On Wed, Jan 21, 2015 at 3:58 PM, Ross Gardler (MS OPEN TECH)
<ross.gard...@microsoft.com> wrote:
> We should not be focusing on who is/is not ticking a box on a report - it's
> a red herring and therefore a distraction.
>
> We should be focusing on identifying and assisting podlings that are not in
> receipt of adequate and appropriate mentoring.
Well, the idea is to identify such podlings with as little effort and fanfare
as possible.
* If a Mentor signs off on a report, trust that they are engaged.
* If they don't check the box (for any number of reasons), ping them
_privately_ to ask politely whether they're still engaged.
I figure that's both more reliable and less intrusive than shepherding.
But let's consider "retiring the shepherd role" and "automated private
followup after missing signoff" as separate initiatives...
The first person to do shepherding was Jukka. He couldn't handle the load by
himself, so he asked for volunteers[1].
The thing is, the IPMC doesn't seem to have the volunteer capacity to provide
senior-level shepherd reviews for all podlings each month, of the kind that a
Jukka or a Dave Fisher might write up. And it has always bothered me to have
outsiders judging podling communities in the context of a Board report, even
when they're experienced.
What the Board shepherds do is fundamentally different from what the Incubator
shepherds do: Board shepherds trust the report content and follow up after
problems, while Incubator shepherds are expected to dive into the mailing list
archives and assess community health proactively. To do that right is
labor-intensive and basically duplicates work already done by the podling's
Mentors. And what are the benefits?
* Because shepherd participation is below 50%, we can't count on them
to flag problem podlings consistently.
* Providing periodic outsider perspectives might be a good thing, but in the
context of a Board report, the stakes are too high.
* The shepherd gets their mind broadened. This is great, but there are
other ways to achieve that.
Fortunately, the Incubator has developed better mechanisms identify and track
problem podlings. Shepherds were essential while the Incubator was cleaning
house in 2012, but that's no longer the case.
Therefore, I support Alan's suggestion to simplify the Incubator by
streamlining away the shepherd role. The "Shepherd/Mentor notes" section of
the report can be changed to "Mentor/IPMC notes".
Marvin Humphrey
[1] http://markmail.org/message/y3pnb6degtnynmc2
http://s.apache.org/v7g
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
For additional commands, e-mail: general-h...@incubator.apache.org
