Motivating developers in OSS projects Veeti Vimpari, Joni Kerkelä, Fanny Vainionpää Abstract 1. Introduction 2. Motivation 2.1 Internal motivation 2.2 External motivation 3. Motivating Developers 4. Conclusions References Abstract The motivation for this paper is to find out why developers seek to join and keep working on Open Source Software projects, what motivates them. The research question forms from the topic as: What motivates developers to join and keep working on Open Source Software projects?. From a research perspective, answering the question requires a literature review on existing works in motivation and Open Source Software development to understand basis of these subjects. Besides reviewing subjects individually, reviewing existing research combining these subjects gives an insight to how these subject are related together and view to the current state in understanding the motivation of developers behind Open Source Software projects. Existing literature focuses heavily on examining narrowly the motivation initiatives in the Open Source Software projects, without examining the basis of motivation from psychological viewpoint. Literature review of existing works on topic, including examining the psychological basis on motivation, aims to give more mature view to the subject in question. The answer may help Open Source Software projects find the developers to join and keep existing developers
working on the project by giving a theoretical answer on the question what motivates developers. A lot of the literature we found was old considering the fast development pace in the software development industry; a lot of motivation literature related to Open Source Software development was ten years old. We found that the motivations of developers in Open Source Software projects are various, most of the sources agreed that internal motivators, such as learning, are the main reasons why developers join the projects in the first place, later on it could be giving something back to the community, and an external motivation could be that the developer needs a modification for their job. The complexity and size of the Open Source Software community means that there are a lot of different participants with different interests. 1. Introduction In general, Open Source Software is not fresh new phenomena as it has been around from the as early as 1950 s, but it has been largely growing phenomena in the last decades ( https://en.wikipedia.org/wiki/history_of_free_and_open-source_software ). Growing popularity has also initiated the research of the subject, to develop understanding of the phenomena. From the Scopus database (https://www.scopus.com/), by analysing the publications on the Open Source Software with search string Open Source Software, it can be seen that activity of publishing has peaked on the last years indicating the popularity of the subject and the relevancy of this article s topic to examine how developers are motivated in the Open Source Software projects. As mentioned afore, the topic of this paper is to examine how developers are motivated in Open Source Software projects. Motivation in software projects will include what initiates developers to join on the project and what makes them to keep on working on the project. Topic is especially relevant as Open Source Software phenomenon is constantly growing in popularity and has already became integrated to the key part of the software industry business, meaning that companies are being part of the Open Source Software projects besides freelance developers (Fitzgerald, 2006). This creates interesting standpoint as the companies have presumably the financial interest towards the Open Source Software project they are involved; and, as opposite, the freelance developers have personal interest, which doesn t include direct financial interest. The interest of freelance developers brings us back to the topic of this article as what motivates developers to join and work on Open Source Software projects. From afore introduced topic the research question is formed as: What motivates developers to join and keep working on Open Source Software projects?. The question is answered through a literature review, with materials from the open source as well as psychological materials. Aim is to provide a more mature view to the motivations of developers behind the Open Source Software projects by including the psychological view of motivation to consideration.
The motivation of the developers to participate in Open Source Software projects has been researched extensively in the past 10 years (Korg, Haefliger, Spaeth, Wallin, 2012). Motivation and Open Source Software projects has been a focus on both case study based analysis and empirical studies, which have researched well known Open Source Software projects. Case studies have been made by Torvalds and Diamond (2001) considering projects such as Apache Web Server and Perl. Empirical study research methods have been used by Hertel et al. (2003) while studying projects like Linux. (Wang, He, Chen, 2005) However, existing studies are focusing heavily on concept on what motivates developers on the Open Source Software projects, without considering solely the motivation from the psychological point of view. This initiates the need for examining the motivation from psychological viewpoint and combining it to the understanding to how motivation initiatives are currently used in the Open Source Software projects. Existing research has considered also the point that there are two very different stakeholders contributing and/or utilizing the Open Source Software projects source code, as freelance developers and companies. Wang et al (2005) claim in their research that actors that are participating in Open Source Software communities can be seen to belong into two different tiers. The first tier considers the developers that work individually to produce the source code for the project. On the other hand there is the second tier, which consists of the companies that use the source code that the first tier has developed in their commercial purposes. The motivation aspect of these two groups can be divided in a similar way as in the previous dichotomy, in which the first tier is participating in the Open Source Software community for their internal motivational reasons, while the second tier is motivated by an external motivator. (Wang et al, 2005). As a summary, the aim of this article is to provide comprehensive view to how developers are motivated in the Open Source Software projects. Method for information generation in the article is literature review, but besides only reviewing literature concerning the motivation initiatives in Open Source Software projects, the article aims to broaden the view by including the focus on the psychological viewpoint to concept of motivation. This way article will complement existing literature, which is mainly solely focusing on the motivation on the Open Source Software projects. The structure of this article includes the introduction of the subject in part 1, the concept of motivation is explained in chapter 2. and in chapters 2.1 and 2.2. Internal and external motivation are discussed as separate items. Chapter 3. considers motivation within the Open Source Software projects, which is the main focus of this article. And lastly, in Conclusions the implications of the characteristics of motivation within the Open Source Software projects for individual developers are discussed.
2. Motivation Oxford dictionary defines motivation as a reason for acting or behaving in some way, or the desire or willingness to do something, whereas the verb motivate is described as giving someone a reason to do something. (Oxford Dictionary, 2016). Maslow (1954) lists five hierarchical levels of needs and suggests that once a lower level need is fulfilled, the individual will be motivated to fulfill the next level of needs, towards higher levels of self-actualisation. Maslow s (1943) five levels of needs driving individuals are physiological needs such as hunger, the safety needs such as the want for an orderly and predictable life, the need for love, the esteem needs which are based on achievements and respect from others, and finally the need for self-actualization which is the need to do what you can (e.g. a programmer writing code). These basic needs can be considered the starting point of motivation. The average desires can be means to an end; having a job to get money to buy something, to promote your status. This means that the actual goals may not be seen from the initial wants, so the goal may be an unconscious motivation. Culture also affects human motivations; in one culture the self-esteem may come from being a good hunter while in another doctors are highly valued. (Maslow, 1954). When considering the motivation deeper from the individual s perspective, excluding what is the target or source of motivation and focusing on the concept of how individual perceives the motivation; the feature of motivation is that it is not a unitary phenomenon. This means that the amount of motivation is hard to measure because it depends on the perception of the individual feeling the motivation. Besides the varying amount of motivation felt by individual, the type of motivation is not a constant either. (Ryan & Deci, 2000) These factors create the complexity of motivation and the challenge to study it in different contexts. As Ryan & Deci (2000) introduce the idea in their research, the amount of motivation felt by an individual can vary largely depending on whether the motivation is internal or external. The challenge of studying motivation can be illustrated by an example case, where two individuals are needed to define how strong their motivation for doing the same task is, in which one has internal motivation, and in comparison, the other has external motivation. In the example s case whichever individual may feel much more motivated than the other, depending on the context and the motivation source. Open Source Software projects have different social structures compared to businesses, the people involved create a type of community where they share interests in using and developing the software. The members themselves assume certain roles depending on their interests, instead of being assigned to a post as is the case with employers. Without these motivated developers, the open source projects can t succeed. (Ye, Y., & Kishida, K., 2003).
2.1 Internal motivation Li et al (2011) describe internal motivation as behaviour performed for the act itself; the pleasure and satisfaction gained from performing it. This makes internal motivation to be very person specific for tasks as people have different preferences obviously (Deci & Ryan, 2000). Internal rewards arise from within the individual, for example pride for a well-executed job (Shanks, 2007). Shanks proposes that while rewards act as external incentives, the real motivation to act comes from the individual. Franken (1994) describes motivation as an internal state of desire that gives a direction for goal seeking behaviour, which is inline with Shank s proposal. Deci & Ryan (2000) have supporting definition for internal motivation as they propose that internal motivation is satisfaction, which comes from doing the act itself rather than doing the act driven by the consequence of act. Aforementioned concepts of internal motivation are appearing in academic material over and over again and contradicting views are basically non-existent. Due that, it is justified to use these arguments as the basis of internal motivation. As going deeper into the concept of internal motivation, Deci & Ryan (2000) point out that internal motivation can be actually affected by the external motivators from the environment. Those external motivators can be for example deadlines, rewards and other external motivators discussed in the chapter: 2.2 External motivation. However, even if external motivators affect a person to feel, for example, better about executing the action, it can arguably be stated that enhanced internal motivation for executing the act by external motivation is still purely internal motivation as it is how person intercepts the actual execution of action. For clarity, in this case, the external motivator will be a present driver for completing the action, but not present in how the person intercepts the execution of action. This is also supported by Deci & Ryan (2000) as they state that action initiated by external motivation can become internally rewarding as they are doing it. Other factors that can enhance the internal motivation are autonomy and competency, according to Deci & Ryan (2000). Feeling of autonomy makes a person assimilate executing the action by free will and thus enhancing the motivation for executing it. The competency makes the actor feel satisfaction by being able to thrive in doing the act. (Deci & Ryan, 2000) The relation of internal motivation and Open source projects originates in the hacker culture, originally seen as altruism and a way to stand up for software freedom, as well as hedonistic nature of the work (enjoying programming). However, the open code also enables feedback and thus technical improvement. (Bonaccorsi & Rossi, 2003). Ye, Y., & Kishida, K. (2003) explain that internal motivations are cognitive and external motivations are social, and they state that the key to motivating developers is the internal satisfaction of being involved in an Open Source Software project but they recognise that purely altruistic work is unsustainable. This is why the developers need social recognition and rewards.
2.2 External motivation The basic idea of external motivation is that the person completes some action to achieve a specific outcome. In this way external motivation differs from the internal motivation as in internal motivation person executes the act independable of the outcome. The external motivation can thus be a motivator for a variance of tasks independable of the person, unlike internal motivation, which depends on the person s personal perception of the act. (Deci & Ryan, 2000) Li et al (2011) describe external motivation as behaviour aiming to achieve some goal; gaining some reward or to avoid punishment. External rewards are given by another person, for example bonuses given to workers when they have performed well (Shanks, 2007). Shanks suggests that employers can offer incentives to employees, such as changing job descriptions, rearranging work schedules, improving work conditions, and team reconfigurations. There are different types of external motivators, B. F. Skinner s reinforcement theory on external motivation states that individuals are motivated when their behaviours are reinforced. There is positive reinforcement which means rewarding positive behaviour, avoidance learning or negative reinforcement refers to actions to reward behaviours which avoid unwanted behaviour. There is also punishment, which is essentially the reducement of unwanted behaviour by having negative consequences. Lastly there is extinction, which means removing rewards in case of unwanted behaviour. There are also some theories on external motivation, Adam s Equity Theory suggests that people are motivated when they are treated equally in comparison with their peers. Vroom s Expectancy Theory proposes that people s expectations on the outcome of their performance motivates them. Locke s Goal Setting Theory says that people are motivated to take action when they are given goals to achieve. (Shanks, 2007). In deeper insight for external motivation based on afore presented concept of external motivation, it can be seen that external motivation is a more complex concept that the internal motivation. As Deci & Ryan (2000) presents, autonomy in executing the act driven by external motivation can vary. For example, very low autonomy can be seen in cases when there is punishment as external motivator, and as an opposite, if there is reward as external motivator, the level of autonomy can be in the same level as actions motivated by internal motivation (Deci & Ryan, 2000). 3. Motivating Developers Krogh et al (2012) state that the categories of motivation for Open Source Software developers can be grouped into ten different subcategories, that can again be separated into three parent categories. Under the internal category, they place motivational factors such as ideology,
altruism, kinship and fun. The next parent category, internalized external motivations, including reputation, reciprocity, learning and own-use. The third and final parent category for motivational reasons is external, which includes career and monetary payment. (Krogh et al, 2012) Hars & Ou (2001) claim that the programmers that participate in the Open Source Software community are not driven or motivated by monetary reasons. The developers are participating for their own preferences and hobbies or alternatively to increase the welfare of others. Based on this, it can be said that the developers are driven by internal motivational factors more than external motivators. (Hars & Ou, 2001) Altruism is considered to be a variant of inner motivation. In these kind of situations, where the participant is driven by the desire to help others in their own cost, is also considered to be a part of internal motivation. Hars & Ou (2001) also say that they believe that altruism is an important factor that motivates developers in the Open Source Software communities and it has a positive influence on the participation level in Open Source Software projects. Another variant of altruism is community identification, which is also considered to be part of internal motivation. (Hars & Ou, 2001) According to Krogh et al. (2012), software developers are collectively accounting to evolve the virtues in social practice, which has also been referred as a school of virtue due to the the ways that software is being developed in the Open Source Software community. Developing Open Source Software can be a valuable provider of training for software developers. This has also been noticed by software companies such as Intel and Red Hat, which have been actively interested in recruiting for software engineers who have been developing high quality contributions in Open Source Software communities. (Krogh et al, 2012) Also in more previous studies, Lakhani & Wolf (2003) have presented that the personal feeling of creativeness is an important factor in what motivates the developers in Open Source Software projects. In their study, more than 61% of the respondents answered that their participation in a Open Source project can be described and considered to be as creative as any other thing that they have done in their lives. This statement can be also supported by the finding, that shows that in their study, Lakhani and Wolf found out that most of the survey respondents were willing to dedicate more additional hours into their project when they had experienced a state of flow, which is considered to be related to creativity related factors. According to the study, internal motivation that is originating from the enjoyment of the contributor is the strongest source of motivation in exerting effort in to a Open Source Software project (Lakhani & Wolf, 2003). Hannebauer & Gruhn conducted a research about the motivations of newcomers to FLOSS projects (GNOME and Mozilla). The reasons developers were motivated to create modifications such as bug fixes or create a specific feature to a FLOSS program were for their own or their employer s needs, for contributing to the project, to learn about the practises or working with other people s code, and a few mentioned the joy of programming. When the participants ranked motivations in order of importance, the joy of programming was ranked highest, second came the need to make the modification, next learning, and finally community. Submission motivation was studied separately, and Hannebauer and Gruhn found that half of the respondents stated altruism as the primary motivation for a submission. The main reasons were
that the developers wanted to foster the projects, give something in return to the FLOSS project, to learn, and some submitted for fun. Economic benefits were also mentioned as motivation, and some wanted to avoid reintegrating a self made modification into every new release, because when the modification is submitted to the main project, it will be integrated by those maintaining the project. When ranked according to their importance, the strongest motivation was learning, then altruism, fostering FLOSS, avoiding re-submissions, giving something back, publicity, external review, fighting closed source, and last came gaining respect. The authors concluded that hedonistic motivations and the need for a modification are usually the drivers for modification motivation, and while altruism has been identified as a less important contributor for motivation, this study shows it to be important with submission modifications. The authors suggest that the developer's own needs are the primary modification motivation for newcomers who are employed. ( Hannebauer & Gruhn, 2016). Monetary payment for the contributors represents external motivation (Alexy & Leitner, 2011). Lakhani and Wolf (2003) researched, that monetary rewards that are paid to the contributors in Open Source Software development affect the level of their involvement. The effect of financial support can be said to be significant, as paid contributors of the project invested more time in their projects compared to non-paid volunteers. Paid contributors participated averagely 17,7 hours in a week in all of their Open Source Software projects, compared to 11,7 hours per week which volunteers used to contribute in their Open Source Software projects. For what it comes to the main project of each developer, similar and even more significant differences can be found from the study, as paid contributors reported to use 10,3 hours per week on their main project, while non paid volunteers contributed only 5,7 hours per week in their primary Open Source Software project. (Lakhani & Wolf, 2003). Alexy and Leitner (2011) found out that the concept of monetary payment as a reward in Open Source Software projects is more complex subject than previously thought for what it comes to the motivation of the contributors. The effect that monetary rewards have on internal motivation is mediated by the level self-determination of the contributor, which is seen to be moderated by payment norms. Payment can potentially have a negative impact on instinct motivation, but the level of total motivation is seen to be affected positively and directly by monetary rewards. Alexy and Leitner also state, that if the payment is fully expected by the contributor, it should lose its potentially positive effect that it directly has on the level of total motivation. (Alexy & Leitner, 2011). 4. Conclusions Most literature we came across seemed to be in agreement that the initial reason why developers join Open Source Software projects is their internal motivation; altruistic ideas of creating something that will benefit others, the enjoyment they get from the tasks and the possibility to learn in the process. The Open Source Software projects can also give rewards in the form of recognition, by giving credit to the developers deserving it, and in many cases the
developers are paid for their work. The motivations could also depend on the developer's current situation in life; a student may be looking for experience and recognition, while some experienced developer may be taking part solely because they need the modifications themselves or for their employer. Based on the findings of Hannebauer & Gruhn (2016) and supported by Hars & Ou (2001), it can be stated that internal motivations are the main reasons for individual developers to join and keep working on Open Source Software projects. This supports the nature of Open Source Software projects as external motivators, such as monetary, career or reputation, are by default diminished on them, as Alexy & Leitner (2011) also propose on their article. This concept of internal motivation factors as the main motivation for developers in Open Source Software projects also explains the observed growing popularity of Open Source Software projects, without Open Source projects giving direct rewards to developers in general. When going deeper to reveal which are in fact the most effective motivators for developers participating in Open Source Software projects, to provide a more thorough answer for the research question, there are a few main motivating factors. A few of the most frequently mentioned internal motivation factors among the quoted research materials are the enjoyment gained from programming and contributing for the community, or reciprocity. The most frequently mentioned external motivators within the research materials are learning and participating to make the program more suitable for the developers own use. These motivating factors are very self sustainable as they don t need direct input from Open Source Software project management to keep up, which also makes the projects self sustainable when considering the developer participation, and on a broader scale, the whole Open Source Software community becomes self sustainable because of this. As a summary, based on the research material found, it can be stated that Open Source Software projects do not have one single motivation factor, which explains the thrive of Open Source Software communities, but rather multiple factors which are together bringing and keeping individual developers working on projects for different reasons. For example, even for individual developer the initial motivation for joining the Open Source Software project may be to learn, but advancing on the project, the motivating factor may change to enjoying to be able to contribute to community. Although, based on the study, it can be said that there are frequently occurring motivating factors present in the Open Source Software projects, which apply for the majority of developers. These frequently occurring motivation factors include learning, enjoyment for programming and being able to contribute for community, as stated afore. This study has limitations regarding the sources; a lot of the literature was older than 10 years and this means that the current situation could be quite different in reality. Old sources mostly emphasise the altruistic nature of Open Source Software Development, whereas it could be the case that more and more developers are paid for their work. This would be possible subject for research, as well as Crowd sourcing in Open Source Software Development. Considering these limitations, this study reveals that some Open Source Software Development research could be
updated, however, it is possible that people still initially look into getting involved with these projects and communities for the internal motivations. Perhaps another aspect to research could be the hobby developers, who contribute on their freetime; how much they affect the projects and how long they remain interested. References Alexy, O., & Leitner, M. (2011). A fistful of dollars: Are financial rewards a suitable management practice for distributed models of innovation?. European Management Review, 8(3), 165-185. Bonaccorsi, A., & Rossi Lamastra, C. (2003). Altruistic individuals, selfish firms? The structure of motivation in Open Source Software. The structure of motivation in Open Source Software. Franken, R. E. (1994). Human Motivation. California, Brooks/Cole. Fitzgerald, B. (2006). The transformation of Open Source Software. Mis Quarterly, 587-598. Hannebauer, C., Gruhn, V. (2016). Motivation of Newcomers to to FLOSS Projects. Proceedings of the 12th International Symposium on Open Collaboration, Berlin, August 17-19, 2016. New York, ACM. Hars, A., & Ou, S. (2001, January). Working for free? Motivations of participating in open source projects. In System Sciences, 2001. Proceedings of the 34th Annual Hawaii International Conference on (pp. 9-pp). IEEE. Hertel, G., Niedner, S., & Herrmann, S. (2003). Motivation of software developers in Open Source projects: an Internet-based survey of contributors to the Linux kernel. Research policy, 32(7), 1159-1177. Lakhani, Karim, and Robert G. Wolf. "Why hackers do what they do: Understanding motivation and effort in free/open Source Software projects." (2003). Li, Y., Tan, C. H., Teo, H. H., & Mattar, A. T. (2006, April). Motivating Open Source Software developers: influence of transformational and transactional leaderships. In Proceedings of the 2006 ACM SIGMIS CPR conference on computer personnel research: Forty four years of computer personnel research: achievements, challenges & the future (pp. 34-43). ACM. Li, Y., Tan, C., Xu, H., Teo, H. (2011). Open Source Software adoption: motivations of adopters and amotivations of non-adopters. ACM, vol 42 issue 2, p 76-94.
Maslow, A.H. (1943). A Theory of Human Motivation. Psychological Review, 50, 370-396. Motivation. (2016). in en.oxforddictionaries.com. Retrieved October 20, 2016, from https://en.oxforddictionaries.com/definition/motivation Motivate. (2016). In en.oxforddictionaries.com. Retrieved October 20, 2016, from https://en.oxforddictionaries.com/definition/motivate Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary educational psychology, 25(1), 54-67. Shanks, N.H.(2007) Introduction to healthcare management. Jones & Bartlett Learning, chapter 2. http://www.jblearning.com/samples/076373473x/3473x_ch02_4759.pdf Von Krogh, G., Haefliger, S., Spaeth, S., & Wallin, M. W. (2012). Carrots and rainbows: Motivation and social practice in Open Source Software development. Mis Quarterly, 36(2), 649-676. Wang, F. R., He, D., & Chen, J. (2005, August). Motivations of individuals and firms participating in open source community. In 2005 International Conference on Machine Learning and Cybernetics (Vol. 1, pp. 309-314). IEEE. Ye, Y., & Kishida, K. (2003, May). Toward an understanding of the motivation of Open Source Software developers. In Software Engineering, 2003. Proceedings. 25th International Conference on (pp. 419-429). IEEE.