A thoroughly articulated, yet potentially misinformed analysis of The DAO which may be worth discussing

[Jupiter00000]

It was suggested that I post this here for discussion.

My Reddit post.

The Blog post that is referenced.

[rileyjt]

Its a very interesting post. I think we could learn a lot from Dan.

Based on his concerns for instance, we need to prep a liquid democracy solution as possibly a first priority, and also be prepared to lower the quorum numbers.

A major question I think would be if the exchanges that own large portions of tokens intend to take part in the voting. We pretty much either need them to, or have a way to exclude their tokens from the quorum calculations.

I may respond more in depth at some point, but a few thoughts:

1. If the fund is not able to come to a consensus and people are not interested in participating, they will split out and cash out, eventually leaving us with a group that is serious about making investment decisions even if it means we have reduced funds to work with. Thats fine (we honestly probably have too much money at this point anyway).
2. The technology involved does matter. The Ethereum network is not the Bitshare network. It already has a lot more activity and investment taking place even without us - and it is not dependent on us.
3. The best thing about Ethereum is that it encourages companies to build and monetize on/in the actual blockchain. Not services sitting on top of or beside of, or on forks of. This should allow us to spend ETH while providing enough value to the network as a whole to offset the expenditures both long and short term. This is the entire reason I am invested here rather than elsewhere, so if that turns out to be wrong, it won't be because of the DAO.
4. If all we do is fund a few major projects, act super conservatively and stake Ether while waiting to see what happens, I am fine with that too - these tokens were my ETH long holdings so I don't mind hodling.

[Teleos]

nice analysis....

i would add that $6M is pretty low and being underfunded is difficult.

[shakedog]

Definitely.

[PH88]

I believe the writer got it wrong on the split mechanism of The DAO. It'd be best if slock.it, the author of the DAO, can come out and clarify. But from what I got from the white paper, you can always split and withdrawal your share of the current ETH balance in DAO whether you've voted before or not. And after you split, you still entitle your share of rewards from proposals approved before the split (but not after). Fair enough to me.

[Chip2]

Unfortunately, the author is right about the inability to split. When you vote "no" you can no longer split.

[PH88]

I believe it's only partially true. i.e. once you vote (yes or no), you cannot split until the conclusion of that particular voting. But after the conclusion of that voting, you can still split.

[Chip2]

But you will still have to invest into the proposal, if it passes. So you can only split after the proposal that you voted "no" is accepted or rejected.

[PH88]

Yeah. But the author of the blog post put it like non-voters have much advantages over voters, which I believe is misleading and incorrect. Basically, for each voting each token holders have 4 options:

1. Vote yes (i.e. support the proposal)
2. Vote no (i.e. not support the proposal, but want to keep the right to invest in future proposals)
3. no action (i.e. follow the crowd)
4. split (i.e. not support this or any future proposals)

Option 1-3, no matter you vote or not, you invest your share of ETH and get your fare share of reward tokens in return when the proposal passes.

Option 4, again no matter you have voted before or not, you exit with your share of remaining uninvested ETH and the reward tokens you earned so far up till the splitting.

This is pretty fair to me.

[islander777]

but there is no option to split if voted NO and the proposal passed.
You can split after the proposal passed, but your funds will already be invested in the proposal, which is not fair if you didnt support in a first place.

[PH88]

Well, if I vote for donald trump but turns out hillary clinton win, I wouldn't feel unfair because it's the rule of the game.
In The DAO it's even better because you have the 'split' option to exit beforehand and no one can force you to invest

[islander777]

ok, a situation:
(lets forget about curators for a min. I want to show a point)
I publicly anounce that Im 51% token holder (can be true or not, noone knows). I submit a proposal to invest all DAO funds into my company

the crowd can
1. Vote yes (=suicide)
2. Vote no (what if I truly am a 51%holder. voting no won't help)
3. no action (=suicide)
4. split (the only option which would save people's funds, but it will seriously hurt/destroy dao, as no ether will be left in it.)

now, if there was an option:
5. Vote No, but split if proposal passes.
it would have fixed the situation completely. People would just vote no with complete peace of mind, and if the proposal passes - they split, or if I am not truly a 51% holder - proposal won't pass, dao is safe, crowd is calm, all is good. Availability of this option will make the whole dao safer. Other options dont offer this.

[PH88]

See your point. I will support if you submit a proposal to add this option 5 One of the nice things about DAO is the ability to change itself and evolve.

[islander777]

Well, if I vote for donald trump but turns out hillary clinton win, I wouldn't feel unfair because it's the rule of the game.

very good comparisson. very unfair situation, I'll explain why. Lets develop this idea further. You have elections coming. Hitler (real one) is a major candidate
you have an option:
1 vote yes Hitler
2 vote no Hitler (but you dont know how many idiots will support him)
3 no action (pretty stupid in this situation)
4. leave the country before the elections (but you dont really want to if you dont have to, you love your country and dont want to leave family)

the system doesnt give you an option to leave the country after elections, if Hitler elected and I honestly dont understand why it has to be this way, when the solution to fix is very simple

[islander777]

See your point. I will support if you submit a proposal to add this option 5 One of the nice things about DAO is the ability to change itself and evolve.

Its already been proposed here, but not many supported. Many people overlooked this - while I believe its a major issue

[vprf]

Wake me up when DAO v3 arrives. By that time all the kinks should be worked out. ZZZzzzZZZ.

[rileyjt]

Here is another (worse) situation. You are a 51% token holder, but you keep it secret. You create a proposal to invest all the DAO funds into your company.

You postpone your vote until the last minute. The crowd inevitably votes No or doesn't bother since 99% of of the votes are No. In the last minute, you put 51% of the tokens to a Yes vote and it passes. No one has the option to split.

We are reliant on the curator to save us from that kind of attack - so please tell me we didn't previously elect the 51% token owner as the curator.

[ranford]

I also agree we can learn a lot from Dan, of course I dont agree with his conclusions, but from 1 of his own responses, I think he makes some good points that we should take as input for improving TheDao:

From dan::

1. only long-term committed users can vote (VESTS).
2. it should be more profitable to be committed than to stay liquid.
3. funding work should be easy with a small minority.
4. spending should be mandatory, the only decision is what to spend on.
5. millions of people should be able to receive funding, not just a few high-profile uses
6. funding should be done “after the fact” not in advance

For me 1, is very important, I believe that just owning a token for any temporary time is insufficient we need to distinguish longterm interest from speculators, i strongly urge us to consider some form of VEST feature being adopted.

point 2 I think can be addressed by 1

I think point 3 is actually pretty much not an issue right now as we have a few large stakeholders making up the top 50% of tokens

on point 4 I Definitely agree, the funds are there for the explicit purpose of creating value by being invested in projects. Sure we may have reserves for certain purposes, but the main rump of capital needs to be used. The precise mechanism im not sure, but we do not want funds to be sat on for too long.

on point 5. I dont yet see this as an issue, as long as the eventual fee to propose and the filtering systems are reasonable

on point 6. This one is already addressed by slockits example where funds are to be dispersed at staged milestones per deliverables, that is fine.

For me all good points by Dan, and especially points 1 and 4, which I hope we can address as early as possible. Lets not worry about Dans perspective, hes an extremely smart thought leader who we should definitely learn from, however he is probably quite frustrated that after all his own great efforts its TheDao that is getting all the huge attention, id say a fair frustration, but that it may have influenced his general position on TheDao.

Finally for me, there are several key proposals that should get funding as early as possible

1. of course slockit in some form
2. a proposal that solves the crowd attention and decision making issues, id like to see blockfeeds proposal as a start
3. a proposal to support us from legal aspects, and im keen to see the proposal from "DAO.LINK"
4. if it needs a proposal, something that can address the points 1 and 4 from Dan , something to differentiate between vested members from speculators and something that prevents funds being sat on indefinitely

[Jupiter00000]

@rileyjt @islander777 @PH88

Alternatively, rather than voting No and locking in your tokens, why not just not vote and then one of these will result:

a) it's voted No anyway, thus doesn't go through;
b) it doesn't meet quorum, thus doesn't go through; or,
c) it goes through, but you didn't vote so your tokens aren't locked and you can just split.

Wouldnt an attacker have to be a curator? Doesn't a proposal require 60 percent to vote? I believe that would drive the price up high enough to make it profitable to trade out rather than split.

[m.patrick.hidalgo]

Does anyone know what happens if:

• We have a hypothetical $100 in the DAO
• We have 50 proposals each asking for $5
• They all propose at the same time
• All of the proposals pass

Since we only have $100 to spend and $250 in proposals that have passed, do we use a first in, first out basis?