Case Study on social software in distributed working
environments
Barbara Kieslinger, Margit Hofer, Yiwei Cao, Karsten Ehms, Sebastian
Fiedler, Anna-Kaarina Kairamo, Ralf Klamma, Beate Krause, Milos Kravcik,
Tommi Ryyppö, et al.
To cite this version:
Barbara Kieslinger, Margit Hofer, Yiwei Cao, Karsten Ehms, Sebastian Fiedler, et al.. Case
Study on social software in distributed working environments. Research report of the ProLearn
Network of Excellence (IST 507310), Deliverable 15.2. 2007. <hal-00591560>
HAL Id: hal-00591560
https://hal.archives-ouvertes.fr/hal-00591560
Submitted on 10 May 2011
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
D15.2 – Case Study on social software in distributed working environments
Network of Excellence for Professional Learning
PROLEARN
European Sixth Framework Project
Deliverable
D15.2
Case Study on social software in distributed
working environments
Barbara Kieslinger, Margit Hofer
Editors
Work Package
15
Status
Draft
Type
Report
Date
22. May 2007
The PROLEARN Consortium
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Universität Hannover, Research Center L3S, Germany
Deutsches Forschungszentrum für Künstliche Intelligenz GmbH (DFKI), Germany
Open University (OU), UK
Katholieke Universiteit Leuven (K.U.Leuven) / ARIADNE Foundation, Belgium
Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. (FHG),
Germany
Wirtschaftsuniversität Wien (WUW), Austria
Universität für Bodenkultur, Zentrum für Soziale Innovation (CSI), Austria
École Polytechnique Fédérale de Lausanne (EPFL), Switzerland
Eigenössische Technische Hochschule Zürich (ETHZ), Switzerland
Politecnico di Milano (POLIMI), Italy
Jožef Stefan Institute (JSI), Slovenia
Universidad Polictécnica de Madrid (UPM), Spain
Kungl. Tekniska Högskolan (KTH), Sweden
National Centre for Scientific Research “Demokritos” (NCSR), Greece
Institut National des Télécommunications (INT), France
Hautes Etudes Commerciales (HEC), France
Technische Universiteit Eindhoven (TU/e), Netherlands
Rheinisch-Westfälische Technische Hochschule Aachen (RWTH), Germany
Helsinki University of Technology (HUT), Finland
1
D15.2 – Case Study on social software in distributed working environments
Document Control
Title:
Case study on social software use in distributed working
environments
Type:
public report
Editors:
Barbara Kieslinger, Margit Hofer
E-mail:
kieslinger@zsi.at, hofer@zsi.at
Origin:
BOKU/CSI
AMENDMENT HISTORY
Version
Date
Authors
Description/Comments
0.1
22/05/2007
Barbara Kieslinger
outline and first version
Contributors
Name
Company
Yiwei Cao
RWTH
Karsten Ehms
Siemens
Sebastian Fiedler
BOKU/CSI
Margit Hofer
BOKU/CSI
Anna-Kaarina Kairamo
HUT
Barbara Kieslinger
BOKU/CSI
Ralf Klamma
RWTH
Beate Krause
L3S, University of Kassel
Milos Kravcik
OUNL
Tommi Ryyppö
Dicole Oy
Marc Spaniol
RWTH
Gerd Stumme
L3S, University of Kassel
Fridolin Wild
WUW
2
D15.2 – Case Study on social software in distributed working environments
Legal Notices
The information in this document is subject to change without notice.
The Members of the PROLEARN Consortium make no warranty of any kind with
regard to this document, including, but not limited to, the implied warranties of
merchantability and fitness for a particular purpose. The Members of the PROLEARN
Consortium shall not be held liable for errors contained herein or direct, indirect,
special, incidental or consequential damages in connection with the furnishing,
performance, or use of this material.
3
D15.2 – Case Study on social software in distributed working environments
Table of Content
1 EXECUTIVE SUMMARY
5
2 INTRODUCTION
6
3 WHAT IS SOCIAL SOFTWARE
7
4 HOW ARE SOCIAL SOFTWARE APPLICATIONS CURRENTLY USED IN
DISTRIBUTED WORKING ENVIRONMENTS
9
5 CASE STUDIES
12
5.1 Case Study 1: Subsidiary of a global pharmaceutical company
5.1.1 Introduction to the case study
5.1.2 Objectives of the study
5.1.3 Methodology
5.1.4 Analysis
5.1.5 Conclusions
12
12
13
14
15
17
5.2 Case Study 2: Siemens
5.2.1 Introduction to the case
5.2.2 Objectives
5.2.3 Methodology
5.2.4 Analysis
19
19
20
20
21
5.3 Case Study 3: Nextspace in iCamp
5.3.1 Introduction to the case
5.3.2 Objectives
5.3.3 Methodology
5.3.4 Analysis
26
26
27
27
28
5.4 Case Study 4: Bibsonomy
5.4.1 Introduction to the case
5.4.2 Objectives
5.4.3 Methodology
5.4.4 Analysis
5.4.5 Conclusion
34
34
35
35
37
44
6 CONCLUSIONS
47
7 REFERENCES
49
4
D15.2 – Case Study on social software in distributed working environments
1 Executive Summary
This document outlines four different case studies on the use of Social Software in
distributed working environments. While two case studies focus on the corporate
world, two other case studies look at the use of Social Software in academic
workplaces.
The concluding remarks identify some common benefits as well as issues with the
use of Social Software and to derive some further research challenges from the
different cases. The findings for future research need to be directed towards
strategies for corporate learning and working environments to effectively integrate
Social Software solutions for very specific needs in different institutional cultures.
Therefore the continuation of social software applications in practice, including some
other case studies from non-corporate or non-academic, will be essential for WP 15.
5
D15.2 – Case Study on social software in distributed working environments
2 Introduction
This deliverable describes four different case studies on the use of Social Software in
distributed working environments. Two case studies focus on the corporate world and
two case studies look at the use of Social Software in academic workplaces.
The general objectives of the case studies can be summarised as following:
•
•
•
to obtain rich descriptions of applications of social software in a variety of
distributed working environments
to provide evidence of (un)successful use of social software in distributed
working environments
to identify possible challenges for further developments in order to support the
use of social software in distributed working processes
Apart from these general objectives each case study also has its own focus and
addresses more specific questions regarding the use of Social Software, mainly
related to knowledge building and knowledge sharing, social network building and
informal learning processes.
Before going into the specific case studies we find it useful and important to dedicate
a section on defining the common grounds regarding the concept of Social Software.
Thus Section 3 will define the shared understanding of our research group regarding
this term.
Section 4 will then give a short overview of current practices regarding Social
Software in distributed working environments with a special focus on Social Software
in the corporate world.
The following sections will be dealing with the different case studies in detail and
finally we will give some concluding remarks trying to identify some common benefits
and problems with the use of Social Software and to derive some further research
challenges from the different cases.
6
D15.2 – Case Study on social software in distributed working environments
3 What is Social Software
In order to discuss case studies on the use of Social Software in distributed working
environments we find it necessary to elaborate on the understanding of this term and
provide a definition that we share amongst our research community.
Whereas there is currently no official agreement on the definition of the term “Social
Software” a number of approaches have come forward. Some people tend to list a
number of software applications that fall according to their understanding under this
term, e.g. e-mails, weblogs, wikis, chat rooms, etc. (Tscherteu, 2003) whereas others
talk more about a philosophical approach that is included in the term “Social
Software”.
Within the PROLEARN community we agree that Social Software concentrates on
the link-up between social entities in digital social networks and their interaction
(Wellman et al., 2002; Shirky, 2003; Bächle, 2006; Klamma et al., 2006). Whereas
community information systems, which have been typically used for communities of
practice (Wenger, 1998), support specific social entities they do not focus on the
relationships among the entities as compared to Social Software. One of the main
outcomes that Social Software supports is a digital social network.
Social software is changing the ways in which people communicate and cooperate.
Physical location, temporal constraints, and face-to-face communication are
becoming less important. The impact and the development wave of social software
make significant sense to the society since the industrial revolution. It changes the
way to make a deal, to travel, and especially to learn (Alexander, 2006).
Digital social networks are networks reflecting the social structures, while social
structures are created by the usage of digital media. Hence, media and communities
are equally involved. Social software is thus innovative as it bridges media and
communities in a seamless way. The status quo of the Internet’s development trend
is “Smarter, Simpler, Social”, which features social software (Bryant, 2003) and the
Web 2.0 technologies (O’Reilly, 2005).
The terms Social Software and Web2.0 are often used together or even
synonymously. Although we do not understand these terms as synonyms, they are
closely related to each other. Basically the term Web2.0, which was coined by
O’Reilly in 2003 and has since then been widely adopted, refers to a new way of how
the World Wide Web is used. According to Tim O'Reilly (2006), "Web 2.0 is the
business revolution in the computer industry caused by the move to the internet as
platform, and an attempt to understand the rules for success on that new platform."
Generally speaking, Social Software employs Web 2.0 technologies. Users of social
software often act on a global stage. Each weblog entry, each uploaded video, each
shared bookmark can be viewed, commented, modified and re-distributed by every
other user if published on the Internet. The complexity induced by this universe of
possible interactions is threatening. It is far beyond the computational efforts of a
single person.
7
D15.2 – Case Study on social software in distributed working environments
Digital social networks are mainly realised by means of computer-mediated
communication (CMC) (Licklider et al., 1968). While social interaction in professional
communication was always there, the new is the persistence and global access to
the social interaction followed by a cultural change of the Internet (Sixtus, 2005), due
to the tremendous number of broadband access in western industrialized countries.
Moreover, every network supports certain types of media. They influence how the
communication links between the members are created. For the purpose of the
following case studies an important difference in the use of media has to be pointed
out here. Since we are looking at the use of Social Software in working environments
- often in private enterprise - the above-mentioned global access to the content and
the social network is restricted and often only granted within the boundaries of the
organisation, the Intranet. In the following chapter we will go into further detail on how
social software is currently implemented in working environments.
Regarding social software in technology enhanced learning the potentials are still not
exploited to its full extent. Companies as well as schools and universities are just
starting to realise the potentials of social software for specific learning purposes.
Dalsgaard argues that social software such as del.icio.us and Wikipedia are useful
tools for personalised learning, together with the traditional learning management
systems (LMS) (Dalsgaard, 2006). Social software is still often considered as an
auxiliary tool for learning. However, a technology enhanced learning scenario for a
scholar who uses just the social software tools can be well imagined. Research
papers and books are searched on Google Scholar instead of in the libraries.
Important research materials are managed at CiteULike (www.citeulike.org), which is
a social bookmarking system for research citations. Professional knowledge can be
explored in weblogs as well as social bookmarking systems. Especially informal
learning processes that usually take place in social networks can be supported by an
internal “blogosphere” within a company. Wiki applications can support the internal
knowledge management processes by providing updated information that is jointly
published by a group of employees and supports a bottom-up approach. Papers
writing can be cooperated in online cooperation-enabled text editors. Researchers
are thinking of the solutions e.g. to cite an article which is not published in the
journals or proceedings, but maybe in a personal weblog.
The case studies to be presented subsequently will show some of the attempts made
so far to make use of social software for some of these scenarios.
8
D15.2 – Case Study on social software in distributed working environments
4 How are social software applications currently
used in distributed working environments
Use of information and communication technology in organizations is self-evident
these days. Several communication technologies, such as email and instant
messaging have been used in organisations for over a decade already (Wellman &
Hampton 1999). The novel tools in the field, social software technologies, have
emerged to support knowledge and information sharing, and we can see a
tremendous jump of the use and interest on them.
These new technologies and services benefit from the wisdom of crowds
(Surowiecki, 2004), which simply means that the many can be smarter than the few.
The role of users changes from passive consumers to active producers that generate
content and form social networks. Some experts even speak about “collective
intelligence” that is quicker, more up to date, deeper, and wider than the traditional
models based on authorities and that derives from innovation mechanisms as
differentiation and integration, competition and collaboration. This collective
intelligence is “emergent” as a self-controlled network knowledge. As Tim O’Reilly
says (Spiegel Special, 2007), the point is to make collective intelligence useful – it is
not just about expression of opinions, but about distributed data development and
real time intelligence.
If we focus now on the corporate world and their distributed working environments,
let us consider how this collective intelligence can be supported by Social Software.
As an example we would like to mention a tool for social bookmarking developed by
IBM which is called “dogear” and has been implemented in the enterprise (Millen et
al., 2005). Generally, applications for shared bookmarking have certain common
features:
•
•
•
individuals can create personal collections of bookmarks and share them with
others,
users can assign keywords or tags for each bookmark – it can belong to more
than one category,
social nature of their use – social browsing according to user names or tags.
Whereas social bookmarking has already been successfully used within the
academic world for many years, the company was interested in whether large
enterprises would also benefit from social bookmarking systems. As the authors of
the experiment claim, it shows great potential for using the application to improve
sharing, expertise location, and support of communities of interest within the
enterprise.
4.1 Current Studies
Academic research on the use of social software in organizations is still relatively
rare. However, lots of organizations have adopted them; for instance, approximately
30 % of companies worldwide are already using or planning to use blogs and/or wikis
at the moment (McKinsey 2007).
9
D15.2 – Case Study on social software in distributed working environments
Although we have noticed an increasing interest of companies in Web2.0 and Social
Software on a global scale, a recent study from the Gartner Group gives a warning
that European enterprises are about to miss this global trend and lag behind the
implementation of Social Software (Computerwoche 11.07.2006).
If we have a look at Australia, which is the continent with the highest Internet usage
by its inhabitants (ITU, 2005) – more than 70% - we seem to spot a different trend. In
March 2007 the Australian Flexible Learning Framework published a report on the
use of social software for knowledge sharing and capability development in
vocational education and training delivery (AFLF, 2007). Their research showed that
social software is valuable in enhancing and enriching knowledge sharing, capability
development and the teaching and learning experience. The successful use of social
software relies on a spirit of openness and a willingness to share and collaborate as
well as an enabling culture. Having an authentic need, being relevant to the context
and appropriate for the client were the critical elements. The trends are indicating
that social software will become ubiquitous largely because it is the technology
already widely used by the ‘Net Gen’ – the new generation of workers and learners,
which is a significant client group of the future. The use of social software in
vocational education and training is in its early days of adoption, being trialled by the
innovators and early adopters.
4.2 Typical Use Patterns
When talking about the current use of social software in companies we can generally
distinguish between the following heavily simplified patterns of internal and external
use:
Fig.1 Use patterns of social software in enterprises
Other authors such as Zerfaß (2005), who has been analysing weblogs as corporate
communication channels, distinguishes between eight functions of corporate weblog
(knowledge transfer blog, contract negotiation blog, CEO blog, public relations blog,
product blog, service blog, customer relationship blog, and crisis blog). This
distinction for one specific social software application would also fit into our scheme
of internal and external use. We prefer to focus on this core distinction that relates to
the actors (authors and audience) than to the content published via certain tools.
10
D15.2 – Case Study on social software in distributed working environments
What we call “Internal use” here is often practiced for knowledge management
purposes within an organisation. There is usually no public access provided as the
content is only published and available on the Intranet, not on the Internet. Within the
internal usage scenario we can distinguish between the use restricted to certain
persons (e.g. the CEO) or to a group of people (e.g. the management team) or the
whole staff members. In the case studies we will see all three types of use patterns
covered.
On the other hand companies are using social software also more and more for
external purposes. This is communication with external partners, customers, and the
general public. In this way a company can receive a valuable feedback from external
bodies. As an example of how an enterprise uses social software externally is
Salesforce.com: they developed software that is now available for other developers
who do adaptations and the company uses social software for transparent knowledge
exchange between customers, external developers, who are also clients, and staff
members. The whole business model has been adapted to this creation of collective
intelligence amongst the different actors. With the right product and the right
community service they have achieved great customer loyalty.
4.3 Outlook
Web 2.0 and social software represent a strong wave that has the potential to
change in a substantial way our society, politics, and economy. Those organizations
that will not stick with the old models and will not fight against the new development,
but that can recognize this new power and benefit out of it, can become successful in
the future. A powerful new economic force in a world where the Internet allows
almost unlimited choice was named as the theory of Long Tail (Anderson, 2006). Its
author Chris Anderson has identified an important truth: the future does not lie in hits,
but in what used to be regarded as misses – the long tail of the traditional demand
curve. The result is a cultural richness when everybody everywhere can find
something to his or her taste.
Since the use of social software in corporate environments is still rather young there
is not much research work available yet on how social software has been
implemented in enterprises yet. In the following case studies we will put our focus on
what is called “internal use” in the above-mentioned scheme. We will not be
analysing applications of social software for marketing, public relations, etc.
purposes. Our focus will be on the “internal use” for the purpose of knowledge
building and knowledge sharing and thus informal learning within distributed working
environments. We will discuss cases in the corporate world, but also in academia. In
the academic world social software seems to have been taken up and integrated into
working processes earlier than in the private sector and may thus give us some
indications for the further implementation in enterprises. Previous technologies such
as text editors, e-mail or more recently instant messaging have gone through a
similar adoption phase starting with students and academia, encountering first
corporate suspicion and finally reaching growing acceptance in the workplaces of
private enterprises (Lovejoy 2003, Perin 1991).
11
D15.2 – Case Study on social software in distributed working environments
5 Case Studies
5.1 Case Study 1: Subsidiary of a global pharmaceutical company
5.1.1 Introduction to the case study
The aim of this case study was to study potential benefits of using social software in
an organization and to identify key elements of a successful social software adoption
process. We try to have a closer look at the key factors of the introduction and use of
social software in companies. The case study is based on research done for a
Master's thesis (Ryyppö 2007) on the same topic. One primary case and two
supporting cases of introduction and testing of social software in companies are
covered. In addition, a comparative analysis based on the primary and supportive
cases is presented.
5.1.1.1 Primary case: Introduction and testing of social software
In February 2006 an internal development project was started at a Finnish subsidiary
of a global pharmaceutical company. The purpose of the internal development
project was to explore new ways to improve information sharing within the company.
The main focus was on developing individual employees’ attitudes and behaviour in
terms of acquiring, organizing and distributing information in the company. It was
decided that social software would be tried out in the company as a sub-project of the
internal development project. One of the focuses of the sub-project was to find out
whether social software use should be promoted on a wider scale in the company in
the future. The social software application introduced and tested was provided by an
application service provider (ASP). The application they offered included blogs, wikis,
discussion forums, aggregators, and some other functionalities.
The following principles and approaches in terms of introducing and testing the
application were decided:
- One work area within the application was created in order to test blogging for
enhancing communication and interaction within and between the
management team and the project manager of the internal development
project. The purpose was to advance the development project by giving the
management team a chance to comment on the blog entries of the project
manager and to promote idea exchange within the management team. The
aim was also to help to document the project.
-
A second work area was created for everyone involved in the project in order
to test the social software application on a wider scale. The testing was done
on a voluntary basis, i.e. only those employees who expressed their
willingness to participate in the sub-project were involved. The purpose was to
familiarize with social software and gain experience of using the application.
The main focus was in the beginning on blogs although other functionalities of
the application used we also introduced and available for use.
The introduction and testing began with a presentation at the end of May 2006. After
the presentation seven people (two members of the management team and 5
12
D15.2 – Case Study on social software in distributed working environments
product managers) expressed their willingness to take part in the testing of the social
software application. It was acknowledged that both of the approaches meant extra
work for the participating parties as no trade-offs between everyday activities and
chores were made. For this group of volunteers a half-day workshop was arranged in
mid-June 2006. The purpose of the workshop was to set-up, to demonstrate and to
gain hands-on experience of using the application for blogging between the members
of the project team. Hence, the main focus was on blog writing, reading, and
commenting. However, the use of the aggregator was also demonstrated and tested.
Wikis and forums were covered very shortly as well. The management team
members were guided individually how to log in into application, read blog entries
and comment on it. After holidays, in August, two more product managers joined in
the testing of the social software application. Both were given hands-on instructions
on the use of application. The testing phase of the social software application took
place till the end of September 2006. At the end of the project an Internet survey was
made.
5.1.1.2 Supporting cases: CEO blogs
The main purpose of the supporting cases was to get supporting data for the primary
case. Furthermore, the supporting cases provided data of already implemented use
of social software for business purposes in an organization. These supporting cases
were CEO blogs open for all the employees of two big companies.
The first supporting case concerns a Finnish airline company in which the CEO of the
company started writing an internal blog. The blog written by the CEO had been
regarded as an interesting media for quite some time before its introduction at the
company. Blogging was considered a well-known and interesting phenomenon, and it
appeared to fit well in the existing interactive communication systems of the
company. Hence, as a new CEO suggested starting to write a blog alongside with the
beginning of his tenure in the beginning of the year 2006, a CEO blog was
introduced. During the first two months of tenure (“trainee period”), the blog was
written on a weekly basis. Later on, as the content of the blog postings concerned
mostly events that had taken place in the company and its competitive environment,
and was written on a monthly basis in intranet.
The second supporting case concerns a leading newspaper publisher in Finland. As
in the first supporting case, the CEO blog was introduced in the beginning of the year
2006 and it coincided with the employment of a new CEO. The idea came from the
new CEO who was aware of similar attempts being implemented elsewhere.
Nevertheless, it took half a year of consideration and a testing phase with a small
number of critical readers before the CEO decided to publish his blog to the whole
organization. After the official launch the CEO blog was written on a regular basis (34 times a month). In the beginning, the postings were relatively short and the content
focused on current events. Later on, the postings became longer and more story-like,
focusing on the CEO’s personal viewpoints. Symbolically this was seen as an
expression of the CEO’s willingness to be more approachable.
5.1.2 Objectives of the study
The main research question in this case study is: What are key issues to consider if
social software is to be successfully introduced and adopted in an established
organization?
13
D15.2 – Case Study on social software in distributed working environments
The research question draws from the recognition that the number of established
organizations adopting social software appears to increase rapidly. Therefore, some
guidelines in terms of its introduction and adoption are needed.
5.1.3 Methodology
The research approach taken was exploratory case study with primary case and two
supporting cases. The approach chosen allowed fieldwork to be undertaken prior to
the final definition of study questions and the research to allow also intuitive paths.. A
narrative strategy for description and analysis of process data was used, following
the ideas described by Langley (1999).
In the primary case, an introduction and testing of a social software application in an
organization was studied. This part of the research took place between April and
September 2006. The supporting cases comprised two cases of blog usage in
organizations. This part of the research took place in October 2006.
The primary research process undertaken had many elements of action research to
realise it in terms of data collection and analysis. The researcher worked in the
organization and with its members. In addition, the matter under research in the
primary case was of genuine concern to the organization and there was intent by the
organization members to take action based on the intervention. (e.g. Eden & Huxham
1997)
In order to provide a strong substantiation of constructs, several data collection
methods were used. In the primary case the social software application introduced
and tested provided many possibilities in terms of quantitative and qualitative data
collection. All blog entries, and any comments to them, were monitored. In addition,
any wiki pages and discussion forum posts created, posted, edited or deleted were
tracked. In addition to the action research process, an Internet survey was carried out
in the primary case among all involved. The questionnaire was sent to 15 people all
of whom had participated in the project. Ten people answered it. It provided both
quantitative and qualitative data on the use of the social software application. In the
supporting cases, phone interview was the method used for data collection. Two
phone interviews were carried out, one for each case. In addition, a serendipitous email correspondence took place in the latter case. Hence, all data collected in the
supporting cases was in qualitative form.
Analysis consisted of within-case analysis and a comparative analysis between the
primary and supportive cases. The strategy chosen for analysis was narrative
strategy which involved construction of a detailed story from the raw data.
Furthermore, the strategy was deemed appropriate as its key anchor point is time
and its focus is on contextual detail. (Langley 1999).
In within-case analysis the idea was to become intimately familiar with each case as
a stand-alone entity. This type of approach allowed the unique patterns of each case
to emerge before patterns across cases were generalized (Eisenhardt 1989). In
practice this meant analyzing the user data of the social software application and the
Internet survey used in the primary case as if trying to answer the research
14
D15.2 – Case Study on social software in distributed working environments
questions. In a similar way, the phone interviews and the e-mail correspondence
were analyzed in the supporting cases. The tactic chosen in comparative analysis
was to select dimensions for which within-group similarities coupled with inter-group
differences could be found (Eisenhardt 1989).
5.1.4 Analysis
5.1.4.1 Results of the primary case
The introduction and testing of the social software application was not a success in
terms of actual usage. After all, the actual usage of the blogs and wikis remained in
this case very low. On the other hand, discussion forums of the application were a
relatively popular media in both work areas even though they had not been
introduced.
Even though awareness of and familiarity with the different tools utilized during the
project increased, the tools were not seen as particularly useful in regard to present
work of the respondents. Furthermore, the application itself was seen as difficult to
use, logging into to it was extra work and required remembering yet another
password, and it was seen as separate from the existing ICT tools used in the
company. Nevertheless, 9 out of ten respondents of the final evaluation perceived
social software to be positively significant in regard to the future activities at the
company. In addition, the respondents had many views on why and how to promote
its use in the organization. Many of these views related to concrete problems in terms
of information sharing within the company. Therefore, as a result of the project, a
recommendation to promote the use of social software in the organization in the
future was given to the management team.
In sum, considering the challenges that had to be faced in the adoption and the
increase in the level of familiarity in terms of social software, a lot was accomplished.
This is especially true in terms of the internal development project, the purpose of
which was to develop the attitudes and behavior of individual employees in regard to
acquiring, organizing and distributing information within the company. Furthermore,
the introduction and testing of social software gave new insights and ideas in terms
of its applicability in the organization that had not been recognized earlier but were
now considered worth pursuing on a wider scale.
5.1.4.2 Results of the supporting cases
Concerning the first supporting case, in the beginning, the main business application
of the blog was to help the new CEO to communicate his first impression of the
company and his vision for it to the employees. The CEO blog was seen as a part of
internal communication system in the company. Its purpose was to support company
strategy and increase the likelihood of successful communication on the CEO’s part
by adding another channel through which to communicate to the employees. In order
to increase the likelihood of successful communication the blog postings were also
published in the next monthly newsletter.
The interest towards and feedback of the CEO blog from the employees of the
company since its introduction had been very positive. The informal and casual style
of blog postings had not only made it easier for the CEO to express his feelings, but
15
D15.2 – Case Study on social software in distributed working environments
had also made it more interesting reading for the employees than formal
communication. It encouraged employees to interact and gave a unique possibility for
them to comment on these postings of the CEO openly. For instance, after the first
three postings approximately 200 feedback messages had been received. The
employees had also understood the informal nature of the CEO blog, and hence had
been more “forgiving” towards the content of it.
In the second supporting case blogging was seen to have many purposes in the
organization. For one, it brought the new CEO as a person closer to the employees
and made his way of thinking visible. Furthermore, it gave an opportunity to the CEO
to interact with employees in a new way; people could ask questions (i.e. comment
on the blog) on which the CEO could answer later in his blog. In other words, blog
was seen as a “soft” media to communicate strategy and give appraisal to people. As
it supported the existing communication systems in the company, it was also
regarded as an addition to the leadership communication “toolbox”. Furthermore, the
blog was seen as a leadership and management tool because it was believed to
increase the level of trust in the management and enhance the development of
organizational culture. The feedback on the CEO blog since its publication had
mostly been very positive and spontaneous, although a few employees had
expressed opposite viewpoints. Overall, blogging had been experienced as a good,
quick, appropriately “soft”, interactive, and modern communication tool.
In both supporting cases, the CEO blog had been a success. However, in both cases
no plans for a more widespread adoption of social software had been made.
5.1.4.3 Comparative analysis of the primary and supportive cases
In the comparative analysis the above described cases were analysed for within-case
similarities and inter-case differences in terms of five dimensions, which were
derived from research questions and existing literature:
- business need/problem/challenge,
-
social software (i.e. what it was),
-
potential areas of applications in organization,
-
implications of social software usage, and
-
issues to consider in introduction and adoption.
The findings of the comparative analysis are summarized in the following table:
Table 1 - Results of the comparative analysis
Primary case
Gain
experience
of
social
Business
software
need/problem/
Increase interaction between the
challenge
project
manager
and
the
management team
Social software Blogs
(one-to-many),
wikis
(many-to-many),
forums,
aggregator
Potential areas Knowledge
management
Supporting cases
Make the new CEO familiar
Support existing communication
strategy
Blog (one-to-many), feedback
channel via comment tool
Interactive
leadership
and
16
D15.2 – Case Study on social software in distributed working environments
of application in (documentation)
organization
Change management tool
Informal learning, just-in-time
learning
Implications of Improved information sharing
social software Increased openness
Lowered change resistance
usage
management communication tool
Interactive feedback tool in
communication strategy
Informal learning, just-in-time
learning
Increased
openness
and
information sharing
Flattened hierarchy
Timing with other events (new
Issues
to Immediate business need
consider
in Integration with existing IT CEO)
introduction
systems and work routines
“Acclimatization”/
and adoption
Slow
“ramp-up”
and testing phase
Active use in the beginning
“backstaging”
(e.g. supportive discourse)
Critical user “mass”
In the primary case the business need/problem/challenge was in nature additional to
the existing business processes and work routines and more explorative for the
possible future needs and challenges. The business need/problem/challenge that
initiated the use of social software in the supportive cases was similar: the change of
the company CEO in both organizations gave a good opportunity to introduce
changes in the communication strategy.
In the primary case the locus of attention was on multiple users of both one-to-many
(e.g. blogs) and many-to-many (e.g. wiki) forms of communication. In supporting
cases the social software application used was a blog. In other words, the focus was
on one person using a one-to-many form of communication, with the opportunity for
readers to comment openly the postings.
When it comes to the potential areas of applications for social software and its
implications in organizations, in the primary case the most potential areas of
application relate to “knowledge management” (e.g. documentation) and change
management. Thus, implications of the use of social software include improved
information sharing, increased openness, and lowered resistance to change. In
supportive cases, social software could be used to increase interaction between the
management and the rest of the organization in change situation. Hence, it might
help to flatten hierarchy and increase openness and information sharing in the
organization. On the other hand, the use of social software may provide a forum for
informal and just-in-time learning for both the management and employees.
5.1.5 Conclusions
Based on the case study, key issues to consider when introducing and adopting
social software in an established organization are numerous. In terms of introducing
and adopting social software in organizations, the primary case presented four issues
to consider:
1. There must be an immediate business need which is tackled with social
software.
2. The use of social software has to be integrated with the existing IT systems
and work routines.
17
D15.2 – Case Study on social software in distributed working environments
3. As novel concepts and applications are introduced, a slow “ramp-up” and
“backstaging” are needed in order to make the new discourse (i.e. terms and
concepts) familiar.
4. The number of users has to be high enough in order for the low proportion of
active users (i.e. “content creators”) to succeed in sustaining a “critical mass”.
On top of this, supporting cases presented three additional issues to be considered:
1. An “acclimatisation”/testing phase before introducing and adopting social
software on a wider scale can prove to be useful if there is uncertainty about
its appropriateness for the situation.
2. Taking advantage of organizational changes that coincide with the introduction
and adoption of social software, and linking them together, appears to give
good leverage.
3. Likelihood of a successful introduction and adoption might increase if special
attention is given to active use of the social software in its early days.
Most important of above-mentioned issues appears to be an immediate business
need to which social software is applied.
From the theoretical perspective, the number of issues to consider in the introduction
and adoption of social software appears overwhelming. Although there are numerous
pitfalls and challenges to be aware of and many choices and success factors to
consider on a detail level, the number of fundamental principles to bear in mind in
terms of a successful introduction and adoption is still relatively small:
1. Social software has to be suitable for the task it is used for.
2. Social software has to fit and support the social context.
3. Immediate business problems through which teams can familiarize
themselves with the concept of social software (i.e. change in attitude) and
learn how to use the applications (i.e. change in behaviour) are crucial.
4. The change agent has to support the “public performance” of change but also
focus on “backstage activity”.
In addition to above-mentioned issues, a key issue to consider in terms of
introduction and adoption of social software relates to organization culture. After all, if
the behaviour induced by the use of social software (e.g. openness and information
sharing) is in conflict with the organization culture, it might turn out to be a potential
hindrance to a successful introduction and adoption of social software. In a similar
way, the bottom-up emergent dynamics of social software necessitates that trust and
control are delegated to the users. However, it must also be ensured that access to
confidential and financial information is restricted in order to avoid concerns about
misuse, abuse, and reliability.
Ultimately, the key issue to consider when introducing and adopting social software
in an established organization is the proposal and introduction of a supporting
discourse through which transformation to new behaviours and routines is reinforced.
There are two reasons for this: Firstly, discourse not only creates, sustains, and
transforms the basic assumptions about organizing but also creates new areas of
18
D15.2 – Case Study on social software in distributed working environments
application for social software in the organizations. Secondly, discourse upholds
organizational metaphors which facilitate the creation and interpretation of social
reality, and thus established behaviour in organizations.
5.2 Case Study 2: Siemens
5.2.1 Introduction to the case
The following case study describes the experiences of Siemens, one of the world's
largest electrical engineering and electronics companies with headquarters in Berlin
and Munich, Germany.
Since the end of June 2006 all employees in Siemens worldwide who have access to
the company’s Intranet have been offered the possibility to write, read and comment
weblogs. This service is owned and sponsored by the central communication
department while the centre for Knowledge Management of Siemens Corporate
Technology is supporting this implementation by providing technical support and
know-how regarding knowledge management aspects.
One of the reasons for introducing weblogs as a new means of corporate knowledge
management has been the unsatisfactory experience with previous (often top-down)
knowledge management solutions. On the one hand, the expected process
integration into daily working practices has only taken place to a smaller degree than
expected. On the other hand, implementations did not manage to transform the more
abstract knowledge processes as postulated e.g. in Probst’s model (Probst et al.
1998) into individual working routines and efficiency increasing tools.
By offering weblogs as a bottow-up approach to corporate knowledge management
employees shall be able to integrate the use of this simple tool into their daily
practices according to their needs and experiences. This implementation represents
an innovative step within this big enterprise as it offers more freedom and requires
specific competencies from many players on an individual and organisational level.
The company has committed to an open publication and networking space for its
employees, which is guided by an internal blogging policy that refers mainly to ethical
issues.
It should be mentioned that Siemens does not consider weblogs as the only solution
for effective knowledge management. It is however an innovative approach
complementing previous top-down methods and as we will see the first
implementation phase already shows some rather promising results.
Generally speaking, weblog posts represent traces of the individual digitalised
working process of an individual and as such an expression of specific topics the
individual is concerned with. Anchoring these expressions in an individual episodic
memory (Tulving 2002) as in the case of a weblog represents an enormous reduction
of complexity for the individual compared to the distributed forms of content
exchange as in the case of other systems such as discussion forums or community
portals. For the author the retrieving is much easier while for the reader the
understanding is supported by the relationship which each post has with its author
19
D15.2 – Case Study on social software in distributed working environments
and is thus better contextualised. The technical support for establishing connections
and creating networks is an additional reason for choosing weblogs over any other
community tool.
The technical solution offers publishing and commenting functions for each
employee. The main access point is via the Siemens Intranet Portal and
authentication is possible via a Single-Sign-On service. Easy access and simple use
were the main priorities when defining the specifications. The Siemens BlogHomepage already includes a tag cloud that gives an overview of the most prominent
topics covered in the Siemens Blogosphere.
Pic.1 Siemens Blog-Homepage as entry point
After the introduction in June 2006 a continuous adaptation of the application based
on user feedback, which is communicated in an own group weblog, has been
foreseen.
5.2.2 Objectives
The objective of this case study is to observe the initial phase of the introduction of
weblogs into a world-wide operating enterprise for knowledge management
purposes. Our observations are concentrating on the specific use of weblogs, on
network building, on knowledge sharing, knowledge exchange and knowledge
building. Since the initial motivation behind the implementation of the Siemens
Blogosphere has been triggered by an innovative approach towards knowledge
management, we are also going to look for any success indicators in this respect.
Both the individual and the group level shall be observed. Especially for the individual
level an objective of this case study is to identify any individual use of the weblog for
personal learning process documentation.
5.2.3 Methodology
The time period covered in this study is from the June 2006 (kick-off) to end of
November 2006. It covers the very first months after the launching of the Siemens
Blogosphere.
Data for this case study has been mainly gathered from anonymous user statistics
and data extracted from the weblogs. Most of the data is concentrating on the
20
D15.2 – Case Study on social software in distributed working environments
observation of the weblog use. Following an exploratory approach without any
specific pre-defined theory the case study includes qualitative and quantitative data.
In addition to the data extracted from the system a few exploratory interviews with
specific user were performed as well. The person performing these interviews is a
staff member of Siemens and also the main responsible for the whole project.
5.2.4 Analysis
5.2.4.1 Statistical overview
First of all, we would like to give a short overview of some user statistics. Until the
end of November 2006 the system counted 8879 registered users. The reach of
people who have at least visited the platform is estimated to cover approximately
25.000 individuals. During that period (June 2006 – November 2006) 309 users
created their personal weblogs and 38 group weblogs with more than one author
were registered. In addition, an open weblog dealing with the system itself was
created in order to collect suggestions for improvements and questions regarding the
use of the system.
By the end of November 2006 the platform had registered 1464 posts and 2556
comments from 482 different authors. The following figure shows the distribution of
the posts and comments over the time period.
90
80
Comments
Posts
70
60
50
40
30
20
10
20
06
-0
630
20
06
-0
712
20
06
-0
722
20
06
-0
803
20
06
-0
81
20
6
06
-0
826
20
06
-0
907
20
06
-0
918
20
06
-0
930
20
06
-1
011
20
06
-1
023
20
06
-1
102
20
06
-1
114
20
06
-1
125
0
Fig.2 Number of posts/comments per day
As expected, after the first months a small decline in comments can be noticed.
However, the number of posts remains rather stable and even increases slightly
towards the end of the period.
When comparing the distribution of posts and comments amongst the active users it
becomes clear that it is a comparably small group of people who does most of the
writing. An exponential increase as it has been observed for many Internet platforms
(Sifry 2006) have not yet been noticed. We notice here some potential for
improvement in attracting more active users.
5.2.4.2 Content
The average length of a post in the Siemens Blogosphere is 970 characters and a
comment includes on an average 370 characters. Most of the entries (approx. 70%)
21
D15.2 – Case Study on social software in distributed working environments
range between 100 and 1.100 characters following a typical power-law distribution
(Cattuto 2006). This corresponds with the average length of weblogs published
openly on the Internet as a study by Brown (2007), who looked at the top 100
Technorati Weblogs, shows.
Rearing the type of posts we differentiate between posts that include a hyperlink and
thus relates to additional sources of information and those posts that do not include
any URL. The first type of posts we would like to called “Resource Posts” because
they relate to additional resources and put the post into a broader context. It
establishes a network of related resources of information. If we look at the length of
the posts without an URL (0) and the resource posts (1) in the graph below it is
interesting to notice that there is a peak at about 500 words for the resource posts.
0
1
100 300 500 700 900 1100 1300 1500 1700 1900
200 400 600 800 1000 1200 1400 1600 1800
100 300 500 700 900 1100 1300 1500 1700 1900
200 400 600 800 1000 1200 1400 1600 1800
length
length
60
Anzahl
40
20
0
Fig.3: post’s length without hyperlink (left) and with hyperlink (right)
As previously stated the authors were given complete freedom in terms of content of
their weblogs. The same applies for tagging. There was no terminological control
(Hammond, et al. 2005) whatsoever regarding the tags that the authors were using
for categorising their weblog entries. Although this fact was not explicitly featured
during the launch of the system the weblog authors made quite extensive use of the
tagging feature. Overall 1.150 different tags were used 3.224 times with an average
of 2,3 tags per weblog post. The frequency scale of the different tags follows the
typical power-law distribution or also called “long-tail” (Cattuto 2006) and has also
been notice for tagging in other social software applications (Barnett 2006) as we will
see in one of the following case studies on the Nextspace (i.e. Section 5.3).
The most frequently used tags are related to the tool itself (e.g. blogs, web2.0,
blogging, wiki, etc.). It shows that during the first phase many users were reflecting
on the new system itself. Other frequently used tags include “innovation”,
“communication” and “technology”. Although a deeper content analysis of the
weblogs is still missing, it shows that at the beginning it was clearly technical
innovations that attracted most authorship.
5.2.4.3 Networking
One of the most important features for networking in the Blogosphere are comments.
Whenever someone writes a comment attention is drawn to the content of the text
22
D15.2 – Case Study on social software in distributed working environments
that is being commented and at the same time attention is also drawn to the author of
the original text. A relationship between the author and the comment author as well
as between the original post and comment is established. The contextualisation of
this relationship is usually supported by hyperlinks.
In the Siemens Blogosphere almost two thirds of the texts are comments to other
people’s posts. During our period of observation 2.556 comments were made,
ranging from 6 characters to 6.370 characters in length. The average text length was
370 characters. If we look at the frequency of comments per post it shows that 655
entries did not receive any comment, 233 entries received 1 comment, 195 entries
received 2 comments, etc.
600
400
200
0
0
1
2
3 4
5
6
7
8
9 10 11 12 13
Fig.4: frequency distribution of comments
Overall we noticed a tendency that the number of comments depends on the topic
and not so much on the length of a post.
As stated above commenting is especially important for social networking and needs
to be followed on further if we want to find out more about learning in social networks.
First exploratory analysis of commenting practice reveal that there is strong
commenting of bloggers from US sites in weblogs of German based bloggers. The
top position is still held by comments of German based bloggers in German based
Weblogs. There is lot of anecdotic evidence that within-country commenting also
leads to new contacts. But since the information if commentator and blogger did
know each other before engaging in blogging cannot be derived from system data
final evidence would require complementing data from a survey, interviews etc.
Nevertheless we have sufficient hints for national and international networking based
on the use of the blogging platform.
5.2.4.4 Blogging-Patterns
According to Technorati data, there are over 175,000 new blogs created every day
on the World Wide Web 1 (including so called “spamblogs). About 55% of the weblogs
that Technorati counts are considered active, which means that they have been
updated at least once in the last 3 months. Of course this definition of an active blogs
is incomplete without taking into account the regularity of posting over a longer period
of time.
1
http://technorati.com/about/ latest visited on 29.06.2007
23
D15.2 – Case Study on social software in distributed working environments
Only if a weblog has been active over a certain period of time experts speak about
the survivability of weblog (Schmidt & Mayer 2006). We would like to stress here
however that the period of analysis within this case study has been rather short with
8 months and the analysis should be repeated after a longer time period.
During the observed time span out of the 309 personal weblogs that were created
112 did not show any posts. This means that approx. one third of the employees who
created a weblog for themselves did not activity make use of it. This percentage is
similar to figures from the Internet (Krüger 2003).
Posting Patterns
10
9
8
7
Posts
6
5
4
3
2
1
0
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Week
Fig.5: posting patterns in Siemens weblogs over time after blog creation
Amongst the thirty most productive weblogs during the period of observation we
counted 10 group weblogs (one third) with distributed authorship. If we look at the
overall percentage of group weblogs amongst the total number of weblogs, which is
11%, we notice that group weblogs show a more stable blogging practice. This might
be due to the fact that multi-authorship triggers more commitment by the individual
author.
The posting patterns visualized in the figure above support these numbers. Only
some bloggers transfer their personal experiment into a stable practice. The
proportion of these employees of course varies depending on the definition of an
“active” blog. Referring to the 3 months mentioned at Technorati we find update
frequencies between 0.5 and 5 posts per week in a typical power-law distribution for
blogs that showed any activity for longer than 12 weeks.
In 2003 Perseus found that blogs are updated much less often than generally
thought. Only 2.6% of the 4.12 millions hosted blogs were updated at least once a
week referring to definition of “active blogs” based on 2 months. Under comparable
conditions this rate is 6% on the Siemens Blogosphere.
A percentage of active blogs between 15% and 20% - depending on the definition of
“acitive” - seems to be state of the art for a voluntary offer of a new medium. This rate
was confirmed by our statistics and numbers from IBM 2 , which are much higher in
2
http://www.cogneon.de/weblogs/bitkom_kem_08_02_2007 retrieved 2007-02-13
24
D15.2 – Case Study on social software in distributed working environments
absolute figures because their service exists since 2003 according to Jackson
(2007).
5.2.4.5 Knowledge-Patterns
From a knowledge perspective we found at least five types of documented
knowledge that can be linked to psychological theories of cognition. Many posts can
be characterized as (simple) declarative knowledge pointing to interrelations between
facts explicated in short checklists etc. Explanatory knowledge (Begründungswissen)
articulated in arguments supporting opinions. Knowledge about resources that links
to other chunks of information, mostly by using URLs. Procedural knowledge as
found in simple tips and tricks for work related tool like MS outlook etc. Finally
narrative knowledge as a rather rare category that can be related to posts that tell a
personal (not privat!) story.
To date some blogs relate to workplace learning either by their authors being part of
the internal training department or – which is even more important from a knowledge
work perspective – by individual employees who deliberately choose to elicit their
personal learning experiences and lessons learned. The example of an employee
publishing his experiences with a certain CAD-Software from installation on is a very
promising and lasting one for learning focused use scenarios.
5.2.4.6 Concluding remarks
The implementation of the Siemens Blogosphere represents an innovative approach
towards knowledge management in this word-wide operating enterprise. Offering
open publishing tools without any editorial control is a big step towards a dialogueoriented communication culture.
The successful use of the platform gives an indication that weblogs are accepted as
an additional working tool that offers employees the possibility of connecting globally
around specific topics and of establishing social networks. It should also be stressed
that the employees are using this service on a completely voluntary basis. With the
growing use of the weblogs a successful indication is given that the blogging has
been successfully integrated into individual working practices. Some posts and
comments also explicitly mention this integration.
Regarding the motivation for the use of the weblog tool we have spotted different,
mainly personally motivated, reasons. For some user the weblog support a reflection
process and is used as a sort of “reflection” or “learning” diary while for some others
the dialogue and the fast feedback that one receives via the comments is most
important. Here again social networking and transparency plays an important role.
Some units within the organisation also make use of RSS feeds from specific topicrelated weblogs and add thus more dynamic content to their websites. Some only
subscribe to specific tags. Overall, the possibilities of linking, RSS feeds and
comments create a new network of information resources that also relates to specific
persons and contextualises the information in a new way. The growing number of
specific topic related posts indicates also that employees better integrate weblogs
into daily working practices than it has been the case with classical top-down
knowledge management tools.
25
D15.2 – Case Study on social software in distributed working environments
Although the Siemens Blogosphere is still very young, the first phase has given some
positive results that shall lead towards a faster, more flexible, transparent and selforganised form of knowledge management.
5.3 Case Study 3: Nextspace in iCamp
5.3.1 Introduction to the case
iCamp (innovative, inclusive, interactive & intercultural learning campus) is a specific
targeted research project under the sixth framework programme of the European
Union. The project develops and evaluates targeted activity sets that make
systematic use of interoperable, networked tools and services in order to support
competence advancement in three areas of challenge: 1) self-directing and selforganising learning projects, 2) collaboration and 3) social networking. For more
information on the specific project objectives, please see: www.icamp.eu
The research team involved in this project consists of 10 partner organisations, which
are geographically distributed across 9 European countries. Each of these
organisations involves around 2-5 people in the project work. Research and
development work are thus performed in a distributed way and have to be supported
by ICT.
For the technologically mediated collaboration and communication of the team the
project management decided to use a platform called Nextspace as the main vehicle.
Nextspace is an innovative software for collaborative knowledge management that
integrates various social software elements. The main features of the Nextspace are:
•
•
•
•
•
•
Chronological content entries (similar to weblog)
Multi-author posting and versioning (similar to wiki)
Tagging
Calendar
Member overview
RSS feeds
Pic.2 iCamp Nextspace Screenshot
26
D15.2 – Case Study on social software in distributed working environments
This case study refers to the use of this software within the iCamp research
community with a special focus on informal learning processes, knowledge exchange
and social networking.
5.3.2 Objectives
Apart from the general research objectives of investigating how social software can
be used in distributed working environments this case study focuses on the following
research questions:
General acceptance of the tool: What is the general acceptance of the software
itself and of the hierarchy-free and unstructured collaboration process that the
Nextspace fosters?
Collaboration process: How is the collaboration process within the project
community perceived? Has there been any perceived change in how the
collaboration process is performed? Have there any perceived gains or
disadvantages compared to previous practices?
Communication process: How is the communication process in the project
community perceived? Has there been any perceived change in how the
communication process is performed; Have there been any perceived gains or
disadvantages compared to previous practices?
Learning process: How is the individual learning process perceived? Has the
individual learning process been influenced positively or negatively by the use of the
Nextspace? Have there been any perceived changes in the learning process? Has
knowledge sharing and knowledge gain been facilitated by the use of Nextspace?
Social behaviour: Has there been any perceived influence by the use of the
Nextspace on the social behaviour of oneself as well as of the other members of the
team? Has there been any change in social behaviour?
5.3.3 Methodology
The time period for the data collection starts with the beginning of the project in Oct.
2005 until June 2006. This covers a time period of approx. 11½ year of the project,
corresponding to the first half of the project duration.
Regarding the number of users on the Nextspace there has been some variation due
to the fluctuation of staff members. From the very beginning each partner
organisation had approximately 2-4 persons registered on the Nextspace. On June
1st 2007 we are now counting 55 members + 1 administrator out of which 13
members have not been active for the last 100 days and are thus either members
who left the team or what we would call “silent followers”, who do not actively
participate in the project work, but still have access to the Nextspace to follow the
project development.
27
D15.2 – Case Study on social software in distributed working environments
Out of this pool of iCamp team members five persons were selected for interviews.
The selection of these persons is based on different interaction patterns that have
been identified via quantitative data analysis from the usage data on the Nextspace.
Via Social Network Analysis (SNA) we identified different interaction patterns and
identified key persons who represent a certain interaction pattern (e.g. strong direct
communication).
After the identification of these exemplary five representatives of Nextspace users
seven interviews were conducted at two different points in time for qualitative data
gathering. Three rather exploratory interviews were performed at the beginning of the
project when the Nextspace had been in use only for a short time period and the
second round of interviews were conducted after more than a year of using the
Nextspace. Out of the five members two persons were interviewed twice. These had
been the two most sceptical users and also as the less active ones regarding their
interaction diagram. The structured interviews at the second point in time included
two of the most active or “core users” of the Nextspace.
The following analysis is thus based on qualitative and quantitative methods used for
data gathering, namely SNA and interviews.
5.3.4 Analysis
In this analysis we will discuss the data gathered in the interviews in combination with
the results from the SNA and relate them to the above-defined research questioned.
5.3.4.1 Interaction patterns
As mentioned above, SNA methods have been applied to identify certain interaction
patterns amongst the members of the iCamp team. When looking at the egocentric
networks of the individual members registered on Nextspace different interaction
patterns can be perceived. On the one hand we can identify some very active
members, for which we can detect frequent direct interaction with other members in
both directions while on the other hand there are members with very little or even no
interaction with other members. Amongst these two extremes there is a continuum of
more or less active members. In the following we show ego-centred network
(Wassermann 1994) of year 1 of the persons that have been interviewed for this case
study. The personal network data indicating 1st order interaction (direct contacts) for
each of them is different and the graphs show clearly that F. and S. are interacting
much more in the Nextspace than e.g.E. ,T. or G.
28
D15.2 – Case Study on social software in distributed working environments
ego-centred network F.
ego-centred network E.
ego-centred network S.
ego-centred network T.
ego-centred network G.
29
D15.2 – Case Study on social software in distributed working environments
Fig.6
If we have a look at the second order contacts of these persons less differences can
be perceived in terms of network density and the network becomes sparser for all of
the actors (see figure 7 below). This can be interpreted by the fact that overall we are
dealing with a closed community of team members who generally interact heavily
within the group in the beginning. With the time life of the project, the project partners
direct their communication not to the entire group any longer as in the beginning of
the project (only those messages which are relevant for all), thus dislocate their
communication tool back to other channels (ie. emails) for messages only relevant
for some of the involved partners.
The direct contacts are represented by the red dots, whereas the blue dots are
indirect contacts.
2nd order F.
2nd order E.
2nd order S.
2nd order T.
30
D15.2 – Case Study on social software in distributed working environments
2nd order G.
Fig. 7
5.3.4.2 General acceptance of the tool
The general tenor regarding the acceptance of the Nextspace can be described as
positive. Specific features could be improved and one bug was commented, but
overall the use of the tool within the project has gained high acceptance.
Overall, there exists a general consensus amongst the interviewees regarding the
usefulness of the Nextspace of project administration and coordination purposes.
Especially the frequent coordination tasks can be handled conveniently via this tool.
To-do-lists, meeting schedules, attendees lists, announcements, etc. are perceived
as being more efficiently and effectively handled via Nextspace compared to previous
workflows e.g. via e-mail.
One person, who also has a coordinating role in the consortium pointed out that the
tool is not very convenient for controlling and monitoring purposes compared to more
traditional project management tools. Another interviewee is missing some
structuring elements such as folders. He pointed out some difficulties with adapting to
the bottom-up folksonomy-based tag structures as the main elements for organising
the content. This approach does not really fit with his personal working habits.
An important aspect addressed by all interviewees is the way in which the use of the
Nextspace integrates with their personal working style. Here we can identify different
attitudes towards the tools. One user e.g. has been working with social software
applications for many years now and has constructed his personal workflow for using
the Nextspace and has integrated it with his personal tools landscape. Three out of
the five interviewees were exploring the potential of the features for adapting to the
personal workflows. The other users seem more reluctant to explore the different
possibilities that Nextspace offers for constructing different workflows and thus better
integrating with the personal working habits.
Although the software provides various options in terms of use and integration with
different workflows and other services (e.g. via RSS) it is mainly the users attitudes
and some basic knowledge on the technical features of the tool that either supports
31
D15.2 – Case Study on social software in distributed working environments
this integration or hinders it. In the iCamp community both cases have been
identified.
5.3.4.3 Communication/Collaboration
As communication is normally part of any collaboration process and cannot be easily
detached from it, we are treating the two aspects together. The Nextspace offers
features for both activities. Weblog-like one-to-many communication structures as
well as joint editing features similar to a wiki are core features of the tool.
The most important change that all interviewees noticed is the increased
transparency compared to previous communication and collaboration in distributed
working environments.
Again, personal attitude and dispositions for open
collaboration and communication play an important role here. Whereas some people
favour complete openness and access for all to any discussion, comment and
content and others are asking for restriction possibilities in order to make some
comments only available to certain persons, but not to the whole group. Thus it is the
personal preferences, attitudes, dispositions and working styles that determine how
much transparency can be achieved.
Overall, an agreement seems to exist amongst the users that the system supports
the emerging knowledge-pool or archive of the project work. Interestingly more
communication regarding content related to special workpackages (WP) was taking
place during the first project year compared to the second. The overall contribution of
content has however not noticed any such decrease.
The following figures 8e.g. show the interaction related to a workpackage (WP) in
year 1 and in year 2.
Year 1
Year 2
Fig. 8
32
D15.2 – Case Study on social software in distributed working environments
A similar pattern can be identified for the communication related to any of the other
WPs. One of the explanations for this is that in the second year the specific WP
members started to use additional tools in smaller groups for specific WP
collaboration and communication (e.g. Skype, etc.).
Generally the Nextspace has not been used extensively for collaboration on a bigger
piece of work, such as a deliverable, or joint editing. According to interviewees the
tool has proven to be more appropriate for communication than for strict collaboration
in the sense of joint artefact production.
5.3.4.4 Individual learning process
The individual perception on the support of the Nextspace for informal learning
purposes cannot be easily derived, as many people do not reflect consciously on
their learning process, especially not on informal learning processes. According to
one user, who is very familiar with the tool, the Nextspace can be a good tool for
supporting this reflection process. The system stores every comment, every
interaction and makes the communication of the group transparent and provides a
good archive for the project and for the involvement of each individual.
Another person explicitly points out that the Nextspace supports his learning process.
He perceives the tool as an important information and knowledge pool that helps him
generate new insights and ideas that are to a certain extent again fed into the system
and might trigger something different for another user.
5.3.4.5 Social behaviour
In terms of social behaviour various interview partners stressed the social
engineering effect that the Nextspace is supporting. Various features add to this
effect and facilitate the group building.
On the one hand the hierarchy free structure prompts the posting of not strictly
project-related content, but also semi-private communication such as the birth of
child from one of the team members or the coordination of some extra (e.g. sports)
activities of the project team before a project meeting. On the other hand, some
specific features such as that online-presence status and the personal image of each
team member supports the group identification and according to one interviewee also
exerts some social pressure.
5.3.4.6 Concluding remarks
Just as any other piece of software Nextspace is only a tool that can either support or
hinder people in their collaboration, communication and social interaction and thus
finally in their informal learning processes. It depends on the personal dispositions
and attitude of the individual user towards the tool and the peers in how far the tool
can support these processes. According to this case study the Nextspace has some
specific affordances that may well support group communication, collaboration,
knowledge sharing and social interaction for distributed working teams.
33
D15.2 – Case Study on social software in distributed working environments
5.4 Case Study 4: Bibsonomy
5.4.1 Introduction to the case
The social bookmarking system BibSonomy is part of the wave of social software
applications developed to enable the collaboration between distributed working
environments. The system focuses on integrating features of bookmarking systems
as well as team-oriented publication management. The social bookmarking capability
allows for storing and sharing bookmarks, and providing entries with keywords (tags)
to structure information and facilitate later retrieval. Besides the publishing of
scientific paper metadata and descriptive keywords, the publication management
component incorporates features such as the creation of bibliographies for
publications or web sites, export facilities in 17 different output formats or automatic
metadata extraction from websites [2, 3].
As the descriptive terms used to describe bookmarks or publications can be freely
chosen, the assignment of tags from different users creates a spontaneous,
uncontrolled vocabulary: a folksonomy. In BibSonomy, the folksonomy evolves from
the participation of research groups, learning communities and individual users,
organizing contents according to their information needs.
After about a year of operation, BibSonomy comprehends 1000 active users,
283.092 bookmarks, and 609.618 different tags. The system is designed to meet the
specific needs and challenges of its target group: scholarly people world wide. In
order to find out about its applicability and to identify possible challenges for further
developments the BibSonomy team of the Knowledge and Data Engineering Group
at the University of Kassel/Research Center L3S conducted a case study considering
researchers and research communities as their unit of analysis.
The study started in the beginning of 2006 and terminated end of April 2006. In order
to improve the reliability of this study, the case builds on multiple sources of evidence
as proposed in [9].
1. Two research communities were selected, which committed to use BibSonomy
to manage their project’s references. Both groups are familiar with social
software: the European Integrated Project “Nepomuk - the social semantic
desktop” and the PROLEARN community. The participants of both projects were
asked to anonymously fill out a questionnaire; additionally, usage statistics were
drawn from the database to find out about the system's acceptance.
2. In the end of March 2007, BibSonomy was part of the CKC challenge in which
social software applications were evaluated by researchers worldwide. The
feedback given was analysed according to the objectives of this case study.
3. Finally, the University Library of Amsterdam conducted a user trial comparing
BibSonomy with Cite-U-Like considering publication management. The
documentation and feedback of the participants contribute to this case study.
The following section is organized as follows. The first sub-section defines the case
study's objectives. The second sub-section outlines the applied methodology, and
34
D15.2 – Case Study on social software in distributed working environments
the third sub-section analyses the results. Finally, a conclusion summarises the
findings and proposes challenges for future research and development activities.
5.4.2 Objectives
The case study underlying this section focuses on evaluating BibSonomy as a social
bookmarking solution for the scientific community. Specific needs and behaviours
regarding information retrieval and publication management of this group need to be
analysed, and BibSonomy's strengths and weaknesses in responding to these
particular requirements are assessed as well.
Three specific research questions can be derived from this general objective:
1. Identify individual benefits and expectations for using social bookmarking tools
in a research environment.
2. Find out about a social bookmarking system's contribution to facilitating
collaboration in scholarly groups regarding literature and reference
management.
3. Explore BibSonomy's usability and applicability as a publication management
tool for scientific authors and readers.
The next sub-section describes the approach chosen to study the three research
questions.
5.4.3 Methodology
The nature of the research questions called for a qualitative, exploratory study
investigating experiences with collaborative publication management. Descriptive
components are included to distinguish tendencies in attitudes. The study uses
documentation, questionnaires and interviews as information sources.
Four representative groups from within the research community were selected to
participate in the study:
•
•
Nepomuk project: The European project aims to develop a comprehensive
solution for “extending the personal desktop into a collaboration environment
which supports both the personal information management and the sharing
and exchange across social and organizational relations” [4]. The project,
started in 2006, uses BibSonomy for organizing their publications. Special
features such as the group functionality to share publication entries only within
the project, the possibility to integrate BibSonomy into a project's web site [5]
and the addition of group specific metadata was developed in conjunction with
this project.
ProLearn project: The 'Network of Excellence' under whose umbrella this
case study was carried out, brings together people working on technology
enhanced professional learning. The project uses BibSonomy for reporting its
publications, but has not integrated the system into its active workflow (e.g.
creating a tag cloud, connecting its web site and BibSonomy). The main
reason why we asked for the participation of ProLearn members, was a
35
D15.2 – Case Study on social software in distributed working environments
workshop on social software in March 2007 [7], in which people were
introduced to the system.
Since both projects involve people across Europe, personal interviews and life
observations were difficult to carry out. Consequently, we decided to distribute an
online questionnaire which consisted of three parts:
•
•
•
Demographic questions
Questions about experiences with information retrieval and reference
management
Questions about system usage, perceived values and usability aspects.
A major risk to be controlled was the bias that people knew the research questions
before responding to the questions. Thus, people were informed of the questionnaire
via a short mail, explaining the basic settings of the case study without giving too
many hints. The approach had the disadvantage that participants were not personally
addressed.
A second risk was the fact, that the authors carrying out the case study were part of
BibSonomy's development team. The development of the questionnaire therefore
might have been biased towards a positive acknowledgement of the system. In order
to receive neutral feedback, we decided to integrate the documentation of two
external challenges in which BibSonomy was involved. The BibSonomy team
contributed to these challenges by presenting the system, and providing help and
guidance during the trials. The development of the survey, the formulations of
questions as well as data collection were conducted externally. The two challenges
are described as follows:
•
•
CKC Challenge: The Collaborative Knowledge Construction Challenge [8]
organized as a workshop in the scope of the WWW 2007, aimed to assess
social knowledge creation tools combining the semantic web and web 2.0
branches. The challenge took place from April 16th to April 30th inviting
participants to try and give comments on a variety of tools. Besides social
bookmarking tools such as BibSonomy, collaborative ontology editing,
visualization and development tools were taken into consideration. A specific
help page [1] for CKC challengers was created which people could access.
Participants were supposed to try the basic features as well as more
advanced functionalities enabling semantic knowledge presentation. For
instance, tags in BibSonomy can be marked as part of a hierarchy.
Participants filled out a web questionnaire consisting of 10 open-ended
questions.
University of Amsterdam Challenge: The Library of the University of
Amsterdam organized a user trial of academic social referencing software with
members of the research group Systems- and Network Engineering of the
University of Amsterdam [6]. Participants compared BibSonomy with Cite-ULike in order to evaluate the potential value of social bookmarking applications
for library services. An external company, Pleiade Management and
Consultancy reported and analysed the feedback given by participants. The
organizers led users to carry out several tasks with both systems. Participants
36
D15.2 – Case Study on social software in distributed working environments
then filled out a questionnaire and log book to report on those tasks. Finally,
an interview rounded up the assessment of social bookmarking systems.
5.4.4 Analysis
Results are presented in three sub-chapters according to the different cases. Each of
the analyses states sources, charts and comments to answer the three research
questions.
5.4.4.1 Results of the research projects
The response rate of the questionnaires was rather low. This may be due to several
reasons: people were not offered incentives and were not addressed personally;
furthermore, not all project members work with BibSonomy. As the structure and
participants of both projects are similar, we decided to merge the results of the
questionnaire in the analysis part. Overall, we had ten respondents, nine male and
one female.
The analysis is grouped into three divisions: Questions considering experience and
habits with search and reference management, questions and findings about the
actual usage of BibSonomy and its value for users, and questions about the system's
usability. The questionnaire can be found in the Appendix.
5.4.4.1.1 Experience and habits
Experience and habits were studied by asking for search behaviour and reference
tools used. Several statements were given which the participants could mark if they
agreed with the statement.
As can be seen in figure 8, conferences and search engines play a major role in
finding literature. Summing up statement 6 and 7 (“Other colleagues outside my
project tell me.” and “Colleagues of my project tell me”) shows a strong influence by
colleagues.
The most common reference management approach (see figure 9) is the local
storage and classification of literature in folders and sub-folders. Local (browser) and
global (social bookmarking systems) are equally interesting for reference
management. Several participants indicated that they use other tools, unfortunately a
specification of the other approach was not given.
Literature Search
1. Other
0
Statements
2. I use domain specific search engines.
4
3. I use social bookmarking tools.
5
4. I get tipps from mailing lists.
5
5. I search in literature lists.
5
6. Other colleagues outside my project tell me.
6
7. Colleagues of my project tell me.
8
8. I use search engines.
9
9. Via conferences.
10
0
1
2
3
4
5
6
7
8
9
10
Number of Users
Figure 8: Literature Search and Reference Management
37
D15.2 – Case Study on social software in distributed working environments
Statements
Reference Mangement
1. I use Excel.
0
2. I use EndNote software.
0
2
3. I use BibTex software.
4. I use social bookmarking systems.
3
5. I bookmark the urls and store them in my browser.
3
4
6. Other
7. I download the documents and store them,
organized in folders and subfolders on my harddisk.
8
0
1
2
3
4
5
6
7
8
9
10
Number of Users
Figure 9: Reference Management Tools
Being asked to which extent they are influenced by others in their literature and
document search, most participants agree or strongly agree with this statement (see
figure 2, statement 1). They are interested in their colleague's literature findings
(statement 4), but not all agree to read the colleague’s comments added to literature
(statement 2). In the contrary to the hypothesis that people prefer to keep their
thoughts about literature in private, people do not mind sharing their literature
classification (statement 3).
1. Advice of others influences
my literature choice.
2. I read comments of colleagues
added to literature.
0
1
4
0
4
1
3. I prefer to share my literature
classification with other people.
4. I am interested in my
colleague's
literature findings.
0
0
Strongly disagree
Disagree
Agree
Strongly agree
5
5
5
0
0
0
5
3
2
7
4
6
8
10
Figure 10: Collaboration in Nepomuk and ProLearn
5.4.4.1.2 Usage and benefits of social bookmarking
Basic usage statistics can be seen in the following table:
Project
Nepomuk
# Group
Members
28
# Group
Posts
6886
# Project
# Group
# Group
Bookmarks Publications Bookmarks
3355
3531
# Project
Publications
# Group
Tags
3631
38
D15.2 – Case Study on social software in distributed working environments
Project
ProLearn
# Group
Members
2
# Group
Posts
395
# Project
# Group
# Group
Bookmarks Publications Bookmarks
4
391
# Project
Publications
4
59
4
391
# Group
Tags
77
Table 1: Usage Statistics
We distinguish two different group statistics: “Group Bookmarks” and “Group
Publications” count all bookmarks and publications (whether or not they are directly
related to the project) contributed by members of the ProLearn and Nepomuk group.
“Project bookmarks” and “Project Publications” are those entries that represent the
project itself (e.g. publications from project members).
Nepomuk is the more active project sharing both bookmarks and publications.
ProLearn focuses on representing its publications, not being an active participant in
collecting bookmarks or using publication reference features. This may be due to the
fact that, within ProLearn, project related publications are reported in Excel sheets
and are then entered into BibSonomy centrally by only few project coordinators,
whereas in Nepomuk, reporting is done directly within BibSonomy by active project
members. The ProLearn decision to organize the reporting centrally was probably
influenced by the fact that the project had already started, and people were used to
reporting in Excel sheets.
The fact that not all of the questionnaire's respondents actually use a social
bookmarking system is also shown in figure 8, where only five respondents include
social bookmarking tools in their literature search and only 3 use such a tool for
reference management.
5.4.4.1.3 Usability of BibSonomy
The usability of BibSonomy was explored in two ways: on the one hand, participants
had to indicate to which extent they agree with different usability statements, on the
other hand, participants were allowed to comment on features they liked/disliked in
BibSonomy.
While people find it easy navigating through menus, the complexity of the system
seems to be a major problem. Half of the participants state that new features are not
easy to find, and that they do not believe learning all that the software can do.
39
D15.2 – Case Study on social software in distributed working environments
BibSonomy Usability
1. Discovery new features is easy.
2. Navigating through the menus and
toolbars is easy to do.
Strongly agree
Agree
Disagree
Strongly disagree
3. I will learn learn how to use all that is
offered in this software.
4. I understand the menus and toolbars.
5. BibSonomy is easy to use.
0%
20%
40%
60%
80%
100%
Figure 11: BibSonomy Usability
Answers given when asked to comment on what people disliked encourage the
assumption that BibSonomy is perceived as being too complex.
What particular aspects of BibSonomy do you dislike?
Too much information on pages; difficulties in finding information as a whole.
The menu bars are too crowded.
The usability of the interface.
Add FOAF support to enable me to connect to my friends without entering them!
Table 2: Suggestions for improving BibSonomy
Positive acknowledge was given regarding the handling and extracting of BibTex
entries for publication management. 80 % agreed that managing reference tasks is
easy. The exporting and importing functionality as well as the extraction of BibTex
from web pages was mentioned.
What particular aspects of BibSonomy do you like?
that people help each other.
the idea, the gui looks ok. The URLs are REST, they look good.
exporting/importing bibtex
extraction of bibtex from web pages (acm…)
5.4.4.2 Results of the CKC Challenge
Eight people world-wide responded to the challenge. Out of the ten given questions
we selected the most representative responses as a data source for our research
questions.
5.4.4.2.1 Experience and habits
To find out about people’s benefits, people could state if they would use the tool for
regular activities.
Do you think you will use the tool in some of your regular activities?
If yes, what for?
40
D15.2 – Case Study on social software in distributed working environments
If no, is there something that tool developers can change to make it more likely that you will
use the tool?
1 yes for bookmarking.
2 I already knew this tool and used it sometime.
so...yes i think it is a great tool for storing paper i want to read but it needs a critical mass of
users yet.
3
Yes, I think as a researcher, I really need an online storage (even if shared to some extent) of
resources (link, pubblications). Moreover, the
ability to export the references in various formats is quite nice, I can link my pubblication page to
bibsonomy and update it dinamically and have
it linked to others (related) collections automatically.
4 1. Yes I shall use together with Web research tools and other social bookmarking tools.
2. I use Firefox and i created a list of tabs with more 5 different social bookmarking tools.
I use them for search. Other search is in blogs.
5 Possibly. It is easy to use and convenient. It will be useful for posting bookmarks online while
away from home.
6 Yes, for storing my bibliography
7 No, I won't use Bibsonomy, but I will continue using Connotea, CiteULike, and Zotero with my
EndNote.
Figure 12: Value of a social bookmarking system
Most participants can think of using a social bookmarking tool in their daily work.
Referring to BibSonomy, one mentions a missing critical mass as a criterion for not
using this tool. One person prefers other existing systems. One suggests the
combination of different bookmarking tools to improve information retrieval. Beside
bookmarking features, most participants positively mention reference management
features such as storing and creating bibliographies and exporting to different
(formatting and reference) languages as a reason to use a social bookmarking tool.
5.4.4.2.2 Collaboration
The group evaluated collaboration with others very differently. One participant entirely
refuses social bookmarking as a collaborative knowledge creation tool in international
environments: “There was no ‘sonomy’ to the ‘bib’. Mob logic and mad conflation of
languages and ideas are worse than ignorance and noise, they’re anti-knowledge.”
Other participants acknowledged the information sharing capabilities when being
asked what they liked about the tool: “the chance to have my own private personal
space together with the opportunity to share the information I want with the
community”. “I shall be glad to collaborate in future as I am collecting a lot of new
information, analyzing it and using in different projects and eLearning.” Two
participants suggested improving the collaboration facility: “Adding users to groups
could be automated to shorten waiting time.” ”A page showing my groups could be
useful”.
5.4.4.2.3 Usability of BibSonomy
Several questions asked for specific information about the system BibSonomy
considering usability aspects. The most representative questions and the
corresponding feedback are given in table 4 and table 5. The frequent mentioning of
relations comes from the fact that many participants of the challenge studied the tag
hierarchy in detail to compare it with the ontology tools of the challenge.
What did you like about this tool?
41
D15.2 – Case Study on social software in distributed working environments
1
1)making relation between tags.
2)making group
3)bibtex
4)importing bookmarks fom del.icio.us
5)exporting bookmarks to xml and RDF
6)exporting bibtex to XML, RDF, HTML, ... .
7)"pick" ing in bibtex
2
the easiness for annotating web resources and publications.
the chance to have my own private personal space together with the opportunity to share the information i
want with the community
3 It allows to categorize both links and publications using social tagging in an homogeneous way. Plus, the way
to provide relations is quite an appealing feature.
4 Very easy saving selected snippet and information source when using postBookmark, easy adding tags
that can be selected and from recommended list. Very usewful feature - easy creation of tags binary relations.
5
The challenge page explained everything in a simple easy-to-understand manner.
The "postBookmark" and "postPublication" buttons that you could bookmark was a very nice touch and made
the whole process of adding bookmarks and posting publications very convenient.
6 Easiness of use, intuitive and simple interface
7 Provided utility like CiteULike, and provided import/export features for use in my EndNote.
Table 4: CKC Challengers: what they liked
What do you think needs to be improved?
1 the mechanism for adding bookmark. like what is in del.icio.us if it has a firefox extension.
2
the automatic annotation of publications from different digital libraries (sometime it does not work)
the difference between a tag and a concept is not clear in the tool
when you insert a relation the name of the concepts have to be without spaces... if they are concept and not
tags i think you should let a concept name composed by more than one word.
3
The presentation layer. The user is immediately prompted with a list of resources, maybe a visual
clustering with some criteria could help in improving the usability. It is not so clear the added features of
having groups and friends (I couldn't use it extensively).
4 Relations editing. It's quite confusing to see all the relations listed with no structure.
5 I have tried about 10 social bookmarking tools and they all lack one feature - saving not only information
source (URL) but a few snippets from the one information source. These snippets will have the same URL
but different tags and different comments. As social bookmarking tools lack this feature I must use other
tools, e.g. Web research tool http://www.macropool.com/en/index.html (German tool) and Net Snippets
(Israel) www.netsnippets.com but in March 2007 they stopped and now they propose to use
http://www.esnips.com/download/ .
Second, it would be useful to visualize tags cloud as a network that will be created using binary relations
between tags.
Third, to implement an advanced search (using Boolean operators).
Fourth, to think about integration social bookmarking, Web research and ontology tools.
Fifth, to think about using semantic tripples instead of simple tags.
6
Adding users to groups could be automated to shorten waiting time.
You are notified in some way if something you posted was edited by another user.
7 Relations - engine for building tags hierarchy
Table 5: CKC Challengers: what they disliked
5.4.4.3 Results of the Amsterdam Librarian Trial
The Library of the University of Amsterdam wanted to explore (dis-)advantages of
social academic referencing tools. Statements and comments regarding the three
research questions are selected from their final report [6]. Overall, the report
indicates that there is a need for organising and sharing literature references and that
42
D15.2 – Case Study on social software in distributed working environments
social bookmarking tools have the potential to fulfil this need. During the trial, the
Amsterdam group switched to Cite-U-Like, another social publication sharing system,
as it was perceived as being more user-friendly. Nevertheless, the trial participants
also reported positive feedback and constructive suggestions to BibSonomy.
5.4.4.3.1 Experience and habits
One focus of the trial was to identify the value of such tools to the academic
community. Two questions from this section give results for our first research
question.
0%
20%
40%
60%
80%
100%
1. Making literature lists.
Statement
2. Using qualifications (comments) of others to select
literature to read.
3. Using qualifications (comments) of colleagues to select
literature to read.
unimportant
somewhat unimportant
neutral
4. Browsing through the personal collection of others.
5. Browsing through the personal collection of colleagues.
somewhat important
important
6. Adding tags to references.
7. Adding my comments to references.
8. Storing literature references.
Figure 13: Expectations from the librarian trial
Most people value “storing literature references” and “using comments of colleagues
for literature retrieval” as the most important features of social bookmarking tools
(combining “somewhat important” and “important” marks). Considering only important
features, colleague’s qualifications and the adding of comments to references is
estimated most important. Some perceive collaboration of others as not important
(statement 2 and 4 have a “somewhat unimportant” component), while none of the
participants perceived advice from colleagues as unimportant (statement 3 and
statement 5).
Figure 13 evaluates expectations of participants regarding the management of
literature. Better management of literature references as such was perceived as
important. Combining the important/somewhat important scale, participants mostly
wish to improve collaboration with colleagues for a better retrieval of literature.
1. I want to use my time spending on literature more effectively.
2. Better collaboration with my colleagues regarding literature.
unimportant
somewhat unimportant
neutral
somewhat important
important
3. I want to find more and better literature references.
4. Better management of my literature references.
5. Better storage of my literature references
0%
20%
40%
60%
80%
100%
Figure 14: Social bookmarking value
43
D15.2 – Case Study on social software in distributed working environments
5.4.4.3.2 Collaboration
In an interview after the user trial, several respondents mentioned that the usage of
such systems supports collaboration in an academic environment. “I think it will also
be very useful for our PhD students. I have advised my PhD students to use it. They
have to read a lot of articles and in this way they preserve their reading better and
make the results of their reading also accessible for the other members of the group.”
Another respondent selected BibSonomy as his social bookmarking system “because
the collaborative functionality was better developed”. However, he states that he
switched to Cite-U-Like since the latter appeared to be more user-friendly. The same
respondent emphasized the importance of colleagues’ and external literature lists.
5.4.4.3.3 Usability of BibSonomy
Usability aspects of BibSonomy were commented in the logbooks. One participant
stated late response times: “I really do hope that they will find a way to enter things
more quickly, because this might get annoying fairly quickly.” The interface was also
commented: “Not totally at easy with the interface, though”. Another positively
marked the ACM extraction capability: “It is very easy to look up the reference list of
an article in ACM. They try to make references into hyperlinks when possible. This
results in a very fast process where you can look up an article, its references and add
it to your literature list in very few clicks”.
5.4.5 Conclusion
This case study explored social bookmarking systems, especially BibSonomy, as
collaborative knowledge creation tools in a scholarly, distributed environment. The
specific characteristics of the unit of study, academics and research groups, led to
specific expectations towards social reference sharing. These expectations were
identified and together with benefits listed.
Four different groups were taken into consideration: two projects using BibSonomy
as their reference management tool, academics from the University of Amsterdam
and participants from the CKC challenge. While the first two groups were questioned
from the authors of this study, the latter ones were external challenges exploring the
value of social bookmarking and collaborative knowledge creation tools. The main
data source consisted of online questionnaires which participants worldwide could fill
out. The Amsterdam Librarian study also involved discussions and log files.
The results of the different cases lead to the following conclusions:
-
Evaluating the experiences and expectations with social bookmarking tools.
Social bookmarking for information retrieval complements traditional methods
such as using search engines or conferences. As the spread of information via
colleagues (e.g. advice, comments, suggestions from the research group)
plays an important role, the sharing and collaboration components in social
bookmarking respond to the specific demands of the research community.
However, a critical mass should be offered. This hypothesis is supported by a
further experiment during the time of the case study: BibSonomy was
introduced to a group of law students which did not accept the system due to
the lack of law content available.
44
D15.2 – Case Study on social software in distributed working environments
Traditional methods (or no methods at all) are still in favor when people
describe their reference management techniques. However, the import/export
facilities, the easiness to proceed with BibTex entries and manage one's
publications were positively acknowledged – making literature management in
the web an alternative for people not having a developed literature
management method yet or for people who want to collaborate and share
publication metadata but keep working with their tool in use, such as EndNote.
(2) Analyzing collaboration behavior
Researchers are aware of each other and their ideas. Several participants
pointed out that they broaden their literature knowledge with the advice of others and
that collaboration is the main motivation for using social bookmarking tools.
Feedback from ProLearn's and Nepomuk's usage of BibSonomy shows that
the acceptance of social bookmarking as a collaboration component for
distributed projects depends on the integration of the system into the project's
referenence management tasks and workflows. While Nepomuk uses
BibSonomy not only to store references, but also integrates the publication
management facility on the web site and allows people sharing project specific
metadata information, the ProLearn decision to report publications centrally,
does not motivate people to get to know the system and use it for their project
related work.
(3) Finding out about BibSonomy’s usability.
Offering literature management capabilities as well as bookmarking services,
BibSonomy covers a broad range of services for the research community.
Positively mentioned were the variety of features the system offers, and the
facility to integrate it with other publication management tools. This also
makes BibSonomy a tool for both, “expert” users, building hierarchical
relations with tags, using multiple functionalities and being part of the semantic
community and “mainstream” users looking for a simple system to manage
and share information.
The different user perspectives within this community implicates two major
further development paths:
Facilitate interaction with the system: a concise interface, clearly arranged
navigation features and personalized result lists shown to individuals might
be a first step to improve the system's usability and make it understandable
to a broader range of users.
Improve specific features: integrate a relations editor and more structuring
facilities within the tag cloud, facilitate the creation and maintenance of
groups, improve search within the system.
Research is already under way to enhance BibSonomy: support for users to find new
and particularly relevant content, new ways to provide better service for communities
of practice and the development of semantics. In further studies we hope to deepen
45
D15.2 – Case Study on social software in distributed working environments
our knowledge of a researcher's search and publication management behavior to
further enhance collaboration efforts in distributed environments.
46
D15.2 – Case Study on social software in distributed working environments
6 Conclusions
Social Software applications gain increasingly importance in the corporate learning
environment. According to some studies, 30% of organizations are already using or
planning to use blogs and/or wikis at the moment (McKinsey 2007). Although we
have noticed an increasing interest of companies in Web2.0 and Social Software on
a global scale, a recent study from the Gartner Group gives a warning that European
enterprises are about to miss this global trend and lag behind the implementation of
Social Software (Computerwoche 11.07.2006).
The discussed case studies demonstrate that the Web 2.0 and social software
represent a strong wave that has the potential to change in a substantial way our
society, politics, and economy. The use of social software in corporate environments
is still rather young and connected with a change in processes and working
environment for the purpose of knowledge building and knowledge sharing and thus
informal learning within distributed working environments.
Although all displayed case studies differ highly in application scenarios and first
conclusions on impact, they demonstrate very high potential for the change of
communication and collaboration processes. But as any other piece of software, the
scenarios have also demonstrated that all the listed tools are only an instrument that
can either support or hinder people in their collaboration, communication and social
interaction and thus finally in their informal learning processes. It depends on the
personal dispositions and attitude of the individual user towards the tool and the
peers in how far the tool can support these processes. For some user the tools
support a reflection process and are used as a sort of “reflection” or “learning” diary
while for some others the dialogue and the fast feedback that one receives via the
comments is most important. Others use the Social Software tools for storing,
sharing and linking with others. Hence, one important finding of the case studies
underpins the bottom up approach by outlining the high dependency of individuals,
their motivation, attitude and personal dispositions with the output of very different
and diverse usage scenarios in corporate environments.
Although social software emphasize the importance of each individual as single
contributor the real value of these contributions lie in the communication and
collaboration with others. A common finding of the case studies is the high
potentiality of social software in supporting group communication, collaboration,
knowledge sharing and social interaction for distributed working teams. First
implementation evaluations show some positive results that shall lead towards a
faster, more flexible, transparent and self-organised form of knowledge management
and networking. Overall, social software creates – provided an ideal setting for the
implementation is enabled - a new network of information resources that also relates
to specific persons and contextualises the information in a new way. Two of the
described case studies have demonstrated a growing number of specific topic related
posts which indicate that employees better integrate weblogs into daily working
practices than it has been the case with classical top-down knowledge management
tools. In this respect a major finding of this report relates to the need of a critical
mass of users and the integration within existing IT systems. Also a clear and
47
D15.2 – Case Study on social software in distributed working environments
immediate (business) need supports the integration of Social Software. In addition
these results raise the question on how corporate management can find structures to
support bottom up knowledge management tools. Still, from all these very ambitious
and successful implementations a key issue relates to the organizational culture.
After all, if the behaviour induced by the use of social software (e.g. openness and
information sharing) is in conflict with the organization culture, it might turn out to be
a potential hindrance to a successful introduction and adoption of social software. In
a similar way, the bottom-up emergent dynamics of social software necessitates that
trust and control are delegated to the users. However, it must also be ensured that
access to confidential and financial information is restricted in order to avoid
concerns about misuse, abuse, and reliability. Ultimately, the key issue for
companies is to understand if the employees’ needs are covered by the Social
Software solution and if the solution is in line with the organizational culture and the
attitude of the employees.
Consequently, the different findings implicate that future research need to focus on
different objectives.
The success of Social Software applications depend on the personal dispositions and
attitude of individuals. Therefore major a research need is identified in the
motivational dynamics of corporate Bloggers and Web2.0 users. The research
towards the individual motivation could give clear indication on how to reach the
necessary critical mass of users in order to implement successfully Social Software
systems within companies.
Web2.0 applications foster a different way to communicate and collaborate than we
have observed from LMS. Consequently companies supporting the (internal)
knowledge exchange between employees with Social Software have to face a critical
change. So far there has not been any evident research on how these changes
influence business processes. Through the changed bottom up and networked
communication structures communication processes are changed as well. Future
research need to be directed towards strategies for corporate learning and working
environments to effectively integrate Social Software solutions for very specific needs
in different institutional cultures.
The findings of the case studies as well as the further research needs will drive the
future work of WP 15. The final deliverable of WP15 will include additional
conclusions and will contribute to a holistic picture of Social Software application in
corporate environments.
48
D15.2 – Case Study on social software in distributed working environments
7 References
Anderson, Ch.: The Long Tail. Random House. 2006.
Alexander, B. : ‘Web 2.0: A New Wave of Innovation for Teaching’, Educause
Review, Vol. 41, No. 2, April/March 2006, online:
http://www.educause.edu/apps/er/erm06/erm062.asp.
Bächle, M.:‘Social Software’, Informatik Spektrum, Vol. 29, No. 2, pp. 121-124. 2006.
Bryant, L.:‘Smarter, Simpler, Social’, Tech. Report, Headshift.com. 2003.
Brown, S.: How Long is the Ideal Blogpost? On Word Counts and Writing for
Entertainment. http://www.modernlifeisrubbish.co.uk/article/how-long-is-the-idealblog-post. 2007.
Barnett, A.: The Long Tail of Tags. 2006.
http://alexbarnett.net/blog/archive/2006/09/16/The-Long-Tail-of-Tags.aspx
Cattuto, C., Loreto, V., & Pietronero, L.: Collaborative Tagging and Semiotic
Dynamics. PNAS 104, 1461 (2007). URL:http://arxiv.org/abs/cs/0605015v1
(03.03.2007).
Challenge instructions for BibSonomy. 2007. 14 May 2007
<http://www.bibsonomy.org/help/events/ckc2007>
Computerwoche.de:
Gartner:
Europas
Firmen
verschlafen
http://www.computerwoche.de/nachrichten/578603/ 2006.
Web
2.0
Dalsgaard, C.: ‘Social software: E-Learning beyond learning management systems’,
European Journal of Open, Distance and E-Learning. 2006.
http://www.eurodl.org/materials/contrib/2006 /Christian_Dalsgaard.htm.
Eden, C. & Huxham, C.: Action Research for the Study of Organizations, in Clegg, S.R.,
Hardy, C. & Nord, W.R. (eds.) Handbook of Organization Studies, pp. 526-542. London:
Sage Publications. 1997.
Eisenhardt, K.M.:Building Theories from Case Study Research. Academy of Management
Review, Vol. 14, No. 4, pp. 532-550. 1989.
Hammond, T., Hannay, T., Lund, B., Scott, J.: Social Bookmarking Tools (I): A
General Review. dlib, 11(4), 2005.
Hotho, A., Robert Jäschke, Christoph Schmitz, Gerd Stumme: BibSonomy: "A Social
Bookmark and Publication Sharing System". 14th International Conference on Conceptual
Structures. 2006.
ITU: International Telecommunication Union. 2005.
49
D15.2 – Case Study on social software in distributed working environments
Jackson, A., Wanda Orlikowski: "Corporate Blogging: Building community through
persistent digital talk," hicss, p. 80, 40th Annual Hawaii International Conference on
System Sciences (HICSS'07). 2007.
Jäschke, R., Grahl, M., Hotho, A., Krause, B., Stumme, G.:“Organizing Publications and
Bookmarks in BibSonomy”, in Proceedings of the Workshop on Social and Collaborative
Construction of Structured Knowledge (CKC2007), WWW 2007, Banff, Canada, 2007.
Klamma, R., Chatti, M. A., Duval, E., Fiedler and S., Hummel, H. et al.: ‘Social
Software for Professional Learning: Examples and Research Issues’, Proceedings of
the 6th International Conference on Advanced Learning Technologies, Kerkrade, the
Netherlands, 5-7 July, pp. 912-914.2006.
Krüger, A.: Massensterben bei den Blogs?: Wie sich die Ergebnisse einer
Untersuchung gegen den Strich bürsten lassen, 2003. URL:
http://www.heise.de/tp/r4/artikel/15/ 15849/1.html (03.03.2007).
Langley, A.: Strategies for Theorizing from Process Data. Academy of Management
Review, Vol. 24, No. 4, pp. 691-710.1999.
Licklider, J.C.R., Taylor, R. and Herbert, E.: ‘The Computer as a Communication
Device’, International Science and Technology, April. 1968.
Lovejoy, T., Grudin, J.: Messaging and Formalitiy: Will IM follow in the Footsteps of
Email? Proc. INTERACT 2003, 817-820. 2006.
McKinsey (2007) How businesses are using Web 2.0: A McKinsey Global Survey.
The McKinsey Quarterly, March. 2007.
Millen, D., Feinberg, J., Kerr, B. 2005. Social bookmarking in the enterprise. Social
Computing, 3/9.2005.
Nepomuk – The Social Semantic Desktop. 2006. 14 May 2007
<http://nepomuk.semanticdesktop.org/xwiki/>
Nepomuk – Publications. 2006. 14 May 2007
<http://nepomuk.semanticdesktop.org/xwiki/bin/view/Main1/Publications>
O’Reilly, T.: ‘What Is Web 2.0 - Design Patterns and Business Models for the Next
Generation of Software’, 2005). www.oreilly.com,
http://www.oreillynet.com/pub/a/oreilly/tim/ news/2005/09/30/what-is-web-20.html,
{3.7.2006}.
Perin, C.: Electronic Social Fields in Burocracies. Comm. ACM, 34, 12 (1991), 74-82.
1991.
50
D15.2 – Case Study on social software in distributed working environments
Pleiade Management en Consultancy. "Academic Social Referencing tools: a user trial with
BibSonomy and Cite-U-Like organized by the Library of University of Amsterdam."
2007. 4 May 2007 <cf.uba.uva.nl/nl/projecten/academic_social_referencing.pdf>
Probst, G. J. B., Raub, S., Romhardt, K.: Wissen managen: Wie Unternehmen ihre
wertvollste Ressource optimal nutzen. FAZ-Verlag, Frankfurt am Main, (2. Aufl),
1998.
Prolearn: Thematic Workshop on Social Software Connecting People. Retrieved 20th May,
2007 from <http://tws.prolearn-project.org/programme.html>
Ryyppö, T (2007) Social Software – Leveraging Virtual Network Organization.
Helsinki University of Technology, Master's thesis. Available online:
http://www.dicole.com/wp-content/uploads/2007/06/thesis_ryyppo_20070605.pdf
Schmidt, J. & Mayer, F.: Wer nutzt Weblogs für kollaborative Lern- und
Wissensprozesse?. Ergebnisse der Befragung 'Wie ich blogge?!' 2005. Bamberg.
URL: http://www.fonk-bamberg.de/pdf/fonkbericht0602.pdf (3.3.2007)
Shirky, C.: ‘Social Software: A New Generation of Tools’, Esther Dyson’s Monthly
Report (10). 2003.
Sifry, D.: State of the Blogosphere, April 2006 Part 1: On Blogosphere Growth. 2006.
URL: http://technorati.com/ weblog/2006/04/96.html (03.03.2007).
Sixtus, M.: ‘Das Web sind wir’, Technology Review, July. 2005.
Spiegel Special:. Das Magazin zum Thema: Wir sind das Netz, 3/2007. 2007
Surowiecki, J.: The Wisdom of Crowds. Why the Many Are Smarter Than the Few
and How Collective Wisdom Shapes Business, Economies, Societies and Nations,
Little, Brown ISBN 0-316-86173-1. 2004.
Tim O'Reilly (2006-12-10). Web 2.0 Compact Definition: Trying Again. 2006.
http://radar.oreilly.com/archives/2006/12/web_20_compact.html
Tscherteu, G.: ‘Social Software – eine Einführung’. 2003.
http://www.realitylab.at/pdf/SocialSoftware.pdf.
Tulving, E.: Episodic memory: From mind to brain. Annual Review of Psychology, 53
(1), 2002, 1–25. 2005.
Wassermann S., Faust, K.: „Social Network Analysis: Methods and Applications.
Cambridge University Press. Cambridge. 1994.
Wellman, B. & Hampton, K.: Living Networked One and Offline. Contemporary
Sociology, Vol. 28, No. 6, pp. 648-654. 1999.
Wenger, E.: Communities of Practice: Learning, Meaning and Identity, Cambridge
University Press, UK. 1998.
51
D15.2 – Case Study on social software in distributed working environments
Workshop on Social and Collaborative Construction of Structured Knowledge. 2007. 20 May
2007 <http://km.aifb.uni-karlsruhe.de/ws/ckc2007/>
Yin, R: Case study research: Design and methods (2nd ed.). Thousand Oaks, CA: Sage
Publishing. 1994.
Yin, R.: Applications of case study research. Newbury Park, Sage Publications. 1993.
Zerfaß, A. Corporate Bolgs: Einsatzmöglichkeiten und Herausforderungen. BIG
BlogInitiativeGermany. 27. January. 2005.
www.bloginitiativegermany.de
52
D15.2 – Case Study on social software in distributed working environments
Appendix
Appendix A: Questionnaire for Bibsonomy Study
5. In what age group are you?
5. 19 and under
6. 20-29
7. 30-39
8. 40-49
9. 50-59
10. 60+
-
Please select your gender. (male/female)
-
z
z
z
z
z
z
z
z
z
What is your main activity in the project?
Administrative Assistant
Professor
Researcher
Technical expert
Student
Ph.D. Student
Administrator
Manager
Other
z
z
z
z
z
z
z
z
z
How do you find new literature?
I use domain specific search engines.
Colleagues of my project tell me.
Other colleagues outside my project tell me.
Via conferences.
I search in literature lists.
I use search engines.
I get tips from mailing lists.
I use social bookmarking tools.
Other
-
Which methods do you have for storing and managing references?
I download the documents and store them, organized in folders and subfolders on my
hard disk.
I use BibTeX software.
I use EndNote software.
I bookmark the URLs and store them in my browser.
I use social bookmarking systems.
I use Excel.
Other.
z
z
z
z
z
z
z
z
z
z
z
z
Please indicate the extend to which you agree or disagree with the following
statements.
--> (Strongly agree, Agree, Disagree, Strongly Disagree)
I am interested in my colleague's literature findings.
I prefer not to share my literature classification with other people.
I spend a lot of time searching for literature alone.
I do not read comments of colleagues added to literature.
Advice of others influences my literature choice.
53
D15.2 – Case Study on social software in distributed working environments
-
Do you use social bookmarking tools other than BibSonomy? (yes, no)
-
How often did you carry out the following activities?
--> (never, 1-10 times, more than 10 times)
Browsing through the collections of my colleagues.
Adding tags to references.
Adding comments to references.
Adding tags to bookmarks.
Exporting literature references with BibTeX.
Adding private comments to my postings.
Exporting with EndNote.
Copying entries from other colleagues.
Browsing the publication list of my project.
Exporting with RSS.
Looking up my references.
Creating literature lists for a publication.
Browsing the general tag cloud to find bookmarks.
Browsing my projects's tag cloud to find information.
Reading comments that others of the team made to a publication.
Using relations to better classify my tags.
Reading the blog to find out about news.
Reading the help pages to get more information.
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
z
Please indicate the extend to which you agree or disagree with the following
statements:
--> (Strongly agree, Agree, Disagree, Strongly Disagree)
I found a certain literature reference of interest to me by browsing the reference of my
colleagues.
We published a paper using BibSonomy for creating the reference list.
I never found interesting bookmarks by browsing the tag cloud.
Due to my personal classification I can refind my bookmarks quickly.
When I followed the link to other users sharing my entries I found people of similar
interests.
I find the keywords in the tag cloud of my project appropriate for the classification of
our literature.
I improved the metadata for my references by looking up other reference entries for
the same publication.
When other people copy one of my bookmarks I am not encouraged to participate
more.
I find the automatic extraction of references by scrapers not very convenient.
Please indicate the extend to which you agree or disagree with the following
statements:
--> (Strongly agree, Agree, Disagree, Strongly Disagree)
BibSonomy is easy to use.
I understand of the menus and toolbars.
I will not be able to learn how to use all that is offered in this software.
Navigating through the menus and toolbars is easy to do.
I can easily navigate to the bookmarks I am searching for.
I often have to re-read the help pages to work with BibSonomy.
I do not mind that others can see my bookmarks.
This software is not flexible.
54
D15.2 – Case Study on social software in distributed working environments
z
z
z
z
z
Finding the options that I want in the menus and toolbars is easy.
I often use the private posting functionality.
Discovering new features is easy.
I get my reference managing tasks done easily.
I am afraid that my personal information is being used by unknown people.
55