Benefitting from Contributions to the Android Open Source Community

The open source community (OSC) is a place to develop collective knowledge available to anyone, thereby inevitably engendering free riders. Despite this, many firms have contributed to OSCs. This study examines 10 Android smartphone manufacturers between 2010 and 2013 with regard to their (a) source code contributions and the relation of those contributions to (b) time to market as measured by the release of their first Android smartphones. The results of the analysis are divided into the following two groups: (A) a group that released smartphones faster than their competitors through source code contributions and (B) a group that made few source code contributions and was slower to release smartphones than group (A). In addition, in a few years, some firms were observed to have move from group (B) to group (A).


Introduction
Many firms today innovate by exploiting external technologies, such as open source software (Altman, Nagle, & Tushman, 2014;Baldwin & von Hippel, 2011;Lakhani, Lifshitz-Assaf, & Tushman, 2013).For example, today's smartphone manufacturers have developed their smartphones using Android operating system (OS) source codes obtained from the Google-led open source community (OSC).However, the issue of inevitable free riders (Gabel, 1987) arises because OSCs keep all information on source codes open (West, 2003).Furthermore, 70% of the contributors in the Linux and Apache OSCs make no more than a single contribution of source code changes (Kogut & Metiu, 2001).Nevertheless, some Android smartphone manufacturers proactively contribute source codes to new versions of the Android OS (Shiu, 2012).Why do these firms exist?
Over the years, existing debates on OSCs have focused on the motivations of individual developers regarding why they would make source code contributions (see David & Shapiro, 2008;Fujita & Ikuine, 2013, 2014;Raymond, 1997;Takahashi & Takamatsu, 2002, 2013;von Krogh, Spaeth, & Lakhani, 2003).These studies have, for example, noted the importance of feedbacks from users (Fujita & Ikuine, 2014), early software releases, and frequent upgrades by OSC leaders (Fujita & Ikuine, 2013;Raymond, 1997;Takahashi & Takamatsu, 2002, 2013).However, few studies have examined the motivations of "firms" that contribute source codes to an OSC (Nagle, 2015).Clearly, firms differ from individuals, and they do not contribute to an OSC simply for their own prestige or to pursue their hobbies.In addition, when contributing source codes to an OSC, fear of spillovers exists whereby a firm's technologies can be used free of charge by other firms.

Firms that participate in an OSC communicate with managers of
Benefitting from contributions to the Android open source community an OSC (Sulayman, Sowe, & Stamelos, 2008), thereby exploiting knowledge with regard to the latest source codes.Concerning this point, this study posits that Android smartphone manufacturers contributing to the Android OSC can release their products to the market faster than other firms. 1 This is because these manufacturers receive early access to knowledge regarding the latest versions of the Android OS by communicating with OSC's manager, Google.

Methods
This study investigates several versions of the Android OS 2 between 2010 and 2013, including Froyo, version 2.2; Gingerbread,

Results
Table 1 lists

Conclusions
This study examined 10 Android smartphone manufacturers.The result demonstrated that they can be categorized into two groups: (A) those that are faster than competitors in releasing products to the market due to their high source code contributions and (B) those that are slower to release products to the market due to their few source code contributions.In addition, this study indicated that some firms have moved from group (B) to group (A).These results showed the possibility of building competitive advantage through abundant source code contributions to an OSC.
In fact, a characteristic of the Android smartphone market is a price decline after the release of a product.For example, in the UK smartphone market, Samsung series smartphones dropped 20% in price three months after its release. 5This decline in prices was attributed to competitor models being released over time and, as a result, competition intensified.Previous studies have noted that becoming a "first mover" in a market enables firms to gain relatively greater rewards (Suarez & Lanzolla, 2007).Such studies have presumed that reinforcing internal R&D helps the acquisition of external technologies faster than competitors (Cohen & Levinthal, 1989, 1990).Meanwhile, collective intelligence (e.g., problem solving in collaboration with multiple external firms in OSCs, von Hippel & von Krogh, 2003) represents an important mechanism for enhancing the knowledge inputs of the firm, which have been shown to contribute to productivity (Hulten, 2010).Nagle (2016) referred to this effect as "learning by contributing" in contrast to the traditional "learning by doing" (Levitt & March, 1988).For example, Nagle (2016) revealed that contributions to a Linux OSC enable a firm to achieve a productivity of approximately 11%. 6 In addition to the relation between OSC contributions and productivity, this study further identified a link between OSC contributions and time-to-market advantage.
Further, Sony has made many source contributions but did not appear to be strictly pursuing time to market alone.Furthermore, Motorola and Sony make many source code contributions at the product software level (application and framework) in contrast to the hardware level (Hard Architecture Layer or HAL and Kernel) wherein Samsung and LG make many of their contributions pursuing time to market.Additional examination of this area is left to future studies.
version 2.3; Ice Cream Sandwich, version 4.0; and Jelly Bean, version 4.3.Furthermore, it examines the relation between (a) the number of source code contributions and (b) the time to market for each firm to release its first Android smartphone.Ten Android smartphone manufacturers with relatively high market shares (HTC, Huawei, LG, Samsung, Sharp, ZTE, Lenovo, Sony, Asus, and Motorola) were the participants of this study.(a) The source code change history (commits) was downloaded from the Android Open Source Project (https://source.android.com)on March 25, 2015, using "Git."Git is a software tool to control software versions.The number of contributions by each of the 10 firms in the Android OSC was calculated using the e-mail addresses of the developers of the source codes for each Android OS version.(b) The Google Nexus is co-developed by Google and specific handset manufacturers.This is the first model released with the most recent version of Android OS. 3 The time lags between the releases of the Google Nexus and the first Android smartphones released by each of the 10 firms were calculated and defined as the time to market.Release dates of each of the 10 Android smartphones were confirmed using product specification data from PDAdb.net 4 in August 2016.
the number of source code contributions to the Android OS version 2.2 (Froyo), 2.3 (Gingerbread), 4.0 (Ice Cream Sandwich), and 4.3 (Jelly Bean).While there were 85,650 source code contributions (counted by source code changes, i.e., commits) to Android OS version 2.2 (Froyo), there were 623,990 source code contributions to 4.3 (Jelly Bean), or 7.29 times more, due to the increasing complexity of the Android OS.While the number of contributions made by Motorola, Samsung, Sony Ericsson, and LG were relatively high, HTC, Huawei, Sharp, ZTE, Lenovo, and ASUS did not contribute much.

Figure 1
Figure 1 illustrates the relation between the level of source code contributions made by the 10 firms and time to market.For any

Figure1.
Figure1.Source code contributions and product releases of 10 Android smartphone manufacturersSource: Authors' analysis Figure 2 indicates the cumulative number of Android OS-based smartphones released by the 10 Android smartphone manufacturers for each version of Android.The total number of models released by these 10 firms for Android OS version 2.2 (Froyo), 2.3 (Gingerbread), 4.0 (Ice Cream Sandwich), and 4.3 (Jelly Bean) were 145, 333, 219, and 663, respectively.For each of these versions, the number of competing models dramatically rose over time.Accordingly, for Android smartphone manufacturers, releasing smartphones with the most recent Android OS version faster than competitors is of the utmost importance from a competitive perspective.

Figure 2 .
Figure 2. The model release of 10 Android smartphone manufacturers Source: Authors' analysis based on the data from PDAdb.net

Table 1 .
Major source code contributors co-developed with Samsung); and July 2012 for the Nexus 7 (version 4.3 (Jelly Bean), co-developed with ASUS) (https://source.android.com/source/code-lines.html).began contributing at the same level as Motorola from Android OS version 2.3 (Gingerbread).In contrast, HTC, ASUS, Sharp, Lenovo, ZTE, and Huawei have made few source code contributions.In other