Some of these ideas have been touched upon here and there, but I think there are some back-of-the-mind assumptions about big farming that are incorrect.
The cryptocurrency space is so used to Bitcoin as the only real-world example we have to gauge against that I think some reflexive thinking gets put in that we should consider.
I’ll hope that @dyamanaka will figure out how to do some number crunching to see if my thoughts bear out.
I’ll start with my conclusion: There no real danger of big farming, at least not on the level we think of when we consider the massive resources of Rackspace, Google, Amazon, etc., entering in as farmers and eclipsing small farmers from getting appreciable safecoin. Further, it is doubtful if it will be profitable for any farmer operation to operate as a dedicated business model on any sizable scale.
My reasoning: The design of the SAFE Network is such that it favors already-owned and operated, consumer-level devices in terms of economy, because these are resources that people will already be owning and using, whether they participate in the SAFE Network or not. For an individual to join and dedicate resources to the network incurs a very, very low marginal cost to be offset by any safecoin earned. So the safecoin profit-bar is very low.
Just that level of contribution to the network, when scaled to even just 100k users is a huge amount of resource. I don’t know what it would take for a server farm to match, but it is considerable. (This is one place where @dyamanaka’s analysis might help.)
The next step up is the “gentleman (or -woman ) farmer”. This is someone who has some spare bandwidth and perhaps gets a raspberry pi or some such, or presses a couple of old but efficient netbooks into service. Or someone who has a pretty powerful machine and runs multiple vaults on it. These are people who can probably even afford and be willing to run at a bit of a loss in terms of current exchange rate for safecoin, because they really just want to participate and/or think whatever extra safecoin they make will be worth more as time goes on. The marginal cost here is still quite low, the profitability is not critical, and it’s a very flexible model, where the individual can adjust the setup easily according to their results, desires, abilities, etc. This class should give the network a huge extra boost in resource availability, again, very decentralized and under the momentary control of contribution.
Then we go beyond that we get into a completely different situation: that of the commercial farmer. Per the design as I understand it, there will be a decreasing utility for vaults above a certain size, so efficiencies of scale are minimized. There may be some earning advantage in having high bandwidth capabilities, speed of processing, etc., but it is doubtful if this will be a huge factor. Such an operation would undoubtedly earn a lot of safecoin, especially early on (when is would represent a higher percentage of network capacity), but let’s look at the downsides.
A commercial operation of any scale would require capital outlay, up front, either in terms of out-of-pocket or borrowed funds. This becomes a big speculation with much more limited ability to shift behavior (and thus expense) by turning equipment off and on based upon profitability.
While individuals will likely soon bypass conversion to other currencies and start spending safecoins amongst each other directly, commercial operations will have to depend on exchanging their safecoins for other means of paying the bills (at least till safecoin becomes accepted more universally). This will expose the enterprise to volatility risks, which also make such a venture less appealing.
Such commercial farming, if profitable at first and thus expanded or adopted by other competing ventures, would force a glut of capacity on the network which would force the farming rate quite low, again making it unprofitable and forcing such commercial farmers out of business because of unprofitability. Individuals and small, excess-capacity farmers would not find the lower farming rate especially challenging because their marginal cost is near insignificant and earning some safecoin would most likely be sufficient to cover their expenses and moderate use of the network. Also, individual and gentleperson farmers can painlessly withdraw from or enter into farming on the network as they are not carrying debt burden, additional rent, utilities, etc.
There is one other case to be considered: The commercial Farmer who is also a commercial End User. This would be something like a major insurance provider putting all their accounting records on the SAFE Network because it was the most secure, and running the necessary equipment to offset the expense, and perhaps earn a bit of bonus safecoin in the bargain. This is not a problem because it is a balanced, sensible use situation.
In examining this, it becomes really apparent that the incentives are directed to balancing the network in a sweet spot of resource-vs-use to minimize perverse incentives which inevitably crop up in hierarchical currency systems.
Not sure what this all means in terms of being a governor against speculative price fluctuations of safecoin, but think it bodes well. I don’t have any idea what the upper limits on safecoin are, but I think these factors will prevent the huge bubble/burst swings we’ve seen in Bitcoin.
Thoughts?