글: 33
언어: English
erinja (프로필 보기) 2011년 2월 19일 오전 1:11:22
danielcg:I agree, and another thing I cant't understand is why they don't at the same time propose to do away with the dot over the i and the j.Personally, I oppose i and j due to their irregularity. I thought Esperanto was supposed to be a regular language with no exceptions???
In that case, how come we use capital I and lowercase i, when obviously the letter pair should be I/ı or İ/i.
This mismatched I/i makes no sense whatsoever. How am I supposed to memorize that the lower-case letter has this whole diacritic on it that the upper-case letter doesn't have??? J/j is similarly mismatched. This is totally unacceptable for this supposedly "regular" language!!!!111
marcuscf (프로필 보기) 2011년 2월 19일 오후 2:02:47
Erinja, when I write my own name in capitals, I use a dotted uppercase I... When I was a child, once my mom told me to write it that way (I don't know why) and I started doing it, but only for my name (MARCUS AURELİUS), LOL!
Art-langs are more likely to have unique alphabets or caracters*, while Aux-langs use the Latin alphabet. The fact that those scripts are for art-langs may contribute for their absence in Unicode too.
*(I know it's spelled characters, but it so illogical and confusing that I can't help myself
)
danielcg:None that I know of, and that includes Esperanto, which circumflex and hook are not unique to it.The combination "circumflex + consonant" is unique, and in computers one often prefers fine-tuned pre-composed caracters* than inventing an algorithm to put any diacritic over any letter making it look good at the same time... (let's not forget that you need to drop the dot above j before adding the hat). Not that Zam could have predicted that.
danielcg:Lack of these letters in the ASCII set of characters is a defect of this set, not of the language.I think it is a shame that no accented letter was included in ASCII. I don't think its designers had any obligation to know Esperanto, but what about their neighboring languages? ASCII extensions were developed to correct this flaw, but they were usually incompatible with each other.
darkweasel:How many conlangs have unique characters at all?Klingon has its own alien alphabet and there is also Tengwar for Tolkien's languages. Neither is in Unicode yet (I was not sure about that when I sent my first message, that's why I asked).
Art-langs are more likely to have unique alphabets or caracters*, while Aux-langs use the Latin alphabet. The fact that those scripts are for art-langs may contribute for their absence in Unicode too.
*(I know it's spelled characters, but it so illogical and confusing that I can't help myself
) marcuscf (프로필 보기) 2011년 2월 19일 오후 2:35:19
Oops, I forgot to say this:
Even tho computers, typewriters and printers do support the hats, before Unicode you sometimes had problems when mixing languages. For instance, in Latin-3 you can write in Esperanto but not in Portuguese (at least ã and õ are missing). In order to work around that you need to either:
- use software-specific features to switch encodings mid-text;
or
- create a patched font with both with both ĉ and ã, which created lots of incompatibilities.
I guess Japanese computers could type their own caracters plus basic Latin caracters, but I dou(b)t that they could cope with any diacritics.
This is not as much of a problem anymore, I know, and we still have good-old H and X when problems do arise. But it was really annoying in those old times. Yahoo groups still messes up sometimes, sending to the browser webpages mixing more than one encoding scheme.
I also want to make it clear that I think the hats were a very smart solution to make international words with G and J recognizable, whether they use the Germanic conventions (G/J) or the Latin conventions (Ĝ/Ĵ) (Italian for Ĝ and French for Ĵ). When I discovered Esperanto, I was very pleased with that design, and I still am!
Even tho computers, typewriters and printers do support the hats, before Unicode you sometimes had problems when mixing languages. For instance, in Latin-3 you can write in Esperanto but not in Portuguese (at least ã and õ are missing). In order to work around that you need to either:
- use software-specific features to switch encodings mid-text;
or
- create a patched font with both with both ĉ and ã, which created lots of incompatibilities.
I guess Japanese computers could type their own caracters plus basic Latin caracters, but I dou(b)t that they could cope with any diacritics.
This is not as much of a problem anymore, I know, and we still have good-old H and X when problems do arise. But it was really annoying in those old times. Yahoo groups still messes up sometimes, sending to the browser webpages mixing more than one encoding scheme.
I also want to make it clear that I think the hats were a very smart solution to make international words with G and J recognizable, whether they use the Germanic conventions (G/J) or the Latin conventions (Ĝ/Ĵ) (Italian for Ĝ and French for Ĵ). When I discovered Esperanto, I was very pleased with that design, and I still am!
erinja (프로필 보기) 2011년 2월 19일 오후 3:04:04
Let's not forget what the A in ASCII stands for.... American.
The creators of ASCII, in their defense, hardly imagined that their little codeset would have such far-reaching implications. When ASCII was invented for telegraph use within the US, the internet was still decades away. How could they possibly think that people sending telegrams within the US in the 60's would need all kinds of accented characters?
The creators of ASCII, in their defense, hardly imagined that their little codeset would have such far-reaching implications. When ASCII was invented for telegraph use within the US, the internet was still decades away. How could they possibly think that people sending telegrams within the US in the 60's would need all kinds of accented characters?
ceigered (프로필 보기) 2011년 2월 22일 오전 12:52:25
marcuscf:What i like best in Eo:Easy - there normally isn't a "ui" diphthong and only a "wi" sound in many of them if the ui part isn't stressed, and if it needs to be a "uy" sound they can just use "oy" to the same basic effect (ui is relatively unstable as a diphthong in mosy languages where "u" is a true "u" and not like the "u" in Japanese, at least from what I've seen).
- 1 letter → 1 sound (if you consider affricates as 1 sound)
- 1 sound → 1 letter
- strictly regular accent (stress)
- tabelvortoj
Other conlangs usually to break one of these features and I can't understand why.
For instance, using I/U as semivowels instead of J/Ŭ or Y/W really annoys me. How can we tell if "ui" is ŭi or uj? In Portuguese it is uj, in Spanish it is ŭi, and these are very closely related languages. A conlang should have a good solution for this.
Ultimately a conlang only needs to deal with what sounds it's planning to use.
I'm personally more a fan of:
1 letter in a certain situation = 1 collection of alike sounds.
More flexibility if the language is designed to take advantage of it (otherwise it's not an entirely useful methodology). I am a fan of those sorts of "sandhi" effects though or "coloured" vowels (like the "ou" in "coloured"), they make a nice flow (like vowel harmony).
As far as most conlangs go though, many were created when computers were rare or nonexistent, which doesn't go well with the ASCII history I guess
. Maybe the US developers never gave a thought to the international community when they started using ASCII in computers, or assumed that other countries would make their own technologies. The world was a much less united place back then I assume? marcuscf (프로필 보기) 2011년 2월 22일 오후 2:08:42
ceigered:As far as most conlangs go though, many were created when computers were rare or nonexistent, which doesn't go well with the ASCII history I guessIt's not ASCII the standard, but the abstract concept of an alphabet with about 26 letters that is shared by all western languages.
Nobody in the 1880's without a time-machine could have guessed that "@" and "\\" would become more universal than "œ" or "ŭ". I guess the backslash didn't even exist (as a distinct caracter) then. But it is easy to see that C is more universal than Ĉ.
Depending on your point of view (or time-machine features), even W could be seen as non-universal, since it is not part of the basic Latin alphabet (J, W and the differentiation between U and V are relatively new). It was added for germanic languages, if i'm not mistaken. But now it is seen as a basic caracter.
ceigered:Maybe the US developers never gave a thought to the international community when they started using ASCII in computers, or assumed that other countries would make their own technologies. The world was a much less united place back then I assume?I think you are right. Just 10 years ago (before WinXP), Microsoft had different operating systems for each part of the world (west, middle-east and east, for example).
See:
21-Unicode_WinXP.pdf
erinja (프로필 보기) 2011년 2월 22일 오후 3:07:32
I have no proof of this, but I suspect that Zamenhof made the conscious decision to use accented letters, even though they are less "universal" than their unaccented counterparts. He was trying hard to adhere to the principle of one letter - one sound, AND keep words recognizable from their roots, AND give Esperanto a relatively large number of phonemes (this is closely related to keeping words recognizable from their roots - if you simplify the sounds too much, as we saw in Volapuk, words are no longer recognizable).
In some respects he was perhaps a visionary in the field of alphabets. At the time he created the language, you would have to have a special typewriter custom made to use Esperanto's unique characters, or to have print specially cast. In spite of the obvious benefits of choosing alphabetic characters that already existed, he chose not to. All of that is considerably more effort than what we have to do today - all we have to do now is create a new font and download it, which is considerably easier than casting new type or refitting a typewriter.
Early Esperanto used orthography based on Polish. For that orthography he wouldn't have needed any special type or equipment.
But "źurnalo" in the orthography of "pra-esperanto" is considerably less recognizable for Western Europeans than "ĵurnalo". I guess he felt that the benefit of having recognizable words was greater than the benefit of being able to type the words easily; I suppose he was also influenced by the fact that he lived in a handwritten world, where personal correspondence was hand-written, and only professionals would ever have the need to typeset language.
In today's world of Unicode, I think the alphabet works just fine. I think Zamenhof would be amazed if he knew what is possible in typing today.
In some respects he was perhaps a visionary in the field of alphabets. At the time he created the language, you would have to have a special typewriter custom made to use Esperanto's unique characters, or to have print specially cast. In spite of the obvious benefits of choosing alphabetic characters that already existed, he chose not to. All of that is considerably more effort than what we have to do today - all we have to do now is create a new font and download it, which is considerably easier than casting new type or refitting a typewriter.
Early Esperanto used orthography based on Polish. For that orthography he wouldn't have needed any special type or equipment.
But "źurnalo" in the orthography of "pra-esperanto" is considerably less recognizable for Western Europeans than "ĵurnalo". I guess he felt that the benefit of having recognizable words was greater than the benefit of being able to type the words easily; I suppose he was also influenced by the fact that he lived in a handwritten world, where personal correspondence was hand-written, and only professionals would ever have the need to typeset language.
In today's world of Unicode, I think the alphabet works just fine. I think Zamenhof would be amazed if he knew what is possible in typing today.
