Committers

[Niclas Hedhman <ni...@hedhman.org>]

Hi,
I am not sure, but can I propose myself as a committer at the Felix project??

I have the feeling that although we have a whole bunch of committers 
officially in the project, that this is essentially a one-man show (sorry 
Tim, I know you have put in efforts as well).

As the list of "Todo" items posted by Richard on Jira indicates, we need a lot 
of hands to pull this bird out of its cage. I know I can help a lot.


Cheers
Niclas

Re: Committers

[Alex Karasulu <ao...@bellsouth.net>]

Ralph Goers wrote:

> Alex Karasulu wrote:

...

> IMO, existing Apache committers deserve a little bit lower bar of 
> entry - especially into an incubator project - simply because they 
> already know how Apache works.
>
The presumption that committers on other projects know how Apache works 
is incorrect.

Yes generally there is a way lower bar for accepting committers on 
incubator podlings and yes this bar is even lower for those that are 
already committers on other Apache projects.  I am familiar with Niclas' 
background since I myself started a vote to grant him karma on another 
Apache project in the past: Apache Directory to be exact.

But this does not mean that every committer understands the Apache way 
and we accept them as is for every project across the ASF.  With that 
said ...

This is not the place to discuss these particulars.  The ppmc list has 
been errected specifically for these discussions.

felix-private@incubator.apache.org

Alex

Re: Committers

[Greg Stein <gs...@apache.org>]

On Tue, Oct 11, 2005 at 02:34:36PM -0700, Ralph Goers wrote:
> Alex Karasulu wrote:
>...
> >         It's also a 
> >good sign of a healthy community.  Frequent patch submissions force 
> >committers to review your patch and get involved.   Increased positive 
> >community interaction is what we want.
>
> Maybe I'm a little overly sensitive, but I found your reply a little 
> condescending.  Niclas is already an Apache committer and I'm sure he 
> already knows all about community at Apache but your response sounds 
> like you don't know that.

While it may sound like that at first blush, there is quite a bit of
history that leads up to Alex's statements. I think that it's fair to say
that Alex's "reminder" of normal Apache process is fine.

> He participates in a number of mailing lists.  
> The sticking point is that he didn't follow the accepted practice of 
> being invited into the community.  However, IIRC all the current felix 
> committers spoke up and asked to be committers when this group was first 
> formed.  So Niclas' "error" here seems to be simply a matter of not 
> speaking up soon enough. 

The community is here, present, and established. Normal meritocratic rules
should start. Yeah, there is the unfortunate "race condition" that is
present; not sure how to solve that really except to possibly consider the
first couple weeks "open". We're a bit past that.

But note that it isn't really a big deal here. Sending a patch is just
'svn diff | mail' :-). And the PPMC can always discuss to make him a
committer.

> And one of the signs of a healthy community is welcoming those who offer 
> to pitch in and help.  IMO, existing Apache committers deserve a little 
> bit lower bar of entry - especially into an incubator project - simply 
> because they already know how Apache works.

That's the normal working assumption, yes. And ASF Members have
practically zero bar for any project since it is fair to say that the
Members represent the ASF which is responsible for all code, so a Member
is simply exercising their responsibility to care for any/all the code at
the ASF.

But back to the point. felix-dev has existed since July. It should be
operating according to meritocratic rules now. And that means: send in
some patches. One of the purposes of doing that for a new committer is to
ensure that the person and the project are in alignment on the concepts
and direction of the project. It is prudent to ensure that everybody in
the community is reasonably aligned on the course for a project, and
especially for one in the Incubator. Unlike (say) httpd, they cannot
really withstand people going in several directions at once, nor would you
want to try that.

Patches serve other purposes, of course, but much of that *is* moot given
Niclas' participation in other Apache projects.

In any case, I'd say to just leave it to the PPMC to figure out. Unless
and until they give somebody commit rights, then the person should just
start submitting patches.

Cheers,
-g

-- 
gstein@apache.org ... ASF Chairman ... http://www.apache.org/

Re: Committers

[Ralph Goers <Ra...@dslextreme.com>]

Alex Karasulu wrote:

>
> I would gauge the size of the contribution as well as the complexity 
> of the task before picking an approach.   However we (Apache) got our 
> name because the httpd server was patched so frequently.  So don't 
> fear sending in several patches: its a tradition :).   It's also a 
> good sign of a healthy community.  Frequent patch submissions force 
> committers to review your patch and get involved.   Increased positive 
> community interaction is what we want.
>
Maybe I'm a little overly sensitive, but I found your reply a little 
condescending.  Niclas is already an Apache committer and I'm sure he 
already knows all about community at Apache but your response sounds 
like you don't know that. He participates in a number of mailing lists.  
The sticking point is that he didn't follow the accepted practice of 
being invited into the community.  However, IIRC all the current felix 
committers spoke up and asked to be committers when this group was first 
formed.  So Niclas' "error" here seems to be simply a matter of not 
speaking up soon enough. 

And one of the signs of a healthy community is welcoming those who offer 
to pitch in and help.  IMO, existing Apache committers deserve a little 
bit lower bar of entry - especially into an incubator project - simply 
because they already know how Apache works.

RE: Committers

["Noel J. Bergman" <no...@devtech.com>]

Alex Karasulu wrote:

> the project just entered the incubator but what's your point?  Are
> you suggesting that using felix commercially gives you a free pass
> as a committer on this project?

I've been meaning to jump in on this, so that's as good a hook as any.
Unrelated to the merits of this request, so neither in favor nor against, it
has been a recommendation that projects adopt a lower bar to entry during
Incubation than they might otherwise employ, in order to help more rapidly
grow their community.  Community building is one of the most important
aspects of Incubation.

	--- Noel

Re: Committers

[Alex Karasulu <ao...@bellsouth.net>]

Niclas Hedhman wrote:

>On Tuesday 11 October 2005 22:11, Alex Karasulu wrote:
>  
>
>>We do need some action but lack of committership does not prevent action.
>>    
>>
>
>Action is not lacking either at my end. 
>
I never commented on your lack of action.  Did I?

>Not sure how many people are trying 
>Felix out in the "real world", but I bet it ain't that many yet.
>  
>
Well the project just entered the incubator but what's your point?  Are 
you suggesting that using felix commercially gives you a free pass as a 
committer on this project? 

Contributions, not just code contributions, and the ability to work well 
with others in a community are  taken into consideration when 
determining whether or not someone deserves karma.  The right attitude, 
as we saw from Avalon, is critical.

>Nevertheless, I intend to complete the "JCL/Log4J LogService" in the next few 
>days and have the bulk of the code in place. Now, please advice on path 
>forward;
>  
>
First off that's great!  ...

> a. I continue to work in a vacuum, and submit everything in one big shabang
>    to the community, when everything is ready.
>  
>
Depends, as you know, on the size of the undertaking.  This specific 
task seems small, simple and straight forward so it could be submitted 
as a single contribution.  The code perhaps is not as important as your 
discussion of your approach on list.  Engaging the community in some 
fashion is "proper" conduct.  Again its not a 0 or 1 where you either 
work in a vacuum or not?  In the end suggestions can be factored into 
the code you submit.  Then the code can be reviewed after you submit 
your patch.  It's much more organic as you will see.

> b. I submit the changes as I go along, perhaps on a daily basis, and allow
>    more "review" (if any) of what I have done and what could be improved.
>
>  
>
I would gauge the size of the contribution as well as the complexity of 
the task before picking an approach.   However we (Apache) got our name 
because the httpd server was patched so frequently.  So don't fear 
sending in several patches: its a tradition :).   It's also a good sign 
of a healthy community.  Frequent patch submissions force committers to 
review your patch and get involved.   Increased positive community 
interaction is what we want.

Just make sure your submission does not break the build or leave the 
code in an inconsistent state.  For larger efforts you can contribute 
islands of functionality until things are ready to be assembled.

>Furthermore, please advice on what build system approach I should take for 
>this. "framework" is plain and simple Ant, but Tim has been working on a 
>Maven2 plugin for OSGi bundles. Not sure of the status of that.
>  
>
This is not my call to make.  It's the responsibility of this community 
to decide that.   Looks like this is a good time to open up another 
thread and get to the bottom if this ;). 

Until the dust settles and an approach is ratified, I guess I would use 
what integrates best with what we have presently in place.  If its ant, 
use ant.  If its maven, use maven.  The logging service does not sound 
like that big of a build undertaking ... scrapping an ant build.xml or 
maven POM would not be that much of a loss.

HTH,
Alex

Re: Committers

[Niclas Hedhman <ni...@apache.org>]

On Tuesday 11 October 2005 22:11, Alex Karasulu wrote:
> We do need some action but lack of committership does not prevent action.

Action is not lacking either at my end. Not sure how many people are trying 
Felix out in the "real world", but I bet it ain't that many yet.

Nevertheless, I intend to complete the "JCL/Log4J LogService" in the next few 
days and have the bulk of the code in place. Now, please advice on path 
forward;

 a. I continue to work in a vacuum, and submit everything in one big shabang
    to the community, when everything is ready.

 b. I submit the changes as I go along, perhaps on a daily basis, and allow
    more "review" (if any) of what I have done and what could be improved.

Furthermore, please advice on what build system approach I should take for 
this. "framework" is plain and simple Ant, but Tim has been working on a 
Maven2 plugin for OSGi bundles. Not sure of the status of that.



Cheers
Niclas

Re: Committers

[Alex Karasulu <ao...@bellsouth.net>]

Daniel Fagerstrom wrote:

> Niclas Hedhman wrote:
>
>> Hi,
>> I am not sure, but can I propose myself as a committer at the Felix 
>> project??
>>  
>>
> As a Apache veteran you should know that propsing one self as a 
> committer generally is supposed to mean that one havn't understand how 
> things work in Apache yet ;)
>
Ditto! I'd like to see some consistent contributions and the 
demonstration of a better understanding of how things work here at the 
ASF before considering Niclas as a committer. 

Also these sort of things are best discussed on the ppmc list hence the 
reason why committers should not be going around asking or suggesting 
they have karma on public lists.

We do need some action but lack of committership does not prevent action. 

-1 for now ... further considerations should be made on the ppmc list in 
the future.

Thanks,
Alex

Re: Committers

[Niclas Hedhman <ni...@hedhman.org>]

On Tuesday 11 October 2005 16:27, Daniel Fagerstrom wrote:
> Niclas Hedhman wrote:
> > I am not sure, but can I propose myself as a committer at the Felix
> > project?? 
> As a Apache veteran you should know that propsing one self as a 
> committer generally is supposed to mean that one havn't understand how
> things work in Apache yet ;)

Granted, and I did put it as a question. I have seen "committer of project X" 
requesting "committership in project Y" before, and that is what triggered 
this move, combined with "I need more action" as we are quickly moving into 
OSGi-land with commercial development.

Cheers
Niclas

Re: Committers

[Daniel Fagerstrom <da...@nada.kth.se>]

Niclas Hedhman wrote:

>Hi,
>I am not sure, but can I propose myself as a committer at the Felix project??
>  
>
As a Apache veteran you should know that propsing one self as a 
committer generally is supposed to mean that one havn't understand how 
things work in Apache yet ;)

>I have the feeling that although we have a whole bunch of committers 
>officially in the project, that this is essentially a one-man show (sorry 
>Tim, I know you have put in efforts as well).
>
>As the list of "Todo" items posted by Richard on Jira indicates, we need a lot 
>of hands to pull this bird out of its cage. I know I can help a lot.
>  
>
Anyway, +1 (non binding, I'm not a committer), we really need to start 
getting some more action.

/Daniel

Re: Committers

[Alex Karasulu <ao...@bellsouth.net>]

Niclas Hedhman wrote:

>Hi,
>I am not sure, but can I propose myself as a committer at the Felix project??
>
>I have the feeling that although we have a whole bunch of committers 
>officially in the project, that this is essentially a one-man show (sorry 
>Tim, I know you have put in efforts as well).
>
>As the list of "Todo" items posted by Richard on Jira indicates, we need a lot 
>of hands to pull this bird out of its cage. I know I can help a lot.
>  
>
Excuse the fact that these emails are out of order.

Please help out and submit patches.  I and others will commit them for you.

Thanks,
Alex