At first when uploaded, all data are equal. Popular data needs to be cached which would give popular data priority and the burden spread across more nodes.
The caching system will be very important. Let’s say pewdiepie uploads a video and millions watch it the first day, 8 vaults will not be able to deliver the bandwidth unless the content is cached. Hopefully vaults will be paid for gets, puts and cached data, atleast for puts and cached data.
Completely agree. At the same time @gan_rl is posing some excellent questions that will need to be tackled overtly sooner or later.
If it turns out at some point in the future, that upload costs somehow become too great, solutions can be implemented. There’s no way to predict how popular some data will be and if we want to preserve data indefinitely, only the pay once per put model works.
There’s already been some suggestions for solution in thread, after a certain amount of GETs, GETs for that specific file could incur a cost or someone has to pay a new PUT fee so the file can be seen X number of times again for free. I think this would probably be a fair model. If someone uploads a popular file and is making money on it, that person, or someone else, can pay once for every million GETs or whatever to keep the file available for free, but if nobody pays the file is still available just with a small GET fee unless somebody pays to make it available for free again. There’s other possibilities also I suppose, but uploading should still just be a one time fee where the cost is the same for anyone who uploads.
In any case, it’s unlikely to be an issue in the beginning and it’s quite possible it won’t be an issue in the future. IF it turns out to be an issue, there will at that time be much more data available to decide on what solution is reasonable.
I differ here. They’re familiar free market idealism, which we’ve heard several times and particularly recently (um, what could that mean?). I don’t see a need to address them (in code) until we have something to test, so it’s fine if they are genuine, but when the ‘questions’ repeat regardless of the response, and become ever more idealistic, it’s time to move on.
This reflects that more resources are needed, not that GET needs to be discouraged. High GET rates should (in my opinion) have no impact on the cost of PUTs. Potential misconceptions of the SAFE economy are just my opinions (ie not any sort of agreed facts or planned design or coded implementation) but to me it makes intuitive sense that a) storecost is only used to rate limit uploading when needed b) reward is only used to bring in more resources when needed and c) membership rules ensure permanence of data.
Point b) is related to high GET rates, and as GETs increase it’s the reward lever that is needed, not the storecost lever.
Hope this helps you understand a little more about the possible mechanisms.
This is how I think about it:
The goal is not to make sure that each individual action (GET, PUT etc.) on the network have their associated true cost. The goal is to give the power from the current centralized entities on the Internet back to the users.
There are a couple foundational ideas of the network:
What do we have left? PUTs. And so really, the only time you CAN charge someone for the data on the network is when it’s being PUT.
The general idea at this time as I understand it is that you pay X amount of coin to put the data, and then that coin will be distributed to farmers over Y amount of time.
The cost of PUTs and the amount paid to farmers will be automatically adjusted by the network so it won’t get overloaded in some way.
The network gives a ton of power to the users because right now, without the network, I literally can’t do what the huge corporations are doing with their data distribution networks, because I’m limited financially. Especially if I want to do something for humanity, and not for monetary gain, it will just be impossible.
But on the SAFE network, I can upload some data, be it a text document, or video or whatever, that I don’t earn anything on, I just do it because I want to do something good, and I pay a one time (hopefully) low fee that I can afford, and the data is accessible for free for everyone forever. Or I can just put my own private stuff there for me or whatever.
And as a developer I can build a video streaming service or some other huge thing like that, and not have to worry about the cost of a huge global data distribution network. The SAFE network takes care of that, and the cost is moved to the people who upload the data, and I can build my service and compete with huge corporations if I want. Again, this is impossible without SAFE.
SAFE works in a fundamentally different way from the current Internet and solves different problems than the current Internet.
So that’s what SAFE network is for as I see it, it doesn’t care how much money you can earn on the uploaded data, that’s not even a consideration. It only cares about giving power to the masses and setting data free.
Yes, this is totally different from the current web. If you want it in the way the current web works then that already exists. Not only in the form of the current web, but I believe there are decentralized storage networks that work more in that way too.
In my opinion because it takes some power away from those with more money and gives it to the masses. But to me the fairness of the network isn’t a measure of the network’s success, a better metric to me is how much good it does for humanity.
As elder nodes will know the price for latest put’s, they could do an average and calculate a get price? let’s say 1/10 of a put.
(If get’s will be implemented to have a cost)
But depending on when the latest put’s have been made that might affect things and or if 1 section talks to other sections.
Get rewards and Put prices are updated independently in real time. The network also has a buffer to support the loose coupling. Too many gets and not enough puts would see the get reward drop more relative to put charge until vaults start to protest. The network buffer can supplement if deemed necessary at that point. Simultaneously, Put costs would adjust to maximize income for the network. When puts are plentiful, put charges rise as high as possible until it just begins to hurt demand and the network buffer refills. It’s a dynamical economic system between clients, vault operators, and “the network”.
I agree, popularity and dominance are the main point for Youtube. But I think it’s not only that. Dtube is very slow. Users today are impatient. If their experience was not wonderful, they will never return. We cannot deny that YouTube has incredible video pre-processing technologies, caching, resources, speed… It is difficult to compete.
Careful Dimitar, getting dangerously close to socialism there. Better keep a tight rein on that mythical horse!!! ;->
Localism is the word you are looking for. Everyone is ready to help their loved ones and close community without thinking. Everyone needs very good reasons to help people on the other side of the world… Look at how difficult it was to find Patreons for the international SAFE forums and sites …
I’ll be honest with you. I’m talking about survival. Virtually every aspect of the SAFE Network has so far pleased me. But this economic model of compensation between PUTs and GETs seems to me to be an extremely dangerous point of failure for the entire ecosystem, putting everything at risk. I noticed that this subject has already been discussed here on the forum before, I am reading, but I have observed that there is a lot of mixture between subjective idealism and technical rigor. The MaidSafe team seems to be very competent, and this community is full of smart people. But this project is like a space rocket. Even with many smart people working, if a small aspect went unnoticed, everything can explode. It is not that I dislike a principle of the project, I am concerned with the fragility of a very relevant point.
I like Bitcoin, although the concept of computers wasting energy to find random numbers bothers me. In that case, I can say that I prefer other models, but it doesn’t mean that Bitcoin will collapse because of that. It’s just my preference, others will argue that this energy expenditure is necessary and justifies everything.
The discussion we are trying here is different, this economic model does not seem sustainable to me. I may be wrong, I am not the greatest expert. But I am a person who came from outside, with a neutral view on the project, studying independently without bias of already being part of the community, and like others, I got stuck in this part of the economic model. I want this project to be successful, sorry if it looked anything different.
Tha might be cos everyone is skint right now. Had I been working full time I would have sent you a few £. As it is you will just need to make do with my BAT tokens – which are due to be paid out in a few days 
The ideas you seem to prefer are not sustainable, but perhaps you haven’t noticed. Nothing is in the long run, it is all evolution, and SAFE can and will make good and bad steps along the way, and be part of the overall evolution of these systems. Bitcoin is not sustainable, nor ETH, and certainly not Google or the style of capitalism that dominates at the moment. We need change, and SAFE’s main contribution to this in the short to medium term is not going to depend on whether its first implementation of resource pricing is tweaked this way or that, but on it filling important needs. So don’t worry. If those are your concerns, don’t expect the network to get those things ‘right’ (in your view) first time. Wait for the test networks, help us examine these questions and refine the solution in the face of any issues that we find.
Likening this to a rocket that could blow up is emotive, not advancing the debate. It certainly might not work first time, but it isn’t going to kill anyone by ‘exploding’ and being a new kind of software system we expect it to have bugs, to need to be refined in the light of unexpected behaviour. By all means join in that process and please don’t take refuge in scare mongering.
We can always find excuses for not helping. After all, we live in a world where we allow children to starve to death…
Thank you for the donations, I collect them and take care of them until the community decides to spend them on something. If you’ve seen the last week update of the SAFE world domination project, we already have 3 Patreons 
We’ve discussed the edge cases in previous threads. Please articulate exactly what you see as the problem.
There will be a natural GET to PUT ratio averaged over all chunks. At the most simplistic level, if the price charged to clients vs the reward paid to vault operators reflects this ratio then there is a sustainable balance.
Mine is the wife reminding me constantly about the mortgage,
Less children would starve if greedy bastards would pay their taxes instead of indulging in a libertarian circle-jerk about it somehow being “robbery”.
I fully agree. This reinforces my point. When it is necessary to reward resources, this model will discourage new uploads instead of encouraging new resources.
I will read the model and the references you gave me again. It’s not possible. I must be missing some point. It’s not making any sense in my head.
Reminds me of the quote from Fight Club - “On a long enough time line, the survival rate for everyone drops to zero”. The same would likely apply for the popularity of any perpetual data
If you can guarantee that there will be enough time to change if a problem occurs, I agree. Ethereum today is facing this problem, started with PoW and decided to migrate to PoS. Until that happens, perhaps the current high fees will make other platforms surpass it. Maybe not. I don’t think the modularity discourse ends the debate.