Hacker News Reader

Comment by kadoban | Hacker News Reader

Depending on what it is, 3-6 is a _lot_. For something like a Steam game, it seems like waiting for 1 should be more than enough.

I say "more than enough" because can't they just revoke the key of whatever was purchased if it gets double-spent? Seems easy enough.

View on HN

> it seems like waiting for 1 should be more than enough.

Not in bitcoin land, orphaned blocks happen a few times a day: https://bitcoin.stackexchange.com/questions/2170/how-often-f...

If you're willing to tolerate this risk then by all means go for it.

(I'm with you I think steam should be able to tolerate this risk.)

FYI the answer in that link is from 2013, not sure how much it's changed since then but I doubt it's the same today.

A feature of bitcoin is it’s stubborn system stability and resistance to change :)

One of the odd benefits of this is that old forum posts and discussions have a good chance of being relevant.

Although I will say lots of progress has been made in p2p message propagation (since this isn’t directly a part of consensus) which could definitely prevent orphan blocks, so you may be on to something in this case.

Yeah, the biggest effects on the number of orphaned blocks should be the number of distinct mining pools and the latency between them. I could see some of those numbers changing enough to matter since 2013, but I'm not all that sure.

But it is much riskier for the fruadster after 1 transaction. With 0 confirmations it is almost certian that a new transaction with a higher fee will replace the older one. With 1 confirmation you have to either swing a lot of mining power or hope that your block becomes orphaned. The fraud likely doesn't make sense anymore if you lose your money 24/25 times instead of almost never.

Then Steam can accept that lower risk and mitigate further by then revoking game access.

Reply on news.ycombinator.com