“`html
Imagine a nearby concert location desires to involve its community by providing social media followers with a simple means to share and discuss new tunes from rising talent. Instead of adhering to the limitations of current social networks, the venue might consider developing its own social app equipped with features tailored to its community. However, constructing a brand new social app from the ground up involves numerous intricate programming tasks, and even if the venue successfully builds a personalized app, the organization’s followers might be hesitant to transition to the new platform as this might require abandoning their existing connections and data.
Currently, investigators from MIT have introduced a framework named Graffiti that simplifies the creation of customized social applications, enabling users to shift between various applications without sacrificing their friends or data.
“We aim to empower individuals to have authority over their own designs rather than having them dictated from above,” states electrical engineering and computer science postgraduate Theia Henderson.
Henderson and her associates developed Graffiti with a versatile structure allowing individuals to craft a wide array of tailored applications, from messaging platforms like WhatsApp to microblogging sites akin to X, to location-centered social networks like Nextdoor, all employing only front-end development tools such as HTML.
The protocol guarantees that all applications can communicate seamlessly, permitting content shared on one platform to be presented on another, even if the designs or functionalities differ. Importantly, Graffiti users maintain control over their data, which is stored on a decentralized framework rather than being controlled by a single application.
While the advantages and drawbacks of deploying Graffiti at scale are yet to be thoroughly assessed, the researchers are hopeful this innovative method can eventually foster healthier online interactions.
“We’ve demonstrated that it is possible to have a vibrant social ecosystem where everyone retains control over their own data and can utilize any applications they prefer to engage with whomever they choose in whichever manner they want. They can craft their own experiences without severing ties with those they wish to remain connected to,” remarks David Karger, professor of EECS and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL).
Henderson, the primary author, and Karger are joined by MIT Research Scientist David D. Clark on a document regarding Graffiti, which will be presented at the ACM Symposium on User Interface Software and Technology.
Customized, integrated applications
With Graffiti, the researchers focused on two primary objectives: to reduce the challenges of creating personalized social applications and to allow those applications to communicate without needing developer authorization.
To simplify the design process, they constructed a shared back-end infrastructure that all applications can access to store and distribute content. This means developers don’t have to write complex server code. Instead, crafting a Graffiti application resembles building a website using familiar tools like Vue.
Developers can also swiftly implement new features and types of content, granting them greater flexibility and nurturing innovation.
“Graffiti is so simple that we utilized it as the backbone for the introduction to web design course I teach, and students were able to easily write the front-end to create all kinds of applications,” Karger notes.
The open, interoperable characteristic of Graffiti implies that no single entity can enforce a moderation policy across the entire platform. Instead, various competing and conflicting moderation services may function, allowing users to select their preferred options.
Graffiti employs the concept of “total reification,” where every action performed within Graffiti, such as liking, sharing, or blocking a post, is represented and archived as its own distinct piece of data. A user can configure their social application to interpret or disregard those data according to their own regulations.
For instance, if an application is established so that a certain user acts as a moderator, posts blocked by that user will not be visible in the application. However, for an application governed by different rules where that individual is not recognized as a moderator, other users might merely receive a warning or no notification at all.
“Theia’s framework allows each person to choose their own moderators, steering clear of the one-size-fits-all moderation policy adopted by major social platforms,” Karger states.
Nonetheless, the absence of a central moderator means there’s no authority to remove content that could be deemed offensive or illegal.
“We must further research to determine if this will result in actual harmful consequences or whether the kind of personalized moderation we’ve created can provide the necessary protections for users,” he adds.
Empowering social media users
The researchers also needed to tackle a challenge known as context collapse, which clashes with their goal of interoperability.
For instance, context collapse would happen if a person’s Tinder profile appeared on LinkedIn, or if a post meant for one group, like close friends, conflicted with another group, such as family members. Context collapse can result in anxiety and have social repercussions for the user and their various communities.
“We realized that interoperability could sometimes be detrimental. Individuals have boundaries across different social contexts, and we did not want to infringe upon those,” Henderson explains.
To prevent context collapse, the researchers designed Graffiti to organize all content into separate channels. Channels are adaptable and can represent a variety of contexts, such as individuals, applications, locations, etc.
If a user’s post is visible in an application channel but not in their personal channel, others utilizing that application will see the post, but those who exclusively follow this user will not.
“People should have the ability to choose the audience for whatever they wish to communicate,” Karger emphasizes.
The researchers developed several Graffiti applications to demonstrate personalization and interoperability, including a community-focused application for a local concert venue, a text-centric microblogging platform modeled after X, a Wikipedia-like application enabling collective editing, and a real-time messaging application featuring various moderation schemes modeled after WhatsApp and Slack.
“It also allows for the creation of numerous social applications that have yet to be conceived. I’m truly eager to see what innovations arise when individuals are given complete creative freedom,” Henderson expresses.
In the future, she and her team aim to investigate additional social applications they could construct with Graffiti. They also plan to integrate tools like graphical editors to enhance the design process. Furthermore, they wish to bolster Graffiti’s security and privacy.
And while much work remains before Graffiti could be widely implemented, the researchers are currently conducting a user study as they examine the potential positive and negative effects the system could pose on the social media environment.
“`