Our design research for a decentralization pattern library has resulted in a report outlining the obstacles to decentralization as a whole.
The potential of decentralization to contribute to a more equal society is huge. Where government and corporate control of information are causing harm, decentralized technologies can give power back through new data ownership and governance models. So why are these technologies not more widely adopted? What are the main obstacles to decentralization, and how can we solve them?
In our latest research report, Decentralization Off the Shelf: 7 Maxims, we interviewed 53 practitioners and led workshops with a total of 85 participants to find out. Here are the 7 key changes we need.
1: We need to collaborate to challenge existing power structures and build usable technologies.
The problem: A focus on bespoke solutions to common problems results in missed opportunities for collaboration, a lack of common vocabulary, and higher access barriers for users.
The solution: Practitioners need to seek collaboration with other projects targeting the same users and use cases, and pattern match with regard to design and architecture choices.
The problem: The threat posed by decentralization to existing power structures means we can expect political and legal backlash as decentralized technologies become more popular. However, many projects are focused on their own products and not on this collective concern.
The solution: In order to effectively challenge the status quo, we need to build a decentralization movement. A movement involves collaborative strategizing and community organizing around a shared mission and values. Decentralization can learn here from adjacent movements.
2: We need to ensure accessibility and usability by employing best practices in human centered design.
The problem: Decentralization introduces concepts that may not be familiar to users who are used to centralized technologies. This can be confusing and discouraging for users. Explanations of decentralized technologies are often not clear and accessible, and there is a general lack of design and product thinking in the space.
The solution: We need to work towards adopting shared language and interfaces across projects. Practitioners need to consistently apply best practices in human-centred design, and test products early and often with end users.
3: We need investment to improve the independent verifiability and resilience of our infrastructure.
The problem: Decentralized protocols are still dependent on centralized points of failure such as cloud infrastructure, web browsers, app stores, and proprietary hardware.
The solution: We need funder investment in solutions for closing these gaps and improving the independent verifiability and resilience of our shared infrastructure. We also need a more collaborative approach: for example, crafting protocols on a variety of modular libraries and collaborating on connectivity toolkits.
4: We need better onboarding, training, and education for developers.
The problem: As the past decade has brought a revolution of content for upskilling and onboarding developers to build client-server applications, there has not been a similar educational movement for decentralized applications. Current educational materials are highly technical and disparate, with a steep learning curve requiring significant time investment.
The solution: To attract and engage more developers, there is a dire need for clear and accessible educational content. We need better user experience for application developers with a focus on toolkits, query languages, and modular libraries. We also need to support projects to share more about their approach through the creation of accessible content.
5: We need better threat modeling for safety and privacy.
The problem: While there is increased interest in decentralized technology as a response to surveillance capitalism, security breaches, and tech monopolization, few projects offer the security and service that end users are looking for from an alternative. Many decentralized projects do not have the resources to implement, communicate, and research good privacy and safety guarantees.
The solution: We need to train more developers and designers on privacy and security approaches and skills, as well as funding research and development on security and anonymity innovations. We need to better communicate with end users about who they are trusting with their data and be more transparent about the privacy trade-offs we make.
6: We need more sustainable funding, business development, and recruiting models.
The problem: Sustainable funding is the biggest factor in the successful delivery and maintenance of any project, yet projects often fail to identify their target audience and develop sustainable funding models. Without sustainable funding, it is difficult to attract talent.
The solution: We need to research, develop, and test community-driven funding models as they relate to the sustainability of people, projects, and protocols. We also need to work with project managers who can introduce business cases, prototyping processes, and more.
7: We need clear governance and decision-making models that foster diversity.
The problem: Decision-making processes are often unclear and poorly communicated, resulting in conflicts in interests and expectations, as well as challenges when it comes to protocol updates.
Practitioners operate within a relatively homogenized environment with respect to demographics and values, resulting in technologies that meet the needs of only a small part of the world’s population.
The solution: We need open governance structures with strong social rules for decision-making that are communicated clearly to contributors and end users. Project leads should focus on diversifying their decision-making teams and the community as a whole.
Making these changes would lead to better usability and wider adoption of decentralized technologies. To find out more about the solutions and how to implement them, read our report. Join the conversation on social media with #decentpatterns
