Hi,
I don't see the problem with people that contribute to a project without
writing source code, as I know we already have these kind of people in
ASF projects.
Just for my information, do we have some numbers about the min
contributors on a project and initial committers to be accepted in the
incubator?
May be it can help new project and people who wants to make a proposal
to have an idea if their proposition could be accepted or not.
regards,
François
On 31/07/2024 14:07, Jean-Baptiste Onofré wrote:
Hi Justin,
Thanks for your email. Much appreciated!
I agree with the requirements to join the incubator, I share the same vision.
The codebase is there, public, under the Apache license. So, I think
we are good here :)
About the community, it's a small one, but present, vibrant, and eager
to grow. We have a lot of enthusiasm around this proposal (and also
Polaris open source). I think it would be a great podling.
The split between PPMC / ocmmitters is unusual because of the
community build goal. The team had a lot of feedback and interest from
different parties, including some Apache projects.
When I prepared the proposal with the team (that included people
listed in the PPMC), we wanted to involve anyone helping the project
to grow. It's not necessary code contributions, it's everything to:
1. Help in the design (for instance Jack Ye is helping on the Apache
Iceberg REST impl vision)
2. Help to grow the community by including other communities (for
instance Anoop proposed to bridge with Apache XTable, John with Apache
Kafka, Robert with Project Nessie)
3. Mentor committers and other PPMC (that's my role mostly :)) in
addition of the regular mentor. For instance, Tyler could be a great
mentor thanks to his experience on Apache Beam.
So about the commits:
- Polaris started on a private repo at Snowflake including code that
it was not possible to push open source (Snowflake specific). So, we
had to change the history to remove these commits and have a clean
open source codebase
- You are right, not all PPMC members have committed on the repo, same
about the committers. The reason they are there is that either:
1. Their contributions are not code
2. Their contributions are not yet there, but they plan to contribute
I think it happens pretty often to have people listed in the initial
committers without commit (for instance, I think on Gravitino, Ashish
Singh is on the initial committer list without a commit afair).
I understand your comment, and it's important to me.
If it's a concern for you, I'm happy to rework the initial committer
list with the team (also removing PPMC list to use the default layout)
and grow the PPMC/committers when in the incubator.
Thoughts ?
Regards
JB
On Wed, Jul 31, 2024 at 5:47 AM Justin Mclean <jus...@classsoftware.com> wrote:
Hi,
In general, projects joining the incubator need two things:
- A codebase
- a community around it
While this isn't a hard requirement, building either while in incubation can be
difficult.
The first issue I see is it’s not clear to me that there is a community here.
The split between PPMC / committers is also unusual.
It also seems there’s a disconnect between the people who have contributed to
the project and those on the initial PMC/committer list. I understand history
has been lost. Is there any reason why history wasn’t preserved? But given it
comes from a single vendor, I would assume that the commits in the new repo
would then give an accurate representation of who is involved from outside that
vendor since the plan/announcement to make it open source. However looking at
the commits, 80%+ of the people involved come from that single vendor which and
most of the proposed PPMC/committers have no activity in the project. How have
these people contributed?
Looking closer, other than jbonofre or snazy, I can see these commits from
non-Snowflake people, and that is it.
anoopj - 1 commit (fixing a basic typo)
ajantha-bhat - 2 commits (minor doc changes)
From the information provided, it seems that a lot of the proposed PMCs have
not been involved with the project at all. Mentors are there to help a project,
adding extra existing PMC members to the PPMC to just help out is also rather
unusual. So both the PPMC/committer split and the composition seem problematic
to me.
Kind Regards,
Justin
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
For additional commands, e-mail: general-h...@incubator.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
For additional commands, e-mail: general-h...@incubator.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
For additional commands, e-mail: general-h...@incubator.apache.org
