Unpaid transactions: Owe a Bro. #occupywallst
Michiel de Jong
DatenSpuren is a very interesting and inspiring unconference. One of the things i talked about with people is the idea of an unhosted IOU system.
bitcoin detaches payments from central banks by introducing a decentralized coin. i think that's not what we need (bear with me here...). Coins are not where the monopolies are. It's banks that have the monopoly. So instead of decentralizing money, i think we should decentralize "I owe you..." letters (IOUs). IOUs are the fundamental transport layer of economy. On top of that is money. On top of that is lending. In fact, bank notes started out as written IOUs. What if we define a protocol for IOUs, on top of remoteStorage. Is there such a protocol already?
We would need three types of messages: invoices, IOUs and receipts. Both PGP-signed, but also manually checked and reconfirmed by both parties (everybody ultimately is responsible to check their own monthly statements and see if it's all correct).
an IOU would need an amount and a currency. I owe you a beer. I owe you 20 euros from the other day. Initially, this would be something you use among friends. Somebody at jsconf (i forgot who it was) showed me an app they used to keep track of who paid for the badminton court that week, and who was there. Simple bookkeeping between friends. It would be an awesome thing to have in a phone app (i'm sure people have already experimented with apps for this use case). fill in the names of 4 friends, one person paid for the gas, the other each owe that friend 13,33 euros. The person who pays for the gas, while getting back into the car, opens his Unpaid app, selects amount '40', currency 'euros', and clicks 'take one for the team'. the others in the car get an alert 'Ben is claiming 13,33 euros from you [agree] / [dispute]'. if you click 'agree', you PGP-sign an IOU from you to Ben. Or, if Ben owed you money from last time, it says 'substracted, Ben and you are now even.' - you get the point. in this last case, Ben would get a receipt from you instead of an IOU, because you're cancelling a pre-existing Owe-A-Bro.
all types of actions (invoice, issue IOU, issue receipt), are confirmed to both parties, and show up in the history of both parties, so you don't get misunderstandings.
in case misunderstandings arise, you resolve them out-of-band.
IOUs can also be resolved out-of-band, for instance, after we come back from holiday, I still owe Peter 320 euros (he paid the car rental, and i didn't buy any of the beer), so i make a POBS (plain-old-banking-system) transaction to Peter.
also dispute resolution would be out-of-band. if we meet through a friend of a friend (or through e-bay for that matter), and when we meet so i can buy your Stratocaster, we get out our phones and i send you an unhosted ("unpaid" ;) IOU. i'm basically saying "I hereby sign a contract, saying that I will pay you for the guitar within reasonable time". and then i don't. then you can take me to court in either the country where you live, or where i live. it's the same as a rental contract. or a labour contract. or any other civil transaction. you just sign "yes, i will pay". that in itself is not the real transaction.
you could build clearing houses on top of this, at a higher level. and you could build lending with interest (the IOUs would need a pay-by-date added into them: "I agree to pay you back before the end of the year").
the system is not linked to any coin. i'm giving examples in euros here, but you could use dollars, or yens. and you could also write your IOUs in bitcoins. or in kWh (electricity is probably one of the most universal goods of our times, as was suggested this morning).
there would be no need to keep the private keys on the client. you can keep your private key on your remoteStorage. in case your remoteStorage gets compromised, people would be able to send whatever type of messages in your name, and it wouldn't legally bind you. also, you would get notified on your statements - weird payments would show up. and you would have to go and dispute them. your private key could also be written on a physical card that you keep in your wallet. if you lose your wallet, you quickly log in online to generate a new keypair (block your creditcard and request a new one). maybe you want to pay 2% of all your payments to a disputes-service, so that you can get your money back should someone steals your "creditcard". all of this would be /on top of/ the actual IOU-transaction system.
by making the underlying IOU-system an open distributed system, in the spirit of the web, and independent of any currencies or dispute-resolution hubs or even interest rates, and make easy phone-apps for it so people can "unpay" each other, we could decentralize some of the power inherent in the current banking system. taxes don't apply here, because it's not the real transactions that are happening. tax would be levied when the actual payment ensues, which from this system's perspective, is out-of-band. And POBS transactions would be the default way to clear IOUs, but no longer necessarily the only one.
let me know if you see any flaws in this. also, let me know if you are the person who showed me the badminton app at jsconf. :)
cheers!
Michiel
Rick van Rein
Bitcoin is nice, but just a start, I agree. What you are proposing
along the lines of an IOU is also what I've been thinking about:
Money should be created at the time and place where value is created,
not in some central bank under government control, like the ECB has
become.
It'd be interested to trade those publicly... someone I never met could
walk up to me and demand a beer from me, heheh :)
I've been thinking along the lines of using physical gold as the basis
for the IOU, like at goldmoney.com, and possibly declare an explicit
counterparty risk on top of that (you might expect a Bit or Grolsch but
end up receiving Heineken or Budweiser, for example). This idea is still
floating in my head and hasn't crystalised.
You won't cut it with storage only, IMHO. You'll need some way to
process payments. This may also call for the type of protocol I was
disucssing yesterday, as that also involves establishing a sequence
from a stream of incoming events from various sources, in such a way
that the distributed world agrees -- which is part of bitcoin too.
The interesting thing about money is that you'd have to care for
reliable, verifiable and non-repudiable processing. That'd sure make
it interesting :)
Other than that, bitcoin is an open call to spend a lot of energy, up
to the break-even point of burning just the Euro's that your bitcoins
are worth. Not very environmentally friendly...
Sorry if I'm a bit responsive on this list -- this is a favourite topic.
Cheers,
-Rick
Melvin Carvalho
> Hi Michiel,
>
> Bitcoin is nice, but just a start, I agree. What you are proposing
> along the lines of an IOU is also what I've been thinking about:
> Money should be created at the time and place where value is created,
> not in some central bank under government control, like the ECB has
> become.
Will reply to OP in a bit ...
Every time an IOU is issued, the money supply increases.
Both systems have merits and attack vectors, but agree decentralized
has been shown to be fault tolerant.
>
> It'd be interested to trade those publicly... someone I never met could
> walk up to me and demand a beer from me, heheh :)
>
> I've been thinking along the lines of using physical gold as the basis
> for the IOU, like at goldmoney.com, and possibly declare an explicit
> counterparty risk on top of that (you might expect a Bit or Grolsch but
> end up receiving Heineken or Budweiser, for example). This idea is still
> floating in my head and hasn't crystalised.
See voucher safe
Melvin Carvalho
> the open web, freedom hacking, and... banking. :)
>
> DatenSpuren is a very interesting and inspiring unconference. One of the
> things i talked about with people is the idea of an unhosted IOU system.
Yes, IMHO this is the single most important idea of the decade! :)
The main operation of a bank is to store a number. Something that
most people can today on a mobile phone.
In a nutshell, if you have access to a network shared file on your
device, you have decentralized the banking system!
>
> bitcoin detaches payments from central banks by introducing a decentralized
> coin. i think that's not what we need (bear with me here...). Coins are not
> where the monopolies are. It's banks that have the monopoly. So instead of
> decentralizing money, i think we should decentralize "I owe you..." letters
> (IOUs). IOUs are the fundamental transport layer of economy. On top of that
> is money. On top of that is lending. In fact, bank notes started out as
> written IOUs. What if we define a protocol for IOUs, on top of
> remoteStorage. Is there such a protocol already?
Work is ongoing in this area.
The best system of its kind is opentransactions: (The PGP of money)
https://github.com/FellowTraveler/Open-Transactions
Here are two wonderful podcasts explaining the the last decade of development:
http://agoristradio.com/?p=234
http://agoristradio.com/?p=246
More work is being done in the W3C payments group, bitcoin, agorist
community and a few others.
>
> We would need three types of messages: invoices, IOUs and receipts. Both
> PGP-signed, but also manually checked and reconfirmed by both parties
> (everybody ultimately is responsible to check their own monthly statements
> and see if it's all correct).
>
> an IOU would need an amount and a currency. I owe you a beer. I owe you 20
> euros from the other day. Initially, this would be something you use among
> friends. Somebody at jsconf (i forgot who it was) showed me an app they used
> to keep track of who paid for the badminton court that week, and who was
> there. Simple bookkeeping between friends. It would be an awesome thing to
> have in a phone app (i'm sure people have already experimented with apps for
> this use case). fill in the names of 4 friends, one person paid for the gas,
> the other each owe that friend 13,33 euros. The person who pays for the gas,
> while getting back into the car, opens his Unpaid app, selects amount '40',
> currency 'euros', and clicks 'take one for the team'. the others in the car
> get an alert 'Ben is claiming 13,33 euros from you [agree] / [dispute]'. if
> you click 'agree', you PGP-sign an IOU from you to Ben. Or, if Ben owed you
> money from last time, it says 'substracted, Ben and you are now even.' - you
> get the point. in this last case, Ben would get a receipt from you instead
> of an IOU, because you're cancelling a pre-existing Owe-A-Bro.
You just need to capture the semantics:
<a> accountEntry
owes <b>
20 EUR
date 1/1/2012
It needs to be REAL simple. No complex markup for the first version, IMHO
I've thought about those for some time and I think you need something
I term NATO
Nettings and Transactions Ontology. Basically capture transactions.
Capture Balances. Allow signing. Allow transactions to occur offline
and sync later (a la git)
>
> all types of actions (invoice, issue IOU, issue receipt), are confirmed to
> both parties, and show up in the history of both parties, so you don't get
> misunderstandings.
>
> in case misunderstandings arise, you resolve them out-of-band.
Yes, this cuts costs.
>
> IOUs can also be resolved out-of-band, for instance, after we come back from
> holiday, I still owe Peter 320 euros (he paid the car rental, and i didn't
> buy any of the beer), so i make a POBS (plain-old-banking-system)
> transaction to Peter.
>
> also dispute resolution would be out-of-band. if we meet through a friend of
> a friend (or through e-bay for that matter), and when we meet so i can buy
> your Stratocaster, we get out our phones and i send you an unhosted
> ("unpaid" ;) IOU. i'm basically saying "I hereby sign a contract, saying
> that I will pay you for the guitar within reasonable time". and then i
> don't. then you can take me to court in either the country where you live,
> or where i live. it's the same as a rental contract. or a labour contract.
> or any other civil transaction. you just sign "yes, i will pay". that in
> itself is not the real transaction.
>
> you could build clearing houses on top of this, at a higher level. and you
> could build lending with interest (the IOUs would need a pay-by-date added
> into them: "I agree to pay you back before the end of the year").
Tons of infrastucture can be build on top of the simple idea..
>
> the system is not linked to any coin. i'm giving examples in euros here, but
> you could use dollars, or yens. and you could also write your IOUs in
> bitcoins. or in kWh (electricity is probably one of the most universal goods
> of our times, as was suggested this morning).
Yes you can have a URL or 'ricardian contract' to describe the asset.
>
> there would be no need to keep the private keys on the client. you can keep
> your private key on your remoteStorage. in case your remoteStorage gets
> compromised, people would be able to send whatever type of messages in your
> name, and it wouldn't legally bind you. also, you would get notified on your
> statements - weird payments would show up. and you would have to go and
> dispute them. your private key could also be written on a physical card that
> you keep in your wallet. if you lose your wallet, you quickly log in online
> to generate a new keypair (block your creditcard and request a new one).
> maybe you want to pay 2% of all your payments to a disputes-service, so that
> you can get your money back should someone steals your "creditcard". all of
> this would be /on top of/ the actual IOU-transaction system.
This is where WebID can be handy! Or Oauth tokens.
>
> by making the underlying IOU-system an open distributed system, in the
> spirit of the web, and independent of any currencies or dispute-resolution
> hubs or even interest rates, and make easy phone-apps for it so people can
> "unpay" each other, we could decentralize some of the power inherent in the
> current banking system. taxes don't apply here, because it's not the real
> transactions that are happening. tax would be levied when the actual payment
> ensues, which from this system's perspective, is out-of-band. And POBS
> transactions would be the default way to clear IOUs, but no longer
> necessarily the only one.
>
> let me know if you see any flaws in this. also, let me know if you are the
> person who showed me the badminton app at jsconf. :)
There's one major ingredient left.
All currencies are based on trust.
Gold is trusted to be pure and have a limited supply
Bitcoin is trusted to open source and secure
National currencies (fiat) are backed by a govt
State currencies (landes, WIR, North Dakota) are backed by a state
The trust infrastructure must grow side by side with the accepted
currencies. This is a project in itself, but I think we're at a
point where things can get started!
>
>
> cheers!
> Michiel
>
Melvin Carvalho
>> The best system of its kind is opentransactions:
>
>
> like cheques and cash, receipts, invoices, markets, etc.
>
> Seems to me that all of the IOU stuff described below would be an
> additional layer on top of this. (Specifically, Ripple.) See the notes
> in my last big release about how to implement Ripple on top of OT
> (coming someday). Basically the Ripple layer is what handles all the IOU
> stuff (and harnesses the web of IOUs to allow for RipplePays based on
> those credit lines.)
>
>
> One more thing...
>
>> The main operation of a bank is to store a number. Something that
>> most people can today on a mobile phone.
>
> for Bitcoin, by "storing a number" to show my Bitcoin balance on the
> MtGox server. But then hackers broke into MtGox and were able to change
> people's balances by simply changing the number.
>
> A transaction processor must go farther than this. It must not be able
> to change your balance simply by changing a number. YOU must sign your
> balance first and then the transaction server must countersign that
> receipt. If your balance is whatever appears on the last signed receipt,
> then the server cannot ever change your balance unless YOU SIGN FIRST.
>
> So the job of the Open-Transactions server isn't just to store a
> "number" as your balance, but to do so in a provable way, that prevents
> the server or any hacker from interfering with that number later on.
>
>
> Bitcoin basically allows all of the nodes to play the role of validating
> transactions and balances by signing off on them. This is why Bitcoin is
> publicly-auditable (not untraceable) and this is also why Bitcoin takes
> 10 minutes to an hour for transactions to clear (because you are waiting
> on enough nodes to sign off on the transaction.)
>
> The only way to have instant finality of settlement, and untraceable
> transactions, is to have a transaction server. However, IMO it is
> possible to use federated servers (like in Diaspora's model) as a layer
> on top of crypto-currencies like Bitcoin, which gives the best of both
> worlds. (That is, the censorship-resistance of Bitcoin, combined with
> the untraceability and instant-finality-of-settlement of OT.)
You're right ... the security part is key in having a bank that you
can trust. Trust is the life blood of any financial system.
As an aside ... it's possible to run person to person 'tally' systems
(ledgers?) using a shared file, say in Ubuntu One or Dropbox (cross
platform). I do think using uhosted as a base system would be
superior than ubuntu one / dropbox, and adding layers of security and
trust can make it that much more robust.
>
> -Fellow Traveler
Jan Wildeboer
Hash: SHA1
On 10/18/2011 01:31 PM, Melvin Carvalho wrote:
> You're right ... the security part is key in having a bank that
> you can trust. Trust is the life blood of any financial system.
The real challenge is to create a decentralised system wher trust is
handled directly and via transfer of trust to a "middleware" layer
called banks.
The real solution would be a P2P system where trust is in the network
and not in the accumulation of financial value (this is why we trusted
banks, which now is in a crisis).
> As an aside ... it's possible to run person to person 'tally'
> systems (ledgers?) using a shared file, say in Ubuntu One or
> Dropbox (cross platform). I do think using uhosted as a base
> system would be superior than ubuntu one / dropbox, and adding
> layers of security and trust can make it that much more robust.
That would be at the max a machine to machine tally system. A true P2P
system would work fundamentally different. Using whatever shared file
system is attractive for abuse (see bitcoin's problems with the
wallet.dat file).
Quite some smart people are thinking about alternatives and we should
be part of that discussion instead of trying to come up with Yet
Another System.
You should meet Georg Zoche, a good friend of mine, who has spent at
least 10+ years in analyizing the current situation, where it went
wrong and what axioms should be defining for a new and better system.
Jan
- --
Jan Wildeboer - j...@wildeboer.net - http://jan.wildeboer.net
Open Source Evangelist. Open Standards Fundamentalist.
Software Patents worst nightmare. Decentralizing whatever exists.
Citizen of the first Transnational Republic
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
iQEcBAEBAgAGBQJOnW56AAoJEK4g/MdBcijwTFcIAKf7zp+Grn9jjocIzymN7ajM
tUQG2nIAj6zBDZOrBDLQLvx0TZMBxuWFHzYxu0tZ0191WJfWI/vXoBxwIlDfpMsE
+MO8jpYB79d2Akdd+Pv6tR2FuIlkw/N6MBnh3AyqL3pqG2AUGjMjUqir2QQ3iE95
H/4brz8Q5kwCJu3XGpARZNLSp9IRPBwQOQSKK+Qe6sR2GTnVZgD2qi7YptTUDFRy
jxn1jKYcJKvtzGe1F45wRt2EqBmjMigS8GhCKSdEpZxG/cFal4esOJ9mMoQVAmlD
wXC0EMjdXRK/mW88J/wmFRpDH7sKVE/2WunlDBoZtx6e+ee2ErAqGFJtonqip5c=
=/PKT
-----END PGP SIGNATURE-----
Rick van Rein
Systems that are based on IOU or tally's have one problem -- they
are founded in debt. This is what goes wrong in the current
banking system, according to the Austrian school of economy. A
very good book on that issue is
http://mises.org/books/desoto.pdf
A Dutch version also exists (many of you here, it seems),
with ISBN 9789033480942.
> Quite some smart people are thinking about alternatives and we should
> be part of that discussion instead of trying to come up with Yet
> Another System.
I've been thinking that too :)
> You should meet Georg Zoche, a good friend of mine, who has spent at
> least 10+ years in analyizing the current situation, where it went
> wrong and what axioms should be defining for a new and better system.
Georg probably knows the book above... and is likely to agree.
Cheers,
-Rick
Melvin Carvalho
> Hello,
>
> Systems that are based on IOU or tally's have one problem -- they
> are founded in debt. This is what goes wrong in the current
> banking system, according to the Austrian school of economy. A
> very good book on that issue is
It is possible to make the argument that all systems of debt will end badly.
However, almost all human interactions are based on some kind of debt
(explicit or implicit), or in fact doing favours for one another.
Sometimes it's not quantified (as often in families and small
communities) sometimes it is, e.g. with a common measurement system
aka money.
A decentralized system is generally more fault tolerant than a
centralized. In practice, you're probably going to need elements of
both. The central models are well established, but decentral less
well investigated.
A shared file space using unhosted could be a valuable start on the
decentralized side!
>
> http://mises.org/books/desoto.pdf
Thanks!
Jan Wildeboer
Hash: SHA1
On 10/18/2011 02:31 PM, Rick van Rein wrote:
> Georg probably knows the book above... and is likely to agree.
He also wrote his own book - Clash of Currencies, online at:
http://www.clashofcurrencies.org/
Don't dismiss it based on the somewhat in-your-face webdesign. It's
the content that counts.
Building a system based on Keynes Bancor principles is something that
is discussed on the highest levels in the financial world ATM.
If the FOSS world could build it, it would be a revolution to the
better ;-)
Jan
- --
Jan Wildeboer - j...@wildeboer.net - http://jan.wildeboer.net
Open Source Evangelist. Open Standards Fundamentalist.
Software Patents worst nightmare. Decentralizing whatever exists.
Citizen of the first Transnational Republic
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
iQEcBAEBAgAGBQJOnXbcAAoJEK4g/MdBcijwueoH/RUoIB5BcMD+3/RSh3ISJZPK
daQrwuypeg3n3T2dPNldAXX+wKJnmzD7tyFv5bhaiRy0nJmkSokw8G7SHntT/S9r
4tDSZ0QWHKF4bv5qSoeNk8QCdrXTRQ+0o07+bmD37U18bzEXCvPbdEBTTc8FppNd
nypmHVnsWUHP2SzNPEOEEV7Z3OKd5I8APZWrz21RTGboT/qDXb7c84sTcY4vK3b3
5tP0IA7/lXUfpaucKRPdE/0CAqyzm3BPKC900NpIba3BYbKfVsy2M6nCPAsUBzHw
iEJsW46aD41iuxeEOy/WQtek1ykN7xEK+xj5hCHX75Qpoim6gWo6nrmQFSsk5j4=
=L9Jr
-----END PGP SIGNATURE-----
Michiel de Jong
Quite some smart people are thinking about alternatives and we should
be part of that discussion instead of trying to come up with Yet
Another System.
ok, so one way to do that is to read all the links that have been mentioned so far in this thread; another is to try out all existing systems. i haven't read all the material that was mentioned so far yet. but so far, my question is: are there any systems, apart from bitcoin, that people are actively using right now?
my default option is to use human-written email:
1, me to you: "IOU - I, Michiel de Jong, hereby owe you, Jan Wildeboer, 10 euros." - ideally signed with pgp
2, you to me: "RE: IOU - OK, next month when I'm in Berlin we'll have a beer and you can give me the 10 euros"
3, me to you: "RE: IOU - Sounds like a good plan, let's do it that way."
So far, I have tried out bitcoin. my practical problem is i want to pay you the equivalent of 10 euros, and bitcoin doesn't solve that problem for me, unless i have been on there for long enough to have "won" 50 bitcoins in a scroll calculation. so as a newcomer to the bitcoin system, it does not (yet) fulfill this need for me, and does not replace the above option of human-written emails.
i looked up open-transactions, but i don't see any usable alternative there either (did i look in the right place?). i see screenshots, and install instructions.
my question is: who is currently using a system that /simplifies/ the sending of IOUs between people who already know each other, compared to the above sequence of 3 emails. i want to testdrive all systems that claims to be better for this than bare email. i am thinking either a website or a phone app.
as i said earlier, i don't mind using euros as a "language construct" for some of my IOUs. I just want to be able to leave IOUs "unpaid" in an easy click-click way.
unless we can find one that already exists, i want to launch a Proof-of-Concept this week, in time to catch some of the Guy Fawkes' Night and #occupywallst momentum. Just a click-click way to send pre-formatted invoice-, IOU-, confirmation-, and receipt- emails. a proof-of-concept. but one that actually lets me authenticate, specify recipient, amount and currency, and simply click 'send'. does this exist anywhere?
You should meet Georg Zoche,
please! that would be wonderful. i have so many questions about this. can you put us in touch?
Michiel de Jong
who is currently using a system that /simplifies/ the sending of IOUs between people who already know each other
some examples i found by searching for 'iou app' and 'iou web app':
web apps:
https://www.billmonk.com/
http://www.buxfer.com/
http://www.billster.net/
iphone apps:
http://www.miragebox.com/
http://itunes.apple.com/us/app/mr-moneybags-lite/id351656592?mt=8
http://itunes.apple.com/us/app/iou/id289106542?mt=8
http://www.macworld.com/appguide/app.html?id=71382&expand=false
http://astarsoftware.com/index.php?option=com_content&view=article&id=46&Itemid=53
http://appshopper.com/finance/iouapp
android app:
http://www.appbrain.com/app/i-o-u-i-owe-you/de.coderarea.android.iou
facebook app:
http://www.whatapps.com/2007/12/06/iou/
Melvin Carvalho
>
>
> On Tue, Oct 18, 2011 at 2:18 PM, Jan Wildeboer <j...@wildeboer.net> wrote:
>>
>> Quite some smart people are thinking about alternatives and we should
>> be part of that discussion instead of trying to come up with Yet
>> Another System.
>
> ok, so one way to do that is to read all the links that have been mentioned
> so far in this thread; another is to try out all existing systems. i haven't
> read all the material that was mentioned so far yet. but so far, my question
> is: are there any systems, apart from bitcoin, that people are actively
> using right now?
>
> my default option is to use human-written email:
>
> 1, me to you: "IOU - I, Michiel de Jong, hereby owe you, Jan Wildeboer, 10
> euros." - ideally signed with pgp
> 2, you to me: "RE: IOU - OK, next month when I'm in Berlin we'll have a beer
> and you can give me the 10 euros"
> 3, me to you: "RE: IOU - Sounds like a good plan, let's do it that way."
Human written is very limited. With a little extra overhead you can
capture the semantics and make it web scale
I think something like this works (using xml for convenience)
<IOU>
<payer>Alice</payer>
<payee>Bob</payee>
<amount>10</amount>
<currency>EUR</currency>
</IOU>
I think something like that and you're almost done.
Michiel de Jong
<IOU>
<payer>Alice</payer>
<payee>Bob</payee>
<amount>10</amount>
<currency>EUR</currency>
</IOU>
I think something like that and you're almost done.
exactly! that's why i'm so excited about this.for payer and payee, we should use full user addresses (user@host). and the whole thing should be sent over PGP.
IOU is when Alice sends this to Bob.
invoice is when Bob sends the same thing to Alice.
receipt is when Bob sends it to Alice, but with a negative amount (-10 euros, cancelling out the outstanding 10 euro IOU from Alice)
notification-of-payment is when Alice sends it to Bob with a negative amount (saying 'hey, i just paid you, so please send me my receipt as soon as you confirm it').
but i'm thinking each IOU should have a unique reference number - maybe made up of a milli-timestamp and with the restriction that you cannot make identical payments within the same millisecond. that way, you get a unique identifier, so you can reference that in other messages, which is how you would link IOUs to other content like comments about the reason the money is owed.
i'll review the list of apps i emailed earlier, but i would be surprised if any of them use pgp signatures. so the app would just be a tool to send pgp-signed messages of the above format. 200 lines of code tops, i would say (excluding pgp library).
Melvin Carvalho
There's two concepts here.
1. Account balance (shows your ledger)
2. Transactions (changes your balances)
Both are very similar concepts.
Balance is an overview,
Transactions have all the same fields as balances, but can have a
timestamp, a note, and unique id (if you can canonicalize it, you can
sign it)
Many different ways to implement, but you can have a queue of
transactions that get processed periodically.
I suggest if starting simple to tackle balances first, then move on to
transactions.
>
Thad Guidry
Melvin Carvalho
> IOUs are just BILLs or an INVOICE
> UBL - Universal Business Language - used throughout Enterprises has a
> general spec for this, which might be helpful for you to think through,
> Michiel :
> http://docs.oasis-open.org/ubl/prd2-UBL-2.1/UBL-2.1.html#BILLING
Thanks for the link.
Looks like quite a complicated spec, perhaps overkill for this idea.
It may be an idea to reuse some of the terms, or "embrace and extend",
but I dont know if OASIS licensing even allows this?
> --
> -Thad
> http://www.freebase.com/view/en/thad_guidry
>
Thad Guidry
Andrew Miller
http://www.nakedcapitalism.com/2011/08/what-is-debt-%E2%80%93-an-interview-with-economic-anthropologist-david-graeber.html
It explains how debt has taken many forms throughout civilization
history. On one hand, many early civilizations involved some form of
'gift economy,' in which the only 'debt' was an implicit social
expectation to reciprocate in some way. On the other extreme, many
civilizations used human slaves as a kind of tradeable asset to pay
off debts. Nowadays, we don't have debtors prisons, but we rely on
impersonal, market based, for-profit-and-not-socially-motivated
lending to drive our markets.
This #futureofmoney concept involves restoring subjectivity, social
context, and personal common-sense as the basis of our economic
activities. What would it take to reframe 'IOUs' so they're valuable
and trustworthy to our friends, and so they're desirable to use
(rather than scary and debt-based)?
Thad Guidry
Thad Guidry
Adrien P.
Andrew Miller
distinct from other kinds of economic transactions like purchasing and
loaning? The fact that gift-giving occurs at all is incomprehensible
from a classical economics point of view. It's a counterexample to #1
- sure there's greed, but there's also lots of generosity. What can we
do to amplify the generosity and cooperation, i.e. 'gift-giving',
while limiting exposure to damage from greedy people?
One inspiration comes from the 'Potlatch,' which is a sort of
competitive gift-giving event that serves as the primary economic
function of several Native American tribes http://en.wikipedia.org/wiki/Potlach
http://homepage.mac.com/allanmcnyc/textpdfs/mauss1.pdf
Maybe an Unhosted social-network-game that rewards, visceralizes, and
enhances the experience of 'gift-giving' (and 'gratitude') would be a
way to start the cycle along. Now as for defending against possibly-
systemic greed - that's at least partially a technical problem. Some
of the nicest features about bitcoin are its 'non-repudiation'
qualities (no one can deny your right to spend a bitcoin if you have
the matching RSA key). #bitcoin-otc uses a GPG web-of-trust to
facilitate trades between not-quite-strangers.
http://serajewelks.bitcoin-otc.com/trustgraph.php?source=amiller&dest=nanotube
These are on the right track, as they are tools for mitigating the
negative impact of dishonesty.
There's nothing that says 'greed' and 'generosity' are conserved
quantities. The awesome web we're working on can serve the function of
promoting generosity and isolating the effects of greed. 'Leaving the
market' might look a lot like 'leaving the cloud.' I'll have to think
more about this analogy. 'Greed' doesn't have to be the driving force
of economics any more than 'hoard the user data' has to drive the
development of the web.
On Oct 18, 12:19 am, "Adrien P." <m.pand...@gmail.com> wrote:
> http://p2pfoundation.net/Ripplehttps://ripplepay.com/
>
> http://p2pfoundation.net/Threebleshttp://www.threebles.com/
>
> ++
> Pandark
Michiel de Jong
<IOU>
<payer>Alice</payer>
<payee>Bob</payee>
<amount>10</amount>
<currency>EUR</currency>
</IOU>
I think something like that and you're almost done.
Jan and I started working on this - if you want to be a part of this, come join in in our chat room: http://webchat.freenode.net/?channels=unhosted
we picked a name as well, and started drafting some wire frames: http://Opentabs.net
Thad Guidry
Michiel de Jong
http://opentabs.net/lifecycle.png
Thad Guidry
Fabio Barone
http://troco.ourproject.org/
haven't tried it myself but reads:
"Troco is a set of wave extensions that aims to provide a decentralized complementary community currency system, that is, a peer-to-peer currency system. Also you can see it as an IOU or promissory note based system."
new on this list; Troco maybe no match but thematically related
Michiel de Jong
i don't think troco can replace Opentabs.net though, because it requires wave. The idea for Opentabs.net is to make it usable when the initiator has a remoteStorage account (which is our standard for decentralizing user data on the web), and the recipient has at least an email account.
thanks for the link!
Evgeni Pandurski
working on for 3 years. It is called "Circular Multilateral barter",
but it is more that a barter system. It is has most of the
functionality that you envison, plus there is an open marketplace for
trading IOUs (which is *very important* im my view). It does not have
the invoice and conter-invoice fuctionality. Also, I guess it is not
as easy-to-use as the concept you described, but the additional
functionallity makes this unavoidable.
You can check it out at https://www.multiswap.net/about/
You can download the source code here: https://sourceforge.net/projects/cmb/
The current implementation (which you see at https://www.multiswap.net/
) is fully centralized (although it allows everybody to start their
own CMB-server). But in principal, the system can be partially or
fully decentralized. I am sure you will find this paper interesting:
http://www.multiswap.net/doc/cmb-general.pdf
Regards,
Evgeni Pandurski
On 16 Окт, 17:22, Michiel de Jong <mich...@unhosted.org> wrote:
> the open web, freedom hacking, and... banking. :)
>
> DatenSpuren is a very interesting and inspiring unconference. One of the
> things i talked about with people is the idea of an unhosted IOU system.
>
> coin. i think that's not what we need (bear with me here...). Coins are not
> where the monopolies are. It's banks that have the monopoly. So instead of
> decentralizing money, i think we should decentralize "I owe you..." letters
> (IOUs). IOUs are the fundamental transport layer of economy. On top of that
> is money. On top of that is lending. In fact, bank notes started out as
> written IOUs. What if we define a protocol for IOUs, on top of
> remoteStorage. Is there such a protocol already?
>
> PGP-signed, but also manually checked and reconfirmed by both parties
> (everybody ultimately is responsible to check their own monthly statements
> and see if it's all correct).
>
> an IOU would need an amount and a currency. I owe you a beer. I owe you 20
> euros from the other day. Initially, this would be something you use among
> friends. Somebody at jsconf (i forgot who it was) showed me an app they used
> to keep track of who paid for the badminton court that week, and who was
> there. Simple bookkeeping between friends. It would be an awesome thing to
> have in a phone app (i'm sure people have already experimented with apps for
> this use case). fill in the names of 4 friends, one person paid for the gas,
> the other each owe that friend 13,33 euros. The person who pays for the gas,
> while getting back into the car, opens his Unpaid app, selects amount '40',
> currency 'euros', and clicks 'take one for the team'. the others in the car
> get an alert 'Ben is claiming 13,33 euros from you [agree] / [dispute]'. if
> you click 'agree', you PGP-sign an IOU from you to Ben. Or, if Ben owed you
> money from last time, it says 'substracted, Ben and you are now even.' - you
> get the point. in this last case, Ben would get a receipt from you instead
> of an IOU, because you're cancelling a pre-existing Owe-A-Bro.
>
> both parties, and show up in the history of both parties, so you don't get
> misunderstandings.
>
> in case misunderstandings arise, you resolve them out-of-band.
>
> holiday, I still owe Peter 320 euros (he paid the car rental, and i didn't
> buy any of the beer), so i make a POBS (plain-old-banking-system)
> transaction to Peter.
>
> also dispute resolution would be out-of-band. if we meet through a friend of
> a friend (or through e-bay for that matter), and when we meet so i can buy
> your Stratocaster, we get out our phones and i send you an unhosted
> ("unpaid" ;) IOU. i'm basically saying "I hereby sign a contract, saying
> that I will pay you for the guitar within reasonable time". and then i
> don't. then you can take me to court in either the country where you live,
> or where i live. it's the same as a rental contract. or a labour contract.
> or any other civil transaction. you just sign "yes, i will pay". that in
> itself is not the real transaction.
>
> you could build clearing houses on top of this, at a higher level. and you
> could build lending with interest (the IOUs would need a pay-by-date added
> into them: "I agree to pay you back before the end of the year").
>
> you could use dollars, or yens. and you could also write your IOUs in
> bitcoins. or in kWh (electricity is probably one of the most universal goods
> of our times, as was suggested this morning).
>
> your private key on your remoteStorage. in case your remoteStorage gets
> compromised, people would be able to send whatever type of messages in your
> name, and it wouldn't legally bind you. also, you would get notified on your
> statements - weird payments would show up. and you would have to go and
> dispute them. your private key could also be written on a physical card that
> you keep in your wallet. if you lose your wallet, you quickly log in online
> to generate a new keypair (block your creditcard and request a new one).
> maybe you want to pay 2% of all your payments to a disputes-service, so that
> you can get your money back should someone steals your "creditcard". all of
> this would be /on top of/ the actual IOU-transaction system.
>
> spirit of the web, and independent of any currencies or dispute-resolution
> hubs or even interest rates, and make easy phone-apps for it so people can
> "unpay" each other, we could decentralize some of the power inherent in the
> current banking system. taxes don't apply here, because it's not the real
> transactions that are happening. tax would be levied when the actual payment
> ensues, which from this system's perspective, is out-of-band. And POBS
> transactions would be the default way to clear IOUs, but no longer
> necessarily the only one.
>
> let me know if you see any flaws in this. also, let me know if you are the
> person who showed me the badminton app at jsconf. :)
>
> Michiel
Ale Fernandez
Michiel de Jong
If we can make opentabs usable for microcredit, then we can improve and decentralize access to the economy on a peer-to-peer basis, and will essentially have achieved "decentralization of opportunity".
but let's try to get the basics working first. sorry for the delay in launching opentabs.net alpha, i'm working hard to get it ready soon.
Hi,It's been really great to discover this thread and see so much information on this kind of thing. I'd just like to suggest that the data that comes from it,
first of all we would have to decide that we even want to log secondary data. i think that should be at most opt-in. people will only pay for secondary data if it can be used for targetting. so a user could opt-in to post their data to a targetted ads server, and receive ads. either the user could get the money earned by displaying the ads back, or, what probably makes more sense, just get very good offers.
a personal friend of mine just started an interesting business based on this sort of idea: http://www.sitepackers.com/ (in Dutch) points potential customers to e-commerce sites, gets money for that, and shares 75% of that commission back with the person clicking.
as we make the web more open and free, banner ads can become opt-in in this way. this also ties in with the whole Vendor Relationship Management (VRM, as opposed to CRM) movement.
rather than being siphoned away by a central organisation as happens these days, can be shared responsibly and democratically amongst the members of the barter or exchange network.By knowing about other transactions and having a say in them, you can then achieve one of the objectives of participatory economics or parecon - self management. This could be as simple as feeding back anonymous data on what is being bought and sold across an area, and training participants on how to interpret that data and visualise it, as that could then help to see what the effect of certain shortages or other problems would have on the whole thing.Ale
you mean establishing market prices without the use of a centralized clearing house? interesting! although after thinking about it for a bit, you would still need a market place, and this market place would be a very logical place to also act as the statistics-publisher. we can see what kind of structures and activities can be built on top of opentabs once we have it working. stuff like that is definitely interesting!
cheers,
Michiel
Ale Fernandez
El domingo 20 de noviembre de 2011 13:21:20 UTC+1, Michiel de Jong escribió:
By knowing about other transactions and having a say in them, you can then achieve one of the objectives of participatory economics or parecon - self management. This could be as simple as feeding back anonymous data on what is being bought and sold across an area, and training participants on how to interpret that data and visualise it, as that could then help to see what the effect of certain shortages or other problems would have on the whole thing.Ale
you mean establishing market prices without the use of a centralized clearing house? interesting! although after thinking about it for a bit, you would still need a market place, and this market place would be a very logical place to also act as the statistics-publisher. we can see what kind of structures and activities can be built on top of opentabs once we have it working. stuff like that is definitely interesting!
cheers,
Michiel
Michiel de Jong
In terms of use of data, I'm more interested in the ethical aspect of people being aware of possible problems in an economy and dealing with them with mutual aid, solidarity or just by being organised.
right, i was assuming we would still use the invisible hand for planning, but maybe it makes sense to rethink that part as well!
There is also the ideal of transparency(for example if you knew something radioactive was being traded in your street, you might want to have a say in it), but I think that needs a lot more thought. I'm only talking about voluntarily opting in, as you say. For example you might choose to share production data to trusted people, or as anonymous stats, knowing that they depend on the product of your labour or of others in your area or sector.
that sounds like a really good concept! a bit like open government (open data), but applied to economics.
it would be like opt-in communism, in a way. so far, i had thought of Unpaid/Opentabs as anti-banking, not anti-capitalism, if that makes sense. So moving the power away from banks, but not necessarily moving the "routing algorithm" of the economy away from "the invisible hand".
i myself have a weird perception of money as a reward for work, which only a very small percentage of people i talk to share. most people want to earn money proportionally to the value of their gross product (whether that is through one of the POBS currencies or through bitcoin or "the bank of time" credits or what-have-you). in fact, i think civilization will evolve towards a point where everything you need is pretty much free (as in, is so cheap, that you will not let your choices be lead by money). an example of this that people can maybe relate to: if you want to sit down for a coffee somewhere, do you go look at the menu first to do a price comparison? Or do you choose the place without knowing the price information, or even factoring in the price information in your choice?
if your choices are not influenced by money, then that means your quality of life is not influenced by money. you just have "enough" money. that means, in turn, that when you're looking for a job, you will choose a job you like, or a job you think you will be good at, and your job search is influenced by differences in salary only very slightly; you just pick your preferred job out of the pool of jobs that earn "enough". Open source software, as well as art (e.g. poetry) are prime examples of economies that have no money-based resource routing.
I hope that explains it a bit!
yes, it does! i hadn't really understood you before, but i do now, i think. i'm also not an economist :)
I'm a techie not an economist, but I'd be very interested in helping develop functionality like this!
splendid!! :) i sent you a separate off-list email about that.
Cheers,
Michiel
Jan-Christoph Borchardt
On Fri, Nov 25, 2011 at 1:23 PM, Michiel de Jong <mic...@unhosted.org> wrote:
> most people want to earn
> money proportionally to the value of their gross product
>
> if your choices are not influenced by money, then that means your quality of
> life is not influenced by money. you just have "enough" money. that means,
> in turn, that when you're looking for a job, you will choose a job you like,
> or a job you think you will be good at, and your job search is influenced by
> differences in salary only very slightly; you just pick your preferred job
> out of the pool of jobs that earn "enough". Open source software, as well as
> art (e.g. poetry) are prime examples of economies that have no money-based
> resource routing.
See »The surprising truth about what motivates us« by Dan Pink:
http://www.youtube.com/watch?v=u6XAPnuFjJc
(might / should be pretty well-known)
»As long as the task involved only mechanical skill – bonuses worked
as they would be expected: The higher the pay, the better the
performance. […] But once the task called for even rudimentary
cognitive skill – a higher reward lead to poorer performance.«
»Pay people enough so the issue of money is off the table.«
»To motivate and satisfy people, we need to maximize purpose instead of profit.«