TechHui

Hawaiʻi's Technology Community

Companies that fail to internationalize their web applications at an early stage often miss out on big opportunities. With China having recently passed the US in terms of total web users and European online advertising revenue rising 40% annually, it is absolutely crazy to ignore international markets. It constantly amazes me how many successful US web app companies wait too long to localize their applications to Chinese, Japanese and European locales and end up conceding these large lucrative markets to local copycats. LinkedIn is a perfect example of a company that really missed the boat in terms of Asia and Europe because they took too long to localize to these markets. I wonder if this was due to technical challenges arising from their application not being properly internationalized (i.e. designed for localization) or a lack of foresight on the part of management. It doesn't take an enormous amount of effort to do proper internationalization of an application if you keep it in mind from the start. To ensure our new ooi social platform is properly internationalized, we are doing simultaneous UI development in Chinese and Japanese. This forces us to deal with two complex locales from the start. This approach ensures English only assumptions don't creep in during the development process. I estimate the total overhead of the internationalization and localization work will be less than 10% of the overall development and Q&A costs. 10% more effort will give us access to more than twice as many users. In terms of cost / benefit, its a no-brainer! Ikayzo - Design • Build • Localize | Web • Desktop • Mobile

Views: 45

Comment

You need to be a member of TechHui to add comments!

Join TechHui

Comment by Daniel Leuck on August 2, 2008 at 1:49pm
John: Question - How do you handle internationalization for content in applications? I understand the aspect of internationalizing the UI and the controls of the applications but what about content. For instance, a site like techhui, even if the app was internationalized, what about the posts and the comments. If the posts and comments are not translated, it would still be hard to get much value out of a site. Am I missing something?
Thats a great question. There are a number of options. One is to integrate web managed human translation work flow. Professional translation is expensive, so you usually want to have a management UI that allows you to flag valuable content for translation. Your web application can then submit the content to companies like SDL to be queued for translation by their network of translation professionals. SDL offers a variety of options including use of translators with domain specializations and the option of adding a proofreader to the process. They offer a variety of multilingual content management products and services.

If you want to go quick, dirty and cheap you can provide machine translation. These days it works reasonably well if you are doing a language pair like English and Spanish, but not so well for pairs like English and Japanese. As I pointed out in last month's Hazards of Machine Translation blog post, this approach has its drawbacks :-) It should only be used for casual correspondence where you are content to convey the general idea of the content, even if it sounds like its being delivered by a rambling drunk.

Another option is to provide a UI to your users and allow them to provide translations. The effectiveness and quality of this approach obviously depends on the motivation and skill of your users.

Finally, you can simply punt on translation and segment your application into different language user bases. Each group of users posts content for consumption by other speakers of their language.
Comment by John on August 2, 2008 at 11:09am
Question - How do you handle internationalization for content in applications? I understand the aspect of internationalizing the UI and the controls of the applications but what about content. For instance, a site like techhui, even if the app was internationalized, what about the posts and the comments. If the posts and comments are not translated, it would still be hard to get much value out of a site. Am I missing something?

People emailed me from France, Greece and Malaysia volunteering to translate content on my site. I am happy that these people reached out to me and if they want to do that's great but it seems that such a process is going to take a ton of ongoing effort. Thoughts?
Comment by Daniel Leuck on July 29, 2008 at 11:07pm
Truman Leung: I found that by working the localization functionality into the UI, it became a marketable feature that allows the client to change any text string to suit their particular style and purpose ... even in the same language.
That is an interesting side effect. It also lets customers in the same locale use different voices (writing styles.) One client may prefer a casual voice while another prefers something more corporate.

The Facebook and Ning strategy of having their users do the localization is an interesting approach. You don't always end up with the highest quality translations, but it does enable any properly motivated individual with the requisite language skills to bring your app to their locale.

John: Companies that win on network effects (like building a network of business contacts) depend on having critical mass. If a foreign copycat gets a critical mass of a country's business contacts, this places a significant obstacle to LinkedIn to overcome. Not all companies win on network effects though so I think the argument is stronger in certain types of business structures.
True. The benefits vary based on the type of company, but I think a cost benefit analysis is likely to come out in favor of internationalization from the start for most general purpose web applications.

John: Also for some markets, people who speak English are worth far more than people who do not speak English.
True, but even in these cases localization is beneficial. Mika speaks English well, but if given a choice between a web app that has been localized to Japanese (her native language) and one that hasn't, she will pick the former every time. Its an issue of comfort and convenience.
Comment by John on July 29, 2008 at 8:30pm
I agree that Linkedin is a good example of a company that should have internationalized early. Companies that win on network effects (like building a network of business contacts) depend on having critical mass. If a foreign copycat gets a critical mass of a country's business contacts, this places a significant obstacle to LinkedIn to overcome. Not all companies win on network effects though so I think the argument is stronger in certain types of business structures.

I believe Hawaii has above average internationalization expertise. Lots of startups do not have anyone with such expertise in house so it makes it more risky to try.

Also for some markets, people who speak English are worth far more than people who do not speak English. If only because they are wealthier. About 1/3 of my current market is from non-English speaking countries but they all seem to read and speak English.

In any event, I do think internationalization is important and that it will increase significantly over the next decade.
Comment by Brooke Fujita on July 29, 2008 at 5:52pm
Almost 10 years now... And I gots the salt-and-pepper hair on my head to show it!
Comment by Daniel Leuck on July 29, 2008 at 5:51pm
Brooke: It was hellish dealing with all of those resources and translators back at WorldPoint, but we still managed to display in over a dozen different languages, didn't we Dan?
Indeed we did. Its a lot easier now than it was back then. How long ago was that? 10 years? We are getting old my friend! :-)
Comment by Brooke Fujita on July 29, 2008 at 5:43pm
> It doesn't take an enormous amount of effort to do proper internationalization
> of an application if you keep it in mind from the start.
Actually, I think it does take an enormous amount of effort to manage it correctly.

But compared the effort and cost of doing it as an afterthought? Best to pay it up front, using people who understand the issues and have the know-how. It was hellish dealing with all of those resources and translators back at WorldPoint, but we still managed to display in over a dozen different languages, didn't we Dan?
Comment by Truman Leung on July 29, 2008 at 7:30am
I found that by working the localization functionality into the UI, it became a marketable feature that allows the client to change any text string to suit their particular style and purpose ... even in the same language.

Sponsors

web design, web development, localization

© 2024   Created by Daniel Leuck.   Powered by

Badges  |  Report an Issue  |  Terms of Service