ELI5: Datawallet White Paper [PART 1]
The Datawallet White Paper, explained in plain English
The reason for our ELI5 White Paper Series
Datawallet has the potential to significantly impact the lives of billions of people: we estimate that by 2022, a person’s data could have an average value of $7,000 per year. By allowing people to be their own data broker, selectively sharing their data with organizations they trust, and reclaiming the value of their data, we effectively have the opportunity to realize the idea of a universal basic income. For organizations, the Datawallet model has similarly impactful implications: by being permissioned by users to leverage their data, organizations get access to the best data ever available. This in turn can fuel the advancement of technologies that are currently still within the realm of fiction, such as a personal Artificial Intelligence (AI) concierge.
But while our mission is bold, it is nothing without a platform that can realize its potential. In order to explain how we intend to build this platform, we wrote the Datawallet White Paper which is entitled
A Data-Ownership Assuring Blockchain Wallet For Privacy-Protected Data Exchange
You can find the full White Paper right here. It is a detailed deep dive into the individual components of our platform and how they work together. While we tried to be as thorough and precise as possible in describing our platform, we realize that many are looking for a more concise and high-level treatment.
We will therefore be publishing a series of posts that will break our White Paper down into approachable, bite-sized chunks. We hope that this encourages more people to read about what it is that we are looking to build, promotes understanding of our platform, and fosters a dialogue in the community around Datawallet about how to best achieve our mission.
So without further ado, here is…
Part 1 of our ELI5(ish) Datawallet White Paper series
The Datawallet ecosystem provides the marketplace where data providers and data consumers can directly engage with one another, removing the predatory data brokers that currently make the data market. Unlike the conventional data brokerage system, the Datawallet market allows users of apps/websites (data providers) and data consumers to transparently, securely, and conscientiously exchange data and services. In the remainder of this article we outline what the requirements and goals of data producers and data consumers are. These requirements inform the ecosystem we detail in subsequent articles; They tell us what success will look like for the stakeholders of the ecosystem.
Fundamentally, data producers want to be able to make their data work for them, which requires them to be able to conscientiously share it with organizations who can offer them value. This value can take many forms. For one, organizations can simply pay for the data. An example for this could be a company undertaking market research to understand what people are buying online versus what they usually buy in brick-and-mortar stores.
A second option is that companies can offer a new service or the ability to for data producers to enhance an existing service. Take for example the Amazon Alexa; currently, users of the Alexa are forced to interact with it on a daily basis, so that it can learn about the user. This is due to the fact that the Alexa needs a data test set in order to train its algorithms with data about who you are and what your preference set looks like. This way, when you say “Alexa, order me 12 eggs”, the Alexa knows that you want eggs from Whole Foods rather than from Walmart. Right now, you are forced to create this data for the Alexa, and it will take you months to do so. However, you already created data about your preferences all over the internet. And with Datawallet, you can now share all of the historic data you created all over the internet (and offline as well) and share it with the Alexa. This way, the Alexa can learn about you within a matter of seconds, rather than months.
A third option is that data producers may want to donate their data to organizations and projects whose missions they believe in. This is the case in situations where a new service is not built, yet, and requires data for the algorithms to be properly trained. In this scenario, anybody who finds the idea of the planned service interesting, can make their data available for the service to be built and get the full personalized experience once it is live.
In order for producers to frictionlessly find and send the relevant data for these exchanges, they need to have a structured and collated data profile, or ‘datawallet.’ This is crucial to ensure that data consumers can build their analyses and products around known endpoints, and to allow data producers to quickly identify if they have the data that is being requested. Data producers want to securely store and access this datawallet, with different users preferring different options along the ease-of-use vs. control/security tradeoff.
Data producers want to be able to search for and engage with the various offerings of the organizations who have built products and services around their collated datawallets — -a contract exchange. Data producers want to know the precise details of what the organization is offering, which can be made explicit in the logic of the smart contract. The contract can clearly specify who created the smart-contract, the precise data points that are being requested, what they will be used for, and what the compensation will be. Data producers also want to be able to easily and securely enact contracts that they choose to engage in.
This constellation of goals is summarized in the paper with this AOM goal model:
Summary Requirements for Data Producers
In order to take data ownership and make their data work for them they want:
- Automatically collated and processed data profiles
- Exchanges hosting clearly defined data/service smart-contracts to engage in
- The ability to securely consent to and execute data/service contracts
Data producers want to be able to do this all as easily and frictionlessly as possible. As we will describe in subsequent articles, our approach is to use native mobile apps as the entry point to all of this functionality.
Data Consumers want to learn from data and use it to build cool stuff. An encouraging percentage of them want to do so with ethically sourced data (they just didn’t have an option before!). They have desires largely complementary to data producers — they want to be able to easily, efficiently, and ethically source specific user data points.
Many of the researchers, developers, AI engineers, and data scientists interested in learning from and building on the data are not blockchain gurus. They want to be able to programmatically generate a smart-contract and deal with the resulting data through a restful API with stable endpoints. Some data consumers want to get data from particular kinds of users by specifying demographic parameters. We have distilled these requirements into a set of mandatory and optional fields that need to be specified to form a valid contract:
The contracts were segmented into two broad types. A data request simply specifies an exchange of data for compensation (in the form of token), while a data product includes the specification of the delivery of a product/service built on top of the data. Both kinds of contracts need to be posted to relevant exchanges where interested and viable data producers can view and potentially engage with the contracts.
The data consumers’ goals is summarized in the paper with this AOM goal model:
Summary Requirements for Data Consumers
In order to efficiently source data:
- (Programmatically) specify the parameters of a data request
- Post data requests on the exchange and manage their viewing and execution
- Provide resulting data in a standardized interface (restful API)
In our next post, we will get into the components of the Datawallet ecosystem which together can fulfill the requirements of data producers and data consumers outlined here.