Design and evaluation of an interface prototype for content
selection for a game that supports user-generated content
ZEHUA SHAN
KTH ROYAL INSTITUTE OF TECHNOLOGY
SCHOOL OF COMPUTER SCIENCE AND COMMUNICATION
content selection for a game that supports user-generated content
Swedish title
Design och utvärdering av en gränssnittsprototyp för innehållsval för ett spel som stödjer användargenererat
innehåll
Zehua Shan
zehua@kth.se
DA223X Degree Project in Computer Science and Communication, Second Cycle, 30 credits Degree Programme in Information and Communication Technology, 300 credits
Master’s Programme, Interactive Media Technology, 120 credits KTH Royal Institute of Technology
School of Computer Science and Communication (CSC) Supervisor at CSC was Helena Tobiasson
Examiner at CSC was Jan Gulliksen 20170619
beyond creating and sharing user’s creations is experiencing experiences created by others. Designing such an interface that support the user to do so poses many challenges. What do the users need? How should the content be displayed? What do users look for when trying to find something interesting? Those are just a few of the questions that need to be answered when creating a solution for such an interface. The question this paper tries to answer is:
What are the main elements for a content selection interface that fulfill the needs of it’s users and meets their expectation for a game that supports vastly different gameplay experiences?
A survey was sent out to get a grasp on what the users needs, with the results from survey and knowledge gained from the background study, an interface prototype design was developed. To get a reality check from potential users, the prototype was evaluated with user testing, and feedback from users was gained from semistructured interviews. Result suggests that the proposed set of function and information available are sufficient for most user, and the structure was clear. However, there are some features that the users wish to see, which could prove useful in helping users find interesting content. The knowledge gained from the study resulted in a prototype and some suggestions for future work regarding similar projects.
SAMMANFATTNING
Spel som stödjer en hög grad av anpassning är få, och deras framgång varierar, men det är en tydlig trend att fler och fler spel strävar efter att erbjuda spelare mer anpassningsbar spelupplevelse. En stor del av den anpassningsbara upplevelsen utöver att skapa och dela användarens skapelser är att uppleva upplevelser som skapats av andra. Att utforma ett sådant gränssnitt som stödjer användaren att göra det ställer många utmaningar. Vad behöver användarna? Hur ska innehållet visas? Vad letar användare efter när man försöker hitta något intressant? Det är bara några av de frågor som behöver besvaras när man skapar en lösning för ett sådant gränssnitt. Frågan som den här studien försöker svara är:
Vilka är huvudelementen för ett innehållsvalgränssnitt som uppfyller användarnas behov och uppfyller förväntningarna för ett spel som stöder väldigt olika spelupplevelser?
En undersökning skickades ut för att få tag på vad användarna behöver, med resultaten från undersökningen och kunskapen från bakgrundsstudien utvecklades en gränssnittsprototypdesign. För att få en verklighetskontroll från potentiella användare utvärderades prototypen med användartestning, och feedback från användarna samlades genom halvstrukturerade intervjuer.
Resultatet tyder på att den föreslagna uppsättningen funktions och information tillgänglig är tillräcklig för de flesta användare, och strukturen var tydlig. Det finns dock vissa funktioner som användarna vill se, vilket kan vara användbart för att hjälpa användare att hitta intressant innehåll. Kunskapen från studien resulterade i en prototyp och några förslag till framtida arbete avseende liknande projekt.
Design and evaluation of an interface prototype for content selection for a game that supports user-generated content
Zehua Shan
KTH Royal Institute of Technology zehua@kth.se
ABSTRACT
Games that support a high level of customization is few, and their success varies, but it is an apparent trend that more and more games strive to offer players more highly customizable gaming experience. One huge part of the customizable experience beyond creating and sharing user’s creations is experiencing experiences created by others. Designing such an interface that support the user to do so poses many challenges.
What do the users need? How should the content be displayed?
What do users look for when trying to find something interesting? Those are just a few of the questions that need to be answered when creating a solution for such an interface. The question this paper tries to answer is:
What are the main elements for a content selection interface that fulfill the needs of it’s users and meets their expectation for a game that supports vastly different gameplay experiences?
A survey was sent out to get a grasp on what the users needs, with the results from survey and knowledge gained from the background study, an interface prototype design was developed. To get a reality check from potential users, the prototype was evaluated with user testing, and feedback from users was gained from semistructured interviews. Result suggests that the proposed set of function and information available are sufficient for most user, and the structure was clear. However, there are some features that the users wish to see, which could prove useful in helping users find interesting content. The knowledge gained from the study resulted in a prototype and some suggestions for future work regarding similar projects.
Categories and Subject Descriptors
I.2.1 [Computing Methodologies] Application and Expert Systems – Games
General Terms
Theory
Keywords
Game, HCI, User Experience, UserGenerated Content, Game Interface Design.
1. INTRODUCTION
In author's experience, most firstperson PC games on the market today has predetermined conditions, generating streamlined and monotonous gameplay[1]. The nature of those games means that the content selection is simple since the game could only play in a certain way. For different gameplay or experience, players would simply have to choose another game that suits their needs better or modify the game(mods).
To keep gamers around, it is becoming a trend for game development companies to develop game as platform or services with more to experience for its users, rather than
singular experiences, as several journals describes[2][27][28].
One model of achieving this is through delivering Downloadable Content(DLC), and the other model is building game with great customization support, delivering more content through community participation. There are some cases of this approach in the industry[21], as some described[29][30], Minecraft is a great example of this. Those games that do support the high level of customization, however, mostly relies on community solutions for sharing of customized gameplay content via forum or websites. Browsing for community websites requires spending time on finding interesting custom game modes on other platforms which might not be known to the user, thus creating a gap between users and the content. For a more convenient experience with higher usability, it is in the industry’s interest to integrate content sharing solutions into the game itself.
Seeing this trend, a game development company where the author conducted this study, started a project which set out to explore the possibilities for creating an environment for personalized gaming experience with a focus on first person games, whether it is a firstperson shooter or a firstperson sports game. The project revolves around creating a game, with the goal that the game content could be adapted by the user in a simple way, also that the user's personalized experiences could be shared easily with players using the same game.
Another part of sharing the experience is for others to find and enjoy the experience shared. With the tools that the game will provide to its users, the users could create everything from a puzzle game to a Massively Multiplayer Online RolePlaying Game(MMORPG); the user has to be able to find game content according to their need. Creating an interface that supports the user to do so poses many challenges, what do the users need?
How should the content be sorted? What do users look for when trying to find something interesting? Those are just a few of the questions that need to be answered when creating a solution for such an interface.
This paper focuses on investigating what features or functions, that are helpful for users for finding interesting game content, and what information do users base their decision about whether to try a new game or not. Lastly, to investigate whether the proposed structure of these features and information is intuitive for users, in the form of evaluation of a design prototype. The investigation is mostly conducted from a content consumer point of view, not from a content creator point of view.
1.1 Game Modification
Game modification, or "mod" for short, is the alteration of the content of a video game. Mods are made by the game's end users or sometimes even game developers, and could range from simple assets to entirely new games, including new maps, units, textures, weapons, items, sound, story line and so on. In this project, the mods are considered as whole new games, and the focus lies on how to present the mods to other users in the
best way. Also, sales aspect of user creations is not taken into consideration during this study.
2. BACKGROUND
In the following section, we will first explore motivation for play and motivation for contributions, since players and content creators or modders are the key user groups of this interface.
Afterward, we will look into related studies regarding usergenerated content and usability design guidelines. Lastly, a few of the related work will be presented, to get a grip of the state in the game industry..
2.1 Motivations
Like any product, for the product to be successful, it is important that designers understand the need of its users.
Although the studies regarding player's motivations are mostly around gameplay experience and are more related to game design rather than interface design at first glance, the interface designer must understand the player's psychological needs while designing interface, to better support player's goals.
Similarly, understanding contributor's motivation is also necessary, for the designer to support contributor's need through interface design.
In an attempt to understand the motivation for computer gameplay, Ryan et al. [3] conducted four studies with online multiplayer gaming communities applying selfdetermination theories. A total of 927 participants were recruited for those studies, and the methods were mostly based on questionnaires, although play tests in laboratories were also conducted in combination in the first three studies. Results suggest that motivation of play rests on the game's ability to satisfies the player's basic psychological needs, namely Autonomy, Competence, Relatedness[4]. Furthermore, Yee conducted a study regarding motivation for playing online games [5]. Data was collected through questionnaires, which was sent out to 3000 Massively Multiplayer Online RolePlaying Game (MMORPG) players. The results indicate that the player's motivation for play could be categorized into three main categories; Achievement, Social, and Immersion. The three main categories consist of 10 subcomponents in total, for example, advancement, mechanics, and competition are subcomponents of Achievement.
Sotamaa [6] studied the modder community around a FirstPerson Shooter game called Operation Flashpoint, about the modders motivation, potential commercialization of their work and their notion of ownership. He suggests that the five main motivations of modding are Playing, Hacking, Researching, Artistic expression and Cooperation. Postigo [7], whom also studied the modder community, identified three central motivations for modding PC games. Firstly, hobbyists that consider modding as a way to creatively and artistically express themselves. Secondly, modders that want to increase their enjoyment of the game and lastly, modders that see modding games as a way of showing off their skills, seeing it as a channel to potentially acquire a job in the game industry.
2.2 UserGenerated Content(UGC)
To handle the user generated gaming experiences, a content selection system that can handle a vast variety of content is needed. A connection could be found between the game's content selection process in this study and currently available online Video On Demand(VOD) services, in the sense that those services also are built with handling the vast variety of content in mind. Cha et al. [8] conducted a study regarding the
popularity cycle of videos on YouTube and other similar UGC video services. In this study, the authors made distinctions between UGC and nonUGC videos and their implications.
Comparing NetFlix and YouTube or similar services, there are clear differences in the contents production cycle and how videos are found by its users. Similarly, usergenerated game content could also be differentiated from the full titles released by game production companies.
Ghosh and Hummel [9] analyzed two mechanisms that use viewer ratings to rank content, to explore which of those two mechanisms incentivizes for higher quality contribution and generates better viewer welfare. One of which is the rankorder based mechanism, where contributions with higher rating are placed on the top of the page, and lower rating contributions on the lower side of the page in falling order. The other mechanism is a proportional mechanism, which places contribution depending their proportion of positive rating. The result suggests that the rankorder mechanism almost always incentivizes higher quality contribution, and thus contributes to better viewer welfare.
Borghol et al [10] studied on factors that impact YouTube video popularity, by collecting and analyzing data regarding video statistics, historical view data, and influential events. When controlling for video content, they found that the total view count is the biggest factor contributing to video popularity, observing a strong "richgetricher" behavior, except for newly uploaded videos. For newly uploaded videos, they found that uploader brand and number of keywords or tags to be the biggest factor impacting video popularity.
Zhou et al [11] conducted a measurement study on a large dataset of YouTube videos, to explore the recommendation system's impact on video views. They crawled the YouTube website to study how videos were discovered by users and what the major sources of the drive are for a video to get views.
They found that the most views are generated through YouTube Search, and the second major source of views comes from Related Videos. Related Videos is the source of 30% of all the generated views, only second to YouTube Search by a small percentage. Also, the evaluation shows that the existence of Related Video recommendation helps to increase the diversity of video views by helping viewers discover videos of their interest, rather than only showing the popular videos.
2.3 Design
As the study is about designing is a graphical interface for a game, general usability aspects, and gamerelated design aspects should be taken into consideration. Lu HsiPeng and Hsu ChinLung conducted a study applying technology acceptance model [12] incorporating social influences and flow experience [13] as parameters, to predict users’ acceptance of online games. The model they suggested was evaluated using data collected from surveys, which was sent out to 233 users.
Some findings from the study indicate that flow experience is important since users intend to play online games to be fully immersed, thus increasing usability through social interaction, good navigation easy access is essential for successfully manage an online game community. Furthermore, Wiles and Johnson [14] conducted a study about whether computer games design comes in conflict with the widely accepted interface design heuristics [15]. One example of which, when game menus generally have an attractive appearance with fancy animations and eyecatching colors, sometimes, this comes at the expense of readability and functionality. Anytime the user is
forced to interact with the menu with such characteristics; it may cause frustration and thus break the flow experience.
Nielsen et al. [16] conducted a study about user’s reading pattern on web solutions using eye tracking. They recorded how 232 users scanned the content of thousands of website.
Results reveal a dominant reading pattern that resembles an “F,”
and the pattern contains three components. They noticed that most of the users would first read in a horizontal movement along the top of the screen, afterward, users will move lower down the screen and scan horizontally again. Finally, users will usually scan the left side of the screen vertically; this is the last element of the “F” pattern.
2.4 Related Work
There are some games or platforms that supports usergenerated content currently in use. In the following section, a few of those systems will be presented.
Dota 2 Arcade is a hub for community made maps in the game of Dota 2 [17]. Players are presented with the most played custom maps and can create their own lobby using those maps.
Beyond most popular maps, players could also see the open lobbies that others have been created by others, or browse maps in a timeline, for example, most popular custom maps in the last month, or the last week. ( see figure 1 ).
Figure 1, Dota 2 Arcade main page.
Figure 2, Starcraft 2 Arcade main page
Like Dota 2 Arcade, Starcraft 2 Arcade[18] is a hub for community made maps. With a large variety of custom made maps, some of the custom maps fundamentally changes the gameplay experience. It offers similar functions as Dota 2 Arcade, allowing players to browse popular custom maps, create an own lobby with the custom maps, or join open lobbies created by others. (see figure 2).
Services like Dota 2 Arcade and Starcraft 2 Arcade are all signs that the players want more out of their game, they like to try out new gaming experiences. To maintain the longevity of a gaming community, a major part of it is to be able to deliver new gaming experience, and usergenerated content could be the answer.
2.5 Research Question
The goal of this master thesis project is to to find out the important aspects of designing an interface that supports UGC, through design and evaluation of an content selection interface prototype. With knowledge from the motivation of gamers and modders alike, usergenerated content, the use of interaction design methods and related work try to answer the research question:
What are the main elements for a content selection interface that fulfill the needs of it’s users and meets their expectation for a game that supports vastly different gameplay experiences?
3. METHODS
A process similar to a UserCentered Design(UCD)[19]
process was used during the development of this project ( see figure 4 ).
Figure 3, a usercentered design process
3.1 Context of Use
The first step in a UCD process is to identify whom the system will be used by, what they will use it for and under what kind of conditions. This was specified together with the team at the company.
3.2 User Requirements
The second step in the UCD process was to define the user requirements; this was done by sending out a survey to potential end users[32] and meetings with project manager and development team at the company. The survey was created to find out what the user’s value in a graphical interface in games, their earlier experience with other game interfaces, and to determine the usefulness of certain functions to be introduced in this project. A pilot test for the survey was conducted to ensure the survey generates enough variation of answers. Meeting with the project manager and development team was carried out to discuss the technical limitations and possibilities with such system if it would be deployed.
3.3 Design Solution
The next step in the UCD process was to design the prototype.
First. A lowfidelity mockup was designed, and was discussed internally with the development team. Based on the feedback, a highfidelity prototype was constructed with InVision prototyping tool. The prototype includes all of the function and information that are intended to be included in the final product.
However, functions are not fully functioning, for example, the search function is not available in the InVision prototype, due to the limitations of the prototyping tool. Using such a prototype for testing in early development stage could discover the obvious usability problems, and show test users a good preview of how the system should work. This could also potentially prevent problems that would prove costly to correct later in the development phase.
3.4 Evaluation
The last step in the UCD process was user evaluation of the design. A round of user evaluation was conducted with a total of 10 test participants to reach a balance between time consumed and coverage of result [20]. The user evaluation was based on a set of tasks that the user had to carry out with the interface[31]. Also, thinkaloud was encouraged, to potentially be able to capture what the users were thinking or feeling [33].
A pilot test for the user evaluation was carried out to see whether the evaluation was sufficiently designed. In addition to this, the test session was recorded with screen capture software, and a semistructured interview was carried out after the test session, which was recorded and transcripted later on.
4. REQUIREMENTS
In this section, result from the survey and takeaways from the meeting with development team will be presented, as a basis for the design.
4.1 Survey
A survey was created with the purpose to find out what the user’s value in a graphical interface in games, their earlier experience with other game interfaces, and finally, to see how users want to find new gaming content. The survey was pilot tested, to see whether the designed questions would collect interesting data. A few questions were edited, and a couple of questions was later added. The survey was sent out via Facebook and Discord, and a total of 60 answers was received.
The survey was divided into three sections; What information users want when looking for new games, “mods” related questions and finally, user experience related questions and comments.
In the first section, questions were asked regarding recommendations of games and what kind of information users look for while deciding to play a new game. Results show that 60% of the participants are interested in games that are similar to the games they already play, and only 5% of participants are not, rest of the participants (35%) answers that it depends.
When asked whether they often check top recommended list of games on platforms like Steam, 45% of the participants answered no, with 26.7% of users answering “Yes”, and the rest answered “Sometimes”(28.3%). Regarding what kind of information users generally look for when deciding to play a new game, the following main source of information was identified: “Gameplay videos” (83.3%), “Friend’s reviews”
(78.3%), “How many or which of your friends are playing”(68.3%) and “User reviews” (55%), as seen in diagram 1. Finally, when asked whether the participants would try a new game without knowing anything about the game, 11.7%
answered “Yes”, and 65% answered “Yes, if it is free”, with the remaining 23.3% answering “No.”
In the second section, participants were asked regarding their view on “mods”. Less than half of the participants have played
“mods”(48.3%). When asked how they find the “mods”, most participants answered “community website” or through Google Search.
In the final section, questions regarding userinterface experiences participants were presented. Participants were to rate how important the fluidness of user interface in games are versus aesthetics, on a scale of 15, where one is not important at all and five is very important. Most of the participants rated 4 or higher in fluidness(81.6%), in comparison, aesthetics received 63.3% ratings of 4 or higher. Finally, participants were asked to give examples of any inconvenience in userinterface in games; the results can be categorized as following:
● Interface do not utilize screen realestate or mouse and keyboard (console ports).
● Poor performance in menus (too much or heavy animations, lag, delay).
● Too much clutter.
● Too many levels of the interface (functions buried deep in menus).
● Incomprehensible structure.
4.2 Development Team Meeting
From the perspective of the company, the requirements that were stated were loosely defined. The interface should be easy to use for the end user, without a specified user group, since they want the product to appeal to as big as a crowd as possible. Also, the interface should support handling of a large amount of varying content. As for technical limitations, the team was very confident in their abilities; Therefore no specific technical limitations regarding the functions within the scope were stated.
5. DESIGN
With knowledge gained from background study and survey, a prototype was designed, as the next step of the UCD design process. In the following section, the design decisions made in different parts of the prototype are presented.
Figure 4, design prototype of the content selection interface
5.1 Core Functionalities
As the main goal of this interface serves to help users find interesting gaming content in a game that supports vastly different gaming experiences, many aspects should be taken into consideration.
5.1.1 Friends
Socializing is a big part of gamer’s motivation of play[3][5][21], this is confirmed by results from the survey. To help users socialize, users can see what their friends are playing. While browsing for new gaming content, users could see if any of their friends are playing or have played the game they are looking for.
5.1.2 Categorization and sorting
One of the most important aspects is how the content could be accessed. With a massive pool of usergenerated content in mind, the content must be sorted in a sensible way, so that the users could find the content they need, without the need of scrolling through a giant everlasting list. Categorizing content by genre is one of many ways to combat this, by dividing content into different genres, it is easier for the users to find the appropriate content which could suit their motivation or needs.
Also, by allowing users to combine different genres, this might increase the likelihood of users getting the content they are looking for, As a content creator, the content creator is responsible for correctly categorizing their content. Also helpful is a search bar, so that the users could search for things they might be interested in straight away.
Beyond filtering by genre, having recommendations, recently created and trending as categories could also help users find
new interesting contents. Recommendations should help users by suggesting content based on genres they have recently played, and trending should base its suggestions on amount of players currently playing the game. Seen from a survey conducted by YouTube[22], Users from different regions may have different trending games, therefore filtering by region could also be helpful in enriching user's experience. To further sort the results after the users have selected a genre or category, sort by date of creation, the rating of the content and player count could also be helpful tools.
5.1.3 Similar games
As seen from the survey, most users are interested in similar games that they already play. It is logical to have a section which offers user suggestions of similar games in the content selection interface, which not only will be able to generate a broader range of experiences for the user, also could potentially create more exposure for the content created by other content creators.
5.1.4 Review and rating
Being able to give a review or read a review for contents are important and are a common practice on digital platforms, ditto for games[23] since many choices made when looking for new games are either based on gameplay videos or reviews. A simple rating based thumbs up and thumbs down are also added into the design; this allows the user to get an idea of the quality of the content quickly. Although it is up to the content creators for which kind of video they upload to promote their mods, it should be suggested to have a gameplay video uploaded as a user guideline.
Furthermore, a more dedicated review page where users could read and write more detailed reviews were created in the design.
In this page, a friend’s review will have priority, positioned on the top of the list. This is done according to the survey, as a friends review seems to have higher importance than a regular user review. Top rated reviews are to be displayed after friends review since they are most likely to be interesting for users beyond the reviews from friends. To counteract biased or reviews without proper experience with the content in mind, time spent on the content is shown alongside with the review.
Also, other users could rate the review, which should eliminate most of the unserious reviews.
In the sense that content creators are responsible for correctly categorizing their content, content that does not meet expectation due to low quality or abuse of categorization system would quickly fall off the top list due to poor rating.
5.1.5 Handling legal issues
Delivering UGC comes with other challenges. Since the end user is the main source of content contributions, and it is likely that they would not have the same production regulation or quality control, unlike game development company. In contrast, most game development companies need to get their products regulated for inappropriate content, before the products could reach the market. Another aspect of this is the potential use of copyrighted material. As an example given by a pc game journal[24], where a game developer broke the geneva convention. This could be solved with employees working with handling the contents that are affected by those issues, but due to the potentially large amount of content, a combination of internal handling and communitybased solution for handling this type of problems is more suitable. Therefore, a report function built into the content selection interface could potentially help manage the issue.
5.1.6 Random content
Finally, a function that randomly selects content for the user is added, after a user has played the content, the user will be prompted to rate the content. This serves to give the user a way to find content when users do not know what to look for, but just want to try out something new. Also, by prompting users to rate the content, it contributes to the general user welfare.
5.2 Information
The most important information a user needs when looking for a new game were placed into the design, with the goal of showing enough information that the user needs to get interested in a game content. This includes rating, review, similar games, review from friends and friends that are playing.
Name of the creator of a certain game content is placed alongside with the name of the content and rating, to give users a quick look at the essential information. (see figure 5) Also, showing the creator’s name up front could also help the creator to get recognition. In the case, that creators wants to acquire a job in the game industry, or that other user could recognize the name and know whether it is a game worth trying or not.
Figure 5, a “card” which shows the most important information of the game content. In this interface, every “card”
represents a game content.
Total voter count is shown alongside the rating in the detailed view, giving users a sense of the validity of the rating.
5.3 General Layout
Like any graphical user interface, having a good structure ensures good interaction. Making sure that the menu system is consistent, sensible and meets the user's expectation is important for the user experience.
5.3.1 The "F" scanning pattern
For providing a convenient browsing experience, the headings and content are placed according to the most common user eyescanning pattern. Placement of those roughly resembles the
"F" scanning pattern, as seen in figure 6 .
Figure 6, the rough resemblance of the "F" scanning pattern
Although the scanning pattern is described as a reading pattern for web content, it is however applicable on other type of user interface as well.
5.3.2 Menu depth vs. breadth
In a study regarding menu structure, where researchers compared the different depth and breadth configurations of the same content, the result suggests that the best result regarding accuracy and speed was achieved with broader and fewer levels of depth[25]. For example, eight choices at each of two levels design were far more superior in comparison to two choices at each six level design. This was taken into consideration while designing the prototype. In this design, for example, a user wanting to find a game content would have to navigate through two levels of menus with 5 or more choices at each level.
5.3.3 Nielsen's heuristics
Details in design were done with Nielsen’s heuristics taken into account. For example, with flexibility in mind, some of the interactions could be executed in different ways, achieving the same result. With consistency in mind, menu and submenus are designed to follow the same structure,
5.3.4 Prioritizing
As seen from the survey, users responds differently to the suggested functions, some are more important for them, some are less. This is reflected in the placement of functions, how they are displayed, and if they are displayed at all. For example, gameplay video, friend’s review and whom and which of their friends are playing, are the most important source of information for users, those are displayed upfront in the interface, while other suggestions mentioned in the survey are not.
5.4 Vision and Readability
One of the reasons to design an interface with black and white aesthetics is to help users with color vision deficiency to navigate through the interface. As the interface in this design is supposed to act as an overlay, and game or other content could be running in the background, a convenient way to ensure good readability is to use a black semitransparent background with white text on top, as seen in figure 7 . When the interface is ready to be implemented, there should be settings in place in the system to turn off the transparency, in case for performance issues or readability issues.
6. USERINVOLVEMENT
The research question concerns two problem areas. Firstly, if the structure of the information and functions meets the user’s expectation. Secondly, Are these information and functions adequate for users to find interesting games?
6.1 User Testing
In this section, findings from the user testing will be presented, to answer whether the structure is logical and meets the user's expectations.
Test participants recruited were all male, between age of 19 and 27. A set of 15 tasks was given to all 10 test participants(here on referred as participant #110), and the users were encouraged to thinkaloud. User testing was conducted with a MacBook Pro 13 inch version with Retina screen. When the participants had issue completing the task, starting to drift far away from the intended navigation route and wasn’t actively thinkingaloud, test participants were asked to explain what difficulties they experienced after the test session, which rarely happened.
The result of the user testing was positive, of the 10 participants, 5 participants had issue completing all 15 tasks, all 5 of those participants only missing one task at most.
Participant #1 and #5 couldn’t locate the “select random content” function, to which both participant later explained:
“I didn’t understand the question, but it made sense that it should be placed under Explore”
Participant #8 had difficulty finding the link to the gameplay video of the specific game, to which he replied:
“I did not see the gameplay video icon clear enough.”
Participant #2 had difficulty locating the report function, which users could use in case of usage of inappropriate materials or copyrighted materials in a certain game, to which he commented after the testing session:
“I have never reported a game, but I realized that it made sense being it was buried in a sub menu, it is not a function
that you would use very often.”
Lastly, Participant #3 had difficulty finding the “other similar”
games section, which is displayed on the same page when user selects a game, to which he later commented:
“I thought it would be found under the “Explore” page.”
Other than the before mentioned occasions, there were no other instances where the users were unable to interpret the menu system, or unable to locate the functions as given in the tasks.
More on whether the menu systems meets the user’s expectations will be explored in the interview section.
6.2 Interviews
In this section, findings from the semistructured interview will be presented. Firstly, what the user's experiences are with the prototype. Next, what the users think regarding the information and functions that are available in this prototype, to answer whether that information and functions are adequate for users to find interesting games or not.
As earlier mentioned, the participants are between 19 and 27 years old. Most of the participants are active PC users, where
most users play many different games, with only one participant being active gaming console user(PlayStation 4).
There are differences in how much they play. Some participants only play a few hours a week, whereas one participant works as a professional gamer, and plays 5 to 8 hours every day.
6.2.1 Experiences with the design
Participants were asked to describe their experience with the interface, to which 8 of the participants had positive feedback regarding the design of the interface. Most users thought the design was simple and intuitive; this is exemplified by a comment received from Participant #5 and Participant #9:
“I like the layout of it. The structure is logical and easy to follow, I also like the design in a sense that it was very easy to
see what belongs to which category. It's very easy to get an overview quick go to different categories or where you can click
and so I would say that it's good. I like the layout and the navigation is intuitive.“
“Everything looked smooth and everything feels nice. Goes quick when you browse through different menus
which is nice.”
Participant #3 was the only one providing only negative feedback, which he answered as following:
“ I wouldn't say complicated but. As a new user, it was different, it wasn't it wasn't too obvious. And that's why I think it like small information popping up could be used at least for
me for the first time. “
Participant #10 also had some negative feedback regarding the design of the menu system, illustrated by the following statement:
“I think it was extremely easy to use, since i found everything i needed very fast. I think the pictures are too big, I’m not a person that are really into pictures, I only need the text so I can
see who is playing what and so on. It feels like the big pictures are limiting me from getting what I need. I don’t like the design,
it feels like it is made for kids. I think everything takes so much place, I would like it to be more compact.”
Participant #1 had similar views regarding the size of the pictures:
“Friends page pictures are very large. It's quite large for just show which friends are playing the game right now. I guess
maybe only have a one row list would be better.”
Other participants had some issues with the size of the text.
When asked about what the participants disliked, 5 participants had nothing particular in mind, other 5 participants disliked the text sizes, for example, participant #8 commented:
“ The text is kind of small, like the "By Valve" and "90% players liked this is” text are a bit too small “
Participant #1 had similar thoughts regarding the text size.
Also, he mentioned about the rating indicator:
“It's just an inconvenience for me to read, it is nice to have the name of the game big and obvious so I don't have to kind of look close to see what it is. For me like this rating indicator, I don't really know if this is this really good or bad, maybe using
an indicator like stars would be clearer for me.”
Participant #2 disliked the placement of the Explore menu, to which he elaborated:
“I think overall it's overall it's good, But. I thought about the Explore button. That I would have probably been easier to find
it if it was closer to favorites because they are more related than friends if it was closer to.”
Also, he struggled a bit finding the report function, to which he explained:
“I did struggle a bit to find the report but I guess that people don't often report things, it makes sense that it's under more and not shown of by default. Overall I didn't feel
like there was any problems navigating.”
To summarize, the user experience with the menu system is generally positive, with 8 participants stating that they could easily find all the function. Positive feedbacks could be summarized as; The users could clearly understand the structure of the interface, The interface was easy and intuitive to navigate through, and the design was clean and tidy.
On the downside, some participants did not find the navigation structure too obvious. The size of pictures and text was also an issue some participants experienced. Some participants think that the pictures were too big, taking much space without showing too much information, and the text was in some cases too small for the participants to comfortably read through.
6.2.2 Feedback regarding information and functions
Regarding the amount of information given to the users, most participants think the amount of information are sufficient for them to decide whether they want to try the game itself or not.
Out of 10 participants, 4 participants expressed that they would like to see more information regarding the game contents before they decided to try it. Participant #7 and participant #9 expressed that being able to see player count would be nice, exemplified by the following statement by participant #9:
“Player count can be nice on certain games I mean if I want to get into a multiplayer game I would like to know how many people are playing it actively so I don't go into a dead game.”
He noticed that there is gameplay video embedded in the interface. However, he expresses his concern regarding the gameplay video;
“The first video that comes up it's often like announcement video or it's like a cool highlight trailer or it's like animated. I
want to see a gameplay video, I don't want to see advertisements. I want to see how it looks like when I play and
maybe I read like a review from the game.”
How long time it takes for a user to experience the full game is also information that some users want to see, which participant
#8 stated:
“I think it is enough. But it may be nice to also see how long time it takes to play a game like if it's like a quick join game and
you play for five minutes or if it's like World of Warcraft where you have to spend many hours to get the full experience. “
Lastly, participant #10 would like to see reviews from their favorite YouTuber or review websites like IGN, exemplified by the following statement of participant #10:
“I would like to see review from people I know, like my friends or favorite YouTubers, I trust their reviews more.”
As the result suggests, beyond these before mentioned information like player count and review from favorite YouTubers, there is no more information that the participants wish to see in such an interface.
When participants were asked about whether “related game”
would be helpful for them to find new interesting gaming content, the feedback was generally positive, with 7 participants replying that it would be useful for them in exploring for new interesting games to play, exemplified by the statement from participant #5:
“ I think it is. Interesting just to see like some suggestions of games and I like that you can watch like a quick gameplay video whatever just to see how it plays, if it looks good i would
give it a try.”
Participant #1, #2 and #3 did not think that they would use such a function, as illustrated by the comment given by participant
#1:
“I don't know really for me. Similar games or related games for me it's not something I would look into that much. Because for me it's more important to see what my friends are playing.
I know some people like it playing some small games and just tests as many games as possible, but it is not really what I
enjoy.”
Another function in the design is a function that selects a random game for the user, where participants gave mixed feedback. Out of the 10 participants, 7 participants expressed that they think it could be helpful, exemplified by answers from participant #9 and #5:
“If I have a LAN party with my mates and we're really bored and we are like let's try out some games and see if we can find
something to play and then we just click it.”
“Yeah I think this could be good to have, I would probably not use it often but it could be interesting to try out when I want to
try something new.”
One suggestion received regarding the random function is to be able to randomly select a game with regards to genre, exemplified by the following comment by participant #7:
“Yes, it could definitely be helpful, but after my preferences, so based on genres or what I have played earlier.”
Three participants could not see themselves using this function at all because they want to play the same games as their friends do and are concerned with the quality of the games, explained with the following statement by participant #2:
“It seems like if it is totally random then it's a high chance that you'll get games you would never play. I don't know this for a fact but I can assume that since there are so many games, let’s
say for example in Steam, the majority of games are probably bad, so the likelihood of getting a bad game by pressing a
random button is big, if that's avoided then yes.“
6.2.3 Other comments
Participants were asked to give examples of things that they particularly liked or disliked, 4 participants expressed that they liked being able to combine category tags, as participant #4 described:
“I like that you're able to combine categories because oftentimes that one category doesn't describe what you're looking for enough. For example, survival horror or survival explorer are two very different kind of games, in this case you
can combine them to find exactly what you want.“
Another suggestion was given by participant #1 regarding information around friends:
“It could be nice to see games my friends have played recently.
Because it's also a part for me while exploring new games. I want to play games that my friends plays.“
Like the comment about what user disliked particularly, the answers were mostly regarding the design aspect of the interface, which was summarized in section 6.2.1.
Beyond those comments mentioned before, users were asked to give suggestions about more ways to find new interesting games; most participants replied that the functions and information in this design are enough, and they do not feel like anything else is needed.
7. DISCUSSION
To answer the question, what are the important aspects of designing a prototype of an interface that will help users find interesting content in a game that supports vastly different gameplay experiences, concerns many areas. What do the users need? What do users look for when trying to find interesting games? Answers to these questions are the cornerstone to designing such a system, and these are the question that this study seeks to answer.
As there are not many studies done in relation to the topic, most of the discussion will be revolved around the result of this study. Result of this study consists of a survey, a prototype design and a round of usertesting. A survey was sent out to get a grasp on what the users needs, with the results from survey and knowledge gained from background study, an interface prototype design was developed. To get a reality check from real potential users, the prototype was evaluated via user testing, and feedback from users was gained from semistructured interviews. Result suggests that the proposed set of function and information available are sufficient for most user, and the structure was clear.
7.1 Design Solution
It was a relief to see that the design was well received. As the result reflects, participants had no problem navigating through the menu and the structure was well understood by the participants.
Regarding the aesthetics of the prototype, most participants expressed that they liked the way the prototype looked, but some concerns were raised by the size of content pictures and the size of content labels. Certainly, issues that should be looked more into, it could, however, be affected by the test setup which was used during the user evaluation. During the design phase, the prototype was developed using a 24inch
display, and the prototype was then tested on a 13inch laptop screen. One issue, however, mentioned by a participant, that it was somewhat difficult to get an overview of the quality of the content fast, because it was hard to tell the difference with a progressbarlike rating indicator, in combination with the rating text being small. This should be addressed in the before moving on into the development phase of this project.
Total played count or active player count is a feature that some participant expressed their interest in, which was not included in this design. This is a reasonable demand since if the game in question is a multiplayer game, nobody would like to play alone[3][4][5]. The reason behind not including the player count in this design is mainly based on two assumptions.
Firstly, using a such a player count would not do some games justice, for example, single player games versus multiplayer games. Secondly, showing a player count could help the popular games get more popular, embracing a richgetricher behavior[10]. However, it could potentially hurt some less popular games severely regarding popularity. In a purely user governed content delivery platform, this might also severely prevent new up and coming content to become popular.
Some other features that participants have expressed interest in are the possibility of looking into what their friends have played and being able to get random game with genre preferences, which could be helpful in providing users with more choices.
A few participants replied in the interview that they wish to see reviews from favorite YouTubers and wellknown review websites like IGN or Metacritic. A way to implement this could be prompting users to enter their favorite YouTubers when the user starts the game for the first time, and the system would then crawl through YouTube and locate those videos, displaying them in the detailed view. Wellknown review websites, however, should be kept out from the design. With knowledge gained from the survey and interview, it is one of the least used ways to find out information regarding games.
Including this in the design may result in most users experiencing unnecessary clutter.
A balance between aesthetics vs functionality is important.
Relating to the systems which were presented in related work[17][18], they seem to be designed around the aesthetics of the game and therefore have limited functionality, to maintain a consistent graphical profile. These system lacks the ability for categorizing content by genre, neither are there a review system in place. This presents several problems for platform longevity and user welfare. Display of content is purely based on user rating, meaning the newly uploaded content have no way of gaining popularity quickly. Since the front page of the content selection page is always occupied by highest rated content and most played, which might give content creator less incentive to create new content, and less refreshing content would, in turn, be delivered to users.
A suggestion made by the development team was to include a Facebook wall style of social page, where users could choose to post their experiences or what they are currently enjoying.
However, there are couple of arguments to be made for not including similar services; Firstly, for the initial stage of the project, it is beneficial to have users share their experience on other social platforms instead of an embedded one, which could lead to more exposure. Secondly, assumptions were made that such platforms are rarely used, implementing such solution could result in unnecessary clutter in the interface.
