Safenetwork sustainability concerns - Bandwidth has an ongoing cost however Safenetwork is a pay once, benefit forever model

EDIT: after a long discussion with several people, I’ve come up with a shorter and simpler way to summarise the concern:

Basically my philosophy and argument simplified is this : it’s actually a valid^ deductive argument -

(^I used “valid” in a technical sense, to only mean that IF the premises are true, then the conclusion must be true. CRITICAL THINKING - Fundamentals: Deductive Arguments - YouTube Here’s a video you can watch to further understand what I mean. Anyway just to clarify, valid doesn’t mean my argument is “right” or even “have any practical significance”)

Premise 1 - If people are actively using the network, then the network should be able to continue to exist and sustain itself.

Premise 2 - Only viewing existing data on the network without storing new data is considered using the network.

Conclusion - Therefore, if people only view existing data without storing new data, the network should be able to continue to exist and sustain itself.

However, the network with it’s current economical model cannot sustain itself(or would have great trouble doing so, as we’ve all previously agreed) if people only view existing data without storing new data. Hence this problem need to be considered.

I challenge anyone to falsify the premises, because if the premises are true, then by logic, the conclusion MUST be true.

LONGER VERSION of the story below: (original post :slight_smile: )

I have a concern for the sustainability of the network as bandwidth isn’t free… so basically, the whole economy of the safenetwork i guess would depend on how much NEW data gets stored on there. But imagine if all data in the world is stored on the safenetwork and NO MORE new data is stored on there(this probably will never happen but let’s just say in a hypothetical scenario to illustrate the point that I think data on there should be able to sustain even without any new data coming in. That is, it shouldn’t have to DEPEND on people storing new data)

And, in order for farmers to continue running vaults, they’d still need to pay for bandwidth, but since safecoins only is paid when someone stores NEW data and not when data is accessed by the person who stored it(which to the farmers cost money as the farmers would have to dedicate their bandwidth, how would it be sustainable? I guess farmers will have to voluntarily donate their bandwidth even if they would lose money? Because a certain site can be as popular as YouTube on the safenetwork, not only will it require a significant amount of data storage but also bandwidth to access exisiting data.

i mean, even if new data DOES get stored on there everyday, as soon as we store everything on safenetwork lets say, the amount of new data would significantly reduce(or lets just say any event has happened that resulted in the amount of new data stored to lower significantly). Then, farmers will get paid less because less safecoins will go to the farmers, as the only come from people who want to store new data, hence increasing the cost of storing data, hence it will discourage the storing of new data further, this positive feedback loop is very dangerous for the network.

In the case with YouTube right now, it’s POSSIBLE for it to sustain if no new videos gets uploaded. But it’s IMPOSSIBLE(at least from my understanding of it so far) for the safenetwork to sustain if they had a similar website however no new videos gets uploaded.

What does maidsafe team think of this issue, and how would we address this issue?


I hope everyone genuinely understand my concerns and not treat me as if I’m trying to challenge the team. I am genuinely trying to help by discussing potentially very important issues."


Then SAFE will eventually die.

But with data generation per year on the rise then having no (little) new data stored to SAFE means that no one is using SAFE anymore to store their generated data and this would (to me) indicate that people are not really using SAFE anymore.

Data generated is massively increasing every year and so I’d expect that even if all the movies & vids are uploaded then each year there will be a huge number more generated to upload. If no more is uploaded then noone is wanting SAFE to store their data and eventually no one (hardly anyone) will download from it

Never happen. Not unless the current trend of doubling or more the data generated each year reverses.


Actually it’s the other way around, farmers only get payed when data is accessed by anyone that requests it.


Yes correct

And as people are using the network less and less as indicated by not uploading anymore, then farmes will be getting less and less and they too will stop and network is effectively dying.


One and a half hour of driving an autonomous car is said to generate 4Tb of data:

Just an example of a data source.
Not at all concerned about data storage needs decreasing.


There has to be a strong relationship between uploads and downloads, as the latter can’t happen without the former. I seriously doubt this relationship will ever change - we generate more and more data every day, just documenting our lives.

If less data is being uploaded, storage costs will decrease, assuming farmers maintain storage increases. This would encourage more uploads.

Ultimately, we have to wait for safe coin and a test network with economics to observe. I am comfortable enough with thus trade off for now though, considering the privacy it provides.


‘Farmers’ (aka everyday users) are already paying for bandwidth. It is a spare resource. I know a lot of people in the US ‘pay as they go’, but most of the rest of the world has unlimited and largely wasted bandwidth. I don’t see how the economics of paying for bandwidth come into it if there is no additional cost in the ‘spare’ economy? :thinking:


To back this up with some primary sources

Farm Attempt RFC (status: agreed)

On receipt of a Get request … [a test is performed] … safecoin is then created

Safecoin Implementation RFC (status: proposed)

Each successful GET will generate a unique identifier … [if it passes a test] … then farming attempt is successful

The hypothetical issue is based on an incorrect assumption that farming rewards are based on PUT, but when based on GET the issue goes away. I still think the incentive structure deserves more probing but this is separate to the concern of the OP.


Um… then you will have a problem of the safenetwork not being free to access like the internet currently is, Given the limited number of safecoins. You have to purchase and then pay safecoins to access the safenet basically.

1 Like

That is not what he was saying

You pay to store data on the network.

Farmers get paid when they retrieve a chunk for a requester. See @mav’s post above


This is what i was concerned about, a data network should depend on the people who are already storing their data and shouldn’t depend on their, or other’s ability to generate new data and pay for it’s storage. Essentially the new data creators and storers are paying for those people that are storing and accessing their own data.

I feel like the people who are storing data should pay for their own data storage and access by making recurring payments. Because if for example we don’t get as much people using the safenetwork, and not much new data stored on there, then it’ll eventually die. This makes no sense and can be fixed so easily by making people pay a recurring payment to store their data, and people will anyway, it makes sense. The idea of one time payment for a lifetime of storing and access of data is nice, however, it’s not sustainable, as every time the data is accessed by the storer, those people that gave bandwidth for their access needs to pay.

Sure, you can argue new data is always being generated and stored there, but the point i’m trying to make is, if the model works like this, then the survival of the safenetwork DEPENDS on people’s ability and willingness to generate new data and pay for their storage on the safenetwork, where it should really depend on people’s ability and willingness to pay for storage of their own data that they’ve stored on the network! Do you guys understand what i mean now? I seriously suggest the team reconsider their current economical model, as it leaves safenetwork open to possible failures, if you change the model it can make it beyond reasonable doubt, infallible.


yes, there has been, but uploading doesn’t necessarily mean you’re storing new data, especially not when the safenetwork is out, uploading to the farmers means they’re giving access to previously stored data.

Changing the economical model to make room for a recurring fee for people to store their data will also give them a peace of mind knowing the safenetwork will be stable and sustainable, because even if they WANT to pay a recurring fee to secure the network and their data they can’t right now. Their data and privacy now do not depend on big companies and corporations, but depend on on theirs and others’ ability and williness to pay to store NEW data on the network. Whilst that may be not as equally as bad, if the vision of the network is to put the data STORER in control of THEIR data, it should make it so that the storer is contributing on a recurring basis so the network is sustained and maintained - a sensible trade off.

The current model as i understand it is making someone else pay for the satisfaction of your needs(namingly, storing your data privately, worry-free and accessing them whenever you like), which will make the satisfaction of your needs essentially depend on others. Philosophically that doesn’t make sense and leaves gap for sustainability failures, is all I’m trying to say.

1 Like

This is like saying the food industry is unsustainable because it depends on people’s willingness to consume food. If you disagree, please explain the scenario where people stop wanting to store data.


Seriously? You’re comparing food to data? People NEED food everyday to live, do people NEED to generate new data everyday to live? No, maybe we SHOULD generate new data everyday if we want to thrive, but surviving and thriving are two different things…

So you can’t provide a scenario.


We actually seem incapable of not storing new data every day… oops, I did it again! :slight_smile:


A scenario where people stop wanting to store data does not matter though, it depends on people’s desires to generate and pay to store NEW data. They may want to store their existing data on there just fine!

And Actually, if you DO what to compare it, this is the current economical model right now. Let’s say we have a company named “SafeFood” who started having a new model to provide people food - You pay for food once, then you get food for the rest of your life, where do the money comes from? it comes from newborn babies or other people also paying ONE OFF for a lifetime consumption of food to SafeFood, do you think that’s sustainable? I don’t know, MAYBE, if we can reproduce fast enough and make more babies who will pay for a lifetime of food consumption for themselves. But the current food model as is, those who are DESIRING the food pay on a recurring basis. They do not pay a one off fee such that they will get food forever and the sustainability of this DEPENDS on others’ willingness to pay this one off fee for a lifetime of food.

In this comparison in the previous paragraph, you can argue that i have left out the fact that with data storage and safenetwork, even if you paid to store one data, you can still pay to store another data. With food in the last paragraph i made it seem like you can only pay once. So fair enough, you can argue on this, and I could change it a bit to make it exactly the same scenario, since there’s an near infinite amount of different food and cooks styles(just like there is new data), lets say SafeFood only allowed you to pay for specific foods, and not all/every food. So you can pay a one off payment for the lifetime supply of raw lobsters, as much as you can eat. And you have to make another payment for lifetime supply of white chocolate, and have to make another payment for a lifetime supply of dark chocolate etc You get the idea. EVERY single customer follows the same payment pattern. Keep in mind the initial payment would probably be somewhat high. And in this comparison, you also have to imagine that the customers will live forever and consume food everyday, just like the safenetwork is one payment for PERPETUAL storage and access of data.

Now, ask yourself, is this model sustainable? Maybe it is, but would it raise genuine sustainability concerns for you? You can argue new people are born everyday, new people are getting to know about “SafeFood” everyday and existing people who use “SafeFood” want to try some new food every so often, if not every day, so they’ll continue paying, to sustain the people that have already paid. But I still feel the current food model makes more logical sense, that YOU pay for YOUR food as you need it, which is on a recurring basis.

I hope everyone genuinely understand my concerns and not treat me as if I’m trying to challenge the team. I am genuinely trying to help by discussing potentially very important issues.

1 Like

That’s a straw man argument. I asked you to provide a scenario for data, and you can’t.


Um… I don’t know if you understand what the logical fallacy ‘strawman’ means… anyway incase you don’t here is from wikipedia…

The straw man fallacy occurs in the following pattern of argument:

Person A asserts proposition X.
Person B argues against a superficially similar proposition Y, falsely, as if an argument against Y were an argument against X.

They even gave an example:

A: We should relax the laws on beer.
B: No, any society with unrestricted access to intoxicants loses its work ethic and goes only for immediate gratification.

So person B has replaced “RELAXING the laws on BEER” with “UNRESTRICTED access to (all)INTOXICANTS” That’s a strawman. I honestly can’t see where I have committed a strawman, so please point it out if you can…

If they want to do their bit to support the network & make sure farmers are incentivised, they don’t need to make recurring payments for stored data but simply keep using the network for uploading new data, and/or offering their own vaults/nodes to contribute bandwidth & storage.

Users would kind of be able to make a recurring payment if they wish by re-uploading the same files to the network again. This will cost Safecoin (paying into the farners’ pay out pot increasing incentives), but due to de-duplication, it won’t cost the farmers any more space or bandwidth.

Data gets cheaper to store and transmit over time, and old data will generally become less interesting & significant over time. This means that old data is unlikely to be a big overhead for the network, so the data generating the most traffic will also be the data most recently uploaded & paid for.

I’ve sometimes felt uncomfortable that bandwidth is currently proposed to be free, but the recent ‘rate limiter’ concept provides some reassurance, and could perhaps be developed to enable a market for bandwidth for anything above a network-set dynamic threshold. This isn’t planned, but I guess it’s possible if such a market became necessary (which it may well not).

These concerns you have aren’t stupid, and have been discussed before with various points of view coming across (worth a serach for those topics).

From what was said previously, I think the team are open to the likelihood that tweaks will need to be made to the economic model in response to testing & seeing how it works in practice, so it’s not all set in stone.