- bug tracking
- We should be able to file and track bugs. We'll start with bugs in the Mozilla Localizations product, Component Other. Make sure to CC l10n@moz on those. We'll create a structure for those bugs once we know what needs to get tracked.
- If you know someone who should be participating in this work as an editor, please convince her or him to apply.
- Tools used
- One part of the specification should be a grammar specification for the fileformat, in a somewhat understandable syntax. Currently, I'm using antlr to create parsers from a EBNF-like grammar. It promised to create code for more than just java, but currently, I'm hitting a few blockers. It might be that I'm hitting bugs in the 3.0b5, or just limitations of the approach in antlr. Anyway, there's currently no descent output other than java, and creating ASTs doesn't work that well without putting AST-related constructs into the grammar. Which is merely noise. Suggestions welcome, in particular for tools that could even generate js.
- Platform dependent strings
- Can l20n help with doing platform-specific strings (think Options vs Preferences) for Mozilla apps?
Linguistic problems common to existing l10n implementations
- Verb usage in yes/no dialogs (Irish/Welsh)
- Can we provide context specific button values for stuff like yes and no? See proposal for that on gnome-i18n. Discussion has been moved to translation-i18n and also brought in next issue.
- Age of a person (Vietnamese)
- This language is lacking completely neutral pronouns, correct usage cannot be derived without age/gender context. See also Tips for your Vietnamese Translation Project
- Honorific speech (Japanese)
- This language uses different verbs to address a person, depending on its relation to speaker. Read more and more. Real-life example from NASA Bugzilla could not be easily (politely) translatable without some knowledge: who is a boss, who is subordinate team member, and who is a valued but external customer...
- Numeric inflection
- silly, widely known and well solved since GNU gettext.
- Vowel harmony
- A feature of many languages
Also we don't have *у* after vowels. It is replaced with *ў*. The sentence "Калі вы знайшлі вашу [% $terms.bug %] ў [% terms.Bugzilla %]" is correct if $terms.bug ends with vowel. It contains a error if $terms.bug ends with consonant.