Received-SPF: pass (sog-mx-4.v43.ch3.sourceforge.com: domain of gmail.com
	designates 209.85.223.171 as permitted sender)
	client-ip=209.85.223.171; envelope-from=gmaxwell@gmail.com;
	helo=mail-ie0-f171.google.com; 
MIME-Version: 1.0
In-Reply-To: <CABjHNoTeMiLWkDBUqdV4HJ=nAhj8wqOjD4cypY9Dv2y9HJWJMg@mail.gmail.com>
References: <552EF785.7000207@sky-ip.org>
	<CAPg+sBgAhdgPPjmT5i0PMYhQo=Hk6Weo8tpX_Wyn-NJ5Ye9D_A@mail.gmail.com>
	<552FDF73.6010104@sky-ip.org>
	<CABjHNoTeMiLWkDBUqdV4HJ=nAhj8wqOjD4cypY9Dv2y9HJWJMg@mail.gmail.com>
Date: Fri, 24 Apr 2015 20:16:57 +0000
Message-ID: <CAAS2fgSay0DqeWXfZwX-sN71sLHdRLD51PBmnJfJ5+TC0BQ8zg@mail.gmail.com>
From: Gregory Maxwell <gmaxwell@gmail.com>
To: William Swanson <swansontec@gmail.com>
Content-Type: text/plain; charset=UTF-8
Cc: Bitcoin Dev <bitcoin-development@lists.sourceforge.net>
Subject: Re: [Bitcoin-development] 75%/95% threshold for transaction versions
Precedence: list

On Fri, Apr 24, 2015 at 7:58 PM, William Swanson <swansontec@gmail.com> wrote:
> On Thu, Apr 16, 2015 at 9:12 AM, s7r <s7r@sky-ip.org> wrote:
>> Thanks for your reply. I agree. Allen has a good point in the previous
>> email too, so the suggestion might not fix anything and complicate things.
>
> The BIP 62 approach to malleability isn't the only option. Another
> approach is to sign the transaction in such a way that the input
> txid's are allowed to change without invalidating the signatures. That
> way, if malleability happens, you just adjust you transaction to match
> and re-broadcast. That proposal is here:

This is not a free choice. There are several concerns, from mild to
severe, that arise when you do not sign enough.

In particular not covering the ID allows for transaction replay which
can result in monetary losses far more severe than any possible
mishandling of malleability could result in. Byzantine attackers can
costlessly replay your old transactions any time anyone reuses an
address, even accidentally (which cannot be easily prevented since
they can race).

Other fun effects also show up like being able to backwards compute
signatures to result in a kind of limited covenant- coins which can
only be spent a particular way which has some implications for
fungibility. (See here for a discussion in general of covenants:
https://bitcointalk.org/index.php?topic=278122.0)

There are no free lunches;  the proposal linked to there is itself a
game of wack-a-mole with assorted masking flags; many of which we have
no notion of if they're useful for any particular application(s); and
it doesn't provide tools to address the replay issue; and in order to
'improve' malleability via that mechanism you must always mask out the
inputs completely; meaning you'd always be exposed to replay and not
just in specialized 'contract' applications where "there won't be
address reuse" could be a strong assumption enforced by the
application.