Back to Contents of Issue: November 1999
by David Doran |
|
The tremendous rise in the
penetration rates for PCs in developed Asian nations such as China, Japan, and
Korea has spurred something of a mad scramble by U.S. software companies seeking
to enter a large and fast-growing market, worth an estimated $20 billion in 1998
[IDC research].
Of course, bringing U.S.-made software to the Asian market requires far more than just slapping a sticker describing the software in Japanese or Mandarin on the box. The software must be localized, a process which requires not only a clear, concise translation of the language used in the interface and documentation, but an overhaul of the single-byte character set used to represent English Latin letters into a double-byte system suitable for Asian languages-which use thousands of possible character combinations. For most smaller software companies, this process usually takes place well after the original English version is complete, but according to Jim Ladd, president and CEO of Kirkland, Washington-based EnCompass Globalization, preparation for the Asian market should begin during the initial design phase of the software. "In order for us to localize the product, the original English documentation must be technically accurate," says Ladd. "If we are given an inaccurate or ambiguous description of the software, there is no way we can improve it to make it work in Asia." EnCompass often has to deal with another aspect of software design that usually never occurs to software engineers-cultural relevancy. "We often have to modify things like icons, pictures, shapes, colors . . . some of these things are not applicable to the target country. For example, if you put someone's picture in a black frame, in Japan, that means that the person is dead." Ladd likes to illustrate this point by describing the subtle changes his company made to e-commerce software that was being localized for the Japanese subsidiary of Recreational Equipment Inc. When a customer in the U.S. chooses a product that is currently out of stock, the website displays a picture of a stylized man giving a "what can you do?" kind of shrug-meaningless to Japanese customers. The Japanese version was modified to show the man bowing in polite apology instead. Of course, there is a highly technical aspect to software localization. "The code of the U.S.-made software must be written from the ground up to be localization-friendly," says Tsugu Yamasaki, EnCompass's vice president. "The original software should not have hard-coded strings, which are English characters embedded into program code. The localized version needs to have Japanese strings in there, so if you want to change the software from English to Japanese, you must create another set of source files, a long and costly process. In order to avoid that, you have to extract all culturally dependent strings, so that the main source code does not need to be modified. We encourage our clients to practice this when designing software. If they design it that way from the start, it saves lots of time later." Another consideration for U.S. software engineers should be the inclusion of pre-packaged third-party components, such as Installshield, in the source code. "These programs must also be localized to work well with localized software," says Yamasaki. "Many times when we get the software, those little programs are missing or not localized, but without this product, the software cannot function properly. If the third-party component is available in Japanese, than it's easy to replace it, but if it's not localized, we have to insert one and make it work with the localized software. Even if a version is available, our client must get a license for the localized version." Choosing the right localization company to modify your software product can mean the difference between a profitable life and an obscure death in the highly competitive Asian market. But Ladd believes that in the case of globalization, bigger doesn't always mean better. "There has been some consolidation in the localization industry as it has matured. The market is pretty much dominated by half a dozen large companies, which usually have a European orientation. I won't say that it's easy, but European languages don't have some of the re-engineering complications that Asian languages do. But we're not trying to do every language in the world. We specialize in Asian languages, and we're trying to be the best in that particular niche. Our specialization allows us to be more competitive, more efficient, and to put out localized products of higher quality." One way to tell a truly specialized localization firm is to find out how much of the translation and re-coding work is outsourced to independent contractors, as well as what percentage of the personnel are native-speakers of the target language. The extra effort required to make software products localizable for the Asian market may seem somewhat daunting for a smaller company, but consider this: Japan, the world's second-largest software market, and China, with huge potential for growth, are voracious consumers of U.S.-made software. Can a software company expect to thrive in a market dominated by giants like Microsoft without keeping a hungry eye on the Asian market?
Jim Ladd: jiml@encomp.com |
Note: The function "email this page" is currently not supported for this page.