1 An introduction to sport informatics
Introduction
Over the past three decades, the discipline ‘sport informatics’ – also called ‘computer science in sport’ – has become a growing discipline. In this chapter the historical roots are reconstructed and some reflections on the nature of this new discipline between sport science and informatics are given.
The term ‘sport informatics’ originates from a congress in Graz (Austria), organized by the International Organisation for Sports Information (IOSI) in 1975. The related proceedings were published by Recla and Timmer (1976) with the German title ‘Kreative Sportinformatik’ (‘Creative Sport Informatics’).
Sport informatics covers all activities that include aspects of computer science and sport science, ranging from simple tools for handling data and controlling sensors up to the modelling and simulation of complex sport-related phenomena. Whereas first applications in the seventies used computers for information and documentation purposes only, current approaches deal, for example, with virtual reality in sport, computer technology for supporting top level sports, e-learning in sports training, the modelling and simulation of biomechanical phenomena, and many more.
Today, computer science in sport is a well-established research and development field. The International Association of Computer Science in Sport (IACSS) was founded in 2002 and promotes research in this area. In many countries such as Austria, Croatia, Germany, Turkey, Great Britain, China, Slovakia and India national workgroups have been established, which represent sport informatics in the national scientific community and contribute new technological innovations to sport. IACSS also maintains good relations with various other sport scientific organizations like the International Association for Sports Information (IASI), the International Council of Sport Science and Physical Education (ICSSPE) or the International Sports Engineering Association (ISEA).
The first chapter of this book reflects the development of the discipline, analyses its current situation and defines the subject area. The roots of the discipline of sport informatics lie in Germany in the early seventies and the first section in this chapter will throw a brief glance on the early developments. It aims to give readers an impression of the prevailing ideas leading to IACSS. The second section discusses the interdisciplinary relation between computer science and sport science and identifies different types of cooperation. Based on this discussion the third section defines the subject area of sport informatics by providing a definition and a structural model of the discipline.
Historical roots
It is rewarding to start a historical view on sport informatics with a glance at the historical development of the two master disciplines involved. Computer science and sport science are dealt with in the first two subsections of this chapter with regard to their historical conceptual structure. The third subsection gives a short overview about the institutional development of sport informatics in Germany.
Computer science
In the sixties and the seventies of the last century, in Germany the term ‘Informatik’ was mainly associated with questions of technology. A popular German encyclopaedia described ‘Informatik’ as ‘the science of the systematic processing of information, in particular the automatic processing using digital computers’ (Engesser, 1988). In terms of this definition, the discipline includes mathematical activities, which deal with algorithmic processes for the description and transformation of information and also engineering activities, concerning aspects of the development and application of computers. This technological perspective is comparable to the common understanding of the discipline ‘computer science’ in the United States or Great Britain (National Research Council, 2004).
In the beginning of the eighties, the importance of computer systems increased in almost every part of modern societies. It became more and more clear that the use of computer systems leads to interactions between system processes and the processes in the real world. To study these interactions, many computer scientists adopted approaches and methods from the social and behavioural sciences. These research fields were accepted as a part of the discipline ‘Informatik’. Today, many countries use the English term ‘informatics’ – derived from the German ‘Informatik’ – for the science of information. Nygaard (1986), for example, defines ‘informatics’ as the ‘science that has as its domain information processes and related phenomena in artefacts, society and nature’. This perspective separates the mathematical/logical part from the technical one and refers to the concepts of cybernetics and systems theory.
Informatics emerges by separation from mathematics and engineering science – later approaches from human sciences were integrated. The discipline is divided into the sub-disciplines of theoretical, technical and practical informatics, which together are called ‘core’ informatics (Claus, 1975). The applications and questions related to the use of computers are studied by applied informatics.
Since computer science is very much appreciated for its support to other sciences, in some cases the combination of technical expertise from computer science and specific domain knowledge led to autonomous research fields like bio-informatics, neuro-informatics and business informatics. Sport informatics could also be seen in this tradition.
In the past there was a debate about whether these research fields should be accepted as integral part of computer science. Some authors, for example, claim a strict distinction between cooperation fields and the core area of ‘Informatik’ (Luft, 1992). Today, the discipline in Germany (also known as ‘Informatique’ in France) is a kind of mixture between computer sciences and Nygaard’s concept of informatics (see Figure 1.1), but nevertheless the question about its limitations is still subject to discussion.
Sport science
Sport science took, at least in Germany, a development that was in many aspects comparable to the development of Informatik. A common definition describes sport science as the set of knowledge, theories and research methods that deal with problems and phenomena related to sport (Röthig and Prohl, 2003). While this definition is evident, a widely accepted definition of the term sport is still an open problem.
Another important issue in the discussion about the nature of sport science is the relationship between the disciplines of sport science. In the late sixties, Germany saw the introduction of sport science to universities. Before, it was taught academically mostly in teacher education institutions. For this – more or less – pragmatic reason the argument was put forward that the complexity of sport could not be investigated by existing research fields. So, the necessity of one unified discipline, with a high degree of interdisciplinarity between its sub-disciplines, was a central argument for the foundation of sport science.
To support this position, Ries and Kriesi (1974) proposed a model showing three phases of the development of sport science: (1) separation from basic disciplines, (2) aggregation of sub-disciplines within a multidisciplinary science and (3) integration of sub-disciplines into a consistent and integrative science (Figure 1.2).
Scientific reality showed that – in contrast to this idealized model – sport was mostly studied through the eyes of each sub-discipline (e.g. sport sociology, sport psychology, exercise science). Due to this fact, today sport science does not describe itself as a ‘unified science of sport’, but as a collection of overlapping research programmes in which interdisciplinarity exists only as temporary, problem centred research projects (Höner, 2001).
Sport informatics
The idea of an interdisciplinary scientific discipline ‘sport informatics’ was promoted initially by Jürgen Perl, himself being a mathematician and a pioneering computer scientist in Germany. Together with Wolf Miethling he published the first monograph in the discipline (Miethling and Perl, 1981) that marked the beginning of sport informatics in the Federal Republic of Germany.
In 1985, Jürgen Perl founded the Institute for Informatics at Mainz University and established a working group in sport informatics. His idea to organize a first workshop on sport informatics in Hochheim (close to Mainz) in April 1989 gave rise to a series of workshops on this topic. The 1989 workshop was attended by many important German sport science groups and, unexpectedly, by some computer science groups in this field as well. This development starting in 1989 resulted in a series of biennial conferences.
This apparent success gave rise to a new strategic aim, which has been pursued since about 1994. The German Association of Sport Science (Deutsche Vereinigung für Sportwissenschaft, dvs) represents German academic sport science with (today) 900 members at sixty-seven universities. It is organized in subgroups (Sektionen) representing the disciplines of sport science and groups (Kommissionen) giving an organizational framework for special interdisciplinary topics. In September 1995, the general assembly of dvs established sport informatics as one of their official subgroups. This may be considered the formal birth date of the scientific discipline, because it meant the acknowledgement of sport informatics as discipline of sport science in Germany.
Since then the biennial conferences of sport informatics in Germany have brought together the two traditions: the most recent workshop at Konstanz, Lake Constance, in 2012, was the 9th dvs Symposium of Sport Informatics as well as the 13th workshop on Informatics in Sport.
Soon after establishing a national association for sport informatics a new aim was targeted, i.e. to establish an international scientific association. Concerning globalization and the world becoming a truly global village through advances in information and communication technologies, it became clear that in different parts of this village, people were addressing the same problems.
In the area of game analysis one could, for example, mention the introduction of digital boards, efforts to enter data by natural language recognition software or the struggle for reliable computer-video couplings. These developments were brought forward independently, for example, at Mainz and Cardiff in the working groups of Jürgen Perl and Mike Hughes, respectively.
After three international meetings at Cologne (1997), Vienna (1999) and Cardiff (2001), the International Association of Computer Science in Sport (IACSS) was founded at Barcelona in 2003 and Jürgen Perl became the first president (Figure 1.3). Since then, a series of biennial international conferences has been organized (Hvar, 2005; Calgary, 2007; Canberra, 2009; Shanghai, 2011; and Istanbul, 2013) and members from different countries and almost all continents have joined the association.
The future prospects of the association are excellent. The unique combination of sport science and informatics with the large application field of sports at any level provides great perspectives. Nevertheless, the remarkable developments of the two sciences make it necessary to reflect episodically on the levels achieved in interdisciplinary cooperation between the fields and the concept of sport informatics.
Interdisciplinarity in sport informatics
First, this section outlines the mutual interests in cooperation of computer science and sport science in common projects. While the motive of sport science is quite obvious, that of computer science needs more elaborate discussion. The second part poses the questions: which quality of interdisciplinarity between sport science and computer science exists today and which quality would be desirable and realistic in future? This is done by discussing existing models of interdisciplinarity and proposing a classification for research activities in sport informatics.
Common fields of interests – why do computer science and sport science cooperate?
It is useful to differentiate between political, scientific and personal motivations for cooperation. From a political perspective one must bear in mind that interdisciplinarity is considered an important research paradigm in most countries. For example, the German Research Foundation (DFG), which is the central research funding organization in Germany, holds the view that scientific progress arises more and more at the borders and intersections of disciplines. In the same vein, the national funding agency for scientific research in sport (German Federal Institute of Sport Science) names interdisciplinarity a ‘key element’ of its funding policy. Announcements on funding initiatives refer to inter- and multidisciplinary approaches, integrated construction of theories, highly specialized choice of research methods and integrative presentation of results. While the precise meaning of such catchphrases is somewhat clouded in jargon, a scientist whose career depends on positive evaluation (and funding) of his research projects is ill-advised to refuse the commitment to interdisciplinarity.
Besides political considerations, sound scientific justification for cooperation does exist (see Figure 1.4). First of all – from a sport science perspective – computer science services are well appreciated in specific technological areas in which sport scientists are not necessarily experts. This applies to data handling and software development, e.g. for the purpose of training documentation, controlling sensors or visualizing data. Second, information technology is an important source of innovations for training and competition. Collaborations with computer science help sport scientists to become aware of new technologies that may lead to improvements in supporting sport activities (Stöckl and Lames, 2011). Third, sport science expects that the approaches and perspectives of computer science should be transferable to the field of sport. For example, the concept of soft computing can assist the understanding of phenomena in sport (see ‘Subject area’).
Seen on a broader perspective, there are also benefits of cooperation with sport science for computer science. Here we face the situation that due to the specific properties of sport science, a long-term benefit is imaginable for computer science.
A special feature of sports that may be attractive to computer science is its degree of complexity. The existing structures in sports are neither too simple to be of interest, nor too complex to be described using mathematical models.
A test field for intention detection, a well-established field of computer science, could well be the analysis of game sports. Here, we have specific action plans that pose challenges to computational requirements (automatic recognition of players, moves and strategies) albeit with reduced complexity (limited degrees of freedom, common rules, tactical invariants) compared to similar problems. More generally speaking, sport could act as an attractive testing field for computer science, in which human behaviour can be observed and studied in simplified, yet authentic settings.
Another motive for computer scientists to engage in sport sciences might be public interest in sport and its huge role in mass media. This may give rise not only to the Basking in Reflected Glory (BIRG) phenomenon (Cialdini et al., 1976), but doing studies in football, for example, releases researchers from the sometimes tedious task of explaining the rules of the domain.
Last but not least, many computer scientists working with sport science are personally involved in sport. Even if collaboration cannot be fully justified on the basis of individual involvement, political considerations and increasing publicity seem to have importance as secondary motives.
Quality of interdisciplinarity – how do computer science and sport science work (or how should they work) together?
There are many ways in which the concept of ‘interdisciplinarity’ has been classified by the philosophy of science. One milestone in categorization was a congress in the year 1972, where the Organisation of Economic Co-operation and Development (OECD) proposed a classification of interactions between disciplines (OECD, 1972). In terms of this definition, multidisciplinarity is a juxtaposition of various disciplines without a connection between them. Interdisciplinarity describes any interaction between disciplines, which can range from simple communication of ideas to the integration of concepts, methodologies and epistemologies. Transdisciplinarity is the highest degree of cooperation and stands for a common set of theories and axioms for a set of disciplines (Figure 1.5). On this basis, enhanced models focusing on different aspects of interaction were developed, e.g. by Heckhausen (1972), Boisot (1972) and Karlqvist (1999).
Unfortunately, a closer look at the practice of sport informatics shows that none of these models is adequate to describe the existing interaction. In this regard, we propose our own classification, using four types of cooperation (see Figure 1.6):
• Type a: Sport science applies existing approaches and tools from computer science. In this case, sport science does not take part in conceptualization and development. Computer science (or – mostly – commercial software developing companies) only acts as anonymous service provider, without contact to sport science.
• Type b: Sport science integrates knowledge from computer science. This happens when sport science needs technical solutions not existing in the market. Knowledge is assimilated either by acquiring the skills necessary or by entering into partnerships with computer science, e.g. by means of students or funded projects. One aspect of this cooperation is that computer science provides nothing but skills in software development. There is no collaboration on a scientific level.
• Type c: Computer and sport science cooperate in research programmes, which are in accordance with the research interest of both disciplines. Examples are the use of artificial neuronal networks for analysing movement patterns or application of image recognition algorithms in sport game analysis. In this case, computer science gets new insights by validating concepts and methods which have relevance for additional – perhaps more complex – problems. Sport science benefits from an improved and faster data acquisition and by getting a different perspective on the structures of sport.
• Type d: This type is comparable to type c with the small but important difference that paradigms and knowledge of sport science are used in computer science. An example would be the use of kinesiological models in controlling the motion of humanoid robots.
‘Borrowing’ methods of computer science (types a, b, c in Figure 1.6) matches to Heckhausen’s concept of auxiliary-interdisciplinarity. The simple usage of pre-defined tools (type a) corresponds to the OECD term multidiscplinarity. The corporate development of tools/methods (type b) can be called pseudo-interdisciplinarity (Heckhausen) or restrictive interdisciplinarity (Boisot). The use of sport scientific knowledge in computer science (type d) accords with the idea of structural interdisciplinarity (Boisot).
A review of the research activities in the last twenty years reveals that many projects are of types a and b, but only very few projects of types c and d can be found. One reason why the popularity of sport informatics in the computer scientific community is not very high (there are computer scientists who never heard about it or do not show interest in any cooperation with sport science) might be that there is often no genuine interdisciplinary research. A deeper concentration on those fields where computer science can profit from sport scientific paradigms and knowledge (types c and d), could improve the situation. This would require better communication of sport scientific expertise and recognition of sport as a fruitful application field for computer scientists.
On the other hand, we must acknowledge that there is basically only low affinity between the disciplines of sport science and computer science. In contrast to other interdisciplinary linkages (like biology and chemistry, astronomy and physics or sociology and psychology), there is no common borderline with common issues and hardly any shared knowledge. Consequently, sport informatics has not yet been established as autonomous interdiscipline like astrophysics or biochemistry.
One-way transfer is also frequently found in other application fields of computer science. Basically, this results from the fact that information processing is fundamental for all sciences, whereas application fields of computer science usually cannot provide any knowledge for the core area of computer science (with exceptions like mathematics and electrical engineering). Moreover, the problems of sport science and computer science in creating real interdisciplinarity within their own sub-disciplines show that creating interdisciplinarity is far from being trivial. They are both heterogeneous sciences without a consistent level of theoretical integration, axioms, methods and terminology (see discussion in ‘Interdisciplinarity in sport informatics’).
What is the final conclusion of these considerations on interdisciplinarity between informatics and sport science? One might advise sport informatics to continue postulating and advancing interdisciplinarity, but also not to overemphasize the idea of integration.
Subject area
Bearing the discussion of the previous sections in mind, we suggest differentiating between sport informatics and computer science in sport. Computer science in sport stands exclusively for the use of computer technology in sport and sport science. Sport informatics also includes the application of methods and paradigms from computer/information science as well as from research programmes, which try to transfer sport scientific knowledge to computer sciences.1 The following definition shows this enhanced self-concept:
Sport informatics is a set of multi- and interdisciplinary research programmes which contain parts of sport science and computer science. The subject area is the application of tools, methods and paradigms from computer science on questions of sport science as well as the integration of sport scientific knowledge in computer science.
Figure 1.7 represents this standpoint: in both disciplines there is knowledge that is potentially useful for the other discipline. Conversely there is a second area, which might be an application field for the findings of the other discipline. The research programmes of sport informatics include parts of both disciplines and can be dedicated to one of the four types, identified in the last section.
The discipline can be described as a set of multi- and interdisciplinary research programmes. Most of these programmes apply the technological and methodological knowledge of computer science to study questions of sport science, but there are also some sport scientific findings, which can be useful for computer science.
Figure 1.8 shows a refinement of this rough structure by using a matrix with four areas. The upper areas give examples for research fields for computer science which are useful for sport and sport science; the lower areas give examples of how computer science can profit from sport science. The next subsections discuss the two different directions of integration.
Computer science in sport science
The table in Figure 1.8 (top left) shows topics in computer science, which may be useful for sport science. According to Perl and Lames (1995) the columns of this table are an open list of research areas of computer science important to sport: (1) acquiring and storing of data, (2) modelling, analysis and simulation, (3) presentation and visualization and (4) communication and networks. The rows of the table illustrate the idea that the research field sport informatics is (or should be) more than just the simple application of tools for the recording, analysis and presentation of data. In addition to the ‘tool level’ (which is more information technology than computer science), there are also methods, theories and paradigms, which have the potential to support sport science (see cooperation types c and d in the last section). The next paragraphs discuss some examples.
In the field of capturing and storing data, information technology provides, for example, database tools, which can be used for the storage of training and competition data (see Chapter 3 ‘Databases and expert systems’). This allows coaches to stay informed about the input load and the performance level of their athletes. Also, many biomechanical devices like force plates and sensors, high speed cameras or laser radars use proprietary software tools, controlling sensors and helping to manage and present measured data (Godbout and Boyd, 2012; see Chapter 4 ‘Data acquisition and processing’). In this context, systems for measuring the movements of players on a court based on computer vision or active sensors became more important (Gomez et al., 2012; Baca et al. 2011; see Chapter 5 ‘Motion tracking and analysis systems’).
For the modelling, analysis and simulation, sport scientists often make use of mathematical software like Matlab, Maple, SPSS or apply software for movement analysis. Sport engineers rely on tools for computer aided design (e.g. AutoCAD or Solid Edge) to develop new sports equipment.
Tools from information technology are also important for the presentation and visualization of data. In sport games for example, the linking of video sequences and databases is useful for game observation (see Chapter 7 ‘Game analysis’). In athletics or gymnastics, coaches use software that enables the showing of additional information like force vectors, torsional movements or speed information on a video or superimposing pictures of two movements. Last but not least communication technology can help to organize training (Lyons, 2011): today Internet blogs and social networking websites are relevant for organizing training sessions and finding partners for game sports. Internet-based groupware tools like synchronous video, video conference and whiteboards support athletes on international championships, by enabling communication with their coach at home (Link and Lames, 2005). In this context mobile devices for coaching become increasingly important (Hummel et al., 2013).
On the method level, computer science, for example, has developed techniques for image recognition, which can be helpful to capture positional and biomechanical parameters directly from video recordings (see Chapter 5). Therefore computer science does not provide readymade tools, but rather general algorithms for colour, texture and shape comparison, which have to be adapted to the specific sport context (see cooperation types b and c in the last section). In the field of analysis and simulation of sports, the methods of soft computing have become more important (see Chapter 6): artificial neuronal networks are used to simulate the relationship between training input and performance output (Perl and Endler, 2012), to analyse tactical or movement patterns (Perl, Grunz and Memmert, 2013; Lamb, Bartlett and Robins, 2010). Genetic algorithms help to find solutions in high dimensional configuration spaces, e.g. to optimize the design for sport equipment (Vajna et al., 2006) or to optimize throwing movements (Bächle, 2003). Last but not least, serious games can have a positive effect on perception, reaction and motor control and are potentially useful for education and intervention in sports (see Chapter 8).
At the theory level the approach of complex dynamic systems is an example that holds a lot of promise for sport science. Many processes in sport seem to rely on non-linear coupling rules, which lead to complex phenomena. The theory of complex systems helps to model and to understand, for example, interaction in sport games as well as the processes of biological adaptation due to training. Successful examples for using concepts from system theory in sport science are perturbations (Hughes, Dawkins and David, 2000), relative phase (Walter et al., 2007), chaos theory (Lames, 1999), and the paradigm of self-organization (McGarry et al., 2002).
The second part of the matrix (see Figure 1.8 top right) shows examples for application fields in sport and sport science. These fields are structured with the headlines ‘theory building’ (getting new theoretical insights into phenomena of sports), ‘intervention’ (improving training and competition) and ‘organization’ (managing activities related to sport). Examples for these application fields have already been mentioned in this section.
Sport science in computer science
The third and fourth parts (see Figure 1.8 bottom left and right) show examples for sport scientific knowledge fields which are potentially useful for research in computer science. We give three examples for clarification.
The first example is software for the safe and autonomous operation of robots. Traditionally the algorithms for navigation, locomotion and the grasping of objects have been based on two different concepts: planning and controlling (Latombe, 1991). Planning methods define the movements and the position of the joints at any time before the movement. This requires complete information about the entire environment and the objects to be manipulated. On the other hand, controlling methods rely on local status information during the movement, based on visual or force feedback. This allows reacting to unexpected events like obstacles, but without global information, it is not guaranteed that the algorithm finds a (the best) solution for a task.
In kinesiology it is known that human motor control does not follow the planning or controlling paradigm, but it would be better described as a self-organizing process, influenced by both aspects. For example, top level dart players show a substantial variability in velocity, joint angles and the sequential timing of body parts from trial to trial (Müller and Sternad, 2004). The presumably most important skill for elite players is to balance parameters during the movement and not to reproduce fixed motor programmes. If sport science could understand this self-organizing process in detail, these findings could also be used in the development of new paradigms for controlling robots.
The second example is cooperation in sport games. An important factor for success in soccer is the quality of interaction between the players of a team. Interaction is needed on the level of the entire team, e.g. by shifting the team formation depending on the tactics and the position of the ball, as well as on the level of subgroups, for example when a striker starts running to receive a pass before the ball was played. Some of these interactions are practiced in training; others spontaneously arise based on the situation. The conditions for successful or non-successful interaction between players (e.g. which are components of decision making in soccer, which agreements are needed for organizing the defence, which cues are used for the timing of a pass?) are of great interest for exercise science and sport psychology. If models could be developed on how cooperation in sport works in detail, these would be valuable results for computer scientific research in fields like intelligent autonomous systems.
A third example can be found in the field of mobile computing. Mobile computing – which means the use of computers during movement – is a fast growing application field for information technology. Examples are the use of handhelds in medicine (documentation of patient records), in the military (geographical information for foot soldiers) and in sport (feedback about physiological parameters). One important aspect is that mobile computers extend demands on the user’s coordination and cognition (Kjeldskov and Stage, 2004). While running or walking, the user has to adjust the movements of the legs with the movements of the hand–arm system. On a cognitive level the user must pay his or her main attention to their forward locomotion and, at the same time, he or she has to look on the screen to coordinate hand movements. Up to now, our knowledge about the interrelationship between walking speed, heart rate, user input, reading performance and interface design has been quite vague. Experiences and research methods from kinesiology and biomechanics can help computer science to develop user friendly human–computer interfaces, which are usable even under physiological stress.
Conclusion
Sport informatics is a reasonable and fruitful liaison between sport science and computer science. Common projects hold a set of advantages for both disciplines, if projects are designed and performed with the objective of genuine interdisciplinary research. As scientific progress in this area is closely connected to technological progress, sport science is well advised to monitor developments and to integrate partners from computer science into its research activities. On the other hand, computer science finds an appropriate application field for testing its methods and also movement-related knowledge which might help to solve its problems. Certainly, it is an important future task for the scientific community in sport informatics to make sure that this potential will be utilized as exhaustively and successfully as possible.
1 What is the definition of sport informatics? Explain the ideas behind this term.
2 You want to establish studies in the field of sport informatics at your university. Which arguments could you put forward?
3 Discuss the concept of interdisciplinarity in sport informatics. Why is it so difficult to achieve high levels of interdisciplinarity?
4 Explain why we suggest differentiating between sport informatics and computer science in sport.
5 Describe and explain the structural model of sport informatics. Give examples for the two directions of collaboration.
Notes
1 The International Association of Computer Science in Sport (IACSS) decided to choose ‘Computer Science in Sport’ instead of ‘Sport Informatics’ for its name, because the term is much better known and easier to understand in most countries.
Recommended further reading
Baca, A. (2006) ‘Computer science in sport: An overview of history, present fields and future applications (Part I)’, International Journal of Computer Science in Sport, 5(special edn 2): 25–35.
Fischer, G. (1998) ‘Transcending cultures – Creating a shared understanding between computer science and sport’, in J. Mester and J. Perl (eds), Sport und Informatik V (pp. 43–52), Köln: Strauss.
Link, D. and Lames, M. (2010) ‘Sport informatics – History, current structure and future perspectives’, International Journal of Computer Science in Sport, 8(2): 68–87.
References
Baca, A., Leser, R. and Ogris, G. (2011) ‘Local Positioning Systems in (game) sports’, Sensors, 11(10): 9778–9797.
Bächle, F. (2003) ‘The optimisation of throwing movements with evolutionary algorithms on the basis of multi-body systems’, International Journal of Computer Science in Sport, 2(special edn 1): 6–11.
Boisot, M. (1972) ‘Discipline and interdisciplinarity’, in OECD (ed.) Interdisciplinarity: Problems of teaching and research in universities (pp. 89–97), Centre for Educational Research and Innovation (CERI), Paris: OECD Publications.
Claus, V. (1975) Einführung in die Informatik [Int], Stuttgart: Teubner.