You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@community.apache.org by Christofer Dutz <ch...@c-ware.de> on 2019/03/20 12:43:02 UTC
Apache Maturity Model Consensus Building contradicts Incubator rules?
Hi all,
I’m currently working on finishing some things in preparation of graduation … one thing we were requested to address, is to do an assessment of the Apache Maturity Model for our project.
Within this, there’s a rule:
CS40 - In Apache projects, vetoes are only valid for code commits and are justified by a technical explanation, as per the Apache voting rules defined in CS30.
This sort of contradicts the rules for incubating projects specified by the Incubator, which says:
https://wiki.apache.org/incubator/DefaultProjectGuidelines
- Consensus Approval<https://www.apache.org/foundation/glossary.html#ConsensusApproval> – Consensus approval requires 3 binding<https://www.apache.org/foundation/voting.html#binding-votes> +1 votes and no -1 votes (vetoes<https://www.apache.org/foundation/glossary.html#Veto>).
So for Consensus Approval there are Vetoes and code changes apply for Lazy Consensus. If there are actually vetoes for code changes, I can imagine that quite some projects would stall instantly.
The incubator guidelines state for adding (or removing) people to(or from) committer and PPMC status, these guidelines claim them being Consensus Approval, which allows vetoes.
Would be cool if this could be streamlined to be more aligned. Right now I claim that the PLC4X project simply fails CS40 as this contradicts the incubator rules.
Chris
Re: Apache Maturity Model Consensus Building contradicts Incubator rules?
Posted by Bertrand Delacretaz <bd...@apache.org>.
Hi Chris,
On Wed, Mar 20, 2019 at 1:43 PM Christofer Dutz
<ch...@c-ware.de> wrote:
> ...CS40 - In Apache projects, vetoes are only valid for code commits and are justified
> by a technical explanation, as per the Apache voting rules defined in CS30....
I wrote that bit as that was my understanding until very recently.
We had fairly long discussions (on members@ IIRC) last year about this
and my conclusions are:
a) vetoes do apply to code as per
https://www.apache.org/foundation/voting.html ("Votes on code
modifications") but that page clearly explains how they can happen,
with strong justifications.
b) vetoes are not recommended for other decisions, due to the
potential for blocking things (especially in small projects) but PMCs
are free to decide for themselves.
Joan linked to http://couchdb.apache.org/bylaws.html which is a great
way to clarify this *for a specific PMC* even though nowadays we
prefer such things to be named "community guidelines" instead of
bylaws.
CouchDB does accept vetoes only for technical decisions as per the
default ASF model, but as per "3.8. Decision Types" they require
stronger 2/3 majorities for some decisions.
That's the kind of things that projects can decide for themselves,
https://httpd.apache.org/dev/guidelines.html has another example. I
have added those links to
https://wiki.apache.org/incubator/DefaultProjectGuidelines
-Bertrand
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Apache Maturity Model Consensus Building contradicts Incubator rules?
Posted by Dmitriy Pavlov <dp...@apache.org>.
Hi Chirs,
I agree some disagreements exist. Practically I suppose adding (or
removing) people to(or from) committer and (P)PMC status is a very
sensitive thing, so there should be an opportunity of veto.
While process changes and releases can't be vetoed, adding committers or
PMCs should be consensus only.
If it is the Majority approval it is always easy for some commercial
company to take full control of any project. In this case, some individual
PMC has almost no influence on PMC members/Committers. All people from a
company will generate as much +1's as needed to elect new PMCs/Committers,
even if this one true volunteer will vote -1.
Sincerely,
Dmitriy Pavlov
ср, 20 мар. 2019 г. в 15:43, Christofer Dutz <ch...@c-ware.de>:
> Hi all,
>
> I’m currently working on finishing some things in preparation of
> graduation … one thing we were requested to address, is to do an assessment
> of the Apache Maturity Model for our project.
>
> Within this, there’s a rule:
> CS40 - In Apache projects, vetoes are only valid for code commits and are
> justified by a technical explanation, as per the Apache voting rules
> defined in CS30.
>
> This sort of contradicts the rules for incubating projects specified by
> the Incubator, which says:
> https://wiki.apache.org/incubator/DefaultProjectGuidelines
>
> - Consensus Approval<
> https://www.apache.org/foundation/glossary.html#ConsensusApproval> –
> Consensus approval requires 3 binding<
> https://www.apache.org/foundation/voting.html#binding-votes> +1 votes and
> no -1 votes (vetoes<https://www.apache.org/foundation/glossary.html#Veto
> >).
>
> So for Consensus Approval there are Vetoes and code changes apply for Lazy
> Consensus. If there are actually vetoes for code changes, I can imagine
> that quite some projects would stall instantly.
>
>
> The incubator guidelines state for adding (or removing) people to(or from)
> committer and PPMC status, these guidelines claim them being Consensus
> Approval, which allows vetoes.
>
> Would be cool if this could be streamlined to be more aligned. Right now I
> claim that the PLC4X project simply fails CS40 as this contradicts the
> incubator rules.
>
> Chris
>
Re: Apache Maturity Model Consensus Building contradicts Incubator rules?
Posted by Isabel Drost-Fromm <is...@apache.org>.
When consolidating, you might also check that things are consistent with
https://www.apache.org/foundation/voting.html
Isabel
Am 20. März 2019 13:43:02 MEZ schrieb Christofer Dutz <ch...@c-ware.de>:
>Hi all,
>
>I’m currently working on finishing some things in preparation of
>graduation … one thing we were requested to address, is to do an
>assessment of the Apache Maturity Model for our project.
>
>Within this, there’s a rule:
>CS40 - In Apache projects, vetoes are only valid for code commits and
>are justified by a technical explanation, as per the Apache voting
>rules defined in CS30.
>
>This sort of contradicts the rules for incubating projects specified by
>the Incubator, which says:
>https://wiki.apache.org/incubator/DefaultProjectGuidelines
>
>- Consensus
>Approval<https://www.apache.org/foundation/glossary.html#ConsensusApproval>
>– Consensus approval requires 3
>binding<https://www.apache.org/foundation/voting.html#binding-votes> +1
>votes and no -1 votes
>(vetoes<https://www.apache.org/foundation/glossary.html#Veto>).
>
>So for Consensus Approval there are Vetoes and code changes apply for
>Lazy Consensus. If there are actually vetoes for code changes, I can
>imagine that quite some projects would stall instantly.
>
>
>The incubator guidelines state for adding (or removing) people to(or
>from) committer and PPMC status, these guidelines claim them being
>Consensus Approval, which allows vetoes.
>
>Would be cool if this could be streamlined to be more aligned. Right
>now I claim that the PLC4X project simply fails CS40 as this
>contradicts the incubator rules.
>
>Chris
--
Diese Nachricht wurde von meinem Android-Gerät mit K-9 Mail gesendet.