Received-SPF: pass (sog-mx-3.v43.ch3.sourceforge.com: domain of gmail.com
	designates 209.85.213.177 as permitted sender)
	client-ip=209.85.213.177; envelope-from=gmaxwell@gmail.com;
	helo=mail-ig0-f177.google.com; 
MIME-Version: 1.0
In-Reply-To: <CALqxMTE1OANaMAvqrcOLuKtYd_jmqYp5GcB4CX77S8+fR05=jg@mail.gmail.com>
References: <CALqxMTE2doZjbsUxd-e09+euiG6bt_J=_BwKY_Ni3MNK6BiW1Q@mail.gmail.com>
	<CANEZrP32M-hSU-a1DA5aTQXsx-6425sTeKW-m-cSUuXCYf+zuQ@mail.gmail.com>
	<CALqxMTFNdtUup5MB2Dc_AmQ827sM-t5yx7WQubbfOEd_bO_Ong@mail.gmail.com>
	<CANEZrP0cOY5Wt_mvBSdGGmi4NfZi04SQ7d6GLpnRxmqvXNArGA@mail.gmail.com>
	<CALqxMTE1OANaMAvqrcOLuKtYd_jmqYp5GcB4CX77S8+fR05=jg@mail.gmail.com>
Date: Fri, 20 Feb 2015 18:20:59 +0000
Message-ID: <CAAS2fgSsXDTzxS29_SZvy1_Tie8=EGDhUjGkyGTXbc=47ta20w@mail.gmail.com>
From: Gregory Maxwell <gmaxwell@gmail.com>
To: Adam Back <adam@cypherspace.org>
Content-Type: text/plain; charset=UTF-8
Cc: Bitcoin Dev <bitcoin-development@lists.sourceforge.net>
Subject: Re: [Bitcoin-development] bloom filtering, privacy
Precedence: list

On Fri, Feb 20, 2015 at 5:59 PM, Adam Back <adam@cypherspace.org> wrote:
> So now they ask a full node for merkle paths + transactions for the
> addresses from the UTXO set from the block(s) that it was found in.

Use of the words "UTXO set" here is probably confusing people as it's
likely to make people think of the complete verification state. In
this case it's simply referring to block-local data. (and thus avoids
the large IO overheads of an actual UTXO set).

It's a straight forward idea: there is a scriptpubkey bitmap per block
which is committed. Users can request the map, and be SPV confident
that they received a faithful one. If there are hits, they can request
the block and be confident there was no censoring.

It's possible to tree structure additional layers to the bitmap, so
one can incrementally query to trade0off map size for false request
overhead, it's not clear to me how much of a win this would be for
normal parameters.. It's also possible to extract the txout list for
the whole block and commit to that too so it's possible to request
just the outputs and get a faithful copy of them, which is _much_
smaller than the block overall.