Anführungsstriche " auf der Tastatur
  sind keine \[Bq]richtigen\[lq] Anführungsstriche.
  (Es sind auch keine Zollzeichen, die sehen so aus: 5\[fm]3\[sd].)
  In deutschen Texten ist es üblich, Anführungsstriche unten und
  oben zu setzen. ...

and

  Das \f[CR]smartq\f[]-Makro macht aus den
  Schreibmaschinen-Anführungsstrichen " mehr oder weniger
  intelligente \[Bq]richtige\[lq] Anführungsstriche.
  (Die Makros sind in diesem Dokument definiert.) ...


> ... definiert die Strings \\Q und \\U (Quote und Unquote).

I'd write these either as just "Q" and "U" (the names of the
strings) or as "\\*Q" and "\\*U" (the escapes to insert the
strings' contents).  "\\Q" and "\\U" is incorrect.