The other option mentioned some time back was the lockbox idea to try to address the whale ownership concern. I should probably wait for more details to comment but some of the issues with the lockbox idea are:
Complexity and its associated security risks
POW and its associated risks. One of these risks being that anyone at all, who didn’t even do the hard work of storing data, could end up mining the safecoins by devising some clever ASIC to overcome any advantage provided by node age in mining.
Network launch will be further delayed.
Are these risks worth “briefly” addressing the whale ownership concern? (Briefly because in the medium to long term, the same distribution will happen again anyway)
Are there other ways to address the whale ownership concern without introducing complexity and delays? Will the whales not sell enough with rising prices? Is it actually a bad thing if these whales had that much since they believed in the project enough to hold this long? Can the whales be called on to voluntarily donate part of their ownership? Etc.
What about popular data versus rarely accessed data - it that handled here or is that something for a caching mechanism down the road, or is it something already worked out elsewhere.
That’s easy - you can simply send them all to me!
Seriously though, that’s tricky I reckon, especially with all the various gov. regs.
On another topic … Is farming algo work coming soon? Does the team think this is a hard problem or relatively easy problem to solve? I understand of course that expectations can change, just curious what the vibe is at the moment.
What is ‘enough’ … at higher prices means less cost in tokens to pay for storage … so kinda balances out to some extent - farming algo work may clarify this further.
Also I think many whales will tend to sell as the price goes up. For many/most/all? of them this is a long term investment and they will sell to invest in new projects as they pop up.
I’m certain donations will occur naturally for promising projects built on the network. It’s an interesting dynamic because any good project built on Safe will attract more people to Safe … which then pushes up demand for tokens and hence the price … so donations to projects may not actually reduce the value of holding much at all - in fact they could increase the value of what you hold!
I’d add that it turns Safe Network into another crypto project with mining. And all the negative impacts that would bring. I cringe every time it is seriously suggested.
It may be a cute and interesting idea, but the crypto mining can be done elsewhere, no need for it to be at the heart of the Safe Network economy. Let that be a crypto project run on the network separate to the network core itself and its token/economy
Yes and I am not a fan of the Safe Network economy turning into part crypto methods/mining. Up to now its been the closest to digital cash out there, but in my opinion introducing crypto style of operation will be the downfall of a clean digital cash system. Even with potential cons, the idea to distribute all the SNT at the start (15% becomes 100%) is much better, a lot better, then any crypto project methodology. Solving a puzzle for a lockbox is mining at the heart, since that is what bitcoin does in essence, just a different puzzle.
I hope whatever is done does not have the feeling or looks of crypto processes. Digital cash is the way to go without any “mining”.
What I’m saying is that’s jumping to conclusions about the proposals, which should be ready for RFC within the next few weeks, and then everyone can evaluate, and chip in.
I think the MAID holders should absolutely be rewarded for their investment, foresight and patience.
However, as someone who holds absolutely no MAID I like the idea of being rewarded in some way for running a node.
From what I’ve read I believe you want lots of individuals to run nodes rather than a small number of individuals being able to run nodes and control being effectively centralised. Having to ‘stake’ anything at all is going to be a complication for most people and having to stake something crazy like $10,000 of tokens will really limit the number of people who can participate.
I also hate the idea of doing wasteful work to prove something and everything I’ve read so far suggests MaidSafe has tried to stay away from that so it would be a shame to put something like that in now.
The only thing I hate more than Bitcoin for waste is Chia. It uses ‘Proof of Space and Time’ which sounds cool but basically means ‘write lots of stuff to a certain amount of storage to prove you’ve got it’ and that is sucking up a lot of SSD and HDD capacity and wasting drives that end up in landfill. It’s affecting the prices and availability of storage.
You will be rewarded for running nodes, which is what they call “farming.” You supply disk space to have data stored to it like an encrypted piecemeal cloud drive. Even if they decided to give the full disbursement to current holders, we still need farmers to store data for the users which will get paid out when people pay to store things.
I would be highly surprised if the Maidsafe team decides to go with a PoW or PoS model.
At the risk of asking the obvious, is this different to ‘client writes to spentbook’?
Is the spentbook stored on the network, or is it held by each node locally like a cache?
To rephrase both these questions, do the elders only provide their authority for the transaction (ie their signature), or do they do other work too like data storage?
Another potentially obvious and stupid question: does auditing the money supply require iterating over the entire spentbook?
It is basically that. What @davidrusu was concerned about was a client not really writing, so perhaps writing to his bad pal Adult and not them all. With the scheme now the elders write the client signed spentbook to the network as they sign the transaction. So at least 1 honest elder will store properly. David can correct me if I am not clear here.
Yes it does, although you cannot see values you can test the out==in value. The keyimage (which is a representation of the input pub key) is visible. So you can see keyimage A spending and the output to say B and then you can see the B keyimage spending later. So you can traverse the whole Dag, but it’s a very messy Dag. So you see many many possible routes to the spending, and most of them are decoys.
Personally, I am not comfy with decoys and feel there should be a better way but right now it is the state of the art. A nice area for consideration though. I like the hide amounts only approach and have tones of microtransactions. So a valid DAG that tells you very little. However, ringct (hide keys and use decoys) appears to cover any/more attack space where attack == reveal spending.
We are implementing the whole ringct right now as it feels safer, but I remain a bit unconvinced it’s strictly necessary and I do feel there is a cleaner / simpler way. In saying that David Rusu has done some incredible work to make al the maths work for us. So a balance
haven’t heard that before. A good one!
