Hacker News Reader

Comment by jws | Hacker News Reader

That is Unicode U+0243 "Latin Capital Letter B with stroke". It is in the Latin Extended-B block, which contains among other mysteries, about half of the latin letters with strokes.

I'd love to see a "Unicode: What where they thinking?" page that documents why they thought a particular codepoint should be in the standard. I mean, capital A, B, and C with a stroke, but not capital D? A and C have their stroke diagonal, B has its horizontal, U also gets a horizontal stroke, but it is named a "bar". All very mysterious.

View on HN

The OP suggests that U+0243 B with stroke is a "capitalized alternate symbol for the voiced bilabial fricative in Americanist phonetic notation"

I would suspect that the other stroke capitals that are present were used in specific actual alphabets (and probably represented in some existing pre-unicode font or code block), whether 'Americanist phonetic alphabet', or other -- and the stroke capitals that are not present, were not.

In general, unicode codepoints are justified by actually existing use.

(There is an interesting metaphysical problem that arises as our lives become increasingly digitized -- things that are not present in unicode _won't_ be used, because they won't be _able_ to be used online... so how will new things be added justified by use, once all the existing pre-digital glyphs have been added? I dunno.)

For what is worth, the voiced bilabial fricative is a sound similar to English "b" and "v". Languages that use that sound normally write it with "b", "v" or "w". It's common in Spanish but also creeps up in certain words in Portuguese, Japanese and others. [1]

In IPA, the standard phonetic alphabet, that sound is notated by "β" instead. The b-with-a-stroke, as well as barred-b (ᴃ) were used by some linguists before IPA became widespread, but they still kept appearing since they were easier to type in most typewriters (B, backspace, hyphen) and apparently also because some linguists didn't like IPA.

In addition, b-with-a-stroke is part of the Jarai alphabet, a variant of the Vietnamese alphabet used to write the Jarai language. [2]

[1] http://en.wikipedia.org/wiki/Voiced_bilabial_fricative [2] http://en.wikipedia.org/wiki/Jarai_language

Everyone can add new things in the PUA, in fact, that's where Emoji were before they were encoded. There is the ConScript Unicode Registry [1], an inofficial effort to map various constructed scripts to the PUA without them clashing. There are lots of pages in Klingon out there that make use of that encoding even though it's not in Unicode. Some of those have been encoded formally by now, so there is a way of creating new characters. It just takes a little longer.

Also, by now with webfonts you can easily create your own font to include characters from the PUA you designed, so you can use them online.

[1] http://www.evertype.com/standards/csur/

> In general, unicode codepoints are justified by actually existing use.

I vaguely remember something about Klingon and unicode related to that... ah, there we go:

"In September 1997, Michael Everson made a proposal for encoding this in Unicode.[2] The Unicode Technical Committee rejected the Klingon proposal in May 2001 on the grounds that research showed almost no use of the script for communication, and the vast majority of the people who did use Klingon employed the Latin alphabet by preference.[3] Everson created a mapping of pIqaD into the Private Use Area of Unicode, which he listed in the ConScript Unicode Registry (U+F8D0 to U+F8FF[4][5])"

http://en.wikipedia.org/wiki/Klingon_alphabets

Everson did manage to get J.R.R. Tolkien's Cirth and Tengwar tentatively included; see http://www.unicode.org/roadmaps/smp/

No. The roadmap indicates that they are formally proposed for inclusion and yet to be accepted (and nor rejected either, as in [1]). From the bottom of the roadmap:

> (Text between parentheses) indicates scripts for which proposals have been formally submitted to the UTC or to WG2. There is generally a link to the formal proposal.

[1] http://www.unicode.org/roadmaps/not-the-roadmap/

According to the phrasing on the bottom of the page I linked to, the blocks are "tentatively allocated." I'm not sure how this is substantially different from the phrasing I used, which was "tentatively included."

Interesting! (The proposals themselves, for anyone else interested: http://std.dkuug.dk/JTC1/SC2/WG2/docs/n1641/n1641.htm, http://std.dkuug.dk/JTC1/SC2/WG2/docs/n1642/n1642.htm)

In a related vein, I suspect ligatures, small caps, and non-lining figures already look weird to "digital natives" because they were dropped during the age of DTP and generally have only been reintroduced as typographic novelties.

It would be interesting to do a study, I would suspect those forms are still seen as 'classy', 'elegant' and 'better' -- non-lining figures and the like. But maybe that's just my own preference/hope, and in fact they are just seen as 'weird'.

But hey, everyone likes an actual em or en dash to two or one hyphen, don't they? No? Say it ain't so! (Not that anyone, including me, bothers to enter them on the keyboard, see above).

They are very easy to use on the mac — shift-option-minus.

There is an interesting metaphysical problem that arises as our lives become increasingly digitized -- things that are not present in unicode _won't_ be used, because they won't be _able_ to be used online...

I remember reading about exactly this problem recently in China: http://www.dnaindia.com/world/report_chinese-villagers-force...

so how will new things be added justified by use, once all the existing pre-digital glyphs have been added?

I see two possible scenarios:

1. They won't be. Instead, existing glyphs will become increasingly overloaded with new meanings.

2. New glyphs will start their life as icons. Once a certain icon becomes wide spread, it may be made into a code point with a glyph.

In practice, it would probably be some combination of the two.

The Unicode people don’t sit around and invent characters – nor do they systematically determine the complete list of characters that should be in the standard.

Instead, they consider proposals from people who want their already-in-use character sets to included in the standard. So if someone proposes SINGING OCTOPUS, DANCING OCTOPUS, and SINGING SQUID, but not DANCING SQUID, well, DANCING SQUID won’t be in Unicode. There are lots of gaps like this.

The dancing octopus will have to wait for the next version.

The unicode consortium is currently only adding the SMILING OCTOPUS: http://codepoints.net/U+1F419

Probably not supported by your browser's font just yet: 🐙🐙🐙

It is in my Firefox Aurora. :D

Thanks for sharing codepoints.net, I didn't know about it before. I wish the campfire :metal: emoji was in 'actual emoji'...

Interestingly enough, the octopus shows up but the proposed Bitcoin symbol doesn't (iOS 6)

Interesting. I'm also on iOS 6 and both show up for me. I wonder why.

Edit: in safari, the symbol shows correctly in the window header, but not the page itself. In news:yc, a hacker news app, the symbol shows up fine. It also shows up in the WebView in news:yc.

I haven't looked at the Supplementary Multilingual Plane: Miscellaneous Symbols And Pictographs [1F300...1F5FF] in a while. My favorite characters are 👹 (Japanese Ogre)👺(Japanese Goblin)👻(Ghost) and 👽(Extra Terrestrial).

The symbol 💣 (Bomb, 1F4A3) is useful for reviewing movies using Leonard Maltin's notation.

Since we're picking favorites, why not the happy pile of poo? 💩

It's not officially supposed to be happy, but Apple seems to have decided eyes and a smile would give it more character.

http://codepoints.net/U+1F4A9

It works in Firefox 19.0.2 and Safari 6.0.3 but not in Chrome 26.0.1410.43.

Interestingly, works on chrome android, but not chrome OSX. If it's copy and pasted elsewhere in OSX though it comes up as a red octopus

Got it in ios 5, which came out a while ago. Title character is still just a square. Weird.

Firefox Nightly reporting in 🐙🐙🐙

The word "smiling" does not appear on the linked page.

Works on Opera too.

And IE10. (!)

My IE10 octopus is missing 3 arms and a mouth, resulting in something that looks more like a 3-legged grey alien trying to surrender than an octopus who is smiling.

🐙🐙🐙 Yes!

I had my "What are they thinking?" moment when I found out about snowman.

http://www.fileformat.info/info/unicode/char/2603/index.htm

After that the only one to shock me was "Pile of poo"

http://www.fileformat.info/info/unicode/char/1f4a9/index.htm

Is 'PILE OF POO' (U+1F4A9) more shocking than 'LOVE HOTEL' (U+1F3E9)?

http://www.fileformat.info/info/unicode/char/1f3e9/index.htm

See: https://en.wikipedia.org/wiki/Love_hotel

You may be interested in this wonderful article [1] about how these and many other unusual symbols ended up in Unicode. I personally had never considered how much of an issue including flags could be. Quite an interesting discussion around it on HN as well. [2]

[1] - http://www.reigndesign.com/blog/love-hotels-and-unicode/ [2] - https://news.ycombinator.com/item?id=3792247

Both were readily in common use on Japanese phones. Most of Unicode is existing character sets being incorporated.

Looks like they're part of the Emoji set

http://en.wikipedia.org/wiki/Emoji#Emoji_in_the_Unicode_stan...

Well, exactly. Emoji are the pictographic emoticons that were in common use on Japanese phones in proprietary encodings.

Well sure, but since when is emoji part of unicode?

Emoji is a lot more mainstream now, too. Both Android and iOS support them, not just 'Japanese phones'.

Uh, what?

>since when is emoji part of unicode?

http://en.wikipedia.org/wiki/Emoji#Emoji_in_the_Unicode_stan...

>Emoji is a lot more mainstream now, too. Both Android and iOS support them, not just 'Japanese phones'.

Um, yes. I said "that were". I realise it is now mainstream now because it is now in Unicode.

Theres not anything seedy about a love hotel. In Japan sometimes it's nicer for couples to go there rather than try and have a romantic evening in a tiny 1 room flat on a futon.

The snowman is an old meteorological symbol for snow. Predates Unicode by decades. I assume it's because a snowman is easier to render by hand than an actual snowflake. (Or like how rain is symbolized not by trying to draw recognizable rain itself, but by an umbrella.)

the diversity in font implementation is interesting to see: http://www.fileformat.info/info/unicode/char/2603/fontsuppor...

Michael Kaplan's "Every Character Has a Story" is kind of like that, but it looks like he's adding to it slower than the Unicode folks are adding characters.

http://blogs.msdn.com/b/michkap/archive/tags/every+character...

There are three capital D letters with strokes within the Latin range: U+00D0 (capital Eth - old English, Icelandic), U+0110 (capital D with stroke - south Slavic, Vietnamese), U+0189 (capital African D - African languages).

Capital D does have a "with stroke" character, Ð. Code point U+00D0. In actuality, it's a capital eth (ð), which is a bona fide letter by itself rather than an existing letter with an extra mark in it.

> Capital D does have a "with stroke" character, Ð. Code point U+00D0. In actuality, it's a capital eth (ð), which is a bona fide letter by itself rather than an existing letter with an extra mark in it.

Actually Unicode has two D with strokes, the "LATIN CAPITAL LETTER ETH" Ð (U+00D0) and "LATIN CAPITAL LETTER D WITH STROKE" Đ (U+0110). The uppercase letters are basically indistinguishable, but they have different lowercase letters: đ and ð.

There is a capital D with stroke: Đ

http://en.wikipedia.org/wiki/D_with_stroke

http://en.wikipedia.org/wiki/Bar_(diacritic)

> A and C have their stroke diagonal, B has its horizontal

That's not something specified by Unicode, it's up to font designers how characters actually look.

Reply on news.ycombinator.com