Re: [Groff] Handling '...'

[Tadziu Hoffmann]

> Looking back further than just a few years ago, we find that
> texts set in metal (hot and cold) tended to use the full word
> space out of pure convenience.  Individually adjusting the
> dots in an ellipsis was simply too much fussy work...

Hmmm.  I have no experience with metal type, but I thought
that's what they had clichés for (if they needed them).


> [...] since typographers and designers had never been happy
> with the full word space [...]

Any citations on this?  The typophile forum seems divided
on the issue.  Chicago apparently specifies a 1/3 em space
between dots.

I prefer a normal word space (with an equal space to any
attached words for consistency) because this preserves the
rhythm of the line.


> In texts of a technical nature, where an ellipsis indicates
> an established sequence that continues, (e.g. a, b, c...),
> the ellipses should, in fact, be fairly tight.

Well, in math they're usually not very tight.  TeX by default
appears to use something like 3/18 em between dots (which
corresponds to troff's "\|"), and as far as I've understood,
this is not based on convenience but rather on a long history
of math typesetting.


> And since you don't want ellipses joined to a word such that
> groff interprets both the word and the ellipsis as a whole
> word and thus breaks the line *before* the word (which can
> lead to gaping holes in the broken line), a useful string
> for ellipses should include a zero-width break point.  Thus
> 
>     .ds ellipsis \:\*[FU4].\*[FU6].\*[FU6].\*[FU8]\"

Wouldn't that leave an unwanted space at the beginning of the
next line?