MIME-Version: 1.0
In-Reply-To: <CABsx9T3NFRO5nw3z=jrs0Hu3caVNkkTTTb1ibqR7LMWsoou9RQ@mail.gmail.com>
References: <CABsx9T2+dG0AE+MgKRAU97KhkHTU1MuxXuwHKv3BgpJswZ5vVg@mail.gmail.com>
	<CABaSBaxcDRzw0X7-fAfxPJyLcWxTHigpHuAPb4aNQ5zk5NoDCQ@mail.gmail.com>
	<CAAS2fgTr-OuL3T6mXX-4xFC_LHnAiogTTcPMbcjsM7WtRisQEQ@mail.gmail.com>
	<CABsx9T3NFRO5nw3z=jrs0Hu3caVNkkTTTb1ibqR7LMWsoou9RQ@mail.gmail.com>
Date: Sun, 27 Sep 2015 01:39:00 +0000
Message-ID: <CAAS2fgRj+fE+znXZzFsXXBivKSxnJ2Lheo_g9us4FXN_yCLhgw@mail.gmail.com>
From: Gregory Maxwell <gmaxwell@gmail.com>
To: Gavin Andresen <gavinandresen@gmail.com>
Content-Type: text/plain; charset=UTF-8
Cc: Bitcoin Dev <bitcoin-dev@lists.linuxfoundation.org>
Subject: Re: [bitcoin-dev] Weak block thoughts...
Precedence: list

On Wed, Sep 23, 2015 at 9:37 PM, Gavin Andresen <gavinandresen@gmail.com> wrote:
>> Avoiding this is why I've always previously described this idea as
>> merged mined block DAG (with blocks of arbitrary strength) which are
>> always efficiently deferentially coded against prior state. A new
>> solution (regardless of who creates it) can still be efficiently
>> transmitted even if it differs in arbitrary ways (though the
>> efficiency is less the more different it is).
>
> Yup, although I don't get the 'merge mined' bit; the weak blocks are
> ephemeral, probably purged out of memory as soon as a few full blocks are
> found...

Unless the weak block transaction list can be a superset of the block
transaction list size proportional propagation costs are not totally
eliminated.

As even if the weak block criteria is MUCH lower than the block
criteria (which would become problematic in its own right at some
point) the network will sometimes find blocks when there hasn't been
any weak block priming at all (e.g. all prior priming has made it into
blocks already).

So if the weak block commitment must be exactly the block commitment
you end up having to add a small number of transactions to your block
above and beyond the latest well propagated weak-blocks... Could still
work, but then creates a pressure to crank up the weak block overhead
which could better be avoided.