Facebook gives users a false sense of control over their data privacy. More recently, it has been evident to the general public that Facebook data might have been sold without any consent of the legitimate owners
limited scalability and high maintenance costs to manage data
Decentralization gives several benefits in terms of privacy
no central entity that has the control on all users’ data
existing terms of service,
exploiting different distributed models.
P2P architectures
several solutions for DOSNs
precursor of current DOSNs can be considered Diaspora
DOSNs, like Tent and Friendica, are based on similar concepts.
One of the main problems is to guarantee the availability of social data, in an environment characterized by a high level of dynamism. Another main problem is related to the development of techniques for propagating social updates in an efficient way. Finally, even if data is no more stored on centralized servers, new privacy issues have to be solved, for instance detecting trusted nodes that may host the profile of off-line users.
classification of the main architectural solutions for DOSNs. Then, we introduce existing approaches for guaranteeing data availability, and the main techniques used to spread social content among the users of the social network.
SN level provides the common social network functionalities, such as chat, mails, wall posts and/or tweets, etc…,
SNS is usually implemented by a P2P network,
CT level consists of Internet, mobile or opportunistic infrastructure
Social Network (SN), a Social Networking Service (SNS) and a Communication and Transport (CT) level
focus of this survey is mainly on the challenges associated to the Data Management Services for DOSNs. With this term we refer to the set of functionalities needed to guarantee data availability, the diffusion of social information, and the protection of social data.
Data availability/persistence
presence information
high level of dynamicity of DOSNs infrastructures makes the problem more challenging
social relationships can change due to the addition and/or the removal of relations
number of users of the DOSN, the structure of the social graph changes during time.
structure of the underlying overlay
Information diffusion
update from one user must be notified to all its direct social contacts and beyond (i.e. friends of friends), depending on the type of social network.
ge users’ updates, i.e. status updates, published posts, comments and so on. Each update from one user must be propagated to all its social connections in a scalable and efficient way.
Privacy
Data can be stored not only on the node of profile owner, but also on others which can be both known and unknown nodes
characterized by different levels of privacy.
Encryption is usually used in DOSNs to keep control over social contents.
Scalability
Mapping a social graph onto a distributed network can be very expensive due to the number of social links for each node, so the cost of mirroring the social network links into distributed network links can be high, and it can be very inefficient due to the high number of inactive links
Topology
information about social relationships is important and nodes may be connected according to their social relationships
finding an overlay topology that exploits social relationships and, as the same time, overcome previous limitations, is a current challenge
Note: Missing in Mastodon, replicas are only realized through personal backups
SO-based systems, each user has a set of replicas, but it keeps its data until it is online and transmits them to the requesting friends through the social overlay. When the user is offline, the set of replica nodes are responsible to manage data on behalf of the owner.
social overlay (SO) (SO-based system), where a logical connection between a pair of nodes corresponds to a friendship relationship
Distributed Hash Table (DHT-based system), whose nodes are those of the users participating to the social network, where the overlay connecting the peers is defined by the specific DHT topology
Physical locality
novel challenges, for instance taking advantage of users’ close proximity for the definition of ad-hoc virtual communities or detecting the periodic physical locality patterns of the users to improve their knowledge about opportunities to socialize.
The Diaspora network is a network of independent, federated servers, that are administrated by users, which are used for storage, communication and access control. Users decide in which servers their information will be stored, and some users choose to maintain their own servers in order to keep complete control of their data.
During the last years, DOSNs have been considered a new paradigm to implement social networks. However, despite the fact that Diaspora (one of the first and most popular examples of DOSN) has collected more than 400,000 users, other DOSNs have still limited impact. The main problem is that a few centralized social service providers like Facebook, managed by big companies, have dominated the OSNs market, attracting millions of users by offering advanced functionalities, and, consequently, people are reluctant to leave a network including so many contacts and advanced functionalities. Although this might change due to the recent emergence of huge data leaks, this problem does not seem so pressing for most users of Facebook, or other centralized OSNs.
Groups and Communities management
main problems that limit the diffusion of DOSN
Groups and communities consist of an aggregation of users that share some common interests and their management is considered of great interest today. People join a community because they care about the common interests that glue the community members together. Some of them may join because they feel the urge to contribute to a cause sustained by the community; others join because they can benefit from being part of the community. Similar to a community, a group is a feature of many OSNs which allows users to share common interest. The management of groups/communities of users in DOSNs, still rises several challenges, in terms of information diffusion, and data management due to the dynamism of DOSNs. Indeed, there is the need of specific algorithms to manage communities or groups in dynamic environments, as described in . A first contribution in this direction is proposed in , where a distributed protocol to detect and manage communities in a DOSN is proposed
Privacy issues
Current DOSNs offer several mechanisms to address the privacy issues. However, there are still open problems. For example that of the impersonation attacks or the privacy
breaches coming from third-party applications accepted by users to get additional functionalities.
Monitoring the evolution of the social network
OSNs offer advanced functionalities to search and add new users to the friend list, like link prediction . This is a useful tool both to understand the evolution of the network and to suggest relationships that may appear in a future network configuration. The prediction of a link (such as the link existence, type, its associated weight, etc.) is a classical problem in complex networks analysis . However, DOSNs currently do not provide functionalities to suggest new friendship relations, because of the lack of distributed algorithms for the prediction of links.
Mobile communication
Note: Mastodon has plenty of mobile client apps.
However, among existing DOSNs, only a few provide support for mobile access
Trust vs. Encryption
Most existing DOSNs use classic encryption techniques (symmetric or asymmetric encryption), to guarantee privacy and anonymity. Encryption adds some overhead in both the time and space domains
An alternative solution is to avoid encryption by storing content, in clear, on trusted nodes. Trust is widely accepted as a major component of human social relationships and is a characteristic of real life where people rely on trusted persons.
The automatic evaluation of trustworthiness can be considered still an open problem strictly tied to the evaluation the nature of an interaction of two users.