Software localization is fast becoming the difference between success and failure, as an increasingly connected world demands products that are adaptable to foreign markets. If you haven’t already implemented a software localization strategy, now’s your time to get started.
But before you dive into the logistics, it helps to first familiarize yourself with common industry terms and the various software localization components you’ll be dealing with. That way, you’ll be prepared for dynamic conversations with your localization service provider on how best to optimize your apps for every user.
Here’s a guide to key terms you’ll need to know, and a clear software localization definition.
- Source File: In translation, this is the original—usually English (US)—language text or content used to create the target file. Not to be confused with source code!
- Target File: The source file text or content translated into a specific (target) language. This localized content features the same shape and format as the source (English) file.
- Bilingual File: Contains corresponding pairs of source and target text segments (for instance, sentences in English and Japanese). Reviewing such a document is part of the initial quality assessment.
Image credit: Wordfast
- Strings: Also known as content strings, these are the words and phrases used within your application. Their order and size vary across languages, which is often a key consideration—we’ll get into that later.
- Computer Aided Translation (CAT): Translation carried out by human translators using specialized software. The image below shows the typical workflow from preparation to final translated file (this infographic goes into more detail about how it works).
- Translation Memory (TM): As you may have noticed, translation memory is part of the computer-aided translation (CAT) process. This is a database that stores translated segments (usually of sentence length) with their source language equivalents; for example, Lionbridge’s cloud-based Translation Workspace.
- Concatenation: Combining two separate text strings to reduce the size of a string, resulting in incorrect word order and grammar. Take the common English concatenation, “Free shipping,” as an example of a software localization mishap:
- Hardcoded text: Text that is embedded in the source code and must be extracted via parser for translation. This is less effective than storing metadata in such a way that strings can be easily extracted.
- Globalization: The process of ensuring your application is internationalized and localized, and able to support users in their preferred languages, platforms, and devices. At the current ratio of U.S.-based to international users, there’s undeniable opportunity to cultivate a global audience:
- Internationalization (abbreviated as i18n): This is the process of ensuring that your code and architecture can handle multiple languages and cultural conventions, making localization possible.
- Localization (abbreviated as l10n): The cornerstone of global reach! This is the next step after internationalization, focusing on adapting a product or content to a specific locale or market. Translation is one of several elements in this process—but global websites don’t rely on this alone. Other examples of software localization elements to consider include:
- Address and phone formats
- Date formats
- Units of measurement
- Societal codes (humor, etiquette, rituals, myths, and symbols)
- Societal values, power, relationships, and beliefs.
- Localization-friendly design: Source code and structure that are ready for localization from the start. This means using design and layout templates that allow for customization, for example. Web design best practices can be applied to app interfaces, such as this clean, simple layout by IKEA:
- UTF-8 Encoding: The most popular Unicode format, this is a software localization component you’ll definitely want to use. UTF-8 is a variable-length, 8-bit character encoding capable of encoding all possible characters, or code points, defined by the Unicode Standard.
Image credit: Scott Hanselman
- Over-localization: Occurs when strings intended to remain in English are erroneously translated, due to incorrectly prepping a file for translation.
- Under-localization: By contrast, this occurs when items which need to be localized fail to be included in the file sent for localization.
- Pseudo-localization: A cost-cutting localizability testing technique where localizable text is replaced with dummy, auto-generated “translations” to reveal potential issues before investing in the real thing.
- Software Localizability: The ability of a software product (text and non-text elements) to be adapted for any local marketing with no changes to the source code. One example of a software localization challenge when designing for localizability is text expansion—it’s important to allow for at least a 30% longer string.
- Terminology Database (Glossary): A type of dictionary used during translation to ensure terminology consistency. A glossary is an essential tool for translation quality and localization effectiveness; here’s how to write one.
Of course, knowing the ins and outs of software localization doesn’t end with knowing your jargon. To become truly global-ready and learn how to put these concepts into practice, see our definitive ebook: The Developer’s Dozen: 12 Best Practices for Software Localization.