MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8;
 format=flowed
Content-Transfer-Encoding: 8bit
Date: Thu, 27 Aug 2015 23:02:47 -0400
From: jl2012@xbt.hk
To: Jeff Garzik <jgarzik@gmail.com>
In-Reply-To: <CADm_WcYXEu2dWCBWJwzSJnp6A4iFsR1eBXrQinQmd_NRXAs9Vw@mail.gmail.com>
References: <CADToNK93kmXjwSAxj3WcPSYgq2uReVNJSpX1a5M7DMzgjgv5vA@mail.gmail.com>
	<CADm_WcbEhMNWKCY6CqsSTRQztTz3Ap8PMkO+csz+jptc9Nbazg@mail.gmail.com>
	<CADToNK972BgCXtZScciF_F+Db_AYWP=obXSqUS7DozY-rAmT7g@mail.gmail.com>
	<CADm_WcYXEu2dWCBWJwzSJnp6A4iFsR1eBXrQinQmd_NRXAs9Vw@mail.gmail.com>
Message-ID: <038c418c354b4b4559137652282e88b4@xbt.hk>
User-Agent: Roundcube Webmail/1.0.5
Cc: Bitcoin development mailing list <bitcoin-dev@lists.linuxfoundation.org>
Subject: Re: [bitcoin-dev] Questiosn about BIP100
Precedence: list

Mode could be ruled out immediately. Just consider this: 34% 8MB, 33% 
1.5MB, 33% 1.2MB

I personally believe the median is the most natural and logical choice. 
51% of miners can always force the 49% to follow the simple majority 
choice through a 51% attack. Using median will eliminate the incentive 
to 51% attack due to this reason. The incentive to 51% attack will exist 
when you use any value other than 50-percentile. The further it is from 
50, the bigger the incentive.

Having said that, I don't think it is an absolutely bad idea to use a 
value other than 50-percentile. The exact value is debatable.

However, if you use something other than median, you should make it 
symmetrical. For example, the block size will increase if the 
20-percentile is bigger than the current limit, and the block size will 
decrease if the 80-percentile is smaller than the current limit.


Jeff Garzik via bitcoin-dev 於 2015-08-27 16:49 寫到:
> 20th percentile, though there is some argument to take the 'mode' of
> several tranches
> 
> On Thu, Aug 27, 2015 at 11:07 AM, Andrew C <achow101@gmail.com> wrote:
> 
>> I have been reading the pdf and one thing I can't figure out is what
>> you mean by "most common floor". Is that the smallest block size
>> that has a vote or the block size with the most votes or something
>> else?
>> 
>> On Mon, Aug 24, 2015 at 10:40 AM Jeff Garzik <jgarzik@gmail.com>
>> wrote:
>> 
>> Great questions.
>> 
>> - Currently working on technical BIP draft and implementation,
>> hopefully for ScalingBitcoin.org. Only the PDF is publicly
>> available as of today.
>> - Yes, the initial deployment is in the same manner as size votes.
>> 
>> On Fri, Aug 21, 2015 at 7:38 PM, Andrew C via bitcoin-dev
>> <bitcoin-dev@lists.linuxfoundation.org> wrote:
>> 
>> Hi all,
>> 
>> Is there any client or code that currently implements BIP 100? And
>> how will it be deployed? WIll the initial fork be deployed in the
>> same manner that the max block size changes are deployed described
>> in the bip?
>> 
>> Thanks
>> 
>> _______________________________________________
>> bitcoin-dev mailing list
>> bitcoin-dev@lists.linuxfoundation.org
>> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev [1]
> 
> 
> 
> Links:
> ------
> [1] https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev
> 
> _______________________________________________
> bitcoin-dev mailing list
> bitcoin-dev@lists.linuxfoundation.org
> https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev