How do software business models used in a software ecosystem differ from traditional software business models?

We figured out the following two sub-questions:

1. What are the differences between the business models of the four main mobile operating systems (Blackberry OS, Symbian, iPhone iOS and Android)?
2. What business models can be identified at 3rd party developers and/or partners?

The second part consists of a case study. We will compare four popular operating systems for mobile phones, namely: Blackberry OS, Android, iOS, and Symbian. We have chosen these four because they are currently the biggest players in the market. For each of those operating systems we will analyze quantitative market data and trends. Furthermore, we will provide a detailed analysis of the different business models we identified that companies currently use for their mobile operating systems. In addition, we will analyze the business models of third party developers. Therefore, a comparison between a software ecosystem orchestrator and the other organizations within a software ecosystem can be made.

Finally, we will try to find similarities between the different business models and compare them with traditional software business models found in previous literature.

• SRQ1: What types are the internal interactions between the different ecosystem’s member types?
• SRQ2: What types are the interactions between the Eclipse Foundation and the Eclipse Ecosystem’s member types?

Research Method:

The proposed research method is descriptive from a qualitative point of view. That is why it requires a substantial amount of data, regarding the Eclipse Platform and its members. In order to answer the proposed research question, the data collection process starts by a thorough examination of the Eclipse Foundation website (www.eclipse.org). After all the members are dumped, their respective websites are analyzed, in order to define relationships between different members of the Eclipse Ecosystem. All these relationships will be categorized according to the relationship’s type.

Due to the high volume and significant similarities of some of the members, the method suggests that a representative sample is selected and analyzed. The selection criteria are to be defined after an exhaustive analysis of the members of the Eclipse Ecosystem.

The aim of the proposed research method is to provide a descriptive visualization of the Eclipse Ecosystem, complying with the Software Supply Network (SSN) diagram technique. It should also visualize the important relationships between the members of the ecosystem themselves (excluding Eclipse Foundation).

Team Keystones

Background

At the moment, the Dutch product software industry is growing steadily (Dialogic, 2010). While the industry is growing, not all the research domains relevant for this industry seem to follow the same path of growth. Little research has yet been conducted on the current state of the Dutch product software industry with regard to supplier dependencies. Awareness of these dependencies is crucial for Dutch product software companies as earlier described by Boucharas, Jansen & Brinkkemper (2009). Heavily depending on a small number of suppliers for components used in software product creation can result in intimate relationships with a supplier from which the product software company can benefit. On the other hand, changes or decisions within this organization can have big consequences.

Problem Statement & Research Question

In order to make a contribution to this field of research, we are planning to do research on supplier dependencies within software ecosystems of Dutch product software companies. We will investigate supplier dependencies and the motives Dutch product software companies have to, for example, choose for a strong dependency on one supplier.  We will conduct this research by making use of empirical data gathered through a survey between September and November 2010. Twenty-five small to medium-sized software vendors completed this survey. The main prerequisite for a software vendor to qualify for participation in this survey was that their number of employees had to be at least ten. Each survey addresses three main topics within the product software industry that are related to the industry or the organization that is participating in this survey. Those three categories include the organizational structure of the software vendor, business and revenue models as well as a software ecosystems examination edged on the Software Supply Network and Product Deployment Context, including the existing dependencies. Given the research possibilities, by making use of the survey data as stated above, we will answer the following research question:

“What is the current state of the Dutch product software industry with regard to supplier dependencies within software ecosystems?”

Research Method

To contextualize the dataset, we will elaborate on various organizational characteristics such as organizational maturity and business and revenue models of the participants. Further contextualizing in this case also includes performing a quantitative analysis on the data, for which we will utilize SPSS. To provide an answer to the research question we will furthermore perform a qualitative analysis on the Software Supply Network (SSN) and Product Deployment Context (PDC) of the participants. The Software Supply Network and Product Deployment Context have been constructed for each participant and for one specific software product they develop and deliver. We will analyze these Product Deployment Contexts and Software Supply Networks of companies who completed the survey to examine supplier dependencies to ultimately identify common patterns. This will be done in the form of a pattern analysis. In addition, these patterns can be used to identify common pitfalls. Alongside answering the research question, the gathered data will also be used for comparison with previous research in the form of a benchmark. By doing so, we will provide further body of evidence for previous findings and will we be able to identify new future research challenges and opportunities.

References

Dialogic (2010). The Dutch software sector – Survey 2010. Utrecht, The Netherlands.

Boucharas, V., Jansen, R.L. & Brinkkemper, S. (2009). Formalizing software ecosystem modeling. In IWOCE ’09: Proceedings of the 1st international workshop on Open component ecosystems (pp. 41-50). New York, NY, USA: ACM.

The research will be qualitative in its essence as most of the data will be collected from application stores companies, and official websites of open source software companies and projects. We use triangulation for the gathering and analyzing of the data. This includes a combination of structured, human and observational techniques for recording key factors when browsing through the data online. Furthermore the observational technique will be direct as in the analysis stage e-mails will be sent to and interviews will be attempted with authors of open source applications where information online is not enough to determine how the business model works or is constructed. The research will also include a couple of case studies that show interesting business models relations between different open source applications.

The analysis spans over companies of different sizes, that satisfy the criteria of producing open source software, have a structured business model and have available information online, from which it can be determined the type and characteristics of the business model.

The assessment criteria of identifying the different business models are taken from Rajala et al.’s framework for analyzing software business models. The four dimensions (categories) of analysis are product strategy, revenue logic, distribution strategy and services and implementation. This research evaluates how components in each of the four categories shape relationships in the ecosystem of open source software.

References:

1. Rajala, R., Rossi, M., Tuunainen, V.K., (2003) A framework for analyzing software business models, Proc. 11th European Conf. Information Systems (ECIS 03)

Research proposal

For software companies it is of strategic importance to understand their own Software Ecosystem (SECO) (Berk, Jansen & Luinenburg, 2010). This applies to start-up companies as well. Being a software start-up choosing which ecosystem to join can have a big influence in the success of the start-up company. For example, would Rovio, the developer of the immensely popular Angry Birds game, have become just as successful if it joined a different ecosystem? We think that it is very important to be aware of the circumstances and conditions of an ecosystem, and to determine what the risk factors are in different SECOs. This results in the following research questions:

What are the characteristics that a start-up product software company should consider when joining a software ecosystem? How do these characteristics influence participants of different software ecosystems?

Proposed Research method

In order to answer the research question a literature research and documentation study will be performed. Based on existing literature about software ecosystems, market-related characteristics that influence participants in a SECO will be identified. Examples of such characteristics are: size of the market that the SECO addresses, entry barriers, quality requirements or information sharing degree.

The reason for choosing market-related characteristics is the nature of the research question. It implies an approach from the outside in to decision making, first analyzing the environment and then basing decisions on that analysis.

When we have formulated the market-related characteristics, we will conduct several case studies for different ecosystems. These case studies will be based on documentation study. These case studies will help to categorize the ecosystems based on characteristics.

Ecosystems that are to be analyzed include: iPhone, Android and Microsoft .Net. These ecosystems have been chosen because they each have a relatively strong keystone player that controls the platform, but these ecosystems differ in the way the keystone influences the ecosystem in terms of policies and platform openness. Information about the ecosystems will be gathered from websites, license agreements, press releases, market data and other publications.

Finally, we will analyze the results and draw some conclusion on how differences in ecosystems regarding market characteristics influences participants in the ecosystem. We will validate these results will by performing expert-interviews with a participant of each ecosystem. With these results, we hope to provide companies with a set of guidelines to be able to choose the best SECO for their specific needs and situation.

A group of people who share a concern, a set of problems, or a passion about a topic, and who enlarge their knowledge and expertise in this area by interacting on an ongoing basis.

Within the Android Ecosystem large communities of developers play an important role in the health of the Ecosystem. We propose to research the effects of new platform/SDK releases on the developer community. We expect that every major Android release has a revitalizing impact on the health of the community and thus on the ecosystem.

The goal of this study is to analyze the dynamics of an online software community in relationship with software releases from a software ecosystems perspective.

The main research question in this survey is: “How do software releases impact developers in an open source software ecosystem?”

In order to answer this question there needs to be a relationship between a software release (seen as an event) and the activity in the community. . A case study with statistical analysis should give insight in the strength of this relationship. This community should offer information about the members and their activities in a significant time range so that we have several releases and release-intervals to measure.

Data about software releases and community members and activity should be available for research purposes. At the time of this research the open source Android system seems to have a community capable of delivering the information needed to perform this research. The Android system is an open source mobile operating systems with eight major releases and dozens of small releases in the past four years. A notably large community is the “Android Community” at androidcommunity.com.  Some numbers on the community: Threads 37,234 Posts 333,370 Members 117,155 Active Members 8,501.

We use a custom-built tool to mine data from this community.

After the data is stored in a database an adjacency matrix is constructed that is required for the social network analysis tool UCINET. A relationship between actors is established when users post a message in the same thread.

Using social network analysis on a date interval on the dataset it is possible to create a representation of a before and after scenario of the community. Mathematical analysis, social network heuristic analysis, and visual analysis of social networks is used to analyze the effect of a software release.

Ideally we would graph the activity in the community and see peeks which pinpoint the Android platform releases.

 Problem statement

Leading web shops (e.g. Bol.com, Pixmania and Amazon) are increasingly integrating the products of third-party sellers in to their own web shop [1] [2]. This requires the transfer of information between two or more web shops. This way, web shops can present competitive offers on their website. By doing this, they are opening up the potential to serve the customer better by selling items outside their own product portfolio. By opening up their website the  leading web shops are transforming from a standard webshop into an open sellers platform. It is not clear if the collaborative webshops are ready for this trend or are able to act on it.

Research questions
Main question:

• What is the readiness of collaborative webshops towards open selling platforms?

Sub questions:

1. What information needs to be shared between an open sellers platform and the collaborative webshop?
2. In what way were the required IS/IT  and business processes of the collaborative webshop altered?
3. How do the capabilities of webshops compare to the requirements of a collaborative webshop?

Research method

two research methods were used. A case study is performed with a CW to determine the requirements and impact. And a survey is performed by sending questionnaires to web shops that haven’t engaged in this trend yet, in order to determine if they are ready for this trend.

Based on these two researches and by answering the three sub questions, we are able to determine the readiness of collaborative webshops towards open sellers platforms.

References

1. (2011, May). Erwin Boogert: Specialisten e-commerce reageren op Bol.com Plaza [Online]. Available: http://www.emerce.nl/nieuws/specialisten-ecommerce-reagere-op-bol-com-plaza
2. (2011, May). Erwin Boogert: Bol.com opent site voor de concurrentie  [Online]. Available: http://www.emerce.nl/nieuws/bol-com-opent-site-voor-de-concurrentie

Scope
The scope of this research will be limited to commercial software ecosystems. Since open source SECO’s are not inclined to have a central body governing and/or controlling the ecosystem. Commercial SECO’s however are characterized by such a central body, as mentioned above. Our research will focus on how strategic, tactical and operational [1][6] decisions, related to openness, made by such keystone parties affect the health of their ecosystem. Health measurements will be derived from den Hartog’s [5] paper on “The health measurement of a business ecosystem”.

Research question & hypotheses
How is the health of a commercial software ecosystem affected by the openness of the platform and how can the ecosystem leader influence it? H1: The degree of openness is positively related with the heath of a SECO. H2: Ecosystems leaders can directly influence the health of an ecosystem. Research method: Case studies Our research method will be a case study research, where a set of similar SECO’s will be chosen to compare how different strategies in openness [6] affect the SECO’s health. The SECO’s for our research will be from the mobile application industry. The mobile app industry is a thriving and dynamic environment in which a large amount of stakeholders cooperate and compete. However the respective SECO leaders vary in how they govern their ecosystem. We believe this variety makes this type of SECO interesting for comparison and will show interesting results on how their openness strategy influences the overall SECO’s health. For case study analysis we will derive techniques proposed by Yan-Ru Li in his analysis on the Cisco ecosystem [4] and retrieve data from Distimo [8] research reports. The case study comparison will be done on the following five SECO’s:

• Apple App Store for iPhone
• Google Android Market
• BlackBerry App World
• Nokia Ovi Store
• Windows Phone 7 Marketplace

Timeline
Our research will follow a strict timeline with activities and deadlines. The research will be performed within a span of nine weeks and the research team will meet at least twice a week where the research status and results will be discussed and activities will be divided.

References
[1] Marco Iansiti, Roy Levien. The KeystoneAdvantage What the New Dynamics of Business Ecosystems Mean for Strategy, Innovation,and Sustainability.

[2] Marco Iansiti, Roy Levien. Strategy as Ecology.
[3] Slinger Jansen, SjaakBrinkkemper and Anthony Finkelstein. Business Network Management as a SurvivalStrategy: A Tale of Two Software Ecosystems.
[4] Yan-Ru Li, The technological roadmap of Cisco’s business ecosystem.
[5] E.den Hartogh, The health measurement of a businessecosystem.
[6] Slinger Jansen, Sjaak Brinkkemper, Lutzen Luinenburg, and Jurriaan Souer. (2011). Shades of Grey: Opening up a Software Producing Organization with the Open Software Enterprise Model.
[7] L. Xu, S. Brinkkemper, Concepts for product software, European Journal of Information Systems 5 (2007) 531–541.
[8] Distimo, “Publications” Distimo. Available: http://www.distimo.com.

We are planning to do research on portfolio management within software ecosystems Portfolio management is a dynamic decision process in (software) product management, whereby a business’s list of active new product (and development) projects is constantly updated and revised [1]. In a software ecosystem a set of actors are functioning as a unit and interacting with a shared market for software and services, together with the relationships among them [2].

As research in (software) portfolio management focuses mainly  on the traditional software product lines, controlled and produced by one company, we are wondering how the shift from software product lines towards software ecosystems [4] has influenced portfolio management. Especially to what extend old challenges have remained or been solved, and to what extend new challenges have emerged. Scientifically, this will contribute to both the research domains, increasing the understanding and insights of these domains and their combination. In practice, this research will contribute to the understanding of the risks and difficulties (concerning portfolio management) that will be faced by companies who are switching, or considering to switch, from a traditional software product line towards an extended software ecosystem.

We have formulated our research question as follows:

“How do the challenges of portfolio management in large software ecosystems differ from those in software product lines?”

Obviously, this will mainly be an exploratory research, which will structure and identify (new) problems in the field of portfolio management and software ecosystems. It will be a combination of a primary and secondary research, i.e. a combination of summarizing existing research and collecting data that does not yet exist. The research will be performed in two steps:

First, a literature study will be conducted in order to create a thorough understanding on software ecosystems, portfolio management and the boundaries of these concepts. In addition, this section will function to position this particular paper in this research area. Beside an overview on the current state of research on the topic, this literature research will result in a conceptual solution to the research question. This solution would comprise known challenges for portfolio management in traditional software product lines, these traditional challenges that will most likely persist in software ecosystems and newly identified challenges for software ecosystems.

After this literature study, case studies will be performed to validate the found challenges on the practical experiences of a company, and to search for more possible challenges and approaches  on the portfolio management. Two or three companies will be examined to conform time limitations of the research, but still allow triangulation. The companies will preferable have changed their business model from software product line (SPL) oriented towards a software ecosystems (SECO) oriented approach recently. This way, much information on old and new challenges can be obtained from one case study.

The results of the literature and case study will be presented in the form of a table where both the portfolio management challenges for SECOs and SPLs are listed. Directions for future research will most likely be towards constructive researches, where a solution to the identified problems can be developed.

References:

[1] Cooper, R., Edgett, S., Kleinschmidt, E. (2001) Portfolio Management for new Products. Perseus Publishing, Philadelphia

[2] Jansen, S., Finkelstein, A., Brinkkemper, S. (2009) A Sense of Community: A Research Agenda for Software Ecosystems. 31st International Conference on Software Ecosystems, New and Emerging Research Track, 187 -190

[3] Pohl, K., Bockle, G., Van Der Linden, F. (2005) Software Product Line Engineering: Foundations, Principles and Techniques. Springer

[4] Bosch, J. (2009) From software product lines to software ecosystems. Proceedings of the 13^th International Conference on Software Product Lines (SPLC). Springer LNCS.