Thesis: The Future of Business Software: Proprietary or Free and Open Source?

Abbreviations

BSD: Berkeley Software Distribution, a free-and-open-source version of Unix.

GNU: Gnu's Not Unix, a recursive acronym for the GNU project for free software.

FOSS: Free and Open-Source Software.

FSF: Free Software Foundation, a non-profit group founded in 1985 to promotes the freedom to study, distribute, create, and modify computer software.

MVP: Minimal Viable Program.

UNICS/UNIX: Uniplexed Information and Computing Service, was the original name for the Unix operating system coined in 1970, and a pun on Multics (Multiplexed Information and Computer Services), from which it was derived. When capitalised as UNIX, it refers to AT&T the trademarked version.

XML: Extensible Mark-Up Language.

VPAC: Victorian Partnership for Advanced Computing.

Chapter 1: Introduction

1.1 A Personal Interest

Before engaging in the research question, whether the future of business software is proprietary or free-and-open-source (FOSS), it is necessary to provide a personal introduction if only to make any motivations and biases explicit, as well as to perhaps suggest some prior experience in the subject. In part, there is a purely academic interest, which notes an apparent conflict between business reasoning, which argues for an organisation weakening the power of competitors, and consumers, and that of economics reasoning, which argues for heightened competition and maximising consumer sovereignty. This academic interest comes from a background in four different degrees, namely a Master of Business Administration, separate graduate degrees in project management and adult education, and an honours degree in politics, philosophy, and sociology.

How the two disciplines of business and economics systems related to software engineering are also of professional interest, from the past twelve years of employment and publications with the Victorian Partnership for Advanced Computing (VPAC) and the University of Melbourne, and over twenty years as an IT professional. A prior position of community advocacy for free-and-open-source software is also readily acknowledged, as a committee member or public officer for Linux Users of Victoria from 2005 to 2019. There is both a theoretical and practical interest in the subject, motivated by what might be conflicting desires for more effective and efficient software to be produced for organisational use, and for software developers and system administrators to be paid well, and for software source-code to be available for re-use, review, and elaboration. The possibility that the thesis is an exercise in cognitive dissonance has not been ignored.

As a matter of a shared anecdotal experience among friends and colleagues in the industry and across multiple organisations, senior management often seems poor at making software choices at the enterprise level. Whilst it is readily recognised that even among technical managers that as one's involvement in management increases one's deep understanding of particular technologies must decrease. This is not a pejorative statement, but simply a function of available time. What is surprising is that core principles of software technology seem to be overlooked. Proprietary "enterprise software" is often selected on the basis of being feature-rich, but is actually complex and fragile, requiring difficult and time-consuming processes if changes are desired. FOSS software systems built on the principle of being a Minimum Viable Program (MVP) that interacts well with other applications are overlooked due to a lack of perceived functionality. The power of irrational decision-making in management choices is, unfortunately, outside the purview of this inquiry, although attention must be drawn to the issue as it will inevitably undermine any rational conclusions that this thesis makes.

A Brief History of the Research Topic

Early computer systems were often sold as primarily hardware with the software provided with source-code, of which the 1951 IBM 701, the 1953 UNIVAC, and the 1959 SHARE operating system are well-known examples. A major change occurred in 1974 when the US Commission on New Technological Uses of Copyrighted Works recommended that software could be subject to copyright, which became a legal norm in 1983 in the United States with the Apple Computer, Inc. v. Franklin Computer Corporation case where it was ultimately determined that a computer software, both source-code and binaries, could be protected by copyright, and in the same year, IBM introduced an object-code-only policy. During this period Unix (originally UNICS) gained prominence as an operating system, with the first commercial version, UNIX System V, released by AT&T also in 1983 and in competition with the University of California, Berkeley Computer Systems Research Group, also known as BSD UNIX and with the GNU Project which was also initiated in 1983 to create utilities and applications that were on Unix systems but as free software.

Whilst competition between various forms of proprietary Unix-based systems occurred during the 1980s and the early 1990s, the so-called "Unix Wars", the competing new desktop computing systems readily adapted to the new legal standards with applications such as WordStar and VisiCalc initially having prominence. As the personal computer market effectively narrowed to IBM-PCs and clones and Apple machines by the early 1990s, so too did operating systems and office applications or, as a suite, "productivity software". Microsoft Office in particular (starting with MS-Word in 1983 with a Macintosh version in 1984) became an integrated suite in 1990, came to dominate usage and has more recently moved from a stand-alone license to a continuing license with a client-server and cloud services model with Office365. Outside this dominance in this immediate suite of applications there is widespread use of proprietary software in various day-to-day applications used by business workers, including the operating system and utilities (such as used by MS-Windows, MacOS X, and the Android utilities), web browsers (Microsoft Edge, Google Chrome, Safari), word processors, spreadsheets, presentation software (MS-Office), and for more specialist cases, databases (MS-SQL, Oracle), customer-relationship management software, enterprise resource planning, and various domain-specific software etc.

However, in the last twenty years, servers, embedded systems, and scientific applications have increasingly made use of free-and-open-source (FOSS) products (GNU/Linux, Open/FreeBSD). In the 1990s the combination of the Linux kernel with the GNU application suite, making GNU/Linux, came to replace the various proprietary Unix systems. In addition, there have also been some notable developments in specialist and general business applications (Firefox, LibreOffice, MariaDB). Between these two general models, there is, of course, software projects and aggregate products may exist in the continuum that has a mix of various license structures. An illustrative example that will be explored is the Office Open XML Ecma/ISO standard, adopted by Microsoft Office 2010. As a whole, software licenses can range from those that are entirely in the public domain to permissive FOSS licenses, protective FOSS licenses, proprietary licenses, and trade secrets.

Simplifying at this stage, public domain licenses are defined as those which have no legal exclusion in use. In comparison, permissive FOSS licenses in software are typically those which allow for redistribution of the software as desired but with attribution. Importantly, they do not require redistribution of the software source-code which is required, in varying degrees, by protective FOSS licenses. The theory behind protective FOSS licenses is that the freedoms that a software user has received from software must be re-distributed to others. In comparison, proprietary software is where copyright is applied to either the source code, the object code, or both. In some cases, a software patent is also applied. Finally, unlike copyright and patents, which have a government record, a trade secret is not disclosed to any public authority.

Organisational and Social Importance of the Research Topic

There can be little doubt of the importance of the continuing contribution of computer software to scientific and technological developments and administrative efficiency. In the engineering fields of fluid dynamics or materials, in the health sciences for ab initio molecular studies, genomics, and systemics, weather forecasting and climate science, the well-established combination of scientific facts feeding into simulation models to explore and then be tested against empirical reality is a well-established and successful method. In terms of efficiency and effectiveness of data collection, storage, and processing, and human communication, it often an expression of wonder among contemporaries on how society even operated at all prior to ubiquitous communications technology. The software which drives these technologies, and continues to do, is, however, subject to different and competing license regimes. Whilst it may not be immediately obvious to end-user, any inquiry which can contribute to improved efficiency and effectiveness of software development will have significant importance on the daily lives of all users.

However, from an organisational perspective, the debate over competing software licenses will vary according to their functional needs. System engineers and software developers prefer access to source code for purposes of efficiency and effectiveness. Software consumers prefer functionality, compatibility, and reduced vendor lock-in, all at the lowest possible price, whereas software vendors prefer to establish their software as the industry default and extract monopolistic profits. These competing needs are also reflected in theoretical models of behaviour across different academic disciplines. The functional needs of a software engineer have their equivalent in the intellectual expectation that software source code is available for development. Choice theory, whether rational or bounded, is at the core of microeconomic decision-making and contrasts with the macro-economic expectations of institutional economics. Monopolistic competition, the quest for "competitive advantage", is a core assumption of business reasoning. Finding a theoretical model that satisfies these disciplinary differences in itself a significant interdisciplinary project.

There are also related flow-on effects. It is not just software that has these conflicting interests and theoretical models which find their legal expression in licensing regimes, but rather all information goods, including books, music, video and so forth. Colloquially and pejoratively referred to as "piracy", the issue of copyright infringement is tangential to the main research question, despite its extremely widespread use. Whereas the thesis seeks to explore how to resolve the different licensing models in a manner that can satisfy developer, vendor, and consumer, interests in business software, the issue of infringement of restrictive versions of such licenses is a different, but related, matter.

Overview and Outline

This thesis seeks to explore the history and development of these various software license models, with attention to how they apply in terms of structural and motivational aspects for competency and innovation, the international landscape especially with regards to existing educational and legal infrastructure, and with ample use of case studies, market trends, and expert interviews. The ultimate research objective of the thesis is to draw upon these resources to determine to what degree the future of business software (also known as productivity software), will follow a free-and-open-source licensing model or whether it will follow a more proprietary model. In exploring this research question this inquiry will (a) conduct a literature review of related works, (b) provide secondary and primary research, analyse the data of the review and research, and (c) provide a conclusion and evaluation and recommendations derived from the results. These sections provide the major chapters for the study.

The literature review will cover four major and diverse areas that are not commonly drawn together. Firstly, there is a discussion of the history and use of various software licenses, which is supplemented by an exploration of the Church-Turing thesis on the nature of computational functions. This has particular relevance in deciding whether software is an 'invention' or a 'discovery', which raises the scope of applicable licenses. The application of particular software licenses is supplemented with the business literature related to business profitability from the licenses applied to information goods and its relationship to competition. From a contrary perspective, the economics of competition, and especially starting with institutional economics and imperfect competition in relation to software licensing, both of which also can be supplemented by an associated connection with the more common and mature research area of the economics of software engineering.

Whilst the literature review provides a theoretical understanding and background to the thesis question, additional research provides evidence which contributes to the answer. In this sense, secondary research which provides existing empirical results is an initial objective standard from logical elaborations that can be derived. These results, which can show trends and changes, do not provide an explanation of why the trends and changes occurred. The distribution of kernels on various operating systems on various platforms and use can thus be provided over time, ranging from the world's most powerful systems, mainly used in scientific research, to servers that make up the core of organisational infrastructure, to consumer-level usage. Several case studies also contribute to the secondary research. This will include the "Unix Wars" in server-level systems, the forking of open-source code when proprietary impositions are put into place, the establishment of the ISO/IEC 26300 Open Document Format for Office Applications and the ISO/IEC 29500 OfficeOpen XML standard, and the adoption of open-source productivity software by the city of council of Munich which switched from proprietary to open-source then back again. Primary research in these areas is derived from semi-structured interviews with system engineers and developers involved in the case studies to find the reasoning and effects of the conscious decisions to change. The data analysis will seek to draw together this secondary and primary data to provide a conclusion which answers the research question in a theoretically grounded manner.

References

Atal, V., Shankar, K. (2014). Open source software: competition with a public good. Atlantic Economic Journal, 42(3), 333-345.

Atal, V., Shankar, K. (2015). Developers’ incentives and open-source software licensing: GPL vs BSD. The BE Journal of Economic Analysis & Policy, 15(3), 1381-1416.

August, T., Tunca, T.I. (2008). Let the pirates patch? an economic analysis of software security patch restrictions. Information Systems Research, 19(1), 48-70.

August, T., Shin, H., Tunca, T.I. (2013). Licensing and competition for services in open source software. Information Systems Research, 24(4), 1068-1086.

August, T., Shin, H., Tunca, T. I. (2017). Generating Value Through Open Source: Software Service Market Regulation and Licensing Policy. Information Systems Research, 29(1), 186-205.

Blind, K., et al (2005). Softwar Patents. Economic Impacts and Policy Implications. Edward Elgar Publishing.

Copeland, B.J., et al. (eds.) (2013). Computability: Turing, Goedel, Church, and Beyond, MIT Press.

Classen, H. W., (2017). A Practical Guide to Software Licensing for Licensees and Licensors (6th ed). American Bar Association.

Gabszewicz J., Thisse J-F. (2000). Microeconomic theories of imperfect competition. Cahiers d'économie politique, 37.

Glass, R.L. (2002). Facts and Fallacies of Software Engineering. Addison-Wesley.

Fink, M. (2003). The Business and Economics of Linux and Open Source. Prentice Hall Professional.

Kirchmer, M. (1999). Business Process Oriented Implementation of Standard Software: How to Achieve Competitive Advantage Efficiently and Effectively (2nd edition, FP 1996). Springer.

Klemens, B., (2005). Math You Can't Use: Patents, Copyright, and Software. Brookings Institution Press.

Herrera, J.T. (2012). El conflicto en el Software: una aproximación a la incidencia del «fork» en los proyectos de software libre. ArtefaCToS. Revista de estudios sobre la ciencia y la tecnología, 5(1), 83-122.

Hewitt, E., (2018). Technology Strategy Patterns: Architecture as Strategy. O'Reilly.

Leith, P., (2006). Software and patents in Europe. Cambridge University Press.

Levine, B. (1992). Sources of competitive advantage in knowledge-based industries. Doctoral thesis. University of Michigan.

Nicols, K., (1998). Inventing Software: The Rise of "Computer-Related" Patents. Quorum Books.

O'Grady, S. (2015). The Software Paradox: The Rise and Fall of the Commercial Software Market. O'Reilly.

O'Regan, G. (2002). A Practical Approach to Software Quality. Springer.

O'Regan, G. (2006). Mathematical Approaches to Software Quality. Springer.

Popp, K.M. (ed), (2011). Advances in Software Economics. A Reader on Business Models & Partner Ecosystems in the Software Industry. Books On Demand.

Porter, M. E. (1985). The Competitive Advantage: Creating and Sustaining Superior Performance. Free Press.

Raymond, E.S. (1999). The Cathedral and the Bazaar: Musings on Linux and Open Source by an Accidental Revolutionary. O'Reilly.

Richardson, G. B. (1998). The Economics of Imperfect Competition: Collected Papers of G.B. Richardson. Edward Elgar.

Robinson, J., (1933). The Economics of Imperfect Competition. McMillan.

Rustad, M., (2010). Software Licensing: Principles and Practical Strategies. OUP USA.

St. Amant, K., Still, B. (2007). Handbook of research on open source software: technological, economic, and social perspectives. IGI Global.

St. Laurent, A. M., (2004). Understanding Open Source & Free Software Licensing: Guide to Navigating Licensing Issues in Existing & New Software. O'Reilly.

Vatiero M. (2009). An Institutionalist Explanation of Market Dominances. World Competition. Law and Economics Review, 32(2):221-6.

Vatiero, M. (2011). The institutional microeconomics of positional goods. Working Paper.

Webb, B., Schlemmer, F. (2008). Information Technology and Competitive Advantage in Small Firms. Routledge.

Williams, S., (2002). Free as in Freedom: Richard Stallman's Crusade for Free Software. O'Reilly.