[freeopinion]

Your statement here has so many qualifiers that it's hard to treat it as a general rule.

"English speaking world"

"untypeable"

"public API"

Even then you exempt Greek symbols. I suppose you might be willing to accept ¢ or €. Of course, "untypeable" is very ambiguous.

Still, not all code is "public API" and not all code is intended for the English speaking world. So I think it's great to support non-ASCII symbols.

よろしく

[jagger27]

I was not trying to argue against its general support, only that for this English-speaking audience (HN) there are very limited use cases.

[freeopinion]

I accept your caution. But it is fun for me to consider the differences in these:

enum Size { P, XS, S, M, L, XL, XXL }

enum Size { Petit, Extra_Small, Small, Medium, Large, Extra_Large, Extra_Extra_Large }

enum Size { 小小小, 小小, 小, 中, 大, 大大, 大大大 }

Which do you prefer: P, Petit, Petite, 小小小, 3小, XXS, 2XS?

English is chalk full of non-English words. And lots of symbols are useful. Sometimes it's worth it to learn the symbol. And often not much harder than learning the English abbreviation.

enum ⏻ { ⭘, ⏽, ⏾ }

Again, "typeable" is ambiguous but obviously an important consideration.

None of this is to argue against your point. Just musing.

[HKH2]

That's not standard Chinese. Is it Japanese?

You could write SSS, SS, LL and LLL in English too.

By the way, it's 'chock-full' not 'chalk-full'.

[freeopinion]

Thanks for the correction. TIL

Are you distinguishing Chinese or Japanese on visual appearance? Or by some other method?

I think 小中大 have an advantage over SML because SML have no meaning by themselves. 小中大 have the exact meaning we need. Granted, 中 kinda ruins that argument. Size::中 is very clear, but perhaps not more so than Size::M.

I'm actually curious. Which do you like better in code? P, SSS, XXS, 小小小, 3S, 3小, whole words, something else?

[HKH2]

I can't read Japanese. I just know they use characters slightly differently, so I can't be sure they wouldn't write that.

The majority of people in the world can read and type 'S' easily, not so for '小'.

It seems you're basically arguing for pictograms. Like coding in Wingdings with a character map or something.

I would prefer to use a number scale, but if I had to choose from those options I would use whatever is most conventional in whatever area I'm in, which would probably be XXS.

[runarberg]

Even if english speaking, it is still ambiguous. For example in my keyboard the micro sign (µ; not to be confused with the Greek small letter mu) is easy to type (AltGr + M). I bet this is the same for the majority of HN users. The US keyboard doesn’t use third level shift (usually AltGr) so most people that use US keyboard exclusively are unaware of this. Whenever I see people using the Latin small letter U (u) instead of the micro sign, all I think about is how restrictive the US keyboard actually is, and how much of a shame it is that the culture which predominantly uses this restrictive keyboard design came to be the dominant when designing computer interfaces.

[jdrek1]

> For example in my keyboard the micro sign (µ; not to be confused with the Greek small letter mu)

The greek letter mu IS the micro sign, where do you think it comes from?

[gourlaysama]

Unicode has both MICRO SIGN (U+00B5) and GREEK SMALL LETTER MU (U+03BC). The former is the one on (most) people's keyboard, and it shouldn't be used to type actual Greek.

[runarberg]

Look the same, but different code points.

MICRO SIGN (U+00B5)[1] vs. GREEK SMALL LETTER MU (U+03BC)[2]

1: https://www.fileformat.info/info/unicode/char/00b5/index.htm

2: https://www.fileformat.info/info/unicode/char/03bc/index.htm

[jdrek1]

My apologies, I did not know about that. Why would they deliberately put duplicates in unicode though?

[runarberg]

Some font-families might want these letters to look different for some reason. Even if they look the same perhaps the boldfaced or italics should behave differently, perhaps the spacing should be different etc.

I’m not a font designer but I can imagine that the micro sign often appears around Latin letters whereas the Greek letter mu doesn’t, so that font designers might space them differently for that reason.

For fun I compared 10µm (upper; using the micro sign) with 10μm (lower; using the greek letter mu) in the comic sans font face. https://imgur.com/qwnnAqD

[dwheeler]

> Why would they deliberately put duplicates in unicode though?

IIRC, to enable adoption. Many of the codepoints were adopted from other encoding systems, and it was often useful to support 2-way conversions without loss while Unicode was being adopted.

What's a "duplicate" is also sometimes in the eye of the beholder.