Mobile Phone Localization in India

languagepanelThe Telecom Regulatory Authority of India (TRAI) is proposing, as part of the government’s Digital India push, to make regional language interface capability mandatory on feature phones. These standards are expected to go into effect in six months.

One of the world’s leading localization companies, Moravia, has weighed in with a blog post entitled “India Gets Serious about Mobile Phone Localization“. Unfortunately, this post reveals a fundamental confusion about almost every aspect of this situation: the difference between feature phones vs. smart phones, between language support on phones themselves vs. applications, and between localization of interface vs. content. The post says that India is:

already one of the fastest-growing smartphone markets in the world, and the second largest overall. India’s vast hinterland is smartphone-equipped and hungry for local language content. The communities are cash-rich, and India’s e-commerce companies will have to reach out to them in earnest if they want to stay in the game.

this single paragraph neatly encapsulating all the confusion.

First, the TRAI proposal is targeting feature phones, not smart phones. Technically it applies to smartphones, but that is largely moot, since smartphones already provide excellent language support, especially at the level of text handling (input and display). Many also provide localized interfaces. This problem is essentially already solved.

Second, what the “hinterlands” may be hungry for is not “local language content”, but primarily rather the ability to read and write (text and mail) in the local languages. They also would be most surprised to hear themselves described as “cash-rich”.

Third, the e-commerce companies are already making economically-informed decisions about what kinds of localization to provide, and in any case the TRAI proposal does not apply to them–a government agency cannot mandate that third-parties provide localized interfaces or content. The point about e-commerce companies needing to “reach out” if they want to “stay in the game” is a little bizarre. It seems to assert that the e-commerce companies are too stupid to understand the linguistic profiles of their own markets, or too incompetent to do anything about it. Neither is the case. The e-commerce companies are acting in economically rational fashion in balancing the cost/benefit trade-offs of localization. At the heart of the trade-off is that fact that the layers of the pyramid with enough money to spend on e-commerce purchases are not just comfortable using English, or to a lesser extent Hindi, but may actually prefer doing so. The e-commerce companies can be counted on to provide localized interfaces and/or content at exactly the point in time when their teams of MBAs determine that doing so makes business sense. Nor do the e-commerce companies need a TRAI directive on feature phone language interface support to alert them to the fact that India has many languages.

The Moravia post goes on to inveigh against the use of machine translation. I assume it is referring to translation of product descriptions and/or user reviews. However, the volumes of both, especially user reviews, are so great as to make human translation, or even post-edited MT, completely unfeasible. Even Western companies such as TripAdvisor use unadorned MT on translated reviews. There would seem to be no viable option here, unless Moravia is proposing that Flipkart spend millions or tens of millions of dollars translating reviews into a half-dozen or a dozen local languages for negligible benefit.

The simplistic view that hey, India has 22 languages, so gosh, we better localize everything into all of them, ignores the nature of the linguistic map in India, the role of Hindi as a lingua franca in northern India and English as a lingua franca in southern India, and the relationship between language proficiency and economic strata.

2 Responses to “Mobile Phone Localization in India”

  1. Vijayalaxmi Hegde Says:

    Bob, many thanks for joining the discussion! This is an exciting topic, and you provide some good insights, however, some of your statements seem more like assumptions rather than being close to what we are trying to say in the blog post.

    1) We aren’t really saying that TRAI is bringing this proposal for smartphones, or that it is forcing the e-commerce companies to localize. Instead, we have tried to show the extent of the demand for local language content by this ruling. And, by extension, we want to show the potential that businesses are currently leaving untapped in India. And yes, we are well aware of the difference between smartphones and features phones, hence the note in the last paragraph about the local language support and Indian language input being effectively resolved in case of non-feature phones.

    2) About MT: perhaps you assumed that we were against MT for user reviews. If so, we’d like to clarify that we were not referring to user-generated content.

    Some ecommerce companies in India, as you may know, had recently applied MT on their sites without even a minimum level of post-editing. This resulted in very poor translations, enough to actually turn away users. We should also note that the content on these sites very often have a messy structure, even when they are just in one language. How then can they hope for the translations to make sense, especially when done through machine translation?

    3) Also, we didn’t say in the post that you will need to translate into all 22 Indian languages. (although we don’t see why not? No one would really think of making such a comment in the EU, for instance, where there is a high awareness about linguistic rights.)

    We are making the point that you can begin with Hindi but can’t stick to only one language and hope that it will suffice for all of India. If Indian companies indeed come to the conclusion that just Hindi and English are enough, then they should get rid of all marketing efforts and customer support that they are currently being provided in multiple Indian languages.

    4) And about India’s hinterland being cash-rich: I don’t need to say much about something well-discussed:

  2. rtm Says:

    Dear Vijayalaxmi, thank you for your comment, the links about the rural super-consumer, and new information about misguided attempts by some e-commerce companies to translate core content with MT, which I had not known about.

    However, my main point is that I have always found it odd that localization companies seem to think it could be useful as part of their marketing to convince companies to do localization. It’s highly patronizing–a we know better than you attitude. The Indian e-commerce companies understand the linguo-economic profile of their target markets very, very well–it’s their job to do so. They would be the last to leave money on the table by failing to reach out to important demographics in their own local language. But there’s a very real cost of the localization work–both initial and ongoing–that they have to factor into their decision. And there’s an opportunity cost for focusing on localization, even if the ROI is positive, since some other initiative could have even higher returns. As I’m sure you realize, there are also compliance and legal issues surrounding localization of product descriptions etc.–a mistranslation could open them up to liability. There is also the issue that there is a dearth of qualified translators for some of the local languages.

    If Moravia or any other company has innovative ideas about how to structure the localization process into additional Indian languages to get more bang for the buck, I’m sure they’d be all ears, but I don’t see merely nagging them to do more languages as a useful endeavor.

Leave a Reply