People who are deaf or hearing impaired (DHI) often struggle with low-paying jobs. Access to education can change their perspectives. Many jobs in the information technology industry are left unfilled due to the lack of skillful candidates. Our lab offers distance learning Java programming courses to the DHI. We wondered how good our java DHI graduates are in finding and fixing program errors and changing program logic. We asked 5 DHI and 5 non-DHI programmers with the same amount of experience and training to perform debugging related tasks. We performed task and situated analysis and found that DHI programmer performance inferior to those without the disability. We argue that a debugging tool based on a mode more kin to the DHI should mitigate the disparity.Resumo. Pessoas surdas ou com deficiência auditiva (SDA), muitas vezes possuem empregos de baixa remuneração. O acessoà educação pode mudar suas perspectivas. Muitos postos de trabalho naárea de tecnologia da informação são deixados ociosos por falta de candidatos hábeis. Nosso laboratório oferece cursosà distância de programação Java para pessoas SDA. Nos perguntamos o quão bons nossos alunos SDA são em encontrar e corrigir erros em um programa. Cinco programadores SDA e cinco programadores ouvintes com a mesma experiência e treinamento executaram atividades de depuração de código. Realizamos uma análise situada e descobrimos que o desempenho dos programadores SDA era menor em relaçãoàqueles sem deficiência. Argumentamos que uma ferramenta de depuração voltada para o público SDA deve atenuar essa disparidade.

No abstract

No abstract

Providing timely feedback on identifier naming to novice programmers can help them to improve their program readability. However, due to the growth in the number of students learning to program nowadays, giving manual feedback on identifier quality become prohibitive. In this paper, we propose a method to automatically give this feedback which is correct 75.0% of the time in contrast to the instructors' assessment. We found that 51.7% of the students who received automated feedback showed their program identifier quality improvement by picking better names. It means that we can help students to improve identifier naming and consequently, their program readability from early coding experiences.

No abstract