IMO, this is the biggest concern. If it makes security weaker, then the goal hasn’t been reached.
It is an interesting deterrent though. Obviously having something to lose is a good deterrent. Node aging and loss of income being made for good behaviour/farming is a pretty good one in theory.
Perhaps a hybrid where sone farmed income is held on deposit would boost this further? So, there would be actually safecoin confiscated if a vault starts misbehaving? Just spit balling anyway.
Thanks for the thought @Stark but speaking for myself I don’t want special treatment, I do what I do not for that or any ‘special’ rewards. It goes against the ‘for everyone’ idea for me for some to be given special status in that way - all those here will be helping by contributing farming nodes on the same basis so that’s what I would favour.
Well I do want special treatment and I’ll happily take @happybeing’s perks too since he doesn’t want them. 
Nah, just kidding. I agree. I’m enjoying this discussion though, some interesting ideas coming up. The autonomous nature of SAFE is curiously limiting when it comes to implementing such an honours system. How could it be done (outside of test networks) without some central authority deciding who deserves favour on the basis of anything other than bandwidth, reliability and CPU? 
The best defense is a large swarm of users.
I’m all for keeping to necessity and doing what is most simple; introducing politics, I wonder is not that.
The boot of the network perhaps could be also in part, some resistance to new nodes becoming elders… until some halving date like bitcoin does, which sees all nodes age. Invoke logarithms and exponentials which I wonder represent the most natural growth. So, the aging becomes not a linear but sees elders become older over a curve and in that way the boot nodes control growth better… the idea then being that a large swarm of new nodes will always be at a disadvantage… yet rewards reasonable similar. The problem then might be keeping the distribution of aged nodes even but perhaps that’s already in place.
Disclaimer: I don’t know enough of what elders actions will be relative to new nodes in decision making. If volume of votes will be a problem, then perhaps this above doesn’t work?
As I understand it the node aging process is already exponential, so it’s quite hard to become an elder. Elders are just the oldest n nodes in a section where n is the quorum. So I think what you suggest is already in place. The trade off will be too hard to become an elder = no-one wants to farm vs too easy = security risk.
In a network in strong growth not so hard because the continuous split of the sections increases the amount of elders.
I’m wondering if the farming factor could be used in the ageing algorithm so a node which is farming when the farming factor is low (assuming low means less rewarded safecoins) it’s aged more rapidly (ageing factor is high 
) than during the periods when the farming factor is high (aeging factor is low), in a way to encourage farmers to always well-behave, and to not leave the network if the reward is low during some periods, as they will get benefited later on? …and I’m not sure if this effectively could help the bootstrapping of the network or not…?..
Interesting idea. Also how can we have mobile devices go off and online according to their charging cycles so that farming can be done by those with only mobile device (phones tablets etc). At this stage those devices will have a much harder time becoming farmers.
I appreciate the innovative approach of this suggestion but do not support it.
Safecoin is ultimately an abstraction of supplied resources so why not keep it closest to the final purpose.
Network health can only be measured from the supply and demand of resources, so proving / staking resources is all that matters.
I’d prefer the network to say “you don’t have enough resources to participate” than “you don’t have enough money to participate”. Do they end up being the same thing? Maybe. But at least resources can be put to work directly on the network, money cannot (or can it?!).
Provocative topic indeed. I’m unconvinced so far but will keep my mind open to more discussion.
Yes, appreciate the out of the box thinking, but if SAFE works as envisioned it won’t need this.
Pure decentralisation can work on its own, if it’s efficient enough
And we wouldn’t be here if we weren’t truly loving this space 
so don’t worry about us
I agree with @mav and @whiteoutmashups that its good to think about various ways we can assist in protecting the network and I may have been a little harsh last night. Sorry @anon40790172
My main objection is that it divides the “haves” and “have not” on money grounds and ability to get the coins.
Whereas providing resources is very much less a hurdle and one that the network will always have by necessity because of what the network is.
Those with and can easily get coins soon after launch is measured in the thousands.
Those who have the resources (computer+storage+bandwidth) is measured in 100’s of millions. The measure then comes down to how many want to.
That is the main objection and I gather a main objection to any proof of stake. SAFE doesn’t need this sort of segregation.
Secondarily by the merest of margins, is the security, by limiting the network (nodes) for the first number of weeks to those with and those who can get coins (many thousand) is less secure than allowing the 10’s to 100 thousand people who will want to try (well at least try).
The issue IMO becomes WHEN is it safe to consume content without risk. China destroys its dissedents. Other governments can be even more cruel. These are the very people we need to jump in. Uncertainty is the limiting factor in the current approach IMO. Years above months? How many will die or be jailed in that time?
I usually don’t participate a lot in the discussions, but this time I have to stand with @anon40790172. In fact I was going to post a thread with the same proposal and then I stumbled upon this one to find some heated arguments.
Honestly @neo I think you are overreacting against PoS as if it were a demon that will put Safenetwork in the hands of the rich and leave the rest out of the pie, even worse, easing an attack by a financially endowed actor.
I believe there are ways in which PoS can make the network resilient without making it unfair.
What I propose, is to keep the aging mechanism as it is, and add a next layer of aging for elder nodes with PoS, in a way in which a vault can choose not to spend the farmed Safecoin and gets an additional vote (and not necessarily a higher reward), this could have many levels of stake with a exponential curve so the oldest vaults would keep gaining votes in a logarithmic progression.
So a new user connects as vault without need of buying coins and could get elder status, but past that point this elder could choose to commit the rewards of farming to have a greater say in the consensus voting.
The amount of a vault can stake should be limited by the resources he has provided to the network, in a way in which you cant just buy coins to UP your stake.
So my proposal is not Stake to Farm at all, and maybe I should move it to another thread.
I would love to read what the team of @maidsafe has to say about these matters, because right now with the introduction of parsec the threshold for an attack is down to 33% and we need all the tools available to reinforce honest voting nodes and differentiate them from attackers. In the end it is a time/cost factor as @dirvine has pointed out. Including the ownership of safecoin in the equation is a smart thing to do, as it can make the network more resistant to a DataCenter attack. The challenge is how to do implement it in a way that does not lead to a passive concentration of wealth.
Not rewarding PoS can be an option, a honest node can risk his stake selflessly because he does not intend to cheat and it is a deterrent to a profit seeking attack.
I wonder if people realise that
In other words to introduce a PoS will mean a major code/feature addition for the Nodes and a weakening of the anonymity and security
Agreed and easy for those who have the means to claim “anyone can do it and it doesn’t give advantage to the few”.
And in the case of SAFE it is less than 10000 who are the advantaged and even over weeks its is only going to be measured in the few thousand extra.
We need everyone and I mean everyone who can spin up a vault to be doing so. It is only in numbers can we have the network size that protects against the bad actors. Not the other way around since the bad actor can easily be one of the relatively few who can add nodes.
They have an address though, so does that force every Farmer to have an account on SAFE as well? That would mean that every Farmer would already have to put up some Safecoin to get an account in the first place.
My understanding right now is that you could have an address on SAFE without an account. So no weakening of someone’s anonymity.
Just a brainstorm: maybe start the Network with another mechanism (like the forum invite system we use now for the authenticator) to allocate vaults and restrict it to a max no. of vaults for one ‘invite’.
And if the network is big enough, remove the ‘training wheels’.
I guess the problem that needs to be solved here is:
So what I hope to solve (in theory) with STF is the idea that for every 100 home users there could be 1 big Farmer providing resources. And there could be another one starting 100 Vaults to be evil. This would add up to:
That’s all like 33%.
Now one could say: Nodeage will prevent this… And I ask: Is that really true?? The Big Farmer and Evil Farmer can do all these CPU-calculations and resource proof tests way more fast than anyone from home.
So it might take days to weeks… But after some time you could get majority in several groups as an Evil Farmer. Now of course I hope that we’ll see way more fair Farmers than evil Farmers… And maybe some evil group doesn’t stand a chance… But a big Evil Farmer could really do harm the moment they have several million USD for example. That’s like a lot of Vaults.
NO it doesn’t
Even if there were an account, the network does not have the ability to look at the account. Unless you remove the account security.
Also the network doesn’t have the ability to see what coins you own even if the account record was there and unsecured. The network would have to scan all 4 billion coin addresses to find yours.
Remember that account can only be decoded with your secrets.
Wrong. You have to have the network know your coins list and this means that the coin list has to be created and the network allowed to see it. Which means bad actors running nodes can see all the other node’s balances. Where is the security in that
Oh and PoS then means PARSEC losses its permission-less attribute and this is MAJOR
Are you still with this idea?
Not only simplify the work for the attacker, but can also make him richer by only going short against Safecoin.
Sorry but it’s not a bad idea, it’s terrible.
Just want to pop in to say I really appreciate the discussion here… taking the time to challenge ideas is a really good way for others (me at least) to better reflect on and appreciate the design decisions that have been taken and their rationale. Cheers all… the SAFE community is hands down one of the best.