Announcing Unicode 7.0

This is the 2nd dupe on the frontpage for today. Should HN do some URL normalization? The dupe about the SSD had an extra trailing slash, and here it is the usual problem with blogspot.com vs blogspot.<countrycode>.

n0body• 12 years ago

now in perl blead

http://www.nntp.perl.org/group/perl.perl5.porters/2014/06/ms...

draegtun• 12 years ago

Yep... here's the commit on the Github mirror - https://github.com/Perl/perl5/commit/09edd81111f67b4331309c8...

edwintorok2• 12 years ago

OCaml got updated libraries for Unicode 7 too: https://sympa.inria.fr/sympa/arc/caml-list/2014-06/msg00081....

beobab• 12 years ago

"Notched Right Semicircle With Three Dots"?

I'm not sure I want to know what this would be used for.

kabdib• 12 years ago

Oh, I'm sure that someone in denotational semantics will use it for something... :-)

(Twice having tried to learn DS, thinking it would be "good for me", only to rage quit after running into the next half dozen undefined hieroglyphics...)

publicfig• 12 years ago

Here's what I was able to find on it.

http://www.unicode.org/L2/L2010/10395-typikon.pdf

Tyr42• 12 years ago

This is a reversed form of the preceding Unicode symbol located at U+1F543 (proposed location): . This symbol is used on right-hand pages to denote an Orthodox Lower Rank feast. As well, this symbol can be used to denote one of the two lower rank feasts in black-and-white printing.

hf• 12 years ago

I value the Unicode effort so much, I consider its inconsistencies a major pain point in my life.

Why, for the love of Zeus, is there no codepoint for

    GREEK SUBSCRIPT SMALL LETTER ALPHA

Am I missing something? Why should I never want to subscript α? I know a few physicists who gladly would.

It doesn't end there, see the article https://en.wikipedia.org/wiki/Unicode_subscripts_and_supersc... which sports a consolidated overview.

Tell me I'm overlooking something very obvious.

jschulenklopper• 12 years ago

Why is there no codepoint for GREEK SUBSCRIPT SMALL LETTER ALPHA?

Could it be that (quoted from that Wikipedia page): The World Wide Web Consortium and the Unicode Consortium have made recommendations on the choice between using markup and using superscript and subscript characters: "When used in mathematical context (MathML) it is recommended to consistently use style markup for superscripts and subscripts."

Inconsistency remains though, since the Unicode standard defines characters for full superscript Latin lowercase alphabet except q, a limited uppercase Latin alphabet, a few subscripted lowercase letters, and some Greek letters.

gioele3• 12 years ago

Unicode is not about what you would like to use, but about what has been used in the past. New characters are avoid as much as possible and old characters are allocated a codepoint only if you can show that they have been extensively used and that they are really different from other similarly looking glyphs that already have a codepoint.

Probably nobody has found enough small alphas used in subscripts and there were no electronic encodings that included it so it has never been considered for inclusion.

smackfu5• 12 years ago

Well, the real reason is that the existing subscripts are to support UPA: http://en.wikipedia.org/wiki/Uralic_Phonetic_Alphabet

Here's where they argued that they needed subscript versions, rather than using the existing greek letters with a modifier: http://std.dkuug.dk/jtc1/sc2/wg2/docs/n2442.pdf

It is kind of a mess though. Maybe it would have been better if they had named the repurposed glyphs like LATERAL POSTVELAR CONSONANT, but that's not generally how Unicode does things.

buster14• 12 years ago

Do we really need small pictures in fonts?

rational-future• 12 years ago

That is what fonts are - sets of small pictures, accessible by an index.

buster14• 12 years ago

I really don't agree on your definition of fonts. When i write a letter i am certainly not painting rows of little pictures on a sheet of paper.

stephen_g• 12 years ago

Why not? It's not like every font of every typeface has to implement every Unicode code point.

You'd probably only have one or two installed that have the emoji characters.

buster14• 12 years ago

Because it's waste. Because if not every font implements it, what is the usefulness of it? Did you ever visit websites that use glyphs your font doesn't know? How would you use those pictures on a website, when you can't even guarantee the style of the picture, because it may be implemented differently or not at all.

It just looks like a big waste for no good reason.

gtirloni• 12 years ago

I had the exact same question pop up in my mind after seeing the emoji's. I understand all characters can be considered nothing but pictures, but a thermometer, really? Is this used in any widespread language to be of significance to Unicode? Is Unicode trying to also offer some kind of clipart capability?

I think we've seen our fair share of derailed projects in IT. Unicode is certainly a very critical project, which makes its focus even more important.

sho_hn10• 12 years ago

Unicode contains a wide range of different scripts.

For example, some are phonographic (the letters represent phonemes, i.e. speech sounds), others are logographic (the letters represent words, or morphemes).

Both of these can be featural: The Korean Hangul script is phonographic, but the letter shapes are not arbitrary. Some of them are visualizations of mouth shape or tongue position when producing the sound. For example the velar consonant ㄱ (somewhere inbetween the English g and k, roughly) is the tongue position as seen from the side.

A logogram can also be featural, i.e. it can look like the thing it represents. This would be the case for a glyph that means "thermometer" and looks to viewers like a thermometer. It could also be arbitrary - some shape that you just agree on to mean thermometer. Perhaps sharing some graphical aspects with topically related logograms.

They're all just spots on a spectrum.

(Edit: I briefly touched on pictograms vs. ideograms a bit more down below.)

elfcard• 12 years ago

That's amusing, because I looked at the thermometer, and thought, oh that's good, I can use that. But a chilli pepper, do we really need a character for a chilli pepper? So to answer your question, unicode is trying to bridge the gap in communication that offers a language independent set of recognizable glyphs that can be used to communicate. And one mans chilli pepper is another mans thermometer.

greggman3• 12 years ago

Is the thermometer symbol used widely? No idea. Is emoji used widely? Yes, In Japan 🇯🇵 (and Korea? 🇰🇷) and has been for > 15 years. A common email/msg composed by a Japanese woman 👧 age 15 to 35 uses many emoji.

I realize this message will not render correctly in Chrome because Chrome, at least Desktop Chrome, does not yet support emoji 😭

JetSpiegel• 12 years ago

Hopefully it will stop the nonsense of having custom fonts in websites for icons. Which can be copied and turn out to be a "t" or something.

scrollaway5• 12 years ago

It won't. Those custom fonts you speak about not only use characters which aren't, and never would be in the unicode database (such as trademarked logos), but also draw from the private unicode blocks: https://en.wikipedia.org/wiki/Private_Use_Areas

Which means there's no sanely written font that would use a "t" for an icon. Wingdings is not a sanely written font.

smackfu5• 12 years ago

That's not really the right question. We already have small pictures in fonts, just look at Wingdings or Webdings. Unicode is just assigning them to unique code points so they can be represented in a document without requiring the font to be specified.

buster14• 12 years ago

And you consider Wingdings a good thing in the first place?! It's an abomination that should never have happened.

TorKlingberg• 12 years ago

They are in Unicode for compatibility with other encodings, to make it easier to switch to Unicode. Most of the emoji comes from Japanese mobile phone carriers or fonts like Wingdings.

greggman3• 12 years ago

Yes, to support 15 years of Japanese communication since they were introduced to cellphones in Japan in 1998 and the majority of users here use them.

sisk• 12 years ago

You might not need them but consider this for a moment:

When I was a kid, there was a boy in my class who was mute. He carried a chalkboard and flashcards to facilitate communication. Eventually, he was chosen to pilot a device that was sort of like a keyboard with oversized keys. Each key had a pictograph on it[0].

Wouldn't it be nice for him if the flashcards and keyboard device had a consistent set of characters that were easily recognizable not only by him but by the people with whom he was communicating?

Now, would he ever need "MAN IN SUIT LEVITATING"? Probably not (you never know!) but I'm glad there is a standard set of otherwise useful pictographs for people like him.

[0] - I'm in my late 20s but this was when few families owned computers, let alone super-helpful handheld devices. Also worth mentioning that this boy knew ASL but attended a regular public school where few students and faculty could so the chalkboard and flashcards were infinitely more practical.

buster14• 12 years ago

And that device would have a keyboard with thousands of keys?

Btw. that device exists today, and it's called a tablet. I'm pretty sure i have seen stuff exactly like that and it serves a great purpose.

The consistent set of "characters" (pictures in this case) would still look very different on each device or application that uses another font. Because a banana in one font could look very different in another font. The only thing that actually would be consistent if the content he uses would use actual pictures! (A website displaying a banana on his touchpad can look different on his laptop, can't it?).

minikomi• 12 years ago

If anyone's wondering what the notched semicircle looks like, you can see it along with quite a few (all?) of the new pictographs on this pdf: http://www.unicode.org/charts/PDF/Unicode-7.0/U70-1F300.pdf

NoMoreNicksLeft• 12 years ago

Dammit, still no glyph for the Artist Formerly Known As Prince.

Don't they realize how much of 1990s music history is unwritable thanks to this glaring omission?

gioele3• 12 years ago

There is no single glyph for The Artist Formerly Known As Prince, but you can use combining chars to obtain it.

    01AC LATIN CAPITAL LETTER T WITH HOOK
    030A COMBINING RING ABOVE
    0335 COMBINING SHORT STROKE OVERLAY
    032C COMBINING CARON BELOW

http://www.unicode.org/mail-arch/unicode-ml/Archives-Old/UML...

moron4hire2• 12 years ago

Towards that artist's motivations, it would be best to keep it this way.

jstalin• 12 years ago

Forgive my ignorance, but how do these actually enter the real world? Do browsers adopt these new characters?

mmastrac2• 12 years ago

Technically the entirety of the unicode range is supported by browsers right now -- they just don't exist in fonts at all. What Unicode does is say that "this codepoint" means "this". At that point, fonts start adding glyphs to mean "this" and authors start adding codepoints that refer to "this".

If you look at FontAwesome, it uses characters in the "unicode private use area" -- which are effectively undefined -- but putting those codepoints in your HTML document while having the appropriate font loaded will show the appropriate icon glyphs.

greggman3• 12 years ago

Actually AFAICT Chrome does not support the entirety of the unicode range right now. Safari and Firefox do.

🙅🎁４👨

dragonwriter• 12 years ago

Forgive my ignorance, but how do these actually enter the real world? Do browsers adopt these new characters?

Browsers don't add characters, fonts do.

sho_hn10• 12 years ago

You should find http://pixelambacht.nl/2014/multicolor-fonts/ an interesting read.

masklinn3• 12 years ago

Do browsers adopt these new characters?

No, fonts ship with them. Browsers don't really care, they see codepoints and ask underlying font-rendering API for the corresponding glyphs, then display that.

And these characters may not have anything to do with browsers, not are they restricted to browsers e.g. "emoji" originate in pre-smartphones cellphone-based text and email messages in japan for instance.

smackfu5• 12 years ago

This page has links to the proposals to add characters: http://babelstone.blogspot.com/2013/10/whats-new-in-unicode-...

Including this one with comments on the Wingdings addition. http://std.dkuug.dk/jtc1/sc2/wg2/docs/n4239.pdf

  a) Ireland recommends adding the following new characters to this table: 
  1F597 REVERSED HAND WITH MIDDLE FINGER EXTENDED 
  1F598 RAISED HAND WITH PART BETWEEN MIDDLE AND RING FINGERS

ximeng• 12 years ago

Love the uk's understated battle with the US over Latin letter middle dot. Man in business suit levitating is another great character. Amazing that this is someone's job.

colanderman• 12 years ago

Note that these glyphs will appear as boxes until font designers get around to adding them to their fonts.

You can have fun with this. Say you're disillusioned with your current job and planning to leave within the year. Insert these into whatever important document you're working on at the time, e.g. as bullet points or decoration around titles. Sometime after you leave, they'll reveal their true nature. A sort of "FU" time-bomb.

moron4hire2• 12 years ago

Nice, but how freaking long is it going to take to get this to show up in software anywhere? We can't currently reliably use a large swath of interesting glyphs in the current version.

jrochkind1• 12 years ago

Everybody's talking about new glyphs/codepoints, but unicode is actually way more than a list of codepoints.

Note the announcement also includes:

* Changes to the unicode collation algorithm for locale-dependent sorting (I haven't figured out what htey are yet)

* New character properties and values, used for case-changing and line-breaking behavior (i think line-breaking might be locale-dependent too).

* some other stuff

The unicode algorithms for _doing things_ with text in appropriate ways are actually way more mind-blowing to me than just the directory of glyphs. When you start thinking about it and get into the weeds and realize how other languages work very differently than English with regard to some of this stuff -- proper "uppercase" or "sort" or "are these two strings semantically the same" across the global universe of glpyhs -- is _really_ challenging, and unicode does a pretty amazing job of it.

It's sadly way more confusing to figure out if/when the unicode library of your choice has been updated to use unicode 7.0 (or even 6.0) text algorithms and full character database with properties, rather than to figure out if a given font includes the new glpyh or whatever.

vonuebelgarten• 12 years ago

And still no tengwar :(

Announcing Unicode 7.0

Related Stories