Economics and Software Engineering: Transdisciplinary Issues in Research and Education

Transcription

1 Economics and Software Engineering: Transdisciplinary Issues in Research and Education Teresa Tharp Valencia Community College 1800 Denn John Lane Kissimmee, FL 34744, USA Janusz Zalewski Computer Engineering University of Central Florida Orlando, FL , USA Abstract This paper presents an approach to combine efforts in Economics and Software Engineering, in order to treat software development from the economics perspective. We define the term transdisciplinary, in the context of multidisciplinary and interdisciplinary research, and use it to derive mutual relationships between Economics and Software Engineering. Finally, we propose a research agenda for connecting both disciplines and enhancements in curricula to include topics of mutual importance. when the method of research is shared by multiple (two or more) disciplines. This is illustrated in Fig 1. Disciplines A and B conduct multidisciplinary research, and disciplines A and Z conduct interdisciplinary research. A similar model has been presented by Mahdjoubi, Introduction In engineering disciplines, the concept of transdisciplinary research and education has been mostly explored in design and process science by Ertas et al., 2000, in relation to their earlier work on multidisciplinary (Jones et al., 1995) and interdisciplinary (Tanik and Ertas, 1997) research. The meaning of transdisciplinary research and education, as used in the literature, is not necessarily very clear. Multiple and sometimes conflicting definitions do not contribute to good understanding of the area. So, let us first define the meaning of the basic terms as understood throughout this paper. Following Nicolescu, 1997, we understand multidisciplinary research, as the situation when the subject of research is shared by multiple disciplines. Also following the same author, by interdisciplinary research we mean the situation Fig. 1. Illustration of multi- (sharing subjects) vs interdisciplinary research (sharing methods). Transdisciplinary research includes both concepts mentioned above, but with a significant extension. By transdisciplinary research, we mean a situation when two or more methods are mutually shared by relevant disciplines. For example, method A from discipline A is adopted by discipline Z, and at the same time (in the same research) method Z from discipline Z is adopted by discipline A, to study the same subject, as illustrated in Fig. 2. This is mutual adoption or exchange of methods and models. An essential component of the meaning of the term transdisciplinary research is "mutual".

2 More generally speaking, transdisciplinary research pertains to the crossing of fields (for example, those which are the subject of this study: computing and economics; software engineering and economics, etc.) in such a way that both disciplines benefit from each other. Once we have resolved the exact meaning of "transdisciplinary research", then we'll be able to talk about transdisciplinary education. Fig. 2 Illustration of transdisciplinary reasearch. The rest of this paper is organized as follows. First, we present mutual relationships between economics and software engineering, based on the above understanding of transdisciplinary research. Then we give a brief background of important elements in both disciplines, and finally present an agenda for joint research, involving both disciplines, and their curriculum enhancement. - Mechanics, whose goal is discovering principles of material objects, feeds Mechanical Engineering, which applies these principles to construct and build houses, factories, roads, bridges, machinery, cars, airplanes, ships and other vehicles - Physics, which focuses on discovering principles of matter at the submolecular level, feeds Electrical Engineering, which applies these principles to construct power plants, appliances, radio stations, radio receivers, telephones, TV sets, etc. - Biology, which studies living organisms, feeds Bioengineering, which applies these discoveries to construct artificial organs, etc. - Computer Science, which deals with computations, feeds Computer Engineering, which applies these discoveries to construct computers, but also feeds Software Engineering, which does apply respective principles to construct software. In this perspective, we may ask a question: what would be the "extension" of Economics, so one could say: Economics, a science which studies production and distribution of goods feeds such knowledge, which applies these principles in a free market economy. We are looking for an applied discipline, extending economics. If we draw a simple diagram that illustrates the role of applied disciplines growing out of basic research, one could see Business as a discipline descendant of Economics, where respective economic rules are applied (Fig. 3). 2. Software Engineering and Economics: Transdisciplinary Relation To study transdisciplinary relations between Economics and Software Engineering, we have to find common grounds between the two disciplines. Sciences, whether natural or social ones, always deal with discovery, but they spawn applied disciplines, which deal with construction. Every science has its own way of applying discoveries into practice, for example, in the technical fields: - Chemistry, which deals with discovering properties of matter, feeds Chemical Engineering, which applies these principles to construct chemical plants Fig. 3. Relationship between basic disciplines and their applied counterparts. Every level of Fig. 3 needs some explanation. The high-level relationship, depicted on this diagram as a horizontal line between Computer Science and Economics, involves sharing scientific methods by both disciplines, one example of which could be game theory. This

3 level of relation, however, is out of the scope of this paper. What we are mostly concerned about is direct relations between Economics and Software Engineering, which involve: - those methods and models which are transferred to and from Economics as a science (depicted by a diagonal connection in Fig. 3), and - those techniques and tools, which are transferred to Software Engineering from the Business discipline as an applied Economics (shown in Fig. 3 as a horizontal line connecting both disciplines). The former include principles and methods, such as economic laws and equations. The latter include subjects, such as job market, investments, labor, etc. These relationships may occur at many levels, for example: - at a software company level, where software engineers need to be prepared to deal with economic and business issues - at the software industry level, where there are rules to be obeyed by companies from the economics and business perspective. In general, mutual relations between economics and software engineering have been studied long ago, beginning with the seminal work of B. Boehm, 1981, and shortly following it the work of W.L. Frank, Relevant issues are summarized in Marciniak, This has not been done, however, in the context of framework of transdisciplinary research. To discover more fundamental principles relating the two disciplines, we present in the next section a brief background on both. 3. Basic Issues 3.1 Economics. Economics is a social science concerned with the study of how society chooses to use and allocate scarce resources to satisfy its unlimited wants. It analyzes the costs and benefits of improving patterns of resource allocation. Of interest to economics are topics such as: production, distribution and exchange, and consumption. This is briefly illustrated on a block diagram in Fig. 4. Focusing on production, there are three fundamental questions economics, and every society, must answer: 1. What products will be produced? 2. For whom will they be produced? 3. How will they be produced? These questions lead us to the structural components of a diagram as follows. By asking the first question, we imply that there is a data flow in the diagram. This could be, of course, material flow, flow of ideas, etc., which is represented by outcoming arrows. The second question implies existence of other structural elements (blocks), such as Consumption, that interact with Production. Finally, the third question implies that each block will have some internal sequence of operations, which constitute its overall activity. The latter is illustrated on a diagram by an example of Distribution. Fig. 4. Illustration of basic economic notions. Having presented the basic diagram, let s look at each of these questions in some more details. What to produce? If we are producing something, for example, software, whoever is doing so must take into consideration their available resources, examples of which are: land, labor, capital (i.e. physical plants, machinery, equipment), entrepreneurial ability. Due to the limited availability of resources, the best alternative allocation of their use determines what is to be produced; which in turn assist us on deciding which product to make. For whom to produce? Among all those desiring the product, who actually receives it? This means that in society there must be a method to decide who will be able to afford the product and who will not. It is in this area where the price system comes into play. The price system is a mechanism that uses the forces of supply and demand to create an equilibrium through rising and falling prices.

4 How to produce? In economics this question refers to how to mix technology and scarce resources in order to produce goods or services. In this area a firm/industry must decide which production technique to use in order to make their products. They must answer the question whether a production technique will be more or less capital-intensive, labor-intensive, etc. Understanding the basic activities and flow of data, we can now concentrate on active elements, participants in the economy. At the microeconomics level, the individual units to consider are represented by households and businesses, possibly involving the government activities, as illustrated in Fig. 5. From the macroeconomics standpoint, we would have to include definitely foreign markets, as well. Fig. 5. Individual economic units. A household consists of the occupants of a housing (economic) unit such as an apartment or a house. Their participation is important because of the projection of consumer demand and spending in goods and services (such as, software programs). A business or a firm is an organization designed to purchase and bring together inputs in order to produce goods and services. Includes sole proprietorship, partnership, or corporation. A meaningful number of privately owned manufacturing or service businesses in a common area are called an industry. Microeconomics focuses on analyzing economic variables such as the prices and outputs of specific firms and industries, the expenditures of consumers or households, wages, rates, competition, and markets. Some of the questions which could be addressed by microeconomics include: * What determines the price and output of individual goods and services? * What are the factors that determine supply and demand of a particular good? * How do government policies such as price control, subsidies, and excise taxes affect the price and output levels of individual markets? The above is a generalization of microeconomics for the purpose of this paper. Many more detailed questions are addressed in textbooks, such as Tucker, Software Engineering. Software Engineering is the application of a systematic, disciplined, quantifiable approach to the development, operation, and maintenance of software (Marciniak, 1994). The development process is usually described in terms of individual activities or phases, the sequence of which leads to the development of a final products. Fig. 6 illustrates the software development process in a simplified diagram. The core activities involve requirements specification, design, implementation, testing, and installation and operation embraced by project management, during the development process, and by maintenance after the development. Multiple other activities are conducted, not listed on the diagram, such as: configuration management, quality control, verification and validation, training, etc. The development process is usually conducted by software engineers at the technical level only. Questions asked are at best related to the precise methods of specifying requirements, optimal methods of designing software components which meet these requirements, best programming languages and operating systems, as well as respective practices to code the designs, and appropriate testing methods to prove that software has as few defects as possible. At the project management level, questions asked are more general. For example, a manager is always concerned about issues such as developing software products on schedule, meeting the budget, personnel qualifications and fluctuations, tool support, etc.

5 which is defined by the research relationships between Economics and Software Engineering. The primary functions of economists are to conduct research and analyze the economic phenomena, and to formulate policies. Economists apply statistics to be used to test the validity of economic theory and to measure the responsiveness of economic variables, which may affect either positively and/or negatively the demand for software. Thus, economists may work hand-in-hand with software engineers on software development by utilizing economic theory regarding specifics of software development. Fig. 6. Simplified view of the software development process. However, even at the management level, questions such as the following are hardly asked: - What is the market demand for this software product? - Who are the potential customers and what is their user profile? - What is the current job market for specialists fluent in this type of software development? - What is the turnaround rate of employees in this particular area? Questions like this are addressed very rarely, if at all, at the project management level. This does not mean, however, that software engineers or project managers should not be aware of respective issues. 4. Where Do Software Engineering and Economics Meet? In a search for more specific mutual relations between Software Engineering and Economics, we have to take now a look back at Fig. 3 and ask a question, how would we apply economics methods to the production process of software? From Fig. 3, it becomes clear that there are two basic paths to Software Engineering: direct path from Economics, and indirect path through Business. Let s first analyze the direct path, Another mean in which software engineers and economists could converge is in the use of successful development of software using econometrics. Econometrics is economic measurement and deals with the quantitative evaluation and analysis of actual economic and business phenomena. Econometrics attempts to quantify economic reality and bridge the gap between the worlds of economic theory and actual business activities, such as software development. One other way of bringing the two disciplines together is to use optimization, for example, to determine optimal allocation of resources in production processes to meet the prescribed goals of consumption. This would be particularly useful in the development of single quantities of software, a case not rare in a real world, especially in software development for the government, but ignored thus far in economics. Economic theories, such as the above, can be applied to quantify various resources to deliver a product: manpower, tools, budget, timescale, etc. In particular, it could be assessed, what are the potential risks related to developing a particular product or using a particular development process. Issues mentioned above connect the two disciplines at the research level. At the practical level, related to the operation of Business, Economics and Software Engineering can mutually assist each other in studying issues such as those: - software as a product - software production process - job market for software engineers - software industry.

6 As a sample framework for such studies, we can consider placing the software development process in the context of economic processes, which is illustrated in Fig. 7. A box marked Engineering embodies both Development and Production of Software, which may be considered separately. Economic processes, as sketched roughly in Fig. 4, are split here into two, Marketing and Distribution, while another software process is added, called Maintenance. Transdisciplinary education pertains to the crossing of both fields and would involve introducing, into the respective curricula, the notions of mutual importance. Examples of those would be a concept of markets and related issues introduced to the Software Engineering education, and a concept of software process introduced into the Economics curriculum. 6. References Boehm B.W., 1981, Software Engineering Economics, Prentice Hall, Englewood Cliffs, NJ Ertas A., Tanik M.M., Maxwell T.T, 2000, Transdisciplinary Engineering Education and Research Model, Transactions of the SDPS, Vol. 4, No. 4, pp Fig. 7. Software development process (shown as engineering) as a part of an economic process. With these processes in the picture (adding more, if necessary, and connecting them properly), we can define variables describing items exchanged between the blocks representing various activities/processes. Such variables should include those, which correspond to basic economic indicators, including: demand, investment, staffing, tools, services, salaries, spending, prices, etc. Having the variables, their flow and ranges defined, we are able to apply economic principles in modeling the software development process by a computer. 5. Conclusion We presented a framework for transdisciplinary research in Economics and Software Engineering, which can be applied in education for the benefit of both disciplines. The primary relationships between these disciplines involve: - direct use of mutually complementary research methods, such as assessment of economic variables, use of econometric measures, and optimization, applied to software development at the company and at the software industry level; - indirect application of economic principles in business practice, by modeling the software development process in the context of economic processes and activities. Frank W.L., 1983, Critical Issues in Software: A Guide to Software Economics, Strategy and Profitability, John Wiley and Sons, New York Mahdjoubi D., Interdisciplinary Study on Knowledge and Innovation Strategy, 2 nd Workshop on Global Transdisciplinary Education, Research and Training, Pasadena, Calif., June 11-13, 2001 Marciniak J.J. (Ed.), 1994, Encyclopedia of Software Engineering, John Wiley and Sons, New York Nicolescu B., 1997, The Transdisciplinary Education of the University: Conditions for Sustainable Development, Int l Congress on Universities Responsibilities to Society, Bangkok, Thailand, November 12-14, 1997, URL /bulletin/b12/b12c8.htm Tanik M.M., Ertas A., 1997, Interdisciplinary Design and Process Science: A Discourse on Scientific Method for the Information Age, Transactions of the SDPS, Vol. 1, No. 1, pp Tucker I.B., 2000, Economics for Today. Second Edition, South-Western College Publishing, Cincinnati, Ohio