Received-SPF: pass (sog-mx-2.v43.ch3.sourceforge.com: domain of gmail.com
	designates 209.85.216.181 as permitted sender)
	client-ip=209.85.216.181; envelope-from=gmaxwell@gmail.com;
	helo=mail-qc0-f181.google.com; 
MIME-Version: 1.0
In-Reply-To: <CANEZrP0nagdAXyMEY5yxyXBjdo78YC16mjUG9=b0AMe4qOS=fA@mail.gmail.com>
References: <CANEZrP0nagdAXyMEY5yxyXBjdo78YC16mjUG9=b0AMe4qOS=fA@mail.gmail.com>
Date: Sun, 24 Jun 2012 14:03:10 -0400
Message-ID: <CAAS2fgSWb2v6F2xbK0ZFNPz5cwgf6zwB8QbRYgyS1u8xmgAUTQ@mail.gmail.com>
From: Gregory Maxwell <gmaxwell@gmail.com>
To: Mike Hearn <mike@plan99.net>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Cc: Bitcoin Dev <bitcoin-development@lists.sourceforge.net>
Subject: Re: [Bitcoin-development] Enforcing inflation rules for SPV clients
Precedence: list

On Sun, Jun 24, 2012 at 8:45 AM, Mike Hearn <mike@plan99.net> wrote:
> d'aniel made a good proposal - having good nodes broadcast
> announcements when they detect a rule that breaks the rules, along
> with a proof that it did so. Checking the proof might be very

Link?

I also proposed this on this list (see the response in the tree
datastructures thread) along with more elaboration on IRC. If multiple
people are coming up with it thats a good sign that it it might
actually be viable. :)

I was going for a slightly different angle and pointing out that the
proofs would mean that a node doing validation with TxOUT tree which
hasn't personally wittnessed the complete history of Bitcoin actually
has basically the same security=E2=80=94 including resistance to miners
creating fake coin in the past=E2=80=94 as a full node today because in ord=
er
to get away with a lie every single node must conspire: It's adequate
that only one honest node wittness the lie because once it has the
proof information is hard to suppress.

To save people from having to dig through the public IRC logs for what
I wrote there:

--- Day changed Thu Jun 21 2012
15:10 < gmaxwell> etotheipi_: amiller: an interesting point with all
this txout tree stuff is that if you join the network late and just
trust that the history is correct based on the headers, any other node
who has witnessed a rule violation in the past can prepare a small
message which you would take to be conclusive proof of a rule
violation and then ignore that chain.
15:11 < gmaxwell> e.g. if someone doublespends I just take the
conflicting transactions out and the segments connecting them to the
chain... and show them to you. And without trusting me you can now
ignore the entire child chain past that point.
15:13 < gmaxwell> This fits nicely with the Satoshi comment "It takes
advantage of the nature of information being easy to spread but hard
to stifle" ...  it would be safe to late-join a txout tree chain,
because if there is only a single other honest node in the world who
was around long enough to wittness the cheating, he could still tell
you and it would be as good as if you saw it yourself.
15:17 < gmaxwell> (this is akin to the provable doublespend alert
stuff we talked about before, but applied to blocks)