The scope and scale of translation work has always seemed daunting to me.
I can’t imagine a way to make it any more efficient, without sacrificing significant accuracy by missing context, than to do manual translation of everything.
I can’t imagine if, every time I added a button to a UI, an entire team of localization translators mobilizing to make sure the right context came through in every language the UI supported. Not to mention the tools that support that workflow of passing context to translators, compiling the translations, and binding them all into a data structure I can use to populate my UI.
And that’s just for language. Iconography and colors have their own localizations.
Screenshots are all you ever need. Just give a screenshot or terminal output to a bilingual SME and have it double checked. It’s not like translators must be given sweat and blood of engineers just for their motivations.
A screenshot of just the dropdown menu could have been generated in CI and could have easily prevented the “postcode file”, so long the translator was presented that image along the string and had recognized that Windows dropdown.
I can’t imagine a way to make it any more efficient, without sacrificing significant accuracy by missing context, than to do manual translation of everything.
I can’t imagine if, every time I added a button to a UI, an entire team of localization translators mobilizing to make sure the right context came through in every language the UI supported. Not to mention the tools that support that workflow of passing context to translators, compiling the translations, and binding them all into a data structure I can use to populate my UI.
And that’s just for language. Iconography and colors have their own localizations.