Localization: Mind the Process

Dr. Nermin Al Sharman

• In localization discourse, the distinction between "locales" and
"languages" is crucial because it emphasizes the importance of
regional, cultural, and contextual variations rather than merely
translating language.

• "Locales" encapsulate both linguistic and cultural elements, which

is essential for accurately adapting digital products to meet the
specific needs of users in different regions. For instance, locales
consider variations such as date formats, currency, and cultural
nuances, making products more relevant and usable for a particular
market segment​.

• In practical settings, such as when changing interface settings on

social media or apps, they are typically labeled as “languages” rather
than “locales.” This is a simplified term for user-friendliness, though
it does not account for the nuanced cultural and regional adaptations
that "locales" would include. In localization discourse, "locales"
would be more accurate because it reflects those broader cultural and
regional specifics beyond language alone.
• Internationalization includes product requirement analysis (global-local), product design, product internationalization, and
product testing and QA. The cycle then moves on to the localization stage, which includes product localization, testing, and
local product marketing support (that is, localization of promotional materials, setting up a support team structure in the
local language, such as phone centers, help pages, chatbots in the local language, etc.).

• The objective of the processes included in the internationalization side of the cycle is what we can call
“internationalization readiness”. This refers to when a product has been properly internationalized for localization,
meaning that known issues that may cause difficulties at later stages have bee resolved. This concept has also been referred
to as “localization-friendly development”
• Some of the issues to avoid are culturally defined, such as whether graphics, colors, icons, or abbreviations
are appropriate for future target locales. For example, the use of colors or icons has been widely studied
because they can have different meanings. Black might be the color of mourning and white is the color of
purity and weddings in the Western world, but in India and China, white is the color of mourning.
Can you research more differences in the use of
colors across cultures and how this can impact
localization? Are there differences across cultures
in the use of emojis?
• Internationalization readiness is mostly concerned with technical issues, such as character encoding, shortcut
keys, user input formats, or the “externalization” of localizable assets.

• Externalization involves the separation of code from locale-specific resources (text, images, videos). This
process involves developing the code itself in a neutral locale, with no actual text immersed in the code. All
the linguistic and visual assets, including those of the locale of development, are kept separate in external
resource files for each version.
• Internationalized products are also prepared to
accept, process, and display data in different
languages and scripts. This applies to things such
as character encoding, text direction (right-to-
left (RTL), left-to-right (LTR), top-to-bottom),
first name-last name combinations,
alphabetical sorting orders, or address
formats.

• For example, if we create an online form that

might be used in Spanish-speaking countries and
in China, we might need to account for the fact
that Spanish speakers normally have two last
names while Chinese speakers usually put their
last name before their first.
• If our application uses alphabetical sorting to organize and display data, the alphabetical sorting order varies depending on the locale.

1. Swedish Alphabetical Sorting with Accented Characters

 The Swedish alphabet has extra characters, including “å,” “ä,” and “ö,” which come after “z” in the alphabet. If you had a list of names
in Swedish, such as "Zara," "Åsa," "Anders," and "Örjan," they would be sorted in this order: A, Z, Å, Ä, Ö

 In Swedish, these accented characters are considered separate letters, not just variations of “a” or “o.” So, in alphabetical sorting, they
appear at the end rather than alongside their base characters (e.g., “a” or “o”).

2. Spanish Alphabetical Sorting with "Ñ"

The Spanish alphabet treats “ñ” as its own distinct letter, which comes after “n.” If you were sorting words alphabetically in Spanish, such
as "nino" (boy), "niña" (girl), and "nuevo" (new), they would be sorted liplace in the sorting order, immediately following “n.”

ke this: nino, niña, nuevo

 Spanish views “ñ” as a separate letter from “n,” so it gets its own
• 3. Hungarian Alphabetical Sorting with Multi-letter Characters

 In Hungarian, certain sounds are represented by two or three letters, like “cs,” “dz,” and “dzs.” These are considered unique
"letters" in the Hungarian alphabet. If you were sorting a list in Hungarian, with words like "cseresznye" (cherry), "cica" (cat),
and "darázs" (wasp), the words would sort in this order: cica, cseresznye, darázs

 Hungarian treats these multi-letter combinations as individual letters, meaning they are sorted differently than a simple "c" or
"d.”

• Why Sorting Rules Matter

 When creating forms, contact lists, or databases for an international audience, it’s important to follow each language's sorting
conventions. If the sorting is done incorrectly (like putting “ñ” next to “n” in Spanish), it can make the list feel unfamiliar or
confusing for native speakers. Adapting these rules in localized products ensures users can quickly and intuitively find names
or words, improving the overall user experience.
• Software products also use what are known as
‘shortcut keys’, that is, combinations of keys that
make sense to the end user. You probably know that
Ctrl + C refers to the ‘Copy’ function in a word
processor, or Ctrl + S to the ‘Save’ function, if you
are using the English language version. Localization
might make shortcuts meaningless to the target
locale user. All the specifications for all languages
and locales can be found in common repositories
such as the Unicode CLDR project
(https://cldr .unicode .org /index)
• Textual expansion during localization has been one of the most significant technical issues. This problem
results from the fact that different languages need more or less space to express the same content. Examples of
languages that require more space to express the same meaning than the English language are German (+40%),
Vietnamese (+30%) Spanish (+25%), Hindi or Burmese (+15%). On the other hand, some are on average
shorter, such as Chinese (varies widely) or Japanese (from −15% to −55%).

• Nowadays, most webs or apps are developed using what is known as “responsive design”, meaning that they are
developed to adjust dynamically to any screen size. This solves many issues of expansion or contraction caused
by differences between writing systems or scripts. However, this might not always be effective. “Pseudo-
localization” or “pseudo-languages” can be used to check for issues of expansion or textual direction or scripts.
• Pseudo-localization, or the use of pseudo-languages, is a process in software and website development to
simulate the localization process by creating artificial or "pseudo" versions of the text. This process allows
developers to test for potential issues with text expansion, directionality, and font support before actual
translation takes place.

• This process helps catch issues early in development, making the actual localization and translation process
smoother and reducing the risk of design and functionality problems in the final product.