Lessons learned about distributed pair programming: what are the knowledge needs to address?

We have realized an experiment aimed at understanding the impact of distribution on time and quality of code developed with pair programming. From this experimental data's analysis the pair dismissal phenomenon has stemmed. After the experiment, we have conducted an assessment discussion guided by a questionnaire. The assessment confirmed the dismissal hypothesis and highlighted knowledge needs to be addressed when managing distributed pair programming in software projects. This paper focuses on the results of the assessment and discusses the lessons learned.