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


#389

Ok, I used “valid” in a technical sense, to only mean that IF the premises are true, then the conclusion must be true. https://www.youtube.com/watch?v=3jvQrpVQaYM 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”

But anyway, this isn’t about likelihood of something happening though, I’ve already mentioned countless times I agree it’s unlikely people won’t use it to store data.

This is about:

  1. The philosophy of the network doesn’t really make sense, aka even if people are using the network, it still can’t(may not be able to) sustain itself, which makes no sense. As outlined by the argument. That’s more philosophically significant than it’s practically.

  2. Second(more practically significant one), is that by the design, even if the network can SURVIVE, it is limited by the storage of new data, it completely ignores the fact that another significant usage case of the network is the demand for people to maintain and access their data on demand. So to put it in argument terms again :slight_smile:

Premise 1 - All significant usage cases of the network should be designed in a way that boosts the growth of the network, for the maximum network growth.
Premise 2 - Accessing the data already on the network would be a significant usage case.
Conclusion - Accessing the data already on the network should be designed in a way that boosts the growth of the network, for the maximum network growth. (So it can THRIVE and not just SURVIVE)


#390

You have shifted focus in the topic. It’s OK :slight_smile:
Anyhow by that we have now left the discussion of sustainability concerns, and gone into a desire to improve growth capability. Fair enough.

  1. Mm, this is a so called “academic” point, based on the artificial situation.
  2. Still, saying it would only “survive” is based on this artificial situation. (I.e.: If there is an unexplicably low ratio of new data stored.)

To continue to spin theories (draw conclusions) based on this condition, requires the background of the condition to be more solid. It is yet not. It seems you do not intend to make it so either. OK.

I see no problem in the desire to improve capability of network to grow. But to motivate the need with an artificial scenario without supporting the scenario with substantial empirical or logical reasoning, would not be one of the best ways to get engineers to pay attention. Let alone to rush to any solutions before even solidifying that fundamental part.

Down to the nuts and bolts, it would need a very clear case on why a specific action is necessary. (And then we often land in measurements by probabilities.)

The desire to improve, that you express in your latest post, is a potentially interesting seed for discussion. But it is still only a seed, without growing in details there will be no fruit coming out of it.
And we, just as plants, come up with plenty of seeds, but very few actually become anything.

Since your conclusion is based on this artificial situation, yet to be supported by logic etc., I would suggest a different formulation, in form of a question:
How can we increase network growth rate by modifying current reward algorithms?
(We assume that we can always, somehow, increase the growth rate)
We would need to know more about what growth rate the algos in their current stage of evolution, would result in. (I.e. read up on them, know them inside out, are you there?) And these are not yet finalized, so it would be all we have to work with.
We would also need to know the cost of every increase from this level. Is it motivated?
Mind you, we are not yet deciding on any solutions.
From what we know today, if at all possible to answer this question now, it is quite a task altogether. The number of unknowns actually make the question weak to start with. But it can be tried. And you have just scratched the surface.


#391

There is one situation where that might happen and that is if most people have migrated to a new network that’s much better/cheaper/faster/safer, while lots of old data hasn’t been migrated yet, so people would use SAFE to access data that isn’t available on the new network yet, but would have no reason to upload anything new to SAFE.

The other argument in this thread that is that bandwidth cost might become an issue and the argument against it is that bandwidth will be practically free as farmers will mainly be home users who has already paid for their bandwidth through their broadband subscription.

If you sign up for a cloud service like Amazon Web Services the price for storing 1 GB for a month is a bit less than downloading 1 GB. If 1000 people in a month download a file on a regular cloud service, that would then cost more than 1000 times more than storing the file for a month. Bandwidth there is quite expensive.

Is there enough free bandwidth around to serve SAFE? If a large number of people suddenly started uploading much more data because they were farming, would ISPs get strained as they don’t actually have enough capacity for a large amount of customers utilizing their bandwidth 100% lots of the time? think there would need to be something like a few hundred million farmers before ISPs potentially would potentially find it a problem though.

This is also interesting in terms of whether high bandwidth costs will makes it unsustainable for many commercial farmers, ensuring SAFE stays decentralized.


#392

Wouldn’t putting a payment model around storing data be disadvantageous.

In an ideal scenario would be is no penalties for uploading; and incentives for storing.

Consumers store; which is primarily what the network needs, providers upload, that’s what the consumers need to make this network appealing. If it isn’t advantages to provide then no-one will have anything to consume.

Let’s say a commercial entity wanted to utilize the network for distributing large amounts of data (videos for instance). Could this network enable this commercial entity be incentivised by this network without the use of ads? If that is possible then BAM! But as i see it, is providers of information are just as disadvantaged (financially) as in a centralized environment.

Is it possible to have a scenario as described, or does it go against all the fundamentals of economics, well I don’t have the answers… but it sure would be pretty interesting.

Especially if you are an early adopter, imagine being penalized for providing content to better and increase the appeal of the network.


#393

How? It’s more likely that you understood it another way before. I think what I was trying to say is relatively consistent throughout. That network is sustained by the amount of new data. Which may be a problem. Even if it’s not a problem, it’s still not good as it’s the only major contributing factor to the networks growth, and a contributing factor is also a limiting factor. I think it’s good if the network has more major contributing factors for its growth. That’s all.

What do you mean? It’s perfectly illustrated. The network is Limited by the amount of new data basically.


#394

Ok, so you can’t form a single argument to support your assertions, beyond the ones I have already highlighted, where rent does not help. That’s fine, but don’t expect me to make your arguments for you - I am happy with the current design.


#395

All good points and I agree. However, I don’t see how adding a rental model is going to change this. Either the network is economical and reliable and people use it or it isn’t and they don’t.

Sure, we could provide an escape route for people abandoning the network, but fundamentally changing the design to achieve this? That is a compromise too far, IMO - people are going to leave with or without a rental model.

The amount of bandwidth required could be an issue, but that should be priced into the PUT costs. So, unless people stop creating new data (which we have established is highly unlikely), bandwidth issues would likely be a symptom of a dying network (unreliable, uneconomical, etc). Would a rental model prevent this? Possibly, as their fees could support the network in this state, but would they be cheaper than alternatives (given people must be storing their data elsewhere…)? Would the network be heading for the grave at this point anyway?

The assertion is that with no new data the network will decay, therefore a rental model should be established. I am not convinced rental will make the network any more viable and it will require significant compromises to enable it. On balance, persistent ownership of data is more compelling for me than providing rental.


#396

Ah, yes. This classical one. It is more likely that all here are not seeing straight, and you are right. Isn’t it so? After all you are the centre in your world, how could it possibly be otherwise? So sure about being right, and all arguments falling off as drops of water on a goose. Most of all that proves an intention other than actually understanding and contributing.

You shifted focus. From serious concerns of sustainability, to admitting that the fundamental condition you base the concern on, is highly unlikely, and therefore you only wish to improve the capability of network to grow (and still without even knowing about current capability, i.e. wanting to change what is not understood).

You seem most of all concerned with repeating the unsupported statement that the network current design is not good.
Like this line below: your serious concerns about network sustainability is now merely a statement that the network is “limited”. To what degree? Is the limitation of any importance? When I walk, I am limited by the air. Is it of importance for my movement forward? If not, and I keep announcing it in high voice, what is my reason for doing that?

No it is not perfectly illustrated. But yes, the conclusions are many times repeated. You repeatedly, over and over, base your conclusions on an artificial situation that you are yourself admitting to is highly unlikely, and of little practical importance. And for that reason you seem not to intend to make the background of it more solid.
This (in bold):


#397

If you are both a farmer and a user, it should pay for your contributions. How long you will need to farm, for how much you plan to upload will be decided by the market though.


#398

Just one observation then I will slink back into the wood work

Not only is the situation highly unlikely, it is made more so by the solution offered.

  • Data no longer being uploaded (just a little but almost nothing)
  • People relying on being able to read data
  • need to keep farmers paid a healthy income
  • rental & bandwidth payments in place to get farmers a lot more income (amounts like other crypto)

So riddle me this if people don’t want to pay for uploading more, then why would they keep paying rental?

This is a proposal to use a 2 dimensional economic model to charge people a lot more for their uploads (per month charge) and charge for download bandwidth and then that is directly paid to the farmers for storing the data. Yea nothing like what SAFE is going to be but a different model altogether and I gleaned this from his “logic” and refusal to accept the current aspects of the model which is more than 2 dimensional. A 2 dimensional model is where uploaders pay the network and as a direct relation the farmers are paid by the network. One dimensional would be uploaders pay farmers without network


#399

Hey, did i ever say I’m right or I am the centre of the world? it’s not helpful if you talk like this, as you’re no longer attacking my arguments. Even if i am a bad person that in no way weakens my arguments. I am only brining some potencial issues up for discussion.

With the current model, it is POSSIBLE(again, not saying it’s likely, but nevertheless possible) for people to be actively using the network yet the network unable to sustain itself. Philosophically, this is significant, as i think, a good model of the network should be able to sustain itself when people are using it, and shouldn’t be POSSIBLE for people to be actively using the network while the network not able to sustain itself. That does not make sense philosophically. Whether it has any practical significance is another question.

Although, if something is possible, then it can become a reality. For example, i could imagine a surge in demand of storage and farmers come on board, everyone starts storing and storing, then the demand for storage of new data reduces, and as farmers quit some people may lose access to their data permanently.

On another note, from how you talked… so you’re so sure that you’re right, that the current model will never have problems even when people access data much more than new data being stored? If so, why? And before you say “all here” first, there has been several people seeing agreeing with me that they may be a problem, second, there’s only about 3-5 people, like you, @Traktion, @ neo @Jabba and I can’t really name a lot more arguing strongly for it. But you are old members of the forum, it may be that you could be unconsciously biased to seeing the network is good the way it is. But nevertheless though, please accept the possibility that you could be also wrong, and the network may face set backs because of its current model. Otherwise, if you think you’re 10000000% right, there’s no longer a point to have any discussion.

Well, what’s your intention in attacking me as a person? I can’t understand why. It does not help the discussion in a meaningful way.

I just want to make the question more clear, so the question is: “if people don’t want to pay for uploading more NEW DATA, then why would they keep paying rental FOR ACCESS TO THEIR EXISTING DATA?”

First, to be honest, i am not even sure if you’re joking or not… because the answer is so simple I don’t see how you don’t see it… If i had some good photos with my family, or did a science research and published my results, I want the data that i’ve put there so far to stay there permanently, let’s say in the research case that i have since stopped doing such research, hence I don’t have any more new data, but nevertheless, i would very much like everyone to have continued access to the existing data I have put there. The network’s model however though, cannot sustain people doing this(if everyone did this)

If you think it won’t have any problems, fine, but don’t attack the person making arguments but the argument itself, and also, you should also please be open to the possibility that you could be wrong, as I am open to the possibility that it could work just fine, because otherwise it’ll be a pointless one sided debate.


#400

This might be a possible concern. Although storing photos only in the phone is risky. People want to store their photos in a secure place that is backed up. And most people don’t want to backup data on their own, so today people store their photos in some cloud service I guess.

The competition for the SAFE network in such use case is then free or cheaper cloud storage. The SAFE network probably can provide storage for a lower cost than most if not all cloud providers. But the question is, will that be enough to make users move to the SAFE network for storing photos and other private files? What if people in a near future expect to be able to store their photos safely on the internet for free? Then the network effect needed for the SAFE network might be absent.

One solution is to make all PUTs and GETs on the SAFE network free, while still rewarding the miners with safecoins. That may require a different inflation model for Safecoin though since coins are then no longer recycled.


#401

You mean they wouldn’t do the common sense thing to do and move their data to the cheaper & better system they are now using to store their new data. (because a old system they don’t want to store their new data on could die any day because others do the sensible thing and move their data and stop paying to a system they no longer want to use to store data) Face it that is so unlikely that I think an asteroid hitting the earth next year is 10000 times more likely

Anyhow now slinking back into the wood work. Have fun trying to get support for charging a lot so farmers can get more income


#402

Well, assuming even if safenetwork asks them to pay recurring, safenetwork will probably be very cheap if not cheaper than other systems anyway. So which systems are you proposing they move to if they wanted it the data to have other benefits of the safenetwork? Like un-censorable, secure, that they can uploaded anonymously, and that it can’t be taken down without taking down the internet, and with great data redundancy, and they have full control of the data?

There IS NO SYSTEMS IN EXISTENCE for them to move their data to. But if a better network comes out, then yeah sure of course they will start using the new one. But you have to realise, with the current model of the network, the network ALREADY promises that they’d pay once for forever access to their data, but if a new network comes out and everyone stopped storing new data on safenetwork, but do not want to pay to store the data they’ve already stored on the safenetwork to store that same data on the new network, they will eventually end up losing their data under the current model of the safenetwork


#403

Never said that. This is how you keep it going and going, answer something people didn’t say. I am done with this. You know that all you are after is farmers to be paid along the rates of other crypto projects at the expense of the uploaders. Of course people move their old data to the new favored system and BTW I didn’t say the new system they moved to was any sort of payment model. And yes there is other systems, what do you think dropbox is?

I am done with you


#404

I am not saying that! I am saying, the model could be changed such that usage of the network boosts the network, instead of having the model in a way such that accessing data, a major usage of the network, actually draining and not helping the network.


#405

I think there’s a misunderstanding, I am a bit confused now, anyway I removed that part the argument still remains.

You know that’s not the case! That’s just one of my proposals, you got to stop assuming all these things about me, but anyway, at least it may explain a bit about why you were quite against me. My intention is simply to bring up the potencial issues and let community come up with some solutions. And the one i have come up indeed have those consequences, but that’s because I am not smart enough to come up with others OKAY? So please help contribute if you can too. I would like the network to be up and running forever and have minimal setbacks as much as you’d like to.

Ok you obviously didn’t read the part of what I said about all the other benefits…

So be it, since you really can’t seem to understand where I’m coming from and at least some of the essence behind my arguments, but at least understand that my intentions are simply to make the network better and try not argue against everything I say unconditionally(which you seem to be doing)


#406

As you say, they are your assertions. You own them. Therefore, you need to provide arguments for them. In fact, I see a lot of arguing and repeating of your assertions, but I see little actual argument for your assertions. If you bring something to the table, expect people to interrogate the owner of them.

It is significant to the same degree as any lifeboat scenario. If you are not in or likely to be in a lifeboat squabbling over food/water, it is all academic though.

Or it could just be that we have thought about the problem for years, digested it, understood the technology and the approach chosen. Thus, we have concluded that your assertions have some validity, but there is no compelling argument for them occurring (you haven’t even presented one!).

What? If everyone did just that, there would be new data perpetually being added to the network!


#407

Yes…

So, Safe Net has become obsolete (uneconomic, unreliable)… rent will not save it. Just move your files to the new fancy pants network. You are hardly likely to start paying rent on an obsolete network, that you already thought you had persistent storage on. It just isn’t logical.


#408

If everyone did that IN DIFFERENT TIMES, if everyone did that at the SAME time then no. So this network then obviously relies on new people keep on doing that, we can never stop doing that otherwise all our data may be at stake.

Yes well could be so also, well said.

First, thank you :slight_smile:

Second, I’m a little confused, I’ve presented my philosophical arguments regarding how a network SHOULD be able to sustain itself when people are using it, and philosophically it shouldn’t be even POSSIBLE for people to use it while the network die at the same time. It does not make sense to me. Could you please help by maybe giving ME an EXAMPLE of a compelling argument that you said I should come up with?