First, I would rather pay a single entity that I'm familiar with that pay
some unknown stranger. Of course, I wouldn't want a corrupt corporation
running micropayments, and there are many individuals that I know and trust.
But I'd have a hard time convincing myself to pay an absolute stranger.
(This is assuming that payments for transfers are direct from node to node).

Second, I don't want to completely disregard the promises of this kind of
system. If each node charged an amount for data (flat rate for all data
types, of course; should rates vary by the amount of data or what? Premiums
for speed?), then the requesting node would fork over that amount of money
for each transaction. Let's put this into an scenario:

I make a request from my node. My node contacts another node, which doesn't
have the data. The second node, Node 2, requests data from Node 3. Node 3
has the data. It charges $0.001 (1/10th of a cent) for each unit of data,
which is, say, 5 1500 byte chunks (we could have a standard splitfile/packet
size to equate payments?). That would be a total of $0.005 for that
transaction. Node 3 tells this to Node 2, and Node 2 forks over the digital
cash. Node 3 sends the data to Node 2. Node 2 tells Node 1 (my node) that
each datum will cost $0.001. Node 1 says okay, sends the cash, and Node 2
sends that data.

Of course, what if Node 2 wanted to charge $0.002 for every datum? Well, if
Node 1 was willing to pay that much, then the transaction would be processed
and the exchange would occur. If Node 1 thought that to be too much, then it
would try to get the data from somewhere else.

So, what is this money? I don't know much about how digital cash works. Is
it possible to create a secure currency that cannot be copied and is still
anonymous? Maybe it doesn't need to be truly anonymous if it works in the
way that Freenet works. Nobody needs to know what the data was, per se. They
shouldn't know, because certain data types should not cost more than others.
All data should cost the same.

How will the money be "verified?" Here's an idea: Individual pieces of
currency are licensed to individual entities, marked by their private key or
whatnot. The pieces of data (I'll call them "credits" just to be creative)
are stamped with the identity (The public or private?) of the original
licensee (which is encrypted with the "banks" public key (only the bank can
read it with it's private key).

Now comes the fun part. The currency is transferred to another party in a
transaction. So, should that node "verify" the currency with a central bank
(a fail point, for sure), should it verify only with  non trusted stranger
nodes (less bandwidth), or what? Would it be better for the currency to die
once it was transferred once (this could waste a lot of bandwidth)? Any
ideas on how currency should work? I know there's a huge body of information
out there. Who knows a lot about digital cash?

So, what do we do about bad nodes? Do we blacklist known bad currency
providers, or should each individual node maintain a personal blacklist of
nodes that it has had bad experiences with (I like this idea better).

Well, that's my two cents.

Wesley.

> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Jim
> Sent: Friday, 05 July, 2002 15:23
> To: [EMAIL PROTECTED]
> Subject: [Tech] Re: Mailing list and Micropayments
>
>
> In the micropayment scheme I'm thinking of, each node would determine the
> prices it would charge for insertions and retrievals.  There
> would not be a
> central organization that would charge people to use freenet.
> There wouldn't
> even have to be a single 'bank' since various 'currencies' could
> be exchanged
> freely.  Also, people wouldn't be required to use cash or credit cards to
> accumulate these currencies.  They could simply set up a node and collect
> fees for insertions and retrievals.  Then they could use that to make
> insertions and retrievals from other nodes.  In short, no one
> would be forced
> to use micropayments, there need not be a single point of failure and
> anonymity should not be compromised.
>
> > I'm no core developer, but I do have an option on micropayments. I think
> > they'd be a great way to raise revenue, and they might help move Freenet
> > development forward a little bit. Also, I'd be willing to pay something
> > (aka, donate) and I'm thinking about doing so. However, not
> everybody has a
> > lot of money to spare, and I think that the contributions
> people are giving
> > that don't include money are already doing a great service for Freenet.
>
> > In English, I don't think that micropayments are a good idea if they are
> > mandatory. However, I think that Christopher's idea of having
> two divisions
> > of Freenet, one for the general public, and one for contributors, might
> > work, and is at least worth a shot to see if it works. I'm
> willing to give
> > it a try. Maybe.
>
> > However, I think that it's important to keep Freenet free. As
> in gratis. At
> > least to use. I think there should still be a public Freenet,
> and I think
> > that the people who run micropayment nodes should still
> contribute some of
> > their node to the general public Freenet. It shouldn't all go
> to some sort
> > of pay network.
>
> > I think that money would make more people want to run a node,
> but I think it
> > would make less people want to use the network. And I'm not sure that
> > micropayments are necessary for the network to actually work.
>
> > Again, I'm willing to give it a thought, and maybe a try.
>
> > Wesley.
>
> _______________________________________________
> Tech mailing list
> [EMAIL PROTECTED]
> http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/tech
>



_______________________________________________
Tech mailing list
[EMAIL PROTECTED]
http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/tech

Reply via email to