Re: [Cellar] [Tools-discuss] non-ascii characters (fwd) Dave Rice: non-ascii characters

[Dave Rice <[e-mail]>]

Hi all,

> On Aug 30, 2019, at 4:00 AM, Carsten Bormann <[e-mail]> wrote:
> 
> On Aug 30, 2019, at 09:39, Julian Reschke <[e-mail]> wrote:
>> 
>>> I’m not aware of an opportunity for that (well, maybe except for the choice between en-dashes and em-dashes for the punctuation mark in the form of a horizontal line used to indicate a pause, to delimit an inserted sentence or phrase, or to indicate a deliberately omitted word).    But given that this is the IETF, I’m sure we’ll find one :-)
>> 
>> AFAIR, there are also different opinions about whether there should be
>> whitespace around it.
> 
> Right.
> 
> Traditional German: em-dash with whitespace
> “Modern” German: en-dash with whitespace
> Traditional US English: em-dash without whitespace, or sometimes with (AP style, i.e., news)
> Traditional UK English: it’s complicated
> Today’s English: hodgepodge
> Laypeople: essentially, use dashes only where Microsoft Word turns a hyphen into a dash automatically, and that involves typing spaces, or so
> 
> (And, of course, there are range and versus dashes as well, which are always en-dashes, but might incomprehensibly have whitespace in English.)
> 
> I don’t have my CMOS with me, or I would cite chapter and verse.
> But the important observation is that this is a well-known editorial style issue to which a copy-editor should have an answer.
> 
> More generally speaking, the fact that Unicode has more characters does not mean we need to spend a lot of time deciding which character to use.  I think we are also old enough to no longer fall into Unicode traps (e.g., Ω vs. Ω).

Thanks very much for this discussion, based on this and also running idnits, I’ve created this pull request for the FFV1 draft, https://github.com/FFmpeg/FFV1/pull/158/files <https://github.com/FFmpeg/FFV1/pull/158/files>. Related to this discussion the pull request changes the unicode minus sign to an ascii hyphen and uses a "ceil(a)" and "floor(a)" function rather than using "⌈a⌉" and "⌊a⌋”. After this PR, it passes idnits with no errors, warnings, or comments.

Thanks much,
Dave Rice