incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Branko ─îibej <br...@apache.org>
Subject Re: [DISCUSSION] Graduate Ignite from the Apache Incubator
Date Thu, 23 Jul 2015 15:38:32 GMT
On 23.07.2015 15:31, Pierre Smits wrote:
> When we're talking about this podling having  gained 'committers many from
> outside the company that donated the code', I wonder who we are talking
> about.
>
> The reports to the IPMC show only numbers (3 in March 2015) and no names.
> And for what it is worth, the June 2015 report shows a link (
> https://ignite.incubator.apache.org/community.html#list) that is pointing
> nowhere. I assume that it should have been this page:
> http://ignite.incubator.apache.org/community/resources.html#people.
>
> So we have (per today):
>
>    - http://people.apache.org/committers-by-project.html#ignite , aka ASF
>    list,
>    - http://incubator.apache.org/projects/ignite , aka Incubator list,
>    - http://ignite.incubator.apache.org/community/resources.html#people
>    (showing affiliation), aka podling list
>
> And there are discrepancies between the pages. E.g.
>
>    - http://incubator.apache.org/projects/ignite shows Ryan Rawson as a
>    committer, but is not in
>    http://people.apache.org/committers-by-project.html#ignite


So there are discrepancies; so what? They'll be fixed eventually, but I
hope you're not implying that a typo somewhere is a barrier to graduation.


> If we substract the mentors (4 according to
> http://incubator.apache.org/projects/ignite, though we can argue whether
> Henry Saputra should also be in there as he is listed in the March 2015
> report to the IPMC as one of the report signers) from any list, we see the
> following changes:
>
>    1. Rayn Rawson (external, Apache Drill committer),
>       1. Incubator list


Ryan is on the list because he's mentioned as an initial commiter in the
original Ignite proposal
http://wiki.apache.org/incubator/IgniteProposal. He's certainly not in
the authz template.

This seems to be the only discrepancy; we have to get to the bottom of
that. But I'll point out that maintaining the Incubator list is not the
responsibility of the podling.

>    2.  Sergey Khisamov (external - Fitech Source),
>    1. ASF list
>       2. Incubator list,
>       3. podling list
>    3. Ilya Sterin (external - ChronoTrack),
>       1. ASF list,
>       2. Incubator list,
>       3. podling list
>    4. Evans Ye (external - TrendMicro),
>       1. ASF list,
>       2. Incubator list,
>       3. podling list
>    5. Ognen Duzlevski (external - Shoutlet),
>       1. ASF list,
>       2. Incubator list,
>       3. podling list
>    6. Gianfrano Murador (external - Engiweb Security)
>       1. ASF list
>       2. Incubator list
>       3. podling list
>
> Adding 5 or 6 new committers isn't many. That is a start (regarding
> diversity). It for sure doesn't scream independence, when the majority (of
> committers, intended PMC members) is affiliated to one company.

We're back to bean counting. How many committers, do you think, is
"enough" and why? I find it more than enough. By the way, if you're
going to be nitpicking, do it right and mention that Evans Ye is a
Bigtop committer.

> As for building the community of the podling, a mentor has the
> responsibility to keep tabs on contributions to ensure that everything goes
> according to the policies of the ASF, of the incubator and of the podling
> and assess (together with the community) everything whether it is in line
> with those policies. And report.
>
> As examples:
>
>    - Statements (on podling pages re StackOverflow, or on external fora,
>    e.g. Nabble) that questions can be raised via those media, isn't in line
>    with how contributions to an ASF project (or podling) should be done. ASF
>    mailing lists are the primary source for non-JIRA (including code patches)
>    / non-wiki contributions. External sources are a nicety, but unreliable
>    when it comes to feeding back into the ASF mailing lists or identifying
>    active contributors or assessing potential additions to list of the
>    privileged few;

Nonsense. It's perfectly fine to follow SO etc. for user questions and
to tell users that such external forums are being followed by community
member. A question from a user is not a contribution.

>    - Community pages should reflect how the on-boarding process is,
>    including pointing out/stressing that an iCLA is required (when
>    contributing anything above the level of question or comment);

As I said above, a question is not a contribution. And it is not true
that an ICLA is required for "anything about the level of question or
comment." I suggest you go and read http://www.apache.org/licenses/
again before making such statements.

Specifically, it's perfectly OK to accept patches from people who've not
signed an ICLA as long as these patches are small and/or trivial.

>    - Committed non-privileged contributors (at least those who have
>    registered an iCLA) should be listed in the pages of the podling, in order
>    to assess who (beyond committers) are enabled to make changes to e.g. wiki
>    pages, and who are eligible for commit privileges.

I suggest you do not not confuse your opinion with actual graduation
requirements. Nothing in our policy says how or if a podling should
track potential committers. It's certainly not a podling's (or a TLP's)
task to track ICLA contributions; that's the domain of secretary@a.o.


> And diversity is not only about the aspect of affiliation.

Exactly!

>  It is also about
> having (privileged) contributors of another kind other than just those who
> contribute code changes (or register with JIRA).

You mean about NOT having such contributors

>  The pages of the podling should reflect such.

That's your opinion again. Evidently you're leaning towards a
bureaucratic keep-lists-of-everything approach. Whilst such an approach
is valid, it's by no means required.

>  And mentors should point that out to the community of
> the podling. Otherwise the podling might imply (through its pages) that the
> other kind of contributors aren't welcome, and/or that principles of the
> ASF (e.g. 'All contributions are equal', 'Contributions buys privileges')
> aren't applicable in the podling.

Well, what's your point? If you're saying that the Ignite community is
doing this wrong, you really should point out specific cases. If you're
just talking about generalities, I'd appreciate your doing that on a
different thread.

> Apart from the above, the podling could and should do a bit more regarding
> building an open, transparent project. Having done a cursory review of the
> mailing list archives of the podling I have found no announcements of
> organisational changes (e.g. adding the new committer/ppmc changes/mentor
> changes).

Please note that discussions and votes about new committers happen on
the private list, which you've obviously not taken a look at. I see no
reason or requirement to announce new committers on the dev@ list,
especially as all committers are (or should be) also PPMC members and
follow the private list, too.

-- Brane

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Mime
View raw message