Content uploaded by Sara Bly
Author content
All content in this area was uploaded by Sara Bly on May 01, 2014
Content may be subject to copyright.
Sharing Encountered Information:
Digital Libraries Get a Social Life
Catherine C. Marshall
Microsoft Corporation
One Microsoft Way
Redmond, WA 98052 USA
cathymar@microsoft.com
Sara Bly
Sara Bly Consulting
24511 NW Moreland Road
North Plains, OR 97133, USA
sara_bly@acm.org
ABSTRACT
As part of a more extensive study of reading-related practices, we
have explored how people share information they encounter in
their everyday reading as a complement to the more traditional
digital library focus on sharing intentionally retrieved materials.
In twenty contextual interviews in home and workplace settings,
we investigated how people encounter and save published
material in the form of paper and electronic clippings. We found
that sharing forms a significant use for encountered materials.
Furthermore, the function of these clippings extends far beyond a
simple exchange of content to inform the recipient; in fact, the
content itself may have little immediate value to the recipient. We
also found the practice to be ubiquitous: all of our participants
had both shared clippings with others and received them
themselves. Specifically, this paper reports on: (1) how sharing
encountered items fits into the broader spectrum of clipping
practices; (2) the function and value of the shared information;
and (3) the social role of sharing the encountered information. We
conclude that from a technological standpoint, we should think
beyond an email model for sharing encountered information and,
from a social perspective, we should attend to how sharing this
sort of material contributes to the strength of social ties outside of
a traditional information needs framework.
Categories and Subject Descriptors
H.3.7 [Information Storage and Retrieval]: Digital Libraries –
User issues; H.5.2 [Information Interfaces and Presentation]:
User interfaces – Evaluation/ methodology; H.5.3 [Information
Interfaces and Presentation]: Group and Organization Interfaces
– Computer supported cooperative work
General Terms
Design, Documentation, Human Factors, Performance
Keywords
Electronic periodicals, reading, browsing, encountered
information, clipping, collaboration, field study, interaction
1. INTRODUCTION
A marketing manager spots an online review of her company’s
new product in the Wall Street Journal and posts a hardcopy of it
on the lunchroom bulletin board; a father emails his daughter a
New York Times op-ed piece about the war in Afghanistan; two
friends exchange recipes cut out from their respective Cooking
Light and Gourmet subscriptions.
What do these vignettes have in common? They are all examples
of sharing information encountered in the course of everyday
activities such as reading the newspaper, browsing the Web, or
leafing through a magazine. These encounters with published
materials are often unintentional, unplanned, and may have
elements of serendipity. The subsequent sharing of the found
materials is usually informal and unbidden.
As part of a more general research program to characterize how
people interact with paper and electronic publications, we have
performed a study of how people save and expect to use the
materials they encounter as they read everyday publications such
as newspapers, magazines, catalogs, and Web sites. Encountered
information forms an important counterpart to information
retrieved as the result of an explicit query or a directed browsing
session [8]. It is thought to foster creativity [20] and be a
significant source of the information people acquire. For our
study, we chose to look specifically at clipping – cutting an item
out of a newspaper or magazine, saving transient published
material from the Internet (either as the literal content or as a
URL), or keeping an entire periodical to have access to particular
content – as a particular form of encounter.
One important use of clippings is to share them with others, the
aspect of our study we report here. Much of the work on sharing
information to date has focused on conducting searches on behalf
of others (for example, [15]) or sharing one’s own knowledge in
electronic forums (for example, [1]) and how this kind of sharing
relates to the development of community. Our investigation
complements this body of research by examining the role of
encountered information in social settings.
To date, digital libraries have not been conceived as a venue for
browsing and sharing material that is not directly connected with
information needs. We suggest that adding this element to digital
libraries will increase their ultimate value and make them more of
a part of everyday life.
Figure 1 shows examples of the sorts of physical and digital
clippings that people share. Figure 1a shows an example of a
physical clipping, a photocopy of a picture that appeared in the
local paper, which has been saved to send to a colleague in
another city. Figure 1b show an example of a digital clipping, a
newspaper article saved from the electronic version of the
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise, or
republish, to post on servers or to redistribute to lists, requires prior
specific permission and/or a fee.
JCDL’04, June 7–11, 2004, Tucson, Arizona, USA.
Copyright 2004 ACM 1-58113-832-6/04/0006…$5.00.
Vancouver Sun. The digital clipping has been saved as a link to
the actual article (as a Favorite). These two examples illustrate
some of the diverse methods people use to clip and share material
from physical and digital media.
(a) Example of a physical
clipping: a photo from a
local newspaper which has
been photocopied to mail to
a colleague in another city.
(b) Example of a digital
clipping: an article from a
distant newspaper which has
been saved as a Favorite.
Figure 1. Clipping examples. The range of clipping forms in
physical and digital media, as well as the ways in which they
are shared, is diverse.
Our data suggested three kinds of questions to help us
characterize how and why people shared the encountered material:
1) Our initial focus was on the general phenomenon of
clipping and saving encountered information, but the
data we gathered indicated that people frequently shared
the materials they found with others. What is the
significance of sharing the materials relative to simply
saving them for one’s own future or immediate use?
2) Much of the sharing that we observed served a variety
of functions in its social settings beyond simply
informing. By carefully examining an extensive set of
instances of this activity, can we understand the ultimate
functions of clippings in various social settings and
evaluate the utility of their content?
3) Information brokers are often portrayed in the literature
as a distinct social type [15]. Is this true of people who
share encountered information as well? Are there people
who tend to encounter and share more materials with
others and, on the other hand, people who tend to be
primarily recipients of the items encountered by others?
To provide a backdrop for this study, we first describe the study
participants, their demographic profile, and the kinds of
publications they typically read. The next section is devoted to our
findings, specifically those that pertain to the areas of inquiry we
outlined above– characteristics of shared clippings; the functions
the shared materials fill for the givers and recipients and whether
their various intents seem to be met; and finally, we examine what
we’ve learned about the givers and recipients themselves. Our
discussion explores the implications of our findings and issues
they reveal; included in this discussion are technology
implications that build on current directions for sharing electronic
material, such as [18].
2. STUDY DESCRIPTION
To characterize clipping practices, we performed a qualitative
study consisting of contextual artifact-driven interviews of 20
diverse individuals at home and at work. The study brought
together narrative accounts and physical examples to investigate
the ways in which people currently collect, save, and use physical
and digital clippings as part of their everyday reading activity in
different social settings. Initially, we were driven by the challenge
of whether clipping is still a necessary kind of interaction in an
age of electronic publications; after all, it is easy to imagine the
function of paper clippings being readily replaced by a digital
library of periodicals with well-designed searching and browsing
capabilities.
Studying how people clip from paper publications (e.g.
magazines, newspapers, and catalogs) and electronic publications
(e.g. online digests and electronic versions of traditional
publications) must take into account both the ubiquity of the
practice and the many genres of periodicals available today in
paper and digital form. Thus, the study was organized around a
number of short field visits in as many different kinds of sites as
possible, both homes and offices.
We conducted the artifact interviews in two West Coast cities.
Half of the interviews took place in homes and the other half in
workplaces; one participant, the head of a commercial nursery,
was interviewed both at her business and in her home. The intent
was to be in locations where people receive, encounter, and read
newspapers, magazines, and catalogs and have access to the
Internet. Participants were selected across a broad range to ensure
that we would understand whether or not clipping practices might
be specific to age, gender, occupation, or genre of reading
materials. Thus, we covered an age range from 16 to over 65; half
of our participants were male and half female; and we made sure
we included people with different levels of education, different
uses of published materials in their jobs, and different reading
interests (either personal or work-related).
Table 1 summarizes the characteristics of the participants in our
study and includes several examples of what kinds of periodicals
each participant reads. In addition to receiving local newspapers
or subscribing to national newspapers online (such as the New
York Times), many of our participants received trade magazines
and higher circulation glossy magazines.
Our primary screening criteria was that the participants read
magazines or newspapers and that they had some form of access
to electronic information. We did not require them to subscribe to
electronic publications. All study participants had some access to
the Web, although there was considerable variation in the
frequency, ease of access, and comfort level with going online.
For example, one participant, a homemaker, had relatively little
experience using electronic information; another (from a low-
income household) accessed the Web from the local public
library; a third, a partner in a commercial nursery, was more likely
to use his business partner as an intermediary for Web access and
did not go online often himself. At the other end of the spectrum,
several of our participants had regular access to Web-based
publications both at work and at home and read electronic
periodicals daily.
The interviews were semi-structured and open-ended. When an
interview took place at a participant’s home, we did not limit our
questions to materials related to personal interests; likewise we
did not limit workplace interviews to work-related materials.
Rather, we focused on any type of material they read at the
interview site. Participants were not told the aim of our study or
the corporate sponsorship of the study until the conclusion of the
interview.
Together we spent 60-90 minutes with each participant to explore
the kinds of published material they encountered, what they
saved, and what they shared or had received from others. Because
reading-related practices like clipping are fairly unselfconscious
and lightweight, we often had to probe their accumulated paper
and electronic “stuff” for examples, and used the examples to
drive the interviews.
3. DATA
The data we collected included notes, audiotapes of the
interviews, and digital photographs that documented the examples
and interview sites. To prepare for the more intensive phase of
data analysis, we reviewed and categorized all of the clipping
examples and stories into collections that reflected the various
forms and functions of clippings, and how they were shared. Our
data show that clippings do not fall into a few simple categories;
the encountered material is saved in multiple forms for multiple
uses. Most people saved information from published material in
some form or other.
Because we decided to focus on shared clippings as our first area
of inquiry, we extracted all of the examples of this phenomenon
that arose during the interviews. These specific instances gave us
a core collection of over 150 entries to analyze. Table 2 illustrates
the types of examples we identified from the interview notes and
transcripts. In P4’s case (in the electronic/work quadrant of the
table), the example discusses a clipping he received; in the other
three examples, the study participants are giving the materials to
others. Each participant had some experience with sharing
Table 1. Overview of Participant Characteristics. Participant IDs are included to clarify quoted material used other places in
this paper.
ID M/F
Job Location
Frequency of
Web Access
Age Education Examples of what the participant reads
P1 F
Teacher
(unemployed)
Home few times/week 45-54
college graduate
Local newspaper; Yoga; Alternative Education
P2 M IT director Home several times/day
35-44
college graduate
Local newspaper; Sunset; Network World
P3 M IT manager Work several times/day
35-44
some college
Storage; Oregon Business Journal
P4 M
Environmental
outreach coordinator
Work several times/day
25-34
some college
Local newspaper; online newspapers;
TidePool (online); High Country News
P5 F Administrator Work several times/day
65+ college graduate
Meetings and Conventions (online and on
paper); American Law Technology
P6 F Student Home several times/day
16 in high school
People; weekly free paper; NY Times (online)
P7 M
Business services
developer
Home several times/day
35-44
college graduate
This Old House; Martha Stewart Living; Fine
Home Building
P8 M Individual consultant
Work several times/day
25-34
college graduate
NYT; WSJ; Fortune; golf magazines;
Hawaiian papers
P9 F
Public relations
coordinator
Work several times/day
25-34
college graduate
Business Week; EuroMoney; NY Times
(online)
P10
F
Environmental
scientist
Home several times/day
25-34
college graduate
UTNE Reader; Backpacker; local newspaper
P11
M
Senior sales
manager
Home several times/day
25-34
college graduate
Instinct; Wired; Pottery Barn catalog; local
newspaper
P12
F Executive assistant
Work several times/day
25-34
some college
Marie Claire; CNN.com; Sacramento; VIA
P13
M
Retail clerk; Army
reservist
Home few times/week 18-24
some high school
Expert Gamer; Maxim; local free weekly;
Soldiers
P14
F Homemaker Home seldom 25-34
some high school
Parenting; Better Homes and Gardens;
Costco catalog
P15
F Secretary Home few times/week 35-44
college graduate
Local newspaper; Cosmo; Time
P16
M
Partner in a design
firm
Work several times/day
35-44
college graduate
local business journal; WSJ; ESPN; NY
Times online
P17
M
Partner in a retail
nursery
Work few times/month
25-34
college graduate
Garden Center Products; Sunset; local
newspaper
P18
F
President of a
commercial nursery
Both several times/day
45-54
college graduate
Cooking Light; Better Homes and Gardens;
People
P19
M Financial advisor Work several times/day
35-44
college graduate
NY Times online; Barron’s; WSJ; Golfsmith
catalog
P20
F
Museum content
coordinator
Work several times/day
25-34
college graduate
Nature (online); Exhibit Builder; local online
paper
clippings and each participant had both given and received
clippings.
Table 2. Examples of how the study participants shared
electronic and paper materials related to both work and
personal interests.
personal work
paper
P10, an environmental
engineer, gave her
friend an entire issue of
UTNE Reader when
they took a camping trip
to Yosemite together
because it had an article
about Edward Abbey
and she knows her
friend likes him… He
read it and gave it back
to her.
P20, a museum content
coordinator, has an article
photocopied from Nature on
Interactive Bus Shelters that
she plans to hand to her
project manager so that she
can talk to him about it. “My
plan is to actually give it to
him and talk to him about it,
rather than just put it in his
in-basket, because he’d kind
of wonder where it came
from or why he was getting
it.”
digital
P11, a sales manager,
emailed Rob Morris’s
article from the
Chronicle on 9/11/02 to
his friends all over –
New York, DC, Boston,
and San Francisco. He
pasted it into an Outlook
message and sent it
without much of a cover
letter. He just prefaced it
with, “Something I found
interesting on the way to
work this morning. Hope
you find it interesting
too.”
P4, a national outreach
associate for an
environmental organization,
received a URL from one of
his colleagues that had been
taken from an online
publication, TIDEPOOL. It
has a paragraph excerpted
from the article as a cover
message, and the title line
says, “from TIDEPOOL.org.”
This example turns out to be
redundant for P4 though – he
already subscribes to this
newsletter himself and had
read the article.
We coded each of the shared clipping examples to reflect the
important characteristics of the example, including whether it was
a given or received item, whether the item was related to personal
or work interests, how it was delivered (for example, by email or
posted on a bulletin board), how much context was provided for
the item (for example, was it a URL, an article, or an entire
section of the newspaper), and the form it was delivered in (for
example, as original material or as a photocopy).
Table 3 illustrates the distribution of the examples according to
whether we observed the sharing at home or at work, whether it
was an example of giving or receiving, and whether the content
itself was related to personal interests or work. Our data suggest
that many people mix personal and work-related reading and
activities seamlessly throughout the day; for example, several of
our participants reported reading online newspapers on the
computer screen when they take a break from their work; several
also reported that they read trade magazines at home, where it is
quiet and they can concentrate.
Table 3. The number of examples we collected of personal and
work-related clippings. The participants’ reading habits were
such that we often encountered personal clippings during
interviews conducted in work settings and work-related
clipping at home. The examples are further divided to indicate
whether the participant gave or received the clipping.
At Home At Work
Total
Personal Work-
related
Personal Work-
related
Giver
82 31 5 13 33
Receiver
69 21 5 11 32
To get a richer perspective on the data and to identify potential
patterns, we also coded more subjective aspects of each clipping
example. These aspects included whether received clippings were
perceived by the recipient as immediately useful, potentially
useful, or not useful at all and how the recipient interpreted the
giver’s intent (if they had verbalized it). For example, sometimes
recipients thought the givers were just saying “hello” by sharing a
clipping with them. We similarly tried to capture the giver’s
primary motive and the intent of the interaction.
We individually coded every example and compared our
interpretations. Where they differed, we reconciled the differences
by either refining our shared understanding of a category, by
creating a new category, or by agreeing one interpretation was
more likely than another.
In 69 examples, our participants told us about clippings they had
received. Figure 2 shows that most of the received material fell
into two main categories: (1) the material was work- or task-
related and potentially (but not immediately) useful; or (2) the
content was not particularly useful as such, but served to
demonstrate a shared interest. A relatively small proportion of the
clippings, regardless of the giver’s intent, represented content that
was perceived as immediately useful to the recipient.
0
5
10
15
20
25
Requested Work/Task Shared
Interest
Immediately Useful
Potentially Useful
Not Useful
Figure 2. Much of the shared material is not considered
immediately useful by the recipient.
Participants referred to 82 examples of clippings that they shared
with at least one other person. We looked at the item’s role (for
example, did it have an identified purpose in a joint project?) and
the sender’s motivation (for example, the material pertained to a
known interest of recipient). Figure 3 shows that less than a
quarter of the clippings were shared as work- or task-related
exchanges; of those, less than half had a known use. In almost half
of the total cases, senders’ motive was to demonstrate knowledge
of the recipients’ interests or emphasize a connection between
sender and recipient (e.g. a shared sense of humor).
4. FINDINGS
As we gathered data about encountering and clipping items from
published material, we identified three general areas of inquiry to
pursue in our analysis:
1. How does sharing encountered items fit into the
broader spectrum of clipping practices?
2. What is the function and value of sharing encountered
information?
3. What is the social role of sharing encountered
information?
We discuss findings related to each of these three areas of inquiry.
In so doing, we will also touch on observations about how the
materials were shared and how much context the sender included.
For example, physical materials might be posted in a central
location or delivered personally. The clipping itself might be a
URL, an extract taken from a longer article, or a whole magazine
or other form that includes substantial content that is not directly
relevant.
4.1 How sharing encountered items fits into
the broader spectrum of clipping practices
Although in this paper we have focused on sharing clippings,
sharing is only part of what people do with the materials they
save. Because clippings can act in a broad range of functions, we
need to put sharing in perspective: Does sharing represent a
important role for encountered material? Is it a significantly
different kind of practice than saving encountered material for
personal use?
We found that the personal uses for clippings include saving
content that might be useful later or might serve as a reference;
setting aside material to read or re-read later; putting the clipping
in view to serve as a reminder for future action (e.g. “buy concert
tickets”); or evoking memories of a personal or historic event (e.g.
a newspaper section saved from a European vacation or a
magazine published in January of 2000 that covered significant
20
th
century events). From our examples, we can infer that
personal clippings are more often content or information-centric
rather than immediately of use. To realize the intended value of
the clippings, people rely on re-encountering them at the proper
time or in the appropriate situation, for example when the
anticipated information need arises. In contrast, shared clippings
often are not so strongly content-oriented. They may serve their
intended function at the time of initial receipt, for example,
strengthening social ties by demonstrating a shared interest. Thus
the shared material’s importance is not tied so tightly to the
anticipated utility of the content, but rather the appropriateness of
the content to the sender’s communicative goal.
How common is the practice? All twenty of our study participants
both saved encountered material for themselves and clipped it to
share with others, although some participants seemed to be more
inclined toward personal use, and others toward shared use. If we
examine the relative prevalence of examples of shared material
and of personal clippings, about forty percent of the examples are
examples of sharing encountered material. Sharing encountered
material is therefore common, reasonably ubiquitous, and likely to
be important in the roles we go on to describe in the next section.
Figure 3. The person sharing the material is more apt to be sharing it for reasons associated with strengthening social ties
than passing on content with a known use. One of the other common categories is material shared to educate and promote
the giver’s values.
4.2 The function and value of sharing
encountered information
It is tempting to take the practice of sharing encountered
information at face value; to look at it simply in terms of passing
on content that would seem to meet the recipient’s information
needs at home or at work. Yet it was apparent to us from the
outset that our participants had material they had received from
others that was not particularly useful for the content itself.
Sometimes they had already encountered it themselves prior to
receiving it from others; other times they said they weren’t going
to actually read it; and sometimes the sharers indicated that it was
something they wanted the recipient to know regardless of
whether they had indicated a prior interest. Nevertheless, the
recipients indicated that they valued receiving the shared material
independent of its immediate utility.
The significance of this paradox is that the shared materials often
appeared to serve a function other than that of an expressed
information need. The content frequently demonstrated a shared
interest that created, strengthened, or renewed the social bonds
between giver and receiver. Several recipients alluded to these
clippings as an indication that the giver was “thinking of you”.
Naturally, this paradox did not reflect a universal characteristic of
the shared materials. Some of the shared information represented
a task or project-based exchange in which the content was of
either known or anticipated utility. However, this was by no
means the dominant kind of sharing that occurred.
We will describe several important categories of shared clippings
from the giver’s perspective since they were each relatively
common:
o News items, articles, and advertisements that helped
establish mutual awareness;
o Articles shared to educate the recipient, particularly in
ways that reflected the giver’s values and concerns;
o Items of peripheral importance to both giver and
receiver, intended as a means of developing rapport; and
o Clippings that reflected known interests of the receiver
(but probably not the giver), where the content itself
may have been of peripheral importance.
In the first two categories, especially the second of the two, the
content is quite likely to be of greater importance to the sender
than the recipient; the interest may not be reciprocated in the
exchange. In the third and fourth categories, the content is clearly
of peripheral importance. That is not to say the content is
irrelevant; quite to the contrary, it must acknowledge the
recipient’s interests for it to serve its social function, which is in
most general terms, to build or strengthen social ties. As such,
these may be anything from recipes to cartoons to sports coverage
to news items that involve people known to sender and recipient.
Sharing to establish mutual awareness. These are usually work-
related clippings, and are a common form of exchange among our
study participants in customer-focused jobs, who were especially
concerned about being aware of what their customers would have
read in the press and would clip accordingly. Alternatively, they
might represent some form of awareness that would help a sales
person understand the customer’s situation. For example, P11, a
senior sales manager, described putting an article on his
colleagues’ desks if the article is about one of their clients –
promotions or layoffs or other significant events in the client
company. He felt no need to put a note on these clippings because
his colleagues would be able to figure out why he had given them
the article and if not, they would ask.
These clippings generally took the form of whole articles; their
form was intended to duplicate the way that customers were
expected to encounter the information. Thus, some of our
participants would look for physical newspapers even though they
had the online form of the article because they would be
concerned that the content would differ slightly. Dissemination
was usually to multiple recipients and occurred in a variety of
ways, from leaving the physical artifact in a shared place, to
putting a copy in mail slots or on desks, to group-wide emails.
Recipients did not always read this kind of material, but they
would save it, at least for awhile, to consult it if necessary.
Sharing to educate or raise consciousness. The information in
this category reflected a value important to the giver that was
perhaps not known or not initially held by the receiver. More
often than not, these were personal topics.
For example, one of the study participants, P15, had a pre-
adolescent son who had been diagnosed with Asperger’s
Syndrome, a form of high-functioning autism. She created a file
of clippings for distribution to people central to her child’s life. In
particular, she had photocopied “a really good” article for her
mother, her sister, and her son’s teacher from a Time magazine.
She told us, “I mailed this to so many people. Because it was very,
very good… It was just last year.” She photocopied the cover of
the magazine as well, since it was the cover story. She put a note
on the photocopy she gave to her mother and to her son’s teacher
(“You need to read this”), but didn’t include a note with the copy
she sent to her sister:
“On my sister’s, she just knew… It was just people I
wanted to share it with… Every once in awhile we’ll go
back to it. [Her daughter] did a speech on it. And she
took some of the pictures from this to her school.”
Although there are fewer examples of work-related clippings of
this sort, we observed them in situations in which an
organization’s values are legitimized by appearing in a reputable
publication. For example, P4, a national outreach coordinator for
an environmental advocacy group, showed us a newspaper-sized
sheet of an article (from “way back in ’97”) that had been
laminated for use when the advocacy group staffs a booth or table
at public events for attendees to look at. He told us:
“Publications are important for us, especially daily
newspapers that reflect opinion and science and
economics. These are tools that we can use to advocate
for [his cause].”
Sharing using common interests to develop rapport. Clippings
that are shared to develop rapport between giver and recipient
have a curious property: the content must appropriately reflect the
interests of both, but it is essentially unimportant to the actual
communicative intent of the act. Rather, the encountered
information is shared to say something like, ‘I’m thinking of you,’
‘we have common concerns,’ or ‘we have the same sense of
humor.’
For example, P8, an individual consultant for a financial services
company, saved an article about a so-called Vice Fund that he had
received from a colleague by email. When we interviewed him, he
had already sent the URL on to a friend (“up his alley”) who had
in turn sent back an email saying, “is this for real?” P8 “stashed”
the email with the URL “in case I wanted to send this to anyone
else.” He said it “cracks me up to see it.”
Sharing to demonstrate knowledge of the recipient’s unique
interests. Another way to develop social ties using encountered
information is to share clippings that reflect the recipient’s
distinctive interests. This mode of interacting through (but not
necessarily over) information was reported by a number of our
study participants.
For example, P10, a young environmental scientist, reported
saving an article from the local paper about the baseball player
Barry Bonds for her aunt to give to her in person. Instead of
clipping the article, P10 saved the entire special section for her.
4.3 The social role of sharing encountered
information
Information brokers are often seen as a distinct social type, people
who desire to act as intermediaries or informal reference librarians
for their friends, relatives, neighbors, and co-workers. For
example, Pettigrew, Durrance, and Unruh assert that "the Internet
has made it easier for researchers to label and identify a particular
social type, one that might be described as 'information gatherers'
or 'monitors' [15]." Most of these discussions in the literature refer
to people who actively perform research – browse or query – on
behalf of another (see, for example, [21] or [14]). Does this
observation hold true for encountered information? Are there
people who tend to share more of the results of their successful
information encounters with others? Are there people who receive
more encountered information than they pass on? Are there others
who, like Erdelez’s non-encounterers, neither give nor receive this
kind of material?
Upon reflection most of us can identify a person in our own lives
who sends us encountered information – a mother who sends us
hometown newspaper clippings about our high school classmates;
a co-worker who warns us of new computer viruses; a friend who
knows we like David Sedaris and points us at articles in this
month’s Esquire. It is easy to come up with examples and
anecdotes that reinforce the notion that there is a distinct
information role – an intermediary for encountered information –
that bears the burden of encountering material on a variety of
topics, tracking the interests of others, and sending on clippings or
other forms of encountered information.
However, when we examined the data we collected from our study
participants, we were surprised to find that there was no bright-
line distinction between people who initiate sharing and those
who are typically recipients; our study participants generally
encountered materials they found interesting or potentially useful
and both shared and received clippings. Table 4 shows the
number of examples participants offered that showed them
sharing or receiving encountered materials as clippings; these
examples came up readily during the interviews and are not
intended as absolute measures, but rather as qualitative
indications of the ease with which they could describe giving or
receiving clippings.
Even given the qualitative nature of these numbers, we can
interpret them as showing:
o All of our participants shared clippings;
1
o All of our participants acted as both givers and receivers
in some instances;
o Younger participants are no less likely to engage in this
practice; and
o People were likely to share clippings either at home or
at work.
Table 4. Tabulation of examples of giving and receiving
encountered information at home and at work. The study
participants’ age group is included to demonstrate that these
trends do not seem to be age-related.
ID Where #Given #Received TOTAL Age
P4 Office 8 5 13 25-34
P8 Office 6 6 12 25-34
P16 Office 3 9 12 35-44
P11 Home 9 2 11 25-34
P20 Office 6 5 11 25-34
P10 Home 5 6 11 25-34
P12 Office 7 3 10 25-34
P15 Home 7 3 10 35-44
P9 Office 5 2 7 25-34
P18 Both 4 3 7 > 45
P1 Home 3 3 6 > 45
P2 Home 3 3 6 35-44
P17 Office 3 3 6 25-34
P3 Office 2 3 5 25-34
P6 Home 2 3 5 15-24
P19 Office 1 4 5 35-44
P5 Office 3 1 4 > 45
P14 Home 2 2 4 25-34
P13 Home 2 1 3 15-24
P7 Home 1 2 3 35-44
Given this set of qualitative observations, we may conclude that
the intermediary role is highly contextual. The circumstantial
nature of the role reinforces the social aspect of the sharing: the
clippings are not just used to share information; they are also
commonly used to “keep in touch” and strengthen social ties by
demonstrating shared interests or values.
5. DISCUSSION
Our findings point in two different, but mutually reinforcing,
directions. First, there are implications for further exploration of
digital library technologies for sharing encountered electronic
materials. The forms and modes of transmission of physical
materials are far richer than we saw with electronic clippings. Yet
electronic clippings hold the promise of greater convenience and
the ability to bridge distances without the intervening steps
1
Our more extensive data that covers the broader practice of
clipping also tells us that each of our study participants had
encountered information that he or she thought was worth saving,
at least temporarily.
inherent in sending a physical item. Thus it is worthwhile to take a
closer look at some examples of ways the intentions behind
physical modes of sharing can translate into technology design.
Second, it is clear that sharing everyday materials like clippings
plays an important social role that we have only begun to
understand. To better gauge the effects of the technologies that
have been proposed for information sharing, the social role of
everyday information bears further examination.
5.1 Technology implications
Field studies of the sort described in this paper are frequently used
to support the design of new technologies. Yet, in this case, we
might think that there are already effective and sensible
mechanisms in place for sharing electronic materials encountered
in everyday venues. For example, most online newspapers and
newspaper archives offer their readers the ability to simply “email
this article” in situ either as text or a link, to keep a copy for
themselves at the same time, and to append a personalized note to
the recipient by way of explanation or greeting. Such mechanisms
could be added to other kinds of digital libraries and electronic
archives as well.
However evaluation of these mechanisms has shown that they are
not widely used; people who share things from the Web
(including articles from online newspapers) most often use their
regular email program even if there are other sharing tools
available. They either clip URLs into a message or (less
commonly) cut and paste whole articles into a message [18].
Rioux asserts that there are many reasons for this, including: the
sharer doesn't have the recipient's email address at hand and has to
go into his or her email environment anyway to get it; it doesn’t
seem as personal as sending the other person email (even though
they may be able to attach a personal note in the sharing tool); the
sharer doesn’t know what the received article will look like (e.g.
will it have advertising? Will it look like spam?); and the sharer is
concerned about privacy, both his or her own and the recipient’s,
since now the third party may retain their email addresses and will
even know about a common interest.
As a first strategy, our findings suggest that by digging deeper
into the diverse forms and functions of shared encountered
information, and by examining senders’ motivations and
recipients’ reactions, there is opportunity for innovation and
refinement in areas such as the representation of collection
elements, the facilities we give to readers for interacting with
interesting material they encounter while they are browsing (and
often looking for something else), and the modes of exchange we
offer them.
The representation of functional elements in an electronic
collection has important consequences for how readers will clip
and share them. At the very least, in a complicated multipart
electronic document like a newspaper or magazine, the ability to
clip at article boundaries and preserve the sense of the article’s
characteristics – how it is laid out, how long it is, and its other
metadata – is central for meaningful sharing. For example, one of
the study participants tried to share a New York Times article
discussing high fat versus low fat diets with two of his co-
workers. He had initially read the article in the physical
newspaper over the weekend, and explained it to a seemingly
interested group of co-workers on Monday. But when he printed
the electronic version, it was far longer than he expected:
“It was a little verbose. I didn’t realize it, because you
see it online, but when it prints out, it was like 20
pages. It was probably a little intimidating. And I don’t
even know if those guys are interested as I was in that. I
thought I’d print it out, it would be like three pages,
and it would be easy to hand to them. And it came out
as this tome.”
Some of our participants clipped and shared article subparts (as
shown in Figure 1a) or entire sections or pages of a newspaper.
This practice of either limiting how much is included (for
example, a photo instead of a whole article) or providing extra
material (for example, the entire page or a complete newspaper
section) is a useful way of focusing the recipient’s attention or
providing additional context, such as a sense of the publication
genre, the clipping’s position in the publication, or the news
locale. Representational requirements are implicit in supporting
this kind of sharing: a photo needs to be marked up as such or
superstructure such as section starts and ends must be adequately
delimited.
Additionally, the examples of sharing in the study strongly imply
that it is frustrating for readers to interrupt their primary activity
(e.g. reading, browsing, or looking for something else) to extract
and immediately enter into a complicated secondary activity (for
example, trying to copy an article from a Web periodical into a
mail message to send it). Both mark-up and interaction design
should support non-intrusive, lightweight sharing of the desired
portion of the reading material.
Modes of exchange of physical materials varied greatly also,
including:
o sending a clipping in the US mail;
o delivering the physical material in person (sometimes to
spur immediate discussion);
o placing it in a conspicuous place in the recipient’s office
or room at home (such as on desk or in an inbox at the
office or on a bed or end table at home);
o putting the clipping in a designated neutral place at
home or in the office (for example, a space on the
kitchen counter or a table in the break room);
o posting the clipping in a central location intended for
this kind of sharing, a place where it might remain for a
longer period of time such as a bulletin board; and
o delivering material through an intermediary, a third
party who can carry the material and act as a surrogate
for hand delivery.
Yet much of the electronic sharing we observed simply took place
over email or the electronic material underwent a transition to a
physical form (i.e. it was printed) and the person sharing the
encountered material used one of the physical delivery
mechanisms listed above.
It is difficult to be prescriptive about electronic means of
supporting what are essentially social activities. For example,
bulletin boards and counters suggest various spins on shared
information spaces, central or agreed-upon places where materials
can be casually left for inspection in the course of daily activities
(as opposed to more formal shared document stores, where
materials are left in a less casual, more purposeful way). While
there are a growing number of tools that promote different kinds
of information-based shared awareness [4] and tools that
implement shared information spaces [6][19], they may need to be
reformulated to have the sense of a lightweight shared place that a
kitchen counter conveys or the kinds of casual encounters with
materials that the break room bulletin board and tabletops
provide. It is an open question how these less formal shared
spaces may be integrated with more intentional collections like
digital libraries or electronic newspaper and magazine archives.
Some exchanges of material – those where the shared clippings
are saved and passed hand-to-hand – have more immediacy and
thus may be more reminiscent of peer-to-peer real-time music
sharing (along the lines of Microsoft’s ThreeDegrees
2
) or artifact-
based chatting (see [5]) than of setting up a situation where the
recipient simply encounters the material as they would on a
bulletin board. The material itself becomes the basis for
interaction and conversation.
A second strategy for using our findings in design is to discover
artificial barriers to familiar modes of sharing; technology can be
designed to avoid these barriers. For example, if Digital Rights
Management software disallows digital library contents to be
printed, then readers will encounter problems sharing electronic
materials as they do now. For example, see P20’s account
(elaborated in Table 2) of handing her project manager an article
on Interactive Bus Shelters so she can have a conversation with
him about it. Most of today’s digital libraries have no such
barriers in place; but if we consider a broader spectrum of
archives and for-pay collections, we often find publishers
choosing to implement more restrictive policies.
Similarly, to make it easier to transmit a clipping across a
distance, it may be important to allow a reader to find the
electronic form corresponding to an article in a paper publication
readily, and to identify the differences between electronic and
paper forms of the “same” article. Several of our study
participants said they checked for print articles after they had
encountered an item in an online publication or received a URL in
email; this practice was particularly important in cases where the
study participants had to be in sync with customers who might see
more or less (or different) information than the online item
contained. We expect digital libraries to continue to have a strong
relationship with comparable physical materials.
5.2 The social roles of information
In addition to having implications for technology design, the
study results also contribute to research on information behavior
and the social roles of information. Much of the existing work that
falls under the rubric of human information behavior has taken a
user needs orientation and tended to focus on search strategies to,
for example, resolve anomalous states of knowledge [3], fill
cognitive gaps [7], match an evolving understanding of how
information needs are met by information resources [1], use
complex digital libraries [12], and re-find the familiar [11].
3
Recently, researchers have been examining the role of
encountered information in everyday media such as
2
http://www.threedegrees.com/
3
This paper will not attempt to summarize many years of work in
this area. Instead, we direct attention to recent rigorous
explorations of information behavior frameworks such as that of
Pettigrew, Fidel, and Bruce in [16].
telecommunications [23], periodicals [9], and the Web [10]; this
research retains a needs focus, although it looks at information
acquisition outside of the normal retrieval framework.
Our work builds on this foundation by exploring the role of
encountered information in a social setting, where the information
content can sometimes play a secondary role to building rapport.
Thus, sharing encountered materials goes well beyond the
information content and may not be connected with information
needs. We can see this by looking at Figure 2: only a small
percentage of the information that is shared meets the immediate
information needs of the recipient. Yet those same recipients turn
around and share information that they themselves have found –
the lack of immediate utility is clearly not the most important
characteristic of the exchanged material. Rather the content
demonstrates a commonality of interests, goals, or values.
In a digital library situation, we tend to rely on users’ needs to
help them both identify a particular information resource to use
and to formulate a specific query expressing their interests. Many
browsing interfaces to digital libraries also assume fairly goal-
directed activities rather than chance encounters with interesting
material.
When we examine the motives that people have for sharing
encountered material, we find it unlikely that in this situation
human intermediaries will be replaced by agents who search
online newspapers, magazines, and other information resources in
anticipation of our interests [13] or that trusted search engine
technology such as Google will render this sort of exchange
unnecessary by virtue of giving us what we want when we need it.
All of our study participants reported exchanging both
encountered digital and physical materials, and many of those
exchanges were not simply information transactions; rather they
were intimately connected to the interaction necessary to share the
material and the social fabric that the sharing strengthens.
Current research on information brokering in a community
network setting shows that the practice “fosters social cohesion”
as people search for materials on “behalf of another person (e.g.,
relative, friend), and not always at that person’s behest” [15].
Similar to the way expertise and help is shared in newsgroups,
finding materials on the Internet relevant to the interests of others
may play an important – if supplemental – part in building social
capital and strengthening social ties within a community [22].
As we sought to define future work based our findings, we began
to think about a broader issue about shared clippings: what role
does the relatively informal exchange of encountered information
play – if any – in developing Putnam’s notion of social capital in
communities and the workplace? What implications does this
interpretation have for technologies designed to support the
sharing of encountered material? Is sharing encountered material
more apt to be used to strengthen weak social ties, where shared
interests are much of what binds people together, or is it used to
maintain strong ones where casual shared interests are better
known? Investigating the role of encountered everyday
information in building social capital – including the mechanics
of how information will be encountered and shared in digital
libraries – is one focus for future work in this area.
It strikes us as crucial that digital libraries become a venue for
encountering and sharing information beyond that which meets
immediate needs.
6. ACKNOWLEDGMENTS
We would like to thank our colleagues in Advanced Reading
Technologies at Microsoft, especially Kevin Larson, and the
study’s participants.
7. REFERENCES
[1] Ackerman, M. Augmenting Organizational Memory: a Field
Study of Answer Garden. ACM Transactions on Information
Systems (TOIS), 16, 3 (July 1998), 203-224.
[2] Bates, M. The Design of Browsing and Berrypicking
Techniques for the Online Search Interface. Online Review,
13, 5 (October 1989), 407-424.
[3] Belkin, N.J., Oddy, R.N., and Brooks, H.M. ASK for
Information Retrieval: Part I. Background and Theory.
Journal of Documentation, 38, 2 (1982), 61-71.
[4] Cadiz, J.J., Venolia, G., Jancke, G., and Gupta, A. Designing
and Deploying an Information Awareness Interface. In
Proceedings of CSCW 2002 (New Orleans, LA, USA, Nov.
16-20, 2002). ACM Press, New York, NY, 2002, 314-323.
[5] Churchill, E., Trevor, J., Bly, S., Nelson, L., and Cubranic,
D. Anchored Conversations. In Proceedings of CHI ’00 (The
Hague, The Netherlands, April 1-6, 2000). ACM Press, New
York, NY, 2000, 454-461.
[6] Cox, D. and Greenberg, S. Supporting Collaborative
Interpretation in Distributed Groupware. In Proceedings of
CSCW ’00 (Philadelphia, PA, USA, December 2-6, 2000).
ACM Press, New York, NY, 2000, 289-298.
[7] Dervin, B. Chaos, Order, and Sense-Making: A Proposed
Theory for Information Design. In Information Design (R.
Jacobson, ed.). MIT Press, Cambridge, MA, 1999, 35-57.
[8] Erdelez, S. Information Encountering: A conceptual
framework for accidental information discovery. In
Proceedings of an International Conference on Research in
Information Needs, Seeking, and Use in Different Contexts
(Tampere, Finland, August 14-16, 1996). Taylor Graham,
Los Angeles, CA, 1997, 412-421.
[9] Erdelez, S. (1999) Information Encountering: It’s More Than
Just Bumping into Information. Bulletin of the American
Society for Information Science, 25, 3 (Feb. 1999), 25-29.
http://www.asis.org/Bulletin/Feb-99/erdelez.html
[10] Erdelez, S. and Rioux, K. Sharing information encountered
for others on the Web. New Review of Information
Behaviour Research: Studies of Information Seeking in
Context 1 (2000), 219-233.
[11] Jones, W., Dumais, S., and Bruce, H. Once found, what
then?: a study of "keeping" behaviors in personal use of Web
information. In Proceedings of ASIST 2002 (Philadelphia,
PA, USA, November 18-21, 2002). Information Today, Inc.,
Medford, NJ, 2002, 391-402.
[12] Marchionini, G., Plaisant, C., and Komlodi, A. The People in
Digital Libraries: Multifaceted Approaches to Assessing
Needs and Impact. In Digital Library Use: Social Practice in
Design and Evaluation (A. Bishop, N. Van House, and B.
Buttenfield, eds.). MIT Press, Cambridge, MA, 2003, 119-
160.
[13] Nardi, B. and O'Day, V.L. Intelligent Agents: What We
Learned at the Library. Libri, 46, 2 (1996), 59-88.
[14] O’Day, V.L. and Jefferies, R. Information Artisans: Patterns
of Result Sharing By Information Searchers. In Proceedings
of COOCS’93 (Milpitas, CA, USA, November 1-4, 1993).
ACM Press, New York, NY, 1993, 98-107.
[15] Pettigrew, K.E., Durrance, J.C., and Unruh, K.T. Facilitating
Community Information Seeking Using the Internet:
Findings from Three Public Library-Community Network
Systems. Journal of the American Society for Information
Science and Technology, 53, 11 (2002), 894-903.
[16] Pettigrew, K.E., Fidel, R., & Bruce, H. Conceptual
frameworks in information behavior. Annual Review of
Information Science and Technology (ARIST), 35 (2001),
43-78.
[17] Quan-Haase, A. and Wellman, B. How does the Internet
Affect Social Capital. Forthcoming in IT and Social Capital
(M. Huysman and V. Wulf, eds.).
http://www.chass.utoronto.ca/~wellman/publications/internet
socialcapital/Net_SC-09.pdf.
[18] Rioux, K. S. Sharing information found for others on the
World Wide Web: A preliminary examination. In
Proceedings of the 63rd Annual Meeting of the American
Society for Information Science (Chicago, IL, Nov. 13-16,
2000). Information Today, Inc., Medford, NJ, 2000, 68-77.
[19] Shipman, F., Airhart, R., Hsieh, H., Maloor, P., Moore, J.M.,
and Shah, D. Visual and Spatial Communication and Task
Organization in the Visual Knowledge Builder. In
Proceedings of GROUP ’01 (Boulder, CO, USA, Sept. 30-
Oct. 3, 2001). ACM Press, New York, NY, 2001, 260-269.
[20] Toms, E. G. Serendipitous Information Retrieval.
Proceedings of the First DELOS Network of Excellence
Workshop on Information Seeking, Searching and Querying
in Digital Libraries (Zurich, Switzerland, December 11-12,
2000). http://www.ercim.org/publication/ws-
proceedings/DelNoe01/3_Toms.pdf.
[21] Vishik, C.M. Internal Information Brokering and Patterns of
Usage on Corporate Intranets. In Proceedings of Group ’97
(Phoenix, AZ, USA, November 16-19, 1997). ACM Press,
New York, NY, 1997, 111-118.
[22] Wellman B., Haase, A.Q., Witte, J., and Hampton, K. Does
the Internet Increase, Decrease, or Supplement Social
Capital? Social Networks, Participation, and Community
Commitment. American Behavioral Scientist, 45, 3 (Nov.
2001), 436-455.
[23] Williamson, K. Discovered by chance: The role of incidental
information acquisition in an ecological model of
information use. Library & Information Science Research,
20, 1 (1998), 23-40.
