We have 3 or 4 threads running on different aspects of this but lets boil it right down to the common denominator: Somebody explain to me how I am wrong here:
The network costs money to run. It needs bandwidth. it needs disk space. It needs CPU time.
If downloads are free, why would anyone refrain from streaming music, watching videos etc to their hearts content?
Those activities are free to the user, but not free to the network. They use bandwidth, they use CPU time, All the GETS getting counted etc… But they are free to the users…
We charge for PUTs – that is the only source of income for the network. But it needs to be adequate to reimburse all of the costs of the farmers. How can this work?
Who would pay to put content on the network if the fees must be high enough to pay for all of the moochers?
I think there needs to be a cost per GET if the network is going to be economically sound – Or maybe high costs will limit the amount of content to things that really need to be on SAFE and cannot be delivered via other means… (Read Bittorent)
I may be incorrect, but my as far as i know the network collectively pays for GETs via inflation by issuing new safecoins to farmers who are hosting chunks that are retrieved via GET.
But then again I’ve also heard that the total supply of safecoin has an upper limit, so if that’s how it works I’m not sure how farming would work once all safecoins exist already…
There are definitely gaps in my understanding of this that I need to fill.
Putting more currency into circulation just makes the price of the coin go down. There has to be an underlying value transaction to balance things out… If people are given value (in the form of data downloads) without exchange (SAFEcoin or donation of services equal or greater value) , more or less the value will just leak out of the system.
This is true. bittorrent has a hit and run type of problem which is solved by using private torrent trackers.
Get = leechers
Put = seeders
Possibility have users micropay for the gets. It’s like bitcoin fees to the miners. If you want your transaction confirmation to be completed less than 10 minutes, pay a high fee. We could take this particular idea from bitcoin. Leechers will have to pay fee to the network to gather the data. Otherwise, their action will be low priority. Micropayment put users on high priority.
Seeders who pay the upload, leechers who pay for the download.
Edited: It’s like a highway with HOV lane. if you want to drive hov line, you need to pay a dollar or whatever, and it bypasses the traffic.
personally I would not use the safe network if I had to pay for downloads. If I had create a website I have to pay for hosting, or if it is free i pay by my computer being filled up with spam.
remember this is a network for the world, do you think the third world can afford to pay for internet and safe network for downloading…If an income comes from farming in this area they be able to upload for free.
Sure – but if you ran netflix off your webhost, what do you figure the bill would have to be? That is the issue that we are facing here…
In our case however a PUT costs what a PUT costs no matter how much strain it will put on the network down the road. So the price will be driven up by basically pooling netflix hosting costs with yours and mine.
I had the thought the safe network was the peoples network, for a place of to express ones self and not a network for big centralized monopolies. well that is my vision any way.
It will be too expensive to upload trivial files. (like you might seed to bittorrent) The more of these are on the network the higher the PUT cost will have to be to balance income vs outgo.
The stuff that will be uploaded to SAFE will be the stuff that really needs to be uploaded to SAFE. Secret stuff that you want to keep out of reach of hackers or snooping eyes.
The sad part it that it probably puts things out of reach of “Everybody”
private data , very rarely requested (GETs) used so will earn virtually zero to farmers of that data, cos no real amount of get requests on data stored for private use, even if this only accounts for a small % of the network, it costs more and will have virtually zero rewards paid out to the farmers . Under your assumption of a network only used to store important private data , then that % would be much much higher of private data and would alone balance it out.
data , some of its useless , some of its popular. some things uploaded to the network will be ridiculously popular(earning more than the cost), but more data will be visited infrequently(costing more than they generate) , meaning the amount spent on the data would unlikely reach the amount paid out for putting the data up.
If there is 1 millionaire app developer who has one of the most popular apps online at the moment , there is going to be thousands if not more, that have got apps that have no real attraction to those viewing and will earn them nothing or very little.
moore’s law . Edit: maybe saying Kryder’s law might be more accurate, but similar expected exponential growth of the network
Pretty sure there has been debates regarding this topic before but it might of been quite a while ago and take Sherlock Holmes to find it.
Edit: in my post it says number 3 before Moore’s law , but shows 1 when i post it, i swear i can count
Powerful stuff @jreighley. The only thing that I can say right now is a similar to @Seneca’s answer in his Vault routing burden thread. This requires a critical mass of users.
What I’m thinking about is the actual interaction ecosphere that APPs will normalize. I forsee APPs PUTting a lot of small bits of information onto the network on behalf of the user. Either as a way for them to customize their experience with that APP privately (think config files/statistics) or publicly posting content related to that APP (think reddit/comment posts/threads).
And that’s about all the answer I can come up with at this point. There simply needs to be many PUTs generated to balance this out.
EDIT: Good points in the post above by @gohan00760 too.
Since the network balances itself out by the available storage space, it can dynamically adjust to how farmers think their long term profitability develops.
It is safe to say that SAFE storage is going to be expensive compared to current cloud storage (which is free for most personal use) so SAFE needs to compete with something else than price.
Only after launch and feature completion can we know if this is feasible.
If the price of safecoin goes up really high, so much so it becomes infeasible to upload new data to the network that just means it’s time to farm a lot on the network and make decent money by selling safecoin. Remember you can easily trade safecoin for bitcoin.
Who wouldn’t farm when PUT fees are increasing to pay for all of the moochers.
Because when PUT fees increase, farmer rewards increase. But that way, the network balances out because there are now more farmers to split up the load.
No, it is a vicious circle. The more people download the harder the farmers work. But none of that increases revenue… The revenue is the same, because nobody pays to GET.
So the farmers are having to work really hard for not enough money their bandwidth is through the roof. – either they shut down or PUT fees go up.
That raises the cost for those of us who are saving the everyday documents… We are now less likely to use the network to store our non-vital stuff… It does nothing to disuade the moochers from downloading though. The network is helpless to stop that. GETS are free… Since I cannot afford to use the network for my everyday stuff, I am less inclined to farm – My ISP is annoyed with me anyway, and payments haven’t been that good.
The raising PUT rate happens not to make us all rich – It happens to keep us from going bankrupt – but it is a lot like Federal Reserve games. It can mask the underlying issue, but unless the issue is fixed the wheels are going to come off.
If somebody else is paying the bill, you are going to use a lot more resources than if you pay yourself. The current model has the PUTers paying the bills for the GETers.
People still need to pay their ISP for their bandwidth. Also, bandwidth is not consumed for it’s own sake, it’s not beer. There is no reason to believe people will download more on SAFE than they do on the current internet. I don’t think the tragedy of the commons applies here.
I think the network can balance itself. If PUT prices are relatively high, people are less inclined to upload huge files. If there are fewer huge files, then bandwidth consumption will be proportionally lower. Moore’s law helps us here, as well as the fact that in general the vast majority of bandwidth usage is from newer content. This makes sure that when PUT prices rise significantly, bandwidth consumption will also start to decline shortly after.
Edit: I must say I appreciate that you started this topic @jreighley. Although I believe it can work, we do need to explore this construction further, if only to better explain why it can work. It’s a very relevant question and will be asked frequently I suspect.
Something else that I’ve been emphasizing in the other threads recently is that content is consumed - meaning that once it’s accessed, there are few reasons to consume it again (there are some, but fewer than the reasons it would not be).
Also content, from what I’ve seen on the internet, breeds more content. Comments on videos. Meme responses to other memes. Replies to threads in forums. These will all contribute to the income of the network.
It’s not the canonical answer, but it certainly chips away at the chance of the network being overwhelmed by GETs.
The farmers do get paid more for working harder. The idea is that price of safecoin will increase with GET demand so the more people GET, the price of safecoin will rise meaning more people will farm. More farmers means an increase in supply for resources bringing the price back down. Farmers split the load and rewards. When farming becomes unprofitable, less people farm creating a lower supply of resources causing the value of the resources to increase. Hence the increase in safecoin price meaning farming will become profitable again. Its all about supply and demand in the economics of safecoin.