Hacker News Reader

Comment by tritium | Hacker News Reader

Oh, this is about the history of calligraphy, typography, alphabets and character sets surrounding the (swedish) letter (which resembles an) “A” with a ring (non)diacritic, and not about the ångström as a unit of measurement.

EDIT: People are vehement about the status of this particular character, and that doesn’t change the fact that I expected an article about a unit of measurement.

  (yes, I understand that disagreement is 
   afoot, but it has nothing to do with 
   what I originally came here for)

View on HN

I think that a Swede would disagree with you that this is an A with a ˚ diacritic; Å is seen here as a seperate letter and granted its own place in the alphabet: https://en.wikipedia.org/wiki/%C3%85#Place_in_alphabet

Any Scandinavian would disagree. Æ, Ø, and Å most definitely are letters in their own right.

The case is slightly less clear for W.

Although a Swedish person might disagree about how to write ø and æ...

Why? Those characters are not used in the Swedish language.

AFAIK they sound exactly the same like København and Köpenhamn.

That's actually not quite the same ø. The Swedish one is slightly lighter. Use the deeper Danish ø-sound in a Swedish context, and you clearly sound like a Dane speaking Swedish.

The same is true for ä and æ. The latter has a broader sound while the former sounds a bit closer to e. It is easy to spot Norwegians talking Swedish (such as myself :)

You try to talk Swedish rather than just speaking Norwegian? My experience is that Scandinavians just speak their native language and expect that other Scandinavians should understand them. Any attempts to speak the other language end up sounding like parodies rather than serious attempts.

Not sure it's different enough to warrant a new letter, but Spanish O and Scandinavian Å has the same sound so I guess it's all very muddled.

"Ä" is worse, it can be three completely different things depending on language.

In for example Swedish, it's a distinct letter of the alphabet, representing a distinct vowel.

In German, it's the letter "A" with an umlaut mark, which is not a distinct letter of the alphabet, but it is a distinct vowel.

And in French, it's the letter "A" with a diaresis mark, which neither makes it a distinct letter of the alphabet, nor a distinct vowel, it's just a mark that signifies that two adjacent vowels should not be pronounced as a diphtong.

Or you have Icelandic:

"A" is a distinct letter and distinct vowel.

"Á" is also a distinct letter and a distinct vowel, and not an "A" with an acute accent.

"À" exists in Icelandic in (very few) French loan-words and is not a distinct letter nor a distinct vowel, instead it's viewed as an "A" with a grave accent.

(And this is why lexical sorting has to be highly localized. I still think Turkish has the funniest gotcha where "I" and "i" aren't the upper/lower-case of the same letter, it's the uppercase of one letter, and the lowercase of a completely different letter.)

Specifically:

ı (capital I; the letter is pronounced as a schwa), and:

i (capital İ)

This is why upper- and lowercasing of letters is also locale-dependant (in addition to collation).

> In German, it's the letter "A" with an umlaut mark, which is not a distinct letter of the alphabet, but it is a distinct vowel.

The umlauts are distinct letters of the German alphabet. The English Wikipedia says so but still lists them as special characters which is confusing at best. The German Wikipedia just lists the umlauts as ordinary characters.

>German uses three letter-diacritic combinations (Ä/ä, Ö/ö, Ü/ü) using the umlaut and one ligature (ß (called Eszett (sz) or scharfes S, sharp s)) which are officially considered distinct letters of the alphabet.[1]

I believe there are just two cases:

- A umlaut: German, Swedish, etc.

- A with diaeresis: French, Catalan, etc.

It is important to distinguish both cases because they differ at least in collation and typography.

In a German dictionary A with diaeresis should be always treated like an A, while A umlaut should be treated according to the different collation standards that exist in German. In a German font the dots are closer to the letter than in a French font.

[1] https://en.wikipedia.org/wiki/German_orthography#Alphabet

Confusingly, there are two different collation standards when treating German umlauts: One for lists of names and one for everything else.

When reciting the alphabet German school kids don't add the umlauts (at least I didn't – I don't think that has changed in the last decades) and if you ask someone how many letters the alphabet has they will answer "26" without hesitation while in Sweden they are treated as distinct characters of the alphabet in every sense.

> Confusingly, there are two different collation standards when treating German umlauts: One for lists of names and one for everything else.

Yes, this alone is confusing. My point is that it is even more complicated because strictly the standard only applies to umlauts but not to letters with diaeresis. Unicode offers a way to treat these two cases differently utilizing the combining grapheme joiner (CGJ).

>> The CGJ can also be used in German, for example, to distinguish in sorting between “ü” in the meaning of u-umlaut, which is the more common case and often sorted like <u,e>, and “ü” in the meaning u-diaeresis, which is comparatively rare and sorted like “u” with a secondary key weight. [1, page 850]

> When reciting the alphabet German school kids don't add the umlauts (at least I didn't – I don't think that has changed in the last decades) and if you ask someone how many letters the alphabet has they will answer "26" without hesitation while in Sweden they are treated as distinct characters of the alphabet in every sense.

You have a point here and maybe the English Wikipedia isn't so wrong in listing them as special characters. Being special doesn't make A umlaut a funky A though. It still is a letter in it's own right.

[1] Unicode Standard 10.0: http://www.unicode.org/versions/Unicode10.0.0/UnicodeStandar...

For what it worth, Dutch has "ij", which could be a letter, or diphthong, or two unrelated letters (in words borrowed from French, for example), and could also be written as "ij", or as "ÿ", or even "y".

There is no dedicated key on computer keyboard for it, but you are supposed to remember it's a unit when capitalizing, for example IJmuiden.

https://en.wikipedia.org/wiki/IJ_(digraph)

Meanwhile, the Swedish alphabet has 29 letters. Since 2006. Before 2006, "W" wasn't considered a letter of the alphabet, since it was just a double "V", and only existed in German and English loan-words and names.

So before that, this would have been the correct sort order of some names: "Valter, William, Viktor", but these days they would probably be sorted like "Valter, Viktor, William"

> I believe there are just two cases:

Nope. :-)

Because in Swedish, ÅÄÖ are distinct letters of the alphabet, they are not treated different than any other vowel, they're not variants of base letters or anything like that.

I was wrong about ÄÖÜ not being part of the German alphabet. They are, but they're not letters (Buchstaben), they're umlauts. So the Swedish alphabet contains 29 letters, but the German alphabet contains 26 letter, 3 umlauts, and the Eszett.

In the end, it's all arbitrary as fuck, with tons of historical reasons for the way things are, and god help the localization engineer who thinks it's all easy peasy. :-)

>> And in French, it's the letter "A" with a diaresis mark, which neither makes it a distinct letter of the alphabet, nor a distinct vowel, it's just a mark that signifies that two adjacent vowels should not be pronounced as a diphtong.

I can't think of an example of a French word where an A takes a diairesis. Can you give one?

Wikipedia has this to say:

En français, le tréma peut se placer sur les voyelles e, i, u (et y dans des noms propres) pour indiquer, normalement, que la voyelle qui précède doit être prononcée séparément et ne fait pas partie d'un digramme.

[https://fr.wikipedia.org/wiki/Tr%C3%A9ma]

https://en.wikipedia.org/wiki/Te%C3%A4n

(Super uncommon, yes, but it exists)

Also, extra extra unnecessary knowledge: The symbol for the Ångström unit has a different Unicode encoding than the capital "Å", in the same way that the Ohm symbol "Ω" is different from Greek capital omega. "W" for Watt doesn't have it's own codepoint in Unicode though, so that's arbitrary.

This has only been allowed into Unicode for compatibility with older encodings and is sort of an exception.

> Multiple Semantics. Because nonspacing combining marks have such a wide variety of applications, they may have multiple semantic values. For example, U+0308 = diaeresis = trema = umlaut = double derivative. Such multiple functions for a single combining mark are not separately encoded in the standard.

[..]

> Occasionally, for compatibility with other standards, a single abstract character may correspond to more than one code point—for example, “Å” corresponds both to U+00C5 Å latin capital letter a with ring above and to U+212B Å angstrom sign.

Both quotes from Unicode Standard 10.0:

http://www.unicode.org/versions/Unicode10.0.0/UnicodeStandar...

This sort of interesting article is exactly what I come to Hacker News for.

It didn’t get interesting to me, until I was quickly corrected by more than one person. Is the phonetic pronunciation... what is the pronunciation for Å in English?

Aww

Similar to English "o" (more so than the Scandinavian o). It is one of the easier vowel sounds. Ø and y and u are far harder for English speakers.

As the o in 'more'

It's satire, and a little funny. They have the correct pronunciation for Norwegian at least, it's similar enough in Danish and Swedish: https://www.youtube.com/watch?v=f488uJAQgmw

That's what all the "non-Swedish physicists" thought.

Reply on news.ycombinator.com