I don’t really understand what you’re saying, but it made me think anyway.
As I understand it, the initial idea was to have some algorithm determine how many safecoins someone gets for providing storage to the network. I’ve always been suspicious of that. Algorithms are pretty bad at determining the value of anything and I believe there needs to be a more subjective component. All value is subjective. I value things differently than the next person and that’s why central planning always fails: One person or committee decides the value for a whole nation. That doesn’t work.
How about an auction system instead?
When someone provides storage they set the price of that storage. They say how many safecoins their HDD space is worth. If their HDD isn’t filling up quickly enough, they need to lower their price. If it fills up immediately, they can raise the price. This allows individuals to respond to the market. The more individuals we have the more smooth this whole process will go. People will figure out how to optimize this.
The danger that someone may fear is that amazon/google/NSA will create a huge server farm and sell out their space for cheap, thus forcing the network to work mostly on their own servers. Once most of the network relies on them, they can just shut it down and effectively destroy it.
My response to that is:
- I’m not convinced that a company would do that. If the network is small, they will not be interested. Once the network is big, it will literally cost them money. If they sell their space for under natural market conditions, it will cost them. They’ll farm safecoins, but not enough to make up for the expenses they incur, running those servers. If they offer their space at market rates, their storage will be worth as much as the storage of everyone else and the network will be distributed relatively evenly across farmers.
A government actor also has budget constraints. So let’s say the three letter agency realizes that SAFE is blowing up. They create a gigantic server farm. I saw on a quora post that facebook spends more than a billion a year on all their storage. I think it’s safe to say that it will take SAFE a while to get to that capacity, and let’s assume the state actor decides to take SAFE down before it reaches that level of use. The NSA has a budget of around $10 billion. A billion dollars is not that easy to just tack on top of that to destroy some exotic new technology, unless there’s a huge public outcry about its danger. Maybe $10-100 million would be possible.
In this case, MAYBE there’s a chance they could pull this off. The network isn’t in its infancy, but it’s not mature either. It keeps expanding ever more rapidly. This is when they need to flood the market with cheap storage, let it get filled up and then pull out the rug under us, thus destroying it.
My thoughts on that would be:
How many times can they do this? There will always be a market for people to use SAFE. So to some extent it will be impossible to kill. For mass adoption, they probably have to do it only once because a lot of people will be put off once their data disappears. So that sucks.
It’s not clear to me that they would win the price battle. If the network becomes popular and especially if the bitcoin people find out about it, there will be a lot of interest in farming and most enthusiasts probably don’t mind donating their space below market value. You will still have the incentive to earn safe coins so that you yourself can upload data. If the market is flooded with cheap storage, more people will have access to storing their data. It will take less for people to earn the safecoins to store their data on the network. I think this is very dangerous for a state actor. They may effectively subsidize the whole network and lead to an exponential increase in its userbase. Many more people may end up creating accounts and donating their own space to upload their data. Whoever at the three letter agency would come up with this idea would be smart enough to think of this as a possibility as well. They may decide that it’s cheaper, less risky and more worth their time and budget to run a propaganda campaign against the network.
I’m sure the developers are planning on doing this: I’m curious to see how much of the network can be switched off at once without affecting its ability to work. Obviously, safecoin prices will skyrocket, but as long as the data can be accessed, that’s not a problem. Depending on the result of those tests and subsequent optimizations, it may end up to be way less easy to pull off this kind of attack.
Finally, I’m still a proponent of the unpopular idea of adding some cost to pulling data. Farmers could set their prices for GETS, thus competing for an audience to download things. We won’t have to worry about paying it forward indefinitely with the pay once-store forever scheme. People would have more of an incentive to actually provide a bit of space to the network. I understand the barrier to entry will be marginally higher, but maybe that’s a good thing. We will have higher quality users. People will actually be more committed to using it. Initially, there won’t be much content. You’ll have people browsing the network, not seeing anything and letting it go. If you can get them hooked to first provide storage before looking at anything, they will have made a bigger commitment. If there’s no content, they may still be content letting their node run in the background to see if/how many safecoins they’ll earn. I remember the first time I got an e-mail address, I signed up with a provider that would physically mail me my password. It took a few days, but that’s just how it was. If a new user needs to wait a bit to earn some safecoins to look things up, that’s not necessarily a bad thing. Also, the more the network grows, the easier and quicker it will be to earn safecoins. It may one day take as little as a minute to have enough to start browsing. GET requests will very likely be dirt cheap, because most of the cost lies in the PUTS.
End of rambling thoughts