JDJ: Could you introduce yourself and tell JDJ readers about your role promoting Java in Japan?
ST: My title is Manager of ISV (Independent Software Vendor) Marketing at Nihon Sun, and my mission is to support ISVs in addition to enterprise and other related consultants. Nihon Sun sells almost all of its products through channels such as CTC, Fujitsu, Toshiba, Unisys, Xerox, Matsushita, NEC, etc., who are all providing consulting, system integration and software development services. But there are many independent software vendors and system integrators, and my role is to support those companies. For Java, I'm fully responsible for promoting localization and the development of original Japanese software. Of course, JavaStation and other products are handled by the product marketing division of Nihon Sun. My mission includes promoting Solaris and Sun platforms and promoting Java computing. But recently I don't feel it's so necessary to promote Java here because it's become so popular.
JDJ: What do you think is behind Java's popularity in Japan that might be different from the US?
ST: Maybe it's similar to the situation in the US, or maybe we're following the US. What's unusual is that Japanese companies normally take a long time to evaluate new technology to find out if it's here to stay or not. That was the case with concepts like client/server computing and data warehousing. But as for Java, the time lag from introduction to acceptance has been quite short. I think it's because Java represents a chance for Japan to catch up in the area of networking. Japanese companies took much longer to install network technologies. Corporations, and even our society as a whole, have become worried about the disadvantages inherent in falling behind in this area. Japanese infrastructure is still behind, but Java computing and Internet/Intranet computing as a whole can be opportunities to narrow the gap. I think it's a good opportunity, and the explanation for Java's popularity here must be that so many others think so too.
JDJ: You mentioned Japanese infrastructure was lagging behind. Can you describe Japanese infrastructure as compared to that in the US?
ST: I'm not as familiar with the US situation, but even in the US, there was a need to deregulate the telecommunications industry and there was some resistance to it. Just now, the Japanese telecommunications industry is struggling with deregulation issues. People in the industry are clever enough so far to dodge these issues. So it means that prices remain high, and bandwidth narrow. Also, Tokyo is so crowded, so it's not so easy to lay cables between new offices. Wireless technology is not widely used due to security concerns.
JDJ: What are the main platforms here on which Java is being used?
ST: We have many of the same platforms here as in the US, but Japanese manufacturers have their own operating systems also. Fujitsu has their own, NEC has their own and Hitachi has several. Of course, all of those vendors are very willing to provide Java virtual machines for their own platforms. I believe convergence to Solaris is proceeding more and more, but the current situation isn't necessarily a bad one. It stems from the pride of Japanese manufacturers - they have their own identities and their own technologies, so they spend a lot of money on very redundant efforts. But to some extent, it invigorates the Japanese software industry and Japanese ISVs. That's very good.
JDJ: JDJ readers would like to know about opportunities in Japan. One might be bringing applications developed in the US to Japan ("Japanization"). Do you see that as a big opportunity for US application developers?
ST: Yes it's a big opportunity, but the difficult part about bringing applications here is modifying them to fit Japanese business practices and behavior. These practices, such as payroll issues and accounting policies, can be quite different from the US. For instance, in Japan the company files income tax returns on behalf of employees, whereas in the States you calculate your income tax and file yourself. That means there's less opportunity for tax deductions, making applications like Quicken less valuable to Japanese individuals. Also, there are different methods for monetary transactions.
Business-to-business transactions often incorporate the use of tegata, a kind of promissory note, and individuals often pay by bank transfer. Checks are almost non-existent here - we don't use them to pay telephone bills or anything like that. So when implementing electronic commerce in Java, a check may be a good payment model, but Japanese people will be unfamiliar with it. So, dealing with these issues can require much greater effort than the technical conversion itself.
JDJ: What about the kanji issue? How difficult is it to take a Java application and make it two-byte capable?
ST: Right now, JavaSoft is finalizing internationalization specifications (finally!). As far as technical porting, I don't see many barriers or difficulties to having applets from overseas support kanji using the internationalized JDK. Before joining Sun one and a half years ago, I was running an ISV that focused on porting overseas software to Japanese platforms. What I found is that supporting kanji and translating menu items and messages into Japanese is the easy part. The hard part is, as I've mentioned, getting a deep understanding of Japanese customs and language.
JDJ: What would you say is the best way for a US developer to bring a hot application to the Japanese market?
ST: I believe the only way is to find a good partner in Japan, a Japanese ISV or SI. Finding porting companies or translators is not a good way. A foreign company should look for a Japanese ISV or system integrator that serves a similar industry segment. For instance, a developer of telecommunications applications should look for a partner in Japan that's already serving customers in that field. As I've said before, you have to know the Japanese market very well. Japanese ISVs and system integrators know their own market. In terms of Java computing and computing in general, the people with the most advanced perspectives are those in the manufacturing sector, rather than retail, etc. These companies direct a lot of resources internally and have a lot of understanding about Java computing. Their names are well-known overseas, and I suggest foreign companies and their partners target them for the best business opportunities.
JDJ: How many people are currently developing Java applications in Japan?
ST: As for Solaris, we have 700 ISVs, and I estimate 60 percent of those are quite active developers. Of those active software developers, about 70-80 percent are involved in or considering Java development projects. We're currently surveying those ISVs to gather some more concrete data to answer that question.
JDJ: Are there opportunities in Japan for skilled Java developers from the US? Or is the language too much of a barrier?
ST:There are some very advanced developers in the US. In Japan, the average skill level is quite high, and there may be more "good" Java developers here than in the US. But the really excellent Java developers are working in the US. That's usually how it is, I think. If such excellent developers are thinking about boosting their salaries, it's a very nice opportunity to come to Japan - partly because of the crazy currency ratio. Of course, your expenses are high, too, but you can come out far ahead. The usual pay in Japan for good or excellent engineers is very high, and I'm sure there are lots of opportunities for them. But if one is looking to develop cutting-edge products, I don't see much difference between the US and Japan. So it depends on the individual, but I think it's an opportunity to make a lot of money if he or she is willing to come to Japan. And Americans and Japanese who can speak both English and Japanese can make much more money. How much is, I guess, up to how well he or she negotiates.
JDJ: Are there examples of US-developed Java applications that have come to Japan successfully?
ST: The earliest example is Applix Anywhere. The reason for their success is that they had a partnership with Ashisuto, a big Japanese company specializing in enterprise-wide mainframe systems and network computing. The key is Ashisuto's technical know-how and experience plus their good contacts with big Japanese companies. It's sometimes difficult, even for Japanese vendors, to get into large Japanese companies, like the electronics and manufacturing giants. To have an account in those companies is a valuable first step. Ashisuto has many business relationships and was able to introduce network computing by Applix Anywhere to those firms. They are now making very big deals with large Japanese corporate customers to use Applix Anywhere and SPARC Ultra Enterprise servers. Some successful US ISVs, successful in the US that is, have started by setting up a Japan office and hiring Japanese staff that speak English well. It is usually the wrong way to start. To have an office is okay. Very expensive, but okay. Still, to find a good partner is the first step. Then, negotiate with them to establish an office here. Another successful case is EPS, a California Java developer that is localizing its product HotForms with the help of ISI Dentsu, a Japanese system integrator with many corporate clients in manufacturing and financial industries.
JDJ: Are there resources available for Java developers looking to build applications for Japan?
ST:Sun is currently setting up such a resource, but the most advanced at this time is the Japanese side of Gamelan. Nihon Sun's home page (http://www.sun.co.jp) also provides links to a lot of resources, but many of those are available only in Japanese.
JDJ: If a foreign developer is interested in finding a Japanese partner, who should be the first person to call?
ST: Please call me! It's one of my biggest roles - to introduce a good partner. But my first and foremost piece of advice is to find a partner that serves the same market in Japan. I, of course, can help introduce such partners. There are several consultants working as brokers. That's another way, but I don't recommend it. Also, some Japanese trading companies, that have their own expertise and computing resources, are very nice partners.Other trading companies are not. But, in general, the key to success in the Japanese market is success in the US first, which generates visibility for Japanese developers and end users. Advanced people are looking, watching the US situation. So if you are visible enough and successful enough, you'll be getting a call. And maybe that's the most cost-effective way, having them call you instead of the other way around.