You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@community.apache.org by Rich Bowen <rb...@rcbowen.com> on 2016/11/16 09:41:59 UTC
Does GSoC help develop communities?
It would be great to have some kind of statistics on how GSoC helps
projects longer term. Do students stick around? Does the code written
actually get incorporated into releases? Does it in fact contribute to
the mission of Community Development, or is it just a nice summer job
for these students?
To be clear, I don't question the value of GSoC. What I'm curious about
is whether these people actually end up participating in our
communities. My experience with GSoC outside of Apache has been that the
students don't stick around, and so the enormous time and effort put
into GSoC, while it helps the skills and career of these students,
didn't help the project longer term.
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
So, since this thread seems to have been perceived as "justify your
existence, you layabouts!" let me rephrase, in the hopes of being a
little more constructive. Apologies to any who took my question as an
attack or accusation. It was neither.
Do we have a comprehensive list of project/particpant going back to when
we started doing GSoC? Or can someone point me to a location where such
a list could be assembled after the fact?
I propose the following concrete tasks/action items. Folks who don't see
the value of doing them are encouraged not to do them. Folks who are
looking for a way to contribute to the mission of Community Development
can consider doing these tasks.
1) Assemble list of participant/project (privately, if there's a privacy
concern here?) across the history of GSoC@Apache
2) Foreach $participant in list, is that participant still active in
$project. (Subtask: State your definition of "active".)
3) Foreach $participant in list, determine whether the code that they
wrote as the C part of GSoC was in fact accepted into $project.
(Subtask: Is it still there?)
4) Present items 2 and 3 above as pretty graphs and charts, showing
outcomes of each year, and so on. Include these graphs and charts on
http://community.apache.org/gsoc and in whatever materials we send to
projects when we encourage them to participate in GSoC.
5) Also offer these statistics back to Google. This is both to help them
in their recruitment (assuming it's good news) and also as a proactive
part of our own community development effort, to draw in the next
generation of GSoC participants, and participants in general.
On 11/16/2016 04:41 AM, Rich Bowen wrote:
> It would be great to have some kind of statistics on how GSoC helps
> projects longer term. Do students stick around? Does the code written
> actually get incorporated into releases? Does it in fact contribute to
> the mission of Community Development, or is it just a nice summer job
> for these students?
>
> To be clear, I don't question the value of GSoC. What I'm curious about
> is whether these people actually end up participating in our
> communities. My experience with GSoC outside of Apache has been that the
> students don't stick around, and so the enormous time and effort put
> into GSoC, while it helps the skills and career of these students,
> didn't help the project longer term.
>
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Stian Soiland-Reyes <st...@apache.org>.
Yes, that was the very case for us in Taverna, where two students got
engaged because of GSOC advertising, even if they didn't participate in the
GSOC programme (we lacked mentor capacity), so they co-mentored each other
instead. Both are now committers.
Related SSI blog post, "Downloading developers":
https://www.software.ac.uk/blog/2016-10-12-downloading-developers-google-summer-code
On 18 Nov 2016 4:58 pm, "Buddhika Jayawardhana" <bu...@cse.mrt.ac.lk>
wrote:
> Also there may be students who have who actually did not take in part in
> GSOC for Apache, but got involved with Apache because of GSOC ,and still
> stick around.
>
> On 16 November 2016 at 15:48, Nick Burch <ni...@apache.org> wrote:
>
> > On Wed, 16 Nov 2016, Rich Bowen wrote:
> >
> >> It would be great to have some kind of statistics on how GSoC helps
> >> projects longer term. Do students stick around? Does the code written
> >> actually get incorporated into releases? Does it in fact contribute to
> the
> >> mission of Community Development, or is it just a nice summer job for
> these
> >> students?
> >>
> >
> > Do we have a list (maybe somewhere in the comdev private svn?) of
> everyone
> > who has taken part in GSoC?
> >
> > If so, it'd be fairly easy to annotate that with apache IDs, then see
> > who's now on PMCs or who's now a member. Producing sharable statistics
> from
> > that automatically is then easy. (I say this as someone who helped update
> > the similar Travel Assistance Committee / TAC ones on Monday!)
> >
> > Not sure if it's easy to find out the "last commit date" for people, to
> > check for the "still around" part (eg for people who got committership
> > during GSoc, maybe on a branch), but I know where infra are to ask...
> >
> > Nick
> >
> > PS FWIW, within Tika we've certainly had a few GSoC people stick around!
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> > For additional commands, e-mail: dev-help@community.apache.org
> >
> >
>
>
> --
> Best Regards
>
> *Buddhika Jayawardhana*
> Undergraduate | Department of Computer Science & Engineering
> University of Moratuwa
> *buddhika.12@cse.mrt.ac.lk <bu...@cse.mrt.ac.lk>* | LinkedIn
> <http://lk.linkedin.com/in/buddhikajay/>
>
Re: Does GSoC help develop communities?
Posted by Buddhika Jayawardhana <bu...@cse.mrt.ac.lk>.
Also there may be students who have who actually did not take in part in
GSOC for Apache, but got involved with Apache because of GSOC ,and still
stick around.
On 16 November 2016 at 15:48, Nick Burch <ni...@apache.org> wrote:
> On Wed, 16 Nov 2016, Rich Bowen wrote:
>
>> It would be great to have some kind of statistics on how GSoC helps
>> projects longer term. Do students stick around? Does the code written
>> actually get incorporated into releases? Does it in fact contribute to the
>> mission of Community Development, or is it just a nice summer job for these
>> students?
>>
>
> Do we have a list (maybe somewhere in the comdev private svn?) of everyone
> who has taken part in GSoC?
>
> If so, it'd be fairly easy to annotate that with apache IDs, then see
> who's now on PMCs or who's now a member. Producing sharable statistics from
> that automatically is then easy. (I say this as someone who helped update
> the similar Travel Assistance Committee / TAC ones on Monday!)
>
> Not sure if it's easy to find out the "last commit date" for people, to
> check for the "still around" part (eg for people who got committership
> during GSoc, maybe on a branch), but I know where infra are to ask...
>
> Nick
>
> PS FWIW, within Tika we've certainly had a few GSoC people stick around!
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
>
--
Best Regards
*Buddhika Jayawardhana*
Undergraduate | Department of Computer Science & Engineering
University of Moratuwa
*buddhika.12@cse.mrt.ac.lk <bu...@cse.mrt.ac.lk>* | LinkedIn
<http://lk.linkedin.com/in/buddhikajay/>
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
Ok, skip it then. However, I'm concerned that what we don't measure, we
don't know how to improve.
On Dec 6, 2016 3:36 AM, "Ulrich Stärk" <ul...@spielviel.de> wrote:
> On 05.12.16 22:24, Rich Bowen wrote:
> > So ... stepping back a bit here. Are you saying that even attempting to
> > measure outcomes is harmful, because we might draw the wrong conclusions?
> >
> > This is a completely new notion to me.
> >
> > You say:
> >
> > "I am reluctant to simply collect some data because I am missing a clear
> > question what we are trying to answer and how the data you want to
> > collect is actually answering that question."
> >
> > I do have a clear question. My question is whether participating in the
> > GSoC contributes to our goal of Community Development. Does it develop
> > our community in any measurable way? I assume, at this point, that it
> > does, or we wouldn't keep doing it. In what way has it had measurable
> > impact on our community?
> >
> > Things that we can clearly measure is adding participants in our project
> > communities, and artifacts added to our projects' revision control
> > repositories. Measuring other things is more difficult, and can be taken
> > as a later task. Easy things first.
>
> And here is what I'm afraid of: By focusing on those two simple dimensions
> only we will from my
> experience end up with a pretty skewed result. The dimensions you propose
> don't cover code living on
> outside our repositories as add-on modules, they don't cover indirect
> contributions, etc.
>
> >
> > The larger question of how and what we measure in the Community
> > Development PMC as a whole hinges on this, too. I'm curious if you're
> > resistant to that, too?
>
> I am resistant to what you are proposing to measure. I am resistant to
> simplifying metrics. Those
> tend to do more harm than good.
>
> Uli
>
> >
> > --Rich
> >
> >
> >
> > On 12/05/2016 04:00 PM, Ulrich Stärk wrote:
> >> On 05.12.16 17:54, Rich Bowen wrote:
> >>>
> >>>
> >>> On 12/05/2016 07:41 AM, Ulrich Stärk wrote:
> >>>>> Or, at the very least, can we make a commitment to track this data
> going
> >>>>>> forward?
> >>>> Let me play the devil's advocate here: What for?
> >>>>
> >>>> GSoC is completely free for the ASF (on the contrary, we even get a
> small amount for every accepted
> >>>> student that we can than put towards fulfilling our goals) and as
> long as we have volunteers willing
> >>>> to organize it and mentor students we can assume that at least those
> volunteers are seeing value in
> >>>> it. Why the stats other than for satisfying our curiosity?
> >>>
> >>> Or, perhaps, let me give a different answer.
> >>>
> >>> I participated in GSoC as a mentor for $WorkProject. While it didn't
> >>> "cost" me anything in dollars, it cost me probably 200 hours of my
> time.
> >>> I know that other projects at work put more time in, and some less.
> >>>
> >>> This is an enormous cost to me, as an employee. So it behooves me to
> >>> measure the benefit to the project. A student received payment to write
> >>> code that was discarded. And I (and several of my colleagues) spent a
> >>> huge amount of time, which I could have spent on other things,
> mentoring
> >>> that student. Benefit to project, pretty heavily negative.
> >>
> >> In that case you should have failed the student pretty early as I tell
> mentors every year and as
> >> documented in the mentor guide.
> >>
> >>>
> >>> So, now, here we are at the ASF, doing GSoC with our projects, and
> >>> promoting it to them as a benefit. Does it actually benefit them, or is
> >>> it merely siphoning off time that could be spent on other things.
> >>
> >> If a mentor feels it is the latter, they should immediately fail the
> student.
> >>
> >>>
> >>> To folks that say we can't measure that, I strongly disagree. There are
> >>> two measures that are obvious and easy.
> >>>
> >>> 1) What % of GSoC student are still active on the project 6 months, 12
> >>> months, 18 months after the project. (We can debate the definition of
> >>> "active" all you like.
> >>
> >> This implies that the program is only successful if a certain number of
> students sticks around. And
> >> this is exactly what I'm arguing against. IMO the program is successful
> as soon as a student has
> >> some exposure to one of our communities.
> >>
> >>>
> >>> 2) What % of code developed by GSoC students actually becomes a part of
> >>> the project codebase at the end of the project?
> >>
> >> This has the same problems as trying to measure productivity from code.
> What is the percentage
> >> telling us about the question we want to answer? Also see above: If the
> code cannot be part of the
> >> project codebase the student should be failed.
> >>
> >>>
> >>> I would maintain that #1 is part of our charter as ComDev, and #2 is
> >>> part of what projects should be made aware of before they sign on.
> >>>
> >>> Again, I'm really not asking for a lot of data here. But I do think
> that
> >>> it's part of a responsible accounting for participating in GSoC. If,
> for
> >>> example, it is actually hindering projects, don't we want to know that.
> >>>
> >>> I did *not* participate in GSoC again, on $WorkProject, because it
> >>> clearly hindered my project.
> >>>
> >>>
> >>
> >> I am reluctant to simply collect some data because I am missing a clear
> question what we are trying
> >> to answer and how the data you want to collect is actually answering
> that question.
> >>
> >> Cheers,
> >>
> >> Uli
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> >> For additional commands, e-mail: dev-help@community.apache.org
> >>
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
>
Re: Does GSoC help develop communities?
Posted by Ulrich Stärk <ul...@spielviel.de>.
On 05.12.16 22:24, Rich Bowen wrote:
> So ... stepping back a bit here. Are you saying that even attempting to
> measure outcomes is harmful, because we might draw the wrong conclusions?
>
> This is a completely new notion to me.
>
> You say:
>
> "I am reluctant to simply collect some data because I am missing a clear
> question what we are trying to answer and how the data you want to
> collect is actually answering that question."
>
> I do have a clear question. My question is whether participating in the
> GSoC contributes to our goal of Community Development. Does it develop
> our community in any measurable way? I assume, at this point, that it
> does, or we wouldn't keep doing it. In what way has it had measurable
> impact on our community?
>
> Things that we can clearly measure is adding participants in our project
> communities, and artifacts added to our projects' revision control
> repositories. Measuring other things is more difficult, and can be taken
> as a later task. Easy things first.
And here is what I'm afraid of: By focusing on those two simple dimensions only we will from my
experience end up with a pretty skewed result. The dimensions you propose don't cover code living on
outside our repositories as add-on modules, they don't cover indirect contributions, etc.
>
> The larger question of how and what we measure in the Community
> Development PMC as a whole hinges on this, too. I'm curious if you're
> resistant to that, too?
I am resistant to what you are proposing to measure. I am resistant to simplifying metrics. Those
tend to do more harm than good.
Uli
>
> --Rich
>
>
>
> On 12/05/2016 04:00 PM, Ulrich Strk wrote:
>> On 05.12.16 17:54, Rich Bowen wrote:
>>>
>>>
>>> On 12/05/2016 07:41 AM, Ulrich Strk wrote:
>>>>> Or, at the very least, can we make a commitment to track this data going
>>>>>> forward?
>>>> Let me play the devil's advocate here: What for?
>>>>
>>>> GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
>>>> student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
>>>> to organize it and mentor students we can assume that at least those volunteers are seeing value in
>>>> it. Why the stats other than for satisfying our curiosity?
>>>
>>> Or, perhaps, let me give a different answer.
>>>
>>> I participated in GSoC as a mentor for $WorkProject. While it didn't
>>> "cost" me anything in dollars, it cost me probably 200 hours of my time.
>>> I know that other projects at work put more time in, and some less.
>>>
>>> This is an enormous cost to me, as an employee. So it behooves me to
>>> measure the benefit to the project. A student received payment to write
>>> code that was discarded. And I (and several of my colleagues) spent a
>>> huge amount of time, which I could have spent on other things, mentoring
>>> that student. Benefit to project, pretty heavily negative.
>>
>> In that case you should have failed the student pretty early as I tell mentors every year and as
>> documented in the mentor guide.
>>
>>>
>>> So, now, here we are at the ASF, doing GSoC with our projects, and
>>> promoting it to them as a benefit. Does it actually benefit them, or is
>>> it merely siphoning off time that could be spent on other things.
>>
>> If a mentor feels it is the latter, they should immediately fail the student.
>>
>>>
>>> To folks that say we can't measure that, I strongly disagree. There are
>>> two measures that are obvious and easy.
>>>
>>> 1) What % of GSoC student are still active on the project 6 months, 12
>>> months, 18 months after the project. (We can debate the definition of
>>> "active" all you like.
>>
>> This implies that the program is only successful if a certain number of students sticks around. And
>> this is exactly what I'm arguing against. IMO the program is successful as soon as a student has
>> some exposure to one of our communities.
>>
>>>
>>> 2) What % of code developed by GSoC students actually becomes a part of
>>> the project codebase at the end of the project?
>>
>> This has the same problems as trying to measure productivity from code. What is the percentage
>> telling us about the question we want to answer? Also see above: If the code cannot be part of the
>> project codebase the student should be failed.
>>
>>>
>>> I would maintain that #1 is part of our charter as ComDev, and #2 is
>>> part of what projects should be made aware of before they sign on.
>>>
>>> Again, I'm really not asking for a lot of data here. But I do think that
>>> it's part of a responsible accounting for participating in GSoC. If, for
>>> example, it is actually hindering projects, don't we want to know that.
>>>
>>> I did *not* participate in GSoC again, on $WorkProject, because it
>>> clearly hindered my project.
>>>
>>>
>>
>> I am reluctant to simply collect some data because I am missing a clear question what we are trying
>> to answer and how the data you want to collect is actually answering that question.
>>
>> Cheers,
>>
>> Uli
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>> For additional commands, e-mail: dev-help@community.apache.org
>>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
So ... stepping back a bit here. Are you saying that even attempting to
measure outcomes is harmful, because we might draw the wrong conclusions?
This is a completely new notion to me.
You say:
"I am reluctant to simply collect some data because I am missing a clear
question what we are trying to answer and how the data you want to
collect is actually answering that question."
I do have a clear question. My question is whether participating in the
GSoC contributes to our goal of Community Development. Does it develop
our community in any measurable way? I assume, at this point, that it
does, or we wouldn't keep doing it. In what way has it had measurable
impact on our community?
Things that we can clearly measure is adding participants in our project
communities, and artifacts added to our projects' revision control
repositories. Measuring other things is more difficult, and can be taken
as a later task. Easy things first.
The larger question of how and what we measure in the Community
Development PMC as a whole hinges on this, too. I'm curious if you're
resistant to that, too?
--Rich
On 12/05/2016 04:00 PM, Ulrich Stärk wrote:
> On 05.12.16 17:54, Rich Bowen wrote:
>>
>>
>> On 12/05/2016 07:41 AM, Ulrich Stärk wrote:
>>>> Or, at the very least, can we make a commitment to track this data going
>>>>> forward?
>>> Let me play the devil's advocate here: What for?
>>>
>>> GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
>>> student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
>>> to organize it and mentor students we can assume that at least those volunteers are seeing value in
>>> it. Why the stats other than for satisfying our curiosity?
>>
>> Or, perhaps, let me give a different answer.
>>
>> I participated in GSoC as a mentor for $WorkProject. While it didn't
>> "cost" me anything in dollars, it cost me probably 200 hours of my time.
>> I know that other projects at work put more time in, and some less.
>>
>> This is an enormous cost to me, as an employee. So it behooves me to
>> measure the benefit to the project. A student received payment to write
>> code that was discarded. And I (and several of my colleagues) spent a
>> huge amount of time, which I could have spent on other things, mentoring
>> that student. Benefit to project, pretty heavily negative.
>
> In that case you should have failed the student pretty early as I tell mentors every year and as
> documented in the mentor guide.
>
>>
>> So, now, here we are at the ASF, doing GSoC with our projects, and
>> promoting it to them as a benefit. Does it actually benefit them, or is
>> it merely siphoning off time that could be spent on other things.
>
> If a mentor feels it is the latter, they should immediately fail the student.
>
>>
>> To folks that say we can't measure that, I strongly disagree. There are
>> two measures that are obvious and easy.
>>
>> 1) What % of GSoC student are still active on the project 6 months, 12
>> months, 18 months after the project. (We can debate the definition of
>> "active" all you like.
>
> This implies that the program is only successful if a certain number of students sticks around. And
> this is exactly what I'm arguing against. IMO the program is successful as soon as a student has
> some exposure to one of our communities.
>
>>
>> 2) What % of code developed by GSoC students actually becomes a part of
>> the project codebase at the end of the project?
>
> This has the same problems as trying to measure productivity from code. What is the percentage
> telling us about the question we want to answer? Also see above: If the code cannot be part of the
> project codebase the student should be failed.
>
>>
>> I would maintain that #1 is part of our charter as ComDev, and #2 is
>> part of what projects should be made aware of before they sign on.
>>
>> Again, I'm really not asking for a lot of data here. But I do think that
>> it's part of a responsible accounting for participating in GSoC. If, for
>> example, it is actually hindering projects, don't we want to know that.
>>
>> I did *not* participate in GSoC again, on $WorkProject, because it
>> clearly hindered my project.
>>
>>
>
> I am reluctant to simply collect some data because I am missing a clear question what we are trying
> to answer and how the data you want to collect is actually answering that question.
>
> Cheers,
>
> Uli
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Ulrich Stärk <ul...@apache.org>.
On 05.12.16 17:54, Rich Bowen wrote:
>
>
> On 12/05/2016 07:41 AM, Ulrich Strk wrote:
>>> Or, at the very least, can we make a commitment to track this data going
>>>> forward?
>> Let me play the devil's advocate here: What for?
>>
>> GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
>> student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
>> to organize it and mentor students we can assume that at least those volunteers are seeing value in
>> it. Why the stats other than for satisfying our curiosity?
>
> Or, perhaps, let me give a different answer.
>
> I participated in GSoC as a mentor for $WorkProject. While it didn't
> "cost" me anything in dollars, it cost me probably 200 hours of my time.
> I know that other projects at work put more time in, and some less.
>
> This is an enormous cost to me, as an employee. So it behooves me to
> measure the benefit to the project. A student received payment to write
> code that was discarded. And I (and several of my colleagues) spent a
> huge amount of time, which I could have spent on other things, mentoring
> that student. Benefit to project, pretty heavily negative.
In that case you should have failed the student pretty early as I tell mentors every year and as
documented in the mentor guide.
>
> So, now, here we are at the ASF, doing GSoC with our projects, and
> promoting it to them as a benefit. Does it actually benefit them, or is
> it merely siphoning off time that could be spent on other things.
If a mentor feels it is the latter, they should immediately fail the student.
>
> To folks that say we can't measure that, I strongly disagree. There are
> two measures that are obvious and easy.
>
> 1) What % of GSoC student are still active on the project 6 months, 12
> months, 18 months after the project. (We can debate the definition of
> "active" all you like.
This implies that the program is only successful if a certain number of students sticks around. And
this is exactly what I'm arguing against. IMO the program is successful as soon as a student has
some exposure to one of our communities.
>
> 2) What % of code developed by GSoC students actually becomes a part of
> the project codebase at the end of the project?
This has the same problems as trying to measure productivity from code. What is the percentage
telling us about the question we want to answer? Also see above: If the code cannot be part of the
project codebase the student should be failed.
>
> I would maintain that #1 is part of our charter as ComDev, and #2 is
> part of what projects should be made aware of before they sign on.
>
> Again, I'm really not asking for a lot of data here. But I do think that
> it's part of a responsible accounting for participating in GSoC. If, for
> example, it is actually hindering projects, don't we want to know that.
>
> I did *not* participate in GSoC again, on $WorkProject, because it
> clearly hindered my project.
>
>
I am reluctant to simply collect some data because I am missing a clear question what we are trying
to answer and how the data you want to collect is actually answering that question.
Cheers,
Uli
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
On 12/05/2016 07:41 AM, Ulrich Stärk wrote:
>> Or, at the very least, can we make a commitment to track this data going
>> > forward?
> Let me play the devil's advocate here: What for?
>
> GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
> student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
> to organize it and mentor students we can assume that at least those volunteers are seeing value in
> it. Why the stats other than for satisfying our curiosity?
Or, perhaps, let me give a different answer.
I participated in GSoC as a mentor for $WorkProject. While it didn't
"cost" me anything in dollars, it cost me probably 200 hours of my time.
I know that other projects at work put more time in, and some less.
This is an enormous cost to me, as an employee. So it behooves me to
measure the benefit to the project. A student received payment to write
code that was discarded. And I (and several of my colleagues) spent a
huge amount of time, which I could have spent on other things, mentoring
that student. Benefit to project, pretty heavily negative.
So, now, here we are at the ASF, doing GSoC with our projects, and
promoting it to them as a benefit. Does it actually benefit them, or is
it merely siphoning off time that could be spent on other things.
To folks that say we can't measure that, I strongly disagree. There are
two measures that are obvious and easy.
1) What % of GSoC student are still active on the project 6 months, 12
months, 18 months after the project. (We can debate the definition of
"active" all you like.
2) What % of code developed by GSoC students actually becomes a part of
the project codebase at the end of the project?
I would maintain that #1 is part of our charter as ComDev, and #2 is
part of what projects should be made aware of before they sign on.
Again, I'm really not asking for a lot of data here. But I do think that
it's part of a responsible accounting for participating in GSoC. If, for
example, it is actually hindering projects, don't we want to know that.
I did *not* participate in GSoC again, on $WorkProject, because it
clearly hindered my project.
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Stian Soiland-Reyes <st...@apache.org>.
I think GSOC is very valuable, both for the students and the ASF projects
that participate.
Even if a student does not hang around after GSOC (were they made
committer?) it is still good for communities to get fresh drive and ideas,
and ask questions which established committers might not have thought that
newbies didn't know. GSOC projects are good ways to kick-start new
prototypes and challenge the existing architecture.
In a way GSOC also teaches/reminds projects how to make its code,
documentation and community welcoming for newcomers, so that is why I think
it is particularly good to encourage fresh incubator podlings to
participate.
This can't be easily measured in numbers.
GSOC mentoring can draw significant project bandwidth, yes, but then
projects and potential mentors can choose to participate or not per year.
On 5 Dec 2016 2:46 pm, "Shane Curcuru" <as...@shanecurcuru.org> wrote:
> Ulrich Stärk wrote on 12/5/16 9:27 AM:
> > On 05.12.16 14:30, Daniel Gruno wrote:
> >> On 12/05/2016 01:41 PM, Ulrich Stärk wrote:
> ...snip...
> >> But this goes beyond GSoC in my mind. We should be looking at ALL ComDev
> >> projects and evaluate what we want to keep, what isn't working, and what
> >> needs a do-over. The task of ComDev is to *develop communities*, it
> >> shouldn't just be a dumping ground for all things cross-project, whether
> >> they work or not. That is at least my opinion.
> >>
> >> We try strategies, give them life, see if they work, and if not, we put
> >> them to sleep or fix them.
> >
> > Geez, we are not maximizing for efficiency here (and that coming from a
> management consultant, how
> > ironic).
> >
> > Let me take GSoC as an example again. As long as we have volunteer
> mentors from our communities that
> > want to mentor students working on their projects than we IMO don't need
> any additional metric or a
> > certain level of usefulness to justify running the program. Our
> communities think it is important -
> > otherwise they wouldn't invest the time -, so should we.
> ...snip...
>
> You both have excellent points.
>
> I believe GSoC is a very valuable program for the ASF and the projects
> that participate, and I really hope the volunteers stepping up to
> organize keep doing it. I'll try to remember to thank you more often!
>
> Separately, it's great when we can also improve things, or at least show
> some sort of progress towards helping our project's communities grow.
> Given the rest of the organization that GSoC brings, and the fact that
> our LDAP and other records are getting pretty easy to script against, it
> would be great if some volunteer wanted to track people who became
> committers from GSoC to see their contributions in the future.
>
> But just because no-one has stepped up to do the metrics doesn't mean we
> should stop GSoC. If people want to volunteer to do something that's
> generally positive, great. Suggestions for improvements are good;
> getting in the way to slow progress because some additional goal hasn't
> yet attracted a volunteer to do it is not as good.
>
> - Shane
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
>
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
On 12/05/2016 01:08 PM, Stian Soiland-Reyes wrote:
> On 5 December 2016 at 17:07, Jacob Champion <ch...@gmail.com> wrote:
>> Perhaps a bit off-topic for this thread in particular, but in all
>> seriousness, I like this idea a lot. As you've noted, older projects and
>> established developers don't always know which parts of their projects need
>> "newbie improvement" -- after all, they've already figured it out for
>> themselves.
>
>
> Yeah, let's start that in another thread.
>
>
> Here is a starting point:
>
> https://svn.apache.org/repos/asf/comdev/tmp/random_projects.py
>
> stain@biggiebuntu:~/src/comdev/tmp$ ./random_projects.py 3
> karaf
> activemq
> oozie
>
>
> (would this belong under comdev/tools? It seems to be meetup-centric)
>
It is at the moment, but it looks like this would go there too. 'tools',
like 'libraries', 'programs' and 'stuff', are terrible directory names.
Feel free to rename. (Yes, I think I created that directory.)
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Stian Soiland-Reyes <st...@apache.org>.
On 5 December 2016 at 17:07, Jacob Champion <ch...@gmail.com> wrote:
> Perhaps a bit off-topic for this thread in particular, but in all
> seriousness, I like this idea a lot. As you've noted, older projects and
> established developers don't always know which parts of their projects need
> "newbie improvement" -- after all, they've already figured it out for
> themselves.
Yeah, let's start that in another thread.
Here is a starting point:
https://svn.apache.org/repos/asf/comdev/tmp/random_projects.py
stain@biggiebuntu:~/src/comdev/tmp$ ./random_projects.py 3
karaf
activemq
oozie
(would this belong under comdev/tools? It seems to be meetup-centric)
--
Stian Soiland-Reyes
http://orcid.org/0000-0001-9842-9718
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Jacob Champion <ch...@gmail.com>.
On 12/05/2016 08:15 AM, Stian Soiland-Reyes wrote:
> Perhaps ComDev can act like a prpject visitor. (Not auditor or inspector!)
"No one expects the ComDev Inquisition!!"
> We can make a wiki page with a randomised list of all ASF projects, a quick
> checklist (smaller than maturity model) and just go through them one by one
> and raise issues/pull requests for any small community encouragement issues
> we find. Is that too intrusive?
Perhaps a bit off-topic for this thread in particular, but in all
seriousness, I like this idea a lot. As you've noted, older projects and
established developers don't always know which parts of their projects
need "newbie improvement" -- after all, they've already figured it out
for themselves.
--Jacob
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Stian Soiland-Reyes <st...@apache.org>.
Perhaps something for ComDev is to work on engagement with software
developers who don't currently consider themselves part of any ASF
community, but use our software without contributing (much) back.
These are generally more experienced than GSOC students, and while they
won't have as much continual time available, they could contribute in other
ways.
You may them moaning on StackOverflow or at big software conferences.
ApacheCon is nice, but is often preaching to the converted (and their
colleagues they managed to lure along :-).
I would think that many times ASF comitters could be at a more specific
conference or venue for their field, and could easily give an ASF/open
source pep talk to people they meet or present to. Golden opportunity!
Perhaps some Comdev PDFs of a generic flyer with "How does ASF work? Join
us!" could be handy to print out, so they don't have to make up everything
on the spot or beg here for trademark guidance to compose their own ASF
promo material.
Knowing myself and my organisation, we used ASF software for at least 10
years, developing our own open source software. It didn't occur to us to
during that time that perhaps we could (should!) also help out on ASF
software we relied on, like httpd, tomcat, commons, derby, Jena, etc.
No-one told us so, no READMEs or announcements, no people or stands at big
conferences.
Now there is GitHub pull requests, which do significantly lower the
barrier, but many ASF projects still don't say much (or at all) on their
pages and READMEs that this is a welcome contribution path.
Many ASF projects don't explain well enough for outsiders how development
happens. Perhaps it looks as if we are all experts and you need to be
specially invited. Perhaps people sign up to dev@ lists, but get scared by
the open development discussions, thinking "Uh, I don't know enough about
this, it's clearly not for me. And what is this VOTE thing?". We can't
know, because those people are not around.
The information is there, but it is often buried, not linked to from
project home pages and email threads - "everybody knows". For instance I
try to include statements like this in VOTE emails:
> Anyone can participate in testing and voting, not just
committers, please feel free to try out the release candidate
and provide your votes.
How to review a release candidate? https://s.apache.org/review-release
ComDev could help with building a generic pages like "what is a release
candidate and what do I do", "How are decisions made?". This then fits into
the ComDev or ASF blog and could be used as a default link or template for
projects to refer to when voting - it should be more observational than
policy.
Could it make sense for ComDev to (sensitively) contribute patches pr
recommendations to ASF projects in this regard? It should not look like
medling or patronising, just like a pull request with friendly suggestions.
Often I find it is that incubator projects get it right (eventually), while
older projects are comfortably established and are effectively outdated in
respect to the the ComDev maturity model and ASF best practice. (E.g. link
to older mbox mailing list archive, no link to Code of Conduct).
Perhaps ComDev can act like a prpject visitor. (Not auditor or inspector!)
We can make a wiki page with a randomised list of all ASF projects, a quick
checklist (smaller than maturity model) and just go through them one by one
and raise issues/pull requests for any small community encouragement issues
we find. Is that too intrusive?
On 5 Dec 2016 3:26 pm, "Daniel Gruno" <hu...@apache.org> wrote:
> On 12/05/2016 03:46 PM, Shane Curcuru wrote:
> > Ulrich Stärk wrote on 12/5/16 9:27 AM:
> >> On 05.12.16 14:30, Daniel Gruno wrote:
> >>> On 12/05/2016 01:41 PM, Ulrich Stärk wrote:
> > ...snip...
> >>> But this goes beyond GSoC in my mind. We should be looking at ALL
> ComDev
> >>> projects and evaluate what we want to keep, what isn't working, and
> what
> >>> needs a do-over. The task of ComDev is to *develop communities*, it
> >>> shouldn't just be a dumping ground for all things cross-project,
> whether
> >>> they work or not. That is at least my opinion.
> >>>
> >>> We try strategies, give them life, see if they work, and if not, we put
> >>> them to sleep or fix them.
> >>
> >> Geez, we are not maximizing for efficiency here (and that coming from a
> management consultant, how
> >> ironic).
> >>
> >> Let me take GSoC as an example again. As long as we have volunteer
> mentors from our communities that
> >> want to mentor students working on their projects than we IMO don't
> need any additional metric or a
> >> certain level of usefulness to justify running the program. Our
> communities think it is important -
> >> otherwise they wouldn't invest the time -, so should we.
> > ...snip...
> >
> > You both have excellent points.
> >
> > I believe GSoC is a very valuable program for the ASF and the projects
> > that participate, and I really hope the volunteers stepping up to
> > organize keep doing it. I'll try to remember to thank you more often!
> >
> > Separately, it's great when we can also improve things, or at least show
> > some sort of progress towards helping our project's communities grow.
> > Given the rest of the organization that GSoC brings, and the fact that
> > our LDAP and other records are getting pretty easy to script against, it
> > would be great if some volunteer wanted to track people who became
> > committers from GSoC to see their contributions in the future.
> >
> > But just because no-one has stepped up to do the metrics doesn't mean we
> > should stop GSoC. If people want to volunteer to do something that's
> > generally positive, great. Suggestions for improvements are good;
> > getting in the way to slow progress because some additional goal hasn't
> > yet attracted a volunteer to do it is not as good.
>
> I don't think anyone has actively suggested we stop GSoC. But it would
> be good to see more cohesion in ComDev and some discussions on what we
> are doing, how it benefits our mission, and what the results are.
>
> Rich's original question was "does it benefit the ASF?". We don't seem
> to have bothered answering that question in a diligent matter over the
> past years, and I think we should do so. If GSoC is as valuable as the
> proponents say, then it should be easy to gather some more non-anecdotal
> information that says "yes, this has helped develop the community".
>
> I'd be super interested to learn what GSoC actually achieves, I have no
> idea if it's just a charitable coding-for-money scheme or if it actually
> helps grow our community in the long run. I also have no idea which
> aspects the projects find most useful, and I'd love to learn that.
>
> What I would be most interested in, however, is (as stated above) more
> cohesion in what ComDev does and how it is done. There should be at
> least some form of point of there being a PMC.
>
> With regards,
> Daniel.
>
> >
> > - Shane
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> > For additional commands, e-mail: dev-help@community.apache.org
> >
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
>
Re: Does GSoC help develop communities?
Posted by Daniel Gruno <hu...@apache.org>.
On 12/05/2016 03:46 PM, Shane Curcuru wrote:
> Ulrich Strk wrote on 12/5/16 9:27 AM:
>> On 05.12.16 14:30, Daniel Gruno wrote:
>>> On 12/05/2016 01:41 PM, Ulrich Strk wrote:
> ...snip...
>>> But this goes beyond GSoC in my mind. We should be looking at ALL ComDev
>>> projects and evaluate what we want to keep, what isn't working, and what
>>> needs a do-over. The task of ComDev is to *develop communities*, it
>>> shouldn't just be a dumping ground for all things cross-project, whether
>>> they work or not. That is at least my opinion.
>>>
>>> We try strategies, give them life, see if they work, and if not, we put
>>> them to sleep or fix them.
>>
>> Geez, we are not maximizing for efficiency here (and that coming from a management consultant, how
>> ironic).
>>
>> Let me take GSoC as an example again. As long as we have volunteer mentors from our communities that
>> want to mentor students working on their projects than we IMO don't need any additional metric or a
>> certain level of usefulness to justify running the program. Our communities think it is important -
>> otherwise they wouldn't invest the time -, so should we.
> ...snip...
>
> You both have excellent points.
>
> I believe GSoC is a very valuable program for the ASF and the projects
> that participate, and I really hope the volunteers stepping up to
> organize keep doing it. I'll try to remember to thank you more often!
>
> Separately, it's great when we can also improve things, or at least show
> some sort of progress towards helping our project's communities grow.
> Given the rest of the organization that GSoC brings, and the fact that
> our LDAP and other records are getting pretty easy to script against, it
> would be great if some volunteer wanted to track people who became
> committers from GSoC to see their contributions in the future.
>
> But just because no-one has stepped up to do the metrics doesn't mean we
> should stop GSoC. If people want to volunteer to do something that's
> generally positive, great. Suggestions for improvements are good;
> getting in the way to slow progress because some additional goal hasn't
> yet attracted a volunteer to do it is not as good.
I don't think anyone has actively suggested we stop GSoC. But it would
be good to see more cohesion in ComDev and some discussions on what we
are doing, how it benefits our mission, and what the results are.
Rich's original question was "does it benefit the ASF?". We don't seem
to have bothered answering that question in a diligent matter over the
past years, and I think we should do so. If GSoC is as valuable as the
proponents say, then it should be easy to gather some more non-anecdotal
information that says "yes, this has helped develop the community".
I'd be super interested to learn what GSoC actually achieves, I have no
idea if it's just a charitable coding-for-money scheme or if it actually
helps grow our community in the long run. I also have no idea which
aspects the projects find most useful, and I'd love to learn that.
What I would be most interested in, however, is (as stated above) more
cohesion in what ComDev does and how it is done. There should be at
least some form of point of there being a PMC.
With regards,
Daniel.
>
> - Shane
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Shane Curcuru <as...@shanecurcuru.org>.
Ulrich Strk wrote on 12/5/16 9:27 AM:
> On 05.12.16 14:30, Daniel Gruno wrote:
>> On 12/05/2016 01:41 PM, Ulrich Strk wrote:
...snip...
>> But this goes beyond GSoC in my mind. We should be looking at ALL ComDev
>> projects and evaluate what we want to keep, what isn't working, and what
>> needs a do-over. The task of ComDev is to *develop communities*, it
>> shouldn't just be a dumping ground for all things cross-project, whether
>> they work or not. That is at least my opinion.
>>
>> We try strategies, give them life, see if they work, and if not, we put
>> them to sleep or fix them.
>
> Geez, we are not maximizing for efficiency here (and that coming from a management consultant, how
> ironic).
>
> Let me take GSoC as an example again. As long as we have volunteer mentors from our communities that
> want to mentor students working on their projects than we IMO don't need any additional metric or a
> certain level of usefulness to justify running the program. Our communities think it is important -
> otherwise they wouldn't invest the time -, so should we.
...snip...
You both have excellent points.
I believe GSoC is a very valuable program for the ASF and the projects
that participate, and I really hope the volunteers stepping up to
organize keep doing it. I'll try to remember to thank you more often!
Separately, it's great when we can also improve things, or at least show
some sort of progress towards helping our project's communities grow.
Given the rest of the organization that GSoC brings, and the fact that
our LDAP and other records are getting pretty easy to script against, it
would be great if some volunteer wanted to track people who became
committers from GSoC to see their contributions in the future.
But just because no-one has stepped up to do the metrics doesn't mean we
should stop GSoC. If people want to volunteer to do something that's
generally positive, great. Suggestions for improvements are good;
getting in the way to slow progress because some additional goal hasn't
yet attracted a volunteer to do it is not as good.
- Shane
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Ulrich Stärk <ul...@apache.org>.
On 05.12.16 14:30, Daniel Gruno wrote:
> On 12/05/2016 01:41 PM, Ulrich Strk wrote:
>> On 03.12.16 23:16, Rich Bowen wrote:
>>> So, this question was asked several weeks ago, and, so far, we have no
>>> statistics. I wonder if, now that there appears to be some interest
>>> around GSoC 2017 starting up, someone could look back at their notes and
>>> answer some of these questions.
>>>
>>> While it's cool that folks are spending time on GSoC, for the sake of
>>> GSoC, and of the students, it would also be awesome to know whether the
>>> time and money that the ASF puts into this is actually doing something
>>> towards the stated goal of Community Development.
>>>
>>> Or, at the very least, can we make a commitment to track this data going
>>> forward?
>>
>> Let me play the devil's advocate here: What for?
>>
>> GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
>> student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
>> to organize it and mentor students we can assume that at least those volunteers are seeing value in
>> it. Why the stats other than for satisfying our curiosity?
>
> Let's flip that around: If the ASF doesn't benefit in any way from this,
> why are we bothering with it? why is it the prominent (and in many cases
> the _only_) thing in our board reports?
I don't get this logic. Can you explain why the ASF isn't benefitting? Apparently our communities
are, otherwise we wouldn't see mentors willing to mentor students, no?
>
> Going back a few reports (those that are public at this date), they all
> seem to be either about starting, doing or concluding (which we have no
> data on) GSoC and then perhaps a small bit of extra data on all the
> rest. It is a VERY prominent thing, so why is that so?
Because apart from GSoC we don't seem to do any other structured efforts to report on. If there has
been something noteworthy, it was in the reports (e.g. helpwanted, diversity efforts, etc.).
Sharan's diversity efforts are a the closest thing to a new structured effort. Her extensive report
was included in our last board report.
>
> The task of ComDev is developing community. If we don't have any data or
> interest in acquiring such to show that this is in fact helping towards
> that, then we should consider whether the current strategy is the right
> thing to focus on.
Again. I strongly believe that there is no suitable metric for answering these kind of questions
except for indirectly through demand and I yet need to be convinced of the contrary.
>
> But this goes beyond GSoC in my mind. We should be looking at ALL ComDev
> projects and evaluate what we want to keep, what isn't working, and what
> needs a do-over. The task of ComDev is to *develop communities*, it
> shouldn't just be a dumping ground for all things cross-project, whether
> they work or not. That is at least my opinion.
>
> We try strategies, give them life, see if they work, and if not, we put
> them to sleep or fix them.
Geez, we are not maximizing for efficiency here (and that coming from a management consultant, how
ironic).
Let me take GSoC as an example again. As long as we have volunteer mentors from our communities that
want to mentor students working on their projects than we IMO don't need any additional metric or a
certain level of usefulness to justify running the program. Our communities think it is important -
otherwise they wouldn't invest the time -, so should we.
Thinking this further we have already won when a student applies. They have interacted with the
community, used the code, read documentation, maybe even already fixed some bug along the way. IMO
that is also an important part of the community building. If a student sticks around than that is
simply the icing on the cake and no metric whatsoever for the success of the program.
Cheers,
Uli
>
> With regards,
> Daniel.
>
>>
>> Cheers,
>>
>> Uli
>>
>>>
>>> --Rich
>>>
>>> On 11/16/2016 05:18 AM, Nick Burch wrote:
>>>> On Wed, 16 Nov 2016, Rich Bowen wrote:
>>>>> It would be great to have some kind of statistics on how GSoC helps
>>>>> projects longer term. Do students stick around? Does the code written
>>>>> actually get incorporated into releases? Does it in fact contribute to
>>>>> the mission of Community Development, or is it just a nice summer job
>>>>> for these students?
>>>>
>>>> Do we have a list (maybe somewhere in the comdev private svn?) of
>>>> everyone who has taken part in GSoC?
>>>>
>>>> If so, it'd be fairly easy to annotate that with apache IDs, then see
>>>> who's now on PMCs or who's now a member. Producing sharable statistics
>>>> from that automatically is then easy. (I say this as someone who helped
>>>> update the similar Travel Assistance Committee / TAC ones on Monday!)
>>>>
>>>> Not sure if it's easy to find out the "last commit date" for people, to
>>>> check for the "still around" part (eg for people who got committership
>>>> during GSoc, maybe on a branch), but I know where infra are to ask...
>>>>
>>>> Nick
>>>>
>>>> PS FWIW, within Tika we've certainly had a few GSoC people stick around!
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>>>> For additional commands, e-mail: dev-help@community.apache.org
>>>>
>>>
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>> For additional commands, e-mail: dev-help@community.apache.org
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
Yes, this would be hugely helpful.
This page - http://community.apache.org/gsoc- should be inward-facing as
well as outward facing. It should tell projects what they should expect
to spend (time, effort, not necessarily money) and what the benefits are
likely to be.
If that's anecdotal, great.
I believe that we (the ComDev PMC) should ask projects to write stories,
and/or give us data, when we assist them with GSoC things. It's not an
unreasonable request, and I think most of these projects would be
willing to do this.
I know that there have been blog posts in the past, but I don't know
where they are. Linking them from this page would be awesome.
But I do think that *numbers* communicate just as well (and better to
some audiences) as stories.
--Rich
On 12/05/2016 12:10 PM, Suresh Marru wrote:
> Hi Rich,
>
> Do you prefer to see cumulative statistics or PMC wise? With a small effort, I can get detailed statistics from Apache Airavata. GSoC has been great for the project and is one of the primary sources to induct fresh blood. We have a decent retention rate, students have stayed around, earned committerships, then became PMC Members, graduated and moved onto real jobs and still continue to contribute (not always by code though).
>
> Engaging the students and mentoring them takes time, but we have spun it around and used it as an opportunity to improve documentation and contribution workflow.
>
> Hope this experience summary helps the discussion, will be happy to elaborate and provide specific examples.
>
> Cheers,
> Suresh
>
>> On Dec 5, 2016, at 12:01 PM, Rich Bowen <rb...@rcbowen.com> wrote:
>>
>> Nobody is suggesting we have a 10 year plan with milestones and
>> deliverables. I'm suggesting that when we do something under the heading
>> of "community development" we have an obligation to make some attempt to
>> measure it to determine if it actually moves us in that direction.
>>
>> Nobody is saying that we shouldn't participate in GSoC. I'm suggesting
>> that before we promote GSoC to our projects, we should have some numbers
>> (we've been doing this for years. surely there's some numbers that we
>> could gather her?) that show projects that it's worth their time. Or
>> warns them that it might not be.
>>
>> I don't care how much time individuals spend on GSoC. I care that we are
>> telling projects that it's a worthwhile thing for them to spend *their*
>> time on, and we don't appear to have actually taken the time to find out
>> of that's true.
>>
>>
>> On 12/05/2016 08:59 AM, Bertrand Delacretaz wrote:
>>> Hi,
>>>
>>> On Mon, Dec 5, 2016 at 2:30 PM, Daniel Gruno <hu...@apache.org> wrote:
>>>> ...The task of ComDev is developing community. If we don't have any data or
>>>> interest in acquiring such to show that this is in fact helping towards
>>>> that, then we should consider whether the current strategy is the right
>>>> thing to focus on....
>>>
>>> I disagree with the need for comdev to have a strategy.
>>>
>>> At the technical level the ASF doesn't have a strategy, it just
>>> provides space for its projects to exist and flourish.
>>>
>>> I think comdev can operate in the same way, as a loose group of
>>> volunteers who collectively help develop communities, without
>>> necessarily having a global comdev strategy to follow.
>>>
>>> Three examples:
>>>
>>> 1) A small group is running GSoC, which as Uli mentions doesn't cost
>>> the ASF anything and actually brings some money in. GSoC clearly helps
>>> our mission by helping a few community members join the ASF every
>>> year. Exactly how many is not very important if volunteers agree to
>>> run it.
>>>
>>> 2) Sharan and others have started work on diversity initiatives -
>>> another subset of folks sharing common interests that match the
>>> overall comdev mission.
>>>
>>> 3) I led a small group to develop our maturity model, I think it's a
>>> very useful tool. I think we made just one change to it in 2016, it's
>>> stable but useful and maintained. Others don't care about that or
>>> didn't have time to help - no problem.
>>>
>>> You could argue that these things are disjoint but they are all small
>>> steps that help towards our overall mission. We don't need much
>>> coordination between them, IMO just making sure the comdev PMC agrees
>>> with these things happening, and doing out best to unify their
>>> communications channels to create synergies is good enough.
>>>
>>> Comdev can just provide a space for volunteers to help develop
>>> communities, that's good enough for me. If others want more structured
>>> activities feel free to do them but don't expect all PMC members to
>>> necessarily join or to feel bad if they don't.
>>>
>>> -Bertrand
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>>> For additional commands, e-mail: dev-help@community.apache.org
>>>
>>
>>
>> --
>> Rich Bowen - rbowen@rcbowen.com - @rbowen
>> http://apachecon.com/ - @apachecon
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Daniel Gruno <hu...@apache.org>.
Why not Zoidberg?
If someone wants to gather stats, that's great - if someone wants to
gather stories etc, that is also great. One doesn't exclude the other,
and considering we have nothing at the moment, anything is a step up.
With regards,
Daniel
On 12/05/2016 06:18 PM, Alex Harui wrote:
> IMO, rather than gathering statistics, it would be better to gather
> stories, tips and advice. It doesn't seem to me that statistics would be
> helpful, folks just need to know that it can have great benefits or great
> cost and some ideas of the reasons why. Even if it hasn't been helpful in
> the past for the vast majority of ASF projects, if it benefits some other
> minority of projects, the important takeaway is that it can work for
> certain reasons, not that it isn't a good idea for your project because it
> didn't work for most other projects. Each project is different.
>
> My 2 cents,
> -Alex
>
> On 12/5/16, 9:10 AM, "Suresh Marru" <sm...@apache.org> wrote:
>
>> Hi Rich,
>>
>> Do you prefer to see cumulative statistics or PMC wise? With a small
>> effort, I can get detailed statistics from Apache Airavata. GSoC has been
>> great for the project and is one of the primary sources to induct fresh
>> blood. We have a decent retention rate, students have stayed around,
>> earned committerships, then became PMC Members, graduated and moved onto
>> real jobs and still continue to contribute (not always by code though).
>>
>> Engaging the students and mentoring them takes time, but we have spun it
>> around and used it as an opportunity to improve documentation and
>> contribution workflow.
>>
>> Hope this experience summary helps the discussion, will be happy to
>> elaborate and provide specific examples.
>>
>> Cheers,
>> Suresh
>>
>>> On Dec 5, 2016, at 12:01 PM, Rich Bowen <rb...@rcbowen.com> wrote:
>>>
>>> Nobody is suggesting we have a 10 year plan with milestones and
>>> deliverables. I'm suggesting that when we do something under the heading
>>> of "community development" we have an obligation to make some attempt to
>>> measure it to determine if it actually moves us in that direction.
>>>
>>> Nobody is saying that we shouldn't participate in GSoC. I'm suggesting
>>> that before we promote GSoC to our projects, we should have some numbers
>>> (we've been doing this for years. surely there's some numbers that we
>>> could gather her?) that show projects that it's worth their time. Or
>>> warns them that it might not be.
>>>
>>> I don't care how much time individuals spend on GSoC. I care that we are
>>> telling projects that it's a worthwhile thing for them to spend *their*
>>> time on, and we don't appear to have actually taken the time to find out
>>> of that's true.
>>>
>>>
>>> On 12/05/2016 08:59 AM, Bertrand Delacretaz wrote:
>>>> Hi,
>>>>
>>>> On Mon, Dec 5, 2016 at 2:30 PM, Daniel Gruno <hu...@apache.org>
>>>> wrote:
>>>>> ...The task of ComDev is developing community. If we don't have any
>>>>> data or
>>>>> interest in acquiring such to show that this is in fact helping
>>>>> towards
>>>>> that, then we should consider whether the current strategy is the
>>>>> right
>>>>> thing to focus on....
>>>>
>>>> I disagree with the need for comdev to have a strategy.
>>>>
>>>> At the technical level the ASF doesn't have a strategy, it just
>>>> provides space for its projects to exist and flourish.
>>>>
>>>> I think comdev can operate in the same way, as a loose group of
>>>> volunteers who collectively help develop communities, without
>>>> necessarily having a global comdev strategy to follow.
>>>>
>>>> Three examples:
>>>>
>>>> 1) A small group is running GSoC, which as Uli mentions doesn't cost
>>>> the ASF anything and actually brings some money in. GSoC clearly helps
>>>> our mission by helping a few community members join the ASF every
>>>> year. Exactly how many is not very important if volunteers agree to
>>>> run it.
>>>>
>>>> 2) Sharan and others have started work on diversity initiatives -
>>>> another subset of folks sharing common interests that match the
>>>> overall comdev mission.
>>>>
>>>> 3) I led a small group to develop our maturity model, I think it's a
>>>> very useful tool. I think we made just one change to it in 2016, it's
>>>> stable but useful and maintained. Others don't care about that or
>>>> didn't have time to help - no problem.
>>>>
>>>> You could argue that these things are disjoint but they are all small
>>>> steps that help towards our overall mission. We don't need much
>>>> coordination between them, IMO just making sure the comdev PMC agrees
>>>> with these things happening, and doing out best to unify their
>>>> communications channels to create synergies is good enough.
>>>>
>>>> Comdev can just provide a space for volunteers to help develop
>>>> communities, that's good enough for me. If others want more structured
>>>> activities feel free to do them but don't expect all PMC members to
>>>> necessarily join or to feel bad if they don't.
>>>>
>>>> -Bertrand
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>>>> For additional commands, e-mail: dev-help@community.apache.org
>>>>
>>>
>>>
>>> --
>>> Rich Bowen - rbowen@rcbowen.com - @rbowen
>>> http://apachecon.com/ - @apachecon
>>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>> For additional commands, e-mail: dev-help@community.apache.org
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Alex Harui <ah...@adobe.com>.
IMO, rather than gathering statistics, it would be better to gather
stories, tips and advice. It doesn't seem to me that statistics would be
helpful, folks just need to know that it can have great benefits or great
cost and some ideas of the reasons why. Even if it hasn't been helpful in
the past for the vast majority of ASF projects, if it benefits some other
minority of projects, the important takeaway is that it can work for
certain reasons, not that it isn't a good idea for your project because it
didn't work for most other projects. Each project is different.
My 2 cents,
-Alex
On 12/5/16, 9:10 AM, "Suresh Marru" <sm...@apache.org> wrote:
>Hi Rich,
>
>Do you prefer to see cumulative statistics or PMC wise? With a small
>effort, I can get detailed statistics from Apache Airavata. GSoC has been
>great for the project and is one of the primary sources to induct fresh
>blood. We have a decent retention rate, students have stayed around,
>earned committerships, then became PMC Members, graduated and moved onto
>real jobs and still continue to contribute (not always by code though).
>
>Engaging the students and mentoring them takes time, but we have spun it
>around and used it as an opportunity to improve documentation and
>contribution workflow.
>
>Hope this experience summary helps the discussion, will be happy to
>elaborate and provide specific examples.
>
>Cheers,
>Suresh
>
>> On Dec 5, 2016, at 12:01 PM, Rich Bowen <rb...@rcbowen.com> wrote:
>>
>> Nobody is suggesting we have a 10 year plan with milestones and
>> deliverables. I'm suggesting that when we do something under the heading
>> of "community development" we have an obligation to make some attempt to
>> measure it to determine if it actually moves us in that direction.
>>
>> Nobody is saying that we shouldn't participate in GSoC. I'm suggesting
>> that before we promote GSoC to our projects, we should have some numbers
>> (we've been doing this for years. surely there's some numbers that we
>> could gather her?) that show projects that it's worth their time. Or
>> warns them that it might not be.
>>
>> I don't care how much time individuals spend on GSoC. I care that we are
>> telling projects that it's a worthwhile thing for them to spend *their*
>> time on, and we don't appear to have actually taken the time to find out
>> of that's true.
>>
>>
>> On 12/05/2016 08:59 AM, Bertrand Delacretaz wrote:
>>> Hi,
>>>
>>> On Mon, Dec 5, 2016 at 2:30 PM, Daniel Gruno <hu...@apache.org>
>>>wrote:
>>>> ...The task of ComDev is developing community. If we don't have any
>>>>data or
>>>> interest in acquiring such to show that this is in fact helping
>>>>towards
>>>> that, then we should consider whether the current strategy is the
>>>>right
>>>> thing to focus on....
>>>
>>> I disagree with the need for comdev to have a strategy.
>>>
>>> At the technical level the ASF doesn't have a strategy, it just
>>> provides space for its projects to exist and flourish.
>>>
>>> I think comdev can operate in the same way, as a loose group of
>>> volunteers who collectively help develop communities, without
>>> necessarily having a global comdev strategy to follow.
>>>
>>> Three examples:
>>>
>>> 1) A small group is running GSoC, which as Uli mentions doesn't cost
>>> the ASF anything and actually brings some money in. GSoC clearly helps
>>> our mission by helping a few community members join the ASF every
>>> year. Exactly how many is not very important if volunteers agree to
>>> run it.
>>>
>>> 2) Sharan and others have started work on diversity initiatives -
>>> another subset of folks sharing common interests that match the
>>> overall comdev mission.
>>>
>>> 3) I led a small group to develop our maturity model, I think it's a
>>> very useful tool. I think we made just one change to it in 2016, it's
>>> stable but useful and maintained. Others don't care about that or
>>> didn't have time to help - no problem.
>>>
>>> You could argue that these things are disjoint but they are all small
>>> steps that help towards our overall mission. We don't need much
>>> coordination between them, IMO just making sure the comdev PMC agrees
>>> with these things happening, and doing out best to unify their
>>> communications channels to create synergies is good enough.
>>>
>>> Comdev can just provide a space for volunteers to help develop
>>> communities, that's good enough for me. If others want more structured
>>> activities feel free to do them but don't expect all PMC members to
>>> necessarily join or to feel bad if they don't.
>>>
>>> -Bertrand
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>>> For additional commands, e-mail: dev-help@community.apache.org
>>>
>>
>>
>> --
>> Rich Bowen - rbowen@rcbowen.com - @rbowen
>> http://apachecon.com/ - @apachecon
>>
>
>
>---------------------------------------------------------------------
>To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>For additional commands, e-mail: dev-help@community.apache.org
>
Re: Does GSoC help develop communities?
Posted by Suresh Marru <sm...@apache.org>.
Hi Rich,
Do you prefer to see cumulative statistics or PMC wise? With a small effort, I can get detailed statistics from Apache Airavata. GSoC has been great for the project and is one of the primary sources to induct fresh blood. We have a decent retention rate, students have stayed around, earned committerships, then became PMC Members, graduated and moved onto real jobs and still continue to contribute (not always by code though).
Engaging the students and mentoring them takes time, but we have spun it around and used it as an opportunity to improve documentation and contribution workflow.
Hope this experience summary helps the discussion, will be happy to elaborate and provide specific examples.
Cheers,
Suresh
> On Dec 5, 2016, at 12:01 PM, Rich Bowen <rb...@rcbowen.com> wrote:
>
> Nobody is suggesting we have a 10 year plan with milestones and
> deliverables. I'm suggesting that when we do something under the heading
> of "community development" we have an obligation to make some attempt to
> measure it to determine if it actually moves us in that direction.
>
> Nobody is saying that we shouldn't participate in GSoC. I'm suggesting
> that before we promote GSoC to our projects, we should have some numbers
> (we've been doing this for years. surely there's some numbers that we
> could gather her?) that show projects that it's worth their time. Or
> warns them that it might not be.
>
> I don't care how much time individuals spend on GSoC. I care that we are
> telling projects that it's a worthwhile thing for them to spend *their*
> time on, and we don't appear to have actually taken the time to find out
> of that's true.
>
>
> On 12/05/2016 08:59 AM, Bertrand Delacretaz wrote:
>> Hi,
>>
>> On Mon, Dec 5, 2016 at 2:30 PM, Daniel Gruno <hu...@apache.org> wrote:
>>> ...The task of ComDev is developing community. If we don't have any data or
>>> interest in acquiring such to show that this is in fact helping towards
>>> that, then we should consider whether the current strategy is the right
>>> thing to focus on....
>>
>> I disagree with the need for comdev to have a strategy.
>>
>> At the technical level the ASF doesn't have a strategy, it just
>> provides space for its projects to exist and flourish.
>>
>> I think comdev can operate in the same way, as a loose group of
>> volunteers who collectively help develop communities, without
>> necessarily having a global comdev strategy to follow.
>>
>> Three examples:
>>
>> 1) A small group is running GSoC, which as Uli mentions doesn't cost
>> the ASF anything and actually brings some money in. GSoC clearly helps
>> our mission by helping a few community members join the ASF every
>> year. Exactly how many is not very important if volunteers agree to
>> run it.
>>
>> 2) Sharan and others have started work on diversity initiatives -
>> another subset of folks sharing common interests that match the
>> overall comdev mission.
>>
>> 3) I led a small group to develop our maturity model, I think it's a
>> very useful tool. I think we made just one change to it in 2016, it's
>> stable but useful and maintained. Others don't care about that or
>> didn't have time to help - no problem.
>>
>> You could argue that these things are disjoint but they are all small
>> steps that help towards our overall mission. We don't need much
>> coordination between them, IMO just making sure the comdev PMC agrees
>> with these things happening, and doing out best to unify their
>> communications channels to create synergies is good enough.
>>
>> Comdev can just provide a space for volunteers to help develop
>> communities, that's good enough for me. If others want more structured
>> activities feel free to do them but don't expect all PMC members to
>> necessarily join or to feel bad if they don't.
>>
>> -Bertrand
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>> For additional commands, e-mail: dev-help@community.apache.org
>>
>
>
> --
> Rich Bowen - rbowen@rcbowen.com - @rbowen
> http://apachecon.com/ - @apachecon
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
Nobody is suggesting we have a 10 year plan with milestones and
deliverables. I'm suggesting that when we do something under the heading
of "community development" we have an obligation to make some attempt to
measure it to determine if it actually moves us in that direction.
Nobody is saying that we shouldn't participate in GSoC. I'm suggesting
that before we promote GSoC to our projects, we should have some numbers
(we've been doing this for years. surely there's some numbers that we
could gather her?) that show projects that it's worth their time. Or
warns them that it might not be.
I don't care how much time individuals spend on GSoC. I care that we are
telling projects that it's a worthwhile thing for them to spend *their*
time on, and we don't appear to have actually taken the time to find out
of that's true.
On 12/05/2016 08:59 AM, Bertrand Delacretaz wrote:
> Hi,
>
> On Mon, Dec 5, 2016 at 2:30 PM, Daniel Gruno <hu...@apache.org> wrote:
>> ...The task of ComDev is developing community. If we don't have any data or
>> interest in acquiring such to show that this is in fact helping towards
>> that, then we should consider whether the current strategy is the right
>> thing to focus on....
>
> I disagree with the need for comdev to have a strategy.
>
> At the technical level the ASF doesn't have a strategy, it just
> provides space for its projects to exist and flourish.
>
> I think comdev can operate in the same way, as a loose group of
> volunteers who collectively help develop communities, without
> necessarily having a global comdev strategy to follow.
>
> Three examples:
>
> 1) A small group is running GSoC, which as Uli mentions doesn't cost
> the ASF anything and actually brings some money in. GSoC clearly helps
> our mission by helping a few community members join the ASF every
> year. Exactly how many is not very important if volunteers agree to
> run it.
>
> 2) Sharan and others have started work on diversity initiatives -
> another subset of folks sharing common interests that match the
> overall comdev mission.
>
> 3) I led a small group to develop our maturity model, I think it's a
> very useful tool. I think we made just one change to it in 2016, it's
> stable but useful and maintained. Others don't care about that or
> didn't have time to help - no problem.
>
> You could argue that these things are disjoint but they are all small
> steps that help towards our overall mission. We don't need much
> coordination between them, IMO just making sure the comdev PMC agrees
> with these things happening, and doing out best to unify their
> communications channels to create synergies is good enough.
>
> Comdev can just provide a space for volunteers to help develop
> communities, that's good enough for me. If others want more structured
> activities feel free to do them but don't expect all PMC members to
> necessarily join or to feel bad if they don't.
>
> -Bertrand
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Daniel Gruno <hu...@apache.org>.
On 12/05/2016 02:59 PM, Bertrand Delacretaz wrote:
> Hi,
>
> On Mon, Dec 5, 2016 at 2:30 PM, Daniel Gruno <hu...@apache.org> wrote:
>> ...The task of ComDev is developing community. If we don't have any data or
>> interest in acquiring such to show that this is in fact helping towards
>> that, then we should consider whether the current strategy is the right
>> thing to focus on....
>
> I disagree with the need for comdev to have a strategy.
>
> At the technical level the ASF doesn't have a strategy, it just
> provides space for its projects to exist and flourish.
>
> I think comdev can operate in the same way, as a loose group of
> volunteers who collectively help develop communities, without
> necessarily having a global comdev strategy to follow.
I wasn't thinking a global strategy in the stringent "we must see a 5%
growth!" sense. But we do have a mission at comdev, it's in the
resolution to establish the committee; "helping people become involved
with Apache projects.". If ComDev can't in any way account for whether
its various projects have any effect, then why are we doing them?
Having a loose management structure is fine, that's what the ASF is
about, it's a very flat organisation. But there should at least be some
form of communication/discussion going on about what we do and whether
it had any effect towards our mission of helping people become involved
in the ASF. If the mission is not considered to matter, then what is the
purpose of ComDev as a PMC? Sometimes, it seems to me - and I'll admit I
fell into that trap as well - that ComDev is just a place you put things
when you can't find another place to put it, regardless of whether it
helps our mission or not. If that's the case, then I think the overall
mission statement of comdev should change to reflect that.
>
> Three examples:
>
> 1) A small group is running GSoC, which as Uli mentions doesn't cost
> the ASF anything and actually brings some money in. GSoC clearly helps
> our mission by helping a few community members join the ASF every
> year. Exactly how many is not very important if volunteers agree to
> run it.
>
> 2) Sharan and others have started work on diversity initiatives -
> another subset of folks sharing common interests that match the
> overall comdev mission.
>
> 3) I led a small group to develop our maturity model, I think it's a
> very useful tool. I think we made just one change to it in 2016, it's
> stable but useful and maintained. Others don't care about that or
> didn't have time to help - no problem.
>
> You could argue that these things are disjoint but they are all small
> steps that help towards our overall mission. We don't need much
> coordination between them, IMO just making sure the comdev PMC agrees
> with these things happening, and doing out best to unify their
> communications channels to create synergies is good enough.
>
> Comdev can just provide a space for volunteers to help develop
> communities, that's good enough for me. If others want more structured
> activities feel free to do them but don't expect all PMC members to
> necessarily join or to feel bad if they don't.
>
> -Bertrand
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
On 12/05/2016 08:59 AM, Bertrand Delacretaz wrote:
> I disagree with the need for comdev to have a strategy.
I'd like to respond to just this remark.
ComDev has a goal. It is stated in the resolution that created the PMC.
That goal is:
to establish a Project Management Committee charged with coordinating
community development efforts.
Now, that's very poorly defined. But there are a number of us here who
have had "community manager", "community liaison", and "community
development lead" on our business cards at one point or another, and
many more of us who have had this role with respect to a project, ASF or
otherwise, over the years. I think that we can, and should, come up with
a definition of what our goal is *as a PMC*.
This is not to say that individuals can't do whatever the heck that they
want, when they want, with whom they want. But as a PMC we have an
obligation to the board and to the Foundation, to work towards the goal
stated in our founding charter. And I think we have an obligation, also,
to measure whether the things that we are doing are actually beneficial
to that goal.
We are a PMC, but we're not a traditional code-based PMC that is
producing a product. And, thus, we have, for the most part, operated in
a "let's do whatever" manner for most of our existence.
Again, I'm not suggesting that we have a 10 year plan and fire people
who don't work towards it. Although, a 10 year plan isn't such a
terrible idea. But I do think that we, as a PMC, need a much clearer
idea of our reason for existence, and that it needs to extend broader
than just participating in GSoC.
GSoC is wonderful but I would like to know whether it actually benefits
"community development." I don't really understand why asking for a
measure of that - ANY measure of that - is controversial.
However, on my other points, look for a separate thread discussing that,
momentarily.
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Bertrand Delacretaz <bd...@apache.org>.
Hi,
On Mon, Dec 5, 2016 at 2:30 PM, Daniel Gruno <hu...@apache.org> wrote:
> ...The task of ComDev is developing community. If we don't have any data or
> interest in acquiring such to show that this is in fact helping towards
> that, then we should consider whether the current strategy is the right
> thing to focus on....
I disagree with the need for comdev to have a strategy.
At the technical level the ASF doesn't have a strategy, it just
provides space for its projects to exist and flourish.
I think comdev can operate in the same way, as a loose group of
volunteers who collectively help develop communities, without
necessarily having a global comdev strategy to follow.
Three examples:
1) A small group is running GSoC, which as Uli mentions doesn't cost
the ASF anything and actually brings some money in. GSoC clearly helps
our mission by helping a few community members join the ASF every
year. Exactly how many is not very important if volunteers agree to
run it.
2) Sharan and others have started work on diversity initiatives -
another subset of folks sharing common interests that match the
overall comdev mission.
3) I led a small group to develop our maturity model, I think it's a
very useful tool. I think we made just one change to it in 2016, it's
stable but useful and maintained. Others don't care about that or
didn't have time to help - no problem.
You could argue that these things are disjoint but they are all small
steps that help towards our overall mission. We don't need much
coordination between them, IMO just making sure the comdev PMC agrees
with these things happening, and doing out best to unify their
communications channels to create synergies is good enough.
Comdev can just provide a space for volunteers to help develop
communities, that's good enough for me. If others want more structured
activities feel free to do them but don't expect all PMC members to
necessarily join or to feel bad if they don't.
-Bertrand
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Daniel Gruno <hu...@apache.org>.
On 12/05/2016 01:41 PM, Ulrich Strk wrote:
> On 03.12.16 23:16, Rich Bowen wrote:
>> So, this question was asked several weeks ago, and, so far, we have no
>> statistics. I wonder if, now that there appears to be some interest
>> around GSoC 2017 starting up, someone could look back at their notes and
>> answer some of these questions.
>>
>> While it's cool that folks are spending time on GSoC, for the sake of
>> GSoC, and of the students, it would also be awesome to know whether the
>> time and money that the ASF puts into this is actually doing something
>> towards the stated goal of Community Development.
>>
>> Or, at the very least, can we make a commitment to track this data going
>> forward?
>
> Let me play the devil's advocate here: What for?
>
> GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
> student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
> to organize it and mentor students we can assume that at least those volunteers are seeing value in
> it. Why the stats other than for satisfying our curiosity?
Let's flip that around: If the ASF doesn't benefit in any way from this,
why are we bothering with it? why is it the prominent (and in many cases
the _only_) thing in our board reports?
Going back a few reports (those that are public at this date), they all
seem to be either about starting, doing or concluding (which we have no
data on) GSoC and then perhaps a small bit of extra data on all the
rest. It is a VERY prominent thing, so why is that so?
The task of ComDev is developing community. If we don't have any data or
interest in acquiring such to show that this is in fact helping towards
that, then we should consider whether the current strategy is the right
thing to focus on.
But this goes beyond GSoC in my mind. We should be looking at ALL ComDev
projects and evaluate what we want to keep, what isn't working, and what
needs a do-over. The task of ComDev is to *develop communities*, it
shouldn't just be a dumping ground for all things cross-project, whether
they work or not. That is at least my opinion.
We try strategies, give them life, see if they work, and if not, we put
them to sleep or fix them.
With regards,
Daniel.
>
> Cheers,
>
> Uli
>
>>
>> --Rich
>>
>> On 11/16/2016 05:18 AM, Nick Burch wrote:
>>> On Wed, 16 Nov 2016, Rich Bowen wrote:
>>>> It would be great to have some kind of statistics on how GSoC helps
>>>> projects longer term. Do students stick around? Does the code written
>>>> actually get incorporated into releases? Does it in fact contribute to
>>>> the mission of Community Development, or is it just a nice summer job
>>>> for these students?
>>>
>>> Do we have a list (maybe somewhere in the comdev private svn?) of
>>> everyone who has taken part in GSoC?
>>>
>>> If so, it'd be fairly easy to annotate that with apache IDs, then see
>>> who's now on PMCs or who's now a member. Producing sharable statistics
>>> from that automatically is then easy. (I say this as someone who helped
>>> update the similar Travel Assistance Committee / TAC ones on Monday!)
>>>
>>> Not sure if it's easy to find out the "last commit date" for people, to
>>> check for the "still around" part (eg for people who got committership
>>> during GSoc, maybe on a branch), but I know where infra are to ask...
>>>
>>> Nick
>>>
>>> PS FWIW, within Tika we've certainly had a few GSoC people stick around!
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>>> For additional commands, e-mail: dev-help@community.apache.org
>>>
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Jacob Champion <ch...@gmail.com>.
I think I agree with both of you...
On 12/05/2016 09:41 PM, Ulrich Strk wrote:
> I have seen way too many metrics being misused to justify the wrong actions
...measuring the wrong thing at the wrong time or in the wrong way can
be *extremely* harmful, yes...
On 12/05/2016 01:07 PM, Daniel Gruno wrote:
> - Do we spend any ASF resources on this (I'm thinking money/paid time here)?
> - Do we enter into any form of legally binding contract with GSoC?
>
> If the answer to both are 'no', then I'll certainly not stand in the way
> of people wanting to do this - but if even one of them is 'yes', then we
> (the PMC at the very least) owe it to the foundation (and ourselves) to
> justify this on a recurring basis.
...but it is still important to justify why the ASF is doing something,
especially if contracts are on the line.
What about a softer metric to start with: "What percentage of past and
current Apache GSoC mentors feel like their involvement has been
beneficial, either for their associated Apache projects or for the
greater open source community?"
Both high and low percentages here are, IMO, instructive and useful in
their own right. A middle-of-the-road answer means we'd have to dig into
the "why do you feel that way?" question more to reach any conclusions,
but again, that's still useful.
(Compare, for example, with the metric "What % of code developed by GSoC
students actually becomes a part of the project codebase at the end of
the project?" In this case, an extremely high percentage could mean that
GSoC is "successful", or it could mean that GSoC code gets into
codebases regardless of its quality, or... Likewise, an extremely low
percentage could mean that GSoC is "unsuccessful", or it could mean that
GSoC is being used for more experimental efforts that are providing
value to projects in some other way, or...)
--Jacob
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Daniel Gruno <hu...@apache.org>.
On 12/05/2016 09:41 PM, Ulrich Strk wrote:
> On 05.12.16 17:45, Rich Bowen wrote:
>>
>>
>> On 12/05/2016 07:41 AM, Ulrich Strk wrote:
>>>> Or, at the very least, can we make a commitment to track this data going
>>>>> forward?
>>> Let me play the devil's advocate here: What for?
>>>
>>> GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
>>> student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
>>> to organize it and mentor students we can assume that at least those volunteers are seeing value in
>>> it. Why the stats other than for satisfying our curiosity?
>>
>> Because part of Community Development is measuring what you do, so that
>> you know if it's effective. This isn't a big request. I'm perplexed at
>> your resistance to it.
>>
>
> I have seen way too many metrics being misused to justify the wrong actions because metrics are
> extremely poor when it comes to more complex questions such as what we are currently trying to
> answer. IMO the only viable approach would be what you call anecdotal but which is in social
> scienses known as qualitative research which provides extremely deep insights but is also extremely
> hard (speak effort) to get right.
I could argue the opposite, but trading anecdotes about metrics isn't
gonna help much when it's missing the comdev context.
>
> How would you measure the impact on communities by students asking questions regarding a GSoC
> proposal but not being accepted? But using the knowledge they gained to promote the project at
> $dayjob later in their career, simply because they once had a short touchpoint. What does it say
> about a community if somebody becomes a committer and sticks with the project after GSoC? Is one
> more important than the other long-term? By which factor?
Those are both excellent data points that would show the value of GSoC
to the ASF. Either would be awesome if we collected them, as they would
show, to both us and others, that this has a benefit to the communities
at the ASF. I am confident we could put together a document on what we
could try to measure (I say _try_ here, as we have to start somewhere
and then see what that brings)
>
> And I'd again argue that we don't have to care too much how effective something like GSoC is as long
> as people are having fun mentoring students. Your experience might not have been fun but then again
> we are telling mentors to fail students early on when they don't think that anything useful will
> come out of it.
Let me ask two questions here:
- Do we spend any ASF resources on this (I'm thinking money/paid time here)?
- Do we enter into any form of legally binding contract with GSoC?
If the answer to both are 'no', then I'll certainly not stand in the way
of people wanting to do this - but if even one of them is 'yes', then we
(the PMC at the very least) owe it to the foundation (and ourselves) to
justify this on a recurring basis. And one way to justify it is to have
something that shows that it's actually providing something useful
towards the mission of this committee.
With regards,
Daniel.
>
> Cheers,
>
> Uli
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Ulrich Stärk <ul...@apache.org>.
On 05.12.16 17:45, Rich Bowen wrote:
>
>
> On 12/05/2016 07:41 AM, Ulrich Strk wrote:
>>> Or, at the very least, can we make a commitment to track this data going
>>>> forward?
>> Let me play the devil's advocate here: What for?
>>
>> GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
>> student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
>> to organize it and mentor students we can assume that at least those volunteers are seeing value in
>> it. Why the stats other than for satisfying our curiosity?
>
> Because part of Community Development is measuring what you do, so that
> you know if it's effective. This isn't a big request. I'm perplexed at
> your resistance to it.
>
I have seen way too many metrics being misused to justify the wrong actions because metrics are
extremely poor when it comes to more complex questions such as what we are currently trying to
answer. IMO the only viable approach would be what you call anecdotal but which is in social
scienses known as qualitative research which provides extremely deep insights but is also extremely
hard (speak effort) to get right.
How would you measure the impact on communities by students asking questions regarding a GSoC
proposal but not being accepted? But using the knowledge they gained to promote the project at
$dayjob later in their career, simply because they once had a short touchpoint. What does it say
about a community if somebody becomes a committer and sticks with the project after GSoC? Is one
more important than the other long-term? By which factor?
And I'd again argue that we don't have to care too much how effective something like GSoC is as long
as people are having fun mentoring students. Your experience might not have been fun but then again
we are telling mentors to fail students early on when they don't think that anything useful will
come out of it.
Cheers,
Uli
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
On 12/05/2016 07:41 AM, Ulrich Stärk wrote:
>> Or, at the very least, can we make a commitment to track this data going
>> > forward?
> Let me play the devil's advocate here: What for?
>
> GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
> student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
> to organize it and mentor students we can assume that at least those volunteers are seeing value in
> it. Why the stats other than for satisfying our curiosity?
Because part of Community Development is measuring what you do, so that
you know if it's effective. This isn't a big request. I'm perplexed at
your resistance to it.
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Ulrich Stärk <ul...@apache.org>.
On 03.12.16 23:16, Rich Bowen wrote:
> So, this question was asked several weeks ago, and, so far, we have no
> statistics. I wonder if, now that there appears to be some interest
> around GSoC 2017 starting up, someone could look back at their notes and
> answer some of these questions.
>
> While it's cool that folks are spending time on GSoC, for the sake of
> GSoC, and of the students, it would also be awesome to know whether the
> time and money that the ASF puts into this is actually doing something
> towards the stated goal of Community Development.
>
> Or, at the very least, can we make a commitment to track this data going
> forward?
Let me play the devil's advocate here: What for?
GSoC is completely free for the ASF (on the contrary, we even get a small amount for every accepted
student that we can than put towards fulfilling our goals) and as long as we have volunteers willing
to organize it and mentor students we can assume that at least those volunteers are seeing value in
it. Why the stats other than for satisfying our curiosity?
Cheers,
Uli
>
> --Rich
>
> On 11/16/2016 05:18 AM, Nick Burch wrote:
>> On Wed, 16 Nov 2016, Rich Bowen wrote:
>>> It would be great to have some kind of statistics on how GSoC helps
>>> projects longer term. Do students stick around? Does the code written
>>> actually get incorporated into releases? Does it in fact contribute to
>>> the mission of Community Development, or is it just a nice summer job
>>> for these students?
>>
>> Do we have a list (maybe somewhere in the comdev private svn?) of
>> everyone who has taken part in GSoC?
>>
>> If so, it'd be fairly easy to annotate that with apache IDs, then see
>> who's now on PMCs or who's now a member. Producing sharable statistics
>> from that automatically is then easy. (I say this as someone who helped
>> update the similar Travel Assistance Committee / TAC ones on Monday!)
>>
>> Not sure if it's easy to find out the "last commit date" for people, to
>> check for the "still around" part (eg for people who got committership
>> during GSoc, maybe on a branch), but I know where infra are to ask...
>>
>> Nick
>>
>> PS FWIW, within Tika we've certainly had a few GSoC people stick around!
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>> For additional commands, e-mail: dev-help@community.apache.org
>>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
On 12/05/2016 07:37 AM, Ulrich Stärk wrote:
> What money are we putting into this?
I have no idea. That would be part of the data that I, as a member,
would really like to know.
>
> Cheers,
>
> Uli
>
> On 03.12.16 23:16, Rich Bowen wrote:
>> So, this question was asked several weeks ago, and, so far, we have no
>> statistics. I wonder if, now that there appears to be some interest
>> around GSoC 2017 starting up, someone could look back at their notes and
>> answer some of these questions.
>>
>> While it's cool that folks are spending time on GSoC, for the sake of
>> GSoC, and of the students, it would also be awesome to know whether the
>> time and money that the ASF puts into this is actually doing something
>> towards the stated goal of Community Development.
>>
>> Or, at the very least, can we make a commitment to track this data going
>> forward?
>>
>> --Rich
>>
>> On 11/16/2016 05:18 AM, Nick Burch wrote:
>>> On Wed, 16 Nov 2016, Rich Bowen wrote:
>>>> It would be great to have some kind of statistics on how GSoC helps
>>>> projects longer term. Do students stick around? Does the code written
>>>> actually get incorporated into releases? Does it in fact contribute to
>>>> the mission of Community Development, or is it just a nice summer job
>>>> for these students?
>>>
>>> Do we have a list (maybe somewhere in the comdev private svn?) of
>>> everyone who has taken part in GSoC?
>>>
>>> If so, it'd be fairly easy to annotate that with apache IDs, then see
>>> who's now on PMCs or who's now a member. Producing sharable statistics
>>> from that automatically is then easy. (I say this as someone who helped
>>> update the similar Travel Assistance Committee / TAC ones on Monday!)
>>>
>>> Not sure if it's easy to find out the "last commit date" for people, to
>>> check for the "still around" part (eg for people who got committership
>>> during GSoc, maybe on a branch), but I know where infra are to ask...
>>>
>>> Nick
>>>
>>> PS FWIW, within Tika we've certainly had a few GSoC people stick around!
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>>> For additional commands, e-mail: dev-help@community.apache.org
>>>
>>
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Ulrich Stärk <ul...@apache.org>.
What money are we putting into this?
Cheers,
Uli
On 03.12.16 23:16, Rich Bowen wrote:
> So, this question was asked several weeks ago, and, so far, we have no
> statistics. I wonder if, now that there appears to be some interest
> around GSoC 2017 starting up, someone could look back at their notes and
> answer some of these questions.
>
> While it's cool that folks are spending time on GSoC, for the sake of
> GSoC, and of the students, it would also be awesome to know whether the
> time and money that the ASF puts into this is actually doing something
> towards the stated goal of Community Development.
>
> Or, at the very least, can we make a commitment to track this data going
> forward?
>
> --Rich
>
> On 11/16/2016 05:18 AM, Nick Burch wrote:
>> On Wed, 16 Nov 2016, Rich Bowen wrote:
>>> It would be great to have some kind of statistics on how GSoC helps
>>> projects longer term. Do students stick around? Does the code written
>>> actually get incorporated into releases? Does it in fact contribute to
>>> the mission of Community Development, or is it just a nice summer job
>>> for these students?
>>
>> Do we have a list (maybe somewhere in the comdev private svn?) of
>> everyone who has taken part in GSoC?
>>
>> If so, it'd be fairly easy to annotate that with apache IDs, then see
>> who's now on PMCs or who's now a member. Producing sharable statistics
>> from that automatically is then easy. (I say this as someone who helped
>> update the similar Travel Assistance Committee / TAC ones on Monday!)
>>
>> Not sure if it's easy to find out the "last commit date" for people, to
>> check for the "still around" part (eg for people who got committership
>> during GSoc, maybe on a branch), but I know where infra are to ask...
>>
>> Nick
>>
>> PS FWIW, within Tika we've certainly had a few GSoC people stick around!
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
>> For additional commands, e-mail: dev-help@community.apache.org
>>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org
Re: Does GSoC help develop communities?
Posted by Rich Bowen <rb...@rcbowen.com>.
So, this question was asked several weeks ago, and, so far, we have no
statistics. I wonder if, now that there appears to be some interest
around GSoC 2017 starting up, someone could look back at their notes and
answer some of these questions.
While it's cool that folks are spending time on GSoC, for the sake of
GSoC, and of the students, it would also be awesome to know whether the
time and money that the ASF puts into this is actually doing something
towards the stated goal of Community Development.
Or, at the very least, can we make a commitment to track this data going
forward?
--Rich
On 11/16/2016 05:18 AM, Nick Burch wrote:
> On Wed, 16 Nov 2016, Rich Bowen wrote:
>> It would be great to have some kind of statistics on how GSoC helps
>> projects longer term. Do students stick around? Does the code written
>> actually get incorporated into releases? Does it in fact contribute to
>> the mission of Community Development, or is it just a nice summer job
>> for these students?
>
> Do we have a list (maybe somewhere in the comdev private svn?) of
> everyone who has taken part in GSoC?
>
> If so, it'd be fairly easy to annotate that with apache IDs, then see
> who's now on PMCs or who's now a member. Producing sharable statistics
> from that automatically is then easy. (I say this as someone who helped
> update the similar Travel Assistance Committee / TAC ones on Monday!)
>
> Not sure if it's easy to find out the "last commit date" for people, to
> check for the "still around" part (eg for people who got committership
> during GSoc, maybe on a branch), but I know where infra are to ask...
>
> Nick
>
> PS FWIW, within Tika we've certainly had a few GSoC people stick around!
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
> For additional commands, e-mail: dev-help@community.apache.org
>
--
Rich Bowen - rbowen@rcbowen.com - @rbowen
http://apachecon.com/ - @apachecon
Re: Does GSoC help develop communities?
Posted by Nick Burch <ni...@apache.org>.
On Wed, 16 Nov 2016, Rich Bowen wrote:
> It would be great to have some kind of statistics on how GSoC helps
> projects longer term. Do students stick around? Does the code written
> actually get incorporated into releases? Does it in fact contribute to
> the mission of Community Development, or is it just a nice summer job
> for these students?
Do we have a list (maybe somewhere in the comdev private svn?) of everyone
who has taken part in GSoC?
If so, it'd be fairly easy to annotate that with apache IDs, then see
who's now on PMCs or who's now a member. Producing sharable statistics
from that automatically is then easy. (I say this as someone who helped
update the similar Travel Assistance Committee / TAC ones on Monday!)
Not sure if it's easy to find out the "last commit date" for people, to
check for the "still around" part (eg for people who got committership
during GSoc, maybe on a branch), but I know where infra are to ask...
Nick
PS FWIW, within Tika we've certainly had a few GSoC people stick around!
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@community.apache.org
For additional commands, e-mail: dev-help@community.apache.org