Exploring the social interactions in GitHub

1. Social Encounter and Interaction Ritual

While Goffman did not coin a clear definition of social encounter, the concept can be formed in his discussion of ‘focused interaction’, as a situation in which two or more indi- viduals come into contact and engage in face-to-face interaction with a shared timeline[1]. For example, people playing a board game, or committing a joint task in a face-to-face manner.[1]. He also argued that human interactions are akin to a theater performance, where individuals engage in specific behaviors and adhere to social norms. [2]

A specific type of social encounter is interaction ritual. According to first Goffman and later Collin [2, 3], interaction ritual is characterized by a heightened sense of collective involvement, shared emotional energy, and a temporary suspension of everyday routines. During an interaction ritual, individuals come together and engage in a coordinated set of behaviors, gestures, and verbal exchanges that follow established social norms and patterns. Generally, key features of interaction rituals include:

• Mutual focus: Participants direct their attention towards a common focal point, whether it is a specific task, event, or shared interest. This shared focus enhances the sense of connection and belonging among individuals. [4, 5, 6, 3]

• Emotional energy: Interaction rituals generate a heightened emotional intensity among participants. This emotional energy can manifest as excitement, enthusiasm, joy, or even tension, depending on the nature of the ritual and the emotions it evokes..[4, 5, 6, 3]

• Shared symbols and meanings: Interaction rituals often involve the use of symbols, gestures, and rituals that hold shared meanings within the particular social context. These symbols and meanings help create a sense of collective identity and reinforce social bonds among participants.[4, 5, 6, 3]

• Temporary suspension of reality: Interaction rituals provide a break from everyday routines and norms. They create a social space where individuals can temporar- ily step out of their usual roles and engage in a different set of behaviors and interactions.[7, 8, 3]

Goffman’s analysis on social interaction emphasizes the physical conditions as the source of behavioral materials. In his face-work discussion [9], he highlighted the body signs like gestures and verbal behaviours in social encounters. The ”face” is the image of the self that reflects social expectations and the ”line” is the acts (verbal and non-verbal) to establish and maintain the face [10, 9]. These two together forms the ”face-work”

- individual’s social actions to fulfill the social expectation reflected by the face, which

is the central principle organizing social interaction [9, 10]. In other words, the root of social interaction is the tacit agreement and cooperation to maintain face[11]. Rituals are one types of face-work containing interactive behaviors that are regular, polite, and enacted by participants in their social encounters[10, 9] Goffman portray the face-work as a stressful part of our social life. The face is in the end a ”loan” to the individual from the society and can be withdrawn anytime if fails the social expectation[9]. In short, the face is constantly under threat and obeying rules to save face is a routine in social encounters.

2. Time-space distanciation and mediated communi- cation in the information age

Traditionally, social encounters were tightly bound to specific physical locations and con- strained by the limitations of time and distance. For instance, the basic unit of analysis in Goffman’s interaction order is confined to physical situation like face-to-face interaction [12]. Nevertheless, with the advent of modern technologies, such as transportation and communication systems, social relations have become more detached from their immediate spatial and temporal contexts. Giddens raised the concept of ”Time-space distanciation”

[13] to reflect this increasing disembeddness of social interactions from their immediate physical context.

The time-space distanciation has several implications for social life, according to Gid- dens [13]. The ”disembedding mechanisms” enables social systems to transcend their local contexts and extend across time and space, allowing creation of global networks and sys- tems of social organization. There is also increased reflexivity of individuals and societies. Reflexivity refers to the capacity to reflect upon and modify social practices based on self- awareness and knowledge. As individuals become aware of the disembedding mechanisms and the ability to transcend time and space, they are compelled to reflect on their actions and make decisions based on this awareness.

The concepts of synchronous and asynchronous communication are not new. They have evolved over time alongside advancements in telecommunication technologies, com- puter networks, and the internet. The main difference is whether the communication is real-time (synchronous) or not (asynchronous). For instance, a face-to-face discussion is synchronous and writing letters is asynchronous [14]. This typology is also commonly used in distinguishing modes of digital communication (also called computer-mediated communication (CMC)) [15]. Some examples of synchronous CMC are e-chat and instant messaging, and asynchronous CMC are email and online discussion boards [15, 16].

Traditionally and commonly, like the one in Goffman’s interaction order regime, the physical context in synchronous mediated communication makes it more effective in col- laboration than asynchronous mediated communication [12]. Using above example of communication, there seems to have a ranking of effectiveness with the highest from synchronous communication (e.g. face-to-face), to synchronous CMC (e.g. instant mes- saging), to asynchronous CMC (e.g. email), and the lowest from other asynchronous like writing letters. However, occasionally in social interactions individuals may prefer the asynchronous than synchronous mode of communication. For example, in some work occasions one may ask the other to send an email rather than talk face-to-face in order to preserve some private time and space. After all, the face-work is social encounter is exhausting and this may be even so in synchronous encounter since an individual needs to react in a more instant manner.

Under the unavoidable trend of time-space distanciation, the information technology seems to be able to introduce some synchronousness in order to increase the effectiveness in the asynchronous mediated communication. While doing this, it still preserve some advantages in asynchronousness for some social encounters when a separation of timeline between individuals is preferred. In the following text, I will try to discuss this with the case of interactions in Github.

GitHub is a widely popular platform for collaboration in the field of software develop- ment. It is created to foster interaction among developers and teams, enabling them to work together effectively and efficiently. Within GitHub, various interaction behaviours and collaborative dynamics emerge.

2.1 Synchronous and asynchronous mediated interaction in Github

Github is a typical product of the time-space distanciation in the post-modern world. The society tends to produce in a global scale without physical and temporal constraints. People want to collaborate anywhere and anytime. What’s even more on the ’anytime’ context is a tendency to create parallel timelines for each individuals so that even when there is a dependency or conflict between each other’s tasks, they do not need to wait for each other in a queue to create. This accelerates the collaboration efficiency and matches the need of the fast-paced post-modern society.

Noting that Github is a complex digital infrastructure with diverse behaviour patterns, I would like to narrow down the discussion to the main interaction behaviour — collabo- ration. Briefly, the collaboration behavour has seven major related functionalities in the platform. Developers can interact through various channels, such as issues, pull requests, and comments. Issues allow users to report bugs, suggest enhancements, or discuss project-related topics.Comments, on the other hand, provide a means for discussions on specific lines of code, issues, or pull requests. These communication channels encour- age transparency and foster a sense of community among contributors. Branches allow developers to work on different features or bug fixes simultaneously without interfering with the main codebase. By creating branches, developers can experiment, iterate, and propose changes in an isolated environment. Forking, on the other hand, allows users to create a copy of a repository under their account, enabling them to make indepen- dent modifications and contribute changes back to the original repository through pull requests. Branches and forks facilitate collaboration by enabling parallel workstreams and promoting experimentation. Pull requests facilitate code review and discussion, enabling team members to propose changes, provide feedback, and discuss potential mod- ifications. Code reviews are a critical step in the collaboration to ensure the quality of work outcome. Developers can review each other’s code, provide feedback, suggest im- provements, and address potential issues. Code reviews encourage a culture of learning, collaboration, and continuous improvement among team members. The merge function is a fundamental feature that allows developers to combine changes from one branch into another. When working on a project, developers often create separate branches to work on specific features, bug fixes, or experiments. The merge function in GitHub enables the integration of these separate branches, bringing their changes into the main branch or another target branch. The merge function essentially combines the commits made in one branch with the target branch, integrating the changes into a cohesive codebase. This is typically done when the changes made in the branch are considered complete, tested, and ready to be incorporated into the main codebase.

The collaboration behaviour in Github is in the CMC context as it is based on the digital infrastructure of Github. Here I do not consider the interaction between individ- uals outside of Github. While the infrastructure does not focus on facilitate synchronous interaction with physical context like face-to-face talks (e.g. provide video call functions), the design seems to blur the line between synchronous and asynchronous mediated in- teraction. For instance, it provides flexibility to the individuals to decide whether they would like to act in real-time matter or not. In the mentioned behaviour like comments, individuals can interact in immediately like in instant messaging when they receive a com- ment, but they also have freedom to leave it and react later. This shares similarity with other digital communication tools like social media apps where people can communicate instantly or make a delay in response. In a broader sense, the infrastructure seems to fa- cilitate a flexibility for modes of communication in order to combine the advantages from both synchronous and asynchronous communication. It is noted that the interactions can be mostly asynchronous, yet the infrastructure brings some level of synchronousness. This is shown in three types of mechanisms that the previous mentioned collaboration-related functions can be briefly categorized into.

1. semi-synchronous mechanism Issue and comment is designed for open interaction behaviors. Individuals can choose to use it as instant messaging (synchronous- mediated) or as forum feeds with delayed reply (asynchronous -mediated).

2. disembedding mechanisms: It is known that communication can be burdensome and affects productivity negatively [17]. Because when one is communicating, you can not at the same time act to produce the detail task. Branching and forking, in the functionality level, allows individuals to break out from the social and concentrate on their own solo actions for efficiency. But the separation is done in a way that individual’s attention is separated but the output of the action can still be con- nected to the social. The funtionality creates parallel timelines that allow multiple individuals to work separately without affecting each other, and yet their actions can collaborate on the same content. For instance, A and B needs to work together in a bug fix task. A and B have different ideas and approaches. Instead of spending time communicating what to do and negotiating details, they can simply create their own branches of the original work, fork what is already in the work, and start work according to their own idea. The parallel timelines increase the possibilities of finding solutions.

3. embedding mechanisms The disembedding mechanisms is paired with embedding mechanisms which reconnect the disembedded ones to the social. For example, the functions of pull request, code review and merge is to map the changes made by the individuals in the disembedded phase and match to the social expectation in the collective level. As an example, A and B finished their own version of bug fix task, and they initiated a pull request. The admin of the project received the request and start reviewing the code and outcome, and select the desired ones to merge to the main project.

2.2 The collaboration in Github as interaction ritual?

While the interaction in Github is without physical context, it seems still present some level of the qualities in the concept of interaction ritual of Goffman, i.e. mutual focus, emotional energy, shared symbols and meanings and temporary suspension of reality

[4, 5, 6, 3, 7, 8]. When individuals conduct actions in Github (e.g. conduct a code review, make a comment), each of them bring they focus in the target tasks which belongs to the target software development projects. This can be seen as shared focus where individuals can feel connected in this digital community with his fellow collaborators. As the actions make contributions to the project outcome, this could generate intense emotions for those who participated. For example in a special type of software project called Open Source Software Development, all the contributors are volunteers and under this nature, it is the emotions like enthusiasm for the hobby or sense of belonging that keep them motivated. The Github platform is composed by a series of shared symbols and meanings that align the users’ behaviours and bring them to a consensus. The buttons in the user interface use terms like fork, branch, review, comments with pre-defined meanings that simulate the symbols and gestures in real life. For example, ’fork’ in real life action is a tool to obtain food from a source location (the common plate) to a destination location (your private plate). It can be understood as an action of transforming something from public to private status. This meaning is used in Github as well — bring a piece of public information in other’s project to your own space. These shared symbols and meanings in Github represent a series of actions the users can take and communicate with others under the ”social rules”. The interaction in Github creates a social sphere for individuals bring in their temporary attention and behaviours that are different from their other roles in other sphere. At the moment one is doing a code review in Github, they are temporarily only a contributor in this task, and his tasks in other realities (e.g. to make a call with his boss) have to be suspended temporarily.

However, compared to the traditional interaction ritual, the one with Github has some levels of deviance. The mutual focus in the interaction in Github may present a shorter duration within each focus episode. There may be a higher frequency of focus episodes in a certain period of time, because one do not need to travel to the agreed avenue to meet or negotiate a time to have a chat in video call. The emotional energe may be weaker in the Github interaction as the emotions can be kept private and is not communicated directly and instantly with each other as face-to-face interaction. As for shared symbols and meanings, the UI features (buttons) in the interface is not a 100 percent copy of the reality, but just a simulation to reflect some degree of meaning. The temporary suspension of reality in Github interaction, compared to a physical encounter, may generally last for a shorter time, and is not strict sense of suspension, for example, when one is focus on commenting in Github project, he is not physically separated from his reality, i.e. he can listen to sometime not related at the same time.

2.3 Is Github interaction faceless?

Goffman has pointed out that the face-work is an unavoidable and frustrating process in social encounters. His statement was based on the unit of analysis as physical social interactions. In the context of interaction in virtual platforms, some indicated that things may be more liberating as one is not left directly for a rebuff by others in social encounters [18]. In terms of interaction in Github, I argued that it can be more subjected to face- work than general social media platforms for entertainments. It is a platform where professionals create and produce an outcome for projects. Hence there is not only a social aspects (e.g. profile information) but also technical aspects (e.g. coding capacity) when people interact and form impressions on each other [19, 20]. The ’face’ is more clear and rigid in this professional platform than other non-professional platforms. It is not just an avatar and some random content that an individual share about their life, but it requires

one to have the knowledge of the specific field and display related expertise in their public image. Otherwise, the face is vulnerable and can be easily rebuffed by actions like a code review. In order to engage with the social there, one needs to act according to the social rules (the ’line’) in the field in order to maintain this public face. For example, actively contributing to reviewing codes, or producing a successful bug-fix tasks and being accepted by a reviewer. The interaction in Github, while less frustrating than the one in a face-to-face work setting, is not a liberated, faceless-work if one wants to be part of the professional world.

References

[1] Erving Goffman. Encounters: Two studies in the sociology of interaction. Ravenio Books, 1961.

[2] Erving Goffman. Interaction ritual: Essays in face to face behavior. AldineTransac- tion, 2005.

[3] Randall Collins. Interaction ritual chains. Princeton university press, 2004.

[4] Marcus Holmes and Nicholas J Wheeler. Social bonding in diplomacy. International Theory, 12(1):133–161, 2020.

[5] Marci D Cottingham. Interaction ritual theory and sports fans: Emotion, symbols, and solidarity. Sociology of Sport Journal, 29(2):168–185, 2012.

[6] Monika Wilin´ska and Pia H Bu˝low. “we are on air now”: the emotionality of video-recording in the institutional setting. International Journal of Social Research Methodology, 20(4):343–355, 2017.

[7] Ilja Simons. The role of event in hybrid communities: An interaction ritual approach. 2022.

[8] Ben Rampton. Interaction ritual and not just artful performance in crossing and stylization. Language in Society, 38(2):149–176, 2009.

[9] Erving Goffman. On face-work: An analysis of ritual elements in social interaction.

Psychiatry, 18(3):213–231, 1955.

[10] Marie L Radford, Gary P Radford, Lynn Silipigni Connaway, and Jocelyn A DeAn- gelis. On virtual face-work: An ethnography of communication approach to a live chat reference interaction. The Library Quarterly, 81(4):431–453, 2011.

[11] Anders Vassenden and Merete Jonvik. Cultural capital as a hidden asset: Culture, egalitarianism and inter-class social encounters in stavanger, norway. Cultural Soci- ology, 13(1):37–56, 2019.

[12] Ruth Rettie. Mobile phone communication: Extending goffman to mediated inter- action. Sociology, 43(3):421–438, 2009.

[13] Anthony Giddens. The constitution of society: Outline of the theory of structuration, volume 349. Univ of California Press, 1986.

[14] Alexander Romiszowski and Robin Mason. Computer-mediated communication. In Handbook of research on educational communications and technology, pages 402–436. Routledge, 2013.

[15] Jannis Androutsopoulos. Multilingualism, diaspora, and the internet: Codes and identities on german-based diaspora websites 1. Journal of Sociolinguistics, 10(4):520–547, 2006.

[16] Roser Beneito-Montagut. Encounters on the social web: Everyday life and emotions online. Sociological Perspectives, 58(4):537–553, 2015.

[17] Ingo Scholtes, Pavlin Mavrodiev, and Frank Schweitzer. From aristotle to ringelmann: a large-scale analysis of team productivity and coordination in open source software projects. Empirical Software Engineering, 21(2):642–683, 2016.

[18] Hugh Miller. The presentation of self in electronic life: Goffman on the internet. In

Embodied knowledge and virtual space conference, volume 9, pages 1–8, 1995.

[19] Denae Ford, Mahnaz Behroozi, Alexander Serebrenik, and Chris Parnin. Beyond the code itself: how programmers really look at pull requests. In 2019 IEEE/ACM 41st International Conference on Software Engineering: Software Engineering in Society (ICSE-SEIS), pages 51–60. IEEE, 2019.

[20] Jennifer Marlow, Laura Dabbish, and Jim Herbsleb. Impression formation in online peer production: activity traces and personal profiles in github. In Proceedings of the 2013 conference on Computer supported cooperative work, pages 117–128, 2013.