Alternative Value Exchange. Anybody interested?

[Michael Grube]

Hey All,

As most of you are probably aware, alternative currency has been becoming a more and more popular topic in the last few years. I see people get excited about BitCoin and  other things, but nobody seems to have implemented a practical system for exchanging value with these alternative currencies. I guess there's an debit card coming out that's supposed to support bitcoins, but why settle for something so limited? BitCoin has its own problems. 

So it'd be nice if we could make a testbed for experimenting with different value exchange systems. It'd be even nicer if I had anybody to work on this with. I'm thinking we'd set up some kind of physical way to exchange value(maybe swipe cards or rfid tags or whatever) and an electronic system to track the exchanges being made.

There is no better place to play with the future of economics than a makerspace.

Mike

[Michael Grube]

Oh hey, I made an incomplete Wiki page.

http://wiki.allhandsactive.com/Value_Networks

[Ryan Hughes]

The bitcoin app for android can initiate transactions with NFC or QR
codes.

https://code.google.com/p/bitcoin-wallet/
http://f-droid.org/repository/browse/?fdid=de.schildbach.wallet&fdpage=3
https://play.google.com/store/apps/details?id=de.schildbach.wallet

You have to have internet available on both phones.

Any other way of doing it, you're going to have to rely on a central
authority to issue the money, and/or track spending, and/or prevent
double-spending, and/or prevent counterfeiting. It's been done. A number
of "local currencies" exist. They usually issue cash in the form of paper
money.
https://en.wikipedia.org/wiki/Local_currency


Now, what *I*'d like to see is a way of computer-assisting the bartering
process.

Someone puts up a thing that's like "I will cut someone's hair but I want
a new blender", and then someone else puts up "I will give someone my
blender but I want a home-cooked dinner" and someone else is like "I will
make someone a home-cooked dinner but I want my hair cut" and then the
computer can get all these people in contact with each other. With a
website, and with a lot of people using it, you could compute long chains
of barter that wouldn't be obvious to people.

Let's cut out the middleman (money) and exchange value directly! by
inserting a new middleman (this website)!

--Ryan

[Michael Grube]

On Thu, Oct 25, 2012 at 3:33 PM, Ryan Hughes <ry...@iheartryan.com> wrote:

The bitcoin app for android can initiate transactions with NFC or QR codes.

https://code.google.com/p/bitcoin-wallet/
http://f-droid.org/repository/browse/?fdid=de.schildbach.wallet&fdpage=3
https://play.google.com/store/apps/details?id=de.schildbach.wallet

You have to have internet available on both phones.

Any other way of doing it, you're going to have to rely on a central authority to issue the money, and/or track spending, and/or prevent double-spending, and/or prevent counterfeiting.

I strongly disagree. Cryptography and a meeting in person should be enough to ensure authenticity.

 

 It's been done.  A number of "local currencies" exist.  They usually issue cash in the form of paper money.
https://en.wikipedia.org/wiki/Local_currency


Now, what *I*'d like to see is a way of computer-assisting the bartering process.

Ripple is a step toward that.  

Someone puts up a thing that's like "I will cut someone's hair but I want a new blender", and then someone else puts up "I will give someone my blender but I want a home-cooked dinner" and someone else is like "I will make someone a home-cooked dinner but I want my hair cut" and then the computer can get all these people in contact with each other.  With a website, and with a lot of people using it, you could compute long chains of barter that wouldn't be obvious to people.

Let's cut out the middleman (money) and exchange value directly! by inserting a new middleman (this website)!

I'm not making a website. 

[Michael Grube]

Lurk moar before you naysay. Computing can happen on a p2p basis too, you know...

[Ryan Hughes]

> Any other way of doing it, you're going to have to rely on a central
> authority to issue the money, and/or track spending, and/or prevent
> double-spending, and/or prevent counterfeiting.
>
> I strongly disagree. Cryptography and a meeting in person should be enough to ensure
> authenticity.

How does the protocol work? I'm imagining a doohickey that stores
transactions on it, and has the current value of the doohickey.

I can sign a transaction and send it to you: "I hereby bequeath $20 unto
you". You see my signature, and you see that my pubkey is signed by a
trusted authority, and so you accept my $20. If you try to give that $20
to somebody else, they'll see that it has your signature, your signature
is valid, and it has my signature, and my signature is valid, so they
accept it.

But what if I "forget" to write down that I gave that money to you, and
then I give it to somebody else? Then one day, bob tries to give carol
$20 and carol says "you can't give me that particular $20. I already have
that $20!" and then I, in the shadows, go "muahahahaha".

Solutions to this problem that I know of are:
* Have physical objects that are gone when you spend them (gold, cash).
In the case of cash, a central authority has to prevent counterfeiting.

* Have a central authority verify each transaction, and disallow
transactions where the spender doesn't have the money they claim to.
Bob won't accept a transaction from Alice until the authority verifies
it. Requires a connection to the internet and a central authority.

* The bitcoin block chain, which implements the central authority in a
peer-to-peer fashion. Requires a connection to the internet.

Do you have another plan?

> Now, what *I*'d like to see is a way of computer-assisting the bartering
> process.
>
>

> Ripple is a step toward that. �

You mean http://ripple-project.org/ ?

I will check that out soon, thanks.


> I'm not making a website.�

Wise decision.

--Ryan

[Ryan Hughes]

> Lurk moar before you naysay. Computing can happen on a p2p basis too, you know...

I'm not saying nay to you. I'm saying yea to bitcoin!

But if you have another p2p authentication plan, I'd love to hear it, cuz
p2p stuff is awezzome.

--Ryan

[Michael Grube]

On Thu, Oct 25, 2012 at 4:29 PM, Ryan Hughes <ry...@iheartryan.com> wrote:

      Any other way of doing it, you're going to have to rely on a central
      authority to issue the money, and/or track spending, and/or prevent
      double-spending, and/or prevent counterfeiting.

I strongly disagree. Cryptography and a meeting in person should be enough to ensure
authenticity.

How does the protocol work?  I'm imagining a doohickey that stores transactions on it, and has the current value of the doohickey.

Everybody has a doohickey. It's honestly a bit complicated to explain here. We should organize a Wiki page.

 

I can sign a transaction and send it to you:  "I hereby bequeath $20 unto you".  You see my signature, and you see that my pubkey is signed by a trusted authority, and so you accept my $20.  If you try to give that $20 to somebody else, they'll see that it has your signature, your signature is valid, and it has my signature, and my signature is valid, so they accept it.

Cash/currency is an antiquated tool that is ready to be replaced by something that can better represent the current value of something. We have data about many things in real time and the amount of this kind of information is going to continue to grow. Value is pretty subjective and we can have something that reflects this better than "dumb" currency. I don't know what I'd call the replacement(other than a value exchange network where value is determined by a set of agreed upon rules in a given social scenario).

Here's the thing: computing itself should be social. I'm not saying people should share the fact that they're pooping more, I'm saying that computing resources could be used more effectively if they proportionately represented the interests of the people who owned those resources(and others, who would agree to carry out computations with/for them).

 

But what if I "forget" to write down that I gave that money to you, and then I give it to somebody else?  Then one day, bob tries to give carol $20 and carol says "you can't give me that particular $20.  I already have that $20!" and then I, in the shadows, go "muahahahaha".

Solutions to this problem that I know of are:
* Have physical objects that are gone when you spend them (gold, cash).
  In the case of cash, a central authority has to prevent counterfeiting.

* Have a central authority verify each transaction, and disallow
  transactions where the spender doesn't have the money they claim to.
  Bob won't accept a transaction from Alice until the authority verifies
  it.  Requires a connection to the internet and a central authority.

* The bitcoin block chain, which implements the central authority in a
  peer-to-peer fashion.  Requires a connection to the internet.

Do you have another plan?

So it turns out that human social networks really do form a "small world" network topology. So you can use trust between individuals to build a global network and keep things "local" at the same time.

More later.

[Alexander Honkala]

Just pointing out that if a concept or a thing is difficult to summarize in text in an email, how is it going to be any easier to summarize on a Wiki page?

-X

[Michael Grube]

On Thu, Oct 25, 2012 at 9:06 PM, Alexander Honkala <alexande...@gmail.com> wrote:

Just pointing out that if a concept or a thing is difficult to summarize in text in an email, how is it going to be any easier to summarize on a Wiki page?

It's not about the size, it's about putting the idea in a place that is easy to reference and modify as needed. Email is not exactly the most convenient format to store and build out ideas. It's like asking why you'd use a google doc if you have email.

[James Valleroy]

      Now, what *I*'d like to see is a way of computer-assisting the bartering

      process.


Ripple is a step toward that.  

You mean http://ripple-project.org/ ?

I will check that out soon, thanks.

I believe Ripple avoids the issue of duplicating currency, because Ripple is not a currency--it's an IOU. If I understand it correctly, Ripple "payments" (cancellation of IOUs) can only be made along chains of trust relationships. The idea is that small-world network effects would allow you to have at least one trust-chain to any random merchant on the web.

The main issue I see with decentralized Ripple is that IOUs in different social networks can have different real-world values. I guess the current approach is to set up IOU exchanges, but how do you determine the relative values of IOU units distributed across the net? It might be a lot simpler to just tie IOUs to a real currency such as the US$.

[Michael Grube]

On Sat, Oct 27, 2012 at 10:24 AM, James Valleroy <james.v...@gmail.com> wrote:

      Now, what *I*'d like to see is a way of computer-assisting the bartering

      process.


Ripple is a step toward that.  

You mean http://ripple-project.org/ ?

I will check that out soon, thanks.

I believe Ripple avoids the issue of duplicating currency, because Ripple is not a currency--it's an IOU. If I understand it correctly, Ripple "payments" (cancellation of IOUs) can only be made along chains of trust relationships. The idea is that small-world network effects would allow you to have at least one trust-chain to any random merchant on the web.

This is correct.

 

The main issue I see with decentralized Ripple is that IOUs in different social networks can have different real-world values. I guess the current approach is to set up IOU exchanges, but how do you determine the relative values of IOU units distributed across the net? It might be a lot simpler to just tie IOUs to a real currency such as the US$.

Could you explain this more? I want to step through an example and maybe you can correct me:

A needs a chair. B has a chair. A has no money, so creates an IOU for B to...I don't know... make a pot for them. Let's keep it simple and say that the thing being owed here is the pot that A could potentially give. 

Now let's look at a case where B is friends with C but not A. B now has an IOU from A offering a pot. If B wants to use A's IOU as credit for C, it is between C and A to barter about what C will give B in exchange for A's pot. The problem is solved at each step. Maybe the pot has a lesser value to C than it does to B, but it is still something to barter with. I suppose I just don't see where the issue is - B must simply offer something more. 

[Michael Grube]

Correction: It is between C and B to barter about the value of the pot, not C and A.

[Alexander Honkala]

Keeping authentication as only F2F will mean that whatever system you make will always stay very small and static.

-X

[Michael Grube]

On Sun, Oct 28, 2012 at 11:56 AM, Alexander Honkala <alexande...@gmail.com> wrote:

Keeping authentication as only F2F will mean that whatever system you make will always stay very small and static.

Small? Did you not see what I wrote about small world networks? I know people that know other people. I don't have to communicate with the directly.

What do you mean static?

[Michael Grube]

My list of people I barter with directly is very small, but the network remains very large. I do not need to change my connections often, but I can if I must.

http://en.wikipedia.org/wiki/Small-world_network

[Michael Grube]

I feel that Xander brought up an important point that needs clarification. In a network like this, the *routing* is f2f, meaning I can communicate with many people without needing to talk to them directly.

[Alexander Honkala]

We're at the point in which interested parties have likely chimed in, time for ahasho...@googlegroups.com so that we don't keep spamming 400 lovely people with this thread.

-X