Ocean Protocol Update 2023. What We’re Doing in 2023, and Why by Trent McConaghy Mar, 2023

What We’re Doing in 2023, and Why

Ocean’s first phase was building the core infrastructure for the open data economy. It’s now entered its next phase: to drive data value-creation loops, by focusing on the users in the last mile.Contents
1. Abstract
2. Introduction
3. Dapp Developers served by team Eagle-Rays (Stream 1)
4. Data Scientists served by team Thresher (Stream 2)
5. cripto-Enthusiasts served by team Sailfish (Stream 3)
6. Conclusion

Ocean was founded to create the building blocks and tools to unleash an open, permissionless and secure data economy. The first phase of Ocean was about building the core infrastructure for the open data economy. Ocean has entered its next phase: to drive data value-creation loops by focusing on the users in the last mile: data dapp developers, data scientists, and data-oriented cripto enthusiasts. To meet these aims, we’ve re-organized ourselves into three specific streams, one for each user type. This post shared the plans for each of the streams and their respective users.

2.1. Ocean Foundations

Ocean Protocol was launched in 2017 with a whitepaper and a promise: to create the building blocks and tools to unleash an open, permissionless and secure data economy.

We shipped Ocean V1 to enable data sovereignty (“your keys, your data”). Ocean V2 enabled privacy via Compute-to-Data. We shipped Ocean V3 in 2020 with ERC20 datatokens to leverage other cripto services: where every wallet could be a data custodian, every DEX a data exchange, every DAO a data DAO. V3 met the whitepaper’s goals.

In 2022 we shipped Ocean V4, which refined V3: data NFTs to clarify intellectual property rights and help publishers and marketplaces monetize. We also shipped Data Farming (DF) and veOCEAN staking. We met all our 2022 product goals. Since then, we’ve further refined the Ocean stack towards reliability and usability.

2.2. A New Phase: Drive Value Creation Via User Focus

The five years of Ocean (2017–2022) was the first phase: building the core infrastructure for the open data economy.

During this phase, hundreds of teams tried out new ideas, 20+ strong teams built data dapps, dozens of data scientists used Ocean in data competitions and more, and hundreds of cripto enthusiasts engaged with Ocean Data Farming — amongst the 45,000 holders of OCEAN.

How do these people sustain and thrive in this burgeoning Data Economy?

Our hypothesis is — to ensure that they can make money! Specifically, create value from data, make money from that value, and loop back and reinvest this value creation into further growth. All the while, Ocean aims to be the orchestration and tooling for deeper and more complex value loops. The focus of the Ocean core team for 2023 and beyond is to realize this hypothesis. Put another way:

Ocean has entered its next phase: to drive data value-creation loops, by focusing on the users in the last mile.

The figure below illustrates.

Figure: The Data Value-Creation Loop. At the top, the user gets data by either buying it, or spending $ to create it. Then, they build an AI model from the data, then make predictions (e.g. “ETH will rise in next 10 minutes”) and choose actions (e.g. “buy ETH”). In executing these actions, they will make $ on average. The $ earned can be looped back into further activities.

In this loop, dapp builders can help their users make money; data scientists can earn directly; and cripto enthusiasts can catalyze the first two if incentivized properly (e.g. to curate valuable data).

DeFi & LLMs. We analyzed dozens of possible verticals with respect to how quickly someone goes through the data value creation loop, and how much impact Ocean can have. The most promising and mature vertical is decentralized finance (DeFi). Already in Q4/2022, the Ocean core team started to bias activities towards DeFi in our data challenges, internal research and funding programs, without constraining the Ocean community or applications elsewhere. With the release of ChatGPT and influx of attention to AI, the vertical of large language models (LLMs) and related AI tech is highly promising. Ocean’s core proposition was to be at the intersection of AI, data and blockchain. The Ocean core team is exploring ways to support the growth of AI and stay ahead of the pack.

Loops, then Scale. Once one or two data-value creation loops have been established on Ocean, where people are sustainably making money, we will help to scale those loops up by extending our technology and integrations so that developers and data engineers can do more and more over time. Our aim is to grow over the long-term, until Ocean is ubiquitous as a tool for orchestration and monetization.

2.3. 2023+ Ocean Core Team Goals & Setup

Ocean core team has set its aims accordingly: help data dapp developers, data scientists, and cripto enthusiasts to drive data value-creation loops.

To meet these aims, we’ve re-organized ourselves into three specific streams: Stream 1 (Eagle Rays) for developers, Stream 2 (Thresher) for data scientists, and Stream 3 (Sailfish) for cripto enthusiasts. The figure below illustrates.

Ocean core team’s aims: help (1) data dapp developers, (2) data scientists, and (3) cripto enthusiasts to drive data value-creation loops. The team is organized around these three aims.

2.4. Key Performance Indicators (KPIs)

Ocean core team has transitioned from building out a product specified by a whitepaper and annual roadmaps, to doing “whatever it takes” to drive data value creation. Accordingly, the measure of success changes from “was X built” to “how well are metrics for the data economy growing?”

Specifically, we track progress via these metrics:

The main metric is data consume volume (DCV) in USD on Ocean, i.e. how much money is spent buying & consuming data and AI assets in a given time period.

Other key performance indicators (KPIs) are:

# Ocean transactions,# dapps deployed using Ocean,# data scientists using Ocean,# assets published on Ocean, and# OCEAN locked.

We have set up internal measures for these metrics and will make them public when ready.

KPIs for Ocean core team. The main metric is $ data consume volume (DCV). It’s flanked by five supporting metrics.

2.5. Outline

This introduction has outlined Ocean’s first phase — creating the building blocks & tools as a foundation for an open, permissionless, secure data economy. And it introduced Ocean’s next phase: driving data value-creation loops, by focusing on the users in the last-mile.

We are now in a position to share the plans for each of the teams, and their respective users. The rest of this post is organized as follows:

Section 2 — DApp Developers served by Team Eagle-Rays (Stream 1)Section 3 — Data Scientists served by team Thresher (Stream 2)Section 4 — cripto-Enthusiasts served by team Sailfish (Stream 3)[Spotted Eagle Ray. Image by John Norton. License: CC-BY]

3.1. Overview

Ocean core team has created building blocks and tools towards enabling an open, permissionless, secure data economy [1]. Dapp developers in the ecosystem can then build last-mile data dapps that use these building blocks and tools. Ocean makes it possible to quickly build & ship decentralized data marketplaces, token-gated dapps, dapps with secure non-custodial user data, and more.

While the Ocean stack has made great progress, there are still challenges. The developer experience is suboptimal and sometimes leads to projects being “lost” in the fray. The core team has only partial visibility on dapps being built, and often fails to provide appropriate support to the development teams. Finally, it can be difficult for developers to see the full potential of possible use cases when given just the building blocks & tools.

We can improve this. The priority for team Eagle-Rays is to help developers build sustainable dapps on top of Ocean Protocol with minimal friction.

The target outcomes for team Eagle Rays are:

A developer can build a sustainable dapp on Ocean stack, within 1 month of interaction with Ocean Protocol development team. Where “sustainable” means that the dapp creates income towards the team supporting itself with a viable business model, and ultimately thriving.Ocean becomes the meeting point for data dapp builders — a place for sharing development tools & templates, for support, and for funding of new ideas.

Team Eagle-Rays will run these activities towards achieving the goals:

Gather community feedbackImprove dapp developer supportRun a dapp dashboard to highlight projects building on OceanCreate demos to expose the capabilities of Ocean to developersHost workshops & hackathons to onboard developers into OceanProvide referrals & funding for promising teams

The following subsections have details.

3.2. Gather Community Feedback

Our first goal is to gain a better understanding of the current state of affairs.

In the coming weeks we will have one-on-one conversations with the 20+ teams building data dapps on Ocean. We will understand the challenges they face and how we can serve them better in terms of tools and the technical support.

From this, we’ll compile a list of learnings, hypotheses and follow-up actions.

3.3. Improve Developer Support

Rather than being merely reactive — responding to requests when something is not working — we aim to be proactive in supporting the developers building dapps on top of Ocean Protocol. To assist the teams from the moment they begin their journey, advising on the design, through to final implementation, and feedback after the dapp goes live.

To better allocate internal resources towards high-potential projects, we will define tiers of guidance & technical support. Allocation is based on the level of engagement the dapp team has with Ocean Protocol. It may be as much as a dedicated core team member being allocated to that dapp team.

We will consolidate existing technical support channels and revamp the support procedures. We will refresh the product documentation, to speed up onboarding of developers.

3.4. Run a Dapp Dashboard

To further raise the prominence of dapps on Ocean, we will launch & run a new Dapp Dashboard, linked from oceanprotocol.com.

Every month, the dashboard will highlight one or two teams building on Ocean, to spread the word and help raise awareness for the teams.

3.5. Create Demos & Scaffolding

Having built the building blocks and tools of the Ocean stack, Ocean core developers intimately know the potential of Ocean to unlock data sharing and monetization.

To accelerate broader understanding and education, core developers will build a portfolio of demos and scaffolding (templates), to showcase the capabilities of Ocean, show what is possible and inspire more developers to build their ideas on Ocean. We aim to release one demo per month, showcasing various business scenarios or use cases where Ocean can be used.

3.6. Host Workshops & Hackathons

To engage more with our developer community, by organizing bi-weekly and monthly technical workshops where we’ll take deep dives into selected technical aspects of our platform.

We will run quarterly hackathons, with the target goals:

Rapid prototyping: To quickly test new ideas for sustainable data dapps.Community building: To bring people together in the Ocean community and together explore what a new data economy can look like.Innovation & idea generation: To generate new use cases for within the new data economy using OceanSkill development: To help dapp developers learn new skills, via a project that pushes them outside their comfort zone.

3.7. Provide Referrals and Funding

Once a dapp team is building on Ocean, the Ocean core team offers a range of non-technical support to help ease launching a viable dapp. This includes referrals to vetted legal, financial and organizational partners.

It also includes potential funding support, as follows:

Ocean Shipyard — funding high quality teams to test out their ideasOcean Ventures — investing between $25,000-$100,000 into Pre-Seed and Seed rounds for dapp teamsOcean Ecosystem Fund — a $10 million pool of capital pledged by Cypher Capital and Faculty Fund to invest in dapp teams[Thresher Shark. Image by Raven Malta. License: CC-BY-SA]

4.1. Overview

Team Thresher’s aim is to help data scientists make $ from their data and algorithms on Ocean.

Let’s unpack. Thresher’s target users are data scientists. They use Python extensively. They’re quite familiar with key AI/ML tools like numpy, and scikit-learn. They love algorithms. They are Web3-curious, though not necessarily Web3 experts. They loathe devops.

The Ocean value proposition for them is: with Ocean, data scientists can sell compelling data feeds powered by an algorithm of their design, while the algorithm stays private, with zero devops.

Accordingly, Team Thresher has these top-level goals:

Make it easy for data scientists to create compelling data feeds and to sell them via OceanGrow the community of data scientists creating & selling compelling data feeds.

Team Thresher will run these activities towards achieving the goals:

Run predict-ETH & other objective data challenges, which guide data scientists to sell compelling data feedsRun subjective data challenges, which drive insight, community, and partnershipsShip open, compelling data feeds, which data scientists can fork and ship as their ownReduce friction in data scientists’ (Python) flows. Especially, make ocean.py “just work”Other tools to demonstrate capabilities and drive traction

Each subsection below elaborates on these points.

4.2. Run Predict-ETH & Other Objective Data Challenges

Team Thresher uses Data Challenges to help walk data scientists down the garden path to monetize. They start by entering competitions, yet end up shipping compelling data feeds that make $$. Here’s how.

We use Data Challenges to help walk guide data scientists towards making money with their algorithms. They start by entering competitions, and end up shipping compelling data feeds that make $$.

Data challenges are competitions that invite participants to solve data-related problems, with prize money attached. They provide a platform to practice solving data problems, to collaborate, and to make $. They have become popular in recent years, being hosted by many companies and research institutions.

In a typical data challenge, a data scientist learns about the challenge, works on it to devise an algorithm, and submits an entry to the competition.

We take this further: the data scientist also submits a data feed powered by the data scientist’s algorithm, where the data feed is live and for sale on Ocean Market. If this algorithm is compelling enough, then people will buy it, and the data scientist will make $.

Predict-ETH. The prime example is the monthly Predict-ETH challenge. This challenge has competitors submit predictions for the price of ETH 1, 2, …, 12 hours into the future. If the data scientist can create accurate predictions, there’s $ to be made from trading. This is compelling data. In upcoming rounds of Predict-ETH, competitors will also be guided to submit C2D data feeds that generate ETH predictions upon request (powered by Ocean C2D stack). These feeds are directly monetizable. If the predictions are good, others will buy them.

Now, the data scientist is making $ by winning the competition, AND making even more $$ by selling the feeds of their algorithms.

When they’re making $, they’re sticking around to make more $, all using the Ocean stack.

This scales. As data scientists make $ and stick around, they share with their friends, leading to organic growth. So an initial handful of data scientists can grow to tens, then hundreds, then thousands. All creating compelling data feeds; all kickstarting an open, permissionless, secure data economy.

4.3. Run Subjective Data Challenges

The strategy of the previous section is best suited to data challenges with an objective quantitative metric like normalized mean-squared error (nmse) like in Predict-ETH.

There are also subjective data challenges — where judges assess qualitative metrics. We are running these too. Here’s why.

Benefits to data scientists:

Tools to demonstrate capabilities and drive adoption: helps data scientists to promote their work. This can include dashboards, visualizations, and other features that help users give insights into data.A means for organizations to partner with Ocean core team and share their data. The partner gets the benefits of people analyzing their data, and learning more about the Ocean stack. E.g. Government of CataloniaProviding more broadly accessible competitions. Subjective challenges appeal to a different demographic of data scientists — ones that prefer the “insight” side of data science over stark objective values

Benefits to Ocean core team:

Qualitative feedback on Ocean stackLearn more about data scientist tools and workflowsLearn more about a new vertical, problem space, or subproblem. Without having to go directly to an objective data challenge.

4.4. Ship Compelling Data Feeds

The best way for competitors to create compelling data feeds is… to fork other compelling data feeds! So, Team Thresher is working to create compelling data feeds that are also open (and therefore forkable). To start, we’re doing this for predictions of ETH. We’ll do this for other objective competitions in the future as well.

Q: if Ocean can make such valuable algorithms, why would it make them open?

A: Ocean aims to kickstart a whole data economy, which means we want many participants shipping compelling data feeds. To catalyze this, we give them a starting point, where they can make $ with little extra effort!

4.5. Reduce Friction in Data Scientist Flows

As described, Thresher’s target users are data scientists, who use Python extensively.

We want to make it as low-friction for them to use Ocean in their Python-based flows. Especially in the flows around monetizing their algorithms, as discussed above.

To reduce friction in their flows, this includes: make ocean.py “just work”. ocean.py is a Python library on pypi. It’s the main interface for Python users to get Ocean capabilities, such as publishing a file or algorithm, creating datatokens (access tokens), and sharing or selling those. We’ve discovered that when people use it, they often run into “gotchas” in OSes or dependencies on Ocean Aquarius / Provider.

This also includes: make ocean.py C2D flow “just work” (C2D = Compute-to-Data). C2D has many moving parts. Can we simplify its UX and make it more reliable?

Furthermore, this includes: making ocean.py more accessible in a broader set of Python workflows. E.g. have it as part of the default Anaconda distribution.

4.6. Other Tools to Demo Capabilities and Drive Traction

We see many opportunities of tools to help potential users better understand Ocean capabilities, try out Ocean, and ultimately drive adoption. Especially in DeFi and LLMs. We keep this open-ended here for now.

[Sailfish. Image by Rodrigo Friscione. License: Public Domain]

5.1. Overview

Last year, the Ocean core team launched Data Farming and veOCEAN. veOCEAN aligns near-term with long-term perspectives in the burgeoning data economy. By Q4, Data Farming was driving up to 1.5M $USD in Data Consume Volume (DCV) per week with up-to 75K $OCEAN in rewards being distributed to data farmers.

Every participant who contributes value to this data economy should benefit from it. This includes work by staking & curating, building dapps that integrate veOCEAN & Data Farming, or otherwise.

Team Sailfish builds on this promising start. Our mission is to help cripto enthusiasts — especially web3-native Ocean participants — to earn in the Ocean ecosystem. These participants do work to reap economic rewards, and the system harnesses this work to further drive Ocean DCV via dapps and data scientists. For example, staking on data assets helps data curation in Ocean Market.

Besides driving DCV, another KPI we focus on is to increase the # participants in Data Farming and their APYs, by specifically targeting (a) OCEAN holders that have not yet engaged with veOCEAN (b) veOCEAN holders whose APYs can be improved, and (c ) Publishers who can benefit from boosting.

Towards these goals, Team Sailfish will run these activities:

Launch & Complete DF MainRefine DF RewardsServe DeFi dapp Builders BetterOther KPI-Driving Activities

The following subsections have details.

5.2. Launch & Complete DF Main

DF has several phases: DF Alpha, DF/VE Alpha, DF Beta, and DF Main. DF Main is soon: it starts in DF Round 29 (Mar 26, 2023).

DF Main is the final “production” phase where the largest amounts of OCEAN are dispensed. It vests OCEAN with a 4-year half-life (like Bitcoin) into rewards for Data Farming.

Given the large amounts of OCEAN, we follow the time-tested approach of ratcheting up value-at-risk over time. Specifically, we’ll first deploy a “canary” vesting contract that will hold just a small amount of OCEAN, and manually top up weekly DF rewards. Once the system has stabilized and contracts are verified (via bug bounty & security audit) then we will deploy the remaining amount.

In completing DF Main, OCEAN vesting & related will be fully automated and on-chain. This will serve the Ocean ecosystem well in the decades that follow, with more transparency, stability, and composability.

5.3. Refine DF Rewards

Improvements to the Reward Function (RF) create new pull-mechanisms that help drive DCV and other metrics. An example is the introduction of Data Farming and how much Data Consume Volume it drove.

Every week DF emits OCEAN rewards, according to a pre-set schedule. We want to maximize the bang-for-the buck of these emissions in driving traction.

Accordingly, we spend effort to refine DF rewards. Recent changes include:

We recently introduced publisher rewards — giving publishers 2x effective stake — to catalyze more data being published into Ocean.For DF29, wash consume will become unprofitable. Until then, it’s profitable because users receive more rewards than the 0.1% fee they pay.

To drive DCV further, we envision more refinements of DF Rewards. Besides tuning the existing function, we can split the weekly active DF rewards budget into sub-streams. New sub-streams could include:

Dapp Data Farming. If a dapp running Ocean uses x$ of gas, pay the dapp developer 25–100% * x (denominated in OCEAN). This could be an excellent way for dapp developers to monetize [2]Competition Data Farming. At some point, some of the Stream 2 data competitions can get streamlined & automated enough to put into weekly DF ops.Other publisher schemes. Eg high-touch schemes to onboard publishers and ensure they get paid. Maybe less automated, but could be vertical helpful in the next 6–18 months.

5.4. Serve DeFi Dapp Builders Better

We will know that we have succeeded when we can observe many products, features, and case studies showing organic demand and validate end-to-end value flows on top of the protocol.

In 2022, it was important for us to build the systems that get us from 0 → 1 with veOCEAN and DF. So, our focus was largely on core software.

For 2023, our focus is core software and customer success — success of (a) OCEAN holders and of (b) DeFi-oriented dapp builders. Previous sections focused on (a) OCEAN holders.

We want to serve (b) DeFi-oriented dapp builders better. Specifically, builders that are building out the DeFi side of OCEAN, veOCEAN, and Data Farming.

A prime example is H2O’s introduction of psdnOCEAN, for liquid staking of OCEAN. psdnOCEAN provides improved UX, higher APY and more liquidity than default veOCEAN. We view this as a success story.

We ask: what other DeFi-oriented dapp builders can we engage with, and serve better? Can OCEAN or veOCEAN integrate with DeFi loan protocols like Aave? Can we guide teams to build fractionalized Data NFTs, revenue-backed loans, or other?

5.5. Other KPI-Driving Activities

We aim to improve these KPIs:

# OCEAN holders participating in veOCEANAPY for veOCEAN and Data Farming participants# Publishers with multiple quality data setsPublisher volume and APY

We have several hypotheses of how to drive these KPIs. We don’t know which will work. We will test ideas, learn, and iterate. Therefore, this section is open-ended. Let’s see where it takes us!

Ocean was founded to create the building blocks and tools to unleash an open, permissionless and secure data economy. The first phase of Ocean was about building the core infrastructure for the open data economy. Ocean has entered its next phase: to drive data value-creation loops, by focusing on the users in the last mile: data dapp developers, data scientists, and data-oriented cripto enthusiasts.

To meet these aims, we’ve re-organized ourselves into three specific streams: Stream 1 (Eagle Rays) for developers, Stream 2 (Thresher) for data scientists, and Stream 3 (Sailfish) for cripto enthusiasts.

This post shared the plans for each of the streams, and their respective users.

[1] Besides building block primitives and tools, the Ocean core team may also build with “baseline apps” like Ocean Market (decentralized data market)

[2] It’s a bit like how NEAR and Canto L1s pay a % of tx fees to dapp developers.