Not-for-Profit Crowdsource-based Translation Technology

Traditional forms of language translation based on a professional pool of translators cannot cope with the future needs and demands of humanity. This is due to many factors, not least of which is the ever-growing user generated content and accessibility of the World Wide Web. Crowdsourcing based on a non-profit translation community has been described as one approach to address the shortcomings of current mainstream translation efforts, allowing the decisions about what should be translated to be shifted from large corporations to the individual or the crowd. To empower language communities, at the Localisation Research Centre (LRC) we have developed a web-based translation platform known as Service-Oriented Localisation Architecture Solution (SOLAS) for volunteer communities. SOLAS was conceived at The Rosetta Foundation Design Fest in San Francisco, February 5-6, 2011. The SOLAS consists of two components: SOLAS Match (matching projects and volunteers) and SOLAS Productivity (a suite of translation productivity tools and language resources). SOLAS is an integral part of a wider generic common online space, owned by everybody, driven by non-market considerations, and managed and protected for the benefit of everybody known as Translation Commons (Trommons).

It is our view that this new model of crowdsourcing may provide equal access to information and knowledge regardless of language. In this paper we describe the current system’s design from the user perspective, particularly focusing on how we have used commonly used concepts in crowdsourcing platforms such as badges, tags and rating systems to deal with more high-level concerns such as motivation, quality, domain classification and related translator knowledge and reputation. In addition, we provide some up-to-date data relating to the current system’s growth and utilisation.