[Battlemesh] IGF 2018 WS #279 Scaling community networks

[pdbinder]

Likewise, if anyone wants me to STFU, I can go back into listen mode. :D This conversation is very relevant to where I am, so I'm enjoying the conversation and will continue. I agree with you that there are strong links, and we aren't even exploring all of them. To me a "trustless protocol" is one in which actors cannot be re-identified within. We are not using the term the same way. In the crypto / blockchain world I've heard and used the term "trustless" to refer to the lack of a centrally planned third-party that must be trusted. For instance, with bitcoin there is no issuer I need to trust - but with fiat I need to trust a State government agency. Another one, with the banking system I have to trust their proprietary security - but with cryptocurrency I can trust the code, not people, and un-censurable information such as network hash power, none of which is centralized. I also used the term "anonymous" to mean there are no names or accounts attached. A good term for your definition is probably "private", to mean "in which actors cannot be re-identified within." There are cryptographic currencies and other projects that are very private. The coin Monero for instance keeps sender, receiver, and amount of a transaction private (only those involved in that ONE transaction can see the details of the transaction). But that is just the protocol. The sender often has other methods of identifying the receiver - but those methods, for censorship reasons, should not be on the blockchain. The "privacy" or "anonymous" aspect of a network can ensure censorship resistance, but that doesn't mean the actors involved aren't known to each other. Here is how Althea deals with this: The payment settlement of an Althea network is going to be Ethereum's eth by default, but could be any currency (fiat, crypto, paypal, cash in hand). Obviously using crypto allows for trustless automation. Eth is "anonymous" but not "private". Transparent, and for parties involved, the actors are known and their actions are identifiable on the blockchain. The governance of an Althea network is ran on the Ethereum blockchain, with voting, finances, etc. Using blockchain to run an organization is called creating a DAO (decentralized autonomous organization). This is where automation occurs. If all dao members vote to approve the latest statements, the accounts will be automatically settle. If there are disputes or bad actors, they can be voted on chain. The governance is broken up into a main DAO with various sub-daos. This can all be forked, but right now there is one main dao (its the one managing nodes peering on the internet (don't like it, fork and peer). The local network is the sub-dao. Each sub-dao is going to be made up of the people who own infrastructure. You join a sub-dao for the purposes of trust, pooling resources, and facilitating payments between trusted parties. That sub-dao is the governance for the local network (as local as member want to keep it). The governance the sub-dao allows is only part of the network, but its a powerful part we haven't had before. As for privacy, I understand the network date is encrypted. And, as mentioned, the payment settlement coin could be another such as Monero that keeps that part private as well. I like the definition of community networks. I think it is important to recognize that we each have maybe an "ideal community network", an archetype, and that any effort will fall on a scale. For me, for instance, a network beholden to State government laws, grants, taxes, regulation, or control, is likely the furthest from my ideal that might also fit the definition given. I think it nearly a given that any network will have users that are not involved in the infrastructure, so we aren't talking about a 100% association there either. if said accumulation occurs - which seems likely as soon as any speculative motive appears - the community network according to our definition may cease to exist. The infrastructure may even stagnate altogether because of a tragedy-of-the-commons situation. I'm not sure from your description what this attack vector is. Who is likely to accumulate in your view? What do you think the goals of the entity or individual who is accumulating might be? If the network starts out decentralized but not anonymous, who could accumulate without the community knowing? More likely an actor might come in with better equipment and accumulate traffic, not existing infrastructure. This obviously leaves the market open to competition with even better infrastructure. I don't think this scenario would fit as a tragedy of the commons example. Though... per the CN definition, its more likely that the abuse would be a tragedy of the commons, but coming from the users who are not invested in the network and are not charged for its use or otherwise care about the network environment. Leeches. I'll be running some Althea tests soon. I'm also actively networking and searching for interested people to form a sub-DAO. So, I'll let you know how it goes! -- Patrick Binder ---- On Tue, 20 Nov 2018 08:19:04 -0600 Paul Fuxjaeger <paul.fuxjaeger at gmx.at> wrote ---- Hey Patrick, On 17.11.18 15:58, pdbinder wrote: > The trustlessness of the protocol and the "*LOCAL* structure of > trust" are two different things. Thanks for the comment, I agree that those two domains are different. But I feel like there is a strong link between them. Let me explain why: (I hope the list forgives a short economic debate ;-) I've seen very fine hackers getting annoyed very fast when the talk goes "crypto" - probably due to speculative aspects.) Some context: Here at Funkfeuer Vienna, we really could use new tools that help us keeping uplinks and public address spaces safely funded. The model we have been using during the last ten years is getting harder and harder to sustain. I see some parallels to our current national government in AT ... but that's another story :( So, we are open to experiments with new economic models, we actually have scratched our heads over the pros/cons quite a bit. Thanks for your hint on Althea, I'm in the middle of reading up on its current state. You wrote: > Althea is using trustless protocols but those who actually > run the infrastructure do not need to be trustless nor anonymous I think it makes sense to check here if we mean the same thing when we say "trustless protocol". To me a "trustless protocol" is one in which actors cannot be re-identified within. E.g. let's assume a community network introduces an anonymous token as a means of economic exchange. We think that in such a case it may become very easy for individual actors - maybe even completely outside the community (e.g. private investor) - to gain control over large parts of the infrastructure. And since the tokens are anonymous, the community can't detect this and therefore can't counteract. Our definition of a community network is something like this: "infrastructure is in control of those that are served by it" So, if said accumulation occurs - which seems likely as soon as any speculative motive appears - the community network according to our definition may cease to exist. The infrastructure may even stagnate altogether because of a tragedy-of-the-commons situation. Evidently, that's very relevant for us to avoid :-D Do you know if such an accumulation is detectable with an Althea network? Sorry for the off-topic digression, but I guess it may be relevant for other communities too... otherwise please feel free to STFU me -paul
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ml.ninux.org/pipermail/battlemesh/attachments/20181120/283e29cb/attachment-0001.html>