bips
bips copied to clipboard
bitcoin
BIP 2: Allow editors to fix typos
Please see https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2017-September/015065.html
@luke-jr If this can't be implemented easily in GitHub's system, I'm willing to help out by watching for these kinds of Pull Requests and then fast-tracking them in when they are only spelling/grammar corrections (after a quick read of them first, of course).
Nobody's approval matters other than the author of the BIP being modified (unless it's s a Final/Active BIP, in which case the usual requirements apply also).
Approving changes you don't have authority to approve is at best a waste of time. :/
I'm in support of this, perhaps requiring sign off from multiple BIP contributors (not necessarily the BIP author) that the meaning is unchanged, as suggested by cdecker.
This would be a good idea since it wouldn't waste people's time to ACK trivial stuff and also if the contributor(s) of the BIP(s) is/are not available, it prevents merging trivial fixes.
Hi @luke-jr, it would be nice to move this forward. Can you please respond to feedback on the mailing list and/or modify wording here if you agree?
Here is a summary:
@kanzure :
Even minor revisions can not change the meaning of text. Changing a single word can often have a strange impact on the meaning of the text. There should be some amount of care exercised here. Maybe it would be okay as long as edits are mentioned in the changelog at the bottom of each document, or mention that the primary authors have not reviewed suggested changes, or something as much; otherwise the reader might not be aware to check revision history to see what's going on.
@Sjors :
Perhaps it's enough to @mention authors in the PR and give them a week to object before merging?
Jean-Paul Kogelman :
Perhaps having authors consent to certain types of changes when they submit their BIP?
@cdecker :
Agreed, I think a sign-off mechanism might be desirable. Currently it must be the original author(s) signing off, but we can probably widen that to be any 2-3 community members. They'd basically be attesting that the meaning did not change.
@petertodd :
As part of this, we may want to say that the BIP editor should cryptographically sign (and ideally timestamp) all their changes as a secondary measure to make it clear who actually made the change.
I got the impression there isn't consensus for this change. @kanzure's requirements in particular seem to make it more effort than it's worth, so I'm inclined to just close the PR...
OTOH, maybe I should point out that these are just documents, and if there's ever a problem, things can be reverted with little harm...?
I'm happy for editors to fix typos even without kanzure's requirements.
Check PGP signatures if you want to be totally sure.
On December 31, 2018 7:25:56 AM UTC, Luke Dashjr [email protected] wrote:
I got the impression there isn't consensus for this change. @kanzure's requirements in particular seem to make it more effort than it's worth, so I'm inclined to just close the PR...
OTOH, maybe I should point out that these are just documents, and if there's ever a problem, things can be reverted with little harm...?
-- https://petertodd.org 'peter'[:-1]@petertodd.org
I'd be fine with simply requiring that changes be pgp signed (although it might reduce participation a bit). Remember, this is only about fixing obvious typos and everything will be recorded in git history.
Oh! I'm ok with both allowing editors to use their judgement in fixing typos, and not requiring PGP signatures on contributions.
These are after all just design docs; leave it up to the authors to decide if they want the security of a PGP signature on their contributions. It's mostly their problem after all as the harm would be to their reputation.
I'm signing (and timestamping) all of mine!
On January 1, 2019 4:37:34 PM UTC, Jonathan Cross [email protected] wrote:
I'd be fine with simply requiring that changes be pgp signed (although it might reduce participation a bit). Remember, this is only about fixing obvious typos and everything will be recorded in git history.
-- https://petertodd.org 'peter'[:-1]@petertodd.org
I would consider it a strict improvement if BIP editors had permission to merge non-meaning changing PRs to BIPs. Permitted should be to fix a typo or a broken link (where it can be adjusted to link to the new location of the original document). Not permitted without sign-off from the authors should be changing any words or adding punctuation.
I would consider it a strict improvement if BIP editors had permission to merge non-meaning changing PRs to BIPs. Permitted should be to fix a typo or a broken link (where it can be adjusted to link to the new location of the original document). Not permitted without sign-off from the authors should be changing any words or adding punctuation.
I agree with this and would like something closer to this language to land in the actual proposed change. As it stands right now the diff suggests something more vague than what @murchandamus writes here.
ACK ce5d831516bfdb34904a765b3f742e38621ec036
I agree with @murchandamus as well.
I'd rather prefer using some kind of layers or ignorelists in order to leave those PRs on hold indefinitely while sticking to process, but I think that would imply migrating the ticket system finally (I appreciate 0xb10c work about that), so for the sake of collaborators sanity, I would hope for allowing BIP editors to merge typos as a temporary measure only in effect as long as the community is still using GitHub as issue tracker
On Tue, Apr 23, 2024 at 6:40 PM murchandamus @.***> wrote:
I would consider it a strict improvement if BIP editors had permission to merge non-meaning changing PRs to BIPs. Permitted should be to fix a typo or a broken link (where it can be adjusted to link to the new location of the original document). Not permitted without sign-off from the authors should be changing any words or adding punctuation.
— Reply to this email directly, view it on GitHub https://github.com/bitcoin/bips/pull/596#issuecomment-2073159256, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMRS4W4ZHLWEDRI7TQCASWLY62TJFAVCNFSM4D4Y6FJ2U5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TEMBXGMYTKOJSGU3A . You are receiving this because you are subscribed to this thread.Message ID: @.***>
ACK
As an author of several BIPs with far too many typos, I wholeheartedly agree with this change.
