Bitcoin Wiki

How the 2010-08-15 overflow incident unfolded: the soft-fork rescue, why a 5-hour response was possible only then, what the transactions themselves show, and the centralization paradox.

Satoshi responds to Hal Finney's botnet/pay-per-send point, proposing that fake mailboxes could 'reverse-spam' spammers by harvesting their POW tokens. He also describes e-gold's 'dusting' problem.

Trammell admits the mislabeling was his error, suggests changing 'Received with' to 'Received payment to' for clarity, and points to PayPal as the closest parallel to multiple addresses.

Satoshi explains the default address label is 'Your Address', suggests the mislabeling was a UI-driven user error, and acknowledges that per-payer receiving addresses have no real-world analogy.

Trammell realizes the confusing 'Satoshi' label was on his own receiving address, confirms he has multiple addresses at home, and asks if 'Satoshi' was the default since he doesn't recall setting it.

Satoshi explains the 'Satoshi' label came from Trammell's own address book, that transactions show the receiving address (not the sender), and recommends per-payer addresses to identify payers.

Trammell reports confusion about a 100 BTC transfer between his own two Bitcoin instances — the transaction details showed 'Satoshi' as a label, leading him to wonder if Satoshi had sent the coins.

Satoshi reveals the wallet location (%appdata%\Bitcoin), explains that its transactional database (DBM) survives crashes and power loss, and says socket cleanup code is in the next release.

Satoshi copies his exchange with Trammell to bitcoin-list and the Cryptography list, sharing his Bitcoin vision including the famous line 'It might make sense just to get some in case it catches on.'

Trammell confirms his IP is static, agrees to CC publicly, says he'll join bitcoin-list, reveals he started mining as an investment after Hal Finney's message, and discusses micropayment uptake.

Trammell argues Bitcoin addresses are more secure than IP-based sending since they can be verified via multiple channels. He proposes an address-advertisement toggle and reports an exit socket bug.

Satoshi responds to Trammell's MITM analysis by classifying attacks into two types (chain-of-communication vs. anyone on the Internet), proposes a combined IP+address fix, and notes wallet encryption.

Satoshi discusses dynamic IPs, asks permission to CC the conversation to bitcoin-list and the Cryptography mailing list, and envisions Bitcoin use cases including pay-to-send email and micropayments.

Trammell's detailed security analysis of Bitcoin's send-to-IP feature, identifying MITM vulnerabilities including ARP poisoning and ISP-level interception. Recommends always using Bitcoin addresses.

Trammell shares his IP for Satoshi to send coins, says proof-hashes is open for posting, names the challenge of getting people to value bitcoins, and compares mining to a roulette wheel.

Satoshi informs Trammell that bugs are fixed in v0.1.3 and offers to send him some coins via the send-to-IP feature. The payment became one of the earliest known direct Bitcoin transfers.

Trammell asks what prevents the most powerful node from generating the majority of bitcoins — one of the earliest questions about mining centralization.

Trammell reports two issues upgrading from v0.1.0 to v0.1.3: the old process wouldn't exit, and all four generated coins showed 'Generated (not accepted)' — likely orphans from the communications bug.

Trammell tells Satoshi about his proof-hashes Google Group, gives feedback on the credit field display, says he is running v0.1.1 and will upgrade, and offers to help test new features.

Satoshi replies to Dustin Trammell, explaining the coin maturity system and recommending an upgrade to version 0.1.3 which had stabilized the software.

Dustin Trammell's first email to Satoshi after running the Bitcoin alpha. He reports how it ran, mentions a public timestamp service, and asks about coin maturity (new coins showing 0.00 credit).