Delegators rewards should be withdrawable to a separate address

[Derkhersh]

@davekaj

if a fix like this is never implemented - it means we continue to build up the sell pressure for a potential huge sell off. Eventually, many indexers will have to unstake and sell. And it would be likely that this would be around the time that Indexers token lock wallets begin to unlock, which is 1 year from network launch.

Wait, why though? If the rewards enter lockup as they are generated, then they will exit lockup one year later- at the same rate the entered. So the difference is selling now as the rewards trickle in, or selling a year from now at the same rate that they trickled in.

It would seem that the 1 year lockup ensures Indexers keep skin in the game for… 1 year, which from a delegator perspective I’d like to see.

I’m confused, why can’t the fix for Indexer cash flow issues be solved by allowing them to unstake a portion of their original stake (assuming it would not move them into an over delegated position, or below the 100k GRT threshhold)?

[dancube]

@Derkhersh

I’m confused, why can’t the fix for Indexer cash flow issues be solved by allowing them to unstake a portion of their original stake

Because the majority of Indexers on the network have their stake vested in token lock contracts which are not upgradeable and require the entirety of their stake to be returned to it before any surplus can be withdrawn.

I thought you were aware of this? Also, isn’t this off topic for this forum thread?

[Derkhersh]

@dancube

I thought you were aware of this? Also, isn’t this off topic for this forum thread?

Dude I literally quoted someone in this thread.

I have seen this explanation, but I find it somewhat unsatisfactory. If we can make one set of changes to the token lock contracts for rewards, surely we could make a set of changes to the token lock contracts for stake, no?

Look, the implications for what could happen if indexers can withdraw rewards in 28 days instead of 365 is obvious, right? An indexer could shut down operations completely, and take their entire stake and rewards with them in a month, enabling a complete cash out of the rewards indexers received before a single additional subgraph gets migrated to mainnet. If the project gets stalled, or heaven forbid was never serious in the first place, this would be highly advantageous for indexers, as it would let them cash out high on GRT before the broader market notices that the project is not going anywhere.

Keeping rewards locked doesn’t add that much safeguard against this, but it is a strong signal of good faith.

[dancube]

@Derkhersh

I have seen this explanation, but I find it somewhat unsatisfactory. If we can make one set of changes to the token lock contracts for rewards, surely we could make a set of changes to the token lock contracts for stake, no?

I think I might have mentioned this in the other thread, but no changes are being made to any token lock contracts to resolve the issue of indexers being unable to retrieve rewards.

@Derkhersh

An indexer could shut down operations completely, and take their entire stake and rewards with them in a month, enabling a complete cash out of the rewards indexers received before a single additional subgraph gets migrated to mainnet.

And what about their delegators? Where does that leave them?

[Derkhersh]

Okay- I’m not sure what you mean here. Are you saying A) there is a workaround in place to allow indexers to draw on their rewards and therefore there is no change necessary to the token lock contracts, or are you saying B) indexers will not be permitted to actively draw on their rewards?

And then as for where would that leave the delegators? Not in a great position, that is for sure. I think many delegators assumed that locked rewards were intended to mitigate the likelihood of such an event, in the sense that collective action from indexers that could tank the price of GRT on the market would, if nothing else, come at the expense of a reduced value of the rewards indexers would have to wait a year to draw on.

I’m not saying this is likley, but from an optics standpoint this looks very rough to anyone with an active delegation who assumed the locked indexer rewards were a feature and not a bug when they delegated.

[dancube]

@Derkhersh

Are you saying A) there is a workaround in place to allow indexers to draw on their rewards and therefore there is no change necessary to the token lock contracts, or are you saying B) indexers will not be permitted to actively draw on their rewards?

A. You can read about the solution here.

With regards to where that would leave delegators, I was talking specifically about those that were staked with an indexer who chose to take their indexer offline in order to retrieve their rewards.

I really think this thread is being steered way off topic now, can we close this loop of conversation out here? I hope you’ve learned a bit more about how the first GIP is proposing to solve the issue that indexers are currently facing, the network was never intended to be this way for them.

[Derkhersh]

Okay. Sure. But you aren’t engaging on the issue of optics, and okay that’s fine.

I’ve heard that the system wasn’t meant to be designed this way, and that’s well and good, but the trouble is that people put money into the system when it WAS designed this way, and so, intentional or not, there is a fundamental change to the incentive structure.

[Oliver]

Feedback from a delegator’s point of view here to the original topic (i.e. immediate withdrawal of rewards): While it may seem inequitable at first glance, I think it has been laid out pretty well why there is a difference between the indexer and delegator group. Indexers require immediate access to rewards in order to financially support their operations. Delegators generally stake GRT with discretionary income and a long-term investment horizon. We don’t have any operational costs associated with delegating GRTs in any way that would provide equal justification to immediate access. I’m ok to leave it as-is for delegators.

I do agree with some thoughts shared around the ability for partial un-delegation vs. the current all-or-nothing approach. However, that’s for a different thread.

[PurplePoring]

@Oliver

Feedback from a delegator’s point of view here to the original topic (i.e. immediate withdrawal of rewards): While it may seem inequitable at first glance, I think it has been laid out pretty well why there is a difference between the indexer and delegator group. Indexers require immediate access to rewards in order to financially support their operations. Delegators generally stake GRT with discretionary income and a long-term investment horizon. We don’t have any operational costs associated with delegating GRTs in any way that would provide equal justification to immediate access. I’m ok to leave it as-is for delegators.

I have no fundamental disagreement with allowing everyone to withdraw subject to the same conditions (whether immediate or 28 days), but I don’t agree with the “immediate access” justification for indexers alone. There’s no logical reason why indexers can’t plan out their finances for a month based on rewards being subject to a 28 day thawing period. And it’s not a new concept, whether you’re a business owner or an employee. My business’s income is basically monthly. We don’t send out invoices to our clients expecting them to pay us daily. We send out monthly (or even quarterly depending on the contract) invoices and they pay them within a set number of days of receipt. Also, if you work in the U.S. (I know people aren’t necessarily in the US here), you don’t have immediate access to your salary as you earn it every day; most commonly your employer will pay your salary every 2 weeks.

[Oliver]

@PurplePoring

And it’s not a new concept, whether you’re a business owner or an employee. My business’s income is basically monthly. We don’t send out invoices to our clients expecting them to pay us daily. We send out monthly (or even quarterly depending on the contract) invoices and they pay them within a set number of days of receipt. Also, if you work in the U.S. (I know people aren’t necessarily in the US here), you don’t have immediate access to your salary as you earn it every day; most commonly your employer will pay your salary every 2 weeks.

Glad you brought up the analogy to the real world, makes it easier to draw parallels. Using your example, I think the fundamental difference is that delegators and indexers are very different type of stakeholders which are not in the same bucket. Indexers are doing the work and are running the operations of the protocol, hence they are actively engaged in the daily operations who also manage the ongoing costs for all of us. Delegators are not involved at all in that, we are more like shareholders of the protocol. I look at rewards similar to dividend payouts of a stock, which are naturally also not paid out on the same schedule as salaries and invoices, but are rather more spread out. I think it’s important to look at this in a more nuanced context.

[PurplePoring]

@Oliver

Glad you brought up the analogy to the real world, makes it easier to draw parallels. Using your example, I think the fundamental difference is that delegators and indexers are very different type of stakeholders which are not in the same bucket. Indexers are doing the work and are running the operations of the protocol, hence they are actively engaged in the daily operations who also manage the ongoing costs for all of us. Delegators are not involved at all in that, we are more like shareholders of the protocol. I look at rewards similar to dividend payouts of a stock, which are naturally also not paid out on the same schedule as salaries and invoices, but are rather more spread out. I think it’s important to look at this in a more nuanced context.

Right, but my point is that budgeting for periodic income (biweekly, monthly, whatever) is normal out in the real world no matter who you are and what side you’re on, so there’s no reason why indexer businesses here can’t do the same thing. So I don’t see indexer business cost overhead alone as an adequate justification. We agree that indexers and delegators perform different roles in the network, but perhaps simply disagree whether that means indexers should be able to withdraw rewards immediately while delegators are still subject to the 28-day thawing period or whether both indexers and delegators should be subject to the same thawing period (whether none or 28 days or anywhere in between).

[Oliver]

@PurplePoring

Right, but my point is that budgeting for periodic income (biweekly, monthly, whatever) is normal out in the real world no matter who you are and what side you’re on, so there’s no reason why indexer businesses here can’t do the same thing. So I don’t see indexer business cost overhead alone as an adequate justification. We agree that indexers and delegators perform different roles in the network, but perhaps simply disagree whether that means indexers should be able to withdraw rewards immediately while delegators are still subject to the 28-day thawing period or whether both indexers and delegators should be subject to the same thawing period (whether none or 28 days or anywhere in between).

I conceptually don’t disagree, but don’t want to sway off topic too much by introducing a discussion that is debated in a separate thread. Let me just share a couple of additional thoughts on that: There are many smaller indexers out there and the protocol overall is still in the beginning stages. Allowing immediate access to rewards for indexers clearly helps the smaller indexer group much more than the larger ones who have deeper pockets anyway. Second, a lot of information has been shared on the current trend towards stake centralization as well as rising indexer costs driven by increasing gas fees. If immediate access of rewards for indexers helps addressing some of these issues (which was part of the original design anyway), I’m all for it.

Back to the topic of this very thread, I think this proposal should be assessed on its own merit. The goal states that it would attract more new delegators to stake GRT. I personally highly doubt that such a marginal enhancement would result in any significant increase in new delegators. Furthermore, I have not read anywhere that we are currently not attracting new delegators fast enough. The numbers have been steadily rising from the day of launch at a rate of growth that looks very healthy for the network overall. Happy to be corrected if I’m mistaken on that.

Subsequent reasoning in the thread reveals another motivation for this proposal which is based on tit-for-tat thinking. Indexers get it, so as a delegator I want it too. To me, that’s simply not a good enough reason to support this idea. The Indexer proposal has an actual rationale behind it, which seems to be missing in this one. Also, the Indexer proposal doesn’t put a delegator worse off in any way, my situation is not impacted at all. I’m still getting the same benefits and access to them as I did at the time I chose to delegate. I knew what I was getting into and I was happy with it. Hence, I don’t see this proposal solving any problems.

[ferosv]

Thanks all for expressing their voice. But I have a feeling, that majority of people who wrote here are indexers (who don’t have interest to help delegator to swipe away their delegated GRTs ) and second group are delegators, who bought tokens in ICO sale and they are locked (vested), so they also don’t have interest to change anything here.

I’ve undelegated 1/2 of my tokens and we will see, how it will go.

[Derkhersh]

@Oliver

Also, the Indexer proposal doesn’t put a delegator worse off in any way, my situation is not impacted at all. I’m still getting the same benefits and access to them as I did at the time I chose to delegate. I knew what I was getting into and I was happy with it. Hence, I don’t see this proposal solving any problems.

Did none of the indexers know what they were getting into? Now I’ve seen it said that indexer rewards being locked for a year was a “mistake” and a simple oversight. That seems a bit suspect to me from an outside perspective, and it’s troubling to find out this mistake was made only after my delegation was committed. But, as an attempt at good faith I’m willing to accept the explanation. Is going from a 365 day lockup to a 28 day lockup (a 93% reduction) not a pretty fair compromise already?

It’s sort of hard to fathom why an actual delegator would be arguing against their interest. If indexer cool down is 28 days and so is yours, you lose nothing. If indexer cool down is 0 days and so is yours, you lose nothing. If indexer cooldown is 0 days and yours is 28, and this change happens 2 months into your delegation that takes a month to get out of, can you still claim you “knew what you were getting into?” Didn’t what you were getting into fundamentally change a little bit, and not in your interest?

@Oliver

If immediate access of rewards for indexers helps addressing some of these issues (which was part of the original design anyway), I’m all for it

Which of these issues does access to rewards after 28 days not help address? And if there are issues that require day-off access to funds, is it not a fair concession to give delegators the same freedom of withdrawal?

If you keep dodging these basic questions then its gonna take us a very long time to get next to nowhere.

[Oliver]

@ferosv

I have a feeling, that majority of people who wrote here are indexers (who don’t have interest to help delegator to swipe away their delegated GRTs ) and second group are delegators, who bought tokens in ICO sale and they are locked (vested), so they also don’t have interest to change anything here.

I myself have not bought any tokens in ICO, but started buying GRT about a week after it. I have not seen you presenting anything that substantiates the positive impact your proposal would have to the network. All I see is a one liner goal statement that reasons on the grounds of a hunch of intuition. What data can be shown that estimates the additional inflow of new delegators this proposal would deliver? How would that increase be a net positive to the network overall relative to the challenges it creates, as previous posts elaborate on?

A lot of people spent a ton of time providing detailed answers in this chat. As a thread originator, I would recommend coming prepared and take this thread seriously. Dismissing the rest of the community with such unfounded claims is rather disrespectful, especially given the lack of substance you yourself have contributed to the discussion thus far. Quantify the benefits of your proposal to the overall network and you have a chance to win people over to your side.

[Oliver]

@Derkhersh

Did none of the indexers know what they were getting into? Now I’ve seen it said that indexer rewards being locked for a year was a “mistake” and a simple oversight. That seems a bit suspect to me from an outside perspective, and it’s troubling to find out this mistake was made only after my delegation was committed. But, as an attempt at good faith I’m willing to accept the explanation. Is going from a 365 day lockup to a 28 day lockup (a 93% reduction) not a pretty fair compromise already?

It’s sort of hard to fathom why an actual delegator would be arguing against their interest. If indexer cool down is 28 days and so is yours, you lose nothing. If indexer cool down is 0 days and so is yours, you lose nothing. If indexer cooldown is 0 days and yours is 28, and this change happens 2 months into your delegation that takes a month to get out of, can you still claim you “knew what you were getting into?” Didn’t what you were getting into fundamentally change a little bit, and not in your interest?

I am not going to address specific questions raised here since they are off topic relative to the goal statement made in the proposal. I think I have established my view that I don’t believe decentralization equals “exactly the same for everyone”. Different network participants have differing degrees of engagement and risk exposure. I therefore believe that the indexer and delegator proposal should be assessed independently from one another.

Thinking about the proposal from a very personal perspective for a moment, I honestly have to say that it makes zero difference to me whether I’m being given faster access to rewards. Delegating to me is a long term investment decision, where such short term benefits have no relevance. I also travel various telegram chats of GRT holders on a daily basis and I have not seen this subject even brought up once by anyone who delegates, let alone expressing heart burn over it. Hence, I don’t think that this is a pressing issue in the delegator community at large.

Didn’t think I would end up having to elaborate so much to further detail the reasons of my thinking. I feel I have shared enough thoughts on this subject and I will not respond to any further follow up questions.

[juanmardefago]

Did none of the indexers know what they were getting into? Now I’ve seen it said that indexer rewards being locked for a year was a “mistake” and a simple oversight. That seems a bit suspect to me from an outside perspective, and it’s troubling to find out this mistake was made only after my delegation was committed. But, as an attempt at good faith I’m willing to accept the explanation. Is going from a 365 day lockup to a 28 day lockup (a 93% reduction) not a pretty fair compromise already?

You’re confusing terms.
Most indexers are either early investors or testnet participants.
Their original stakes are locked for 1/2 years, if I recall correctly, depending on whether it’s part of the testnet rewards or an early investor’s tokens, this isn’t going to change, tokens distributed in any of those cases will still be locked for the full amount of time that they were supposed to.

What was discussed was how to extract the inflation/indexing rewards, since currently indexers cannot withdraw the rewards from inflation (indexing rewards) without unstaking all of their GRT (and thus, stopping operations on their indexer for 28 days, which doesn’t only impact the indexer, who will miss 28 days of rewards, but also all its delegators, since unstaking is the indexer equivalent to undelegating, and is subject to the 28 day thawing period).

This means that indexers don’t have access to all the money they are generating right now, which was supposed to be used to pay for infrastructure, maintenance, and generally all costs related to indexing. That wasn’t intended, and this is happening because the token lock contract wallets check that the balance in the contract wallet needs to be higher than the initial balance to be able to withdraw, which leads to the situation described above, where you need to unstake fully to withdraw only the surplus, since the rest of the GRT is still under strict lock until the end of the vesting period (1-2 years).

That issue with the token lock contract wallets also affects wallets from the Curator’s Program (albeit curators get quarterly unlocked tokens during a 4 year period, which is a little bit different than the vesting parameters for indexers and early investors), and most people with those rewards are delegators, hence why I said that the change for indexers should also be made to impact curator’s contracts, and both should be subject to the 28 day period.

I’m not up to date with how the current implementation for indexers being tested works, but from what I understood, there were some technical issues that meant that in order to avoid the situation described above, the change would need to bypass the thawing period. It also meant that all rewards generated up to this point won’t be accessible with the fix, so it would only works for newly generated tokens. If anyone has the full details on the implementation, feel free to correct me!

[PurplePoring]

I’m willing to accept at face value what Council and other people have said about indexer rewards not being withdrawable at all being unintended. That’s fine (although I believe there are still unintended consequences that will occur, such as eventual overdelegation if an indexer consistently withdraws their rewards but delegators consistently compound theirs).

There are many smaller indexers out there and the protocol overall is still in the beginning stages. Allowing immediate access to rewards for indexers clearly helps the smaller indexer group much more than the larger ones who have deeper pockets anyway.

Indexers have already been operating for months without withdrawing their rewards. I have sympathy for the proposal of allowing indexers to withdraw their rewards generally, but my sympathy doesn’t extend so far to an indexer who might literally go out of business within 28 days after the change is implemented to let indexers withdraw their rewards…

Back to the topic of this very thread, I think this proposal should be assessed on its own merit.

I don’t think we should isolate proposals like this when they bear some relation to another proposal or to a change already being implemented.

The goal states that it would attract more new delegators to stake GRT. I personally highly doubt that such a marginal enhancement would result in any significant increase in new delegators. Furthermore, I have not read anywhere that we are currently not attracting new delegators fast enough. The numbers have been steadily rising from the day of launch at a rate of growth that looks very healthy for the network overall. Happy to be corrected if I’m mistaken on that.

The problem here is how can you ever research or quantify the number of potential delegators who decided not to delegate? It’s not like we have some mailing list where potential delegators who don’t delegate receive a survey asking why they decided not to delegate.

We’re getting both indexers and delegators, but I don’t know what the numbers truly look like. I believe many of our recent delegations in the past month have been very big, which suggests to me early backers who decided to delegate rather than run an indexer (and, given the numbers, who likely wouldn’t care whether they could withdraw rewards or not–just as some big indexers might not care as much).

Subsequent reasoning in the thread reveals another motivation for this proposal which is based on tit-for-tat thinking. Indexers get it, so as a delegator I want it too. To me, that’s simply not a good enough reason to support this idea. The Indexer proposal has an actual rationale behind it, which seems to be missing in this one. Also, the Indexer proposal doesn’t put a delegator worse off in any way, my situation is not impacted at all. I’m still getting the same benefits and access to them as I did at the time I chose to delegate. I knew what I was getting into and I was happy with it. Hence, I don’t see this proposal solving any problems.

Actually, I believe equity itself is actually quite an important goal, especially for what seems like a relatively harmless proposal such as this one. Inequity creates or further exacerbate existing yet unnecessary conflicts between indexers and delegators. It puts delegators in a place where indexers can shut future delegator proposals and ideas down by simply suggesting, “we disagree, and we’re more important than you, so you should just shut up and sit down.” I already mentioned this dynamic and flawed thinking about the indexer-delegator relationship in another thread–it’s very harmful.

[Derkhersh]

Yeah, I generally accept all of this, and I understand that mistakes happen. What I am suggesting though is just that from a optics standpoint, it looks rough to make fixes after 1.9 billion in delegations have been accepted. Now, sometimes you must do something to right a wrong even if it looks bad, and that’s okay. What I think folks are suggesting is especially because there’s an optics problem here, indexers should take steps to signal good faith to their delegators by ensuring any “fix” has a level of stakeholder equality in mind.

Governance is politics, and in politics you sometimes need to put lipstick on a pig so that your constituents stay happy. This is really all anyone is really driving at, from a delegator perspective.

[juanmardefago]

Yeah, it might look a little bit rough, but to be fair, it’s only been ~2 months since launch, and the problem itself has been discussed for at least a month, plus development, testing, auditing, all those things take time so, all in all, it was actually quite a fast fix

That’s why I proposed a middle ground solution which would allow indexers to get some money flow to pay for stuff right now, but “bind” that to a resolution for the fairness of the original fix (whether adding a thawing period to the fix and/or letting delegators choose to restake in the same way, whichever is best for all).

I still think that unstaking/undelegation should be subject to the 28 day thawing period, while the rewards could potentially be withdrawn with no thawing period (or at least a shorter one), for both delegators and indexers, after the permanent fixes are deployed.