Your importance (social=market) driven inbox
Each msg comes with arbitrary (of objective xor subjective value) currency attached
You receive the currency if you reply
A -> B
-
A sends a msg to B some coins attached
-
A can cancel anytime, whereby the msg disappears from B's inbox and A gets the currency back
-
B receives the currency by replying to the msg
demo: https://youtu.be/qu6TXRGHSr8
FairInbox uses FairMarket for the inbox ordering and as the trustless escrow and FairFX as the trustless FX provider
The default order of common inboxs nowadays is chronological. In the 90's, Google created an ordering algorithm according to relevance to a search term FairMarket is an algorithm that orders msgs fairly according to market driven importance
Currently, most people do not open their inboxes to the world because they would be overwhelmed by the amount of messages
Respecting the recipient's time is acknowlading that time (replying to msgs) costs energy. Hence it is respectful to attach currency to msgs to replinish the energy of the recipient and respect it's time
The currency can be arbitrary, of objective value (e.g. USD coin, Gold coin, etc.) xor of subjective value (e.g an art NFT, ticket NFT, etc.)
The ordering of such an inbox is according to social importance. Society/others determine how important a msg is for the recipient The order is not simply more-is-better. The FairMarket algorithm fairly interweaves those msgs that offer the minimum required, those that offer more and those that offer subjective value
Evolve on the current email system
We could solve X's revenue problem by allowing dms to come with currency. Trivial to monetize by taking cuts of transactions. Immediately jump starts a new economy whether influencers replying to followers, advisers giving advice, friends respecting each others time
This dapp is useful for mainstream and can bring real adoption
Smart contracts were written using tealish and make use of modern AVM features like box
es
Smart contracts make this dapp zero-credit-risk: send large value currency without fear
FairInbox will exist as a FairEnterprise to give trustless and fair access to value to the community
The frontend part of the dapp is minimalistic
We use a mathematical proven trustless computer called Algorand that proves to the user the exact code that runs when the dapp is run. The trustlessness also allows for secure transfer of energy and information, without any need to trust any authority Eventually, Algorand will connect via bridges to other trustless computers, giving users elsewhere access to it's FairInbox everywhere; alternatively, FairInbox can also be ported and made available on other trustless computers if that strategy is superior. Algorand would be considered the base barring unforeseen information
Note that the world trustless refers to systems that do not require the user to trust any operator, whilst the word trusted refers to the opposite, systems that require trust to the operator, as ironic as that sounds.
The non smart contract code of the dapp (trusted computing) is live on IPFS, as a bundle of wasm/js/html, currently served by fleek
Serverless, trustless, permissionless
Whitelisted user provided css will be used for dapp. Email: [email protected]. The plan is to support this as soon as the service is used, for its function.
todo: Use e.g. TinyGo to build dApp WASM. Depends on whether the trustless computer sdk can be compiled with TinyGo
npx webpack
./node_modules/.bin/http-server -a localhost dist
npx webpack && ./node_modules/.bin/http-server -a localhost dist
build to wasm:
tinygo build -o ./dist/FairInbox.wasm -target wasm .
need cmd to serve wasm+html+js
a golang module with multiple packages that builds to wasm+html+js
tests to run. no main.
(future) CI: push wasm+html+js to ipfs and change ipns
The trusted computer code contains vanilla js and html only (future: use WASM instead of js). No js frameworks or libs are used besides the trustless computer sdk, js-sha512 and the wallet connector sdk. This minimizes the risk of injecting malicious code
Smart contract are secured by asserting conditions early. That work was live streamed as well
potential todo: As the theory asks, we choose the learning algorithm (information) as an LLM because our data consists of text. LLMs can understand, given enough examples, and with privacy (your own LLM) that a user can choose this forecasted strategy vs random, which gives a partial ordering overall
golang -> WASM -> IPFS -> IPNS
user: IPNS -> WASM -> WASM runtime -> API -> ACTION -> VIEW
action: API -> trustless computer
views: WASM -> HTML WASM -> ???
Dev of the project was/is streamed best-effort: https://www.youtube.com/@1m1-yt/streams