Bitcoin Forum
April 16, 2024, 02:58:19 PM *
News: Latest Bitcoin Core release: 26.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 [2]  All
  Print  
Author Topic: A Decentralized Exchange?  (Read 3221 times)
albus
Member
**
Offline Offline

Activity: 82
Merit: 10


View Profile
November 19, 2014, 05:41:12 AM
 #21

and I would add that all these new - Yes-I-found-the-solution - "decentralized" exchanges are just pushing to the user the burden of:
1) KYC (and all the risks it carries) and of
2) doing the bank to bank transfers,

while they keep centralized the most important part: trust.

In my view, being a trusted, centralized, escrow (and arbitrator in the very probable case of some issue arising) does not solve anything. Or I am missing something and in that case please enlighten me...
BitcoinCleanup.com: Learn why Bitcoin isn't bad for the environment
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
1713279499
Hero Member
*
Offline Offline

Posts: 1713279499

View Profile Personal Message (Offline)

Ignore
1713279499
Reply with quote  #2

1713279499
Report to moderator
DumbFruit
Sr. Member
****
Offline Offline

Activity: 433
Merit: 254


View Profile
November 19, 2014, 03:44:31 PM
 #22

In my view, being a trusted, centralized, escrow (and arbitrator in the very probable case of some issue arising) does not solve anything. Or I am missing something and in that case please enlighten me...
I don't think you're missing anything. Decentralized "exchanges" can decentralize communication between buyers and sellers but when the rubber meets the road, when the actual transaction takes place, there either needs to be a centralized escrow, or the "exchange" just dumps the responsibility on the end users to take care of it themselves.

By their (dumb) fruits shall ye know them indeed...
pajak666
Hero Member
*****
Offline Offline

Activity: 746
Merit: 502


Looking for advertising deal


View Profile
November 19, 2014, 04:00:20 PM
 #23

http://www.coinffeine.com/
is trying something like that
DumbFruit
Sr. Member
****
Offline Offline

Activity: 433
Merit: 254


View Profile
November 19, 2014, 04:03:40 PM
Last edit: November 19, 2014, 04:14:27 PM by DumbFruit
 #24

http://www.coinffeine.com/
is trying something like that
No...

Quote
Fiat transactions fingerprinting

The fiat financial system is based on trust: you are trusting your central bank authority not to devalue your money and you are trusting your bank to keep your money safe and not lose it in bad investments. Coinffeine is designed to minimize or even eliminate trust wherever possible, but since the fiat financial system is based on trust, you still need to chose [sic] a trusted payment processor to handle your fiat money.

The fingerprinting attack involves the payment processor blocking the Coinffeine-related transactions. This is possible since Coinffeine transactions will have a distinct pattern (lots of small, identical transactions between two accounts). We can mitigate this attack by supporting several payment processors in the client. If one of the payment processor chooses to block Coinffeine payments, then users will be able to move to another payment processor, which will increase its revenue with the flux of new users.
Emphasis added.

https://github.com/Coinffeine/coinffeine/wiki/Exchange-algorithm

In other words, coinffeine decentralizes the communication between buyers and sellers and then offloads the responsibility of actually performing the transaction on normal fiat transaction avenues.

By their (dumb) fruits shall ye know them indeed...
waxwing
Sr. Member
****
Offline Offline

Activity: 469
Merit: 253


View Profile
November 19, 2014, 08:10:03 PM
 #25

DumbFruit, you're greatly oversimplifying.

First, we have multisignature in Bitcoin. This means that a third party can arbitrate on the result of a transaction *without* holding funds. So they would have to be trusted to not collude with buyer and seller, and would have to be trusted to make a sound judgement on whether the fiat side of an exchange was carried out correctly, but they would *not* have to be trusted to hold funds. In this way it is dramatically different from a traditional escrow, in which the funds are transferred to and held by the escrow agent/arbitrator during the period of the transaction.

Then there are two problems remaining to be solved: how can you accurately assess whether a fiat transaction has taken place or not, and second, how (if at all) is it possible to decentralize this function of arbitration. The project I'm involved with, TLSNotary, attempts to address the first problem. And without what TLSNotary attempts (cryptographic proof of fiat transfer), you already have many ways used today to try to verify that a transaction took place (see for example the way that localbitcoins attempts to address this problem - there are lots of methods, although outside of tlsnotary they don't have *cryptographic* soundness). The second problem, decentralization of arbitration, is also extremely difficult to address perfectly, but that doesn't mean it isn't worth trying. Take a look at bitrated.com for an example of how you could have a market in arbitration, using multisig as mentioned above. This is clearly a step forward compared with Paypal, for example, where you have only one, centralized arbitration service available - Paypal themselves.

With regard to Coinffeine, don't ignore the game theoretic aspect of their attempted solution. It takes away the problem of having all the funds at risk at one time. However, I have never been convinced as I have always thought that you will end up with centralization anyway, as there is no real 'fiat micropayment channel' without centralizing that part. So I kind of agree with you on that, but maybe for a different reason.


PGP fingerprint 2B6FC204D9BF332D062B 461A141001A1AF77F20B (use email to contact)
altsay
Sr. Member
****
Offline Offline

Activity: 359
Merit: 250


View Profile
November 19, 2014, 08:58:47 PM
 #26

http://voluntary.net/bitmarkets

This one is using Bitmessage protocol.
Pages: « 1 [2]  All
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!