Localization Testing

Ministry of Testing

Localization could mean different things to different people:

  • Translating text, including using characters not in the pre-localization version of the software

  • Text direction

  • Formatting dates, times, numbers, money amounts

  • Calendars

  • Time zones, leaving/entering daylight saving

  • Currencies - conversion between them, possibly having >1 per user or business entity e.g. a discount of $50 across a EUR account and a GBP account

  • Tax

  • Etc.

-- Anonymous

Suggestion 1: To add more context to my original question - was give a series of Chinese and Japanese text translation for a site and need to figure out how to ensure the text remains the same through the user experience on the site.

e.g. Chinese and Japanese. Things like Chinese need multi-byte characters e.g. UTF8. If the code was originally built on single-byte characters, some of the conversion to multi-byte might be missed. The places where this is most likely to show up is where text is split or joined together. E.g. given name + family name (stored separately in e.g. db.) => whole name.

Suggestion 2: Also, don't use long strings (say 50 characters) with no spaces. I've seen a lot of people do this and it's just not realistic (With one possible exception if you're from Anglesey where technically you could use 'Llanfairpwllgwyngyllgogerychwyrndrobwllllantysiliogogogoch', but everybody shortens it anyway, for obvious reasons...)

Reference:

Last updated