[Kinrany]

We need a crypto that builds the network on trust between real people. If A and B know each other and B and C know each other, then A and C can trade or give credit to each other by going through B. No need for a global consensus.

[dgb23]

Vouching systems are generally very solid if you also have some way to expell parts of your network after the fact. You need some way to slow down and curate your network or all the benefits are lost.

The benefit of these: it mimics in person interactions and describes a simple process.

But The trade off is that they only work if they grow slowly and organically, which seems to be contrary to what the crypto space is trying to do.

[BlueTemplar]

Hawala is widely used, but since it does the same job as a cryptocurrency (decentralized ledger), what would be gained by adding a cryptocurrency on top of it ?

[Kinrany]

This is indeed the same principle, it just needs better UX.

[pyrryh]

This is the case where the worse UX is just a better security in disguise (note: "better", not "perfect"; see multiple US reports on trying to investigate the terrorist funding through hawala). No electronic transactions cuts a lot of SIGINT (though not all, people carry their phones with them, call each other, satellites make images, etc.) and sometimes HUMINT is more difficult in countries like Afghanistan or Pakistan.

[Kinrany]

Security can be improved later, the same way Monero came out after Bitcoin. Having the option to transact without intermediaries and locally is enough of an improvement, the same way Bitcoin is Pareto-optimal when compared to SWIFT.

[pyrryh]

Hawala has kinda been out there for a long time, and I'm not sure that there is a space for incremental improvement since, as I already mentioned, its main security properties lie in hawala being out of modern fully electronic systems. This is the case when we are talking about the general mechanism, of course; a group of 2, 5, 10, N people | N < some abstract big number where we can't tailor solutions easily and need to scale, there some customization may surely be devised as an improvement.

[candiddevmike]

https://www.gnu.org/software/gnu-crypto/

[rglullis]

https://trustlines.network

[Kinrany]

Huh! The marketing materials look like exactly the same thing so far.

> However, since Trustlines is designed to be implemented on a public blockchain, making transactions requires the use of a native cryptocurrency to pay transaction fees.

But this one part seems wrong. With such a design there should be no need for global consensus. :(

[rglullis]

You still need to avoid double-spending issues.

[Kinrany]

Why? Everyone is only ever dealing with people they trust. And every single agent is in charge of confirming or rejecting every operation they agree to.

[rglullis]

Alice trusts Bob, who trusts Charlie. Alice has no idea who Charlie is.

Now Charlie gets $10 dollars as credit (not the same as cash) from Alice (mediated through Bob) and wants to buy $10 items from David and Eve. He shows the letter saying that Alice is good for the cash. Without a distributed consensus method, how can David or Eve confirm that they will be able to go to Alice to collect the cash? How can they even know that Alice has the cash in the first place? What if something happens to Bob?

The only way to do this without a blockchain would require something like Paxos, but Paxos only works if the participants are selected a priori. Every new participant would have to be vetted by everyone else, or everyone else would have to follow some central authority that can grant access to the system. If you are going this route, you are just re-inventing a credit cooperative.

[Kinrany]

Charlie cannot get a 10$ IOU from Alice, he can only get it from Bob. If Alice wants to give a 10$ IOU to Charlie, she will have to give it to Bob and ask him to give an equivalent IOU to Charlie.

If something happens to Bob, too bad. Both Alice's and Bob's IOUs are void because Bob can neither pay up nor demand payment. Both Alice and Charlie are sad because they knew Bob, and their ability to trade was more valuable than the current balance.