incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dave Fisher <>
Subject Re: Too many late reports
Date Tue, 08 Oct 2013 04:10:43 GMT

On Oct 6, 2013, at 9:19 PM, Marvin Humphrey wrote:

> On Sun, Oct 6, 2013 at 11:29 AM, Dave Fisher <> wrote:
>>>   Dave Fisher       DeviceMap
>>>   Dave Fisher       Spark
>> Although there are no reports yet.
> On Sun, Oct 6, 2013 at 11:45 AM, Matt Franklin <> wrote:
>> Missing the ODF Toolkit report for this month.
> Lots of late or missing reports this month!
> As of the due date last Wednesday, 8 podlings out of 16 had not reported:
> *   Chukwa
> *   DeviceMap
> *   Helix
> *   ODF Toolkit
> *   Ripple
> *   Samza
> *   Spark
> *   Stratos
> Since then, Chukwa and Helix filed their reports late Saturday; Spark filed a
> report today -- *after* Dave Fisher had filed his shepherding commentary,
> which concludes with the sentence "I'm not sure why they have not reported
> yet."

I've read the report. Good info there about JIRA. I updated my shepherd's note, but left Marvin's/

I still wonder about DeviceMap. I suspect a lot of people were hit with conference travel
of various types.


> It's inconvenient to receive podling reports late.  Mentors have less time to
> sign off and possibly comment; shepherds don't get to take the podling's own
> self-examination into account when conducting their reviews.
> Perhaps it would be better if we ask late podlings to file next month instead.
> That's what the Board asks TLPs to do when their reports show up too late.
> In the past, the Incubator PMC has not done a good job of following up when
> reports are missed entirely, arguably contributing to our losing track of
> low-activity podlings and delaying intervention.  I've started doing it a bit,
> though I've missed at least one (NPanday did not file in August but there was
> no followup AFAIK).  It might be nice to build a remedy into the reporting run
> book -- something like this:
>    After filing the report, perform the following steps for each podling
>    which did not report:
>    *   Edit podlings.xml to set the "monthly" reporting attribute for the
>        podling.
>    *   Send a message to the podling's dev list requesting that they
>        file an out-of-cycle report next month as a makeup.
> One possible contributing factor is that podlings are still not getting a
> full week's notice to file their reports.  Sure, reporting isn't a huge task,
> but when you're new to creating reports it takes much longer.
> The argument that podlings ought to just know these dates and file on time
> rather than relying on reminders[1] simply isn't getting the job done.  I
> think we'd get better results if we expect punctuality from podlings, set a
> good example by demonstrating punctuality ourselves, and express lenience by
> requesting that a podling report next month rather than by accepting a
> report too
> late to give it the attention it deserves.
> The problem is that at present the reminders have to be sent out manually[2],
> and despite the best of intentions, humans are not ideal replacements for cron
> jobs.  It's time to ask again: what would it take to get those reminders
> firing with unerring reliability three weeks before the Board meeting?  Is it
> realistic to attempt a patch eliminating the need for an svn password?
> Marvin Humphrey
> [1]
> [2]
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message