Social Structure on YouTube: an Online Community or a Celebrity-growing Platform? A Digital Methods Approach to Studying the Use of Social Affordances on YouTube

(1)

Social Structure on YouTube: an Online

Community or a Celebrity-growing Platform?

A Digital Methods Approach to Studying the Use of Social Affordances on

YouTube

Britt Boss

Submitted in partial fulfilment of the requirements for the degree of Master of Arts in the Program New Media and Digital Culture at the University of Amsterdam.

22th of August 2016

(2)

ABSTRACT

The purpose of this research is to analyse how users on YouTube engage with the platform’s social affordances and the meaning hereof for the social structure of the platform. YouTube has always proclaimed to be a community. While during the first years of the platform’s existence there was a lot of emphasis on this community character articulated through community features such as ‘My Friends’, ‘Groups’ and

‘Community’, the platform as it is today does not entail these features anymore. Despite throwing out these community features, YouTube still describes itself as a community. This invites for a clarification of the social structure of the platform. This research will be conducted drawing on concepts of social structure, online communities and platform affordances. I will analyse the three most important ways for users to connect on the platform as it is today: featured channels, subscriptions and comments. For this I will be using a digital methods approach to turn data into findings. Some of the most important findings were that different user types on YouTube have different levels of engagement and therefore make different use of the social features of the platform. Some highly engaged ‘YouTube celebrities’ seem to mainly articulate their community of friends through these features, while slightly smaller channels seem to make use the social features in order to promote their own channel. Subsequently, YouTube celebrities turned out to have little interaction with their viewers in the comment section, while smaller content creating channels did make use of the comment section to maintain their loyal viewers. The social structure of the platform does show some community aspects in a social way, while at the same time promotional practices are at stake.

KEYWORDS

(3)

LIST OF FIGURES

Figure 2: YouTube’s About-page as seen through the Wayback Machine. ... 23

Figure 3: Implementation of the 'My Friends' section as seen through the Wayback Machine. ... 23

Figure 4: Implementation of 'Groups' feature as seen through the Wayback Machine. ... 24

Figure 5: Explanation of the 'Groups' feature as seen through the Wayback machine February 2006. ... 24

Figure 6: Introduction of a 'Community' tab as seen through the Wayback Machine. ... 26

Figure 7: Percentage of Travel Channel Subscriptions by Format according to Google (Crowel et al. 5). .. 31

Figure 8: Content that travellers are most likely to engage with according to Google (Crowel et al. 5). ... 32

Figure 9: YTDT Channel Network tool developed by Bernhard Rieder. ... 34

Figure 10: YTDT Channel Network tool - ranking options when using a search query as seed ... 34

Figure 11: YTDT Video Infro and Comments tool designed by Bernhard Rieder. ... 35

Figure 12: Self-curated list of 30 active channels that are all associated with travel vlogging. ... 39

Figure 13: Overview of the four differences for the networks I visualised. ... 40

Figure 14: Settings Channel Network tool ... 40

Figure 15: Featured channels network of a self-curated dataset of 30 travel vlogger channels on YouTube. Crawl depth 0. Node size based on In-Degree. ... 41

Figure 16: Collaborative videos by multiple travel vloggers within the dataset. ... 44

Figure 17: FunForLouis with his girlfriend RayaWasHere. ... 45

Figure 19: Featured channels ánd subscriptions network of a self-curated list of 30 travel vlogger channels on YouTube. Crawl depth 0. Node size based on In-Degree. ... 46

Figure 21: Featured channels only network of a self-curated list of 30 travel vlogger channels on YouTube. Crawl depth 1. Node size based on In-Degree. The blue nodes are the 30 initial travel vlogger channels. The pink nodes are the channels retrieved in a crawl depth of 1. ... 48

Figure 23: Featured channels and subscriptions network of a self-curated list of 30 travel vlogger channels on YouTube. Crawl depth 1. Node size based on In-Degree. The blue nodes are the 30 initial travel vlogger channels. The pink nodes are the channels retrieved in a crawl depth of 1. ... 52

Figure 24: Zoomed-in image of the large cluster on the top right side of the larger network that can be seen in Figure 23. ... 54

Figure 25: Zoomed-in image of the large cluster on the bottom right side of the larger network that can be seen in Figure 23. ... 55

Figure 26: Zoomed-in image of the large cluster on the bottom right side of the larger network that can be seen in Figure 23. ... 56

Figure 27: A numerical overview of the comment section for two of the most popular travel videos by Jacks Gap. Consulted on the 23th of May 2016. ... 58

(6)

Figure 28: A numerical overview of the comment section for two of the most popular travel videos by Fun For Louis. Consulted on the 23th_{of May 2016. ... 58}

Figure 29: A numerical overview of the comment section for two of the most popular travel videos by Hey Nadine. Consulted on the 23th_{of May 2016. ... 58}

Figure 30: A numerical overview of the comment section for two of the most popular travel videos by vagabrothers. Consulted on the 23th of May 2016. ... 59

(7)

1. INTRODUCTION: HOLDING ON TO THE

COMMUNITY CHARACTER

VidCon, Playlist Live, Summer in the City, Amity Fest and Buffer Festival; just a small selection of YouTube conventions that take place all over the world on a regular basis. Thousands and thousands of YouTube-enthusiasts come together here to attend talks, meetings or just to ‘hang out’ with like-minded people. These events are mainly focused around so-called YouTube celebrities (De Lange, Commenting on Comments n. pag.); YouTube creators that have become celebrities in their own right, receiving millions of views and masses of fans who follow them wherever they go (Stone n. pag.). The distinction between celebrity YouTubers and ‘traditional’ celebrities is becoming increasingly blurred. Not only because YouTube stars are more popular than traditional celebrities among American teenagers (Ault n. pag.), but also because YouTube celebrities are no longer flying below the radar of the mainstream media (Stone n. pag.). YouTube celebrities are said to be genuinely influential authorities (Dredge n. pag.) and are hence “changing the face of youth culture” and altering the future of marketing and advertising (Chawla n. pag.). This image established by the mainstream media mainly portrays YouTube as a star-growing business that allows for a new type of celebrity to emerge and grow their fan base.

Contrary to the image picked up by the mainstream media is the platform description that YouTube provides on its About-page: “remember that this is your community. Each and every person on YouTube makes the site what it is, so don't be afraid to dig in and get involved. […] We hope you'll find something new to love as you get to know the community” (Community Guidelines n. pag.). This

characterisation of YouTube as a community implies a different social structure than the characterisation of YouTube being a platform for YouTube celebrities. Whereas a celebrity platform denotes a hierarchical structure in which some users would be influential and others susceptible (Van Dijck, Culture of Connectivity 129), a community implies a heterarchical structure (Bruns & Schmidt 5) that invites for all users to actively engage.

From its emergence in 2005, YouTube has always proclaimed to be a community. During the first years of the platform’s existence, a lot of community

(8)

features were articulated on the platform, which emphasised YouTube’s aim to be a community platform. However, YouTube has been through enormous changes since its emergence more than ten years ago. The acquisition by Google was an important starting point for the platform to grow from a project by three like-minded individuals into a million dollar business. This evolution of course also carried its consequences for the social structure of the platform. The YouTube of today does not have

community features like friends, groups or a community section anymore. Despite of this, YouTube is still firmly holding on to this community-oriented description of the platform. This thus invites for a clarification of the social structure of YouTube. It is important to know how users are forming social structures on an understudied

medium like YouTube (Rieder, Introducing the YouTube Data Tools n. pag.) because only when fully understood, YouTube’s social and cultural implications can be detected and discussed. YouTube is increasingly more being acknowledged in the media field, as also showed from the above-discussed mainstream media articles. With more than a billion users (YouTube, Statistics n. pag.) and a second rank position1 on the global top viewed and visited websites (Alexa n. pag.), YouTube has firmly established its significant presence within our contemporary digital media landscape. A clear vision of how users are actually connecting on the platform, what these relations mean and what kind of social structure this generates has however not yet been established. This is what I would like to change by performing this research. In order to do this, I defined the following research question:

How do users on YouTube engage with the platform’s social affordances and what does this mean for the social structure of the platform?

In order to conduct this research, I made use of tools that were developed by media scholar Bernhard Rieder for the Digital Methods Initiative. I will be deploying a digital methods approach to turn data about connections between users on YouTube into findings about the social structure of the platform. Before I will get to the actual analysis, I will first, in chapter 2, define and establish a clear understanding of the concepts of social structure and (online) community for which I will borrow from

1

(9)

social sciences. These two concepts will be used throughout this research to refer to the organisation of relations between users of YouTube that results from their engagement with the most important features for interaction on the platform. In this chapter I will also discuss the concept of platform affordances as discussed by media scholars Taina Bucher and Anne Helmond. Platform affordances, or the symbols and implications that platform features carry as well as the type of user activity they enact and encourage, will be a conceptual basis for the way I will be discussing the social features on YouTube.

Hereafter, in chapter 3, I will properly introduce YouTube as being the object of this research. I will situate the emergence of the platform within the Web 2.0 environment, which has been very telling for YouTube’s social character (Burgess & Green 93). Besides this, I will discuss how different user types can be distinguished on the platform. These user types show different levels of engagement with the platform and with each other, which makes some more prominent than others when it comes to the formation of a social structure on YouTube. Also, I will discuss how vlogging went from an amateur practice to a profession. This shows the changing character of the social structure, which is in line with the professionalization of the entire platform, mainly induced by the Google take-over.

What will follow in chapter 4 is a discussion of the most important community features that YouTube entailed during the earlier years of the platform. Features like ‘My Friends’, ‘Groups’ and ‘Community’ showed how YouTube was trying to establish itself as a community. These specific community features are not anymore incorporated in YouTube as it is today but they do provide a valuable insight into how YouTube would (have) like(d) for its users to connect on the platform. The three most important social features through which users can connect on the current version of the platform are: featured channels, subscriptions and comments. I will discuss the social affordances of these features as well as what they implicate about the social structure of the platform in chapter 5.

Chapter 6 contains the methodology in which I will explain the steps that I have taken in order to conduct the analysis of how users are engaging with the social features of YouTube and the meaning hereof for the social structure of the platform. As already discussed, I will be deploying a digital methods approach which will allow me to perform a close research of the social connections between users based on

(10)

network visualisations. Since the magnitude of the platform I was forced to perform a case study analysis for which I designed a set of 30 travel vlogger channels that, because of their high levels of engagement (Crowel et al. n pag.), will allow for interesting social research.

Chapter 7 shows the findings and discussion of this case study analysis. Divided into multiple sections that cover the relations between users based on the three most important social features on the platform. I will discuss how different user types make different use of these features, which allows for different social structures to be formed. I will come back to the main motivation of this research - the fact that YouTube claims to be a community - in chapter 8 by discussing if and how

community structures are discernible from the way users are engaging with the social affordances.

In chapter 9 I will mention some limitations of the method, as the tools that I have been working with to conduct the data but also the way I have made use of them have shown some limitations for this research. Hereafter I will conclude this research by presenting a tripartite conclusion in chapter 10 that will give an answer to my research question. Here I will also provide some suggestions for further research.

2. SOCIAL STRUCTURES, (ONLINE)

COMMUNITIES AND PLATFORM

AFFORDANCES

The organisation of the social in the offline world is often described with the help of the term social structure. New media studies and other disciplines that focus on life in the online world have borrowed the term that originated in social sciences to discuss how the social is organised in an offline world. To understand where this term that is now also used to describe social organisations in the online world, comes from I will first address sociological literature providing insights in the concept. This exploration will lead to a new definition of social structures so that it is more applicable to online platforms. Subsequently, the specific social structure of a(n) (online) community will be addressed, as YouTube has called itself a community from their early days. This section will address the similarities between offline and online communities as well as their differences and what that implicates. Lastly the roles of platform affordances in

(11)

the formation of online communities and online social structures are subject of discussion. The concept of ‘affordances’ as discussed by Bucher and Helmond, addresses the implications in the design of social media platforms.

2.1 Social structure: a definition

The term ‘social structure’ has been used within multiple disciplines2, but mainly within social sciences. Many famous theorists have contributed to the

institutionalisation of this term and field of research. It is one of the most central concepts within social sciences, however, the term is conceptualised in many different ways. Social structure is a term that can be used to analyse all kinds of social groups. Subsequently, a social group can be defined as networks of individuals bounded by a particular relationship and who share a common social identification of themselves (Turner 15). Since there is not really one shared understanding of what ‘social structure’ actually means, sociologist Douglas V. Porpora presents four different conceptions of social structure that are, according to him, the most prominent within the field:

- Patterns of aggregate behaviour that are stable over time - Law-like regularities that govern the behaviour of social facts - Systems of human relationships among social positions - Collective rules and resources that structure behaviour

(Porpora 339)

These conceptions, produced to portray the offline world, have a lot of overlap, but also represent very different social organisations. As not one of these is directly applicable to online platforms, I will propose a new conception, based on Porpora’s conceptions, to better portray the social structures present within social platform environments. Social structures in a platform are based on law-like rules and

resources alongside which relations between platform users are formed as a result of their patterns of aggregated behaviour with each other. Within this definition it

becomes clear that social constructs on an online platform are subjected to a strict and limited set of rules, namely the actions that the platform allows its users to perform.

(12)

Next to that this definition also points out the importance of regular interactions for bonds to form between different users.

2.2 (Online) communities as a social structure

As mentioned before, YouTube has always self-proclaimed to be a community. ‘Online community’ or ‘virtual community’ are no unusual terms when it comes to describing the social structure of online platforms. But what does it mean to be a ‘community’? How do they emerge? How does a platform allow for a community to be formed? And what kind of social relations does a community entail?

According to information studies scholar Caroline Haythornthwaite, the term ‘community’ used to be related to a physical location or “geographic touchstone” (121). Communities in the original sense of the word, which would now be referred to as ‘offline communities’, have been a popular object for study within social science in particular. David W. McMillan and David M. Chavis, scholars in the field of

community science and psychology, argue that in order for people to experience a ‘sense of community’, there are four characteristics that a social group must experience:

- Feelings of membership: the feeling of belonging to, and identifying with the community (9).

- Feelings of influence: the feeling of having influence on, and being influenced by the community (11).

- Integration and fulfilment of needs: the feeling of being supported by others in the community while also supporting them (12).

- Shared emotional connection: the feeling of relationships, shared history, time together and a ‘spirit’ of community (13).

Despite the fact that the article on the above mentioned characteristics of a ‘sense of community’ was published in 1986 and referred to offline communities, these four characteristics have proven to hold as important characteristics for a sense of

community in virtual or online communities (Blanchard & Markus 5). When it comes to discussing the social structure of the Internet of today, the community metaphor

(13)

still continues to influence the way we think about and study the social web and platforms.

In one of the first books about virtual communities, Howard Rheingold states the following about virtual communities: “virtual communities are social aggregations that emerge from the Net when enough people carry on those public discussions long enough, with sufficient human feeling, to form webs of personal relationships in cyberspace” (6). The public discussions that Rheingold is talking about are possible because of computer-mediated-communication or CMC. He even argues that

“whenever CMC technology becomes available to people anywhere, they inevitably build virtual communities with it, just as microorganisms inevitably create colonies (7). Meaning that online communication will always lead to the formation of communities and thus always be present on the web. This view is supported by Haythornthwaite, who states that online communities proved to entail the same kinds of “strong emotional and social bonds associated with local community” and their members proved to be “sharing the resources of stories and information, enjoying their time together online and working toward common goals” (121). So, freed from the geographical constrains, Haythornthwaite argues that an online community emerges “when the cumulative impact of interactions among individuals adds value above the level of pairwise interactions” (121). These can be interactions based on the four needs and characteristics as outlined above by McMillan and Chavis, such as the exchange of information and advice, social support, mutual help and provision and receipt of services. These interactions ultimately create trusting bonds between community members and in achieving that create a shared history, language and known expectations in order for the community to achieve its common goals (Haythornthwaite 121).

Technology used to be designed for usability3 and was mainly work or task-related (Preece 349), the rise of online communities thus led to new issues for researchers and developers. Just designing for usability was not enough anymore, so an understanding to how technology can support social interaction, or in the words of computer science scholar Jenny Preece ‘sociability’4, needed to be developed (349). With sociability she refers to “developing software, policies and practices to support

3_{Usability focuses on the interaction across the human-computer interface (Preece 349).} 4_{Sociability focuses on human-human interaction supported by technology (Preece 349)}

(14)

social interaction online” (349). Good sociability thus determines how a platform can contribute to and facilitate the formation of a community. Preece discusses some key components that contribute to good sociability on a platform:

- Purpose. The focus on a shared interest or need that serves as a reason for members to belong to the community (349).

- People. The people form the community and interact with each other. They all have individual, social and organisation needs. They might take on different roles within the community (leaders, protagonists, comedians, moderators, etc.) (349).

- Policies. The language and protocols that guide the user’s interactions. These can be informal and formal policies that provide community governance. Examples hereof are folklore and rituals but also registration policies and codes of behaviour (349).

She argues that decisions about these three key components by platform developers help to determine the initial sociability of an online community (349). Decisions about these sociability components thus allow for online communities to be formed. Preece identifies two types of relations within online communities: strong-tie relationships that “satisfy important needs and produce closely-knit groups as in family relationships” (348) and weak-tie relationships that “occur when people do not depend on each other for life supporting resources. The focus of these weak-tie

relationships is often information exchange, as in many special interest groups” (348).

2.3 Platform affordances: how features are objects of intense

feelings

The social structure of a platform comes into being by the way users are interacting or engaging with certain features of the platform. To understand and make sense of the kind of relations that are being formed on the platform, it is important to look further than the technical constrains of the features. As Bucher and Helmond argue: “a feature is […] not just a feature. The symbols and implications that they carry matter.” (2). They state that the concept of ‘affordances’ is a “key term for

(15)

understanding and analysing social media interfaces and the relations between technology and its users (3). Ganaele Langlois implies this as well as she states that software can be understood as a “communicational actor, as an entity that produces meanings and meaningfulness, an entity that interacts with us” (52). The specific interest of a platform is often embedded in these symbols and the implications of certain platform features. It could even be stated that the platform’s affordances dictate what social interactions and connections are possible and thus are the main influence on how the communities on the platform organise themselves. This is aside from platform affordances that are repurposed by a community to serve a purpose that was not intended by the platforms developers. Understanding how users are

employing features on a platform can thus be very telling for drawing conclusions on what those affordances mean for social structure on that platform.

Exactly how important the symbols and the users’ connotations with that symbol are was proven when Twitter changed its star icon for favourites to a heart icon for likes. According to Twitter, the heart was universally better understood as a sign that could say “congrats, adorbs, LOL” (Twitter n. pag. and Kumar n. pag.), especially for new users on the platform as the star “could be confusing” (Kumar n. pag.). Twitter users however were not impressed and felt that “a heart is a terrible symbol to use when you mean ‘you're totally owning that guy in this argument I'm too cowardly to participate in’” (Meyer n. pag.). Becoming aware of a platform’s

affordances will thus help to make sense of the way users are engaging with certain features and forming a social structure.

Although the concept of social structures and more specifically online communities originates from sociological theories and research, these classic views on how the social organises itself has proven to be as applicable then in offline situations as now in online social environments. The affordances of the platform have an undeniable role in the creation of those communities, as they are the dictators of what is possible and what is not possible in terms of making social connections and bonds, but also in terms of communication possibilities.

(16)

3. INTRODUCING YOUTUBE: USER TYPES

AND THE PROFESSIONALIZATION OF

VLOGGERS

As this thesis is an exploration of the social structures formed on the video sharing network site YouTube it is important to understand the website as a whole first. So, after having introduced and discussed the most important concepts that I will be working with throughout this thesis, I will introduce the object of my study: YouTube. This chapter is structured as an inverted pyramid (see Figure 1), which means that I will start broad by discussing the Web 2.0 environment, which

influenced the character of the platform when it emerged. Also, I will briefly discuss the founding and evolution of the platform. Then, I will discuss the users of YouTube and their different levels of engagement with the platform and with each other. Finally, I will describe how the popular practice of vlogging went from an amateur practice to something people now do as a profession as this shows the changing character of the platform’s social structure.

(17)

3.1 Web 2.0 and the emergence of YouTube

Due to technical innovation as well as the development of new user habits and requirements, the Internet has gone through some major changes over the past decades. Whereas the Internet used to be a place of one-way communication, from producers to consumers, we now use the Internet as a place of many-to-many communication, or ‘connected communication’ (van Dijck, Culture of Connectivity 5). Technical writer Tim O’Reilly referred to this evolving Internet landscape as ‘Web 2.0’ (O’Reilly n. pag.). The term mainly provided a useful conceptual framework that analysts, marketers and others involved in the tech field could use to categorise and comprehend the “new generation of internet applications and businesses that were emerging to form the ‘participatory web’ as we know it today” (Fox & Madden n. pag.). Besides a means to categorise and comprehend certain new media

developments, Web 2.0 can also be referred to as an ideology or a change in discourse about which new media scholar Alice Marwick argues the following:

Web 2.0 is more accurately described as a set of applications and general philosophy of information and technology than a technical development. This philosophy espouses transparency, openness, creativity, participation, and freedom. (86)

Web 2.0 and the emergence of YouTube

User types and different levels of engagement

Vloggers

(18)

It has especially been social media platforms that introduced new possibilities for participation and are being called one of the most typical Web 2.0 applications (Beer & Burrows n. pag.). Social media platforms were positioned as “a group of Internet-based applications that build on the ideological and technological foundations of Web 2.0, and that allow the creation and exchange of user generated content” (Kaplan and Haenlein 60). Characteristic for this emerging participatory web were websites and social media platforms such as YouTube, Flickr, del.icio.us, Digg, Facebook,

Instagram and Twitter where an online participatory culture is formed (Helmond 6).It is within this realm of ‘the Web 2.0’ that a lot of social media platforms emerged at a fast pace, with YouTube thus being one of them.

YouTube was officially launched in June 2005, after being conceived in a Silicon Valley garage by former PayPal employees Chad Hurley, Steve Chen and Jawed Karim (Van Dijck, The Culture of Connectivity 110). The initial objective of the website was to serve as a platform for sharing self-made amateur videos. The website was one of multiple competing services5 that aimed to make the sharing of videos online easier. From the beginning YouTube entailed community features that I will elaborate on in chapter 4. Jawed Karim, one of the co-founders, argued in 2006 that one of the essential features that determined YouTube’s success were the video comments and other social networking and interaction tools (Gannes n. pag.).

The real success story however, began in October 2006 when Google acquired YouTube for $1.65 billion. From this moment on, the website gained a lot in

popularity. This acquisition by Google has of course been a crucial moment and probably determined a large part of YouTube’s success. By this time Google was rapidly growing and became increasingly dominant in the world of networked

sociality. This maturing image also reflected on YouTube (Van Dijck, The Culture of

Connectivity 111). YouTube started as a fun project from a group of like-minded

individuals and grew out to be a large-scale business organisation. This enormous change has of course carried some implications for the social structure and the users of the platform.

(19)

3.2 User types and levels of engagement

In this paragraph I will go into some different types of users that can be distinguished on YouTube. As is the case for every website and platform, users visit YouTube with different intentions and have different levels of engagement. They might interact with the platform in diverse ways and use different parts of the platform (Smith et al. 104). Digital media and anthropology scholar Patricia de Lange distinguishes five different types of YouTube users, based on an analysis of interviews with YouTube users:

1. Former participants

These users no longer post videos but do maintain an account, watch videos and occasionally post a comment.

2. Casual users

These users mostly do not have a YouTube account. They tend to only view videos when they specifically search for something or when they get sent a link to a particular video.

3. Active participants

These users have an account and usually upload videos or at least participate by leaving comments on other people’s videos. Active participants may be aware of the issues and people that are important in the YouTube community.

4. YouTubers or ‘Tubers’

These users have a more intense engagement with YouTube in terms of the amount and type of their participation. They mostly spend an hour or more on the website when visiting, which happens daily or at least weekly. Many of these users upload videos and promote these within as well as outside

YouTube. They pay close attention to and participate in YouTube debates and discussions.

5. YouTube celebrities

These users share some characteristics with the previous group, YouTubers. However, they are also quite well known both within and often outside of the site. YouTube celebrities influence the discourse, goals, and activities on YouTube through their videos, comments, bulletins, and other forms of interaction.

(20)

Another approach for distinguishing user types leads to four types of users: lurkers6, fans, influencers and pros (Levine n. pag.). What I want to show by discussing these user types is that users on YouTube have different levels of engagement. This varies from users who occasionally visit the platform to professional vloggers who make their living from being a YouTuber. Users that are more engaged with the platform will probably also be more engaged with the community and thus fulfil a more prominent role within the social structure of the platform. Also, most lurkers, former participants and casual users will not be discernible within the social structure when they do not make use of the social features of the platform. So, a large part of the platform’s users is kind of invisible when it comes to analysing the social structure. Also, it is very hard to draw a line between the different user types. A channel allows users to both display content that they uploaded, and keep track of the videos of other users that they are subscribed to. The various possible uses of a channel blur the boundaries between creators and consumers of content (Susarla 24). For the analysis of the social structure on YouTube, I will be looking at the social connections between the creators of content. Simply because these users or, active participants, YouTubers and YouTube celebrities, as de Lange calls them (Commenting on

Comments 5) are more prominently present in in the social structure because of their

active role on the platform.

3.3 Vlogging: from amateurs to professionals

The stars of YouTube are the highly popular vloggers, these video bloggers create video content on a regular basis about their daily lives. Here I will introduce the vlog and the makers hereof. The vlog (which stands for ‘video log’) is most often

associated with the initial objective of YouTube because vlogs started as pure ‘amateur’ video production (Burgess & Green 93). Since the word ‘vlogging’ has been derived from the word ‘blogging’, these two have quite some similarities in their base. Information and computer sciences scholar Bonnie Nardi argues back in 2004 that blogs are intrinsically social, as they reveal the blogger’s personality, passions and point of view (Nardi et al. 42). Also, a blog provides a platform for interaction between bloggers, their readers and fellow bloggers. Therefore, blogs that share a

(21)

similar focus, interest or topic are usually interconnected and part of a virtual

community (Nardi et al. 45). The same goes for vlogs, since they entail a certain type of interpersonal face-to-face communication because the vlogger is directly talking to the camera (Burgess & Green 94). As argued by media and digital culture scholar Will Luers, vlogging is also a very social practice because most vloggers tend to look to fellow vloggers for feedback and support (n. pag.). Luers identified some social needs that vlogging seems to fulfil: “being connected, finding validation for one’s experience and ideas, and being a producer as well as a consumer” (n. pag.). So, like the blog, the vlog seems to be intrinsically social. Burgess and Green described the vlog as following back in 2009:

Typically structured primarily around a monologue delivered directly to camera, vlogs are characteristically produced with little more than a webcam and some witty editing. The subject matter ranges from reasoned political debate to the mundane details of everyday life and impassioned rants about YouTube itself. [...] The form has antecedents in webcam culture, personal blogging and the more widespread “confessional culture” that characterises television talk shows and reality television focused on the observation of everyday life. (94)

Even though this description of Burgess and Green was written seven years ago, the basis of vlogging has remained the same: a person talking into the camera and

showing their day-to-day life. However, a large group of vloggers on YouTube cannot be called ‘amateurs’ anymore. A lot of vloggers nowadays use professional cameras and microphones and have developed some impressive editing skills. A lot of this professionalization has to do with the YouTube partnership program, introduced in 2007 (YouTube, Partnership n. pag.) YouTube recognised the growing popularity of some community members and declared the following in a blog post: “many of you have gone from creating videos in bedrooms and living rooms on webcams to

becoming YouTube celebrities with fans and audiences all over the world” (YouTube,

Partnership n.pag.). What the partnership program entails is that partners can

“participate in the same revenue sharing and promotional opportunities that are available to YouTube’s other partners” (YouTube, Partnership n.pag.). As a result, a

(22)

lot of vloggers nowadays can make their living from being a YouTube partner and so are definitely not to be called ‘amateurs’ anymore.

Vlogging in general has become increasingly popular over the last few years (Warmbrodt et al. 1). Because of the YouTube partnership program and the

professionalization of vloggers in general, vloggers have established a firm role within the media field and gain more exposure. The possibility to generate an income from being a professional vlogger on YouTube may have changed the way users are connecting on the platform. Since they are dependent of the exposure and popularity of their videos, they will make different use of the social features by deploying them for promotional practices.

4. GOING WAYBACK: YOUTUBE’S

COMMUNITY AFFORDANCES

To understand the origin of YouTube proclaiming to be a community and how YouTube has changed its affordances and thus the accompanying connotations of those affordances it is important to map out the changes that the platform went through. Therefore, I will discuss the most important community aspects that the platform entailed during its earlier years. For this I will make use of the Internet Archive’s7_{search interface: the Wayback Machine (www.waybackmachine.org).}

Looking into the history of a platform is helpful for better understanding the platform as it is today (Brügger 753). Furthermore, shifting priorities and commitments of the group, organisation or institution behind the website might be recognisable via changes in the interface (Rogers 70). This piece of historiography serves as a relevance to why it is interesting to analyse the social structure on YouTube as it is today; does this still resemble a ‘community’ even when a lot of the community features that once seemed a crucial part of the platform’s social structure, have disappeared?

7

This non-profit Internet archive was established in 1996 and strives to archive Internet based entities, mainly for research purposes. The Internet Archive can be used to visit previous versions of website interfaces.

(23)

4.1 Statement on the About-page - 2005

As can be seen in Figure 2 below, which shows the earliest captured About-page, YouTube is from the beginning describing itself as an ‘online community’.

Figure 2: YouTube’s About-page as seen through the Wayback Machine.

This statement on the About-page suggests that YouTube would have an emphasis on social connections and would rely on the collaborations and interactions of users to make the website more valuable, which is one of the most important characteristics of a community (Haythornthwaite 121). By using this term to describe the platform’s social structure, YouTube encourages its users to contribute to and become a member of this community. So this self-proclaimed community character interpellates the user as an active and engaging member of this community. Moreover, by calling

themselves a community, a user will be more likely to have a sense of community when engaging with the platform.

4.2 My Friends - 2005

As can be seen in Figure 3 below, about three months after YouTube’s launch, the website implemented a new section called ‘My Friends’.

(24)

YouTube stated the following about this feature:

Friends, Family, and any other lists you decide to create allow you to quickly share videos with a large group of people. Once you’ve added someone to your contact list, you can move them to your Friends list or any other list you’ve made. […] You can also set privacy settings for your contact lists, which allows you to make videos visible only to members of that list. This makes it easy to share personal videos only with friends or family, and other YouTube users will not be able to see them. (YouTube’s Help Center as seen through the Wayback machine. August 2005)

According to social media scholars danah boyd and Nicole Ellison, having an articulated list of friends, who are also users of the website is an important technical feature of a Social Network Site (213). By introducing this ‘My Friends’ feature, YouTube thus seems to position itself as a Social Network Site.

4.3 Groups - 2006

Shortly hereafter, another community feature called ‘Groups’ was introduced, as can be seen in Figure 4 below.

Figure 4: Implementation of 'Groups' feature as seen through the Wayback Machine.

In the glossary that was provided on the website at the time, YouTube explained this feature as can be seen in Figure 5 below.

(25)

So besides having social groups of friends or family, a user could also join a group based on a shared interest to expand his or her social connections on the platform even further. Whereas the ‘My Friends’ feature will most likely contain a lot of the user’s offline friends and/or family, the ‘Groups’ feature will probably contain users that might not know each other offline, but became acquainted on YouTube because of a shared interest. Media scholars Dana Rotman and Jennifer Preece argue that

communities on YouTube are maintained through shared interests and a sense of belonging (325).

So the ‘My Friends’ and ‘Groups’ features on YouTube seemed to be crucial for a sense of community on YouTube because these were the main sites on the platform where interaction between users took place. What can also be seen in Figure 4 is that YouTube already entailed a ‘My Subscriptions’ feature. Being subscribed to a user thus denoted a different relation than being friends with a user. A friend-relation is a bi-directional tie, meaning that both parties need to agree on being friends. A subscriber-relation is a one-directional tie. In December 2011 however, the subscriptions and friends were merged on the platform. YouTube stated the following on their help forum:

Friends and Subscribers have overlapped in many cases and it’s been

confusing how they’re different. Some people both ‘subscribe to’ and ‘friend’ a channel at the same time, making it hard to tell why you’re getting an update. Now, you will have all the information you’re looking for from one source, making your experience much more relevant, easy to view, and (hopefully) more entertaining. (YT Kendall n. pag.)

This would imply that users perceived their connections based on subscriptions as being similar to connections based on friends. This would also explain why there are no ‘friends’ anymore on the platform as it is today, because the subscriptions would have taken over that feature.

(26)

4.4 Community tab - 2006

At the end of 2006, a ‘Community’ tab was introduced, as can be seen in Figure 6.

Figure 6: Introduction of a 'Community' tab as seen through the Wayback Machine.

In this Community tab, the Group section could be found, even as a section with Contests and Colleges. The Community tab left the platform in August 2009, so for almost three years, this feature was in use on the platform and was very telling for how YouTube was establishing itself as being a community or allowing for

communities to be formed on the platform. Of course there have been numerous other changes on the platform, but I chose to only include the most important features that were significant for YouTube’s aim to be perceived as a community. These features show how YouTube really put an emphasis on the social connections that users could make on the platform. Users could already rate videos and subscribe to other users, but YouTube still felt it was necessary to use terms like ‘community’, ‘my friends’ and ‘groups’ to create a sense of community on the platform.

So, during the first years of YouTube’s existence, the platform tried to really establish itself as being a community by articulating this through platform features such as friends, groups and communities. These features were very telling for the community character of YouTube during the first few years. On the platform as it is today, all of these features are not longer available. As mentioned before, YouTube is still

referring to itself as being a community, which allows for interesting research as to how the social structure of YouTube is today and if this resembles a community, even without the presence of these typical community features. Therefore, I will proceed by looking into the three most important current community features, or rather ‘social features’, as I have not established whether these features actually allow for

(27)

connection on the platform as it is today are featured channels, subscriptions and comments. In the next chapter, I will explore these features by discussing their social affordances.

5. YOUTUBE’S CURRENT SOCIAL

AFFORDANCES

In this chapter, I will discuss the three most important current social features on YouTube. In the previous chapter I discussed some of the most important community features that the platform entailed during the earlier years. Features like ‘My Friends’, ‘Groups’ and the Community-tab were important sites for users to connect, but have all disappeared from the platform. The most important features for users to connect to other users on the current version of YouTube, are: featured channels, subscriptions and comments. I will thoroughly discuss these three features in the following paragraphs, as well as explore the possible uses and what they implicate about the social structure of the platform. For this, I will deploy the concept of platform

affordances by Bucher and Helmond as introduced earlier. These features can be seen as ‘low-level affordances’. Bucher and Helmond state that “low-level affordances are typically located in the materiality of the medium, in specific features, buttons, screens and platforms” (12). Instead of only describing how the features work on a technical level, I will discuss their symbolic values, uses and implications that they carry. Because “pressing a button means something; it mediates and communicates and […] relates to different affordances (Bucher & Helmond 2).

5.1 Self-presentation and promotional practices through

featuring

The featured channels section is one of the ways in which users can actively connect to each other on YouTube. This feature allows for users to add other users’ channels that will then be displayed in this section. This featured channels section is based on the way YouTube creates navigational pathways by offering click-throughs (Lovink and Niederer 12). The idea is that users are more likely to visit a channel that is being

(28)

featured, or recommended, by a channel that they like. Rogers states about making links to another site, or in this case to another channel, that “making a link [...], not making a link, or removing a link may be viewed, sociologically or politically, as acts of association, non-association, or disassociation” (44). This featured channels feature is something that channels use to express them feeling associated with another

channel. This is something that Marwick calls “publicly articulated relational ties” (To See and Be Seen 146). Featured channels are based on one-directional ties, so if channel A is featuring channel B, channel B does not have to confirm this connection, or feature channel A ‘back’ (boyd and Ellison 213). ‘Featuring’ denotes a strong emphasis on promotion. So, users can promote other users in this section. However, this also works the other way around. The public displaying of connections to other channels allows users to actively negotiate their self-presentation. The featured channels section can thus function as ‘identity markers’ for the channel owner (boyd and Ellison 220). Associating your channel with other users’ channels could have an influence on the reputation of your channel. The featuring other channels does not only benefit the featured channel, but thus also functions as a form of reputation management or brand building for your own channel.

5.2 Subscriptions as expressions of fandom

Like featuring channels, the subscription feature is also a way for users to actively connect to other users. As media and technology scholars Anjana Susarla, et al. describe: “subscriber ties could exist for informational purposes to obtain content based on users’ tastes and interests” (25). So besides an expression of feeling associated with another channel, subscribing to a channel is also functional. By subscribing to a channel, every new uploaded video will appear on the user’s subscription page which makes it easier to keep up with all the new videos of one’s favourite channels. Something that endorses the assumption of subscriptions being mainly a functional practice is the fact that subscriptions are usually not displayed on user channels. Subscriptions are always visible as a number on every channel page, but not with (user)names and profile pictures as is the case with featured channels for instance. Sometimes a channel chooses to display its descriptions, because this is possible, but most of the time, users set their subscriptions to private. Also, it does not work the other way around: you can never see a list of the channels that are

(29)

subscribed to a specific channel. While the featured channels feature mostly represents connections between fellow content creators, because users most likely recommend another channel for the content, subscriptions represent a one-directional tie between a content creator and its viewers8_{. Because of this, subscriptions can be}

considered as a social connection that enhances a social structure that Van Dijck calls a “fan-centric media circuit with a YouTube star” (Publicly Private 377).

The fact that the subscriptions mechanism is called a ‘subscription’ is interesting on a symbolical level. Subscribing usually represents an arrangement in which the user must periodically pay a fee to get access to specific content, a service or a product. One might get a subscription to a newspaper, or for instance a phone or Internet provider. For being subscribed to YouTube channels however, a user does not have to pay. Also, the content of a channel is still available when a user is not subscribed to the channel. So this type of subscription does not work entirely the same as ‘normal’ subscriptions do. The choice for the word ‘subscribe’ on the buttons of a channel instead of for instance ‘join my community’ or ‘become fan’ or ‘follow’ to give a few options, is interesting because ‘subscribe’ represents a service that usually is quite anonymous and impersonal. Furthermore, the association with subscribing to television services installs a feeling of the platform being a site for broadcasting practices similar to television, with the YouTube celebrities as broadcasters.

As already mentioned, subscribing usually represents an arrangement between a company and an individual, and so does not denote a personal or friendly

connection. Even though subscriptions might be ‘weaker’ social connections than featured channels because they mostly represent a practical use, they still are important components for the general social structure of the platform because they denote a certain hierarchy between content creators and their viewers.

5.3 Maintaining viewer loyalty through the comments

Besides featuring and subscribing, the comment section is another way in which users can actively connect and interact with each other and thus contribute to the formation of a social structure. This section is displayed on the video page right below the video.

8

As mentioned before, among these viewers can of course also be content creators. But even though if they regularly upload videos themselves, if they are subscribed to a channel, they can be considered to be viewers of that channel.

(30)

Together with the discussion tab9, and private messages, this is the only way users can communicate with each other in in actual words on YouTube. This section can either be organised based on top comments, or from newest to oldest. Users can mention each other and reply to each other. Comments can also be liked, disliked or reported. What makes the comment section a different social connection than featuring and subscribing, is that this type of connection is more open to a user’s own interpretation and is an interaction that can even be conversation-like. After all, featuring mainly functions as a click-through to another channel and subscribing instructs a script to fetch all new uploaded videos for the channels that a user is subscribed to. These functionalities can be seen as engagement techniques that computer scientist Philip Agre calls ‘grammars of action’. Grammars of action can be defined as a set of rules and constraints that are embedded in the software’s interface (745). They structure the pathways a user can take within the software and define certain modes of interaction and engagement. So, in short, the grammars of action define what a user can and cannot do within the software. The act of placing a comment is also a grammar of action, or an option in the interface that a user can follow; however, the user can decide for himself what to comment.

The comment section is an important site for content creators to engage with their viewers. This practice brings YouTube celebrities closer to their fans. Besides this socially orientated use, this type of interaction also benefits the content creator. This can be seen as a form of maintaining and intensifying fan-relations, which will result in more loyal fans or viewers. Besides a site for interaction between content creators and their viewers, the comment section also serves as a discussion board where viewers can engage with each other. By doing this, they create community dynamics around a certain video (Rieder, Introducing the YouTube Data Tools n. pag.). So, the comment section allows for different user types to engage with each other in order to intensify existing social connections or establish new ones.

The featured channels, subscriptions and comments thus allow for different kinds of social connections to be established. These features of course carry different technical actions, but they also suggest or afford different social actions and thereby establish

9

(31)

different social relations. These social relations all together can be analysed in order to say something about how the social structure of YouTube is constructed.

Therefore, I will use the connections based on these three features to analyse the social structure on the platform. In the next chapter, I will explain all of the steps I have taken and the tools I have used in order to perform this analysis.

6.METHODOLOGY

In this chapter I will discuss the methodology and all the steps I have taken in order to conduct the case study analysis of the social structure on YouTube as seen through its three most important social features: featured channels, subscriptions and comments.

6.1 Demarcation of the dataset: travel vloggers

Because of the scope of the platform, it would be nearly impossible to perform an analysis of the social structure by analysing data for all users of the platform.

Therefore, I decided to perform a case study based on a dataset of travel vloggers on YouTube, which can be seen as a subgenre of vlogging in general. A Google report from August 2014 shows that travel vlogs receive a lot of social engagement (Crowel et al. n. pag.). The pie chart in Figure 7 shows the total percentage of subscriptions to travel channels. Nearly half of all the subscriptions are to travel vlog channels that feature personal travel experiences.

(32)

Not only are travel vlogs very popular when looking at the number of subscribers, travel vlogs receive four times more social engagement (likes, comments, shares, favourites and subscriptions) than other types of travel content on YouTube, which can be seen in Figure 8.

Figure 8: Content that travellers are most likely to engage with according to Google (Crowel et al. 5).

These numbers thus indicate that there are lots of social bonds and social actions present within the network surrounding travel vloggers. This makes the travel vlogger an excellent subgroup for analysing the social structure and habits, as it is present on YouTube.

6.2 Digital Methods

In order to map out YouTube’s social structure I will repurpose data that is freely available on the web and I will repurpose that data so that it can give insights in the organisation of the social on YouTube. This repurposing of online available data is in line with the Digital Methods Initiative10 way of doing research. By deploying

platform specific methods and tools, digital methods make use of the forms and materials of specific digital media (Rogers 15). Richard Rogers, founder of the Digital Methods Initiative, argues that digital methods as a research practice, “strive to follow the evolving methods of the medium” by continually thinking along with the devices and the objects they handle (1). Also, for studying web services and for developing tools and interfaces, digital methods search for ways in which digital objects can be combined and recombined until they provide interesting insights and questions for social and cultural research (1). Another important principle of digital methods is that

10_{For more information about the Digital Methods Initiative:}

(33)

they build upon the existing dominant devices themselves. Digital methods create ‘views’ by combining data in specific ways that allow for analysis, “turning what was once familiar […] into indicators and findings (Rogers 3).

I will be using digital methods to perform a medium focused research. Instead of studying platform documents on how users can or should use specific features, or performing a survey-research as to how users are saying they are making use of specific features, the tools developed by the Digital Methods Initiative allow me to actually see how users are making use of these features. The tools that I will be using are built on the YouTube API. The tools make calls to the API to extract certain data. By doing this, the tools create ‘views’ by combining data into specific sets that can be processed in software and allow for analysis. To make sense of the social structure on YouTube, I will analyse the data with the following questions in mind:

• Who are connecting?

• Through which social feature(s) are they connecting? • What kind of clusters are being formed?

Studying the data with these questions in mind, allows me to draw conclusions on the social structure of the platform by looking at the way users are connecting to each other by making specific use of the three social features.

6.3 Gathering the data: YouTube data tools

In order to gather the data that I needed for this case study of the social structure on YouTube, I made use of the YouTube data tools developed by media scholar

Bernhard Rieder. I used two tools for obtaining the data of the connections based on featured channels, subscriptions and comments. For the connections based on featured channels and subscriptions, I used the Channels Network tool. For the connections based on comments, I used the Video Info and Comments tool.

6.3.1 Channel Network tool

The first tool that I used is the Channel Network tool, which can be seen in Figure 9 below.

(34)

Figure 9: YTDT Channel Network tool developed by Bernhard Rieder.

This tool extracts data that can be used to show the connections between channels. The connection is based on featured channels, with the option to also show the connections based on subscriptions. The tool offers two options to use as a starting point:

1. Search query

When using a search query11_{as starting point, the YouTube algorithm will provide the}

results that will end up in the dataset that results from using the tool. These results can be ranked based on different values, which can be seen in Figure 10. Using a search query as seed can provide insight in how YouTube orchestrates access to a specific topic and what users encounter when they search for specific content.

Figure 10: YTDT Channel Network tool - ranking options when using a search query as seed

11_{The world “query” refers to}_{a precise request for}_{information retrieval}_{with database and information} systems.

(35)

When using a search query as seed, the number of iterations can be adjusted. When using an iteration of 1, the tool will retrieve the first 50 results. This can go up to 1000 results. There is the option to also include the connections based on subscriptions. The last variable that can be adjusted is the crawl depth12_.

2. Seeds

When using a self-designed dataset, the seed13 can be one or more channel ID’s14. The option to also include the connections based on subscriptions and to set the crawl depth are the same as when using a search query

The tool will present the data in a .gdf file that can be used to visualise a network15. I used both a search query and a self-designed list of seeds as a starting point for the tool, but, as I will elaborate on in paragraph 6.5, I ended up only using the dataset based on the self-designed list of seeds for this analysis.

6.3.2 Video Info and Comments tool

The second tool that I used is the Video Info and Comments tool, which can be seen in Figure 11.

Figure 11: YTDT Video Infro and Comments tool designed by Bernhard Rieder.

12

This can be set to 0, 1, or 2. A crawl depth of 0 will only retrieve the data - so the featured channels (and/or subscriptions) - for the seeds. A crawl depth of 1 will retrieve the data for the seeds as well as their featured channels (and subscriptions) that are thus not in the initial seeds. Crawl depth 2 will go one layer deeper again and retrieve featured channels (and subscriptions) for the featured channels of the seeds.

13

The word "seed" refers to a piece of data that is used to generate other data.

14

The channel ID can be found in the URL or in the source code of a channel. Every channel has a unique channel ID, which is a combination of letters and numbers.

15

A network here refers to a graph structure “which represents people or objects as nodes, and

relations with undirected edges (for symmetric relations) or with directed arcs (for relations that are not necessarily symmetric)” (Salah et al. 415). Each node thus harbours a certain amount of information and is related to other nodes by directed or undirected edges. These graphs can be analysed to detect certain densities, connections and reciprocities (Rieder, De la communauté à l’écume 42).

(36)

This tool retrieves all available comments, both top level and replies, for one single video. The input is the video ID, which can be found in de URL of the video. The tool creates multiple outputs, of which I used the tabular file containing comment authors and their comment count. By analysing this, active users can be found and community dynamics that develop around videos can be mapped and understood.

6.4 Visualising networks: Gephi

For creating network visualisations of the social connections on YouTube I used Gephi (Bastian, Heymann & Jacomy 2009). Gephi is an open source network

exploration and visualisation software tool (Bastian & Heymann 361). I imported the .gdf files that contained the data extracted by the YouTube Channel Network tool into Gephi in order to create network visualisations. All of the visualisations were based on the Forceatlas2 algorithm. Also, I used the LinLog mode for every network to make the clusters more tight. I will be using Gephi to look into the social connections based on subscriptions and/or featured channels, these are both one-directional ties and will thus be treated as a directed graph.

6.5 Trial and error: the path to the eventual dataset of travel

vloggers

As already mentioned, I first used a search query as starting point for the Channel Network tool in order to map the connections between travel vloggers. I used a search query because this would show me the kind of results that any user would encounter when searching for this topic. I decided on using the query [“travel vlog”]16_because

this would retrieve the most relevant results for this specific query. This query would retrieve the 50 most relevant results for travel vloggers on YouTube, based on the YouTube algorithms. I performed multiple visualisations of the networks based on featured channels only, and based on featured channels and subscriptions. What was remarkable for these networks is that there were very few connections between the channels. Another remarkable thing was that the algorithms retrieved a lot of very small or even inactive channels. Only 10 out of the 50 retrieved channels had more

16_{I used the double quotation marks in order to exclude results that are based on only ‘travel’ or}

(37)

than 100 subscribers, which makes the remaining 40 channels very small. This is of course not per se an inaccurate reflection of users on YouTube in general, since only a small section of all users are active participants, YouTubers or even YouTube

celebrities, as discussed by de Lange (Commenting on Comments 5). However, because of the many inactive channels, this dataset was not an accurate representation of travel vloggers on YouTube.

Performing this analysis did show me that the YouTube algorithms do not allow ‘easy access’ to specific subcultures, social groups or communities on the platform. As David Beer argues:

Algorithms are inevitably modeled on visions of the social world, and with outcomes in mind, outcomes influenced by commercial or other interests and agendas. As well as being produced from a social context, the algorithms are lived with, they are an integral part of that social world; they are woven into practices and outcomes. (4)

What this means for YouTube is that the algorithms on this platform do not rank channels higher in the search results based on connections with other channels. So, channels that are active in terms of engaging with other channels are not necessarily ‘promoted’ by being ranked in the top results. This makes it more difficult for users to find any clusters or social groups on the platform.

6.5.1 A self-curated dataset

Because the Channel Network tool did not provide me with the data I needed in order to conduct an analysis of the connections between channels when using a search query, I decided to design my own dataset. This method is of course quite subjective because I choose which channels to include and which not. But, at the same time, this method will probably result in a network with more connections between the

channels, because there will be no inactive channels within this dataset. Inactive channels are less likely to be connected to other channels, so by designing my own dataset of travel vlogger channels, I will be able to see if travel vloggers on YouTube are maybe way more connected than the networks based on the search query

Social Structure on YouTube: an Online Community or a Celebrity-growing Platform? A Digital Methods Approach to Studying the Use of Social Affordances on YouTube