Economic reasoning for Storage Emissions in White Paper

[TylerAbeoJordan]

@JimCollinson

Isn’t the opposite true though? Subsidising in a smaller network leads to lower barriers to entry due reduced risk, and more incentive to enter earlier, therefore higher decentralisation due to more diversity of supply?

Edit: I really want to bring @dirvine into this convo too as the more I think about it, the scarier a scenario it is.

The whole post is heavily edited … sorry about that, but I think most of you were still sleeping anyway

To truly subsidize the network, we have to have something of real genuine value. In the early days in particular the network has little relative value (just hope). What the 70% represents is nothing different from what a government does when it prints money. We are diluting value, so not honestly subsiding, we are diluting the value of the network and the value that users hold and redistributing it to random users.

1. The excess tokens they use on the network will drive up storage costs and inhibit/deter new users from coming onboard.
2. Tokens generated that are excess to need will be sold on the open market driving down the price of the token and giving an opening for big fish to game the whole system (as they will know that in the future the supply will decrease as they will know the supply curve). They can use this to take advantage of the market for manipulation and hence facilitate more centralization of token ownership.

This is really important to think about - as we have a network mechanism that is actually working COUNTER to the idea of diluting to reduce the value to encourage new users. Unlike with bitcoin and others, such a dilution will make the network more unstable. As it drives up the cost to store, which in turn reduces the amount of data stored, which in turn reduces the incentive for node operators (especially as the price on the open market is also being reduced by the shift in value) … this means the 70% will effectively shrink the network and risk the loss of data – something that could give us a huge black eye at the start.

This is why I suggested a bell curve release, so that the max tokens given out will happen when the network is more mature and capable of handling the added burden that such a dilution will incur on all users and market-makers. Although the more I consider it, the more I think the 70% as a random redistribution is a bad idea altogether given that the network adjusts to token value automatically. @DavidMc0 had a good idea that I commented on in a later post.

edit: although economically, it would be best and least stressful to the network if there is no 70% at all, winning the hearts and minds of new users is perhaps helpful here – although most of them have no clue about economics, so I accept the premise that such a dilution will attract users in the same way that the promises of a government giving away free stuff attracts people. In truth this is akin to a scam as, economically speaking, we are simply taking value from everyone and giving that value to random users - sort of a lottery. It would be far better to give it as a reward instead.

[TylerAbeoJordan]

@DavidMc0

For example, just chuck the whole 70% into the Royalties Pool with some vesting schedule, so the market for resources can work without interference, and all of the dilution only comes into being by adding serious value to the network ecosystem. A vastly better funded ecosystem might be far better for node operators than a boosted number of tokens earned!

@JimCollinson This would would be great – if the 70% were used as a reward mechanism for adding value to the network. A dilution takes value from the holders of tokens and the users of the network, so asking to give something in return for such a dilution is a much better way forward. --and we don’t risk breaking the network.

[JimCollinson]

@DavidMc0

just chuck the whole 70% into the Royalties Pool with some vesting schedule, so the market for resources can work without interference, and all of the dilution only comes into being by adding serious value to the network ecosystem.

This means the Foundation controlling 85% of the supply from day one, I don’t think anyone really wants that, do they?

[TylerAbeoJordan]

So it’s impossible to have the distribution curve as previous, but the release goes to the royalty pool over time?

[JimCollinson]

You’ll be pleased to know, we have parameters that can be tweaked and adjusted over time… we (and by we that I mean Network participants / community) could change the curve if we liked.

@TylerAbeoJordan

but the release goes to the royalty pool over time?

That would still be the foundation controlling 85% of the supply though, just over an extended period.

Better that it’s decentralised, and goes to people actively supporting the network.

[TylerAbeoJordan]

@JimCollinson

Better that it’s decentralised, and goes to people actively supporting the network.

Yes, better that it’s decentralized but over a long long period of time, means no big pile that would attract bad actors.

Also those actively supporting the network are also already being rewarded. So only taking value from those using the network and randomly giving it back to them and in the process lowering the value of the token possibly causing havoc with the network mechanics.

Better to reward those adding value to the network who aren’t already being rewarded. That grows the network. So even though diluting and raising the cost of storage, it’s still bringing in new people to counter that negative.

edit: I feel like this is probably a tiring topic for you Jim, and it is rehash after rehash (I know you’ve worked on this area for a long time and put in a lot of effort) … I only want what’s best for the network which is why I am pushing on this. I just hope that you and Bux and David will think this through some more before pushing whichever button you decide to push.

[JimCollinson]

@TylerAbeoJordan

I feel like this is probably a tiring topic for you Jim, and it is rehash after rehash (I know you’ve worked on this area for a long time and put in a lot of effort) … I only want what’s best for the network which is why I am pushing on this. I just hope that you and Bux and David will think this through some more before pushing whichever button you decide to push.

It’s ok, you develop quite a tolerance for long discussions that get hashed and re-hashed (self encrypted even ) over the years.

But this is also the year of action. We have to do things now!

Thankfully, though, it’s just not down to me… I am but a plan wrangler! We’ve got folk modelling things, testing strategies, and asking tough questions. Yourself included.

We all want what’s best for the Network! So don’t feel a bother.

[DavidMc0]

@JimCollinson

That would still be the foundation controlling 85% of the supply though, just over an extended period.

The foundation controlling 85% may not be ideal, but if they can be trusted with about 50% of the initial supply, why not 85% of the total supply subject to vesting, potentially over a decade or so?

@JimCollinson

Better that it’s decentralised, and goes to people actively supporting the network.

Is it though?

The downside of this suggestion is that the foundation would control a lot of supply (85% vs 14%), which obviously isn’t ideal.

But the plus sides are huge:

• a better functioning market for Autonomi resources (better price signalling and less price distortion as I’ve previously outlined)
• A vastly better funded ecosystem (better core development, better apps, better marketing, better resourcing of partnerships, more significant impact fund etc etc etc)
• Due to the greater investment detailed in the previous point, the network would have far greater resilience due to more users, better code, more diverse applications etc etc.

With the current proposal, as Storage Emissions
are issued to node operators in year 1, those emissions will be actively reducing the value of the 14% Royalties Pool through dilution, therefore weakening investment in the ecosystem.

Honestly, I know which of these options sounds better to me, and it’s not close.

But yes, 85% is very significant. Why not reduce it, but still not implement the potentially harmful Storage Emissions & direct supply to something far more productive & impactful?

I’d prefer a doubling of the Royalties Pool + eradication of Storage / other resource emissions, as it would lead to a massively better funded Autonomi Ecosystem, and avoid the probably harmful issues with the Storage Emissions.

Node operators don’t need the emissions, so it’s no loss to them. But the whole ecosystem could be hugely impacted positively with this change.

@JimCollinson

We all want what’s best for the Network! So don’t feel a bother

Absolutely. I don’t want to be a pain either, but want to thrash these thoughts out while they can still have an impact

[Anselme]

@neo

But can the issue of people knowing the address prefixes and then generating chunks whose hash matches the prefixes

Indeed, rewards mining is definitely something we want to avoid but since we can’t stop people from doing it, what we can do is make it harder and unprofitable.

One way to make it harder for example is to only allow rewards for chunks that are MAX_SIZE, as @mav suggested: sha3 of 32B is about 7000x faster than sha3 of 1MB. This comes with no extra effort for regular nodes, but makes rewards mining extra difficult.

We can also adapt the rewards to be lower than the mining cost by adjusting the ‘difficulty’ of this mining.

Hopefully this gives a few pointers to how we can counter such behavior.

[neo]

@Anselme

One way to make it harder for example is to only allow rewards for chunks that are MAX_SIZE, as @mav suggested

Hopefully you have already seen the problem with this.

There is no guarantee that the addresses generated will require the hashing of a full size node.

Since, as Jim said, they are not the full addresses, but prefixes, since full 256 bit address has its own problems, which means there will be many hashes/addresses what will match so my easy “mine” will work without issue.

If you use full addresses then its unlikely that many of the hashes/addresses will ever be found, 256 bits is a big number

[neo]

Another question. Where are these hashes/addresses (prefixes?) supposed to be stored? Will they be stored at an address on the network?

And how much overhead will there be checking each and every store again a subset of this list? Will that be kept in memory and read in when the node starts (and it creates a memory store of the subset around its XOR address)

[neo]

@Anselme It might be better to use the hash of the payment data block rather than the address.

Because then a gamers would need to be getting quotes and generating a spend meaning

• generate a chunk
• get a quote off a node close to the hash of the chunk
• generate a spend

Makes it a lot more difficult task and well and truly limits the number of attempts they can do in a day.

If it took 100mS to get a quote (often longer) then they could only do 840,000 attempts maximum on each thread.

[Anselme]

@neo

It might be better to use the hash of the payment data block rather than the address.

Indeed! Linking to other related data like the payment further increases the difficulty for miners, I think that’s an excellent suggestion!

[neo]

This is why I ask questions and try and point out dead easy ways to break the system. Not as a hacker but just as an engineer (with 50+years experience)

Yea the simple chunk address prefix is too easy to game

Just need the payment data block once encrypted to be random enough, be better still if it has to be spent, then that would be a huge barrier to game, very costly

[DavidMc0]

Just coming back to this again, I had a thought on a possible improvement in how tokens could be issued to far greater effect than just giving it to node operators, but in a way that still involves them in the process.

Potential value of emissions
Firstly, to put a scale on the potential value of issued tokens during the first few years, I did some calculations / estimates based on the emissions rate given in the White Paper along with my own rough assumptions of possible token price and market cap:

Year:	Value of issued tokens:	Assumed SNT Price:	Assumed market cap:
1	$275,391,307	$2.11	$3,000,000,000
2	$475,730,345	$3.89	$6,000,000,000
3	$692,321,754	$6.04	$10,000,000,000
4	$914,803,664	$8.51	$15,000,000,000
5	$70,351,694,820	$697.46	$1,300,000,000,000

Obviously nobody knows what the token price will be at any point in the future, but these seem like plausible figures that can be useful for the purposes of this discussion.

With an SNT price of just over $2 in the first year, emitted tokens (~130m tokens if my calculations are correct) would be worth around $275m.

The 5th year value I chose was to show what the value of the emissions might be if the Market Cap of Autonomi in 5 or 6 years time were equal to where Bitcoin is today, which is possible if the network is successful, and Bitcoin is at a multiple of its current price.

$70bn is a huge amount of value, and if this is a possibility, it’s worth being very careful about how that value is leveraged to do the most it possibly can for the Autonomi ecosystem.

Huge value of tokens, what benefit?
In the current proposal, the benefits of the emission of new tokens only go to node operators, providing no further ecosystem benefit beyond the dilution of MAID holders, Shareholders, and the Royalties Pool.

I previously suggested that instead of the new tokens being issued in a way that doesn’t benefit the ecosystem significantly, it could be put to better use if it were added to the royalties pool instead.

I feel it would be far better if, e.g. $275m in the first year were invested into promising projects, core development, marketing efforts, partnerships, and other activities that provide huge benefit to the Autonomi ecosystem, rather than distributed to node operators with possibly zero benefit to the Autonomi ecosystem.

But, an issue @JimCollinson mentioned with this is that the foundation would then be controlling an additional 49% of the token supply, which has its risks.

Foundation crowd-funding idea
Might it make this less of an issue if the additional resources provided to the foundation through emissions were democratically allocated by node operators? Here’s an example of how this could work:

1. Instead of receiving SNT token emissions, node operators receive Foundation Voting Tokens

2. The foundation lists projects and activities that are eligible to receive Foundation Voting Tokens

3. Node operators who have earned Foundation Voting Tokens allocate them to project that they want to see happen

4. Once a project has received sufficient Foundation Voting Tokens, they can be exchanged for SNT from the foundation’s 49% fund, and their project / activity can commence.

This would mean that while the foundation does hold more tokens, the allocation of those tokens is decided by node operators. It allows the possibly billions of dollars of value from token emissions over the years to do loads of stuff that will be hugely beneficial to the ecosystem, rather than entering circulation in a way that is potentially a net-negative to the Autonomy economy and ecosystem (by diluting Royalties Pool and distorting pricing signals in the market for resources).

Pros and cons

White Paper token emissions plan:

Pros:

• Effectively dilutes early holders (MAID holders, Shareholders, Royalties Pool) for more equitable token distribution

Cons:

• Benefits of new token supply goes to node operators only, and benefit is short-term (diluting value of prior earnings)
• Price distortions may reduce efficacy of market for resources
• No significant boost to ecosystem despite possible hundreds of millions of dollars worth of tokens entering the supply each year
• Diluting effect on Royalties Pool holdings lessen the impact the Royalties Pool will have in benefiting the Autonomi ecosystem

Expanded royalties pool + node operator voting:

Pros:

• Effectively dilutes early holders (MAID holders, Shareholders, Royalties Pool) for more equitable token distribution
• Avoids any price distortion in the core market for resources
• All new tokens entering the supply over the years will add value as they enter the ecosystem
• Node operators get to allocate funding to projects that they think will best serve the ecosystem, in line with their own self-interest to boost the value of their earnings / holdings

Cons:

• Additional 49% of supply is held by the foundation

If this idea isn’t workable, I think it’s well worth looking for ways to leverage the massive value of tokens planned to enter circulation in the coming years / decades to do great stuff for the ecosystem vs doing negligible good and possible harm, which would be a huge wasted opportunity in my view.

[neo]

@DavidMc0

I previously suggested that instead of the new tokens being issued in a way that doesn’t benefit the ecosystem significantly, it could instead be put to better use if it were added to the royalties pool instead.

Have you missed the point that to some extent the node receiving emissions does help the ecosystem as well. It provides a few things. Now I would work on a system that does both your suggestion and the current, rather than viewing it as all one way or the other way.

• during times of lower data uploads, it encourages more node operators to keep their nodes online
• encourages more nodes to be run for each operator and tips the balance in the minds of others encouraging them to run nodes.
• both those improves traffic flow by having less churning
• node operators will spend their tokens, either selling them or uploading data themselves improving token flows.

I do not see giving emissions to nodes as a total negative and has many necessary benefits.

I can see your argument that the foundation receiving some of the emissions, but disagree with all emissions as stated above.

Also have you considered that eventually as tokens flow around the foundation will effectively receive at least 15% of the emissions as data is uploaded in royalties.

Have you considered the foundation receiving the 21% and leave the 49% as emissions to nodes. (70%x70%=49% of whole and 70%x30% = 21% of whole) knowing that some of the 49% will go to the foundations anyhow to be distributed.

[Bogard]

@neo

the point that to some extent the node receiving emissions does help the ecosystem

I’d go even further. Once the protocol has been released and hardened, the node is doing all the work in the network. It’s the base that everything else reposes on. So the node should get all of the rewards/coins and circulate it thereon (as is the case in bitcoin).

@neo

Also have you considered that eventually as tokens flow around the foundation will effectively receive at least 15% of the emissions as data is uploaded in royalties.

On top of its genesis allocation. That 15% perpetual rent is effectively raising prices by ~18% on clients by raising node operator’s expenses by ~18% which nodes will pass on to clients (i.e., what should be a price of 85x is being raised to 100x).

@neo

Have you considered the foundation receiving the 21% and leave the 49% as emissions to nodes. (70%x70%=49% of whole and 70%x30% = 21% of whole) knowing that some of the 49% will go to the foundations anyhow to be distributed.

Yep, which still gives me the ick. Everything should go to the nodes (the foundation already got more than it needs at genesis, no need to further double dip). Anything else will eventually see nodes (and clients) revolt.

[DavidMc0]

@neo

Have you missed the point that to some extent the node receiving emissions does help the ecosystem as well.

No. Earlier in this thread I argued for why I see node emissions as providing no benefitial role in the ecosystem, and possibly causing harm.

Firstly I argued that, assuming a functional market for resources, there shouldn’t be any significant dips in demand that would lead to nodes losing incentives to run, as any temporary dips would lead low-priority upload / sync jobs to trigger to start as store-cost dips.

Secondly I argued that even if a scenario occurred where demand were dropping in a way that threatens nodes, dilution of the token supply wouldn’t be a very good incentive, as it would reduce the future value of current earnings, and won’t do anything to stimulate demand to solve the lack of demand issue.

Node operators will be willing to provide resources if they’re confident in the future value of current earnings, and I expect they will be more confident in this if there is greater investment in the ecosystem, vs just being given a higher number of tokens within a lower investment ecosystem.

Also, in a low-demand situation, token price would presumably be dipping, so increasing supply likely won’t be a good incentive as operators will dump tokens exacerbating the issue by discouraging new nodes from joining, rather than solving the problem.

So, when there is healthy demand for network resources there is no need for token emissions to incentivise nodes, and where there is low demand for network resources, token emissions to node operators seem to be the wrong medicine.

@neo

Now I would work on a system that does both your suggestion and the current, rather than viewing it as all one way or the other way.

Yes, this would also be an improvement on the white paper suggestion.

Though, unless there is a good argument for why node emissions would be beneficial, I hope they are scrapped and better uses for token emissions are found.

@neo

Also have you considered that eventually as tokens flow around the foundation will effectively receive at least 15% of the emissions as data is uploaded in royalties.

That’s a good point, as it will prevent the emissions from eroding the overall royalties pool, but keep it even.

@neo

Have you considered the foundation receiving the 21% and leave the 49% as emissions to nodes. (70%x70%=49% of whole and 70%x30% = 21% of whole) knowing that some of the 49% will go to the foundations anyhow to be distributed

That is also a possibility that would lead to greater investment. I doubt the 29% will be needed for any future incentivisation, as markets for any resources should work without additional token issuance.

[neo]

@Bogard

Yep, which still gives me the ick. Everything should go to the node

Just to be clear the whole 70% is going to nodes according to Jim and WP. 49% as emissions and the 21% is yet to be known by me how that will get to nodes, but have been assured it is

[DavidMc0]

@Bogard

the node is doing all the work in the network. It’s the base that everything else reposes on. So the node should get all of the rewards/coins and circulate it thereon

Nodes are obviously essential, but they aren’t the sole providers of value.

Nodes couldn’t earn anything if their software isn’t working well, or if the network isn’t developing over time to ensure it’s relevant to users to keep demand from users up. If the foundation is doing a good job, node operators may prefer to have them getting their cut than not.

Also, I seem to remember Jim saying about the possibility of a governance system, where nodes could vote to change the allocation of rewards to the foundation, so if that’s the case, we’d get to see what they decide.

@Bogard

On top of its genesis allocation. That 15% perpetual rent is effectively raising prices by ~18% on clients by raising node operator’s expenses by ~18% which nodes will pass on to clients (i.e., what should be a price of 85x is being raised to 100x).

That is true, and removing these would improve the price signalling in the market for resources, but might harm longer term ecosystem development.

If token emissions were to go to the foundation / another way of stimulating broad ecosystem development, it might mean that the 15% of node rewards going to the foundation could be scrapped, as there will be more than sufficient funding from the Genesis allocation + emissions.