Received-SPF: pass (sog-mx-1.v43.ch3.sourceforge.com: domain of gmail.com
	designates 74.125.82.45 as permitted sender)
	client-ip=74.125.82.45; envelope-from=mh.in.england@gmail.com;
	helo=mail-wg0-f45.google.com; 
MIME-Version: 1.0
Sender: mh.in.england@gmail.com
In-Reply-To: <CAFVoEQQHVcY0Ad-4c2wnH+WF_7M-o5SNwVr-nce_9bQ794cwDQ@mail.gmail.com>
References: <etPan.54d0b945.46e87ccd.7f23@Williams-MBP>
	<CAFVoEQQHVcY0Ad-4c2wnH+WF_7M-o5SNwVr-nce_9bQ794cwDQ@mail.gmail.com>
Date: Tue, 3 Feb 2015 22:01:47 +0100
Message-ID: <CANEZrP29oJU1i5mN=TmxRsVk4m5mJy9jf-=4gYvRKRfq11pkBg@mail.gmail.com>
From: Mike Hearn <mike@plan99.net>
To: Adam Weiss <adam@signal11.com>
Content-Type: multipart/alternative; boundary=f46d041826be01d8f0050e3561b5
Cc: Bitcoin Dev <bitcoin-development@lists.sourceforge.net>,
	Will <will.madden@novauri.com>
Subject: Re: [Bitcoin-development] Subject: Re: Proposal to address Bitcoin
	malware
Precedence: list

--f46d041826be01d8f0050e3561b5
Content-Type: text/plain; charset=UTF-8

>
> TREZOR like devices with BIP70 support and third party cosigning services
> are a solution I really like the sound of.  I suppose though that adding
> BIP70 request signature validation and adding certificate revocation
> support starts to balloon the scope of what is supposed to be a very simple
> device though.
>

Yes, X.509 is ....... unfortunate. We'll have to wait and see how the
TREZOR team get on with implementing it. TREZOR doesn't have any OS at all
at the moment, so an implementation of PKIX will probably end up being
larger than their existing codebase.

That said, X.509 parsing is so security critical that the existing
codebases for it are by now pretty robust. Touch wood. So just having a
super stripped down OpenSSL implementation is probably good enough.

W.R.T revocation, BIP70 doesn't support this. If your private key leaks
you're currently hosed, identity wise, until the certificate expires. This
is obviously suboptimal. In a world where we all have infinite time and
resources the right fix will be to piggy back on an X.509 extension being
proposed in the browser world called "Must Staple". It's a bit in the
certificate flags that tell the client to expect a stapled OCSP response
and to hard-fail if none is provided. By requesting the CA set this flag
when you get your certificate issued, you sign up for more pain but more
security.

An OCSP stapling extension to BIP70 would probably not be very hard to
implement, but it'd be pointless today because the client has no idea
whether to expect it or not. The absence of a certificate changes the UI by
showing you a random Bitcoin address instead of a human readable name, but
the absence of stapled OCSP would not result in any UI change.


> Regardless, I think a standard for passing partially signed transactions
> around might make sense
>

I'm hoping that the hardware wallet world just standardises on the TREZOR
protocol. It's well designed and these devices all have fairly similar
capabilities.

--f46d041826be01d8f0050e3561b5
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div class=3D"gmail_extra"><div class=3D"gmail_quote"><blo=
ckquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #c=
cc solid;padding-left:1ex"><div dir=3D"ltr"><div class=3D"gmail_extra"><div=
 class=3D"gmail_quote"><div>TREZOR like devices with BIP70 support and thir=
d party cosigning services are a solution I really like the sound of.=C2=A0=
 I suppose though that adding BIP70 request signature validation and adding=
 certificate revocation support starts to balloon the scope of what is supp=
osed to be a very simple device though.</div></div></div></div></blockquote=
><div><br></div><div>Yes, X.509 is ....... unfortunate. We&#39;ll have to w=
ait and see how the TREZOR team get on with implementing it. TREZOR doesn&#=
39;t have any OS at all at the moment, so an implementation of PKIX will pr=
obably end up being larger than their existing codebase.=C2=A0</div><div><b=
r></div><div>That said, X.509 parsing is so security critical that the exis=
ting codebases for it are by now pretty robust. Touch wood. So just having =
a super stripped down OpenSSL implementation is probably good enough.</div>=
<div><br></div><div>W.R.T revocation, BIP70 doesn&#39;t support this. If yo=
ur private key leaks you&#39;re currently hosed, identity wise, until the c=
ertificate expires. This is obviously suboptimal. In a world where we all h=
ave infinite time and resources the right fix will be to piggy back on an X=
.509 extension being proposed in the browser world called &quot;Must Staple=
&quot;. It&#39;s a bit in the certificate flags that tell the client to exp=
ect a stapled OCSP response and to hard-fail if none is provided. By reques=
ting the CA set this flag when you get your certificate issued, you sign up=
 for more pain but more security.</div><div><br></div><div>An OCSP stapling=
 extension to BIP70 would probably not be very hard to implement, but it=
9;d be pointless today because the client has no idea whether to expect it =
or not. The absence of a certificate changes the UI by showing you a random=
 Bitcoin address instead of a human readable name, but the absence of stapl=
ed OCSP would not result in any UI change.</div><div>=C2=A0</div><blockquot=
e class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc sol=
id;padding-left:1ex"><div dir=3D"ltr"><div class=3D"gmail_extra"><div class=
=3D"gmail_quote"><div>Regardless, I think a standard for passing partially =
signed transactions around might make sense</div></div></div></div></blockq=
uote><div><br></div><div>I&#39;m hoping that the hardware wallet world just=
 standardises on the TREZOR protocol. It&#39;s well designed and these devi=
ces all have fairly similar capabilities.=C2=A0</div></div></div></div>

--f46d041826be01d8f0050e3561b5--