ArticlePDF Available

The Benefits of Networking

Authors:

Abstract

A network of 18 competence centres, called ESPINODES, supports companies conducting software process improvement (SPI) experiments under the European Commission's ESSI programme. Three Nordic ESPINODES report on experiences from participating in this network. We focus on general issues using examples from our local activities. The benefits of the network as seen from the sponsor, the participants as well as their customers are discussed.
The Benefits of Networking
Jørgen Bøegh1, Mads Christiansen1, Ebba Þóra Hvannberg2, Tor Stalhane3
1DELTA, Danish Electronics, Light & Acoustics
Venlighedsvej 4
2970 Hørsholm (Hoersholm), DK - Denmark
{jb, mc}@delta.dk
2University of Iceland, Hjardarhaga 2-6,
IS-107 Reykjavik, Iceland,
ebba@kerfi.hi.is
3SINTEF Telecom and Informatics
N-7465 Trondheim, NORWAY
Tor.Stalhane@informatics.sintef.no
Abstract. A network of 18 competence centres, called ESPINODES, supports
companies conducting software process improvement (SPI) experiments under the
European Commission’s ESSI programme. Three Nordic ESPINODES report on
experiences from participating in this network. We focus on general issues using
examples from our local activities. The benefits of the network as seen from the
sponsor, the participants as well as their customers are discussed.
1 Introduction
There are many kinds of networks in the world: franchises and networks of companies
that share some service. Examples of these are McDonalds hamburger chain, airline
alliances, and real estate chains. In this paper we will focus on networks of
excellence in a technical area. Our experiences come from participating in a network
of organisations supporting Software Process Improvement (SPI). Each node has
clear responsibilities: serving regional clients that run an experiment on Software
Process Improvement; serving the local industry for dissemination and general
information on SPI, and finally sharing knowledge and collaborating between nodes.
The intended target audience for this paper is current and future customers and
participants in networks. The type of questions we would like to address is:
Customers of networks:
What will customers gain by contacting a network?
Is it important for customers to have a one-stop-shop, e.g. their local node or the
network’s aggregated information or will they benefit from looking at all the nodes
in the network?
What type of tasks will customers ask the network to do? How does the network
best serve them and reach them?
Participants in a network:
What types of resources can participants draw from the network?
Can the existence of the network make participants more visible?
Can participants improve their services?
Can participants offer more services?
2 ESPINODE network
Since 1993, the European Commission’s ESSI (European Systems and Software
Initiative) programme has supported almost 400 Process Improvement Experiments
(PIEs), designed to improve the quality and cost-effectiveness of the software
development process [1]. In 1998 the European Commission established a network of
competent partners, the so-called ESSI PIE Nodes. 18 ESPINODEs distributed
throughout Europe plus Israel participate in the network (see Figure 1).
Fig. 1. ESPINODEs
The ESPINODE network is characterised by:
ESPINODEs get funding from the European Commission.
ESPINODEs do not have a direct commercial goal and operate only for a limited
period.
Each of the nodes has a great freedom to operate the way it likes within the
framework specified in the contract with the Commission. The network is not
centrally managed, and it does not have a steering committee or a board.
There is no way to force nodes to work in a similar way or together. Common
interests and our wish to fulfil our contract with the European Commission
motivate us to work together.
The customers of the network get to know the network through the individual nodes.
This means that once a customer contacts a node of the network, we (the nodes)
introduce the network to the customer. Another approach would be to brand the
ESPINODE network. Then we would attract the customers through some common
web-site or reference, e.g. www.espinodes.org, where the home region node is
introduced to the customer. This latter approach is more natural when the network is
charged with producing some specific deliverable. This is for instance the case with
the European Support Network of Usability Centres (URL
http://www.lboro.ac.uk/research/husat/eusc/). The EUSC is a commercial network
sponsored by the EC Telematics Application Programme.
Nations and unions of nations have funded networks of excellence. Examples of
these are COST or European Co-Operation in the field of Scientific and Technical
research, and clusters that are created within the 4th and 5th framework programmes of
the European union. SPIN or Software Process Improvement Networks throughout
Europe and USA have also been successful. SPIN Groups are different in many
aspects from the ESPINODE network, since the SPINs themselves mostly serve their
region and do not co-operate closely between themselves.
The ESPINODE network has a rich diversity because the regional influence is
considerable. We have to serve local customers that may be quite different from one
region to another. In addition, the nodes have different expertise and goals, which
influence the actual approach to the customer. Some networks have the goal of
serving a homogeneous customer base. For the ESPINODE network it is important
that we respect our cultural differences and the needs and expectations of our
customers.
3 Regional experiences
In the following subsections we highlight the individual regional experiences of three
nodes in the network. There is no particular reason for choosing those three and they
are not model nodes. The presentation of each node will explain the context in which
the work has started; the focus of the activities or their characteristics; and finally
their experiences. We try to articulate the experiences from both the participant’s and
our customer’s point of view. A more thorough description of the activities of each
node can be found in a consolidated report [1].
3.1 Norway and Sweden
QIS is the ESPINODE that services Norway and Sweden and is organised by SINTEF
Telecom and Informatics, which is a division of SINTEF - an independent non-profit
R&D organisation situated in Trondheim, Norway. At the time when the European
Commission established the ESPINODE network, there was no interest among
Swedish firms and SINTEF was thus assigned the responsibility of PIEs in both
countries.
SINTEF Telecom and Informatics was at that time engaged by the Norwegian
Research Council in a program called SPIQ - Software Process Improvement for
better Quality. As part of this we were already building up a network of Norwegian
software companies doing process improvement and saw this as an opportunity to
enlarge the network and thus get a larger area of impact. In addition, we had already
considerable experience in PIE work since we had been consultants to several
Norwegian companies doing SPI partly financed by the EU Commission. The
ESPINODE idea provided an additional opportunity for dissemination of SPI
knowledge and experience as well as marketing SINTEF services.
The node started with four PIEs - two in Norway and two in Sweden. Later a fifth
Swedish PIE was added, thus giving us responsibility for a total of five PIEs. Our
experience being a node is mixed. We expected a lot of requests for assistance in
technical matters but received mostly requests for help with administrative matters
and help in tackling the EU bureaucracy. We expected a good turnout for our
regional workshops and started a rather heavy marketing blitz. It turned out, however,
that only the PIEs participated. On the other hand, the PIEs have stated that they
found the workshops important to them both for picking up new ideas and for being
kept informed on the SPI scene. This is consistent with the result of a survey
performed by the Commission, which showed that contact with other companies
doing SPI was of high importance to all companies engaged in SPI.
Our “How-to-do” memos, newsletter and web pages have turned out to be
successful in the sense that they are read and some of the material downloaded by
people doing SPI in Scandinavia.
One of the concerns of the Commission when they started the ESPINODE network
for Norway and Sweden was that there had been no PIEs in Sweden before. Thus, the
Norwegian-Swedish ESPINODE was assigned a special responsibility to market SPI
and PIEs in Sweden. The Commission’s assumption of little SPI activity in Sweden
turned out to be wrong documented by looking at the Eurex web page
(http://www.sisu.se/projects/eurex/ASADEP.html).
Both SINTEF and the Commission assumed that the two Nordic countries Norway
and Sweden were so similar that it would be easy for SINTEF to gain access to fora in
Sweden which could be used to push the message of SPI to the local industry. Our
experience shows that this assumption was wrong. We first tried to arrange a SPI
workshop in Gothenburg, but with no success. After the EU reviewers’ suggestions,
we then tried to piggyback on events arranged by the Swedish Computer Society in
Gothenburg and Stockholm, but got only lukewarm responses. The only Swedish
forum where we were able to get access was the Trondheim-Jemtland forum
“Vindue”. The lesson learned is that it is difficult to operate without a network
especially if you start to operate in another country. Even among the Scandinavian
countries there are great cultural differences that must be carefully dealt with to obtain
success. It is also mandatory to have access to a local network.
Even so, we have been able to make a considerable amount of industrial contacts
with Swedish industry through contacts at Nordic events like the EuroSPI
conferences. The situation was simpler in Norway, where we already had a large
network established through the SPIQ program and a good working relationship with
the Norwegian Computer Society - DND.
All considered we feel that the work of our ESPINODE has given value for money
to the software industry in Norway and Sweden in the area of SPI. There are three
reasons for this:
The companies in Norway and Sweden have access to a large amount of SPI
knowledge through the QIS newsletter, our web pages and all the other
ESPINODE web pages that are linked to it.
Through the workshops, the PIEs have met other companies doing SPI and have
been exchanging information and practical experiences.
By encouraging and helping Scandinavian industry to participate in SPI
conferences, the industry has been able to get many contacts and access to a large
amount of information, which will help them in their regular SPI work.
Since participation in the ESPINODE network carried a considerable cost for
SINTEF - the Commission only pays about 40% of our commercial rates - we also
need to see what we have got out of the participation in the ESPINODE network. In
general, we think that the participation has had a positive bottom line for SINTEF.
This conclusion builds on three observations:
We have enlarged our industrial network, thus giving us more marketing
opportunities. This goes both for Norway and Sweden - and to some degree
Denmark in areas were we are not marketing services in competition with the
Danish ESPINODE (PRIMERS).
We have received valuable inputs from Scandinavian industry pertaining to the
work in the turbulent border area between research and practicalities where
consulting usually takes place.
We have had the opportunity to sharpen and document our industrial experiences
and have moved more theory and methods from research to practical, industrial
application.
3.2 Denmark
The Danish ESPINODE - PRIMERS - is hosted at DELTA Danish Electronics, Light
& Acoustics. DELTA is a private self-governed company, approved by the Danish
Ministry of Trade and Industry as a GTS (Approved Technology Service) Institute
located in the northern part of Copenhagen. As a GTS Institute DELTA has an
obligation to contribute to innovation and dissemination to industry. Participation in
International R&D projects and networks ensures up-to-date information and
knowledge.
The obligation to disseminate knowledge is done partly by organising courses and
seminars and partly by running an experience exchange network. The experience
exchange network (Datatechnical Forum) was established in 1973 and comprises
about 75 Danish companies. The network runs 10 experience exchange groups
covering areas like methods and tools, quality management, software process
improvement (i.e. the Danish SPIN group), project management, and testing.
In addition, DELTA is a leading consulting company in the area of software
process evaluation and improvement. Software process evaluations are based on the
Bootstrap approach. DELTA has conducted more than 50 commercial evaluations.
This has lead to a detailed understanding of the state-of-practice in the Danish
software industry. DELTA was also the leading partner in a large (2.6 MEuro)
national SPI effort [2].
DELTA was already deeply involved in SPI projects for Danish industry and had
long-term experience in network activities. In addition, we felt that the experiences
already accumulated at DELTA could be valuable for SPI projects in the framework
of ESSI projects. This explains the background and motivation for creating a Danish
ESPINODE.
In the ESSI context DELTA supported three PIEs. The type of support was
twofold. Firstly, a general support was related to the specifics of European
Commission funded PIEs, including identification and refinement of project goals,
reporting to the Commission and presentation of project results. Second, DELTA was
a subcontractor to the PIEs and carried out specific consulting services related to
process improvement. These two approaches to the PIEs provided a very close
relation and enabled identification of problems early in the projects.
Early in the process, we formed a PIE-club - a small network for experience
exchange among the PIEs. We believed that this could be a forum of considerable
inspiration for the PIEs in order to show new possibilities in their daily work.
Although the projects were quite different, we knew that the difficult parts of SPI
projects are related to communication, diffusion and adoption, and not so much to the
technical issues. As with SPI it also takes time to build a network that creates benefits.
The first meetings were used to tell the others about their projects. The purpose was
also to try to focus on areas of common interest. Unfortunately one of our PIEs was
very impatient and didn’t feel they could benefit from listening to technical
discussions outside the scope of their project. They would rather work at home if the
time spent in the network were just added to their work-hours. This attitude to
networking resulted in a network that never really provided the anticipated benefits
and having only two PIEs in a “network” didn’t really make any sense. Conclusion
networks are totally dependent on the members and their interest in and commitment
to the network. The two PIEs interested in networking continued in the Danish SPIN
group, one of the experience exchange groups organised by DELTA within
Datatechnical Forum. The 4 to 6 annual full-day meetings attract 15 to 25 delegates
from a variety of companies in Denmark.
The web pages provided by PRIMERS included a “Paper of the month” which has
been very popular among the other nodes and the attached PIEs although they were
published less frequent than every month.
The PIEs may only regard the ESPINODE as a service centre, but there are more
benefits. Being part of the international network DELTA had access to knowledge
available at other ESPINODES. In cases where DELTA did not immediately have the
answer to questions raised, it was possible to seek information and assistance
elsewhere. These extra resources considerably strengthened DELTA’s support to the
PIEs.
DELTA also benefited from being part of the ESPINODE network at a more
general level. First, a good network of personal contacts was established which again
resulted in closer collaboration with other ESPINODES. In the heterogeneous group
of ESPINODES, one cannot expect to find a common interest in everything.
Therefore, we found it most efficient to establish collaborations on a case by case
basis. These common activities included setting up a joint workshop and other
dissemination actions. Only the future will show how far the ESPINODE network
will be developed, but until now, it has at least provided new opportunities.
3.3 Iceland
The ESPINODE in Iceland, called ESPICE, is at the University of Iceland. At the
university there is considerable experience in software engineering. There was no
formal association with industry, and no formal activity of software process
improvement was going on when the node was established. Because of the small
society it was relatively easy to reach the clients and associations and people that
could disseminate the activities of the node. Individual professors have done
consulting for industry but there is no institute on software engineering. There is also
a close relationship with the Continuing education institute of the University of
Iceland that can offer training in software engineering in co-operation with the node.
There was one PIE attached to the node. The major activities of the Icelandic node
have been workshops that each focus on one single theme. Each workshop consists of
a lecture, tutorials, and demonstrations of tools. Exchange of views and opinions
among members of the workshop is an important component and the workshops are
pragmatic and hands on. Although many of the companies are well aware of the
importance of standard, and defined methods, only a few have an ISO certificate and
most are newcomers to software process improvement. Little consulting is presently
offered in Iceland in this area.
The experience of the ESPINODE network has been positive. As a network
member, we have established new relationships with mature service providers of SPI
and setup a basis for a technical and commercial co-operation. We have also relied on
other network members to answer questions on behalf of our clients, thus taking
advantage of the expertise of the network.
As a national node we have succeeded in creating an awareness of SPI and 22
companies have attended our workshops. We have also introduced the activities to
quality associations and a governmental advisory board on information technology.
However, at this time of writing we can not measure to what extent our customers
have improved their software processes or attempted to do so.
In a small society, it is important that companies have access to foreign resources.
Because of the size of the industry, say in Denmark, Norway or Sweden, there is
access to more experienced consultants in the area of SPI. Therefore, the ESPINODE
network proved to be a good opportunity to transfer knowledge from other more
experienced countries.
4 Benefits
We believe that networking is the future. In Silicon Valley people quite openly share
information even between companies that are in direct competition. One reason is that
employees change job more often in Silicon Valley than generally is the case in
Europe. If a company cannot prevent information from “leaking” it may just as well
use it as an opportunity. It is well known that the growth rate in Silicon Valley is
higher than average and one reason is the culture of open exchange of information
that makes companies more competitive. If I share my knowledge with 10 people,
they are more likely to share their information with me. This means that I have the
experiences from 10 people. When a network is based on mutual trust and common
interests it will lead to a win-win situation.
At the beginning of our work in the ESPINODE network we analysed whom the
beneficiaries of our networks were, what types of activities we wanted to carry out,
who would do it and how we would measure our success. We also did a brief analysis
of the environment and the context of our work including assumptions, hindrances
and opportunities. The analysis was carried out using the Soft Systems Methodology
[3]. This analysis was important since it helped us to focus our work. As a result of
this analysis we formed a few working groups with defined objectives. Furthermore,
informal activities between participants take place, for example by asking for
resources, experiences, advice etc. It is important that the activities of the network
directly support the primary activities of each of the nodes. It is difficult to motivate
the nodes to do something only for the network's sake.
Both the customers and the participants of the network are beneficiaries. The
sponsor of the network, i.e. the European Commission should also benefit. We think
the Commission benefits, since the nodes support many of the administrative
activities. Also, research activities funded by the Commission are more likely to be
widely disseminated and exploited with the aid of a network.
It is evident that the nodes differ greatly. Some nodes have long experience of
providing services while others have little experience. Some nodes have a clientele
that is capable and motivated. Those nodes often have the objective to service
customers that are already convinced of the importance of SPI. Others have customers
with low awareness pertaining to the importance of SPI. These nodes have the
objective to raise awareness of SPI and to increase the number of companies that are
involved in SPI. The ESPINODEs have benefited from the network in having access
to a broader set of competencies and opportunities to share ideas. The customers of
the ESPINODEs have benefited in terms of the increased attention and focus from the
ESPINODEs.
   
   
 
 
  !"#$% &'
( &)* + &'   
  !"#$-, . /
( 0 ( &'1 
2436587:9<;>=<2
?<@ACB)D-EF@G58B)B'H
Fig. 2. Resources
The inter ESPINODE services have been fairly well structured. A discussion forum
and a pool of references to resources were established (see Figure 2). The resources
include links to relevant web pages, books, articles and experts within the community.
The ESPINODEs are responsible for adding information to the resource pool by
filling in and submitting a web-based form to the maintainer of the specific topic. The
maintainer - another ESPINODE - is responsible for keeping the information up-to-
date. Currently no information on the number of accesses to these resources is
available to us.
In order to understand the needs of our customers we did a marketing survey
during the first year of the network. The major benefit of the network is that
customers get better service from their local node because they can focus on specific
local problems and they can consult other nodes. Examples of questions have been:
What resources are there on teams and teamwork in software development?
How do governments in other countries support their institutes and department
with the procurement process?
We have found that customers of the network mainly access their local node.
Analysis of web-logs for all three ESPINODEs have, however, shown that local web-
sites on SPI are accessed from abroad, although there is no evidence that they are
from customers of other nodes. If customers should benefit directly from other nodes
then the nodes and the network would have to add services:
Each node could advertise its focus or speciality. Currently, there is most often
only a list of upcoming events.
The network has to provide a search tool in order to look for something specific at
the nodes.
The network could provide an on-line help-desk where participants place questions
that are redirected to appropriate nodes.
5 Recommendations
In this section, we try to summarise from our experience, our recommendations to
organisers of networks:
Personal relations are important for the success of a network; the Internet is not
enough.
The network’s success depends on the willingness of its participants to share
information.
Marketing has to be one of the primary tasks, especially in the beginning. Don’t
expect this to be easy, since the network has to decide what it needs to market first.
An analysis of the beneficiaries, goals, actors, activities, including deliverables and
the context has to be done early on and updated regularly.
Within a network, a subset of nodes is bound to work together on some tasks.
Some of the tasks are highly structured and decided by the network as a whole, i.e.
top-down. Other tasks are more bottom-up, that is unstructured and initiated by
some subset of nodes.
High performance requirements are a driving force for people to work together in a
team, and a network is organised by a team of people [4]. At the beginning, the
nodes set performance targets for themselves. As a network, we have been aware
of the requirement that the co-operation should deliver results.
To potential customers of the network, we can only recommend that they actively
search for networks that fit their needs. It is also the responsibility of the
customers to let the network know how it can provide better services and what it
does well.
It is difficult to decide or recommend what type of structure or organisation for a
network is the most important. Doing scenarios as a part of an analysis can help the
network to visualise the best organisation.
6 Conclusions
We have shared our experience from being participants in a network and hopefully
shed a light on how it has benefited our customers. We hope that the paper will both
motivate discussions on networks and also convince those that haven’t used a network
to do so in their work.
While it is not necessary for a network to live forever, there is a drawback that
many of the networks have a lifetime of 2-3 years. This is sometimes apparent from
the web-sites that haven’t been updated for some time, although the original material
continues to be valuable.
Our network is different from the McDonalds, airlines and real estate networks.
They are based on a certain level of control and common discipline in order to work.
Our network is purely based on interests in SPI and to improve our local business.
There is only a minimum of controlling elements built into the network and we
entirely rely on finding common interests and good personal relations.
To create a network, a framework and a clear scope is needed. This must be
developed among the key interested parties and communicated to others with the
same interests in order to attract their attention to the network. Setting up measurable
goals for the network up front is difficult, as the common interests may not be clear at
that time. Identifying the beneficiaries and clarifying their interests in an open way
helps. But we believe that one of the major benefits of networking is the unpredictable
mixture of common interests and good personal relations. Then spontaneous
formation of subgroups based on common interests eventually emerges and the
network shows its true value.
If we look at the story told, we can see the necessity of networks. The Norwegian
ESPINODE had difficulties in penetrating the Swedish community mainly due to lack
of access to a local network. In Denmark the PIE-club network became obsolete much
too early because the benefits of networking was not realised among all the members
of the network. If the interests are not clear to the entire network or the commitment is
not present, the network will fail. The size of the network is also important. At the
European level the network proved to be a good opportunity to transfer knowledge
from more experienced countries in the area of SPI such as Denmark, Sweden and
Norway and to smaller societies such as Iceland where there is less variety of
expertise because of the small population.
References
1. ATB-Bremen, with contributions from all ESPINODEs, ESPINODE, Consolidated report.
Issue No. 2, Period May, 1998-April, 1999, http://www.atb-bremen.de/spinode-ng/ecr/
2. J. Johansen and L. Mathiasen. Lessons Learned in a National SPI Effort. The Danish SPI
Initiative: Centre for Software Process Improvement. Paper for ISCN 1998.
3. Checkland, P. and Scholes, J. Soft systems methodology in action. John Wiley & Sons, Ltd.
1999.
4. Katzenbach, J. R., Smith, D. K., The Wisdom of Teams. McGraw-Hill, 1998.
ResearchGate has not been able to resolve any citations for this publication.
with contributions from all ESPINODEs, ESPINODE, Consolidated report
  • Atb-Bremen
ATB-Bremen, with contributions from all ESPINODEs, ESPINODE, Consolidated report. Issue No. 2, Period May, 1998-April, 1999, http://www.atb-bremen.de/spinode-ng/ecr/
Lessons Learned in a National SPI Effort. The Danish SPI Initiative: Centre for Software Process Improvement
  • J Johansen
  • L Mathiasen
J. Johansen and L. Mathiasen. Lessons Learned in a National SPI Effort. The Danish SPI Initiative: Centre for Software Process Improvement. Paper for ISCN 1998.