If more storage resource is required, it means someone has been paying the network Safecoin to put data onto the network. The network can use this Safecoin to incentivise farmers to provide more space by increasing farming rewards, and at the same time start increasing the price of putting data on the network as space becomes more scarce.
Assuming some people start farming out of curiosity, there should be sufficient storage in the very early days to kick-start a market.
Farming wouldn’t need to rarely create a coin; if it is receiving Safecoin from those who want to store, it can distribute Safecoin to those who are offering resources.
Reasons for Safecoin increasing in value from there could be anything that increased demand for Safecoin, including growing use of the Safe network, or Safecoin as a crypto currency. If demand goes up with a fixed supply, the price will also go up.
Not having an increasing supply will in no way prevent the rise in value of Safecoin - with a given level of demand for Safe network services, the value of each coin will be higher the smaller the supply.
Investors will factor in future anticipated dilution when calculating a present value for an asset like Safecoin to make decisions on whether to invest or not as a matter of course. Investors making such decisions will have a very significant impact on price.
Saying this isn’t the case is like saying investors in shares don’t take into account anticipated future dilution, or that investors in fiat currencies don’t take into account inflation.
Even with all coins issued at the start, people would still need to pay for network resources with Safecoin, and would need to farm in order to receive Safecoin, which people would want to do, as Safecoin would be valuable for use in paying for network resources and as a crypto currency. So all required incentives would be in place for a functioning market.
Yes, I think markets will have factored in the anticipated dilution to some degree. The price would very likely be significantly higher if there were no anticipated dilution because big investors would divide their future projected network value over a smaller number of future coins, that when discounted to today would result in a larger present value. This means they would likely buy at higher prices, and only sell at higher prices.
I’d be very interested if you could point to any actively traded coin unexpectedly increasing or decreasing its ‘hard cap’ significantly without impacting the coin’s price.
IIRC safecoin is “hardcapped” to 2^32, and each will be divisible to at least 2^32 as in the original proposal. This maximum total supply that is able to circulate won’t change, unless the private keys for coins are permanently lost by human owners. Much like when bitcoins are “burned”. In that case the effective hard cap will be reduced, thereby permanently increasing the relative value of each coin.
As far as the total amount of coin in circulation (within the network) at any particular instant, this is supposed to fluctuate. An analogy is how supplies of agricultural commodities fluctuate depending on whether rice farmers had agreeable weather pattern which gave them a bumper crop, or there was a drought which didn’t. Investors profit greatly from commodity speculation in traditional stock markets all the time. I think safecoin will feel familiar in that regard.
But will only pay out at 1/1000th or 1/430th or 1/200th the rate depending if billions or 1000s of people start uploading large or smallish amounts each. This is because you gave out all the coins.
Be like paying all workers collectively all the money the company has when when they start the job. A very small few will stay and toil and most will relax. You need more than meager incentives. And then you made the coin worth 1/10th at the start so 1/10th the incentive too.
Even if you paid out quicker by removing the scarcity factor then all the coins are rewarded quickly and farmers are then left being paid only when a “PUT” balance of uploads is done. Thats like living hand to mouth all the time and never enough.
Considering I didn’t say that. I said they have taken the future increase over 2 decades into account from the start.
You can only “hope” that this will happen and safecoin worth 1/10th of what it would be. You need to change the reward model for SAFE in order to succeed. The current model simply would not work if you gave all coins out to start with.
SAFEcoin will never have this happen either. There cannot be a situation where there is even a 10% increase/decrease “suddenly” occur.
Are you saying the network can only pay out a fraction of what it receives in Safecoin back to farmers? That makes no sense to me. Surely if the network receives 5000 Safecoin on day 1, it will have 5000 Safecoin to reward farmers with to incentivise resource provision as required?
I think we’re in agreement on this one - the planned supply increase will be priced in today. When I stated this previously you replied that it was a false argument without merit, hence why I didn’t think you agreed (perhaps my wording could have been clearer).
We seem to be misunderstanding each other here - basically what I want to know is: what is the reason for the 90% of the coins that the network holds? What function will it play? Why is it required?
A simple economic answer of what role the network’s subsidy will play & why it is required would be appreciated, as I can’t see any need for it to create a healthy, functioning market for Safe network resources.
If you read my previous post then you would see that its not paying out a fraction of what it receives. But rather the CHANCE of receiving it is 1/430 (if 10K coins available). Over time the random probability would be that the whole 10K will be distributed. But EACH time there is only a 1/430th of a chance of the farming attempt succeeding. So farmers would be pissed off to say the least.
This is why I said to give out all coins to start with, you need to redesign the reward system.
And if you read further I qualified that
I think though that I treated the “impact” as meaning something different to what you meant it to be
The reasons are
future incentives The network can reward people when uploads are low.
Fast response to low spare space conditions. The network is able to reward much faster then spending to network is.
When spare space is low, then rewards are higher (faster), but the safecoin cost to buy resources increases also (to slow down uploading). This creates a problem if working hand to mouth since the network runs out of coin very quickly, but with the “reserves” then the network can very easily ride out this situation.
Generally people (yes not all) naturally lose interest if everything has been given out and new people feel they have little chance of getting a coin and indeed they would. Only those creating large farms would have a chance of success and if all given out at the start then the returns are 1/10th $$$ wise of what they would have been if only 10% given out.
It allows the network to change rewards/costs by large amounts without causing a loss of ability to reward.
Not as an investment. There are more reasons than investments when something rises in price.
For SAFEcoin the rewards reduce (very slowly though) as more coin exists. So for a farmer in the early days they might get “X” coins per year and 10 years later it is “X/8” coins per year for the same “work” This is due to the scarcity of the coins function in rewards
So the thought based on just that factor alone is that its not worth farming in later years.
What the graph shows is that the coin in $$$ value will be rising ans so the “X/8” will be worth a lot more $$$ wise than what the “X” was 10 years prior. And thus still worth farming.
I thought that Safecoin & farming hasn’t even been coded or the specs finalised?
I think it would be worthwhile letting desired market dynamics shape the algorithms / code, rather than the other way around, so hopefully the reward system can be adjusted or redesigned if improvements are possible.
Thanks for these points.
When uploads are low (demand for network resources is low), the network should reduce farming rewards in proportion to the degree it reduces the price per put. If a subsidy is required, then the market is not functioning - the point of reducing rewards when demand is low is to disincentivise further resource provision when it is not required, and encourage people to use the network more.
Having a fast response to low-capacity should not be required, as the price will gradually increase as demand increases relative to supply, leading to a gently increasing incentive to farmers to increase space being made available, at the same time as reducing demand through increasing put price.
With a well functioning market, the rate of new resource being made available should not stray far off the rate of demand for new resource, and if it does, it will balance out due to the pricing mechanism the network manages.
I don’t think people will lose interest in providing sufficient resource to the network as long as the reward for doing so is makes providing resources worthwhile. I don’t think returns should in any way be affected by how many coins there are to give out - they should only be determined by supply and demand for resources. There will always be rewards to give out as long as people are uploading data to the network.
If none of this makes any sense, let’s just leave this for now, and when I have time I’ll put together a basic outline of the network economics as I see them to make discussion easier
@Deadloch, have you managed to progress your idea of an article on the economics of the Safe network? I’d be very interested to hear your thoughts & insights.
I’m confused. I’m not sure if you are speaking in terms of external world markets or the internal “closed cycle” safecoin ecosystem when you talk about pricing. I also think subsidy is not a fair “coin of phrase” for describing the process. The external market of human creation will decide the price of a safecoin based on normal supply/demand market fundamentals. The internal safecoin network fundamentals consist of 3 players ( consumers, producers, and the network which facilitates the exchange based on supply and demand of physical network resources). As I mentioned before, it seems rather obvious that the initial condition of ~10% humans vs. 90% network safecoin ratio was in order to allow a supermajority of resources so that the network could have the opportunity to rapidly expand to steady state as it deems necessary. From the network’s perspective I see it as kind of like having startup funds for a new venture, or analogous to having a good layer of fat in the rare case of famine.
There is an RFC which has the basis for the coding. The basics have been around for quite a while now.
Ummm that is what was done. It is the desired dynamics that shaped the RFC and what I explained.
The costs at this time are determined by the primary variable of Farming_Rate. This rate is influenced by the amount of spare space.
It might be more difficult to determine when demand (uploads) is low, because you have to have a metric for that. The metric at this time is how fast spare space is reducing and the amount of the spare space is used to measure that.
I am a bit worried about you use of subsidy. That is not a part of the RFC and suggests there is a base rate for payments of resources. There is no base rate unless you consider 10^-18 as the base rate and you cannot subsidy that.
The use of that terminology suggests you have an alternative view of the dynamics.
Fast and gradual are not defined by you. Fast as in hours or fast as in days/weeks??
When the network is small to medium fast will NEED to be in terms of hours since in a 10000 node network a small outage of a country could mean the network is low on space immediately. Or if people react en mass to $$$ price changes to safecoin then the network could be low on space in hours. As the network grows the problems reduces a lot and the fast becomes days then weeks.
The issue is that if you give out all coins then the smaller the network is the worse off it is in its ability to react. If farmers are unlikely to get a coin then only the diehards will be farming and that may just not be enough. You need to reward farmers early on to grow the network. Giving out all the coins is a self defeating system since only diehards will farm and if they don’t increase their space regularly then they eventually end up with all the coin then no more space.
This is a typical control system scenario and giving out all the coin is like putting a huge damping block into the feedback loop.
I still hold to the statement
If you want to give out all the coins to start with you NEED to change the basic model SAFE is going to implement.
Only then could it work and I do not deny that. I just think its a difficult task and the 10% ~ 90% model that David devised seems to be a good one that has a lot of benefits.
I better get cracking on this, just back from holiday in sunny Finland. Keep up the good work @DavidMc0 I’m liking the discussion you’re fuelling here, I feel we might be on the same page on some of these issues.
I’ve only been talking about the internal ‘closed cycle’ ecosystem. I’m only concerned about the internal market of the SAFE network, which can work whether a coin is $0.10 or $1000 (assuming divisibility at some stage).
The issue I have with this concept is that invariably when resources are demanded, people are paying the network for storage, so the network will have the Safecoin required to incentivise farmers to provide more resource by increasing rewards with no need for a buffer.
On the flip side, if there is plenty of spare capacity, the network will reduce the price paid to farmers until the rate of new storage being provided is closely matched with the rate of new data being stored.
Why would it be beneficial to incentivise the provision of excessive resource to the network?
If the network pays farmers above the market rate for providing resources to the network, the result will be cheaper than market rate PUT cost.
Surely this is a subsidy? My fear is that if putting data on the Safe network costs less than the price farmers are willing to provide resources for because of a subsidy, people will begin using the Safe network in an unsustainable way.
Why not let the network facilitate a fair and real market between producers (farmers) & consumers (uploaders) so that the price reflects the costs involved?
I get the concept of having some money in the tank can be useful when talking about a startup, but the network’s job is maintaining a strong network, which I think will be best served with accurate pricing signals from a well functioning market.
What ‘famine’ situation could the network face that justifies holding prices artificially low (which is what paying farmers above what users are willing to pay for storage does)? What benefit would it have in the long term?
I’m just looking at it from a high, simplistic level; if farmers end up getting paid more than what users are paying the network to upload data, then the network must be subsidising farmers, which could be unhelpful when trying to create a well functioning market.
I’ll have a look at the RFC if it’s understandable to a non-coder, but right now my thoughts are not about specific implementations but high-level economics of the system.
It generally takes a couple of days to order and set up a new bigger hard drive, so responding within hours may be tough, unless people withhold some of their free space until the farming rate increases. I imagine that farmers will be more likely to maximise what they offer to the network to maximise their possible rewards.
The network should grow organically based on supply & demand.
If the network subsidised farming to the degree that uploads cost far less than they otherwise would, the result would likely be more storage provided, but there would be a corresponding reduction in the cost per put, so people would upload things that they wouldn’t if paying the true cost to farmers.
The result would just be a distorted market during a growth phase that encourages unsustainable use of the network that won’t continue in the same way once the subsidy stops.
This may be disruptive as users who found use cases for the Safe network that no longer make sense once the network stops subsidising farming, and another down side is that it inflates the Safecoin supply, reducing its value compared to if this inflation did not occur.
Apologies, I don’t understand this at all. Why would non-diehards stop farming because the network doesn’t have a huge stash of Safecoin?
Why wouldn’t people increase their space offered to the network if the reward / price of providing storage increased in line with any increase in demand?
No more space would require the network to completely fail in passing on income from PUTs to farmers. Doing this well doesn’t require a huge (e.g. 90%) stash of coins, though a small degree of smoothing could keep things predictable, and require only a tiny buffer.
This is fine - I’m looking at the model and trying to get my head around the proposed, and optimal economics of it. Technical details are more likely to get in the way at this stage, though some will be important (such as knowing which metrics the network has to judge the pricing levels for PUTs and farming rewards).
Apologies if I tend to repeat myself in the following set of comments.
By allowing the network to have a large buffer, it has the capability to absorb a lot of real-time volatility. Another analogy might be like having a large flywheel to smooth out transient torque loads, or a large capacitor to smooth out ripple voltage. More grease for the wheels to overcome friction,the yolk in the egg for the baby chick, etc. Who wants volatility when it comes to keeping their data safe?
No. You are forgetting the fact that “the network” is also a self-governing / autonomous market player. You can’t have a “cheaper than market rate” because “the market rate” is the set by the interaction between the network and other market players in order to satisfy it’s prime directives, which are the prime directives we have given it; like keeping all of the data under it’s care safe and sound forever, and maintaining or improving communications performance for example.
Again, you are forgetting that there are more than just producers and consumers involved, the middle man between those two is also a market player, which in this case is the network itself. The price does reflect all real costs involved for the internal economics of the SAFE network.
The pricing signals are not only accurate, they are exact. Again you are forgetting about the network as a market player. The constant interaction between the network and human players form a symbiotic relationship. The algorithms themselves will continuously optimize PUT/GET prices with respect to storage supply, demand, and the networks own available supply of safecoin in order to maximize data safety and network performance / resiliency. This is no different than any self-governing human service provider. The difference is that the network’s prime directive is to protect your data and communications above all else, rather than generate safecoin profit. If it happens to accumulate safecoin in the process that is serendipity and it doesn’t really care unless experiments show that a particular safecoin buffer size has positive effects on network performance and it may try to push a little towards that goal. This is all yet to be decided based on testsafecoin. Initial conditions happen to make the SAFE network the biggest safecoin whale in the ocean, with good reason, since initially it needs to be well endowed “financially” before it matures and becomes physically strong enough to take human abuses in stride.
There are various scenarios. One case is when large sums of safecoin are held by savers in paper wallets or “lost” and not spent on resources for long periods of time. Yes, over near to long term the external fiat markets will adjust and divisibility may kick in, but the initial super-majority of resources allocated to the network allows it the most leverage while it is young, small, and relatively weak. You also need to consider fiat manipulation of the safecoin price, so the network needs the flexibility to overcome those hurdles depending on instantaneous needs and conditions. PUT and GET price adjustments will keep things in balance, but there is always a lag or time delay between when prices change and new resources come online.
After the network matures, it’s physical size and capacity provide the needed inertia/buffer in order to allow for a higher percentage of resources to be in human hands without causing too much volatility, and divisibility may/will be more in play at that time. Volatility may be what the typical day-trader desires, but who among actual users of the network wants large price swings for consumable resources? The network is continuously trying to optimize in order for everyone to have data that is safe, sound, and accessible. Far better for things to be relatively stable with free market forces leading to long term price trends based on average market prices and general technological/farmer improvements. We’re telling the network that it’s sole job is to keep data safe and secure, so why not give it all the help we can initially? At least that’s my perspective on it.
Remember that a 90/10 or 80/20 or 99/1 split between the network and the human population is only a set of initial conditions, because the whole point of the project is to have an autonomous network free of direct human intervention that is able keep data safe… not something that serves as a real time stock market surrogate. Could you arbitrarily choose a different split ratio for the initial conditions? Sure, but I think the approximate 90/10 or 85/15 split was a very very good choice in order to follow human custom and psychology. (Ever hear of the 80/20 rule?) The network would be better off with a 99/1 split but initial investors would have likely complained and then the network wouldn’t ever get built. Anything more than a 51/49 split between the network and all other humans would make people question whether or not the network would have enough resources to protect itself right after launch while it is still an infant.
It’s a complex system and the current set of initial conditions make the network a market player who is the biggest whale in the ocean. If you just increase every maid holder’s stash by 10x, the network now starts off as the smallest amoeba in the sea, susceptible to the whims of human whales and/or other high-roller adversaries. What benefit would that serve other than to increase volatility and risk network collapse due to human attack/abuse? None. Who care’s how much safecoin we have if we don’t have SAFE?
Because you have to restrict the functions to a unity fnction. safecoin in === safecoin out on a time basis. In other words you restrict the network to how it can attract farmers.
Imagine a time (holidays), or global event that sees uploads dry up (almost) for a month or more. The network would start a downward turn with farmers not being paid after a while. Then farmers leave, the farming rate rises and the few uploaders slow down due to higher upload costs. Farming rate is high but no coin to pay farmers so no new farmers join.
The network will spiral downwards. And nothing to solve the issue.
Simple control theory. Only when the network has grown to sufficient size can scarcity of coin be tolerated.
BS. Its a wrong concept. Its simple pricing to attract or not attract. Subsidy is a different kettle of fish and may explain why you think giving all coin out to start with is a good idea.
NO NO NO the network is the one paying the farmers. The uploaders are NOT paying the farmers. If you use the model (hand to mouth) to pay farmers then the network paying more is a subsidy, but with the current model it is not. It is simple pricing to suit the conditions. The harder the conditions the more workers are paid, the easier the conditions the less they are paid. The worker in the harder conditions is not being subsidised but being pay for the work he/she is doing. Subsidy would be artificially paying more for the same work
If you give out all the coins then the uploaders are indeed paying the farmers since that is what the model is CHANGED to. But the current model is the network pays the farmers;
The uploaders pay the network. But there is a distinct disconnect between the network being paid and the farmers being paid by the network. The only connect is the scarity is reduced by uploaders. Hell the coins paid to the network are physically/digitally destroyed. If this is not a disconnect then what is. The network wil be able to pay farmers for a decade or so without uploads. Even after 2 decades there would be coins available to pay.
Yes the uploads being paid for help the scarcity.
As I said in another thread. The price wil not drop because only 10% are given out up front. The network grows to accommodate the number of coins existing and thus the price grows. To have all coins existing means that the coin is worth 10% of what it was on day one. Less farmers are willing to farm a network that is living hand to mouth. To give out all the coins at start is to create the poverty trap that people who live hand to mouth suffer, except its the network that will be a poverty network never quite having enough to survive the up and downs that always occur
To give out all coins at the start is also akin to sending your newborn, or toddler, or young child or teenager out to design and build rockets. No you grow into things and so as the network grows in size then too does the number of coins existing grow,
Thanks jlpell and neo! You made me understand the dynamics on a deeper level. Your answers should be saved for future reference.
This is quite an important explanation IMO and I feel it deserves to be fleshed out in very simple terms and especially with visual aids (illustrations) to explain the concept of the network as the middle man, how coins get recycled and rewarded, how if affects the price (or actually how it is more like disconnected from it) etc. I expect this subject to come up again and again in the future (as some other subjects in this forum do) as we get nearer to launch.
This will dispel the misconception that investors share/coin worth gets diluted to 10% as the network launches and also proves that having no inflation - that means giving out 100% now to the investors - will probably mean a crippled network unable to mature.
One of the initial misconceptions I had was that that I was thinking in terms of the “hand to mouth” model too, neglecting the role of the network as a player. So I had imagined a few possible “gotchas” and problems. Once people start seeing it as a real human-machine interactive relationship (game theory?) the beauty of the proposed system comes into focus. It also brings in interesting debates for futurists such as “personhood of a network” or the “personhood of an AI”.
I think some of the terminology strewn throughout the forum makes it a bit hard to see this relationship at first. Some people have used the term “burn” when a coin is paid to the network for storage, the other original term used was “recycling”, and then there is just “paying” the network for PUTs. The burn terminology is incorrect because that should be used only in the case when a safecoin is lost forever by its owner’s negligence or misfortune. The recycling term is too passive, and makes it easy for people to forget about the service the network is constantly trying to provide for us. So as I’ve mentioned before I think the easiest way to view the situation it is to just pretend “the network” is a market agent. It just happens to be an AI, but I suppose that’s not much different than competing with robo-traders, accept for the fact that SAFE is programed by design to help all of us as much as networkly possible rather than extract profits from people.
My issue is trying to understand what role the network should undertake as a player other than balancing supply and demand between producers & consumers.
In my current thinking, the network only needs to make sure there’s a healthy amount of resource available. It can do this by facilitating a market between those who provide resources, and those who consume them, which requires no significant reserve.
What role should the network take as a market player other than ensuring sufficient resource availability?