Beyond Traditional Ethics when Developing Assistive Technology for and with Deaf People in Developing Regions - Ethical Engineering for International Development and Environmental Sustainability - page 303

perceived overlap between Voice over Internet Protocol (VoIP) research and the
Teldem, a locally produced text telephone, both of interest to the primary donor,
Telkom. We perceived the need for such technology and proceeded with a series of
projects that ultimately ended up unused. However, the requirements for the
SignSupport project, a mobile tool to assist communication between a Deaf person
and a medical doctor, were initiated from DCCT participants (Looijesteijn 2009 ).
This project was further refi ned by Chininthorn et al. ( 2012 ) to focus on a more
limited communication domain, with a pharmacist.
Because of the long-running nature of the wider project on AT with and for this
Deaf community since 2001, we are able to devise a number of future work ideas
each year and are able to take in new postgraduate students each year to continue
work on the project. However, even with the underlying requirements set by the
Deaf community, we still encounter the dilemma of whether each particular project
is acceptable by the community or not. For example, we know by interviewing both
Deaf and pharmacist participants after a mock trial that SignSupport may indeed
require a VRS when the pre-recorded dialogue on the phone is unclear or insuffi -
cient to convey critical communication. Thus we enlisted a PhD student to include
the relay work and mobile video work we had done with earlier projects. His par-
ticular spin, though, for a research topic was to secure the relayed communication.
So, the need for mobile relay has emerged from participants, and not the security
aspect of it. Is it right to pursue security as a priority when it was not identifi ed as
such by the Deaf community? We are working under the premise that by educating
Deaf people about Internet and mobile security, there might be a good offshoot of
the technical development. But then, the Deaf community had not asked for such
instruction either. We therefore view part of the research programme to empower
our Deaf partners to be able to participate more strongly, e.g. with English and
computer literacy training and accreditation. With regard to this particular decision,
we have decided to continue with the project because a video relay service was
prioritised by the community, and the security aspect provides both community
empowerment and Computer Science research merit, illustrating a parallel research
agenda that aims to satisfy and empower both major stakeholders.
Consider another example illustrating a similar decision. A postgraduate student
had modifi ed a pattern passcode for SignSupport as a fi nal year project. The pattern
passcode enhancement was identifi ed by observing that Deaf trainees for the ICDL
course routinely forgot text-based passwords. We thought that a more visual pass-
code, more similar to the visual nature of signed language, would be more appropri-
ate for them. However, the Deaf users had not explicitly asked for this, perhaps
because they did not possess enough computer literacy to warrant requesting such a
modifi cation. When the pattern passcode was shown to Deaf people, we received
encouraging feedback. This particular student wanted to continue with the project
for an MSc, yet we convinced him to switch to a project more prioritised by the
Deaf community: to enable SignSupport for other scenarios, as multiple scenarios
were identifi ed by Looijesteijn ( 2009 ). Then it becomes our challenge, as computer
scientists based at a tertiary institution, to devise a research topic out of needs priori-
tised by the Deaf community and to recognise and accommodate that their priorities