On 19 November 2013 17:01, Gregory Maxwell <gmaxwell@gmail.com> wrote:

On Tue, Nov 19, 2013 at 8:53 AM, Drak <drak@zikula.org> wrote:
> It's quite normal for standards bodies to allocate numbers when in draft
> status. If they don't pass, they don't pass - they are clearly labelled
> DRAFTs.
>
> +1 on having things in a github repository. Much better for collaboration,

The IETF makes a clear distinction between individual proposals and
documents which have been accepted by a working group. The former are
named after their authors. Work is not assigned a number until it is
complete.

I believe it is important to distinguish complete work that people
should be implementing from things which are incomplete, and even
more important to distinguish the work of single parties.

Otherwise you're going to get crap like BIP90: "Increase the supply of
Bitcoins to 210 million" being confused as an earnest proposal
supported by many that has traction.

I wasnt suggesting people add drafts willy nilly to the repository.

When working on a proposal you can work on it in your own fork and create a PR. When it's ready to be accepted as a working draft by the WG, then it can be merged into the draft folder. At which point, PRs are made to that draft copy until it gets into a ready state to become final. If passed, it's moved to the accepted/ folder.

This way random BIPS cannot be added to the drafts/ folder in the official repo. They are only added once they are accepted as a working draft proposal by Gavin or whatever. Now you get all the niceties of github workflow for collaboration and tweaking of the draft proposal.

Drak