bug#16292: 24.3.50; info docs now contain single straight quotes instead of `'

[Paul Eggert]

Eli Zaretskii wrote:
> These are very rare (and I would argue will look ugly any
> way you typeset them).

They're not that rare: for example, I count 949 info lines
containing both straight and the curly apostrophes, with
many opportunities for confusion.  ‘C-x '’ may look ugly on
some displays, but it's typically legible, and it's
definitely easier to grok than 'C-x ''.  The primary goal
here is conveying information, not beauty.

> I don't see the problem: just don't edit any letters, only edit
> apostrophes, quotes, and arrows.  What am I missing?

Info generates lots of special characters like that, in
response to ASCII markup.  From the unibyte reader's point
of view, why should the output of "``", "@quoteleft{}",
"@expansion{}", "@result{}", etc. be converted from mojibake
to ASCII, while the output of "--", "@bullet{}", "@minus{}",
"@equiv{}", "@~n", etc. remains mojibake?  I don't see any
systematic principle to distinguish between the two sets of
characters.

> > The point of cp-ascii is to not put mojibake on unibyte users'
> > screens, so why not fix all the mojibake while we're at it?
>
> To make it more acceptable to UTF-8 locales.

UTF-8 locales work just fine (actually, better) with the
original UTF-8 characters, so it's not a priority for
cp-ascii's output to be more acceptable to UTF-8 locales.
It's fine if cp-ascii's output is just as acceptable for
UTF-8 locales as it is for ASCII locales.