This is a segment from the Supply Shock newsletter. To read full editions, subscribe.
Today’s topic is a touchy subject for anyone who’s been in Bitcoin for the past decade.
Eight years ago exactly, a key voting threshold was met that would lead to the official adoption of Segregated Witness (SegWit), a backwards-compatible scaling upgrade that had been in the pipeline since late 2015.
SegWit would make it technically possible to fit up to 4MB of transaction data within Bitcoin blocks by separating out related witness data into a different construct. A neat workaround to spiritually stretch the block-size limit beyond the Satoshi-imposed 1MB.
Miners had been required to signal their support for SegWit adoption following its original proposal, BIP141, in November 2016. Voting “yes” on the initial proposal meant mining “bit 1” blocks — blocks that had a special variable in their headers.
But after around 18 months, less than 40% of blocks were signaling support. In May 2017, Bitcoin Core contributor James Hilliard submitted a supplementary proposal, BIP91, which added an element of time pressure.
Miners had to signal their support for BIP91 above the threshold by midnight Nov. 15, 2017, or else SegWit would be dropped for good. BIP91 reduced the mark for miner support from 95% to 80% over a period of around 2.33 days (269 blocks out of a 336-block epoch).
On July 21, 2017, enough pro-SegWit blocks had been mined to kickstart the soft fork, to be fully activated by Aug. 24, about a month after the threshold was met.
And this is where things get messy. On a technical level, support for SegWit was cut and dry. Miners either supported it or they didn’t.
It was much muddier at the social layer. The infamous New York Agreement signatories — which included most mining pools, Coinbase, Circle, ShapeShift and a large portion of Digital Currency Group portfolio companies — pushed for an additional hard fork if BIP91 would activate.
The hard fork, SegWit2x, would effectively double the impact of SegWit, allowing for up to 8MB of transaction data to fit in each block rather than 4MB. SegWit2x was ultimately cancelled months after the launch of Bitcoin Cash which features 8MB blocks.
There have been literal books written about the Blocksize War, so I’ll leave it to you to pick those up for more detail about what happened in between the acceptance of BIP91 and the failure of SegWit2x.
Whatever your view of the matter, it’s clear that the definition of Bitcoin for many operators shifted around this time. A notable contingency sought to explore Bitcoin’s “money” properties on a separate chain in Bitcoin Cash, which never really took off.
Did that make Bitcoin — the one with persistent consensus — less “money” as a result?
Obviously not, but considering how the ever-rising price has put a dampener on that particular use case, it’s sometimes hard to tell.
Get the news in your inbox. Explore Blockworks newsletters:
