|
|
|
|
Web-only Exclusives November 30, 2000 From Our Correspondent: Hirohito and the War A conversation with biographer Herbert Bix From Our Correspondent: A Rough Road Ahead Bad news for the Philippines - and some others From Our Correspondent: Making Enemies Indonesia needs friends. So why is it picking fights? |
OCTOBER 8, 1999 VOL. 25 NO. 40
Language Lessons
Software is finding more than one voice
By WARREN CARAGATA
ALSO IN ASIAWEEK
| |
|
 Illustration by Simon Wan |
|
Until recently, lessons have been incomplete. Most commercial software is English-centric, written by Western programmers for Western users. Entering Chinese characters into a word-processing application or corporate database programmed for the English alphabet is a waste of time - the computer simply doesn't recognize Mandarin. Only when software is exported is it modified for the local language, and once modified, it tends to lose its mother-tongue capabilities.
For people who work in different languages and for companies with worldwide operations, this monolingual nature of software complicates life. Multinational companies must buy different versions of the same program to suit users in various countries, but that's a relatively minor headache. A larger problem arises when exchanging information and ideas internationally via networks. People using Chinese software can't access vital corporate data stored in English-speaking computers. A document written in Beijing can't easily be shared with employees in New York.
There's a solution, and thanks to the ongoing globalization of the software industry and commerce in general, it is rapidly being implemented. The biggest software makers, including Microsoft, are now incorporating into their products something called Unicode, an emerging industry standard that enables computer programs to handle virtually all languages without radical modification.
Simply put: Unicode is an improved format for translating human-language characters and symbols into machine language. The current standard, called ASCII, uses seven-digit combinations of ones and zeros to represent characters. Thus the letter "a," for example, is represented in machine language as "1100001." ASCII is sufficient for the 26-letter English alphabet. But seven "bits" (a bit is a one or a zero, the atom of computation) aren't enough to represent all the characters in complex lexicons such as Chinese and Japanese. Unicode is a 16-bit system, vastly increasing the possible combinations of ones and zeros. Software programs written under the Unicode standard can handle more than 65,000 unique characters, including those for Japanese, Korean, Chinese, and Taiwanese, as well as the mathematical symbols for each.
The switch to Unicode, already underway, may not be obvious to most users, but it is nevertheless significant. The standard has already been endorsed by companies such as Oracle, maker of industrial-strength databases used by big companies, and Unicode is used in several operating systems that run powerful corporate servers. Microsoft is bringing Unicode to the masses by adopting it in Office2000, the popular software package that includes Word and Excel. Indeed, Unicode will be spread throughout the company's product line. "We're going Unicode everywhere," says Richard Kartawijaya, president of PT Microsoft Indonesia.
That means someone using Office2000 can store and display Chinese characters correctly even if they are not using the Chinese version of the product. Different versions for each market are still sold but the only changes in the program deal with minor issues such as the interface and error messages. With a single installation, users can now choose to work in whatever language they need, with the appropriate tools like spellcheckers ready to deal with any kind of script that is on the screen.
Ricoh Hong Kong is one company that has already purchased Office2000, which went on sale in June. The Japan-based producer of copiers, printers and other office automation equipment previously had to install add-on programs called double-byte enablers to handle Japanese and Chinese scripts. Now, says Paul Lee, the company's systems supervisor, employees can write in different languages in the same document. Databases and spreadsheets can contain English and Chinese or Japanese data. One shrink-wrapped box fits all.
Of course, Unicode will not magically erase language barriers - it can't translate a Chinese sales report into Japanese. But many analysts believe that the benefits ensure Unicode will eventually supplant ASCII as the standard character-coding format. The forces of globalization have already spurred businesses to think in more than one language. It's about time that the software started to catch up.
 |
