Topic: transaction fees for high entropy transaction? (Read 1314 times)
Well, entropy has a meaning for a number, at least in a particular base (100000 has very little entropy in Base-10, for instance, but much higher in base-8).
But anyway, I totally forgot that a transaction amount can indeed be the result of some arithmetic operation (currency exchange, division into several parts, and so on).
That was silly of me. I lock this thread.
I was just looking at bitcoin explorer and I saw that some high decimal precision transactions already occured.
Such as [utl=http://blockexplorer.com/t/AVJUJWnahm]82.68708455[/url] for instance.
Such a complex amount for a transaction is probably holding data, not just an amount.
Why do you think so? This is a transfer from a wallet that contains generation from luke-jr pool that creates amounts with 8 digit precision:
http://blockexplorer.com/address/14rJYeAPkXGDRVeTcEn7czE9xNaU5aRi64
Nothing sinister is going on. And even if so, why should we care? Bitcoin cannot hold much of entropy and even if such trasfers would become more common, it's not a bigger problem than transaction spam.
Now, could you inspect transactions and try to make a guess at whether they're conveying data through a side channel? Perhaps. But it would be an unreliable guess. For example, the presence of .01 transaction, made to a particular address at a roughly particular time, could convey significant information. Is 2.71828 a high-entropy amount, or is it just an estimate of e or a result of an exchange-rate conversion? Etc.
Why on earth would anyone use e as an amount for a monetary transaction ?
You are right about exchange rates conversion, though. I've just realised it. So you have a point.
I was just looking at bitcoin explorer and I saw that some high decimal precision transactions already occured.
Such as [utl=http://blockexplorer.com/t/AVJUJWnahm]82.68708455[/url] for instance.
Such a complex amount for a transaction is probably holding data, not just an amount. Satoshi had already expressed his concern about people using the block chain to store data, and I think he considered this as some kind of an attack on the network.
I wonder if miners could not consider requesting special transaction fees for amounts that have high entropic value.
By entropic, I mean that for instance an amount such as 105.43908821 has much more entropy than 0.01, even if it is much bigger. To me, a transaction of the former amount should be more likely to require a transaction fee than the latter.
I think it doesn't even have to need any modification of the bitcoin code, does it? I guess it is just a request we could address to the miners.
Jump to: