Creating scalable peer-to-peer trust and replacing firms with smart contracts
In our previous post we shared with you our vision on the future of knowledge work and collaboration.
To refresh your memory, we believe that the future of data-driven knowledge work lies not in firms and centralized networks, but in Blockchain-based, decentralized information networks that enable teamwork at scale.
In this post we will dive more into the current state of data-driven knowledge work, its challenges and the gap that Covee Network is designed to fill.
Crowdsourcing and collaborative platforms of the Web 2.0 era
The past ten years with the rise of data science, numerous centralized platforms and firms have emerged. Their goal is to make sense of the huge amount of data available and how to best leverage them to create value. However, trust and coordination at scale are the two crucial issues we need to address.
Winner-takes-it-all
As of today there are more than 1.5 million people signed up on competitive crowdsourcing platforms, submitting as individuals their solutions to data-driven problems. However, on these kind of platforms 99.99% of their users have never won a data science competition or were rewarded for their work and output.
Our message is simple: most complex problems cannot be solved by an individual. Even if a great Python coder spends months learning about a domain like cancer diagnostics, it is not the best way of using his time. And for a trained radiologist it is also more productive to apply her medical expertise than spending half a year trying to become a computer scientist.
When solving complex, data-driven problems collaboration, division of labor and complementary skills are key.
Free-riding on open collaboration platforms
According to studies and empirical data, on open collaboration platforms such as Wikipedia 1% of users create content, 9% make occasional edits and 90% only consume information. On Reddit, the distribution appears to be more extreme: only 0.1% create content, 1.9% comment while 98% only read.
Free-riding is the dominant user strategy in open collaboration networks.
Only with the right incentives and mechanism design in place, will each individual be motivated to contribute his/her share of work and trust that the rest of the team members will also contribute their part.
Web 3.0 era: A new way for effective team coordination
The current state of talent coordination is imperfect. Firms find, hire and coordinate talent. Therefore, they have the power to form a team that satisfied their best interests and not yours.
On the other hand, competitive crowdsourcing platforms largely avoid these coordination problems by focusing on competing individuals. Some platforms offer the option of inviting others to form teams, but even on platforms that strongly encourage this, the vast majority compete as individuals. Team formation is limited to people who already know each other.
Personal trust is the least scalable form of interaction hence the scale of effective collaboration on such platforms is extremely limited.
Another coordination challenge is: how do you find people who are interested in collaborating and how do you identify people who have complementary skills? Competitive crowdsourcing platforms do not answer these questions.
This is the reason we are building Covee Network, a global network giving you access to a large scale of users. You will be able to find other like-minded people with complementary skills and solve complex data-driven problems together. Therefore, team coordination becomes more efficient without the need of bosses or managers that exist in traditional firms.
Our solution
But how do we get people to trust their teammates? How do we avoid free-riding and winner-takes-all? And how do we achieve better team coordination than in traditional firms?
The answer is simple: With the right combination of technology, mechanism design and incentives. A combination of smart contracts, cryptographic tokens and a unique mechanism design, leading to a scalable peer-to-peer trust network with automated coordination.
Smart contracts (i.e. software applications that can enforce an agreement without human intervention)
They enable decentralized governance of our network by replacing traditional legal contracts that exist in firms and automating the coordination process required for the completion of the project.
Our vision is to replace personal trust that currently exists in corporations with Blockchain-powered trust. Therefore, enabling team members to interact via smart contracts, drastically improves the motivation/coordination trade-off.
Cryptographic tokens
Create entirely new incentive structures. Our decentralized tokenized network gives users a financial stake in the network’s success via ownership of cryptographic tokens, which increases participants’ motivation.
Unique mechanism design
Enables scalable trust in decentralized teams. If the future of knowledge work is to be collaborative, decentralized and scalable then a set of governance mechanisms is needed to incentivize contribution, signal commitment and mitigate free-riding (i.e., if a knowledge worker does not contribute sufficiently to the project).
Below we explain the three mechanisms that we designed in order to enable this blockchain-powered trust and disincentivize free-riding.
Mechanism design for decentralized governance of knowledge work
As illustrated below, being part of our network as a knowledge worker includes a sequence of steps.
1. Team formation and staking mechanism
Are you an expert on machine learning, coding, designing, dev ops, or some other speciality longing for applying your skills on cool data projects? But instead of competing with other individuals, you want to contribute with your knowledge and skills and get fairly rewarded?
If you answered yes to the above questions, then Covee Network is the right place for you. You can pick the data-driven project that you like the most and collaborate with other smart people.
Are you an expert on genomics looking for data scientists to help you with Python coding required to analyze your lab results? You looked for people willing to team up on existing data science forums or other websites but no luck in terms of commitment and contribution?
Again, Covee Network is your place. You can initiate your own genomics project and find motivated people interested in helping you with their coding skills.
So how do you join a team? We aim to assemble a team of users to complete each project. We use a staking mechanism to achieve the following goals in the team formation process:
i. Create synergistic teams matching the needs (skills) of the project
ii. Select users most motivated/committed to a particular project
iii. Select the best user for each role in the project
iv. Select team members that can work well together
How does the staking mechanism work?
We use a staking mechanism to select the team members for the project. The project initiator defines how many tokens must be staked to be eligible to join the team, and then users “apply” to individual roles by staking the required tokens.
The mechanism has three parts:
a. Project creation/definition phase
There are two types of team initiators. An individual who can create and manage the project or a client (a financial institution, an academic institution, etc.).
Team initiators can set minimum requirements for a set of variables (reputation score, minimum stake, experience). Additionally, he/she defines how much (in %) each role is expected to contribute to the success of the overall project and briefly describes what tasks will be expected in the project. And also defines how many tokens a user must stake to “apply” for each role in the team.
b. Application/staking phase
All users on the platform can see the project with all of its details and can apply to it. Once they find the role that they want to apply for, they submit their tokens to the smart contract and thus everything is visible to every user.
c. Team member selection phase
The team initiator has access to the full list of applicants, i.e., those users that have staked the required amount of tokens. For each of those users, the team initiator can review the reputation scores, introduction videos, previous projects, etc. and select the “winner” for each role.
2.1 Peer-to-Peer (P2P) Review Mechanism
The P2P review mechanism is a key building block in our overall design. All team members working on a project together review each others’ work at pre-defined moments (milestones) during the project, indicating how much each other team member has contributed to the progress of the project.
The P2P review mechanism is designed in such a way that it is optimal for each team member to be truthful when providing reviews about all other team members.
2.2 Stake payout and client fee payout mechanisms
Once a team has formed and a project has started, we want to create a good work environment and encourage the team members to collaborate.
In order to ensure contribution and client fee payout, we design a system that relies on milestones. At each milestone, two review processes are kicked off:
i. The team initiator reviews the team’s progress towards the project
ii.The team members are asked to review each others’ contributions
Thus, those team members who contribute most towards the progress of the project obtain the largest share of the staked tokens and of the client fee.
3. Reputation mechanisms
In essence, a reputation system collects feedback from participants as automated “word-of-mouth”. Reputations systems on Covee Network are intended to foster good behavior amongst users and between users and clients. In the long-run the reputation score will be the decisive factor in bringing teams together. There will be three sets of reputations:
i. Peer reputation: a peer-to-peer reputation system for users that results from Covee users’ mutual ratings
iii. Client reputation: a client-to-peer reputation system for users that results from clients’ ratings of their Covee users
iii. Market reputation: a peer-to-client reputation system for clients that results from Covee users’ ratings of the clients
Conclusion
We are convinced that data-driven knowledge work is teamwork. Complex data-driven problems in various industries like engineering, AI, genomics, algorithmic trading, require multiple, complementary skills and cannot be solved by an individual. Therefore, we are building Covee Network, the ideal place for smart people to team up and work on data-driven projects. Collaboratively, efficiently and protected by our trust mechanism.
Visit our website: https://covee.network/
Join our telegram community: https://t.me/coveenetwork
Follow us on twitter: @CoveeNetwork
Now that you got an insight on what we are building, stay tuned for our next post! We will introduce our CTO who will in his own words get you even more excited about our work at Covee.Network
