Re: Solidus variations

From: Hans Aberg <haberg-1_at_telia.com>
Date: Fri, 7 Oct 2011 20:34:17 +0200

On 7 Oct 2011, at 19:39, Jukka K. Korpela wrote:

>> There are several solidus (slash) variations.
> > What is the intent of those, in as much there been expressed,
>> in a mathematical context?
>
> Unicode mostly encodes characters that are in use or have been encoded in other standards. While not semantically agnostic, it is much less oriented towards semantic clarifications and distinctions than many people might hope for (and this includes me, some of the time at least).

I am aware of that, but by tracing the origin, one can get hints of usage.

>> For example, is U+2044 intended for rational numbers,
>
> Yes, and the idea behind it is that programs may format such a number in a manner typographically suitable for fractions, so that the adjacent numbers (digit sequences) are affected. This means that e.g. 1⁄2 may create a rendering similar to some of the glyphs for the character ½. You probably won’t see this happening in your favorite email program, text editor, web browser, or even word processor—it’s just the underlying idea and a possibility, not a requirement (or commonly implemented).
>
> On the practical side, it may happen that even by virtue of the different shape of U+2044 (vs. U+002F)—it’s typically in a 45° angle, though the implementation could be more complicated, even implementing it as horizonal—, fractions may look somewhat better. But there’s the risk that U+2044 is not present in the font that will be used (or cannot be transmitted when using many legacy non-Unicode encodings).

I am playing around with a small parser on top GMP, so input semantics is the main issue for me. Then I realized that if I let the lexer parse rational numbers, then 1/2/3/4 will parse as (1/2)/(3/4), and not as ((1/2)/3)/4 as expected in ASCII computer programs.

But using ⁄ U+2044 FRACTION SLASH as in 1⁄2/3⁄4 = (1/2)/(3/4) becomes unambiguous with 1/2/3/4 = ((1/2)/3)/4.

>> and U+2215 a long variation of U+002F,
>
> I wouldn’t call it long. Visually, it might be expected to differ from U+002F by looking specifically like a division operator (as it _is_ a division operator), as opposite to the semantically ambiguous U+002F. If it’s longer, I think it’s longer as a consequence of extending from the baseline to a specific height in a different slope than U+002F.

If it is not longer, then it unusable as a divisio operator with lower precedence that the ASCII "/".

>> which can be used to disambiguate a/b/c/d as in a/b∕c/d = (a/b)/(c/d)?
>
> I don’t quite see what you mean, but if I understand the idea correctly, it’s not the kind of thing you’re supposed to do. U+2215 is semantically less ambiguous than U+002F, but the latter too can be used as a division operator. The choice between U+002F and U+2215 does not affect operator precedence.
>
> In fact, the relatively new standard on mathematical notations, ISO 80000-2, which identifies the operators by their Unicode numbers, explicitly says that the symbol “/” used for division is SOLIDUS U+002F. Maybe they just didn’t think of other possibilities, but in any case this indicates that U+2215 cannot be expected to the normal, or even normative, symbol for division.
>
>> And is U+FF0F intended for non-math use?
>
> As the name FULLWIDTH SOLIDUS says, it’s meant for use instead of SOLIDUS in East Asian writing systems. It’s just a wide variant of U+002F. So it may have math and non-math use, just as SOLIDUS may.

One can use the characters how one pleases, but it would not be consistent with its past history. Which is what I suspected and wanted to know. Thanks.

Hans
Received on Fri Oct 07 2011 - 13:38:58 CDT

This archive was generated by hypermail 2.2.0 : Fri Oct 07 2011 - 13:38:59 CDT