Contest! Best comparison of Etherium, Storj and MaidSafe

Contract, vice implicit social contract (which many don’t accept) amounts to private law with the aim supposedly of making society more predictable. But its still coercion even if the remedies for breach generally amount to performance. We need system that avoid contract where possible. Contract as formal promises signify agreement and cooperation but they also tie knots that bind when it makes no sense. We need trust-less systems where possible as a means to trust. To me the aim of contract in the IT domain is likely going to be enclosure. To contract, to bind, even where it is at least semi-voluntary, closes down opportunities and reduces flexibility. Its needed where we intend to charge interest because at some future point we may need to coerce when the interest in reality looks unfair. In the Islamic system three people shake on an agreement and its binding. A billion dollars can change hands. That’s probably good enough as that system also avoids the tourniquet ratchet of interest.

Also contract is like a sword or a pair of them, it represents a will and the point where two swords clash- a meeting of two hard human wills. Deeper cooperation may do more to align wills. Contract seems more important in contingency where there is agreement on what to do if plans fail. Still limit where possible as we can’t predict the future and contract can work to make that problem worse vice better when there is power inequity between parties. Now we hear all the time of freedom from contract. As I suggest this I am not suggesting that people won’t bargain as hard as they can. Trustless systems can help them bargain honestly as hard as they can without have to worry as much about acting in good faith. Trustless systems may make behavior more predictable and readable from a future point that the paper trail of contract.

Ethereum=SUM(Storj,MaidSafe)+x

19Ena78yTq2Rw3o1rHhvgKVZQpzUWgJ9Kg

There’s an error in your equation:

Maidsafe -X = Etherium

Storj doesn’t come into the equation…lol

MaidSafes can run as contracts on Ethereum.

@kurt Project SAFE can execute contracts too, in a less restrictive instruction set and no need for ether. SAFE eliminates blockchain and the ledger/audit functionality, replacing them with true anonymity. Neither bitcoin nor Ethereum can do anything more, or more efficiently, than SAFE afaik. I don’t see any technical or functional reason for choosing either over SAFE, unless you actually want a ledger/audit capability and non-anonymity.

3 Likes

The problem is verifiability. The ledger is necessary for the provision of trust.
I am all for SAFE but I don’t see it being a viable solution to relay trust.
I am nowhere done with studying the domain but execution would depend again on trusting some 3rd party who is connected to the SAFE network.
Ethereum, admittedly very inefficient at the moment, promises to remove this layer. My contract is only limited by the amount of ether I can provide to run it. Even if half of the network goes down it’s still not issue due to the blockchain while my SAFE datas integrity is easily gone in the same case.
No question, even a simple contract will be horrendously expensive initially never mind storing data. So SAFE can fill that gap. So it could be some sort of back-end of your contract.

What makes you think this is the case? To me it is like a flat world thinker saying the world needs to be flat. This is saying for A to trust B then the whole planet must see it as if it will make a difference. A ledger is good for public recording though and that is where I see it being used, not that public ledgers in SAFE are not possible (far from it :slight_smile: )

SAFE is predicated on the fact that not every node needs to know everything. Imagine if each neuron in your brain had to hold the same info all others did. I think nature shows us a centralised ledger is not a design chosen by evolution. If you forget computers and look at the world around us, centralised public ledgers are not how things work. It’s the opposite. When you see that then all you need to do is wonder, how can we have systems like this, then SAFE becomes more apparent.

For SAFE data integrity assuming global coverage 4 continents would have to disappear almost simultaneously before the network would begin to potentially lose data (hopefully temporarily). A huge subject that could take years to prove, tests will show this though. It’s down to how the XOR network works, imagine poking a hole in the sea as an example.

In terms of integrity and attacks, here are some figures from test runs I am doing right at this moment. They assume many things that are unlikely, such as:

  • All nodes being added are bad nodes (no good nodes at all, the
    attacker has managed to disable anyone from joining)
  • Rank is not calculated correctly and in line with the real world
    (favouring an attacker)

There are some more technical details, but this is worst case scenario and does not include consensus chaining (which makes this insanely secure).

# of runs | good network size | bad nodes count for attack | average attack percent
----------|-------------------|----------------------------|-----------------------
    20    |       1000        |            2957            |        71.0116
    20    |       2000        |            4626            |        70.1686
    20    |       3000        |            7946            |        70.268
    20    |       4000        |            9428            |        71.3206

The attack nodes size here is the number of nodes you need to add sequentially and as a single attacker with no other person getting on the network (groups of attackers will actually out attack each other :slight_smile: ) This test has been running all week, it’s pretty thorough, but the last three weeks have seen a lot of these tests, which we will make public (already is in the source code)

So to attack a network of 7000 nodes then you need to create (and have behave properly and grow rank, over time) a further ~21,000 nodes (yes 2-3 times the size of the network you are attacking). If you did manage this attack you would control potentially a single safecoin [importantly not the network or distribution/creation of coins] (you will have earned millions of them though behaving properly (to get rank and create a group)). Now imagine the network of millions of nodes, the costs are pretty high and the reward miniscule.

So this is all possible and much more, but it is inordinately difficult to envisage and also to get your head around a new way of thinking. It is however, as nature seems to create and evolution favours.

I hope this helps a little, but this is not a small change to how we work or competition with a blockchain (they serve different purposes), but potentially a mind shift of significant proportions in IT. SAFE II and SAFE III whatever they are will take this to amazing places, SAFE I will protect our data and communications and give us a currency that’s fast free and anonymous, I hope. Lots of work and lots and lots of testing though.

[EDIT] I should add one of the most important fact this tests assumes is nobodies computer goes off and/or on again as this make this attack significantly more difficult. Sorry I forgot to mention it.

7 Likes

Just to play devils advocate there is DNA which is a little bit like a block chain :slight_smile:

1 Like

This is incorrect. The ledger provides an audit trail, trust is ensured through concensus - the blockchain allows this concensus to be recorded and traced back through all transactions - but that is not needed to ensure the transaction happened and can be trusted. SAFE handles the concensus which profides trust, and records the last transaction which provides verifiability.

This is not true either. It sounds to me like you understand how the blockchain works, and you trust it because of that, but you have not understood how SAFE works, and so you are imagining it is not trustable, verifiable, reliable etc. It is.

2 Likes

Yes a few folks say that, I wonder though :smiley:

1 Like

As mentioned mentioned a little earlier in this thread about Storj and SJCX;

Storj’s crowdsale is over. Sending BTC to the crowdsale address will not allow you to receive early access to DriveShare and Metadisk or receive SJCX. As of right now, the exchange with the highest volume is Poloniex, but you can also trade SJCX on Melotic and the Counterparty DEX.

Yeah @Luke @dirvine, DNA isn’t like the blockchain because everyone’s DNA is slightly different to allow for improvement over time but the whole point of the blockchain is that it’s exactly the same across all the copies everywhere

I am not a big fan of those comparisons but think about it more like H2O, it’s a prerequisite so is the ledger for further evolution in the autonomous space.

I think this is important.
I don’t have time to quarrel or wast yours with my lack of knowledge.
I’ll PM dirvine with some pointers from the pirate party mailing list on this subject.

1 Like

Thank you for this topic, I wonder if the definitions of Safe and Etherium have changed now we have had time to better interpret them?

I could not give a great debate on the difference of Eth and Safe…

I wonder if we could collect answers stating…

A. How they are similar…

B. How they are different…

C. The purposes they perform and why one will be better positioned than the other.

I see that Safe wants to remove servers but does not Eth do the same?

Safe wants privacy as key is not Eth the same?

Persistence of data… Well ETC is the same…

Ability to place functions, contracts and program apps, both seem to be able to do this.

I do not fell at the moment i could talk to a person faithful in Eth and give him the Safe perspective.

Any help in me grasping this better is appreciated.

3 Likes

Sorry to bump old thread. Just interested in convergence of goals between Etherium and SAFE with the new phases of Etherium being proposed and a better sense of the speed of SAFE. In particular refactor from C++ to Rust over speed etc., then dropping PARSEC because if I have this right because of speed, a centralizing factor (in a decentralized system) and memory leak(creep) and realization that almost all of it is unneeded because as a reward for effort a much simpler approach presented without compromise. Then Vitalek talking of increasing speed and going serverless to a degree and trustless- above I cluelessly criticize trustless systems implying they are not trustless not knowing they are trustless because I strangely equate contracts (agreements) with less than trustable outcomes and because my understanding of the material doesn’t run deeper than buzzwords a lot of the time.

I don’t think BT is book ended by Etherium but suspect both are by SAFE. Have a sense that none of these systems will be able to maintain on the basis of scarcity of compute or storage or
energy inputs- that is I don’t trust those. But really am interested to know if there is a converging consensus and vision at least on aims or goals. I know SAFE was more ambitious always. I was never that interested in the coin aside from a means to bootstrap a stable secure network into existence.

I am suddenly more optimistic that these networks will be allowed to develop to potential and maturity where before I worried they would be opposed and oppressed. That change has come because I finally see (maybe not) the magic in sovereign fiat currency as in the MMT explanations so it seems that most or all of the reasons I thought we had to have a new internet are fictions and surely known by those who keep those currencies and system. Top of the system is all grants, bottom always trickle down through rents and the most undesirable out come is merely slightly higher than expected inflation (never hyper) and scaring ourselves with nonsense to justify spending when we just need to let go of the myths about the system which seems to be inevitable and happening.