Neither SegWit Nor BU Can Activate: Interview With Rhett Creighton
Cointelegraph spoke to Rhett Creighton, Blockchain engineer and hacker, who with the Bitcoin Unlimited Camp, but left a few weeks ago about the current situation.
There is no indication that the scaling battle is coming to an amicable end anytime soon. Can you imagine buying a mobile phone credit for 48 cents and having to pay 32 cents in transaction fees? The fact is many were attracted to Bitcoin because of its lower transaction fees. The expensive and destructive cul-de-sac is taking away a lot from the Bitcoin ecosystem.
Cointelegraph spoke to Rhett Creighton, a Blockchain engineer and hacker who was with the Bitcoin Unlimited camp but left a few weeks ago due to the current situation.
Cointelegraph: Is Bitcoin Unlimited anti-Bitcoin?
Rhett Creighton: I don't think it is fair to say BU is anti-Bitcoin. The goal of BU is to create a new consensus mechanism for enabling protocol upgrades in Bitcoin. The near-term goal of BU is to increase Bitcoin block sizes to 2 MB. Almost nobody believes 2 MB blocks would be destructive to Bitcoin if that's what everyone used.
BU may not be what some people want in Bitcoin. Some people would argue that no protocol upgrades in Bitcoin, while altcoins take over more market share, is anti-Bitcoin.
CT: Has BU closed its source, releasing binary only versions now?
RC: Right now in Bitcoin, there is no protocol specification. We only have the implementation which started from Satoshi's original codebase, "bitcoind," which is hosted on GitHub. I believe 2-5 people maybe have commit access to this repository.
There are a few other implementations of Bitcoin. For example, "btcd" is written in golang. However, "bitcoind" developers assert that "bitcoind" is the true version of Bitcoin, and if people want to run other software like "btcd" it better be bug-for-bug compatible with "bitcoind."
This is not the case with Ethereum, for example. In Ethereum, there is a reference specification and anyone can make a client as long as it follows the specification. The client could be closed or open source.
Right now, a small number of developers are de-facto custodians of "bitcoind." BU's goal is to decentralize reliance on a single codebase and more towards a model where miners agree on the specification for what Bitcoin is. At that point, a patched version of "btcd" or "bitcoind" would work as a full node as long as it met the network's consensus parameters.
I don't work on or follow the development of the Bitcoin Unlimited implementation closely, so I don't know exactly what plans there might be with respect to closed source development.
CT: How true is the allegation that exchanges will not list BU as Bitcoin and businesses are signaling that they want nothing to do with it?
RC: People will follow the money. If a minority hash rate chain tries to split off and does 2 MB blocks, of course, many exchanges will want nothing to do with it.
In the case where 70+ percent mining hash rate wants to do something, the majority chain has the excess power it could use to attack the minority chain. It could even launch an attack where it will only ever append empty blocks to the minority chain. Because difficulty only resets every two weeks under normal conditions in Bitcoin, a loss of 70 percent mining hash rate would mean now there are only blocks every 33 minutes instead of every 10 minutes and those blocks will always be empty because miners on the majority chain are 51 percent attacking the minority chain.
It's absurd to think that exchanges would support the chain where no transactions are possible for months when there is a perfectly functional Bitcoin chain with the majority of hashing power and 2 MB blocks.
Under this scenario, "bitcoind" developers have suggested changing the proof-of-work algorithm Bitcoin uses. People would then decide if they want to use the chain with 2 MB blocks and SHA256 proof-of-work, or 1 MB blocks and a new kind of proof-of-work that will likely be CPU/GPU friendly.
This case would be similar to the way ETH/ETC played out where people who held their ETH coins received ETC coins also, and people can use one, both or neither.
Eventually, I think a chain split is the most likely outcome.
CT: Are assertions like the BU dev team has taken on a very difficult task that only a handful of men on earth can handle properly and they have run into many quality control problems true?
RC: Increasing the block size limit to 2 MB is one of the easiest changes you could make to "bitcoind," so I don't think that should be an issue. Again, the goal would be for all the best developers to work on different implementations of Bitcoin in different languages: go rust, c++, etc. If 2 MB Bitcoin had the majority of the market share, it would attract the best developers.
CT: What makes SegWit better than BU?
RC: I don't see how SegWit can activate. It needs 95 percent signaling to activate. If it gets 94 percent, it does not activate. There are risks associated with activating SegWit without an overwhelming majority consensus.
Of course, if 100 percent of Bitcoin miners agreed to switch to SegWit and none of the risks outlined by Bitcoin Core was an issue. It's widely known that some of the benefits of SegWit include fixing a transaction malleability bug in Bitcoin and allowing for more transactions to fit in a 1 MB block. The Bitcoin transaction malleability bug severely limits what can currently be done with Bitcoin smart contracts. There are likely different parties with different motivations.
CT: But why do you say Bitcoin Unlimited was likely set up not to succeed, but simply to block any other progress from happening with Bitcoin?
RC: There are likely different parties with different motivations and may be people funding BU simply to block SegWit, which is a threat to AsicBoost. Many of the people who work on BU believe they are trying to fix scaling Bitcoin. I think it's unlikely at this point that BU will get a supermajority of mining hash rate power, enough to launch an attack on a minority chain and take over Bitcoin. However, a rational miner may still want to fund BU simply to block SegWit if it was secretly using AsicBoost. Like I also said in a recent blog article, I think it's possible that people like Roger Ver just want bigger blocks, and didn't know anything about AsicBoost.
CT: Finally, what will be the outcome of the scaling battle?
RC: If I have to predict something, I think there is going to be a split, which most of the Bitcoin community doesn't want right now. I think it's unlikely at this point that BU will get a supermajority of mining hash rate power enough to launch an attack on a minority chain and take over Bitcoin.
The way I look at it is what happened after Ethereum split to ETH and ETC. The combined market total value is over 400 percent greater than what it was before the split.
Years ago, Blockstream received its initial funding largely due to a white paper on "side chains" which promised trustless two-way-pegs as a way to innovate with new types of Blockchains without disrupting the main Blockchain. Now with almost $70 mln in funding raised later, we still don't have the details of how a trustless two-way-peg could work in practice.
Blockstream has released something they call a sidechain, "Alpha," which operates on a federated peg. This means you could set up a Blockchain where three out of five banks or exchanges act as a gateway to their private Blockchain. The fact that they even call that a side chain is an insult to the initial sidechain white paper their company was founded on.
In engineering, there is a widely known KISS principle, which is an acronym for "keep it simple, stupid." Chain splits are effective and work because they are just so simple and stupid.