You are viewing a read-only archive of the Blogs.Harvard network. Learn more.

The Longest Now

The Tower of Babel : normalizing language representation
Sunday August 23rd 2009, 3:27 pm
Filed under: chain-gang,international,popular demand,SJ,wikipedia

Part of a series on difficult topics from the Wikimedia community

There are some perennial projects that take more than a single barnraising to understand and plan for. One is the issue of supporting different languages equally — the world’s largest and smallest languages are both underrepresented among the projects.  While I would like to see Wikimedia become a model for the rest of the online world in this area, how a global community can provide support, bugfixes, and advice to different/new language groups is an issue for many multilingual projects.  So I offer these questions to all readers – feel free to answer them for the projects you are most familiar with.

  • What technical and other support do various language projects need to become awesome?
  • What variations are needed for projects whose main goal is language and cultural preservation?
  • What sharing of advice or practices would make starting new projects easier?
  • How can established projects help new projects with outreach, communication, and planning?

Let me offer one example of how this has been difficult to grasp within Wikimedia: discussions on the early international list were generally in English.  This led to a certain founder effect among participants, and in how the projects are today framed to the world, from elaborations of the vision to interface design.  And this has forked discussions of what language projects need – those in the language of the project, which can happen easily and fluidly among its participants and contributors, and those meta-discussions in one or two shared languages with the potential of setting Wikimedia-wide policy or affecting all projects.

As another example: non-Latin character sets, and cultural differences about editing and participation across different parts of the world, have always been part of discussions about how Wikipedia and its sister projects should advance.  Nevertheless, the early language communities drawn to the project were largely European, and issues that only affect non-Latin readers can still take a while to fix (for instance, replacements for Roman-alphabet captchas, or fixes to javascript and css layouts in corner cases).

What are your examples? What am I leaving out?  How can the global community and the Foundation better support small and underrepresented languages?  Feel free to leave links to current or historical discussions about problems and opportunities.

Thank you Samuel… this is the kind of document I am lookling for.

Comment by GerardM 08.24.09 @ 12:50 pm

Bad Behavior has blocked 233 access attempts in the last 7 days.