Author Topic: BSIP: DFX Protocol  (Read 3945 times)

0 Members and 1 Guest are viewing this topic.

Offline R

  • Hero Member
  • *****
  • Posts: 993
    • View Profile
How do you plan to proceed?

Had any conversations with committee or core devs regarding this?

Can you commit this BSIP to the bitshares BSIP repo?

Worker proposal before start of work? You'll need to supersede the blocking worker proposals, probably will require some campaigning and more community involvement.

Offline litepresence

How many people do you estimate will be temporarily employed for this worker proposal?

I can see various "Roles"

Project Coordinator
Documentation Specialist
Gamified Oracle Lead Dev
Simplex ACCS Lead Dev
Mint/Burn Lead Dev
Ring Signature Lead Dev
Code Audit, Unit Testing, Quality Assurance
Javascript UI Developer
Pybitshares Developer

As few as 4 individuals, as many as 9 maybe.   It depends on the level of funding, the willingness of developers to work for the funding level authorized.   

I'd like to be the Coordinator.   I'm open to Docs and Pybitshares.   The rest is beyond my payscale.   Obviously we have lots of unused talent; hopefully we can encourage them to return.   The budget in the BSIP was just kicking tires.   For this to become more serious we'd need to consult with individual developers and get a sense of their willingness to get in on the action.     I would like for the contract to be in BTS terms; original crypto gangsta style; where developers believed in the project and were willing to work for a contract that presumed bullish BTS:BTC and BTS:USD action upon delivery.



Also, do you think this DFX minting mechanism might enable NFT minting? Like one way burn NFT mint, maybe randomly from pool of possible results? Or are DFX UIA not planned?

Mint and Burn would be a "primitive", which could be reused.  In the context of DFX Haven Swaps they would be price oracle mediated to ensure value of quantity A = value of quantity B.    Random pool of possible NFT's sounds like a cool alternate use case... but is beyond the scope of my intentions.   I think it could be an interesting feature and could leverage the underlying technology; but require additional BSIP and funding.
« Last Edit: June 04, 2023, 09:04:39 pm by litepresence »

Offline R

  • Hero Member
  • *****
  • Posts: 993
    • View Profile
How many people do you estimate will be temporarily employed for this worker proposal?

Also, do you think this DFX minting mechanism might enable NFT minting? Like one way burn NFT mint, maybe randomly from pool of possible results? Or are DFX UIA not planned?
« Last Edit: June 01, 2023, 01:44:18 pm by R »

Offline litepresence

The scope is quite large, is this BSIP all or nothing? Or could it be broken down into multiple BSIPs?

could be broken down

1) currently the oracle is either "witness cartel" (eg. bitCNY) or "private cartel" (eg. HONEST.USD).   I would think first step is to gamify the oracle so that it could actually be trusted as a decentralized protocol rather than an hierarchical elected clique.

2) then we get into what exactly are we to do with this new legit oracle.   seems there is considerable pressure to keep the bitasset MPA's as they stand... so they do not necessarily benefit from the new oracle.   So step two would be to give that oracle purpose; "mint and burn haven swaps" become the second phase.

3) they're not really "haven" swaps in the absence of stealth.    3rd phase then... we add requirement that swaps occur blind that blind transfers are signed with ring signatures.   

4) finally we can resolve the gateway honey-pot issues by leveraging the new oracle procedures to provide gateways for major cryptos


The "Simplex Oracle Mediated Atomic Cross Chain Swaps" tech sounds good for liquidity, however I'm concerned about haven protocol considering Terra's collapse when using a similar underlying protocol. Do you believe planned attack mitigations would prevent such a scenario? Would Terra have collapsed like it did had such mitigations been in place?

The Terra protocol and the Haven protocol swaps are quite different.   

Haven is very simple.  You have DFX.  DFX has some oracle rate it exchanges for USD... you burn your DFX and get USD at any time at oracle spot rate.   The process takes some time and staking to mitigate the possibility arbitrage attacks.   

Terra on the other hand has a "Supply and Demand Dynamics" algorithm: When the demand for UST increases, the protocol automatically mints new UST tokens. Conversely, when the demand decreases, the protocol burns UST tokens to reduce supply. This algorithmic adjustment mechanism helps stabilize the value of UST.  Terra Protocol also employs a mechanism called "Seigniorage Shares" to manage the stability of its stablecoins. When the protocol mints new UST tokens to meet increasing demand, these newly minted tokens are distributed to holders of Terra's governance token, LUNA. This distribution is intended to incentivize LUNA holders to stake their tokens and participate in the stability of the Terra ecosystem.   So the exploit vectors of the two protocols are not the same.




How much more complex does making everything stealth ring signature based make it? Would initially forgoing this stealth tech reduce complexity and cost?

By far the consensus mechanism behind a secure decentralized oracle is the most complex, unique, and dev hours taxing element of the design.   

Ring signatures, while complex in implementation... the path has largely been forged on other chains and the "blind balances" already exist.    So while it requires high level technical expertise... the lines of code involved should be much more limited.

Tokenization contracts for both Haven Swaps and ACCS swaps should be far less challenging both technically and in dev hour terms. 

Approx 30m BTS for this work is quite a lot, breaking the BSIP down to the most profitable bits for less might make it more feasible?

Sure that could be one path.   IMO... with value of the chain falling below $20M... 30M BTS probably will no longer cover what developers will ask to be involved.   So... I'd expect "either you're doing it or your not" that is... we view DFX Protocol as a Hail Mary to save a dying institution... or we let it bleed... Part of what creates value for "the developers" is that getting paid in tokens that have developer based "novel crypto magic" behind them makes them worth developing for.   That's the OG way things were done and that's the way I'd like to see this unfold.   That is... the whole package makes each BTS the dev gets paid worth more.   In cash terms... I think the build approaches $1M.    In crypto terms... I think it can get paid out closer to $300k in embued "magic tokens" that hold new and novel dev.

Is DFX only distributed to oracle data providers? Or is there a distribution planned?

There is mention in the paper of an initial "federation" of approved providers and there may be necessity to give away some "dust" to kick off the mechanics of the system.   that dust would go to those approved data providers; bitshares historical github contributors.

Offline R

  • Hero Member
  • *****
  • Posts: 993
    • View Profile
The scope is quite large, is this BSIP all or nothing? Or could it be broken down into multiple BSIPs?

The "Simplex Oracle Mediated Atomic Cross Chain Swaps" tech sounds good for liquidity, however I'm concerned about haven protocol considering Terra's collapse when using a similar underlying protocol. Do you believe planned attack mitigations would prevent such a scenario? Would Terra have collapsed like it did had such mitigations been in place?

How much more complex does making everything stealth ring signature based make it? Would initially forgoing this stealth tech reduce complexity and cost? Approx 30m BTS for this work is quite a lot, breaking the BSIP down to the most profitable bits for less might make it more feasible?

Is DFX only distributed to oracle data providers? Or is there a distribution planned?

Offline litepresence

Gamified Secure Oracle
Mint and Burn Haven Smartcoins
True Stealth Balances and Transfers
Simplex Oracle Mediated ACCS

https://github.com/litepresence/Bitshares-DFX-Protocol/blob/main/DFX_Protocol.md