You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by Nicola Ken Barozzi <ni...@apache.org> on 2003/11/12 10:28:46 UTC
[RT] Multiple Mentors
From the current incubations, it's becoming quite clear IMO that a
single Mentor is not enough for most, if not all, incubated projects.
As PMCs are composed of more than one member, we need more Mentors or
each project, so that there should be *at least* one of them active in
every moment. In the projects where there are more than one Mentor
already things have been going better in this regard, in fact.
What has been said though is that if more than one have to do something,
then nobody does it, as they assume others did. From what I see at
Apache this is generally false, as since all is done on mailing lists,
it's easy to see if things are being done or not.
What I still don't know is if we need a "chair" Mentor, but since I
really don't see ATM the real need for it, I would keep it simple and
deny a different status between Mentors.
In practice, it all boils down just to change our docs to have more than
one Mentor, and that all project members that abide to the Mentor
definition should become Mentors.
What do you think?
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Nicola Ken Barozzi <ni...@apache.org>.
Leo Simons wrote:
> Nicola Ken Barozzi wrote:
>
>> What do you think?
>
> +1 to redundancy. -0 on making it an 'absolute' requirement.
Of course.
1 Mentor is a requirement.
>1 Mentor is a *strong* suggestion.
> We need to make sure we ourselves don't get so completely
> lost in figuring out the right process that there is no energy
> left to apply it.
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Leo Simons <le...@apache.org>.
Nicola Ken Barozzi wrote:
> What do you think?
+1 to redundancy. -0 on making it an 'absolute' requirement.
We need to make sure we ourselves don't get so completely
lost in figuring out the right process that there is no energy
left to apply it.
cheers!
- Leo
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
New Status file template ( was Re: [RT] Multiple Mentors)
Posted by Nicola Ken Barozzi <ni...@apache.org>.
Stephen McConnell wrote:
...
> Keep in mind that I can see the use of JIRA for an incubation iniative
> as something seperate for a projects own use of JIRA for bug tracking.
Same here. Projects can use Jira to track their Incubation status and
issues.
...
> Ask the following question:
>
> * what are the top three things for FTP or Directory or Xxxx to move it
> towards exit
There are no top-rated issues in Incubating a project. All are relevant
and have to be addressed.
> I could spend an afternoon trying to put the answer together - and I'd
> be guessing alot along the way.
Just go in order. The first issues are the first to check (or at least
it should be this way).
> Irrespective of the mechanisms used -
> providing a way for incubating projects to push actions up to the
> incubation community is going address the concern you raised about
> sharing the load.
Having better and clearer status files is something we definitely need,
as someone that wants to help should really be eased in it. I believe
that a single file is a sufficient base, but it needs to be better than
what we have.
Let's start here:
http://nagoya.apache.org/wiki/apachewiki.cgi?NewProjectSTATUSTemplate
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Stephen McConnell <mc...@apache.org>.
Nicola Ken Barozzi wrote:
> Stephen McConnell wrote:
>
> ...
> > An important
>
>> aspect here is visibility of status and pending actions. This takes
>> me back to the discussion about using JIRA - that's the sort of
>> example where I could checkout open-actions on any incubator project
>> - and based on that information I could jump in and help out.
>
>
> Click on the project you want to help out:
> http://incubator.apache.org/projects/index.html
>
> Are these files not so nice? Sure, if someone crafts up nicer status
> files I'll be happy to switch to them.
>
> As for using an issue tracker for this, single projects are free to do
> so, but it's their decision; what I have linked to is the baseline.
Keep in mind that I can see the use of JIRA for an incubation iniative
as something seperate for a projects own use of JIRA for bug tracking.
An incubator view is all about actions facilitating exit (exit bugs). I
bet that if incubator projects could push up and maintain their top
three open actions - then more people could help out simply because they
can see what a project is asserting at the topics it needs incubation
help on.
Ask the following question:
* what are the top three things for FTP or Directory or Xxxx to move it
towards exit
I could spend an afternoon trying to put the answer together - and I'd
be guessing alot along the way. Irrespective of the mechanisms used -
providing a way for incubating projects to push actions up to the
incubation community is going address the concern you raised about
sharing the load.
Cheers, Stephen.
>
>
--
Stephen J. McConnell
mailto:mcconnell@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Nicola Ken Barozzi <ni...@apache.org>.
Stephen McConnell wrote:
...
> An important
> aspect here is visibility of status and pending actions. This takes me
> back to the discussion about using JIRA - that's the sort of example
> where I could checkout open-actions on any incubator project - and based
> on that information I could jump in and help out.
Click on the project you want to help out:
http://incubator.apache.org/projects/index.html
Are these files not so nice? Sure, if someone crafts up nicer status
files I'll be happy to switch to them.
As for using an issue tracker for this, single projects are free to do
so, but it's their decision; what I have linked to is the baseline.
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Stephen McConnell <mc...@apache.org>.
Nicola Ken Barozzi wrote:
> Berin Lautenbach wrote:
>
>> Nicola Ken Barozzi wrote:
>>
>>> Berin Lautenbach wrote:
>>>
>>>> Nicola,
>>>>
>>>> I suppose the only slight reservation would be who is accountable?
>>>> The old "Fred Bloggs is looking after that" can kick in. I think
>>>> the Incubator PMC also wants to be able to hold people accountable
>>>> for inubation activities. Gets harder with multiple people.
>>>
>>>
>>> If the accountable person fails, what can we do? Nothing.
>>
>>
>> Replace them!
>
>
> Too late, the bad is done. I want to minimize the possibility of it
> happening.
I understand you point - but I think you focussing on the wrong place.
The single point of accountability is IMO an important criteria -
instead - I think what you should be looking at is the framework within
which that reposibility becomes drop dead easy based on a framework of
support.
As a simple example - the process description establishes a sequence of
events with responsibilities on a project to move itself out of
incubation. The process will involve actions by the Mentors and the PMC
and we don't these actions to be slowing down the works. An important
aspect here is visibility of status and pending actions. This takes me
back to the discussion about using JIRA - that's the sort of example
where I could checkout open-actions on any incubator project - and based
on that information I could jump in and help out. That results is
greater community contribution to the process - less overhead oper
Mentor - etc. etc. I.e. think more about facilities to enable members
of the community to facilitate the exit of project under incubation.
Cheers, Steve.
--
Stephen J. McConnell
mailto:mcconnell@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Nicola Ken Barozzi <ni...@apache.org>.
Berin Lautenbach wrote:
> Nicola Ken Barozzi wrote:
>
>> Berin Lautenbach wrote:
>>
>>> Nicola,
>>>
>>> I suppose the only slight reservation would be who is accountable?
>>> The old "Fred Bloggs is looking after that" can kick in. I think the
>>> Incubator PMC also wants to be able to hold people accountable for
>>> inubation activities. Gets harder with multiple people.
>>
>> If the accountable person fails, what can we do? Nothing.
>
> Replace them!
Too late, the bad is done. I want to minimize the possibility of it
happening.
> That's what the board would do to a PMC chair for a PMC
> that was not delivering what the board requires.
Having the PMC ping Mentors regularly is another point that I will have
to get started.
>> Our goal is not to have one person accountable, as we are all
>> "accountable" as PMC members, but to have more eyes and hands working
>> on it that are also on our PMC.
>
> Yes and no. (My favourite kind of answer.)
:-)
> As far as the board is concerned (which is where the PMC reports) -
> there *is* an accountable person. The chair of the PMC. That doesn't
> mean that the entire PMC doesn't get involved in any given issue though.
> As I think Noel said in a later e-mail, this is all a community process.
>
> So - I have no issue at all with having multiple mentors. Completely
> the opposite - I think it's a good thing. I'm just not sure it's worth
> making it overly official.
There has already been some confusion whether multiple mentors are even
possible, so I'd like it to be specified.
Furthermore I'd also add that it's better if there is more than one
Mentor, and that it's recommended that all the ones that can become
mentors (given the definition) do so.
> A single "official" mentor with multiple helpers?
Not needed IMHO. From time to time there may be one person that stands
out and feels more "in charge", but IMHO it's not needed to be
formalized, exactly how project leads are not a formal Apache figure but
are nevertheless there.
> As I said in the original post - it's only a slight reservation. I've
> just always found that if there is some single person who has official
> responsibility for something, it's generally more likely to happen.
The current real results from actual incubations has shown us the
opposite, so I'd prefer that we try this route as it seems better based
on actual happenings.
If you want, go ahead and change our docs based on this, as I see that
there is no strong opposition from Incubator PMC members (rather the
opposite).
As for the fact of having more docs, the HOWTOs, and all that stuff, I
just dropped everything and will be happy to work with the existing
docs. :-)
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Berin Lautenbach <be...@ozemail.com.au>.
Nicola Ken Barozzi wrote:
> Berin Lautenbach wrote:
>
>> Nicola,
>>
>> I suppose the only slight reservation would be who is accountable?
>> The old "Fred Bloggs is looking after that" can kick in. I think the
>> Incubator PMC also wants to be able to hold people accountable for
>> inubation activities. Gets harder with multiple people.
>
>
> If the accountable person fails, what can we do? Nothing.
>
Replace them! That's what the board would do to a PMC chair for a PMC
that was not delivering what the board requires.
> Our goal is not to have one person accountable, as we are all
> "accountable" as PMC members, but to have more eyes and hands working on
> it that are also on our PMC.
>
Yes and no. (My favourite kind of answer.)
As far as the board is concerned (which is where the PMC reports) -
there *is* an accountable person. The chair of the PMC. That doesn't
mean that the entire PMC doesn't get involved in any given issue though.
As I think Noel said in a later e-mail, this is all a community process.
So - I have no issue at all with having multiple mentors. Completely
the opposite - I think it's a good thing. I'm just not sure it's worth
making it overly official. A single "official" mentor with multiple
helpers?
As I said in the original post - it's only a slight reservation. I've
just always found that if there is some single person who has official
responsibility for something, it's generally more likely to happen.
Cheers,
Berin
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: [RT] Multiple Mentors
Posted by "Noel J. Bergman" <no...@devtech.com>.
> > I think the accountable individual is an important criteria!
> > If the accountable person fails
No. We are a Community. We do not ALLOW an "accountable person" to fail.
We are supposed to be here on behalf of the ASF and on behalf of the
podling. It isn't a matter of an accountable person failing. If we see
things not getting done, we need to take some responsibility.
Yes, there will be too many projects for everyone to be everywhere. That is
why we need to make sure that the PMC is large enough. But we must
structure things so that there are enough eyes, and trust that if any ONE
person sees something, that they will make sure it gets addressed. That may
mean taking care of it, or it may mean making sure that it gets taking care
of, but either way, we are all responsible.
You may not be able to directly address something, but you can still raise
the issue with fellow PMC members, solicite advise and/or participation,
make sure that you get some, raise it to the Board if you don't, and get the
Committers involved.
--- Noel
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Nicola Ken Barozzi <ni...@apache.org>.
Stephen McConnell wrote:
> Nicola Ken Barozzi wrote:
...
>> If the accountable person fails, what can we do? Nothing.
>
> I think the accountable individual is an important criteria!
>
> If the accountable person fails
We have to make it so that he does not fail. This proposal is based on
the fact that in Apache, having more people makes failure less likely.
Think about OS in general: having tons of beta testers makes buggy
software less likely.
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Stephen McConnell <mc...@apache.org>.
Nicola Ken Barozzi wrote:
> Berin Lautenbach wrote:
>
>> Nicola,
>>
>> I suppose the only slight reservation would be who is accountable?
>> The old "Fred Bloggs is looking after that" can kick in. I think the
>> Incubator PMC also wants to be able to hold people accountable for
>> inubation activities. Gets harder with multiple people.
>
>
> If the accountable person fails, what can we do? Nothing.
I think the accountable individual is an important criteria!
If the accountable person fails - then the individual that represents
the responsible PMC is accountable for ensuring that an accountable
person is established (e.g. the individual is the Chair and the problem
is locating a new Mentor).
Cheers, Steve.
--
Stephen J. McConnell
mailto:mcconnell@apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Nicola Ken Barozzi <ni...@apache.org>.
Berin Lautenbach wrote:
> Nicola,
>
> I suppose the only slight reservation would be who is accountable? The
> old "Fred Bloggs is looking after that" can kick in. I think the
> Incubator PMC also wants to be able to hold people accountable for
> inubation activities. Gets harder with multiple people.
If the accountable person fails, what can we do? Nothing.
Our goal is not to have one person accountable, as we are all
"accountable" as PMC members, but to have more eyes and hands working on
it that are also on our PMC.
As for the idea that more people makes things less accountable, it's
vice versa at Apache. Put more committers on a project, and you will
have more that will eventually do.
All have to do the same thing, so none of them should assume that
someone else is looking over it. Things are done on mailing lists, and
if things do not progress, it's evident enough IMO.
> Does it have to be formally done? There can be a single "Mentor" with
> others involved and helping out.
It does, as these Mentors would all be on the Incubator PMC, and all
participate as Mentors in full.
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
RE: [RT] Multiple Mentors
Posted by "Noel J. Bergman" <no...@devtech.com>.
> who is accountable?
The PMC is accountable. There has been much expecting someone else to take
care of things, and too little of making sure that things get done.
Everyone who has seen something that needs to be done, and hasn't at least
made sure that it did, gets to share the problem. And everyone needs to
make themselves the solution. If things aren't getting done, contribute.
Push the issue. Make sure that the discussion happens. If you don't think
that you're the right person to DO a thing for some reason, still take some
responsibility for helping to make sure that the THING gets done.
Folks like you, Berin, generally do just that, so I am sure you understand.
> Gets harder with multiple people.
It only gets harder if people are waiting for "Fred Bloggs" to deal with it.
If everyone takes responsibility, it should not be a problem.
Sometimes it takes an entire Community to raise a project*. :-)
--- Noel
* American pop-culture reference.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Berin Lautenbach <be...@ozemail.com.au>.
Nicola,
I suppose the only slight reservation would be who is accountable? The
old "Fred Bloggs is looking after that" can kick in. I think the
Incubator PMC also wants to be able to hold people accountable for
inubation activities. Gets harder with multiple people.
Does it have to be formally done? There can be a single "Mentor" with
others involved and helping out.
Cheers,
Berin
Nicola Ken Barozzi wrote:
>
> From the current incubations, it's becoming quite clear IMO that a
> single Mentor is not enough for most, if not all, incubated projects.
>
> As PMCs are composed of more than one member, we need more Mentors or
> each project, so that there should be *at least* one of them active in
> every moment. In the projects where there are more than one Mentor
> already things have been going better in this regard, in fact.
>
> What has been said though is that if more than one have to do something,
> then nobody does it, as they assume others did. From what I see at
> Apache this is generally false, as since all is done on mailing lists,
> it's easy to see if things are being done or not.
>
> What I still don't know is if we need a "chair" Mentor, but since I
> really don't see ATM the real need for it, I would keep it simple and
> deny a different status between Mentors.
>
> In practice, it all boils down just to change our docs to have more than
> one Mentor, and that all project members that abide to the Mentor
> definition should become Mentors.
>
> What do you think?
>
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Nicola Ken Barozzi <ni...@apache.org>.
Geir Magnusson Jr. wrote:
> On Wednesday, November 12, 2003, at 05:46 AM, Nicola Ken Barozzi wrote:
>
...
>> In any case, the Mentors are on the PMC, so the PMC responsiveness
>> would increase with more Mentors that are active on the projects being
>> there.
>
> Am I on the incubator PMC? I'm now assisting Jim in mentoring Geronimo...
Not yet officially, I sent a mail about this on teh PMC list, I have
just now forwarded it to you.
In any case, sign up to the pmc list so you can start participating
there too.
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by "Geir Magnusson Jr." <ge...@optonline.net>.
On Wednesday, November 12, 2003, at 05:46 AM, Nicola Ken Barozzi wrote:
> Jochen Wiedmann wrote:
>
>> Nicola Ken Barozzi wrote:
>>> From the current incubations, it's becoming quite clear IMO that a
>>> single Mentor is not enough for most, if not all, incubated >>> projects.
>> As far as JaxMe is concerned, we had so far more problems with
>> missing replies on general@incubator or pmc@incubator.
>
> The pmc issues have partially been addressed, and other changes are on
> the way. Now we have more real active committers and things are
> changing.
>
> In any case, the Mentors are on teh PMC, so the PMC responsiveness
> would increase with more Mentors that are active on the projects being
> there.
Am I on the incubator PMC? I'm now assisting Jim in mentoring
Geronimo...
>
>> Dims, our mentor, is always replying.
>
> Excellent :-)
>
> I don't want to imply that Mentors are not doing their work, just that
> more of them would ease the burden.
>
> --
> Nicola Ken Barozzi nicolaken@apache.org
> - verba volant, scripta manent -
> (discussions get forgotten, just code remains)
> ---------------------------------------------------------------------
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>
--
Geir Magnusson Jr 203-247-1713(m)
geirm@optonline.net
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Nicola Ken Barozzi <ni...@apache.org>.
Jochen Wiedmann wrote:
> Nicola Ken Barozzi wrote:
>
>> From the current incubations, it's becoming quite clear IMO that a
>> single Mentor is not enough for most, if not all, incubated projects.
>
> As far as JaxMe is concerned, we had so far more problems with missing
> replies on general@incubator or pmc@incubator.
The pmc issues have partially been addressed, and other changes are on
the way. Now we have more real active committers and things are changing.
In any case, the Mentors are on teh PMC, so the PMC responsiveness would
increase with more Mentors that are active on the projects being there.
> Dims, our mentor, is always replying.
Excellent :-)
I don't want to imply that Mentors are not doing their work, just that
more of them would ease the burden.
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [RT] Multiple Mentors
Posted by Jochen Wiedmann <jo...@ispsoft.de>.
Nicola Ken Barozzi wrote:
> From the current incubations, it's becoming quite clear IMO that a
> single Mentor is not enough for most, if not all, incubated projects.
As far as JaxMe is concerned, we had so far more problems with missing
replies on general@incubator or pmc@incubator. Dims, our mentor, is always
replying.
Jochen
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org