You are probably right, some market mechanism will arise, I just can't imagine yet how the propagation of transactions could be monetized. Bitcoin does not provide a mechanism for that.
My guess is that online wallet services (like instawallet etc.) or similar services will have an incentive to speed up the transaction propagation, but then again they might choose to only propagate transactions that their clients are involved in.
So the generalized question is: could the incentive to leach by choosing to propagate none or only a subset of the transactions become dangerous to the bitcoin network? why should any "rational" node propagate any transaction that does not directly benefit you?

It is possible that being a regular node will not be so easy when the bitcoin network has grown and a node needs to receive and forward thousands of transactions per second. So the only ones that will have the incentive to be full nodes will be the miners. And everybody else might be using a "light" client that does provide this functionality to the bitcoin network.

In the not so distant future when the miners reward consists mainly of transaction fees, what is the incentive of a miner to forward received transactions to other miners?
To clarify, imagine you are a big miner (maybe a pool operator) and a transaction comes in from another node with a (considerable) fee. If you do not forward this transaction to the nodes you are connected to, you will reduce, admittedly by just a tiny amount, the chance that some other miner will include this transaction in the next block and reap those rewards if they find the block before you. By keeping it to yourself you can include the transaction in the next block that you will find.
If only one miner does this, the effects will be barely noticeable given the interconnectedness of the bitcoin nework, but if many miners have this behavior, could this potentially be a problem?

3

I am trying to figure out the size of the bitcoin economy and I have 2 questions for the bitcoin analyzers:

1) Does anybody have any idea how many bitcoins are actually used to buy products and services rather than buying other currencies? Any close approximation of that would provide a good insight of the utility of bitcoin. Perhaps this could help in understanding whether the recent peak is a bubble or not.

2) Has anybody analyzed the bitcoin transaction chain to see what percentage of the existing bitcoins are actually involved in transactions after their creation? Or, in other words, can one see how many bitcoins are actively participating in the bitcoin economy? Perhaps it is possible to see the percentage of not exchanged bitcoins in the last x months or days.

Nice calculation. certainly it is not worth building a machine for bitcoin creation. However if you already have this computing power sitting idly then you should calculate only the costs of the extra electricity that you need to consume to generate a block. Has anybody done this?
If you want to build a bitcoin generation machine you are probably better off buying a FPGA board and designing a dedicated hash calculating hardware monster.

Interesting note: If one puts 4100000 hash/sec in the calculator then you get 10 minutes average.
10 minutes is the actual target average block production rate.
This means that currently the the total computing power working on solving hashes is 4.1 Mhash/s
or about 900 quad cores. or 3-4 thousand laptops. Pretty impressive!

This means that there are some people who think it is interesting to produce bitcoin.
Who knows why, I stopped several difficulties ago.

There has been a similar thread elsewhere: http://bitcointalk.org/index.php?topic=721.0;all

I guess the idea of producing useful work is good if it could be implemented. But this is difficult since there must be a "proof of the work" in order to earn your bitcoins. The question is who is going to provide this proof?

Proof of work bitcoins should satify (at least) two criteria:
a) must be verifiable by everybody (otherwise the verifying entity could gain control of minting if it turns malicious).
b) must be much easier to verify than the work itself (otherwise too much work would have to be done for verification)

I dont think the algorithms which run in the the BOINC client satisfy these criteria.

The "proof of work blocks" of bitcoin satisfy these criteria. But if you could find an algorithm that does something useful while at the same time satisfies these criteria then i suppose it could be included into the client. Maybe something like calculating the nth Pi digit, which is useful (for mathematicians I guess) and difficult, but i have no idea if it is satisfies the 2nd criterion. Also I don't know technically how the transactions should be included in the work.
But I believe that is is theoretically possible to create other useful work except the obvious which is the creation of bitcoins.

Not necessarily, digging out a little bit more gold just makes the existing gold a bit less valuable since the supply is increased. That is because its value is mainly derived from its scarcity. Its utility IS its scarcity. However, it is not a waste for the miner if the cost of mining can stay below the price of gold.


I suppose that if anybody (as opposed to a single trusted entity) can produce bitcoins then the production costs will converge close to value of the produced bitcoin itself. If that were not true then it would make sense for someone to invest resources to produce as many bitcoin as possible. Since we dont want to let the value of bitcoin to go down then the production costs need to rise to the value of bitcoin.

As far as I understand, this is not entirely true. See http://www.bitcoin.org/wiki/doku.php?id=transaction_fee

The minting should be seen as digging for gold in a gold mine. This is similarly wasting energy and resources for actually not doing anyone any favor since you increase the supply of gold. Except for yourself of course, and that why people will always will dig for gold as long as the price of gold dug will be higher than the cost of resources needed to dig. However, this difficulty is what gives scarcity and therefore value to gold.

Similarly, if and when bitcoin becomes more mainstream and follows economic sense (rather than beeing driven by curiosity, fun and excitement for participating in something new) the cost of electricity (in dollars or equivalent bitcoins) to create a bitcoin will never exceed the value of the created bitcoin. If that would happen, many will shut down their bitcoin generation for it would make no economic sense (I already stopped mine!). Luckily, and unlike gold mining, this behavior will adjust the cost of creating bitcoin to a sensible level (below the value of the generated bitcoin)

In the long run i expect that bitcoin minting will indeed be left to the supernodes which will probably have specialized hardware  that are excellent in hash calculations and can do that way better than anyone with a general purpose pc. Similarly to specialized gold digging machines.

So although you are right that it would be nice to find a system that provides scarcity without cost (I dont know whether this is possible, it seems not), the costs are certainly connected and less to the value of the created bitcoin. And it is a self regulated system that will increase the cost of creation only if the use of bitcoin actually adds value (in the form of ease of use, anonimity etc.) and therefore is worth it.
The cost of creation is therefore is no limitation of the growth but more and indication of the added value of the use of bitcoin.

fuuny note: Following this logic the cost of minting will be somewhat less than 21million bitcoins which with the current prize of bitcoin is  around 1 million dollars.

------
1neNcqV4yRUNwP9xCdcgRvBrNwWdFo3Wv