Hi Rich,

I think you might have captured the very essence of the issue. I appreciate
how open-minded and inviting this discussion is, and I wish more
conversations started this way. It is refreshing to see a focus on
identifying the root cause rather than jumping straight to solutions and
tools before we fully understand the problem.

In my view, the main reason for prolonged graduations might indeed be the
classic, "If it's everyone's job, it's no one's job."". On one hand,
mentors want PPMCs to mature and recognize when to ask about graduation. On
the other hand, PPMCs that aren't aware of the need to be self-directing
won't ask because they expect mentors to tell them when they are ready.
This is a bit of a chicken-and-egg problem.

This comes down to an active versus passive dynamics. When both sides
expect the other to take the lead, things naturally stall. If even one side
is active, progress happens—whether toward graduation or retirement.

There probably isn't a single definitive answer regarding whether mentors
or PPMCs should drive the graduation readiness. Depending on the PPMC's
stage, current challenges, level of drive, and mentor availability, the
driving force will naturally shift - and likely many times during the
graduation process. That flexibility is fine, and we should simply adapt to
it.

Perhaps all we need is a prompt in the reporting process. Aside from the
self-assessment, the key question in the report could be: "Do you think the
next steps for graduation require action from the PPMC or the mentors, and
what specific actions are needed?"

Unspoken assumptions often lead to friction or stagnation. When
expectations are explicitly stated, both the PPMC and mentors can align (or
realize where they are misaligned) on who expects whom to do what. The next
actions and who is driving them. This simple procedural update could make a
huge difference. It would also reinforce that moving toward graduation or
retirement is the core purpose of every incubator report—which is what
distinguishes incubator reports from standard PMC reports.

I definitely believe we can achieve this by iterating on and improving our
current process, without needing to dismantle the incubator.

I am really glad we are having this discussion this way.

Best,
J.


On Sun, Nov 8, 2026 at 8:21 PM George <[email protected]> wrote:

> Just an "outsider"
>
> Measures is for me tick boxes, do this, then do that, then this and there
> you go. it's a conveyer belt, it's a Project manager taking you through a
> gant chart and being surprised when you get to the end and it does not
> work.
>
> A Mentors project for me is hold the hand of the new project, guide them,
> some grow at one rate, others at another, part of this is naturally, whats
> being done atm, whats needed, how do we get from A to B.
>
> if these are on a "page" thats there to help the Podling and the menter,
> for sure, but don't add tools, i.e. LLM's and then say well we've added
> this, this should fix it.
>
> G
>
> On Mon, Nov 9, 2026 at 3:49 AM Dave Fisher via general <
> [email protected]> wrote:
>
> > Hi -
> >
> > > On Nov 5, 2026, at 2:44 PM, Ed Espino via general <
> > [email protected]> wrote:
> > >
> > > Hi,
> > >
> > > You are not wrong. I have seen the same split. I would not close it by
> > > pushing a project out early.
> >
> > What’s early? There’s a VOTE to retire a project because it was not
> pushed
> > out at all.
> >
> > >
> > > A few things only get learned if the project has to do them while it is
> > > still here. Security is one. It may not come up at all. If it does,
> > someone
> > > has to answer the reporter and figure out what the report actually is.
> > That
> > > mail belongs on the private list. A page that points people at
> > > [email protected] is not enough.
> >
> > Having dealt with a security issue is not a graduation requirement. The
> > ASF Security team helps both podlings and PMCs handle security reports.
> > Podlings just need to know that it is their duty just like trademarks.
> > Having dealt with a trademark issue should not be a graduation
> requirement.
> >
> >
> > > I would leave the issues out of the podling report. The counts are the
> > > useful part: how many are open, how long until someone answered, how
> long
> > > until one was fixed. The project can see that without the bugs being
> > > described in a board report.
> > >
> > > The same report can carry the other easy things. Blank rows on the
> status
> > > page. How long since the last release. New committers, new PPMC
> members,
> > > and who from the original PPMC is still around. None of that graduates
> a
> > > project. It just makes what is left visible.
> >
> > The podling report is about progress in the incubator and should not be
> > practice towards a board report that may not need to be as you describe
> in
> > the last two paragraphs.
> >
> > >
> > > One more thing. The person who writes the report should change from one
> > > report to the next, the same way the release manager should change. You
> > > have to look at the whole story to write it. That is how more people
> find
> > > out what is left. A tool we run over here does not do that for them.
> >
> > A better practice is to review the answers on the dev mailing list. The
> > report is meant to be the codling’s view. I think doing that and always
> > thinking about graduation will be the best way to make graduation happen!
> >
> > It could be a good discussion to review what the requirements actually
> are
> > as a simple list.
> >
> > (1) IP transfer as needed. This is really the first thing and should not
> > wait for the first release attempt.
> > (2) Establish development at the ASF.
> > (3) Make at lease one valid ASF Release.
> > (4) Grow the community. In my experience even a singular new PPMC member
> > is enough.
> >
> > Is there really anything more? Yes, the Maturity Model has more, but big
> > picture those are the (4).
> >
> > The IP Transfer plan(1) should be practically complete by the time the
> > Champion asks the Incubator to accept the podling. If IP transfer is a
> > complex event then that can prevent the podling from ever graduating.
> >
> > Best,
> > Dave
> >
> > >
> > > -=e
> > > --
> > > Ed Espino
> > >
> > > On Thu, 5 Nov 2026 at 16:38 -0500, Rich Bowen <[email protected]>
> > wrote:
> > >
> > >> Is there a middle ground where the Incubator has mechanisms (scripts,
> AI
> > >> agents, GitHub actions, interns, who cares the actual mechanism) that
> > send
> > >> out those prods?
> > >>
> > >> --Rich
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [email protected]
> > For additional commands, e-mail: [email protected]
> >
> >
>
> --
> You have the obligation to inform one honestly of the risk, and as a person
> you are committed to educate yourself to the total risk in any activity!
>
> Once informed & totally aware of the risk,
> every fool has the right to kill or injure themselves as they see fit!
>

Reply via email to