Enhancing Privacy, Censorship Resistance, and User Engagement in a

Blockchain-Based Social Network

Myo Thiha

1 a

, Halil Yetgin

1 b

, Luca Piras

1 c

and Mohammed Ghazi Al-Obeidallah

2 d

Department of Computer Science, Middlesex University, The Burroughs, London, NW4 4BT, U.K.

Department of CS and IT, Abu Dhabi University, Abu Dhabi, U.A.E.

Keywords:

Blockchain, Decentralised Social Network, Privacy, Censorship, User Engagement.

Abstract:

In the contemporary digital era, online social networks have become integral to global communication, facil-

itating connectivity and information dissemination. Millions of people use centralised social media platforms

today, which raises concerns about user control, privacy, and censorship. These platforms proﬁt from user data

and content, as the single authority of these platforms has complete control over user data. Although peer-to-

peer decentralised online social networks were developed to address the weaknesses in centralised platforms,

they still have signiﬁcant limitations in terms of securing privacy, and handling censorship resistance issues. In

this work, we propose a novel decentralised online social network leveraging blockchain technology to address

these pressing issues in centralised and peer-to-peer online social networks. The proposed system prioritizes

user control by decentralizing data storage and network governance, thereby reducing the issues associated

with centralized control. By employing blockchain technology, individuals maintain ownership of their data

and gain greater control, thereby enhancing user privacy protection. Additionally, the cryptographic security

and immutable ledger of blockchain technology protect freedom of expression and information exchange by

resisting censorship. Moreover, with the integration of incentivization mechanisms, users are incentivized to

contribute to the network’s growth and sustainability, as well as promoting engaging content and encouraging

ownership among users. The evaluation results show that our blockchain-based decentralised online social

network (DOSN) accomplishes the aim and objectives for preserving privacy, censorship resistance and en-

hancing user engagement in online social network with the use of blockchain technology.

1 INTRODUCTION

Nowadays, social networks are the most common

platforms on which people share their personal in-

formation. There are over 4.5 billion Internet users

worldwide, and over 4 billion people utilise social

networks. In the upcoming years, social networks

will continue to grow and bring in more users. More

signiﬁcantly, they have impacted how people inter-

act with one another and communicate. However,

this impact is not always positive (Mohammed, 2022).

The business strategy of these platforms is based on

providing users services for free in an attempt to at-

tract more users. In return, users’ behaviour and ev-

erything they share on the platform are used by the

https://orcid.org/0009-0002-3684-2093

https://orcid.org/0000-0003-0994-5839

https://orcid.org/0000-0002-7530-4119

https://orcid.org/0000-0003-4976-1380

owners in order to make proﬁt through selling ad-

vertising (Guidi, 2021). Therefore, the increasing

dominance of major industry players such as Meta

and Twitter has led to the suppression of opposing

viewpoints and discussions about user privacy due

to the vulnerability of personal information to data

breaches and misuse, as well as the centralisation

of control over user interactions and content (Yerby

et al., 2019). Furthermore, the centralization of power

in a small number of organisations raises concerns

about responsibility and the possibility of improper

inﬂuence over public opinion. One of the primary

causes of these issues is because of the dependence

on the client-server architecture with proprietary and

centralised servers (Tran et al., 2016).

In an attempt to address this issue, researchers

have suggested alternative platforms. Over the past

ten years, various decentralised online social net-

works have been developed. These networks are built

on various peer-to-peer system architectures. How-

Thiha, M., Yetgin, H., Piras, L. and Al-Obeidallah, M. G.

Enhancing Pr ivacy, Censorship Resistance, and User Engagement in a Blockchain-Based Social Network.

DOI: 10.5220/0013238500003928

In Proceedings of the 20th International Conference on Evaluation of Novel Approaches to Software Engineering (ENASE 2025), pages 67-79

ISBN: 978-989-758-742-9; ISSN: 2184-4895

ever, due to the limitations and lack of success of peer-

to-peer decentralisation techniques, blockchain tech-

nology has been considered as a decentralised alter-

native in a number of research areas to address chal-

lenges arising from centralised and peer-to-peer de-

centralised social networks.

Blockchain is a distributed, decentralised ledger

that operates without a central authority to facili-

tate secure and transparent transactions. Although its

main application has been in the ﬁeld of cryptocur-

rency, its prospective beneﬁts are not limited to this

sector. The application of blockchain technology to

social network has been investigated recently in an

effort to solve the problems with current centralised

platforms. Decentralised social networks provide an

alternative to traditional social network by promoting

trust and transparency in online interactions, ensur-

ing privacy and communication without restriction or

censorship, and giving users more control over their

personal information (Zhan et al., 2022).

The principal aim of this paper is to conceptu-

alise, design, and implement a decentralised online

social network application by integrating blockchain

technology. The integration aims to enhance privacy,

provide users with greater control over their personal

data, and resistance against censorship, as well as pro-

mote an engaging and democratic online community.

This will alleviate a great deal of relief from many

data-related issues of traditional centralised online so-

cial networks. Moreover, the proposed application

aims to grow user communities and engaging con-

tents by incentivising them monetarily for their con-

tributions in sharing valuable content. This results in

highly personalised feeds and more engaging content.

By means of this aim, it will signiﬁcantly contribute

to the decentralisation of social networks, as well as

user-centred online communication platforms.

In line with the aim outlined above, the following

research questions (RQ) are individuated:

RQ1. How does the blockchain architecture ad-

dress potential privacy concerns in centralised OSNs

and peer-to-peer decentralised online social networks

(DOSNs)?

RQ2. How does the decentralised nature of

blockchain-based DOSN provide the censorship resis-

tance?

RQ3. How can the blockchain-based DOSN en-

hance the contribution of engaging content than cen-

tralised OSNs and peer-to-peer DOSNs.

To answer the research questions, we performed

different phases: (i) decentralisation with blockchain

technology. Initial research steps with literature re-

view for privacy, censorship-resistance and user en-

gagement in decentralised online social networks and

how blockchain combined with other decentralised

technology could offer a more efﬁcient solution. (ii)

Requirements Analysis and Design. Based on the in-

formation collected from the literature review, func-

tional and non-functional requirements are analysed

and the design of the system architecture is cre-

ated. (iii) Implementation. Based on the requirements

and design, the actual implementation of blockchain-

based DOSN application is conducted. (iv) Evalu-

ation. The viability and effectiveness of the imple-

mented blockchain-based DOSN application is evalu-

ated via observation against the research questions.

The rest of the paper is organised as follows: Sec-

tion 2 describes the literature review of background

and initial research steps (relating to point (i) above).

Section 3 describes the requirement analysis and de-

sign for the application as well as the actual imple-

mentation process by using blockchian and related de-

centralisation technology (point (ii) and (iii) above).

Section 4 describes the evaluation via observation

against the research questions, and related works.

Section 5 concludes the paper.

2 BACKGROUND AND

LITERATURE REVIEWS

In this section, we will discuss the relevant litera-

ture which other researchers have done in the ﬁeld

of decentralised online social networks and explain

the initial research steps on how blockchain technol-

ogy could offer a more efﬁcient decentralised solu-

tion. First, we will discuss the overview of the his-

tory of online social networks, along with the evo-

lution of peer-to-peer decentralised online social net-

works using speciﬁc examples. Second, we will ex-

amine the pertinent studies that have been conducted

on the subjects of preserving user privacy, and resis-

tance to censorship in decentralised online social net-

works with various architectures, outlining the limi-

tations and challenges of each. Next, we will depict

the evolution of blockchain-based decentralised on-

line social networks. Finally, we will explain the core

features of our blockchain-based decentralised DOSN

that, in comparison to previous peer-to-peer decen-

tralised online social networks, is more effective at

addressing the issues of user privacy, and censorship

resistance arising from centralised OSNs, as well as

enhancing user engagement.

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

2.1 History of Online Social Networks

and Their Problems

A new era of networked communication was ush-

ered in 1991 when Tim Berners-Lee successfully inte-

grated hypertext applications with the Internet, which

led to the development of the World Wide Web. In

addition to revolutionizing the development of on-

line communities, this technology provided ofﬂine

groups with support. The launch of websites includ-

ing SixDegrees.com, Classmates.com, and GeoCities

in the middle of the 1990 indicated the beginning

of innovative social networking sites. Online social

networks, or OSNs, have become extremely popular

since the mid-2010s. Large corporations such as Meta

and X have emerged, allowing users to share text mes-

sages, digital images, and videos with their friends.

Meta is a well-known online social network that ap-

peals to users of all ages, and currently, it has over

2.9 billion members. OSNs have become an essential

component of many users’ life (Shilina, 2023).

However, it is also discovered that user privacy

in those OSNs is readily compromised. While most

OSNs allow users to adjust privacy settings to prevent

other users from seeing their data, there are no practi-

cal technological ways to restrict OSN providers’ ac-

cess or prevent them from sharing user data with third

parties. Massive amounts of personal data, such as

demographics, interests, and online activity, are gath-

ered by dominant social networks from their users;

this data is frequently breached by attackers. In 2020,

almost 25% of records that were exposed came from

database breaches on social networking sites such as

Facebook and Twitter (Shilina, 2023). But since then,

the problems have become worse; in 2022, the per-

centage increased to almost 41%, showing a substan-

tial increase in data breaches and social networking

platform theft. Censorship is another issue with the

centralised OSNs that exist currently. Facebook, for

instance, prohibits certain content that does not ad-

here to the platform’s policies (Shilina, 2023). Nev-

ertheless, the fact that a single authority determines

these rules compromises users’ freedom of speech.

2.2 Evolution of Peer-to-Peer

Decentralised Online Social

Networks

In order to accommodate these issues in centralised

OSNs, DOSNs have been explored. A DOSN is

typically implemented by a peer-to-peer (P2P) man-

ner, whereby independent users collaborate together

to form a platform architecture with the goal of

empowering them to have greater control over the

information they store and can access. According

to the survey of (Schwittmann et al., 2013), they

have distinguished three main categories of P2P

architecture in DOSNs for preserving privacy, user

control and censorship resistance.

Federated Architecture, which is made up of sepa-

rate servers, is the most well-known P2P architecture

for DOSN. With federated networks, users can

choose which server to register with to gain access to

the full network, which is dispersed among numerous

servers. It allows content sharing with particular

contacts or contact groups through a ﬁne-grained

access control system. Secure sockets layer/transport

layer security (SSL/TLS)-like security is provided

via the server-to-server protocol, which regulates

the transfer of encrypted and signed messages.

By facilitating decentralised cooperation among

separate groups, this architecture preserves user

autonomy and privacy while giving users control

over their data. Furthermore, as the network is

federated and lacks a single point of control, it

is impossible for a single body to censor or man-

age the entire network, offering censorship resistance.

End-to-End Client-Side Data Encryption Archi-

tecture, which implements encryption on the end

hosts to protect data content privacy from curious or

hostile service providers. In a centralised OSN, the

server is responsible for enforcing access control lists.

When using user-to-user encryption, access control

via cryptography falls under the responsibility of the

end host. Persona, Vegas, and SoNet are examples

of such types of social networks. To begin with, the

Persona network employs an encryption approach

called attribute- based encryption (ABE), which

permits the sharing of the symmetric content key

with speciﬁed groups or sets of groups. Compared

to conventional hybrid techniques, which encrypt the

group key for every recipient, this is more efﬁcient.

Persona conceals metadata from the general public

as well, although it still permits server providers to

view social network graphs (Baden et al., 2009a).

Next, to make the process more convenient for users,

the Vegas social network leverages QR codes for

key exchange (Paul et al., 2014). Additionally, it

describes a coupling technique that enables a user

to ensure the accuracy of two of his trusted friends’

identities by launching friendship authentication

between them. Vegas hides the social graph with

a rigorous approach. It employs public servers’

storage with random names of ﬁle and concealed

the structures of directory, and it does not let users

Enhancing Privacy, Censorship Resistance, and User Engagement in a Blockchain-Based Social Network

view their friends’ contact lists. By doing this, the

social graph is successfully conceled from friends

and server providers. Lastly, SoNet uses aliasing for

hidding the social graph from server providers. Users

only communicate to their storage provider, which

also serves as a proxy for accessing the storage of

other users. In order to prevent the storage server

from resolving aliases of other storage services to

cleartext usernames, both parties generate random

IDs when establishing acquaintances (Schwittmann

et al., 2013).

Distributed Hash Table Architecture, which is

utilised to search data in a P2P system with loga-

rithmic routing complexity. By distributed user data

among a network of peers, each of whom is in charge

of a portion of the data, this architecture is used in

PeerSoN, LifeSocial, and Cachet decentralised social

networks to assure privacy and censorship resistance

(Schwittmann et al., 2013). In a peer-to-peer system

with logarithmic routing complexity, PeerSoN em-

ploys a distributed hash table (DHT) to locate data.

Every participating peer and every data object to be

stored is assigned an ID within the DHT key space.

Each peer is in charge of the data stored in a speciﬁc

area of the key space, which includes the IDs that are

closest to them. This strategy makes sure that author-

ity and responsibility are distributed among a diverse

group of peers. Next, all user content in LifeSocial

is stored using a DHT. References to other data items

or the actual payload data itself can be contained in

data objects (Grafﬁ et al., 2011). Finally, Cachet

hides the recipients of data items by using object ref-

erences. The public keys are stored in encrypted form

at their parent objects rather than with the data objects

(Nilizadeh et al., 2012). Only friends with permission

can view the receivers of the referenced items when

they are given a well-known object as an entry point

(Urdaneta et al., 2011).

2.3 Limitations and Challenges of P2P

DOSNs

Each of the three architectures — federated server,

end-to-end client-side data encryption, and dis-

tributed hash table — has limitations and challenges

speciﬁc to its implementation and use.

2.3.1 Limitations of Federated Server

Architecture

Even though the architecture in federated server is

decentralised, individual user data is not always re-

tained on the server where it was initially stored. If

friends from other servers ask for this information,

it will be transferred to their servers and made ac-

cessible in plaintext to the service provider. Users

must therefore have trust in both their friends and their

own service provider (Bielenberg et al., 2012). More-

over, as each user is uniquely identiﬁed, the service

provider can see every interaction and can use this in-

formation to determine the social graph of the people

it hosts. Therefore, only having a theoretically de-

centralised architecture won’t be sufﬁcient to protect

user data privacy in a single location (Schwittmann

et al., 2013). Even with server-side imposed ac-

cess control and user data distribution across multiple

servers, users remain susceptible to privacy attacks.

Furthermore, users need to have trust in both the ser-

vices they receive from friends and their own service

provider (Schwittmann et al., 2013). Users have less

control over their data since user privacy depends on

providers not disclosing user data without authoriza-

tion.

2.3.2 Limitations of End-to-End Client-Side

Data Encryption Architecture

World-readable encrypted data stores are nonetheless

susceptible to metadata leaks, including timestamps,

data object sizes, data structures, and header informa-

tion. This enables the inference of social graphs and

reveals trends of user behavior. Third-party attackers

may be able to obtain sufﬁcient information from the

metadata linked to the data objects even though the

content is encrypted (Greschbach et al., 2012). The

computational cost of the encryption procedures is

high, particularly when using more intricate systems

such as Attribute-Based Encryption(ABE). ABE op-

erations might be 100–1,000 times slower than RSA

public-key method operations (Baden et al., 2009a).

This can be a major performance bottleneck, espe-

cially on mobile devices. Discussion groups and com-

ments that may be viewed by other contacts are not

supported by some DOSNs, such as Vegas, which

concentrate on obscuring the social graph. This re-

striction results from the requirement to avoid com-

munication types that disclose interpersonal connec-

tions, which forces a trade-off between functionality

and privacy functionality (Schwittmann et al., 2013).

2.3.3 Limitations of Distributed Hash Table

Architecture

Each peer in a DHT is in charge of the informa-

tion stored in a certain area of the key space. This

may lead to an unfair situation where some peers

have far more hard drive space allocated to them

than the actual amount of data they store in the DHT

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

(Schwittmann et al., 2013). Data objects can contain

pointers to other data objects or actual payload data in

DOSN such as LifeSocial (Grafﬁ et al., 2011), which

keeps all user content in a Pastry DHT. However, ev-

ery reference must be resolved in a DHT query, which

is typically transmitted to different peers. Since the

assignment of data objects to peers seems to be ran-

dom, neither authorship nor network proximity are

taken into consideration. This may impact the re-

trieval of data, hence decreasing its efﬁciency. Peers

must always communicate with one another accord-

ing to the DHT architecture, especially while they are

signing on and off. When the DHT is employed as

payload storage rather than just a data index, there is

an overhead in comparison to server federations. As

a result, there is a chance that performance will be

slower and network trafﬁc will increase (Schwittmann

et al., 2013).

2.4 Blockchain-Based DOSN and Main

Features

Recent years have seen a major transformation to-

wards a more decentralised and user-powered inter-

net ecosystem. Distributed ledger technologies, such

as blockchain, can be utilised for addressing privacy

and censorship concerns associated with centralised

OSNs. The integration of blockchain technology into

online social networks has garnered signiﬁcant atten-

tion from researchers in recent times, due to its rapid

advancement and enduring popularity. In an attempt

to improve privacy, and censorship resistance while

addressing the shortcomings of early P2P DOSNs,

several research projects are investigating a new gen-

eration of decentralised social networks that incorpo-

rate blockchain technology.

A blockchain is an expanding distributed ledger

that maintains an unchangeable, secure, and chrono-

logical record of every transaction that has ever oc-

curred. A person or group going by the name

of Satoshi Nakamoto created the ﬁrst block of the

blockchain, known as the gensis block, in 2009 and

employed it to create the cryptocurrency known as

Bitcoin. The main goal is to protect electronic ﬁles

from manipulation by encrypting their bit sequence

using a cryptosystem. Blockchain is one of the most

innovative and promising technologies that powers

decentralised applications by giving the network a

strong infrastructure. Overall, blockchain is an im-

mutable distributed ledger that records all network in-

teractions and transactions (Yaga et al., 2018).

The main features of the blockchain based DOSN

are explained as follow:

User Privacy - Online social networks built on

blockchain technology eliminate a single point of

failure due to decentralisation. In contrast, centralised

social networks are susceptible to data breaches be-

cause they primarily rely on centralised servers. In

fact, because blockchain technology is decentralised,

it removes the possibility of control by a single entity.

Additionally, because all transactions are encrypted

and tracked, user data cannot be tampered with,

guaranteeing its integrity and conﬁdentiality (Guidi,

2021).

Censorship Resistance - One of the most signiﬁcant

problems with centralised OSNs is censorship. Users

speciﬁcally oppose the restriction of sensitive topics

in order to maintain their freedom of speech and

expression. On the other hand, content in centralised

OSNs is reviewed and removed if it violates speciﬁed

guidelines. Blockchain-based DOSNs, in contrast

to centralised ones, let users publish and exchange

content without worrying about censorship. Content

that has been stored on the blockchain is irreversible

and cannot be changed or deleted by a central entity.

This encourages freedom of speech and offers a po-

tential solution for the censorship issue (Guidi, 2021).

Reward for Contribution in Engaging Content

- In blockchain-based social networks, a content

creator or a regular user might be compensated with

incentives for high-quality content. All transactions

are documented and audited by everyone, making the

rewarding phase transparent thanks to the blockchain.

In addition to being one of the main characteristics of

a blockchain-based DOSN, rewarding is also thought

to be essential for success in adding value to content

and creating economic models (Guidi, 2021).

3 ANALYSIS AND DESIGN

In this section, the analysis and design phases per-

formed for the blockchain-based DOSN applica-

tion will be discussed. Firstly, it will discuss the

blockchain and its related technologies. Secondly, the

system architecture and prototype of the application

will be discussed. Finally, it will explain the algo-

rithm and mechanism which are implemented to ad-

dress the user control, privacy, and censorship resis-

tance limitations in centralised and P2P decentralised

OSNs.

Enhancing Privacy, Censorship Resistance, and User Engagement in a Blockchain-Based Social Network

3.1 Blockchain-Based Decentralised

Technologies

Blockchain is a unique data structure that stores data

across numerous nodes or computers in the form of a

sequence of blocks. According to Figure 1, the initial

block, known as the genesis block, has its data trans-

formed into a ﬁxed-length hash value using a hash al-

gorithm. The following block then stores this hash

value. Every time a new block is generated, this pro-

cess is repeated, creating a ﬁxed-length hash value

based on the data in each block, which includes the

hash value of the block that preceded it, to be stored

in the subsequent block. The blocks are chained in

this way, so any changes made to one will also af-

fect all of the blocks that come after it. As a result,

once recorded data is placed on the blockchain, it can-

not be changed because the original blockchain is im-

pervious to manipulation. Blockchains fall into three

primary categories, according to (Guidi, 2021) Public

Blockchains (everyone can join the network), Private

Blockchains (members are selected subject to certain

criteria), and Consortium Blockchains (semi private

blockchains restricted to a group). Furthermore, per-

missioned (private blockchain), permissionless (pub-

lic blockchain), or both (consortium blockchain) can

be used for all three types. Therefore, blockchain

is characterised by audibility, privacy, conﬁdentiality,

consistency, decentralisation and integrity.

3.1.1 Ethereum Blockchain and Smart Contract

Ethereum’s blockchain is particularly appropriate for

implementing blockchain-based social networks due

to a number of signiﬁcant features and beneﬁts. To

begin with, Ethereum is decentralised, meaning that

no single entity can control the network as a whole.

Because of this, it is resistant to interference and cen-

sorship, which is a major beneﬁt for social networks

where freedom of speech and expression are essen-

tial. Second, social networks built on Ethereum could

provide users with better privacy and more control

over user data. It is impossible to gather and utilise

personal data without permission in the absence of a

third-party central authority. All data is stored in sep-

arate blocks as unique immutable hashes, consider-

ably decreasing the risk of data breaches and identity

theft. Furthermore, social networks and other sophis-

ticated applications can be created with Ethereum’s

smart contracts. These contracts can automate a num-

ber of tasks, including user transactions and content

monetization, among others. In addition, the interop-

erability of Ethereum facilitates data sharing and in-

teraction between various apps, building a more inte-

grated and effective ecosystem. Finally, Ethereum en-

ables the generation of native tokens that can be em-

ployed for a variety of purposes inside the social net-

work, such as incentivising individuals for their con-

tributions or facilitating network transactions. This

may encourage user involvement and engagement.

Due to the aforementioned reasons, the Ethereum

Blockchain is selected for our decentralised online so-

cial network based blockchain network.

3.1.2 InterPlanetary File System (IPFS)

Interplanetary File System (IPFS) is a decentralised

protocol to address, route, and transfer content-

addressed data in a decentralised network. The way

it operates is by using a hash to identify ﬁles and

enabling a peer-to-peer network to locate the near-

est copy of the ﬁle using that hash. Data is repre-

sented by IPFS as content-addressed content identi-

ﬁers (CIDs), which are speciﬁc to the data they were

computed from. This facilitates the retrieval of data

by considering its content instead of its location. The

protocol employs interplanetary linked data (IPLD)

to operate with CIDs to represent relationships be-

tween data. Blockchain-based social networks can

be implemented using IPFS thanks to its features,

which include data control, portability, and integrity.

It can be used to store messages and posts in a de-

centralised online social network, allowing for decen-

tralisation and censorship resistance. Furthermore, in-

tegrating IPFS with blockchain technology can facil-

itate the management of encryption/decryption keys

and the recording of associations of user credentials,

both of which are critical for developing secure so-

cial networks (Xu et al., 2018). In our blockchain-

based DOSN, all the user’s media data such as im-

ages, and videos will be stored on the IPFS, and only

the hash value returned from the IPFS is stored on

the blockchain. In this way, it allows the application

to become more scalable, and reduces the storage de-

mand on the blockchain. In other words, by lever-

aging Ethereum’s smart contracts, IPFS enhances se-

cure and cost-efﬁcient storage solutions within the

blockchain ecosystem, thereby boosting the overall

performance of Ethereum.

3.2 System Architecture and Smart

Contract Design

3.2.1 System Architecture

The proposed blockchain-based DOSN application

forms a three-layer architecture according to the

Figure 2.

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

Figure 1: Blockchain consisting of a continuous sequence of blocks.

Figure 2: Three-layer system architecture encompassing presentation, application and data layers.

Presentation Layer - The presentation layer of

the application is made up of its user interface. To

provide rapid and consistent access to the applica-

tion, a decentralized content distribution network

(CDN) will serve the user interface. Through the

user-friendly and intuitive interface, users can easily

interact with features and functionalities of the

application.

Application Layer - Application Programming Inter-

face (API) requests from the frontend application to

the backend blockchain and IPFS storage are handled

by the Web3 and Pinata IPFS APIs at the application

layer. User input is sent to the appropriate Web3

API and Pinata IPFS API, including user proﬁles and

post data from the frontend client. The underlying

blockchain system will then communicate with the

Web3 API to provide function calls, contract deploy-

ment, and fund transfers. When a user creates a post

with an image or video, the Pinata IPFS API will up-

load that image or video ﬁle to the IPFS cloud storage

and return only the IPFS hash value to the blockchain.

Data Layer - The application’s data layer consists of

a decentralised storage network that stores user data,

post content, and social interactions. The storage

network, which will be accessible via the Ethereum

blockchain, will make use of an IPFS decentralised

protocol to guarantee data consistency and availabil-

ity. To protect user privacy and control, data will be

stored in an encrypted format with the user’s private

keys.

Enhancing Privacy, Censorship Resistance, and User Engagement in a Blockchain-Based Social Network

3.2.2 Smart Contract Design

The smart contract that is deployed on the Ethereum

blockchain for this application consists of two main

structs, which is an object data structure in Solidity

programming language, in order to store information

about user, and post. First of all, the ”Proﬁle” struct

will store information about each user including their

account address, username, biography, hash value re-

turned from the IPFS storage for the proﬁle picture,

the time at which the account is created, unique Id,

total number of posts created, followers, followings,

account addresses of followers, and followings. Sec-

ondly, the ”Post” struct will store information about

each post including post id, creator’s account address,

hash value returned from the IPFS storage for the pro-

ﬁle picture of the creator, post type, hash value re-

turned from the IPFS storage for the image or video

ﬁle contained in the post, description of the post, the

time at which the post is created, the total tip amount,

likes and dislikes.Next, the smart contract includes

two arrays: one is for the list of users who created

account on the application and the other is for the list

of posts created by users. Also, it includes four map-

pings, which are similar to a hash-table or dictionary

data structure in other programming languages. They

are used to store the data in the form of key-value

pairs. The ﬁrst mapping “proﬁles” is to map from the

user account address to related “Proﬁle” struct. This

is to make sure that each user account is registered

only once and to quickly search user information by

address, similar to locating a person by ID in the re-

lational database. The second mapping “posts” is to

map from the account address of the post creator to

the array of related “Post” struct so that it stores the

list of related posts that the speciﬁc user created. The

last two mappings: “likedPosts” and “dislikedPosts”

are to map from the account address of the user to the

array of related post IDs that are liked or disliked by

that user. In this way, it can store separately the list

of related posts that the user liked or disliked by the

account address.

3.3 Algorithm and Mechanism

The login and authentication process in the applica-

tion is validated with the user Ethereum account wal-

let address provided via Metamask rather than tradi-

tional username and password validation used in cen-

tralised OSNs. When the user attempts to login, the

wallet address of the user will be directly connected

with the application and validated. The authentication

will be carried out with the account wallet address,

to decide whether the username already exists or not.

If the username already exists, the user information

will be fetched from the blockchain. If the username,

which attempts to login, has not been registered on the

blockchain, the application will ask to create a user

account. In order to handle the user account regis-

tration, all the user information including username,

biography and proﬁle picture will be stored on the

blockchain. However, to store the proﬁle picture of

the user to the blockchain, the ﬁle of the proﬁle pic-

ture is ﬁrst uploaded to the IPFS decentralised cloud

storage. Then, only the hash value returned from the

IPFS storage is stored on the blockchain. In this way,

it prevents the size of the blocks to store.

In terms of post creation, the user can create post

including post description, and media contents such

as images or videos. Unlike, centralised OSNs, all

the information regarding the post is stored on the

blockchain and decentralised storage, IPFS so that

there is no central entity who can alter or censor user

posts. Moreover, users can give likes as well as dis-

likes to the posts which mean the content modera-

tion is only powered by community driven approach.

Also, the users can tip the post which is engaging, so

that the owner of the post can acquire incentives. In

this way, it promotes the creation of more engaging

content and ﬁltering the inappropriate content. Also,

the feed algorithm of this social network is based on

popularity sorting based on three main factors, likes,

dislikes, and tips. Popularity sorting ensures that

high-quality and relevant content is prominently fea-

tured, while low-quality or inappropriate content is

pushed down in the feed. This helps users discover

valuable content more easily and reduces the visibil-

ity of harmful or irrelevant content.

4 EVALUATION AND

DISCUSSION

In this section, we will assess the viability and effec-

tiveness of our blockchain-based DOSN application

in relation to the research questions. This evalua-

tion intends to examine whether the project accom-

plishes the aim and objectives for preserving privacy,

and censorship resistance as well as enhancing user

engagement in online social network with the use of

blockchain technology.

(RQ1). How Does the Blockchain Architecture

Address Potential Privacy Concerns in Centralised

OSNs and Peer-to-Peer DOSNs?

The underlying blockchain technology of the applica-

tion can address privacy concerns in centralised OSNs

and peer-to-peer DOSNs through several mecha-

nisms. In traditional centralised OSNs, user data is

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

stored on centralised servers controlled by a single

entity, making them vulnerable to data breaches and

unauthorized access. Also, while peer-to-peer net-

works aim to distribute data across multiple nodes,

there’s still a risk of data leakage if one or more nodes

are compromised. In this blockchain- based DOSN,

as the Ethereum blockchain maintains an immutable

ledger of transactions or data regarding user inter-

actions on the blockchain in a tamper-proof manner.

Once data is recorded on a blockchain, it is extremely

difﬁcult to alter without consensus from the network.

Therefore, it provides a strong level of transparency

and integrity to the data stored on the blockchain,

which is helpful for preserving the privacy of social

network interactions and transactions. Peer-to-peer

technologies and blockchain technology are both de-

centralized, however the blockchain network offers a

more advanced level of decentralisation. It also uses

strong encryption methods to protect user communi-

cations and data. End-to-end encryption protects user

privacy from unauthorised access by ensuring that

only the intended receivers may access and decrypt

information. Moreover, the combination of end-to-

end encryption and smart contract logic used for so-

cial connection between users help secure the privacy

between users’ information. For example, according

to the Figure 3, users cannot view the content of the

other user, who does not have any social connection

between them. Unless the post of the user is available

public, other users cannot view the content. The con-

tent can only be decrypted and viewed by the autho-

rized user, who is the creator of the content, and other

users who share social connections with that particu-

lar creator can view his or her posts. This is achieved

through the use of Ethereum smart contract validation

rules. In this way, our solution addresses the privacy

problems which are vulnerable in centralised OSNs,

and peer-to-peer networks.

(RQ2). How Does the Decentralised Nature of

Blockchain-Based DOSN Provide the Censorship

Resistance.

Underlying technologies: Ethereum blockchain,

IPFS storage, and smart contract which are used to

implement this online social network, provide cen-

sorship resistance for users. As Ethereum blockchain

and IPFS are decentralised systems, they don’t rely

on a single central authority to function. Instead,

they operate on a network of nodes spread across

the world. Once data is written onto the Ethereum

blockchain, it becomes immutable. Similarly, con-

tent stored on IPFS is distributed across multiple

nodes, and once uploaded, it remains immutable.

This immutability ensures that no central authority

can censor or remove content on blockchain and

IPFS without user’s consent. Also, social network

functionalities are deployed as smart contracts on the

Ethereum blockchain, users can interact with the net-

work without relying on a centralised intermediary.

Moreover, Ethereum blockchain network is governed

by decentralised consensus mechanism, Proof of

Stake (PoS), meaning that decision-making power is

distributed among network participants rather than

being controlled by a single authority. As shown in

Figure 4, users can participate in the governance of

the network, inﬂuencing decisions related to data

management policies, such as giving likes to valuable

content, and dislikes to harmful content. Based on the

number of likes and dislikes, all the content is sorted,

and only the valuable content will be on top of the

news-feed, and harmful content will be pushed down

and unable to view. In this way, this reduces the risk

of censorship by eliminating single points of control,

and enhances the community driven approach on

content moderation and ﬁltering.

(RQ3). How Can the Blockchain-Based DOSN En-

hance the Contribution of Engaging Content than

Centralised OSNs and Peer-to-Peer DOSNs.

The implemented Ethereum blockchain-based DOSN

enhances the contribution of engaging content com-

pared to peer-to-peer DOSNs through several mech-

anisms. In peer-to-peer DOSNs, there is no chance

to integrate incentivisation for social interactions, and

thus, user’s engagement is only enhanced through

traditional ways of liking, commenting, and sharing

posts. However, in this blockchain-based DOSN, in-

centivisation mechanism is implemented to reward

users with incentives for creating engaging content.

As shown in Figure 5, users incentivise each other

through giving likes, dislikes, and tips for valuable

contributions. By incentivizing users to contribute

valuable content, the network motivates users to ac-

tively participate in the network and produce engag-

ing content. Moreover, in this social network, de-

centralised content curation algorithm is implemented

that rely on user interactions and community consen-

sus to surface engaging content. This algorithm prior-

itizes content based on factors such as likes, dislikes,

tips, and reputation scores, ensuring that the most rel-

evant and interesting content is prominently featured.

Also, through decentralized decision-making process

or incentive-based rewarding system, users can col-

lectively determine what constitutes engaging content

and how it should be promoted without relying on

centralised authorities, which addresses unfair mon-

etization mechanism, such as boosting in centralised

OSNs.

Enhancing Privacy, Censorship Resistance, and User Engagement in a Blockchain-Based Social Network

Figure 3: Feed showing that users cannot view posts of other users unless they are publicly shared.

Figure 4: Users participate for the governance of the network, such as adding likes and dislikes, which reduces the risk of

censorship by eliminating single point of control.

Figure 5: Users can incentivise other users to encourage participation.

5 RELATED WORK

In order to accommodate the privacy, censorship re-

sistance and user engagement issues in centralised

OSNs, various decentralised solutions have been ex-

plored. The ﬁrst detailed architecture for the de-

centralised peer-to-peer (P2P) online social network,

PeerSon, was proposed by Sonja Buchegger in 2009

(Buchegger et al., 2009). They developed the Peer-

SoN system, which combines a P2P infrastructure

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

with encryption and direct communication between

users’ devices, in an effort to strengthen OSNs while

protecting users’ privacy. The two-tier system archi-

tecture of the prototype consists of a distinct look-up

service and peers interacting with one another. They

created protocols for sharing data directly between

peers as well as between peers and the lookup ser-

vice. However, since the service provider can view

every interaction and use this data to ascertain the so-

cial graph of the users it hosts, maintaining the mech-

anism in PeerSoN is a crucial problem. Therefore,

protecting user data privacy in a single location won’t

be possible with just a theoretically decentralized de-

sign. Moreover, Randy Baden and Adam Bender pro-

posed a decentralised OSN called Persona, which al-

lows user to have more control over their information

(Baden et al., 2009b). Persona uses attribute-based

encryption (ABE) to hide user data, giving users the

ability to impose strict guidelines on who can access

their information. Persona offers a powerful tool for

developing apps where users, not the OSN, set access

rules for personal information. They present novel

cryptographic techniques that improve ABE’s broader

application. Although, the Persona OSN secure user

information, metadata including timestamps, data ob-

ject sizes, data structures, and header information still

leak. This makes it possible to infer social graphs and

exposes user behavior patterns. Even when the con-

tent is encrypted, third-party attackers might still be

able to gather enough information from the metadata

attached to the data objects. Two initiatives that inves-

tigate blockchain technology for social networks are

Akasha7 and Synereo (Chakravorty and Rong, 2017).

These solutions are all still in their initial stages of

development. Using Ethereum as its blockchain net-

work, Akasha seeks to create a knowledge architec-

ture for social human advocacy within the framework

of social networks, freedom of speech, creative perpe-

tuity, and privacy for an improved Internet that ben-

eﬁts all people. A distributed, decentralized social

network built for the attention economy is offered

by Synereo (Chakravorty and Rong, 2017). It pro-

vides more of a social marketplace platform. These

blockchain-based solutions protect user privacy and

withstand censorship, but they still have limits when

it comes to moderation and content ﬁltering.

Unlike these above decentralised solutions, the in-

tegration of Ethereum blockchain, smart contract, and

IPFS storage solution in our blockchain-based DOSN,

offers several advantages. In centralised OSN, the sin-

gle authority owns the user data. Although users may

be able to modify their privacy settings, the network

controls what information is gathered and how it is

used. Also, in peer-to-peer DOSNs, although it al-

lows users to have control over data, users still need to

trust peers and service providers and also distribution

of user data may not ensure complete control. In this

blockchain-based DOSN, users have complete own-

ership of their data with the use of cryptographic keys

that grant them control over who can access their data

and under what conditions. Because the user’s crypto

wallet is connected to this social network, providing

public and private keys for network access, authen-

tication and authorization are implemented through

that connection. Hence, only authorized users with

the corresponding private keys can decrypt and view

the content. Public keys stored on the blockchain

facilitate secure communication between users. In

contrast, users of peer-to-peer DOSN and centralised

OSN have to share personal information in order to

participate, as authentication and permission are han-

dled through email and password mechanisms. Ad-

ditionally, because Blockchain transactions are im-

mutable and transparent, users can simply validate

how their data is being accessed or used. On the

blockchain, every transaction is recorded and cannot

be changed without the users’ permission. This trans-

parency gives users greater control over their data

compared to peer-to-peer networks where data han-

dling practices might not be as transparent, and de-

pends on the network operator of a speciﬁc node.

This social network application utilises smart con-

tracts to give users control over how they participate.

Users can specify rules for accessing their data or in-

teracting with their content, and the smart contract

rules will be automatically enforced without the need

for intermediaries. Also, the smart contract manages

core functionalities of this social network applica-

tion, by facilitating privacy-enhancing rules such as

zero-knowledge proofs or selective disclosure of in-

formation. Therefore, users can selectively share only

the necessary details required for speciﬁc interactions,

preserving their privacy and having more control over

data.

Moreover, to address content moderation and ﬁl-

tering inappropriate content issues in peer-to-peer and

other blockchain-based DOSNs, community-based

moderation is used in this social network, where

users can collectively give likes, dislikes, incentives

to posts, and sorting by popularity. Users can indi-

cate their approval of content by liking it. This serves

as a positive signal indicating that the content is rele-

vant, valuable, or appropriate. Conversely, users can

express their disapproval of content by disliking it.

This serves as a negative signal indicating that the

content may be inappropriate, misleading, or harmful.

Moreover, users can ﬁnancially reward content cre-

ators by tipping their posts. This incentivizes the cre-

Enhancing Privacy, Censorship Resistance, and User Engagement in a Blockchain-Based Social Network

ation of high-quality content and encourages users to

contribute positively to the network. Then, posts are

sorted based on its popularity, which are determined

by factors such as the number of likes, dislikes, and

tips. Popularity sorting ensures that high-quality and

relevant content is prominently featured, while low-

quality or inappropriate content is pushed down in

the feed. This helps users discover valuable content

more easily and reduces the visibility of harmful or

irrelevant content. Also, with the proof-of-stake con-

sensus mechanism of Ethereum blockchain, it helps

prevent spam, and other forms of abuse. By lever-

aging the community-driven moderation mechanism,

the blockchain-based social network empowers users

to collectively ﬁlter and moderate content according

to community standards and preferences, while pre-

serving censorship resistance.

6 CONCLUSIONS

In conclusion, the development of a decentralised on-

line social network powered by Ethereum blockchain

is an important advancement toward protecting users’

privacy, and encouraging resistance to censorship.

With the help of blockchain technology’s integrated

security and transparency, we have developed this de-

centralised online social network that allows users

to actively participate in creating their online expe-

riences rather than just being passive consumers. The

network ensures that user data is secure and tamper

proof by mitigating the weaknesses of peer-to- peer

and centralised OSNs through the use of blockchain

technology. The implemented online social network

gives users control over who can access their in-

formation and under what circumstances by utiliz-

ing smart contracts, cryptographic and decentralized

techniques in authentication and interaction. Further-

more, users’ trust is increased by the immutability of

blockchain transactions, which ensures accountabil-

ity and transparency. Moreover, the social network

promotes censorship resistance by offering an envi-

ronment where people may express themselves freely

without worrying about repression or manipulation.

Additionally, by integrating incentives and promot-

ing user engagement, the implemented blockchain-

based DOSN has successfully addressed issues with

both centralised and peer-to-peer OSNs. This is ev-

idenced by the results of the evaluation against re-

search questions. Overall, the principal aim of our

work is successfully accomplished. In the future, in

terms of features improvement of this social network

application, more advanced functionalities such as de-

centralised commenting, sharing, messaging, creating

stories, creating Non-fungible Tokens (NFTs), and in-

tegrating with other decentralized applications, will

be included.

REFERENCES

Baden, R., Bender, A., Spring, N., Bhattacharjee, B., and

Starin, D. (2009a). Persona: An online social network

with user-deﬁned privacy. volume 39, pages 135–146.

Baden, R., Bender, A., Spring, N., Bhattacharjee, B., and

Starin, D. (2009b). Persona: An online social network

with user-deﬁned privacy. volume 39, pages 135–146.

Bielenberg, A., Helm, L., Gentilucci, A., Stefanescu, D.,

and Zhang, H. (2012). The growth of diaspora - a

decentralized online social network in the wild.

Buchegger, S., Schi

oberg, D., Vu, L.-H., and Datta, A.

(2009). Peerson: P2p social networking: early ex-

periences and insights. In Proceedings of the Second

ACM EuroSys Workshop on Social Network Systems,

SNS ’09, page 46–52, New York, NY, USA. Associa-

tion for Computing Machinery.

Chakravorty, A. and Rong, C. (2017). Ushare: user con-

trolled social media based on blockchain. In Proceed-

ings of the 11th International Conference on Ubiq-

uitous Information Management and Communication,

IMCOM ’17, New York, NY, USA. Association for

Computing Machinery.

Grafﬁ, K., Gross, C., Stingl, D., Hartung, D., Kovacevic,

A., and Steinmetz, R. (2011). Lifesocial.kom: A

secure and p2p-based solution for online social net-

works. pages 554 – 558.

Greschbach, B., Kreitz, G., and Buchegger, S. (2012). The

devil is in the metadata—new privacy challenges in

decentralised online social networks.

Guidi, B. (2021). An overview of blockchain online social

media from the technical point of view. Applied Sci-

ences, 11:9880.

Mohammed, A. B. (2022). Decentralised social media plat-

form using blockchain technology.

Nilizadeh, S., Jahid, S., Mittal, P., Borisov, N., and Ka-

padia, A. (2012). Cachet: A decentralized archi-

tecture for privacy preserving social networking with

caching. CoNEXT 2012 - Proceedings of the 2012

ACM Conference on Emerging Networking Experi-

ments and Technologies.

Paul, T., Famulari, A., and Strufe, T. (2014). Survey on

decentralized online social networks. Computer Net-

works, 75, Part A:437 – 452.

Schwittmann, L., Boelmann, C., Wander, M., and Weis, T.

(2013). Sonet – privacy and replication in federated

online social networks. pages 51–57.

Shilina, S. (2023). The Promise of Blockchain-Based De-

centralized Social Networks: Enabling Privacy, Cen-

sorship Resistance, and User Control, pages 172 –

198.

Tran, M., Nguyen, S., and Ha, S. (2016). Decentralized

online social network using peer-to-peer technology.

REV Journal on Electronics and Communications, 5.

ENASE 2025 - 20th International Conference on Evaluation of Novel Approaches to Software Engineering

Urdaneta, G., Pierre, G., and van Steen, M. (2011). A sur-

vey of dht security techniques. ACM Comput. Surv.,

43:8.

Xu, Q., Song, Z., Goh, R., and Li, Y. (2018). Building an

ethereum and ipfs-based decentralized social network

system. pages 1–6.

Yaga, D., Mell, P., Roby, N., and Scarfone, K. (2018).

Blockchain technology overview.

Yerby, J., Koohang, A., and Paliszkiewicz, J. (2019). So-

cial media privacy concerns and risk beliefs. Online

Journal of Applied Knowledge Management, 7:1–13.

Zhan, Y., Xing, X., and Xiong, Y. (2022). A conceptual

model and case study of blockchain-enabled social

media platform. Technovation, 119.

Enhancing Privacy, Censorship Resistance, and User Engagement in a Blockchain-Based Social Network