> On Jun 19, 2015, at 8:48 PM, Luke Dashjr wrote: > > On Saturday, June 20, 2015 1:23:03 AM Aaron Voisine wrote: >> They don't need to be made cryptographically safe, they just have to be >> safer than, for instance, credit card payments that can be charged back. As >> long as it's reasonably good in practice, that's fine. > > They never will be. You can get a decent rate of success merely by making one > transaction propagate fast (eg, 1 input, 1 output) and the other slow (eg, > 1000 inputs, 1000 outputs) and choosing your peers carefully. The only reason > unconfirmed transactions aren't double spent today is because nobody is > seriously *trying*. > > Luke > Newspapers are often sold in vending machines that make it possible for anyone to just pay the price of one and take them all…and most of the time they are not that carefully monitored. Why? Because most people have better things to do than try to steal a few newspapers. They probably were much more closely monitored earlier in their history…but once it became clear that despite the obvious attack vector very few people actually try to game it, vendors figured it wasn’t really that big a risk. Same thing applies to people trying to steal a piece of bubble gum at the cash register at a convenience store by double-spending. - Eric Lombrozo > ------------------------------------------------------------------------------ > _______________________________________________ > Bitcoin-development mailing list > Bitcoin-development@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bitcoin-development