I've studied exchanges and exchange software for 10 years.
unfortunately things are pretty complicated. I would suggest try and
code together an exchange first. you will then find you need
time-ordering. matching by hand?? well.... that gets you 1 transaction
per 10 minutes. how does one node even know about prices from the
other node? it doesn't. an exchange is nothing else than a price
generation machine based on order inputs.

On Wed, Feb 19, 2014 at 1:34 PM, Manfred Karrer <mk@???> wrote:
> Have you read the paper?
>
> The project is like localbitcoin but without the company/single point of failure part. It is pure p2p. Any node has exactly the same rights/capabilities.
> Order matching is done by the people. Some publish an order, you recieve the orders you are interested in (filter) and accept any order you like in that list.
> Banks are only included in a private bank transfer between 2 private people. As long BTC is not illegal there is no justification to attack that (I know banks does not need a justification, but in large scale they would need, otherwise they lose their reputation). You dont hav one bank acount of localbitcoin or wikileaks, you have 1000s of individual ones and you have no proof that they do btc trades.
> Also it would be very hard to track those tx.
> The bank has to monitor all private bank accounts for tx which could match with the tx on the blockchain (impossible to do that without faults). That will produce more problems than it will helpt them to attack.
> If banks acts too costumer unfriendly they will loose them.
>
> br,
> Manfred
>
> Am 19.02.2014 um 13:16 schrieb Benjamin Cordes <benjamin.l.cordes@???>:
>
>> some good points here.
>>
>> * on the international level we have irreversible SWIFT, yes. but,
>> SWIFT is owned by an international regime. so its very unlike the
>> internet as a network. its an important tool for sanctions etc. you
>> can't just join the network. if you're very rich you might get a
>> license and join as a member. up to now the international financial
>> network and the internet evolved separately.
>>
>> * you also have the CLS group settling between fiat money. to trade
>> any type of money on the wire you need special accounts with banks or
>> central banks. quite a few big corporations have their own banks, but
>> usually you can't interface to their API's. large volume you can't do
>> by hand.
>>
>> * there are some attempts at combining financial networks and the
>> internet via credit cards. that's what eCommerce is largely using (
>> paypal is a separate case). stripe.com is an internet frontend for the
>> credit card network. but this network is owned by Visa/Mastercard and
>> used as a political tool as well. if you're Ku-Klux-Klan you're ok,
>> but not if you're a political activist the US has on a list.
>>
>> * if you do something like localbitcoins for fiat (you operate a small
>> exchange), you are going to get shutdown if you do any significant
>> volume. you have to comply with AML. as a maker of a market you match
>> buys and sells for other people. it is in fact illegal to exchange
>> things for other people, as you need a license. if you match them only
>> for yourself you have localbitcoins for that. bitcoin.de is very
>> similar (only that they take fees for I don't know what).
>>
>> * exchanges which are scattered are totally inefficient. there is just
>> no way to match orders not in a central point. what you can do is have
>> gateways to that central point, but centrality is what are exchanges
>> are about. imagine you have 100 exchanges - which one should I chose?
>> that with the lowest fees and highest throughput. the two are related.
>> high volume means little transaction cost. p2p exchange would mean
>> that everyone node operates an exchange. but 99/100 from those nodes
>> are uninteresting as they don't have volume. and if you do this over a
>> block-chain its even worse. p2p exchanges are failed projects and will
>> continue to fail. it really helps if you study auction systems. this
>> stuff looks trivial, but it is not. kind of like digital cash.
>>
>> On Wed, Feb 19, 2014 at 12:49 PM, Manfred Karrer <mk@???> wrote:
>>>
>>> Am 19.02.2014 um 12:27 schrieb Benjamin Cordes <benjamin.l.cordes@???>:
>>>
>>>> the problem you have here is called the double spending problem. every
>>>> bank transaction is reversible. it clears through institutions called
>>>> central banks. banks and central banks interact through a complex
>>>> inter-banking network.
>>>
>>> You are right that there is no 100% protection agains a chargeback (in case of criminal involvement like stolen bank accounts), but there are diffent levels of hardness for the chargeback.
>>> Paypal is the worst and someothers are pretty safe with chargebacks (OKPay, Perfect Money, Bank wire,...)
>>>
>>> See also BTC wiki Money hardness:
>>> https://en.bitcoin.it/wiki/Payment_methods
>>>
>>> Also a user doing chargebacks frequently will get problems with his bank I assume. It means additional effort and make him suspicious.
>>> Reputation system can help as well against that attack class.
>>>
>>>
>>>> so its quite easy for some authority to shut
>>>> down certain transactions, nodes,
>>>
>>> The tx is betweenn 2 privat people and has no visible conneection to a btc trade (there is no annotation on the bank tx like "BTC trade id:43535". people can write on it what they like...)
>>> There could be some data analysis including the analysing of the blockchain to match bank tx with btc trades, but that will produce many false positives and will be hard.
>>> As bank make it more difficult to use their services they will loose power as well. So it is not the best strategy for them...
>>> Also people could switch to mail, if it gets too cumbersome. Mail will not need much more time than a bank tx (what a joke...).
>>>
>>>> etc. in the underground people use
>>>> cash and money mules for rooting the banking world.
>>>> what big multi-nationals do is they circumvent rules by inventing
>>>> clever legal schemes, channeling money through offshore locations. so
>>>> the one strategy is illegal and can likely land you in jail, the other
>>>> strategy is legal and will let you earn applause for your cleverness.
>>>> its what accountants, bankers do all day long. if they are big enough
>>>> they just buy the laws they need.
>>>
>>> I dont see any legal issue.
>>> As long BTC itself is not illegal to do a private person to person trade without inclusion of any third party is not subject of any regulation.
>>> The tool does not take any fees or earn any money with the trades. So the tool providers (open source project, probably backed by a NGO) are not subject of any legal issues here as well.
>>>
>>>>
>>>> For ATM's you need a banking license. which is why the mafia don't get
>>>> to operate their own ATM's for white washing their money. all of this
>>>> should be obvious enough, but there is little that surprises me in the
>>>> bitcoin world in terms of naivety of how the financial and economic
>>>> systems works.
>>>
>>> It does have nothing to do with ATMs or the like.
>>>
>>>> if you really want to operate such a system you are
>>>> moving in the underworld. the problem that every anarchist has, he is
>>>> on the same side of the criminal, gangster, terrorist, certainly in
>>>> the eyes of the power structure upholding "the rule of law". on the
>>>> other hand you have a different category of criminals running the
>>>> world in board rooms of cooperations and in parliament. the best thing
>>>> is to understand laws and use them to ones own advantage in every way
>>>> possible, to change political systems so that they work better. not to
>>>> ignore all institutions and be naive about it. they will not get you
>>>> very far.
>>>
>>> I am not ignoring the legal framework. The he whole concept is designed to work inside that framework.
>>> There may be countries like russia where that is not possible.
>>> We will not recommend to use it there, that would be irresponsible as the threat of undercover agents is a real one in such circumstances.
>>> If our world is moving more in the direction that any private interaction gets suspicious we have to work hard on it to change that.
>>>
>>>
>>>>
>>>> On Wed, Feb 19, 2014 at 12:01 PM, Mats-Erik Pistol <meapistol@???> wrote:
>>>>> Manfred Karrer,
>>>>>
>>>>> Please include me in the upcoming forum/wiki also. I am highly interested in
>>>>> following how this pans out. I cannot contribute much, since I have a
>>>>> daytime job.
>>>>> However I think you should put more emphasis on sending the fiat in the mail
>>>>> and avoid bank transfers. The sender can then be anonymous and the problem
>>>>> of undercover agents more or less disappears. Another thought is that if
>>>>> trading btc becomes illegal in one country one should encourage
>>>>> international trading which induces competition between countries.
>>>>> Can you estimate how much funding is needed?
>>>>>
>>>>> Best,
>>>>> Mats-Erik Pistol
>>>>>
>>>>>
>>>>> On Fri, Feb 14, 2014 at 5:53 PM, Manfred Karrer <mk@???> wrote:
>>>>>>
>>>>>> I just created a project development thread
>>>>>> (https://bitcointalk.org/index.php?topic=465326.0) intended for open and
>>>>>> transparent communication for the development process.
>>>>>>
>>>>>> If anybody interested to join/contribute to the project please join there,
>>>>>> so we can start working on it.
>>>>>>
>>>>>> I am in contact now with about 7 people so lets see how to get the thing
>>>>>> rolling...
>>>>>>
>>>>>> Topics to discuss:
>>>>>> - Check the overall concept and test it there are open flaws
>>>>>> - Structure the development process
>>>>>> - Design the basic architecture
>>>>>> - Choose the technology to implement
>>>>>> - Funding models
>>>>>> - Organisational structure
>>>>>> - ...
>>>>>>
>>>>>> As soon as we have agreed on a project name, I will register a domain and
>>>>>> we can setup our own forum/wiki,... there. The thread is just intended for a
>>>>>> first start....
>>>>>> I am working on an update for the paper to clearify and address some open
>>>>>> questions/issues.
>>>>>>
>>>>>> br,
>>>>>> Manfred
>>>>>>
>>>>>>
>>>>>> Am 14.02.2014 um 02:23 schrieb Washington Sanchez
>>>>>> <washington.sanchez@???>:
>>>>>>
>>>>>> @Manfred:
>>>>>>
>>>>>> Not on OT dev team, just working on the conceptual design of the
>>>>>> TradeNet/Bazaar. Will post my proposal soon.
>>>>>>
>>>>>> @Adam:
>>>>>>
>>>>>> Unsystem forum?
>>>>>>
>>>>>> @Everyone:
>>>>>>
>>>>>> First name is Washington (drwasho), last name is Sanchez... don't mind
>>>>>> which of these you call me!
>>>>>>
>>>>>>
>>>>>> On Fri, Feb 14, 2014 at 11:06 AM, Adam Gibson <ekaggata@???> wrote:
>>>>>>>
>>>>>>> Sanchez, Manfred,
>>>>>>> This is what dansmith is doing now (integration of fiat transfer into
>>>>>>> OT).
>>>>>>>
>>>>>>> Adam/waxwing on the forums.
>>>>>>>
>>>>>>> On 02/14/2014 07:58 AM, Manfred Karrer wrote:
>>>>>>>> Thanks Sanchez!
>>>>>>>>
>>>>>>>> Open Transaction I was not considering yet as an implementation
>>>>>>>> technology, but definitely worth to evaluate.
>>>>>>>> Until yet I was mostly focussed on the conceptional part.
>>>>>>>> Are you working with OT?
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> Am 14.02.2014 um 00:40 schrieb Washington Sanchez
>>>>>>>> <washington.sanchez@???>:
>>>>>>>>
>>>>>>>>> Hey Manfred,
>>>>>>>>> First off, I want to give you sufficient kudos for taking the time and
>>>>>>>>> effort to think about this problem... we need more people like you tackling
>>>>>>>>> the big problems.
>>>>>>>>> It appears that most of this can be executed on top of Open
>>>>>>>>> Transactions... which I believe should be your approach here, rather than
>>>>>>>>> trying to reinvent the wheel.
>>>>>>>>> Open Transactions is fundamentally peer to peer, as in transactions
>>>>>>>>> between OT wallets is p2p as well as transactions between a wallet and
>>>>>>>>> Federated Server (which is necessary in the case of a central issuer)...
>>>>>>>>> combined with BitMessage (as you already discuss), you have a powerful
>>>>>>>>> combination of platforms to achieve your aims.
>>>>>>>>> The big problem at the moment is discovery of participants willing to
>>>>>>>>> perform this type of transaction, which is something that the OT dev team
>>>>>>>>> are working on (called the 'Bazaar')
>>>>>>>>> Here are a few resources RE the last point that should help:
>>>>>>>>> OT + BitMessage for p2p exchanges:
>>>>>>>>> https://bitcointalk.org/index.php?topic=212490.0;all
>>>>>>>>> Lex Cryptographia (don't this miss one, it's a work of art and will
>>>>>>>>> give you ideas for reputation systems and managing counter-party risk in a
>>>>>>>>> near-trustless manner)
>>>>>>>>> http://bitcoinism.blogspot.com.au/2013/12/lex-cryptographia.html
>>>>>>>>> - drwasho
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Fri, Feb 14, 2014 at 4:51 AM, Manfred Karrer <mk@???> wrote:
>>>>>>>>> I have written a concept for a P2P Fiat-Bitcoin Exchange.
>>>>>>>>> I am looking for developers so if you are interested, let me know!
>>>>>>>>> Otherwise any comment or review is highly appreciated!
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> BANK RUN! - P2P Fiat-Bitcoin Exchange
>>>>>>>>>
>>>>>>>>> Basics:
>>>>>>>>> - Pure P2P, cannot shut down by a central authority, no 3rd party
>>>>>>>>> involved.
>>>>>>>>> - Collateral as warranty for fair behavior
>>>>>>>>> - Fiat money will be transferred from one users bank account to the
>>>>>>>>> others bank account without any intermediary party. No identification that
>>>>>>>>> the money transaction is connected to a BCT trade.
>>>>>>>>> - No fees beside the standard BTC transaction fees.
>>>>>>>>> - Open source
>>>>>>>>>
>>>>>>>>> Here is the paper:
>>>>>>>>>
>>>>>>>>> https://docs.google.com/document/d/1d3EiWZdaM89-P6MVhS53unXv2-pDpSFsN3W4kCGXKgY
>>>>>>>>>
>>>>>>>>> High-level overview of the protocol:
>>>>>>>>> http://nucleo.io/bankrun/flow/bankrun.png
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Bitcointalk:
>>>>>>>>> https://bitcointalk.org/index.php?topic=462236.0
>>>>>>>>>
>>>>>>>>> Reddit:
>>>>>>>>>
>>>>>>>>> http://www.reddit.com/r/Bitcoin/comments/1xsona/bank_run_p2p_fiatbitcoin_exchange/
>>>>>>>>>
>>>>>>>>> br,
>>>>>>>>> Manfred
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> unSYSTEM mailing list: http://unsystem.net
>>>>>>>>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> -------------------------------------------
>>>>>>>>> Dr Washington Y. Sanchez
>>>>>>>>> Post-doctoral research officer
>>>>>>>>> Therapeutics Research Centre
>>>>>>>>> University of Queensland
>>>>>>>>> Princess Alexandra Hospital
>>>>>>>>> Woollongabba, 4102
>>>>>>>>> Queensland, Australia
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> unSYSTEM mailing list: http://unsystem.net
>>>>>>>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> unSYSTEM mailing list: http://unsystem.net
>>>>>>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> -------------------------------------------
>>>>>> Dr Washington Y. Sanchez
>>>>>> Post-doctoral research officer
>>>>>> Therapeutics Research Centre
>>>>>> University of Queensland
>>>>>> Princess Alexandra Hospital
>>>>>> Woollongabba, 4102
>>>>>> Queensland, Australia
>>>>>> _______________________________________________
>>>>>> unSYSTEM mailing list: http://unsystem.net
>>>>>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>>>>>>
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> unSYSTEM mailing list: http://unsystem.net
>>>>>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> mvh,
>>>>> Mats-Erik
>>>>>
>>>>> _______________________________________________
>>>>> unSYSTEM mailing list: http://unsystem.net
>>>>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>>>>>
>>>> _______________________________________________
>>>> unSYSTEM mailing list: http://unsystem.net
>>>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>>>
>>>
>>> _______________________________________________
>>> unSYSTEM mailing list: http://unsystem.net
>>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>>>
>> _______________________________________________
>> unSYSTEM mailing list: http://unsystem.net
>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>
>
> _______________________________________________
> unSYSTEM mailing list: http://unsystem.net
> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>