From: Luke Dashjr <luke@dashjr.org>
To: Dave Scotese <dscotese@litmocracy.com>, Ryan Grant <bitcoin-dev@rgrant.org>
Date: Tue, 2 Feb 2016 07:54:29 +0000
User-Agent: KMail/1.13.7 (Linux/4.1.13-gentoo; KDE/4.14.8; x86_64; ; )
References: <201602012253.18009.luke@dashjr.org>
	<CAGLBAhffm+1m=DAph-ac8mA9ytLpKqTT45XG1r6UFGFoUvJ+PA@mail.gmail.com>
In-Reply-To: <CAGLBAhffm+1m=DAph-ac8mA9ytLpKqTT45XG1r6UFGFoUvJ+PA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: Text/Plain;
  charset="iso-8859-15"
Content-Transfer-Encoding: 7bit
Message-Id: <201602020754.31734.luke@dashjr.org>
Cc: Bitcoin Dev <bitcoin-dev@lists.linuxfoundation.org>
Subject: Re: [bitcoin-dev] BIP Process: Status, comments,
	and copyright licenses
Precedence: list

On Tuesday, February 02, 2016 5:50:29 AM Dave Scotese wrote:
> The section that starts "Should two software projects need to release"
> addresses issues that are difficult to ascertain from what is written
> there.  I'll take a stab at what it means:
> 
> Would bitcoin be better off if multiple applications provided their own
> implementations of API/RPC and corresponding application layer BIPs?
> 
>    - While there is only one such application, its UI will be the obvious
>    standard and confusion in usability will be avoided.
>    - Any more than a single such application will benefit from the
>    coordination encouraged and aided by this BIP and BIP 123.

The original question is intended to answer both: a) why only one 
implementation is insufficient for Final status, and b) why two is sufficient.

If every application had its own BIP (how I understand your version), none of 
them would be standards and it wouldn't make sense to have a BIP at all - just 
project documentation would be sufficient.

> "To avoid doubt: comments and status are unrelated metrics to judge a BIP,
> and neither should be directly influencing the other." makes more sense to
> me as "To avoid doubt: comments and status are intended to be unrelated
> metrics. Any influence of one over the other indicates a deviation from
> their intended use."  This can be expanded with a simple example: "In other
> words, a BIP having  the status 'Rejected' is no reason not to write
> additional comments about it.  Likewise, overwhelming support for a BIP in
> its comments section doesn't change the requirements for the 'Accepted' or
> 'Active' status."

Extending this to "influence" is probably too far - after all, comments may 
discourage implementations, which can very well result in the Status 
eventually becoming Rejected rather than Final. How about:

"To avoid doubt: comments and status are intended to be unrelated metrics. In 
other words, a BIP having the status 'Rejected' is no reason to write (or not 
write) additional comments about it, nor would a status of 'Final' preclude 
comments discouraging [further] implementation. Likewise, overwhelming support 
for a BIP in its comments section doesn't change the requirements for the 
'Final' or 'Active' status."

> Since the Bitcoin Wiki can be updated with comments from other places, I
> think the author of a BIP should be allowed to specify other Internet
> locations for comments.  So "link to a Bitcoin Wiki page" could instead be
> "link to a comments page (strongly recommended to be in the Bitcoin
> Wiki)". 

Hmm, I wonder if this could be too easily abuse to discourage comments 
(because the commenter does not wish to register with yet another forum), 
and/or censor negative comments (because the author has made his own forum 
specifically for the purpose).

On Tuesday, February 02, 2016 6:35:07 AM you wrote:
> For section "Formally defining consensus",
> 
> Where objections were not deemed substantiated by the community, clear
> reasoning must be offered.

I have integrated this into the draft.

> For section "BIP Comments",
> 
> Comments should be solicited on the bitcoin-dev mailing list, and
> summarized fairly in the wiki; with notice of summarization and time
> for suggesting edits on the mailing list.  Wiki registration and
> monitoring should not be a required hurdle to participation.

The intent is for the commenter to edit the wiki page himself. I have updated 
it to reflect this.

Luke