Free and Open Source Are Not Necessary Conditions of Successful Development

Abstract

Abstract: Free and are not requirements for successful software development. The argument that and frequent are one indicator of successful software development does not imply the necessity of users updating the source code but the importance that they function as testers and debuggers. In the development of Japan's non-open source, non-free shareware Mail, the users perform the functions of testers and debuggers. It is evident from the case of Hidemaru Mail that early and frequent upgrades are neither result nor performance indicators, but rather the way to intrinsically motivate users so that they can function as testers and debuggers. In actuality, the developer of Hidemaru Mail motivates users to submit requests and reports more often by meticulously responding to almost all users' requests and reports and upgrading early and frequently. Existing studies are overly particular about whether the source code should be open, and lack the perspective of organizational theory about how developers should motivate users to function as testers and debuggers. While there is the definite possibility of motivating users through open source, this does not indicate that open source is a necessary condition. The necessary condition is to motivate as many users as possible to cooperate in the development. If this is accomplished, the development of software will succeed, regardless of it being free or open.Keywords: open source, software development, early and frequent releases, intrinsic motivation, Hidemaru MailIntroductionThere is little research disputing that the reason for the success of Linux was due to it being free and open source. DiBona, Stone, and Ockman (1999) in particular, emphasized the excellence of open source. The fact that subsequent open source research can be largely divided into the following two categories is proof of this. One set of research is related to the motivation of developers for open source software (OSS) and the open source community (Bitzer, Schrettl, & Schroder, 2007; David & Shapiro, 2008; Hertel, Niedner, & Herrmann, 2003; Oreg & Nov, 2008; von Krogh, Spaeth, & Lakhani, 2003). Another set deals with open source strategy of companies (Capra, Francalanci, Merlo, & Rossi-Lamastra, 2011; Dahlander & Magnusson, 2005, 2008; Grand, von Krogh, Leonard, & Swap, 2004; Sasaki & Kitayama, 2000; Stam, 2009; West, 2003). Both types of research assume that software, which is superior in terms of function and quality as well as stability and robustness, is created in a short period, provided it is developed as open source. Hence, interest is solely devoted to how to ensure the establishment and survival of open source projects and communities, and how companies utilize the results well.Open source does not automatically lead software development to success. For example, Takahashi and Takamatsu (2002, 2013) assert that the miraculous timing of the release of Linux and developer's motivation were the reasons behind its success, and it did not matter whether it was open source or closed source, or free or charged software.1 If early and frequent releases of a new version are an indicator of software development success, then there is no need to be particular about free and open source. It becomes important not to motivate users with high levels of development competence to modify the code through open source, but to motivate mere general users to function as testers and debuggers. If developers can entrust the testing and debugging to users, the developers can release new versions very quickly. A good example that illustrates this is the development case of Japan's representative mailer, Mail. Hidemaru Mail is non-free shareware. While the developer, Hideo Saito, did not publicize the source code, new versions have been released at a pace of approximately once every four days for twelve years from the initial version release. …