=Paper=
{{Paper
|id=None
|storemode=property
|title= Towards improving structure and content of information demand patterns
|pdfUrl=https://ceur-ws.org/Vol-963/paper2.pdf
|volume=Vol-963
}}
== Towards improving structure and content of information demand patterns ==
https://ceur-ws.org/Vol-963/paper2.pdf
Towards Improving Structure and Content of
Information Demand Patterns
Kurt Sandkuhl[1,2], Dirk Stamer[1]
[1]
Rostock University, Institute of Computer Science
Albert-Einstein-Str. 22, 18059 Rostock, Germany
[Kurt.Sandkuhl, Dirk.Stamer]@uni-rostock.de
[2]
Jönköping University, School of Engineering
Box 1026, 55 111 Jönköping, Sweden
Kurt.Sandkuhl@jth.hj.se
Abstract. Work presented in this paper originates from the field of information
logistics and aims at the reduction of information overload. Among the
approaches for reducing information overload, the concept of information
demand patterns (IDP) was developed for capturing organizational knowledge
on how to improve information flow in enterprises. The paper investigates how
the structure of IDPs is perceived by potential users and which enhancements of
the structure could be made for improving content and quality of IDPs. For this
purpose, an investigation including two steps is performed. The first step
involves students in a university course who apply the IDP structure for
developing IDP candidate descriptions. The intention is to explore whether the
IDP structure reached a level of maturity to transfer it to non-experts in the IDP
field. The second step evaluates the IDP developed by the students with respect
to the quality of the different parts of the IDP structure. The conclusion of the
investigation is that the IDP structure in general is applicable and useful, but
consistency between the different IDP parts needs to be improved by providing
aids and guidelines.
Keywords: Information demand, information demand pattern, demand
modeling, validation.
1 Introduction
Work presented in this paper is a contribution to the field of information logistics,
which aims at improving information flow in enterprises and organizations [1]. The
general intention is to contribute to reducing information overload, which more and
more is perceived as problem in enterprises [2]. Among the many approaches for
achieving a more demand-oriented information supply, modeling and analysis of
information demand have been proposed [3] and the concept of information demand
patterns has been developed. Information demand patterns are considered as a way of
capturing organizational knowledge about what information is required for specific
roles in an organization.
Application and validation of information demand patterns so far was based on
industrial case studies, like in collaborative engineering [4], and surveys in higher
education [7]. This paper aims to extend work on information demand patterns by
addressing the aspect of how to improve the inner quality of information demand
patterns, i.e. the completeness, accuracy and pertinence of the pattern content, which
is structured into different parts. For this purpose, an investigation including two steps
is performed. The first step involves students in a university course who apply the
IDP structure for developing IDP candidate descriptions. The intention is to explore
whether the IDP structure reached a level of maturity to transfer it to non-experts in
the IDP field. The second step evaluates the IDP developed by the students with
respect to the quality of the different parts of the IDP structure. The main
contributions of this paper are (1) results from validating the concept of information
demand pattern in a university course, (2) results from evaluating the different parts
of information demand patterns regarding their quality and (3) conclusions how to
improve the concept and structure of IDP.
The remaining part of this paper is structured as follows: section 2 describes the
background for the work by introducing the concept of information demand and
information demand patterns. Sections 3 and 4 contain the two-step investigation
performed: Section 3 focuses on the first step of the investigation, i.e. the set-up, data
collection and results of applying IDP in a university master course for developing
IDP candidate descriptions. Section 4 covers the second step of the investigation,
which is a quality assessment of the different parts of the IDP candidate descriptions
produced in step one. Section 5 draws conclusions and presents an outlook on future
work.
2 Background
Work on understanding the nature of information demand and on identifying and
structuring patterns of information demand form the background for this paper. This
background will briefly be summarized in this section. Furthermore, the section
summarizes the results of a previous IDP validation exercise performed in higher
education in 2010 (section 2.3).
2.1 Information Demand
The notion of information demand is closely related to work in the area information
logistics, which considers understanding information demand as key aspect of
information logistics solutions [4]. Information demand usually includes different
dimensions, like the content required, the time of delivery, the location, the
presentation and the quality of information. The research field information logistics
explores, develops, and implements concepts, methods, technologies, and solutions
for the above mentioned purpose.
Lundqvist explored the nature and characteristics of information demand in an
enterprise context in an empirical investigation [5]. The conclusion from the study is
information demand of employees in an organization is to a large extent based on the
organizational role and the tasks an employee has. This role-centric perspective with
tasks and responsibilities as primary characteristics has been the starting point for
developing a method for information demand analysis [3].
Lundqvist defines information demand as: “Information demand is the constantly
changing need for relevant, current, accurate, reliable, and integrated information to
support (business) activities, when ever and where ever it is needed.” [5, p. 61]
2.2 Information Demand Patterns
The concept of information demand pattern originates from work in the research and
development project Information Logistics for SME (small and medium-sized
enterprises) (infoFLOW). infoFLOW included seven partners from automotive
supplier industries, IT industry and academia. The objectives were to develop a
method for information demand analysis [6] and to identify recurring elements in
information demand, i.e. patterns of information demand. The general idea of
information demand patterns (IDP) is similar to most pattern developments in
computer science: to capture knowledge about proven solutions in order to facilitate
reuse of this knowledge. In this paper, the term information demand pattern is defined
as follows: An information demand pattern addresses a recurring information flow
problem that arises for specific roles and work situations in an enterprise, and
presents a conceptual solution to it.
All information demand patterns are supposed to have a uniform structure, which
consists of a number of essential parts used for describing the pattern:
• The pattern name usually is the name of the role the pattern addresses.
• The organisational context explains where the pattern is useful by identifying
the application domain or the specific departments or functions in an
organisation forming the context for pattern definition.
• The problems of a role that the pattern addresses are identified. The tasks and
responsibilities a certain role has are described in order to identify challenges
and problems, which this role usually faces in the defined context.
• The conceptual solution describes how to solve the addressed problem. This
includes the information demand of the role, which is related to the tasks and
responsibilities and usually consists of different elements; quality criteria for
the different elements of the information demand, like the importance of
accuracy, completeness and timeliness; the timeline indicating the points in
time when the different information demand elements should be available.
• The effects that play in using the proposed solution are described. If the
different elements of the information demand should arrive too late or are not
available at all this might affect the possibility of the role to complete its task
and responsibilities. Information demand patterns include a description of
potential economic consequences; time/efficiency effects; effects on
increasing or reducing the quality of the work results; effects on the
motivation of the role responsible; learning and experience effects; effects
from a customer perspective.
The above parts of a pattern are described in much detail in the textual description
of the pattern. Additionally, a pattern can also be represented as a visual model, e.g. a
kind of enterprise model. This model representation is supposed to support
communication with potential users of the pattern and solution development based on
the pattern. An example for an actual pattern for the role of “Material Specification
Responsible” in a manufacturing enterprise can be found in [4].
2.3 Validation of IDP in Higher Education
As a contribution to increasing the maturity of IDP, a validation activity was
performed in 2010 in the context of higher education [7]. We investigated the use of
information demand patterns in higher education by performing an exercise with 22
students divided into 12 groups in a master course on information logistics. The
intention was to explore whether the IDP structure has reached a level of maturity to
transfer it to actors outside the development team and whether it also can be applied
for capturing information demand in general. The exercise consisted of a practical
task and a questionnaire designed to capture experiences and impressions of the
students. The practical task was to describe the information demand for a task or a
role, where the students felt that they are experts or have at least a lot of experience.
The main result of the validation activity was that the respondents participating in
the exercise managed to produce information demand descriptions using the pattern
structure, which indicates that they understood the structure as such and were able to
apply it. The work also resulted in some proposals for improving the IDP structure,
e.g. by putting more weight on actual information sources, and in improving the way
of teaching information demand modeling, e.g. by spending more efforts on enterprise
modeling aspects such as role modeling. These recommendations were implemented
in the IDP structure and taken into account in teaching activities.
3 Study on Development of IDP Candidate Descriptions
In order to improve the maturity of IDP structure introduced in section 2.2, a two-step
investigation was designed focusing on the “inner” quality of information demand
patterns, i.e. the completeness, accuracy and soundness of the different parts of a
pattern (e.g. context, problem, information demand, effects, etc.). The first step
introduced in this section basically is a repetition of the validation exercise presented
in section 2.3 in a different context: again, students in a university course are asked to
apply the IDP structure for developing IDP candidate descriptions. The intention of
this repetition is twofold: we wanted to confirm the result of the first exercise that the
pattern structure can be applied by non-experts in the IDP field and we wanted to
create a larger sample applicable for the second step of the investigation, which
focuses on the quality of the different parts of the IDP structure and is presented in
section 4. The remainder of this section describes the set-up for the investigation,
shows selected results of the data collection and compares the results with the
previous validation exercise.
3.1 Set-up for Data Collection
The data collection was performed in an exercise within the above mentioned master
course in Riga, Latvia. The information system development course was attended by
18 students. The participating students did their bachelor degree in Riga. As a
preparation for the exercise, the students were introduced in several sessions into the
area of information logistics on basic principles of information logistics, demand
modeling approaches and typical applications. This included an introduction to the
concept of information demand pattern including examples. This initial introduction
into the field of this work was essential to increase the validity of this work.
Afterwards, the students were given an exercise consisting of a practical task and a
questionnaire to be filled in after the practical task. The students had to accomplish
the task on their own, which resulted in 18 solutions handed in. The task was
introduced by the teacher, but there was no further guidance provided during the work
on the exercise.
The practical task was to describe the information demand for a task or a role,
where the students felt that they are experts or have at least a lot of experience. They
were encouraged to consider different areas when deciding on the task or role they
want to describe, including their private or social background, the university context
or any other field. The description of the information demand had to be structured like
an information demand pattern (see section 2.2). A template was provided for this
purpose as well.
The questionnaire included 9 different questions, four of them with a five-point
Likert scale, one with a nominal scale, one with an interval scale and three for free
text answers.
3.2 Data Analysis
All 18 students, which in the following will be called respondents, submitted
information demand descriptions following the pattern template and all filled in the
questionnaire. The information demand descriptions addressed the following subjects:
• Human resources manager
• Client / credit manager
• Employee of logistics department
• Data update responsible
• Automation of scheduled request output saving in project accounting module
• Responsible for warehouse operations
• Responsible for coordination of audit project team
• Responsible for organizing a team building event
• Responsible for cleaning clothes
• Business visionary
• Bus terminal accounting software administrator
• Helpdesk employee
• Website editor
• Responsible for online shop
• Change administrator
• Change administrator for game software
• Administrator for server load balancing
In the questionnaire, the respondents were asked whether the elements in the IDP
structure are needed in order to describe information demand or whether there are
superfluous elements. Sixteen respondents answered that all elements are needed, two
of them responded that one element – timeline – was not needed.
When asked whether anything is missing in the IDP structure, two respondents
demanded: more visualization of information to understand the pattern easier and an
appendix to provide more information about filling in the pattern in detail. The other
sixteen students considered the structure as complete.
Table 1. Response distribution for “How difficult to identify were the different parts?”.
Very Neither Very
Element of difficult difficult difficult easy easy
IDP structure nor easy
Context 1 2 9 4 2
Problem 1 3 10 2 2
Tasks/responsib. 1 2 12 3
Info demand 1 9 7 1
Quality Criteria 3 8 6 1
Timeline 2 3 7 6
Effects 1 9 6 2
When asked “How difficult to identify were the different parts of structure?”, the
following distribution of answers was observed (see table 1). This distribution follows
in most rows the Gaussian distribution. The information demand, quality criteria and
effects are believed to be more difficult than the other elements of the IDP structure.
The question “How difficult to describe were the different parts of structure?”
resulted in the following response distribution (see table 2). Again, most of the
responses follow Gaussian distribution with exception of one aspect. To describe the
effects of receiving information too late or not receiving is considered difficult by half
of the participants. As seen one participant did not answer this question.
Table 2. Response distribution for “How difficult to describe were the different parts?”.
Very Neither Very
Element of difficult difficult difficult easy easy
IDP structure nor easy
Context 3 10 3 1
Problem 5 9 2 1
Tasks/responsib. 1 2 9 5
Info demand 1 5 8 3
Quality Criteria 2 5 7 3
Timeline 3 8 6
Effects 1 7 6 3
Regarding the aspect “How much time was needed to identify and describe the
different parts?”, table 3 shows the distribution of responses. Here, describing the
information demand was considered the most time consuming activity, as 51% of the
respondents answered that “very much” or “much” time was needed. Describing
quality criteria and effect also were considered time consuming by roughly 47% of
the respondents.
Table 3. Response distribution for “How much time was needed for the different parts?”.
Very Neither Very
Element of much much much nor little little
IDP structure little
Context 1 5 9 3
Problem 1 5 9 3
Tasks/responsib. 1 5 9 2
Info demand 1 9 7 1
Quality Criteria 2 8 7 1
Timeline 4 7 5 2
Effects 8 7 1 1
When it comes to the time needed to develop the information demand description,
the field is located close to each other except one participant needing more than 12
hours (see table 4).
Table 4. Response distribution for the overall time needed for completing the exercise.
Time needed Number of responses
less than 2 hours
2 hours – 4 hours 8
4 hours - 6 hours 7
6 hours - 8 hours 2
8 hours - 10 hours
10 hours - 12 hours
More than 12 hours 1
3.3 Discussion
The main conclusion from the performed exercise is that there are clear indicators
to believe that
• the respondents understood concept and structure of information demand
patterns and were able to apply it on their own, and
• the structure proposed for information demand patterns is also suitable for
describing information demand in general.
Both conclusions are supported by the fact that 18 complete information demand
descriptions for different tasks or roles were developed by the participants in the
exercise without guiding or supporting them in the actual development process of
these demand descriptions. This indicates that the students learned how to use the
information demand pattern structure for capturing information demand descriptions.
The pattern structure was judged suitable and complete by the clear majority of the
students. The weight of these observations is relatively low due to the limited
experience level of the students, but nevertheless contributes to the validation of the
IDP concept.
The list of developed information demand descriptions shows a wide bandwidth
from more socially oriented tasks (like “organizing a team building event”) to strictly
business-oriented (“human resources manager”) or IT-oriented roles (e.g.
“administrator for server load balancing”). The quality of these patterns was only
evaluated by the teaching team in the course who checked the consistency between
the different parts of the description, the understandability of the different textual
elements, the completeness of the description, and whether the descriptions were
sound and reasonable. This “perceived” quality of the patterns might be sufficient as
initial check, but would have to be complemented with an “in-use” check of the
description, i.e. applying them in a real-world situation for performing the task/role
under consideration.
With respect to the perceived quality, the impression was that those patterns who
addressed quite small and very specific roles or tasks were the better ones. Examples
are “organizing a team building event” or “responsible for cleaning clothes”.
Furthermore, it was observed that some of the business related demand descriptions
were very detailed and high-lighted interesting challenges. An example is the
“helpdesk employee”, which was reflecting a lot of experience, since the respondent
developing it had several years of job experience on this position from his time before
starting the master education. Thus, the results of the exercise also include some
information demand descriptions, which can be considered as candidates for future
information demand patterns.
The answers from the questionnaire regarding difficulty and time needed to
identify and to describe the different elements of a demand description can be used as
basis when improving the exercise and the course on information logistics. More
emphasis in next year’s teaching should be put on repeating organizational concepts
like “role” and “task” and how to actually describe them in a proper way. This is in
line with the conclusion from the study performed in 2010 [7].
Furthermore, the quality of the information demand descriptions probably could be
improved by several measures. One aspect could be to add more examples for
information demand patterns and perform the stepwise development together with
students in order to improve the understanding of the concept and the different
elements of an IDP. Furthermore, the evaluation of the patterns could be made part of
the course, by providing a guidance for the evaluation process and letting the students
evaluate another group’s demand description.
When evaluating the free form questions of the questionnaire, an interesting
observation surfaced: the students had to learn that they really have to be experts in
the domain in order to be able to describe the information demand. More than half of
the groups indicated that the hardest task in the exercise was to scale down the role or
task under consideration to a scope which really reflected their own area of expertise.
3.4 Conclusions from the first investigation step
The results of the investigation done in Riga have been presented in section 3.2 and
have been discussed in section 3.3. It is clearly to see that most of the achieved results
are very similar to the first investigation done in 2010. The results of the first
investigation are confirmed. Some noticeable differences will be discussed in the
following.
The students in Riga needed less time to accomplish their task filling out the
information demand pattern template than the students from the first investigation
done in 2010. A possible explanation is quite simple because the teacher in the 2010
study accepted the pattern only after having a look at the pattern handed in. Therefore
some pattern descriptions were given back to the students in order to improve the
quality.
Noticeable is as well that information demand and effects were the parts which
were most difficult to identify and to describe. Furthermore information demand and
effects were the most time consuming part of the pattern. A relation between time an
quality can be supposed.
As shown above this study was useful to confirm the result from the first study.
The IDP pattern can be applied by non-experts without further guidance, but was not
suitable to indicate how to improve the pattern quality.
Therefore one more investigation step is needed, which is described in section 4.
4 Quality Assessment of the IDP Candidate Descriptions
As illustrated in section 4 both exercises were performed in order to get indications
about the usability of the IDP concept by non-experts in the field and about where
improvements would be recommendable. Starting from the results presented in 3.3
and 3.4 we decided to investigate in more detail why participants perceive the parts
“information demand” and “effects” as more difficult as the other parts. An
interesting question in this context is, whether the quality of the “information
demand” description and the “effect” description is worse than the quality of the other
parts and the overall pattern, since the respondents consider these two aspects as more
difficult. If this conjecture could be confirmed, more efforts should be spent on
supporting the developers of IDP descriptions with respect to these two aspects:
information demand and effects. This section describes how this second step of our
investigation was performed, what results were achieved and which conclusions to
draw.
4.1 Data Collection Process
As already pointed out there were 12 respondents in the first exercise and 18
respondents in the second. We decided to draw a sample of one third of all
information demand pattern. It is reasonable to consider the ratio of the respondents
from both investigations. Therefore 4 IDP’s should be from the first exercise and 6
from the second one. We numbered the IDP’s from the first investigation from 1 to 12
and the IDP’s from the second from 1 to 18. To draw the random sample we used 2
respectively 3 hexagonal dices.
When evaluating the quality of the ten patterns, we decided to apply and extend
criteria addressing the quality of the pattern content originating from document
engineering [8]. We defined the criteria as follows; “content” refers in this context to
the textual parts in the pattern description:
• Comprehensibility (co): The content is well-structured and formulated
clearly. The reader is able to understand the content easily.
• Completeness (cm): The content includes all information to completely
describe the subject under consideration.
• Consistency (cn): No contradictions within the content or with respect to
other parts of the pattern description can be detected.
• Soundness (so): The given information fits to the subject addressed and
is considered realistic.
• Clarity (cl): The content or the information demand pattern can be
applied without modifications.
In order to evaluate the presented criteria we decided to use marks from 1 to 5
being common practice in German Universities. In this context the marks are used as
follows:
• 1 (excellent quality): The IDP is applicable without restrictions.
• 2 (good quality): The IDP is applicable with minor restrictions.
• 3 (average quality): The IDP is applicable with restrictions and needs
minor revision.
• 4 (limited quality): The IDP is applicable with major restrictions and
needs much revision.
• 5 (unusable quality): The IDP is not applicable.
Every IDP was evaluated according to the above criteria differentiating five
specific parts of an IDP: context and problem, tasks and responsibilities, information
demand, effects, and the pattern in total. This leads to 25 marks per IDP, i.e. five
criteria for each of the five parts. Furthermore, the IDPs were evaluated by two
researchers independently from each other to increase reliability. After the individual
evaluation the results were compared and discussed. There were three differences
about the rating between the investigators. After a discussion the identified reasons
were as follows: two misunderstanding about the content of the IDP and one
difference in the expectation about the results. The investigators were able to clarify
the differences which leads to a joint view presented in section 4.2.
4.2 Results
The results are presented in table 5 and table 6, respectively. Table 5 shows the
results for the “pattern in total” and the part “context and problem”. The results for
the other parts had to be omitted due to space limitations in this work. The missing
parts are tasks and responsibilities, information demand and effects. The ten evaluated
IDPs are named from A to J in order to save space and since the names of the IDPs
are not important for the further analysis.
Table 5 Example results in excerpts
IDP
Pattern
in
total
Context
and
Problem
...
co
cm
cn
so
cl
co
cm
A
2
1
1
1
2
1
1
B
3
5
5
2
5
3
3
C
3
3
3
2
3
3
3
D
3
2
3
1
2
2
2
E
3
4
4
2
3
2
4
...
F
3
4
5
2
4
3
4
G
1
1
1
1
2
1
1
H
2
3
3
1
2
3
2
I
3
3
4
2
4
3
4
J
3
3
3
3
3
2
2
Table 6 shows the overall results about all investigated parts of the IDPs. The
arithmetic average is presented for each rated part of the IDPs.
Table 6 Overall results
Context
Tasks
Pattern
Information
IDP
and
and
Effects
in
total
Demand
Problem
Responsibilities
A
1,4
1,0
1,6
1,0
1,6
B
4,0
3,0
2,0
3,8
3,8
C
2,8
3,0
2,4
1,6
2,8
D
2,2
2,8
2,0
2,0
2,0
E
3,2
3,0
3,0
3,2
2,6
F
3,6
3,4
3,0
3,6
3,8
G
1,2
1,0
1,0
1,0
1,6
H
2,2
2,4
2,8
1,6
1,2
I
3,2
3,4
4,0
1,4
1,4
J
3,0
2,6
5,0
2,4
2,4
4.3 Discussion
The results of investigation did not confirm the conjecture that the quality of the
parts “information demand” and “effect” is worse than the quality of the other parts
and the overall pattern quality. The evaluation results did not show any patterns of
dependencies at all between information demand, effects, context & problem or tasks
& responsibilities. This means it cannot be concluded that a certain quality level of
one pattern part causes a certain quality level for another pattern part.
But the results showed that consistency and soundness got the lowest quality marks
of the five evaluation criteria and that the “tasks & responsibilities” part gets the
worst quality rating among all the parts of an IDP.
Our conclusion from the above results is that we need to support the construction
of IDP in general and the development of the part “task and responsibilities” much
more. Regarding the overall IDP, consistency and soundness between the different
parts need to be improved. This could be done by offering checklists and practices for
interconnecting the different IDP parts. Regarding “tasks & responsibilities”, more
examples and a thorough introduction into the concept of a role and the implications
towards responsibilities from organization theory could be suitable measures.
5 Summary and Future Work
This paper extended work on IDP by addressing the aspect of how to improve their
“inner” quality, i.e. the completeness, accuracy and pertinence of the pattern content.
For this purpose, an investigation was performed with two steps. The first step
involved students in a university course who apply the IDP structure for developing
IDP candidate descriptions. The second step evaluated the IDP developed by the
students with respect to the quality of the different parts of the IDP structure.
The main conclusion from the first step is that the respondents seem to have
understood concept and structure of information demand patterns and were able to
apply it on their own. The students learned how to use the information demand pattern
structure for capturing information demand descriptions. The pattern structure was
judged suitable and complete by the clear majority of the students. The significance of
these observations is relatively low due to the limited experience level of the students,
but nevertheless contributes to the validation of the IDP concept.
The conclusion from the second step is that consistency and soundness got the
lowest quality rating and that the part “tasks & responsibilities” needs improvement.
Such improvements could be reached by checklists and practices for interconnecting
the different IDP parts.
This motivates continuous work into at least three directions:
• A method development effort should be initiated in order to create a
systematic and integrated IDP development method.
• A similar validation effort should be made outside the academic context, i.e. to
transfer the concept of information demand patterns to an industrially oriented
community and evaluate the results of modeling information demand in such a
setting
• In order to validate the IDP concept, the quality of the actual demand
descriptions developed with this concept also has to be evaluated in a more
systematic way than what was done in the exercise described
References
1. Sandkuhl, K.: Information Logistics in Networked Organizations: Selected Concepts and
Applications. Enterprise Information Systems, 9th International Conference, ICEIS 2008.
LNBIP, Springer. (2008)
2. Öhgren, A. & Sandkuhl, K.: Information Overload in Industrial Enterprises - Results of an
Empirical Investigation. Proceedings ECIME 2008, London, UK. (2008)
3. Lundqvist, M.; Sandkuhl, K.; Seigerroth, U. & Stirna, J.: Method Requirements for
Information Demand Analysis. 2nd International Conference on Adaptive Business
Systems, ABS 2008. (2008)
4. Sandkuhl, K.: Supporting Collaborative Engineering with Information Supply Patterns.
Euromicro PDP 2010. IEEE/CS, Pisa, Italy, 2010. (2010)
5. Lundqvist, M.: Information Demand and Use: Improving Information Flow within Small-
scale Business Contexts. Licentiate Thesis, Dept of Computer and Information Science,
Linköping University, Linköping, Sweden, ISSN 0280-7971. (2007)
6. Lundqvist, M.; Holmquist, E.; Sandkuhl, K.; Seigerroth, U. & Strandesjö, J.: Information
Demand Context Modelling for Improved Information Flow: Experiences and Practices. 2.
IFIP Working Conference on Practices of Enterprise Modelling, LNBIP, Springer Verlag.
(2009)
7. Sandkuhl, K.: Validation and Use of Information Demand Patterns in Higher Education. BIS
2010 International Workshops, Berlin, Germany, May 3-5, 2010. LNBIP Volume 57,
Springer. (2010)
8. Arthur, J.D.; Stevens, K.T.: Assessing the adequacy of documentation through document
quality indicators, Software Maintenance, 1989., Proceedings., Conference on , vol., no.,
pp.40-49, 16-19 Oct 1989. doi: 10.1109/ICSM.1989.65192