It's theoretically possible for transactions per block to become so high, that in order to get in a block at all you have to put in a transaction fee, right? Well what happens, in that environment, if I put a transaction fee of 0, and my transaction is never ever picked up? How would I correct that situation?
If I increased the transaction fee amount, and tried again, and that one went through, would an attacker then be able to create his own block, with the first payment inside? Or is there a way to revoke transactions permanently before they end up in a block? Is there a 'deadline' on a signed transaction? i.e. 'after block X, this transaction would not be valid'?
- See https://www.bitcoin.org/smf/index.php?topic=3411.0 theymos 20:13, 17 February 2011 (GMT)
- That link is broken. Swiftpaw 23:08, 4 September 2011 (GMT)
Transactions are confusing me as well. Along with the above question which was a great one to ask, another one I have is that if transactions are the incentive for miners to create blocks in the block chain, and blocks which aren't in the chain aren't verified to have occurred and thus prevent counterfeiting, then what happens in the future when it becomes incredibly hard to create blocks? If transaction fees are designed to deal with this problem, how can they do that when the difficulty of mining blocks increases exponentially? Wouldn't that mean that the fees will get unbearable, or do I have the exponential function labelled backwards? If the response to that, which would create a huge problem, is that instead the fees vs. computing power costs mathematically designed to level off at a comfortable ratio, at least given the future predictions of electricity + computing costs, and the difficulty of creating blocks stops increasing very much as you approach infinity on the exponential curve, won't that mean that computing power has a chance of vastly overcoming the computing difficulties? If the entire bitcoin model is ultimately dependent on the difficulty of computing, then you can try as hard as you can to predict computing and electricity cost, but of course your predictions will be off and you will eventually have to re-adjust your math as reality veers away from your predicted path, right? So no matter what "computing difficulty" does, if it increases exponentially (y), levels off (x), or if it is instead some fairly linear function, won't re-adjusting it always be a problem? Thanks! ^^ Swiftpaw 23:08, 4 September 2011 (GMT)
I asked this over at the Bitcoin Q&A site (should be open to the public in about 2-3 days)