Link Between Gaming Communities in YouTube and Computer Science

Lassi Haaranen and Rodrigo Duran

Department of Computer Science, Aalto University, Helsinki, Finland

Keywords:

Computer Science Education, Game-based Learning, Informal Learning.

Abstract:

Playing games has become a permanent part of popular culture, and the number of players keeps increasing.

Part of this phenomenon is the act of recording gameplay and sharing the videos creating online gaming

communities. We describe different types of gaming videos that intersect with learning computer science

(CS). In addition, we looked at the discussions those videos spurred and found a rich interaction with CS

topics. Since games can act as an engaging environment for informal learning, we conclude that the gaming

community has relevant connections to learning CS and we as the research community should pay more

attention to this phenomenon.

1 INTRODUCTION

Digital games are an important part of the modern day

culture. Whereas once video games were viewed as a

pastime for the children, this picture has dramatically

shifted in the past decades. In the United States alone,

it is estimated that the amount of money consumers

spent on games exceed 22 billion dollars, the aver-

age age of gamers is 35 years, and approximately 155

million Americans play video games (Entertainment

Software Association, 2015).

But just because the average age of a gamer keeps

climbing up, it does not mean that children are not

playing. New York Times published a long arti-

cle describing the ’Minecraft Generation’ (Thomp-

son, 2016) – a generation building their own visions in

virtual block worlds. And the culture of video games

is not just contained in the games and the worlds they

create. Different fan sites and discussion forums have

existed for a long while, but in more recent years

video sharing services, YouTube in particular, have

become an important part of the cultural phenomenon

of video games.

Videos related to gaming have spawned their

own industry. Google, owner of YouTube, it-

self acknowledged the importance of games cre-

ating its ﬁrst spinoff platform from YouTube ex-

clusively dedicated to games, YouTube Gaming

(https://gaming.youtube.com/). As an example,

there are dozens of YouTube channels focusing on

Minecraft that have more than a million subscribers

and correspondingly new videos from those chan-

nels receive millions of views. Typically these chan-

nels focus on gameplay videos, usually in the form

of “let’s play” where the author records him/herself

playing a game and narrating the events at the same

time. Whilst majority of these channels and videos

are purely for entertainment, there are some that have

educational value as well. For example, user Seth-

Bling has created a four-part video tutorial series

how to replicate a Turtle-like programming environ-

ment in MineCraft. The ﬁrst part of the series detail-

ing the use of if-else statements has gathered over

700 000 views, so the potential impact on learning to

program could be substantial.

Given how popular gaming and the videos re-

lated to it are, and especially since quite a few of

them contain instructional material regarding comput-

ing and programming, we wanted to investigate this

phenomenon further. Our goals are summarized by

the following research questions:

• RQ1 What kind of gaming videos there are that

are related to computer science?

• RQ2 How are the commenters discussing these

videos?

The overall goal is to describe and examine the

phenomena of gaming videos related to learning com-

puting. We are particularly interested in informal

learning from the videos, in other words, their use

outside of classrooms and formal education.

https://www.youtube.com/playlist?list=PL2Qvl4gaBge

02Eh4AqtDSWg3sojt3jeRO

Haaranen, L. and Duran, R.

Link Between Gaming Communities in YouTube and Computer Science.

DOI: 10.5220/0006267000170024

In Proceedings of the 9th International Conference on Computer Supported Education (CSEDU 2017) - Volume 2, pages 17-24

ISBN: 978-989-758-240-0

The rest of the article is structured as follows: In

Section 2 we provide relevant research linking games

and learning computer science. Section 3 describes

our data collection and analysis and Section 4 summa-

rizes our results. Finally, we conclude by discussing

our ﬁndings in Section 5.

2 BACKGROUND

In this section, we ﬁrst present relevant literature re-

garding the use of games in computer science educa-

tion. After that, we discuss previous work on informal

learning contexts. Finally, we explore research done

on YouTube videos and comments done in other do-

mains.

2.1 Games and Computer Science

Education

There has been very little research in learning pro-

gramming or computing in informal contexts. How-

ever, in more formal context the research on using

games to teach has been focused on programming and

computer science. Hayes and Games reviewed the

use of games in education and identiﬁed four differ-

ent goals for using games: making games as a con-

text to learn programming, as a way to interest girls

in programming, as a way to learn in other academic

domains, and as a way to understand design con-

cept (Hayes and Games, 2008). They state that the

most commonly sought after learning goal was related

to learning programming or concepts related to it.

Use of game elements (achievements, badges,

rankings) in non-game environments (such as projects

and classroom lessons) has attracted a great interest

of education community in recent years embedded in

the concept known as gamiﬁcation (Deterding et al.,

2011). Although being used in many diverse aspects,

gamiﬁcation applied to programming attempted to in-

troduce computing skills, especially to younger stu-

dents and novice programmers, in a more palatable

fashion with different interfaces, objectives, and out-

comes. Tillmann et al. describe how their new gami-

ﬁed platform, Code Hunt (http://www.codehunt.com)

can help beginners to understand computing concepts

in the form of numerical output puzzles using Java

and C# (Tillman et al., 2014) . Engagement to the

platform comes from the introduction of a ranking

system that rewards programmers that create elegant

programs with fewer attempts.

Using full games, instead of gamiﬁed appli-

cations, as an engaging environment to promote

learning of computer skills has been discussed for

decades (Kelleher and Pausch, 2005). When it comes

to formal education and computer science courses,

four ways of utilizing computer games as part of the

learning process has been described (Wallace et al.,

2010). In the ﬁrst approach, students create their own

games in order to learn (1). And relatedly, in the sec-

ond group they implement a critical part of a given

game (2). In the third approach, students learn by

implementing an agent to a game (3). This group is

of particular interest given that this can be done in

many of the current games meant for entertainment.

In the last category, students learn by playing a seri-

ous game that has been designed to teach particular

concepts (4).

Code.org (https://code.org/learn) iniative hosts

several games that support programming learning

through games. Most of these games use mechanics

similar to the popular Lightbot (https://lightbot.com)

game where users need to provide a sequence of com-

mands so a robot can reach a speciﬁc point on the

map. To address syntax issues that young users might

have most of these games use a block-based interface.

Figure 1 shows an example of a Minecraft-themed

game by Code.org, where users solve speciﬁc tasks

like constructing or moving in the map using blocks

interface.

Figure 1: Code.org Minecraft themed sequence commands

game.

2.2 Games Enhanced by Programming

Although the literature of using games to teach pro-

gramming in a formal setting as a development tool

is rich, in recent years a new approach has emerged:

using programming inside games as a way to provide

context to programming, as well as, to enhance the

play itself. This approach is distinct from the pre-

vious experiences using games not as a development

environment or an educational game to teach a spe-

ciﬁc subject, but as a skill to improve features on the

CSEDU 2017 - 9th International Conference on Computer Supported Education

act of play itself.

Minecraft, from Mojang, now the 3rd most sold

game of all time, allows deep modiﬁcations of the

game so researchers use its features as building blocks

to provide a constructionist experience to learn pro-

gramming. A tailored version of Minecraft called

ComputerCraftEdu provides insight on how block-

based programming performs versus textual program-

ming in a game environment, showing that block-

based programming increases interest in program-

ming compared to textual in the same setting (Saito

et al., 2015).

Zorn et al. developed their own block-

representations of programming constructs that func-

tion similarly to normal Minecraft blocks (Zorn et al.,

2013). This enabled users to create code in the same

manner they construct originally in Minecraft. As in

ComputerCraftEdu, the goal of the research was to

better understand how block-based versus textual lan-

guage performs in terms of students appreciation and

if learning style affected this appreciation. Results

show that programming appreciation increased with

the use of their blocks programming tool.

It is interesting to highlight that those features de-

scribed in Saito et al. (2015) and Zorn et al. (2013),

although studied as teaching tools, could be used

to automatize tasks executed extensively by players,

such as mining and constructing. This intrinsic cor-

relation with in-game functions creates the potential

use of those features in diverse aspect from the ex-

periences previously described in the literature, pro-

moting more engagement and even reaching a distinct

audience that incidentally discovered computing sub-

jects.

Another example of games that use programming

as an intrinsic part of its mechanics is CodeSpells

(https://codespells.org/). Starting as a research project

and later evolving into a full commercial game, the

concept relies on creating powerful magic spells us-

ing programming (Esper et al., 2014). A program-

ming language is used to describe the properties and

actions performed by a spell that users can freely cre-

ate and manipulate. In that sense, learning program-

ming happens due to the built-in game mechanics.

2.3 Learning Computer Science in

Informal Settings

The previous examples are from a formal context,

i.e. courses or outreach activities speciﬁcally aimed

at teaching or improving the perception of program-

ming and computer science. However, we are specif-

ically interested in informal learning in computer sci-

ence but previous research on the subject is very

sparse. One example where informal learning in

computing has been studied comes from a short re-

port that looked at how elementary school students

acquired programming knowledge (Boyle and Mc-

Dougall, 2003). Although this report looked at the

informal sources used to gain programming knowl-

edge, the knowledge was used and practiced in a for-

mal classroom setting as well.

McCartney et al. investigated what motivates

computing students to learn in informal and self-

directed settings. Using questionnaires with 17 se-

lected students, a thematic analysis was performed

and ﬁve different topics emerged (social inﬂuence,

projects, joy of learning and fear) to understand

why students engaged in this particular learning set-

ting (McCartney et al., 2016). Evidence points to that

learning with projects, in different nuances, is pivotal

in self-directed learning and could have an impact in

classroom pedagogies.

Whilst the actual learning process in informal set-

tings has not been studied in detail, there is at least

evidence that gaming communities and forums can

host a high-level discussion and scientiﬁc argumen-

tation. Investigation on discussion forum of a popular

massively multiplayer online game World of Warcraft

and found that at least in that one particular forum,

more than 80 percent of the discussion was centered

around social knowledge construction (Steinkuehler

and Duncan, 2008). And importantly, over a half of

the forum posts contained systems based reasoning,

as well as, a few posts containing model-based rea-

soning of game mechanics that the players were try-

ing to uncover.

2.4 YouTube as a Community for

Learning

One of the important features of YouTube is the low

barrier to entry to produce and consume videos (Chau,

2010). Videos and comments to those can be easily

added and circulated amongst peer groups. And one

of the main categories of videos he identiﬁed was the

’informal mentorship’ in the form of how-to videos.

In other domains, there is some research on

using YouTube as an educational channel. In

medicine, publishing videos on clinical skill train-

ing on YouTube instead of peer-reviewed services has

been investigated, in hopes of attracting a consider-

ably larger audience (Topps et al., 2013). They found

the results promising and the number of views encour-

aging, but naturally assessing educational impact is

difﬁcult with anonymous viewers.

Link Between Gaming Communities in YouTube and Computer Science

3 METHODS

To gain an understanding on how gaming videos and

learning programming are connected we used a mixed

method approach. We ﬁrst collected and categorized

videos based on their content to gauge the prevalence

of game videos that related to programming or com-

puter science in some way. After this step, we se-

lected two videos and used a grounded theory based

analysis to have a more detailed picture of the phe-

nomena. This section ﬁrst describes the classiﬁca-

tion of the videos and then explains the method of the

comment analysis in more detail.

3.1 Classiﬁcation of Videos

Video categorization began by both authors reviewing

together an initial set of 50 gaming videos selected us-

ing the ’programming’ query. Through discussion of

the titles and content of the videos, ﬁve categories of

gaming videos that pertain to programming were cre-

ated. This categorization scheme was used to catego-

rize a larger sample of videos. The reliability of this

categorization was tested by evaluating inter-rater re-

liability. The ﬁve categories to classify the larger set

of videos were:

• Gaming enhanced by programming Using pro-

gramming and/or computer science concepts in-

side a speciﬁc game, to improve gameplay

• Game programming tutorial General tutorials

for game programming, teaching game develop-

ment whilst not being related to any speciﬁc game

• Modding tutorials Tutorials teaching how to cre-

ate a modiﬁcation (mod) for a speciﬁc existing

game

• Game programming discussion Discussions on

general level what it means to be a game program-

mer and what the ﬁeld is like

• Other Unrelated videos or videos that did not ﬁt

any of the above categories

We used the query ‘programming’ and gathered

the ﬁrst 400 videos for classiﬁcation. Whilst search-

ing we used the default parameters available (sorted

by relevance and no limits to when the video was up-

loaded). However, it should be noted that the informa-

tion on how exactly YouTube orders the search results

is not available.

Both authors independently labeled the 400 pres-

elected videos. After labeling, we resolved any con-

ﬂicts through discussion. Consensus and the results

of the categorization are presented in Section 4.1.

3.2 Grounded Theory and Comments

In addition to looking at the videos as a whole, we an-

alyzed the discussions in the comment sections to pro-

vide a richer description of the phenomena. To do this

we adopted a grounded theory approach. Grounded

theory as a method aims at providing a rich descrip-

tion and explanation of a phenomenon and it is also

useful for generating hypotheses. At the core is the

idea of building a theory that is grounded in the

gathered data. The procedure of grounded theory is

to analyze the data by ﬁrstly coding it openly and

then returning to the codes and doing axial coding.

In axial coding, the codes are categorized to extract

themes and concepts of interest. For a discussion on

grounded theory speciﬁcally in computing education,

see (Kinnunen and Simon, 2010).

We started our open coding by selecting four

videos from the games enhanced by programming

category and one video from the modding category.

The selection of these two categories highlights the

informal setting we are investigating. These types of

videos enable a diverse type of content where learning

is incidental to the task. The initial open coding re-

sulted in ten different categories for comments, such

as ‘joke’, ‘reﬂect’, ‘question’ etc. Whilst this open

coding did describe the contents of the discussions it

did so in very broad terms. However, it did not high-

light the importance of discussing computer science

and programming concepts which were common in

the comments.

To improve the coding and produce more useful

axial coding, we selected two videos where program-

ming and games were mixed and that had a large

number of comments. The key difference this time

was that we only included comments that were di-

rectly related to computer science or programming,

thus enabling us to focus on coding the relevant con-

tent. Based on the original set of codes and the codes

generated during this step, we created the axial coding

that is presented in Section 4.2.

After the open and axial coding we had the key

themes of the discussions, or more precisely in our

case we had identiﬁed different types of discussions.

The third analysis phase of grounded theory is the se-

lective coding, where a core category is selected that

describes the phenomena and the rest of the categories

are integrated and reﬂected in light of this core cat-

egory. However, since the categories of discussions

that we found were so different from each other we

decide to forgo this step and focus on providing a de-

scription communication in the comments.

CSEDU 2017 - 9th International Conference on Computer Supported Education

4 RESULTS

4.1 Categorizing Game Videos Related

to Computing

Both authors coded the list of 400 videos individu-

ally, whenever the title of the video clearly showed

the category it belonged to it was used. In the cases,

where the title was not enough to classify the video,

the content of the video was reviewed. After the indi-

vidual coding (Cohen’s κ = 0.619) the conﬂicts were

resolved through discussion. Table 1 summarizes the

distribution of the videos.

Table 1: The distribution of the videos.

Category # %

Games enhanced by Programming 38 9.50

Game programming tutorial 242 60.50

Modding tutorials 7 1.75

Game programming discussion 56 14.00

Other 57 14.25

As can be seen in Table 1, the majority of the

videos are in game programming tutorials which usu-

ally come in longer series having multiple videos.

However, almost tenth of the videos were about com-

mercial games meant for entertainment (not serious

games) that incorporated programming in some form.

4.2 Discussions Related to The Videos

To provide a richer description of the gaming commu-

nity intersecting with computer science, we looked at

the discussions in the comment sections of two differ-

ent videos. The videos were chosen to highlight two

different types of interaction with computing. Any

of the popular videos could have been selected but

we felt that these two videos provided a good variety

in both discussion and the content of the video itself.

The content of the selected videos are detailed next.

V1 BASIC Interpreter in Minecraft

The ﬁrst video chosen is called “BASIC Program-

ming Language in Minecraft” from the channel Seth-

Bling, seen in Figure 2. At the time of writing (July

2016), the video has garnered over half a million

views and hundreds of comments. The running time

of the video is 13’45” and it focuses on how the author

implemented an interpreter for BASIC programming

language inside MineCraft.

The ﬁrst part of the video focuses on how the in-

terpreter was created in the game and how to use it to

Figure 2: BASIC Interpreter in Minecraft.

run programs. The author showcases the interpreter

by writing a simple function that calculates whether

a number is prime and also touches some advanced

topics, such as abstract syntax tree, the difference be-

tween compiled and interpreted language. The video

also contains a section with the author explaining how

this BASIC interpreter can be used to control a turtle

inside MineCraft, demonstrated by a loop with com-

mands for the turtle to move in a particular pattern and

mine a route through the virtual blocks of MineCraft.

V2 Programming Rockets in Kerbal Space

Program

Figure 3: Programming Rockets in Kerbal Space Program

Link Between Gaming Communities in YouTube and Computer Science

In the second video (seen in Figure 3) the author, Scott

Manley, describes the usage of kOS mod for Kerbal

Space Program. The game consists of designing and

ﬂying rockets and kOS is a modiﬁcation for the game

which brings a programmable computer that can be

used to automate tasks. The total length of the video

is 8’11”, and at the time of writing (July 2016), it has

gathered over 120 thousand views and more than 400

comments.

The video showcases a simple script written for

the kOS that guides the rocket to an orbit around the

planet. First, in the video, the rocket is launched. Af-

ter a while, it reaches the highest point in the ballis-

tic trajectory, and the script takes control of the ship.

The script then ﬁnalizes the launch by guiding the

rocket into a stable orbit. The contents of the script

are shown in the video and the various lines of codes

are explained as they happen. In the scripting lan-

guage various commands are given that wait until cer-

tain conditions (e.g. high enough altitude) are met and

then another command is given (e.g. turn the rocket to

face a particular direction or turn the engines on). The

author also talks about how the mod can be used for

more complex things such as automated rovers and

discusses the similarities with the system and the use

of computers in real life during space missions.

Overview of The Discussion

After the open coding of the ﬁrst 350 comments

from both videos, we started axial coding and found

that the majority of comments abstracted to three

categories: programming languages, efﬁciency, and

learning experience. We use V1 and V2 to denote

from which video the comment came from. The num-

ber of relevant, pertaining to computer science, com-

ments for V1 was 139 (ca. 40%) and for V2 73 (ca.

21%).

From the comments, it is clear that some of the

commenters were already familiar with computer sci-

ence and programming. How big this population is

compared to those who are novices is hard to estimate,

and the only method for doing so is to infer from the

contents of the comments.

Programming Languages

Referring to Douglas Adam’s Hitchhiker’s Guide to

Galaxy, someone started a posted a comment: Now

we need a command block that can calculate the

answer to life the universe and everything. (V1).

This gathered several replies, where people showed

their knowledge and shared simple programs in many

languages to show how printing number 42 was

achieved, e.g. “ this is it in HTML <p>42 </p>”

(V1), “In scratch: When green ﬂag clicked: Say

42” (V1), “Here is in python print(42)” (V1), and

other commenters point out a way to replicate this in

multiple languages, including for example Java, and

JavaScript.

Since V2 featured a custom language created for

the game that is not in use elsewhere, it received

plenty of comments reﬂecting on learning a new and

highly speciﬁc language versus more general pro-

gramming languages. This was reﬂected in comments

such as: “I just love the idea of kOS, and would gladly

spend humongous amount of time playing with it...

but, i also hate the programming language itself. It

would be so awesome to see something like kOS which

is using JavaScript/Python/Ruby/C# .. or any sensible

programming language out there (thus object orien-

tation would probably be very useful feature)” (V2)

and “Yeah, a C or Python (maybe even Lua) version

would have been nicer, but basic isn’t hard to learn, it

is very basic” (V2)

Some of the commenters also pointed out the lim-

itations of the Kerboscript (the name of the language

used in the kOS mod): “Yeah I really wish they had a

more mainstream language, but then again it’s still a

beta. We might see something like that later hopefully,

having functions and classes would be very cool.”

(V2). This, in turn, prompted a discussion on how lan-

guages work on a lower level: “Whatever program-

ming language and style you like, it still gets turned

into machine code which doesn’t inherently support

functions and classes” (V2).

Overall, the discussion on both videos was fo-

cused on the authenticity of programming languages.

This was especially prevalent on V2 since the video

portrayed a custom programming language that was

then used to complete ‘real’ programming tasks.

Efﬁciency

Much of the discussion related to computer science in

comments to V1 revolved around the topic of algo-

rithmic efﬁciency. In the video, the author presented

a simple ISPRIME function which computed whether

a number was prime in a suboptimal way. This was

pointed out in many comments, often with sugges-

tions on how to improve, for example: “Instead of

X-1 wouldn’t 1/2 X be a better cutoff point for ﬁnding

primes? (Since no number can be evenly divided by

a number greater than 1/2 itself?) That would save

some cycles, especially as the numbers get higher.

...” (V1) , and “Yeah, computing the square root of

something is usually ’expensive’, but dividing by two

is usually cheap. It’d be an easy optimization.” (V1)

As a classic algorithm efﬁciency problem, some

experienced users provided the best cutoff helping

CSEDU 2017 - 9th International Conference on Computer Supported Education

to improve the author’s solution. However, some

users also reﬂected about the problem and discussed

the methods. In V2, efﬁciency discussion centered

around certain parameters in the code, which some

users felt could be optimized further.

Learning Experience

The third and perhaps the most interesting category is

the discussion on learning experience regarding pro-

gramming. One long thread in the comments involved

the use of these types of games and their affordances

in teaching younger population: “Does this have the

potential to help kids learn to program? IE if it was

directly applicable in the world like you showed the

turtle? :-)” (V1)

This spurred many replies, giving suggestions on

how to start learning programming using various ap-

proaches, such as game modding: “Could try learn-

ing to mod the game; modding can range from need-

ing almost no programming experience to advance

programming, depending on how far you take it.”

(V1) and watching programming tutorials on video:

“You can watch video tutorials though this is the

hardest way to learn programming.” (V1)

The different tools and aspects of learning pro-

gramming were discussed as well. With one com-

menter noting that block-based programming has

gained popularity: To be honest, logical-block-style

programming is taught very very early in some places

now. Because it teaches logic without needing to

learn a programming language syntax.” (V2). There

were thoughts also on the difﬁculties of using profes-

sional tools when starting out in programming: Well,

the main reason it was hard for me to get into pro-

gramming was because the IDE was made for more

experienced people and was packed with features ...

(V1)

Other Observations

We initially suspected, based on the literature (Ra-

malingam et al., 2004), that there would be nu-

merous comments portraying poor self-efﬁcacy when

confronted with programming and computer science.

However, to our surprise these comments were almost

completely missing, with only one concrete example:

“Now i feel stupid ..... This is insane! :D Wow” (V1).

While it is difﬁcult to verify, we suspect that there are

more feelings of poor self-efﬁcacy but those individ-

ual just do not comment, or even tune out the video as

soon as something complex is presented.

5 DISCUSSION

In this article, we have examined the intersection of

computer science and computer games in an informal

setting. We approached this by describing the gaming

content that includes programming as well as discus-

sion related to it. While there has been many different

approaches to incorporate games into classrooms, es-

pecially programming ones, very little discussion has

taken place regarding informal learning from games.

We approached this phenomenon through analyzing

gaming videos that relate to programming – a popular

way of disseminating content and ideas in the gaming

community.

Looking at different kind of gaming videos that

relate to computer science (RQ1), we discovered a

plethora of programming-related videos associated

with a variety of different games. Additionally, we

sought to provide a more comprehensive picture of

the phenomena by analyzing the discussions on these

videos (RQ2). We conclude that a fair portion of the

discussion is relevant to computer science and pro-

gramming. The vast majority of programming re-

lated discussion can be described to belong to one

of the three categories: programming languages and

their capabilities, efﬁciency in algorithms and pro-

grams, and the experience of learning programming

and computer science.

Incorporating computing and programming edu-

cation into the K-12 curriculum has been heavily de-

bated and discussed all over the world. At the same

time, games and the whole gaming culture have risen

to be a massive cultural phenomenon. And it at least

seems that the number of games that contain pro-

gramming within them is increasing. These games,

where programming is a way to interact with the game

world, could provide a rich and engaging environment

to learn programming. Because the need to program

arises from a problem within the game, they provide

a motivating environment to learn to program.

However, based on the data, we have no way of

knowing who the commenters are and what is their

previous experience with computer science. Simi-

larly, based on the discussion it is hard to tell how

much actual learning is going on, and whether any of

the possible learning is transferable to other contexts.

As we have shown, some of these exchanges

are relevant to learning programming and computing

skills. Given that, so many, especially young, people

do play these games, we argue that the informal side

warrants more discussion and research. We hold the

view that it is crucial to understand when, where, and

how the ﬁrst contact with programming happens.

Link Between Gaming Communities in YouTube and Computer Science

5.1 Future Work

In the future, we plan to do a more detailed investiga-

tion of the users involved in this kind of informal way

to learn programming through gaming videos. Since

it is the background of users’ is not available and the

fact that most of the comments are very short, the phe-

nomena is difﬁcult to analyze using methods applied

to other discussion contexts (e.g. forums). Develop-

ing methodology to investigate informal learning in

general is required.

Additionally, it would be interesting to ﬁnd out

whether similar phenomena is happening in other sub-

jects. While it was not the focus of this research, we

did note a portion of the discussion related to physics

happening in the comments of V2.

ACKNOWLEDGEMENTS

Rodrigo Duran was supported by CNPq–Brazil.

REFERENCES

Boyle, M. and McDougall, A. (2003). Formal and infor-

mal environments for the learning and teaching of

computer programming. In Proceedings of the 3.1

and 3.3 working groups conference on International

federation for information processing: ICT and the

teacher of the future-Volume 23, pages 11–12. Aus-

tralian Computer Society, Inc.

Chau, C. (2010). Youtube as a participatory culture. New

directions for youth development, 2010(128):65–74.

Deterding, S., Dixon, D., Khaled, R., and Nacke, L. (2011).

From game design elements to gamefulness: deﬁn-

ing gamiﬁcation. In Proceedings of the 15th inter-

national academic MindTrek conference: Envisioning

future media environments, pages 9–15. ACM.

Entertainment Software Association (2015). Essential

facts about the computer and video game industry.

Available Online: http://www.theesa.com/wp-

content/uploads/2015/04/ESA-Essential-Facts-

2015.pdf.

Esper, S., Foster, S. R., Griswold, W. G., Herrera, C., and

Snyder, W. (2014). Codespells: bridging educational

language features with industry-standard languages.

In Proceedings of the 14th Koli Calling International

Conference on Computing Education Research, pages

05–14. ACM.

Hayes, B. and Games, I. (2008). Making computer games

and design thinking: A review of current software and

strategies. Games and Culture.

Kelleher, C. and Pausch, R. (2005). Lowering the barriers to

programming: A taxonomy of programming environ-

ments and languages for novice programmers. ACM

Computing Surveys (CSUR), 37(2):83–137.

Kinnunen, P. and Simon, B. (2010). Building theory

about computing education phenomena: a discussion

of grounded theory. In Proceedings of the 10th Koli

Calling International Conference on Computing Edu-

cation Research, pages 37–42. ACM.

McCartney, R., Boustedt, J., Eckerdal, A., Sanders, K.,

Thomas, L., and Zander, C. (2016). Why comput-

ing students learn on their own: Motivation for self-

directed learning of computing. ACM Transactions on

Computing Education (TOCE), 16(1):2.

Ramalingam, V., LaBelle, D., and Wiedenbeck, S. (2004).

Self-efﬁcacy and mental models in learning to pro-

gram. In ACM SIGCSE Bulletin, volume 36, pages

171–175. ACM.

Saito, D., Washizaki, H., and Fukazawa, Y. (2015). Work

in progress: A comparison of programming way:

Illustration-based programming and text-based pro-

gramming. In Teaching, Assessment, and Learning for

Engineering (TALE), 2015 IEEE International Con-

ference on, pages 220–223. IEEE.

Steinkuehler, C. and Duncan, S. (2008). Scientiﬁc habits of

mind in virtual worlds. Journal of Science Education

and Technology, 17(6):530–543.

Thompson, C. (2016). The Minecraft Genera-

tion. The New York Times. Available online:

http://www.nytimes.com/2016/04/17/magazine/the-

minecraft-generation.html.

Tillman, N., Halleux, J., Xie, T., and Bishop, J. (2014).

Code hunt: Gamifying teaching and learning of com-

puter science at scale. L@S ’14 Proceedings of the

ﬁrst ACM conference on Learning @ scale confer-

ence, pages 221–222.

Topps, D., Helmer, J., and Ellaway, R. (2013). Youtube as a

platform for publishing clinical skills training videos.

Academic Medicine, 88(2):192–197.

Wallace, S. A., McCartney, R., and Russell, I. (2010).

Games and machine learning: a powerful combination

in an artiﬁcial intelligence course. Computer Science

Education, 20(1):17–36.

Zorn, C., Wingrave, C. A., Charbonneau, E., and LaVi-

ola Jr, J. J. (2013). Exploring minecraft as a conduit

for increasing interest in programming. In FDG, pages

352–359. Citeseer.

CSEDU 2017 - 9th International Conference on Computer Supported Education