AGP-106 Discussion: Develop Aragon Chain

This thread is for discussing the Develop Aragon Chain AGP intended to create an interoperable Aragon Chain on Cosmos or Polkadot.

8 Likes

We (ChainSafe) will be releasing the feasibility study shortly, at which point we are looking for community feedback. If you have any questions before then feel free to ask away :slight_smile:

7 Likes

I really like this idea! I do think off-chain offers the opportunity to fix UX issue, slowness,ā€¦

Just to be clear, this would be an alternative layer 1 blockchain (akin to Ethereum mainnet).

:o so how would this work? Would it be a Either/Or situation (the DAO would be deployed on one or the other) or would we store data on either chain according to their nature? IIRC tendermints are very promising and fast but less resistant to majority attacks.

Yes, DAOs would run on either Ethereum or Aragon Chain. We are working on ways for a DAO on one chain to control an Agent instance on the other chain.

IIRC, Tendermint supports up to 1/3 of stake being byzantine

1 Like

Regarding the deliverables, since the feasibility study isnā€™t public itā€™s hard to evaluate what exactly is being funded in this proposal. As-is, I recommend that this proposal be modified to only fund the production of the spec, and then once itā€™s more clear exactly what the Aragon Chain will look like then the community can make an informed decision in a future ANV about whether or not to fund the actual development and launch of the chain.

We are planning to release it today/tomorrow ahead of the deadline! We are just performing some edits :slight_smile:

1 Like

From github: @Igorbarinov
You can use our setup for your own EVM chain on DPOS with POAā€™s stack

1 Like

What are the plans pertaining to privacy on the Aragon Chain? On Ethereum, great progress is being made on this side so I expect that sooner or later it will be possible to ā€œdecentlyā€ anonymise ETH & ERC-20 tokens.

If and when Aragon becomes important enough, one of the major use cases are oppressed organisations that couldnā€™t exist anywhere else, (e.g. a DAO that would manage funding of HK defence against China). If that DAO becomes big enough, there will be sophisticated attempts to find out who are the people managing the DAO and more broadly, who are the ANT token holders.

Thanks!

1 Like

From github: @Igorbarinov
You can use our setup for your own EVM chain on DPOS with POAā€™s stack

First and foremost, the tooling that POA has created has shown a lot of promise (I was at ETHDenver, xDai was a great way to pay for food!). As you can tell, the feasibility study was written between Polkadot and Cosmos, for a few reasons:

  1. Itā€™s not just about the EVM, that is just a bonus since it allows for contracts to be migrated with ease :slightly_smiling_face:

  2. By utilizing one of these two protocols it gives access to a wide variety of other toolsets that are unique to each of them (cosmos-SDK, substrate, etcā€¦ offer a high-level of customizability)

  3. PoS vs DPoS, mainly put - governance mechanism for choosing your validator sets can be less decentralized (what if all the validators end up being from the Association or Aragon One?). Ultimately, its important to mimic the decentralization that Ethereum has, anyone can join and mine, similarly anyone can join and validate, DPoS just doesnā€™t offer that. Cosmos already has a fairly large user base of staking infrastructure (pools, individuals, etcā€¦), and Polkadot from our knowledge will have the same upon launch (Kusama has shown this already).

  4. Ultimately we were given a set of questions by Aragon, and were requested to create an unbiased report between the two networks (since we are very much involved in both). We handed the report over, and they made a decision based on the information provided.

Does this mean we canā€™t use POA at all? Absolutely not, blockscout is EVM compatible, and does the job (we interface with it for our work on Ethereum Classic). We havenā€™t written a full specification for the Aragon Chain yet, ultimately we didnā€™t find out the decision to go with Ethermint until well after we handed off the report. The beautiful part of all this, is that we have a large design space to work with, and a lot of things have yet to be finalized.

3 Likes

Hey, after basically starting the discussion on Twitter, I would like to learn more about the advantages of the tooling you mention.
Does it have big advantages over Ethereum?
The second question would be - why is a not in Cosmos integrated Aragon Chain any different regarding threat to centralization than a sidechain to Ethereum? As far as I have read from @jorgeā€™s comment, the node infrastructure is running separated from Cosmos.

Do you mind linking the tweet in mention?

regarding aragon chain and the ARA token. it would be nice if i ANT holders can stake their ANT behind block producers and then there is a 1-3% inflation of ANT each year which go into the ARA bonding curve and ARA tokens get minted. the block producers get those ARA tokens and they can burn those tokens by exchanging them for ANT. By staking ANT behind block producers you are still able to vote with this ANTs on proposals. Perhaps even possible that the more ANT you stake the more resources you get from the Aragon Chain, so Dapps which build on Aragon Chain need ANT. The appreciation in value of the ANT through the need of holding ANT to use the Aragon Chain will outperform the 1-3 % inflation I think (perhaps not short term but long term). what i donā€™t understand is why do you need ARA token when you can just use ANT inflation?

I really like the proposal, and maybe I am missing something, also I know Jorge mentioned some candidate dates for the Aragon chain release, but in terms of AGP document, I think time should be an important factor in the equation, this proposal does not mention any reasonable roadmap or due dates.

How are holders supposed to assess the utility of this proposal without a single date on it? It could perfectly be 3 months or 3 years as per proposal textā€¦ it should include deadlines to be considered a real deal, In my opinion, in its current state, this proposal seems too much vague for the amount of funds it asks.

This is a thought/question on my own as an ANT holder.

2 Likes

@ottodevs Thank you for your concerns, upon acceptance of the proposal we will commence finalization of the specification, that should be completed within 30 days after the grant is accepted. Commencement of the Aragon Chain would begin thereafter, and we expect initial POCā€™s to be completed early Q1 2020, with the full chain delivered end of Q1 2020. Naturally this assumes that there are no foreseeable changes or major roadblocks from the original vision. During the entire process we will be working closely with Aragon One and relevant parties to ensure that roadmap, timelines and deliverables will be met accordingly. As noted in the grant, we will be invoicing on a bi-weekly basis, the $500k was simply an upper bound that we will not exceed.

3 Likes

Iā€™m generally in favor of this. I think the Q1 2020 is ambitiousā€¦ I hope there isnā€™t too much scope creep and it can actually happen!

Iā€™m probably going to vote yes on this, but I REALLY hope that it wouldnā€™t preclude a sidechain deployment. If it did I would vote against it.

@jorge @luis do you guys oppose a sidechain deployment? Would you facilitate a team if they wanted to make an xDai Deployment happen? Would you oppose an AGP that would make this possible?

Iā€™m sure it would cost much less than 500kā€¦

Sidenote: I do think this is a great dealā€¦ It makes me want to look deeper into what is possible with Cosmosā€¦ building a custom chain from spec for <500k in 3-6 months with a badass team like chainsafe behind itā€¦ WOW! Dapp chains are going to be a thing for sure.

That saidā€¦ iā€™m sure we could get on xDai in <1 month for <50k

3 Likes

Hey @GriffGreen :wave:

Sidenote: I do think this is a great dealā€¦ It makes me want to look deeper into what is possible with Cosmosā€¦ building a custom chain from spec for <500k in 3-6 months with a badass team like chainsafe behind itā€¦ WOW! Dapp chains are going to be a thing for sure.

Just want to clarify, weā€™re not asking for $500k, weā€™re just adding that as an upper-limit incase of scope creep, and unforeseeable roadblocks. At the end of the day we will be invoicing based on actual work and hours completed, and compensation is at the discretion of the Aragon Association (per the grant document).

That saidā€¦ iā€™m sure we could get on xDai in <1 month for <50k

I did make a note above that talks about my personal concerns surrounding DPoS, my main point discussing how decentralized a DPoS chain is.

1 Like

Hey Greg!!! :smiley:

Absolutely, I think this is an amazing opportunity to work with a badass team to get our own chain!

Different use cases need different levels of decentralization. Iā€™m a both/and guy on this oneā€¦ Revolutions should not be made on xDAIā€¦ but small community orgs that want a stable currency and donā€™t want to pay $0.05-$1 every time they vote on a decision would work great on xDAI.

Maybe Chainsafe would be down to do an xDAI deployment as well? :wink:

I just donā€™t want making our own chain to be a blocker from deploying on other sidechainsā€¦ and xDai is not the only one! Flora looks epic as well! Discount-tokens ftw!

3 Likes

I just donā€™t want making our own chain to be a blocker from deploying on other sidechainsā€¦ and xDai is not the only one! Flora 1 looks epic as well! Discount-tokens ftw!

Thereā€™s definitely nothing stopping someone from deploying the whole stack onto xDai. The main difference though comes down to security. Now weā€™ve built our fair share of side chains, some that have strong security guarantees and some that completely disregard security entirely. It all comes down to the use case youā€™re looking to achieve. Something I should have added to the above exert, (I may edit it in although i hope everyone reads the whole thread :laughing:) the lower order of decentralization poses a security issue that the user needs to be aware of.

IMO (now this is just a contrast) - If you have good set of validators that you ā€œtrustā€, then I donā€™t see any major differences than just launching a private PoA (Proof of Authority) network using something like Goerli, and any generic bridge. The security guarantees are practically the same.

Yes, the cost of gas can be annoying, although we have the ability to really set the rules on Ethermint, we have the ability to completely alter the way gas works, but we should be careful due to security issue within the evm.