Sada Vehra: a framework for crowdsourcing Punjabi language content

Abstract

We present preliminary research leading to a prototype for Sada Vehra, an online resource for crowdsourcing Punjabi language content. We argue that languages less represented online benefit from crowdsourced volunteer-contributed translation, and present some challenges around effective crowdsourced non-expert translations. These include interface challenges around handoffs and ownership, language-specific challenges such as nuance and standardization, and infrastructure-specific challenges such as bandwidth and media quality.