StreamSpeak
UX Design
This case study explores the initial research, requirement gathering, and envisionment done in the domain of online game streaming and audience participation. We found that many people turn to games to socialize and thus a large gaming community is formed. An important part of this community is the spectators: people who watch other people play games, either in co-located environments on digital streams. On reading literature, we reach the problem area of limited agency and interactive design in spectatorship, and the scope for innovating and designing interactive systems that promote audience interaction and participation in the stream. After rigorous ideation, prototyping, and evaluation we develop StreamSpeak, a Chrome extension to aid communication in game streams.
View Live →View Report →View Code →View Prototype →We started with a very broad problem area and defined a problem statement for us. This was done after validating our initial research and need finding.
Game spectators have limited modes of communication and social interaction while watching streams and thus need an improved agency to participate in a more meaningful way.
While researching more in this space, we ended up on Twitch oberving different streams to identify pain points and potential areas to dig deeper into. What we found was that it was tough to keep up with the chat and we had difficulties in decoding the language and context of chat, and most of us were pretty clueless about the tons of slang used in the chat. After validating this pain point we started to brainstorm for solution ideas keeping in mind the following question -
We follow the ideation planning process discussed in This is Service Design Doing. We generated over 100+ ideas using two brainstorming sessions, sampling the world and brainwriting. The ideas were then evaluated based on scalability, technology, control, omnipresence, and novelty. After evaluating we were left with ten ideas, for which we created very low fidelity to test the feasibility internally.
Out of the huge list, we chose 10 solutions after following various methods to vote. We reached a final list of 10 ideas which were further ranked according to various parameter.
We decided to create low fidelity prototypes for our top 4 ideas.
We tested them out by various methods such as paper prototyping, clickable prototypes, acting scenarios out and storyboarding. Through this prototyping process, we learned not to go too deep or define too many features in the solution at this stage. We got valuable insights about our prototypes from the users, which made us rethink our perspective on some of the solution spaces.
After evaluating the 4 solutions, we had StreamSpeak which ranked the highest. Hence we went on to further design and develop StreamSpeak. The use of slang words has increased so much both on and off the stream, that it is harder for the newbies to feel a part of the community if they don’t understand the meaning of the everchanging gaming vernacular. Different stream communities develop their own vernacular, so it is even harder to keep track. StreamSpeak helps the audience understand the slang by interacting with the text, and thus gives the user an instantaneous context for the word so that they can continue to participate in the conversation.
With the advent of new technology, higher broadband speeds, and increased accessibility of technology, the world is going more and more remote. People are more incentivized to work from home, along with students at schools and colleges opting for e-learning options. Our brains are wired a certain way and there are special physiological responses to being around other people that current modes of technology cannot elicit. Thus, as the remote nature of the world increases, so does social isolation. This becomes especially important in the current context of the COVID-19 pandemic when many of us have been forced to stay at home, leading to workplaces, schools and other institutions making changes to account for the new normal.
We found that many people turn to games to socialize and thus a large gaming community is formed. Through the use of effective storytelling, artistic choices, game design and mechanics, games are highly engaging. An important part of this community is the spectators: people who watch other people play games, either in co-located environments on digital streams. People are attracted to watching streams to fulfill informational, social and escapist needs. This rise in gaming has also led to changes in urban vernacular, where we see people using terms as ’clip that’ or ’GGez’ in everyday conversations as well.
Game spectators have limited modes of communication and social interaction while watching streams and thus need an improved agency to participate in a more meaningful way.
Procedure: We conducted preliminary surveys to develop a broad understanding of the problem space we are trying to explore. The participants of the survey were in the age group of 18-23 and mainly comprised of students, with a small set of working individuals as well. We had a total of 104 participants in the survey. The questions we asked in our survey revolved around life before lockdown, social isolation during the lockdown, and some insights about the impact of technology and games in people’s lives.
Insights: Though video calls and texting (>93%) were the highest, we discovered that a surprisingly large percentage (91.5%) of the people we surveyed employed gaming as a means of communication, even beating out voice calling (82.9%). About a third of the people (30%) felt the need to socialize, going as far as to break certain protocols of the lockdown. With this knowledge, we felt sufficiently confident with the increased role of technology and gaming in people’s life, especially in a remote and isolated context like that of the lockdown.
Procedure: Our goal for conducting the interview was to understand the social aspects of gaming and streaming. To do that we formed questions about people’s gaming habits, their choice between multiplayer/solo games, technological constraints they face and more questions about their motivation to watch streams or other people play. Before conducting the interview, the procedure was explained and verbal consent was taken to record the interview. The interview was preceded by casual everyday conversation to establish neutral grounds and succeeded by some general open ended questions about their favourite games. We conducted 8 semi-structured interviews with a varied set of people. Among the interviewees there was a professional gamer and streamer, a serious gamer and streamer, 4 avid gamers and 2 occasional gamers. Out of the 8 people interviewed, 2 were female.
Insights: Although everyone had different motivations, we found that all the interviewees used gaming as a way to socialise. Some preferred to play with friends whereas others did not have any preference. Following were the common themes we gathered from the interviews:
While researching more in this space, we ended up on Twitch oberving different streams to identify pain points and potential areas to dig deeper into. What we found was that it was tough to keep up with the chat and we had difficulties in decoding the language and context of chat, and most of us were pretty clueless about the tons of slang used in the chat. After validating this pain point we started to brainstorm for solution ideas keeping in mind the following question -
We follow the ideation planning process discussed in This is Service Design Doing. We generated over 100+ ideas using two brainstorming sessions, sampling the world and brainwriting.
Out of the huge list, we chose 10 solutions after following various methods to vote. We reached a final list of 10 ideas which were further ranked according to same parameters.
We decided to create low fidelity prototypes for our top 4 ideas.
We tested them out by various methods such as paper prototyping, clickable prototypes, acting scenarios out and storyboarding. Through this prototyping process, we learned not to go too deep or define too many features in the solution at this stage. A solution need not necessarily cover all aspects of our problem statement. The rapid nature of this prototyping forced us to focus less on implementation and spend more time on ideating and testing. The most important learning was the feedback given by users during prototype evaluation. We had the assumption that we would not get precious/actionable insights due to the lower fidelity, but our assumption turned out to be wrong. We got valuable insights about our prototypes from the users, which made us rethink our perspective on some of the solution spaces.
We collectively decided to go ahead with StreamSpeak as the idea to implement out of the four prototypes due to it’s novelty in better integrating new viewers into the community.
The concept for the extension was to provide a quick context/meaning for the stream chat without requiring the viewer to navigating away from the stream which could lead to distraction or loss of information. We wanted the interaction to feel natural. We observed that when people try to look up words online, they tend to copy-paste it into the search bar. To copy the word, the instinctive reaction is to double click the word to select it, hence we used the same interaction to activate the extension.
We developed a Chrome extension that works with the stream chat on Twitch. Here, the extension can recognize the selection of words (via double click on the word or through dragging the cursor over it). Then the extension pings an API of a pre-existing dictionary that returns the meaning of the word. Once we have acquired the word, we inject HTML code into the web-page, that creates a pop-up under the word selected, giving us the definition of our word.
The goal of our evaluation is to fine tune the interface of the extension. We want to evaluate the intuitiveness of the design, ease of learning and subjective satisfaction. We also want to examine the usability of the current application from the perspective of users who have not used the system previously. We also want to evaluate the effectiveness of content presentation in the extension. As a broader goal, we hope to check if the solution lowers the barrier of communication and socialising during game streams. We devised two tasks for users and a post study survey -
The users were pleased with the ease of use of the extension. Some users felt that examples and citations would also be beneficial. A user expressed concern regarding the singular definition being a hit-or-miss situation, and would prefer availability of alternate definitions too for better clarification of the meaning in the case that the initial meaning wasn’t accurate. They also expressed a need for finding the meanings of the slang the streamer may speak, or to search their own words using the extension. The user also was curious to know about how this extension would differentiate between a normal word and a slang.
Following the design processes, we learned from the various sources, and implementing StreamSpeak as a tool for viewers to feel better included greatly augmented our design skills in the context of various interactions that we participate in. We explored the various ways people engage in socialising and interacting on streaming platforms and were able to arrive at some great ideas to augment the stream viewing experience. StreamSpeak is one example of the various ways we identify stream viewing experiences to be made more engaging, interactive, and non-intimidating while keeping usability and convenience as the top priority.
With the advent of new technology, higher broadband speeds, and increased accessibility of technology, the world is going more and more remote. People are more incentivized to work from home, along with students at schools and colleges opting for e-learning options. Our brains are wired a certain way and there are special physiological responses to being around other people that current modes of technology cannot elicit. Thus, as the remote nature of the world increases, so does social isolation. This becomes especially important in the current context of the COVID-19 pandemic when many of us have been forced to stay at home, leading to workplaces, schools and other institutions making changes to account for the new normal.
We found that many people turn to games to socialize and thus a large gaming community is formed. Through the use of effective storytelling, artistic choices, game design and mechanics, games are highly engaging. An important part of this community is the spectators: people who watch other people play games, either in co-located environments on digital streams. People are attracted to watching streams to fulfill informational, social and escapist needs. This rise in gaming has also led to changes in urban vernacular, where we see people using terms as ’clip that’ or ’GGez’ in everyday conversations as well.
Game spectators have limited modes of communication and social interaction while watching streams and thus need an improved agency to participate in a more meaningful way.
Procedure: We conducted preliminary surveys to develop a broad understanding of the problem space we are trying to explore. The participants of the survey were in the age group of 18-23 and mainly comprised of students, with a small set of working individuals as well. We had a total of 104 participants in the survey. The questions we asked in our survey revolved around life before lockdown, social isolation during the lockdown, and some insights about the impact of technology and games in people’s lives.
Insights: Though video calls and texting (>93%) were the highest, we discovered that a surprisingly large percentage (91.5%) of the people we surveyed employed gaming as a means of communication, even beating out voice calling (82.9%). About a third of the people (30%) felt the need to socialize, going as far as to break certain protocols of the lockdown. With this knowledge, we felt sufficiently confident with the increased role of technology and gaming in people’s life, especially in a remote and isolated context like that of the lockdown.
Procedure: Our goal for conducting the interview was to understand the social aspects of gaming and streaming. To do that we formed questions about people’s gaming habits, their choice between multiplayer/solo games, technological constraints they face and more questions about their motivation to watch streams or other people play. Before conducting the interview, the procedure was explained and verbal consent was taken to record the interview. The interview was preceded by casual everyday conversation to establish neutral grounds and succeeded by some general open ended questions about their favourite games. We conducted 8 semi-structured interviews with a varied set of people. Among the interviewees there was a professional gamer and streamer, a serious gamer and streamer, 4 avid gamers and 2 occasional gamers. Out of the 8 people interviewed, 2 were female.
Insights: Although everyone had different motivations, we found that all the interviewees used gaming as a way to socialise. Some preferred to play with friends whereas others did not have any preference. Following were the common themes we gathered from the interviews:
While researching more in this space, we ended up on Twitch oberving different streams to identify pain points and potential areas to dig deeper into. What we found was that it was tough to keep up with the chat and we had difficulties in decoding the language and context of chat, and most of us were pretty clueless about the tons of slang used in the chat. After validating this pain point we started to brainstorm for solution ideas keeping in mind the following question -
We follow the ideation planning process discussed in This is Service Design Doing. We generated over 100+ ideas using two brainstorming sessions, sampling the world and brainwriting.
Out of the huge list, we chose 10 solutions after following various methods to vote. We reached a final list of 10 ideas which were further ranked according to same parameters.
We decided to create low fidelity prototypes for our top 4 ideas.
We tested them out by various methods such as paper prototyping, clickable prototypes, acting scenarios out and storyboarding. Through this prototyping process, we learned not to go too deep or define too many features in the solution at this stage. A solution need not necessarily cover all aspects of our problem statement. The rapid nature of this prototyping forced us to focus less on implementation and spend more time on ideating and testing. The most important learning was the feedback given by users during prototype evaluation. We had the assumption that we would not get precious/actionable insights due to the lower fidelity, but our assumption turned out to be wrong. We got valuable insights about our prototypes from the users, which made us rethink our perspective on some of the solution spaces.
We collectively decided to go ahead with StreamSpeak as the idea to implement out of the four prototypes due to it’s novelty in better integrating new viewers into the community.
The concept for the extension was to provide a quick context/meaning for the stream chat without requiring the viewer to navigating away from the stream which could lead to distraction or loss of information. We wanted the interaction to feel natural. We observed that when people try to look up words online, they tend to copy-paste it into the search bar. To copy the word, the instinctive reaction is to double click the word to select it, hence we used the same interaction to activate the extension.
We developed a Chrome extension that works with the stream chat on Twitch. Here, the extension can recognize the selection of words (via double click on the word or through dragging the cursor over it). Then the extension pings an API of a pre-existing dictionary that returns the meaning of the word. Once we have acquired the word, we inject HTML code into the web-page, that creates a pop-up under the word selected, giving us the definition of our word.
The goal of our evaluation is to fine tune the interface of the extension. We want to evaluate the intuitiveness of the design, ease of learning and subjective satisfaction. We also want to examine the usability of the current application from the perspective of users who have not used the system previously. We also want to evaluate the effectiveness of content presentation in the extension. As a broader goal, we hope to check if the solution lowers the barrier of communication and socialising during game streams. We devised two tasks for users and a post study survey -
The users were pleased with the ease of use of the extension. Some users felt that examples and citations would also be beneficial. A user expressed concern regarding the singular definition being a hit-or-miss situation, and would prefer availability of alternate definitions too for better clarification of the meaning in the case that the initial meaning wasn’t accurate. They also expressed a need for finding the meanings of the slang the streamer may speak, or to search their own words using the extension. The user also was curious to know about how this extension would differentiate between a normal word and a slang.
Following the design processes, we learned from the various sources, and implementing StreamSpeak as a tool for viewers to feel better included greatly augmented our design skills in the context of various interactions that we participate in. We explored the various ways people engage in socialising and interacting on streaming platforms and were able to arrive at some great ideas to augment the stream viewing experience. StreamSpeak is one example of the various ways we identify stream viewing experiences to be made more engaging, interactive, and non-intimidating while keeping usability and convenience as the top priority.
