Social networking platforms (SNPs) are complex distributed software applications exhibiting many challenges related to data portability. Since existing platforms are propriety in design, users cannot easily share their data with other SNPs, however decentralisation of social networking platforms can provide a solution to this problem. There is a difference of opinion, the way the research and developer communities have pursued this issue. Existing approaches used in decentralisation provide limited structural detail and lack in providing a systematic framework of design activities. There is a need for an architectural framework based on standardised software architectural principles and technologies to guide the design and development of decentralised social networking platforms in order to improve the level of both data portability and interoperability. The main aim of this research is to develop an architectural solution to achieve data portability among SNPs via decentralisation. Existing proposed decentralised platforms are based on a distributed structure and are mainly for a specific aspect such as access control or security and privacy. In addition to this, existing approaches lack in practicality due to underdeveloped and non-standardised design. To solve these issues a new architectural framework is needed, which can provide design and development guidelines for the decentralised social networking platform. The goal of this thesis is to study, design and develop an architectural framework for social networking platforms that can incorporate the requirements of the decentralisation, to make portability possible. The synergies between the software engineering principles and social web technologies are investigated to create a standard approach. The proposed architecture is based on component-based software development (CBSD) and aspect-oriented software development (AOSD), a unified approach known as CAM (Component Aspect Model). The foundations of the proposed architecture are based on decentralised social networking architecture (DSNA), architectural style which is derived from CAM. Components and aspects are the building blocks of the proposed decentralised social networking platform architecture. From a development perspective, each component represents a social network functionality and aspects represent the properties and preferences that are used to decentralise the functionality. The model for the component composition is a major challenge because the use of CAM for social networks has not been attempted before. The proposed architecture comprehensively integrates the DSNA architectural style into each architectural component. Portability among SNPs by means of decentralisation can be summarised into three steps. (1) Definition of the architectural style, (2) implementation of the architectural style into components and (3) integration of the component composition. To date component composition approaches have not been used for social networks as a way to develop social network functionality. The concept of middleware has been adapted to achieve the composition feature of the architecture. In the architecture Social Network Support Layer (SNSL) functions as middleware to facilitate component composition. Existing middleware solutions still lack integration of CBSD and AOSD concepts. This limitation is characterised by, a lack of explicit guidelines for composition, a lack of declarative specification and definition model to express component composition and a lack of support for role allocation. This research overcome these limitations. The application of the architecture is based on the W3C SWAT (Social Web Acid Test) scenario. A Messaging application is developed to evaluate the scenario based on the Design Science Research Methodology. The architectural style is defined in the first stage of design followed by the component-based architecture. The architectural style is defined to guide the architecture and the component composition model. In the second stage, the design and implementation of composition technology (that is SNSL) are developed with architectural style and the rules defined in the first stage. The refined version of the architecture is evaluated in the third stage, according to WC3 SWAT test. The definitive version of the proposed architecture with the benchmarked result can be used to design and build social networking platforms, allowing users to share and collaborate information across the different social networking platforms.

Social networking platforms have grown to become the most popular websites on the Internet,and are the most widely used tool for job seeking nowadays. In this study the researchersinvestigate the job seekers perspective on what they perceive as the benefits of using socialnetworking platforms when used as a job seeking method. The benefits are how socialnetworking platforms are superior to traditional job seeking methods. The researchers refer totraditional job seeking methods as methods used before the popularization of social media,such as handing in paper resumes to employers and also the Internet before social media.Although the major focus is on the Internet which can nowadays be seen as the newtraditional method, before people started using social networking sites. This study thereforefocus on the transition from the Internet into Web 2.0 with social networking platforms andhow the labor market and job seeking methods has changed with this new informationtechnology. This study is a qualitative research study that is based on a hermeneuticperspective, an inductive approach and a cross-sectional research design. To be able to findanswers for this study’s purpose and research question, a utilization of in-depth semistructuredinterviews on five respondents have been conducted. What was found in the studywas that job seekers indeed perceive that there are benefits to using social networkingplatforms for job seeking, over traditional methods. What was also found was that the jobseeking process has not changed that much, although it has become more effective by beingavailable online and on social networking platforms. The perceived benefits that wereconcluded based on the analysis and discussion of the theoretical and empirical data were:globalization, more effective, easier networking, increased personal visibility and the accessto social networking platforms niched for specific professions.

垂直性社群平台將定位聚焦在特定主題，匯聚對特定主題有興趣的人。由於 使用者在不分眾社群平台與分眾社群平台之間，表現出多棲的行為。在Facebook崛起後的台灣社群平台市場，分眾是進入社群平台市場的機會點。然而，對平台企業而言，成功經營的關鍵挑戰是能夠以平台創造的價值建立收益流。因此本研究探討以下問題： 一、 垂直性社群平台如何獲取價值？ 二、 影響垂直性社群平台營收模式發展方向的關鍵因素為何？ 三、 驅動垂直性社群平台營收持續成長的關鍵策略為何？ 為回答上述三大問題，本研究採取多個案研究法，針對三家個案平台依序分 析其營收模式的建立、主要營收驅動力的作用過程、營收成長策略。從分析結果，本研究得到以下的結論： 一、 垂直性社群平台獲取價值的方式主要有兩種型態：其一是藉由掌握對大量 高同質性「社群成員邊」及「外部的資訊搜尋者邊」購買決策的影響力，吸引以此群人為目標客群的廠商，並與廠商在新市場進入策略或行銷或銷售等面向合作。其二是藉由對目標客群偏好需求的瞭解，推出實體商品。 二、 社群成員的特質是影響垂直性社群平台營收模式發展方向的關鍵因素。 三、 社群成員成長策略是驅動垂直性社群平台營收持續成長的關鍵。 / The position of a vertical social networking platform focuses on a specific topic, and members of the vertical social networking are interested in this specific topic. Due to multi-homing behavior of users between social networking platforms which focus on specific topics and social networking platforms which do not focus on specific topics, after the growing of Facebook in Taiwan, focusing a specific group gives platform-based companies opportunities to enter the social networking platform market. However, what the biggest challenges in running a successful platform is to establish revenue streams by the value that the platform creates. The present study tries to answer the following three questions: 1. How does a vertical social networking platform capture value？ 2. What are the key factors that affect the development direction of revenue model for a vertical social networking platform？ 3. What are the key strategies a vertical social networking platform use to drive revenue growth？ By analyzing the platform value capture mechanisms, this study concludes that: 1. Vertical social networking platforms capture value by two way mainly: first, on the basis of influence on consumer decision-making, vertical social networking platforms attract firms to cooperate on market entry strategies or marketing. Second, vertical social networking platforms could launch physical products based on understanding needs and preferences of users. 2. The common characteristic of the community members is the key factor that affects the development direction of revenue model for a vertical social networking platform. 3. The community member growth strategy is the key to drive community platform revenue growth.

In South Africa, many young children from poor social and economic backgrounds are cared for at home by parents or guardians who are themselves illiterate. This leads to poor educational outcomes later in life. Yet there are many privileged teenagers with access to mobile technologies who spend a greater portion of their spare time interacting on ubiquitous social media platforms. This presents an opportunity whereby the poor educational outcomes referred to previously could be addressed by applying a technology solution providing social media-based homework support by privileged teenagers to underprivileged younger children. However, most applications designed for use by children are designed by adults, with little understanding of the user requirements of the target end users. This research explores the following question: How can a cross-age tutoring system be designed for implementation on a social networking platform to support numeracy and literacy skill acquisition? The main contribution of this research was the definition of the Cooperative design by Children for Children (CD2C) Design Framework, a blueprint of how a cross-age tutoring system could be co-designed by children of different age groups and life circumstances. The CD2C Design Framework was derived as an abstraction of the second contribution of this research, the TitanTutor, an artifact designed using co-operative inquiry method and the Design Science Research approach. The third novelty of this research was contribution to Design Science Research theory, with the addition of new theory that states that cooperative design by children from different age groups and life circumstances is tempered by socio-environmental context and power relations between the co-design partners. This work provided important contributions to researchers in the areas of Cooperative Inquiry (CI), Human Computer Interaction (HCI), and Design Science Research (DSR). Future researchers could extend the CD2C Design Framework to make it even more abstract, thereby making it universally applicable to any co-design scenario. / Computing / Ph. D. (Information Systems)