Mixed Language,' a characteristic pattern of language use among African township residents in South Africa, may well include words or full constituents from several languages. However, from both a structural and a social perspective, such speech has a systematic nature. In reference to grammatical structure, within any CP (projection of COMP) showing codeswitching, only one language (the Matrix Language) provides the grammatical frame in the data studied. Also, while speakers from different educational levels engage in codeswitching with similar frequencies, the types of codeswitched constitutents they prefer are different. In reference to the social use of language, we argue that specific patterns of codeswitching indicate how language is both an index of identity and a tool of communication in South Africa. In the codeswitching patterns they use, speakers exhibit strong loyalty to their own first languages. Yet, because they recognize that codeswitching facilitates communication with members of other ethnic groups, they use a number of codeswitching strategies as a means of accommodating to their addressees and simultaneously as a means of projecting multiple identities for themselves.
Research using voice-based services as a technology platform for providing information access and services within developing world regions has shown much promise. The results for design and deployment of such voice-based services have varied depending on the application domain, user community and context. In this paper we describe our work on developing a voice-based service for obtaining feedback from school children, a previously unexplored user community. Through a user study, focus group discussions and observations of learners' interaction with multiple design prototype versions, we investigated several factors around input modality preference, language preference, performance and overall user experience. Whilst no significant differences were observed for performance across the prototypes, there were strong preferences for speech (input modality) and English (language). Focus group discussions revealed rich information on learner's perceptions around trust, confidentiality and general system usage. We highlight several design changes made and provide further recommendations on designing for this user community.
In this paper we introduce a solution for disease surveillance and monitoring in the primary animal health care (PAHC) domain that uses inbound voice-based services and voice-and text-based outbound services for connecting rural veterinarians and livestock owners with a PAHC service provider. We describe our findings from the ongoing pilots, where we found that it is crucial to close the loop between data collection and information dissemination. /01…$15.00 health tips, and newsletters, sent out based on user communication preferences gathered at registration. The goal is to afford Afrivet a one-stop place from where the information that is being received from different users can be disseminated to the multiple PAHC role-players. In the design we focus on capabilities to customize (e.g. send via SMS only to cattle owners), localize (disease outbreaks info sent only to affected provinces), and pre-schedule (set-up reminders, seasonal tips beforehand) such messages.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.