Content uploaded by Vladimir Mazalov
Author content
All content in this area was uploaded by Vladimir Mazalov on Jan 23, 2014
Content may be subject to copyright.
167
Do Digital Libraries satisfy Users’ Information Demand?
Findings from an Empirical Study♣
© Magnus Lundqvist1, Vladimir Mazalov2, Kurt Sandkuhl1,
Vladimir Vdovitsyn2, Evgeny Ivashko2
1School of Engineering at Jönköping University, Sweden
[kurt.sandkuhl, magnus.lundqvist]@jth.hj.se
2Karelian Research Center, Russia
[vmazalov, vdov, ivashko]@krc.karelia.ru
Abstract
Digital libraries are an important information
source of high quality information for various
user groups in education, research and
industry. With an exponential growing amount
of digital content, digital libraries face the
challenge of enhancing the support for
information seeking. This paper takes the
users’ perspective and investigates whether the
users of digital libraries perceive that their
information demand is satisfied. The approach
taken is an empirical qualitative study with
various user groups in two different countries.
From an information demand perspective, the
main result is the support for the conjecture
that there is a coupling between the user’s
context and the information demand.
Furthermore, a usability questionnaire was
used to identify shortcomings and propose
improvements in the digital library systems
applied at the two study locations.
1 Introduction
During the last decade, the amount of information
available on the Internet, in digital libraries or in
enterprise information systems has been growing
exponentially. The main challenge of the information
society is no longer that the needed information does
not exist electronically [8], the challenge rather is to
find and provide the right information. Among the
research activities working on this challenge are
approaches from information filtering and information
retrieval [1, 4], context-based ubiquitous computing [2],
context-based decision support and problem solving [9]
and information logistics [6].
Digital libraries are an important information source
of high quality information for various user groups in
education, research and industry. Recent developments
in this area aiming at meeting the challenge of the
growing amount of digital content include the
enhancement of meta-data, enrichment of content or
meta-information systems. This paper takes the users’
perspective and investigates selected aspects of the
users’ perception of digital libraries. The guiding
question is: Do the users of digital libraries perceive
that their information demand is satisfied? This subject
can be divided into two aspects: (1) the users’
awareness of the own information demand and (2) the
usability of the retrieval tools.
The approach taken in this paper is an empirical
qualitative study investigating the above questions in
various user groups and two different application
scenarios: the digital library at Jönköping University
(Sweden) and the digital collections of the Karelian
Research Center (Russia).
The remaining part of the paper is structured as
follows: section 2 introduces the general design of the
study. Section 3 presents results form an information
demand perspective. Section 4 focuses on findings from
a usability perspective. Summary of the work and
conclusions are presented in section 5.
2 Study Design
The study consisted of two parts: the first part was
performed in 2007 at Jönköping University in Sweden,
included development of interview guidelines and a
usability questionnaire as preparatory activities, and
consisted of a pilot study and end user studies. The
second part was carried out in 2009 at the Karelian
Research Center in Petrozavodsk and used the
guidelines and questionnaire from the first part.
2.1 Study at Jönköping University
The first part in Jönköping focused on a meta-
information system used in the library of Jönköping
University, called Samsök. This meta-information
system offers a common interface for retrieving
information in various “underlying systems”, like
library catalogues, online archives and full-text
Proceedings of the 11th All-Russian Research Conference
«Digital Libraries: Advanced Methods and Technologies,
Digital Collections» - RCDL’2009, Petrozavodsk, Russia,
2009.
168
literature databases. Queries entered by the user in the
Samsök user interface are transformed to the interfaces
(i.e. query language/format and service interface) of the
underlying systems, executed in these systems, and the
results are presented in the common Samsök user
interface with possibility to continue navigation into the
underlying systems.
The purpose of the study was to investigate four
main questions:
1. How does Samsök support the end-users, in
particular in satisfying the end-user’s information
demand?
2. How does Samsök support the library’s activities
and services?
3. What are the results of evaluating Samsök from a
usability perspective?
4. What improvement potential can be identified
based on the results from the first 3 questions?
The scope of the paper is limited to the end-user
perspective, i.e. question 2 will not be discussed and for
question 3 and 4 only the end user related aspects are
included. A complete account of the results is available
in [5].
Based on the above questions, guidelines for data
collection and a questionnaire were developed for use in
end user studies consisting of sessions of the evaluator
with one individual user (respondent) at the time. The
guidelines had three purposes: to define the tasks to be
performed by the respondents, to structure the session to
be performed and to support the evaluator during the
observation. The purpose of the questionnaire was to
collect data about the usability and usefulness of
Samsök from the respondents perspective. The
questions used were a sub-set of the Questionnaire for
User Interface Satisfaction (QUIS) [7]. The selection of
questions was guided by two principles: 1) the question
should be relevant in the context of Samsök and 2) it
should be questions directed to end-users, i.e. the
respondent should be able to answer them.
In the next step, guidelines and questionnaire were
evaluated in a pilot study with two respondents. The
results were documented by recording the screen events
and recording of the users’ oral comments while using
the system according to the thinking-aloud approach
[3]. Within this study, thinking-aloud means that the
user was encouraged by the evaluator to say what
he/she is thinking and doing when using the system.
This leads to a richer set of data for the analysis work.
The results of the pilot study were used to improve both
the interview guidelines and the questionnaire. The end
user studies were performed with in total 12 users, 2 in
the pilot study and 10 in the main study part. Among
these 10 persons were 5 students, 1 researcher, 3 PhD
candidates and 1 subject teacher. The objective was to
observe a number of end-users with different roles and
background in order to get a rich set of information
regarding Samsök’s application and potential
improvements. It should be observed that the intention
was to capture qualitative data, and not to collect data
with statistical relevance.
2.2 Study at Karelian Research Center
The second part in Karelia used the second and
improved version of guidelines and questionnaire from
the Jönköping study (see 2.1). The study in Karelia was
performed using two sets of digital collections located
at the scientific digital library (http://dl.krc.karelia.ru)
and at the section “Publications” of web-portal
(www.krc.karelia.ru) KarRC RAS. These collections
are the result of long-term researches developed at
KarRC RAS.
The main questions of our study were aligned with
questions 1, 3 and 4 of the Jönköping part:
1. What are the usability results of working with
digital collections?
2. How does the digital collections' infrastructure
meet the users' information demand?
3. How can we improve the digital collections'
infrastructure?
The study was performed with 10 users. Among
these users were 5 fourth-year students, 3 PhD students
and 2 PhD researchers. Such scope of respondents
should give us the different answers to our questions.
3 Information Demand Perspective
This section summarizes the results from the end
user study from an information demand perspective.
For each respondent in the end user study, the
session which was part of the end user study started
with a pre-interview. In this pre-interview, the
respondent had to briefly describe her/his role at the
university, how familiar she/he was with using
computers, and the information demand he/she has,
which shall be the basis for the information searching.
During the information searching, the screen events and
the oral comments were recorded. After the use of the
system, the respondent was asked to what extent her/his
information demand was met by the results from the
information searching. Due to the qualitative character
of the study we chose to capture this “perceived”
relevance rather than to evaluate the found information
from a recall/precision perspective.
3.1 Observations from Jönköping
Awareness of the own Information Demand
In general, there is a clear tendency in the interviews
that the research and teaching personnel has a more
specific and better defined information demand as the
students seem to have. Two statements taken from the
interviews can serve as example. The first statement is
from a student:
I thought we should look into the area … We are
currently developing a web-portal. I thought I could
maybe search something about communities and
usability. Some theories. Input to the theoretical frame,
short theoretical presentations which I then might
apply. [Respondent 2]
The second statement is from a PhD candidate
defining his information demand as follows:
169
My research area is something called Discharge
Care Planning, which is discharge from hospitals if
someone has been patient at a hospital. In Sweden, the
term coordinated care planning is also used. This is a
quite specific track within nursing care. Currently, I am
investigating how some sort of IT system or software –
called Medics - is used for this purpose. Could be
interesting to check whether there are some
publications in the domain, but I have no idea which
keywords to use as I didn’t search for such material
before. [Respondent 6]
This obvious difference is not really surprising,
since research and teaching personnel often has a quite
well-defined, often narrow and specific work area,
which makes it easier to define the information demand.
Furthermore, the experience in using libraries is higher.
But this difference illustrates the challenges to be met
when improving usability of library systems. There
seems to be a necessity to take the user’s background
into account in order to support information searching
really well.
Another important aspect of the study was to
investigate to what extent the information demand was
met. Among the respondents, only a few perceived the
support from the Samsök system as satisfactory for
finding (enough) information meeting their information
demand. This can be illustrated with some statements
like:
Ehhh.. No, you can’t really say that. But I found a
book. I wanted to have something that is connected to
both, dialect and trust. And from Sweden. But, no. I am
not really satisfied with what I found. [Respondent 10]
No, I didn’t. I did not find any article, but I found a
book. [Respondent 4]
Is work context important for information demand?
From an information demand perspective, the main
result from the study is the support for the conjecture
that there is a tight coupling between the user’s context
and the information demand: The analysis of the data
collected in the interviews and of the observations made
during the system use shows a tight connection between
the respondent’s role (teacher, researcher, student, etc.)
and the activities for which the searched information is
needed (assignment, lectures, scientific work, etc.).
3.2 Observations from Karelia
From an information demand perspective the main
results are the following.
First of all, we haven't a significant difference in the
user groups’ awareness of their specific information
demand. Some of the students had a more specific and
well-defined information demand because they tried to
find the information helpful in their scientific work.
Experienced scientists tried to find any information that
they are interested in. It is also a consequence of
specialization of our digital libraries because
experienced scientists knew about specific areas of
publications at both sites.
Another consequence of specialization of digital
libraries is a quite big number of unsuccessful search
queries. Some of the respondents tried to find an
information in areas that haven't been studied by
scientists of KarRC RAS.
4 Usability Perspective
From a usability perspective, the data collected
during the end user study and the results of the usability
questionnaire were evaluated. The next two sections
will present the user study observations from Jönköping
and Karelia, respectively. Section 4.3 will summarize
the usability questionnaires.
4.1 Observations from Jönköping
The analysis of the data collected in Jönköping has
been structured into different categories reflecting the
activities to be supported by the Samsök system:
• Perform the selection (identify keywords for
search, combine them, etc.)
• Interpret search results
• Get full-text version of publication
This section will summarize the above results.
Perform Selection
The study showed a number of problems when
deciding about the keywords to use while searching, in
what sequence to apply them during the search and how
to express combinations (e.g. by using “and” or “or”) of
keywords.
Several informants point out the importance of
knowing in advance how a search will be performed in
order to achieve a good result. This, in combination
with an understanding of the language (the syntax used
for formulation the search condition), seems to be
essential prerequisites for having real use of Samsök.
Since the informants in this study received no training
on the tool they lack such an understanding, resulting in
an unmanageable amount of search hits (the bulk of
them being irrelevant). As a consequence, the
informants request functionality for filtering results
based on language and date. Several informants express
a need for higher competence and better support in the
search process.
One problem that was observed with most of the
informants was connected to preexisting knowledge
about databases and different concepts used in the
Samsök interface:
”But if one searches such broad fields as this there
is a risk that there will be to much, at least that is the
feeling. But at the same time this perhaps is
unavoidable and then one has to sort. And there is the
possibility to narrow in, that I saw. If I get in to this
different ones where I could choose databases one
could exclude a lot. But that requires you to know what
to exclude on beforehand”
[RE3]
The statement above exemplifies the library clients’
need for pre-existing knowledge regarding the different
databases used when searching. The observations also
revealed some practical problems with the automatic
170
selection of databases i the quick search – some users
did not realise that the search only was performed in a
selection of databases.
The observations also revealed that the informants
had problems with interpreting a number of terms in
Samsök’s interface. Meta search is not an intuitive word
and thus means that it is not obvious to the user that this
is where more advanced searches can be performed.
The meta search is appreciated after some use but is
perceived as unclear at a first glance. It is not clear how
the left part, where the databases are chosen, is to be
used. Most of the time the users click on the first list
where ”categories”, ”quick groups” and ”combine” are
instead of selecting a topic category in the lower list
despite this being functionality they ask for. Functions
for creating personal groupings of databases are
requested.
The observations also revealed misinterpretations
regarding the following terms; Topic terms was
confused with search terms, the formulations search
database and search electronic magazine were
interpreted as searching content rather than on names of
databases/publications.
To summarize the problems observed, we can group
them according to the cause of the problem:
• Database knowledge – Knowledge of relevant
databases and how to select the databases to be
used during the search
• Search competency (general) - General knowledge
about structured information search
• Search competency (Samsök) – Knowledge about
Samsök and how to express queries
• Functionality – to be able to express the selection
based on criteria important for the users (e.g.
language or time interval)
• Terminology – regarding difficulties to understand
the available choices for meta-search, including the
term meta-search as such
Interpret Search Results
Several informants commented on the large amount
of hits in the search results. This is directly connected to
the difficulties with performing selection.
A lack of understanding of the link ”view collected
hits” in the quick search results in the abortion of the
search. This is inconsistent with the interface in the
meta search. When performing a meta search no
indication of collected hits is showed in the result list
until all results are collected.
Better support for the user when deciding on
relevance is needed. This is connected to a lack of
understanding of the underlying databases:
”No, I found this a bit hard, that they are showed
like this, OK, this magazine has so and so many hits
and this has so many. It would have been much better to
just get them listed in a row and not having to continue
again by clicking in to an article or magazine because I
do not know the magazines. If I had know that I might
have been able to select in a different way but now it is
just names to me. They could just as well been named
1234567 or blue, red, green because I have no idea
what it is. It felt a bit, OK but which one should I
choose? I take the one with most hits?”
[RE10]
In the part of the interface where the search results is
listed a certain amount of problems regarding the
navigation between different views were identified.
When the view full post is shown use of the web
browsers back functionality does not return the user to
the previous page but rather to the previous post in the
list, hence it is hard to return to the list view.
Conceptually it is also reasonable to question the
use of the term weight with respect to search results as
this is not a obvious term for describing relevance,
something that contributes to confusion.
The problems observed in interpreting the search
results can be categorized as follows:
• Database knowledge – lack of knowledge
regarding the databases makes the interpretation of
the search results difficult
• Incomplete Searches – users tend to misinterpret to
what extent a search is ”completed” when they start
to look at the hits
• Navigation between views – some users had
problems to navigate between the list of search
results and the view showing details for one search
result
• Terminology – respondents had difficulties to
interpret certain system terms, like meaning of
”weight” in search results, significance of different
databases, meaning of ”get more hits”
Get Fulltext
It is not obvious how one should go about to get a
full text version of an article. Sometimes, this is done
by following an ordinary hyperlink while in other cases
it is done by means of the SFX screen. The symbol used
for SFX is unintuitive and in some views its
functionality is not explained and it is therefore
consequently unused. Furthermore, the SFX screen
gives no feedback on the existence of the article leading
to the users using JULIA instead. Many of the problems
seems to be related to the users’ lack of understanding
of the library domain, hence they do not understand the
use of and need for ”LIBRIS web search”.
Sometimes when navigating to full text versions the
user is transferred to external websites. This requires the
user to interpret and understand additional
environments to perform a successful search. As the
appearance of these sites are not a part of Samsök this is
hard to influence but there is nevertheless important to
realise that these different systems are a part of the
overall user experience.
To go from search result to full-text of a publication
caused some problems for the respondents, which can
be summarized in three categories:
Unclear how to get full-text – a quite general
problem was to access the full-text versions of
publications found during the information searching
171
Terminology – the users do not connect the SFX-
symbol with the possibility to get the full-text. The
users don’t know the terms used in the SFX-window.
Navigation between views – Samsök offers and
requires different ways to navigate to the full-text
version. This confuses the respondents who would
prefer one clearly defined way to go from search result
to full-text
4.2 Observations from Karelia
There are three main categories of results from a
usability perspective:
1. usability of the site;
2. performing the search;
3. interpretation of the search result.
Both studied sites estimated by respondents by a
single mark if they haven't seen significant difference.
Usability of the site
Marks made by respondents show that the usability
of both sites is good enough. The most part of
respondents made good marks for convenience,
usefulness, design and so on. These characteristics are
important for stimulating users to further looking for
needed information.
Terms used by the sites also didn't cause any doubts.
Performing the search
Characteristics related to performing the search have
been estimated by respondents in different ways.
Respondents hadn't a single opinion about complexity
of the search system, functionality and flexibility. It is
interesting that the more experienced users made the
higher marks for these characteristics.
Opposite estimations made by respondents for time
needed to learn about basic and additional search
functions. There is also a similar difference between
more experienced and less experienced users.
Interpretation of the search result
Respondents pointed that there is enough amount of
visualized information of the search result, but the
number of documents is very small. It is also a
consequence of specialization of digital libraries. Any
attempts to find out the areas that haven't been studied
by researchers of KarRC RAS were not successful.
4.3 Results from the Usability Questionnaire
The results from the usability questionnaire are
summarized in the following two tables. Table 1 reflects
the answers regarding the general impression. Table 2
addresses the user interface impression.
The questionnaire results gave some indications
regarding the users’ impressions of Samsök, represented
in table 1 below. It should be pointed out that the
selection in the survey is too small to derive statistically
valid conclusions about a larger population. Instead we
view the results as an indication of how the users of
Samsök perceive the application. The underlying reason
for the not so positive remarks done by the informants
(shadowed cells in the table) is according to our
perception that most of them were unsuccessful in
finding the type of material they were looking for. It
should also be noted that there were users that valued
the application as simple, powerful and rewarding
despite the fact that they never before had used it.
Perception 1 2 3 4 5
1.1 Terrible - Wonderful 2 7 1
1.2 Frustrat. - Rewarding 3 3 4
1.3 Boring - Stimulating 4 4 2
1.4 Difficult – Easy 2 2 5 1
1.5 Insufficient - Powerful 4 1 5
1.6 Rigid – Flexible 7 2 1
Table 1 – Respondents’ general perception of Samsök
The following table describes results of the study at
KarRC RAS. These results show that the search system
meet the user's purposes.
Perception 1 2 3 4 5
1.1 Terrible - Wonderful 2 8
1.2 Frustrat. - Rewarding 3 2 5
1.3 Boring - Stimulating 1 2 3 4
1.4 Difficult – Easy 2 1 2 3 2
1.5 Insufficient - Powerful 1 1 5 3
1.6 Rigid – Flexible 2 7 1
Table 2 – Respondents’ general perception of KarRC
RAS' digital libraries
The questionnaire also contained a number of
questions regarding the design and learnability of
Samsök, the results from which is listed in table 2
below. Parts of the table has been shadowed to point out
the cases where opinions strongly various between
different informants.
Perception 1 2 3 4 5
2.1 Design 2 4 2 2
2.2 Terminology 3 1 3 3
2.3 Graphic symbols 2 4 1 3
2.4 System status 1 3 5 1
2.5 Feedback (content) 2 4 3 1
2.6 Feedback (visibility) 2 3 5
2.7 Search results – amount of
information 2 3 4 1
2.8 Learning - basic 1 5 4
2.9 Learning - advanced 1 1 3 3 2
2.10 Navigation 1 1 2 5 1
2.11 Response time (search) 2 3 3 2
2.12 Response time (navigation) 1 3 1 2 3
Table 3 – Respondents’ impression of Samsök’s user
interface
172
The terminology in the interface was perceived as
relatively clear while the graphical symbols was
considered harder to interpret. Regarding the systems
status, i.e. how easy it is to understand what the system
is doing at the moment, the answers are polarised. This
is most likely due to the users’ different experience of
using web applications. The feedback given by the
system gets a vaguely positive judgment. The users
generally think that it is relatively simple to learn the
simpler parts of Samsök while the more advanced parts
(meta search) is perceived as more difficult to
understand. The navigation was by most perceived as
relatively simple to handle while the response times
when searching and navigating indicates certain
problems. Especially the response time for searches
hints that involved servers have different response times
– some users have not experienced this as a problem
while others have.
Themes that stands out in the survey is according to
us that Samsök suffers from less than stable response
times and that the users’ impressions on a whole leans
towards less positive judgements such as boring and
frustrating.
The following table shows marks made by
respondents of KarRC RAS.
Perception 1 2 3 4 5
2.1 Design 3 2 5
2.2 Terminology 1 5 4
2.3 Graphic symbols 3 7
2.4 System status 1 7 2
2.5 Feedback (content) 1 5 4
2.6 Feedback (visibility) 1 9
2.7 Search results – amount of
information 2 1 2 3 2
2.8 Learning - basic 2 1 4 3
2.9 Learning - advanced 2 1 3 3 1
2.10 Navigation 3 5 2
2.11 Response time (search) 2 3 5
2.12 Response time
(navigation) 1 1 4 4
Table 4 – Respondents’ impression of KarRC RAS' digital
libraries user interface
The strict design with absence of superfluous
elements of design and functions makes the interface
convenient. But the lack of information makes users
dissatisfied.
5 Conclusions
Six categories of experiences form the use of
Samsök were discussed in section 3.3 Three of these
are directly connected to different phases in the search
process; Selection, Interpretation, and Collecting full-
text. The remaining three categories are more connected
to the overall use of Samsök; General opinions, reasons
for contacting the library, and proposals for further
development. Within each category a number of
problematic themes have been generated from the
interviews and observations. The following themes has
been identified:
Database knowledge – a basic understanding of
academic databases is required to utilise Samsök. This
introduces problems to the activities, selection and
interpretation.
Search competence (general) – information
searching requires some general knowledge that many
of the survey’s informants do not have. An example of
this is competence in evaluating the quality of different
types of publications as well as the competence to, for a
given problem, identify relevant topics and search
terms.
Search competence (Samsök) – viewed as a tool
Samsök requires its users to have some knowledge
regarding how to formulate search terms and selecting
suitable databases for the meta search. Several
informants had problems with these parts.
Unclear access to full-text – there is at the moment
several different ways to access full-text versions of
articles, something that confuses users and in the worst
case scenario means that they do not understand that a
full-text version is available.
Terms and symbols – there is a number of terms and
symbols used in Samsök that is hard to understand for
the uninitiated users. One such term is meta-search,
another is weight and the symbol used for SFX a third.
Requested functionality – some functionality, with
respect to the users’ information demand, is missing or
hidden in Samsök. An example of this is the possibility
to search based on language and dates. Other examples
of the same problems identified by the informants are
the possibility to search within search results as well as
gaining simple access to search history.
Unfinished searches – it is possible to view
collected hits despite the fact that the search still is
ongoing. This is not obvious to the user in the current
design. Furthermore, it is not obvious that more hits and
then with higher relevance can be collected with the
function ”collect more hits”.
Navigation between views – a number of problems
in Samsök is related to the navigation. One such
problem is that the use of the browsers back-button
breaks the expected behaviour of taking the user back to
the previous screen. Another is that the linking to
external documents is inconsistent and unintuitive.
Response times – the response times of the system
vary depending between the different observations,
resulting in negative judgements from the informants.
Overall perception – The survey showed on a
frustration amongst some of the informants regarding
the use of Samsök. Some of them even perceived the
system as a boring tool.
The study of the digital libraries of KarRC RAS
highlighted a different main problem – the lack of
content. On the one hand our digital libraries aims to
make public the work of the researcher of KarRC RAS,
but on the other hand the small amount of content
makes users dissatisfied in looking for specific
173
information. The consequence of this is a small number
of users who regularly use the digital libraries. With a
growing amount of content, other usability issues might
be raised, like for example navigation in large lists of
hits for a query.
6 Summary
This paper investigates whether the users of digital
libraries perceive that their information demand is
satisfied. The approach taken is an empirical qualitative
study with various user groups in Jönköping and
Karelia. This study includes two aspects: the users’
awareness of the own information demand and the
usability of the retrieval tools.
From an information demand perspective, the main
result from the study is the support for the conjecture
that there is a coupling between the user’s context and
the information demand: The analysis of the data
collected in Jönköping shows a tight connection
between the respondent’s role (teacher, researcher,
student, etc.) and the activities for which the searched
information is needed (assignment, lectures, scientific
work, etc.). With respect to usability, there seems to be
a necessity to take the user’s background into account in
order to support information searching really well.
Furthermore, in the Jönköping study there is a clear
tendency that the research and teaching personnel has a
more specific and better defined information demand as
the students seem to have. This observation from
Jönköping that researchers seem to be more aware of
their information demand was not confirmed in the
Karelian part.
The usability questionnaire was helpful in
identifying shortcomings and proposing improvements,
both for the Samsök system in Jönköping and the digital
collections in Karelia. However, the two systems are far
too different regarding user interfaces, functionality and
amount of content that a comparison of the findings
should be considered. A commonality between both
cases is that we observed that usability was graded
worse by those users who were not successful in
retrieving content meeting their information demand.
The main limit of the research presented here is the
limitation to just two digital libraries/collections and to
just groups of 10 end users in every part of the study. It
would be worthwhile and interesting to include a larger
number of both digital libraries and users.
References
[1] Belkin, N. J. and Croft, W. B. (1992). Information
filtering and information retrieval: Two sides of the
same coin? Communications of the ACM, 35 (12),
29–38
[2] Dey, A. K. (2000). Providing Architectural Support
for Building Context-Aware Applications. PhD
thesis, College of Computing, Georgia Institute of
Technology.
[3] Lewis, C (1982) Using the ”thinking-aloud”
method in cognitive interface design (IBM
Research Rep. No. RC 9265 [#40713]). Yorktown
Heights, NY: IBM Thomas J. Watson Research
Center
[4] Palme J. (1998). Information filtering. In
Proceedings of the 12th Biennial ITS (International
Telecommunications Society) Conference,
Stockholm.
[5] Samsök: Handlingsbarhet, behovsinriktad
användarstudie och tutorial. Rapport för projekt
med utvecklingsbidrag från Kungl. biblioteket /
BIBSAM; Högskolebiblioteket i Jönköping. Dnr.
vid Kungl. biblioteket / BIBSAM: 63-612-2005.
[6] Kurt Sandkuhl: Information Logistics in
Networked Organizations: Selected Concepts and
Applications. Enterprise Information Systems, 9th
International Conference, ICEIS 2008. LNBIP,
Springer.
[7] Shneiderman, B, Plaisant C (2005) Designing the
User Interface – Strategies for Effective Human-
Computer Interaction. Addison-Wesley. ISBN 0-
321-26978-0.
[8] Skjong, R., Johnsen, Ö., Weitzenböck, J.,
Brynestad, S., Mestl, T., 2004. Technology
Outlook, p.109. Det Norske Veritas, Norway, ISBN
82-515-0300-0
[9] Smirnov, A., Pashkin, M., Chilov, N., and
Levashova, T. (2005). Ontology–Based Knowledge
Repository Support for Healthgrids. In Proceedings
of Healthgrid 2005: From Grid to Healthgrid,
(Solomonides T., McClatchey R., Breton V.,
Legré Y. and Nørager S. Eds.), 47—56. IOS Press.
Удовлетворяют ли электронные
библиотеки информационным запросам
пользователей? Эмпирическое
исследование
М. Ландквист, В. Мазалов, К. Сенкюль,
В. Вдовицын, Е. Ивашко
Электронные библиотеки являются важным
источником информации для различных групп
пользователей в области промышленности,
образования и науки. Взрывной рост объемов
информации, представленной в цифровом виде,
ведет к тому, что пользователи электронных
библиотек все чаще сталкиваются с проблемами
поиска информации. В статье представлено
исследование, цель которого — оценить насколько
хорошо пользователи могут удовлетворить свои
потребности в информации с помощью
электронных библиотек. Объектом исследования
данной эмпирической работы являются различные
группы пользователей двух стран. С точки зрения
удовлетворения информационных запросов,
основной полученный результат — это
подтверждение наличия связи между контекстом
поиска и информационными запросами. На основе
специально разработанного опросника, среди
174
пользователей, участвующих в исследовании, было
проведено анкетирование для определения
недостатков и возможностей улучшения систем
поддержки электронных библиотек.
♣ Part of the work was supported by RFFR (grant No 08-07-
00085а). Another part was financed by The Swedish Royal
Library in context of the Samsök project, The authors wish to
thank their co-workers from Jönköping International Business
School and Jönköping University Library, in particular
Thomas Albertsen and Jonas Sjöström.
