The farmers don’t need to care about whether they store spam or precious bank account data. The farmers will be rewarded regardless of what kind of data is stored.
Let’s look at an extreme case where 90% of the data stored on SAFE will be spam and attack data. And only 10% ordinary data. Then I estimate there will still be a lot of extra storage space available in the SAFE network. Because the actual usage of the farmers’ total storage capacity will only be like 50% or something like that.
Neo, by now you should see the logic and stop arguing: if the spammers fill up as much as 99.9999% of an arbitrary amount of available storage, they will never use all the resources!
In economic science this is known as Anders’ Paradox of the Fast Spammer and Slow Ant.
The upshot is that the fast spammer can never use up the entire capacity of SAFE network.
Now you are not even answering the quote. See ya later. Your beliefs override logic. Sorry cannot discuss on those terms.
That would be problematic yes. But think of how much spam email servers handle every day. My estimation is that safecoin will by itself fuel the SAFE network above the actual usage of the data storage. Ok, 90% spam is perhaps a bit over the top, ha ha, but say 50% spam will be manageable I think. And in reality the amount of spam can be reduced significantly by the proposal I mentioned earlier:
Make storage of both private and public data free and introduce a new storage type ‘public content’ that will allow people to earn safecoins for the content. The public content type will cost safecoins to store. So a spammer who wants to earn safecoins on the data will have to use the public content type which means it will cost the spammer safecoins to upload that kind of data.
I don’t get it. A certain percentage of spam storage can simply be taken into account in the economic model. To punish ordinary users and demand that they should pay for storage just to remove some spam is irrational.
You do not even realise you are not responding to what was said. Strong belief in an idea which that has been defended against logic for a while begins to override logic. Too difficult to discuss logic.
Your model will fail due to free uploading without any safeguards.
I proposed a safeguard. In the beginning Maidsafe can set up large farms as buffers. And then there will be other attacks that would be more worrisome early on. For example different types of DDOS attacks are probably possible in the beginning when the network is still small.
Please post in one thread so that people don’t have to copy responses into 2 threads. Choose one and stick to that thread please.
Yes, and maybe a new thread would be better. Something like “Safecoin making free data resources possible”
But I’m too unsure about my own idea myself, so I’d better wait or let someone else start it (someone with more knowledge about SAFE than me).
Subscription services are quite different. They involve payments to reimburse another party for storage expenses.
If I’m a farmer, I cannot enter into an exchange with another user that involves committing a portion of my vault to that user. That’s barter. Anything else involves a monetary exchange.
Ah, so you want to remove the dynamic farming rate algorithm as well. That would make the network even more unstable, because it would lose it’s ability to increase farming rewards when supply of resources falls short.
In Bitcoin it doens’t matter much if the total mining power drastically reduces all of a sudden. The mining difficulty would be reduced and the network functions just like before. It can do this because the continuity of the services offered by Bitcoin isn’t dependent on the amount of resources put into mining.
In SAFE on the other hand, a high enough reduction in capacity due to farmers leaving would result in data loss, i.e. network failure. This is why we need a farming rate algorithm that can increase it’s rewards to attract more farmers during such a crisis. This leads to an increase in the SafeCoin supply, which can be decreased again after the crisis by setting upload costs to such a level that network income is slighly higher than network expenditures (total rate rewards to farmers). Without upload costs, the network would have no income, and thus the expansion of the SafeCoin supply would be permanent, which would lead to the inflationary spiral I described earlier.
I understand the idea of “free” storage seems great in theory, but in practice it is harfmul. It leads to misallocation of resources and market distortions. Just look at the consequences of long term zero interest rate policies (free loans/money).
Products and services should only be free if there is true abandunce. In that case a market doesn’t exist in the first place. In any other case, someone is always paying the price. In your proposal, the price is paid by devaluation of the currency. It’s essentially a tax on all owners of SafeCoins. The idea that this loss in tax will forever be compensated by infinite growth of the ecosystem is a pipedream.
If there will indeed be a permanent over-supply of resources by farmers, the network will adjust the upload costs to match. This means that one might get hundreds of GB of storage for a single SafeCoin. There really is no good reason at all to abolish upload costs. If there’s abundance, the cost will be extremely low, as good as free. Whenever that ideal situation ceases to be the case, the network can increase the costs to keep the balance. Without upload costs, there is no control, and collapse inevitable.
No, that sounds like a clever algorithm. So it would be better I think to keep that algorithm and just modify it. I will not describe it again here, but check out my compromise in the other thread: Price difference between public and private data - #95 by Anders
Sorry for the late reply @happybeing. This thread is more about economic ideology rather than analysis. I’ll provide my opinion, for what it’s worth. I haven’t kept up with the whole thread. Some of my answers may not be in line with the questions.
Yes, I believe the SAFE Network can function using free storage (Zero PUT costs) after it gains mass adoption. However, I was concerned about the Network being vulnerable in the early stages, and posted this older thread DOS Attack caused by Data Overload.
Ideological Rant
The problem is not the system, it’s the people. We have to acknowledge there are some people who, for lack of a better word, mess it up for the rest of us. They over consume, abuse, exploit, and intentionally disrupt the balance of things. Certain forum members are able to see, understand, and expect those kinds of behavior. We should heed their warnings.
Analytical Rant
The SAFE Network is designed to be incredibly resilient. De-duplication creates a HUGE opportunity for near unlimited storage. If we truly believe this to be true, then PUT costs should eventually go to zero. But other factors (bandwith) affect the whole eco system. And perhaps, that is why we settled to “incentivize” farming behavior with Safecoin.
Many systems end up being 20/80, where 20% does the work for the other 80%. I don’t have a scientific reason why this is, but it tends to ring true universally. The SAFE Network may end up with 20% farmer population no matter what.
Summary
If we think of the SAFE Network in terms of growth stages (Baby, Toddler, Teen, Adult) then it makes sense to put safeguards (PUT costs), until the Network grows strong enough to not rely on them anymore. This means Safecoin will buy so much storage, you only need 1 SC for all your storage needs. Basically, you create/fund your new account and you’re done paying for PUT storage.
In addition, I hope SAFE GB will be implemented so we can have an inflationary turtle currency. It also removes the need to inflate Safecoin in the future. But that is another topic.
I’m not sure its possible to match the level of randomness by including a storage fee. By recycling the coins back to the main pool and promoting equality across nodes, the network can ensure the chunks are dispersed equally to reduce potential for some nodes to have more power than others.
It’s a difficult problem partially because you need to verify advertised behavior (e.g. vault lives on mirrored disks) or reward it (e.g. by paying more for faster response, because the issue is you always get the chunks at the speed of the slowest node (paying for speed would be wasteful) and another problem is if you could do that, it would encourage pooling/concentration, which is seen as negative).
One could run a separate (SAFE-s/w based) network and try to guarantee high SLA and sell their own coins to those willing to pay for better quality of service and support.
Thanks @ioptio and @janitor, that makes it clearer for me. It would only be something I could support if it could be made to work, as the privacy/anonymity/security is the highest priority, and the economics of it are what will make it sustainable 
[quote=“Anders, post:96, topic:4960, full:true”]
Something that could be tried is to have four storage types. I know, adding more complexity is usually not a good sign but anyway. A suggestion for four data storage types:
Private data - 100% cost to store.
Public data - 25% (or 100%) cost to store.
Free private data - Same as private data except without cost to store.
Free public data - Same as public data except without cost to store.[/quote]
This possibly has some merit if some changes are made to both simplify and suggest things.
So then it is reasonable to expect that as far as upload costs are concerned public and private will cost the same, and that providing free is problematic based on the nature of some abusive humans out there.
That seems to me to be a reasonable goal. So how can we achieve that?
May I introduce a new idea that you can explore or find fault with. It needs work to make it usable, but I expect that its doable. (remember I assume, with some confidence, that private and public upload costs are equal)
Thus we have a system that allows newbies to both learn of the benefits and get there first coin (as a certificate) gifted to them. It would explain how to get more coins when needed, how to setup a vault and the advantages of running one/more, etc.
Would this satisfy your desire for providing (some) free storage. If need be the gift certificate could be worth 2 or more coins to help out further. But any more than 3 coins would need some real attack vector analysis and reason for it to be so.
The gift certificates are paid for by SAFEcoins so the network economy is maintained. As suggested by @BenMS elsewhere it maybe possible to freeze the coins used for purchase until the certificate is used or returned to the system if account is not made permanent, and then the coin is returned to the system. This would prevent any inflationary effects from having coins+certificates > max coins
This wouldn’t be a bad new topic for discussion…
I especially am a fan of the tutorial aspect of this.
(EDIT: Maybe with several major ideological tweaks that is…)
You can if you think it has merit
But it is mainly for Anders to consider, since its his thread and hopefully this satisfies his desires since totally free storage is problematic due to the “if I can abuse it I will try”
This reminds me of a situation when the government starts messing with the market which causes both inefficiencies (suboptimal allocation of resources) as well as obvious problems, both of which obviously require even more messing with the market.
Dream on… I am only surprised Neo fell for it.
Guys, the good news is there is no way devs will have time to implement these fancy redistributionist schemes. (I may create a new topic on that later.)
But back to this socialist utopia: how can anyone - especially those who think socialism sucks - argue for these complicated and frankly ludicrous redistribution ideas? We already had discussions, more than once, on the supposed need to “stimulate” demand with freebies. At first these plans were merely meaningless and naive (up to 5GB, or 50GB, of free storage to anyone, which is the same, or less, that one gets from Google or MS), but because there were naive and unworkable their proponents have been coming up with plans that are even more complicated and - because redistribution creates ever more problems - more dangerous for the stability of the system.
So now we have this idea to not only give away farmers’ investment away (because that is what the idea is calling for - to redistribute farmers’ investment to users and SAFE app vendors), but to also make it more complicated, increase MaidSafe dev costs, further delay (or worse) release of SAFE and make the system game-able (which it will be - example: Wall Street and the Fed).
What good product or service needs that? Only those that have a high premium (price-cost). SAFE doesn’t.
So now I can only ask (for the fifth time): which part of this plan cannot be done by a private entity such as SAFE application vendor or private charity?
What unique skill and planning does SAFE coin giveaway requires, that private sector cannot provide?
Luckily development is already slow and behind schedule so it’s unlikely fancy redistributionist ideas will make it to v1.0.
Edit: fixed spelling