To: Luke Dashjr <luke@dashjr.org>, bitcoin-dev@lists.linuxfoundation.org
References: <ea63ed5a-4280-c063-4984-5bc8a4b2aafa@gmail.com>
	<201702052302.29599.luke@dashjr.org>
From: Andrew C <achow101@gmail.com>
Message-ID: <03b80a7b-6c8c-cdff-1826-6535bef12993@gmail.com>
Date: Sun, 5 Feb 2017 18:53:03 -0500
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:45.0) Gecko/20100101
	Thunderbird/45.7.0
MIME-Version: 1.0
In-Reply-To: <201702052302.29599.luke@dashjr.org>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: quoted-printable
Subject: Re: [bitcoin-dev] A Modified Version of Luke-jr's Block Size BIP
Precedence: list


On 2/5/2017 6:02 PM, Luke Dashjr wrote:
> My BIP draft didn't make progress because the community opposes any blo=
ck size=20
> increase hardfork ever.
=46rom what I have observed, it seems to be that people are more so
opposed to a hard fork when there is a comparable soft fork available
than simply opposed to any block size increase hard fork ever. From the
various threads discussing your proposal, it seemed that many would
favor it if it increased over 1 MB sooner or if it never even decreased
in the first place.

>  Your version doesn't address the current block size=20
> issues (ie, the blocks being too large).=20
Many users are of the opposite opinion, that the block size is too
small. I understand that the decrease is to allow the blockchain size to
grow more slowly thereby allowing users to be more likely to run full
nodes. Unfortunately, I think that we are way past the point of no
return on that. The blockchain is already 100+ GB. Decreasing the block
size is not going to make that any smaller and is not going to make it
any less painful to run a full node. Given that in order to start up a
new full node will still require downloading at least 100 GB of data, I
don't think that decreasing the block size will better facilitate full
node creation. Furthermore, the current trend with ISPs (at least in the
US) is implementing data and bandwidth caps so users are still unlikely
to start up new full nodes regardless of any changes that we can
currently do.

> So you've retained the only certain-
> DOA parts of my proposal, and removed the most useful part... I'm not s=
ure the=20
> point. Also, your version is now EXCLUSIVELY a hardfork, so it makes no=
 sense=20
> to keep the BIP 9 deployment at all - either it gets consensus or it do=
esn't,=20
> but miners have no part in deployment of it.
Yes, I know deployment needs to be fixed. I was more proposing this for
comment on the modified block size schedule. I just kept the deployment
as it was originally. However, we could use a modified version of BIP 9
by using one of the top three bits and a longer locked-in period as a
grace period for all users to upgrade.
>
> On Sunday, February 05, 2017 9:50:26 PM Andrew C via bitcoin-dev wrote:=

>> Hello all,
>>
>> Many people have expressed discontent with Luke-jr's proposed block si=
ze
>> BIP, in particular with the decrease in size that would occur if it we=
re
>> to be activated prior to 2024.
>>
>> I have decided to modify the proposal to instead begin the increase
>> steps at the current 1000000 byte limit. The increases and the time sp=
am
>> of each increase will remain the same, just that the increase begins
>> from 1000000 bytes instead of 300000 bytes.
>>
>> Furthermore, instead of a fixed schedule from a fixed point in time, t=
he
>> increases will instead be calculated off of the MTP of the activation
>> block (the first block to be in the active state for this fork).
>>
>> While this proposal shares many of the same issues with the one it
>> modifies, I hope that it will be slightly less controversial and can
>> allow us to move forward with scaling Bitcoin.
>>
>> The full text of the proposal can be found at
>> https://github.com/achow101/bips/blob/bip-blksize/bip-blksize.mediawik=
i.
>> My implementation of it is available at
>> https://github.com/achow101/bitcoin/tree/bip-blksize
>>
>>
>> Andrew
>>
>> _______________________________________________
>> bitcoin-dev mailing list
>> bitcoin-dev@lists.linuxfoundation.org
>> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev