They received a mathematical assurance that it would work. The relying nodes, blockchain, miners and priority rules are all well-known and straightforward to account for. If you pay enough fees given your transaction priority, it'll get into a block eventually. If you want it to be mined sooner, you can pay higher fees.

And even if it won't be mined, the worst case scenario is that you would keep your Bitcoins and can try re-sending them later, they won't disappear to anywhere.

Except for that pesky blockchain split that happened once...

There have been more than one block chain splits. I think you are talking about the hard fork that happened in May, which was planned for and not an unexpected accident. If you conduct a transaction and it is included in the shorter block chain, then by definition the network doesn't recognize the transaction and you keep your money and just send again.

I'm not sure what you're talking about, exactly none of the blockchain splits that have happened so far were expected to happen.

That hard fork was rolled out over a period of two months, with warning (whilst older clients were sent messages to upgrade) and smaller splits are part of the protocol and happen quite regularly. as expected (like when two miners hit the same block within seconds and they wait for the network to recognise the block).

That is the justification Satoshi Naakamoto gave in the original paper for waiting for 6 blocks for a full confirmation.

"They received a mathematical assurance that it would work"

No they did not. Mathematics is about proofs; Bitcoin does not even have the rigorous definitions needed for that.

To put it another way, what you said is the equivalent of saying that Hacker News gives you mathematical guarantees of security because it happens to use TLS.

I hear you, but I also notice the developers still have not assigned bitcoin-qt a 1.0 version number. I might go ahead and transfer those millions anyways but the beta status would give me pause for thought.

Huh, if you make a valid transaction there is no reason why it would fail.

Not true. The transaction has to be verified by multiple other people before it can be trusted. If you spend the same bitcoin twice in a short time period, it's a toss-up as to which one gets accepted into the official blockchain.

>The transaction has to be verified by multiple other people before it can be trusted.

That's not how it works, sorry. If you don't double spend your transaction and you attach a proper fee it's going to get in the chain, period.

If you were worried about how it would go through, you could break the transaction into many small transfers (which would've been very smart in this context).