WeRelate:Suggestions/Content Language Neutrality via Wikipedia/Wikidata

Note: The following appeared originally as an idea on the Watercooler. I would like to think it makes sense to at least preserve the idea and discussion as a suggestion. --jrm03063 18:27, 6 January 2014 (UTC)

Consider the Place page for the US State of Alabama. When opened for editing, the initial content appears as:

== History ==

The extract from English Wikipedia is presently obtained via the template {{wp-Alabama}}. That template begins thus:

<!-- This text is copied from wikipedia. Any changes made will be overwritten during the next update. -->
'''Alabama''' is a state located in the [[Wikipedia:Southern United States|southeastern region]] of the [[Place:United States|United States]].

My proposal, is that we find a way to make the existing content templates smarter, such that they are apt to present content in the preferred language of the user. For example, if we had the Multilang extension (which claims to work with 1.7 through 1.16 - we're at 1.7)? Perhaps, {{wp-Alabama}} would contain something like the following:


Presumably, we could re-generate the existing template pages, and the language-specific extracts, entirely via software. That software would start by examining any given English language WP page at refresh time. If the English WP page is found to be the subject of a Wikidata object, then we may be able to find different language versions of the page (If not, the extract would be regenerated exactly as it is at present). Knowing what languages support a version of the page, and what the page is named in that language's WP, we can determine the options to provide in the "multilang" section, as well as what language-specific content templates ({{en-Alabama}}, {{de-Alabama}}, etc.) to generate.

Just to be clear - this is not a discussion of Internationalization and localization, which is already richly supported by mediawiki.