At this week's technical meetup, we discussed:
1) "What is Bitcoin" - finding a (meetup) consensus definition (non-technical)
2) TX censorship attack with 30% hash power
3) UHS
4) Node_BLOOM
5) Eltoo
TL;DRs coming soon.
1) "What is Bitcoin" - finding a (meetup) consensus definition (non-technical)
2) TX censorship attack with 30% hash power
3) UHS
4) Node_BLOOM
5) Eltoo
TL;DRs coming soon.
TL;DR: Eltoo
Eltoo uses SIGHASH_NOINPUT to enable Lightning Network channels where the latest state can always be published (as opposed to penalty based invalidation). It also simplifies multi-user channels (e.g. Lightning Channel Factories).
blockstream.com/eltoo.pdf
Eltoo uses SIGHASH_NOINPUT to enable Lightning Network channels where the latest state can always be published (as opposed to penalty based invalidation). It also simplifies multi-user channels (e.g. Lightning Channel Factories).
blockstream.com/eltoo.pdf
TL;DR: "What is 'Bitcoin'?"
*consensus was not achieved. Discussion continues*
But Gavin's articles were interesting:
1) gavinandresen.ninja/a-definition-o…
2) gavinandresen.ninja/other-definiti…
*consensus was not achieved. Discussion continues*
But Gavin's articles were interesting:
1) gavinandresen.ninja/a-definition-o…
2) gavinandresen.ninja/other-definiti…
TL;DR: UHS
- "Rather than storing all unspent outputs, store their hashes."
- There is a storage/bandwidth tradeoff. Doing this might save storage and increase bandwidth. At this time, Bandwidth seems to be the more expensive, so UHS is probably getting sidelined for now.
- "Rather than storing all unspent outputs, store their hashes."
- There is a storage/bandwidth tradeoff. Doing this might save storage and increase bandwidth. At this time, Bandwidth seems to be the more expensive, so UHS is probably getting sidelined for now.
Thread:
1/ TL;DR: "TX censorship attack with 30% hash power at a 1/8 success rate:
Relevant Reddit thread on the topic: reddit.com/r/btc/comments…
See: Bitcoin Whitepaper p.7-8 for technical workouts
bitcoin.org/bitcoin.pdf
1/ TL;DR: "TX censorship attack with 30% hash power at a 1/8 success rate:
Relevant Reddit thread on the topic: reddit.com/r/btc/comments…
See: Bitcoin Whitepaper p.7-8 for technical workouts
bitcoin.org/bitcoin.pdf
2/ Six confirmations are treated as safe today as a thumb rule by many exchanges. There's around 1/8 chance a malicious miner can rework 6 confirmations. How? The malicious miner mines on his secret chain without broadcasting it unless-- (cont)
3/ unless he's able to successfully rewrite 6 blocks in history. A malicious miner with 30% hash power can find 6 blocks before the rest of world (70% hash power) can find 6 blocks at about 1/8 probability. This is extremely expensive in practice as it would fail 7 out of 8 times
4/ The miner has incentive to do this attack if he wants to double spend a tx with value more than 12.5*6*8 = 450 BTC. 12.5*6 = 75 BTC is the electricity cost he loses. And he succeeds only 1 out of 8 times, hence the factor of 8.
5/ Important takeaway: if you r dealing with high value transactions, 6 is not a safe number. Some miners have direct or indirect control over a high percentage of hash power. One should wait for a longer period depending on the value of the transaction.
