6 ms·

The way the Bitcoin community handled this mornings fork in the bitcoin blockchain is nothing short of incredible. This event significantly increases my confide

by yuvadam 14y ago

The way the Bitcoin community handled this mornings fork in the bitcoin blockchain is nothing short of incredible. This event significantly increases my confidence in the Bitcoin ecosystem to thrive even when faced with critical issues. If only the global finance institutions had a fraction of bitcoins capacity to resolve crises with the same unconditional transparency and full cooperation with all agents in the ecosystem.

swhitt 14y ago
It was great to see in action, but given the way the system is right now it wasn't that incredible. The maintainers of the biggest pools are all very active in the community; it would just take 2-3 of the big mining pools to say "hey, let's go back to 0.7" for this to be resolved automatically.
skore 14y ago
I agree. In the past few weeks, Bitcoin has really picked up lots of speed. So much in fact, that I'm wondering whether we're witnessing an exponential adoption where years of people doubting Bitcoin were simply the beginning of the curve. That would also mean that things are going to explode pretty soon. I sincerely hope that the aftermath of said explosion will leave Bitcoin as the strong contender to established monetary systems that it always was - just "proven by reality" now. That would certainly be all I need to finally jump on board.
betterunix 14y ago
"This event significantly increases my confidence in the Bitcoin ecosystem to thrive even when faced with critical issues." It does the opposite for me, because it demonstrates that Bitcoin is extremely vulnerable to double spending. Either everyone uses the same version of the software, or you have to deal with the possibility that this event will happen again. Worse still, an attacker might use some kind of malware to modify Bitcoin clients, until they are able to fork the network and double spend. Yes, making cryptosystems that are secure against malicious majorities is difficult, and experts have published tons of papers and several books on the subject. If Bitcoin's developers want people to trust their system for anything non-trivial, they really need to fix this problem. To break David Chaum's digital cash systems, an attacker would have required exponential computing power in a security parameter that could be scaled arbitrarily, while the users only require polynomial computing power in that same parameter; to break Bitcoin, an attacker only requires a little more than half the sum of all the computing power in the Bitcoin network. The Germans learned the hard way that just because an attack seems too pricey for anyone to bother with does not mean that the attack actually is too pricey for anyone to bother with.
- mozboz 14y ago
  The ability for clients to erroneously or maliciously fork the chain can't be designed out whilst keeping the system open and decentralised. The blockchain is a distributed copy of a shared truth which takes time to propagate and has to be agreed upon by all participants. If the network is open, then me, or me and 500,000 friends, can join the network and attempt to propagate a different truth. If you compare the way the Bitcoin community responded to this with the response you might have expected from even a very well run company, I think the Bitcoin community comes out ahead. It would be useful and or interesting to compare the Bitcoin community response with a specific failure and response of a for-profit closed-technology company, but I'm not sure what case to use for comparison. Maybe the recent cloudflare outage?
  - betterunix 14y ago
    "The ability for clients to erroneously or maliciously fork the chain can't be designed out whilst keeping the system open and decentralised" That is not true. Cryptographers have published many thousands of papers on securing multiparty computation against malicious participants over the past twenty years. Here is a very brief introduction: https://en.wikipedia.org/wiki/Secure_multi-party_computation https://en.wikipedia.org/wiki/Secure_multi-party_computation The relevant citations in this field, like the relevant citations about digital cash protocols, are conspicuously absent from Satoshi's writeup about Bitcoin. This is one of the things that bugs me the most about the Bitcoin community: nobody seems to have done any background research at all. There is a huge volume of work that has been done in this field, but little mention of it among Bitcoin's developers. The Bitcoin developers do not even seem to be worried about the fact that the work required to attack Bitcoin scales linearly with the work required to run the Bitcoin system: https://en.bitcoin.it/wiki/Weaknesses#Attacker_has_a_lot_of_computing_power https://en.bitcoin.it/wiki/Weaknesses#Attacker_has_a_lot_of_... "If you compare the way the Bitcoin community responded to this with the response you might have expected from even a very well run company, I think the Bitcoin community comes out ahead." OK, the community is great at admitting that there are bugs and finding workarounds. That does not count for much in a secure multiparty system, where subtle problems in a protocol can undermine the security of the entire system. Security in a system like Bitcoin is about more than just "avoid X, remember to do Y, and patch bugs quickly!" Malicious clients are going to have to be dealt with in Bitcoin, and they are going to have to be dealt with better than what we saw today.

by yuvadam 14y ago

swhitt 14y ago

It was great to see in action, but given the way the system is right now it wasn't that incredible. The maintainers of the biggest pools are all very active in the community; it would just take 2-3 of the big mining pools to say "hey, let's go back to 0.7" for this to be resolved automatically.

skore 14y ago

I agree. In the past few weeks, Bitcoin has really picked up lots of speed. So much in fact, that I'm wondering whether we're witnessing an exponential adoption where years of people doubting Bitcoin were simply the beginning of the curve. That would also mean that things are going to explode pretty soon. I sincerely hope that the aftermath of said explosion will leave Bitcoin as the strong contender to established monetary systems that it always was - just "proven by reality" now. That would certainly be all I need to finally jump on board.

betterunix 14y ago

"This event significantly increases my confidence in the Bitcoin ecosystem to thrive even when faced with critical issues." It does the opposite for me, because it demonstrates that Bitcoin is extremely vulnerable to double spending. Either everyone uses the same version of the software, or you have to deal with the possibility that this event will happen again. Worse still, an attacker might use some kind of malware to modify Bitcoin clients, until they are able to fork the network and double spend. Yes, making cryptosystems that are secure against malicious majorities is difficult, and experts have published tons of papers and several books on the subject. If Bitcoin's developers want people to trust their system for anything non-trivial, they really need to fix this problem. To break David Chaum's digital cash systems, an attacker would have required exponential computing power in a security parameter that could be scaled arbitrarily, while the users only require polynomial computing power in that same parameter; to break Bitcoin, an attacker only requires a little more than half the sum of all the computing power in the Bitcoin network. The Germans learned the hard way that just because an attack seems too pricey for anyone to bother with does not mean that the attack actually is too pricey for anyone to bother with.

mozboz 14y ago
The ability for clients to erroneously or maliciously fork the chain can't be designed out whilst keeping the system open and decentralised. The blockchain is a distributed copy of a shared truth which takes time to propagate and has to be agreed upon by all participants. If the network is open, then me, or me and 500,000 friends, can join the network and attempt to propagate a different truth. If you compare the way the Bitcoin community responded to this with the response you might have expected from even a very well run company, I think the Bitcoin community comes out ahead. It would be useful and or interesting to compare the Bitcoin community response with a specific failure and response of a for-profit closed-technology company, but I'm not sure what case to use for comparison. Maybe the recent cloudflare outage?
- betterunix 14y ago
  "The ability for clients to erroneously or maliciously fork the chain can't be designed out whilst keeping the system open and decentralised" That is not true. Cryptographers have published many thousands of papers on securing multiparty computation against malicious participants over the past twenty years. Here is a very brief introduction: https://en.wikipedia.org/wiki/Secure_multi-party_computation https://en.wikipedia.org/wiki/Secure_multi-party_computation The relevant citations in this field, like the relevant citations about digital cash protocols, are conspicuously absent from Satoshi's writeup about Bitcoin. This is one of the things that bugs me the most about the Bitcoin community: nobody seems to have done any background research at all. There is a huge volume of work that has been done in this field, but little mention of it among Bitcoin's developers. The Bitcoin developers do not even seem to be worried about the fact that the work required to attack Bitcoin scales linearly with the work required to run the Bitcoin system: https://en.bitcoin.it/wiki/Weaknesses#Attacker_has_a_lot_of_computing_power https://en.bitcoin.it/wiki/Weaknesses#Attacker_has_a_lot_of_... "If you compare the way the Bitcoin community responded to this with the response you might have expected from even a very well run company, I think the Bitcoin community comes out ahead." OK, the community is great at admitting that there are bugs and finding workarounds. That does not count for much in a secure multiparty system, where subtle problems in a protocol can undermine the security of the entire system. Security in a system like Bitcoin is about more than just "avoid X, remember to do Y, and patch bugs quickly!" Malicious clients are going to have to be dealt with in Bitcoin, and they are going to have to be dealt with better than what we saw today.