Counterattack is a naive strategy. It is a knee-jerk reaction to an attack in progress. Yes it can reduce or wipe out the attacker's profits, if the counterattacker is ready to go and has the money to spend in the hashrate market. Who will spend time preparing to counterattack and being ready with a minute's notice? Who will spend money buying hashes to counterattack, for no reward?
The article suggests that because exchanges are the victims, that exchanges should be the counterattackers. This might work if an exchange only trades BTG pairs and attackers only 51% shadow-mine BTG. But in reality, exchanges have hundreds of cross-pairs and BTG is not the only coin vulnerable to attack
So far, it has observed 40 reorgs at least six blocks deep on Expanse, Vertcoin, Litecoin Cash, Bitcoin Gold, Verge and Hanacoin
In the long run, if these reorgs become more common, there is an opportunity to redesign PoW mining to resist attacks. There was a discussion on Reddit a few months ago suggesting that a coin can have multiple separate PoW hashing algorithms - maybe 10 or more - and a rule requiring each block to be mined with a different algorithm, maybe by cycling through the algorithms in turn, or calculating difficulty values so that each algorithm has equal probability of winning, whether it's a CPU on a Raspberry Pi or a 100TH SHA-256 ASIC, or a simple rule which blocks the same algorithm from winning 2 blocks in a row. This would mitigate 51% attacks because an attacker should find it impractical to switch into a different algorithm for every block
barnz3000Platinum | QC: BCH 141, CC 116, BTC 51 | TraderSub2 weeks ago
I would be interested to know the approximate costs required to commit these attacks on minor chains.
What is the $risk to reward here?
I'm sure a lot of miners whos equipment is going to be economically non via me in 2 months, would like to know.
> the approximate costs required to commit these attacks on minor chains
The articles which covered the most recent BTG attacks said the price was about $800-$900 per hour to shadow mine for 2 separate 2-hour periods
> a lot of miners whos equipment is going to be economically non viable in 2 months
SHA-256 mining to attack the BCash and BCash-SV forks could become viable
One interesting aspect only mentioned in passing in the article is that the coin is not the victim of the attacks. The profit comes from attacking an exchange. The earlier BTG attack won millions because the attacked exchange had no measures to recognize the reorgs, not enough confirmations. The most recent attack only won thousands
One possible future scenario has occasional hit-and-run 51% attacks spending a few thousand to make a few thousand. The 51% attack risk moves from being a disaster which can destroy a currency, to a minor annoyance which can be minimized with mitigation techniques
I have been asking Medium to fix this problem for 2 years. Until Medium get around to fixing it, can writers please boycott Medium and find somewhere else to publish