Content uploaded by Oliver L. Haimson
Author content
All content in this area was uploaded by Oliver L. Haimson on May 09, 2017
Content may be subject to copyright.
What Makes Live Events Engaging on Facebook Live,
Periscope, and Snapchat
Oliver L. Haimson1,2
ohaimson@uci.edu
1Microsoft Research
Redmond, WA, USA
John C. Tang1
johntang@microsoft.com
2Department of Informatics
University of California, Irvine
Irvine, CA, USA
ABSTRACT
Live streaming platforms bring events from all around the
world to people’s computing devices. We conducted a mixed
methods study including interviews (N = 42) and a survey (N
= 223) to understand how people currently experience events
using Facebook Live, Periscope, and Snapchat Live Stories.
We identified four dimensions that make remote event
viewing engaging: immersion, immediacy, interaction, and
sociality. We find that both live streams and the more curated
event content found on Snapchat are immersive and
immediate, yet Snapchat Live Stories enable quickly
switching among different views of the event. Live streams,
on the other hand, offer real time interaction and sociality in
a way that Snapchat Live Stories do not. However, the
interaction’s impact depends on comment volume, comment
content, and relationship between viewer and broadcaster.
We describe how people experience events remotely using
these social media, and identify design opportunities around
detecting exciting content, leveraging multiple viewpoints,
and enabling interactivity to create engaging user
experiences for remotely participating in events.
Author Keywords
Live streams; events; video; user engagement; social media;
Facebook Live; Periscope; Snapchat.
ACM Classification Keywords
H.5.m. Information interfaces and presentation (e.g., HCI):
Miscellaneous.
INTRODUCTION
In June 2016, thousands watched as U.S. House of
Representatives Democrats and C-SPAN used the Periscope
live streaming platform to broadcast the Democrats’ sit-in to
protest gun control political inaction after the official camera
feed was turned off [61]. Outside, supporters also used
Periscope to live stream their rally. Several weeks later, a
Minnesota woman used Facebook Live to live stream the
aftermath of a police officer shooting her boyfriend
Philandro Castille [53]. In August, a New Yorker’s Periscope
stream of a man climbing Trump Tower using suction cups
reached over 225,000 live viewers [27]. These are just a few
well-publicized events in 2016 that demonstrated the rise of
live streaming as a tool for sharing exciting or contentious
events. Beyond these national stories, many events were live
streamed at a smaller scale: concerts, birthday parties,
political marches, karaoke nights, gaming conventions, and
baseball games, just to name a few.
Snapchat also emerged in 2016 as a major media platform
[19,31] both for social sharing and event experiences via the
Snapchat Stories feature, recently mimicked by Facebook,
Instagram, and others [24]. Snapchat “Live” Stories are not
truly live, instead appearing 45 minutes to 16 hours after an
event, in our observation. They are human or algorithmic
curations of user-submitted video or photo “Snaps.” While
many breaking news events, such as the House sit-in and
Castille’s murder, did not appear as Snapchat Live Stories,
others, including the Trump Tower climber, did. Snapchat
Live Stories tend to focus on planned events like sports
games, festivals, and conventions, but do sometimes feature
breaking news stories (though not in real time).
While live streams and Snapchat Live Stories offer popular
but different experiences for viewing events, together they
provide a window into how people use social media to
remotely experience events today. To understand this
emergent sociotechnical phenomenon, we conducted a
mixed methods study including interviews (N = 42) and a
survey (N = 223). We examined how people use the most
prominent live streaming tools, Facebook Live and
Periscope, along with Snapchat Live Stories, to view and
share event experiences. We asked, what is engaging about
viewing events live? How do live streamed events compare
with viewing Snapchat Live Stories? We define user
engagement as a quality of user experience characterized by
four of O’Brien and Toms’ [41] user engagement attributes
that are most relevant to viewing events remotely: focused
attention, endurability (i.e., satisfaction), novelty, and felt
involvement [39]. By understanding current tools’ strengths
and limitations, we identify design opportunities for new
user experiences around remotely participating in events.
Permission to make digital or hard copies of all or part of this work for personal o
r
classroom use is granted without fee provided that copies are not made or distribute
d
for profit or commercial advantage and that copies bear this notice and the full citatio
n
on the first page. Copyrights for components of this work owned by others than the
author(s) must be honored. Abstracting with credit is permitted. To copy othe rwise, o
r
republish, to post on servers or to redistribute to lists, require s pri or sp eci fic p ermi ssio
n
and/or a fee. Request permissions from Permissions@acm.org.
CHI 2017, May 06 - 11, 2017, Denver, CO, USA.
Copyright is held by the owner/author(s). Publication rights licensed to ACM.
ACM 978-1-4503-4655-9/17/05…$15.00
DOI: http://dx.doi.org/10.1145/3025453.3025642
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
48
We find that both live streams and Snapchat Live Stories are
engaging for viewers. However, people desire to view a
future event via live streams more than Snapchat Live
Stories, largely due to the interaction live streams afford with
the streamer and other viewers. We contribute an empirical
understanding of how people use social media to view events
and what makes this experience engaging.
BACKGROUND
Despite the growing popularity of live streams, any
individual live stream can be boring. Zooming into any given
region on Periscope’s or Facebook’s live map interface will
likely reveal many streams of talking heads, “Untitled”
broadcasts of people going about their day or showing their
surroundings, and even broadcasts titled “Bored.” Even
when a viewer’s friends’ broadcasts show up in their
Facebook feed, they may find that friend’s commentary or
even vacation footage boring.
However, during events, live streams can provide engaging
viewing experiences. From the perspective of a person in the
stadium, a sports fan can watch her favorite baseball team
win the game. An activist can view a political march
thousands of miles away, from the perspective of someone
in the middle of the crowd. These can be powerful
experiences, but may also be punctuated with down times.
Perhaps the team has not scored in a few innings. Maybe the
protesters are losing steam toward the end of a long march.
Something could happen at any moment, but the viewer
cannot anticipate whether anything will happen. This keeps
them watching. Live streamed events are engaging but dull.
Snapchat Live Stories, on the other hand, tend to be fast-
paced and exciting. Curated to only include videos and
photos from lively periods during an event, they give viewers
a quick snapshot of event highlights. The sports fan can still
watch her favorite team win the game and skip the uneventful
innings; however, she can only view the event after the game
has already been won. The activist likely cannot view the
political march via Snapchat Live Stories unless the march
was high-profile enough to be on Snapchat’s radar, which
many smaller events are not.
We compare these two platform types, live streams and
Snapchat Live Stories, to understand how to better design
user experiences for remote event viewing. How can a
system optimize the engaging parts and work around the dull
moments in event live streams? How can it deliver exciting
event experiences in real time? We first provide background
on the three platforms of interest (see Figure 1), which we
chose due to their popularity, making it more likely that they
offer multiple streams for events. We detail the state of each
of these apps at the time of this writing, which we expect to
change with the quick pace of technological development.
Facebook Live
Originally released in 2015 as a platform for celebrities and
public figures to broadcast live content, Facebook Live was
made available to all U.S. users in January 2016 [9]. While
Facebook Live streams can be broadcast publicly, people
typically only come across streams broadcast by people they
follow, either via notification or when a stream appears in
their newsfeed. Users may also find event live streams by
browsing using the Facebook Live Map [16]. Facebook Live
streams offer several types of interaction: comments, which
are persistent and appear to the right of the video for viewers
and below the video for broadcasters; and reactions,
including Like, Love, Haha, Wow, Sad, and Angry, which,
when clicked or touched by the viewer, first show a small
photo of the viewer’s face which turns into the reaction’s
emoji while floating across the video screen.
Periscope
Periscope, owned by Twitter, is a live streaming platform
launched in March 2015. The app operates separately from
Twitter, and users can have separate social graphs on each,
but live streams are typically shared publicly on Twitter
using #Periscope when the broadcast begins. Event live
streams are discoverable by browsing or searching event
hashtags on Twitter or within the Periscope app, browsing in
the app’s map interface, or via notification if the broadcaster
is a connection. Periscope live streams allow text comments
(though only from the first 100 viewers), which overlay the
video starting at the bottom of the screen before floating up
and disappearing, and hearts, originating at the bottom right
screen corner when the viewer taps and then floating up.
Snapchat Live Stories
Snapchat launched its “Live” Stories feature in June 2014
[11]. The feature allows event attendees to contribute to a
collective story that can then be viewed by remote Snapchat
users. Live Stories were displayed in the app under the
“Live” heading starting in August 2014 [37]. After a user
records a “Snap” (a video or photo, sometimes with overlaid
annotations) at an event, users may be given the option of
sharing to an event story, which appears above the list of
people to whom the Snap can be sent. Snapchat then curates
a selection of submitted Snaps into an event story, organized
in a particular order (usually chronological), sometimes with
added contextual information. Each Snap lasts for 10
seconds at most, and viewers can skip to the next by tapping
(a)
(b) (c)
Figure 1. Screenshots showing user interfaces and interaction
types for event viewers on Facebook Live (a), Periscope (b),
and Snapchat Live Stories (c). (Please zoom in for detail).
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
49
the screen. At the time of this writing, Snapchat Live Stories
afford no interactivity, and the Live Story is available for
roughly 24 hours before disappearing.
RELATED WORK
Much as video communication tools enable people to share
experiences across distance [6], mobile video media allow
people to experience events as a group, rather than
individually [28,29]. When live events are broadcast to
remote audiences, “distributed liveness” occurs as new
configurations of audience, broadcasters, and technologies
combine various physical and social environments [59]. Live
streams can make shared event experiences especially vivid
by intertwining physical and digital experiences [12]. To
effectively support event spectatorship, technology must be
designed to consider people, context, and the event [38]. This
is especially important for remote viewers, who lack
situational context of an event’s physical surroundings.
Yet both present and remote event spectators face
challenges. Event spectatorship is tricky, because the closer
a spectator is to a particular active area, the more she loses
context for the larger event [15]. Live streaming systems can
enable broader context via multiple streams from different
event areas. Dezuli et al. designed a system that allowed
event spectators to share videos of different perspectives
with each other [12]; however, users had to already be friends
to share content, a limitation that diminishes when content is
shared publicly on live stream platforms.
Live event attendees often record event videos as a way both
to relive the experience later, and to give a gift to those who
could not attend in person; however, videos of live events
cannot replace the experience of being there in person [34].
Several studies have capitalized on event audiences’
willingness to record video, creating systems for
collaborative video productions to enhance crowd-recorded
event content [4,49,50].
Live streaming is an important resource for sharing
information during political, breaking news, and crisis
events, and can be a means for civic engagement [13] and
citizen [17,33,48] as well as professional journalism [17].
For political events, live streaming enables what Andén-
Papadopoulos [1] calls “citizen camera-witnessing,” a means
for people to document and spread images of oppression, and
what Gregory [20] calls “co-presence,” the shared sense of
space and time that bridges the gap between event
participants and audience. Thus, remote audiences can
witness and even participate in events [2,20].
Live streams have become popular in part because of the
opportunity for viewers to interact with and participate in
streams, and even to build informal, impromptu communities
through shared viewing [21,22,55]. Broadcasters appreciate
interactivity, and often let viewers’ comments influence their
streams’ content [51,55]. While Twitter offers some
interactivity during live events through text [52], live streams
offers that interaction in sync with visual event footage.
Two-way interaction between viewers and broadcasters can
be powerful, but brings challenges on both sides. Live
streamed event experiences suffer when remote audiences do
not feel acknowledged by streamers, or when event
performers or broadcasters cannot tell whether their audience
is engaged [45,59]. Broadcasters could improve live stream
experiences for viewers by avoiding repetitive content and
making stronger connections with their audiences [46].
Technical challenges also detract from both broadcasters’
and viewers’ live stream experiences [30], and most streams
currently draw few viewers [54].
Viewer experiences are improved when they have multiple
live streams to choose from. Tang et al. [55] found evidence
of multiple live streams from events, signifying the potential
to aggregate multiple streams into a rich event experience for
viewers. Hamilton et al. [22] prototyped this idea using
Rivulet, a system offering viewers a multiple-stream
interface for events, and found that multiple streams engaged
users through interaction and a shared experience.
While a large body of literature has examined Facebook as a
social network site (e.g., [14,32]), no research has been
published yet about Facebook Live as an event-viewing
platform. Previous literature on Snapchat has investigated it
as a communication platform, to understand types of content
people post and how their behaviors relate to self-
presentation, privacy, and ephemerality [5,44,47,56–58,62].
Yet no research to date has studied Snapchat Live Stories.
We examine Facebook and Snapchat in a new light, along
with Periscope, to understand how people use these popular
social media platforms to view events remotely.
METHODS
We conducted a mixed methods study involving interviews
(N = 42) and a survey (N = 223) to understand how people
use Facebook Live, Periscope, and Snapchat Live Stories to
view and share event experiences. We took an iterative
approach to the study design, adjusting our methods as we
learned from our interview data. The study was approved by
the ethics review committee at our institution.
Data Collection: Interviews
Interviews took place in June – August 2016 in three phases.
In all phases, interviews were semi-structured to allow
participants to focus on topics most salient in their viewing
or broadcasting experience. Viewer interviews focused on
how viewers experienced events remotely, what they found
engaging about event viewing on the different platforms,
what was missing from their experience, and how they
interacted with broadcasters and other viewers. Each
interviewee received a $20 Amazon gift card gratuity.
Phase 1: Within-subjects approach
In this phase, we asked participants (N = 8) to view content
from one of two popular events (an NBA Finals game and
the gaming convention E3 Expo) on each of the three
platforms. We chose events that we anticipated would offer
multiple live streams as well as a Snapchat Live Story. We
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
50
matched participants to events based on their interest in
recent events they had attended, elicited by a screening
survey. Participants were instructed to view 15 minutes total
of content within a designated timeframe and to switch
between the three platforms as they wished, while making
sure to view at least two different live streams of the event
on both Facebook Live and Periscope, and the event’s
Snapchat Live Story. Different participants may have
watched different streams, reflecting a realistic user
experience in which each chose what appealed most to them.
They were also asked to interact with at least one live stream
on each platform. To make events as consistently
discoverable as possible between Facebook Live and
Periscope, we directed participants to use the Facebook Live
Map [16] and Periscope’s map view to find streams. We then
interviewed participants for approximately 30 minutes
within 48 hours after their event viewing experience.
Phase 2: Between-subjects approach
Phase 2 was similar to Phase 1, but instead we employed a
between-subjects approach, where some participants (N = 6)
viewed live streamed content and others (N = 4) viewed
Snapchat Live Stories for two popular events (San Francisco
Pride parade and VidCon convention for online video). The
between-subjects approach allowed us to understand
participants’ experiences with the different platforms
without them making their own comparisons.
Phase 3: In-the-wild approach
In Phase 3, we interviewed people (N = 14) who had chosen
to view event live streams “in the wild” (i.e., on their own
accord) within the last three weeks. This allowed us to
understand viewing motivations and experiences more
holistically by focusing on people who were self-motivated
to seek out live streamed event content.
Broadcaster interviews
Throughout the three phases, we also recruited and
interviewed broadcasters (N = 10) from the same events
viewed by Phase 1 and 2 viewers, and from events similar to
those viewed by Phase 3 viewers.
Recruitment
To recruit, we created a screening survey that we sent out to
email lists for interns and full-time employees at our
company, shared via our personal networks on Facebook and
Twitter, and posted on event and live stream-focused groups
on Facebook and forums on Reddit. We chose participants
for Phases 1 and 2 based on their interest in the event types
we wanted them to view and their willingness to download
and use the platforms studied. We chose Phase 3 participants
based on the type of events they had viewed or broadcast,
aiming to get a wide range of event types. We limited our
sample to people between ages 18-44 to capture the
demographics common for live streaming platforms [3].
Data Analysis: Interviews
We analyzed interviews using iterative open coding,
allowing codes to emerge from the data, and the constant
comparative method to develop themes and organize codes
within themes [10]. We focused our coding on factors that
make live streams and Snapchat Live Stories engaging,
limitations to viewing, and differences among the platforms.
We quickly learned that live streams and Snapchat Live
Stories are vastly different experiences – the former is a way
to view events remotely, while the latter is a summary of
event content – and we treated them as such for the remaining
data collection and analysis. We met regularly to
collaboratively discuss the emerging codes and to organize
them into larger themes, an iterative process that evolved as
we analyzed more data. We ultimately settled on four themes
that contribute to viewer engagement and limitations to
viewing, and with which we can compare across the different
platforms: immersion, immediacy, interaction, and sociality.
Each theme included a subset of codes, described in the
results section. After reaching saturation, we coded the
remaining interview data using this coding scheme.
Data Collection: Survey
Based on the themes and codes that emerged in our interview
data analysis, we created a survey instrument to validate the
findings among a larger population of live stream viewers.
We wanted to quantify user engagement across the platforms
and understand which factors contributed most to user
engagement. We asked participants to recall a specific event
when they viewed one or more live streams. For this event,
we measured engagement using a subset of questions from
O’Brien and Toms’ validated user engagement scale [41]
(see Table 1). Additionally, we developed questions to
measure the presence of each of the codes within our four
themes (immersion, immediacy, interaction, and sociality)
for that event. For example, “The [live stream(s)/Snapchat
Live Story] enabled me to be one of the first people to see
the event” helped us measure the immediacy theme. Users
responded on a 5-point Likert scale ranging from Strongly
disagree to Strongly agree. Participants were required to
A B C
Live
streams
(full sample)
Live
streams
(those who
also viewed
Snapchat)
Snapchat
Live
Stories
Diff.
btwn.
A & C
Diff.
btwn.
B & C
N = 223 N = 63 N = 63
Scale or variable Mean (Standard deviation)
Engagement 3.85 (0.62) 4.06 (0.72) 3.96 (0.72) † n.s.
(average of Focused Attention, Endurability, Novelty, Involvement)
Focused Attentiona3.43 (0.98) 3.86 (0.83) 3.68 (0.98) * †
Endurabilitya
(Satisfaction)
4.02 (0.73) 4.18 (0.75) 4.15 (0.72) † n.s.
Noveltya4.07 (0.63) 4.18 (0.64) 4.01 (0.79) n.s. n.s.
Involvementa3.88 (0.80) 4.02 (0.79) 3.98 (0.94) n.s. n.s.
Wanting to
experience a future
event via
p
latform
4.22 (0.73) 4.40 (0.61) 4.03 (0.97) n.s. *
Immersionb3.82 (0.59) 4.02 (0.51) 3.93 (0.70) † n.s.
Immediacyb3.57 (0.58) 3.58 (0.57) 3.48 (0.57) n.s. n.s.
Interactivityb3.47 (0.60) 3.63 (0.58) - - -
Socialityb3.75 (0.74) 4.05 (0.75) - - -
asubset of questions from [37]
baverage of responses to several questions about factors associated with theme
† p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001
Table 1. Descriptive statistics for scales and variables used in
survey instrument.
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
51
have viewed a live streamed event within the past three
weeks. Those who had also viewed a Snapchat Live Story
within the past three weeks were given additional questions
about that experience. The Snapchat Live Stories survey
section did not include questions about interactivity or
sociality since the platform does not afford such interactions
between event story viewers and those who submit Snaps.
Recruitment
We used SurveyGizmo as an intermediary to recruit panel
participants from Cint, a firm specializing in panel
recruitment. As with the interview sample, we recruited
people ages 18-44. The survey was active for 1.5 weeks in
August 2016. We received 376 complete surveys, which we
cleaned using both manual inspection and SurveyGizmo’s
data cleaning tools. We removed 153 responses for one or
more of the follo wing reasons: v iewing a liv e str eam t hat was
not an event; viewing an event via a platform other than those
being studied (e.g., on television); failing our “trap”
questions; speeding; straightlining or patterned responses;
fake or gibberish text responses. This left 223 completed,
clean responses, for which we paid $15 each.
Data Analysis: Survey
We analyzed the quantitative survey data using descriptive
statistics and Wilcoxon tests to understand differences in
engagement and viewing experience on the platforms. We
built linear regression models to identify factors that most
contributed to (1) viewer engagement and (2) wanting to
experience a future event via live streams. We used a type of
backwards feature selection called recursive feature
elimination in R’s caret package [7] to determine the optimal
features to include in the models (see Table 2).
Participant Data
Interview participants
Interview participants were 55% women, 43% men, and 2%
genderqueer and had a mean age of 25 (SD = 6.15). 76%
viewed and/or broadcast on Facebook Live, 64% on
Periscope, 50% on both live stream platforms, and 10% on
Snapchat only. 64% had viewed a Snapchat event story
recently. Because many of the interviews took place in July
2016 during and after the Republican and Democratic
National Conventions (which we classified as conventions
rather than political rallies or breaking news, though they fit
in multiple categories), the convention category is most
prevalent at 35%, followed by political rallies, marches, and
protests (17%), breaking news (15%), and public social
events (15%) (see Figure 2). The interview event type
percentages add up to more than 100% because many
interview participants had viewed or broadcast more than
one type of event, while in the survey we asked participants
to recall one particular event.
Survey participants
Survey participants were 63% women and had a mean age of
31 (SD = 6.86). After removing major outliers, survey
participants had viewed on average 2.91 live streams (SD =
2.01, median = 2) for an average viewing time of 38.39
minutes (SD = 41.08, median = 25). 93% viewed live streams
on Facebook Live, 16% on Periscope, and 9% on both
platforms. 28% had also viewed a Snapchat Live Story
within the past three weeks. Due to the survey’s timing, the
most common event type was Olympics-related (22%)
followed by concerts/performances (16%), breaking news
(16%), and public social events (15%) (see Figure 2).
Limitations
For the survey, we recruited live stream viewers, then asked
the subset of them who had viewed Snapchat Live Stories
about that experience. Recruiting specifically for Snapchat
Live Stories viewers may have led to different results.
Additionally, recruiting for some interviewees internally at
our institution skewed participants toward being young and
tech-savvy; a skew that may also hold for users of the tools
we study. For the survey, however, we recruited a more
general population via the panel sample, and still found
support for our interview findings. Finally, in the regression
models we controlled for event type, but particular streams
that participants watched may have been more engaging than
others, a potential confound.
RESULTS
In this work, we first show that live streamed events, as well
as Snapchat Live Stories (a more curated and after-the-fact
way to view events), engage viewers. We then detail what
makes remote event viewing engaging. We find that while
immersion and immediacy make remote event viewing
engaging for both live streams and Snapchat Live Stories,
live streams are unique in providing interactive and social
experiences. Even without interactivity, Snapchat Live
Stories are engaging because they offer a simple and concise
way to browse through multiple rich pieces of content, a
feature currently missing from live stream platforms. Though
interactivity is key to engaging live streamed event viewing
experiences, interactivity can drive or detract from
engagement depending on the content and volume of
comments and the relationship between viewer and
broadcaster. We elaborate on these results, and present
statistical results from the survey and quotes from the
interviews.
Remote event viewing is engaging.
Live streams are engaging for most viewers. Live stream
engagement was, on average, 3.85 on a 5-point Likert scale
(SD = 0.62). To add some context, this metric is higher than
O’Brien and Toms’ user engagement scale [41] in the
Figure 2. Percent of participants who viewed each event type.
Olympics concert news public sports political convention
pe
r
cent
0
10
20
30
40 survey
interview
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
52
context of video games (M = 3.66) [60], online shopping (M
= 3.38) [42], and multimedia webcasts (M = 3.05 after
translating to a 5-point Likert scale) [40], when taking the
average of subscales similar to those we used (see Table 1).
In our interviews we learned that engagement is a broad
construct made up of many dimensions. Beyond engagement
more generally, in this section we detail how engagement for
live streams compared to engagement for Snapchat Live
Stories, and what factors make remote event viewing
engaging for viewers.
Live streams are as engaging as Snapchat Live Stories.
To understand how engaged viewers were for live streams
compared to Snapchat Live Stories, we conducted a paired
Wilcoxon signed-rank test with continuity correction. The
Wilcoxon test is a non-parametric statistical test used to
compare means for two samples of ordinal data, appropriate
for Likert-scale data such as ours. We find no significant
difference between live stream engagement (M = 4.06, SD =
0.54) and Snapchat Live Stories engagement (M = 3.96, SD
= 0.72), V = 879.5, p = 0.24, for those survey participants
who viewed events on both platforms. Thus, we conclude
that viewers find live streams, on average, just as engaging
as Snapchat Live Stories.
Immersion and immediacy make event viewing engaging.
Four dimensions that make event viewing engaging emerged
inductively from our qualitative interview analysis:
immersion, immediacy, interactivity, and sociality. Using
these dimensions, we can compare and contrast across
platforms. While both live streaming platforms in this study
(Periscope and Facebook Live) exhibit all four of these
characteristics, Snapchat Live Stories do not afford
interaction or sociality. Interestingly, both interview and
survey participants attributed immediacy to Snapchat Live
Stories despite the fact that the stories are not truly
immediate. We now discuss immersion and immediacy, the
two dimensions that make event viewing engaging for both
live streams and Snapchat Live Stories.
Immersion
Immersion is the feeling of “being there,” an experience that
video systems have attempted to provide for remote parties
for a number of years [18,43] and which is significantly
associated with event viewing engagement, r(221) = 0.65, p
< 0.001. As one interview participant described her event
viewing experience:
You're sort of like being there without necessarily being
there. I get really into watching [sports events], and I really
like watching the hockey games or the soccer games live so
I can be like, ‘I was there when he scored.’ – Phase 2, F, 20
Watching a sports event remotely via live stream gave an
experience immersive enough that this person can recount
being there when the athlete scored. But what contributes to
this sense of being there? We identified five factors that
make remote event viewing immersive. First, a viewing
experience is immersive if the energy and excitement of the
event comes through in the video. Seeing and hearing an
excited and energetic crowd is a second, closely related
factor that enables immersion for viewers. Getting another
person’s perspective allows for an immersive experience, as
does seeing the event from a privileged or special viewpoint
(e.g., front row or backstage). Finally, when a viewer can see
multiple views and experience multiple people’s
perspectives, the event becomes even more immersive.
In the survey, several of these factors were significantly
associated with engagement in Model 1 (see Table 2): energy
and excitement (on a 5-point Likert scale, a 1 point increase
in a live streamed event’s energy and excitement increases
engagement by 0.19, p < 0.001), access to other people’s
perspectives, and multiple viewpoints.
Snapchat Live Stories are similarly immersive; an
interviewee described gaining other people’s perspectives,
special viewpoints, and multiple views while watching a
Story for a fashion event:
It was not only the designers; it was the models, it was the
actual people who were doing the clothes, the fashion artists,
the makeup artists, whatever. It was cool because you could
see just a lot of perspectives. You could see from the
perspective of someone walking on the runway. Then the next
story was someone watching the same walk from the
audience. So it was really fun. We could see all of the
different angles. You could see all of the different stories
there. – Phase 3, F, 20
Model 1:
Live stream
engagement
Model 2: Wanting
to experience a
future event via
live streams
Variable Coefficient (Standard error)
Immersion variables
Energy and excitement 0.19*** (0.04) 0.24*** (0.07)
Crowd’s presence 0.19** (0.06)
Other people’s perspectives 0.14** (0.04)
Multiple viewpoints 0.09** (0.03)
Immediacy variables
Unpredictability 0.14*** (0.03) 0.11* (0.05)
A
ccess to information not
available elsewhere 0.14*** (0.03)
Being the first to see an
event 0.09** (0.03)
Content not censored or
edited 0.08* (0.04)
Interactivity variables
Other viewers responded to
comments/
q
uestions 0.12* (0.05)
Many comments -0.11* (0.05)
Control variables (included in model; none statistically significant;
details omitted for space)
A
ge; Gender; Time viewed; Used Periscope; Also viewed
Snapchat; Event type; Broadcaster relationship to viewer
A
djusted
R
2 0.54 0.29
† p < 0.10; * p < 0.05; ** p < 0.01; *** p < 0.001
Table 2. Linear regression models examining factors
associated with engagement (Model 1) and wanting to view a
future event via live streams (Model 2).
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
53
Thus, Snapchat Live Stories enable multiple views in a
quick, easy, and exciting user experience.
Immediacy
By virtue of being live, live streams give viewers a sense of
immediacy. Viewers can see what is happening in real time,
and this immediacy is associated with a more engaging
viewing experience, r(221) = 0.53, p < 0.001. A major factor
of immediacy is unpredictability; that is, the sense that the
viewer, and the broadcaster, do not know what will happen
next. An interview participant described the unpredictable
nature of viewing a Black Lives Matter rally:
You know, we watch so many videos online of things getting
out of hand or things turning violent... And so watching it
remotely I think there was constantly this sense of what’s
going to happen. Is something going to go wrong? Is
something going to turn violent? – Phase 3, M, 32
Additionally, immediacy involves having access to
information that one could not find elsewhere, and being one
of the first to learn about an event as it unfolds. Many
interview participants described viewing live streamed
events, such as political rallies or concerts, that were not
covered by mainstream news sources. In these instances,
people could stay current on the event’s developments via
live stream. Because the content was also not censored or
edited by any media company, live streams enabled viewers
to see “what really happened.”
Statistical modeling helps us understand how immediacy
factors relate to engagement. In Model 1 (see Table 2),
unpredictability, access to information not available
elsewhere, and being one of the first to see an event were
significantly positively related to live stream engagement.
Although Snapchat Live Stories are not truly live, many
interview and survey participants reported viewing them as
immediate. They do offer several of the characteristics that
we found relate to immediacy: an inability to predict what
one will see next, and access to information not available
elsewhere. Surprisingly, although content often appears on
Snapchat hours after it happens, 62% of survey respondents
who had viewed Snapchat Live Stories reported that they
enabled them to be one of the first to see an event. Snapchat
Live Stories may not be truly immediate, but they are
immediate enough for some users, perhaps because some do
not check their phone often enough for true immediacy to
matter [8]. Snapchat Live Stories also appear unedited and
uncensored to 43% of Snapchat-viewing survey participants,
despite being professionally or algorithmically edited and
curated (a costly and time-consuming endeavor). Snapchat’s
ability to appear immediate without being truly live gives
insight into what makes Snapchat Live Stories so engaging.
The quick switch is important.
Snapchat Live Stories provide an immersive and immediate
experience in a very different way than live streams. On
Snapchat, the viewer switches between viewpoints very
quickly: no video or photo is displayed for longer than 10
seconds, and users can tap the screen even before a Snap ends
to get to the next one. On Facebook Live and Periscope,
viewers can become immersed in an event experience for a
longer timespan without the view switching so quickly.
However, to get multiple viewpoints, viewers must exit out
of one live stream, interrupting their viewing experience
while searching or browsing for another stream. This detracts
from immersion and immediacy because people lose
concentration and worry that they might miss an important
event moment. Thus, many interview participants described
desiring a fast, easy way to switch between streams. One
interviewee described wanting to switch between live
streams as easily as changing television channels:
Imagine that some big event happened and you are surfing
channels on the TV. Sometimes if you're watching the news,
you would be like, ‘This news channel isn't covering enough
details about this.’ Then you see what the other guys are
covering. So I can imagine in my head the same thing playing
out [with live streams]. – Phase 3, M, 27
Another interviewee described the importance of being able
to switch quickly between different event content on
Snapchat, enabling immersion and immediacy:
I love that I can just go back and forth between the content
they show. That, to me, is really awesome, especially if …
let's say Xbox and Sony both had their press conferences at
the same time. How can I pick? I want to be doing it all at
the same time. I think that quick switch is really important. –
Phase 1, M, 20
The quick switch is missing from Periscope. Some Facebook
Live streams include a subtle arrow (see Fi gure 1) that allows
switching to another stream, but the feature remained
undiscovered by participants in our study, and the “next”
streams are often unrelated. Providing the ability to quickly
switch between different viewpoints of an event, in real time,
could be a powerful way to increase user engagement.
People want to experience future events more via live
streams than via Snapchat Live Stories.
Despite the absence of a quick switch, people were
nonetheless eager to experience future events via live stream.
We found that survey participants wanted to experience a
future event significantly more via live streams (M = 4.40,
SD = 0.61) than they did via Snapchat Live Stories (M = 4.03 ,
SD = 0.97), V = 457.5, p < 0.05. While Snapchat Live Stories
offer a compelling way of quickly and easily managing
multiple viewpoints, with a sense of immediacy despite not
showing content in real-time, we argue that interactivity is
the important aspect missing from the Snapchat Live Story
experience that leads people to be more likely to turn to live
streams for future event viewing. We discuss the exciting and
challenging nature of real-time interactivity for designing
engaging real-time event experiences in the next section.
Interactivity is key in remote event viewing experiences.
While watching video content is often a passive experience,
attending an event is typically interactive: the crowd cheers
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
54
together for the musician on stage; the people checking out
the same booth at a convention make small talk. Thus, to
fully experience an event remotely, viewers must have the
ability to interact with broadcasters and with other viewers.
One interview participant described the excitement and
involvement that real time interaction affords:
I am able to ask [a] question through [the broadcaster’s]
involvement, which is amazing to me, but I also think having
that live feed, especially to breaking events or anything that
is on the cusp of enacting change or that has personal
meaning to me is – it’s exciting. – Phase 3, G, 30
In this section we detail what we mean by interaction and
sociality in live streamed events, how they relate to user
engagement, and some of the challenges in providing
interactive viewer experiences.
Live streams are interactive and social in a way that Snapchat
Live Stories are not.
Interaction and sociality, two important factors contributing
to engagement for live streams, are absent from Snapchat
Live Stories. Participants noticed this difference and many
(70% of survey respondents who had viewed Snapchat Live
Stories) reported wanting to be able to interact with Snapchat
Live Stories. As one interviewee stated:
Snapchat’s a lot different ‘cause Snapchat’s like… you can't
interact with a story. But with Periscope and Facebook, I
think that's something I did enjoy from them. – Phase 2, F,
19
Interaction
Interaction is an important aspect of engagement for live
streams, r(221) = 0.41, p < 0.001. Fichet et al. [17] found that
interactions between live stream viewers and broadcasters
importantly allow the audience to participate in crisis events
as they unfold, and we find that this is also true for non-crisis
events. Interactions on live streams include text comments
and questions, to which broadcasters and other viewers often
respond. A specific interaction type that people reported was
requests, where viewers can often influence the broadcaster’s
actions at the event. As one interview participant who viewed
a gaming expo described:
Someone asked like…, ‘Oh, what are the specs on this
game?’ And then the person there could answer just by
looking at the booth display, or… read it to them and just
walk up closer. – Phase 1, F, 20
Interactions also include likes and reactions on Facebook
Live and hearts on Periscope. Interview participants
described these reactions as lightweight ways to show
appreciation to broadcasters, particularly after an exciting
moment in the stream. Viewer interactions encouraged other
viewers to interact with the live streams:
So when that happened, there was tons of hearts, and then I
felt like I probably clicked more than I would have because I
saw more people commenting and, you know, their comments
– or their interactions, rather, in both types – encouraged
more interactions on my part. – Phase 1, M, 20
Model 2 (see Table 2) shows that when other viewers
responded to comments or questions made on a live stream,
survey participants were significantly more likely to want to
experience a future event via live stream. This demonstrates
the importance of interactivity in the live streaming
experience, which was echoed in our interview data. One
interview participant described that broadcasters answering
viewer questions enabled “a personal engagement with the
individual who is providing that live video” (Phase 3, G, 30).
Another described how interactions with broadcasters and
other viewers led to an immersive experience:
I think questions are very important. It was very cool to be
able to see other people's questions, answer them, have the
broadcaster answer them. It really feels like you are with
them at the moment, if they see what you are saying and they
reply. – Phase 2, F, 19
Interactivity is a key difference between event content that is
broadcast live in real time, and content that is made available
afterwards. Interactive live streams make viewers feel, as one
interview participant put it, like “you actually have a say and
you feel like you have a voice that could be heard” (Phase 1,
M, 25). Interactions on live streams may help explain why
live streams are as engaging as Snapchat Live Stories,
despite often including long stretches when nothing
particularly interesting occurs. Live streams are an active,
rather than a passive, means of viewing video.
Sociality
By sociality, we mean the ways that live streams can be
social even without the viewer interacting in any way.
Sociality is positively associated with engagement, r(221) =
0.59, p < 0.001. Sociality occurs particularly when the
broadcaster is a friend of the viewer’s, and enables friends to
connect remotely. Interviewees described using live streams
as a way to remotely attend events with a group of friends,
or “catching up with them, even without greeting them”
(Phase 3, M, 27).
Even when people did not know other viewers of the same
event, live stream viewers sometimes created impromptu,
short-term groups of people who were all viewing the same
content, as found in previous research [21,22,55]. In this
way, people could share the viewing experience with others,
further adding to sociality and engagement.
Finally, live stream viewing was often motivated by
emotional proximity [26], a term we extend from crisis
informatics to apply to events more broadly. Emotional
proximity is defined as “an emotional connection to people
who were affected by a crisis, or a sentimental association to
the crisis location” [26]. Interview participants reported
feeling emotionally and socially connected to live streams
involving a person or place that they cared about:
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
55
It helps that I am from Toledo and that's where [the live
stream] was from, so I think there's this kind of emotional
connection with growing up and hearing about this special
event that happened here. So yeah, definitely kind of felt that
emotional connection. – Phase 3, F, 21
Emotional proximity to both Toledo and to her friend who
was broadcasting the event compelled this participant to
view several songs by a country singer, despite the fact she
was not typically a country fan. Other interview participants
described feeling motivated to view live streams from the
Turkish military uprising because they felt connected to
friends from Turkey. Sociality makes live streams
emotionally relevant and, together with interactivity,
contributes to engaging live stream viewing experiences.
Comment volume and content affects interactivity.
As much as interactivity overall makes live streams
engaging, certain interactive aspects can cause challenges. In
particular, the volume and content of comments can make
interactivity exciting or frustrating, especially on Periscope,
where the text comments overlay the video. Many interview
participants reported disliking streams with overwhelming
amounts of text, which participants described as distracting:
Most of the time it was just all those comments popping up,
distracting if I'm really trying to watch it … if I want to read
the comments I'll read them, but I just want to watch a video
without distractions. – Phase 2, F, 20
Others remarked on the overwhelming and unruly nature of
comments on live streams, often expressing frustration that
broadcasters would not get the chance to see or respond to
their comment in the sea of text. Even if text overload could
be decreased, the problem of which text should be displayed
remains. Many interview participants remarked on the boring
nature of some comments (e.g., “Hi from [location]!”).
While mundane comments can be annoying for viewers,
harassing or offensive comments are even worse.
Particularly with political events or events celebrating
marginalized identities, online harassment occurred in real
time, as described by an interview participant:
The one thing I hate, though, about the bubbles just popping
up like that is that there were a lot of trolls,… and it being a
Pride Festival there was a lot of hate speech, and so… that
was kind of ruining the experience on some videos for me. –
Phase 2, M, 21
Finding the right balance between a stream that is interactive
enough vs. overwhelming text, or a stream with exciting
debate vs. online harassment, is a challenge for live stream
platform designers and moderators. This difficult balance
may account for the fact that in Model 2 (Table 2), more
comments on a live stream is negatively associated with
wanting to experience a future live streamed event, coef. = -
0.11, p < 0.05. The tradeoffs live stream viewers face
between interaction and engagement highlight the need to
design for interactive experiences that recognize the nuances
of real time commenting. This requires handling comment
volume and content in real time, whether through human,
crowdsourced, or automated moderation.
Who is broadcasting matters for interactivity.
In addition to nuances around comments, the nature of a
viewer’s relationship to the broadcaster also affects a live
stream’s interactivity. Viewers find live streams engaging no
matter who is broadcasting; in Model 1 we found no
significant effects for broadcaster type. People also desire to
view a future event via live streams the same amount no
matter who is broadcasting (Model 2). However, when it
comes to interaction, the viewer’s relationship with the
broadcaster matters.
Barriers to interaction arise on live streams. Many viewers
are lurkers on social media more broadly, and live streams
are no exception. Others reported not knowing a tool’s norms
well enough to feel comfortable commenting. However, if a
broadcaster is a friend, barriers to interaction often dissolve.
People liked the opportunity to interact more on live streams
broadcast by friends (M = 4.12, SD = 0.82) than those live
streams not broadcast by friends (M = 3.78, SD = 0.85), W =
6024.5, p < 0.01. One interview participant described feeling
comfortable interacting when she knew the broadcaster:
If it were people that I knew, like someone live streaming a
birthday party or something like that,… I would totally
interact. That barrier would absolutely be gone. With
strangers, that could be like corporations, it feels like there
are so many comments and so much traffic that is being
generated that whatever you put is just lost in the void
almost. – Phase 3, F, 20
This quote highlights the fact that it is not only a viewer’s
relationship to the broadcaster that encourages interaction;
viewers must also be sure that their comments will be seen.
Thus, to increase interactivity, live streaming platforms must
cleverly manage text volume without making viewers feel
that their voices are not being heard.
While a viewer may prefer to interact when a friend is
broadcasting, many events that she would want to view are
not attended by her friends. It is not realistic to expect that
one’s friends would be present at, for instance, the Cubs
game in Pittsburgh, or the Adele concert in London.
Remotely attending these events via live stream would
require streams broadcast by strangers or professional
sources. Finding ways to increase interaction among
strangers might enable live stream platforms to sustain
interaction during events, which is key to engaging live
stream experiences.
DISCUSSION AND IMPLICATIONS FOR DESIGN
In this work we showed how interaction with broadcasters
and other viewers engages viewers of live streamed events.
At the same time, though interactivity is missing, people love
the experience of viewing events on Snapchat and the way
the platform presents an immersive experience without the
dull segments often present in live streams. We identify
opportunities for designing event viewing tools that offer
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
56
interactivity and exciting, Snapchat-like user experiences.
We apply our results to explore how to provide interactive,
social, immersive, real-time event viewing experiences.
Detecting exciting content
Snapchat requires skilled content curators, an algorithm that
identifies compelling content, or a combination of the two,
to deliver Live Stories. Such content curation is costly, and
adds a time window between when content is recorded and
viewed. In contrast, we suggest that a similarly compelling
event experience could be achieved by delivering live
streamed, user-generated content in an easily-browseable
format, using the crowd to generate and highlight interesting
material. The value of interactivity and sociality suggests
ways of creating a more satisfying experience. Because most
live stream experiences currently focus on individual
streams, not the overall event, we need to help people find
which stream within an event is of most interest to them.
Thus, we need to develop digital equivalents for the cues
people use to decide what acts in a three-ring circus to attend
to: they look to see which act is attracting the biggest crowd
(number of viewers) and which is getting the most
engagement via applause and laughter (hearts and text
comments). As Mostafa et al. [36] found, detecting exciting
streams and moments in streams is feasible by analyzing the
volume of user actions: views, comments, and likes/hearts.
We recommend that designers leverage this crowd-generated
information to enable early detection of engaging content,
and make it available in an experience in which viewers can
quickly discover and switch among different views.
Managing multiple views
Spectators, media sources, and even performers record many
videos during events, but this vast collection of content is not
well organized or easily consumable by viewers. Even the
collection of live streamed content for a particular event on
a particular platform (e.g., the Olympics on Facebook Live),
is not easy for viewers to discover, browse, and view. Thus,
a live streaming tool should allow viewers to easily and
quickly browse through different live streams from the same
event – the “quick switch” or “surfing channels” approach
participants mentioned in interviews. This is something that
Snapchat does well, by allowing viewers to tap on the screen
to access the “next” Snap. However, when browsing real-
time event content, it is less clear how to specify the “next”
logical stream. Crowd-generated information, along with
geographical proximity, could be used to organize streams
into a ribbon of content that viewers could flip back and forth
among. Such an approach would increase immersion and
decrease the possibility of missing an important event
moment by displaying multiple streams in an easy-to-
navigate manner and eliminating the need to exit out of one
stream and browse for another.
Capitalizing on interactivity
Live streaming enables a new form of “active spectatorship”
[29] in which spectators can be geographically remote; thus,
live streaming platforms must be designed specifically for a
collective, rather than individual, viewing experience.
Interactivity is a key component of active spectatorship. We
described three limitations to interactivity in live streams:
comment volume, comment content, and barriers to
interacting with broadcasters who are not friends. To address
these issues, we discuss opportunities for creating the right
kind of interactions that add to, rather than detract from,
immersion and engagement. Live stream platforms should
employ clever ways to manage comment streams on popular
live streams to tackle comment content and volume issues.
One solution is to use crowdsourcing to highlight the most
important comments [35]. Another is to segment comments
by grouping viewers based on social graphs or shared
interests. Our results indicate that interaction increases when
the streamer is a friend; thus, grouping viewers according to
their social network connections could lead to a more
comfortable context for commenting. When watching
television remotely with friends and family, social awareness
and lightweight messaging makes people feel more involved
in viewing [23]. This likely holds for live streams: making
small, personal chat channels available for people to view
live streams with and interact with friends, family, or
existing online communities (rather than strangers) could
increase viewer interactivity and engagement. Additionally,
providing an interface in which users could add live streams
into existing group text or video chats would provide an
exciting, personal, and social event viewing experience.
Another optimal strategy may be to group commenters by
shared interest or allegiance (e.g., Cubs fans in one comment
channel and Indians fans in another), with the caveat that this
may lead to filter bubbles.
If designed well, live stream platforms that enable multiple
views and promote interactivity can meet needs that are not
currently met even by physical event attendance, enabling
remote event experiences to go “beyond being there” [25].
By highlighting design opportunities, we hope to influence
live stream system design to create new, engaging
experiences around attending events remotely.
CONCLUSION
We empirically examined people’s experiences with viewing
videos of remote events across three current platforms. By
understanding the dimensions that make live streamed events
engaging (immersion, immediacy, interaction, and sociality)
and current tools’ limitations (the cumbersome nature of
finding exciting content, viewing multiple streams, and
managing comments), we outline design opportunities for
live streaming platforms. With this work, we contribute a
nuanced understanding of remote event viewing, and inform
the continued emergence of this sociotechnical phenomenon.
ACKNOWLEDGEMENTS
We thank study participants for sharing their experiences
with us. We also acknowledge Microsoft Research
colleagues (Gina Venolia, Kori Inkpen, Matthew Miller, and
Gerard Wilkinson), Nazanin Andalibi, and anonymous
reviewers for providing valuable feedback on this work.
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
57
REFERENCES
1. Kari Andén-Papadopoulos. 2013. Citizen camera-
witnessing: Embodied political dissent in the age of
“mediated mass self-communication.” New Media &
Society: 1461444813489863.
https://doi.org/10.1177/1461444813489863
2. Kari Andén-Papadopoulos. 2013. Media witnessing and
the “crowd-sourced video revolution.” Visual
Communication 12, 3: 341–357.
https://doi.org/10.1177/1470357213483055
3. Salman Aslam. 2015. Periscope by the Numbers: Stats,
Demographics & Fun Facts.
http://www.omnicoreagency.com/periscope-statistics/
4. Tom Bartindale, Guy Schofield, and Peter Wright. 2016.
Scaffolding Community Documentary Film Making
Using Commissioning Templates. In Proceedings of the
2016 CHI Conference on Human Factors in Computing
Systems (CHI ’16), 2705–2716.
https://doi.org/10.1145/2858036.2858102
5. Joseph B. Bayer, Nicole B. Ellison, Sarita Y.
Schoenebeck, and Emily B. Falk. 2016. Sharing the
small moments: ephemeral social interaction on
Snapchat. Information, Communication & Society 19, 7:
956–977.
https://doi.org/10.1080/1369118X.2015.1084349
6. Jed R. Brubaker, Gina Venolia, and John C. Tang. 2012.
Focusing on Shared Experiences: Moving Beyond the
Camera in Video Communication. In Proceedings of the
Designing Interactive Systems Conference (DIS ’12),
96–105. https://doi.org/10.1145/2317956.2317973
7. caret. Recursive Feature Elimination.
http://topepo.github.io/caret/rfe.html
8. Yung-Ju Chang and John C. Tang. 2015. Investigating
Mobile Users’ Ringer Mode Usage and Attentiveness
and Responsiveness to Communication. 6–15.
https://doi.org/10.1145/2785830.2785852
9. Josh Constine. 2016. Facebook Takes On Periscope By
Giving Live Streaming To All U.S. iPhoners.
https://techcrunch.com/2016/01/28/comfortable-
ephemerality-vs-reach/
10. Juliet Corbin and Anselm Strauss. 2008. Basics of
qualitative research: Techniques and procedures for
developing grounded theory. Sage.
11. Jordan Crook. 2014. Snapchat Launches Collaborative
Timelines Based On Events.
https://techcrunch.com/2014/06/17/snapchat-launches-
collaborative-timelines-based-on-events/
12. Niloofar Dezuli, Jochen Huber, Elizabeth F. Churchill,
and Max Mühlhäuser. 2013. CoStream: Co-construction
of Shared Experiences Through Mobile Live Video
Sharing. In Proceedings of the 27th International BCS
Human Computer Interaction Conference (BCS-HCI
’13), 6:1–6:10.
http://dl.acm.org/citation.cfm?id=2578048.2578057
13. Audubon Dougherty. 2011. Live-streaming Mobile
Video: Production As Civic Engagement. In
Proceedings of the 13th International Conference on
Human Computer Interaction with Mobile Devices and
Services (MobileHCI ’11), 425–434.
https://doi.org/10.1145/2037373.2037437
14. Nicole B. Ellison, Charles Steinfield, and Cliff Lampe.
2007. The Benefits of Facebook “Friends:” Social
Capital and College Students’ Use of Online Social
Network Sites. Journal of Computer-Mediated
Communication 12, 4: 1143–1168.
https://doi.org/10.1111/j.1083-6101.2007.00367.x
15. Mattias Esbjörnsson, Barry Brown, Oskar Juhlin, Daniel
Normark, Mattias Östergren, and Eric Laurier. 2006.
Watching the cars go round and round: designing for
active spectating. In Proceedings of the SIGCHI
conference on Human Factors in computing systems,
1221–1224. http://dl.acm.org/citation.cfm?id=1124955
16. Facebook. 2016. Facebook Live Map.
https://www.facebook.com/livemap
17. Elodie Fichet, John Robinson, Dharma Dailey, and Kate
Starbird. 2016. Eyes on the Ground: Emerging Practices
in Periscope Use during Crisis Events. In Proceedings of
ISCRAM 2016.
http://faculty.washington.edu/kstarbi/ISCRAM2016_Per
iscope_FINAL.pdf
18. Robert S. Fish, Robert E. Kraut, and Barbara L.
Chalfonte. 1990. The VideoWindow System in Informal
Communication. In Proceedings of the 1990 ACM
Conference on Computer-supported Cooperative Work
(CSCW ’90), 1–11. https://doi.org/10.1145/99332.99335
19. Sarah Frier. Snapchat Passes Twitter in Daily Usage.
Bloomberg.com.
http://www.bloomberg.com/news/articles/2016-06-
02/snapchat-passes-twitter-in-daily-usage
20. Sam Gregory. 2015. Ubiquitous witnesses: who creates
the evidence and the live(d) experience of human rights
violations? Information, Communication & Society 18,
11: 1378–1392.
https://doi.org/10.1080/1369118X.2015.1070891
21. William A. Hamilton, Oliver Garretson, and Andruid
Kerne. 2014. Streaming on Twitch: Fostering
Participatory Communities of Play Within Live Mixed
Media. In Proceedings of the 32Nd Annual ACM
Conference on Human Factors in Computing Systems
(CHI ’14), 1315–1324.
https://doi.org/10.1145/2556288.2557048
22. William A. Hamilton, John Tang, Gina Venolia, Kori
Inkpen, Jakob Zillner, and Derek Huang. 2016. Rivulet:
Exploring Participation in Live Events Through Multi-
Stream Experiences. In Proceedings of the ACM
International Conference on Interactive Experiences for
TV and Online Video (TVX ’16), 31–42.
https://doi.org/10.1145/2932206.2932211
23. Gunnar Harboe, Crysta J. Metcalf, Frank Bentley, Joe
Tullio, Noel Massey, and Guy Romano. 2008. Ambient
social tv: drawing people into a shared experience. In
Proceedings of CHI, 1.
https://doi.org/10.1145/1357054.1357056
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
58
24. Alex Heath. Here are all the times Facebook copied
Snapchat in 2016. Business Insider.
http://www.businessinsider.com/all-the-times-facebook-
copied-snapchat-in-2016-2016-12
25. Jim Hollan and Scott Stornetta. 1992. Beyond Being
There. In Proceedings of the SIGCHI Conference on
Human Factors in Computing Systems (CHI ’92), 119–
125. https://doi.org/10.1145/142750.142769
26. Y. Linlin Huang, Kate Starbird, Mania Orand, Stephanie
A. Stanek, and Heather T. Pedersen. 2015. Connected
Through Crisis: Emotional Proximity and the Spread of
Misinformation Online. In Proceedings of the 18th ACM
Conference on Computer Supported Cooperative Work
& Social Computing (CSCW ’15), 969–980.
https://doi.org/10.1145/2675133.2675202
27. Gerald Iorio. 2016. #TrumpClimber. Periscope.
https://www.periscope.tv/w/1vOxwpkOjYLKB
28. Giulio Jacucci, Antti Oulasvirta, Tommi Ilmonen, John
Evans, and Antti Salovaara. 2007. Comedia: Mobile
Group Media for Active Spectatorship. In Proceedings
of the SIGCHI Conference on Human Factors in
Computing Systems (CHI ’07), 1273–1282.
https://doi.org/10.1145/1240624.1240817
29. Giulio Jacucci, Antti Oulasvirta, and Antti Salovaara.
2006. Active construction of experience through mobile
media: a field study with implications for recording and
sharing. Personal and Ubiquitous Computing 11, 4:
215–234. https://doi.org/10.1007/s00779-006-0084-5
30. Oskar Juhlin, Arvid Engström, and Erika Reponen.
2010. Mobile broadcasting: the whats and hows of live
video as a social medium. In Proceedings of the 12th
international conference on Human computer
interaction with mobile devices and services, 35–44.
http://dl.acm.org/citation.cfm?id=1851610
31. Hope King. 2016. Teens say Snapchat is “most
important social network.” CNNMoney.
http://money.cnn.com/2016/04/13/technology/snapchat-
instagram-facebook/index.html
32. Cliff Lampe, Nicole Ellison, and Charles Steinfield.
2006. A face(book) in the crowd: social Searching vs.
social browsing. In Proceedings of the 2006 20th
anniversary conference on Computer supported
cooperative work (CSCW ’06), 167–170.
https://doi.org/10.1145/1180875.1180901
33. Ben Lenzner. 2014. The emergence of Occupy Wall
Street and digital video practices: Tim Pool, live
streaming and experimentations in citizen journalism.
Studies in Documentary Film 8, 3: 251–266.
https://doi.org/10.1080/17503280.2014.961634
34. Jessa Lingel and Mor Naaman. 2012. You should have
been there, man: Live music, DIY content and online
communities. New Media & Society 14, 2: 332–349.
https://doi.org/10.1177/1461444811417284
35. Matthew K. Miller, John C. Tang, Gina Venolia, Gerard
Wilkinson, and Kori Inkpen. 2017. Conversational Chat
Circles: Being All Here Without Having to Hear It All.
In Proceedings of ACM CHI Conference on Human
Factors in Computing Systems.
36. Ahmed E. Mostafa, Kori Inkpen, John C. Tang, Gina
Venolia, and William A. Hamilton. 2016.
SocialStreamViewer: Guiding the Viewer Experience of
Multiple Streams of an Event. In Proceedings of the
19th International Conference on Supporting Group
Work (GROUP ’16), 287–291.
https://doi.org/10.1145/2957276.2957286
37. David Nield. 2014. Our Story event streams go live for
all Snapchat users. Digital Trends.
http://www.digitaltrends.com/mobile/story-event-
streams-go-live-snapchat-users/
38. Andreas Nilsson, Urban Nulden, and Daniel Olsson.
2004. Spectator Information Support: Exploring the
Context of Distributed Events. In The Interaction
Society: Practice, Theories and Supportive
Technologies: Practice, Theories and Supportive
Technologies. Idea Group Inc (IGI).
39. Heather L. O’Brien and Elaine G. Toms. 2008. What is
user engagement? A conceptual framework for defining
user engagement with technology. Journal of the
American Society for Information Science and
Technology 59, 6: 938–955.
https://doi.org/10.1002/asi.20801
40. Heather L. O’Brien and Elaine G. Toms. 2010. Is there a
universal instrument for measuring interactive
information retrieval?: the case of the user engagement
scale. In Proceedings of Information Interaction in
Context, 335. https://doi.org/10.1145/1840784.1840835
41. Heather L. O’Brien and Elaine G. Toms. 2010. The
development and evaluation of a survey to measure user
engagement. Journal of the American Society for
Information Science and Technology 61, 1: 50–69.
https://doi.org/10.1002/asi.21229
42. Heather Lynn O’Brien. 2010. The influence of hedonic
and utilitarian motivations on user engagement: The
case of online shopping experiences. Interacting with
Computers 22, 5: 344–352.
https://doi.org/10.1016/j.intcom.2010.04.001
43. Gary M. Olson and Judith S. Olson. 2000. Distance
Matters. Hum.-Comput. Interact. 15, 2: 139–178.
https://doi.org/10.1207/S15327051HCI1523_4
44. Lukasz Piwek and Adam Joinson. 2016. “What do they
snapchat about?” Patterns of use in time-limited instant
messaging service. Computers in Human Behavior 54:
358–367.
45. Stuart Reeves, Steve Benford, Claire O’Malley, and
Mike Fraser. 2005. Designing the Spectator Experience.
In Proceedings of the SIGCHI Conference on Human
Factors in Computing Systems (CHI ’05), 741–750.
https://doi.org/10.1145/1054972.1055074
46. Stuart Reeves, Christian Greiffenhagen, Martin
Flintham, Steve Benford, Matt Adams, Ju Row Farr, and
Nicholas Tandavantij. 2015. I’D Hide You: Performing
Live Broadcasting in Public. In Proceedings of the 33rd
Annual ACM Conference on Human Factors in
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
59
Computing Systems (CHI ’15), 2573–2582.
https://doi.org/10.1145/2702123.2702257
47. Franziska Roesner, Brian T. Gill, and Tadayoshi Kohno.
2014. Sex, Lies, or Kittens? Investigating the Use of
Snapchat’s Self-Destructing Messages. In Financial
Cryptography and Data Security, Nicolas Christin and
Reihaneh Safavi-Naini (eds.). Springer Berlin
Heidelberg, 64–76.
http://link.springer.com/chapter/10.1007/978-3-662-
45472-5_5
48. Adam Rugg and Benjamin Burroughs. 2015. Periscope,
Live-Streaming and Mobile Video Culture. In
Geoblocking and Global Video Culture. Institute of
Network Cultures, Amsterdam.
49. Marco Sá, David Shamma, and Elizabeth Churchill.
2014. Live mobile collaboration for video production:
design, guidelines, and requirements. Personal &
Ubiquitous Computing 18, 3: 693–707.
https://doi.org/10.1007/s00779-013-0700-0
50. Guy Schofield, Tom Bartindale, and Peter Wright. 2015.
Bootlegger: Turning Fans into Film Crew. In
Proceedings of the 33rd Annual ACM Conference on
Human Factors in Computing Systems (CHI ’15), 767–
776. https://doi.org/10.1145/2702123.2702229
51. David A. Shamma, Elizabeth F. Churchill, Nikhil Bobb,
and Matt Fukuda. 2009. Spinning Online: A Case Study
of Internet Broadcasting by DJs. In Proceedings of the
Fourth International Conference on Communities and
Technologies (C&T ’09), 175–184.
https://doi.org/10.1145/1556460.1556486
52. David A. Shamma, Lyndon Kennedy, and Elizabeth F.
Churchill. 2009. Tweet the Debates: Understanding
Community Annotation of Uncollected Sources. In
Proceedings of the First SIGMM Workshop on Social
Media (WSM ’09), 3–10.
https://doi.org/10.1145/1631144.1631148
53. Catherine E. Shoichet. 2016. Facebook Live video offers
new perspective on police shootings. CNN.
http://www.cnn.com/2016/07/07/us/facebook-live-
video-minnesota-police-shooting/index.html
54. Matti Siekkinen, Enrico Masala, and Teemu
Kämäräinen. 2016. A First Look at Quality of Mobile
Live Streaming Experience: the Case of Periscope. In
Proceedings of Internet Measurement Conference.
http://arxiv.org/abs/1605.04270
55. John C. Tang, Gina Venolia, and Kori M. Inkpen. 2016.
Meerkat and Periscope: I Stream, You Stream, Apps
Stream for Live Streams. In Proceedings of the 2016
CHI Conference on Human Factors in Computing
Systems (CHI ’16), 4770–4780.
https://doi.org/10.1145/2858036.2858374
56. Sonja Utz, Nicole Muscanell, and Cameran Khalid.
2015. Snapchat Elicits More Jealousy than Facebook: A
Comparison of Snapchat and Facebook Use.
Cyberpsychology, Behavior, and Social Networking 18,
3: 141–146. https://doi.org/10.1089/cyber.2014.0479
57. J. Mitchell Vaterlaus, Kathryn Barnett, Cesia Roche, and
Jimmy A. Young. 2016. “Snapchat is more personal”:
An exploratory study on Snapchat behaviors and young
adult interpersonal relationships. Computers in Human
Behavior 62: 594–601.
https://doi.org/10.1016/j.chb.2016.04.029
58. Jon M. Wargo. 2015. Spatial Stories with Nomadic
Narrators: Affect, Snapchat, and Feeling Embodiment in
Youth Mobile Composing. Journal of Language &
Literacy Education 11, 1.
http://files.eric.ed.gov/fulltext/EJ1061110.pdf
59. Andrew M. Webb, Chen Wang, Andruid Kerne, and
Pablo Cesar. 2016. Distributed Liveness: Understanding
How New Technologies Transform Performance
Experiences. In Proceedings of the 19th ACM
Conference on Computer-Supported Cooperative Work
& Social Computing (CSCW ’16), 432–437.
https://doi.org/10.1145/2818048.2819974
60. Eric N. Wiebe, Allison Lamb, Megan Hardy, and David
Sharek. 2014. Measuring engagement in video game-
based environments: Investigation of the User
Engagement Scale. Computers in Human Behavior 32:
123–132. https://doi.org/10.1016/j.chb.2013.12.001
61. Nicky Woolf. 2016. Democrats stream gun control sit-in
on Periscope after Republicans turn TV cameras off | US
news | The Guardian.
https://www.theguardian.com/technology/2016/jun/22/d
emocrats-sit-in-periscope-facebook-live-gun-control
62. Bin Xu, Pamara Chang, Christopher L. Welker, Natalya
N. Bazarova, and Dan Cosley. 2016. Automatic
Archiving Versus Default Deletion: What Snapchat
Tells Us About Ephemerality in Design. In Proceedings
of the 19th ACM Conference on Computer-Supported
Cooperative Work & Social Computing (CSCW ’16),
1662–1675. https://doi.org/10.1145/2818048.2819948
Interruptions and Email
CHI 2017, May 6–11, 2017, Denver, CO, USA
60
