Fri 30 Mar 2018 9:48PM

Milestone payout insurance - further discussion before next gov meeting

KI Kris is Public Seen by 322

During the governance meeting we discussed a 'merge' of two related proposals on paying out people in time.
A decision was tabled as we need more input. Below you can find the 2 proposals (plz also review comments) plus the notes from our governance meeting. Please take the time to read this through and feel free to comment below, so we can come to a decision during next week's meeting.

RJ - Likes Satya’s proposal but this isn’t a quick answer. For the time being we need to just do payouts on time. This should extend to everything with a Fiat budget. MultiSends are a lot of work, monthly makes sense. Having an incentive layer will ease this tension. Use a stable coin.

Edu - No reaction; is distracted

Yalor - I like the idea of the payouts being closer AFAIK. Agrees the insurance policy will take longer.

Emex - Likes vojtechs because it’s easier to implement. Might have to abandon the MultiSend unless we can fix it.

Q: Use a stable coin, multisend with monthly payouts sounds good. Smart contract insurence is novel but likely too technical.

Kris - Bi-weekly or monthly would be great. We should hire somebody just for finance or budgeting. Not okay with using a stable coin. Would prefer to get paid in Ether. Doesn’t like the monthly rate, having the rate at the moment the milestone goes out protects people from volatility. Would prefer not to change the DApp. Agree with the idea of making an additional bonus milestone if we’re not getting paid in time.

Kay - Good feedback so far. The most imporant thing is to get payouts out quickly. Make a milestone, get paid out quickly. As long as we have this manual process it’s hard. Averaging out volatility is good for regular rewards.

Griff - Budget started on the 15th of Dec, he always offered to front any ether to anyone, this was the first proposal to fix the issue. We’ve been delayed due to feature creep since the DApp sucks and we have to deal with that. The issue is due to ether crashing and milestones not coming as fast as they need to. This wouldn’t be an issue if we didn’t move to Rinkeby. The actual price right now is cheap, I would think satya would want to make these milestones. There is a centralization issue on the multisend. Need to get more people to learn how to do the multi-send. There is a risk in V’s idea, if we have a dip on the one day we pay out, we get fucked, as Kris mentioned.

Lindsay - Handing it off

BowenSanders - Agrees with Kris and Griff

Amend and Clarify (Satya)
Most people seemed to be in favor of Vojtech’s solution. Not sure what to say. A stable coin is too ambitious(?). Let’s try to implement Vojtech’s solution as soon as possible but there may be an issue with sample size, we don’t want milestones on milestones. Let’s do another Loomio and do another discussion.


Griff Green Sat 31 Mar 2018 6:02PM

Yeah! So what do you think about going the social coding route and making this a project. To build a volatility insurance DApp... honestly it would surprise me if it hasnt already been done, I know Satya was looking at oracles...

Until then... I am always happy to front ether to anyone with a milestone up!


Vojtěch Šimetka Mon 2 Apr 2018 6:25AM

I would like to see a commitment from our side to some payout dates as a first step. Could we say the we pay every 2 weeks on wednesday and if not there will be a penalty which we invlude in the payout? I relly like Satya’s idea and would love to see it happen in as well as soon as possible.


Griff Green Mon 2 Apr 2018 6:43AM

I would love that. We have a slow step in the fact that no one knows how to run the thing that scrapes the milestones... instead everyone wants to automate it more... (SCOPE CREEP) it just needs to work asap... in whatever ghetto way that is, and more people need to be empowered to be able to initiate it.
If we want to add more features... cool, but let's do that after, 3 people know how to run it and the repo is open and in giveth :-D @quazia Thoughts?


Quazia Mon 2 Apr 2018 2:15PM

I would rather get the DApp working to the point where we have the option to do payouts in a stable coin (which of course could immediately be converted to ether for people who want that volatility) than invest labor in trying to build some insurance contract. MakerDAO already works as a futures market designed to ensure volatility, wouldn't we just be re-inventing MakerDAO in a way... Also, the milestone CLI tool has been open source for a long time it's just not under Giveth as it's not up to standards and isn't something we'll maintain long term. I've included a link to the repo. Should be fairly easy to get running if you have node and stuff loaded but it's not working the way I want it to. Still this is what I've been using to get the payments out. I think making the changes me and RJ talked about actually refine the scope not increase it. It's not a matter of adding new features just changing the way they work currently to require less work and go out more regularly. In the meantime, I would feel comfortable committing to a specific payout schedule and can make sure that unless there's another bug (which I doubt) the multi-send component of the payments will happen regularly.



Kris is Tue 3 Apr 2018 6:40PM

ok, so as I hear it there is no multisend issue any longer, and if there still is, quazia commits above to be available for the scraping. To me the issue does not lie here, the issue is that someone in the governance team needs to take the responsability to get the payments out every two weeks. This means someone who rrrrreally dedicates time to do get the triple check done (eg do one him/herself and delegate the others + chase those people), then chase quazia, then chase the multisig signers. Last time it took THREE weeks for the triple check to get done. I still see no need for a stablecoin, I would like to keep our dapp as simple as possible. If people are paid out every two weeks there is not a lot of risk of volatility. The Governance team just needs to assure that this actually gets done, and plz let's do this ASAP, looking fwd to the next triple check doc (why is it not there yet?)


RJ Tue 3 Apr 2018 7:57PM

I 100% disagree that there isn't much of a risk of volatility w/ payouts every 2 weeks. Things can drop 30%-50% in 1 day, how is that not much volatility?

While I think we can improve payout speed, there will always be a risk of volatility w/ eth. The best case scenario for payouts I see being 2-3 days (from creation to receiving payment) for at least the next 6-12 months, which is too much volatility risk for me. Plans have always been to integrate a stable coin and it will happen soon after or as part of the bridging solution.


Kris is Tue 3 Apr 2018 9:43PM

with little risk I mean it is only a risk for that week's payment, which I think is a risk we can/should take, it's part of working in this community. volatility also happens in the opposite direction. in any case, had no idea it has always been the idea to integrate a stable coin, haven't seen that mentioned since october (when I started). Personally in any case a fan of keeping it all in eth (as we are giveth) and building on the ethereum platform.


Griff Green Wed 4 Apr 2018 9:03AM

Yeah i mean i guess thats just been implied by "multitoken functionality" which is inclusive of DAI and ERC20 Tether
When thats an option and technically feasible, then everyone can ask for it :-D... I for one will stay with ETH tho ;-)


Satya Thu 5 Apr 2018 7:28AM

Kriss, if you want to speculate with your payment, then by all means do so. You can keep your payment in Ether or buy more Ether with your fiat once you've received it. Your choice.

But what you want, and what we do at Giveth, is forcing people who work full time on Giveth to speculate with their income. I 100% disagree that that is part of working in this (which?) community. It's not part of it, it is a decision currently enforced by Giveth. And it violates budget agreements.

I did not agree on speculating. I want us all to build a world changing platform and not worry about our income. It is distracting, we waste time, we will loose people and we don't reach our goals. Just look at this madness in Loomio for such a simple problem.

So please stop making decisions for other people.

Weekly, bi-weekly, monthly payouts are all bad solutions if they don't compensate for volatility.

And regarding that, I'm not asking for more Ether as you suggest in your comments. What you don't seem to understand is that Ether is worthless, until you exchange it for fiat. As such the budget is in fiat, and Giveth didn't pay out according to budget.

I'm asking for Giveth to pay according to the budget we agreed on. And that means compensating. If that is solved by paying out more Ether, gold or EUR, I couldn't give a shit really. But if we don't solve this, what are the agreements I make with Giveth worth?

If Giveth can't afford spending that amount in fiat, or worse, it doesn't want to spend that amount because it would reduce our precious Ether stash - as you suggest - then Giveth shouldn't hire and reward such a huge amount of people.

This is called business, and what it takes to run a company. There's a runway, there are agreements, there's a burnrate, and there's output. If they are not in balance, one or more of these will loose.


Kris is Thu 5 Apr 2018 11:20AM

Hey Satya, this is not a business, this is not a company. This is us building the ground layers of a new ecosystem (which is bigger than the DApp). And yes, to me this is a Community, even a strong one, and the DApp team is invited to join us.

I cannot make decisions for other people, none of us can, only thing I'm looking into is how we can keep on moving forward and avoid - indeed - all this loomio & gov craziness.

It makes me sad that when we are building a system that should take money out of the equation (by putting the admin transparently into the dapp and bring the talk back to the actual action), that the only thing that's been happening is that the people building that DApp only talk ... about money.

I live full-time off of what Giveth pays me, which is a fraction of what the devs get (or what i earned before) and also work full-time for Giveth (and more). I'm not in here for the money, I'd prefer to work on solutions and hope you want to do so too. Hence the new proposal griff wrote tonight, this came after a call we had, I keep on looking for constructive practical solutions that keep everyone happy.

Volatility is part of this space right now and it will be for a long while. This is a take or leave as far as I understand it, and you are 100% free to make that decision, nobody is making it for you.


Kris is Tue 3 Apr 2018 6:43PM

(We're currently two rewardDAO's behind if I see it correctly)


Griff Green Wed 4 Apr 2018 9:04AM

I thought some of the last reward dao was paid out.


Justjen Tue 3 Apr 2018 6:46PM

Hello guys, i'm just trying to make my way true all the discussions....I think weekly payouts are fair and we should definitely (in terms of being trustable, accountable and reliable as an organization) pay our people in time...I am also a fan of keeping the Dapp as simple as possible, for us and others...If there is a dedicated person needed to do payouts or controls milestones, triple check, chasing down people etc, i am happy to do so( for a little ETH).....


Yalda Mousavinia Tue 3 Apr 2018 10:02PM

Near term solution:
1) Determine who in the org wants their regular reward paid in DAI. Maintain a spreadsheet where only regular reward folks have write access
2) Based on this, determine what the monthly DAI budget is
3) [x days] before the target payout date, use oasisdex.com to trade ETH for DAI in the amount that is needed for the DAI payouts
4) Check the spreadsheet the next cycle to see if anyone changed their DAI preferences. It’s up to the individual to update the spreadsheet if their preference changes.

With this method, monthly payouts would be easier to maintain as the DAI exchange just happens once a month. The amount of DAI per person should be the same as if the Euro salary was converted to USD.


Grace Wed 4 Apr 2018 6:14AM

It seems to me that we have forgotten who we are building this app for. It is very oriented to solve the Giveth issues without realizing if these issues are actually the same for our intended user. Payout are extremely important and so is ‘stability’. When you start looking are the options from our intended users perspective, you will start solving the problems differently.
With that said: stability is important and so is reliability of the pay outs. Frequency of the payout may not be an issue the majorly of users will encounter as they are truly milestone based. A good and efficient review system and a accurate disbursement of the funds - yes very important. Think of your small decentralized altruistic community helping veterans who suffer from ptsd...what do they need to make it work? Whatever Giveth DAC needs, is not the answer.


Griff Green Wed 4 Apr 2018 10:10AM

Yep, what we need is the multitoken functionality in the DApp for DAI and the ability to have a check on the creation of the milestones (which happens when we can donate to milestones and campaigns directly). We are mitigating the issue that we created this DApp for mainnet and then had to move it to test net...
RJ is ATTACKING a ghetto bridge solution and Adria is starting to look over the newest version of the code.

The original plan was that I front ETH to whoever needs it when they have a good milestone... I think that is still valid till we finish the real solution, and we get the payment creation centralization out of quazia's hands


RJ Wed 4 Apr 2018 3:42PM

another proposal for the current situation: https://www.loomio.org/p/9n7U41bV/payouts
comments before the gov meeting are appreciated


Griff Green Thu 5 Apr 2018 1:25PM

Our agreements have always been in ETH, I would like to see somewhere where that wasn't the case... We haven't received any DAI donations... only ETH donations. Giveth is not incorporated in some country, it lives on the Ethereum Blockchain. We have implemented fiat numbers to help hedge against volatility, but only ETH numbers have ever been guaranteed and it is up to each contributor to hedge their position, just as a foreign company working in Mexico doing long term deals has to hedge against the volatility of the Mexican Peso. This is of course more dramatic, but that is the state of our DApp.

The simple solution, which i have done with several people is similar to what Yalda suggested and what we have been doing for ever. If you don't want to speculate on ETH, you ask someone to front you the ETH for the milestone, and I am always up to do it, the only thing i ask is for the milestone to be done and to look nice.

A better solution is that we build a milestone plug in to fit your desired out come... thats the point of dogfooding. Most of the push here is coming from the DApp team. If our DApp isnt working the way you want... then make it work the way you want. Let's take this problem and be constructive with it, not destructive. In the mean time take me up on my offer to avoid having to speculate with ETH.

Constructive Solution: Make a milestone plugin that works the way we need it to work. the reviewer can set the exchange rate date with a couple of small tweeks to the code base i'm sure, then we just need the dapp to be able to be aware of which milestone plugin is being used.


Satya Thu 5 Apr 2018 1:44PM

"Our agreements have always been in ETH, I would like to see somewhere where that wasn't the case..."

Like here?


[deactivated account] Tue 8 May 2018 3:27PM

Why don't we just started requesting payments in ETH then and stop use a Fiat calculator ?
If someone on the DApp Team could add ETH to ETH functionality then we could say "I request 2.5 ETH for the work of _______ week" If one ETH is one ETH in Giveth then why can't we just scrap the whole conversion speculation situation for Giveth internal payments ?


Satya Thu 10 May 2018 6:54AM

That's already been added during the Barca meetup. But I see that it is broken in the dapp


[deactivated account] Thu 10 May 2018 2:42PM

Maybe we can fix ;)


Griff Green Wed 9 May 2018 10:59PM