Establishing Trust in Reputation Systems: Challenges for Blockchain-Based Reviews
Reputation systems are an essential part of the online world. We buy products, book hotel rooms and hire handymen based on the millions of reviews we find online. What can the blockchain add to the trustworthiness of these reviews? This third and last blog in my 3-part series exploring blockchain in practice attempts to answer this question.
Current reputation systems are easy to defraud. By creating a new online identity with a new e-mail address, anyone can write bad reviews about a restaurant or store whose owner they don’t like or good reviews about those owned by family or friends. And if you’re a business owner who wants to swing for the fences, you can even hire an army of bots so that thousands of fake accounts boost your product.
How can we know if the reviews we’re using to make buying decisions are reliable? We can’t - and that makes even the best and most trustworthy reviews less trustworthy.
The Benefits of Blockchain Technology for Review Systems
A decentralized reputation system using blockchain technology can help. Blockchain technology works with unique identities, which means that bots and individual fake accounts are ruled out if the accounts are verified. Most importantly, reputation information cannot be manipulated because the recording of the information is decentralized and therefore immutable.
The “know your customer process” is the biggest challenge for blockchain-based reputation systems. How can you ensure that every account corresponds to an identity without needing a central authority that recognizes that you’re an individual based on your passport? And how do you guarantee anonymity? Several ideas exist to verify accounts; one of them is called proof-of-individuality.
The Proof-of-Individuality Protocol
This protocol uses the concept of pseudonym parties coined by Brian Ford in his article, “Pseudonym Parties: An Offline Foundation for Online Accountability:”
Ensuring accountability does not, in general, require identifying users. However, it only requires enforcing a principle of one person, one persona for a given online service. This paper proposes pseudonym parties, a decentralized scheme that combines technical tools (pseudonymous online accounts) with in-person social occasions (parties) to provide online accountability while preserving the ability of users to participate anonymously in online services (Ford, 2007).
The Proof-of-Individuality Anti-Sybil Project wants to use online pseudonym parties, video chats in which an individual has to show up at a certain given time. It is based on the idea that you can only be in one video chat at the same time. After attending an online pseudonym party, users are verified by the other users in that hangout and get a token (POI) that proves their individuality. That token is valid for a month, after which there is a new hangout. The proof of concept is now complete for 80 percent, and I’m very curious about how it will work out.
Tokenization of Reviews and Connections
Another idea is to let people vouch for each other. If you’re part of a network, you can invite friends—people you trust—to that network. Those friends can then invite their friends as well. So it’s like a social network with high-quality content where users anonymously leave reviews. Geo-based locations could e.g. be used to verify if a user visited a place. The system would gain value by collecting, verifying and providing these reviews. Similar to providers of content, e.g., on Steem.io, creators of accurate reviews could be rewarded so that the platform value increases and incentivizes great reviews and diminishes ‘bad’ ones.
Current system designs plan to tokenize reviews and social connections, although still facing mentioned challenges. For example, Lina.review positions itself as the “ecosystem for reviewers, professionals, and businesses all around the world, powered and connected by blockchain” and issued a token for reviews. The Know Your Customer process is currently circumvented by requiring a passport scan. Besides, there is DREP, “a decentralized reputation ecosystem comprised of a public chain, a reputation-based protocol and the tools for internet platforms to leverage its power.”
Immutability of Blockchain
Nonetheless, the challenges in this field are not to be underestimated. So far, only a few startups dare to tackle such a new service. Immutability of blockchains are its great advantage but can also create problems with current or new regulation, like the right to be forgotten. Assume that a user published personal data on a blockchain but now wants to make use of the right to be forgotten. That’s impossible with common blockchain protocols.
The blockchain is moving into all kinds of fields: e-commerce is under attack by Openbazaar, a distributed market platform; the Dutch startup Volareo wants to change and tokenize music streaming, and Donkey Bike enables a shared economy of bicycles.
Review systems pose more difficulties than other fields but won’t be immune to the attraction of blockchain benefits.
Related Blockchain Technology Information
This three-part series explores ways to find Blockchain Technology use cases. Published by Xebia & GoDataDriven.
In the previous two articles we cover your business transactions in depth and discuss the different ways technology can impact the business model of platform companies. The other blog posts are:
- How to Find Blockchain Use Cases - Part I;
- How Blockchain Technology Will Make Platforms Obsolete - Part II.
Blockchain Technology Foundation - Training
Sometimes it can be hard to distinguish the different elements of blockchain and its surrounding phenomena. You can learn to differentiate between hype and real potential! This one-day training course at Xebia Academy is available on 26 September, 12 November, and 17 December 2018: Blockchain Technology Foundation.
Blockchain Survey 2019
Free download of Blockchain Survey 2019! Together with Computable, we recently conducted a survey about the use and expectations of #blockchain, and we condensed all the results into one comprehensive report.
Want to learn more about our vision about blockchain technology and what we could add to your business? Feel free to get in touch!
Follow us for more of this
How to build your first image classifier using PyTorch
July 18, 2019
Data Science Podcast Recommendations
July 12, 2019
The Analytics Translator Part 3: Characteristics of an Analytics Translator
July 10, 2019
The Analytics Translator Part 2: The Problems an Analytics Translator Solves
July 03, 2019
GoDataDriven Open Source Contribution for May and June 2019
June 28, 2019
Deploying Apache Airflow on Azure Kubernetes Service
June 28, 2019