A Micro-Crowdsourcing Implementation: the Babel Software Project
|Title||A Micro-Crowdsourcing Implementation: the Babel Software Project|
|Publication Type||Journal Article|
|Year of Publication||2010|
|Authors||Exton, C, Spillane, B, Buckley, J|
|Journal||Localisation Focus - The International Journal of Localisation|
|Keywords||babel, digital divide, java, localisation, micro crowdsourcing, micro-crowdsourcing, open source, real time localization, server|
Today a lack of access to localisation resources and the cost of these resources contribute significantly to the digital divide. Therefore, finding ways to reduce cost and increase the rate at which software can be localised must be given high priority. While technologies such as Translation Memory and Machine Translation represent a major leap forward in improving this rate, they are, however, viewed mostly as enabling technologies and will not replace the human translator in most cases. This means that, the ability to acquire and manage an extremely large pool of human translators/post-editors is a major stumbling block in improving the rate at which software can be translated, as part of localisation. A Micro-Crowdsourcing architecture, proposed by Exton et al. (2009), could significantly reduce cost and increase capacity within the localisation space. For this reason the authors have created an exploratory implementation based on this architecture to investigate its feasibility.
This paper describes the exploratory implementation of the Micro-Crowdsourcing architecture called Babel Software. This contains two major parts, the development of the Babel Client Library and the Babel Server. The development process was demanding, yet found that the development of full Micro-Crowdsourcing should be feasible. It also found that, although it was possible to retro fit an existing application with the Babel architecture it is not advisable and so Babel should be incorporated into a client application's design during the initial development phase. An extremely important aspect of the implementation was based around the exploration of management tools on the Babel Server. These tools can be used by a project owner managing the localisation community. This helped provide great insight into the infrastructure which the Babel Server should provide in future implementations. Babel Software is an open source implementation which can be found on Souceforge.net. It is hoped that this exploratory implementation will spawn future releases; which will see Micro-Crowdsourcing complement other technologies in the localisation workflow. Together, these technologies can further reduce cost and increase the rate at which software is localised in an effort to reduce the digital divide.