Architecture design

This section of the book shows the software architecture of DarkFi and the network implementations.

For this phase of development we organize into teams lead by a single surgeon. The role of the team is to give full support to the surgeon and make his work effortless and smooth.

Release Cycle

gantt
    title Release Cycle
    dateFormat  DD-MM-YYYY
    axisFormat  %m-%y
    section Phases
    Dcon0            :done, d0, 06-01-2023, 120d
    Dcon1            :done, d1, after d0,   120d
    Dcon2            :done, d2, after d1,   120d
    Dcon3            :done, d3, after d2,   60d
    Dcon4            :      d4, after d3,   14d
    Dcon5            :      d5, after d4,   7d

The first 10 weeks overlap with the Dcon3 & Dcon4 phases of the previous release, and many developers will focus on bug fixing in those first weeks.

Developers dedicate a steady 1-2 days/week to the bug tracker, focusing on triaging and newly introduced bugs.

If a new feature is too unstable or incomplete, it will be reverted before the end of this phase. Developers spend 2-3 days/week in the bug tracker, triaging, fixing recently introduced or prioritized module bugs.

Development moves to the stable stabilizing branch. In master Dcon1 for the next release starts. stable is regularly merged into master.

High priority bugs dictate how much time developers will spend in the tracker as oppose to work on the next release Dcon1 features.

Release candidate and release builds are made. Developers spend a short time 5 days/week with an eye in the tracker for any unexpected high priority regression.

The final switch is flicked on dark.fi for the new release to show up on the Download page.

Mainnet Roadmap

High-level explanations and tasks on the mainnet roadmap, in no particular order. Some may depend on others, use intuition.

DAO Smart Contract

The DAO needs to have a parameter that defines the length of a proposal and the time when it is allowed to vote. Could be a start and end time or just end time. After end time has passed, new votes should be rejected, and only DAO::Exec would be allowed.

The DAO also has to implement ElGamal-ish note encryption in order to be able to be verified inside ZK. darkfi-sdk already provides an interface to this, although not providing an interface to the zkVM, just external. (See ElGamalEncryptedNote in darkfi-sdk). The cryptography also has to be verified for correctness, as this was just a proof of concept.

Smart Contract

Client API:

The native contracts should have a unified and “standard” API so they’re all the same. Perhaps it is also possible to define some way for contracts to expose an ABI so it becomes simpler and easier for clients to get the knowledge they need to build transactions and chain contract calls with each other.

Testing Environment:

There is a tool called Zkrunner that takes the zkas circuit and the private inputs, then generates a proof and verify it.

It’s like an interactive environment for zkas circuit developer. Without Zkrunner, the developer needs to manually program, and feed the private and pulibc inputs and drive the verification. It needs some code cleanup and documentation on how to use it.

Passive APR/APY

Consensus participants should be incentivised to stake by getting rewards for participation. We need to find something useful for them to do in order to receive these rewards, and also we have to find a way to ensure liveness throughout the entire epoch. The solution to this should not be something that congests the consensus/transaction bandwidth or increases the blockchain size a lot.

Non-native Smart Contract Deployment

There is a basic smart contract called deployooor that is used as a mechanism to deploy arbitrary smart contracts on the network. We need to evaluate the best way to support this. The WASM needs to be verified for correctness (e.g. look through the binary and find if all symbols are in place) so that at least here we disallow people from writing arbitrary data to the chain.

Transaction Fees

TBD

drk

UX!

We need to handle confirmed and unconfirmed transactions, make things prettier and better to use. When broadcasting transactions, if they pass locally, the wallet should be updated to represent the state change but things should stay unconfirmed. The DAO SQL schema gives a nice way to do this, where there’s a tx_hash, etc. which can be used to evaluate whether the transaction/coins/whatever was confirmed.

We also discussed about having clients handle their own wallets, and not providing a sink through darkfid where there’s a single API for interfacing with the wallet, and having to be closely integrated over JSON-RPC <-> SQLite glue. darkfid will only ever have to manage secrets for the consensus coins that are being staked and won’t have to deal with the entire wallet itself.

darkirc

Write documentation about usage. We need proper tutorials about running on mobile. Both weechat-android setups, and local ones.

A simple Android app can be made drawing inspiration from Orbot where there is a big On/Off button that can start/stop the node on the phone and then the user can use any IRC client they prefer.

tau

TBD

p2p (anon) git

The motivation is to move off of centralised platforms like Github. Additionally, it would ideally have the capability keep contributor information private.

P2P

The P2P library needs a complete test suite in order to more easily be able to make changes to it without introducing regressions. This includes bad network simulations, latency, etc. The P2P stack also needs to be able to heal itself without restarting the application, much in the way like when you unplug an ethernet cable and then plug it back in. Currently when this happens, all the P2P hosts might be dropped from the known hosts db as they’re considered offline/unreachable, so we might want to implement some kind of “quarantine” zone instead of deleting the peers whenever we are unable to connect to them.

In the TLS layer of P2P communication, the client-server certificate logic needs to be reviewed for security and we should define a protocol for this.

zkVM

The zkVM has to implement dynamic self-optimising circuits. The first part and the scaffolding for this is already in place, now we need to come up with an optimisation algorithm that is able to optimally configure the columns used in the circuit based on what the circuit is doing.

All the zkVM opcodes need to be benchmarked for their performance and we need to see how many columns and rows they use so we’re able to properly price them for verification fees.

Documentation

• Create beginner level tutorial to introduce contract development and tools.
• Create a list of outstanding work before mainnet.