Translate: Constant Output Language

[Catt0s]

Scenario: A webpage's main content is in a foreign language (say, Japanese), but the page's UI (headers, footers, content headers) are in a language that does not need translation (eg, English). The user will attempt to manually trigger a translation of the page using Firefox's built-in Translate tool (BETA). 

Current behavior: The translate tool reads the page's language, and always puts that as the source language. If the source language is a preferred language, the output language will be blank, and needs a dropdown item to be selected. 

Suggested behavior: The translate tool should ALWAYS put the user's preferred language in the output by default. If a user is attempting to translate a page, it is very likely they are trying to translate it INTO their preferred language, not out of it. The input language should be blank (ie, how the output currently behaves) - perhaps it should even attempt to detect what Unicode blocks are present on the page, and use it to guess the language the user wants to translate from. 

Why: Trying to translate a page that claims to be English (but isn't) into English requires the user to first change from English in the source language to the actual language (which they would need to determine and find in the dropdown), and then select English in the output. It adds lots of user friction that could be easily avoided.

[Jon]

Thanks for submitting an idea to the Mozilla Connect community! Your idea is now open to votes (aka kudos) and comments.

[marco]

We are planning some improvements for language identification, we'll take this into account.