Sharpe Platform: UI Core Concept Designs
Today we are pleased to reveal initial designs for the mobile app that will underpin the Sharpe Platform. This includes a viewer for our ‘trustless ledger system’ (TLS) for monitoring Sharpe’s investment activity, our platform for crowd-sourcing investor sentiment, and our motion tabling-based democratic community governance structure.
Once a user is logged into their Sharpe Platform account, which is paired with the Ethereum address in which they hold Sharpe Utility Tokens (SHP), they are greeted with the Dashboard view.
This view presents the user with all closed positions to date, and provides details on the fund size and year-to-date growth (YTD). These details are stored immutably on the blockchain, as described in our white paper (Section 3, pp. 11–14). Utilising our TLS, we encode all our trades on the blockchain using RSA encryption. When a position is closed, the keys required to decrypt the details of the trade are also added to the blockchain. This creates a trustless, immutable record of all our previous trading history that anybody can view through our web/mobile platform or directly on the Ethereum blockchain, at any time. This permits any interested parties to freely audit our activity, without requiring us to reveal sensitive information, such as currently open positions.
The first section - ‘Sentiment’ - provides a platform for users to receive Ether (ETH) rewards in exchange for telling us whether they feel bullish or bearish on a particular asset. This system utilises a ‘proof-of-stake’ through SHP token ownership and a ‘proof-of-reputation’ with each user having an immutable reputation score stored on the blockchain. Rewards are calculated from the Sharpe Capital discretionary Ether reward pool, the user’s proportion of SHP owned, and their ‘reputation score’.
The reputation score is a reflection of historical accuracy in providing sentiment towards assets. Each user begins with a reputation score of 0.5, and this moves towards a maximum of 1.0 and a minimum of 0. Every correct prediction increases the reputation score. Likewise, every incorrect prediction decreases it. The precise calculations, with real-world examples can be found in our white paper (Section 6, pp. 33–37). The reputation score allows us to weight user provided sentiment according to their accuracy in different sectors, improving the returns of our fund.
The key distinguishing feature of this platform with respect to ‘prediction markets’ is that there are no losses for incorrect predictions, and no ‘odds’ given. An incorrect prediction merely results in a reduction in future reward payments due to a reduced reputation score. This incentivises users to consider each sentiment indication they provide, as achieving a maximum reputation score of 1.0 would double the size of reward payments received.
The following interface is one of several methods that users will have available to them to provide sentiment indications or predictions:
For example, a user who provides correct sentiment 70% of the time when shown technology companies, but only 20% of the time when shown pharmaceutical companies, will be shown more assets in the technology sector and less in the pharmaceutical sector. This mechanism ensures that we’re crowd-sourcing the highest quality sentiment, adapted to each user's experience. This further serves to increase the rewards users will see, as our algorithm works to show them assets it expects the user to perform better on.
Users are intentionally not shown charts of prices for assets. This is because we wish to gauge investor sentiment and not simply ‘chart watching’. There is little-to-no evidence that chart watching, sometimes referred to as ‘technical analysis’, has any efficacy in forecasting prices beyond extremely short time-scales. For this reason, we also ask users only to provide binary indications of ‘positive’ or ‘negative’, as opposed to alternatives such as a visual-analogue scale score in which a user provides a relative assessment from ‘very negative’ to ‘very positive’. Instead, we will construct a quantitative measure of overall sentiment by aggregating the sentiment of many participants.
Sharpe Platform Participant Wallets
The ‘Wallets’ option in the menu provides an interface for the users Ethereum wallet used for storing SHP tokens.
The ‘Wallets’ view shows each user’s current ownership of SHP tokens, ETH, and, once available, Sharpe Crypto-derivative tokens (SCD). Future funds created using our consensus-based or democratic motion tabling-based community governance structures will have their own token identities and will also appear in this view.
Community Governance through Motion Tabling
The final view we want to share with our community today is that of our ‘community governance’ motion tabling and voting mechanism. This is presented below:
Votes are weighted according to each participant’s SHP ownership. For this reason, we are utilising dynamic ceilings for the duration of our public crowd-sale; this serves to ensure that no one participant can accumulate a large majority of SHP tokens – this would fundamentally undermine the democratic nature of the Sharpe Platform.
We are working tirelessly to improve on these designs, and will have more to share with the community soon on the development of the Sharpe Platform. We hope that these designs help make many of the complex concepts described in our white paper in a somewhat abstract manner more intuitively understood, by visualising how they will operate in practice.
We encourage community feedback on every faucet of our development process, and welcome feedback from the community on what we have presented to you in this article.
The Sharpe Capital White Paper, describing our financial markets protocol and investment platform, is available for download here.
Due to the significant demand for our token and product, we are encouraging interested participants to register for our pre-sale by getting in touch at email@example.com. To help us cope with demand, we are formalising the sign up process and will release a form for this on our website on the coming week.
Note: All white paper page numbers cited in this post are based on version 1.0, released 2017–08–23.