Encouraging participation

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Encouraging participation

Rob Weir
Renaming the thread, since the topic has shifted greatly.

On Thu, Nov 1, 2012 at 4:21 PM, jan iversen <[hidden email]> wrote:

> Hi Dave.
>
> Even though I have stopped my companies, I still have many other things to
> do than working on AOO, and when I had my companies I had limited time, so
> I can for sure follow you. Today I am just trying to help open source as
> such, because it has helped me a lot in my career.
>
> And to answer your question, yes I do have some ideas (but they might be
> wrong), I have listed some of the important ones below:
> - We need to focus more on people who want to help, instead of using all
> the legal stuff (which are necessary) as a buffer not to move things. (e.g.
> I got 2 volunteers working on a danish translation, highly motivated, now
> we are discussing details about how to release the stuff). I think Rob is
> having a lead here with his new web pages.

I don't think anyone is using "legal stuff' to prevent things from
moving forward.  At least I hope not.  But there is some minimal
amount of process that needs to occur before  release is published.
This is done Apache-wide.  To the extent building release candidates
is a semi-manual task this will be painful and take time.  The
solution (or part of it) is to more fully automate.

But it is certainly true that a new volunteer is encouraged the best
when they can contribute today and see their results released
tomorrow.  The closer we approximate that kind of feedback loop the
better it is for volunteers.

On the other hand, we know that creating a release requires a
significant amount of build effort, review work (licenses, etc.), QA,
voting on a release, website update, announcement preparation, etc.
It really is currently a 4-6 week cycle of activities just to release
a new version.  Until we automate most of this, it is rather painful
to release.

It might help if we explain why the work, beyond coding, that goes
into a release is important, how it benefits our users, etc.  This
isn't just a formality without purpose.

> - We do NOT want a war of religions between AOO and others, ASF is well
> known, upper end of free software, so we should be publicly asking for
> collaboration.

Agreed.  But sometimes we find ourselves in a situation that we did
not create nor ask for,

> - I think events like ApacheCon is nice, but events like FOSDEM is quite a
> lot more important for the "ordinary" openSource developer.

And we are planning a dev room at Fosdem for that reason.

> - I would like to see more "marketing" for developers, instead of
> businesses...I think we need to get back to roots where a developers think
> its fun, and pride to develop AOO. We could easily e.g. make challenges
> like "who can solve this problem".
>

Good idea.  But it does presuppose a more intelligible build system as
well.  For new developers to enjoy working in this code base we need
to get to a point where they get enjoyment from focusing on problems
that are relevant to them.  Focusing on the build system is not
relevant.  It is friction, effort wasted, not accomplishment   So we
can move this forward with documentation as well.

> I am new to AOO (so I am either interfering or bringing in new views), but
> I have quite some years of experience with openSource and I am a strong
> believer of ASF. The "apache way" is in many ways a limitation, but at the
> end it is the guarantee for a better end-user product.
>

"When forced to work within a strict framework the imagination is
taxed to its utmost — and will produce its richest ideas. Given total
freedom the work is likely to sprawl." – T.S. Eliot

> Please accept my apologies, if I have broken n-policies, but I think the
> question from Dave was well placed, and well formulated so it deserved a
> straight answer.
>
> Jan.
>
>
>
>
>
> On 1 November 2012 20:51, Dave Fisher <[hidden email]> wrote:
>
>> Hi Jan,
>>
>> We are all here as individuals with various and different amounts of time
>> and energy. Many are employed to work on OpenOffice, but many like me are
>> volunteers who have demanding day jobs. The key part of the Apache Way is
>> that leadership comes from DOING and COMMUNICATING.
>>
>> You are new here with lots of admirable energy and work! This is what
>> acquires merit in an Apache project!
>>
>> Since we ultimately can only control ourselves, do you have any
>> suggestions about how we can more actively encourage participation?
>>
>> Best Regards,
>> Dave
>>
>> On Nov 1, 2012, at 9:38 AM, jan iversen wrote:
>>
>> > Please excuse me, I think I know the difference between hooligans and
>> > people who are just blowing hot air.
>> >
>> > To be honest, at the moment AOO does NOT have a great deal of momentum,
>> and
>> > have (I think) lost a quite a lot of reputation among developers. That is
>> > something we have to remedy, not by glittering folders, or smart
>> marketing,
>> > but by showing the developers, that we really care about their
>> > contributions.
>> >
>> > If I may say so, some developers might see "the apache way" as a
>> > limitation, which my experience during the last month somewhat confirms,
>> I
>> > think we really need to focus on "the community" instead of telling
>> people
>> > about legal issues, but about getting a product that still can out beat
>> the
>> > big (costly) products out there. Do NOT forget some state institutions in
>> > EU choose OpenOffice against other, but today I would not be so sure !!!
>> >
>> > Sorry for the outburst, but I am used to say what I think, and I really
>> > really want AOO to be the opensource project, as it was in the past. Lets
>> > not forget why we are all here.....
>> >
>> > Jan
>> >
>> > On 1 November 2012 17:20, RGB ES <[hidden email]> wrote:
>> >
>> >> 2012/11/1 Rob Weir <[hidden email]>
>> >>
>> >>> I'm hearing that some project volunteers, especially new ones, are
>> >>> being contacted by certain external parties, who then try to
>> >>> discourage them from contributing to the Apache OpenOffice project.
>> >>> I'm hearing that similar notes have been sent out to those who
>> >>> submitted listings to our new Consultants Directory, also discouraging
>> >>> them from involvement in the project.
>> >>>
>> >>> This is my personal view on this matter, for what it is worth.
>> >>>
>> >>> I think we all would agree that such techniques are deplorable and
>> >>> bring disrepute to the individuals involved, and to the project that
>> >>> sanctions such techniques.  If you recall we had a similar wave of
>> >>> such unprofessional behavior a few months ago, when certain external
>> >>> parties were contacting journalists who mentioned OpenOffice and
>> >>> telling them that it was no longer being developed and to link to a
>> >>> different product instead.
>> >>>
>> >>> I any case, if you are receiving such FUD yourself, I'd encourage you
>> >>> to simply post it to this mailing list, or to your blog, or some other
>> >>> public website.  "Daylight is the best antiseptic" as they say.  I am
>> >>> not a medical doctor, but I do believe that FUD exposed to public
>> >>> scrutiny loses its potency.   But FUD ignored is FUD that spreads.
>> >>>
>> >>
>> >> There is and always will be people who do not understand what an
>> opensource
>> >> project is and behave like hooligans "defending" their soccer team. I
>> hope
>> >> they are just individuals and nothing more, but I fully agree to put
>> each
>> >> case under daylight.
>> >>
>> >> Regards
>> >> Ricardo
>> >>
>> >>
>> >>
>> >>>
>> >>>
>> >>> Regards,
>> >>>
>> >>> -Rob
>> >>>
>> >>
>>
>>
Reply | Threaded
Open this post in threaded view
|

Re: Encouraging participation

Andrea Pescetti-2
On 01/11/2012 Rob Weir wrote:
> On Thu, Nov 1, 2012 at 4:21 PM, jan iversen wrote:
>> - We need to focus more on people who want to help, instead of using all
>> the legal stuff (which are necessary) as a buffer not to move things. (e.g.
>> I got 2 volunteers working on a danish translation, highly motivated, now
>> we are discussing details about how to release the stuff).  ...
> I don't think anyone is using "legal stuff' to prevent things from
> moving forward.

There is a bit of confusion here. One thing is allowing volunteers to
have feedback on their work, the other one is releasing their work. For
feedback we needn't focus on legal issues. So the Danish translation as
discussed in
https://issues.apache.org/ooo/show_bug.cgi?id=121179
will be integrated in any next 3.4.x (informal, i.e., "snapshots")
builds. The "legal stuff" is not playing any roles here.

> But it is certainly true that a new volunteer is encouraged the best
> when they can contribute today and see their results released
> tomorrow.

I'd focus on "used" rather than "released": it is more motivating to see
their results used (i.e., a snapshot build) soon than to see them
released after months. And this is where we should improve. To give
volunteers feedback we only need a very lightweight process, ideally zero.

What is delaying us with the current translations, for example, is just
that we need to determine a suitable deadline for translators to check
in their PO files, integrating them on
http://svn.apache.org/viewvc/incubator/ooo/branches/AOO34/ and building
snapshot for AOO34. At the moment this is indeed quite demanding on
Juergen and Ariel.

>> - I think events like ApacheCon is nice, but events like FOSDEM is quite a
>> lot more important for the "ordinary" openSource developer.
> And we are planning a dev room at Fosdem for that reason.

By FOSDEM (and ideally much earlier) we must be ready to integrate new
volunteers in a way that fully satisfies them and the project. This is a
priority for OpenOffice as a project.

We are getting close to this for what concerns localization: I expect
that in a couple weeks we will be able to involve, engage and satisfy
localization volunteers with an established process. We must then do the
same for QA, development, Marketing...

An important result we should achieve is that nobody should feel
frustrated by not having committer privileges: it is also up to us to
define tasks that can be done without depending too much on a committer
helping the contributor. At least we should warn them: if someone wants
to rebuild an entire section of the OpenOffice website, like it is
happening with Jan, he should be told in advance that this contribution
is really welcome (and that, for most sections, we really need it!) but
that at a certain point he might feel frustration for not being a
committer. There are hundreds of tasks that can be done by
non-committers, and we should keep the distinction clear when we
advertise tasks for volunteers. (That said, the "privileges" of being a
committer or a PMC member are greatly exaggerated at times... it's not
that much really; but when this is the only obstacle to getting things
really done, I can understand the impatience).

Regards,
   Andrea.
Reply | Threaded
Open this post in threaded view
|

Re: Encouraging participation

Ariel Constenla-Haile-2
On Fri, Nov 02, 2012 at 05:54:56PM +0100, Andrea Pescetti wrote:
> What is delaying us with the current translations, for example, is
> just that we need to determine a suitable deadline for translators
> to check in their PO files, integrating them on
> http://svn.apache.org/viewvc/incubator/ooo/branches/AOO34/ and
> building snapshot for AOO34. At the moment this is indeed quite
> demanding on Juergen and Ariel.

Note that building is not that demanding: as long as there is no build
breaker, it's simply running a script, because the local build
environment has already been tested on previous builds. The same goes
for uploading packages, a simple scp command.


Regards
--
Ariel Constenla-Haile
La Plata, Argentina

attachment0 (853 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Encouraging participation

jan iversen
In reply to this post by Andrea Pescetti-2
On 2 November 2012 17:54, Andrea Pescetti <[hidden email]> wrote:

> On 01/11/2012 Rob Weir wrote:
>
>> On Thu, Nov 1, 2012 at 4:21 PM, jan iversen wrote:
>>
>>> - We need to focus more on people who want to help, instead of using all
>>> the legal stuff (which are necessary) as a buffer not to move things.
>>> (e.g.
>>> I got 2 volunteers working on a danish translation, highly motivated, now
>>> we are discussing details about how to release the stuff).  ...
>>>
>> I don't think anyone is using "legal stuff' to prevent things from
>> moving forward.
>>
>
> There is a bit of confusion here. One thing is allowing volunteers to have
> feedback on their work, the other one is releasing their work. For feedback
> we needn't focus on legal issues. So the Danish translation as discussed in
> https://issues.apache.org/ooo/**show_bug.cgi?id=121179<https://issues.apache.org/ooo/show_bug.cgi?id=121179>
> will be integrated in any next 3.4.x (informal, i.e., "snapshots") builds.
> The "legal stuff" is not playing any roles here.
>
>
>  But it is certainly true that a new volunteer is encouraged the best
>> when they can contribute today and see their results released
>> tomorrow.
>>
>
> I'd focus on "used" rather than "released": it is more motivating to see
> their results used (i.e., a snapshot build) soon than to see them released
> after months. And this is where we should improve. To give volunteers
> feedback we only need a very lightweight process, ideally zero.
>
> What is delaying us with the current translations, for example, is just
> that we need to determine a suitable deadline for translators to check in
> their PO files, integrating them on http://svn.apache.org/viewvc/**
> incubator/ooo/branches/AOO34/<http://svn.apache.org/viewvc/incubator/ooo/branches/AOO34/>and building snapshot for AOO34. At the moment this is indeed quite
> demanding on Juergen and Ariel.
>
>
>  - I think events like ApacheCon is nice, but events like FOSDEM is quite a
>>> lot more important for the "ordinary" openSource developer.
>>>
>> And we are planning a dev room at Fosdem for that reason.
>>
>
> By FOSDEM (and ideally much earlier) we must be ready to integrate new
> volunteers in a way that fully satisfies them and the project. This is a
> priority for OpenOffice as a project.
>
> We are getting close to this for what concerns localization: I expect that
> in a couple weeks we will be able to involve, engage and satisfy
> localization volunteers with an established process. We must then do the
> same for QA, development, Marketing...
>
> An important result we should achieve is that nobody should feel
> frustrated by not having committer privileges: it is also up to us to
> define tasks that can be done without depending too much on a committer
> helping the contributor. At least we should warn them: if someone wants to
> rebuild an entire section of the OpenOffice website, like it is happening
> with Jan, he should be told in advance that this contribution is really
> welcome (and that, for most sections, we really need it!) but that at a
> certain point he might feel frustration for not being a committer. There
> are hundreds of tasks that can be done by non-committers, and we should
> keep the distinction clear when we advertise tasks for volunteers. (That
> said, the "privileges" of being a committer or a PMC member are greatly
> exaggerated at times... it's not that much really; but when this is the
> only obstacle to getting things really done, I can understand the
> impatience).
>

I think I got ample warning ahead of doing the rewrite of l10n, what
surprised was the discussion going on right now, that is quite frustrating,
especially because I opened the theme before I did the work, and nobody
complained, on the contrary many said "yes please do".

If you things like I do it can be quite frustrating not to have committer
status, not at all for the privilege, but because I have to waster a
committers valuable time, doing simple jobs. So the sentence "it's not that
much really", is not quite correct, it can be quite time saving.


> Regards,
>   Andrea.
>
Reply | Threaded
Open this post in threaded view
|

Re: Encouraging participation

Dave Fisher

On Nov 2, 2012, at 2:12 PM, jan iversen wrote:

> On 2 November 2012 17:54, Andrea Pescetti <[hidden email]> wrote:
>
>> On 01/11/2012 Rob Weir wrote:
>>
>>> On Thu, Nov 1, 2012 at 4:21 PM, jan iversen wrote:
>>>
>>>> - We need to focus more on people who want to help, instead of using all
>>>> the legal stuff (which are necessary) as a buffer not to move things.
>>>> (e.g.
>>>> I got 2 volunteers working on a danish translation, highly motivated, now
>>>> we are discussing details about how to release the stuff).  ...
>>>>
>>> I don't think anyone is using "legal stuff' to prevent things from
>>> moving forward.
>>>
>>
>> There is a bit of confusion here. One thing is allowing volunteers to have
>> feedback on their work, the other one is releasing their work. For feedback
>> we needn't focus on legal issues. So the Danish translation as discussed in
>> https://issues.apache.org/ooo/**show_bug.cgi?id=121179<https://issues.apache.org/ooo/show_bug.cgi?id=121179>
>> will be integrated in any next 3.4.x (informal, i.e., "snapshots") builds.
>> The "legal stuff" is not playing any roles here.
>>
>>
>> But it is certainly true that a new volunteer is encouraged the best
>>> when they can contribute today and see their results released
>>> tomorrow.
>>>
>>
>> I'd focus on "used" rather than "released": it is more motivating to see
>> their results used (i.e., a snapshot build) soon than to see them released
>> after months. And this is where we should improve. To give volunteers
>> feedback we only need a very lightweight process, ideally zero.
>>
>> What is delaying us with the current translations, for example, is just
>> that we need to determine a suitable deadline for translators to check in
>> their PO files, integrating them on http://svn.apache.org/viewvc/**
>> incubator/ooo/branches/AOO34/<http://svn.apache.org/viewvc/incubator/ooo/branches/AOO34/>and building snapshot for AOO34. At the moment this is indeed quite
>> demanding on Juergen and Ariel.
>>
>>
>> - I think events like ApacheCon is nice, but events like FOSDEM is quite a
>>>> lot more important for the "ordinary" openSource developer.
>>>>
>>> And we are planning a dev room at Fosdem for that reason.
>>>
>>
>> By FOSDEM (and ideally much earlier) we must be ready to integrate new
>> volunteers in a way that fully satisfies them and the project. This is a
>> priority for OpenOffice as a project.
>>
>> We are getting close to this for what concerns localization: I expect that
>> in a couple weeks we will be able to involve, engage and satisfy
>> localization volunteers with an established process. We must then do the
>> same for QA, development, Marketing...
>>
>> An important result we should achieve is that nobody should feel
>> frustrated by not having committer privileges: it is also up to us to
>> define tasks that can be done without depending too much on a committer
>> helping the contributor. At least we should warn them: if someone wants to
>> rebuild an entire section of the OpenOffice website, like it is happening
>> with Jan, he should be told in advance that this contribution is really
>> welcome (and that, for most sections, we really need it!) but that at a
>> certain point he might feel frustration for not being a committer. There
>> are hundreds of tasks that can be done by non-committers, and we should
>> keep the distinction clear when we advertise tasks for volunteers. (That
>> said, the "privileges" of being a committer or a PMC member are greatly
>> exaggerated at times... it's not that much really; but when this is the
>> only obstacle to getting things really done, I can understand the
>> impatience).
>>
>
> I think I got ample warning ahead of doing the rewrite of l10n, what
> surprised was the discussion going on right now, that is quite frustrating,
> especially because I opened the theme before I did the work, and nobody
> complained, on the contrary many said "yes please do".
>
> If you things like I do it can be quite frustrating not to have committer
> status, not at all for the privilege, but because I have to waster a
> committers valuable time, doing simple jobs.

You are not wasting a committers valuable time. The committer's time is spent evaluating your contribution. When the committer(s) begin to feel that their time is beginning to be wasted that is the point they ought to suggest to the PMC that it is time DISCUSS giving the individual committers rights. This discussion occurs in private, the discussion is then followed by a private VOTE that lasts at least 3 days. EIther or both of these processes can be public on the dev list.

If the community thinks that a private DISCUSS followed by a public VOTE would encourage contributors then I would be for changing the policy. Perhaps to the following.

(1) Private DISCUSS on potential new Committers lasting at least 72 hours until there is clear CONSENSUS.
(2) Chair contacts the contributor to make sure they are interested and to let them know a VOTE is coming.
(3) Potential Committer confirms interest to private.
(4) VOTE is started on dev @ oo.a.o. Lasts 72 hours.
(5) ICLA, Account creation and Karma grant occur as now.

Regards,
Dave

> So the sentence "it's not that
> much really", is not quite correct, it can be quite time saving.
>
>
>> Regards,
>>  Andrea.
>>

Reply | Threaded
Open this post in threaded view
|

Re: Encouraging participation

jan iversen
On 2 November 2012 22:31, Dave Fisher <[hidden email]> wrote:

>
> On Nov 2, 2012, at 2:12 PM, jan iversen wrote:
>
> > On 2 November 2012 17:54, Andrea Pescetti <[hidden email]> wrote:
> >
> >> On 01/11/2012 Rob Weir wrote:
> >>
> >>> On Thu, Nov 1, 2012 at 4:21 PM, jan iversen wrote:
> >>>
> >>>> - We need to focus more on people who want to help, instead of using
> all
> >>>> the legal stuff (which are necessary) as a buffer not to move things.
> >>>> (e.g.
> >>>> I got 2 volunteers working on a danish translation, highly motivated,
> now
> >>>> we are discussing details about how to release the stuff).  ...
> >>>>
> >>> I don't think anyone is using "legal stuff' to prevent things from
> >>> moving forward.
> >>>
> >>
> >> There is a bit of confusion here. One thing is allowing volunteers to
> have
> >> feedback on their work, the other one is releasing their work. For
> feedback
> >> we needn't focus on legal issues. So the Danish translation as
> discussed in
> >> https://issues.apache.org/ooo/**show_bug.cgi?id=121179<
> https://issues.apache.org/ooo/show_bug.cgi?id=121179>
> >> will be integrated in any next 3.4.x (informal, i.e., "snapshots")
> builds.
> >> The "legal stuff" is not playing any roles here.
> >>
> >>
> >> But it is certainly true that a new volunteer is encouraged the best
> >>> when they can contribute today and see their results released
> >>> tomorrow.
> >>>
> >>
> >> I'd focus on "used" rather than "released": it is more motivating to see
> >> their results used (i.e., a snapshot build) soon than to see them
> released
> >> after months. And this is where we should improve. To give volunteers
> >> feedback we only need a very lightweight process, ideally zero.
> >>
> >> What is delaying us with the current translations, for example, is just
> >> that we need to determine a suitable deadline for translators to check
> in
> >> their PO files, integrating them on http://svn.apache.org/viewvc/**
> >> incubator/ooo/branches/AOO34/<
> http://svn.apache.org/viewvc/incubator/ooo/branches/AOO34/>and building
> snapshot for AOO34. At the moment this is indeed quite
> >> demanding on Juergen and Ariel.
> >>
> >>
> >> - I think events like ApacheCon is nice, but events like FOSDEM is
> quite a
> >>>> lot more important for the "ordinary" openSource developer.
> >>>>
> >>> And we are planning a dev room at Fosdem for that reason.
> >>>
> >>
> >> By FOSDEM (and ideally much earlier) we must be ready to integrate new
> >> volunteers in a way that fully satisfies them and the project. This is a
> >> priority for OpenOffice as a project.
> >>
> >> We are getting close to this for what concerns localization: I expect
> that
> >> in a couple weeks we will be able to involve, engage and satisfy
> >> localization volunteers with an established process. We must then do the
> >> same for QA, development, Marketing...
> >>
> >> An important result we should achieve is that nobody should feel
> >> frustrated by not having committer privileges: it is also up to us to
> >> define tasks that can be done without depending too much on a committer
> >> helping the contributor. At least we should warn them: if someone wants
> to
> >> rebuild an entire section of the OpenOffice website, like it is
> happening
> >> with Jan, he should be told in advance that this contribution is really
> >> welcome (and that, for most sections, we really need it!) but that at a
> >> certain point he might feel frustration for not being a committer. There
> >> are hundreds of tasks that can be done by non-committers, and we should
> >> keep the distinction clear when we advertise tasks for volunteers. (That
> >> said, the "privileges" of being a committer or a PMC member are greatly
> >> exaggerated at times... it's not that much really; but when this is the
> >> only obstacle to getting things really done, I can understand the
> >> impatience).
> >>
> >
> > I think I got ample warning ahead of doing the rewrite of l10n, what
> > surprised was the discussion going on right now, that is quite
> frustrating,
> > especially because I opened the theme before I did the work, and nobody
> > complained, on the contrary many said "yes please do".
> >
> > If you things like I do it can be quite frustrating not to have committer
> > status, not at all for the privilege, but because I have to waster a
> > committers valuable time, doing simple jobs.
>
> You are not wasting a committers valuable time. The committer's time is
> spent evaluating your contribution. When the committer(s) begin to feel
> that their time is beginning to be wasted that is the point they ought to
> suggest to the PMC that it is time DISCUSS giving the individual committers
> rights. This discussion occurs in private, the discussion is then followed
> by a private VOTE that lasts at least 3 days. EIther or both of these
> processes can be public on the dev list.
>

I think I formulated myself badly, there is a process for being invited to
be committer and I have NO opinion on that process, except it sounds
reasonable to me !!

The part about time waste (regarding the  l10n website), is currently a
discussion on l10n, so we should not also discuss it here.


> If the community thinks that a private DISCUSS followed by a public VOTE
> would encourage contributors then I would be for changing the policy.
> Perhaps to the following.
>
> (1) Private DISCUSS on potential new Committers lasting at least 72 hours
> until there is clear CONSENSUS.
> (2) Chair contacts the contributor to make sure they are interested and to
> let them know a VOTE is coming.
> (3) Potential Committer confirms interest to private.
> (4) VOTE is started on dev @ oo.a.o. Lasts 72 hours.
> (5) ICLA, Account creation and Karma grant occur as now.
>
> Regards,
> Dave
>
> > So the sentence "it's not that
> > much really", is not quite correct, it can be quite time saving.
> >
> >
> >> Regards,
> >>  Andrea.
> >>
>
>