Bitcoin Core Development’s Game of Thrones


With the settling of the blocksize “debate,” Bitcoin Core development is evolving into what appears to be a more decentralized structure.

At the beginning, of course, there was only Nakamoto who made changes as he pleased and then sent a warning to all node operators to upgrade if necessary.

Then, there were a number of developers who worked on the open source code for free, something which caused a problem because coders need to eat too.

The first attempt to solve that problem was to create a Bitcoin Foundation, which unravelled. Leading to companies, like BitPay, Coinbase, MIT and others, directly funding one or two coders each.

That led to a number of them going without funding, so Adam Back and others came up with the idea of raising money from VCs to create a for-profit corporation which hired the vast majority of developers.

With Blockstream being centralized, and therefore more organized and more single minded while furnished with a marketing and PR team, they could easily outmaneuver other disparate coders or companies who had diverse interests and thus never managed to speak in one voice at the same time.

That gave the corporation immense influence, amplified by the moderator status of some of their employees on social media, like IRC where much development discussion takes place.

However, one could easily copy their business model. Another company therefore rose, Chaincode Labs, to itself hire a number of Bitcoin Core developers. Then Lightning Labs gained funding.

Suddenly, Bitcoin Core developers no longer speak with one voice. The most recent example is a new proposal to change the Bitcoin Core protocol to accommodate an Asicboost implementation.

There appears to be a relationship between that Asicboost implementation and Blockstream, which may have developed it.

As such Luke-jr, a Blockstream employee, is quick to approve the new proposal, stating “concept ACK, assuming community support.”

But Matt Corallo, who now works for ChainCode Labs, takes a more measured approach:

“Without some kind of review in the form of several mining hardware manufacturers weighing in that they feel comfortable building in asicboost into their hardware, I don’t think we can give this kind of implied approval of it’s use,” he said.

Marco Falke, a Bitcoin Core committer and founder of Chaincode Labs, further says: “this needs a bip and mailing list discussion, I guess.”

For some time now there has been a sign of some competition of sorts between these three development teams.

In this context we can perhaps make sense of a very interesting character, Cobra, who says:

“This change is designed to make it easier for a business to use patented technology on the network… Rather than improve the Bitcoin network and its functionality, this pull request is designed to aid outside interests and give competitive advantage to a mining hardware business owned by the OP.”

Money has now clearly made it to bitcoin development, with most key players, perhaps even Theymos, most likely employed by one company or another.

Companies, so being for-profit motivated, have their own business interests and aims which can at times conflict with wider ecosystem interests.

It might have been very different a few years ago for example if many developers were not speaking with one voice either because they were paid by one check or were hoping to become paid by it.

Because now with hindsight, it appears just 2MB would have been sufficient to accommodate all transactions and provide the many millions that flocked last year with an excellent experience of internet money that is almost instant and almost free.

That 2MB would have been sufficient because the backlog reached only around 300,000, which suggests the network needed only 600,000 transactions a day to provide an optimal user experience.

Segwit has a max size of 4MB, or around 1.2 million transactions. Had that been instantly implemented in 2016, rather than so incredibly slowly it was not implemented at all in time for last year’s boom, the network would have been perfectly able to handle it while the demand would have been no more than the provided 4MB capacity.

Now, with LN almost out to a usable extent, capacity would have further been increased through second layers so maintaining at all times a decent user experience without causing a division, a chain-split and without causing a reverse merchant’s adoption.

Few therefore can say with hindsight that the requested temporary relief would have not been the right choice, just as few can really say there were not plenty of mistakes on all sides.

“When will a transaction fee market ever develop?” Bram Cohen, investor in Lightning Labs, asked yesterday at a time when the average blocksize has fallen to nearly 0.5MB.

It is difficult to really see how one can think such fee market would develop in a highly competitive environment except for very temporarily, but that doesn’t mean LN doesn’t have its uses.

The protocol can be more convenient where there is a centralized service provider such as a coffee shop or Amazon, but it is also necessarily limited because you can’t just put up an LN donations address, for example.

As such the idea that everyone will be forced to use LN rather than competing blockchains because of high fees rather than because it does have some advantages appears to us to be misguided as increasing prices reduces demand according to econ 101, rather than vice versa.

Now we are finally starting to see such matters raise not only among what they have portrayed as detractors, but also among “themselves.”

That’s presumably because there are now a variety of checks paying them, which has with it brought a much welcomed variety of approaches.

Because we find some of these coding teams to be a lot more accommodating and far too reluctant to fall so low as to employ outright censorship.

Bitcoin, therefore, might once more start moving a bit quicker as their development becomes a bit more decentralized so allowing for more feedback and thus wiser approaches.


Leave a Reply

Notify of