Really, I am hooked. I have opened this thread in Reddit, but got no answer there.
I think the chances to deliver are small, but if the vision is implemented and achieved, bitcoin and other blockchain-based tech will get obsolete. So I am ready to place my bet.
I have a few questions though: How do you reconcile the fact that SAFE coin will be deflationary with a limited supply, while data is inflationary by nature ? It is estimated that the world will produce 7 zettabytes of data every year by 2020. What if SAFE Is successful and we come upon the 2^32 limit after a couple of years ?
The coin has to be divisible, because you can’t sell to the mass decentralized storage, if the quantity of storage is limited. I understand why we put a hard limit to the supply: Every crypto has to have incentives and reward early adopters, but I feel that when we reach the limit, the money supply has to be increased to handle capacity needs through some sort of mechanism (ideally proof of stake).
However how do you divide coins, when data is dependent on them ? The data will already be fragmented on closed group, so it will be impossible to devide the data into further bits.
This is the main issue I see from all my reading. if someone could address my concerns I would be very grateful.
Yes, I’m more than a bit baffled by the people who think SAFE Network could function as described and not be a HUGE success. I think we all have to have our own position on how likely it is that the final product actually works as we all hope, but IF it does… well, I don’t see how it doesn’t go viral and take over. The formula is perfect and it’s desperately needed by everyone for a million different reasons…
I’m more optimistic than you that it will work, although some sacrifices and compromises to the functionality we think we’ll have are still possible for sure.
Enjoy reading, there’s plenty to get your teeth into and I can’t possibly do any of the proposed solutions justice with my half-arsed understanding of how they work. There are various possibilities though and it’s worth reading the links.
No you cannot if the quantity of storage is limited. That is why the storage cost algorithm is dynamic and the cost of storing data depends on the amount of storage currently available. And storage is added when more farmers come online. One coin does not buy a set amount of storage but rather the amount is determined by the algorithm and when you use up your balance that the coin bought.
In other words even without divisibility the coin is divided by the storage algorithm and storage subsystem. So you spend one coin and the network gives you a “PUT” balance. “PUT” is the term used for storing one block/chunk. And when you store a block/chunk the network reduces your “PUT” balance by the current price for storage. One coin usually will go a long long way when storing.
The reason we need divisibility is not for the farming/storing of data, but because (we hope) the price of a coin will be going higher as time goes on and the only way to do micropayments to/from other people will be divisibility.
Yes you are and actually its not “needed” but the benefits of having it for farming/storing will be great. I was also trying to contrast the divisibility issue so that the OP would be able to understand things better. And what you say is very valid too.
The network will hum along nicely even if we didn’t have divisibility of the coin. BUT we humans will be frustrated when the price goes high without divisibility.
If the price of safecoin goes high, will attract more farmer with more space and the storage price decrease. The possibility of a high price and limited space is, from an economic point of view, contradictory.
And you forget that the safecoin is an recyclable coin. You destroy the safecoin with PUTs and create them with GETs. In fact the supply is unlimited.
I know you understand this, but to clarify in case anyone gets the wrong end of the stick - technically the supply will be limited to a maximum quantity, but there will be a constantly available stream of coins to incentivise farming due to the constant supply of coins being received by the network from users putting data onto the network (the recycling of coins as you say).
So coins will be flowing constantly, but their quantity is constrained to a maximum limit.
The amount of storage is determined by the number of farmers and the sizes of their vaults.
As the spare space reduces the rewards to the farmers increases which attracts more farmers thus more space and the PUT cost drops as does the farmer rewards. It becomes a balance between farmers and prices, there will be a point where rewards are high enough for the spare space.
There are a few topics on this very subject, and a search should help you here.
Since the design is for farmers to use their spare resources there is almost zero cost to them, so any rewards is a profit for a lot of farmers
Coins are not tied to data in any way - they’re just a payment required to upload data to the network.
Coins paid to the network in order to upload data are recycled to pay farmers - they are not held against the data that was uploaded. The coins are free to flow back to farmers as soon as they’ve been paid to the network - deleting data isn’t required to free up coins to flow.
As the network store new data they destroy Safecoin so this Safecoin are available for farmer. The only possibility that no more safecoin are created is that nobody uses the network to put new data. In that case, of course, the network is a failure and the price of safecoin must tend to zero.
There are other very little possibility, exactly the opposite, and is that the safecoin, as currency, is a resounding success and everyone wants to save and use it, exclusively, as a transfer of value. In this case the network would only serve as a support for the safecoin that would become the new bitcoin.
I’m not sure if this is impossible or if it makes sense at all, maybe someone with expertise in finance, numerical analysis and simulations could help me understand.
I was wondering if it could be possible to make a simulation of the economic model of the SAFE network to try to understand, verify and/or confirm how all this dynamic would/could behave?
If it’s possible it would be a nice topic for an academic degree thesis I would say.
I did about 18 months ago, but things change and the code was too immature to use with the new parameters and it would be better to rewrite it (reusing code where possible). But also it showed that it will be difficult to do proper simulations until the code base and interactions between data storage systems is more established. For instance the change from SD to MD is quite a large change in terms of simulation. Especially since we don’t even know yet the charging model that is going to be used to pay for creation and for updating MDs