What would be the point though? The cost of copying wikipedia to safe would mostly be a one-off large fee + small cost maintain … if you host your own version of Safe, then you are going to have costs as well - in terms of managing and marketing (to potential farmers to donate) … there are always costs involved and a smaller network will very likely have an overall higher cost because of economies of scale.
I can’t argue that, governments don’t have to be efficient as they use the gun to get their money - taxation. But such things aren’t really competing with the original Safe Network.
There will be people who will not want to provide a resource to a network where porn is stored, but would donate a resource to a partially regulated Safe Network (modified to work only with Wikipedia)
Everything that takes a computer resource is a competition of the first Safe network, if this computer resource can be in the first Safe network
This is some great precedence to work with, thanks for the links! Looks like we should not be surprised if up to 55% of the network resources depart suddenly for up to 55 hours… I’m pretty surprised and impressed by those numbers, especially considering bitcoin was quite well established at this time, it wasn’t exactly a newcomer.
The part that interests me is it took 53 hours for resources to return to bitcoin because that’s how long it took for the difficulty adjustment in the alternative network. If the difficulty had changed faster or slower presumably this would have been a shorter or longer movement in resources.
It’s also interesting how the ‘real world’ financial incentive (rather than theoretical / instantaneous incentive) to move resources turned out to be quite small (a lot of resources moved despite being only a small gain, so seems like it could happen quite easily):
I do not know, so I raised this topic. There are much smarter people here than me to discuss the topic.
All I know is that we have not discussed the economic attacks enough, only the technical ones. And these are just economic attacks on the bags of the current holders of MAID. As a Safe Maximalist, I am not worried if I lose everything I have invested, I am interested in having a Safe Network in the world.
EDIT: In fact, I have some ideas on how we can defend ourselves against such attacks. If we have a Foundation that receives part of the inflation of the currency in the network. This Foundation can use this money to protect the interest of all token holders and to promote our Safe Network.
A few specific examples on how can a Foundation protect us?
A new Safe Network appears, which pays twice as much (a third party subsidizes this network). Our Foundation can use its funds to temporarily increase the payments to our farmers to be competitive against the new Safe Network.
A new Safe Network is launching Google ads to attract new farmers. Our foundation may also run an ad that precedes the ad on the other network.
A new Safe Network creates low liquidity in a decentralized exchange to manipulate the price of its token. Our Foundation again can increase farming payments to make our farmers more safecoin.
Different difficulty retargeting approaches have been attempted for many years. I remember KGW for example tried to retarget every block. I haven’t followed super closely, but as far as I know none of these have turned out to be stable or secure (time warp attacks, etc). Bcash’s EDA was a disaster at launch (fixed now?
Don’t really hear about it anymore…).
I’m not so sure there really is an element of the design that contributed to the resilience here. The mining death spiral is still a concern. I’d be inclined to attribute the resilience more to “economic inertia”.
Also being the dominant coin for a given PoW also is the best protection. Forked chains that do not change the PoW are inherently vulnerable (see recent ETC 51% attacks).
That is why we need economic momentum too. In order to have one, our currency must be traded on the largest decentralized exchanges as soon as the network is launched…
Foresight is a wonderfull thing, but we don’t have it.
No one would have known how much hash power would have jumped across, so also not know how fast the difficulty would adjust.
Also the fiat gain maybe was minimal, but the actual coins substantially more.
Another thing to keep in mind is just how centralized bitcoin mining is. The BTC/BCH fork marked a great schism in a fairly concentrated pool of miners. In a world without truly distributed resources (I.e. limited decentralization) such quick and dramatic shifts transpire easily.
Those shifts correct easily as well. Ideology more so than route profit maximization may have been at play with regards to miners shifting resources to BCH. When those miners realized that this wasn’t going to be an ETC/ETH situation, they quickly turned back to BTC—a profit maximizing decision.
At the end if the day, so long as SAFE reflects farmers’ value-based (rather than simply cost-based) perspective to set the price/reward for PUTs, I think these concerns about competing networks are largely moot. The price of a PUT should balance what users think secure and private data storage is worth with what farmers believe supplying their resources is worth. The Network may need some mechanism to take into account the appreciation of Safecoin price due to multifold use cases (i.e. how does the reward amount change if the price of Safecoin goes from $100 to $500 due to the emergence of some new SAFE dApp?).
All this chatter about multiple networks looks like a solution in search of a problem but even in the case of a corporation or city wanting a network they control, it doesn’t make a lot of sense - it wouldn’t be stable in the way the wider network assures.
The reason for so many copy coins in cryptocurrency, was and still is cowboyz wanting a quick profit, without interest in the product or assuring any utility going forward. Those are for the most part just a scam. I’ve yet to see one cryptocurrency deliver in a way that necessarily will last… even BTC available to mainstream users, is not 10years old!.. and its prospect is not stable wrt energy… it’s just speculation; as is this talking up separate networks.
In theory data could be challenged from one network to another, siphoned off or made use of but the cost of that prohibitive; there’s no benefit for a separate network - except to the owners - and then the stink of corruption about why they must be the ones in control… politiks is rot. I default cynical of anyone wanting to spawn new currencies or control resources without hard arguement requiring it.
So far I see little prospect for separate networks beyond novelty and perhaps testing of new features… some subnetwork that is dev or cert, might be useful?
It’s not going to be so easy, likely not economically rewarding, for farmers to move to a competing network. Two most obvious Safe N design attributes that help there are 1) node age and 2) rewards being paid only upon node’s move to another section after having performed its duties to its original section.
Edit: “not easy” is the wrong way to put it. But potentially expensive and to those wanting to depart but also want payment for their work, slow as they must wait.
I would frame this a bit differently though. “Economic” has too narrow scope. I would call it as “human behaviour” attack, or just behaviour attack for the sake of shortness. We don’t actually care if people leave the network for economical incentives of another network, or any other better use of their resources. Just them leaving is the problem, no matter what causes it. It might as well be some big smear campaign against the network. Then people may want to leave, because they don’t want to be part of “that”. (This of course goes off-topic right now.)
On the other hand, they may also choose that they want to part of “this” even if they would get better money from some other network.
Basically I argue that the choise to farm or not, has other significant factors, than just pure economical gain. All these other factors can make the network stand or fall, working in conjuction or even against the economical incentives.
I would say that at least one of the most significant factors is if the network is seen as “our” network or network of “them”.
But no matter how we turn this, the loss of farmers is something that can happen. Maybe not for economical reasons alone, but for any reason. The question is, how much shrinkage can the network handle? How can the mass run of farmes be prevented? How can the resililiency be increased?
One failure scenario in my mind would be if network grows on the resources of many naive farmers, who think that Safe Network has only “innocent” content - and the they leave after a smear campaign pointing out all the creepiest stuff they help keeping up. This could happen at any point of networks history, before people are educated enough to know that they give resources for all that stuff too.
It seems that people not participating is not problem, but people changing they mind is. “We” cannot have “our” network, if large enough group of us decides to not want to be part of us anymore.
And in the scenario you describe, a Foundation that gets some part of the inflation will be able to take decisive action against such a campaign aimed at discrediting the network.
This is going to be a very interesting dynamic to watch and study: the relationship between market price of Safecoin and reward amount. I expect it will be very fluid in the beginning and tend to stabilize as time goes by. Brings up an interesting question: Could a large group of farmers, if they band together (a “pool”), influence the market price in a predictable way?
All going well it will be fine. So a farmer will farm for say $5 per day, this might mean the reward is 5 safecoins one day and then next it might mean 3 safecoins. The network will try to continually reduce $ amount rewards until farmers start to leave. AS the network needs more resources it will increase $ amount rewards.
The $ amount is purely a function of the market value of safecoin, but the network does not know that, it just knows if I pay 3 safecoin and farmers leave I will increase to 4 safecoins, then 5 etc. until farmers join again.
So while the network does not know $ it does know safecoin numbers required in rewards for farmers to farm.
The more interesting point I took from VaCrunch’s post there, perhaps misreading it, was the prospect of cost of different services fluxing… unclear if price might flux as different services affect pricing. Look forward to finding out as it evolves in the real world.
But the market might very well know what the current farming reward is. I am wondering if the market will constantly adjust depending on the reward payout to farmers. Will the two have a symbiotic relationship?
I wonder how quickly pricing information will travel between the market and the network itself. On the one hand, I suppose one determinant will be how quickly farmers start joining the Network if the market price of Safecoin spikes, say, 100%. On the other hand, the question of Network sensitivity remains, i.e.
will the adjustment in rewards be more continuous or discrete?
in what increments will rewards be adjusted?
how much time does it take for the new rewards amount to come into play?
Safe certainly requires a much more complex and dynamic “algorithm” than bitcoin.
Not so much controlled at all by farmers as it’s set as a reward that is balanced by the amount Clients are paying to mutate/store etc. If we ignore initial payments (the start up algo) then the farmers are getting paid relative to the costs charged to clients to store.
So when we need more farmers costs to clients increases to give us the cash to pay out. As the rewards decrease then cost to store decreases for clients. This is the constant balance we are trying to achieve. As clients will vastly outnumber farmers the the cost differential for clients is likely very small, but at the other side (farmers) the rewards differential is much larger, just related to size of both groups.
These are the key points we need to test in beta. The network should react very quickly to price changes (likely measured in minutes, possibly seconds). The increments need tested for sure but should be quite quick as well, this may be good to be slower, we cannot tell just yet. However the reward will come into play almost immediately as the rewards alter.
When we get this part in play we will simulate it to let folk see and play with parameters, except the main parameter of course which is human behaviour.
