Localization that understands the users

In the localization industry, it is crucial to have an in-depth understanding of end users’ linguistic preferences. Only when we understand the nuances of the language that is used by the users – in their homes, outside their homes and the many other variants other than the two forms – can we localize in a way that suits them and may be even touch their hearts. But, the users we hope to reach, number in crores and are spread across multiple demographic categories such as gender, age, education / economic status, area of residence etc. Each of these categories affects their linguistic preferences. Just like there are different categories among people, many products are developed and localized, targeting a specific category or categories of the public.

Gender

Telugu grammar rules have the Masculine and Feminine genders and different versions of words in accordance with that. However, like in majority of the world’s languages, centuries of patriarchy has affected it in the form of defaulting to the masculine gender in many cases. Since the masculine versions are more popular, when we do use the feminine versions, readability gets affected. Eg. – The masculine version నిపుణుడు (nipunudu) for ‘expert’ is well known unlike the feminine version నిపుణురాలు (nipunuraalu). So, in order not to affect readability by using the feminine version, the plural ‘నిపుణులు’ (nipunulu) is often used. Using a plural adds an honorific to that person as well. But, it creates a problem. In cases when we have to refer to a singular female expert, we can’t use this form and have to stick to the feminine version.

Another way to address this problem is to go for a transliteration. The word ‘expert’ when transliterated is gender-neutral and can be used as ‘expert’ for singular and ‘experts’ for plural. However, like we discussed in the previous posts, we would encounter a resistance to the usage of transliterated words. Even if we set aside opposing viewpoints around transliteration, it is possible that the transliteration may not be understood by telugu users in every case and therefore can’t be used.

Based on the end users and context, we have to choose between a plural, a transliteration or going for the feminine gender. 

Like most languages, Telugu recognizes only two genders among human beings. Transgender people who do not conform to these gender binaries were officially recognized by the Supreme Court of India and it is high time that Telugu recognizes this fact. The way we have ‘అతడు’ (athadu) for masculine and ‘ఆమె’ (aame) for feminine genders, there are no words yet to respectfully refer to transgender people. For example, a gender-neutral pronoun ‘they / them’ (third person singular) has been recently approved and is being implemented in the English language. Such changes need to be brought in Telugu as well. We should evolve the language in a direction that corrects the centuries of injustice meted out to this section of humanity.

If we evolve our language in this manner and deploy it during localization, that would receive appreciation from end users irrespective of their gender.

Age

Let’s have a look at apps that provide online education services. Their target audience is likely to be children and occasionally their parents. In such cases, the language used in localization needs to be simple and easily understandable by kids.

Eg. – A string such as ‘Topics for revision’ – if it is translated as పునశ్చరణ చేయాల్సిన అంశాలు (punascharana cheyalsina amshaalu), kids will not understand it. Don’t be surprised if their parents don’t either. Instead, if it is translated as రివైజ్ చేయాల్సిన టాపిక్స్ (revise cheyalsina topics), that would be keeping the end users in mind. Products with such localization would be well appreciated by the users.

In a similar vein, another example – apps that sell medicines online. Usually, their target users are above 50 years of age. Therefore, the language there needs to be more formal.

Education

Imagine a product being localized, that is directed at experts in a specific field, or at those who at least have a minimal knowledge of the field. In such cases, the terminology being used needs to be consistent with what is being already in vogue in that field. Only then can confusion be prevented when the users of this product need to discuss with others in the same field. 

Eg. – Imagine a web search engine’s User Interface (UI). A string such as ‘There were no results for your search query’, instead of translating it as మీ శోధన పదబంధానికి ఫలితాలేవీ రాలేదు (mee shodhana padabandhaaniki falithaalevi raaledhu), if it is translated as మీ సెర్చ్ క్వెరీకి రిజల్ట్స్ ఏవీ రాలేదు (mee search query ki results yevi raaledu), users understand it more easily. Moreover, they will not have to learn new terminology just to use the localized version of the app. Additionally, when they have to discuss the UI with others, the latter can easily understand what the former are trying to convey.

Region

Depending on the region, there are multiple dialects that are currently in vogue among the telugu states. They are – Telangana, Rayalaseema, Godavari, North Coastal Andhra, Nellore etc. While these are widely recognized as unique, there are finer variations among these dialects as well, as one travels from a district to the neighboring district. Eg. – The Rayalaseema dialect used in Anantapur is slightly different from that used in Chittoor.

By and large, there has not been a question of deploying these dialects in the mainstream media. This is because, official Telugu was always the one in which any regional dialect has been stripped away. It has been the norm to use such formal telugu typed out in newspapers and spoken out loud by television news anchors. However, after the formation of Telangana state, the corresponding dialect has been encouraged and deployed prominently in newspapers and television programmes.

Any product that hopes to connect strongly with the users, it would be good idea to keep these dialects in mind as well. When a regional dialect is used correctly and at the correct user demographic, the product would distinguish itself from the wide majority of products that use the same boilerplate language. It would satisfy the user in the same away one feels on seeing a person speak one’s own language in a foreign land. Instead of every single product, dialect-based-localization is more effective when it is used in products that directly speak with users, such as virtual assistant apps.