After another mammoth session trying to arrive at consensus on how to progress with the scaling of Bitcoin the newly convened Bitcoin Roundtable has arrived at a plan. The five largest mining pools (Antpool, F2Pool, BTCC Pool, and Bitfury), which together represent more than 75% of the Bitcoin network's hashing power, reached an agreement with several of the Bitcoin Core contributors on a roadmap for scaling bitcoin.

The group has eschewed Bitcoin Classic or any other scaling solution that is not endorsed by the Bitcoin Core developers. The Hong Kong meeting backed Bitcoin Core's introduction of a hard fork in July 2016, after adopting Segregated Witness in April – a way of compressing transactions within blocks by leaving out their digital signatures.

A statement issued by the Bitcoin Roundtable following the talks said: "This hard-fork is expected to include features which are currently being discussed within technical communities, including an increase in the non-witness data to be around 2 MB, with the total size no more than 4 MB, and will only be adopted with broad support across the entire Bitcoin community."

Some commentators said of the unfolding process that what was once a decentralised system had turned into a political game of chess.

The concept of decentralisation

Samson Mow, COO of BTCC and co-organiser of the scaling Bitcoin meeting, agreed the fight between Bitcoin Classic and Bitcoin Core was perhaps more about control of the protocol than block size. He told IBTimes UK: "Bitcoin Classic is another fork; people have to run that before it will activate the hard fork. So you fork on to another software not maintained by Bitcoin Core. So you can call it a coup."

BTCC chief operating officer Samson Mow

Mow was philosophical about the concept of "decentralisation". He said: "First of all, there are levels of decentralisation. You could say the development team is centralised; you could say mining is centralised; you could even say bitcoin merchants services are centralised into BitPay or Coinbase.

"The thing is, people are free to use what services they want. The miners in a pool are free to mine individually; they are free to move to another pool. Decentralisation is a very loaded word I think."

A recent article in Forbes raised some interesting cultural questions about how Chinese stakeholders in Bitcoin are not concerned about the sort of libertarian ideas that inform much of the Western Bitcoin community. However, Mow pointed out the Bitcoin community is very active in China and they do value decentralisation, which is equated with robustness.

He said: "It means something is durable because it cannot be broken or destroyed easily, so people do value that aspect of it. Now people don't really talk about 'decentralisation' because I think the word is not taken up here; people don't use that in debates.

"Chinese culture is much more collaborative, much more focused on harmony, so people working together, such as miners working in a pool. You could call it centralisation or you can call it cooperation to reach a goal."

Soft hard fork

Mow said a Bitcoin improvement proposal (BIP) would follow with technical details of the coming "soft, hard fork" (a new concept), adding people criticised the initial letter from the roundtable because it lacked technical details. "It would hard fork but then it wouldn't increase the cap. The activation date would require a super-majority of 95% and the plan right now is to have some voting mechanism where coins can vote but this is not finalised yet because we are waiting for a BIP. We are hoping that will come in a short time."

Mow said Classic had resonated with a lot of people because it appeared to be very simple: increase block size from 1MB to 2MB. "They market it that way," he said. "They say, it's so simple and these core developers are morons, they can't increase a simple thing from one to two. But what is going to be in the BIP that comes out of this meeting is much more complex because SegWit will be the first stage."

Mow pointed out there will other compression techniques such as Schnorr multisig, and the new block size will have two components: the witness portion and non-witness portion. He said: "These two things will combine together to give you a final block size. The meeting was hung up on the concept of the effective block size because you have all these technical people in the room talking about how the end result will be calculated and what is the effective block size.

"That is why the letter says around 2MB to 4MB, because at this time, we don't know. To say that would be irresponsible because we haven't done all the maths and calculations and we don't know all the methods of compression."

Scaling ceiling

A recent academic paper on scaling Bitcoin stated a basic rethinking of technical approaches is required, adding a 4MB block size corresponds to a maximum throughput of at most 27 transactions per second.

Mow said: "I have had a lot of conversations with developers and technical experts and the general consensus from the people I recognise to be the experts in the space is that Bitcoin cannot scale indefinitely on-chain as it is today. We have to face the network constraints.

"A lot of these proposals just disregard very pragmatic aspects of the network like is the block going to propagate, is it going to be orphaned, are we going to force nodes off the network because the blockchain becomes too big, or are we going to force miners out the network – smaller miners or miners behind the great firewall.

"The important thing is to want to scale, but scale according to the constraints that we face today. So if we can go to 2MB that's good; if we do the proposal from the Bitcoin Roundtable meeting, we go SegWit and then we have a hard fork increase for a maximum effective size of 4MB, that's good too. Then if Lightning comes out in the next year, sometime in 2017, we will use Lightning.

"We have to do is use whatever is available at our disposal to scale safely whether on-chain or off chain. We don't live in a world where we can make mutually exclusive decisions to choose one or the other. We have to be pragmatists and use everything at our disposal."