Quote from: niner on May 10, 2014, 05:00:03 AM

Quote from: MrBig on May 10, 2014, 12:07:59 AM

Quote from: MrBig on May 05, 2014, 07:27:36 PM

Hi there I got a few questions about safecoin.

I was reading about it and what I understand so far is that it's proof-of-resource, so basically farmers gain coins by providing storage and processing power on the network for the benefit of other people. Now bitcoin is proof-of-work, so the miners get a percentage of the coins based on how much work they contribute. As we know, the value of bitcoin goes up and down due to speculators and other reasons. When the value dips a lot and mining isn't profitable, people start taking their miners offline reducing the total hashing power.

Now let's compare this to the proof-of-resource model, say safecoin has 1000 terabytes of cloud storage and the price drops to a point where it's not profitable to keep farming machines running so many take them offline. Now let's say that 950 terabytes of storage was being used to store data, but after a lot of farmers went offline there's only 750 terabytes available. What would happen to the excess data? Is a scenario like this possible? I'm still trying to understand this whole concept as it does seem a little confusing.

Also are 51% attacks a possibility on proof-of-resource networks? It would seem a lot easier for a giant corporation to gain 51% of the network I guess.

Can someone with more knowledge give their insight please?

Thanks

Bump, any insight would be helpful, thanks.

I believe that having 51% of the storage/cpu/bandwidth resources would give you 51% of the farming income.
Although I don't see how that would allow you to double-spend safecoins or mess with other people's data.

It is not mentioned as an attack vector in this topic: https://groups.google.com/forum/#!topic/maidsafe-development/fFY34cKXIMs

If I understand your first question, the network would need to have enough redundancy to handle a certain percentage of farmers going offline.
Each shard of data would be stored on a large number of nodes.

Caveat: Just my thoughts after reading the white paper a few weeks ago and having forgotten most of it.
https://github.com/maidsafe/Whitepapers/blob/master/Project-Safe.md

Regarding my first question, do you think it would be cost-prohibitive to provide such storage redundancy when compared to centralized cloud-based storage services. I'd imagine that it would be cheaper to have one online instance of data with an offline backup.

Thanks for the link, it did answer a few other issues I was wondering about.