Topic: ASICBOOST Aftermath: What Now Must Be Done? - page 6. (Read 3802 times)

within the recent Bitmain expose you only see blockstream? lol
hahahahahahahahahahahahahahahahaha

Thanks for your own comments and insights. My last 2 points just distinguish between (1) making the fix to the current problem and (2) making it clear that similar actions will be taken in the future to prevent the problem from recurring.

(I'm a quality engineer, and this breakdown just reflects our habitual thinking as we write corrective action reports and answer a checklist of questions about countermeasures and preventing recurrence.)

or blockstream can just admit that segwit isnt perfect by admitting that its 'fixes' are not guaranteed and that it does nothing to stop native spammers etc.

and for blockstream to reset their own snobbery and try to fix their own internal issues and try something the community can and will happily accept

however using PoW nukes, mandatory activations, fee discount bribes and deadlines, bypassing node consensus.. looks very desperate. they are too blind to ven ask why should they need to resort to such tactics..

Fully in agreement. We may have averted a real disaster for Bitcoin in the shape of BU, but the plans for the next coup have already been made public, and all the blocksize obsessives on the forums have started the cycle all over again.

The honest miners should not be penalised for maintaining their decorum faced with adversity, but the risk of mining centralisation is still too significant. A staged move away from SHA-2 hashing must happen sooner rather than later to mitigate the risk. Miners can take the opportunity to halt their development of mining ASICs, and concentrate on making the transition to CPU mining, there's no reason why they cannot still compete in the marketplace using their other marginal advantages, which will still exist after chaging the PoW algorithm.


Isn't this the same as point 1, though? If the Bitcoin developers write the code, that is the policy, and it couldn't be any more clear written as code. In cryptographic networks like Bitcoin, the code is the policy, the code is the law.


Further to this, it's possible that UASF might not even be needed. Bitmain's loss in hashrate share resulting from gmaxwell's neutering of ASIC Boost might be enough to force Bitmain's hand. Segwit signalling might rise to above 50% from removing ASIC Boost's viability, and 50% BIP acceptance has a tendency to move to 95% very quickly, it makes miners and pools nervous that they may start losing block solution races when 51% threshold is passed.

For those who've missed it:

https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2017-April/013996.html

https://bitcoinmagazine.com/articles/mining-manufacturer-blocking-segwit-benefit-asicboost/

Summary of the situation: Bitmain has been producing and mining with ASICs using an exploit that gives them a dramatic advantage over competing miners. This exploit is incompatible with Segwit, so Bitcoin Unlimited was created to gin up the appearance of public opposition to Core and place monopoly control over bitcoin mining and the protocol in the hands of Bitmain and associates.

The exploit also incentivizes miners to mine empty blocks, which has aggravated the TX bottleneck situation and contributed to high fees.

While they have been exposed, it is not clear to me that there is any reason they will be shut down in the near term. In other words a conspiracy has been unmasked, but not defeated.

ISSUE: ASICBOOST provides a ~30% mining advantage over non-ASICBOOST miners.

NEAR-TERM RISKS:
1. BitMain keeps mining empty blocks. TX fees remain elevated.
2. BU keeps blocking Segwit from reaching 95% consensus. TX capacity remains throttled.
3. By extension, the Lightning Network continues to be blocked/delayed/impaired insofar as SW provides the best implementation for it.

LONG-TERM RISKS:
1. Bitmain runs all opposition out of business and assumes monopoly control over mining.

COUNTERMEASURES:
1. Patent infringement? This is outside of the bitcoin communities' control, but one countermeasure is if Bitmain is infringing on the original ASICBOOST patent and it can be upheld in a Chinese court. I'm not optimistic here but it has to be listed as a possibility.
2. UASF for Segwit/Gregory Maxwell's countermeasures (see link above). A UASF introducing Segwit would force Bitmain to either mine BTC while abandoning their ASICBOOST exploit, or hardfork away with their own altcoin.
3. ? Other options?

LONG-TERM COUNTERMEASURE:
This episode has taken the issue of mining centralization from a relaxed discussion about future risks to bitcoin to an imminent crisis. BU was close to 50% hashrate in the weeks leading up to this expose. A craftier pair than Wu and Ver might pull off a coup next time. Long term no cryptocurrency can survive and thrive in an environment where monopoly control is threatened or possible.

At the same time I want to emphasize that a raw majority of miners did not join BU and should not be unfairly penalized for their actions. They've made tremendous investments in specialized hardware and should not be bankrupted in a rash reaction to this situation.

Fortunately, the ROI on bitcoin mining hardware is pretty short - less than a year from all I've seen. The means the Core development community can, after due process, provide a roadmap covering several years. One that leads back to decentralization of mining. Thus I urge Core to:

1. Provide a roadmap away from the current ASIC-based paradigm back to mining with desktop-based hardware, over a timeframe that allows the mining community to maintain profitability (without the pressure of Bitmain's advantage weighing on them, through SW deployment).

2. Formalize a public policy that such a decentralized mining paradigm will be maintained.  Either by altering algorithms or through other mitigating actions to defeat any ASIC or other specialized hardware implementation that significantly centralizes mining.

Just my 2 satoshis,
ebliever