Why did Satoshi leave Bitcoin?

“I’ve moved on to other things. It’s in good hands with Gavin and everyone.”

These famous lines come from one of Satoshoi’s last communications, sent to Mike Hearne on 23 Apr 2011. The very last email (ignoring a suspect message about Dorian Nakamoto) was sent to Gavin Andresen on 26 Apr. In reply Gavin told Satoshi that he had accepted an invitation to give a talk about Bitcoin at a CIA event. This has been seized on as the reason for Satoshi leaving Bitcoin. But Satoshi had announced his departure before knowing of the CIA meeting.

For most Bitcoiners Satoshi had already left months before. As Hearne wrote in an email to Satoshi in March “Hope you can find the time/energy to rejoin us soon!”. He had departed the forum abruptly in Dec 2010 and was handing over reasonability through private communications with a few key individuals.

Satoshi leaving the project was a complete surprise. In December he had been active in the forum with 21 posts in the first eleven days. He was fully engaged and gave no warning signs of anything wrong. On 8 Dec, he published an update of the software, v0.3.18, featuring some bug fixes and enhancements. Everything appeared normal. But just four days later, Satoshi issued a surprise new release, v0.3.19, which wrapped up all his own work in progress:

“There’s more work to do on DoS, but I’m doing a quick build of what I have so far in case it’s needed, before venturing into more complex ideas.” (12 Dec 2010)

This was Satoshi’s last post. On SourceForge he amended the copyright notice for this release from the previous “Satoshi Nakamoto” to “Bitcoin Developers” a clear indication of his intention to withdraw.

Something happened between 8 Dec and 12 Dec to cause his departure. Most obviously, the WikiLeaks affair was attracting attention to Bitcoin. The issue led to Bitcoin’s first mention in a mainstream publication when PC World published an article titled “Could the Wikileaks Scandal Lead to New Virtual Currency?” The title was the most controversial thing about the piece, which gave a balanced, non-sensational account of Bitcoin as the invention of a Japanese man called Satoshi Nakamoto. Satoshi was not pleased:

“It would have been nice to get this attention in any other context. WikiLeaks has kicked the hornet’s nest, and the swarm is headed towards us.” (11 Dec 2010)

Is this why Satoshi left the forum? It was posted only eighteen hours before his final post. Would that have given him enough time to make his decision and produce his final software update? Perhaps, but it would have been an overreaction. It was only PC World and just one article.

There was something else in the crucial four-day period which would have caused Satoshi far more alarm. A new poster called farmer_boy made a negative comment about Satoshi’s programming style:

“Another client is useful, especially since the current Bitcoin client is a big mess. I was *shocked* that cryptography code looked like this.” (10 Dec 2010)

Among those who posted back supporting Satoshi was Hal Finney: “To me it looks like an impressive job [….] This is some powerful machinery.” (11 Dec 2010)

But farmer_boy continued posting about Satoshi’s code:

“The cryptography code that competent people use, doesn’t look like a mess, btw. That 99% of the world uses this kind of obfuscated cryptography only shows that users don’t know what they use. Have you ever looked at TrueCrypt for example? It’s the exact same kind of mess.” (10 Dec 2010)

TrueCrypt was heavily based on Paul Le Roux’s E4M product. Some 80%-90% of the TrueCrypt code was Le Roux’s work, including all of the cryptography elements that Farmer Boy is criticising. Farmer Boy claims that Bitcoin is the “exact same kind of mess” as code that Le Roux wrote.

Two days after farmer_boy’s post Satoshi departed the forum for good without any farewell or explanation.