Provide an API that allows to push and pull resource files.

I didn't find an option allowing translators to modify master language strings together with target language strings. It could be often useful to paraphrase original strings to make them sound better on both languages.

Developers often insert comments before groups of resource strings, but these comments are not displayed by Amanuens.

It would be good to use these comments.

If a resource files strings were displayed unsorted (see [1]) it would also be possible to display comments from the resource file before each group of strings.

If strings are sorted (as now) it may be possible to calculate which strings a comment is associated with and display it by each associated string.

[1] http://threeplicate.uservoice.com/forums/44829-amanuens/suggestions/1761551-provide-option-to-prevent-sorting-of-resource-file

Since the change in user creation interface, we have to wait for user login confirmation before assigning the translation.

Provide better support for GitHub. The current git support doesn't seem to work with the SSH URLs that GitHub uses. The setup process is also much more complicated than it needs to be.

It should be possible to lock resource entries so they cannot be translated. This should be possible either explicitly (on a specific key) or with a regular expression on the key name.

Provide a HTML widget that allows to translate resource for open-source apps directly from the project's website (a sort of crowdsourced translation).

Are there any plans to support .rc files (created by pre-.NET MS Visual Studio projects)?

Strings in Android strings.xml files are not always represented by a <string> tag e.g.
<string name="developer">Martin</string>

they are often represented by a <plurals> tag which contains <item> tags e.g.
<plurals name="person">
<item quantity="one">%d person</item>
<item quantity="other">%d people</item>
</plurals>

Amanuens caters only for <string> tags and ignores <plurals><item> tags.

Create a report that compares the language specific resources that have been marked as translated against the neutral resources and raise warnings for all strings that are still the same as the neutral language.

Add a simple way to add reviewers and approve/reject translations as well as make minor changes.

If a translator does not translate a string, the string is saved to the translated file and overrides any changes to the original string in the untranslated file.

Also some translated files are only partially translated because of variation in translation difficulty and importance, but English text is saved in the translated file, which is wrong and confusing. We deliberately don't commit partially translated files for a library we use for this reason.

Also, we have many translators who come and go and often untranslated strings are stuck on old versions for a considerable time.

Our git repository has several branches but Amanuens can currently only use "master". This should be selectable.

There is no support for Xliff :-(

I often forget which translator knows which language, so some way of filtering the translators based on the selected language or showing the translator's languages in the translator list would make it easier to pick the right translator for a translation.

We pre-process all our resource files to identify duplicates that are shared across files. We then extract these duplicates into a shared file and mark them as shared in the original file. The translator therefore only needs to translate the shared file. It would be great if Amanuens could check for a tag within a resource text whereby it will automatically mark the resource as already translated (we use [## SHARED ##]. That way we don't have to manually go through all our neutral resource files to mark these shared resources as already being translated.

I've uploaded some Java .properties files (ResourceBundles) and was planning on using your tool, but during evaluation I noticed you don't support UTF-8 encoding. I know that is not the standard Java ResourceBundle encoding, but that is what we are using. Can you add support for different encoding types?

When new files are added from a sync, automatically add them to every language