You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@airflow.apache.org by Kaxil Naik <ka...@gmail.com> on 2020/03/16 10:50:25 UTC
Let's agree to guidelines for AIP (Airflow Improvement Proposal)
Hey all,
I feel like we have good guidelines on creating an AIP, however, there we
don't have "clear" guidelines on the following (We might already do, in
which case please correct me):
1. How long should the *Vote *on AIP go on?
2. Minimum number of votes required to marked the AIP as "accepted"
3. What happens when the minimum number of votes is not reached within
the deadline we decide for (1)? Should we consider it an implicit "YES" or
just wait! Or is it an implicit "not interested in this AIP"?
4. Can someone veto an AIP?
We can adopt the *Release Approval*
http://www.apache.org/legal/release-policy.html#release-approval guidelines
for AIP too.
My Proposed Answer (similar to Apache Release Process):
- A *[DISCUSS]* thread is created to discuss the approach and idea. If
there is a general interest in the idea and unless there are security
concerns or a veto from a PMC member, this will go to a VOTE.
- A *[VOTE]* thread is created that would last for at least 3 days
*and *until
3 *+1* *binding votes* are obtained.
- Binding Votes: PMC and Committers
- Non-binding Votes: Members of the community
I am happy to document the process once we finalize it.
Regards,
Kaxil
Re: Let's agree to guidelines for AIP (Airflow Improvement
Proposal)
Posted by Daniel Imberman <da...@gmail.com>.
Everything else looks good to me, except for the idea of “lazy consensus.” I think if you can’t get 3 binding +1’s that might mean there isn’t enough interest around your idea.
Daniel
On Mar 17, 2020, 6:42 AM -0700, Kaxil Naik <ka...@gmail.com>, wrote:
> What do you think about the other guidelines/questions?
>
> On Tue, Mar 17, 2020 at 1:37 PM Daniel Imberman <da...@gmail.com>
> wrote:
>
> > I agree with Jarek. All -1’s should be considered bur need to be qualified.
> >
> > Daniel
> > On Mar 16, 2020, 4:35 AM -0700, Kaxil Naik <ka...@gmail.com>, wrote:
> > > Yes, I like the procedural issues one (that includes lazy consensus) too.
> > >
> > >
> > >
> > > On Mon, Mar 16, 2020, 11:31 Jarek Potiuk <Ja...@polidea.com>
> > wrote:
> > >
> > > > (and BTW when we vote on this procedure we should follow voting
> > process on
> > > > procedural issues (same link - above)
> > > >
> > > > "Votes on procedural issues follow the common format of majority rule
> > > > unless otherwise stated. That is, if there are more favourable votes
> > than
> > > > unfavourable ones, the issue is considered to have passed --
> > regardless of
> > > > the number of votes in each category. (If the number of votes seems too
> > > > small to be representative of a community consensus, the issue is
> > typically
> > > > not pursued. However, see the description of lazy consensus for a
> > modifying
> > > > factor.)"
> > > >
> > > > J.
> > > >
> > > >
> > > > On Mon, Mar 16, 2020 at 12:29 PM Jarek Potiuk <
> > Jarek.Potiuk@polidea.com>
> > > > wrote:
> > > >
> > > > > Agree with the proposal in general.
> > > > >
> > > > > However I think this is about code modification, so we should rather
> > > > > follow Votes on code modifications rather than releases:
> > > > >
> > > > >
> > https://www.apache.org/foundation/voting.html#votes-on-code-modification
> > > > >
> > > > > That means:
> > > > >
> > > > > - we sum all votes and positive means "passed"
> > > > > - qualified -1 is a veto but it needs strong explanation and good
> > > > > reason otherwise veto is invalid
> > > > > - there are fractional votes - -0.5 and -.0.9 as well as +0.9 with
> > > > > implications described above.
> > > > > - minimum 3 '+1' votes are required- without it we should continue to
> > > > > discuss and vote (unless we declare lazy-consensus).
> > > > >
> > > > > J.
> > > > >
> > > > >
> > > > > On Mon, Mar 16, 2020 at 11:50 AM Kaxil Naik <ka...@gmail.com>
> > wrote:
> > > > >
> > > > > > Hey all,
> > > > > >
> > > > > > I feel like we have good guidelines on creating an AIP, however,
> > there
> > > > we
> > > > > > don't have "clear" guidelines on the following (We might already
> > do, in
> > > > > > which case please correct me):
> > > > > >
> > > > > >
> > > > > > 1. How long should the *Vote *on AIP go on?
> > > > > > 2. Minimum number of votes required to marked the AIP as "accepted"
> > > > > > 3. What happens when the minimum number of votes is not reached
> > > > within
> > > > > > the deadline we decide for (1)? Should we consider it an implicit
> > > > > > "YES" or
> > > > > > just wait! Or is it an implicit "not interested in this AIP"?
> > > > > > 4. Can someone veto an AIP?
> > > > > >
> > > > > >
> > > > > > We can adopt the *Release Approval*
> > > > > > http://www.apache.org/legal/release-policy.html#release-approval
> > > > > > guidelines
> > > > > > for AIP too.
> > > > > >
> > > > > > My Proposed Answer (similar to Apache Release Process):
> > > > > >
> > > > > > - A *[DISCUSS]* thread is created to discuss the approach and idea.
> > > > If
> > > > > > there is a general interest in the idea and unless there are
> > security
> > > > > > concerns or a veto from a PMC member, this will go to a VOTE.
> > > > > > - A *[VOTE]* thread is created that would last for at least 3 days
> > > > > > *and *until
> > > > > > 3 *+1* *binding votes* are obtained.
> > > > > > - Binding Votes: PMC and Committers
> > > > > > - Non-binding Votes: Members of the community
> > > > > >
> > > > > >
> > > > > > I am happy to document the process once we finalize it.
> > > > > >
> > > > > > Regards,
> > > > > > Kaxil
> > > > > >
> > > > >
> > > > >
> > > > > --
> > > > >
> > > > > Jarek Potiuk
> > > > > Polidea <https://www.polidea.com/> | Principal Software Engineer
> > > > >
> > > > > M: +48 660 796 129 <+48660796129>
> > > > > [image: Polidea] <https://www.polidea.com/>
> > > > >
> > > > >
> > > >
> > > > --
> > > >
> > > > Jarek Potiuk
> > > > Polidea <https://www.polidea.com/> | Principal Software Engineer
> > > >
> > > > M: +48 660 796 129 <+48660796129>
> > > > [image: Polidea] <https://www.polidea.com/>
> > > >
> >
Re: Let's agree to guidelines for AIP (Airflow Improvement Proposal)
Posted by Kaxil Naik <ka...@gmail.com>.
What do you think about the other guidelines/questions?
On Tue, Mar 17, 2020 at 1:37 PM Daniel Imberman <da...@gmail.com>
wrote:
> I agree with Jarek. All -1’s should be considered bur need to be qualified.
>
> Daniel
> On Mar 16, 2020, 4:35 AM -0700, Kaxil Naik <ka...@gmail.com>, wrote:
> > Yes, I like the procedural issues one (that includes lazy consensus) too.
> >
> >
> >
> > On Mon, Mar 16, 2020, 11:31 Jarek Potiuk <Ja...@polidea.com>
> wrote:
> >
> > > (and BTW when we vote on this procedure we should follow voting
> process on
> > > procedural issues (same link - above)
> > >
> > > "Votes on procedural issues follow the common format of majority rule
> > > unless otherwise stated. That is, if there are more favourable votes
> than
> > > unfavourable ones, the issue is considered to have passed --
> regardless of
> > > the number of votes in each category. (If the number of votes seems too
> > > small to be representative of a community consensus, the issue is
> typically
> > > not pursued. However, see the description of lazy consensus for a
> modifying
> > > factor.)"
> > >
> > > J.
> > >
> > >
> > > On Mon, Mar 16, 2020 at 12:29 PM Jarek Potiuk <
> Jarek.Potiuk@polidea.com>
> > > wrote:
> > >
> > > > Agree with the proposal in general.
> > > >
> > > > However I think this is about code modification, so we should rather
> > > > follow Votes on code modifications rather than releases:
> > > >
> > > >
> https://www.apache.org/foundation/voting.html#votes-on-code-modification
> > > >
> > > > That means:
> > > >
> > > > - we sum all votes and positive means "passed"
> > > > - qualified -1 is a veto but it needs strong explanation and good
> > > > reason otherwise veto is invalid
> > > > - there are fractional votes - -0.5 and -.0.9 as well as +0.9 with
> > > > implications described above.
> > > > - minimum 3 '+1' votes are required- without it we should continue to
> > > > discuss and vote (unless we declare lazy-consensus).
> > > >
> > > > J.
> > > >
> > > >
> > > > On Mon, Mar 16, 2020 at 11:50 AM Kaxil Naik <ka...@gmail.com>
> wrote:
> > > >
> > > > > Hey all,
> > > > >
> > > > > I feel like we have good guidelines on creating an AIP, however,
> there
> > > we
> > > > > don't have "clear" guidelines on the following (We might already
> do, in
> > > > > which case please correct me):
> > > > >
> > > > >
> > > > > 1. How long should the *Vote *on AIP go on?
> > > > > 2. Minimum number of votes required to marked the AIP as "accepted"
> > > > > 3. What happens when the minimum number of votes is not reached
> > > within
> > > > > the deadline we decide for (1)? Should we consider it an implicit
> > > > > "YES" or
> > > > > just wait! Or is it an implicit "not interested in this AIP"?
> > > > > 4. Can someone veto an AIP?
> > > > >
> > > > >
> > > > > We can adopt the *Release Approval*
> > > > > http://www.apache.org/legal/release-policy.html#release-approval
> > > > > guidelines
> > > > > for AIP too.
> > > > >
> > > > > My Proposed Answer (similar to Apache Release Process):
> > > > >
> > > > > - A *[DISCUSS]* thread is created to discuss the approach and idea.
> > > If
> > > > > there is a general interest in the idea and unless there are
> security
> > > > > concerns or a veto from a PMC member, this will go to a VOTE.
> > > > > - A *[VOTE]* thread is created that would last for at least 3 days
> > > > > *and *until
> > > > > 3 *+1* *binding votes* are obtained.
> > > > > - Binding Votes: PMC and Committers
> > > > > - Non-binding Votes: Members of the community
> > > > >
> > > > >
> > > > > I am happy to document the process once we finalize it.
> > > > >
> > > > > Regards,
> > > > > Kaxil
> > > > >
> > > >
> > > >
> > > > --
> > > >
> > > > Jarek Potiuk
> > > > Polidea <https://www.polidea.com/> | Principal Software Engineer
> > > >
> > > > M: +48 660 796 129 <+48660796129>
> > > > [image: Polidea] <https://www.polidea.com/>
> > > >
> > > >
> > >
> > > --
> > >
> > > Jarek Potiuk
> > > Polidea <https://www.polidea.com/> | Principal Software Engineer
> > >
> > > M: +48 660 796 129 <+48660796129>
> > > [image: Polidea] <https://www.polidea.com/>
> > >
>
Re: Let's agree to guidelines for AIP (Airflow Improvement
Proposal)
Posted by Daniel Imberman <da...@gmail.com>.
I agree with Jarek. All -1’s should be considered bur need to be qualified.
Daniel
On Mar 16, 2020, 4:35 AM -0700, Kaxil Naik <ka...@gmail.com>, wrote:
> Yes, I like the procedural issues one (that includes lazy consensus) too.
>
>
>
> On Mon, Mar 16, 2020, 11:31 Jarek Potiuk <Ja...@polidea.com> wrote:
>
> > (and BTW when we vote on this procedure we should follow voting process on
> > procedural issues (same link - above)
> >
> > "Votes on procedural issues follow the common format of majority rule
> > unless otherwise stated. That is, if there are more favourable votes than
> > unfavourable ones, the issue is considered to have passed -- regardless of
> > the number of votes in each category. (If the number of votes seems too
> > small to be representative of a community consensus, the issue is typically
> > not pursued. However, see the description of lazy consensus for a modifying
> > factor.)"
> >
> > J.
> >
> >
> > On Mon, Mar 16, 2020 at 12:29 PM Jarek Potiuk <Ja...@polidea.com>
> > wrote:
> >
> > > Agree with the proposal in general.
> > >
> > > However I think this is about code modification, so we should rather
> > > follow Votes on code modifications rather than releases:
> > >
> > > https://www.apache.org/foundation/voting.html#votes-on-code-modification
> > >
> > > That means:
> > >
> > > - we sum all votes and positive means "passed"
> > > - qualified -1 is a veto but it needs strong explanation and good
> > > reason otherwise veto is invalid
> > > - there are fractional votes - -0.5 and -.0.9 as well as +0.9 with
> > > implications described above.
> > > - minimum 3 '+1' votes are required- without it we should continue to
> > > discuss and vote (unless we declare lazy-consensus).
> > >
> > > J.
> > >
> > >
> > > On Mon, Mar 16, 2020 at 11:50 AM Kaxil Naik <ka...@gmail.com> wrote:
> > >
> > > > Hey all,
> > > >
> > > > I feel like we have good guidelines on creating an AIP, however, there
> > we
> > > > don't have "clear" guidelines on the following (We might already do, in
> > > > which case please correct me):
> > > >
> > > >
> > > > 1. How long should the *Vote *on AIP go on?
> > > > 2. Minimum number of votes required to marked the AIP as "accepted"
> > > > 3. What happens when the minimum number of votes is not reached
> > within
> > > > the deadline we decide for (1)? Should we consider it an implicit
> > > > "YES" or
> > > > just wait! Or is it an implicit "not interested in this AIP"?
> > > > 4. Can someone veto an AIP?
> > > >
> > > >
> > > > We can adopt the *Release Approval*
> > > > http://www.apache.org/legal/release-policy.html#release-approval
> > > > guidelines
> > > > for AIP too.
> > > >
> > > > My Proposed Answer (similar to Apache Release Process):
> > > >
> > > > - A *[DISCUSS]* thread is created to discuss the approach and idea.
> > If
> > > > there is a general interest in the idea and unless there are security
> > > > concerns or a veto from a PMC member, this will go to a VOTE.
> > > > - A *[VOTE]* thread is created that would last for at least 3 days
> > > > *and *until
> > > > 3 *+1* *binding votes* are obtained.
> > > > - Binding Votes: PMC and Committers
> > > > - Non-binding Votes: Members of the community
> > > >
> > > >
> > > > I am happy to document the process once we finalize it.
> > > >
> > > > Regards,
> > > > Kaxil
> > > >
> > >
> > >
> > > --
> > >
> > > Jarek Potiuk
> > > Polidea <https://www.polidea.com/> | Principal Software Engineer
> > >
> > > M: +48 660 796 129 <+48660796129>
> > > [image: Polidea] <https://www.polidea.com/>
> > >
> > >
> >
> > --
> >
> > Jarek Potiuk
> > Polidea <https://www.polidea.com/> | Principal Software Engineer
> >
> > M: +48 660 796 129 <+48660796129>
> > [image: Polidea] <https://www.polidea.com/>
> >
Re: Let's agree to guidelines for AIP (Airflow Improvement Proposal)
Posted by Kaxil Naik <ka...@gmail.com>.
Yes, I like the procedural issues one (that includes lazy consensus) too.
On Mon, Mar 16, 2020, 11:31 Jarek Potiuk <Ja...@polidea.com> wrote:
> (and BTW when we vote on this procedure we should follow voting process on
> procedural issues (same link - above)
>
> "Votes on procedural issues follow the common format of majority rule
> unless otherwise stated. That is, if there are more favourable votes than
> unfavourable ones, the issue is considered to have passed -- regardless of
> the number of votes in each category. (If the number of votes seems too
> small to be representative of a community consensus, the issue is typically
> not pursued. However, see the description of lazy consensus for a modifying
> factor.)"
>
> J.
>
>
> On Mon, Mar 16, 2020 at 12:29 PM Jarek Potiuk <Ja...@polidea.com>
> wrote:
>
> > Agree with the proposal in general.
> >
> > However I think this is about code modification, so we should rather
> > follow Votes on code modifications rather than releases:
> >
> > https://www.apache.org/foundation/voting.html#votes-on-code-modification
> >
> > That means:
> >
> > - we sum all votes and positive means "passed"
> > - qualified -1 is a veto but it needs strong explanation and good
> > reason otherwise veto is invalid
> > - there are fractional votes - -0.5 and -.0.9 as well as +0.9 with
> > implications described above.
> > - minimum 3 '+1' votes are required- without it we should continue to
> > discuss and vote (unless we declare lazy-consensus).
> >
> > J.
> >
> >
> > On Mon, Mar 16, 2020 at 11:50 AM Kaxil Naik <ka...@gmail.com> wrote:
> >
> >> Hey all,
> >>
> >> I feel like we have good guidelines on creating an AIP, however, there
> we
> >> don't have "clear" guidelines on the following (We might already do, in
> >> which case please correct me):
> >>
> >>
> >> 1. How long should the *Vote *on AIP go on?
> >> 2. Minimum number of votes required to marked the AIP as "accepted"
> >> 3. What happens when the minimum number of votes is not reached
> within
> >> the deadline we decide for (1)? Should we consider it an implicit
> >> "YES" or
> >> just wait! Or is it an implicit "not interested in this AIP"?
> >> 4. Can someone veto an AIP?
> >>
> >>
> >> We can adopt the *Release Approval*
> >> http://www.apache.org/legal/release-policy.html#release-approval
> >> guidelines
> >> for AIP too.
> >>
> >> My Proposed Answer (similar to Apache Release Process):
> >>
> >> - A *[DISCUSS]* thread is created to discuss the approach and idea.
> If
> >> there is a general interest in the idea and unless there are security
> >> concerns or a veto from a PMC member, this will go to a VOTE.
> >> - A *[VOTE]* thread is created that would last for at least 3 days
> >> *and *until
> >> 3 *+1* *binding votes* are obtained.
> >> - Binding Votes: PMC and Committers
> >> - Non-binding Votes: Members of the community
> >>
> >>
> >> I am happy to document the process once we finalize it.
> >>
> >> Regards,
> >> Kaxil
> >>
> >
> >
> > --
> >
> > Jarek Potiuk
> > Polidea <https://www.polidea.com/> | Principal Software Engineer
> >
> > M: +48 660 796 129 <+48660796129>
> > [image: Polidea] <https://www.polidea.com/>
> >
> >
>
> --
>
> Jarek Potiuk
> Polidea <https://www.polidea.com/> | Principal Software Engineer
>
> M: +48 660 796 129 <+48660796129>
> [image: Polidea] <https://www.polidea.com/>
>
Re: Let's agree to guidelines for AIP (Airflow Improvement Proposal)
Posted by Jarek Potiuk <Ja...@polidea.com>.
(and BTW when we vote on this procedure we should follow voting process on
procedural issues (same link - above)
"Votes on procedural issues follow the common format of majority rule
unless otherwise stated. That is, if there are more favourable votes than
unfavourable ones, the issue is considered to have passed -- regardless of
the number of votes in each category. (If the number of votes seems too
small to be representative of a community consensus, the issue is typically
not pursued. However, see the description of lazy consensus for a modifying
factor.)"
J.
On Mon, Mar 16, 2020 at 12:29 PM Jarek Potiuk <Ja...@polidea.com>
wrote:
> Agree with the proposal in general.
>
> However I think this is about code modification, so we should rather
> follow Votes on code modifications rather than releases:
>
> https://www.apache.org/foundation/voting.html#votes-on-code-modification
>
> That means:
>
> - we sum all votes and positive means "passed"
> - qualified -1 is a veto but it needs strong explanation and good
> reason otherwise veto is invalid
> - there are fractional votes - -0.5 and -.0.9 as well as +0.9 with
> implications described above.
> - minimum 3 '+1' votes are required- without it we should continue to
> discuss and vote (unless we declare lazy-consensus).
>
> J.
>
>
> On Mon, Mar 16, 2020 at 11:50 AM Kaxil Naik <ka...@gmail.com> wrote:
>
>> Hey all,
>>
>> I feel like we have good guidelines on creating an AIP, however, there we
>> don't have "clear" guidelines on the following (We might already do, in
>> which case please correct me):
>>
>>
>> 1. How long should the *Vote *on AIP go on?
>> 2. Minimum number of votes required to marked the AIP as "accepted"
>> 3. What happens when the minimum number of votes is not reached within
>> the deadline we decide for (1)? Should we consider it an implicit
>> "YES" or
>> just wait! Or is it an implicit "not interested in this AIP"?
>> 4. Can someone veto an AIP?
>>
>>
>> We can adopt the *Release Approval*
>> http://www.apache.org/legal/release-policy.html#release-approval
>> guidelines
>> for AIP too.
>>
>> My Proposed Answer (similar to Apache Release Process):
>>
>> - A *[DISCUSS]* thread is created to discuss the approach and idea. If
>> there is a general interest in the idea and unless there are security
>> concerns or a veto from a PMC member, this will go to a VOTE.
>> - A *[VOTE]* thread is created that would last for at least 3 days
>> *and *until
>> 3 *+1* *binding votes* are obtained.
>> - Binding Votes: PMC and Committers
>> - Non-binding Votes: Members of the community
>>
>>
>> I am happy to document the process once we finalize it.
>>
>> Regards,
>> Kaxil
>>
>
>
> --
>
> Jarek Potiuk
> Polidea <https://www.polidea.com/> | Principal Software Engineer
>
> M: +48 660 796 129 <+48660796129>
> [image: Polidea] <https://www.polidea.com/>
>
>
--
Jarek Potiuk
Polidea <https://www.polidea.com/> | Principal Software Engineer
M: +48 660 796 129 <+48660796129>
[image: Polidea] <https://www.polidea.com/>
Re: Let's agree to guidelines for AIP (Airflow Improvement Proposal)
Posted by Jarek Potiuk <Ja...@polidea.com>.
Agree with the proposal in general.
However I think this is about code modification, so we should rather follow
Votes on code modifications rather than releases:
https://www.apache.org/foundation/voting.html#votes-on-code-modification
That means:
- we sum all votes and positive means "passed"
- qualified -1 is a veto but it needs strong explanation and good reason
otherwise veto is invalid
- there are fractional votes - -0.5 and -.0.9 as well as +0.9 with
implications described above.
- minimum 3 '+1' votes are required- without it we should continue to
discuss and vote (unless we declare lazy-consensus).
J.
On Mon, Mar 16, 2020 at 11:50 AM Kaxil Naik <ka...@gmail.com> wrote:
> Hey all,
>
> I feel like we have good guidelines on creating an AIP, however, there we
> don't have "clear" guidelines on the following (We might already do, in
> which case please correct me):
>
>
> 1. How long should the *Vote *on AIP go on?
> 2. Minimum number of votes required to marked the AIP as "accepted"
> 3. What happens when the minimum number of votes is not reached within
> the deadline we decide for (1)? Should we consider it an implicit "YES"
> or
> just wait! Or is it an implicit "not interested in this AIP"?
> 4. Can someone veto an AIP?
>
>
> We can adopt the *Release Approval*
> http://www.apache.org/legal/release-policy.html#release-approval
> guidelines
> for AIP too.
>
> My Proposed Answer (similar to Apache Release Process):
>
> - A *[DISCUSS]* thread is created to discuss the approach and idea. If
> there is a general interest in the idea and unless there are security
> concerns or a veto from a PMC member, this will go to a VOTE.
> - A *[VOTE]* thread is created that would last for at least 3 days
> *and *until
> 3 *+1* *binding votes* are obtained.
> - Binding Votes: PMC and Committers
> - Non-binding Votes: Members of the community
>
>
> I am happy to document the process once we finalize it.
>
> Regards,
> Kaxil
>
--
Jarek Potiuk
Polidea <https://www.polidea.com/> | Principal Software Engineer
M: +48 660 796 129 <+48660796129>
[image: Polidea] <https://www.polidea.com/>