Re: [groff] [patch] modernize -T ascii rendering of opening single quote

[John Gardner]

*> It's if they choose, or are coerced, into a LC_CTYPE=C locale causing
-Tascii that they see asymmetric `'.*

Uhm. Is this issue more of a "would like to have" instead of something
objectively problematic....?

For a start, I see ``this'' in project readmes and plain-text docs all the
time, and Emacs Lisp even uses quote-pairs in docstrings to denote symbol
references.

Which means that plugging this issue is really a drop in the ocean —
there's countless more documentation out there, in plain text, which
readers have to put up with seeing on a regular basis. If all we're
catering to is ASCII-shackled users who hate the ``weird'' looking quotes,
well… these folks are free to define an alias, set MANPAGER to a wrapper
script that pipes formatted output through tr '`' "'", or hack on their
troffrc files with something clever like `.if #\*(.T#ascii# .char \(oq "'`
or something...

As somebody in this thread stated before me, we're spending a little too
much thought on what's essentially bikeshedding.

As an aside, I wrote a package for Atom
<https://github.com/Alhadis/Atom-XTermQuotes> that uses plain old CSS and
HTML to mask `this' to look like real directional quotes (and only making a
habit of targeting prose, not source code). Amazing how well it works. ;-)
Maybe every terminal should be supporting CSS instead?
— John

On Tue, 19 Feb 2019 at 23:14, Ralph Corderoy <address@hidden> wrote:

> Hi Jeff,
>
> > > Thus one that makes 60 and 27 look like a left/right pair of single
> > > quotes, and also grave and acute accents for over-striking.
> >
> > This is yet a third possibility, which the editor of the 1967 standard
> > said was also true of the earlier version.  They gave possibilities,
> > but intended no recommendations on which options to choose.
>
> Agreed.  But industry, to my recollection and Googling, seems to have
> settled on symmetric quotes that could double as accents to get the most
> use out of those two bytes.  The lineprinter we had, don't know the make
> other than it wasn't IBM nor DEC, also made them symmetric.
>
> > Perhaps an even more relevant consideration: what percentage of
> > displays today are in conflict with the ISO 646 encoding?  Shouldn’t
> > the objective be the greatest good for the greatest number?
>
> It's not the displays that matter.  Those seeing asymmetric `' that
> would like that changed have displays that can, and normally do, render
> with -Tutf8 and so see symmetric ‘’ instead.  It's if they choose, or
> are coerced, into a LC_CTYPE=C locale causing -Tascii that they see
> asymmetric `'.
>
> --
> Cheers, Ralph.
>
>