Bitcoin took two upgrade paths in 2017
On 1 August 2017 the Bitcoin network activated a protocol upgrade raising the base block-size limit. On 24 August a separate protocol upgrade — SegWit — activated on the same network. The convention that calls one of these "Bitcoin" and the other "a fork of Bitcoin" describes the market's naming decision, not the underlying event. Both are upgrades activated on chains descended from the pre-2017 protocol.
Nothing was created in 2017
Bitcoin was released in 2009. On 1 August 2017 (block 478,559) the network activated a protocol upgrade, and two groups of nodes took different upgrade paths. One took the SegWit upgrade path, activating three weeks later on 24 August (block 481,824), segregating witness data to raise effective throughput within a 1 MB base block. The other took the block-size upgrade path, raising the base block first to 8 MB and later to 32 MB. Both are protocol upgrades activated on chains descended from the pre-2017 protocol — neither was created, and neither was released.
The block-size raise was Bitcoin's original scaling plan
Raising the block size was not a novel or breakaway idea. Bitcoin's creator Satoshi Nakamoto wrote in October 2010 that the limit could be phased in with a simple height-triggered change:
if (blocknumber > 115000) maxblocksize = largerlimit
It can start being in versions way ahead, so by the time it reaches that block number and goes into effect, the older versions that don't have it are already obsolete. When we're near the cutoff block number, I can put an alert to old versions to make sure they know they have to upgrade. — Satoshi, bitcointalk, 4 October 2010
The block-size raise activated on 1 August 2017 is the mechanism Bitcoin's designer described seven years earlier as the ordinary way to scale. There was nothing novel about the upgrade itself; what was novel was that a coordinated activation of Satoshi's own scaling plan met resistance from a separate group preferring a different approach.
Why the block-size upgrade activated first
The three-week gap between the two activations was not accidental. The part of the Bitcoin community that supported scaling by raising the base block size chose the earlier date to activate that upgrade on the network. They were not trying to establish an alternative chain; they were activating what they held to be Bitcoin's ordinary and planned way to scale, before a different rule change took effect on the same network.
Their position was that the two upgrades are not comparable in kind. Raising a numeric cap is the mechanism Satoshi described. Segregating witness data changes where transaction signatures live relative to the block that carries them — and, in doing so, changes what a node running the pre-2017 rules can still verify.
The whitepaper's chain of signatures
Section 2 of the Bitcoin whitepaper defines an electronic coin as "a chain of digital signatures". Each owner transfers the coin by digitally signing a hash of the previous transaction and the public key of the next owner. A payee can verify the signatures to verify the chain of ownership. That definition — that a coin is its signature chain, and that any payee can verify it — is the whitepaper's operational description of what a Bitcoin transaction is.
Under the SegWit rules, the signatures for a SegWit transaction live in a separate witness section that a node running the pre-2017 rules cannot see. That legacy node still accepts the block — which is what makes SegWit a soft fork rather than a hard one, in the standard telling — but cannot verify the signatures the whitepaper says the chain of ownership depends on. Supporters of the block-size upgrade considered that a departure from the whitepaper's design. Whether it makes SegWit itself technically a hard fork rather than a soft one is a longer argument in its own Journal piece.
The naming asymmetry
The convention that treats one 2017 upgrade path as "Bitcoin" and the other as "a fork of Bitcoin" is inheritance by ticker, not by protocol history. Applied consistently, Bitcoin's own SegWit activation later that same month would read as "a fork of Bitcoin with a protocol upgrade to segregate witness data" — because that is what it technically was — but nobody writes it that way, because the ticker stayed on the same chain. The framing follows the market's naming decision; it doesn't describe the underlying event.
What the counting rule does
This site's counting rule has no preferred side to the naming question. Both chains satisfy the two conditions (own blockchain, own consensus) and both are counted. Both are listed in the ranking as cryptocurrencies with their own consensus — see the counting rule applied to BTC, BCH and BSV for the detailed case, and the distinction between hard fork and chain split for the underlying mechanics.
Where the site does have a position is on the framing. Calling one chain "Bitcoin" and the other "a fork of Bitcoin" is a naming choice, not a technical fact. Both chains are descendants of the same pre-2017 network taking different upgrade paths — one holding the 1 MB base block, one raising it — and each is worth reading as what it is.
Activation dates and block heights are stable historical facts; the whitepaper and the bitcointalk archive are the primary sources cited. This post argues about framing, not figures — it does not carry a data stamp.