Proposal for an i18n/I10n Project

[sebrock]

User Story:

As a user living in a given country and using a browser set to the appropriate locale, I want to see UI Text in my language and see time and number formats according to the locale.
AND
be able to override the above settings by choosing a locale from a dropdown menu or profile setting.

This can be achieved by

• querying and evaluating the browser locale
• switch language and number formats to browser locale specific formats.
  There are internationalization (i18n) and localization (i10n) methods/functionalities available in most browsers. At least in Chromium-based and Mozilla browsers.

Prerequisites:

• Identify and list UI text items to be translated
• Utilize POEditor as a tool to crowdsource translation (I have experience in this and translators for ca. 20 languages at hand)
• Implement language files (preferably .json format (supported by POEditor, other formats work as well).
• Implement query and evaluation of browser locale
• Implement switching to locale specific version

Recommended approach:

Define a Project (in github) to track activities and use issues as activities for the back log.
I am volunteering to

• Plan and facilitate the project,
• Create all issues to cover activities
• Set up and maintain project in POEditor
• Assign translators
• Collect translations
• Provide .json files to the main project.
• Follow up with user feedback regarding translations and take corrective action as needed.

What do you think?

[gbl08ma]

I discussed this with the moderators and the majority agreed that it is not a priority. As much as I wouldn't mind i18n on a "normal" service/tool, the fact that JungleTV largely revolves around user-submitted content/having a community of sorts, means that localization has implications that go beyond the technical aspects. Due to moderation and community management concerns we'd like to keep the website mostly English-centric, and a localized interface would send mixed signals about that.

The way I see things, fully separate, unaffiliated JungleTV "clones" for each language/culture might be the best way to handle this. This way, each one can develop its own subculture, set of rules, staff team and moderation guidance that are the most appropriate for the communities that would spin up in each of these "clones" and the respective cultures.

On the technical front alone, i18n brings challenges that I am really not looking forward to dealing with, at least not as long as I am the one doing most of the development and frontend design work on my free time. I have experience developing an Android app with support for 5 languages, and waiting for translations, validating them and testing the UI for every language took a considerable amount of time and development effort - and this was an Android app designed to be localized pretty much from the first minute, which is not the case with JungleTV; adapting the current JungleTV frontend code to support i18n would take at minimum a couple dozen hours alone. And this is even before we get to the "boss levels" of i18n, like dealing with e.g. right-to-left languages.

I will keep this issue open to encourage further discussion on this topic, but I would not expect our stance on this to change any time soon, as long as the number of users on the website remains on the same order of magnitude.

[sebrock]

Fair enough, but yeah keep it open for discussion.