Empirical evaluation of collaborative support for distributed pair programming

Pair programming is an Extreme Programming (XP) practice where two programmers work on a single computer to produce an artifact. Empirical evaluations have provided evidence that this technique results in higher quality code in half the time it would take an individual programmer. Distributed pair programming could facilitate opportunistic pair programming sessions with colleagues working in remote sites. In this paper we present the preliminary results of the empirical evaluation of the COPPER collaborative editor, developed explicitly to support pair programming. The evaluation was performed on three different conditions: pairs working collocated on a single computer; distributed pairs working in application sharing mode; and distributed pairs using collaboration aware facilities. In all three cases the subjects used the COPPER collaborative editor. The results support our hypothesis that distributed pairs could find the same amount of errors as their collocated counterparts. However, no evidence was found that the pairs that used collaborative awareness services had better code comprehension, as we had also hypothesized.