Re: Raw UTF-8?

[Ingo Schwarze <schwarze@usta.de>]

Hi Jonathan,

J.C. Roberts wrote on Sat, Jul 10, 2010 at 11:11:18AM -0700:

> UTF-8 support is very important and many consider it a "requirement"
> these days.

Not for manual pages, i just don't see the point.

That said, i wouldn't oppose a -Tlatin1 or -Tutf8 output mode merely
for groff compatibility, as long as it is not too intrusive (which
probably it need not be, implementation will be quite local in one
corner of the terminal output frontend).  But i have no plans to
implement, use or maintain it, and i would test it only in so far
as it must not break anything else.

Also, i would continue urging people to not use it, as manual pages
relying on it would sacrifice portability for no good reason.

And, of course, i will strongly oppose 8-bit-character input.
I'm not willing to deal with multi-byte or wide character support
functions anywhere in mandoc's code.

> Personally, I think the \(:o syntax is nonsense.

Agreed, the reason being that there is no reliable way to render it.
Again, i consider it provided for backward compatibility.

> Since we now have UTF, it seems better to error out on the archaic \(:o
> syntax to prompt change, rather than support it to prolong a bad idea
> and yet another syntax everyone needs to learn.

No, it is used in too many places, and it would not be nice to deny
rendering just because some piece of mdoc(7) or man(7) source code
contains syntax we don't like.  We are not defining new standards
right now, we are re-implementing an existing language.

> There's probably an existing IF-THEN-ELSE which could be leveraged
> without undue overhead, but you get the main idea... --make the author
> provide both the THEN and the ELSE.

You mean, like in
http://www.openbsd.org/cgi-bin/cvsweb/src/share/man/man4/sppp.4#rev1.20

I don't consider that viable, and jmc and myself agree to remove
it from the tree when we find it (unless it comes from upstream).

I doubt that people will regularly provide alternatives.
Typing in special characters at all is already tedious,
providing alternatives will not get done.

And even if it is done, the result is incredibly ugly
and hardly maintainable.

> In many situations, even when a terminal is capable of displaying the
> UTF-8, it could still be beneficial to also display the ascii, possibly
> in parenthesis.

In my opinion, you are vastly overestimating the importance of special
characters in manual pages (beware, i'm not talking about typesetting
of mathematical papers here!) and you are vastly underestimating the
importance of portability and simplicity.

Yours,
  Ingo
--
 To unsubscribe send an email to discuss+unsubscribe@mdocml.bsd.lv