One thing worth mentioning is that Safe will NOT be “printing money” in any sense of the word long-term. Yes, there are going to be more SNT than MAID by as much as 10X. But once that transition occurs and the number of SNT stabilizes the Safe economy will run with mild deflation as the value of a SNT will go up due to fixed supply. This will be very much like a gold standard where the total gold supply goes up slower than the economy grows (mild deflation rewarding savers).
This is such an important point. We sorely need uber-popular sites with heavy data content like Wikipedia and the Internet Archive to be early uploaders to the Safe Network. @JimCollinson indicated MaidSafe is on it, though, so I take comfort in that. Would it be possible for sites like that to upload to Safe Network and a copycat? Not likely, imo. Maybe a contract is in order.
There is 450 million MAID and can be upto 4 billion SNT.
SNT is recycled because every coin used to buy resources becaomes available for farming rewards.
So over 20 years there could be 40 billion or even 400 billion SNT farmed since there would be similar amount spent to buy resources. By having a buffer the rate of spending and farming can have large changes without the need to have 100x or 1000x or 100000x price differences in store cost (SNT amount) and rewards every minute or two depending on haow much storing is done.
This is such an important point that it can not be stated enough.
Only public data can be copied over to a new fork. But can you imagine the effort to copy over that much data just to try and become a viable fork. Getting as much public data on Safe will be so important early.
The problem is that farmers do not decide on what they store. The uploader cannot know which farmer will store any particular chunk. Its the design of the protocols. Thus its not viable to let the uploader know of farmers offering discounts.
The the other more significant problem is that farmers can only offer discounts on GETs since that is all they are paid for. They charge ZERO for PUTs into their store and the network decides who stores the chunk.
The software is open source, literally they can instantly copy any improvement we make…
Yes, I wrote it at the beginning of the topic of vampire attacks, although at the time my main concern was the lack of access to the liquid market and the claim of some members of the community that we did not need such access at all and that everything would be fine without it - this terrified me then.
I completely agree with that. What we don’t know is how long it will take. Currently, the common opinion in the forum is that the Safe Network will be like a revolution - we are launching and conquering the world in a few years. We don’t know that.
Look at Sia - they have had a working network and trade for millions of dollars for years, but the network use is minimal. Look at Filecoin - the ratio between trade and usage is just as bad.
If the same thing happens in the Safe Network and the usage is low, we become vulnerable to the following:
- large super farms line up to enter the network and gradually displace small farms from home because they have a more stable uptime and due to the initial low use of the network their influence in the network remains predominant for years.
- this opens up an opportunity for a corporation like Filestar to offer a copy of Safe with a better economy - whit a scheme like this:
- Filestar is also preparing to launch a staking program that would enable community members to earn the network’s native STAR tokens, cast votes for miners and participate in “important decisions for Filestar’s future.”).
If these super farms move, it can destroy our network (data loss) and we lose people’s trust.
That’s why we really need simulations for the growth rate of the network and plans to introduce if the growth rate does not meet our expectations.
That is a good analogy. And the real estate market has a huge buffer and why it works even if its profiteering. Safe will not do the profiteering but the dynamics still are there.
As Store price increases over time (not minutes) people will put off storing non-urgent things more and more. This increase in store cost reflects the decrease in available space (as indicated by some Nodes becoming full). Like housing there will still be some buying while prices at the peak, but those who can put it off will do so. Then as the store cost drops then people will upload what they put off.
Its an interesting concept but from what I understand SAFE will have one of the only network based demurrage systems in play in the entire space. It will be interesting to see how the network handles the costs associated with the space vs how users perceive that value in a real (trading) space against other currencies.
Which is why Safe should eventually become a mere open protocol with a core team implementation.
This is one of the reasons why I argue that the imposition of domination of the Safe Network worldwide will be a process similar to that of Bitcoin and will continue for decades.
They can literally only offer a stacking program outside the code like Filestar or to allow farms into the network only after they have carried out a KYC…
What does this mean?
I googled it and got an answer thats its a fine for holding cargo in a loading bay for to long, so I still don’t understand.
I’ve seen the term used before, here I think, but dont understand it.
I wasn’t deliberately understating anything, but sorry if I wasn’t clear.
I wasn’t suggesting only 5% of people would be away from their computers, but that a 5% dip in StoreCost could be a trigger where people / companies could have queued lower priority uploads ready to go to take advantage of such situations (and even lower priority jobs queued at 10 / 20 / 30% StoreCost dips), which together would all limit the drop off in network demand even if everyone is away from their computers for a few hours (or even days) in your scenario.
I also don’t see a big problem even if farmer rewards dropped for a relatively short period, e.g. a day or two out of a month, as the average over time is what will likely matter to them.
I’m suggesting that there is likely to be significant automatic buffers based on user and farmer behaviour that mean there may be no / little need for a built-in network buffer. I’m genuinely interested to hear reasoning for where a buffer might be needed, as I don’t currently see a problem with having a pretty raw market between farmers and uploaders.
I don’t understand how the StoreCost could change significantly minute to minute if it’s based on the proportion of full nodes. How could a network with e.g. 50% good nodes see that number change significantly in a very short time?
Of course it would, but that sounds like an impossible occurrence on a network of a fair size (anything except newborn) with sufficient spare capacity and a StoreCost based around a market price that matches supply and demand.
Anyway, I agree, this is fruitless as we don’t get each other’s perspective on this (after all these years!).
Real testing in live networks is where the reality will come through vs what people thought might happen. I’m looking forward to seeing that.
Both inflation and demurrage reduce the purchasing power of money held over time, but demurrage does so through fixed regular fees, while inflation does it in a variety of ways
The theory is that if you keep Safe tokens, theirs value will go up because you will be able to buy more resources with 1 safe token over time. That is why there is no demurrage in Safe.
It is needed against vampire attacks and social attacks against the network, for example, religious leaders suddenly tell millions of people professing this religion that Safe is evil and its use is a sin.
That was not the way I imagined it, I will try to see if I can put it in another way.
Let’s try some form of step by step:
Let’s say elders ping farmers every second to see if they are Still there.
Farmers reply to elders that they are online and price for storing data.
A client connects to an elder and want to upload data, it asks elders for farmers price of storing data.
The user had at earlier moment set minimum/maximum price for storage. The client automatic randomly chooses prices within the range of limits and then reply to elders that it want to store at prices:
and so on
Elders sends message to farmers reserving total 75MB at selected prices, to farmers x1, x2, x3.
Client sends payment to elders shared group wallet.
Elders redirect the client upload selfencryoted/chunks data to farmers.
When farmers have stored the chunks then they get payed by the shared group wallet by the elders.
I know this might not be possible and you know much more about what an work and how the routing works and so on. But I hope this discussion as a concept can bring something interesting.
Do you guys know if there are simple illustrations and text that shows the process step by step on how it looks from a network perspective on buying and storing data? It don’t have to be fully technical correct as long as it shows the process in a simplyfied step by step way.
Also If you guys have the possibility and interest look at my previous post, I might had some interesting thoughts comparing a client buying data like a gpu allocates Vram memory to be able to deliver what a game needs for every 30-60 seconds gameplay.
I hope you guys have a great weekend.
Bravo, @Antifragile! You are my new hero! Couldn’t have said it better myself.
I can see how a big exodus of farmers would be a problem. But if that were to happen for any reason, how would diluting the token supply limit the harm of such an attack?
Have you read the topic of vampire attacks? In short, a copy of the network can steal our farmers, if we have a buffer of tokens we can temporarily increase the payment to farmers and keep them in our network.
My assumption is that people who are farmers will earn very little over their costs, but will speculate on the future value of the token, ie. when you are a farmer it is like buying a token at a discount. Bitcoin miners speculate on the same thing - they believe they can get more bitcoin as miners than if they buy it - certainly no one is a bitcoin miner to protect the bitcoin network
In the Safe network it will be different, for example I will be a farmer even at a loss and I do not intend to sell tokens, but only to use them. The question is what will be the ratio between people like me and between people who are just speculation farmers.
I haven’t in detail, but that makes some sense. However, if the appeal of a ‘vampire’ network were harder monetary policy (less token dilution), I doubt many farmers would be persuaded to stay by having an increase in token issuance. I’ll need to have a look at the topic.
I think this is likely, and why that would make you concerned about token dilution combined with vampire networks (which I think you are if I picked that up from other comments?).