Digipl's battle. Read the assault and help provide clarity. Call to arms!

[hillbicks]

And the 2016 award for Mr Smarty Pants goes to…

[Marin]

@remonster

I understand that the nodes act as a higher trustable authority that will validate new groups. It would be helpful to explain the mathematical reasons behind the “sizing” of these groups that you propose and why this work. In particular, since there is no PoW, what is the PoS or similar that puts a cost into creating multiple identities that attemps to create malicious groups. That originally the groups are trustable it doesn’t mean that they can’t be tricked. If you have a link for this part of the protocol it would be helpful to understand why it works.

What do you mean by saying there is no POW?

[dirvine]

@hillbicks

With regards to the RFCs, which ones should be included?

Yes I think all accepted and also active, perhaps these can be shown as two groups as exist in the rfc repo, mind you a link to the repo may be enough I suppose. It’s just many folk don’t want to search for docs. I suppose we could link to a google search for all citations and papers, but not sure how far this should go. It does seem an issue, I suppose we need to decide how far we go in supplying all the links and papers. I say this as github links for accepted will change when an RFC goes active. Maybe for ease just a link to active RFC’s and one link for accepted rfc’s root dirs ?

[remonster]

@dirvine

Not nitpicking but here do you mean the Byzantine Generals Problem ? Bitcoin is an example of such a possible solution (proving itself so far, but shows where it would break). There are also a great many Byzantine tolerant networks (check many MS patents and papers on this subject).

I am not aware of a Nakimoto Generals problems, it’s perhaps a variant, but if you have some links it will help. Thanks for the tip if it is unknown to me.

In terms of papers I hope you have found some so far. I do really appreciate papers and work therein, but yes as part of a larger picture. Until it is a law it’s a theory and until it’s a working product it’s a hope, is the way I treat much of this. With the exception of cryptographic algorithms and similar principles (which only after many eyes on and in the wild testing can be trusted).

As even bitcoin shows without network effect bitcoin does not solve the problem, but you cannot really mathematically prove network effect, only show sybil resistance etc. We used some maths (some posted on here) for this and also many simulation iterations to back that up. It’s not formalised in a paper but is available for all to see, so far anyway.

Hope this all helps a bit.

“Nakamoto Byzantine General Problem” is the extension of Lamport’s Byzantine General problem for an open network (this means without a fixed number of “generals”) which is obviously harder to solve and for which I am not aware of a complete solution without PoW.

I do not understand what you mean by “without network effect bitcoin does not solve the problem”. Do you mean that a minimal critical computation power is necessary to achieve security?

Anyway what would be relevant are the mathematical ground for your new approach, this is why I am curious about the magic numbers you give for the size of the groups. In particular how the security improves with larger groups (at which rate?).

[dirvine]

@remonster

Anyway what would be relevant are the mathematical ground for your new approach, this is why I am curious about the magic numbers you give for the size of the groups. In particular how the security improves with larger groups (at which rate?).

There was a load of tests and results, I think posted here for this. If you grab the address_space_tool from our archive in the common library you can run the simulations. There is also a statistical formula which agrees with the simulations, not a proof though. Here it is.

[Edit - sorry about scale, please message for lyx copy if desired]
We also ran a load of tests with “Eureka” (the digital scientist, math tool) to try and calculate a more formal proof, but to no avail, it could not do it, even after running and a large machine for 5 weeks. The closest it got was

x = 6577.35444491108/(2.884820904 + G + 13M) + 1500.62923471823G/(3.029780114 + G - 0.9398090401*M) - 1540.9094305201

With x as attack size, G == GROUP_SIZE and M == Majority. I have the simulation results in csv for anyone who wants the,

So the address space tool gives over 70% attack resistance at Group size > 19 (well quorum actually) and the difference between GROUP_SIZE And QUORUM is the number of nodes out of sync during the lifetime of a churn event. So take QUORUM_SIZE as a security parameter and GROUP_SIZE as the number above quorum to maintain integrity. Testing is easiest otherwise you need to calculate a wide range of parameters (cpu usage for encryption, wire format of messages, number of messages etc.). Some of this we are testing through the MVP roll out to tailor for efficiency.

Hope that helps a bit

@remonster

Nakamoto Byzantine General Problem

Yes I have not heard this description previously. I do know that with a small network and small hash rate then this is not secured. With network effect then it becomes secured with the aid of economic incentive. This was seen from the many clones that were “out hashed” and died. I digress though. Thanks for the description, that helps.

[Marin]

@remonster

“Nakamoto Byzantine General Problem” is the extension of Lamport’s Byzantine General problem for an open network (this means without a fixed number of “generals”) which is obviously harder to solve and for which I am not aware of a complete solution without PoW.

Again why do you say there is no PoW in Maidsafe?

PoW in Maidsafe is called PoR(proof of resource) and is measured by the amount of storage (and by proxy bandwith) that nodes provide to the network.

Because the safe network is not in an open and hackable internet it has it’s advantages over Bitcoin, and one of them is that it can forego the use of blockchain and ASIC mining as it’s security mechanism.

[janitor]

@Marin

Because the safe network is not in an open and hackable internet it has it’s advantages over Bitcoin, and one of them is that it can forego

Bitcoin can work over different networks (including Tor), but more importantly that is completely unrelated to the PoW/PoR debate, why is this even being mentioned here?

It is “intends to forego”, not “can forego”. SAFE isn’t available while Bitcoin has withstood years of scrutiny. Pay a bit of attention to these meaningful differences.
It wouldn’t mean that SAFE cannot work securely, but it would mean that you understand the difference between “in testing” and “generally available”, which automatically means any comparison between the two has no real life importance.

[Marin]

@janitor

Bitcoin can work over different networks (including Tor), but more importantly that is completely unrelated to the PoW/PoR debate, why is this even being mentioned here?

It’s being mentioned here because my reply is answering why PoR is possible in SAFE and doesn’t need to use PoW. If SAFE network wasn’t secure then there is no chance of PoR working…

@janitor

It is “intends to forego”, not “can forego”. SAFE isn’t available while Bitcoin has withstood years of scrutiny. Pay a bit of attention to these meaningful differences.It wouldn’t mean that SAFE cannot work securely, but it would mean that you understand the difference between “in testing” and “generally available”, which automatically means any comparison between the two has no real life importance.

seriously?

What you think anybody reading my post is going to conclude that SAFE network is already built and running because I used can instead of intends?

I’m pretty sure that everyone is well aware of the stage SAFE is in right now and I certainly won’t preface every post I make with “If the SAFE network works as planned” - because that is just common sense by now

[janitor]

@Marin

It’s being mentioned here because my reply is answering why PoR is possible in SAFE and doesn’t need to use PoW.

PoR in SAFE is used to reward farmers for storing and delivering data, not for security. It’s wrong to equate it with PoW, which is a reward for security work that prevents double spends. I think those are very different things and therefore have different names.

@Marin

What you think anybody reading my post is going to conclude that SAFE network is already built and running because I used can instead of intends?

You overestimate the level of technical understanding among general readership, and while they won’t conclude the SAFE is up and running (although they do ask, recently almost once per day, just check out the newbies category if you aren’t aware of what’s going on on this forum) they may conclude that PoW equals PoR and that it’s been “proven” that it’s just as safe. It may turn out to be, but it hasn’t been proven in theory or practice.

[Marin]

@janitor

PoR in SAFE is used for rewards, not for security. It’s wrong to equate it with PoW, which is a reward for security work that prevents double spends.

If SAFE network isn’t secure the PoR doesn’t work.

If SAFE network isn’t secure then anybody that can hack the network can change the amount of safecoins that any node (or wallet) has and therefore security is essential for Safecoin to work.

That is what Bitcoin uses mining for - security! if SAFE network has security as a baked in feature it doesn’t need blockchain.

That is why I specifically mentioned that SAFE network is not in an open and unsecured internet protocol.

or should I say intends to…

[stupidcache]

There is a lot of altcoins without POW model but all those have a blockchain so the key here is not POW vs POS, the key is the consensus mechanism.

[cl0ck3d]

I’m so absolutely repulsed at the anti-innovative attitude that went on in that thread I WENT AHEAD AND SOLD ALL MY ETHEREUM.