Re: Mentor disengagement - a suggestion

[Rich Bowen]

On Nov 4, 2015 12:54 AM, "Justin Erenkrantz"  wrote:
>
> I will just note that I disagree with adding bureaucracy like this.

It's not bureaucracy. It's suggesting a tool by which a mentor might
measure a podling's progress. It answers the question "how can I tell if
they're ready." It also gives a better-than-gut-feeling way to communicate
to others that they're ready.

When a project is recommended to the board for graduation, most of the time
all we have to go on is a list of names and a vague description. It'd be
nice to have more. But this isn't a requirement or policy. It's just a
suggestion.

> We already require podlings to submit reports as frequently as
> monthly.  (Geode somehow had to report monthly for no discernible
> reason.)
>
> This further adds to the burden on being a mentor - probably to
> something like being a teacher enforcing a pedagogical structure on
> the podlings.  I don't think that is what we should be striving for.
> For the two podlings I currently mentor (Geode and HAWQ), I keep an
> eye on the mailing lists and try to ensure that any process questions
> that are raised are addressed.  In the early days of Geode, there was
> a bit of that - less so now as the community is finding its way.  HAWQ
> seems to be doing well - nothing surprising that I can tell so far.
>
> I believe in a "big tent" foundation - we should welcome new projects
> of any stripe.  If they fail within the Incubator to graduate, so be
> it.  But, that is separate from asking the mentors to somehow be
> "responsible" for the podlings.  I have an interest to see them
> succeed, but if it doesn't, *shrug* and we move on.  I don't see any
> value in adding more bureaucracy as it will further sap any motivation
> to truly "mentor" projects.  I view the mentor as someone that can be
> called upon, but doesn't necessarily require active involvement on a
> daily basis.  If I wanted that, I would be a committer to the project.
> Let's not confuse the two.
>
> My $.02.  -- justin
>
> On Mon, Oct 12, 2015 at 7:18 AM, Rich Bowen  wrote:
> > Fellow mentors,
> >
> > There was a conversation at ApacheCon about the Incubator. I'll leave
it to
> > the other participants to champion the particular parts that they are
> > passionate about, but I was particularly concerned with mentor
> > disengagement, and suggestions for improving it.
> >
> > A mentor's role is to help a project learn the ropes at the ASF, and
that
> > mentor might not necessarily be deeply versed in the particular
technology
> > that the podling works with. As such, it can frequently be the case
that the
> > mentor becomes disengaged from the daily conversation of the lists, and
> > eventually with the entire process.
> >
> > As a means of refocusing the mentors' efforts, and keeping them
engaged, I'd
> > like to encourage each mentor (or group of mentors) to consider writing
a
> > running report (ie, evolving, updated every quarter) based on
> >
https://community.apache.org/apache-way/apache-project-maturity-model.html
> > where they evaluate each point on the maturity model, as a path towards
> > graduation. This gives a concrete target, and a lens through which to
view
> > the podling's progress towards that target.
> >
> > This could be kept in the incubator wiki, and linked from the official
> > project report, or it could be maintained just for your own benefit. I
think
> > it would be particularly useful to attach to a graduation
recommendation, as
> > a sign that the recommendation is more than just checking the various
boxes,
> > but is a glowing endorsement of the project's readiness to be TLP.
> >
> > As a side-note, I'd also encourage mentors who are mentors in name
only, and
> > not reality, to consider cleaning up the paperwork by removing
themselves
> > from the roster. It doesn't look great when a podling can't get mentor
> > signoff on their reports.
> >
> > --
> > Rich Bowen - rbo...@rcbowen.com - @rbowen
> > http://apachecon.com/ - @apachecon
> >
> > -
> > 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
>

Re: Mentor disengagement - a suggestion

[John D. Ament]

Hi Justin,

Just to clarify.

On Wed, Nov 4, 2015 at 12:54 AM Justin Erenkrantz 
wrote:

> I will just note that I disagree with adding bureaucracy like this.
> We already require podlings to submit reports as frequently as
> monthly.  (Geode somehow had to report monthly for no discernible
> reason.)
>

It was requested that Geode provide an extra report as they did not report
for their first 3 months.  New podlings, new TLPs are all required to
report monthly to the (board/IPMC) as a way to ensure they're getting
started properly.  Roman agreed with the request and they reported 1 extra
month.


>
> This further adds to the burden on being a mentor - probably to
> something like being a teacher enforcing a pedagogical structure on
> the podlings.  I don't think that is what we should be striving for.
> For the two podlings I currently mentor (Geode and HAWQ), I keep an
> eye on the mailing lists and try to ensure that any process questions
> that are raised are addressed.  In the early days of Geode, there was
> a bit of that - less so now as the community is finding its way.  HAWQ
> seems to be doing well - nothing surprising that I can tell so far.
>
> I believe in a "big tent" foundation - we should welcome new projects
> of any stripe.  If they fail within the Incubator to graduate, so be
> it.  But, that is separate from asking the mentors to somehow be
> "responsible" for the podlings.  I have an interest to see them
> succeed, but if it doesn't, *shrug* and we move on.  I don't see any
> value in adding more bureaucracy as it will further sap any motivation
> to truly "mentor" projects.  I view the mentor as someone that can be
> called upon, but doesn't necessarily require active involvement on a
> daily basis.  If I wanted that, I would be a committer to the project.
> Let's not confuse the two.
>
> My $.02.  -- justin
>
> On Mon, Oct 12, 2015 at 7:18 AM, Rich Bowen  wrote:
> > Fellow mentors,
> >
> > There was a conversation at ApacheCon about the Incubator. I'll leave it
> to
> > the other participants to champion the particular parts that they are
> > passionate about, but I was particularly concerned with mentor
> > disengagement, and suggestions for improving it.
> >
> > A mentor's role is to help a project learn the ropes at the ASF, and that
> > mentor might not necessarily be deeply versed in the particular
> technology
> > that the podling works with. As such, it can frequently be the case that
> the
> > mentor becomes disengaged from the daily conversation of the lists, and
> > eventually with the entire process.
> >
> > As a means of refocusing the mentors' efforts, and keeping them engaged,
> I'd
> > like to encourage each mentor (or group of mentors) to consider writing a
> > running report (ie, evolving, updated every quarter) based on
> >
> https://community.apache.org/apache-way/apache-project-maturity-model.html
> > where they evaluate each point on the maturity model, as a path towards
> > graduation. This gives a concrete target, and a lens through which to
> view
> > the podling's progress towards that target.
> >
> > This could be kept in the incubator wiki, and linked from the official
> > project report, or it could be maintained just for your own benefit. I
> think
> > it would be particularly useful to attach to a graduation
> recommendation, as
> > a sign that the recommendation is more than just checking the various
> boxes,
> > but is a glowing endorsement of the project's readiness to be TLP.
> >
> > As a side-note, I'd also encourage mentors who are mentors in name only,
> and
> > not reality, to consider cleaning up the paperwork by removing themselves
> > from the roster. It doesn't look great when a podling can't get mentor
> > signoff on their reports.
> >
> > --
> > Rich Bowen - rbo...@rcbowen.com - @rbowen
> > http://apachecon.com/ - @apachecon
> >
> > -
> > 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
>
>

Re: Mentor disengagement - a suggestion

[Roman Shaposhnik]

On Sun, Nov 1, 2015 at 11:24 AM, Rich Bowen  wrote:
> I look to the ipmc chair to say what we MUST or MUST NOT do. I've learned
> my lesson.

Makes two of us. Still waiting on actionable guidance.

Thanks,
Roman.

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

Re: Mentor disengagement - a suggestion

[Justin Erenkrantz]

I will just note that I disagree with adding bureaucracy like this.
We already require podlings to submit reports as frequently as
monthly.  (Geode somehow had to report monthly for no discernible
reason.)

This further adds to the burden on being a mentor - probably to
something like being a teacher enforcing a pedagogical structure on
the podlings.  I don't think that is what we should be striving for.
For the two podlings I currently mentor (Geode and HAWQ), I keep an
eye on the mailing lists and try to ensure that any process questions
that are raised are addressed.  In the early days of Geode, there was
a bit of that - less so now as the community is finding its way.  HAWQ
seems to be doing well - nothing surprising that I can tell so far.

I believe in a "big tent" foundation - we should welcome new projects
of any stripe.  If they fail within the Incubator to graduate, so be
it.  But, that is separate from asking the mentors to somehow be
"responsible" for the podlings.  I have an interest to see them
succeed, but if it doesn't, *shrug* and we move on.  I don't see any
value in adding more bureaucracy as it will further sap any motivation
to truly "mentor" projects.  I view the mentor as someone that can be
called upon, but doesn't necessarily require active involvement on a
daily basis.  If I wanted that, I would be a committer to the project.
Let's not confuse the two.

My $.02.  -- justin

On Mon, Oct 12, 2015 at 7:18 AM, Rich Bowen  wrote:
> Fellow mentors,
>
> There was a conversation at ApacheCon about the Incubator. I'll leave it to
> the other participants to champion the particular parts that they are
> passionate about, but I was particularly concerned with mentor
> disengagement, and suggestions for improving it.
>
> A mentor's role is to help a project learn the ropes at the ASF, and that
> mentor might not necessarily be deeply versed in the particular technology
> that the podling works with. As such, it can frequently be the case that the
> mentor becomes disengaged from the daily conversation of the lists, and
> eventually with the entire process.
>
> As a means of refocusing the mentors' efforts, and keeping them engaged, I'd
> like to encourage each mentor (or group of mentors) to consider writing a
> running report (ie, evolving, updated every quarter) based on
> https://community.apache.org/apache-way/apache-project-maturity-model.html
> where they evaluate each point on the maturity model, as a path towards
> graduation. This gives a concrete target, and a lens through which to view
> the podling's progress towards that target.
>
> This could be kept in the incubator wiki, and linked from the official
> project report, or it could be maintained just for your own benefit. I think
> it would be particularly useful to attach to a graduation recommendation, as
> a sign that the recommendation is more than just checking the various boxes,
> but is a glowing endorsement of the project's readiness to be TLP.
>
> As a side-note, I'd also encourage mentors who are mentors in name only, and
> not reality, to consider cleaning up the paperwork by removing themselves
> from the roster. It doesn't look great when a podling can't get mentor
> signoff on their reports.
>
> --
> Rich Bowen - rbo...@rcbowen.com - @rbowen
> http://apachecon.com/ - @apachecon
>
> -
> 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

Re: Mentor disengagement - a suggestion

[Rich Bowen]

>
> But, I've got to wonder: as long as it is just a suggestion what's to
compel
> a mentor to actually spend quite a bit of time on doing that?
>
>

A mentor's desire to do their job thoroughly and serve the project and the
Foundation. This is one tool to assist with reaching that end. Use it, or
don't. It's just a suggestion.

I look to the ipmc chair to say what we MUST or MUST NOT do. I've learned
my lesson.

Re: Mentor disengagement - a suggestion

[Ted Dunning]

On Mon, Oct 19, 2015 at 6:35 PM, Roman Shaposhnik 
wrote:

> > As a means of refocusing the mentors' efforts, and keeping them engaged,
> I'd
> > like to encourage each mentor (or group of mentors) to consider writing a
> > running report (ie, evolving, updated every quarter) based on
> >
> https://community.apache.org/apache-way/apache-project-maturity-model.html
> > where they evaluate each point on the maturity model, as a path towards
> > graduation. This gives a concrete target, and a lens through which to
> view
> > the podling's progress towards that target.
>
> +1 this will definitely help with both responsibilities.
>
> But, I've got to wonder: as long as it is just a suggestion what's to
> compel
> a mentor to actually spend quite a bit of time on doing that?
>

Nothing.

Is compulsion the only logic here?

Re: Mentor disengagement - a suggestion

[Sam Ruby]

On Thu, Oct 15, 2015 at 3:24 AM, Bertrand Delacretaz
 wrote:
> On Thu, Oct 15, 2015 at 12:24 AM, Martijn Dashorst
>  wrote:
>> ...we should
>> strive to make incubation take not longer than a year or so. Or when
>> incubation takes longer, have a plan to switch mentors, perhaps a new
>> mentor might bring a new graduation spirit into the project...
>
> I like the idea of revisiting the mentors roster from time to time.
>
> We might just add an item to the podling's reports, asking if they
> (mentors + community) think rotating some of their mentors would be
> useful, or if they're happy with the current roster. And optionally
> make that a private comment that does not to in the public minutes.

Agreed.

I don't believe that speeding up incubation would be desirable for
other reasons (being done in a year is wonderful, but if not, then
what?  Killing the podling would seem unwise).  Socializing the idea
of an annual rotation of mentors would be a good thing; as long as
plenty of room is left for people who are both active and wish to
remain.  Sign-offs are one way to determine potential activity, and
reports like the one I provided previously but organized in a
different manner might be helpful as a catalyst for the right
discussions to occur.  I'll note that podlings already have a
reporting schedule and a start date.  This could be used to spread out
the notion of an annual review of mentors.  Applying this consistently
to all projects (as opposed to singling out individual projects) would
introduce less social friction.

> -Bertrand

- Sam Ruby

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

Re: Mentor disengagement - a suggestion

[Bertrand Delacretaz]

On Thu, Oct 15, 2015 at 12:24 AM, Martijn Dashorst
 wrote:
> ...we should
> strive to make incubation take not longer than a year or so. Or when
> incubation takes longer, have a plan to switch mentors, perhaps a new
> mentor might bring a new graduation spirit into the project...

I like the idea of revisiting the mentors roster from time to time.

We might just add an item to the podling's reports, asking if they
(mentors + community) think rotating some of their mentors would be
useful, or if they're happy with the current roster. And optionally
make that a private comment that does not to in the public minutes.

-Bertrand

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

Re: Mentor disengagement - a suggestion

[Bertrand Delacretaz]

On Wed, Oct 14, 2015 at 3:15 PM, Marko Rodriguez  wrote:
> ...Once we got one of our mentors engaged -- Daniel Gruno -- its been smooth 
> sailing. You
> know how we got him engaged? HipChat. He set us up an account and now we
> can "@Humbedooh" with questions and get responses

That works for mentors who spend a lot of their time on such channels.

Other mentors work in a more asynchronous way, so this would not help
in all cases.

-Bertrand

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

Re: Mentor disengagement - a suggestion

[Sam Ruby]

On Wed, Oct 14, 2015 at 10:07 AM, Ted Dunning  wrote:
> On Wed, Oct 14, 2015 at 6:46 AM, Jim Jagielski  wrote:
>
>> My point is that with 1 mentor, everyone knows where the buck stops.
>> With >1 nobody knows. A flat hierarchy for mentors does not seem
>> workable or, at least, optimal.
>>
>
> That's a fine point.
>
> But it is counter to my experience.
>
> For instance, on Kylin, I was extremely active early on. Lately, Julian
> Hyde has taken much more of the questions.  It has worked very well.  We
> have a third mentor (whose name I won't mention) who has been nearly
> completely missing, but that hasn't been a problem since everybody knows
> that they are totally MiA.
>
>
>> If we wish to address this, and not "force" mentors to leave,
>> we could simply add the idea of "lead mentor" and have the PPMC
>> vote on which mentor they want to be the lead mentor (pseudo
>> PPMC chair); the remaining mentors would remain as co-mentors
>> but the intent is that the lead mentor would be the primary person
>> responsible.
>>
>
> I think that this is largely process without clear need. But others may
> have different experience.

The following is from the definition of "champion" on the incubator
roles and responsibility page[1]:

---
During incubation, the Champion:

 * Coordinates the creation and timely delivery of the podling's board reports.
 * Keeps an eye on the mentors' activity and takes action (ask for new
mentors, talk to the Incubator PMC) if they don't seem to provide
enough oversight or mentorship to the podling,
---

First question: does this description match current practice.  Second
question: is there something in that description that needs to get
done, but isn't getting done?

There also is a separate thread about the number of podlings that the
Incubator can handle.  That number would tend to increase if some of
the work were decentralized more.

My take is that (at least for this month) Marvin is effectively doing
the first bullet for all podlings.  Now I don't see any signs of
burnout in Marvin, so that's not the immediate concern, but
scalability and sustainability would benefit from decentralizing this
more.

As to the second bullet, it isn't clear to me that that is being done.
Even if it were only a one time thing, reverifying (or in many cases,
naming) the champion for every podling, and then asking each to
reverify the active status for every mentor for podlings that they
champion might go a long way toward easing some of the concerns that
people have raised.

Ultimately the definition of champion, as posted, should match reality.

- Sam Ruby

[1] 
http://incubator.apache.org/incubation/Roles_and_Responsibilities.html#Champion

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

RE: Mentor disengagement - a suggestion

[Ross Gardler]

Further to Sam's suggestion and observations below see 

Suggestion 0.1.8 at 
http://wiki.apache.org/incubator/IncubatorIssues2013#Suggestions

-Original Message-
From: sa3r...@gmail.com [mailto:sa3r...@gmail.com] On Behalf Of Sam Ruby
Sent: Wednesday, October 14, 2015 7:26 AM
To: general@incubator.apache.org
Subject: Re: Mentor disengagement - a suggestion

On Wed, Oct 14, 2015 at 10:07 AM, Ted Dunning <ted.dunn...@gmail.com> wrote:
> On Wed, Oct 14, 2015 at 6:46 AM, Jim Jagielski <j...@jagunet.com> wrote:
>
>> My point is that with 1 mentor, everyone knows where the buck stops.
>> With >1 nobody knows. A flat hierarchy for mentors does not seem 
>> workable or, at least, optimal.
>>
>
> That's a fine point.
>
> But it is counter to my experience.
>
> For instance, on Kylin, I was extremely active early on. Lately, 
> Julian Hyde has taken much more of the questions.  It has worked very 
> well.  We have a third mentor (whose name I won't mention) who has 
> been nearly completely missing, but that hasn't been a problem since 
> everybody knows that they are totally MiA.
>
>
>> If we wish to address this, and not "force" mentors to leave, we 
>> could simply add the idea of "lead mentor" and have the PPMC vote on 
>> which mentor they want to be the lead mentor (pseudo PPMC chair); the 
>> remaining mentors would remain as co-mentors but the intent is that 
>> the lead mentor would be the primary person responsible.
>>
>
> I think that this is largely process without clear need. But others 
> may have different experience.

The following is from the definition of "champion" on the incubator roles and 
responsibility page[1]:

---
During incubation, the Champion:

 * Coordinates the creation and timely delivery of the podling's board reports.
 * Keeps an eye on the mentors' activity and takes action (ask for new mentors, 
talk to the Incubator PMC) if they don't seem to provide enough oversight or 
mentorship to the podling,
---

First question: does this description match current practice.  Second
question: is there something in that description that needs to get done, but 
isn't getting done?

There also is a separate thread about the number of podlings that the Incubator 
can handle.  That number would tend to increase if some of the work were 
decentralized more.

My take is that (at least for this month) Marvin is effectively doing the first 
bullet for all podlings.  Now I don't see any signs of burnout in Marvin, so 
that's not the immediate concern, but scalability and sustainability would 
benefit from decentralizing this more.

As to the second bullet, it isn't clear to me that that is being done.
Even if it were only a one time thing, reverifying (or in many cases,
naming) the champion for every podling, and then asking each to reverify the 
active status for every mentor for podlings that they champion might go a long 
way toward easing some of the concerns that people have raised.

Ultimately the definition of champion, as posted, should match reality.

- Sam Ruby

[1] 
https://na01.safelinks.protection.outlook.com/?url=http%3a%2f%2fincubator.apache.org%2fincubation%2fRoles_and_Responsibilities.html%23Champion=01%7c01%7cRoss.Gardler%40microsoft.com%7c02e20d0e08f54f9c81f408d2d4a356cc%7c72f988bf86f141af91ab2d7cd011db47%7c1=q10OQ8l3N4%2bAPk%2bXJYxZVF1mEREPIOYNpQcD9Frr0rI%3d

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

Re: Mentor disengagement - a suggestion

[Konstantin Boudnik]

On Wed, Oct 14, 2015 at 06:10AM, Ted Dunning wrote:
> If it can work, that is very good. With intermittent availability, I have
> often seen the need for a spare.

Exactly! I've been out for 6 weeks back in May/June and missed all the reports
and other activity on the projects I am/was a mentor to. But I didn't fret as
there were a couple of other guys to make sure the podlings aren't hanging
high and dry

Cos

> On Wed, Oct 14, 2015 at 5:53 AM, Jim Jagielski  wrote:
> 
> > Agreed. My only comment would be that I still think that the
> > optimal number of mentors is 1.
> >
> > > On Oct 14, 2015, at 12:45 AM, Julian Hyde  wrote:
> > >
> > > It's not activity on the dev list, or even report signoffs, that
> > > matter most. Podlings, especially new podlings, have lots and lots of
> > > questions, especially about infrastructure. Without at least two
> > > responsive mentors to field those questions you feel like banging your
> > > head on the wall. And you start wondering why you left the comfort and
> > > convenience of github and whether Apache itself is fascinated by its
> > > own brand.
> > >
> > > Before you ask, you won't get podlings to send their questions to
> > > another list, because we're all too proud to ask questions which in
> > > retrospect always turn out to be dumb questions.
> > >
> > > It's not possible to measure that kind of mentor activity, so I think
> > > people are inclined to measure the "public" forms of activity as proxy
> > > indicators.
> > >
> > > Julian
> > >
> > >
> > > On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski  wrote:
> > >> For me, I consider being a mentor as I do being a member of a PMC.
> > >> Occasionally one simply lacks cycles to be actively involved, but
> > >> one is involve enough to see that others *ARE* involved, and so I
> > >> am "unconcerned" about my inactivity during those times.
> > >>
> > >> My understanding is that this is OK and its one of the reasons
> > >> why we *have* multiple mentors.
> > >>
> > >> "Shaming" inactive mentors would be akin to "shaming" PMC members who
> > >> didn't post on the dev@ list this month, or who didn't vote on a
> > release
> > >> or etc...
> > >>
> > >> I am not, of course, referring to mentors who are truly MIA month in and
> > >> month out. But, as someone said, if you remove those from the equation,
> > >> the list of "active" mentors is pretty constant.
> > >>
> > >> So the question is "Is there a difference or problem between a podling
> > >> with 10 mentors, of which 4 are 'active', as compared to a podling with
> > >> 4 mentors, all of which are 'active'"??
> > >>
> > >>> On Oct 13, 2015, at 2:29 AM, Ted Dunning 
> > wrote:
> > >>>
> > >>> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby 
> > wrote:
> > >>>
> > > Sounds like reaching out to the inactive mentors is a great idea and
> > I
> > > think we have a great example here of how complicated it can be.
> > 
> >  Nope.  I posted that link knowing that my name would be on it, and
> >  advocated that we should be having exactly this discussion.  I should
> >  either become more active on this, or (and probably more likely)
> >  remove myself as a mentor for this podling.
> > >>>
> > >>>
> > >>> And possibly by so doing become a great example to others of us who
> > can't
> > >>> admit to ourselves that we are over-extended.
> > >>
> > >>
> > >> -
> > >> 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
> >
> >


signature.asc
Description: Digital signature

Re: Mentor disengagement - a suggestion

[Martijn Dashorst]

The biggest issue with mentor disengagement IMO is that when
incubation starts to take longer than 9 months it is hard to maintain
focus and engagement if you are not a user/member of the incubating
community (and its code).

So while we can start measuring AWOL mentors and try to fix that by
adding more administrative duties, perhaps we might think about
reasons mentors loose interest.

For example, my interest for a project I'm not involved in or invested
in in my $dayjob can only last for a couple of months, a year tops.
Longer than that, you might see my interest wane. So IM we should
strive to make incubation take not longer than a year or so. Or when
incubation takes longer, have a plan to switch mentors, perhaps a new
mentor might bring a new graduation spirit into the project.

That brings me to my second understanding of mentor (human) nature,
which is that not everybody is created equal and one person is better
in starting things, the other is better in making things manageable
and the other is better in getting results. In the life time of a
podling different focuses are needed, and perhaps not all mentors are
good in all these phases of incubation. When I'm not very good in
something, my interest often times wanes...

So we have setting up a podling which can be quite invigorating the
first few months, but then we have to start cranking out a release,
have the i's dotted and t's crossed, add new members etc. Getting a
project running is fun for some, others dread having to go through the
signing up for new mailing lists, getting a git repository, having to
explain that github is not the canonical repository for ASF projects,
etc.

So in short:
 - strive to keep incubation short
 - use different types of mentors for different types of incubation phases

Just my thoughts.

Martijn






On Mon, Oct 12, 2015 at 1:18 PM, Rich Bowen  wrote:
> Fellow mentors,
>
> There was a conversation at ApacheCon about the Incubator. I'll leave it to
> the other participants to champion the particular parts that they are
> passionate about, but I was particularly concerned with mentor
> disengagement, and suggestions for improving it.
>
> A mentor's role is to help a project learn the ropes at the ASF, and that
> mentor might not necessarily be deeply versed in the particular technology
> that the podling works with. As such, it can frequently be the case that the
> mentor becomes disengaged from the daily conversation of the lists, and
> eventually with the entire process.
>
> As a means of refocusing the mentors' efforts, and keeping them engaged, I'd
> like to encourage each mentor (or group of mentors) to consider writing a
> running report (ie, evolving, updated every quarter) based on
> https://community.apache.org/apache-way/apache-project-maturity-model.html
> where they evaluate each point on the maturity model, as a path towards
> graduation. This gives a concrete target, and a lens through which to view
> the podling's progress towards that target.
>
> This could be kept in the incubator wiki, and linked from the official
> project report, or it could be maintained just for your own benefit. I think
> it would be particularly useful to attach to a graduation recommendation, as
> a sign that the recommendation is more than just checking the various boxes,
> but is a glowing endorsement of the project's readiness to be TLP.
>
> As a side-note, I'd also encourage mentors who are mentors in name only, and
> not reality, to consider cleaning up the paperwork by removing themselves
> from the roster. It doesn't look great when a podling can't get mentor
> signoff on their reports.
>
> --
> Rich Bowen - rbo...@rcbowen.com - @rbowen
> http://apachecon.com/ - @apachecon
>
> -
> To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
> For additional commands, e-mail: general-h...@incubator.apache.org
>



-- 
Become a Wicket expert, learn from the best: http://wicketinaction.com

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

Re: Mentor disengagement - a suggestion

[Andy Seaborne]

On 14/10/15 16:21, Ross Gardler wrote:

Further to Sam's suggestion and observations below see

Suggestion 0.1.8 at 
http://wiki.apache.org/incubator/IncubatorIssues2013#Suggestions


(Summary: Champion => PPMC chair)

I agree with the comments that champion-as-chair is a negative to the 
community self-goverance.


Some kind of phased transition seems better, running down the process 
commitments of mentors as the community ramps up.


"Champion" => "lead mentor" and the incoming community nominates one of 
their own as "community point of contact" or "community lead".


"Community lead" picks up responsibilities like reporting (usually in 
the 3 months phase??) and is there to make sure sufficient mentors are 
active enough.


Andy



-Original Message-
From: sa3r...@gmail.com [mailto:sa3r...@gmail.com] On Behalf Of Sam Ruby
Sent: Wednesday, October 14, 2015 7:26 AM
To: general@incubator.apache.org
Subject: Re: Mentor disengagement - a suggestion

On Wed, Oct 14, 2015 at 10:07 AM, Ted Dunning <ted.dunn...@gmail.com> wrote:

On Wed, Oct 14, 2015 at 6:46 AM, Jim Jagielski <j...@jagunet.com> wrote:


My point is that with 1 mentor, everyone knows where the buck stops.
With >1 nobody knows. A flat hierarchy for mentors does not seem
workable or, at least, optimal.



That's a fine point.

But it is counter to my experience.

For instance, on Kylin, I was extremely active early on. Lately,
Julian Hyde has taken much more of the questions.  It has worked very
well.  We have a third mentor (whose name I won't mention) who has
been nearly completely missing, but that hasn't been a problem since
everybody knows that they are totally MiA.



If we wish to address this, and not "force" mentors to leave, we
could simply add the idea of "lead mentor" and have the PPMC vote on
which mentor they want to be the lead mentor (pseudo PPMC chair); the
remaining mentors would remain as co-mentors but the intent is that
the lead mentor would be the primary person responsible.



I think that this is largely process without clear need. But others
may have different experience.


The following is from the definition of "champion" on the incubator roles and 
responsibility page[1]:

---
During incubation, the Champion:

  * Coordinates the creation and timely delivery of the podling's board reports.
  * Keeps an eye on the mentors' activity and takes action (ask for new 
mentors, talk to the Incubator PMC) if they don't seem to provide enough 
oversight or mentorship to the podling,
---

First question: does this description match current practice.  Second
question: is there something in that description that needs to get done, but 
isn't getting done?

There also is a separate thread about the number of podlings that the Incubator 
can handle.  That number would tend to increase if some of the work were 
decentralized more.

My take is that (at least for this month) Marvin is effectively doing the first 
bullet for all podlings.  Now I don't see any signs of burnout in Marvin, so 
that's not the immediate concern, but scalability and sustainability would 
benefit from decentralizing this more.

As to the second bullet, it isn't clear to me that that is being done.
Even if it were only a one time thing, reverifying (or in many cases,
naming) the champion for every podling, and then asking each to reverify the 
active status for every mentor for podlings that they champion might go a long 
way toward easing some of the concerns that people have raised.

Ultimately the definition of champion, as posted, should match reality.

- Sam Ruby

[1] 
https://na01.safelinks.protection.outlook.com/?url=http%3a%2f%2fincubator.apache.org%2fincubation%2fRoles_and_Responsibilities.html%23Champion=01%7c01%7cRoss.Gardler%40microsoft.com%7c02e20d0e08f54f9c81f408d2d4a356cc%7c72f988bf86f141af91ab2d7cd011db47%7c1=q10OQ8l3N4%2bAPk%2bXJYxZVF1mEREPIOYNpQcD9Frr0rI%3d

-
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

Re: Mentor disengagement - a suggestion

[Jim Jagielski]

Agreed here as well.

My point is that with 1 mentor, everyone knows where the buck stops.
With >1 nobody knows. A flat hierarchy for mentors does not seem
workable or, at least, optimal.

If we wish to address this, and not "force" mentors to leave,
we could simply add the idea of "lead mentor" and have the PPMC
vote on which mentor they want to be the lead mentor (pseudo
PPMC chair); the remaining mentors would remain as co-mentors
but the intent is that the lead mentor would be the primary person
responsible.

> On Oct 14, 2015, at 9:13 AM, Daniel Gruno  wrote:
> 
> And sometimes, s/spare/sparring partner/ :)
> 
> I find it extremely useful to have a fellow mentor to bounce ideas and
> perceptions off on. Sometimes having a really engaged mentor and a more
> loosely engaged works well, as you get both a view from the inside and
> the outside.
> 
> With regards,
> Daniel.
> On 10/14/2015 03:10 PM, Ted Dunning wrote:
>> If it can work, that is very good. With intermittent availability, I have
>> often seen the need for a spare.
>> 
>> 
>> 
>> On Wed, Oct 14, 2015 at 5:53 AM, Jim Jagielski  wrote:
>> 
>>> Agreed. My only comment would be that I still think that the
>>> optimal number of mentors is 1.
>>> 
 On Oct 14, 2015, at 12:45 AM, Julian Hyde  wrote:
 
 It's not activity on the dev list, or even report signoffs, that
 matter most. Podlings, especially new podlings, have lots and lots of
 questions, especially about infrastructure. Without at least two
 responsive mentors to field those questions you feel like banging your
 head on the wall. And you start wondering why you left the comfort and
 convenience of github and whether Apache itself is fascinated by its
 own brand.
 
 Before you ask, you won't get podlings to send their questions to
 another list, because we're all too proud to ask questions which in
 retrospect always turn out to be dumb questions.
 
 It's not possible to measure that kind of mentor activity, so I think
 people are inclined to measure the "public" forms of activity as proxy
 indicators.
 
 Julian
 
 
 On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski  wrote:
> For me, I consider being a mentor as I do being a member of a PMC.
> Occasionally one simply lacks cycles to be actively involved, but
> one is involve enough to see that others *ARE* involved, and so I
> am "unconcerned" about my inactivity during those times.
> 
> My understanding is that this is OK and its one of the reasons
> why we *have* multiple mentors.
> 
> "Shaming" inactive mentors would be akin to "shaming" PMC members who
> didn't post on the dev@ list this month, or who didn't vote on a
>>> release
> or etc...
> 
> I am not, of course, referring to mentors who are truly MIA month in and
> month out. But, as someone said, if you remove those from the equation,
> the list of "active" mentors is pretty constant.
> 
> So the question is "Is there a difference or problem between a podling
> with 10 mentors, of which 4 are 'active', as compared to a podling with
> 4 mentors, all of which are 'active'"??
> 
>> On Oct 13, 2015, at 2:29 AM, Ted Dunning 
>>> wrote:
>> 
>> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby 
>>> wrote:
>> 
 Sounds like reaching out to the inactive mentors is a great idea and
>>> I
 think we have a great example here of how complicated it can be.
>>> 
>>> Nope.  I posted that link knowing that my name would be on it, and
>>> advocated that we should be having exactly this discussion.  I should
>>> either become more active on this, or (and probably more likely)
>>> remove myself as a mentor for this podling.
>> 
>> 
>> And possibly by so doing become a great example to others of us who
>>> can't
>> admit to ourselves that we are over-extended.
> 
> 
> -
> 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
>>> 
>>> 
>> 
> 
> 
> -
> To unsubscribe, e-mail: general-unsubscr...@incubator.apache.org
> For additional commands, e-mail: 

Re: Mentor disengagement - a suggestion

[Jim Jagielski]

Agreed. My only comment would be that I still think that the
optimal number of mentors is 1.

> On Oct 14, 2015, at 12:45 AM, Julian Hyde  wrote:
> 
> It's not activity on the dev list, or even report signoffs, that
> matter most. Podlings, especially new podlings, have lots and lots of
> questions, especially about infrastructure. Without at least two
> responsive mentors to field those questions you feel like banging your
> head on the wall. And you start wondering why you left the comfort and
> convenience of github and whether Apache itself is fascinated by its
> own brand.
> 
> Before you ask, you won't get podlings to send their questions to
> another list, because we're all too proud to ask questions which in
> retrospect always turn out to be dumb questions.
> 
> It's not possible to measure that kind of mentor activity, so I think
> people are inclined to measure the "public" forms of activity as proxy
> indicators.
> 
> Julian
> 
> 
> On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski  wrote:
>> For me, I consider being a mentor as I do being a member of a PMC.
>> Occasionally one simply lacks cycles to be actively involved, but
>> one is involve enough to see that others *ARE* involved, and so I
>> am "unconcerned" about my inactivity during those times.
>> 
>> My understanding is that this is OK and its one of the reasons
>> why we *have* multiple mentors.
>> 
>> "Shaming" inactive mentors would be akin to "shaming" PMC members who
>> didn't post on the dev@ list this month, or who didn't vote on a release
>> or etc...
>> 
>> I am not, of course, referring to mentors who are truly MIA month in and
>> month out. But, as someone said, if you remove those from the equation,
>> the list of "active" mentors is pretty constant.
>> 
>> So the question is "Is there a difference or problem between a podling
>> with 10 mentors, of which 4 are 'active', as compared to a podling with
>> 4 mentors, all of which are 'active'"??
>> 
>>> On Oct 13, 2015, at 2:29 AM, Ted Dunning  wrote:
>>> 
>>> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby  wrote:
>>> 
> Sounds like reaching out to the inactive mentors is a great idea and I
> think we have a great example here of how complicated it can be.
 
 Nope.  I posted that link knowing that my name would be on it, and
 advocated that we should be having exactly this discussion.  I should
 either become more active on this, or (and probably more likely)
 remove myself as a mentor for this podling.
>>> 
>>> 
>>> And possibly by so doing become a great example to others of us who can't
>>> admit to ourselves that we are over-extended.
>> 
>> 
>> -
>> 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

Re: Mentor disengagement - a suggestion

[Daniel Gruno]

And sometimes, s/spare/sparring partner/ :)

I find it extremely useful to have a fellow mentor to bounce ideas and
perceptions off on. Sometimes having a really engaged mentor and a more
loosely engaged works well, as you get both a view from the inside and
the outside.

With regards,
Daniel.
On 10/14/2015 03:10 PM, Ted Dunning wrote:
> If it can work, that is very good. With intermittent availability, I have
> often seen the need for a spare.
> 
> 
> 
> On Wed, Oct 14, 2015 at 5:53 AM, Jim Jagielski  wrote:
> 
>> Agreed. My only comment would be that I still think that the
>> optimal number of mentors is 1.
>>
>>> On Oct 14, 2015, at 12:45 AM, Julian Hyde  wrote:
>>>
>>> It's not activity on the dev list, or even report signoffs, that
>>> matter most. Podlings, especially new podlings, have lots and lots of
>>> questions, especially about infrastructure. Without at least two
>>> responsive mentors to field those questions you feel like banging your
>>> head on the wall. And you start wondering why you left the comfort and
>>> convenience of github and whether Apache itself is fascinated by its
>>> own brand.
>>>
>>> Before you ask, you won't get podlings to send their questions to
>>> another list, because we're all too proud to ask questions which in
>>> retrospect always turn out to be dumb questions.
>>>
>>> It's not possible to measure that kind of mentor activity, so I think
>>> people are inclined to measure the "public" forms of activity as proxy
>>> indicators.
>>>
>>> Julian
>>>
>>>
>>> On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski  wrote:
 For me, I consider being a mentor as I do being a member of a PMC.
 Occasionally one simply lacks cycles to be actively involved, but
 one is involve enough to see that others *ARE* involved, and so I
 am "unconcerned" about my inactivity during those times.

 My understanding is that this is OK and its one of the reasons
 why we *have* multiple mentors.

 "Shaming" inactive mentors would be akin to "shaming" PMC members who
 didn't post on the dev@ list this month, or who didn't vote on a
>> release
 or etc...

 I am not, of course, referring to mentors who are truly MIA month in and
 month out. But, as someone said, if you remove those from the equation,
 the list of "active" mentors is pretty constant.

 So the question is "Is there a difference or problem between a podling
 with 10 mentors, of which 4 are 'active', as compared to a podling with
 4 mentors, all of which are 'active'"??

> On Oct 13, 2015, at 2:29 AM, Ted Dunning 
>> wrote:
>
> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby 
>> wrote:
>
>>> Sounds like reaching out to the inactive mentors is a great idea and
>> I
>>> think we have a great example here of how complicated it can be.
>>
>> Nope.  I posted that link knowing that my name would be on it, and
>> advocated that we should be having exactly this discussion.  I should
>> either become more active on this, or (and probably more likely)
>> remove myself as a mentor for this podling.
>
>
> And possibly by so doing become a great example to others of us who
>> can't
> admit to ourselves that we are over-extended.


 -
 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
>>
>>
> 


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

Re: Mentor disengagement - a suggestion

[Marko Rodriguez]

Hello,

I concur with this. Once we got one of our mentors engaged -- Daniel Gruno -- 
its been smooth sailing. You know how we got him engaged? HipChat. He set us up 
an account and now we can "@Humbedooh" with questions and get responses. No 
more "Hello?! Please answer our emails..."

So yes, optimal number of mentors -- 1. And if that mentor doesn't know the 
answer to the question, he can route you appropriately.

Marko.

http://markorodriguez.com

On Oct 14, 2015, at 6:53 AM, Jim Jagielski  wrote:

> Agreed. My only comment would be that I still think that the
> optimal number of mentors is 1.
> 
>> On Oct 14, 2015, at 12:45 AM, Julian Hyde  wrote:
>> 
>> It's not activity on the dev list, or even report signoffs, that
>> matter most. Podlings, especially new podlings, have lots and lots of
>> questions, especially about infrastructure. Without at least two
>> responsive mentors to field those questions you feel like banging your
>> head on the wall. And you start wondering why you left the comfort and
>> convenience of github and whether Apache itself is fascinated by its
>> own brand.
>> 
>> Before you ask, you won't get podlings to send their questions to
>> another list, because we're all too proud to ask questions which in
>> retrospect always turn out to be dumb questions.
>> 
>> It's not possible to measure that kind of mentor activity, so I think
>> people are inclined to measure the "public" forms of activity as proxy
>> indicators.
>> 
>> Julian
>> 
>> 
>> On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski  wrote:
>>> For me, I consider being a mentor as I do being a member of a PMC.
>>> Occasionally one simply lacks cycles to be actively involved, but
>>> one is involve enough to see that others *ARE* involved, and so I
>>> am "unconcerned" about my inactivity during those times.
>>> 
>>> My understanding is that this is OK and its one of the reasons
>>> why we *have* multiple mentors.
>>> 
>>> "Shaming" inactive mentors would be akin to "shaming" PMC members who
>>> didn't post on the dev@ list this month, or who didn't vote on a release
>>> or etc...
>>> 
>>> I am not, of course, referring to mentors who are truly MIA month in and
>>> month out. But, as someone said, if you remove those from the equation,
>>> the list of "active" mentors is pretty constant.
>>> 
>>> So the question is "Is there a difference or problem between a podling
>>> with 10 mentors, of which 4 are 'active', as compared to a podling with
>>> 4 mentors, all of which are 'active'"??
>>> 
 On Oct 13, 2015, at 2:29 AM, Ted Dunning  wrote:
 
 On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby  wrote:
 
>> Sounds like reaching out to the inactive mentors is a great idea and I
>> think we have a great example here of how complicated it can be.
> 
> Nope.  I posted that link knowing that my name would be on it, and
> advocated that we should be having exactly this discussion.  I should
> either become more active on this, or (and probably more likely)
> remove myself as a mentor for this podling.
 
 
 And possibly by so doing become a great example to others of us who can't
 admit to ourselves that we are over-extended.
>>> 
>>> 
>>> -
>>> 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
> 

Re: Mentor disengagement - a suggestion

[Ted Dunning]

If it can work, that is very good. With intermittent availability, I have
often seen the need for a spare.



On Wed, Oct 14, 2015 at 5:53 AM, Jim Jagielski  wrote:

> Agreed. My only comment would be that I still think that the
> optimal number of mentors is 1.
>
> > On Oct 14, 2015, at 12:45 AM, Julian Hyde  wrote:
> >
> > It's not activity on the dev list, or even report signoffs, that
> > matter most. Podlings, especially new podlings, have lots and lots of
> > questions, especially about infrastructure. Without at least two
> > responsive mentors to field those questions you feel like banging your
> > head on the wall. And you start wondering why you left the comfort and
> > convenience of github and whether Apache itself is fascinated by its
> > own brand.
> >
> > Before you ask, you won't get podlings to send their questions to
> > another list, because we're all too proud to ask questions which in
> > retrospect always turn out to be dumb questions.
> >
> > It's not possible to measure that kind of mentor activity, so I think
> > people are inclined to measure the "public" forms of activity as proxy
> > indicators.
> >
> > Julian
> >
> >
> > On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski  wrote:
> >> For me, I consider being a mentor as I do being a member of a PMC.
> >> Occasionally one simply lacks cycles to be actively involved, but
> >> one is involve enough to see that others *ARE* involved, and so I
> >> am "unconcerned" about my inactivity during those times.
> >>
> >> My understanding is that this is OK and its one of the reasons
> >> why we *have* multiple mentors.
> >>
> >> "Shaming" inactive mentors would be akin to "shaming" PMC members who
> >> didn't post on the dev@ list this month, or who didn't vote on a
> release
> >> or etc...
> >>
> >> I am not, of course, referring to mentors who are truly MIA month in and
> >> month out. But, as someone said, if you remove those from the equation,
> >> the list of "active" mentors is pretty constant.
> >>
> >> So the question is "Is there a difference or problem between a podling
> >> with 10 mentors, of which 4 are 'active', as compared to a podling with
> >> 4 mentors, all of which are 'active'"??
> >>
> >>> On Oct 13, 2015, at 2:29 AM, Ted Dunning 
> wrote:
> >>>
> >>> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby 
> wrote:
> >>>
> > Sounds like reaching out to the inactive mentors is a great idea and
> I
> > think we have a great example here of how complicated it can be.
> 
>  Nope.  I posted that link knowing that my name would be on it, and
>  advocated that we should be having exactly this discussion.  I should
>  either become more active on this, or (and probably more likely)
>  remove myself as a mentor for this podling.
> >>>
> >>>
> >>> And possibly by so doing become a great example to others of us who
> can't
> >>> admit to ourselves that we are over-extended.
> >>
> >>
> >> -
> >> 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
>
>

Re: Mentor disengagement - a suggestion

[Daniel Gruno]

I'll note that this is mostly questions about legalities of things and
technical tidbits - we are not having community discussions off-list per
se, just questions from individuals on how to word this, phrase that,
what RTC/CTR is etc. And everything is brought back to the list for a
thorough discussion if it impacts the project/community in any way.

With regards,
Daniel.

On 10/14/2015 03:15 PM, Marko Rodriguez wrote:
> Hello,
> 
> I concur with this. Once we got one of our mentors engaged -- Daniel Gruno -- 
> its been smooth sailing. You know how we got him engaged? HipChat. He set us 
> up an account and now we can "@Humbedooh" with questions and get responses. 
> No more "Hello?! Please answer our emails..."
> 
> So yes, optimal number of mentors -- 1. And if that mentor doesn't know the 
> answer to the question, he can route you appropriately.
> 
> Marko.
> 
> http://markorodriguez.com
> 
> On Oct 14, 2015, at 6:53 AM, Jim Jagielski  wrote:
> 
>> Agreed. My only comment would be that I still think that the
>> optimal number of mentors is 1.
>>
>>> On Oct 14, 2015, at 12:45 AM, Julian Hyde  wrote:
>>>
>>> It's not activity on the dev list, or even report signoffs, that
>>> matter most. Podlings, especially new podlings, have lots and lots of
>>> questions, especially about infrastructure. Without at least two
>>> responsive mentors to field those questions you feel like banging your
>>> head on the wall. And you start wondering why you left the comfort and
>>> convenience of github and whether Apache itself is fascinated by its
>>> own brand.
>>>
>>> Before you ask, you won't get podlings to send their questions to
>>> another list, because we're all too proud to ask questions which in
>>> retrospect always turn out to be dumb questions.
>>>
>>> It's not possible to measure that kind of mentor activity, so I think
>>> people are inclined to measure the "public" forms of activity as proxy
>>> indicators.
>>>
>>> Julian
>>>
>>>
>>> On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski  wrote:
 For me, I consider being a mentor as I do being a member of a PMC.
 Occasionally one simply lacks cycles to be actively involved, but
 one is involve enough to see that others *ARE* involved, and so I
 am "unconcerned" about my inactivity during those times.

 My understanding is that this is OK and its one of the reasons
 why we *have* multiple mentors.

 "Shaming" inactive mentors would be akin to "shaming" PMC members who
 didn't post on the dev@ list this month, or who didn't vote on a release
 or etc...

 I am not, of course, referring to mentors who are truly MIA month in and
 month out. But, as someone said, if you remove those from the equation,
 the list of "active" mentors is pretty constant.

 So the question is "Is there a difference or problem between a podling
 with 10 mentors, of which 4 are 'active', as compared to a podling with
 4 mentors, all of which are 'active'"??

> On Oct 13, 2015, at 2:29 AM, Ted Dunning  wrote:
>
> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby  wrote:
>
>>> Sounds like reaching out to the inactive mentors is a great idea and I
>>> think we have a great example here of how complicated it can be.
>>
>> Nope.  I posted that link knowing that my name would be on it, and
>> advocated that we should be having exactly this discussion.  I should
>> either become more active on this, or (and probably more likely)
>> remove myself as a mentor for this podling.
>
>
> And possibly by so doing become a great example to others of us who can't
> admit to ourselves that we are over-extended.


 -
 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
>>
> 
> 


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

Re: Mentor disengagement - a suggestion

[Ted Dunning]

On Wed, Oct 14, 2015 at 6:46 AM, Jim Jagielski  wrote:

> My point is that with 1 mentor, everyone knows where the buck stops.
> With >1 nobody knows. A flat hierarchy for mentors does not seem
> workable or, at least, optimal.
>

That's a fine point.

But it is counter to my experience.

For instance, on Kylin, I was extremely active early on. Lately, Julian
Hyde has taken much more of the questions.  It has worked very well.  We
have a third mentor (whose name I won't mention) who has been nearly
completely missing, but that hasn't been a problem since everybody knows
that they are totally MiA.


> If we wish to address this, and not "force" mentors to leave,
> we could simply add the idea of "lead mentor" and have the PPMC
> vote on which mentor they want to be the lead mentor (pseudo
> PPMC chair); the remaining mentors would remain as co-mentors
> but the intent is that the lead mentor would be the primary person
> responsible.
>

I think that this is largely process without clear need. But others may
have different experience.

Re: Mentor disengagement - a suggestion

[Sam Ruby]

On Wed, Oct 14, 2015 at 9:46 AM, Jim Jagielski  wrote:
> Agreed here as well.
>
> My point is that with 1 mentor, everyone knows where the buck stops.
> With >1 nobody knows. A flat hierarchy for mentors does not seem
> workable or, at least, optimal.
>
> If we wish to address this, and not "force" mentors to leave,
> we could simply add the idea of "lead mentor" and have the PPMC
> vote on which mentor they want to be the lead mentor (pseudo
> PPMC chair); the remaining mentors would remain as co-mentors
> but the intent is that the lead mentor would be the primary person
> responsible.

Perhaps this could be accomplished by tweaking the definition of champion?

http://incubator.apache.org/incubation/Roles_and_Responsibilities.html#Champion
http://incubator.apache.org/incubation/Roles_and_Responsibilities.html#Mentor

- Sam Ruby


>> On Oct 14, 2015, at 9:13 AM, Daniel Gruno  wrote:
>>
>> And sometimes, s/spare/sparring partner/ :)
>>
>> I find it extremely useful to have a fellow mentor to bounce ideas and
>> perceptions off on. Sometimes having a really engaged mentor and a more
>> loosely engaged works well, as you get both a view from the inside and
>> the outside.
>>
>> With regards,
>> Daniel.
>> On 10/14/2015 03:10 PM, Ted Dunning wrote:
>>> If it can work, that is very good. With intermittent availability, I have
>>> often seen the need for a spare.
>>>
>>>
>>>
>>> On Wed, Oct 14, 2015 at 5:53 AM, Jim Jagielski  wrote:
>>>
 Agreed. My only comment would be that I still think that the
 optimal number of mentors is 1.

> On Oct 14, 2015, at 12:45 AM, Julian Hyde  wrote:
>
> It's not activity on the dev list, or even report signoffs, that
> matter most. Podlings, especially new podlings, have lots and lots of
> questions, especially about infrastructure. Without at least two
> responsive mentors to field those questions you feel like banging your
> head on the wall. And you start wondering why you left the comfort and
> convenience of github and whether Apache itself is fascinated by its
> own brand.
>
> Before you ask, you won't get podlings to send their questions to
> another list, because we're all too proud to ask questions which in
> retrospect always turn out to be dumb questions.
>
> It's not possible to measure that kind of mentor activity, so I think
> people are inclined to measure the "public" forms of activity as proxy
> indicators.
>
> Julian
>
>
> On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski  wrote:
>> For me, I consider being a mentor as I do being a member of a PMC.
>> Occasionally one simply lacks cycles to be actively involved, but
>> one is involve enough to see that others *ARE* involved, and so I
>> am "unconcerned" about my inactivity during those times.
>>
>> My understanding is that this is OK and its one of the reasons
>> why we *have* multiple mentors.
>>
>> "Shaming" inactive mentors would be akin to "shaming" PMC members who
>> didn't post on the dev@ list this month, or who didn't vote on a
 release
>> or etc...
>>
>> I am not, of course, referring to mentors who are truly MIA month in and
>> month out. But, as someone said, if you remove those from the equation,
>> the list of "active" mentors is pretty constant.
>>
>> So the question is "Is there a difference or problem between a podling
>> with 10 mentors, of which 4 are 'active', as compared to a podling with
>> 4 mentors, all of which are 'active'"??
>>
>>> On Oct 13, 2015, at 2:29 AM, Ted Dunning 
 wrote:
>>>
>>> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby 
 wrote:
>>>
> Sounds like reaching out to the inactive mentors is a great idea and
 I
> think we have a great example here of how complicated it can be.

 Nope.  I posted that link knowing that my name would be on it, and
 advocated that we should be having exactly this discussion.  I should
 either become more active on this, or (and probably more likely)
 remove myself as a mentor for this podling.
>>>
>>>
>>> And possibly by so doing become a great example to others of us who
 can't
>>> admit to ourselves that we are over-extended.
>>
>>
>> -
>> 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
>


 

Re: Mentor disengagement - a suggestion

[Julian Hyde]

It's not activity on the dev list, or even report signoffs, that
matter most. Podlings, especially new podlings, have lots and lots of
questions, especially about infrastructure. Without at least two
responsive mentors to field those questions you feel like banging your
head on the wall. And you start wondering why you left the comfort and
convenience of github and whether Apache itself is fascinated by its
own brand.

Before you ask, you won't get podlings to send their questions to
another list, because we're all too proud to ask questions which in
retrospect always turn out to be dumb questions.

It's not possible to measure that kind of mentor activity, so I think
people are inclined to measure the "public" forms of activity as proxy
indicators.

Julian


On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski  wrote:
> For me, I consider being a mentor as I do being a member of a PMC.
> Occasionally one simply lacks cycles to be actively involved, but
> one is involve enough to see that others *ARE* involved, and so I
> am "unconcerned" about my inactivity during those times.
>
> My understanding is that this is OK and its one of the reasons
> why we *have* multiple mentors.
>
> "Shaming" inactive mentors would be akin to "shaming" PMC members who
> didn't post on the dev@ list this month, or who didn't vote on a release
> or etc...
>
> I am not, of course, referring to mentors who are truly MIA month in and
> month out. But, as someone said, if you remove those from the equation,
> the list of "active" mentors is pretty constant.
>
> So the question is "Is there a difference or problem between a podling
> with 10 mentors, of which 4 are 'active', as compared to a podling with
> 4 mentors, all of which are 'active'"??
>
>> On Oct 13, 2015, at 2:29 AM, Ted Dunning  wrote:
>>
>> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby  wrote:
>>
 Sounds like reaching out to the inactive mentors is a great idea and I
 think we have a great example here of how complicated it can be.
>>>
>>> Nope.  I posted that link knowing that my name would be on it, and
>>> advocated that we should be having exactly this discussion.  I should
>>> either become more active on this, or (and probably more likely)
>>> remove myself as a mentor for this podling.
>>
>>
>> And possibly by so doing become a great example to others of us who can't
>> admit to ourselves that we are over-extended.
>
>
> -
> 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

RE: Mentor disengagement - a suggestion

[Ross Gardler]

+1000 (though I would argue a single highly committed mentor is sufficient)

-Original Message-
From: Julian Hyde [mailto:jh...@apache.org] 
Sent: Tuesday, October 13, 2015 9:46 PM
To: general@incubator.apache.org
Subject: Re: Mentor disengagement - a suggestion

It's not activity on the dev list, or even report signoffs, that matter most. 
Podlings, especially new podlings, have lots and lots of questions, especially 
about infrastructure. Without at least two responsive mentors to field those 
questions you feel like banging your head on the wall. And you start wondering 
why you left the comfort and convenience of github and whether Apache itself is 
fascinated by its own brand.

Before you ask, you won't get podlings to send their questions to another list, 
because we're all too proud to ask questions which in retrospect always turn 
out to be dumb questions.

It's not possible to measure that kind of mentor activity, so I think people 
are inclined to measure the "public" forms of activity as proxy indicators.

Julian


On Tue, Oct 13, 2015 at 4:19 AM, Jim Jagielski <j...@jagunet.com> wrote:
> For me, I consider being a mentor as I do being a member of a PMC.
> Occasionally one simply lacks cycles to be actively involved, but one 
> is involve enough to see that others *ARE* involved, and so I am 
> "unconcerned" about my inactivity during those times.
>
> My understanding is that this is OK and its one of the reasons why we 
> *have* multiple mentors.
>
> "Shaming" inactive mentors would be akin to "shaming" PMC members who 
> didn't post on the dev@ list this month, or who didn't vote on a 
> release or etc...
>
> I am not, of course, referring to mentors who are truly MIA month in 
> and month out. But, as someone said, if you remove those from the 
> equation, the list of "active" mentors is pretty constant.
>
> So the question is "Is there a difference or problem between a podling 
> with 10 mentors, of which 4 are 'active', as compared to a podling 
> with
> 4 mentors, all of which are 'active'"??
>
>> On Oct 13, 2015, at 2:29 AM, Ted Dunning <ted.dunn...@gmail.com> wrote:
>>
>> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby <ru...@intertwingly.net> wrote:
>>
>>>> Sounds like reaching out to the inactive mentors is a great idea 
>>>> and I think we have a great example here of how complicated it can be.
>>>
>>> Nope.  I posted that link knowing that my name would be on it, and 
>>> advocated that we should be having exactly this discussion.  I 
>>> should either become more active on this, or (and probably more 
>>> likely) remove myself as a mentor for this podling.
>>
>>
>> And possibly by so doing become a great example to others of us who 
>> can't admit to ourselves that we are over-extended.
>
>
> -
> 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

Re: Mentor disengagement - a suggestion

[Nick Kew]

On Mon, 2015-10-12 at 16:10 +0200, Emmanuel Lécharny wrote:

> Regarding inactive mentors, this is quite simple : we have a monthly
> report that has to be signed off by mentors, if one mentor does not sign
> it three time in a raw, shouldn't we consider that this mentor has
> already stepped down ?

Hmmm.  I can think of at least one occasion when that would've
saved me sinking into limbo from which it gets ever harder to
extricate oneself.  I express interest in a podling, get signed
up as mentor without really meaning to.  Then my personal
circumstances get difficult, and all Apache activities get shelved
while I'm caring for a serious invalid.  She dies, but it's
a year on and I can't re-engage with the podling.  Or indeed,
for quite some time, the Incubator.

A well-defined withdraw-by-default process would've been a
blessing.  And I expect many disengaged mentors could tell
similar stories.

-- 
Nick Kew


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

Re: Mentor disengagement - a suggestion

[Ted Dunning]

On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby  wrote:

> > Sounds like reaching out to the inactive mentors is a great idea and I
> > think we have a great example here of how complicated it can be.
>
> Nope.  I posted that link knowing that my name would be on it, and
> advocated that we should be having exactly this discussion.  I should
> either become more active on this, or (and probably more likely)
> remove myself as a mentor for this podling.


And possibly by so doing become a great example to others of us who can't
admit to ourselves that we are over-extended.

Re: Mentor disengagement - a suggestion

[Jim Jagielski]

For me, I consider being a mentor as I do being a member of a PMC.
Occasionally one simply lacks cycles to be actively involved, but
one is involve enough to see that others *ARE* involved, and so I
am "unconcerned" about my inactivity during those times.

My understanding is that this is OK and its one of the reasons
why we *have* multiple mentors.

"Shaming" inactive mentors would be akin to "shaming" PMC members who
didn't post on the dev@ list this month, or who didn't vote on a release
or etc...

I am not, of course, referring to mentors who are truly MIA month in and
month out. But, as someone said, if you remove those from the equation,
the list of "active" mentors is pretty constant.

So the question is "Is there a difference or problem between a podling
with 10 mentors, of which 4 are 'active', as compared to a podling with
4 mentors, all of which are 'active'"??

> On Oct 13, 2015, at 2:29 AM, Ted Dunning  wrote:
> 
> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby  wrote:
> 
>>> Sounds like reaching out to the inactive mentors is a great idea and I
>>> think we have a great example here of how complicated it can be.
>> 
>> Nope.  I posted that link knowing that my name would be on it, and
>> advocated that we should be having exactly this discussion.  I should
>> either become more active on this, or (and probably more likely)
>> remove myself as a mentor for this podling.
> 
> 
> And possibly by so doing become a great example to others of us who can't
> admit to ourselves that we are over-extended.


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

Re: Mentor disengagement - a suggestion

[Tim Williams]

On Mon, Oct 12, 2015 at 6:37 PM, Sam Ruby  wrote:
> On Mon, Oct 12, 2015 at 11:06 AM, Rich Bowen  wrote:
>>
>> On 10/12/2015 10:10 AM, Emmanuel Lécharny wrote:
>>>
>>> Regarding inactive mentors, this is quite simple : we have a monthly
>>> report that has to be signed off by mentors, if one mentor does not sign
>>> it three time in a raw, shouldn't we consider that this mentor has
>>> already stepped down ?
>>
>> No, it's not simple. Actively removing people from volunteer roles is much
>> more complicated than you might suppose. I'd rather focus on making myself a
>> better mentor than any measures against other mentors which might be seen as
>> punitive.
>
> Getting some things out of the way:
>
> I agree that it is not simple.  I agree that it is complicated.  I
> disagree that it needs to be viewed as punitive.  I suggest it would
> be worthwhile to do.  I respect that this may not be something you
> personally would want to volunteer for.  I believe that you can't fix
> what you don't measure.  I acknowledge that measuring may lead to
> gaming, though in this case I think the rewards are so vanishingly
> small that that is unlikely to be a major issue.
>
> Whew!
>
> Now on to the substance of my reply:
>
> https://whimsy.apache.org/incubator/signoff

Hi Sam,
A small request - any chance you can adjust the colors for that page?
David added some notes to Clutch[1] for details but I reckon
#009e73 and #d55e00 would be better.

Thanks,
--tim

[1] - http://incubator.apache.org/clutch.html#notes-cud

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

Re: Mentor disengagement - a suggestion

[Sam Ruby]

On Tue, Oct 13, 2015 at 7:19 AM, Jim Jagielski  wrote:
> For me, I consider being a mentor as I do being a member of a PMC.
> Occasionally one simply lacks cycles to be actively involved, but
> one is involve enough to see that others *ARE* involved, and so I
> am "unconcerned" about my inactivity during those times.
>
> My understanding is that this is OK and its one of the reasons
> why we *have* multiple mentors.

Agreed.

> "Shaming" inactive mentors would be akin to "shaming" PMC members who
> didn't post on the dev@ list this month, or who didn't vote on a release
> or etc...

I do not advocate shaming.  I do advocate (1) making changes to policy
based on actual data, and (2) using that data to have personal
discussions.

What spawned this period of introspection was a proposal made
elsewhere that to me seemed like the essence of truthiness[1].  I
would like to replace that, wherever possible, with actual data;
however uncomfortable that might be.  I encourage those that gather
such data to be sensitive when using that data.  In particular, Ted's
noting that I haven't been a visibly active mentor for odftoolkit is
probably something that should have been done privately.  It worked
out fine with me, but it might not with others.

> I am not, of course, referring to mentors who are truly MIA month in and
> month out. But, as someone said, if you remove those from the equation,
> the list of "active" mentors is pretty constant.

I think it would be helpful if we had an accurate picture of
mentorship.  This will help with things like evaluating graduation
resolutions.

> So the question is "Is there a difference or problem between a podling
> with 10 mentors, of which 4 are 'active', as compared to a podling with
> 4 mentors, all of which are 'active'"??

I know many mentors personally.  Knowing which podlings you (Jim) are
on top of makes me comfortable that I can focus elsewhere.  Having me
as a mentor of record on a podling that I am not focusing on may be
misleading.

- Sam Ruby

[1] https://en.wikipedia.org/wiki/Truthiness

>> On Oct 13, 2015, at 2:29 AM, Ted Dunning  wrote:
>>
>> On Mon, Oct 12, 2015 at 11:05 PM, Sam Ruby  wrote:
>>
 Sounds like reaching out to the inactive mentors is a great idea and I
 think we have a great example here of how complicated it can be.
>>>
>>> Nope.  I posted that link knowing that my name would be on it, and
>>> advocated that we should be having exactly this discussion.  I should
>>> either become more active on this, or (and probably more likely)
>>> remove myself as a mentor for this podling.
>>
>>
>> And possibly by so doing become a great example to others of us who can't
>> admit to ourselves that we are over-extended.
>
>
> -
> 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

Re: Mentor disengagement - a suggestion

[Tim Williams]

On Tue, Oct 13, 2015 at 2:32 PM, Sam Ruby  wrote:
> On Tue, Oct 13, 2015 at 8:28 AM, Tim Williams  wrote:
>> On Mon, Oct 12, 2015 at 6:37 PM, Sam Ruby  wrote:
>>> On Mon, Oct 12, 2015 at 11:06 AM, Rich Bowen  wrote:

 On 10/12/2015 10:10 AM, Emmanuel Lécharny wrote:
>
> Regarding inactive mentors, this is quite simple : we have a monthly
> report that has to be signed off by mentors, if one mentor does not sign
> it three time in a raw, shouldn't we consider that this mentor has
> already stepped down ?

 No, it's not simple. Actively removing people from volunteer roles is much
 more complicated than you might suppose. I'd rather focus on making myself 
 a
 better mentor than any measures against other mentors which might be seen 
 as
 punitive.
>>>
>>> Getting some things out of the way:
>>>
>>> I agree that it is not simple.  I agree that it is complicated.  I
>>> disagree that it needs to be viewed as punitive.  I suggest it would
>>> be worthwhile to do.  I respect that this may not be something you
>>> personally would want to volunteer for.  I believe that you can't fix
>>> what you don't measure.  I acknowledge that measuring may lead to
>>> gaming, though in this case I think the rewards are so vanishingly
>>> small that that is unlikely to be a major issue.
>>>
>>> Whew!
>>>
>>> Now on to the substance of my reply:
>>>
>>> https://whimsy.apache.org/incubator/signoff
>>
>> Hi Sam,
>> A small request - any chance you can adjust the colors for that page?
>> David added some notes to Clutch[1] for details but I reckon
>> #009e73 and #d55e00 would be better.
>
> I did a small amount of research, and made a different color
> choice[2].  For extra measure, I varied both the font-weight and
> font-style.  Let me know if this doesn't work for you.

That's fantastic!  Thanks for taking the extra time Sam, it's just
great to see a distinction.

Thanks again,
--tim

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

Re: Mentor disengagement - a suggestion

[Sam Ruby]

On Tue, Oct 13, 2015 at 8:28 AM, Tim Williams  wrote:
> On Mon, Oct 12, 2015 at 6:37 PM, Sam Ruby  wrote:
>> On Mon, Oct 12, 2015 at 11:06 AM, Rich Bowen  wrote:
>>>
>>> On 10/12/2015 10:10 AM, Emmanuel Lécharny wrote:

 Regarding inactive mentors, this is quite simple : we have a monthly
 report that has to be signed off by mentors, if one mentor does not sign
 it three time in a raw, shouldn't we consider that this mentor has
 already stepped down ?
>>>
>>> No, it's not simple. Actively removing people from volunteer roles is much
>>> more complicated than you might suppose. I'd rather focus on making myself a
>>> better mentor than any measures against other mentors which might be seen as
>>> punitive.
>>
>> Getting some things out of the way:
>>
>> I agree that it is not simple.  I agree that it is complicated.  I
>> disagree that it needs to be viewed as punitive.  I suggest it would
>> be worthwhile to do.  I respect that this may not be something you
>> personally would want to volunteer for.  I believe that you can't fix
>> what you don't measure.  I acknowledge that measuring may lead to
>> gaming, though in this case I think the rewards are so vanishingly
>> small that that is unlikely to be a major issue.
>>
>> Whew!
>>
>> Now on to the substance of my reply:
>>
>> https://whimsy.apache.org/incubator/signoff
>
> Hi Sam,
> A small request - any chance you can adjust the colors for that page?
> David added some notes to Clutch[1] for details but I reckon
> #009e73 and #d55e00 would be better.

I did a small amount of research, and made a different color
choice[2].  For extra measure, I varied both the font-weight and
font-style.  Let me know if this doesn't work for you.

> Thanks,
> --tim
>
> [1] - http://incubator.apache.org/clutch.html#notes-cud
[2] http://bconnelly.net/2013/10/creating-colorblind-friendly-figures/

- Sam Ruby

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

Re: Mentor disengagement - a suggestion

[Sam Ruby]

On Tue, Oct 13, 2015 at 12:39 AM, Ted Dunning  wrote:
> On Mon, Oct 12, 2015 at 11:37 AM, Sam Ruby  wrote:
>
>> Now on to the substance of my reply:
>>
>> https://whimsy.apache.org/incubator/signoff
>>
>> If we can get some volunteers to split this list up, perhaps we can
>> reach out to those that haven't been participating in signoffs and see
>> what changes need to be made.
>
> Uh... Sam, I see you haven't been signing off on odftoolkit. Is this
> something we should be concerned about?

Yes, indeed.

> Sounds like reaching out to the inactive mentors is a great idea and I
> think we have a great example here of how complicated it can be.

Nope.  I posted that link knowing that my name would be on it, and
advocated that we should be having exactly this discussion.  I should
either become more active on this, or (and probably more likely)
remove myself as a mentor for this podling.

- Sam Ruby

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

Re: Mentor disengagement - a suggestion

[Emmanuel Lécharny]

Le 12/10/15 13:18, Rich Bowen a écrit :
> Fellow mentors,
>
> There was a conversation at ApacheCon about the Incubator. I'll leave
> it to the other participants to champion the particular parts that
> they are passionate about, but I was particularly concerned with
> mentor disengagement, and suggestions for improving it.
>
> A mentor's role is to help a project learn the ropes at the ASF, and
> that mentor might not necessarily be deeply versed in the particular
> technology that the podling works with. As such, it can frequently be
> the case that the mentor becomes disengaged from the daily
> conversation of the lists, and eventually with the entire process.
>
> As a means of refocusing the mentors' efforts, and keeping them
> engaged, I'd like to encourage each mentor (or group of mentors) to
> consider writing a running report (ie, evolving, updated every
> quarter) based on
> https://community.apache.org/apache-way/apache-project-maturity-model.html
> where they evaluate each point on the maturity model, as a path
> towards graduation. This gives a concrete target, and a lens through
> which to view the podling's progress towards that target.
>
> This could be kept in the incubator wiki, and linked from the official
> project report, or it could be maintained just for your own benefit. I
> think it would be particularly useful to attach to a graduation
> recommendation, as a sign that the recommendation is more than just
> checking the various boxes, but is a glowing endorsement of the
> project's readiness to be TLP.
>
> As a side-note, I'd also encourage mentors who are mentors in name
> only, and not reality, to consider cleaning up the paperwork by
> removing themselves from the roster. It doesn't look great when a
> podling can't get mentor signoff on their reports.
>
Sounds liek a good idea. I would also encourage mentors to actually add
a comment in each report they are signing.

Actually, when discussing a podling potential exit from incubation, a
[DISUSSION] thread is started on the podling mailing list, and the
mentors can express their opinion, which can be aggregated in a report
when the podling is proposed for exit, or on the opposite, when the
podling is seen as not ready, and send this report to the board (within
all the other podling reports).

Regarding inactive mentors, this is quite simple : we have a monthly
report that has to be signed off by mentors, if one mentor does not sign
it three time in a raw, shouldn't we consider that this mentor has
already stepped down ?

My 2cts...


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

Re: Mentor disengagement - a suggestion

[Rich Bowen]

On 10/12/2015 10:10 AM, Emmanuel Lécharny wrote:

Regarding inactive mentors, this is quite simple : we have a monthly
report that has to be signed off by mentors, if one mentor does not sign
it three time in a raw, shouldn't we consider that this mentor has
already stepped down ?



No, it's not simple. Actively removing people from volunteer roles is 
much more complicated than you might suppose. I'd rather focus on making 
myself a better mentor than any measures against other mentors which 
might be seen as punitive.



--
Rich Bowen - rbo...@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon

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

Re: Mentor disengagement - a suggestion

[Sam Ruby]

On Mon, Oct 12, 2015 at 11:06 AM, Rich Bowen  wrote:
>
> On 10/12/2015 10:10 AM, Emmanuel Lécharny wrote:
>>
>> Regarding inactive mentors, this is quite simple : we have a monthly
>> report that has to be signed off by mentors, if one mentor does not sign
>> it three time in a raw, shouldn't we consider that this mentor has
>> already stepped down ?
>
> No, it's not simple. Actively removing people from volunteer roles is much
> more complicated than you might suppose. I'd rather focus on making myself a
> better mentor than any measures against other mentors which might be seen as
> punitive.

Getting some things out of the way:

I agree that it is not simple.  I agree that it is complicated.  I
disagree that it needs to be viewed as punitive.  I suggest it would
be worthwhile to do.  I respect that this may not be something you
personally would want to volunteer for.  I believe that you can't fix
what you don't measure.  I acknowledge that measuring may lead to
gaming, though in this case I think the rewards are so vanishingly
small that that is unlikely to be a major issue.

Whew!

Now on to the substance of my reply:

https://whimsy.apache.org/incubator/signoff

If we can get some volunteers to split this list up, perhaps we can
reach out to those that haven't been participating in signoffs and see
what changes need to be made.

Any takers?

- Sam Ruby

P.S.  Despite this being pulled from either public or soon to be
public data, I have limited access to IPMC members and ASF members.
At a minimum, this should keep the page out of search engine results.


> --
> Rich Bowen - rbo...@rcbowen.com - @rbowen
> http://apachecon.com/ - @apachecon
>
> -
> 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

Re: Mentor disengagement - a suggestion

[Ted Dunning]

On Mon, Oct 12, 2015 at 11:37 AM, Sam Ruby  wrote:

> Now on to the substance of my reply:
>
> https://whimsy.apache.org/incubator/signoff
>
> If we can get some volunteers to split this list up, perhaps we can
> reach out to those that haven't been participating in signoffs and see
> what changes need to be made.
>

Reaching out is a great idea.

(not volunteering yet)

Re: Mentor disengagement - a suggestion

[Ted Dunning]

On Mon, Oct 12, 2015 at 11:37 AM, Sam Ruby  wrote:

>
> Now on to the substance of my reply:
>
> https://whimsy.apache.org/incubator/signoff
>
> If we can get some volunteers to split this list up, perhaps we can
> reach out to those that haven't been participating in signoffs and see
> what changes need to be made.


Uh... Sam, I see you haven't been signing off on odftoolkit. Is this
something we should be concerned about?

Sounds like reaching out to the inactive mentors is a great idea and I
think we have a great example here of how complicated it can be.