[Cryptography] Transactions, was BitCoin bug reported

John Levine johnl at iecc.com
Sun Feb 16 14:53:07 EST 2014

>Now, this may seem to be just pedantry about wordage, but the issue is
>that every user, caller, business, accountant, auditor, competitor,
>regulator, and indeed person understands that a transaction is the ONE.
> There is the transaction, full stop.

Sheesh.  Here in the quaint United States, people still write a lot of
paper checks (that's check with a "k".)  Back in my youth, the actual
paper checks were routinely returned with my statements, but these
days if you're lucky you'll get images of the checks on the statement,
or sometimes just the check numbers.  

The recipient can deposit the physical check at his bank, or he can
make an image of the check using a device attached to a point of sale
terminal, a scanner provided by the recipient's bank, or even by taking
a picture of both sides with a mobile phone app.  In some cases where
the check is processed by someone that still handles paper checks, the
image is turned back into a substitute check, a piece of paper with a
printout of the images and the marginal legend SUBSTITUTE CHECK.  Then
the substitute is itself imaged and a picture of it shows up on my

Despite all that, all of those are the same check and the same
transaction, even though the form of the transaction varies all over
the place.  If I were to deposit a check in one account by taking a
picture, and then physically deposit the check in another account, I
am quite confident that at least one of those would bounce.  (I
haven't tried it.)

If a bitcoin app doesn't recognize that two forms of the same
transaction are in fact the same transaction, that's just a bug,
not a philosophical event.


More information about the cryptography mailing list