The war for control over Bitcoin's future took a bizarre turn yesterday when
Greg Maxwell, a developer with Bitcoin Core project and CTO of Blockstream,
announced that a certain chip contained the ASICBOOST optimization, and
suggested a BIP that involves a Segwit-style commitment, which in practice
would be coupled for a call to activate Segwit.
Here's my quick response to this turn of events:
So far, there's absolutely no independently confirmed evidence of ASICBOOST actually being used on any chip.
If ASICBOOST was actually used, we'd see ample evidence on the blockchain.
One type of evidence would be in the form of transactions that are picked according to an algorithm that shuffles the transactions to create a sought collision, instead of sorting them by fee-per-byte. I do not have the time to perform a thorough study, but I manually examined a randomly picked block mined by Antfarm a few days ago. As you can also examine for yourself, the transactions are ordered essentially by fee-per-byte, which is not what we would see if they were shuffled to create collisions.
You can stop reading here. If #1, #2 and #3 are false, then there's absolutely nothing but baseless accusations flying around. Someone needs to do a thorough study to instantiate the claims.
On the offchance that Antpool is actively utilizing ASICBOOST, there's still quite a lot to be said about the claims:
Building more efficient mining chips is what miners are supposed to do, as it secures the blockchain. Framing an optimization as an attack is disingenuous. Going from 28nm to 16nm was not an attack on the network. Better mining algorithms have never been considered an attack.
You can stop reading here as well. I almost stopped writing here. But someone
will inevitably make the argument that patents can confer an uneven advantage
to certain miners, and that this can skew the marketplace for mining equipment.
Or more likely, they will not make this case, because it is highly unlikely
that this patent actually is insurmountable, or that it will likely have
catastrophic consequences that require a call to arms. Instead, they'll
make an emotional argument for why patents are bad. For those,
here are some additional thoughts.
There are countless patents involved in the manufacture of ASICS, spanning layout, doping, lithography, all the way to packaging. Layout alone can confer advantages that far exceed the paltry 30% possible with ASICBOOST. The Bitcoin community believes it understands hashing algorithms, but that's no reason to ignore patents related ASIC manufacture in general. If a case is to be made against patents, then it needs to be made properly, across all patents related to ASIC manufacture.
Taking action to block this is similar to the government taking measures to block Elon Musk's more efficient electric cars. Specifically prosecuting a chosen miner, in the current political climate, would send a terrible message of absolute centralization in the hands of one particular developer group, and it would severely damage Bitcoin mining and the coin's security.
This is another point where one can take a break.
Segwit is being offered as a solution to ASICBOOST. Yet Segwit has at best a tenuous connection as a response to ASICBOOST mining. There are countless other, non-controversial solutions that can disable ASICBOOST, should one choose (wrongly) to do so.
The fact that Segwit is being offered as an inevitable solution to ASICBOOST, at this point in time in the blocksize debate, suggests something about the true motivations behind this reputational attack.
There is a history of reputational attacks against people who have spoken up against Segwit. This is toxic and not healthy for the space.
Some people have asserted that ASICBOOST forms a conflict of interest (CoI) that explains why Jihan Wu opposes Segwit. This is based on the false premise that everyone would adopt Segwit if it weren't for a CoI. There are many other publicly discussed reasons to oppose Segwit, based on technical and philosophical grounds.
If commercial "conflicts of interest" are showstoppers that should cause a company to abdicate its voice in the ecosystem, then practically every significant company would have to have no say in Bitcoin's evolution. Every other company is trying to influence the coin to evolve according to their own vision, while taking steps to maximize its role within that vision. We call this "capitalism." If we suddenly instituted draconian measures to cut companies out due to CoIs, one of the first companies with CoIs would be Blockstream.
Overall, the community seems to be at a point of rapture. As we all move forward, let's make sure that we take every step for good, substantiated reasons backed by science. Reality is too important to let be swayed by arguments to emotion and by the noise generated by troll armies, on either side.
Let's do our due diligence first, and dispassionately find a coherent strategy that represents the best path forward for all.
Empty blocks indicate ASICBOOST usage, though they can also arise out of datacenter connectivity issues or from headers-first mining. Counter to popular opinion, empty blocks help bury other blocks and thus provide security. (Thanks to the ever insightful Dan Robinson of Chain for bringing this up!)
There are other ways of grinding through blocks to populate the tables in ASICBOOST, besides permuting transactions. Each grinding scheme exacts a cost for creating diverse inputs. These costs eat into the benefits of ASICBOOST. It would be interesting to perform a quantitative comparison of different grinding schemes.