Re: [Groff] Tweaking font metrics

[Ted Harding]

On 22-Oct-06 Gunnar Ritter wrote:
> (Ted Harding) <address@hidden> wrote:
> 
>> In other words, 
>>
>>   .kp Font c1 d1 n1  c2 d2 n2 ...
>>
>> would extend the kern-pair table for font Font by the lines
> 
> I have introduced a ".kernpair" in Heirloom troff. It is
> similar to your proposed groff request but also allows to
> specify different fonts for the characters in a kerning
> pair, i.e. the syntax is
> 
>   .kernpair font1 char1 font2 char2 distance
> 
> The added functionality is quite useful e.g. to solve the
> problem of an italic "f" character that intersects with
> roman parentheses surrounding it (and contrasting to any
> half-solutions such as inserting "\/" or a motion command,
> one does not have to write it every time the problem occurs
> in the document text).

Yes, that is a useful "extension" to my proposal. And you
are well on the point when it comes to \fIf\fP)!

>> zero length, then
>>
>>   .shc \[h0]
>>
>> would have the effect that the hyphen at the end of a hyphenated
>> line would protrude be the full length of \[hy].
> 
> This proposal does not actually solve the problem of hanging
> punctuation since e.g. an em dash should certainly also extend
> beyond the right (or even the left) margin but should of course
> retain its width when it appears between two words on a line.
> Since you do not know in advance where the line breaks, you
> cannot use a static width change here.

Yes, I'm well aware of that aspect! (Which is why I confined
the discussion to the "hanging hyphen" since we do know where
that will come in).

Indeed, I'm experimenting with combining normal hyphenation
with "hanging" hyphenation in the following contexts:

a) A line containing a naturally hyphenated word (or rather
   word-pair) is broken at the hyphen. In this case, the hyphen
   does not "hang out".

b) A word is broken in the middle and a hyphen is introduced.
   In this case the hyphen does hang out.

Example: Compare

  ...                                   It investigates the
  associations  between four aspects of family life in mid-
  adolescence (family structure, reason for family disrupt-
  ion,  time spent in  joint family activities  and parent-
  adolescent conflict) and a range of indicators of current
  lifestyle  and future  life  chances  (health-related and
  delinquent behaviours; ...

[normal hyphenation] with:

  ...                                   It investigates the
  associations  between four aspects of family life in mid-
  adolescence (family structure,  reason for family disrupt-
  ion,  time spent in  joint family activities  and parent-
  adolescent conflict) and a range of indicators of current
  lifestyle  and future  life  chances  (health-related and
  delinquent behaviours; ...

This looks better in print than it does on screen!]


The point is that in the first version the eye (at least, my eye)
is given a false expectation when it encounters "mid-" and "parent-"
at the ends of the lines, somehow anticipating a broken word
(e.g. "middle" or "parental"), perhaps put on the wrong foot by
the hyphenated "disruption". This false expectation does not arise
(at least for me) in the second version. (And, as I say, it looks
a lot better when printed).

As I say, this is an experiment on "perception of print".
The point for the present discussion is that the soft hyphen
has to behave differently from the normal hyphen.

Thanks for the contribution!
Ted.

--------------------------------------------------------------------
E-Mail: (Ted Harding) <address@hidden>
Fax-to-email: +44 (0)870 094 0861
Date: 22-Oct-06                                       Time: 20:15:23
------------------------------ XFMail ------------------------------