PoE at Internet is achieved by storing a file fingerprint (sha26) on the BTC blockchain. It is quite straight forward, though expensive, and leave the file as such outside. Nearly all users complain about that because they think the file itself ought to be part of the PoE . However The BTC blockchain doesn’t have space for that. (there might be other solutions I am not aware of)
So would not Safenet stored files be an ideal solution?
Please give some thoughts about this and correct me if I am thinking about the Safenet in a faulty way.
I’m not sure how a hash on a Blockchain proves the existence of a file that isn’t also stored there - perhaps it’s not quite as stated?
Regardless, I’m sure that you are correct to look at Safe Network for a better way of storing data such as files and being able to make definitive statements about that data.
For example, content based addressing means that anybody with the address of a public file will be able to access that file, know that the file cannot be altered and will be there forever. The integrity of the file can be verified by anyone because the address is a hash of the content.
The only thing that may need to be accounted for additionally to the existence generally is something adding up to a time stamp, which the blockchain hash provided. At least one can know that the file with that hash existed “at the time that block was minted.” Not sure how that would work on SAFE.
Edit: Could a sequence (time substitute) be determined by the proofs related to section storing authority?
It proves a file with specified hash existed at time the given blockchain block was generated. Less useful than actually storing the file, but i believe it has its use cases.
Thanks both of you. I expect there will be other pseudo time mechanisms on Safe. Time itself is a problem for Blockchain as well as Safe because there’s no way way to prove the time a block was created unless you keep track of this yourself, but that isn’t proof of time either.
It may be enough to know which block in a Blockchain, and something similar in Safe based in a fundamental characteristic of the network - section trees perhaps. Time itself is tricky
It Isn’t simple IMO, and I think it’s much easier to determine trust based on other factors so I’d trust those over time. IMO time would be a can of worms so best kept out of the core and implemented in other ways.
Perhaps later when something like smart contacts are being added above the core this could be a part of that.
If we consider time to be a progress of events (any event, a decay of an atom, an energy transition) then there is a psysical end of time in the universe when maximum entropy has occurred (according to one popular hypothesis).
Within safenetwork there are events during its lifetime, and those are what can form the safenetwork conception of time.
Multiple events with various granularity can be used.
But if we need to measure progress, and compare one instant to another, there will be some ways to do that (with various restrictions, implications and uses). It will just not be the wall-clock-way.
With that said, it’s absolutely possible that external oracles will try to find a way to map that time to our concept of time.
On Safe the file is stored at its various fingerprints.
File is chopped into chunks and the address of the chunks is their fingerprint.
The datamap that connects together the chunks is also similarly stored.
