Requirements document for LinkedTV user interfaces

Deliverable 3.5 Requirements Document LinkedTV User Interfaces
(Version 2)
Lilia Perez Romero, Lynda Hardman, Michiel Hildebrand
11/10/2013
Work Package 3: LinkedTV Interface and Presentation Engine
LinkedTV
Television Linked To The Web
Integrated Project (IP)
FP7-ICT-2011-7. Information and Communication Technologies
Grant Agreement Number 287911

Requirements Document for LinkedTV User Interfaces (Version 2) D3.5
© LinkedTV Consortium, 2013 2/59
Dissemination level1
PU
Contractual date of delivery 30th
September 2013
Actual date of delivery 11th
October 2013
Deliverable number D3.5
Deliverable name Requirements Document for LinkedTV User Interfaces
(Version 2)
File LinkedTV_D3.5.docx
Nature Report
Status & version Final draft
Number of pages 59
WP contributing to the
deliverable
WP 3
Task responsible CWI
Other contributors
Author(s) Lilia Perez Romero, Lynda Hardman, Michiel Hildebrand
(CWI)
Reviewer Veselina Milanova (Universitat St Gallen)
EC Project Officer Thomas Küpper
Keywords User centred design, second screen, information needs
Abstract (for dissemination) This report describes the process of design and evaluation of
LinkedTV News, a second screen companion for interacting
with hyperlinked television in the domain of newscasts. Our
primary goal was to obtain knowledge about potential users
of LinkedTV’s technology regarding their information needs
and an indication of how they perceive the technology. We
performed two initial studies: a focus group and a series of
interviews. These allowed us to identify our target group, the
context of use, and the requirements with which we created
the concept of the application. The design of the application
was refined through a series of design iterations and a high
fidelity prototype was produced. After creating the LinkedTV
News prototype, we evaluated it with a task-based study
performed with a selection of participants of the initial studies
that closely matched the target profile.
1
• PU = Public
• PP = Restricted to other programme participants (including the Commission Services)
• RE = Restricted to a group specified by the consortium (including the Commission Services)
• CO = Confidential, only for members of the consortium (including the Commission Services)

Table of contents
1 Introduction.................................................................................................................... 5
1.1 Role of the study within LinkedTV .......................................................................... 5
1.2 Challenges of Web-enriched broadcast video and the role of a second screen...... 6
1.3 Method and structure of the report ......................................................................... 6
2 Overview of related work ............................................................................................... 7
2.1 Enhanced interactive news .................................................................................... 7
2.2 Second screen....................................................................................................... 8
3 Understanding the context of use .................................................................................. 9
3.1 Focus group........................................................................................................... 9
3.1.1 Focus group setting and method........................................................................ 9
3.1.2 Focus group participants...................................................................................10
3.1.3 Findings of the focus group and implications for the design of an interactive
news companion...........................................................................................................10
3.1.4 Study Limitations (focus group).........................................................................13
3.2 Interviews..............................................................................................................13
3.2.1 Method of the interviews ...................................................................................13
3.2.2 Participants .......................................................................................................14
3.2.3 Interview data analysis......................................................................................14
3.2.4 Findings of the interviews and implications for the design of an interactive news
companion....................................................................................................................14
3.2.5 Reflections about the interviews........................................................................17
4 Conclusions about Context...........................................................................................17
4.1 Target group .........................................................................................................17
4.2 Choice of device ...................................................................................................18
4.3 Context of use.......................................................................................................18
4.4 Personas...............................................................................................................18
4.5 Scenarios..............................................................................................................19
4.6 User requirements ................................................................................................19
5 Design solutions ...........................................................................................................20
5.1 LinkedTV News, the design concept .....................................................................20
5.2 Prototyping............................................................................................................21
5.2.1 Wireframes........................................................................................................21
5.2.2 User testing with screenshots and Paper prototyping........................................22
5.3 LinkedTV News UI design.....................................................................................23
5.4 Explanation of the five browsing routes by contents..............................................26
6 Evaluation.....................................................................................................................31

6.1 Videos, entities and related content ......................................................................32
6.2 Objectives.............................................................................................................32
6.3 Set up ...................................................................................................................33
6.4 Method..................................................................................................................33
6.5 Participants...........................................................................................................34
6.6 Results..................................................................................................................34
6.6.1 Usability and ease of use (SUS scores) ............................................................34
6.6.2 Perceived usefulness (USE scores) ..................................................................35
6.6.3 Satisfaction (USE scores) .................................................................................35
6.6.4 Assessment of the slides...................................................................................36
6.6.5 Perceived mental effort .....................................................................................36
6.6.6 Ability to answer freely formulated questions.....................................................37
6.6.7 Realism of the tasks..........................................................................................37
6.6.8 Completion time tasks.......................................................................................37
6.6.9 Usability findings ...............................................................................................37
6.6.10 Results of the interviews ...............................................................................38
6.6.11 Proposed changes ........................................................................................39
7 Conclusion....................................................................................................................40
8 Discussion of current work............................................................................................41
9 Future work ..................................................................................................................41
10 References...................................................................................................................42
11 Appendix A: News Items presented in the focus group.................................................44
12 Appendix B: Focus group protocol................................................................................45
13 Appendix C: Extended list of focus group findings ........................................................47
14 Appendix D: Qualitative interview findings ....................................................................49
15 Appendix E: Personas ..................................................................................................51
16 Appendix F: Scenarios..................................................................................................53
17 Appendix G: Requirements related to participants’ comments ......................................55
18 Appendix H: List of the functionalities implemented in the hi-fi prototype ......................56
19 Appendix I: Example of task types included in the task-based study.............................57
20 Appendix J: SUS OF INDIVIDUAL PARTICIPANTS .....................................................58
21 Appendix K: USE usefulness scores.............................................................................59
22 Appendix L: USE satisfaction scores ............................................................................59

1 Introduction
This report describes the process of design and evaluation of LinkedTV News, a second
screen companion designed for interacting with hyperlinked television in the domain of
television newscasts.
The motivation of the study is two-fold: to analyze user’s information needs in relation to
hyperlinked broadcast video in news context; and to investigate the role that a second screen
could have in dealing with some of the challenges that emerge from the interaction with
hyperlinked broadcast video.
Desired outcomes of the study are to:
 obtain knowledge about potential users of LinkedTV’s technology regarding their
information needs and an indication of how they perceive the technology;
 extract initial design guidelines for future work within LinkedTV and in the field of second
screen interfaces for web-enriched television newscasts;
 create a high-fidelity prototype to facilitate communicating a future vision of hyperlinked
broadcast news.
1.1 Role of the study within LinkedTV
An objective of the LinkedTV project in its second year was to understand the requirements
of potential users of Linked TV’s technology, which up until this point had been mainly driven
by technology concerns. The Linked TV News application was created with this in mind not
as a final product, but as a dialogue initiator between the LinkedTV technologies and users,
and as a tool for research. It was guided and constrained by the LinkedTV project goals, but
not limited by the current development of its technology. We purposely did not start with the
presentation engine and the generic second screen application that is built on top of it for the
LinkedTV scenario demonstrators (D6.2). To support our iterative design process we needed
a flexible environment, and at the time of the development of the LinkedTV News prototype
the presentation engine was not yet suited for quick prototyping. Therefore, we opted for the
development of a separate lightweight second screen prototype.
By now the presentation engine in the form of the multi screen toolkit has matured (D3.6),
and aims to support rapid prototyping. In year three of the project we plan to transfer the
LinkedTV News application to the multi screen toolkit. This will provide a more solid
infrastructure than is currently provided in the LinkedTV News prototype, for example for the
synchronization between screens. It needs further investigation to determine how the user
interface components of the generic second screen application can be reused for LinkedTV
News.
In year 3 we plan to continue the study into the "antiques iterative" scenario that was started
at Sound and Vision, as reported in D6.2. This work should result in a second screen
application that is tailored for this scenario. This work will also provide a test case for the
prototyping facilities of the multi screen toolkit.
By reflecting on the studies in these two domains we want to collect first insights into the
reusability of interface components and functionalities for second screen applications. At the
end of the project we should have a better understanding on what is needed to effectively
tailor second screen applications for different domains.
The LinkedTV news prototype does not use the LinkedTV pipeline to generate annotations
and related content. For the user testing we wanted complete control of the content shown in
the application, as we did not want the quality of the content affect the users perception of
the application. An important goal for the LinkedTV project in year 3 is to investigate how the
automatic annotation and hyperlinking algorithms developed in WP2 (D2.4) can generate

content suited for the LinkedTV News application. In year 3 we plan to work closely together
with partners in WP2 to improve and evaluate the annotation and hyperlinking algorithms.
This starts with a PhD student from Eurecom visiting CWI for two weeks in November.
Studying the process of automatic content generation for the LinkedTV News application will
also lead to a better understanding of what is required from a curator using the editor tool
(D1.3). We want to use this knowledge to investigate, in cooperation with Sound&Vision, how
to improve the editor tool.
1.2 Challenges of Web-enriched broadcast video and the role of a
second screen
Hyperlinked broadcast video implies combining content intended for a lean-forward medium
(the Web) with a lean-back one (TV). It offers the advantage of providing smooth access to
additional information associated with objects depicted or mentioned in the video. However, it
also poses new challenges and motivates new requirements from the point of view of user
system interaction.
Two challenges for supporting hypermedia are [1]:
 Tension between lean-forward content and lean-back consumption.
 Tension between multiple and single users.
We design LinkedTV News as a second screen interface, e.g. for a smart phone or a tablet,
that is to be used together with a main TV screen to investigate whether a second screen is
useful for dealing with these challenges.
1.3 Method and structure of the report
Due to the user-focused nature of the proposed research, we addressed the design of the
intended interface, LinkedTV News, using a commonly employed user-centered and iterative
design model, namely the one described in the international standard ISO 9241-210.
After an initial planning stage, iterative processes usually start by researching the context of
use of the product that is being designed product. From the results of this context research,
requirements are extracted and design solutions are produced, see Figure 1. Prototypes of
diverse degrees of fidelity are created to illustrate these solutions and evaluated until the
user requirements are met.
We follow this iterative design process structure. After a brief view of related work (Section
7), we describe the method, process, and results of the two (Section 3). Next, we define our
target group, context of use, and information needs (Section 4) .We propose a design
concept to satisfy a subset of these requirements (Section 5). The interaction design and the
process of improvement of the concept through prototyping are then explained (Section 5.3).
Section 6 describes the user evaluation of the high-fidelity prototype and concludes by
proposing a direction for future improvements. In the conclusion, the obtained results are
briefly analyzed and guidelines for future work are proposed (Section 7).

Figure 1 User-centred Design Cycle [19]
2 Overview of related work
2.1 Enhanced interactive news
Interactive news is a subject that has been researched as a use case in various studies
concerned with hypermedia and information retrieval. In 1997, Hauptmann and Witbrock [2]
presented News-on-Demand, an application that monitors news from TV, radio and text
sources and allows the user to retrieve stories of interest. News-on-Demand employed
speech and image recognition techniques for scene segmentation and automatic tagging of
archival data. The interface was also based on speech recognition and the accuracy of this
technology constituted the focus of the study’s user evaluations. News-on-Demand can be
seen as an early example of hypermedia and automatic linking to the topic of the news.
Another example of content augmentation of news broadcasts is the work of Dowman et al.
[3]. Their system identifies individual stories in news broadcasts and annotates them with
content from the World Wide Web. Annotations are used to produce summarized texts that
can be employed potentially for enrichment of electronic program guides. The focus of
Dowman’s study is the technology that enables annotation. It presents an interface designed
for the postproduction (curating and editing) phase of the enriched hypermedia. This work
relates to LinkedTV News in its integration of television newscasts with Web content.
However it differs in that LinkedTV News is concerned with news consumption and not news
postproduction.
Ardissono et al. [4] investigated personalization in the context of interactive news broadcasts
through tracking and inferring user content interests and media preferences. The authors
describe the design and implementation of a broadcast news generator. This system
combines automatic video analysis and extraction together with user modeling to provide
individualized personalcasts. We coincide with this work in that personalization should be an
important ingredient in any end-user hyperlinked news interface.

MyNewsMyWay, Lindstedt et al. [5], is a system for nonlinear and interactive news
broadcasts with an emphasis on the user needs. With the intention of recreating the
television experience that users are most acquainted with, MyNewsMyWay uses the remote
control as input device and displays the system interface on a TV screen. A similar remote
control, single screen set up was produced and evaluated by Olsen et al. [6] to enable users
to access opinions over stories headlines, skip back and forth over stories, and select stories
from a playlist.
One of Lindstedt and Olsen’s motivations to use the remote control for interacting with news
is the users’ familiarity with this device [5]. It is reasonable to think that basing new
technologies on old habits, such as the use of a remote control along with the TV, will
promote a smooth adoption of these technologies. However, we believe that the remote
control capabilities are too limited and don’t meet the demands of Web-enriched broadcast
video in terms of interaction. Moreover, remote controls usually depend on TV screen
overlays of hierarchical menus that may disrupt the TV viewing experience. We would like to
propose to take advantage of other more recent familiarity and more current habits: the
familiarity of users with mobile devices and the recent habit of mobile multitasking. With this
motivation, we would like to explore the role of a dual screen set-up (TV plus laptop, tablet or
smartphone) in the delivery of Web-enriched broadcast news.
2.2 Second screen
The use of mobile devices by viewers alongside the TV is a typical behavior according to
recent studies of user habits. The Razorfish and Yahoo! survey [10], for example, found that
mobile multitasking during TV watching is common practice among 80% of the 3000
participants. The idea of replacing these currently disconnected multitasking activities with
integrated interaction through the coupling between the second screen and the TV screen
has motivated various design-driven research studies.
In [11] a prototype of a second screen web application running on a laptop was distributed to
11 households and evaluated for three weeks. Basapur et al. found that the adoption of
second screen systems is facilitated by their resemblance to current behavior patterns
inasmuch as they are perceived as natural extensions of these behaviors. The authors
reported that users valued the effortless information gathering and felt more connected with
their TV shows through experiencing them with a second screen. They also found that the
experience promoted engagement with the program even after it was over by inspiring
outside activities (e.g. attending a certain movie recommended by the system) and enriching
sociality around TV (expanding the circle of friends with whom information is shared).
Prompting and control strategies for secondary devices were explored in [12]. Participants in
this study showed a clear preference for a separation between content and control. They
agreed that TV content belongs in the primary screen and activities demanding interaction in
the secondary device.
The employment of secondary devices for control of TV content was also explored in [12],
where four major usages of the secondary screen in interactive television are identified and
discussed: control, enrich, share, and transfer television content.
Second screen multitasking is a growing practice that has triggered the interest of industry
and stimulated the proliferation of commercial applications for dual device set-ups. One of
the best known examples is Yahoo’s IntoNow [15]. IntoNow uses audio fingerprinting
technology to automatically detect the program that users are watching and deliver related
content. Like many of its kind, IntoNow is mainly focused on delivering the experience of
“social TV”. It detects what friends are watching and connects them so that they can chat
about the show. In this study we do not study the social aspects of interconnected TV
viewing. We focus on the user information needs with respect to related content.

3 Understanding the context of use
In order to gain understanding of users’ current habits and requirements in relation to news
broadcasts and news consume in general, we conducted a focus group and a series of semi-
structured interviews. This activities relate to the first step “understand and specify context of
use” in Figure 1 about user-centered design cycles.
3.1 Focus group
The goal of the focus group is to get a first idea of people’s interests and behavior when
watching news programs. We used an exploratory approach combined with a feature
prioritization approach [16]. The exploratory approach aimed at discerning general attitudes
on TV news watching and relation of the users with the news in general, focusing on finding
the user’s information needs. The feature prioritization approach aimed at determining the
features that users would find important to have in the application and the reason for their
preferences.
3.1.1 Focus group setting and method
The study took place in an experiment room that recreated an old-fashioned living room
environment (Figure 2). The environment consisted of a large television screen, a clock, a
center table, a couch and diverse decorative objects. The motivation behind using such a
space was to reproduce a setting similar to where participants would most likely watch TV in
real life.
Four groups of, in average, three participants were scheduled in four sessions that each
lasted 45 minutes. After being welcomed in the living room and given relevant information
related to their role in the study, participants were asked to complete a questionnaire on
demographics and current news watching habits.
Figure 2 Example of focus group environment
On top of the center table, three stacks of paper were placed with printed mockup frames
representing a laptop, a smartphone or a tablet. All the frames had blank space for writing in
the center.

Once seated, participants were shown six reports from BBC’s One Minute Online World
News. The choice of video fragments was made trying to emulate the diversity of news
categories included in popular news broadcasts, and was designed after careful observation
of the latter. A list of presented news items is given in appendix A.
After each fragment, participants were asked to think whether the news piece had elicited
any question or if there was any additional information that they would look up in a similar
situation in real life. Then they were asked to choose a paper frame according to the device
they would most likely employ to find the answer, if any, and write down the information need
they had. Information needs were then discussed with the group and the moderator.
Following the news broadcast exercise participants were asked to freely brainstorm about
the functionalities that they imagined would be nice to have in a second screen news
companion. Their suggestions were written on a flip-board for the group to see.
Next, each participant was interviewed individually and asked to give his or her vision of how
the activity of watching the news would change in the future.
Finally, participants were given a cinema voucher as a “thank you” gift for their help.
The focus group sessions were recorded on video, which along with the mockup frames
containing the participant notes and the flip-board pages were kept for later analysis. The
main researcher, acting as a technical facilitator, and a second researcher, acting as
moderator, conducted the focus groups. A pilot session proceeded the four group sessions.
A complete copy of the protocol can be consulted in appendix B.
3.1.2 Focus group participants
We had no requirements for demographic characteristics of the target group, because the
product was unprecedented. Therefore we started with a diverse sample hoping that a
secondary effect of the study would be to collect information that would be useful to further
refine what the target audience should be. However, a necessary condition for participants to
be included in the group was that they were regular viewers of news broadcasts.
The recruiting took place at a university and a research center. To ensure a range of
participants representative of potential users, participants that were neither students nor
researchers were included.
Eleven participants were selected. The mean age of participants was 43 with a standard
deviation of 11.5 due to the inclusion of some young students.
According to PEW’s 2010 Media Consumption Survey spectators of this age group consume
in average more television news than younger audiences [17]. We wanted to investigate if
these habitual consumers of television news would also be a good target audience for a
news related application.
The population was gender balanced including 6 female and 5 male participants. All
participants are residents of the Netherlands. Four of the participants live alone, two live with
one other person; one lives with two other people; and the remaining four live with three
more.
3.1.3 Findings of the focus group and implications for the design of an
interactive news companion
In this section findings obtained from the focus group are presented. The first subsection is
dedicated to the questionnaire results (qualitative data), followed by the findings of the focus
group discussions and brainstorming. Each finding is followed by possible related
implications. To clearly denote the implications, they have been enclosed in frames.

3.1.3.1 Focus group questionnaire results
Devices owned and preferred
Figure 3 Shared and personal devices used by participant
Although smartphones are the devices that users most frequently own, when it comes to
searching for additional information about the news while watching TV, half of the
participants preferred laptops and the other half tablets. This corresponds to the findings of
Nielsen Connected Devices Study Q2 2012 [18], which suggest that second screen
multitasking with smartphones is a practice of users within the age group of 18-24 while older
audiences (as in this study) prefer other devices.
For a user group of middle age tablets and laptops would be the preferred devices to design
for when designing a second screen companion for watching the news.
Location and time
Out of 11 participants 9 watch the news in the evening and sit in a living room.
Chances are that users will watch the news sitting on a couch, and therefore in a relaxing
environment and that, if a table is present, it will not be at an adequate level to offer typing
support. Most likely, users will be holding the device on their lap. This suggests that a tablet
is a more suitable device than a laptop.
Social interaction
Out of 11 participants 6 watch the news in the company of other people. From these 6
participants 5 interact with their companions while they watch.
A second screen companion for watching the news should be able to adapt to group as well
as to solitary television watching.
Multitasking
Out of 11 participants 4 eat or drink while watching the news.
From the 11 participants 6 perform second screen multitasking (email reading and Internet
browsing) while watching the news. 3 participants didn’t answer this question.
A second screen companion should allow diverse levels of engagement in order to
accommodate some of the tasks that people perform concurrently to watching newscasts. It
should ideally have a “hands-free” mode to allow people to continue interacting with the
application while eating or performing tasks that are unrelated to the use of the device.
The fact that some users are already employing their mobile devices to browse the Internet
or check their email while watching the news indicates that there is an existing behavior that
can allow a smoother adoption of second screen applications.
0
0,2
0,4
0,6
0,8
1
1,2

3.1.3.2 Focus group discussions and brainstorming results
Information needs
The following information needs were identified through the coding of the focus group
transcriptions. Quotes have been included with the purpose of illustrating each need and the
number of the participant who expressed it is indicated with a “P”. These information needs
are the ones more frequently mentioned. A more comprehensive list of focus group findings
is included in appendix C.
 Need for more in-depth specific information
(11 comments / 7 participants + 1 brainstorm list)
P6 About the bird flu - I was wondering just how expensive it is to make all those vaccines,
you know? I was just wondering if it is a good idea to even make a lot of it or how hard it is to
make.
 Overview of related events that occurred in the past
P7 About North Korea’s state of war - my search would be something like "South Korea state
of war ramp up” (…) I would like to know what happened two years ago and how they solved
it…
 Search for different opinions
P7 About North Korea’s state of war - I'd start with Wikipedia, probably, and then I always
look at AP and Al Jazeera to get two different opinions…
 Information about how a news event ends, updates when important changes occur
P8 About “Human parents for a baby gorilla”- I would like to know if the story succeeded, if
the gorilla was saved …
 Local effect of global news – What is the impact on my life?
(4 comments / 2 participants + 2 comments in 1 brainstorm list)
P1 About the bird flu – Yes, then I would look it up, what is the effect in the Netherlands… (of
the bird flu virus)
People have complex information needs that can’t always be transparently traced back to a
named entity (in LinkedTV the name of a location, person or organization). It is important to
identify some of the most usual information needs and try to model the application’s
functionalities and the semantic linking accordingly.
Actions triggered by information needs
Different kinds of information trigger different actions. For example, the information about the
weather may trigger warning a friend about the rain or comparing the weather in two
locations while information about the cinema may stimulate users to buy movie tickets.
However, in this study the action that participants executed in order to look for the answer to
a given doubt or question was in most of the cases to look it up in a search engine.
We recommend looking at the tasks related to searching information in a search engine as
some of the tasks that may be integrated and simplified in the second screen interactive
news companion.
Sources currently used
Although some sources, such as Wikipedia, seem to be very often consulted, different
questions or information types are best answered by different sources. For example,
participants mentioned using Buienradar for finding out about the weather or Pathé when

they wanted to look for information about movies currently shown in cinemas. Also, users
typically have their set of preferred sources.
It would be advisable to have a certain flexibility or degree of personalization in terms of the
sources of information included in the application.
Functionalities mentioned during the brainstorm sessions
Participants seemed to expect a second screen application to include extended remote
control functionalities, and one of the most frequently mentioned desires with regard the
application was to be able to skip the part of the news that they don’t like or go back to what
they think is interesting. Web related functionalities like bookmarking, electronic transactions,
getting alerts, or sharing information that are possibly inspired by the Internet, were also
mentioned, although more sparingly.
In order to comply with user expectations and to simplify television control, a second screen
companion should include video control capabilities as one of its basic features. The role of
Web-related features should also be studied.
Perceptions about television and newscasts
Some participants were reluctant to adopt a technology that will require them to be active
during TV watching, because they considered this activity as mainly relaxing.
Integrating different levels of engagement through the second screen, including some that
are almost linear (not interactive), could be a strategy to promote smooth adoption.
At the other hand some participants indicated they miss interaction in traditional TV only
settings. TV may be perceived as outdated because it is less flexible than other “newer”
media.
Interactive applications for TV may help change the perception of television as outdated.
3.1.4 Study Limitations (focus group)
The focus group provided a first view on people’s current habits while watching the video
news broadcasts. It also gave an indication about what kind of additional information people
would like to have while watching the news. However, some aspects were detected that
could have been improved in favor of the flow of the discussion with the participants. The first
one is the language. The focus group was held in English, and the news program shown was
also in English. Although most Dutch people are fluent speaking it, apparently some
members of our group of participants were not completely comfortable watching British
television without subtitles. There was also room for improvement regarding the subject of
the news. International news, used in the study, is not interesting for all audiences. Some of
the members of the focus group had strong preference for national and local newscasts.
3.2 Interviews
Parallel to the focus group, a series of semi-structured interviews were conducted and
complemented the study. This technique was chosen in order to gain a more individual in-
depth view at users’ habits and perceptions. The interviews were especially helpful in trying
out a more specific target group with a smaller standard deviation within the ages of
participants and a more specific profile in terms of skills and relation to the news.
3.2.1 Method of the interviews
The interviews were semi-structured. The same interviewer, the researcher, conducted all of
them. They were loosely organized around a 45-item questionnaire. The questionnaire
served as a script, but probes, or follow-up questions were improvised depending on the

respondent’s answers in order to promote a more fluid conversation about the relevant
topics. The main topics of the interview were:
a. Second screen and multitasking user habits
b. Habits and social context regarding news broadcasts
c. Habits when searching for additional information (frequency, tools, processes)
d. Opinions, suggestions and desires regarding the application and functionalities
Questions related to items b and c were defined in such a way that they could map to what
Maguire [19] calls the five context factors: user goals and characteristics, tasks, technical
environment, physical environment and social or organizational environment. In other words,
what do people want to achieve (when watching the news or looking for additional
information), what do they do to achieve it (steps that they currently take to find information),
which tools do they use (which software and which devices), what is the setting where they
perform these tasks, and what is their situation in social terms (do they perform these
activities alone or in a group).
An emphasis was made in dealing with not only how participants relate to newscasts, but
also in dealing with how participants relate with the news in general and thus with other
information sources and how these sources relate or differentiate from the TV.
The interviews were conducted in an isolated room within the respondents’ workplace. They
were all digitally recorded with the interviewer’s smart phone. The length of the interviews
varied from 24 to 45 minutes.
Before starting the interviews, interviewees were asked to fill out a consent form and a
demographics questionnaire. Once the interview was concluded the respondents were given
a list of functionalities extracted from literature and previous findings to rate in accordance to
their perceived usefulness.
3.2.2 Participants
The number of interviewees was 8 with a gender distribution of 3 females and 5 males. The
mean age of the participants was 33 with a standard deviation 5. All participants were
university graduates or postgraduates. All participants were digitally skilled and acquainted
with online media and digital technologies as well as mobile devices. All participants were
regular viewers of newscasts and declared spending on average 30 minutes daily getting
information related to the news through different means.
3.2.3 Interview data analysis
After concluding all of the interviewing sessions, the eight interviews were transcribed with
time code stamping using the software F5. Transcriptions of the interviews were then
summarized. Data that could be translated into quantitative terms was translated to those
terms. Qualitative data was extracted and grouped according to affinity of the remarks in the
following broad categories: attitudes toward media (TV and newspapers) based on habits,
desired functionalities, and personalization. Some individual remarks that could not be
clustered were also kept and reported because they are considered relevant to understand
current user habits and perceptions regarding news consumption.
3.2.4 Findings of the interviews and implications for the design of an
interactive news companion
In this section findings obtained from the interviews are presented. The first subsection is
dedicated to the quantitative data results. This subsection is comparable to section 3.1.3.1
about the focus group questionnaire results. It is followed by qualitative findings obtained
from the interviews. The latter are divided according to the categories described in section

3.1.3 and used for coding. Each finding is followed by possible related implications. To
clearly denote the implications they are displayed in a frame. Finally, the list of functionalities
ranked by the users during the last part of the interview is presented.
3.2.4.1 Quantitative findings of the interviews related to users’ current habits
Devices used by participants, Figure 4.
Figure 4 Personal and shared devices used by participants
Location
Out of 8 participants 6 watch the news in a living room occasionally from the dining table.
The remaining 2 watch the news in an office environment and a bedroom.
Social interaction
Out of 8 participants 5 watch the news accompanied. All of these 5 participants interact with
their companions while they watch.
Multitasking
All participants perform second screen multitasking (email reading and Internet browsing)
while watching the news.
The interview’s quantitative data findings about user’s habits when watching the news are
comparable to the findings obtained in the focus group (section 3.1.3.1) and therefore, the
same implications apply.
3.2.4.2 Qualitative data obtained from the interviews
The following user perceptions, habits, and preferences were identified through the coding of
the interview transcriptions. One quote has been included at the end of each group of
statements to serve as example. The number of the participant who expressed each quote is
indicated with a “P”. This is a selection of the ideas that were more frequently expressed or
seemed more relevant to our work. A more comprehensive list of qualitative interview
findings and quotes is included in appendix D.
Attitudes toward media (TV and newspapers)
While reading online newspapers and watching TV broadcasts are different ways of
consuming the news, these activities are not mutually exclusive. Our contextual research
showed that users frequently perform both because they believe that each one of them offers
different advantages, for example some users believe that:
 broadcasts are useful in getting an overview of the news, and online newspapers are
good for searching in-depth information about specific news.
 it is possible to be more selective and control time better when you are reading (P3, P4,
P8).
 watching video news is a social act while newspaper reading is a solitary act (P2).
0
0,2
0,4
0,6
0,8
1
1,2

P2 - So having the video footage, that you have more in the television news, I think it adds
value there, and maybe also the fact that you can watch it together, adds value, so when you
read the newspaper I think that you really read alone.
Desired Functionalities regarding information
Users asserted that they believe that most media are biased and emphasized the importance
of getting a complete factual view on a news topic as opposed to partial or predisposed ones.
The strategy that some of them currently use to compensate for bias and to get a
comprehensive view of the news is to contrast the news headlines or topics of their interest
within different information sources and from the point of view of different people. In this
sense, when asked about important functionalities for a second screen news companion,
users stated that they would like to:
 be able to compare different opinions about a subject, different scenarios, and different
sources. (P8, P7)
 access and compare the international view on local news. (P2)
 access opinions of locals, regular people, preferably friends that are where the news take
place. (P8)
P8 - I think that the German news are still kind of OK, but still, I mean everybody is biased in
their opinion, so no matter which news you read you always have some bias. It is good to
have different and diverse opinions and first-hand information.
In order to facilitate this contrasting of information within different sources, and in order to fill
in general information gaps some users proposed the integration of the television with other
sources such as Wikipedia or online newspapers (P1, P2, P6).
P5 - I think, I would expect that you would include like the main links to what they're talking
about like if it was about Hugo Chávez, links to Wikipedia or some articles about him, the
country, and stuff like that because I think that's the thing that I would look for myself.
Personalization
Most interviewees preferred an effortlessly configurable user-controlled personalization that
allows adding personal filters to categories or topics and sources. Personalization should not
override the importance or diversity of the news. Users don’t want to miss important
information because of personalization (P1, P2, P8, P3, P7).
P8 - So, not because in the past I always clicked on a specific news article, I want to have
more of these, I would still want to have more diverse angles.
An integration between media, mainly television newscasts and online newspapers, could be
a suitable solution for providing an enriched and hyperlinked news experience because it
would incorporate activities that users currently perform and it is potentially simplifying them.
Regarding personalization, it should be targeted at preserving user empowerment in making
decisions about which media to consult.
Ranking of potential functionalities by interview participants
After the interview had ended, participants were given a list of functionalities and were asked
to rank them on the following scale:
1= not useful, 5 = very useful
Table 1 shows the results. The columns labeled with number contain the ratings from each
participant. The last column contains the average over all participants. The results closely
match the opinions expressed during the open questions part of the interview.
Functionalities 1 2 3 6 4 5 8 7 Average rating
Access different opinions about a topic in the news 4 4 4 4 3 4 5 5 4.1
Follow news about a particular subject throughout sources 3 5 3 4 5 5 5 1 3.8
Follow a news story throughout time 5 4 2 5 5 3 4 2 3.7

Functionalities 1 2 3 6 4 5 8 7 Average rating
Track what a person said in the past about a certain subject 3 2 5 4 3 4 2 5 3.5
Share news with your friends or colleagues 2 2 4 3 3 5 3 4 3.2
Get more information about people in the news 3 4 1 4 4 4 3.5 1 3.0
Follow what an expert thinks about a topic 2 2 3 3 3 1 5 5 3
Look up archived and past news 2 2 2 4 3 4 3.5 3 2.9
Get an information guide about the program's contents 1 2 3 4 2 4 3 3 2.7
Follow what your friends think about a topic 1 2 4 3 2 1 4 2 2.3
Table 1 Ranking of potential functionalities by interview participants scale: 1=not useful 5= very useful
3.2.5 Reflections about the interviews
The interviews allowed us to communicate with participants in a one to one basis, which lent
itself to personalized explorations into specific habits and beliefs. Interviews were based on
people’s memories as opposed to the focus group where participants were asked to react on
given contents. Therefore, people could more easily relate to the information needs they
were expressing in the interview. However, participants may value information differently
when confronted with real needs than they do when reflecting upon past needs. Therefore, in
the evaluation we complement the interviews with a task-based study and compare the
participant’s reflections with their actions.
Only one researcher coded the interviews. In the future, it would be advisable to include
more coders to avoid bias.
4 Conclusions about Context
4.1 Target group
One of the objectives of conducting the preliminary studies was to define the target group for
LinkedTV News, which was not known in advance due to the novelty of the application. After
comparing the results of both context studies, the main target user group was defined as
follows:
LinkedTV News is targeted at users between 25 and 45 years of age; highly-educated; who
like to be up to date about the international news, watch news broadcasts regularly; and own
a tablet computer or share it with someone in their household.
These conditions were defined on the following basis:
Education
According to PEW 2010 Media Survey, highly educated people spend more time with the
news: in 2010 in USA an average of 81 minutes daily [17]. In addition, according to their
2013 Tablet ownership report [20], they constitute the group most likely to own a tablet
computer compared to adults with lower levels of education.
Age
The age of 25 to 45 years corresponds to the age when users are “technology savvy” but
also are most likely to have the financial means to buy the kind of technology in which our
application would run (a tablet computer). According to the previously mentioned Tablet
ownership report [20], 37% of adults 25-34 and 49% of adults 35-44 own a tablet. This
percentage is higher than older and younger audiences.

Interest for news
LinkedTV News is a tool intended to assess users’ information needs regarding the news. It
is meant for people who want in-depth information and it is a TV companion. Thus, our target
group should be avid news consumers and regular newscasts spectators.
4.2 Choice of device
The preliminary studies (focus group and interview) indicated that for a user group of middle
age the preferred devices for second screen multitasking are laptops and tablets (see 3.1.4
and 3.2.4). It was also shown that the setting where people watch the newscasts is mostly in
the living room. Although from a distribution point of view, designing for a laptop could seem
like the best option (more people own laptops than tablets), from the ergonomic point of view
(users are sitting on a couch with the device on their lap) tablets seem to be a better choice.
Furthermore, according to PEW’s 2013 Tablet ownership report [20], tablet ownership has
doubled within the past year and is on the rise. Therefore, developing for this kind of device
seems to be a good investment. Finally, we consider that a touch screen based kind of
interaction is better suited for second screen multitasking because it requires less focused
attention than the mouse and pointer based one. These considerations led us to choose for a
tablet as a target device to design LinkedTV News.
4.3 Context of use
According to sections 3.1.3 and 3.2.4 and the results of the qualitative interviews, the
environment where users consume television broadcasts and, therefore where they will use
LinkedTV News, is the following:
Physical Environment
A living room (home environment) with a couch where users lay or sit possibly in the
company of another person (most likely partner) or a pet. Often users are eating or drinking.
Occasionally, users sit in front of a dining table. Users also sometimes watch the news in a
desktop environment (office or hotel).
Social or organizational environment
Users who live with others will most likely be in their company while they watch TV. Users
will likely discuss the topics of the news program with the people they’re watching with.
Accompanied users tend not to pause or rewind the program even if the technology they use
allows them to as opposed to users watching alone.
Users will often multitask while watching TV either alone or in company of others although
the latter are more reluctant to multitask.
Technical environment
The preferred device for watching newscasts is television, although, occasionally, PCs,
mainly laptops with or without external monitors are also used.
4.4 Personas
Based on the ethnographical data obtained during the studies, and in our defined target
group, two personas, Gina and Alonso were created as a communication strategy and in
order to give a name and a face to our users. Gina was adapted from LinkedTV’s deliverable
D3.1 [21] while Alonso is a completely new creation. See appendix E for a description of
these personas.

4.5 Scenarios
Two Scenarios were modeled after the context of use described above. They were written
taking into account the characteristics, behaviors and preferences of the users and,
therefore, the personas. Their intention was to envision the role of LinkedTV News in the
persona’s environment (current user environment) in order to extract requirements. See
appendix F for a description of these scenarios.
4.6 User requirements
The second phase of the user-centered design methodology, after the analysis and definition
of the context of use, is the specification of system requirements (Figure 1). This
specification is a product of the observation that took place during the first phase and will
guide the next phase, the production of design solutions.
Information about users’ potential requirements was gathered from the initial exploratory
studies (sections 2.1.4 and 3.1.4). Some of them were directly expressed by participants and
some of them were deduced as implications of their current practices. Requirements related
to information needs were given priority while requirements about social networking or
commercial transaction were not always taken into account because they are not within the
scope of this study. The following list of requirements summarizes the findings that serve as
constrains and guides for the design of LinkedTV News. This list of requirements with an
indication of the findings that originated them can be found in appendix G.
Requirements about information needs
1. Give in-depth information about the news on TV
2. Give an overview of past-related news
3. Illustrate a multiplicity of opinions about the news
4. Allow users to consult news from diverse sources
5. Build a link between the TV and user’s habitual newspapers
6. Update users about important changes in news
7. Give contextual information about the news setting
8. Give information about people and organizations mentioned in the news
9. Allow to access opinions of locals
General requirements
10. Programmed and designed for running on a tablet PC
11. Automatically synchronized to user’s program choice
12. Allow for group or single television watching
13. Enable synchronous and asynchronous modes of interaction (bookmarking)
14. Allow some hands free mode of viewing or interacting
15. VCR capabilities pause, replay, skip
16. Portable
17. Share information through the application
18. Notifications turned on/off on demand
Requirements about personalization
19. User-controlled personalization
20. Easy configurable with a couple of clicks

21. Selection of preferred sources or influence the ranking of sources
22. Selection of preferred categories
23. Personalization does not override the importance of the news
In the next section we will indicate the requirement that motivated each design decision with
an “R” enclosed within parenthesis.
5 Design solutions
After the requirements are gathered, and the context of use is defined, the next step in the
user-centered design cycle is to propose design solutions for the intended application (as
depicted in Figure 1). This section describes the design of Linked TV News, its concept
(Section 5.1), the process of refinement of its interface (Section 5.2) the look and feel of its
UI (Section 5.3), and the distribution of its contents (Section 5.4).
5.1 LinkedTV News, the design concept
LinkedTV News is a second screen application for tablets that acts as a companion to
viewers when watching the news broadcasts (R10, R16). Its main goal is to enrich television
newscasts by integrating them with other media thus, integrating and concentrating the
different activities related to getting informed about the news in one interactive multiple
screen, and potentially mobile experience (R5). It is designed to accommodate two viewing
modes in terms of interaction: a lean-back mode and a lean-forward mode.
The lean-back mode
In this mode, the application presents the user with summarized additional information about
elements of the news, for example, basic information about the location or people involved
(R7, R8). This happens in the form of a slideshow with short texts and images that acts as a
sort of news ticker to the broadcast appearing in the main screen. It automatically
synchronizes with the TV (R11), and requires no action from the user, but allows the user to
browse slides, save them or bookmark news for later exploration (R13). This lean-back mode
of the application intends to be as close as possible to a hands free and passive experience
(R14). The slides are presented without need for user interaction, so the user can relax,
share with her company and pay attention only to what really calls her interest while easily
bookmarking her favorite news for later reading (R12). This favors an asynchronous mode of
interaction that respects passive TV watching and allows for postponing activities that
demand more attention by using the TV program as an index for deciding what to explore in
a more suitable time.
The lean-forward mode
The lean-forward mode of the application acts as a hub where the user can access different
news sources at different levels of information depth and communicated through different
media formats (video, text, audio). This mode offers the following browsing alternatives:
 Follow a news story throughout time (R2).
 Follow news throughout sources (R4).
 Access different opinions about a topic in the news (R3).
 Geo-localized information (common people’s remarks, tweets, according to their country of
origin) (R9).
 In-depth curated information about a subject (R1).

Some simple manual configuration of preferences (personalization) is possible, and the user
can choose her favorite news categories and her preferred authors or news sources (R19,
R20, R21, R22). The application counts with an alert about important updates in users’
bookmarked news and about other relevant events (R6). Alerts can be turned on and off at
user’s will (R18). The application is loosely linked with social media through geo-localized
Twitter comments in the “global to local” section, and the possibility to share slides or articles
through diverse channels like email, Facebook or Twitter (R17, R9).
The application LinkedTV News synchronizes its content automatically with the television
without any action required from the users (R11). It has television remote control
functionalities (R15), so it is possible to use it for pausing or browsing the television program
as well as for skipping news sections or news stories.
LinkedTV News is different from news aggregators because it is structured around news
headlines other than a news category, a source, or trends in the news. Every headline can
be explored from different levels, in the lean-back mode by dissecting the headline in its
actors and settings, and in the lean-forward mode by showing it from different viewpoints.
Main Assumptions (obtained through the studies conducted)
 TV viewers value the possibility to decide whether to multitask while they watch TV or
simply sit and relax.
 TV newscasts viewing is regarded as a potentially social activity, while online newspaper
reading is an act best performed individually.
 The television newscasts are useful for the users in that they present a panoramic view
about the most important recent happenings, while online newspapers are better suited
for in-depth information searching.
How LinkedTV News addresses the challenges inherent to supporting hyperlinked media
In the introduction section 1.2 three challenges of web-enriched broadcast video are
mentioned. LinkedTV News addresses the first two of these challenges through the use of a
second screen as follows:
 Tension between lean-forward content and lean-back consumption. By allowing
synchronous as well as asynchronous interaction through the use of bookmarks and by
including two interaction modes, LinkedTV News gives its users freedom to decide about
how to distribute their attentive resources and which level of engagement they want to
have towards the enrichments of the TV contents. This facilitates a smooth transition that
will ease the tension between contents and the habits related to the context and medium
where the content are consumed (Web contents in a living room – television
environment).
 Tension between multiple and single users. By presenting enrichments in a second
screen, LinkedTV leaves the main screen free from overlays, thus keeping the
experience of multiple viewers undisrupted. It delivers a personalized version of the TV
contents and a personalized experience without directly affecting the shared one.
Through the use of bookmarks, users can choose to delay online newspaper reading
related to the broadcasted news, for a later moment and meanwhile enjoy the social
experience.
5.2 Prototyping
As part of the iterative design phase of the process, prototypes are created for
communication purposes with the team and stakeholders as well as for initial user testing.
5.2.1 Wireframes
Wireframes were created using a mockup tool, Balsamiq. The wireframes were discussed
with several project partners. Figure 5 below shows one of the Balsamiq wireframes that

were created. Using the feedback on the wireframes the concept and browsing routes were
approved.
5.2.2 User testing with screenshots and Paper prototyping
The updated wireframes served as the base for the design of static mockups with a polished
design that was specifically created for tablet browsing. These mockups were printed to
create paper prototype. These prototypes were informally tested with 4 users who were
asked to perform simple tasks: select a slide, swipe a slide, bookmark the current news,
share a slide, play, pause, and browse the video, and browse the in-depth information.
Although small changes were made after this iteration, the basic navigation metaphors were
well understood. However, it was hard to reproduce the conditions that happen
simultaneously when interacting with moving images and multitasking. These aspects would
be much better experienced for evaluation in the functional high-fidelity prototype.
Figure 5 Example of Balsamiq wireframes for LinkedTV News

5.3 LinkedTV News UI design
This section contains detailed descriptions of the UI and interaction design of LinkedTV
News.
Lean-back mode (slideshow screen)
When a user launches LinkedTV News and once the application has automatically synced to
the chosen program, the first thing that she sees is the lean-back screen, Figure 6.
The lean-back mode presents bits of information about key elements of the news item in the
form of slides, in a similar way to a photo carrousel. The elements of this screen are
distributed in 4 main areas: the menu area (A), the slide navigation or thumbnail strip area
(B), the main slide or information area (C), and the television area (D).
At the bottom of the screen, in the television area (D) can be found information about the
show, for example, the name of the program (1), and the news heading or story currently
being told (2). The TV control (3) is also in that TV area. Moving the slider control or tapping
at any point in the line can advance the program back or forth. The TV can also be paused
with the play/pause button. The bookmark button (4) is placed inside the TV area to the right
of the news heading to indicate that when it is pushed, the whole news heading or episode is
bookmarked (Figure 7). The bookmark icon was placed in this corner of the screen in order
to facilitate bookmarking with the thumb while holding the tablet with the hands. Thus making
it possible users to bookmark without need of looking at the app’s screen.
Figure 6 The five areas of the LinkedTV News lean back mode screen: the menu area (A), the slide
navigation or thumb strip area (B), the main slide or information area (C) and the television area (D)

Figure 7 Bookmarking a news story with the
thumb
Figure 8 LinkedTV News contextual menu
The information shown in the main slide is mainly taken from Wikipedia and it corresponds to
the first lines (short summary) that are standard in most Wikipedia pages and to the facts that
appear in the main fact rectangle at the right of the Wikipedia entries. The image frame is
intended for displaying an illustration representative of the entity.
The slides in the carrousel change automatically. The transition between one slide and the
next is a horizontal push transition where the new slide “pushes” the old one from right to left
and takes its place. It is also possible to browse through the slides manually and this is done
either swiping the slides or selecting slides in the thumbnails strip.
The pace at which the slides change depends on the number of slides available and the
duration of the news item, and it varies between 12 and 16 seconds. When users select a
slide, the slideshow pauses for 15 seconds, and then, if the system senses no activity from
the user, the carrousel advances to the slide coinciding with the current moment in the video.
The main slide area contains its own controls at the bottom right of the slide (8). These
controls are from left to right: the timer that shows how much time a slide will be shown in the
automatic mode of the slideshow, the pause/play button which can be activated to stop the
movements of the slides, the share button used to access the options for sending an email,
posting on Facebook or tweeting, and the save icon used for saving the slides within the
application.
The thumbnail strip area (B) consists of a series of thumbnails that correspond to all the
available entities in the news story or chapter (9). A triangle marker (10) indicates the start of
a news story, and therewith indicates which slides correspond to a news story. The
thumbnail strip acts as navigation and selection device, selecting a slide’s thumbnail
advances the carrousel to the point where the corresponding slide is shown. The selected
slide is indicated with a blue rectangle (11). The thumbnail strip shows one news story at a
time and thumbnails are loaded into it only when the beginning of the story is playing on TV.
When a news story has ended, the thumbnails of the next story substitute the existing
thumbnails. The thumbnails and slides remain in the left section of the thumbnail strip and
can be browsed through. In other words, it is possible to see slides from past news stories
but not future news stories. The navigation and selection in the thumbnail strip happens
through swipe and tap motions.
The menu area (A) is a bar on the top of the screen. It contains the menu icon (12), the
application’s name (13), the date (14), and the settings icon (15).
When tapped, the menu item (Figure 8) displays a contextual menu. This menu enables the
user to change screens from the lean-back mode to the lean-forward mode (“slideshow
mode” to “explore current program”) and gives access to the bookmarked news and to the
saved slides.

Lean-forward mode (in-depth browsing routes)
Figure 9 shows the lean-forward mode. Once the user has chosen the option in the menu to
browse more in-depth information (by selecting “explore current program”) she switches to
the lean-forward screen from where she can access all of the program news through five
different browsing routes or in five different sections. The lean-forward screen has a standard
design through the sections. Only the section for “global to local” has an alternative design.
Figure 9 LinkedTV News lean-forward screen: the news item column A, the article selection column B, the
contents column
The structure of the lean-forward screens is based in a three-column design as shown in
Figure 9. The left column (D) presents a list of the news headings. When exploring the
current program, these headings correspond to the program now playing. When exploring
bookmarked news, those headings correspond to the bookmarked news. To select an item of
this column the user taps on it. Selected items are highlighted (1). The column contents scroll
vertically when needed with a tap and drag gesture.
The center column (E) is formed by two main sub-sections: the menu tabs (2) and the article
selection list (3). The menu tabs present the 5 browsing routes. When the user taps on one
of the sections, the menu tabs below that section slide down to the bottom of the page to
uncover the selected section’s article list as shown in Figure 10. When the user taps a menu
tab that is already at the bottom of the page, the menu tab slides up to show its contents.

Figure 10 Menu tabs sliding down upon
selection through tapping
Figure 11 Example of airplay and video controls
The contents of the articles list vary depending on the selected section and the selected
news. However, all article cells (4), have a similar basic structure. This structure is composed
of a square thumbnail icon representing a person or the logo of a media source (5), a short
text (6) (usually containing the name of the article and the author or author’s details), date
(7), and a video or audio icon when the contents of the article are available in one of those
media formats (8). Identical to the news item column, to select an item of this column the
user taps on it. Selected items are highlighted. The column contents scroll vertically when
needed with a tap and drag gesture.
The third column of the lean-forward mode (F) presents the articles and media content. This
column has a title area (9) for the name of the article, the date and author of the article (10)
and a link to the article in its original source (11). When clicking on this link, the external
source is opened within the application. The app menu remains visible and a “close” button
appears to go back to the lean-forward screen.
The article and media content column also contains an area where video/audio
representations, or images illustrating the contents appear (12). This media frame or
container has a space for caption text when the caption is available.
When video or audio material is presented in this media container, playing controls appear
including an “airplay” button, Figure 11. When the “airplay” button is tapped, a contextual
menu offers the option of showing the media content on the main TV screen. Selecting this
option (by tapping the TV icon) interrupts the program being shown on the TV and replaces it
with the content of the tablet’s media frame. Tapping the button again and selecting the
tablet icon returns the media content to the application’s media frame and the TV returns to
the point of the video where it left.
The last area of the right column (F) is where the body or main text of the article is presented
(13). This area extends beyond the bottom of the screen as much as needed to adjust to the
length of the article and the user can scroll vertically to access the hidden parts of the text.
The whole article column can also be swiped horizontally. Swiping horizontally to the right
will show the next article on the list and to the left the previous one.
5.4 Explanation of the five browsing routes by contents
In other Media
The “in other media” section of the interface is dedicated to showing the selected news as
they are reported in other newspapers, radio, or TV programs (Figure 12). This section
should be devoted to the narration of facts rather than the presentation of opinions. The

system should, for this reason, look for close semantic match between all the terms in the
news headline itself and the articles to be included here.
In the implementation of the application, users should be able to decide which sources they
want to include or influence their ranking in case that the broadcaster uses a predefined list
of sources. This section is different from the “opinions” section in that it doesn’t emphasize
the author of the news, but its source.
Figure 12 In other media

Opinions
This section shows opinions from different authors about the selected news, see Figure 13.
In this section the most relevant element is the author. In the implementation of the
application, users should be able to decide which authors they want to include or influence
their ranking in case that the broadcaster uses a predefined list of authors. It should be
possible to include public figures like politicians as well as bloggers. This section
corresponds to the “opinion articles” section included in many newspapers.
Figure 13 Opinions

In-Depth
The “in-depth” information (Figure 15) includes in-depth coverage of articles related to the
selected headline that could form part of the lean-back mode “entities”, but offer a more
extensive view. It also includes archive articles. In-depth covers the details of the facts:
background, actors and their backgrounds, context, and prediction about possible future
outcomes. It also includes profiles of the persons involved in shown events.
Figure 14 In-depth

Timeline
The timeline section, Figure 15, shows an overview of the events related to the selected
news item chronologically. Ideally, the timeline should support different degrees of zooming
(years, months, days, or minutes) depending on the type of event. Sources and authors are
not relevant for this section. Timelines are often presented in well-known newspapers or
posted by lay authors.
Figure 15 Timeline

Global to local (geo-localized tweets)
This section includes live feeds from Twitter, Figure 16. The tweets included here, however,
are filtered by subject and by geo-location. The intention of this section is to show people’s
opinion in the place where the news happens as well as near the user’s current location.
The underlying motivation is to give an overview of the perceived local effect of the
international news in the user locality as well as a first-hand opinion of people located where
the story takes place.
This section is different from the rest in terms of interaction design because instead of the
article, the right column contains a map that can be zoomed and panned. The selection of
the information can be made by tapping on the center column or by tapping the Twitter icons.
Balloons with the tweets then appear as overlays to the map. Two buttons on the bottom of
this section with the legend “near me” and “near the news” serve as shortcuts to navigate the
map with one click to the current user location (near me) and read tweets of neighbors about
the subject of the news; or to the place where the event in the news occurs (near the news),
and read the testimonies and opinions of locals.
Figure 16 Global to local
6 Evaluation
Evaluating design solutions against user requirements is the last step of the user-centered
design cycle before deciding whether to consider the cycle concluded or restart a new
iteration as denoted in Figure 1. This section describes a task-based evaluation of the
LinkedTV News hi-fi prototype conducted with the aim of testing its usability, perceived
usefulness, appeal, and most importantly its ability to fulfill user- demands in terms of

information needs. Section 6.1 discusses the implementation of the hi-fi prototype. Section
6.2 the objectives of the evaluation; Section 6.3 its setup; Section 6.4 the method; Section
6.5 its participants; and Section 6.6 the results obtained.
Figure 17 Participant interacting with LinkedTV News during the evaluation
6.1 Videos, entities and related content
Five short news videos were selected and edited to form an eight and a half minutes long
newscast. An HTML5 functional website was created according to the InDesign screenshots
and specifications. This website was adapted so that it could seamlessly pass for a native
IOS application seen in full screen. Contents were added manually through json and HTML
coding to each of the news stories and each of the browsing routes. This resulted in 84 html
pages containing articles, 40 summarized descriptions and 467 image files. The look and feel
of the application obtained was very realistic except that the contents are not live and
therefore, not updated. The experience of feeding the interface with real contents helped to
adjust some of its characteristics such as space in the diverse text fields.
Due to time and technology constrains, not all the functionalities intended for a final
application were implemented and the ones implemented worked only for the horizontal
screen position. However, core functionalities were all included.
A list of the functionalities implemented in the hi-fi prototype can be found in appendix H.
6.2 Objectives
The objective of the evaluation was to investigate how users experienced the information
needs in practice, how these are supported by the prototype. In addition, we evaluated the
interaction with the application and the pertinence of the elements of the interface. The
following 12 points were used as items of interest to design the study:
Information needs

 How well does the system meet the user’s information needs and how can it be improved?
 Is the choice of information elements in the slides adequate?
 Is the information contained in the slides interesting?
 Are the five browsing routes of the lean-forward mode useful and suitable?
Interaction
 How do users perceive the experience of having additional information presented in a
second screen while viewing a television newscast?
 Is the pace at which slides are shown adequate for users to decide whether to consume
the information that they contain?
 Is the amount of information contained in the slides adequate?
 Do users prefer synchronous or asynchronous interaction with the application?
Interface
 Are the elements of the interface clear to the user?
 Are they adequate for achieving their intended goals?
 Is it easy to learn to navigate the interface?
 Is the application perceived as useful?
 Do users find interacting with the application pleasant and satisfying?
Tasks
In order to investigate the points described above, a task based evaluation format was
chosen. Three basic types of tasks were included: exploratory tasks, simulated information
needs tasks, and free formulated question tasks. For an overview of the task types and
examples see appendix I.
Participants were asked to perform some of the tasks synchronously and some
asynchronously in order to assess their experience in both cases. The actual evaluation was
preceded by a pilot in which we test the wording of the tasks, the length, and other aspects of
the protocol.
6.3 Set up
The evaluation was conducted in the same setting as the focus group, reported in Section
3.1.1, in a living room like experiment room provided with a sofa, a center table, and a large
television. The television was used to broadcast the news program and an iPad was given to
participants as a second screen. The application was installed and functioning in a full screen
mode in the tablet. Two cameras were included in the set-up: one situated on top of the
participant pointing at the second screen device and transmitting live feed to the researcher’s
computer and one on top of the television recording the general setup. The first was intended
for direct observation and creation of usability logs by the observer and the second one will
be used in future research (out of the scope of this deliverable) to analyze the direction of the
gaze while interacting with the application.
6.4 Method
After being greeted, users were given a task book. This task book contained consent forms
and demographic questionnaires in its first part and the rest were instructions for completing
6 tasks. Many of the tasks were followed by ASQs (after scenario questionnaires) to assess
the ease of fulfilling the task, the pertinence of the time demanded and how realistic the task
seemed to the user in terms of resembling real tasks that she would perform outside the
studio environment.

Each task and its corresponding instructions were explained in one page and participants
were asked not to turn the page until they were done with the task. Tasks were loosely timed,
there was no limit as to how soon a task could end, but participants were asked to stop
solving a task after having invested five minutes in it. At the end of the task book two
questionnaires were added: the SUS (System Usability Scale) [22] and the USE [23]
(Questionnaire for User Interface Satisfaction). These assess perceived usefulness and ease
of use, learnability and satisfaction. Only a part of the USE was employed (perceived
usefulness and satisfaction) in order not to overload users with questions. Furthermore, SUS
already measures learnability and ease of use and is a more widely employed and validated
tool. After finishing the questionnaires a follow up interview about the experience was
conducted by the researcher and recorded for analysis.
At the end of the evaluation, participants were rewarded with cinema vouchers. Each session
lasted approximately one hour.
6.5 Participants
This study constituted a follow up to the initial interviews, and the participants are the same.
Figure 18 Researcher's view of participant (usability study)
6.6 Results
6.6.1 Usability and ease of use (SUS scores)
SUS, System Usability Scale, is a widely validated test. It was designed to measure one
single dimension, ease of use. However, recent studies have suggested that it assesses also
learnability [24].
SUS yields a single number representing a composite measure of the overall usability of the
system being studied [22]. The scores for individual items are not meaningful on their own.
The scale of the averaged items ranges from 0 to 100.
The mark obtained by LinkedTV News in the SUS questionnaire is 83 with a standard
deviation of 10.

The individual participant average rating can be seen in Figure 19.
Figure 19 individual participants' SUS score
The average SUS systems grade is 68% [24], and therefore the obtained grade of 83 is
above average. Furthermore, it corresponds to the highest grade, a letter A in the letter
scale. For a complete table with details about the SUS scores in the evaluation see appendix
J.
6.6.2 Perceived usefulness (USE scores)
The perceived usefulness of the system was assessed through 6 items of the USE
questionnaire using a 7-point Likert scale where the highest values are the most positive.
The resulting score was 5.6 with a standard deviation of .7. This would correspond to a
percentile of 80. Table 2 shows the questions and the averaged user responses.
USEFULNESS
1 It helps me to find information more efficiently. 6.2
2 It is useful. 6.2
3 It makes the things I want to accomplish easier to get done. 5.8
4 It saves me time when I use it. 6
5 It meets my needs. 5
6 It does everything I would expect it to do. 4.5
Table 2 Usefulness
As can be seen the higher marks are for the efficiency of the system that is perceived as
time-saving.
The lowest scores, although still tending towards the positive side of the scale are the ones
related to user expectations regarding the system and its functionalities. For a complete table
with details about the USE usefulness scores in the evaluation see appendix K.
6.6.3 Satisfaction (USE scores)
The user satisfaction with respect to the system was assessed through 7 items of the USE.
The resulting score was 5.5 with a standard deviation of .5. This would correspond to a
percentile of 78%. Table 3 shows the questions and the user’s averaged responses.
SATISFACTION
7 I am satisfied with it. 5.7
8 I would recommend it to a friend. 6.3
0
20
40
60
80
100
P1 P2 P3 P4 P5 P6 P7 P8

9 It is fun to use. 6
10 It works the way I want it to work. 5.1
11 It is wonderful. 4.7
12 I feel I need to have it. 5
13 It is pleasant to use. 5.7
Table 3 Satisfaction
For a complete table with details about the USE satisfaction scores in the evaluation see
appendix L.
6.6.4 Assessment of the slides
In order to assess the suitability of the slides and their components, the participants
completed a questionnaire composed by opposite adjectives. The participants received this
questionnaire after the task where they were presented with additional information in the
slides of the lean-back. The following are the averaged results of the questionnaire. In this
questionnaire a 6-point Likert scale with no neutral or “undecided” position was used.
The information contained in the slides was found to be useful, interesting, and easy to
browse. Figure 20 summarizes the overall results. On the other hand the simultaneous
presentation of TV newscast and the slides was found to be distracting and the pace in which
the slides appeared was described as somewhat fast.
1 2 3 4 5 6 Average StdDev
Useless Useful 4.8 0.8
Trivial Interesting 4.8 1.0
Distracting Engaging 3.0 1.0
Cumbersome Easy 5.6 0.5
Excessive Info Lacking Info 3.6 1.0
Fast Pace Slow Pace 2.9 0.8
Value
Figure 20 Averaged results of the slideshow questionnaire
6.6.5 Perceived mental effort
Task number 3 was a simulated need task where users were requested to look for certain
information while at the same time watching the television program. After completing the
task, they were asked to rate the activity through 2 pairs of opposites in a 6-point Likert
scale.
It should be noted that participants found it easy to browse the information, but at the same
time they tend to describe the activity as distracting as shown in Figure 21.
Figure 21 Evaluation of the multitasking search task
1 2 3 4 5 6 Average StdDev
Distracting Engaging 2.8 1.0
Difficult Easy 4.0 1.3
Value

6.6.6 Ability to answer freely formulated questions
In the free formulated question tasks every participant asked a question about 2 topics
voluntarily selected from the ones included in the study newscast. Sixteen questions were
formulated, and participants found answers for ten of them within the application. This means
that 62% of their questions were found without the need to consult the Internet in the given
amount of time. From the remaining six questions, four were found with ease through
Google, and one was partially answered by a Google suggested link and one was not found
in the given time. Table 4 shows a list of the questions and indicates whether LinkedTV
News answered them according to the participants.
Participant
Number 1 2 3 4 5 6 7 8
News choice 1 Snowden
Fukushim
a Snowden Snowden Snowden Snowden
Fukushim
a
Fukushim
a
News choice 2
Zimmerma
n
Zimmerma
n
Fukushim
a Zimmerman
Fukushim
a Egypt
Royal
baby
Fukushim
a
Question 1 Short
overview
of what
happened.
What kind
of person
was the
preacher?
What are
his
political
positions
(also on
other past
events)
More
technical
backgroun
d
More
NSA
related
story
Who is
Anatoly
Kucherena?
Where
else did
he apply
for
asylum?
Details
about
extraditio
n treaties
between
US and
Russia
When did
the leak
start?
Map of
current
radioactiv
e levels
of Japan
Q1 answered? no yes yes yes yes yes yes no
Question 2 Snowden:
Legal
conditions
for
applying
for an
asylum in
Russia?
More
opinions
How big
is the
leak
Where is
Zimmerman
's family
from?
How do
they think
to solve
the
problem?
What
were
Morsi's
political
decisions
that
made the
process
start
What is
the baby's
name?
Interview
s with
Japanese
living in
Japan
about the
situation
Q2 answered? yes no yes yes no no yes no
Table 4 free formulated questions
6.6.7 Realism of the tasks
Participants’ evaluation of the tasks in terms of their realism was very positive, on average
5.5 of realism on a 6-point scale, representing 91% of perceived realism.
6.6.8 Completion time tasks
The average satisfaction with the time that it took to solve the tasks using LinkedTV News
was 4.9 on a 6-point scale equivalent to 81.8%.
6.6.9 Usability findings
According to the usability logs and observations, the following findings were extracted from
participant interactions with the interface. These findings will be taken into account as
opportunities for improvement:

 The difference between the concept “save” and “bookmark” is not clear, and users are not
always sure of whether their bookmarking will affect the entire news item or just a slide.
 Some users would like to have the VCR controls for the TV present at all times, also when
exploring the in-depth information.
 It is not clear to many users what the wording in the title “global to local” refers to.
 The buttons in the “global to local” section are not clearly visible and therefore not often
used.
 Some users expect the slide thumbnails to act as markers to navigate the video.
6.6.10 Results of the interviews
Each user evaluation session of LinkedTV News ended with an open interview guided by a
15-item questionnaire. The interview assessed participants’ general impressions after
interacting with the application as well as their suggestions for improvement. The main goal
of the interviews was to recognize whether user requirements identified in the preliminary
user study had been fulfilled. The fact that the participants were already acquainted with the
researcher and some of the research goals due to the previous interview facilitated their
involvement in a dialogue that resembled a co-creation exercise.
The interviews were summarized and transcribed, and comments were clustered according
to affinities. We present the most interesting ones.
Positive aspects
 Overall, participants’ reception of the application was very positive. It was described as
useful and easy to use. 7 out of 8 participants declared that it had met or surpassed their
expectations (all but P7).
P8 I like it, it’s useful. It's so much better to have this tool than not to have this tool. The
best is the easiness, it’s easy to use; it gives you easy access to different kinds of
information with just one click.
 7 out of 8 participants stated that they would use it if it were available (P1, P2, P3, P4, P5,
P6, P8).
P8 Yeah I would download it for sure and use it if it’s matched to my news channel I would
definitely use it.
 Users appreciated the time-saving aspect of the interface (P3, P5).
P3 The list with limited options is of course much better because it’s quicker. Something
with limited options, it limits you, but on the other hand it’s quicker so there is a balance,
it's your choice.
 They were satisfied with the density of the information that it contains (P1, P4, P6).
P6 It really covers most of the things that, usually, the things that you can look for when
you look up a news item.
 All participants appreciated the overview of the temporal background of the news through
a timeline, and the opportunity to have multiple views and multiple opinions of a subject
gathered in one place.
P4 The timeline I found very interesting. Sometimes we hear about the news late, and
miss things in between. The timeline updates you about the process of the news.
 Users indicated that they would be likely to use the bookmark functionality to be able to
postpone browsing for the time during the advertisement break or after the show. They
also said that they would browse during the show if the following news was not interesting
(P1, P2, P3, P4, P5, P7).

P7 I would bookmark, both things at once are not compatible. I second screen multitask
only when I am not interested.
 Many participants described the presentation of short fragments of information through
slides simultaneously with the broadcast as useful and interesting (P4, P5, P8).
P5 I liked the additional information. Sometimes information is needed to understand or go
in depth in the news. I saw a person in the Snowden story and I didn't know who he was,
but I clicked the slide and found basic information to calm my curiosity and help me
understand.
Negative aspects
 Overall, participants perceive interacting with the application simultaneously with watching
the news as distracting (P1, P2, P4, P5, P7).
P2 I like the functionalities but not the distraction.
P2 I prefer a screen where nothing moves and I request info. If something moves you
have to watch.
 Global to local (geo-localized tweets) was found to be the least useful functionality (P1,
P2, P3, P4, P5, P6, P7, P8).
P5 Least useful functionality for me: tweets, though for other people might be good.
 Some participants felt that the time to view the slides was too short (P5, P6, P7, P8).
P8 I don't like to have a specific amount of time to read if I select something.
 The difference between “save” and “bookmark” is not clear (P1, P2, P4, P5, P6, P7).
P1 Had confusion at the beginning between bookmark and saving.
Suggested functionalities and improvements
 Use static slides or use a more subtle transition for the slides (P1, P2, P7).
 VCR TV controls visible at all times (P6, P8)
 Control the video with the slides (P2, P6)
 Allow users to deviate in multiple directions (like the related articles in the newspapers)
(P3, P4, P5)
 Read people's comments (as in online newspapers) (p6, p7)
 Map to place of the news geographically (P5, P6)
6.6.11 Proposed changes
In a future version of LinkedTV News, in order to improve the interaction, the following
changes should be performed:
 The TV area should become permanently visible and easily accessible in both main
screen modes (lean-back and lean-forward).
 The difference between “save” and “bookmark” should be clarified and more widely
understood icons should replace the current icons. Otherwise, a label should reinforce the
icon meaning.
 The functionalities of the VCR controls of the television area should be expanded to “skip
chapter” or to navigate by chapter.
 The thumbnails should allow two modes of interaction, one where selecting a thumbnail
affects only the second screen slide show, and one where selecting a thumbnail advances
the main TV program to the point where the object depicted on the thumbnail is
mentioned. Such double functionality should be thoroughly tested.

 “Related articles” should be added as a browsing route or embedded as a sidebar or
footnote to the articles.
 The section “global to local” should be renamed. A good strategy would be to use a card
sorting evaluation to find consensus among participants. The buttons of this section
should be relocated for clarity. If further evaluations prove that this section is still not used
despite the changes, the section should be eliminated.
 A “search” field to perform searches outside the application could be considered, but this
requires further study.
7 Conclusion
The work described in this deliverable was driven by two motivations: our primary goal was
to design a communication "device" to understand user information needs and integrate
them in a technology-driven process. A secondary aim was to explore the role of a second
screen interface in coping with the challenges that are inherent to the presentation and
consumption of hypermedia.
The LinkedTV News companion is an example of the application of hyperlinked video and
LinkedTV's technology to the news. It integrates two activities that are related by subject, but
currently often take place through different devices and at different times: watching TV
newscasts and consulting online newspapers and videos. These activities are not mutually
exclusive. Our contextual research shows that users frequently perform both because they
believe that each one of them offers different advantages, for example, some of them watch
TV broadcasts to get an overview of the news, and read online newspapers when they want
to get in-depth information about specific news.
Our task-based evaluation with a hi-fi prototype showed that the proposed integration is
appealing to users who are interested in news and daily dedicate a certain amount of time to
get updated.
Participants in the study were generally enthusiastic to have different types of information
related to the news concentrated in one application and considered that, in this sense, the
interface would be time-saving with respect to their current equivalent practices (e.g.
consulting several independent sources). The proposed interface sections were described as
very thorough in covering possible user information needs. In particular, the most valued
sections were:
 the chronological overview of news delivered through a timeline
 the opportunity to read the same news headline as it was presented in different media
sources.
 Participants also valued the possibility of news bookmarking. They would use it to
postpone the more cognitively demanding tasks related to in-depth information
exploration for after watching the television program.
The possibility of accessing geo-localized information through tweets was, unexpectedly, the
functionality that participants liked the least.
In terms of understanding user needs, evaluation with a hi-fi prototype showed that the
application succeeds in fulfilling many of the needs identified in the preliminary study.
Overall, there seems to be interest from users in an integrated hypermedia solution for the
news that resembles the one we proposed, and there is reasonable evidence (see Section
6.6.10) to expect that an application like this would be welcomed by users.
Furthermore, through LinkedTV News a continuing dialog was established with the members
of the LinkedTV team about what types of links users would be interested in and how to deal

with choosing relevant subsets of them, thus, contributing to guiding the next technological
steps of the research project.
With respect to understanding the role of a second screen interface in hypermedia
presentation, the goal was partially fulfilled. From the two challenges of web-enriched
broadcast video presented in the introduction, consuming active contents through a passive
medium was most influential in designing the application. The lean back mode of the
application was conceptualized to accommodate the passive mode of interaction that is usual
when watching TV. The intention behind this was to allow users to choose which level and
type of engagement (active or passive) they want to employ in interacting with the
hypermedia enrichments.
Though most users valued the additional information presented through a slideshow during
the program, and understood that they could use the application to delay the in-depth
exploration, they complained about being distracted by the spontaneous movements that
occurred on their second screen. These movements forced them to look at the second
screen involuntarily, thus introducing an element of disturbance in their peripheral view and
their experience of media consumption.
8 Discussion of current work
The initial contextual studies were performed with nineteen participants in total. However, we
decided to continue working only with the eight who more closely matched our target
audience. We performed in-depth open interviews with each of them, and gathered
information about their experiences and their information needs. They suggested
functionalities and helped envision an application that would potentially fulfill those needs. In
a second iteration, they assessed to what extent the application met their initial motivations
and validated our proposal.
This iterative approach with one group of users lent itself to in-depth exploration bordering on
a co-creation exercise. This in-depth character of the study was intended to compensate for
the small number of participants. There is a possibility, nonetheless, that the familiarity of the
users with the project might cause some bias. Therefore we recommend testing the
application with a group of independent participants in order to compare results.
Another element that could be improved is the temporal aspect of the news used in the
study. The news articles were manually fed in the system by a process that took a few
weeks, and thus they were outdated as the evaluation was performed. Ideally, the evaluation
should be repeated with “fresh”, up-to-date news.
9 Future work
In order to explore further the role of a second screen and the cognitive load that a dual
screen set-up demands, the lean-back mode screen should be studied separately in a series
of controlled experiments. The number of enrichments, length of the information contained,
pace in which they change, topics to which they refer, colors and animations they use should
be studied in order to prevent overload of users’ attention.
Group evaluations should also be performed in order to test the performance of the second
screen application when watching the news collectively. This refers to the second challenge
of supporting Web-enriched broadcast mentioned in the introduction (Section 1.2).
LinkedTV News could be turned into a valuable commercial application if the following
conditions are met:

 Lean-back and lean-forward modes are both available and switching between them is
easily possible.
 The cognitive load of the lean-back mode with respect to peripheral vision is reduced.
 Necessary link types are implemented and high quality information is provided.
 Accurate synchronization of the second screen with the television is achieved.
 Easily configurable user-controlled personalization is enabled.
10 References
1. Pérez Romero, L., Traub, M. C., Leyssen, M. H. R. 2013. Hardman, L., Second Screen
Interactions for Automatically Web-enriched Broadcast Video. In Workshop proceedings
of EuroITV 2013, Como, Italy. TO APPEAR
2. Hauptmann, A. G. and Witbrock, M. J. 1997. Informedia: News-on-demand multimedia
information acquisition and retrieval. In Intelligent Multimedia Information Retrieval, AAAI
Press, Cambridge, MA.
3. Dowman, M., Tablan, V., Ursu, C., Cunningham, H., And Pobov, B. 2005. Semantically
enhanced television news through web and video integration. In Proceedings of the
Workshop on Multimedia and the Semantic Web at the European Semantic Web
Conference.
4. Ardissono, L., Kobsa, A., and Maybury, M., (eds.) 2004. Personalized Digital Television:
Targeting Programs to Individual Users. Springer Verlag, London.
5. Lindstedt, I., Löwgren, J., Reimer, B., and Topgaard, R. 2009. Nonlinear news production
and consumption: A collaborative approach. ACM Comput. Entertain. 7, 3.
6. Olsen, D. R., Bunn, D., Boulter, T., and Walz, R. 2012. Interactive television news. ACM
Trans. Multimedia Comput. Commun. Appl. 8, 2.
7. Pulse, https://www.pulse.me/ [Accessed: 10 Sep 2013]
8. Flipboard, https://flipboard.com/ [Accessed: 10 Sep 2013]
9. Hanson, G., Haridakis, P. 2008. YouTube Users Watching and Sharing the News: A Uses
and Gratifications Approach. JEP, Journal of Electronic Publishing 11, 3.
[http://www.journalofelectronicpublishing.org/]
10. Lockhorn, J. 2011. Forget Mobile, Think Multiscreen in Razorfish Outlook, Report 10
[http://razorfishoutlook.razorfish.com]
11. Basapur, S., Harboe, G., Mandalia, H., Novak, A., Vuong, V., and Metcalf C. 2011. Field
Trial of a Dual Device User Experience for iTV. In 9th European interactive TV conference
(EuroITV), pp. 127-135. ACM.
12. Fleury, A., Pedersen, J. S., Baunstrup, M., & Larsen, L. B. 2012. Interactive TV:
Interaction and Control in Second-screen TV Consumption. In Adjunct proceedings of the
10th European interactive TV conference (EuroITV), pp. 104-107. Fraunhofer Institute for
Open Communication Systems, FOKUS.
13. Cesar, P., Bulterman, D.C.A., and Jansen, A.J. 2008. Usages of the Secondary Screen in
an Interactive Television Environment: Control, Enrich, Share, and Transfer Television
Content. Springer-Verlag Berlin Heidelberg.
14. Cesar, P., Bulterman, D.C.A., and Jansen, A.J. 2009. Leveraging the User Impact: An
Architecture for Secondary Screens Usage in an Interactive Television
Environment. Springer Multimedia Systems Journal 15, 3, pp127-142, Springer.
15. IntoNow, http://www.intonow.com/ci

16. Goodman E., Kuniavsky M., Moed A. 2012. Observing the User Experience, A
Practitioner's Guide to User Research, Morgan Kaufmann (eds.)
17. PEW 2010 Media Consumption Survey; Princeton Survey Research Associates
International for the Pew Research Center, Princeton, NJ.
18. Nielsen 2012. Nielsen State of the media: cross-platform report Q2 2012
[http://www.nielsen.com/us/en/reports/2012/state-of-the-media--cross-platform-report-q2-
2012.html]
19. Maguire, M. 2001. Context of Use within usability activities, Int. J. Human-Computer
Studies 55, pp 453-483
20. PEW 2013. Tablet Ownership Report. Princeton Survey Research Associates
International for the Pew Research Center.
21. Leyssen, M., Hardman, L., van Ossenbruggen, J., Baltussen, L., Abreu, N. 2012.
LinkedTV Deliverable 3.1: Specification of Functional Requirements Satisfying User
Information Needs, CWI, Amsterdam, NL
22. Brooke, J. 1996. "SUS: a "quick and dirty" usability scale". In P. W. Jordan, B. Thomas, B.
A. Weerdmeester, & A. L. McClelland. Usability Evaluation in Industry. London: Taylor and
Francis.
23. Lund, A.M. .2001. Measuring Usability with the USE Questionnaire. STC Usability SIG
Newsletter, 8:2
24. Lewis, J.R. & Sauro, J. 2009. The factor structure of the system usability scale.
International conference (HCII 2009), San Diego CA, USA.

11 Appendix A: News Items presented in the focus group
NEWS SECTION FILE NAME AND TOPIC Track Length
International News Summary One-minute World News.FLV 1 00:01:00
International News Floods wreak havoc in Mauritius.FLV 2 00:01:15
Bird Flu in China
Politics N Korea in 'state of war' with South.FLV 3 00:01:26
Local News 'Psychedelic' tulip show in the Netherlands.FLV 4 00:02:33
Arts and Entertainment Iran planning to sue over Argo.FLV 5 00:01:44
Science and Technology Touchscreen ‘that reacts even without touching’ .FLV 6 00:01:08
Sport Messi UEFA Best Player of the Year.mp4 7 00:01:30
Finance/business Why do banks go bust?.FLV 8 00:01:10
Human interest story Human parents for a baby gorilla .FLV 9 0:01:17
Weather Europe weather forecast.FLV 10 00:01:20
TOTAL 0:14:23

12 Appendix B: Focus group protocol
WORKSHOP/FOCUS GROUP APRIL 5 @CWI – BY LILIA PÉREZ AND KRISTINA ANDERSEN
 Goal: gathering input and inspiration for formulating a concept design (high-level
functionalities for a second- screen companion intended to deliver additional information
while watching the news broadcasts).
 Total duration: 45min
 Location: living room setting, Pampus room M366 at CWI
 Reward for participants: digital voucher for Pathé movies.
 Date and time: 5 of April from 13 to 17h.
 Target user profiles of focus group participants:
o 12 users (4 groups of 3) or 16 users (4 groups of 4). At least 60% of them should be
TV newscasts viewers. The other 40% can be on-line news readers even if they don’t
regularly watch the news.
 Who: two moderators, one acting as host (Kristina) and the other one as technical facilitator
(Lilia)
INTRODUCTION / COLLECT PARTICIPANTS OUTSIDE ROOM (5 MIN PREVIOUS TO START)
(KRISTINA + LILIA)
1) A poster is placed outside the room to indicate that participants have reached the right location.
It indicates that they should wait there for the others to arrive.
2) Participants wait outside the room or are collected at the entrance of CWI (external visitors).
START OF WORKSHOP
FILLING OF QUESTIONNAIRE AND CONSENT FORM (10 MINUTES)
3) Once all participants have been gathered outside the room, they are introduced in the living
room (by both researchers) and given:
a) Consent form
b) Brief questionnaire
 Documents obtained: filled questionnaires and consent forms
WELCOME AND INSTRUCTIONS (2 MINUTES)
KRISTINA
4) After collecting the questionnaire, participants are shortly briefed by the host. They are then
given a pack comprised by pens, colored post-its, and a folder with the following documents:
a. Blank A4 sheets of paper with mock up frames of ipad, iphone and laptop, labeled
according to the videos (annex C, D, E)
VIDEO + ADITIONAL INFORMATION EXERCISE (15 TO 20 MINUTES)
KRISTINA AS HOST + VIDEO DOCUMENTER / LILIA AS VIDEO JOCKEY AND TECHNICAL FACILITATOR
5) Participants are asked to take a seat on the sofa in front of the TV with the pens and the blank
mock up frames ready. They then receive the following instructions: You will now watch 6 to 10
blocks of news (see annex F for a specification of the video news fragments). Each one lasts
between 1 to 3 minutes.

a. Please think about the additional information that you would like to have about each
piece of news. Feel free to ask, it doesn’t matter how difficult or how strange the
request for additional information seems.
b. Choose the blank sheets with the mock up frame corresponding to the device that
you think you would more likely use in real life to look for the answer to your
questions.
c. Write down the question on the blank paper. For each video, use the paper labeled
with the corresponding number.
d. After the video has finished playing, you will be given a couple of more minutes to
finish writing the questions, so don’t worry if you don’t finish before the video does.
You can also choose to first watch and then write.
e. Rate the video from 1 to 5 stars according to your preference ( 1 low 5 is high).
f. If you have things that you would like to be able to do with the video besides looking
for information (for example, you would like to be able to bookmark the video) write
down those ideas too.
6) After each video host approaches the participant with a camera and asks him/her to explain, in a
mode similar to the think aloud method, what they wrote and why. Whenever a participant does
not feel comfortable with her face being shown on video, only her hands will be recorded.
 Documents obtained: lists with set of questions (types of additional information
desired) and an indication of second screen devices and sources used. Videos
rated according to preference. Video + audio documentation of the whole
process and video shots with people explaining their choices and needs aloud.
HOSTED BRAINSTORM: FUNCTIONALITIES (10 MINUTES)
KRISTINA AS HOST – VIDEO DOCUMENTER
7) Participants remain seated on the sofa. A flip board and markers are introduced into the living
room space. Participants are given the following instructions:
a) Reflect on the previous exercise and think about functionalities that you would like a second
screen companion for the news broadcasts to have. The functionalities should not be
constrained by what is possible in the present, by existing applications or by what you know
about technology.
b) Make a brainstormed list of functionalities with your group.
c) Vote within the group for your favorite functionality.
 Documents obtained: rated lists of desired functionalities.
WILDCARD : TELEVISION OF THE FUTURE (5 MINUTES)
KRISTINA AS HOST
8) Participants are asked to speak to the camera and briefly explain how the news of the future look
like for them, or how we will access the news in the future (Head shot, vox –pop, blue sky
thinking).
 Documents obtained: video snapshots of futuristic views on the news.

13 Appendix C: Extended list of focus group findings
1. Need for more in-depth specific information.
P6 About the bird flu - I was wondering just how expensive it is to make all those
vaccines, you know? I was just wondering if it is a good idea to even make a lot of it
or how hard it is to make.
2. Would like an overview related events that occurred in the past.
P7 About North Korea’s state of war - my search would be something like "South
Korea state of war ramp up” (…) I would like to know, what happened two years ago
and how they solved it…
3. Search for different opinions.
P7 About North Korea’s state of War - I'd start with Wikipedia, probably, and then I
always look at AP and Al Jazeera to get two different opinions…
4. Information about how this ends, updates when important changes occur
P8 About “Human parents for a baby gorilla”- I would like to know if the story
succeeded, if the gorilla was saved …
5. Local effect of global news – What is the impact in my life?
(4 comments / 2 participants + 2 comments in 1 brainstorm list)
P1 About the bird flu - Yes then I would look it up, what is the effect in the
Netherlands… (of the bird flu virus)
Secondary information needs expressed by participants:
6. More information about people in the news - the ones provoking the events (2
comments / 2 participants)
7. Information about watching a film (2 comments / 2 participants)
8. How can I help? (1 comments / 1 participants + 1 comment in 1 brainstorm list)
9. Access-related publications, documentaries and news from that time. (1 comments / 1
participant + 1 comment in 1 brainstorm list)
10. Opinion from a specific expert-thinker (1 comments / 1 participant + 1 comment in 1 brainstorm
list)
11. Access-related information selected by experts and suggested by broadcasters. (1
comments / 1 participant + 1 comment in 1 brainstorm list)
12. Information about who is behind the news (1 comments / 1 participant)
13. Where is that location? (1 comments / 1 participants)
Actions triggered by information needs
14. Google the information (12 comments / 6 participants)
15. Consult different sources (4 comments / 4 participants)
16. Look it up (no specified source) (3 comments / 3 participants)
17. Check the newspaper (printed) (2 comments / 2 participants)
18. Make a note for later sharing (2 comments / 2 participants)
19. Go to Wikipedia (2 comments / 2 participants)
20. Compare weather in two locations (2 comments / 2 participants)
21. Look up movie schedule (2 comments / 2 participants)

22. Wait to see the outcome (2 comments / 2 participants)
Sources currently used
23. Google (6 participants)
24. Wikipedia (3 participants)
25. BBC (2 participants)
26. US government (1 participant)
27. AP (1 participant)
28. Al Jazeera (1 participant)
29. Volkskrant (1 participant)
30. Buienradar (1 participant)
Functionalities mentioned during the brainstorm sessions
31. VCR capabilities, pause, replay, skip (5 comments / 2 participants + 3 brainstorm lists)
32. Bookmark for later viewing (2 comments/ 1 participant + 2 brainstorm lists)
33. Get an alert or update when something changes in that particular news (2 comments / 2
+ 1 brainstorm list)
34. Buying tickets, reserving (1 comment/ 1 participant + I brainstorm list)
35. See longer versions of stories that I like / or make playlist of similar stories – to keep
on watching (2 comments / 2 participants)
36. Share information like warnings (about bad weather) (1 brainstorm list)
37. Recommender system (1 brainstorm list)
Perceptions of technology
38. Perception of TV as a passive medium that will remain so
(3 comments / 2 participants + 2 in 1 brainstorm)
P9 - TV works very well because it sort of blocks you out. Second screens just
distract from just watching the TV. So although it's news it's also sort of a relaxing
moment if you don't have to do anything, don't have to think about it.
39. News in the future will not be linked to a certain space and time
(2 comments/2 participants)
P1 - I think that it's more that when you want to look the news you can everywhere
see the news and not at a specific time or place…
40. TV is outdated
(2 comments/2 participants)
P6 - I think that television in itself is kind of outdated

14 Appendix D: Qualitative interview findings
Attitudes toward media (TV and newspapers) based on habits
1. It is possible to be more selective and control time better when you are reading.
(P3, P4, P8)
P4 - When you read, you manage your time and read what interests you versus when
you watch a video you have to watch all the video, and you cannot select video stuff.
When you see a report […] one look and you can know what you need, and then it is
much faster to find information…
2. TV is sometimes perceived as a way to get a general overview of the news, and
newspaper a medium that goes more in-depth. (P1, P6)
P1 - I think the main difference (between news on TV and online newspapers) is that
when I read it I only read the articles that I am really interested in and I watch it on TV to
get a general overview on what is happening in the world.
3. Some users consider that watching video news is a social act while newspaper
reading is a solitary act (P2).
P2 - So having the video footage, that you have more in the television news, I think adds
value there, and maybe also the fact that you can watch it together, adds value, so when
you read the newspaper I think that you really read alone.
4. Newscasts lend themselves to a passive behavior and relaxation. (P8)
P8 - I guess this is one of the benefits of having just this 50 minutes newscast, that you
can have a really passive behavior, right? You can just be in bed, listen to everything and
you don't have to concentrate, just focus in what they are talking about. This is also nice.
Desired functionalities regarding information
5. Some users would like to have a link between TV and other sources (P1, P2, P6)
P5 - I think, I would expect that you would include like the main links to what they're
talking about like if it was about Hugo Chávez, links to Wikipedia or some articles about
him, the country, and stuff like that because I think that's the thing that I would look for
myself.
6. Some users would like to be able to compare different opinions about a
subject, different scenarios, and different sources. (P8, P7)
P8 - I think that the German news are still kind of OK, but still, I mean everybody is
biased in their opinion, so no matter which news you read you always have some bias. It
is good to have different and diverse opinions and first-hand information.
7. To access and compare the international view on local news. (P2)
P2 - Typically I will look up for some confirmation I think sometimes I'm interested for
instance on national news, how international news would cover the same thing.
8. Access opinions of locals, regular people, preferably friends that are where the
news take place (P8)
P8 When I think about this now, about this communication aspect, what really is
interesting for me is to talk with people preferably friends who are in the situation. For
example in Korea if there is this tension between north and south, I want to talk with
someone from there, to see like the real opinion of people there, because it's very
different if you have a news reporter talking about this and if you just have an ordinary
people talking about it.
Other desired functionalities (not related to information needs)
9. Notifications and the app itself should be turned on and off on demand. (P3,
P1, P5)
P1 - I think I would like to open it for myself and request information. I am a control
freak.
10. Bookmark news and delay reading. (P1)

P1 - Maybe, ahh some sort of bookmark, so that just like a red button that I could hit in
the moment that I think something is interesting and that I could get back to it later. And
then maybe, in the meantime it could collect some information for me.
11. P5 would like to follow a story through time, but would also like it to contain an
indication about what he has already seen. (P6)
12. When watching with others through Skype, P6 would like to have video
synchronization to be sure both persons are watching exactly the same. (P6)
13. Having a small lateral screen showing short information pieces about what appears in
the text. For example the context.
Personalization
14. Users prefer pull and user controlled instead of push.
15. In general a user-controlled personalization that is easy configurable and doesn’t
personalize regarding political inclinations, only categories or topics and sources
16. Personalization, but easy to configure, general categories and known sources.
17. People don’ t want to miss important information because of personalization.

15 Appendix E: Personas
Nina (Mother
of Lisa)
Occupation: teacher
Age: 32
Nationality / place of residence: German / Prenzlauer Berg
Digital literacy: interested in new media and “hip” technology
Nina is a typical inhabitant of Berlin’s hippest quarter, Prenzlauer Berg. She is a well-
educated and well-informed young mother. She really likes discussing things. She especially
likes to talk about politics and culture. Therefore, she likes to get deeper and not just
superficial information. When a subject is interesting for her, she will take the time to
understand it properly. She likes Berlin with its constant changes and she feels very much at
home in her family-friendly neighborhood. She likes to visit exhibitions and also to go to the
theatre and readings. Because she is very interested in culture, she would like to be informed
about the city life, current events and new galleries. She does not like it at all when things are
complicated and take a long time to understand. She can get impatient very quickly, so every
application or service has to be smooth and easy. Nina is not interested in technological
background information on how a system works, it should just work well and adapt to her life.
Time-independence is also very important for her, because of her little girl. She only has time
to watch infotainment programs whenever her daughter is asleep.
Alonso (Maria’s
husband)
Occupation: civil engineer
Age: 42
Nationality / place of residence: Spanish / Amsterdam, the Netherlands
Digital literacy: loves gadgets, has some knowledge of programming
Alonso is a civil engineer specialized in hydraulic systems. He lives in the Pijp quarter in
Amsterdam. He has a postgraduate degree in engineering and an MBA. He works for an
international company in the Netherlands. His job is well paid, and he likes it very much
although he really misses Spain. Alonso lives with his wife, Maria, who is also Spanish and
studied art history. Alonso is a workaholic and stays in the office until late. He uses a car as a
means of transportation when going to work. In Spain, Alonso used to listen to the news on
the car’s radio, but he can’t understand Dutch, so he can’t do that in his new place of
residence. Instead, every night at nine, he watches the Spanish news in TVE. He likes that
moment when he can relax and sit with Maria watching as they both have dinner.
Alonso is crazy about gadgets; he buys and tries them all. He also likes to read about tech
news in blogs like End Gadget or Gizmodo. He doesn’t have time to read his blogs or twitter
during his working hours, so he does so at home, mainly multitasking while he watches a TV
series. Alonso is also interested in politics and likes political satire, he reads El Pais daily in
the morning, while Maria is taking a shower. He secretly loves this time for himself.

Alonso and Maria have no kids. They like going to the movies together and like to travel a lot.
They both are fond of building things so often they get involved in bricolage projects and
decorate the house together.

16 Appendix F: Scenarios
The numbers within parenthesis indicate the requirement that each part of the scenario
relates to. The user requirements are found in section 4.6 of the report.
Urban mom
Nina's baby has fallen asleep after feeding, so Nina switches on the TV to get informed.
Browsing the television guide in her tablet EPG (electronic program guide), she sees that
"rbb AKTUELL" evening edition is available and starts the program.
She sits comfortably on her favorite couch with her tablet on her lap. LinkedTV News knows
what she has chosen and loads the enrichments for rbb AKTUELL (11). The first news is,
again, about the failed construction of the Berlin Brandenburg airport. While she listens to
what the anchor says, a picture of Willy Brandt shows up automatically in the LinkedTV News
enrichment screen (8). He is the Nobel Prize winner after whom the airport will be named.
There are a few things about him that she didn’t know and finds out with one quick look at
the short information on her tablet screen. He was a city mayor in the sixties, she wasn’t
aware, that was before her time, but now she knows. The next enrichment tells her about the
new company taking over the construction (8); she gets a nice and quick overview of whom
these people are and what they did before. Nina is amazed at how new this old news feels
when she uses her second screen companion.
Watching a news spot about Berlin's Green Party leader, Volker Ratzmann, who withdrew
from his office yesterday, Nina is kind of frustrated as she voted for him and feels her vote is
now "used" by someone she might not have voted for. She would like to know what other
politicians and people who voted for him think about Ratzmann's decision to resign. She
uses LinkedTV News to pause the TV (15), because she doesn’t want to lose any detail of
this story that concerns her a lot and looks for such information in her LinkedTV News
application sitting on her lap (13). She goes to the “opinion” section and checks what is
available there (3). She chooses to read one article from a special contributor for Der Spiegel
and watches a selection of video statements of politicians and voters. She finds all kinds of
opinions that help her understand better what is happening.
Watching a news spot about a debate on the mandatory change of uniform for police officers,
Nina wonders why the usual green uniform should be changed. Out of interest, she looks at
the application and browses the in-depth section (1) where she quickly finds the answer. In
the news over time overview (2) of the app she even finds out how German uniforms looked
before being green and who the designer of the current uniforms was. Some of the uniforms
are very classy. This makes her think of her friend Gama who is a fashion designer, so she
decides to share the article with him. She clicks “share” in the LinkedTV News app and
chooses to send him an email with the link (17).
Technology savvy couple
On Monday, Alonso comes back as always around 8:00PM in the evening. After greeting
Maria, who has also just arrived, they start cooking dinner together. Soon it’s time for the
evening news. They set the table in front of the TV and sit in order to eat and watch. Next to
them they place a tablet with the LinkedTV News application they use the smart cover so that
the tablet stands on the table within sight while they eat (14,12,16). The app synchs
automatically with the TV (11). Soon, his mood changes while he listens to the anchor
speaking about how king Juan Carlos got injured while hunting elephants during an
expensive trip to Botswana. The injury hadn’t been serious, but the trip for Alonso was a
serious matter. How can the king go on luxury vacations with his lover while the country is in
recession? Alonso wants to know what people in Spain think about this, and if anyone is as

angry as he is, but he is busy eating now, so he bookmarks the news and continues
watching (13).
The following news is interesting for Maria, it is about a movie that they both watch together:
Argo. The makers of the movie are being sued by the Iranian government who claims that
the film distorts the facts and the country’s image. Maria asks Alonso to bookmark this news
for her (13).
After dinner, Alonso starts his usual multitasking while the TV plays some comedy show. He
does so by browsing the LinkedTV News application’s twitter feed where he filters the tweets
in relation to a location: Spain (9). Because he is already browsing his bookmarked news, he
doesn’t need to filter the subject. There he finds many that think like him. He also consults
other media, and what they are saying about this news (4). He expects El Pais to agree with
him, but El Mundo for sure will be trying to speak in favor of the king (5). He always
compares those two newspapers of opposed tendencies that represent the two main political
parties. He used to do this before by going to the online versions of the papers, but know it is
much easier and faster. He configured EL Pais and El Mundo in his personal list of sources
(20,21,22) with a couple of simple steps and now it is all already there in LinkedTV News. He
has also included some international newspapers in his configuration, so he can even read
about how the rest of the world sees this royal scandal.
Maria postpones reading about Argo until the next day after work (13), but activates alerts
because she wants to know if anything changes while she’s distracted with other things (18).
When she is waiting for Alonso to come home, she gets a message alerting her of an update
in this bookmarked news (6). Apparently Isabelle Coutant-Peyre, a lawyer of a very
questionable reputation who also represented Venezuelan terrorist Ilich Ramirez Sanchez,
will act as Iran’s lawyer. “This is getting serious” she thinks, and sits to read as she waits.
Maria wonders to what extent the movie was really lying. She browses the in-depth
information in the application (1,2) and checks the timeline, and that allows her to decide for
herself.

17 Appendix G: Requirements related to participants’
comments
The following list shows the user requirements of LinkedTV News on the left column and on
the right column it states which comment or report section originated the requirement. The
letter A stands for Appendix, it is followed by the Appendix letter (e.g. Appendix C = AC) the
number refers to the number of item in that appendix that corresponds to the relevant
remark.
Requirements about information needs Related user comment
1. Give in-depth information about the news on TV AC1
2. Give an overview of past-related news AC2
3. Illustrate a multiplicity of opinions about the news AC3
4. Allow users to the consult news in diverse sources AC15, AD46, AD47
5. Build a link between TV and user’s habitual newspapers AD45
6. Update users about important changes in news AC4
7. Give contextual information about setting of the news AC13/AD53
8. Give information about people and organizations mentioned in
the news
AC6
9. Allow to access opinions of locals AD47
General requirements
10. Programmed and designed for running in a tablet PC Report Sections 3.1.3
and 3.2.4
11. Automatically synchronized to user’s program choice Report Sections 3.1.3
and 3.2.4
12. Allow for group or single television watching Report Sections 3.1.3
and 3.2.4
13. Enable synchronous and asynchronous modes of
interaction (bookmarking)
Report Sections 3.1.3
and 3.2.4
14. Allow some hands free mode of viewing or interacting Report Sections 3.1.3
and 3.2.4
15. VCR capabilities pause, replay, skip AC31
16. Portable AC39
17. Share information through the application AC36
18. Notifications turned on/off on demand AD49
Requirements about personalization
19. User-controlled personalization AD54
20. Easy configurable with a couple of clicks AD55
21. Selection of preferred sources or influence the ranking of
sources
AD55
22. Selection of preferred categories AD55
23. Personalization does not override the importance of the
news
AD57

18 Appendix H: List of the functionalities implemented in
the hi-fi prototype
Functionalities implemented in the hi-fi prototype:
Lean-back mode
 VRC type controls for the main screen (TV or PC) through a second screen controller
interface.
 Presentation of additional information in the form of thumbnails and slides synchronized
to the main screen program.
 Browsing through the slides
 Bookmark news
 Slide timer (to update slides when there is no activity after selection)
 Pause slide presentation
 Save slides (partially implemented)
 Menu
 Lean-forward mode
 Browse bookmarked news
 Browse current program
 The browsing routes:
 In other Media
 Opinions
 In-Depth
 Timeline
 Global to local (geo-located tweets)
 Manually browse map and select tweets
 Browse map through buttons
 “Near me” and “near the news” buttons
 Select a news heading
 Select a browsing route
 Go to original article
 Swipe through articles
 Airplay videos
 Menu
At this stage functionalities were implemented only in the horizontal screen mode though it is
contemplated to use also the vertical mode in the future.

19 Appendix I: Example of task types included in the task-
based study
Exploratory tasks
The purpose of exploratory tasks was to allow participants to get acquainted with the system
and to experience it freely without being directed. These tasks were combined with small
subtasks intended to evaluate specific elements of the interface.
…Today you are in a relaxed mood so you decide to sit and watch the television program and postpone other
activities.
 Watch the first three news items of the program without stopping or leaving the “slideshow” screen.
 Bookmark at least one news item that you consider interesting.
 Feel free to browse the slides, if you wish and to save the ones that you would like to explore later.
 When the three first news items are over, please press “pause” then return to this booklet and turn the
page. The researcher will remind you when to do this.
Please press play to start the task
Simulated information needs tasks
The purpose of simulated information need tasks was to evaluate whether the browsing
route’s purpose was clear and whether they match participants’ structure of thought when
performing an information seeking related task. Another goal was also to find out whether
participants could find their way through the interface in a given amount of time.
Find out the opinion of a local about the riots in Egypt:
You have heard a lot in the news about riots going on in Egypt. You know that there are two main groups of
people in conflict: those that are in favor of the president, and those who are against. However, you mistrust the
information presented by foreign media and would like to have some first hand information and testimonies from
common people living in Cairo.
See if you can find such information with the help of the application.
Return to this booklet and turn the page when you have completed the task. You have 3 to 5 minutes to find
information.
Free formulated question tasks
The purpose of free formulated question tasks was to approach as much as possible a real
life (non-study) situation in which the participants would have an authentic, self-motivated,
doubt about the news and the need to solve it, and to determine to which extent the
application would be useful to solve such doubts in a given amount of time. A secondary
intention of these tasks was to observe the sequence of actions that participants would take
to search for an answer on the Internet in case that they couldn’t find it in the application.
From this list of the news that you have just watched, please mark the two that you find more interesting.
[ ] Snowden applies for Russia asylum
[ ] Egypt's Morsi vows to stay in office
[ ] Fukushima leak causes Japan concern
[ ] Rallies in US over Zimmerman verdict
[ ] Royal baby prince named George
Which additional information would you like to have about these news items?
For each of the chosen news items, please write down a question or information need that you have:
Please turn the page when you finish the task.
(Next page) Please search the application for an answer to your first question.

20 Appendix J: SUS OF INDIVIDUAL PARTICIPANTS
Question P1 SUS P2 SUS P3 SUS P4 SUS P5 SUS P6 SUS P7 SUS P8 SUS
1
I think that I would like to use this website
frequently:
5 4 5 4 4 3 4 3 5 4 4 3 3 2 4 3
2 I found the website unnecessarily complex: 1 4 1 4 2 3 1 4 1 4 1 4 2 3 1 4
3 I thought the website was easy to use: 4 3 4 3 4 3 4 3 5 4 4 3 3 2 4 3
4
I think that I would need the support of a
technical person to be able to use this
website:
1 4 1 4 2 3 1 4 1 4 1 4 2 3 1 4
5
I found the various functions in this website
were well integrated:
5 4 2 1 4 3 4 3 4 3 3 2 3 2 3 2
6
I thought there was too much
inconsistency in this website:
1 4 4 1 2 3 1 4 1 4 2 3 1 4 1 4
7
I would imagine that most people would
learn to use this website very quickly:
4 3 4 3 4 3 5 4 5 4 4 3 3 2 5 4
8
I found the website very cumbersome to
use:
1 4 2 3 2 3 1 4 1 4 1 4 2 3 1 4
9 I felt very confident using the website: 4 3 4 3 4 3 5 4 4 3 4 3 3 2 4 3
10
I needed to learn a lot of things before I
could get going with this website:
2 3 1 4 2 3 1 4 1 4 1 4 3 2 1 4
Individual SUS Score 90 75 75 92.5 95 82.5 62.5 87.5
P = participant scores
SUS = sus value

21 Appendix K: USE usefulness scores
USEFULNESS 1 2 3 4 5 6 7 8 Average
1 It helps me to find information more efficiently. 6 6 6 7 6 7 6 6 6.3
2 It is useful. 7 6 6 6 7 7 5 6 6.3
3
It makes the things I want to accomplish
easier to get done. 6 6 6 6 6 6 5 6 5.9
4 It saves me time when I use it. 7 7 4 6 6 6 6 6 6.0
5 It meets my needs. 5 5 4 5 6 6 4 5 5.0
6 It does everything I would expect it to do. 7 5 3 4 4 5 3 5 4.5
22 Appendix L: USE satisfaction scores
SATISFACTION
7 I am satisfied with it. 7 5 5 6 6 6 5 6 5.8
8 I would recommend it to a friend. 7 6 5 7 7 7 6 6 6.4
9 It is fun to use. 6 6 6 7 7 5 6 5 6.0
10 It works the way I want it to work. 7 5 4 5 5 6 4 5 5.1
11 It is wonderful. 6 4 5 6 6 4 3 4 4.8
12 I feel I need to have it. 6 4 5 6 6 5 4 4 5.0
13 It is pleasant to use. 7 5 5 5 7 6 5 6 5.8

Requirements document for LinkedTV user interfaces

More Related Content

What's hot (20)

Viewers also liked (6)

Similar to Requirements document for LinkedTV user interfaces (20)

More from LinkedTV (15)

Recently uploaded (20)

Requirements document for LinkedTV user interfaces