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


#121

Ok like i said there WILL be people who do it for free. But for it to be the majority of people, it can’t be like that. Do you see majority of people uploading for other bittorrent users for free? Do you get my point? And this will limit the expansion of the network, cos the economical model may not be favourable to the majority.


#122

Yes, if that’s the case then we will probably be fine, but that may not be the case, in fact it is not, you have to pay for bandwidth, which is included in your monthly fee, and usually they have either limited speed or bandwidth. Taking australia for example, either you pay for NBN for about $140 per 4TB per month, which is $0.035 per GB. Or you can choose ADSL which have ‘unlimited bandwidth’ as they claim, but it’s speed is around about 500KB/s(it’s upload speed is slower than that), but even taking 500KB as the upload speed, so .0005GB/s hence 1296GB per month in total, and it cost $59/month hence it’s $0.046 per GB. Bandwidth cost money, you have to accept it. And if bandwidth cost money, AND recurring money, to make it so that you pay ONCE for infinite time AND infinite bandwidth to own and freely access the data logically doesn’t seem to work…

Pay once forever data is a good concept, but is it really possible?


#123

Ummm they do it NOW for free, so I don’t get your question. You seem to contradict yourself here. And if they do it for free and to help others then more will do it if they are rewarded for doing it. So if anything that says the opposite of what you contend.

We will find out won’t we. But the big experiment of bittorrents suggests it can succeed with flying colours. My thought is though is that some tweaking of the rewards will be needed, like the how quickly to change the rates depending on spare resources at any given point of time.


If we tried this 10 years ago then I’d say that no way it will work since the availability of very high caps and unlimited were rare compared to today and the actual links speeds were probably too low on average to sustain it.

But with link speeds increasing in leaps and bounds compared to storage size increases and with the cost of links dropping too and with hardware costs dropping it is a much better proposition.

17 years ago I was being charged 15 cents per mega byte on top of my internet connection. Then I got cable and went to 3 GBytes per month and 10Mbit/sec down, 512K up. 10 years ago I had 200GB cap per month and soon after when to unlimited for cheaper with the same company. Now I get cable TV included for free and still unlimited 100MBits/sec down 2Mbits/sec up. Next year I am due to get the UP increased to 40MBits/sec.

I said that as an example that if you take today as all that there will ever be then one model may seem better. But as real costs drop and link speeds increase the pay once stor forever model becomes better and better and obviously more sustainable.


#124

Not the majority of people… majority of people just downloads and don’t upload. And i was talking about the farmers when the safenetwork launches when i said will.


#125

Yes, it will depend entirely on the cost of bandwidth and hard disk drives and the amount of new data stored on the network. I am suggesting implementing a safety net, it can have a threshold such as if you store a certain amount of new data every month or contributing a certain amount of GBH or GBGB to the network then you WILL only have to pay once for a lifetime storage of data. But the model as is now isn’t the best and have sustainability concerns, that’s all I’m saying.


#126

Yes and since bittorrent can and does thrive with most people just downloading so to can SAFE which will be similar. SAFE will reward those who farm (upload) as well, so it should work better.


#127

It does NOT thrive though! For unpopular data it’s absolutely shit! If you stored your encrypted btc private keys using bittorrent then deleted them from your hard drives, good luck ever seeing your bitcoins again!


#128

And I am confident that with many times more farmers than bittorrenters and SAFE rewarding farmers then that will not be an issue. Bittorrents is at best a uncoordinated, costly to uploaders (fear of losing connection, time etc and no rewards). SAFE solves those issues and adds rewards which should cover the cents a day to run a dedicated rig+HD including replacement costs.

I still contend that most if not all farmers will be those on very high caps/unlimited because of the cost factor, so bandwidth costs are basically free for those who can farm.


#129

So like i said, a safety net can be implemented without hurting the philosophy of the network or do damage to the current model at all, especially not with the new GBGB concept! So WHY NOT? It makes NO SENSE for you to not do it. Everyone can still enjoy the benefit of paying once for perpetual data, however they could be just required to donate a certain amount of their resources to the network every so often(proportional to what they’re using off the network) in order to get that status, otherwise they can pay for it too!


#130

Losing some inbuilt features of not trying data to people/accounts.

To do a GBGB you need SAFE to data log your (data) usage and know when to charge or not charge.

At the moment SAFE does not know what you upload or download, there is no way to have a “datacollection” node in the network to log this information. If you don’t collect then no one can spy on it.

The so called datacollection node is simply one that does all the correct functionaing but also grabs the data you want it to. So if you have to keep data logs of what is accessed by whom then a node can be built to collect this data.

In other words if enough of these nodes are included in the network then a group could start to use it or a government could too.


#131

Log your usage? Can’t farmers just log how many times the encrypted chuck of data gets uploaded by them? I’m not good on the technical side btw


#132

No this would be open to abuse. I could then as a farmer say that a chunk was grabbed a billion times to upset the applecart.

It would have to be a group/section kept statistic that has consensus. So a node (vaults also act as nodes) that records this statistic could then give it to whomever is running that “datacollection” node.

Also to provide this functionality then the chunk has to be tied to whom ever is to pay or receive compensation. So now you have a link to the owner and the downloader of the chunk.

This really upsets the security model that allows SAFE to keep the network having no knowledge of who is doing what with whichever data including how many times it was accessed. All SAFE has to do now is charge for storing chunks and the owner of the chunk is striped and not recorded. Then if you request a chunk the network supplies it without recording that fact.

This means that the network does not have to keep logs on any activity of any user or chunk in order to know when and whom to compensate or charge.

In my opinion the best way to implement any sort of compensation for bandwidth is to introduce a reward for bandwidth usage Obviously the cost to upload data is going to have to cover it at some stage.

Remember that the charging of data is disconnected from rewarding farmers. SAFEcoin is like a big buffer/well that allows data to be charged independently of what farmers are receiving even if the uncreated coins is relatively small.

Oh did you know that if two people upload the same file then de-dup means only one copy is stored but both are charge for uploading it. De-duplication occurs because the same file generates the same chunk (& its hash & its network address), so the network knows inherently not to actualy store it again when it reaches the vaults to store it.

This means that popular files will usually be uploaded multiple times so there is some compensation to the network for popular files. Popular files means that farmers will be rewarded more often for those popular files.


#133

I have 2 servers and both of them cost me Way less than 10 dollars per month - both have traffic unlimited (being throttled to 10mbps from 100 after some high amount being used)… My worrying level about sustainability is ‘not at all’


#134

Well, Look i’m not as technical but i assume if every farmer logs this information, it’ll just be the consensus of majority of the farmers? Even if you data collection nodes won’t really compromise much at all as they just collect randomly encrypted data pieces and log how many times each one is accessed. But anyway, it all depends on the bandwidth cost and the demand for NEW data to be paid-for for storage on the safenetwork.

While it can work if the cost of bandwidth is so low and the amount of people storing new data everyday is a good amount, i think it’s useful to have a safety net in a way that doesn’t compromise the existing network as it will only make it better. My proposals are just the start, there can be other ways to do this. Like you said for example, a reward for bandwidth usage.

But anyway, i have got to go for maybe quite a while now as I’ll be busy, I’ll join back the discussion when i can. And just want to say I feel like a better solution for the bandwidth problem can be worked out. Maybe people will pay to access the data they’ve stored on the network, so website owners can choose to pay for bandwidth used by visitors to their site, or otherwise visitors pay for themselves sometimes, but it’ll be miniscule, like maybe leaving $10 worth of safecoin in your wallet can give enough bandwidth for a family to access all the non-paid-for websites within the network for a month, so if they pay their ISP like $60/month then $10/month to use the safenetwork isn’t a terrible deal, i don’t know. But anyway… I do hope the founder @dirvine will have a think about this, as this can be a limiting factor to the network’s growth.


#135

No money required and new demand has to be taken as a given. The spare capacity, bandwidth, HD etc already exists out there. There is no cost beyond a tiny bit more electricity. SAFE releases spare capacity from the world, it does not create an incentive model to build new infrastructure, the infrastructure is already there, it is just providing an incentive to use it (which is as close to free as you can get).

RE incentives to farm, post and use bandwidth - Take me as an example, my pc is on 24/7 anyway, my connection is unlimited (as is almost everyone else’s, except in the US) and I even have solar panels so I need to use up my extra electricity. I don’t need a financial incentive to keep farming as long as I have a utility incentive because my cost to contribute is near zero. Many people won’t know or care about crypto, they won’t be using exchanges or ‘selling’ their coins for fiat. The vast majority of people just want something simple that works. They don’t care much about making 25c a day and they don’t care about it dropping to 15c, as long as it is enough to do what they want to do on the network. If you want a non-passive relationship with the network you have to use SafeCoin and you have to burn/destroy your coins when you do. That is the incentive to farm as I see it. Buying crypto is a huge barrier to entry, if I can farm with 3 clicks then Joe Public can finally ‘use’ crypto without it just being an investment/gamble. Likewise that is also the incentive to spend. If I have farmed 0.5 safecoin this week for free and it costs me 0.00001 safecoin to post an image to my new SAFE social media site then that won’t put me off doing it. The coins were free, the charge is so tiny as to not even be noticeable. I’d rather that than having adverts and being dependent on a single organisation (never mind the thought of being their ‘product’ and moving toward a surveillance state /shudder).

The reasonably large percentage of users (in the US) who have limited bandwidth may want to buy SafeCoin rather than farming, but that shouldn’t be applied as a generalisation for the economic incentives for farmers. Really, limited bandwidth is a giant scam and the rest of the world doesn’t have to deal with it. :wink: Design for tomorrow… I doubt ‘limited’ bandiwidth will last even in the US personally.

Having a 30 year+ storage seems pointless to me. All the data on the planet today will surely be an actual drop in the ocean by the time we reach any of those proposed rental expiration points anyway? It feels like selling buckets of water at the bottom of Niagara Falls. Sure, water is a precious resource, but in the context of abundance there’s no point in charging more just because the water originated in a little tributary way upriver. Even if our data capacity expands at a fraction of the rate predicted, it will still likely be more than we will need. There is a big market incentive for hardware providers to keep providing more than we need after all.

I only speed read all the stuff above so sorry if this had all been said, but it didn’t seem like the incentives you were seeing really tallied up with the design. Not from my non-technical perspective anyway.


#136

I’ve been idly skimming through this thread as it developed, and my 2 cents is that what @foreverjoyful thinks is a bug is actually a feature.

From what I gather, especially in the most recent comments, you seem to be concerned that the majority of people will be more motivated to make a profit farming rather than altruistically supporting the network. Fair enough, I think so too.

Thing is, SAFE’s economic model, as far as I can tell, is architected to encourage little guys rather than big guys. In other words, 1000 people supplying 1 TB is better than 2 people supplying 500 TB, even though the total space for both cases is the same.

Of course the person trying to farm with 500TB in their datacenter that they own will struggle to turn a profit. They’re probably not using those servers for anything else, have to pay a lot of electricity/cooling/bandwidth/etc. So they might have to drop out.

I see this as a design success, not a flaw.

Because the person farming on 1 TB (first scenario) is using their personal conputer. They’re using it anyway for music, work, games, and browsing, so they’d be paying those electricity / bandwidth costs anyway. These are the people farming is meant for.

So, imo, this is a success cuz it discourages the formation of a “farming cartel” (similar to the socalled “mining cartel” of btc miners in china) and keeps data more distributed.

Also, if the people in it for-profit drop out for not making a profit, then the farming rewards for everyone remaining would increase.


#137

The 20%-above-average ceiling also plays a large role in limiting the cost-effectiveness of a farming cartel, doesn’t it?


#138

Yes I think so.

Also @foreverjoyful, early in this thread you mentioned “Nobody wants to pay to send Snapchats”. Yes! This is exactly the point!

I would hate to have to pay per Snap I send out… Luckily, I wouldn’t have to! This is the point of farming, so that people can upload their data without using any of their hard-earned pocket money. Then sending snaps is “free” because u got the safecoins from just running the computer you were already using

This is the true incentive to farm, profit is secondary to this incentive i think.


#139

can we downgrade the topic title to the removal of “serious” to just “safenetwork sustainability concerns due to bandwidth”.

Thanks. Just getting a little tired of alarmist headlines to get eyeballs.


#140

I’m not a mod, but screw it, we can all edit titles and I’ve been thinking the same thing. Edited

:stuck_out_tongue_closed_eyes: