On 6/11/12 02:14 AM, danimoth wrote:
On 03/11/12 at 07:32pm, ianG wrote:
Better to avoid that and come up with a payment system that doesn't
need reputation - or at least one that doesn't lean so heavily on
it.
I don't know how split these problems. In everyday life, when I use a
payment system (like cash) I accidentally use a reputation system: for
example, I know the reputation of sellers, and buy things from high
rated ones. The same applies to the sellers: they are more inclined to
sell to high rated buyers (but losses happen anyway).
Right - the money you use in real life is entirely independent of the
reputation of the seller. So when you do a real-life trade, you have to
predict the reliability of the result of the trade from somewhere else.
Reputation, size, nature of trade all play a part.
Which leads to a second observation: there is a huge difference between
a payment and a trade. So much so that conflation in design will likely
be fatal. Here's a rant that explores what trade is:
http://financialcryptography.com/mt/archives/001247.html
as compared to a payment, amongst other things.
My two points here being perhaps:
* you need both a payments system and a trade system
* the payments system should be entirely independent of the trade
system (which doesn't imply the converse).
The only way which seems, at this point, to be able to avoid the use of
reputation systems is a fair exchange. A real life example is easy to
setup.
In real life, you can exchange cash for goods, because (a) you are
physically in the same place and (b) both parties will engage in the
process to the full, because of rule of law. On the net, the former
isn't true and the latter is not essentially true. If you model it out
as a protocol, you'll find that it is a non-trivial problem to do a
reliable exchange.
[1] Enforcing Collaboration in Peer-to-Peer Routing Services
(by Tim Moreton and Andrew Twigg)
That's an unfortunate turn of phrase there, which rather strikes at
the heart of the problem you are trying to solve :)
It focuses on the routing layer of p2p networks, using a reputation
system. What I propose is complementary to that approach: users are
incentived to route well by their work, and they are incentived to
stay active (e.g. share) because they get money in exchange on mine.
So, their work should be extended to include payments on
the computation. Or I misunderstood your statement?
Well, 'Enforcing Collaboration' is a contradiction in terms.
Collaboration is a voluntary thing, it cannot be enforced, as then it
would be compliance.
Incentivising Collaboration might have been a closer to what you wanted
to convey. But sometimes drama in titles is a good thing.
Anyway, maybe I catched the whole point after reading that from James A. Donald:
But yes, there is no point in fixing a problem which is hard to solve
with a solution to a problem that no one has yet adequately solved.
A last question come up to my mind: what about fixing the price? Maybe better
than nothing, without consider fair exchange problem.
Fixing price works well for say Second Life. That works if the prices
of the goods can vary independently and frequently, as James pointed
out. It works well if you are the supplier of the good or otherwise in
control of some important component.
It doesn't work well if you are envisaging many suppliers and many
consumers. Many of those suppliers will prefer a price of zero, many
will hope for a price well above what you set. Consumers equally will
not be happy, so you'll split both consumers and suppliers along price
lines.
iang
_______________________________________________
p2p-hackers mailing list
[email protected]
http://lists.zooko.com/mailman/listinfo/p2p-hackers
