diff --git a/.gitbook/assets/data-farming/predictoordf_main.png b/.gitbook/assets/data-farming/predictoordf_main.png new file mode 100644 index 00000000..319c67e3 Binary files /dev/null and b/.gitbook/assets/data-farming/predictoordf_main.png differ diff --git a/.gitbook/assets/gif/making-money-is-fun.gif b/.gitbook/assets/gif/making-money-is-fun.gif new file mode 100644 index 00000000..f1930f3d Binary files /dev/null and b/.gitbook/assets/gif/making-money-is-fun.gif differ diff --git a/.gitbook/assets/gif/money-robot.gif b/.gitbook/assets/gif/money-robot.gif new file mode 100644 index 00000000..4ea8aa78 Binary files /dev/null and b/.gitbook/assets/gif/money-robot.gif differ diff --git a/.gitbook/assets/gif/my-mood-plot.gif b/.gitbook/assets/gif/my-mood-plot.gif deleted file mode 100644 index e19cccb2..00000000 Binary files a/.gitbook/assets/gif/my-mood-plot.gif and /dev/null differ diff --git a/data-farming/apys-plot.md b/data-farming/apys-plot.md index 4a7d56a3..5e9e5127 100644 --- a/data-farming/apys-plot.md +++ b/data-farming/apys-plot.md @@ -3,7 +3,7 @@ description: >- Modeling APYs --- -

+

# Estimated APY vs Time diff --git a/data-farming/claim-rewards.md b/data-farming/claim-rewards.md index e9abfb24..74206bff 100644 --- a/data-farming/claim-rewards.md +++ b/data-farming/claim-rewards.md @@ -6,9 +6,11 @@ description: Unbox your OCEAN rewards
-To claim OCEAN rewards, go to [DF Rewards page](https://df.oceandao.org/rewards) and click the "Claim OCEAN rewards" buttons. +**This page describes how to claim DF rewards.** (Predictoor DF is an exception, [see here](predictoordf-guide.md).) -A detailed guide is below. +To claim rewards, go to DF webapp's **[Rewards page](https://df.oceandao.org/rewards)** and click "Claim OCEAN rewards" button. + +Detailed instructions are below. ### Step 1 - Navigate to the Data Farming Rewards page diff --git a/data-farming/predictoordf-guide.md b/data-farming/predictoordf-guide.md index 41c1ab5d..74d7fd0b 100644 --- a/data-farming/predictoordf-guide.md +++ b/data-farming/predictoordf-guide.md @@ -1,10 +1,27 @@ --- description: >- - Guide to run prediction bots to earn continuously. + How to earn $ via predictoor DF --- # Guide to Predictoor DF -In Predictoor DF, you run prediction bots to earn continuously. Starts on DF62. +
+ +In Predictoor DF (and Predictoor proper), you run prediction bots to earn continuously. + +## How to be eligible for Predictoor DF + +**Running a predictoor bot will automatically make you eligible for Predictoor DF rewards.** + +Therefore, you simply need to become a predictoor! See [Predictoor docs](../predictoor/README.md) or go directly to the [predictoor quickstart README](https://github.com/oceanprotocol/pdr-backend) + +## How to claim Predictoor DF Rewards + +Claiming comes in two parts: OCEAN and ROSE. + +1. **For OCEAN rewards:** recall that these act as more sales coming to you (as a predictoor). So you claim your OCEAN from sales in the usual way, by running the [OCEAN payout script](https://github.com/oceanprotocol/pdr-backend/blob/main/READMEs/payout.md). + +2. **For ROSE rewards:** run the ROSE payout script, linked from [here](https://github.com/oceanprotocol/pdr-backend/blob/main/READMEs/predictoor.md). + +NOTE: this may be updated before Predictoor DF launch. -(Note: this doc will get fleshed out soon.) diff --git a/data-farming/predictoordf.md b/data-farming/predictoordf.md index ed75400a..51b48c02 100644 --- a/data-farming/predictoordf.md +++ b/data-farming/predictoordf.md @@ -1,16 +1,88 @@ --- description: >- - Run prediction bots to earn continuously. + Baseline sales for predictoors --- -# Predictoor Data Farming - +
**This page** is about Predictoor DF, and [this page](predictoordf-guide.md) is a guide. # About Predictoor DF -In Predictoor DF, you run prediction bots to earn continuously. Starts on DF62. +**Predictoor DF** is a substream of Active DF that amplifies predictoors’ earnings, via extra sales to Ocean Predictoor data feeds. + +Predictoor DF has 37K OCEAN weekly rewards (ongoing) and 100K ROSE rewards (first 4 weeks). + +The higher baseline sales makes Volume DF and Passive more attractive. + +## Introduction + +**[Ocean Predictoor](../predictoor/README.md)** data feeds predict whether BTC, ETH etc will rise or fall 5min or 1h into the future. These feeds are crowdsourced by “predictoors”: people running AI-powered prediction bots. + +**[Data Farming (DF)](../data-farming/README.md)** is Ocean’s incentive program, that rewards OCEAN to people who lock OCEAN or do active DF activities. + +You should be familiar with both Predictoor and DF before reading on. + +**Predictoor Data Farming is a substream of Active DF. It amplifies existing predictoors’ earnings based on their accuracy and stake.** + +## Predictoor DF Timing + +Predictoor DF starts counting on Nov 9, 2023, at the beginning of Data Farming Round 62 (DF62). It runs indefinitely. + +## Predictoor DF Rewards + +Predictoor DF has two components: [OCEAN](https://oceanprotocol.com/about-us/ocean-token) rewards and [Oasis ROSE](https://www.coingecko.com/en/coins/oasis-network) rewards. + +### OCEAN Rewards + +OCEAN payouts are 37,000 OCEAN/week, ongoing. +- A special “DF buyer” bot purchases Predictoor feeds. It starts operating on Nov 9. Every day, it spends 1/7 of the weekly Predictoor OCEAN budget for another 24h subscription. It spends an equal amount per feed. (Currently there are 20 feeds: 10 x 5min, 10 x 1h.) +- The OCEAN comes from the Ocean DF budget, as part of the 75,000 OCEAN/week for Active DF. The Volume DF budget has been adjusted to 37,000 OCEAN/week, and Challenge DF to 1,000 OCEAN/week. Here are details. + +### ROSE Rewards + +ROSE payouts of 100,000 ROSE/week for the first 4 weeks of Predictoor DF. +- Payout is at the end of the DF round. Therefore there will be payouts at the end of DF62, DF63, DF64, and DF65. +- Payout for a given predictoor is pro-rata to the net earnings of that predictoor over that DF round, specifically (total sales $ to the predictoor) minus (predictoor stake slashed due to being wrong). +- The ROSE comes from a generous contribution of[ Oasis Protocol Foundation](https://oasisprotocol.org/) 👪🙏. -(Note: this doc will get fleshed out soon.) +## How to Earn $ Via Predictoor DF + +**Running a predictoor bot will automatically make you eligible for Predictoor DF rewards.** + +The [Predictoor DF user guide](predictoordf-guide.md) tells how to get started as a predictoor, and how to claim rewards. + +## How to Earn More $ Via Passive DF & Volume DF + +Predictoor DF makes [Active DF](../data-farming/activedf.md) more attractive, and in turn [Passive DF](../data-farming/passivedf.md). You should be familiar with both before reading on. + +**Predictoor DF makes Volume DF more attractive than status quo**, because the volume-based bounds on weekly rewards trends tend to be markedly higher for prediction feeds. (The Appendix has details.) + +**Curating is straightforward.** Ocean assets with high DCV are easy to identify: it's the 20 OPF-published prediction feeds. This makes the choice of veOCEAN allocation easy: point to those 20 assets. (Only OPF-published feeds are eligible.) + +**Doing Volume DF gives Passive DF rewards too**, of course. + +The net result: Predictoor DF means earnings potential from Predictoor DF, Volume DF, and Passive DF. + +## Appendix: On DCV Bounds of Prediction Feeds + +**Predictoor DF makes Volume DF more attractive than status quo.** Active DF's rewards are bounded by `DCV_bound` which in turn is bound by sales volume and fees. In Predictoor DF, both those factors are raised; this raises the bounds in Volume DF; this in turn means higher earning potential. Let's elaborate. + +At one time, Volume DF had a "wash consume" problem, where people published and consumed their own datasets. [DF9](https://blog.oceanprotocol.com/data-farming-df8-completed-df9-started-reward-function-tuned-d74b5134b5d1) onwards address this, by putting a bounds on weekly rewards: + +> `DCV_bound = DCV * m` + +Where `m` = `DCV_bounding_multiplier` = `Ocean community fee (0.1%) + publish market fee` + +This stopped wash consume 💪 because it became unprofitable to _do_ wash consume: fees eat up all potential profits. + +A low `DCV` (data consume volume) or a low `m` (publish market fee) mean low `DCV_bound`. + +Predictoor DF makes both `DCV` and `m` higher! + +- **`DCV` is higher:** the 37K OCEAN/week counts as consume volume +- **`m` is higher:** publish market fee is 20% for prediction feeds + +Therefore `DCV_bound` is higher. Specifically: `DCV_bound = 37000 * (0.001 + 0.20) = 7437 OCEAN`. In other words, at least 7437 OCEAN is available for Volume DF in any given week. + diff --git a/predictoor/pdr-earn.md b/predictoor/pdr-earn.md index 95918398..732e7bef 100644 --- a/predictoor/pdr-earn.md +++ b/predictoor/pdr-earn.md @@ -13,7 +13,7 @@ Typical steps as a Predictoor: The actions as a predictoor give the following ways to earn: * **Feed sales.** At an epoch, sales revenue (minus fees) for that epoch goes to predictoors. It’s distributed pro-rata by stake among the predictoors who predicted the true value correctly. The revenue for an epoch is the fraction of sales, spread uniformly across subscription length. A price of 3 OCEAN, 5m epochs, and 24h (1440m) subscriptions gives a revenue of (# subscribers) * (3 OCEAN) * / (1440m / 5m). * **Stake reshuffling.** At an epoch, incorrect predictoors have their stake slashed. This slashed stake is distributed to the correct predictoors pro-rata on their stake. -* **Predictoor DF.** The third phase of Predictoor rollout will have an incentives program that amounts to additional earning for predictoors. 37,000 OCEAN/week rewards. +* **Predictoor DF.** This amounts to additional earning for predictoors. 37,000 OCEAN/week rewards. [Here are details.](../data-farming/predictoordf.md) Don’t expect to be 100% accurate in your up/down predictions. Marginally better than 50% might be enough, and be skeptical if you’re greatly above 50%, you probably have a bug in your testing. Every week there’s [150,000 total OCEAN rewards](../data-farming/df-intro.md#reward-schedule) for Ocean Data Farming. This will increase in early 2024, and more yet later.