GitHub and Stack Overflow: Analyzing Developer Interests Across Multiple Social Collaborative Platforms

Abstract

Increasingly, software developers are using a wide array of social collaborative platforms for software development and learning. In this work, we examined the similarities in developer’s interests within and across GitHub and Stack Overflow. Our study finds that developers share common interests in GitHub and Stack Overflow; on average, 39% of the GitHub repositories and Stack Overflow questions that a developer had participated fall in the common interests. Also, developers do share similar interests with other developers who co-participated activities in the two platforms. In particular, developers who co-commit and co-pull-request same GitHub repositories and co-answer same Stack Overflow questions, share more common interests compare to other developers who co-participate in other platform activities.

Appendix 1: Example for Cross-Platform Similarity Score Calculation

Figure 4 shows an example for the calculation of cross-platform similarity score $Sim^{SO\text {-}GH}(d)$. Consider developer d who has participated activities in GitHub and Stack Overflow. d has forked 2 repositories; Repository A which description contains the tag set {Java, Android}, and Repository B which description contains the tag set {Java}, and watched Repository C which description contains the tag set {C#}. d also favorited 2 Stack Overflow questions; Question D which are tagged with {Android}, and Question F which are tagged with {iOS}, and answered Question E which are tagged with {Java}. We can estimate d’s interests in GitHub (i.e. $I^{GH}(d)$) as {Java, Android, C#} and d’s interests in Stack Overflow (i.e., $I^{SO}(d)$) as {Android, iOS}. The common interests of d (i.e., CI(d)) would be {Java, Android}. Therefore, $Shared^{R}(d)$ would include repositories A and B, while $Shared^{Q}(d)$ would include questions D and E. Thus, $Sim^{SO\text {-}GH}(d) = \frac{|2| + |2)|}{|3| + |3|}$.

Fig. 4.

Example of cross-platform similarity score calculation

1.1 Appendix 2: Example for Co-Participation Similarity Score Calculation

Figure 5 shows an example for the calculation of co-participation similarity score for watch activity $Sim^W(d,co^W(d))$ for developer d. Let us consider two developers d and $d'$ and assume that there are no other developers. Developer d watched repositories A and B. Developer $d'$ co-watched B with d. Thus, $co^W(d)$ is {d’}. In addition to B, developer $d'$ also watched repositories C and D. $Shared^W(d,d')$ would then include B and C as both of the repositories share common interests with the repositories that d watched. $Sim^W(d) = \bigg [\sum _{d'\in Co^W(d)} \frac{|2|}{|3|}\bigg ] / |1| = 0.67$.

Fig. 5.

Example of co-participation similarity score calculation for watch activity

It is important to note that the co-participation similarity scores only consider the similarity in interests between pairs of developers who have co-participated in at least one common repository or question with each other but the developers may have participated in many other repositories and questions different from each other. For example, developers d and $d'$ only watched one common repository but they had watched many other repositories which were different from each other. Also, when computing the co-participation similarity measure between developers who participated in a particular activity, we only consider the interests of the developers in that target activity. For instance, when computing $Sim^W(d)$, we consider how similar are the interests between developers based only on the watch activities, i.e., we do not consider repositories forked by the developers or questions answered and favorited by the developers.