The Cardano treasury paper refers to the Dash treasury as an example, but doesn't address any of the issues that can be observed from the history of the Dash DAO. Some statements it makes about Dash are inaccurate and unsourced

English General Discussions
1

I posted about this on [reddit](link removed as I’m only allowed 2) and twitter, and was encouraged to post about it here too.

I was looking forward to reading the Cardano treasury paper, but I was quite disappointed with it. It references Dash and some theoretical issues with its MasterNode voting, which just happen to be addressed by features Cardano will add: private ballots and delegated voting.

However, the paper demonstrates little familiarity with the history of Dash’s treasury and does not address the most obvious problems it has faced.

The Dash treasury DAO transmits funds to the wallets of successful proposals as soon as the vote has concluded, payment is in advance. This is open to abuse, especially while the DAO cannot enter into binding contracts with proposers.

Decentralizing decision-making about which proposals should be funded is just the first step, if the story ends with “successful proposals receive funding” Cardano will have the same problems that Dash has/had. Decentralizing control of escrow, or ensuring that satisfactory progress is being made to justify continued funding, is important and probably quite difficult to do well.

I have written an article about the history of the Dash treasury DAO. There is plenty that other projects with autonomous funding can learn from Dash’s experience.

There are some other issues with the Cardano treasury paper’s treatment of Dash.

iii) The voting rule and the decision to allow only masternodes to vote in the election makes it “unfairly” difficult for proposals that do not have the support of the founder and core team to succeed because a considerably large amount (about 33%) of masternodes are owned/controlled by the founder and/or core team.

I’m surprised to see a statement like this, about a controversial aspect of a competitor project’s history, appearing in an academic paper with no source/reference.

Amongst other drawbacks, only operators of MasterNodes are allowed to propose projects and vote and about 73% of all funded proposals have been proposed by two members of the DASH community.

It is incorrect to say that only MasterNodes are allowed to propose projects, but that contradicts an earlier statement that “The DGS allows regular users on the Dash network to participate in the development process of the Dash cryptocurrency by allowing them submit project proposals” so looks like a typo.

There are a number of issues with the second statement about 73% of all funded proposals being proposed by two members of the Dash community (I assume this refers to eduffield and babygiraffe).

That statement needs a date, the system is live and the proportion changes all the time (it’s been dwindling rapidly over the last year)

Based on an observation period of August 2015-January 2018, these users submitted 109 proposals which received funding, out of a total 321 funded proposals, so 34%. I’m not going to figure out when 73% was true but I’m guessing it’s a long time ago.

Number of funded proposals means very little, as the amount of funding requested differs markedly between proposals. By my calculations, proposals from eduffield and babygiraffe together have received 64% of the total DASH distributed by the treasury. However, much of this DASH was distributed while the currency had a relatively low value. As the $USD price of DASH has increased, the proportion going to proposals from these two users has decreased. Using a simple method of attaching $USD values to proposals based on their time of submission, proposals from eduffield and babygiraffe have received around 15% of treasury funds.

6 Likes
2

We thank the author of the post for analyzing the paper’s short discussion of Dash Governance System (DGS).

As we did not have an aim to provide any detailed analysis of DGS, some given numbers had been taken from our previous DGS report dated Oct, 2016 (the report is here: https://files.zotero.net/18675828634/Dash-Governance-System.pdf, it is present in reference list; the Dash Core Response for our report is here: https://dashpay.atlassian.net/wiki/spaces/OC/pages/96370774/Dash+Governance+System+-+Dash+Core+Team+Response ). We admit that statistics on DGS proposals has changed from time of report publishing, as well as various metrics can be applied. As we continue to work on our paper, we will update our treasury paper on eprint.

We insist on the importance of utilize the knowledge of community experts in the decision-making process via delegation and ballot privacy to the voters to provide the soundness of funding decisions, coercion resistance and other desirable properties of the proposed voting scheme.

With regard to a concern raised in the post, “Decentralizing control of escrow, or ensuring that satisfactory progress is being made to justify continued funding, is important and probably quite difficult to do well”: it may implicitly assume another architecture with enough high abuse risks (our Treasury proposal takes into account DGS experience, as well as other decentralized governance systems). In our proposal, we have lower risks that exist only for one treasury period with limited funding; inappropriate realizations of a proposal supposes downvoting it for the next period. In our opinion, 100% efficiency is unreachable in the real world. For a long-term self-sustainable cryptocurrency platform, allowing probes and small scale mistakes in the evolution process is more efficient than attempts to put constant control on each small action of proposal implementation, with the need of extra verification of control level quality, etc.
We believe in honest majority in the community of successful long term running cryptocurrency and the existence of qualified experts involved in the proposal analysis.

4 Likes
4

Thank you for this response. I replied to your comment on reddit and will paste my response here too.

I see that the paper has been updated on July 4th and the points about Dash which I mentioned here have been removed?

The report on DGS you refer to appears to have a dead link (https://files.zotero.net/18675828634/Dash-Governance-System.pdf). I did not see this report in the reference list and have not read it (yet). If this report had been referenced within the body of the text (as other citations are), and it presents a rationale for the figures used, I probably wouldn’t have commented on that aspect (although I feel it is a little churlish to throw shade on a competitor project in a paper like this).

We insist on the importance of utilize the knowledge of community experts in the decision-making process via delegation and ballot privacy to the voters to provide the soundness of funding decisions, coercion resistance and other desirable properties of the proposed voting scheme. The description of the Cardano treasury system itself is interesting. I am looking forward to seeing this application of liquid democracy in action. In our proposal, we have lower risks that exist only for one treasury period with limited funding; inappropriate realizations of a proposal supposes downvoting it for the next period.

Forgive me if this is in the paper, but how much ADA will this system be distributing per period initially?

I don’t see how Cardano’s treasury periods are different to Dash’s (where multi-month proposals can also be “downvoted” if they are perceived to be progressing poorly). Consequently, I don’t follow how the risks are lower than for Dash, as I am assuming the USD$ value available per period will be much higher than it was in Dash’s infancy.

In our opinion, 100% efficiency is unreachable in the real world.

I agree.

For a long-term self-sustainable cryptocurrency platform, allowing probes and small scale mistakes in the evolution process is more efficient than attempts to put constant control on each small action of proposal implementation, with the need of extra verification of control level quality, etc.

I agree on the need for probes and mistakes, but I’m not sure what your statement about this being more efficient is based on.

My view on payment up-front for winning proposals is that it will make trust in the proposer a major factor in decision-making. This will advantage entities that the community already trusts (IOHK presumably being chief among these).

Proposers who are relatively unknown to the community are unlikely to be successful with big proposals unless some trusted entity can vouch for them. Based on Dash’s history, they will either have to prove themselves by executing smaller proposals or make an arrangement with a trusted entity to provide some sort of escrow service - which may incur a fee, and would give that entity power that they can exercise in an opaque way.

For treasury decision-makers, it seems sub-optimal to me that they would have to consider whether the proposer is likely to follow through on the proposal or deliver something half-baked, or abscond with the funds, alongside the proposal’s merits if it were to be funded and executed well. A lack of recourse if a proposer keeps the funds and does nothing will influence decision-making and, I would guess, the aggregate productivity of funded proposals.

If the treasury paper is to be further iterated, I would suggest that it could be strengthened with an articulation of why payment up-front is preferred (such as you provided in this comment). I think this is a defensible position, but when it’s not mentioned it looks like it hasn’t been considered.

1 Like
5

I didn’t realize this form is pre-moderated, you can reject my first reply as I improved the formatting in the second attempt.

6

I also consider it is very beneficial and incentive to be optimistic and see a fundamental good in every person/humanity (which too often requires the ability to see quite deep through the surface).

However, it is useful to recall that a fundamental good does not eliminate the need for having the best systemic measures and solutions for any possible adverse situation that can happen in the future, that can be imagined. As the world’s everyday events clearly show the humanity is, generally speking, still far away from being able to realize its high internal potential, which has been demonstrated by a very small number of great examples of mankind.

In this regard I would like to refer to the interview of 22 june 2018
(see: http://www.huoxing24.com/newsdetail/2018062223420574778.html)
with Vitalik Buterin who, when called (see »Question 3«) to respond regarding Dan Larimer’s believes that human nature is good and that the main differences bewtween him and Dan seem to lie on the basic assumptions, explained:

»I think my philosophy is that we want base-layer blockchains to work under as wide a range of situations as possible, and it really is difficult to predict what kind of interests and values coin holders will have in the future. (…)
That said, economic incentives make much less sense in environments where we have close relationships with each other and know each other well, and in those cases relying on goodwill can generally work much better.
(…) is a base-layer blockchain platform for the world, so it cannot make assumptions about who is participating in it or who is running the proof of stake validator nodes. (…)”

I am confident that the Cardano team will not rely on the humanity’s fundamental good and will incorporate in the system all required measures, according to different situations, being envisaged to possibly arise in the future.

1 Like
7

I see that the paper has been updated on July 4th and the points about Dash which I mentioned here have been removed?

yes, we currently removed them; we’ll consider them and level of those detals in our paper.

The report on DGS you refer to appears to have a dead link

Prof. Bingsheng Zhang provided a correct link, thanks for pointing out that.

Forgive me if this is in the paper, but how much ADA will this system be distributing per period initially?

We still work on that analysis.

I don’t see how Cardano’s treasury periods are different to Dash’s (where multi-month proposals can also be “downvoted” if they are perceived to be progressing poorly).

We also use different incentive mechanism for participants and expect quite different level of participation in treasury elections (these mechanisms are still being researched and we have not published any final version of them yet).

Proposers who are relatively unknown to the community are unlikely to be successful with big proposals

I agree with that. It’s a property of modern society, and it is not easy to give opposite examples from the world economy. In general, a team/person/company/institution/party/etc. needs to initially get reputation before being trusted for significant support. There is always tradeoff between efficiency and stability. One can get the same service for lower price from less known small company, but the right selection of such a company is difficult and mistakes are costly. Thus, well known teams with good reputation will get big proposal, and there elevators for new teams for their mistakes and probes with smaller proposals.
I can accept specific counterexamples and their limited consequences. But they work in short term periods.
On the approach taken in treasury: let highly qualified professionals to do their job, and the community will decide if it is happy with enough big milestones. Writing reports for a wide audience on selection of specific bilinear pairing transformation in the development process of a complex crypto protocol is really inefficient and gives no solution at all. When it is appropriate, big milestone results should get independent professional analysis and reviews. But professionals should be concentrated on the main result, not on the audience sympathy in the reality show. When one have doubts, she/he should provide a small scale test task or select another professional who proved his competence.

I would suggest that it could be strengthened with an articulation of why payment up-front is preferred

Thanks for recommendation, we’ll consider it.

2 Likes
8

yes, we do our analysis in the model of the rational behavior of honest participants who form honest majority. We’re familiar with the tragedy of the commons :slight_smile:

4 Likes