Skip to content
This repository was archived by the owner on Nov 9, 2017. It is now read-only.
This repository was archived by the owner on Nov 9, 2017. It is now read-only.

Buffer constructor: should we backport zero-filling? #91

Closed

Description

@Trott

Barring multiple last-minute vote changes, we're going to be zero-filling the Buffer constructor in Node.js 8.0.0

Final question from the set of votes held at this week's CTC meeting: Do we backport this change?

No one expressed approval for the idea of backporting to some release lines but not all. So I'm going to assume we're talking about backporting to all currently-supported release lines: 7.x, 6.x, and 4.x. If you favor backporting only to some release lines, leave a comment and we'll figure out how to account for it.

YES: @Fishrock123 @Trott @jasnell @MylesBorins @trevnorris

NO: @mscdex @mhdawson @evanlucas @fhinkel @thefourtheye @misterdjules @rvagg @indutny

ABSTAIN: @targos @cjihrig @addaleax @ofrobots @ChALkeR @shigeki

Activity

  1. Trott commented on Mar 31, 2017

    @Trott
    MemberAuthor

    @nodejs/ctc If you have not been accounted for above, please leave your vote or abstention in a comment. Thanks.

  2. thefourtheye commented on Mar 31, 2017

    @thefourtheye

    For example, if we backport to 6.x.y, the servers running older versions of 6 will not get this change. And when they eventually upgrade, this would be a surprising behavioural change. So, voting no.

  3. misterdjules commented on Mar 31, 2017

    @misterdjules

    My vote is no.

  4. shigeki commented on Mar 31, 2017

    @shigeki

    ABSTAIN

  5. trevnorris commented on Mar 31, 2017

    @trevnorris

    sure.

  6. ChALkeR commented on Mar 31, 2017

    @ChALkeR
    Member

    @trevnorris, just to be sure — is that a YES?

  7. rvagg commented on Mar 31, 2017

    @rvagg
    Member

    No, partly because it's a surprising behaviour change and is technical breaking but also because 4 is in maintenance as of today and this is not "critical".

  8. Trott commented on Apr 1, 2017

    @Trott
    MemberAuthor

    I think we can call this as rejected, or at least not passing.

    6 abstentions means that it needs 8 YES votes to pass.

    Even if both the remaining outstanding votes (from @bnoordhuis and @indutny) come in as YES, it's a 7-YES/7-NO tie and the proposal does not pass.

    I will note that we can still choose to backport this at a later date, or someone can raise a proposal (right now or later) to backport only to some lines.

    But for the moment: We will not backport this behavior.

    (Feel free to re-open if you think my vote-counting is in error or if anyone is switching from ABSTAIN or NO to YES.)

  9. Trott commented on Apr 1, 2017

    @Trott
    MemberAuthor

    I do want to comment how nice it is that votes are not divided along company lines or anything like that.

    Three NodeSource employees can vote. Two of them voted yes and one of them voted no.

    Three Google employees can vote. One voted yes, one voted no, and one abstained.

    I wouldn't expect any less than people to vote their consciences and all that, but it's nice to see it manifest anyway.

  10. indutny commented on Apr 1, 2017

    @indutny
    Member

    NO

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions