[KS] The Mystery of the Breve

Otfried Cheong otfried at airpost.net
Mon Sep 14 04:18:23 EDT 2009


Frank Hoffmann wrote:
>> Just because it is in Unicode doesn't mean you can actually use it
>> in a given web application,
> 
> Please point out why not so. I see no reason why this would not be
> possible, other than, as pointed out, in outdated scripts.

Please think it through for a moment.  The address I enter into a form
on amazon.com's website is used to print an address label, which will
guide a package through the US mail system to Korea.  You are
essentially saying that the system printing labels should be able to
handle arbitrary Unicode characters, and that the US mail system should
be able to deliver it!   In principle, you are right:  As long as I
indicate the country name in English, I should be able to write the
address in Hangul, the address label should contain correct Hangul, and
all would be fine - and in fact no romanization would be needed at all! 
But that's not how it works, and that's not a purely technical issue:
apparently Amazon wants to be able to read the address, for instance if
there are issues with the credit card payment.  So even if their system 
is entirely modern and uses Unicode everywhere, they will still restrict 
the range of input characters they will allow.

Similarly, when I apply for an entry permit or visa on the web, the visa
officer wants to be able to read what I have written.   So the web
application must do some sanity check on my input.  Where to draw the
borderline?  You say roman letters with diacritics are fine, but Greek
and Cyrillic are not?  How many diacritics?  One, two, or full Vietnamese?

These are _not_ purely technical problems.

In my opinion, the fact that Koreanists in 1999 were so adamant that
this was all a technical issue that (a) was already solved, or (b) would
be solved shortly, or (c) would be solved in due course, was the main
reason that their input was greatly ignored - every Korean who had used
an overseas web site knew better, and they were just making up their own
romanization on the spot.

Ideally, a romanization system should be such that Koreans would feel
comfortable to use it to spell their name.  RR falls short of this goal,
as nobody would spell 김 as "Gim", and "Bak" is still quite uncommon
(but on the rise - in general, you see far more young people now using
the official romanization for their name).

But this goal means that diacritics are out.  Nobody would voluntarily
use diacritics in their name, and I doubt Korean passports can actually
be printed using diacritics (a technical issue?).  Certainly air tickets
cannot handle diacritics.  I know a number of, say, German scholars
working in the US, who have entirely given up on their umlauts or ß
letters, so, say, "Günter Groß" will be "Guenter Gross" or even "Gunter
Gross" on his social-security card and even on his scientific papers.

>> ...in the Netherlands a few years ago, I found that their
>> administrative system could not handle umlauts, and so they had to
>> type AE instead of Ä.  And this in a European country where
>> diacritics are used!
> 
> Yes, and in the 2nd grade of primary school I wrote "school" with one
> "o" and there may still be kids doing that now, and so we better
> should not implement orthography rules? (Unfortunately they are
> already implemented).

My point was that administrative systems all over the world, including 
resident registers, IATA software, banking software, etc, will not 
handle anything outside the good old 26 roman letters A-Z, even in 
countries where diacritics are used to write the national language.

I fail to see the relevance of your answer to this point. Please try to 
convince your bank or national visa office to update their software - 
good luck!

For what it is worth, I am a professor of computer science, and I have 
done consulting from 1998 to 2004 for companies implementing 
Unicode-based input-methods for the CJK languages.  I believe I know a 
little bit about Unicode, and about the amount of work it will take to 
convert existing systems.  But if you tell me that it's easy and will be 
fixed soon, I will of course bow to your superior wisdom.

Best wishes,
  Otfried






More information about the Koreanstudies mailing list