How will pricing for services be calculated?
Let’s says someone wants to store files on the Safe Network. Obviously he has to pay with Safecoin. But what will be the price? Suppose person X pays 1Safecoin for 10MB storage.
If he pays this amount he agrees with the price. But what will happen when Safecoin skyrockets 10-fold? He will no longer be happy to pay 10 times more.
Is there something I missed?
1 Like
There are really three ways that they could code a fix to this problem; with external controls, strictly network only controls, or a mix.
The first would be with external controls, sort of like a Tether token. If they had some awareness of the price of the coin versus, say, vs USD, they could adjust the price of storage accordingly, so what used to cost 1 coin per MB now costs .01 coin per MB. This has a downside, though, as it does not account for how much available space is in the network.
The second option would be internal only controls, where the network recognizes the supply of coin (and thus vault space) is too low for demand, and adjusts accordingly by increasing the award of coin to draw more people to supply vaults. This also has a downside of having a delayed effect, as the network cannot instantly grow to meet demand.
The third way would be some mix of the two, involving an algorithm to determine the trade price of SafeCoin->disk space in some tether token or directly versus USD (or whatever currency of their choosing) versus total supply of space on the network. This would allow the price to be more flexible, as they could try to, more or less, fix the price to some external force, but have the flexibility to make disk space substantially cheaper or more expensive depending on the availability of the network. Meaning, they could aim for a fixed 1 USD per 10 MB (in whatever that would be in SafeCoin at the time), but if there is a plethora of available space, they could lower that to .1x the going rate of SafeCoin to $1 USD, or vice versa if not much space is available. They could jump the price to 10x the going rate of SafeCoin to USD. They could then restrict or increase the rate of release of SafeCoin via farming to get the network back into some balance at a time.
I assume they will aim for an internal only mechanism, but I don’t see how they could account for the delay in response to increase the supply of vaults, or how the network could be aware of the fact that it’s disk space is way too expensive.
2 Likes
Storage is not paid for on a quid pro quo basis.
It hasn’t been coded yet, but the mechanism, as I understand it, is more or less as follows:
When one wishes to store a file, the client software checks to see if one has enough available PUT credits to cover it. If so, the total needed is deducted and the action is taken (store, etc.). If not, the wallet uses the users safecoin to purchase enough PUTs to do its business. The number of PUTs per safecoin is determined by the storage available to the network compared with the amount of already-stored data. This figure is calculated in the local address-region, with the assumption that, since storage effectively random, it will be more-or-less uniform across the network. (This is all algorithmically determined.)
Unused PUTs remain in the users account for further use, till they run too low to do the action they want, at which an additional chunk of PUTs will be purchased at the prevailing rate.
It is anticipated that one safecoin will purchase a large amount PUTs, so it should be cheap especially considering that storage data is permanent.
6 Likes
This would make it very difficult to build a business model arround the Safe Network.
If you sell thing you need to have prices that are more or less fixed. Small price fluctuations can be allowed, but not constantly. Customers like stability.
I am not sure I agree. Prices of commodities fluctuate depending on demand. This should be normal. A fixed price is depreciating asset in an inflationary world etc. All kinda fancy pants stuff, but tl;dr is that cost of something will be what folk will pay AND what folk will sell it for. The network will require people to sell space for the cheapest it can at any time.
So in good times, it will be cheap, in lean times, more expensive. This is quite natural.
15 Likes
Sure, but in essence, it is, you are just moving it a layer down. This still comes down to, will the network simply exist outside of any external influence, or will it be price aware versus a traditional fiat currency.
The existence of an exchange could make the price of SafeCoin skyrocket without the network itself having any awareness of this effect. Now, eventually it would realize and lower its prices for PUT purchases, but this will have a delayed effect as supply and demand works its magic. This means at any given time, someone who wants to desperately drop some data onto the network (think a new whistleblower) could be faced with exhorbinant prices to place their data because the network has not yet caught up to its externally affected price.
Thus having a completely closed system, irrespective of any outside controls, while possibly prudent, has some real downsides as well.
1 Like
sure … stability is the most important thing for many people … but nobody can sell stuff chaper than production price and it’s very common to have cycles … some years you get many oranges … some years only a few … so if something has a stable price that is because someone who is okay with fluctuating prices is buying it cheaper than you every single year and selling it to you with different profit margins …
4 Likes
I think you are making an assumption here - that Safecoin will be volatile to a similar extent as other cryptocurrencies. I agree in the short term this will be the case, but long term probably much less.
For one thing, the network itself may well provide a stabilising influence. Most coins today don’t have a very popular, widely sourced commodity that is directly linked to them. Think USD v Oil. Each stabilises the other. So given that data storage, arguably an even more universal commodity than oil, is going to be bought and sold by many who know very well what the price should be, the impact of speculation and so on in the price is Safecoin is going to be dampened considerably - why pay twice the price for something that you think of as data storage? You are likely to pause or find an alternative in the short term and upload to SAFE later when the price is more reasonable. Similarly on the sell side.
So I think Safecoin will eventually be more stable than people expect.
8 Likes
Will the network credit us if say an upload does not complete for some reason?
What if I have just enough to upload a file at current price. Then the price suddenly jumps above what I can afford and the upload cannot finish?
What if a bunch of nodes suddenly drop out before my file is uploaded making it impossible for the network to store it?
What if my connection is lost half way through the process of uploading a large file?
1 Like
I agree with you to some extent, I do think the rate at which the decline in purchasing power of disk/compute will settle into a more reasonable supply/demand situation over the long term. But if Safecoin were to make a steady increase to becoming the defacto transactional method for day-to-day activities, like some here speculate, and not to just purchase disk space or later, compute, the price of disk or compute would be constantly behind the curve. It would be more expensive than necessary, as the network is in an neverending state of catch-up on price of its commodities versus the exchange price.
That’s not necessarily the worst thing over the long term, but in the short term, it could kill adoption as its actual designed utility is essentially not usable at times because prices are in the stratosphere for resources due to influences external to the network driving up the cost of Safecoin in terms of fiat exchange rates.
1 Like
This actually seems like a business opportunity to me. There are businesses the world over that assume the risk of commodity price fluctuations in order to provide a more predictable and stable purchasing experience for customers.
Some people using the network will be willing to go with the ebb and flow of PUT costs, and perhaps balance these through providing resources (and we intend to provide them with the tools to make that straightforward and understandable).
But then there will be others who will be willing to perhaps pay a slightly higher but fixed cost, and not think about it. Hence the business opportunity, to cater for these individuals.
And this is what is so exciting; the multitude of different approaches, ideas, individualised solutions, and cottage industries that will spring out of this thing—and all to the benefit of the end user.
10 Likes
What if you let farmers(humans) set the price they want to receive?
Farmers can compare prices with competitors, know the Safecoin price on external markets, etc.
The network will start with including the cheapest vaults till it has enough space, and can then dertermine the average ask price.
Farmers may adjust the price when they are excluded, or try to get more when they are included. This will create a balance somewhere, and will result in a quite stable and competitive price I guess.
I think you’re forgetting the fact everyone and anyone can farm safecoin and that the more resources devoted to the network will ALSO affect the price of safecoin AND the number of safecoin distributed by the network. Say 1 safecoin cost a million dollars over night. That means everyone and his brother would try to farm safecoin. And they could! But that would also mean the value of that safecoin would go through the basement floor quite quickly because everyone could get ahold of it and since it was so easy to get it would no longer be as valuable. Why buy what you can farm yourself? But then the price would stabalize because the network would produce less safecoin. And you’ve also got to factor in the price of hardware as well because if safecoin costs a million dollars maybe it’s just cheaper to stash your data on a hard drive? That would mean less uploads and again drive the price down. See what I’m getting at?
1 Like
What if we weren’t talking about safecoin? What if we were talking about potable water or gold or land? Would your objection to fluctuating prices still make sense? As I see it the safenet economy will be more like a barter economy with trading of raw resources. One will need to know what the actual value of things are not rely on fixed prices. But if one is worrying about stability that I think will come from competition in values (price of hardware vs uploads vs farming rewards for example) rather than government regulation.
What if you let farmers(humans) set the price they want to receive?
Farmers can compare prices with competitors, know the Safecoin price on external markets, etc.
The network will start with including the cheapest vaults till it has enough space, and can then dertermine the average ask price.
Farmers may adjust the price when they are excluded, or try to get more when they are included. This will create a balance somewhere, and will result in a quite stable and competitive price I guess.
This isn’t really the model though… and would require a systemic redesign. The whole point is that by keeping the markets separated the whole thing acts in a way vaguely analogous to a PID controller. Expensive Safe = more farmers trying to earn it = more storage space. More available space = more puts/safe = cheaper network access. Further, many farmers will be quite rational about extrapolating future demand.
The key will be tuning the constants for the two circuits. Given 1) the overall availability of excess storage and 2) other network features such as the possibilities of deduplication, I think it’s rather more likely that the network will start with virtually zero cost to store and gradually move in punctuated equilibriums towards some sort of fair market price for the service.
6 Likes
The lottery and the safecoin reward mechanism could remain. The reward is then just the amount you asked. Yes this needs some smart distribution mechanism as the price someone paid is not the same as you asked.
StoreCost = FR * NC / GROUP_SIZE (source)
NC = number of clients which is unpredictable but relatively stable
GROUP_SIZE = 8
FR = farm rate = 1; farm rate is determined by spare space (source). Since it’s likely that farmers will maximize their returns that means FR = 1
This is the price of storage in safecoin. Then add exchange rates to USD etc for the price in local currency.
I think this is a really unsatisfactory answer, but it’s the best we’ve got for now.
There may be usd/safecoin futures to reduce volatility.
There may be people who exchange safecoin for putbalance in excess of their immediate needs to lock in their future price. They may then use that to provide a service of stable pricing to others.
People can wait until the price reduces in the future.
People can get cheap storage now before it rockets 100 times in value.
6 Likes
If you have enough PUTs in your account to store the file, you’re good. That’s already-owned storage ability. Your wallet purchases the PUT credits at whatever price is extant, then you use them. If you don’t have enough, you can’t store the file, and have to get more. Any remainder remains as PUT credits for further usage.
1 Like
The term “lottery” is misleading here. It’s just a way of saying that not every opportunity to earn a safecoin results in getting one. All this stuff happens under the hood, out of sight to the user. It’s all a matter of the vault just doing its business, participating as expected in network function. From time to time safecoin will appear in your wallet. The user takes no part in any of the factors except installing the vault and letting it do what it is supposed to do. Over time, the user decides if he finds it desirable to participate or not. If the vault is taxing his limited bandwidth or storage, he’ll likely stop participating. If he sees no downside, he’ll continue letting it run whether earning much safecoin or not.
3 Likes
Then it is no longer an autonomous network is it. Humans are controlling on of the most important control features of the network. And human greed will destroy it and few people will use the network because of it.
We are not after a profit making network where human greed and counter to that a race to the bottom conditions occur. History shows these systems either cause the product to fail or a 2 tier system occurs, the manipulators and the poor.
Simple - its based on the autonomous network’s requirements.
If spare space is low then PUT cost in SAFEcoin is higher and if spare is plentiful then PUT price in safecoin is lower. The Algo will determine the price at the time the PUT occurs. You PUT balance is not # of puts but what value is left for PUTs.
SO the beauty is it is a control system with built in feedback to ensure it works. Obviously it has to be tuned reasonably well.
Now as the $$$ price rises people will be encourage to farm because each coin they farm is worth more to them. This causes the spare space to rise and the PUT cost (in safecoin obviously) drops. So the $$$ price is high but the PUT cost in safecoin is low. Thus you can see that it evens out a lot. The $$$ price will fluctuate a lot but the actual $$$ cost of each put will fluctuate a whole lot less.
EDIT: I should read first. Beaten by a few people. The rfc uses factors that are not really there any more so the actual method of spare space will change and some tweaking could occur to encourage at home farming vs centralised data centre farming. But in essence the basics it seems will remain.