[Pulp-dev] PUP Process: "obvious consensus"

Mon Jun 19 17:48:17 UTC 2017

After some in-person discussion, we will have a call to discuss ideas and
options regarding the pup1 process. We will use this etherpad [0] for
notes, and we will recap the information to the list also. In preparation,
please continue to share ideas, perspectives and concerns via this list.

When: June 22nd, 1pm UTC. See this in your local timezone here [1]. The
call will last no longer than 1 hour.

How to connect:
video chat:    https://bluejeans.com/697488960
phone only: + 800 451 8679   Enter Meeting ID: 697488960

[0]: http://pad-katello.rhcloud.com/p/Pulp_PUP_Process_Revisited
[1]: http://bit.ly/2rJqegX

-Brian

On Mon, Jun 19, 2017 at 9:23 AM, Michael Hrivnak <mhrivnak at redhat.com>
wrote:

> Back to where we started, having digested the discussion here and
> references cited, it seems clear that we have a system based on consensus,
> and that there is strong desire for decisions about process to continue
> being made with consensus. In terms of "obvious consensus", I'll propose
> that if any core member thinks it has not been reached, it has (perhaps by
> definition) not been reached.
>
> PUP0001 simply states in that case, "If obvious consensus is not reached,
> then the core devs decide." We don't need to over-complicate this. We've
> had reasonable success for many years at making process changes and
> agreeing on them. The PUP system should be a tool that helps us define a
> proposal as best we can, while providing a focal point for discussion. It
> should not unduly impede our ability to make decisions.
>
> So in a case where consensus is not obvious, can we talk it out among the
> core devs, particularly those with reservations, and make it our collective
> responsibility to find a path forward? Do we need to define it in more
> detail than that?
>
> On Fri, Jun 16, 2017 at 9:22 AM, David Davis <daviddavis at redhat.com>
> wrote:
>
>> I like centos model but personally I’m not a fan of the lazy consensus
>> option (X=0). Instead, I like the idea of having X be greater than 1
>> (preferably 2). I feel like if there’s at least two people driving a change
>> (i.e. X=2) then if one person leaves the project, we’ll still have someone
>> who is able and motivated to take on the maintenance and evolution of the
>> change. That said, I am happy to test out the model where X=0.
>>
>>
>> David
>>
>> On Thu, Jun 15, 2017 at 3:20 PM, Brian Bouterse <bbouters at redhat.com>
>> wrote:
>>
>>> I asked about some of these governance questions to a group of community
>>> managers from several open source projects that I meet with weekly. They
>>> said that if you don't have a BDFL (Pulp does not) the other very popular
>>> model is the lazy consensus model. I think lazy consensus is the spirit of
>>> pup1. I asked for some examples and they pointed me at the CentOS
>>> governance model [0][1].
>>>
>>> Also @daviddavis and I were talking and codifying the problem as what
>>> value should X be if X are the number of +1s required to pass a decision
>>> with zero -1 votes (vetos)? The CentOS governance model sets X = 0 by
>>> stating "There is no minimum +1 vote requirement". I'm also advocating for
>>> X=0 for the reasons I wrote in my earlier email. Practically speaking, I
>>> don't think an X=1, or X=2 will prevent many proposals that would have also
>>> passed with X=0.
>>>
>>> Regardless of the X value, we should continue the discussion so we can
>>> arrive at a decision on both pup1 and pup3. Thanks for continuing the convo.
>>>
>>> [0]: https://www.centos.org/about/governance/appendix-glossary/#c
>>> onsensus-decision-making
>>> [1]: https://www.centos.org/about/governance/voting/
>>>
>>> -Brian
>>>
>>>
>>> On Tue, Jun 13, 2017 at 11:46 AM, Ina Panova <ipanova at redhat.com> wrote:
>>>
>>>> And if we would remove all 'shades of grey' and go back just to +1 and
>>>> -1 where people would need to make their mind up *clearly* which would lead
>>>> stronger arguments of doing or not doing this.
>>>>
>>>>
>>>>
>>>> --------
>>>> Regards,
>>>>
>>>> Ina Panova
>>>> Software Engineer| Pulp| Red Hat Inc.
>>>>
>>>> "Do not go where the path may lead,
>>>>  go instead where there is no path and leave a trail."
>>>>
>>>> On Tue, Jun 13, 2017 at 5:30 PM, David Davis <daviddavis at redhat.com>
>>>> wrote:
>>>>
>>>>> In this model of where only -1 votes stop the PUP from passing,
>>>>> wouldn’t it mean that there needn't be any consensus at all? In other words
>>>>> we could effectively strike the language about consensus from PUP-1. This
>>>>> model makes me worried that people other than those casting -1 won’t bother
>>>>> to vote or participate since only -1 votes matter.
>>>>>
>>>>> I personally like the idea of having at least 30% that are +1 or +0.
>>>>> This means that enough -0 votes can still block the vote, and also +0 votes
>>>>> goes towards helping the PUP pass. Thus +0 and -0 would both matter. I
>>>>> think this is a good compromise between the extremes of "broad buy-in" and
>>>>> "default to change."
>>>>>
>>>>>
>>>>> David
>>>>>
>>>>> On Tue, Jun 13, 2017 at 10:36 AM, Brian Bouterse <bbouters at redhat.com>
>>>>> wrote:
>>>>>
>>>>>> We should (I thought we did) adopt a process that favors change and
>>>>>> does not have a "broad buy-in requirement". Any change that doesn't harm
>>>>>> the project should be allowed without broad buy-in. This empowers even a
>>>>>> single individual to enact change. This makes Pulp better because:
>>>>>>
>>>>>> * Everyone is empowered. A single individual can have a meaningful
>>>>>> impact.
>>>>>> * Anyone can stop an idea that will negatively affect the project or
>>>>>> community via veto.
>>>>>> * We avoid the tyranny of the majority [0] or supermajority.
>>>>>> * It avoids politics. If we start averaging, or counting votes
>>>>>> for/against in an offsetting way, there will be politics. Counting votes
>>>>>> for/against will create inequality because influential project members will
>>>>>> likely see their ideas adopted but others won't. Having a "default to
>>>>>> change and any core dev can veto" approach creates equality.
>>>>>>
>>>>>> Regarding how "obvious consensus" works with the "veto-or-it-passes"
>>>>>> model, if there are zero -1 votes cast, that means no one wanted to stop
>>>>>> the process. If no wants to stop it, and at least one is for it, then the
>>>>>> most sensible thing to do is to pass it. Since someone took time to write
>>>>>> the PUP there is obviously someone giving it a +1. If one person really
>>>>>> wants to go to place X for dinner (aka a +1), and there are no
>>>>>> counterproposals (aka a -1 with a suggestion) or strong preferences against
>>>>>> (aka -0 or +0) then the group will probably go to place X for dinner by way
>>>>>> of "obvious consensus".
>>>>>>
>>>>>> In summary, adopting a "default to accept or reject with even a
>>>>>> single veto" system creates an equal system. A system where, a single
>>>>>> individual can make a difference, and anyone can stop a bad idea from
>>>>>> occurring. To @mhrivnak's point about a change not meeting a broad range of
>>>>>> needs, I expect -1's to be cast in those cases, so this system is still
>>>>>> very safe in terms of protecting the projects needs and interests.
>>>>>>
>>>>>> [0]: https://en.wikipedia.org/wiki/Tyranny_of_the_majority
>>>>>>
>>>>>> -Brian
>>>>>>
>>>>>> On Mon, Jun 12, 2017 at 7:53 PM, David Davis <daviddavis at redhat.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Not sure this is true. I actually abstained from voting on PUP-3
>>>>>>> because I was somewhere between a +0 and a -0.
>>>>>>>
>>>>>>>
>>>>>>> David
>>>>>>>
>>>>>>> On Mon, Jun 12, 2017 at 11:43 AM, Ina Panova <ipanova at redhat.com>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Having at least one  +1 is not impartial approach just because the
>>>>>>>> developer who , as you said, found the time for the research and writing
>>>>>>>> down the proposal obviously will vote as +1 :)
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> --------
>>>>>>>> Regards,
>>>>>>>>
>>>>>>>> Ina Panova
>>>>>>>> Software Engineer| Pulp| Red Hat Inc.
>>>>>>>>
>>>>>>>> "Do not go where the path may lead,
>>>>>>>>  go instead where there is no path and leave a trail."
>>>>>>>>
>>>>>>>> On Mon, Jun 12, 2017 at 5:35 PM, Austin Macdonald <
>>>>>>>> amacdona at redhat.com> wrote:
>>>>>>>>
>>>>>>>>> This reminds me of the concept of a "Do-ocracy".
>>>>>>>>>
>>>>>>>>> If developers take the time to research and write up a proposal,
>>>>>>>>> they have "done". It seems completely reasonable to default to the opinion
>>>>>>>>> of the people that cared enough to do the work. If it isn't the right
>>>>>>>>> decision, then someone must actively block it, simple as that.
>>>>>>>>>
>>>>>>>>> I think the rule should be "PUP passes if we have at least one +1
>>>>>>>>> and no -1s".
>>>>>>>>>
>>>>>>>>> _______________________________________________
>>>>>>>>> Pulp-dev mailing list
>>>>>>>>> Pulp-dev at redhat.com
>>>>>>>>> https://www.redhat.com/mailman/listinfo/pulp-dev
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> Pulp-dev mailing list
>>>>>>>> Pulp-dev at redhat.com
>>>>>>>> https://www.redhat.com/mailman/listinfo/pulp-dev
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Pulp-dev mailing list
>>>>>>> Pulp-dev at redhat.com
>>>>>>> https://www.redhat.com/mailman/listinfo/pulp-dev
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>> _______________________________________________
>> Pulp-dev mailing list
>> Pulp-dev at redhat.com
>> https://www.redhat.com/mailman/listinfo/pulp-dev
>>
>>
>
>
> --
>
> Michael Hrivnak
>
> Principal Software Engineer, RHCE
>
> Red Hat
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/pulp-dev/attachments/20170619/64fa67e3/attachment.htm>