Re: [Groff] rationale for italic correction mechanism?

[Dave Kemper]

> I want the typography that *I* set, not someone else's. If the price
> of this freedom is more manual intervention then it is a price I shall
> gladly pay.

Robert,

This is absolutely a worthwhile principle.

However, if you are using groff, you are already using typographic
principles that someone else has determined.  You're allowing your
typesetting software to paginate, to fill lines, to kern, to convert
character sequences to appropriate ligatures, to stretch spaces for
justification, to hyphenate.  If you wanted to do every single one of
those things manually, you wouldn't be using groff; you'd be writing
raw PostScript code.  (-:

The power of groff is that it gives you total control over these
things -- any of those features can be turned off, and you can manually
adjust almost anything you desire -- while still producing *by default*
typography following what are widely considered to be best practices.

We use groff, in general, because we don't want to have to manually
insert a horizontal movement command every time a capital A and capital
V appear next to each other.  But for anyone who *does* want to do that,
groff will happily let you turn off kerning and manually adjust away.

I have not heard of anyone championing italic and roman glyphs that
overlap each other.  If there are people who do so, they can be easily
accommodated by providing a request that turns off italic corrections,
just as groff provides requests to turn off kerning, ligatures, and
other things that most people want on by default.

My argument is not to take control away from users, but to provide
sensible default behavior.  If you can make a case that overlapping glyphs
in certain roman/italic transitions is sensible default behavior, and that
every such case should require individual correction rather than a global
setting saying "fix all of these," I would be quite interested to hear it.