References: <1327812740.41242.YahooMailNeo@web121002.mail.ne1.yahoo.com>
	<CAAS2fgRWJb+p=gM4eG_p4fr1+UsT-EO_YZOvkgh6SnktRRTyEw@mail.gmail.com>
	<1327835941.47827.YahooMailNeo@web121006.mail.ne1.yahoo.com>
Message-ID: <1327835984.12365.YahooMailNeo@web121002.mail.ne1.yahoo.com>
Date: Sun, 29 Jan 2012 03:19:44 -0800 (PST)
From: Amir Taaki <zgenjix@yahoo.com>
To: "bitcoin-development@lists.sourceforge.net"
	<bitcoin-development@lists.sourceforge.net>
In-Reply-To: <1327835941.47827.YahooMailNeo@web121006.mail.ne1.yahoo.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: [Bitcoin-development] Fw:  Quote on BIP 16
Precedence: list
Reply-To: Amir Taaki <zgenjix@yahoo.com>

(oops sorry greg- replied to you by mistake)=0A=0AThat address he gives is =
77 characters/bytes (same thing). What I'm asking is how can it be so small=
. I know that it's encoding a script, but then I started trying to imagine =
what kind of script and to me it seems that 2 public keys are too large for=
 those 77 characters when encoded.=0A=0AThat is the example quoted on the f=
orums:=0A57HrrfEw6ZgRS58dygiHhfN7vVhaPaBE7HrrfEw6ZgRS58dygiHhfN7vVhaPaBiTE7=
vVhaPaBE7Hr=0A=0A=0ACould it be a mistake?=0A=0A=0A----- Original Message -=
----=0AFrom: Gregory Maxwell <gmaxwell@gmail.com>=0ATo: Amir Taaki <zgenjix=
@yahoo.com>=0ACc: "bitcoin-development@lists.sourceforge.net" <bitcoin-deve=
lopment@lists.sourceforge.net>=0ASent: Sunday, January 29, 2012 5:19 AM=0AS=
ubject: Re: [Bitcoin-development] Quote on BIP 16=0A=0AOn Sat, Jan 28, 2012=
 at 11:52 PM, Amir Taaki <zgenjix@yahoo.com> wrote:=0A> How could you have =
a 70 byte long address without a P2SH scheme? Is this a mistake?=0A=0A...=
=A0 No it's not a mistake.=A0 P2SH _prevents_ needing long addresses.=0A=0A=
Lets unpack the acronym "pay to script _hash_".=A0 Hashes only need to=0Abe=
 128-256 bits in size or so to have acceptable security, so you=0Adon't nee=
d something longer than that for paying to a hash.=0A=0ANote that gavin is =
saying 70 characters, not bytes.=0A=0AWithout some form of P2SH then only w=
ay for you to make a personal=0Achoice of asking people to pay to a two-fac=
tor protected account or=0Atwo a multiparty trust that manages the finances=
 of an organization=0Ais using some form of "P2S", pay-to-script.=0A=0AIn o=
ther words, you'd have to have an address that encodes a full=0Ascript spec=
ification for the sender to pay to,=A0 instead of just=0Aencoding its hash.=
=A0 As a result these addresses would be much longer=0A(and potentially ver=
y long).=0A=0AThe minimum size of a two address involving encoded script wo=
uld be on=0Athat order, but they get bigger quite quickly if you add more o=
ptions=0Ato the script (actually 70 sounds quite small, it should be more l=
ike=0A100 for a minimum two pubkey script).=0A=0AIn addition to the unworka=
bility of very long addresses as described=0Aby gavin (amusingly I am unabl=
e to copy and paste the quoted example=0Ain one go) a P2S solution has seve=
ral problems which you might=0Aconsider more or less important:=0A=0A=0A(1)=
 They are highly vulnerable to invisible substitution.=A0 E.g. I can=0Atriv=
ially take a P2S address, change one or two characters and get a=0Ascript w=
hich is redeemable by anyone.=A0 With P2SH you have to do=0Acomputation whi=
ch is exponential in the number of unchanged digits to=0Aget a look alike a=
ddress.=0A=0A(2) The sender is fully responsible for fees related to the en=
larged=0Atransactions. Even if _you're_ willing to take the txn-processing =
time=0Aand fee burden of a 30 person joint trust address,=A0 random e-comme=
rce=0Asites will not be and will randomly reject your addresses.=0A=0A(3) T=
hey create another input vector for non-trivial data which must=0Abe inspec=
ted and validated, potentially presenting an attack surface.=0A=0A(4) They =
leave the complicated (long) release rules in the transaction=0Aoutputs.=A0=
 When a transaction is mined we can't be sure if it will ever=0Abe redeemed=
. The outputs are unprunable.=A0=A0 In a future world where=0Amany nodes pr=
une output space is far more important than input space=0Aand it would make=
 sense to require more fees for it because we're=0Anever sure how long it w=
ould need to be stored (making it an=0Aattractive target for someone who wa=
nts to make Bitcoin unusable by=0Aspamming it with worthless data).=A0 P2SH=
 reduces output sizes to the=0Aabsolute minimum without inflating the total=
 data size.