Received-SPF: pass (sog-mx-1.v43.ch3.sourceforge.com: domain of gmail.com
	designates 209.85.214.173 as permitted sender)
	client-ip=209.85.214.173; envelope-from=mh.in.england@gmail.com;
	helo=mail-ob0-f173.google.com; 
MIME-Version: 1.0
Sender: mh.in.england@gmail.com
In-Reply-To: <CABsx9T0Q8g9KYRbAvCV=35x5Rb5HFnrNkrwwMZ=Mv-namMEPpg@mail.gmail.com>
References: <20141001130826.GM28710@savin.petertodd.org>
	<CABbpET8_FMCcnh0dELnHsYmF+YP05Gz=nZ3SPkLZuqXYV3JUpQ@mail.gmail.com>
	<1987325.zKPNeYyO8K@crushinator>
	<201410031750.27323.luke@dashjr.org>
	<CANEZrP1eGi-AHgciQiKUuUB7WwqKsMOyTjCQAAO=RWEkPC2Uiw@mail.gmail.com>
	<CAJHLa0NRNEQLqA2E=ysXsKw6hWS-H9X_AFYK4ckC4-_Bk=qbSA@mail.gmail.com>
	<20141004003850.GA23202@muck>
	<CANEZrP0_jDouDCLn9BvxUd7UYiZLbVsaGGkkxwjcOYxZryBDPQ@mail.gmail.com>
	<CABsx9T0Q8g9KYRbAvCV=35x5Rb5HFnrNkrwwMZ=Mv-namMEPpg@mail.gmail.com>
Date: Tue, 7 Oct 2014 18:08:13 +0200
Message-ID: <CANEZrP2Xp7ene+KDw_L_YnNW=hDt9K-UigvZ6PLb3oUviOr_Tw@mail.gmail.com>
From: Mike Hearn <mike@plan99.net>
To: Gavin Andresen <gavinandresen@gmail.com>
Content-Type: multipart/alternative; boundary=001a11c32f940702da0504d7689f
Cc: Bitcoin Dev <bitcoin-development@lists.sourceforge.net>,
	Flavien Charlon <flavien.charlon@coinprism.com>
Subject: Re: [Bitcoin-development] [BIP draft] CHECKLOCKTIMEVERIFY - Prevent
 a txout from being spent until an expiration time
Precedence: list

--001a11c32f940702da0504d7689f
Content-Type: text/plain; charset=UTF-8

>
> That is easy to change; I'll submit a pull request.
>

That's certainly a useful improvement. It won't help the existing userbase
though - assuming CHECKLOCKTIMEVERIFY is to go in to the next major
release. If there's going to be an intermediate release (6 months?) which
lays the groundwork for future rule changes, it helps more.

It would be good if getblocktemplate was updated at the same time to serve
errors if the fork warning is active. I'd hope miners have some way to
automatically handle IBD/getting forked off the chain, but I guess some
(newer) pools might not, and refusing to serve work should be the safest
option that shuts them down.

I don't have any opinion on the hard- versus soft- fork debate. I think
> either can work.
>

P2SH was a soft fork and the sky did not fall, but miners did lose money
and waste electricity mining blocks on the wrong side of the chain:

https://bitcointalk.org/index.php?topic=75294.0

Presumably they didn't notice for longer because it looked like a run of
unusually bad orphaning luck. It seems safer to have a clean fork, with
alerts telling people during the lockin period before new rule enforcement
starts (and possibly automated termination if there's no upgrade by the
flag day?). Miners who ignore it would still risk losing money, but
merchants who wait for a block at least would not be at risk.

One open question is how can you actually trigger a hard fork? Coinbase
scriptSigs are not executed, so putting some ignored but failing opcode
sequence there wouldn't work. One possibility would be to have a special
invalid tx in the block that marks the start of new rule enforcement. New
nodes would know to ignore it. But this risks corrupting block explorers.
Alternatively the coinbase outpoint structure could have its hash set to 1
instead of 0.

--001a11c32f940702da0504d7689f
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:0px 0px 0px 0.8ex;border-left=
-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;paddi=
ng-left:1ex"><div dir=3D"ltr"><div class=3D"gmail_extra"><div class=3D"gmai=
l_quote"><div>That is easy to change; I&#39;ll submit a pull request. </div=
></div></div></div></blockquote><div><br></div><div>That&#39;s certainly a =
useful improvement. It won&#39;t help the existing userbase though - assumi=
ng CHECKLOCKTIMEVERIFY is to go in to the next major release. If there&#39;=
s going to be an intermediate release (6 months?) which lays the groundwork=
 for future rule changes, it helps more.<br></div><div><br></div><div>It wo=
uld be good if getblocktemplate was updated at the same time to serve error=
s if the fork warning is active. I&#39;d hope miners have some way to autom=
atically handle IBD/getting forked off the chain, but I guess some (newer) =
pools might not, and refusing to serve work should be the safest option tha=
t shuts them down.</div><div><br></div><blockquote class=3D"gmail_quote" st=
yle=3D"margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb=
(204,204,204);border-left-style:solid;padding-left:1ex"><div style=3D"font-=
family:arial,sans-serif;font-size:12.7272720336914px">I don&#39;t have any =
opinion on the hard- versus soft- fork debate. I think either can work.</di=
v><div class=3D"" style=3D"font-family:arial,sans-serif;font-size:12.727272=
0336914px"></div></blockquote><div>=C2=A0</div><div>P2SH was a soft fork an=
d the sky did not fall, but miners did lose money and waste electricity min=
ing blocks on the wrong side of the chain:</div><div><br></div><div><a href=
=3D"https://bitcointalk.org/index.php?topic=3D75294.0">https://bitcointalk.=
org/index.php?topic=3D75294.0</a><br></div><div><br></div><div>Presumably t=
hey didn&#39;t notice for longer because it looked like a run of unusually =
bad orphaning luck. It seems safer to have a clean fork, with alerts tellin=
g people during the lockin period before new rule enforcement starts (and p=
ossibly automated termination if there&#39;s no upgrade by the flag day?). =
Miners who ignore it would still risk losing money, but merchants who wait =
for a block at least would not be at risk.</div><div><br></div><div>One ope=
n question is how can you actually trigger a hard fork? Coinbase scriptSigs=
 are not executed, so putting some ignored but failing opcode sequence ther=
e wouldn&#39;t work. One possibility would be to have a special invalid tx =
in the block that marks the start of new rule enforcement. New nodes would =
know to ignore it. But this risks corrupting block explorers. Alternatively=
 the coinbase outpoint structure could have its hash set to 1 instead of 0.=
</div></div></div></div>

--001a11c32f940702da0504d7689f--