Smart Learning Objects for Smart Education in Computer Science

The aim of this chapter is twofold. The first aim is to describe the context to ease the understanding of the subsequent topics. Here, by the context, I mean the analysis of research trends in the e-learning and domains (the latter is treated as a very significant branch of e-learning) so that the reader could be able first to understand the essence of the domain and then be aware of the intention of our approach which focuses on two novel concepts, smart learning object and smart educational environment, to teach computer science. The second aim is to present the primary knowledge on those smart items to be considered in detail later throughout the book.

The aim of this chapter is to introduce and discuss a taxonomy-based framework to understand the CS LO domain in large. I motivate the need of such a framework by the following reasons: (1) LO domain is commonly recognized as the heart of e-learning in general; (2) the LO concept is accepted and its role well understood for teaching CS as well; (3) the LO domain is continuously evolving in horizontal (meaning in general) and vertical dimensions (meaning in CS); and (4) a taxonomy-based approach is fundamental in many aspects (knowledge and artefacts systemizing, standardizing, sharing, gaining and teaching). Here, within the introduced framework, we highlight and consider (to some extent only) the following tasks: (1) concept-based modelling and experimentation using a restricted database of literature sources (about 500) and (2) creation of ontology-based models among those concepts that are most likely relevant to our approach.

The aim of this chapter is to discuss the LO reusability aspects to the much broader extent that it was done so far. There are many reasons for that. Let us remind some of them. First, reuse principles are universal and general to be applicable in many fields. Reuse experience taken from the other domains with a higher maturity level (e.g. hardware, software) can be easily transferred and adapted to e-learning. Second, reuse is a very promising approach because of well-defined objectives to design software-oriented educational systems (higher productivity, better quality, shorter time to market). Third, reuse might be seen as a very simple and attractive subject if it is considered at the individual level (say a teacher or learner) as the following paradigms: copy-paste and use-as-is. Fourth, reuse is a very complicated area if it is considered at the organization or cross-organizational levels because there should be taken into account both the technical and non-technical (social, pedagogical, economical, organizational, etc.) reuse aspects. These aspects are extremely broad and complex, indicating on managerial, social, technical and other issues.

In previous chapters (see Chaps. 1 and 2), I have outlined the most general issues of e-learning and CS teaching on the basis of the LO concept. The main focus was given to understanding of the issues through conceptual analysis of the domain literature at the level of LO concepts and their taxonomies. In Chap. 3, I have analysed all these with the emphasis on pedagogical reusability using the software reuse approaches. In this chapter, I provide more in-deep analysis of modelling CS learning and teaching using a systematic approach which is a synthetic product of some domain analysis methods well known in SWE as well as in e-learning domains. In general, the aim of modelling, as it is conceived in the large, for example, in software engineering, is to extract and represent artefacts and knowledge needed to build a software system. As a rule, the extracted artefacts from the domain to be modelled should be represented at a higher level of abstraction. Often we refer to those artefacts as a domain model. Modelling is a primary stage in developing systems.

So far we have discussed smart LOs (SLOs) mainly from two perspectives: (1) conceptual understanding which focuses on definitions (examples) only (Chaps. 3, 4) and (2) model-driven specification with the focus on variability aspects. Though the model-driven view provides the reader with the fundamental knowledge on SLOs, this knowledge was presented at a higher level of abstraction with many details missed. Knowledge representation and knowledge gaining on SLOs are the different processes. As, according to Bloom’s taxonomy, teaching and learning (roughly knowledge gaining) are most effective when learners are involved in doing something, here accordingly we try to involve the reader in the process of constructing SLOs. Therefore, the aim of this chapter is to deliver the next part of the design methodology, i.e. how the SLOs should be coded, tested and redesigned (if needed).

Smart LOs being reusable items in terms of generative capabilities may also offer new opportunities to create individual and highly adaptable content for learning processes. As it was shown in the previous chapters, reusability is a central topic in LO research. However, reusability cannot be generally understood without the educational context. The main goal of reusability is to adapt the teaching content to the context of use in some learning processes. The adaptive aspects of reusability should be discussed from a wider perspective than it was done so far. We need to have a framework enabling to connect reuse issues with the educational context in order we could be able first to specialize and then having the specialized SLO to consider the adaptability problem in some well-defined manner. Therefore, the aim of this chapter is to introduce such a framework and discuss the SLO problem.

We consider adaptation as a bridge to connect generative reuse aspects with the educational context to create opportunities for adaptive personalized learning. To achieve this aim, we have already made an essential move – we have created conditions for that. Indeed the learning variability discussed at the modelling level in Chap. 4 can be seen also as the conceptual background (in terms of creating a space of possible variants) for adaptive personalized learning on the basis of using SLOs. Even more, the specialized SLO considered in the previous chapter is the methodological background for automatic adaptation.

Here, by the designer’s perspective, we mean the representation of the data and processes related to the functioning and design of smart LOs (SLOs) as fully and abstractly as possible. As we use the model-driven approach for designing SLOs, a formal definition and representation of the adequate models play a significant role. Thus, our focus is directed to the precise and complete representation of the SLO models here. The aim is to provide the motivated and sound background of the approach. However, the educational software designers typically tend to work with the informal scenarios of an application domain for its implementation. To resolve this contradiction, we also use informal scenarios (motivating example) to explain the essential details of the approach. Therefore, we hope that different forms of representing the design models at the different level of abstraction are helpful for a variety of designer flavours.

Nowadays the teaching and learning processes are widely supported by the adequate authoring tools. In general, the aim of using the tools is to gain the technological value in the first place, i.e. efficiency, flexibility, etc. (of course, the pedagogical value comes together if the tools are applied properly). Our approach is different in many aspects from those analysed throughout the book. The main distinguishing feature is the realization of the concept of producing and adapting the teaching content automatically. The automation, however, never comes for free. The process of developing smart LOs (SLOs) is the time-consuming and error-prone activity. It requires specific knowledge, competency and some experience of working with meta-programming. Of course, it is possible to write the meta-programming-based SLO specifications manually (by the knowledgeable CS teacher or even by knowledgeable students). Our practice shows that, at the initial phase of adoption of the approach, it is even recommended to apply the manual development. On the other hand, the human efforts are highly dependable on the complexity of SLOs (simply, it might be measured by the number of parameters and their dependency, i.e. model complexity). The more complex SLOs are, the more efforts to develop them are needed. In this case, the use of the adequate tools is highly desirable. Such a situation is with the development of SLOs.

The main distinguishing feature of the smart LO approach is the realization of the concept of producing and adapting the teaching content semi-automatically, or even automatically. The automation never comes for free. On the other hand, the use of the SLO design tool enables us to develop the highly reusable entities. At the development stage, for example, we are able to ensure reusability due to the use of the design paradigm known as design for and design with (see Chap. 9). In this chapter, design with reuse can be technologically interpreted as the SLO problem. As the designed SLO, in fact, is the context-driven meta-specification implementing a wide scale of learning variability, indeed there is a large space for adaptation. In the pure technological sense, the adaptation is a specific transformation process. In the case of using SLO specification, we are able to carry out adaptation through refactoring or specialization (see the discussion on the term issues in Sect. 7.​2).

First, the term ‘smart educational environment’ should be defined. There are standard educational environments that are based on using the Internet-based technology along with some e-learning-oriented systems such as Moodle. In the widest sense, the word ‘environment’ should be understood as the overall technological support (hardware, software and networking with remote terminals) and the infrastructure of the methodological support, including databases or digital libraries with the teaching content, management facilities and teaching instructions (for teachers and students) to support e-learning. The base actors (teachers and students), maintenance facilities and personnel might be also treated as components of the environment. In the narrow sense, by the educational environment, we mean the facilities for functioning e-learning processes to achieving teaching goals within the teaching organization. Using the m-learning paradigm, for example, on the smartphones basis, perhaps, one can treat as being the smart environment too.

Today, computer science (CS) is regarded as a fundamental course (similarly to mathematics, physics, etc.), which is delivered in both universities and schools. Its importance has been recognized far ago because it is a source of the primary and fundamental knowledge needed for our lives and activities, which are highly penetrated by the use of computers, the Internet and other modern technologies. On the other hand, CS can be also seen as an interdisciplinary course, for example, with respect to its relation to robotics and e-learning domains. Furthermore, combining CS topics with the use of robots in learning adequately, it is possible to make a significant contribution to the STEM (science, technology, engineering and mathematics) paradigm, a new interdisciplinary approach to learning and teaching for the twenty-first century. Though we have not considered this paradigm explicitly so far, in fact, by introducing and combining two novel approaches, smart LOs and robot-based smart educational environments, we have paved a way for researching and studying the STEM approach too. But first, we need to show how smart LOs and smart educational environments interact among themselves and to approve this interaction in the real learning and teaching setting.