V6 TRAC Tokenomics Discussion - Node Runner Perspective

[Valcyclovir]

The purpose of this discussion is to address the current issues facing V6 node runners. This is by no means a confrontation and any non constructive criticism will be promptly removed. We recognize and appreciate the hard work and dedication of the team, and the reason you are here to discuss is because you care about the future of the project. It is important to keep the discussion respectful and focused on constructive feedback.

Let's start off with some context.

Node runners have played a crucial role in supporting the OriginTrail ecosystem over the past few years, even when the network numbers were not as high as they had hoped. Node runners continued to run nodes at a deficit to secure the system, support it, encourage it. They were told that running nodes when TRAC was low would allow then to accumulate more TRAC and that the price would increase with adoption, but this did not happen, yet. Instead, the TRAC price went down and payouts decreased by 76%. The V5 bottlenecks also negatively affected the network numbers.

Despite these challenges, the node runners remained supportive of the ecosystem and looked forward to the liftoff event, which promised a shift from supply chains to a DKG, through running our own StarFleet Chain (SFC). We quickly amassed a groundbreaking 100M TRAC in the SFC staking contract, ready for liftoff. However, the launch was unfortunately delayed by over a year, which was difficult for the node runners who were already facing challenging market conditions and a 90% drop in the TRAC price.

At last, V6 has officially launched. Or did it?

That was the general reception of the community - optimistic confusion. It is unclear whether it has lived up to the expectations and hopes of the node runners. The community understands the difficulties the team has faced over the past year, but we cannot ignore the key issues that were missing despite the V6 launch. Many of us were left feeling confused and undervalued by the launch event, which seemed to be more focused on celebrating the team's success rather than providing specific details or timeframes to the community. The decision to run 30 nodes and publish 20k assets into the DKG without involving any community nodes was particularly unpopular, and the lack of announcements or visibility made it even more confusing. There were no announcements made on the same day that instructions were available on the docs to run a node. Unless a GitHub user followed the dkg-docs page, there were absolutely no ways for the community to learn that the instructions were indeed updated. Additionally, there were a number of questions regarding profitability, staking, Houston, Magnify, TRAC tokenomics, OTP collators, Ledger support, and publish numbers that have yet to be fully addressed.

It has come to our attention that the average V6 node runner is currently making just 0.35 TRAC per day, or approximately $1.78 per month at current prices. This is a major concern, as it is not sustainable for node runners to operate at such low earnings. We are facing a chicken-and-egg dilemma, where the team needs node runners to increase their stake and drive up the price, but node runners want to see higher earnings before committing. The node community has previously warned the team about the potential for "race-to-the-bottom" tokenomics, and it appears that this is what we are currently experiencing. Node runners are even testing ask prices 0.00005 above the current ask of 0.00375, and are not winning any commits, which further confirms this. To make the current tokenomics work, node runners may be forced to form a union in order to manipulate the price, but this is not a desirable outcome for anyone. We are at a crossroads, and it is clear that the current situation is not satisfactory for node runners which is reflected by the amount of V6 nodes ran by the community. If we take out the 12 nodes that myself and 3 other members who share the same sentiments run, we are down to less than 5 from the community. A significant number of potential node runners are waiting on the sidelines in the hope of seeing improvements.

It is now time for the community, node runners and the team to work together as one and take action to address these issues.

The floor is open for discussion.

From BRX/Valcyclovir,
OriginTrail community admin, OriginTrail Club admin, Deep dive author and V6 node runner

[calr0x]

I have the capability to run multiple nodes but have chosen not to until several issues are resolved, one of them being how profitable nodes are.

I wasn't happy to see OT silently reduce node rewards in 2022 and the ONLY reason several of us remained profitable was due to multinodes (running multiple nodes on one VPS).

Guys, OT and noderunners are a symbiotic relationship. We need each other for this to succeed.

I would like to propose forming a group that represents node runners in the community that can meet and work with OT to help solve node runner issues beyond technical support such as tokenomics etc.

Thanks!

[Valcyclovir]

@calr0x Thanks for your feedback, I believe this is the intention of OriginTrail Club and you are part of it ;D

[mucke12]

Good morning Tracers!

I agree with everything that was said so far. It's not sustainable at all to run a node. We also can't life from promises anymore ( we already did to long in the past years).

So lets work TOGETHER on a solution and trace on!!

[drMurlly]

I also have the capability to run multiple nodes but have chosen not to,
because of the open questions that Valcyclovir described beautifully.
I am waiting to run the v6 nodes; until this, several issues are resolved, one of them being how profitable nodes are, how to retire the v6 nodes properly, and how long it would take to payout collateral without being slashed.
Thanks!

[hottogo]

The community understands that adoption will build over time. That is not an issue.

The issue is that it is currently unprofitable (and in the past not worth it) to run nodes. This makes the project a failure if this is not rectified, as to have a decentralized knowledge graph you need decentralized nodes.

The solution is not for just raise our ask price, mainly because it may ruin future onboarding of DC's and impact adoption growth. The price to upload to the ODN needs to be a fair value. The missing part of the equation is volume of publishes.

While we wait for adoption to build over the months and years, there needs to be incentive for individuals to run nodes to promote decentralization. Other projects/ecoystems do this in the form of inflation of the token, paying tokens to node runners and inflating the total supply. We do not have this option because Trac is a fixed supply. You could use OTP but the community don't see much value in this token and would only be incentivised by payment in Trac.

Therefore my suggestion is that a small part of the dev funds be used to incentivise nodes. The simplest implementation of this would be to pay a top up amount of Trac per month to node runners to make running a node viable. My suggested value is $1,000 per year (say $83 per month) per node, which at current pricing would mean a payment of approximately 6,000 Trac per year (500 Trac per month).

This could be capped at 200 nodes maximum, being a total outflow of 1.2m Trac per year. If over 200 nodes are spun up, the same total outflow could be split across the total nodes. So if there are 300 nodes, they would get 4,000 Trac each instead of 6,000. Anyone staking to make up the 50,000 Trac balance would get their share of this payment if technically possible. Having too many nodes trying to join in would be a great problem to have.

I propose this be implemented for a 12 month period and the total payment to node runners be re-assessed at that point to see if publishings have grown significantly to the point of removing it or at least reducing it significantly. It could also phase out earlier if the growth of publishings comes in strong earlier.

This initiative would strengthen the ecosystem and have a lot more people incentivised to run nodes, build scrips and interact with the protocol. 1.2M Trac seems a small price to pay per year until we hit our growth targets, given the ecosystem dev fund sits at approximately 139M Trac if I am not mistaken.

[hottogo]

Actually I should add that the easiest way to implement this is for the team to just start posting their own publishes using dev Trac. That way the ecosystem can react as it normally would, with nodes, staking etc.

We just need some more publishes pushed through to supplement the ecosystem until it reaches some level of adoption.

[Valcyclovir]

I really like your suggestion hot. Subsidize node runners in the form of a bounty at the beginning to cover server costs, maintenance and the risk they take for running an early node, and slowly phase out incentive as adoption increases.
If I recall OTP was supposed to do just that, incentivize growth of the DKG on OT Parachain but I also agree that the community values TRAC a lot more for this situation. Node runners run nodes to accumulate more TRAC, not OTP.

[mucke12]

This makes the project a failure if this is not rectified, as to have a decentralized knowledge graph you need decentralized nodes.

I couldn't agree more.

[botnumberseven]

Don't be surprised with such a strong statements about nodes profitability, as v6 nodes profitability discussion does not start from a neutral position, we had the v5 prior to it. And from node runner point of view v5 was a huge disappointment. Let's see how it looked like from a node runner point of view (I'm talking only about xDAI portion here).

In the first few months following v5 launch we saw a significant decline in TRAC locked per job per holder (from 8 TRAC in the first month to 2.6 TRAC by Aug'21, so 56% drop, $5.9 =>$1). But we also saw a significant increase in jobs created per day (up to 150 jobs in Aug'21). So these things sort of cancelled each other out. Within this period total amount of TRAC spent on jobs per day was around 900 (~$350). This is what was shared among all node runners.

Then things went south rapidly.

1. Unlocked TRAC from Starfleet lead to a significant increase in number of nodes. That is on the community, too many nodes were created too fast.
2. Total number of jobs stopped growing, and then started to decline.
3. Amount of TRAC locked per job per holder collapsed from 2.6 TRAC to 0.38 TRAC by Dec'21, additional 85% drop. And it stayed that way since then.

So the total amount of TRAC spent on all jobs per day went down from 900 to 200 by Dec'21 (to $200), much less had to be shared between many more nodes. This is when it became clear that nodes are breakeven or negative unless you play 3D chess and run dozens of nodes on one VPS.
The community was very vocal about it, but nothing changed. Number of jobs per day continued to decline, total amount of TRAC/USD locked per day continued to decline. By Nov'22 we talk about $10 spent on all jobs per day by all DCs combined.

One of the most concerning to the community was the fact that job price changed for all DCs at the exactly same time. Below you can see TRAC locked per job per holder over time. That led community to doubt all the discussions about market forces to drive job price, it was clear as day that job prices are controlled by a single entity.

And the same chart but split by DC ID.

So you see what baggage we have while entering v6. I'm completely skipping any discussion about tech, potential use cases, partners etc... This discussion is purely about economics. How does v6 node looks to a potential node runner?

1. Required capital to run a node increased 16x, even at current rock bottom prices 50k TRAC is ~$8000. This is not something a random crypto enthusiast from reddit would easily play with. Number of community members who can afford to run a node reduced significantly, while the average age and quality of judgment increased.
2. Risks increased as well with slashing mechanism (it's not there yet, but will be eventually). Now the risk is to lock a significant portion of capital, instead of losing a few TRACs on a missed job.
3. Baggage from v5.
4. No clarity about current amount of TRAC/USD spent on publishing per day. No clarity on what to expect in the next 8 quarters.

Hence we have just a handful of folks in the community who runs v6 nodes out of pure enthusiasm and excess of free time. There is just no good business justification to do that.

Potential solution
So, my suggestion to the team is to start treating node runners as businesses. Any business has risk and reward. Right now running a node is a complex, high risk and no reward business. Why would anyone do it?
If community nodes are critical for the OT and TRAC to thrive, they need to be treated as a critical element of the system. E.g. to bootstrap the network an artificial incentive can be used, while there is not enough real use of the network.
Something like publishings dummy data by OT for a period of time. It's debatable how much USD needs to be spent on that per day and how long these charity publishings need to exist.

Personally, I'd start to consider running a node if I see a strong path to make at least 20% on investment on a node for at least 5-8 quarters guaranteed (especially with lack of clarity about node retirement process). Anything below is not worth it, as there are better options around with lower risk and higher reward.

[Valcyclovir]

Thanks for providing numbers as usual! I agree with most things that have been said except the final part. The 20% ROI seems way too high for a non inflationary token such as TRAC. You have provided a very similar suggestion as hottogo but I am more aligned with his numbers. It just seems more realistic to me. If we are able to reach 20% ROI, it would definitely have to be a combination of TRAC and OTP.

[botnumberseven]

I do not see it unreasonable.
250k v6 bounty was announced on Jan 6th 2022, at that time it was ~$425k. So there was a commitment to spend this budget within ~1 year on the community for them to perform a specific function..
If you'd use the same budget to bootstrap a network for 1 year, that would be ~2.6M TRAC (at today's prices)
That's enough to provide 20% ROI for 1 year for 260 nodes (assuming 50k stake per node). I do not even think we will have that many community nodes.
Seems rather reasonable to me.

[Larsk97]

As mentioned earlier by others about roi, I am also waiting on the sideline. Right now the risk/reward do not justify spinning up nodes for me. If we want this thing to thrive, we atleast need to break even.

[mucke12]

Other projects with fixed supply incentive stakers via a pool. (like EWT)
We could max this incentive to 50k token per node with an 10-15% apy on that ( for 1-2 years). On top the node runners have also the normals publishing rewards

[Valcyclovir]

There seems to be several community members suggesting node runner subsidy while the V6 network is still early. The team definitely needs to be careful with local laws and policy and a good way to go at it is through a bounty program just like on testnet. The team spoke about a new bounty briefly and here's hoping our perspectives are aligned.

[mucke12]

Might be a bit far fetched. But there is also on other benefit of this APY incentive:
We can even make sure that delegator will get some rewards from noderunners. Why should anyone share his 2$ rewards wen hes at a loss with VPS. This incentive would be just to kickstart, get people excited and onboard new people

[harveysimpsondata]

I also have the capability to run multiple nodes but have chosen not to.
I did not think we would be here discussing this today.
I thought these issues would have been resolved with the v6 launch.

So currently the price of assets is priced as the third quartile of neighborhood (20 nodes) asks. The TL community has 90 nodes.
I thought all this testing all year would have come up with some kind of mechanism to entice to community to run nodes on their own. I guess the only solution moving forward could be for the community to calculate what their node ask should be so their node could be profitable or break even. It is not ideal to force us to come to a consensus and manipulate the price but we might have to. $8,500 (50k TRAC) to run a node is not chump change compared to v5 and it looks like we are in the same situation as before.

I am waiting to run the v6 nodes; until this, several issues need to be resolved: how profitable nodes are, how to retire the v6 nodes properly, and how long it would take to payout collateral without being slashed.

[CosmiCloud]

My thoughts:
Peg jobs to usd price? - Comes with its own issues like apis, and DCs over paying (which might not be a bad thing at first)

White list until network grows or allocate a pool of growth funds that get paid to node operators over x amount of time?

The old mechanism of dropping off the network when there are too many nodes doesnt really work with v6. With v5 dropping off meant you might lose a few trac from not hosting jobs, now with v6 you will lose 2.5k trac no matter the ask price for each piece of data you dont have. There needs to be a retire function for the tokenomics to be able to claim this feature.

[Valcyclovir]

Using a stable currency instead of TRAC to dictate the price of a publish has long been what the node community has been hoping for. nOS deals with enterprises who don't want to keep cryptocurrency on their books, and node runners want a predictable return as well.

As for the stake slash, you don't "lose" 2.5k TRAC, it gets locked for 2 years but we all know 2 years in the crypto world is a long time. I agree that there needs to be an easy way to retire responsibly so node runners are not stuck for 3 months before they can retrieve their funds. This is a long time compared to all other staking options available in the market.

[CosmiCloud]

lose = locked :) Im aware its not lost, but locked for 2 years is as good as lost! :D

[botnumberseven]

Aye, publishing price needs to be pegged to USD, not to TRAC price.
From DKG user point of view, my publishing should cost about the same, and should not depend on crypto market volatility.

Also agree about retirement, I highlighted that during RFC discussion that reasonably fast retirement process is a must have. But i do not know if anything changed based on FRC comments.

[Valcyclovir]

I believe the whole node community is behind the idea of pegging the service to the dollar.

Here is a suggestion I just came up with:

We create a TRAC/USD pool through nOS or using DeFi on Polkadot.
The service creator decides the dollar value of an asset he wants to publish.
For this example, let's assume 1$ for a 10kb asset for 10 epochs, which comes down to 0.10$ payout per epoch.
At the end of each epoch, proofs are submitted and the node runner is rewarded 0.10$ worth of TRAC at that moment.
The smart contract will initiate a market buy using the TRAC/USD pool.
Repeat that for the remaining 9 epochs.

This will create an additional use for TRAC holders, as they will be able to provide liquidity to this TRAC/USD pool, and earning fees every time a payout is initiated.

[AzorinAZM]

First of all I would like to say that I do not agree with most of what I have read here.

I think that this ecosystem should be driven by market forces, the invisible hand or supply and demand.

I do not share the idea of ​​intervening in a market at all, I think that when market forces are intervened by an external force, they are harmed in the long run, that is, where I live it is called bread for today, hunger for tomorrow. The only case in which I would consider an intervention is in the case of a monopoly or oligopoly in the market, in which one or several actors begin to alter the market using bad practices that harm the rest of the participants in the grid.

It is clear that there is a problem with the profitability of the nodes at the moment, but it is the same market forces that are producing this. It's easy for me to understand:
-Nodes ↑ + publications ↓ = Price per publication ↓

-Nodes ↓ + publications ↓ = Price per publication ↓/=

-Nodes ↓ + publications ↑ = Price per publication ↑

-Nodes ↑ + publications ↑ = Price per publication ↑/=

This is a simple way of explaining, although we should also take into account some other factors.

At the moment both the nodes and the publications are scarce, the network at the launch of V6 is practically not having activity and we should give it some time and see how the next months/weeks go.

Maybe you (node ​​runners) should approach everything in a different way.

If the nodes are not profitable, why do you continue to support the network?

On the other hand, I would also like to argue that instead of using funds to subsidize nodes, why don't we use those funds to attract developers? Why not use the funds to teach how to build solutions with the DKG?

That is, we could use the funds to keep the nodes operational, but for how long? what good would it do? What if in a year we continue with the same number of jobs? What if we don't manage to attract developers and the ecosystem doesn't grow? What use would it have been to keep the nodes artificially connected to a little-used network?

Let's focus on growing this, not short-term profits.

[DalSlacker]

With TL running the vast majority of the nodes AND controlling all of the publishes, how can you expect to see market forces at work?

"The only case in which I would consider an intervention is in the case of a monopoly or oligopoly in the market, in which one or several actors begin to alter the market using bad practices that harm the rest of the participants in the grid."

So you must've not been around for V5? With lambda steadily stepping down over time?

[AzorinAZM]

With TL running the vast majority of the nodes AND controlling all of the publishes, how can you expect to see market forces at work?
"The only case in which I would consider an intervention is in the case of a monopoly or oligopoly in the market, in which one or several actors begin to alter the market using bad practices that harm the rest of the participants in the grid."
So you must've not been around for V5? With lambda steadily stepping down over time?

Yes, I've been here for the whole of V5 and I didn't set up a node precisely because they weren't profitable.

Add that I'm not sure that the drop in the price per publication was due to bad actors. In my opinion they have been the market forces:

-Nodes ↑↑ + Publications ↑/= = Price per publication =/↓

[botnumberseven]

@AzorinAZM I hear you about market forces and we might get to the point when they control publishing prices, number of nodes and nodes profit.
But they were simply not applicable for the whole 2 years of v5 existence.
I showed it clearly that single entity controlled price of all jobs of all DCs, complete monopoly on job creation.

Moreover it was enough to have ~20-30 nodes for the network to function. So node runners had no tools influence the price at all.
All node runners could shutdown their nodes and TL would continue to run their two dozen nodes and that's it. Network would still be functional.

And without any significant change that's how I see the short-term future. There will be a KG, but all nodes will be run by TL. And maybe it's ok.

[AzorinAZM]

If the entire network requires only 30 nodes to function correctly, any amount that exceeds those 30 nodes will cause the price per publication to decrease. In the opposite case, if the publications increase and the network requires 100 nodes, the price per publication will rise, making it more interesting to have nodes working for the network.

That is why I say that it is counterproductive to subsidize the nodes, what is necessary is to build and attract developers who build solutions so that more DCs appear in the market and consequently more publications.

[DalSlacker]

If TL provides 30 nodes in a network that only requires 30 nodes, and additional nodes are discouraged (though financial means or otherwise) I feel like we are missing the whole point here

[jalopy1]

Just adding my voice and concerns to the conversation.
I had been running a few nodes since 2020 and found that the race to the bottom issue has put me off from starting up a v6 node(which I have been very excited for),
I can't afford to run at a loss plan and simple.
I love the vision and trajectory of this project which is why I'm still here and why I ran nodes at a loss. But the time has come for a solution to this issue as its holding many back from entering into node running and supporting the project.

I would also like to acknowledge that it has only been a few week since the inception of V6.

As for the solution to the issues I'm not sure the right course of action but will add my thoughts down when I have ample time to think on it.

[branarakic]

Hi everyone,

Thanks for the comments and feedback on this thread. The core development team values all constructive opinions shared within the community and the effort in the direction of improvement of the OriginTrail ecosystem.

The majority of the points mentioned in the discussion above show that it’s not fully clear how DKG V6 works, how different the publishing system is compared to previous versions, how node retirement works, slashing etc. The gap in familiarity with details of these concepts should significantly decrease in the coming period as more resources will become available, such as additional documentation, pending RFCs, updates in Houston etc. Additional resources are aimed to allow anyone to use the technology in the best way they see fit.

One example of the gap in familiarity with the v6 novelties is the comment on publishing 20k assets at the network launch while having few community nodes present as an act of pursuing centralization. In this particular example, the gap appears in familiarity with details on how DKG assets are hosted in their respective neighborhoods. Contrary to pursuing centralisation of publishings of any assets, all new nodes joining the network will always have the opportunity to download all relevant assets (“sync” with the DKG, a feature coming in the next releases) and then become eligible for rewards, even if they weren’t hosting the assets from the beginning.

Similarly, about the new V6 pricing market mechanics - as a brief response before the additional resources are available - the amount of the network fees paid is determined by the node asks (set by node runners). Everyone needs to set their node service asks at the level they see fit, and therefore collectively set the price in their respective neighborhoods to achieve profitability. The fresh V6 network is expected to see growth in activity with the transformation of existing implementations, as well as new projects being deployed, the expected market conditions are also to be influenced by this and we expect to see a different volume of publishing as well.

In summary, the existing gaps clearly show the need for providing more information, which will be addressed and should help remove most of the confusion, as well as enable active collaboration through the existing structured process. We’re confident that this way the network will continue growing in adoption to the satisfaction of asset publishers, asset consumers, system runners or token holders.

The recent V6 launch reaffirmed that OriginTrail ecosystem is at the cutting edge of innovation, delivering value and solving problems for a vast variety of users and organizations impacting trillions of dollars of value. Inclusiveness, neutrality and usability remain to be the key guiding principles keeping our ecosystem future proof and we received plenty of commendation from ecosystem partners who understand the significance of these norms that underpin our joint achievements. We are pleased to see how some of the brightest minds in various sectors accelerate value creation with OriginTrail infrastructure and our team is excited to continue to stretch abilities in supporting them.

We are looking forward to seeing the number of humanity’s most important knowledge assets grow to make the global economy work sustainably, and we express our immense gratitude to OriginTrail ecosystem stakeholders that keep on making this happen day in - day out.

Trace on!

[Valcyclovir]

Hello Brana,

Thanks for spending your precious time to respond to your node community, very appreciated. While I have no doubts that the community are missing many pieces of the puzzle, I believe in giving constructive feedback throughout the way so the team knows where the node community stands. Right now, the node community is mostly feeling exhaustion and confusion. I am confident the sentiment will change once we receive more instructions and information on current development.

I will keep this brief. In summary, node runners want the price of each publish to be in $/kb-epoch, not TRAC/kb-epoch. Businesses don't want to deal with TRAC volatility to determine the price to use the ODN, and the same applies to node runners. Node runners want to be paid the same amount for a service - TRAC going down should not give node runners less reward and node payouts should not be affected by market conditions. In other words, we don't want our "salary" to be paid in TRAC, but a predictable stable currency. If this is fulfilled, the node community will be very satisfied.

I am glad we agree that communications can be worked on between the core devs and the community, and as a member of OriginTrail Club, OTC, I feel like we could play a role in limiting misunderstandings and helping provide support to the community.

Here's hoping for more discussion and collaboration!

BRX

[DalSlacker]

Thanks for the input, Brana
I'm happy that you are in the fold here, and at least seeing for yourself how the community feels

Just a couple very quick points:
1- The team shouldn't expect many community nodes to spin up until they are financially viable (like Dmitry said, the 50k TRAC required will change the target demographic here, and we aren't sold on the setup yet)

2- Given that the team was bootstrapping V5 for the entirety of its existence, it shouldn't seem like a crazy idea to bootstrap V6 for a while too, IF (and only if) you guys actually care to have community nodes and a decentralized knowledge graph. If you want that decentralization, the yearly/monthly cost should be easy to justify.

Cheers!

[DalSlacker]

In other words, we don't want our "salary" to be paid in TRAC, but a predictable stable currency. If this is fulfilled, the node community will be very satisfied.

Almost
Pay us in TRAC, of course. But pay us a specific USD-value of TRAC, based on publish/ask price and then-current USD/TRAC value.
So if TRAC price doubles, we receive half as much TRAC but an equal value. And if TRAC price halves we receive twice as much TRAC but an equal value.

[hottogo]

In other words, we don't want our "salary" to be paid in TRAC, but a predictable stable currency. If this is fulfilled, the node community will be very satisfied.

Almost Pay us in TRAC, of course. But pay us a specific USD-value of TRAC, based on publish/ask price and then-current USD/TRAC value. So if TRAC price doubles, we receive half as much TRAC but an equal value. And if TRAC price halves we receive twice as much TRAC but an equal value.

This could work, but only at the service start. Not at each of the epochs, no need to re-calculate and convert - especially if Trac is rising over the year. New data being uploaded yes, but not existing data that is getting paid out evenly over the epochs it paid upfront for.

Just my opinion.