hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Lefty Leverenz <leftylever...@gmail.com>
Subject Re: Fwd: [DISCUSS] Allow any jira user to assign HIVE bugs to them self
Date Wed, 06 May 2015 03:08:07 GMT
Thank you for raising these points about the ICLA, Sushanth.  I'm looking
forward to the discussion in another thread.

-- Lefty

On Tue, May 5, 2015 at 9:05 PM, Sushanth Sowmyan <khorgath@gmail.com> wrote:

> I accept that this need not be linked to the jira-users issue itself,
> except that it made it much more difficult to enforce that
> contributors have ICLAs on file if we choose to go down that route. I
> will send out another mail asking for the project's position on ICLAs
> for all contributors, and what the committers should be responsible
> for in a separate mail.
>
> Also, yes, even if ICLA is on file, that does not imply that the
> contributor had rights to contribute, but it does put the legal
> responsibility on the contributor, rather than on the project or the
> committers.
>
> On Tue, May 5, 2015 at 5:46 PM, Thejas Nair <thejas.nair@gmail.com> wrote:
> > I would like to emphasise again that this change in adding jira-users,
> > does not change Hive's policy regarding ICLA.  In hive, we never
> > required people to file ICLA before submitting a patch.
> >
> > Your question regarding ICLA requirements merits a discussion on its
> > own. Even if ICLA is on file, that does not automatically imply that
> > the contributor had all rights to contribute. It just means that such
> > a contributor has lied, if he didn't have rights to contribute.
> >
> >
> >
> > On Tue, May 5, 2015 at 5:36 PM, Sushanth Sowmyan <khorgath@gmail.com>
> wrote:
> >> I will defer to the larger community's opinion on this, and from the
> >> looks of it, Apache does suggest, but not require (but does heavily
> >> suggest as desired) an ICLA from contributors, but I kinda agree with
> >>
> https://julien.ponge.org/blog/in-defense-of-contributor-license-agreements/
> >> in the place ICLAs have with projects.
> >>
> >> The relevant portion, as I see it, is this:
> >>
> >>>> Grant of Copyright License. Subject to the terms and conditions of
> this Agreement, You hereby grant to the Foundation and to recipients of
> software distributed by the Foundation a perpetual, worldwide,
> non-exclusive, no-charge, royalty-free, irrevocable copyright license to
> reproduce, prepare derivative works of, publicly display, publicly perform,
> sublicense, and distribute Your Contributions and such derivative works.
> >>> This is, I think, the first key point. Contributors explictly grant a
> license to the upstream project maintainers to use contributions.
> Sublicensing is important, too, as it opens licensing under new terms in
> the future, even if the contributor is out of reach.
> >>
> >> I feel like without having an ICLA requirement for contributors(and
> >> yes, I acknowledge that being a jira-user and requesting in the
> >> mailing list did not already cover this - it was my mistaken memory
> >> that felt like it did from back when the jira had a UI element
> >> granting ASF rights), committers open themselves to the possibility
> >> that we +1 and accept a contribution that we will wind up being
> >> responsible for that should not have been legally acceptable.
> >>
> >> I also agree with Lefty that taken to an extreme, this could apply for
> >> docs and wiki, etc, and that does sound ludicrous, but still a place
> >> we open ourselves to legal responsibility. If $COMPANY sues apache
> >> because we have some content in our wiki that we should not have,
> >> removal is not hard. If that happens with our git repo, we're in for a
> >> not-fun exercise in rewriting git history.
> >>
> >> I also concede the advantages in being more "open" by making it easier
> >> to contribute, and indeed the link I paste above does refer to people
> >> that will not contribute to a project that has a CLA requirement, but
> >> I'm not completely satisfied by not addressing this issue in some
> >> manner either.
> >>
> >> This is not a -1 for this move, and indeed, would/could not be a
> >> binding one even if it were so, but I would like to understand what
> >> the hive project's legal position is on the cases where a committer
> >> commits a patch that a contributor contributed that they did not have
> >> rights to contribute.
> >>
> >>
> >> On Tue, May 5, 2015 at 11:11 AM, Thejas Nair <thejas.nair@gmail.com>
> wrote:
> >>> I guess the limit is around the number of entries in the contributor
> >>> group, and adding a jira-user group would not count towards that.
> >>> Let me give it a try.
> >>>
> >>> That INFRA jira is another good reason to add jira-users group to
> contributors!
> >>>
> >>>
> >>> On Mon, May 4, 2015 at 11:31 PM, Carl Steinbach <cwsteinbach@gmail.com>
> wrote:
> >>>> It turns out there's a limit on the number of people you can list as
> >>>> "contributors" for any given JIRA project. I bumped into this a couple
> >>>> months back when I tried adding someone to the list and found this:
> >>>>
> >>>> https://issues.apache.org/jira/browse/INFRA-7293
> >>>>
> >>>>
> >>>>
> >>>> On Mon, May 4, 2015 at 10:02 PM, Lefty Leverenz <
> leftyleverenz@gmail.com>
> >>>> wrote:
> >>>>
> >>>>> Sure, go for it.
> >>>>>
> >>>>>
> >>>>> -- Lefty
> >>>>>
> >>>>> On Mon, May 4, 2015 at 5:33 PM, Thejas Nair <thejas.nair@gmail.com>
> wrote:
> >>>>>
> >>>>> > As Lefty noted, we don't require anyone being made a jira
> contributor
> >>>>> > or uploading a patch to have ICLA on file. Apache does not
require
> >>>>> > that, though that is encouraged.
> >>>>> > So allowing any user to be a contributor without asking for
> permission
> >>>>> > does not change things with respect to ICLA.
> >>>>> >
> >>>>> > Looks like people are on board with this. I will change the
> settings
> >>>>> > in another day as long as there are no objections.
> >>>>> >
> >>>>> >
> >>>>> >
> >>>>> >
> >>>>> > On Sat, May 2, 2015 at 11:44 PM, Lefty Leverenz <
> leftyleverenz@gmail.com
> >>>>> >
> >>>>> > wrote:
> >>>>> > > Hive only requires committers to sign ICLAs.  That doesn't
seem
> to
> >>>>> > provide
> >>>>> > > any legal protection when non-committers contribute patches.
> >>>>> > >
> >>>>> > > In days gone by, JIRA made us assign rights to Apache
when we
> attached
> >>>>> a
> >>>>> > > patch to an issue.  That's still in the instructions for
> Contributing
> >>>>> > Your
> >>>>> > > Work
> >>>>> > > <
> >>>>> >
> >>>>>
> https://cwiki.apache.org/confluence/display/Hive/HowToContribute#HowToContribute-ContributingYourWork
> >>>>> > >:
> >>>>> > >  "Please note that the attachment should be granted license
to
> ASF for
> >>>>> > > inclusion in ASF work" although the JIRA GUI doesn't have
that
> option
> >>>>> > > anymore.
> >>>>> > >
> >>>>> > > See Apache's page on licenses <
> http://www.apache.org/licenses/#clas>:
> >>>>> > "The
> >>>>> > > ASF desires that all contributors of ideas, code, or
> documentation to
> >>>>> the
> >>>>> > > Apache projects complete, sign, and submit (via postal
mail, fax
> or
> >>>>> > email)
> >>>>> > > an Individual Contributor License Agreement" *(highlighting
> added)*.
> >>>>> > >
> >>>>> > > So documentation in the wiki should also be covered by
ICLAs.
> Carried
> >>>>> to
> >>>>> > > extremes, anyone who participates on a mailing list, comments
on
> a JIRA
> >>>>> > > issue, or reviews a patch should sign an ICLA.
> >>>>> > >
> >>>>> > >
> >>>>> > > -- Lefty
> >>>>> > >
> >>>>> > > On Sun, May 3, 2015 at 12:30 AM, Sushanth Sowmyan <
> khorgath@gmail.com>
> >>>>> > > wrote:
> >>>>> > >
> >>>>> > >> I seem to remember something on the lines of that
the
> traditional
> >>>>> reason
> >>>>> > >> was so that a project could be sure that the contributor
had an
> ICLA
> >>>>> on
> >>>>> > >> file with apache so as to not expose the project to
legal risk
> of code
> >>>>> > that
> >>>>> > >> was contributed that the contributor did not have
any rights
> to. We
> >>>>> > should
> >>>>> > >> probably check with folks from other projects who've
had
> experience
> >>>>> > dealing
> >>>>> > >> with stuff like this?
> >>>>> > >>
> >>>>> > >> Maybe Owen?
> >>>>> > >> On May 2, 2015 17:08, "Thejas Nair" <thejas.nair@gmail.com>
> wrote:
> >>>>> > >>
> >>>>> > >> > Sending again, didn't make to the list for some
reason.
> >>>>> > >> >
> >>>>> > >> >
> >>>>> > >> > ---------- Forwarded message ----------
> >>>>> > >> > From: Thejas Nair <thejas.nair@gmail.com>
> >>>>> > >> > Date: Fri, May 1, 2015 at 1:53 PM
> >>>>> > >> > Subject: [DISCUSS] Allow any jira user to assign
HIVE bugs to
> them
> >>>>> > self
> >>>>> > >> > To: dev <dev@hive.apache.org>
> >>>>> > >> >
> >>>>> > >> >
> >>>>> > >> > I am not sure why a user needs to ask to be added
as a
> contributor
> >>>>> in
> >>>>> > >> > HIVE jira to be able to assign jiras to themselves.
I don't
> see it
> >>>>> > >> > adding any value. Also the jira ADMIN UI for
adding this is
> usually
> >>>>> > >> > flaky.
> >>>>> > >> >
> >>>>> > >> > I think we should let any jira users assign the
bugs to them
> self.
> >>>>> > >> > Looks like adding jira-users group to contributions
would do
> it.
> >>>>> > >> >
> >>>>> > >> > Thoughts ?
> >>>>> > >> >
> >>>>> > >> > Thanks,
> >>>>> > >> > Thejas
> >>>>> > >> >
> >>>>> > >>
> >>>>> >
> >>>>>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message