The Good Left Undone: Advances and Challenges in Decentralizing Online Social Networks
Document Type
Article - Abstract Only
Publication Date
2017
Keywords
Online social networks, Decentralization, Survey
Abstract
Billions of users are now inter-connected in Online Social Networks (OSNs) and, as they interact with each other, massive amounts of potentially private data are collected at the OSN providers’ (e.g., Facebook or Twitter) premises. Unfortunately, provider-initiated privacy violations on this data are frequent and there is little chance that the providers will grant users effective data-protection means. To address these issues and to help users regain the control over their data, decentralized OSNs (DOSNs) have lately been introduced as a competitive paradigm to provider-controlled, centralized OSNs. DOSNs are built to function without the participation of a provider and with the intent to prevent any misuse of private user data. However, all proposed DOSNs still lack widespread adoption. While challenging the market-leading OSNs is difficult for many reasons, in this paper, we set out to understand the technical deficiencies behind the absence of a successful DOSN. We focus on the major technical challenge of DOSNs: they need to substitute the datacenter-based infrastructure of centralized OSNs. We first review recent advances in decentralizing OSNs based on how they approach that challenge. In a next step, we analyze the advantages and disadvantages each approach yields, and then derive a series of challenges that a successful DOSN will have to fulfill. Finally, we discuss options of moving forward in designing a new DOSN that could be successful in doing so.
Citation
Koll, D., Li, J., & Fu, X. (2017). The good left undone: Advances and challenges in decentralizing online social networks. Computer Communications, 108, 36-51. doi:https://doi.org/10.1016/j.comcom.2017.04.008
Comments
Principal Investigator: Xintao Wu
Acknowledgements: This material is partially based upon work supported by the National Science Foundation under Grant no. 1564348 and the Lindemann Foundation (Grant no. Az. 12-2016).