Return-Path: Received: from smtp1.linuxfoundation.org (smtp1.linux-foundation.org [172.17.192.35]) by mail.linuxfoundation.org (Postfix) with ESMTPS id 77654ED0 for ; Fri, 4 Sep 2015 21:37:03 +0000 (UTC) X-Greylist: whitelisted by SQLgrey-1.7.6 Received: from mail-ig0-f172.google.com (mail-ig0-f172.google.com [209.85.213.172]) by smtp1.linuxfoundation.org (Postfix) with ESMTPS id DED57124 for ; Fri, 4 Sep 2015 21:37:02 +0000 (UTC) Received: by igbni9 with SMTP id ni9so23427545igb.0 for ; Fri, 04 Sep 2015 14:37:02 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:sender:in-reply-to:references:from:date:message-id :subject:to:content-type; bh=Ml0RMFjQZxfcQzQfJu2Y3+qtAQe2b/3U4ibQnfWVrac=; b=ViGXR9XB74eaonOAgnFH92Dov6VRhl1jtVfWJ6vU0BLNDjw/g6WIV8hhHkWAtOeErE R/sZfKMt3jFk5eG6XLhBaJ4OFp3OQ6KqFD/ScEFGmgHxkETHgHUiSoyBtt/KPhhDNVsa HIbRQhqoItTZDiAX+urhBDUqDk3MLGSoZiyDF5zIarkW4xOMCPBvELDbvkpkmvLpHhBN rlZA/pLPcLGNJBQ8DyKGVYUwbo883vDQ+C35vfFzk4CfUy0uLF8uiyLYPPirYhtUgJ07 cuNAZdb+ldtMWBbsGNJKw+hYmqLydQtUTnObotPXtlegtc+wTwnf3AvPDC4Ow0uGuo4S of8g== X-Received: by 10.50.118.33 with SMTP id kj1mr685737igb.97.1441402622382; Fri, 04 Sep 2015 14:37:02 -0700 (PDT) MIME-Version: 1.0 Sender: asperous2@gmail.com Received: by 10.50.3.33 with HTTP; Fri, 4 Sep 2015 14:36:42 -0700 (PDT) In-Reply-To: <201509042101.11839.luke@dashjr.org> References: <64B72DF6-BE37-4624-ADAA-CE28C14A4227@gmail.com> <201509042101.11839.luke@dashjr.org> From: Andy Chase Date: Fri, 4 Sep 2015 14:36:42 -0700 X-Google-Sender-Auth: 0To_8VVFQvr0C0XIgFwVeAhV6WM Message-ID: To: Luke Dashjr , bitcoin-dev@lists.linuxfoundation.org, pete@petertodd.org Content-Type: multipart/alternative; boundary=089e012947524918ac051ef2b3ef X-Spam-Status: No, score=-2.4 required=5.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,FREEMAIL_ENVFROM_END_DIGIT,FREEMAIL_FROM, HTML_MESSAGE,RCVD_IN_DNSWL_LOW autolearn=ham version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on smtp1.linux-foundation.org Subject: Re: [bitcoin-dev] [BIP/Draft] BIP Acceptance Process X-BeenThere: bitcoin-dev@lists.linuxfoundation.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Bitcoin Development Discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 04 Sep 2015 21:37:03 -0000 --089e012947524918ac051ef2b3ef Content-Type: text/plain; charset=UTF-8 I understand your concerns. What kinds of changes do you think should go through a process like this? Just hard forks? I was thinking that an advantage of making all BIPs use this process is that it makes it familiar and well used. Kinda like a muscle grows stronger with use. If only hard forks go through the process then there's the risk that the process has to be spun up whenever they happen which might cause confusion. Another reason I was thinking is that even small, local changes, it doesn't hurt to have a few more people take a look at it and approve it. The bureaucracy only applies to BIPs, not PRs. There's only been 18 approved/final/accepted BIPs in 4 years since BIP-0001. That's only about ~5 per year. I get that bureaucracy is often a waste of time, but I just don't think every second counts for these things. On Fri, Sep 4, 2015 at 2:01 PM, Luke Dashjr wrote: > On Friday, September 04, 2015 8:13:18 PM Andy Chase via bitcoin-dev wrote: > > Who makes high-level Bitcoin decisions? Miners, client devs, merchants, > or > > users? Let's set up a system where everyone has a say and clear > acceptance > > can be reached. > > For hardforks (removing consensus rules), economic consensus: people who > accept payment in bitcoins weighted by their actual volume of such > payments. > A supermajority subset may arguably be sufficient for some hardforks (which > don't violate Bitcoin's social contract) since they can effectively compel > the remaining economy to comply. > > For softforks (adding consensus rules), a majority of miners: they can "51% > attack" miners who don't go along with it. > > Anything else does not necessarily need universal agreement, so are > completely up to the whim of individual software projects. If someone > doesn't > like a decision in Core (for example), they can safely fork the code. If > any > significant amount of people use their fork, then the BIP is accepted > whether > or not Core later adopts it. > > Note this "system" is really describing a lack of a system - that is, what > naturally must happen for changes to occur. Softforks have a relatively > mature technical method for measuring support and deploying (which I > believe > someone else is already working on a BIP describing), but the same thing is > impractical for hardforks. Some formal way to measure actual economic > acceptance seems like a good idea to study, but it needs to be reasonably > accurate so as to not change the outcome from its natural/necessary result. > > Luke > --089e012947524918ac051ef2b3ef Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
I understand your concerns. What kinds of changes do you t= hink should go through a process like this? Just hard forks?

=
I was thinking that an advantage of making all BIPs use this process i= s that it makes it familiar and well used. Kinda like a muscle grows strong= er with use. If only hard forks go through the process then there's the= risk that the process has to be spun up whenever they happen which might c= ause confusion.

Another reason I was thinking is t= hat even small, local changes, it doesn't hurt to have a few more peopl= e take a look at it and approve it.

The bureaucrac= y only applies to BIPs, not PRs. There's only been 18 approved/final/ac= cepted BIPs in 4 years since BIP-0001. That's only about ~5 per year. I= get that bureaucracy is often a waste of time, but I just don't think = every second counts for these things.

<= div class=3D"gmail_quote">On Fri, Sep 4, 2015 at 2:01 PM, Luke Dashjr <luke@= dashjr.org> wrote:
On Friday, Septe= mber 04, 2015 8:13:18 PM Andy Chase via bitcoin-dev wrote:
> Who makes high-level Bitcoin decisions? Miners, client devs, merchants= , or
> users? Let's set up a system where everyone has a say and clear ac= ceptance
> can be reached.

For hardforks (removing consensus rules), economic consensus: people= who
accept payment in bitcoins weighted by their actual volume of such payments= .
A supermajority subset may arguably be sufficient for some hardforks (which=
don't violate Bitcoin's social contract) since they can effectively= compel
the remaining economy to comply.

For softforks (adding consensus rules), a majority of miners: they can &quo= t;51%
attack" miners who don't go along with it.

Anything else does not necessarily need universal agreement, so are
completely up to the whim of individual software projects. If someone doesn= 't
like a decision in Core (for example), they can safely fork the code. If an= y
significant amount of people use their fork, then the BIP is accepted wheth= er
or not Core later adopts it.

Note this "system" is really describing a lack of a system - that= is, what
naturally must happen for changes to occur. Softforks have a relatively
mature technical method for measuring support and deploying (which I believ= e
someone else is already working on a BIP describing), but the same thing is=
impractical for hardforks. Some formal way to measure actual economic
acceptance seems like a good idea to study, but it needs to be reasonably accurate so as to not change the outcome from its natural/necessary result.=

Luke

--089e012947524918ac051ef2b3ef--