SUPPORTING NEWS READING AND DISCOVERY THROUGH
DIRECTED ONSITE INFORMATION SHARING
Pia Tukkinen
1
, Mikko Pohja
1
, Katarina Segerståhl
2
, Suvi Silfverberg
2
and Petri Vuorimaa
1
1
Department of Media Technology, Aalto University, P.O. Box 15400, 00076 Aalto, Finland
2
Helsinki Institute of Information Technology, Aalto University, P.O. Box 19800, 00076 Aalto, Finland
Keywords: Social Link Sharing, Reading News, Social Navigation, Human Computing.
Abstract: Sharing has become an integral part of online content discovery. In many cases, content is relevant to
members of a close-knit group, rather than all Facebook friends or Twitter followers. To explore emerging
practices of directed content sharing, we implemented an integrated sharing platform associated with an
online newspaper that allows members of close-knit groups to share and discover articles onsite as a part of
their news consumption. The application was evaluated in a field trial. Participants reported lower sharing
barriers and increased sharing frequency. They perceived value in sharing and receiving personally targeted
news content. Based on these outcomes, we present design recommendations for such systems. Specifically,
we contribute to the design requirements of group interfaces and directed information sharing.
1 INTRODUCTION
Although information has never been so easily
accessible, many people feel overwhelmed by the
amount of information available (Purcell, 2010). A
wide variety of tools have been developed to aid in
information retrieval and discovery, and such tools
often emphasize personal information needs.
However, in many cases, information is relevant to a
dyad or a group. It has been shown that people often
rely on friends or colleagues to pass relevant links
along as they encounter them (Bernstein, 2010). In
the future, content discovery will increasingly rely
on sharing and suggestions (Purcell, 2010).
Recent studies (Gupte, 2009); (Leino, 2011)
indicate that Social Network Services (SNS) are
already bypassing email, which has been the most
popular channel for directed information sharing
(Bernstein, 2010). However, the increasing flood of
content in SNSs is already beginning to overwhelm
users (Koroleva, 2010). Keeping in touch and
maintaining relationships is a strong motivator
behind the sharing behavior.
So far, research has underlined the importance of
weak ties (Granovetter, 1073) rather than sharing of
mundane content between close connections.
Compared to social interactions in real life, there is a
clear gap between sharing methods provided by
social broadcasting and email. It has been shown,
that in directed link sharing services, the importance
of the connection is more important than social
capital (Bernstein, 2010). Personal and directed
recommendations are also consumed more likely
than the ones that have been broadcasted (Leino,
2011).
Even though content sharing is gaining
popularity as a means of finding content, most
people still find and consume online news directly
from the news site (Leino, 2011). The aim of the
media companies providing the news service is to
seek new ways to increase stickiness and
engagement to their content portfolios.
In this paper, we explore practices and propose
design requirements for online tools that facilitate
directed onsite sharing in close-knit groups. Onsite
sharing allows users to share and consume articles
directly in the online newspaper domain.
We define requirements and introduce a design
for an onsite toolbar prototype. To evaluate the
perceived value of directed onsite sharing and the
implications for social practices, a field trial was
carried out over a period of ten days. We organized
semi-structured interviews to gain deeper
understanding about the features users would
appreciate.
During the trial period, users actively shared and
reported a clear perceived contrast between sharing
practices in broadcast media and privately among a
367
Tukkinen P., Pohja M., Segerståhl K., Silfverberg S. and Vuorimaa P..
SUPPORTING NEWS READING AND DISCOVERY THROUGH DIRECTED ONSITE INFORMATION SHARING.
DOI: 10.5220/0003907903670373
In Proceedings of the 8th International Conference on Web Information Systems and Technologies (WEBIST-2012), pages 367-373
ISBN: 978-989-8565-08-2
Copyright
c
2012 SCITEPRESS (Science and Technology Publications, Lda.)
close group. We contribute design recommendations
for facilitating directed sharing based on the results.
2 RELATED WORK
Systems designed to connect individuals exploit
voluntary willingness to share information, but they
should do it in such a way that the cost of sharing,
receiving and processing of content is minimal
(Dearman, 2008); (Milic-Frayling, 2008). A small,
clearly defined audience works best to increase the
interest and expectations of shared content and the
discussion around it (Bernstein, 2010). Strongly
connected pairs are also more likely to share their
resources (Haythornthwaite, 2001); (Wellman,
1992).
Social bookmarking and collaborative filtering
have been presented as enablers of social navigation
already as early as 1995 (Höök, 2003); (Maltz,
1995). These services are based on users tagging
content with selected keywords. Popular services
such as Diigo provide platforms for social tagging in
which effectiveness depends on finding and re-
finding information (Golder, 2006). By automating
parts of the recommendation process, collaborative
filtering aims to provide personalized content for a
user with relatively low effort (Resnick, 1994).
However, users know that recommendations are
automatically created, rather than being direct
communications.
Online chat systems, such as Internet Relay Chat
(IRC) and Skype are also popular tools for sharing
links during discussions. Due to their synchronous
nature, instant feedback is available (Nardi, 2000).
However, chats do not provide information
management capabilities, and sharing is not their
primary function. Chats provide a strong model for
communication with close contacts and strong
feeling of community. Users perceive the content
created in IRC to be very communal (Olsson, 2008).
Therefore, we aim to adopt this model, modify it
further and move it to the web.
Bernstein et al. introduced an application called
FeedMe for directed sharing interesting links from
RSS feeds by email. They found that people are
more likely to share directed links about interesting
topics between close contacts whose interests are
known to them (Bernstein, 2010).
However, the problem with most of the presented
solutions is that the majority of Internet users do not
use tagging services or RSS readers.
3 REQUIREMENTS
Based on the previous research, we present a
scenario of use and requirements for a system that
would facilitate sharing in relatively small close-knit
groups.
3.1 Scenario: Reading News
Mary is sitting in the metro returning home from
work. She reads the online news of a local
newspaper from their website. Mary is using a
service that allows her to share content with her
friends and to comment on links. The service has
been embedded on the site. When she enters the site,
she sees the latest headlines and, moreover, sees the
headlines that her selected friends have shared. With
her husband, she is following a discussion on a new
daycare center that might be built nearby. She is
following the news about an airline strike with her
friends because she is about to spend the following
weekend in Paris with them.
3.2 Requirements of the System
Based on the above scenario and previous research,
we defined the following requirements.
• Onsite Submission: It must be possible to share
a link on the domain of an online newspaper where
the news content is originally published.
• Persistent Storage: Shared links must be stored
persistently.
• Collective Storage: Shared links must be stored
based on a group (not by a single user). Members of
the group must be able to jointly manage the stored
links.
• Directed Sharing: Users must be able to define
the visibility of their recommendations.
• Onsite Viewing: Users must be able to
simultaneously browse a website and the
recommended links that are related to it.
• Categorization: Users must be able to
categorize the recommended links based on social
connections.
4 NEWS TOOLBAR CASE STUDY
To validate the concept of directed onsite sharing,
we designed and implemented a toolbar that was
integrated to an online newspaper. The aim of the
system is to provide an efficient way to share and
WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies
368
discover content in close-knit groups. The onsite
toolbar allows users to share news items in various
channels with varying levels of privacy. These
channels can be defined by users and shared with
particular contacts. When shared, the channel may
be simultaneously viewed and updated by its users
while they are browsing and reading the news.
Contacts can be invited via email or they can be
searched among registered users or Facebook
friends.
We wanted to increase stickiness and
engagement in the news portal. Therefore, we
decided to implement the toolbar as a website add-
on, which is available for everyone without
installation.
4.1 User Interface
The prototype consists of a toolbar integrated into
the online version of a Finnish newspaper. Figure 1
illustrates the user interface of the toolbar at the top
of the newspaper website. The service is
implemented in the Finnish, but the screen captures
have been translated into English.
Figure 1 displays the list of links that have been
shared to the selected channel. Channels equal the
channel concept in IRC: one user creates a channel
and invites people or lets them join freely. Everyone
in the channel is able to see who else are in that
channel and can access the content. Infinite number
of Channels can be created and managed in the
system for collecting direct links to news items
around a specific theme or purpose for a defined
audience. Only members of the particular channel
can see the content.
Figure 1: The channels of the specific user in the onsite
toolbar.
Users can read articles by clicking the links in
the toolbar. The article is then opened in the online
newspaper website. The link text is the headline of
the article in question.
The system is connected to Facebook. Facebook
users can easily share selected items with the wider
audience of Facebook as well as exploit their user
account for login and invite friends.
The right-hand section is for notifications
provided by the system, such as invites to channels.
The left-hand section of the toolbar (presented in
Figure 2) is for managing channels. The middle
portion is allocated for presenting the actual content,
which contains collaboratively collected items. The
list can be opened and closed, and the top row serves
as a status bar.
4.2 Comparison to Existing
Approaches
To highlight the novelty of our approach, we
compare it with existing sharing methods through
the requirements defined in Section 4.2. The starting
point of our service was to support onsite viewing of
the recommendations, and this distinguishes the
toolbar from other methods.
Social Bookmarking is the closest method to our
system; some systems, such as Diigo, fulfill all the
requirements except for onsite viewing. “Tell a
friend” via e-mail and FeedMe are similar to each
other, although FeedMe’s onsite submission happens
on a feed reader and not on the original website. The
recommendation e-mails can be stored persistently,
but not collectively, and the receivers are always
defined particularly. SNS fulfills only the onsite
submission requirement, whereas IM fulfills only the
directed sharing requirement. Blogs are typically
stored persistently and the entries can be
categorized. However, they lack onsite submission
and viewing, collective storage, and directed
sharing. In the light of this comparison, it can be
said that our toolbar is a new type of social
bookmarking tool that enables onsite viewing.
5 EVALUATION
Our analysis is based on semi-structured participant
observation interviews.
5.1 User Study
Ten participants were recruited purposively to
participate in a ten-day usage period. The criteria for
the purposive sampling were that they were already
reading the online version of the newspaper and that
they were adults. The set of ten users consisted of
three couples and one group of colleagues. At the
SUPPORTINGNEWSREADINGANDDISCOVERYTHROUGHDIRECTEDONSITEINFORMATIONSHARING
369
Figure 2: Screen capture of the onsite toolbar presenting the most recently added articles in the shared channel called “Mary
and Joe”, which is shared by two people.
beginning of the ten-day usage period, participants
received a simple instruction letter. The purpose of
the usage period was to observe how the participants
would use the service and how the system could
extend interpersonal communication.
After the ten-day period, participant-observation
interviews were conducted. The in situ interviews
were organized in the natural browsing environment
of the participants, such as at home or at work. The
interviews were video recorded to obtain accurate
data, such as contents of the different channels from
the computer screen. The interviews were semi-
structured in order to follow interesting paths
indicated by the participants. The aim was to
extensively explore the contents of each personal
account. After the interview, each participant filled
in a questionnaire. The video recordings were
transcribed verbatim, and the computer-screen
information was transformed into tables. The data
were analyzed using qualitative content analysis in
conjunction with the screen information in table
form.
5.2 Participants
The participants were selected to represent three
different types of existing social groups. They were
recruited through word-of-mouth.
The mean age of the participants was 28 years;
many of the older people who we tried to recruit
seem to prefer the paper version. Based on the
questionnaire data, six participants out of nine were
members of Facebook, and few of them used IM
services. They did not use social bookmarking
services, and only one had a Twitter account. None
of our participants was very active in recommending
content online. Sharing information about news
articles face-to-face, or asking the other person to
come to view the screen, were common practices.
Team 1 consists of a married couple living in a
long-distance relationship during the last eight
months. They had shared apartment announcements
by email but had not shared news items. They used
the system to keep in touch and often continued to
discuss the items on Skype.
Team 2 consists of two sisters and their spouses.
One of the couples is married. And the other couple
lives together. They have shared news face-to-face
but never with online services. They used the system
to distribute interesting, locally relevant articles and
to continue discussions.
Team 3 consists of four colleagues working together
in a department store. They have also become
friends and enjoy discussing personal matters. One
of them was sick during the interviews. All of the
group members used the system to share work-
related articles and articles related to their private
life. They also shared amusing articles to uplift the
spirit at the workplace.
6 RESULTS
Participants mostly shared news related to local
issues or to familiar people, culture or entertainment;
they shared items that they knew would interest the
recipient. Participants told that they always logged-
in to the service immediately when they accessed the
news site.
In directed sharing, content is often targeted
towards the recipient. Participants found this
approach suitable for intimate sharing for mundane
items that supported their existing interpersonal
communications. Content was selected carefully by
thinking of the receivers and their perceived
interests.
Links were often used as references to back up
WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies
370
earlier statements from unmediated discussions.
Couples were most likely to continue offline
discussions by using the content as references. Many
participants reported that they used to share and
discuss articles by showing them to the recipient in
person from the computer screen.
Participants were sharing links related to local
issues and to people they knew. In many cases, local
issues were considered interesting to the group but
not relevant enough for larger audience.
The problem with sharing links in public
broadcast media, such as Twitter or Facebook, is the
untargeted nature of the sharing method. The
perception of interesting content varies widely
between individuals. Many of our participants would
like to discuss content between close-knit groups
rather than have such content curated by distinct
acquaintances.
Participants frequently compared the system to
Facebook. They felt that there were two different
categories for sharing news content. Some items
would be important to everyone and therefore were
good Facebook material. The second category was
more private and was intended for a small number of
people. Some items were considered to be generally
important to a wider target audience. It is important
to note that only a small number of shared links
belonged to this category.
In SNS or collaborative sharing, the amount of
likes or recommendations is considered to be the
measure of an interesting article. During the field
trial, participants shared items that they considered
particularly interesting. The relevance of the
recommendation was considered as the most
important factor, not the amount of links available.
Contrary to public commentary, many
participants confirmed that the barrier for adding
comments to news items and discussing them online
would be significantly lower if commenting was a
self-moderated activity between small and
meaningful groups of people. It was also considered
important that all participants would be interested in
the topic in question. Facebook provides a
possibility to comment on recommendations among
friends, but the overall visibility of the comments is
unknown to participants, except to the original
sharer.
Shared news articles can function as reminders
of intended action. The reminder can be there for the
person reminding, the recipient or both. In many
cases, such items are still referred to as ambient
information and not as a critical to-do item.
Notifications and awareness of synchronous use
were frequently mentioned in the interviews.
One popular use case was quick access to the
most relevant news in a busy situation. Directed
sharing was considered as a suitable means of
filtering.
7 DISCUSSION
We suggest that sharing as a part of direct, onsite
content consumption will help users to discover
interesting news content more efficiently. The aim is
to provide access to personalized pre-filtered content
that is known to be interesting. As the motivation for
using the service is to consume news, users are not
very cautious about spamming in comparison to user
interfaces that forward the content to others without
knowing their motivation.
As expected, content was largely shared based on
the interests of the recipients, and in many cases, it
was relevant only to that particular group in the
context in which it was mentioned. Many
participants felt that the tone in active public
discussion around news articles was usually
negative. They felt that private discussion inside the
group would be more positive and constructive.
Our current solution had two problems. The first
was the lack of a possibility to privately discuss
items, and that content was limited to one source of
news. If the whole range of services provided by the
media company were available enhanced with
commenting, then many participants would use the
application. Reasons for not using this type of
system were preferences for paper versions and face-
to-face communication.
An alternate solution would be to allow users to
add external links, and thus provide a system that
would work as a universal bookmarking application.
However, during the interviews, users were more
concentrated on news content than on web content in
general. Clearly specified domain might help users
to construe content better. Interviews also
highlighted the problem of information overload.
People would like to receive content suggestions,
but the current flood of content, even in SNSs,
overwhelms them.
Based on our findings, we contribute design
recommendations for future systems to facilitate
directed onsite sharing.
• Support for groups should be easily available.
Users should be aware with whom they share and
who will see the content. The cost of selecting
groups while sharing should be kept minimal.
• Efficient for everyone. System should be easy
to adopt and should not require installation. Users
SUPPORTINGNEWSREADINGANDDISCOVERYTHROUGHDIRECTEDONSITEINFORMATIONSHARING
371
should remain logged-in constantly while they are
online.
• Provide Notifications. Users would like to be
aware when someone has added new content to the
system even when they are not browsing the news.
• Support for discussions. Applications should
support voluntary online discussions.
• Semi-synchronous feedback. Users would
appreciate the possibility of using the discussion
channel as a semi-synchronous chat if there were
more than one user online at a given moment.
• Persistency for asynchronous use. All
information, including links and chat conversations,
should be available to users who were offline during
the activity.
We were able to test our application with a limited
number of users and within limited amount of time.
This research was conducted in an explorative way
to investigate users’ thoughts, and our contribution
is more with regard to implications for future
designs. To reach a full understanding of wider
audiences, the system should be a publicly available
and it should be tested during a longer period of
time. Further data collection is required to determine
how the service affects interpersonal relationships
and whether social connections enable users to
construe content more efficiently.
8 CONCLUSIONS
Directed onsite sharing enables users to share and
discover mundane content among close-knit groups,
to simultaneously browse the news and to broadcast
publicly relevant articles to SNSs. Users perceive
broadcasted and directed sharing approaches
differently, which leads to different practices and
needs. The native toolbar was used to share mutually
important and local content with members of the
group. Recipients’ interests were the focus.
Particularly in contrast with the broadcast nature and
self-presentational aims of many link-sharing
services, the barrier for sharing is lower and the
reported frequency is higher. Users want to remain
aware of with whom they share and what others
have shared. They are more eager to discuss within
small circle of friends about local and current things
that are relevant to them, instead of contributing to
the public discussion. Based on our results, we
presented recommendations for design for future
applications facilitating directed onsite sharing.
REFERENCES
Bernstein, M., Marcus, A., Karger, D. and Miller, R.,
2010, Enhancing directed content sharing on the web.
In CHI ’10: Proceedings of the 28th international
conference on Human factors in computing systems,
pages 971–980, ACM.
Gary, M. R. Optimal binary identification procedures.
SIAM J. Appl. Math. 23, 2 (Feb. 1972), 173–186.
Dearman, D., Kellar, M. and K. N. Truong, 2008, An
examination of daily information needs and sharing
opportunities. CSCW ’08, pages 679–688, ACM.
Golder, S., and Huberman. B., 2006, Usage patterns of
collaborative tagging systems. In Journal of
Information Science, volume 32, page 198.
Granovetter, M., 1973, The strength of weak ties. In
American Journal of Sociology, 78(6), pages 1360–
1380.
Hong, L., Chi, E. H., Budiu, R., Pirolli, P. and Nelson L.,
2008, Spartag.us: a low cost tagging system for
foraging of web content. In Proceedings of the
working conference on Advanced visual interfaces,
AVI ’08, pages 65–72, ACM.
Höök, K., 2003, Designing Information Spaces: The
Social Navigation Approach. In Designing
Information Spaces: The Social Navigation Approach.
Springer.
Koroleva, K., Krasnova, H., and Gunther, O., 2010, Stop
spamming me! – Exploring information overload on
Facebook. In AMCIS 2010 Proceedings, AMCIS ’10
Leino, J., Räihä, K. and Finnberg, S., 2011, All the News
That's Fit to Read: Finding and Recommending News
Online, In Proceedings of INTERACT '11, Springer.
Maltz, D. and Ehrlich, K., 1995, Pointing the way: active
collaborative filtering. In Proceedings of the SIGCHI
conference on Human factors in computing systems,
CHI ’95, pages 202–209, ACM.
Milic-Frayling, N., Hicks, M., Jones, R., and Costello. J.,
2008, Exploring personal broadcast channels for rich
media sharing in close relationships. In Proceedings of
the 7th International Conference on Mobile and
Ubiquitous Multimedia, MUM ’08, pages 23–30,
ACM.
Nardi, B., Whittaker, S. and Bradner E., 2000, Interaction
and outeraction: instant messaging in action. In
Proceedings of the 2000 ACM conference on
Computer supported cooperative work, CSCW ’00,
pages 79–88, ACM.
Olsson, T., Soronen, H., and Väänänen-Vainio-Mattila, K.
User needs and design guidelines for mobile services
for sharing digital life memories. In MobileHCI ’08:
Proceedings of the 10th international conference on
Human computer interaction with mobile devices and
services, pages 273–282, New York, NY, USA, 2008.
ACM.
Purcell, K., Rainie, L., Mitchell, A., Rosenstiel, T. and
Olmstead, K., 2010, Understanding the participatory
news consumer: How internet and cell phone users
have turned news into a social experience. Technical
Report, Pew Research Centre.
WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies
372
Resnick, P., Iacovou, N., Suchak, M., Bergstrom, P. and
Riedl, J,. 1994, Grouplens, An open architecture for
collaborative filtering of netnews. In Proceedings of
the ACM conference on Computer supported
cooperative work, CSCW ’94, pages 175–186, ACM.
Wellman, B., 1992, Which types of ties and networks give
what kinds of social support? In Advances in Group
Processes, 9, pages 207–235.
SUPPORTINGNEWSREADINGANDDISCOVERYTHROUGHDIRECTEDONSITEINFORMATIONSHARING
373
