Update EIP-1: Remove eip preamble entry

[Pandapip1]

Nothing really depends on the EIP number being in the preamble anymore, so we can safely remove it.

[eth-bot]

File EIPS/eip-1.md

Requires 2 more reviewers from @axic, @gcolvin, @lightclient, @SamWilsn

[abcoathup]

Nothing really depends on the EIP number being in the preamble anymore, so we can safely remove it.

Where is the EIP number specified if not in the preamble? Is it in the filename?

[Pandapip1]

Is it in the filename?

Yes.

[lightclient]

There is no need to remove the eip from the preamble.

[Pandapip1]

There is no need to remove the eip from the preamble.

Actually, it makes eip-review-bot slightly simpler, and also removes redundant information. So my stance is 'there is no need to keep the eip in the preamble.'

[SamWilsn]

I'm generally in favour of single-sourcing information as much as possible, so in this case I'd lean towards removing eip: from the preamble.

@lightclient are there any scenarios where removing the header is problematic?

[lightclient]

It's a long standing piece of the preamble. I think it is a unique piece of the EIP process and I would be sad to see it go. I would rather increase the complexity of the bots to deal with it than remove it. However, it's not something I feel terribly strongly about. What I do feel strongly about is the constant change in the repository. Things like this are marginal improvements that will likely cause more confusion to authors than do good in the short term.

I think it is time to start make changes at more spaced intervals and batching many of these changes together at once.

[Pandapip1]

What I do feel strongly about is the constant change in the repository.

Would you mind explaining what exactly you dislike about continuous improvement? How would batching changes address those concerns?

[lightclient]

The vast majority of contributors to this repository only do so occasionally. I am messaged frequently about "why is this like this now" and "when did this requirement get added / go away". If software is expected to batch breaking changes into major release, it should go doubly so for processes IMO.

[SamWilsn]

I don't think it significantly increases the technical complexity of the bots to include the EIP number in the preamble. We'd just need to figure out exactly how it should work.

On the eipw side:

1. If the filename matches eip-[0-9]+\.md:
   • The eip: header must exist.
   • The eip: header must match the filename.
2. If the filename doesn't match eip-[0-9]+\.md:
   • The filename must match the title header, with underscores.
   • The status must be Draft.
   • The eip header must not exist.

Sound right?

[Pandapip1]

Nope, I think that the eip header shouldn't exist, ever.

[Pandapip1]

If software is expected to batch breaking changes into major release, it should go doubly so for processes IMO.

How about versioned processes? So that older EIPs can get "grandfathered" and use older formatting/rules?

[xinbenlv]

My position is against of removing EIP number in preamble in the content.

Reason: the piece of information "number" is important enough to be replicated in the content and hence could be double checked, carried in publication, spread etc.

[lightclient]

Closing as there wasn't sufficient support in the EIPIP meeting.