Re: ActiveMQ Graduation From Incubator

2006-03-17 Thread David Blevins

There are some good ideas in here.  Though I don't see Alan complaining.

I do see that Alan did compiled a list (STATUS file), pointed to it,  
and sent the list out to people asking for feedback and discussion.


Seems like a positive start.

-David

On Mar 17, 2006, at 9:33 AM, Leo Simons wrote:


Alan,

Incubation is something incubating communities have to do, and  
something

incubating communities are responsible for. Those communities get some
help and guidance from their mentors and the people on the
[EMAIL PROTECTED] mailing list, but never enough since most of  
those people

are volunteers with other things to do with their free time.

(...)

What is not fair is casting aside a few months of e-mail and face- 
to-face
history of various people trying to help with this incubation  
thing, stamp

your feet once every few weeks, and demand that people go and make a
specific list of specific tasks you need to do. This is now the  
third time
I've seen you do this and it is the third time I'm telling you this  
is not

how it works.

(...)

Here's a list of things to do (subjectively, none of these are easy):

 * stop complaining. Right now. It is not fair.

 * compile your own list, try to make it as extensive as possible.
   mail-archives.apache.org is your friend, people have spent hundreds
   of hours writing hundreds of e-mails to explain this to you and to
   those that came before you.

 * send the list out to people (like [EMAIL PROTECTED]) for feedback
   and discussion.

 * work to address the list.

 * keep a record of this work.

 * point to the record (STATUS file).

 * spend time explaining concisely in a format processable by humans
   during a concall, what is in this record, what changed, etc, and
   send this in time when Noel asks for a report for the board  
meeting.


 * look back on this process and document what you learned so others
   can benefit from it.

The idea that ActiveMQ as a community (not the software, I have no
clue about the software) is ready to leave the incubator, is well,
awkward. The very fact that there are long e-mail threads like this
everytime I look at [EMAIL PROTECTED] should be enough indication that
it is not.

LSD

On Wed, Mar 15, 2006 at 06:28:12PM -0800, Alan D. Cabrera wrote:

Noel J. Bergman wrote:

Alan D. Cabrera wrote:



I only see infrastructure issues in your list of concerns
that would prevent the graduation of ActiveMQ.



Look again, but also at comments from Dims, Henri and others.



At the moment, only Dims has taken the time to enumerate a list of
concerns.  Henri and the others have provided well thought out  
points on
the definition of umbrella projects and whether AMQ should be a  
TLP or

subproject; these not really being impediments to graduation but the
necessary discourse about the final disposition of AMQ when it  
graduates

that I was looking for when I initially sent out my email.

You express an opinion that it should be a TLP but mention that  
it has a

long way to go before it's ready for that.  Can you enumerate what
remains, aside from the infrastructure issues



See my reply to Dain.  And I do feel that some of it does come  
down to

being
able to convey a subjective confidence to the Incubator PMC that the
community really does "get it" regarding ASF principles and  
practices.  And

that is supposed to happen before, not after, a community leaves the
Incubator.



There are a number of definitions for the word "subjective".  If
subjective means that your concerns may be peculiar to yourself,  
can you

not explicitly state what you'd like to see?  If you are unable to
communicate what those are, we may not unable to address them.  Is  
that

fair to the AMQ community?


If AMQ has less inspiring aspirations and was to initially land
as a sub-project



I am not sure how much difference there ought to be, but some of  
that comes

down to the landing PMC.  I do have a concern an issue of fairness.

Consider David Blevin's well-stated views, including "We've more  
or less
been running as TLPs [for] the past two plus years already."  So  
if we have
some community that has been autonomous, and it becomes part of  
another TLP
within the ASF, how fair would it be for the members of that  
community to
lose their decision making ability?  I would say not, so are they  
going to
be made part of the destination PMC, which would be required for  
them to

have binding votes?

This is a generic issue.  I would have to cross-reference in  
detail the PMC
and committer lists for ActiveMQ and Geronimo to be specific to  
this case.

I do realize that there is overlap, but also others who are part of
ActiveMQ
and are not part of Geronimo.  Is Geronimo prepared to welcome  
them as

Committers on the Geronimo TLP and members of the Geronimo PMC?

Related comment will go as a reply to David Blevins.




If I take away the list of infrastructure issues, I only see the  
need to
have a thorough discussion as to where

Re: ActiveMQ Graduation From Incubator

2006-03-17 Thread Hiram Chirino
Leo,

Many of the folks in the ActiveMQ project already have been through
the incubation process once before when we put Geronimo though.  It's
not like this is our first rodeo.  So in our eyes we really do think
we are very close to having satisfied the incubation requirements.  I
think Alan was opening up the discussion to get constructive feed back
on what people feel is missing.

For example, you have an opinion that "The idea that ActiveMQ as a
community is ready to leave the incubator, is well, awkward."  You are
aware that this was a community that was started by Apache committers
and run very much in the Apache meritocratic style when it was the
codehaus right?  So in sort, it would be nice for you to explain this
opinion a little more.

Regards,
Hiram

On 3/17/06, Leo Simons <[EMAIL PROTECTED]> wrote:
> Alan,
>
> Incubation is something incubating communities have to do, and something
> incubating communities are responsible for. Those communities get some
> help and guidance from their mentors and the people on the
> [EMAIL PROTECTED] mailing list, but never enough since most of those people
> are volunteers with other things to do with their free time.
>
> (...)
>
> What is not fair is casting aside a few months of e-mail and face-to-face
> history of various people trying to help with this incubation thing, stamp
> your feet once every few weeks, and demand that people go and make a
> specific list of specific tasks you need to do. This is now the third time
> I've seen you do this and it is the third time I'm telling you this is not
> how it works.
>
> (...)
>
> Here's a list of things to do (subjectively, none of these are easy):
>
>  * stop complaining. Right now. It is not fair.
>
>  * compile your own list, try to make it as extensive as possible.
>mail-archives.apache.org is your friend, people have spent hundreds
>of hours writing hundreds of e-mails to explain this to you and to
>those that came before you.
>
>  * send the list out to people (like [EMAIL PROTECTED]) for feedback
>and discussion.
>
>  * work to address the list.
>
>  * keep a record of this work.
>
>  * point to the record (STATUS file).
>
>  * spend time explaining concisely in a format processable by humans
>during a concall, what is in this record, what changed, etc, and
>send this in time when Noel asks for a report for the board meeting.
>
>  * look back on this process and document what you learned so others
>can benefit from it.
>
> The idea that ActiveMQ as a community (not the software, I have no
> clue about the software) is ready to leave the incubator, is well,
> awkward. The very fact that there are long e-mail threads like this
> everytime I look at [EMAIL PROTECTED] should be enough indication that
> it is not.
>
> LSD
>
> On Wed, Mar 15, 2006 at 06:28:12PM -0800, Alan D. Cabrera wrote:
> > Noel J. Bergman wrote:
> > >Alan D. Cabrera wrote:
> > >
> > >
> > >>I only see infrastructure issues in your list of concerns
> > >>that would prevent the graduation of ActiveMQ.
> > >>
> > >
> > >Look again, but also at comments from Dims, Henri and others.
> > >
> >
> > At the moment, only Dims has taken the time to enumerate a list of
> > concerns.  Henri and the others have provided well thought out points on
> > the definition of umbrella projects and whether AMQ should be a TLP or
> > subproject; these not really being impediments to graduation but the
> > necessary discourse about the final disposition of AMQ when it graduates
> > that I was looking for when I initially sent out my email.
> >
> > >>You express an opinion that it should be a TLP but mention that it has a
> > >>long way to go before it's ready for that.  Can you enumerate what
> > >>remains, aside from the infrastructure issues
> > >>
> > >
> > >See my reply to Dain.  And I do feel that some of it does come down to
> > >being
> > >able to convey a subjective confidence to the Incubator PMC that the
> > >community really does "get it" regarding ASF principles and practices.  And
> > >that is supposed to happen before, not after, a community leaves the
> > >Incubator.
> > >
> >
> > There are a number of definitions for the word "subjective".  If
> > subjective means that your concerns may be peculiar to yourself, can you
> > not explicitly state what you'd like to see?  If you are unable to
> > communicate what those are, we may not unable to address them.  Is that
> > fair to the AMQ community?
> >
> > >>If AMQ has less inspiring aspirations and was to initially land
> > >>as a sub-project
> > >>
> > >
> > >I am not sure how much difference there ought to be, but some of that comes
> > >down to the landing PMC.  I do have a concern an issue of fairness.
> > >
> > >Consider David Blevin's well-stated views, including "We've more or less
> > >been running as TLPs [for] the past two plus years already."  So if we have
> > >some community that has been autonomous, and it becomes part of another TLP
> > >within the ASF, how f

Re: ActiveMQ Graduation From Incubator

2006-03-17 Thread Leo Simons
Alan,

Incubation is something incubating communities have to do, and something
incubating communities are responsible for. Those communities get some
help and guidance from their mentors and the people on the
[EMAIL PROTECTED] mailing list, but never enough since most of those people
are volunteers with other things to do with their free time.

(...)

What is not fair is casting aside a few months of e-mail and face-to-face
history of various people trying to help with this incubation thing, stamp
your feet once every few weeks, and demand that people go and make a
specific list of specific tasks you need to do. This is now the third time
I've seen you do this and it is the third time I'm telling you this is not
how it works.

(...)

Here's a list of things to do (subjectively, none of these are easy):

 * stop complaining. Right now. It is not fair.

 * compile your own list, try to make it as extensive as possible.
   mail-archives.apache.org is your friend, people have spent hundreds
   of hours writing hundreds of e-mails to explain this to you and to
   those that came before you.

 * send the list out to people (like [EMAIL PROTECTED]) for feedback
   and discussion.

 * work to address the list.

 * keep a record of this work.

 * point to the record (STATUS file).

 * spend time explaining concisely in a format processable by humans
   during a concall, what is in this record, what changed, etc, and
   send this in time when Noel asks for a report for the board meeting.

 * look back on this process and document what you learned so others
   can benefit from it.

The idea that ActiveMQ as a community (not the software, I have no
clue about the software) is ready to leave the incubator, is well,
awkward. The very fact that there are long e-mail threads like this
everytime I look at [EMAIL PROTECTED] should be enough indication that
it is not.

LSD

On Wed, Mar 15, 2006 at 06:28:12PM -0800, Alan D. Cabrera wrote:
> Noel J. Bergman wrote:
> >Alan D. Cabrera wrote:
> >
> >  
> >>I only see infrastructure issues in your list of concerns
> >>that would prevent the graduation of ActiveMQ.
> >>
> >
> >Look again, but also at comments from Dims, Henri and others.
> >  
> 
> At the moment, only Dims has taken the time to enumerate a list of 
> concerns.  Henri and the others have provided well thought out points on 
> the definition of umbrella projects and whether AMQ should be a TLP or 
> subproject; these not really being impediments to graduation but the 
> necessary discourse about the final disposition of AMQ when it graduates 
> that I was looking for when I initially sent out my email.
> 
> >>You express an opinion that it should be a TLP but mention that it has a
> >>long way to go before it's ready for that.  Can you enumerate what
> >>remains, aside from the infrastructure issues
> >>
> >
> >See my reply to Dain.  And I do feel that some of it does come down to 
> >being
> >able to convey a subjective confidence to the Incubator PMC that the
> >community really does "get it" regarding ASF principles and practices.  And
> >that is supposed to happen before, not after, a community leaves the
> >Incubator.
> >  
> 
> There are a number of definitions for the word "subjective".  If 
> subjective means that your concerns may be peculiar to yourself, can you 
> not explicitly state what you'd like to see?  If you are unable to 
> communicate what those are, we may not unable to address them.  Is that 
> fair to the AMQ community?
> 
> >>If AMQ has less inspiring aspirations and was to initially land
> >>as a sub-project
> >>
> >
> >I am not sure how much difference there ought to be, but some of that comes
> >down to the landing PMC.  I do have a concern an issue of fairness.
> >
> >Consider David Blevin's well-stated views, including "We've more or less
> >been running as TLPs [for] the past two plus years already."  So if we have
> >some community that has been autonomous, and it becomes part of another TLP
> >within the ASF, how fair would it be for the members of that community to
> >lose their decision making ability?  I would say not, so are they going to
> >be made part of the destination PMC, which would be required for them to
> >have binding votes?
> >
> >This is a generic issue.  I would have to cross-reference in detail the PMC
> >and committer lists for ActiveMQ and Geronimo to be specific to this case.
> >I do realize that there is overlap, but also others who are part of 
> >ActiveMQ
> >and are not part of Geronimo.  Is Geronimo prepared to welcome them as
> >Committers on the Geronimo TLP and members of the Geronimo PMC?
> >
> >Related comment will go as a reply to David Blevins.
> >
> > 
> 
> If I take away the list of infrastructure issues, I only see the need to 
> have a thorough discussion as to where AMQ will land when it graduates.  
> Once this settles down and we, hopefully, reach a consensus we will be 
> ready to vote, imho.
> 
> 
> 
> Regards,
> Alan
> 
> 


Re: ActiveMQ Graduation From Incubator

2006-03-16 Thread Dain Sundstrom

+1

I couldn't have said it better myself.

-dain

On Mar 15, 2006, at 4:27 PM, David Blevins wrote:

If you ask me what my opinion on OpenEJB's future or James' opinion  
on ActiveMQ's future, we'll both probably tell you TLP is a good  
goal eventually.


We've more or less been running as TLPs in relation to Geronimo for  
the past two plus years already, just at Codehaus.  We've seen how  
that plays out and we'd like to try a much more unified front for a  
while and see what shakes out.  We're all ready for a change in the  
status-quo.


The Geronimo world is awkward and unbalanced.  We have too tight  
integration with OpenEJB such that the standalone version of that  
completely disappeared.  We have too little integration with  
ActiveMQ such that the things you can do with standalone ActiveMQ  
you can't do with the Geronimo ActiveMQ.  We have parts of Geronimo  
which could very well become separately reusable components, like  
the transaction manager or the XBean code.  We're aren't  
successfully leveraging each other's communities to the fullest.   
All in all, we don't make decisions together and lean on each other  
as much as we could.


I'd really like to see all our projects rolled up, balanced out,  
then split up again along possibly different lines with potentially  
more standalone pieces than we see now.


TLP is a good goal, but I really see value in the journey.

-David




Re: ActiveMQ Graduation From Incubator

2006-03-16 Thread Dain Sundstrom


On Mar 15, 2006, at 3:32 PM, Noel J. Bergman wrote:


Dain Sundstrom wrote:


Our goal when starting the incubation process of ActiveMQ, OpenEJB,
ServiceMix, WADI, and XBean, was to consolidate the Geronimo
community.


Consolidating the community is a good thing.  I've long wanted to  
see a

number of those projects at the ASF.

The vision was to have a single community focused on building a  
modular

server architecture based on a single core.


No disagreement.  The question at hand is simple and specific.  Are  
you
going to actually make one community from the bunch, with everyone  
having

access to work on every piece of code?  Are you going to have a large,
single, PMC with everyone having binding decisions over every  
aspect of the

project?

THAT is a TLP.


Yes, that is exactly what we are talking about.


each of the sub projects would be deliverable as a standalone
(basically the core with one plugin installed).


Separate from destination, but that sounded fine right up to the  
last point.
What is the core?  If I just want ActiveMQ, or I just want OpenEJB,  
or I
just want ... can I get it, or do I have to take some code Geronimo  
stuff,
too?  From what Alan Cabrera said last night, it sounded as if  
there have
been some complaints about that, specifically in his example with  
OpenEJB.

Or in my case, if we want to use ActiveMQ for JMS in James, what other
non-ActiveMQ bits would we required to deal with in order to get  
the code

that was apparently separable earlier?


The problem is not sharing a single core, it is that the core we  
have, GBean, is too intrusive.  XBean was designed to not be intrusive.



end up with just a bunch of separate TLPs because of some
unknown apache rules.


I'm expressing a personal opinion that the projects would be better  
off as
TLPs that collaborate with Geronimo.  From what I am being told,  
that is not
an uncommon view within those communities, although there are  
questions as
to whether to go TLP directly or via Geronimo.  Perhaps that ought  
to be put
forward for the individuals who make up the communities to directly  
answer?

:)


I think you are starting to see the responses. :)

-dain




Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Alan D. Cabrera wrote:
>
> I don't see any rush here.

I think your initial 'what do we need to work on in order
to eventually graduate?' message got interpreted by some
- -- probably myself included -- as a 'what are the last
items to check off so we can graduate?' message.  And
from there emotions flared.  Sorry to say, I think Dain's
comment:

> I understand this concern and agree with the solution, but we should
> remember that AMQ entered the incubator before this was a rule, so I
> for one didn't think it appled to them, since they are so close to
> graduation.

unluckily came at the perfect time to reinforce the wrong message.

> Is it not a natural question to ask what else is left to do?

It absolutely is.

> Is there a compelling reason to keep this in the incubator if there
> is no real need other than historical precedence of previous
> podlings' matriculation?

No.  But that's the wrong question.  'Is there a real need to keep
this in the incubator, and, if so, what is it?'
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBjiw5rNPMCpn3XdAQL6lQP+Pa7TWNs6BO57HAbSkeerPFS3k7ZRfqRV
U/MyaPM2iiCaaXQaIhEBdJrXC/tlT3SKOOgJ9KzxZyfBJkC/9Vu77z35LgI+uIEq
Dt+2/rQvU6EeENYiz0eN6RG1RhthHBfremhM1ONr/SMND35QaxZQQIAycm4zBS0h
guuwMdwjxxY=
=ZQYY
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Alan D. Cabrera




Noel J. Bergman wrote:

  Alan D. Cabrera wrote:

  
  
I only see infrastructure issues in your list of concerns
that would prevent the graduation of ActiveMQ.

  
  
Look again, but also at comments from Dims, Henri and others.
  


At the moment, only Dims has taken the time to enumerate a list of
concerns.  Henri and the others have provided well thought out points
on the definition of umbrella projects and whether AMQ should be a TLP
or subproject; these not really being impediments to graduation but the
necessary discourse about the final disposition of AMQ when it
graduates that I was looking for when I initially sent out my email.


  
You express an opinion that it should be a TLP but mention that it has a
long way to go before it's ready for that.  Can you enumerate what
remains, aside from the infrastructure issues

  
  
See my reply to Dain.  And I do feel that some of it does come down to being
able to convey a subjective confidence to the Incubator PMC that the
community really does "get it" regarding ASF principles and practices.  And
that is supposed to happen before, not after, a community leaves the
Incubator.
  


There are a number of definitions for the word "subjective".  If
subjective means that your concerns may be peculiar to yourself, can
you not explicitly state what you'd like to see?  If you are unable to
communicate what those are, we may not unable to address them.  Is that
fair to the AMQ community?


  
If AMQ has less inspiring aspirations and was to initially land
as a sub-project

  
  
I am not sure how much difference there ought to be, but some of that comes
down to the landing PMC.  I do have a concern an issue of fairness.

Consider David Blevin's well-stated views, including "We've more or less
been running as TLPs [for] the past two plus years already."  So if we have
some community that has been autonomous, and it becomes part of another TLP
within the ASF, how fair would it be for the members of that community to
lose their decision making ability?  I would say not, so are they going to
be made part of the destination PMC, which would be required for them to
have binding votes?

This is a generic issue.  I would have to cross-reference in detail the PMC
and committer lists for ActiveMQ and Geronimo to be specific to this case.
I do realize that there is overlap, but also others who are part of ActiveMQ
and are not part of Geronimo.  Is Geronimo prepared to welcome them as
Committers on the Geronimo TLP and members of the Geronimo PMC?

Related comment will go as a reply to David Blevins.

	


If I take away the list of infrastructure issues, I only see the need
to have a thorough discussion as to where AMQ will land when it
graduates.  Once this settles down and we, hopefully, reach a consensus
we will be ready to vote, imho.



Regards,
Alan






Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Alan D. Cabrera




Noel J. Bergman wrote:

  James Strachan wrote:

  
  
What other issues are there?

  
  
A number of infrastucture issues.  Votes from the Incubator PMC and Geronimo
PMC.  To do that responsibly, I'd say that we would want to see communities
having demonstrated that they understand how to practice as an ASF
community.  Such things are subjective, and typically have taken some time.

Looking at the historical record: Derby took a year; Nutch took 5 months;
JaxME and jUDDI took 6 months; JDO took 7 months.  Those are just a few
projects that have graduated to another TLP, and every community is
different, but ActiveMQ and ServiceMix haven't been in the Incubator for 3
months, and are still moving over their infrastructure.  Anyone see a need
to rush to judgment?


I don't see any rush here.  Is it not a natural question to ask what
else is left to do?  Is there a compelling reason to keep this in the
incubator if there is no real need other than historical precedence of
previous podlings' matriculation?  There were concrete reasons for the
duration of their incubation.  If they are the same issues as AMQ, then
let's focus on those concrete issues.


Regards,
Alan







RE: ActiveMQ Graduation From Incubator

2006-03-15 Thread Noel J. Bergman
David Blevins wrote:

Lots of good stuff, thanks.  :-)

> If you ask me what my opinion on OpenEJB's future or James' opinion
> on ActiveMQ's future, we'll both probably tell you TLP is a good
> goal eventually.

> We've more or less been running as TLPs in relation to Geronimo for
> the past two plus years already, just at Codehaus.  We've seen how
> that plays out and we'd like to try a much more unified front for a
> while and see what shakes out.

> The Geronimo world is awkward and unbalanced.  We have too tight
> integration with OpenEJB such that the standalone version of that
> completely disappeared [, and too little integration with ActiveMQ.]

> We're aren't successfully leveraging each other's communities
> to the fullest.  All in all, we don't make decisions together
> and lean on each other as much as we could.

How is merging the communities into the Geronimo TLP going to correct these
problems?  Is the Geronimo PMC prepared to add all of you as Committers on
the TLP, and add all or most of you to the Geronimo PMC?  Is this going to
result in a single community, or more integrated, better, meeting place for
multiple communities?  If the latter, that would indicate to me that you
really should be a TLP.  See Henri's e-mails for a fuller discussion of
Federation.

And what about community growth?  When TLPs have sub-projects, how do they
go about building up each sub-project?  How quick would the target PMC be to
grant TLP karma and PMC status to people interested in working on just one
of these disjoint projects?  Or would it have disjoint karma, and a smaller
number of people with binding votes?  See Henri's and Robert's comments on
these issues.

> I'd really like to see all our projects rolled up, balanced out, then
> split up again along possibly different lines with potentially more
> standalone pieces than we see now.

So you see putting everything together in Geronimo as a first step towards a
significant reorganization, with multiple TLPs?

> TLP is a good goal, but I really see value in the journey.

So do I.  The question appears to be where/how to take that journey.  :-)

--- Noel



RE: ActiveMQ Graduation From Incubator

2006-03-15 Thread Noel J. Bergman
Alan D. Cabrera wrote:

> I only see infrastructure issues in your list of concerns
> that would prevent the graduation of ActiveMQ.

Look again, but also at comments from Dims, Henri and others.

> You express an opinion that it should be a TLP but mention that it has a
> long way to go before it's ready for that.  Can you enumerate what
> remains, aside from the infrastructure issues

See my reply to Dain.  And I do feel that some of it does come down to being
able to convey a subjective confidence to the Incubator PMC that the
community really does "get it" regarding ASF principles and practices.  And
that is supposed to happen before, not after, a community leaves the
Incubator.

> If AMQ has less inspiring aspirations and was to initially land
> as a sub-project

I am not sure how much difference there ought to be, but some of that comes
down to the landing PMC.  I do have a concern an issue of fairness.

Consider David Blevin's well-stated views, including "We've more or less
been running as TLPs [for] the past two plus years already."  So if we have
some community that has been autonomous, and it becomes part of another TLP
within the ASF, how fair would it be for the members of that community to
lose their decision making ability?  I would say not, so are they going to
be made part of the destination PMC, which would be required for them to
have binding votes?

This is a generic issue.  I would have to cross-reference in detail the PMC
and committer lists for ActiveMQ and Geronimo to be specific to this case.
I do realize that there is overlap, but also others who are part of ActiveMQ
and are not part of Geronimo.  Is Geronimo prepared to welcome them as
Committers on the Geronimo TLP and members of the Geronimo PMC?

Related comment will go as a reply to David Blevins.

--- Noel



Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread David Blevins
If you ask me what my opinion on OpenEJB's future or James' opinion  
on ActiveMQ's future, we'll both probably tell you TLP is a good goal  
eventually.


We've more or less been running as TLPs in relation to Geronimo for  
the past two plus years already, just at Codehaus.  We've seen how  
that plays out and we'd like to try a much more unified front for a  
while and see what shakes out.  We're all ready for a change in the  
status-quo.


The Geronimo world is awkward and unbalanced.  We have too tight  
integration with OpenEJB such that the standalone version of that  
completely disappeared.  We have too little integration with ActiveMQ  
such that the things you can do with standalone ActiveMQ you can't do  
with the Geronimo ActiveMQ.  We have parts of Geronimo which could  
very well become separately reusable components, like the transaction  
manager or the XBean code.  We're aren't successfully leveraging each  
other's communities to the fullest.  All in all, we don't make  
decisions together and lean on each other as much as we could.


I'd really like to see all our projects rolled up, balanced out, then  
split up again along possibly different lines with potentially more  
standalone pieces than we see now.


TLP is a good goal, but I really see value in the journey.

-David



RE: ActiveMQ Graduation From Incubator

2006-03-15 Thread Noel J. Bergman
Henri Yandell wrote:

> Justin Erenkrantz wrote:

> > The APR spin-off from HTTP Server was probably the first federation
> > (although it wasn't called that).  HTTP Server depends upon APR and
> > they have a large committer and PMCer overlap (but not total), but
> > from the Foundation/Board's perspective - they are independent TLPs
> > and are treated as such.  The same technical arguments seem to be made
> > for ActiveMQ and ServiceMix in their dependency relationship with
> > Geronimo.  -- justin

> Right. I would suggest having individual TLPs, individual mailing
> lists, and individual websites, but with a lot of cross-linking,

Something in that realm, sure.  I am sure that there are a variety of ways
to structure Federation.  People are still exploring them.  We also talked
about how ontology might be addressed to support cross-project discussion.
And the work that Dave & Co are doing with projects.apache.org can help.

--- Noel



RE: ActiveMQ Graduation From Incubator

2006-03-15 Thread Noel J. Bergman
James Strachan wrote:

> What other issues are there?

A number of infrastucture issues.  Votes from the Incubator PMC and Geronimo
PMC.  To do that responsibly, I'd say that we would want to see communities
having demonstrated that they understand how to practice as an ASF
community.  Such things are subjective, and typically have taken some time.

Looking at the historical record: Derby took a year; Nutch took 5 months;
JaxME and jUDDI took 6 months; JDO took 7 months.  Those are just a few
projects that have graduated to another TLP, and every community is
different, but ActiveMQ and ServiceMix haven't been in the Incubator for 3
months, and are still moving over their infrastructure.  Anyone see a need
to rush to judgment?

--- Noel



Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Henri Yandell
On 3/15/06, Justin Erenkrantz <[EMAIL PROTECTED]> wrote:
> On 3/15/06, Henri Yandell <[EMAIL PROTECTED]> wrote:
> >...snip good stuff..
> > Have TLPs and have each TLP's website be at geronimo.apache.org.
> > Investigate federations. Even investigate sharing mailing lists.
> >...snip good stuff..
>
> Jakarta and XML have gone that 'federation' route with a bunch of
> 'friendly' TLPs.  HTTP Server may eventually spin off more
> sub-projects to TLP status as well (such as mod_python).
>
> The APR spin-off from HTTP Server was probably the first federation
> (although it wasn't called that).  HTTP Server depends upon APR and
> they have a large committer and PMCer overlap (but not total), but
> from the Foundation/Board's perspective - they are independent TLPs
> and are treated as such.  The same technical arguments seem to be made
> for ActiveMQ and ServiceMix in their dependency relationship with
> Geronimo.  -- justin

Right. I would suggest having individual TLPs, individual mailing
lists, and individual websites, but with a lot of cross-linking,
brand-logo [Geronimo federation image on each of the other TLP sites] 
and with the Geronimo PMC providing space (open to the other TLPs) on
the Geronimo website for the integration details of each of these
plugins.

Just an idea though :)

Hen


RE: ActiveMQ Graduation From Incubator

2006-03-15 Thread Noel J. Bergman
Dain Sundstrom wrote:

> Our goal when starting the incubation process of ActiveMQ, OpenEJB,
> ServiceMix, WADI, and XBean, was to consolidate the Geronimo
> community.

Consolidating the community is a good thing.  I've long wanted to see a
number of those projects at the ASF.

> The vision was to have a single community focused on building a modular
> server architecture based on a single core.

No disagreement.  The question at hand is simple and specific.  Are you
going to actually make one community from the bunch, with everyone having
access to work on every piece of code?  Are you going to have a large,
single, PMC with everyone having binding decisions over every aspect of the
project?

THAT is a TLP.

> each of the sub projects would be deliverable as a standalone
> (basically the core with one plugin installed).

Separate from destination, but that sounded fine right up to the last point.
What is the core?  If I just want ActiveMQ, or I just want OpenEJB, or I
just want ... can I get it, or do I have to take some code Geronimo stuff,
too?  From what Alan Cabrera said last night, it sounded as if there have
been some complaints about that, specifically in his example with OpenEJB.
Or in my case, if we want to use ActiveMQ for JMS in James, what other
non-ActiveMQ bits would we required to deal with in order to get the code
that was apparently separable earlier?

> end up with just a bunch of separate TLPs because of some
> unknown apache rules.

I'm expressing a personal opinion that the projects would be better off as
TLPs that collaborate with Geronimo.  From what I am being told, that is not
an uncommon view within those communities, although there are questions as
to whether to go TLP directly or via Geronimo.  Perhaps that ought to be put
forward for the individuals who make up the communities to directly answer?
:)

--- Noel



Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Justin Erenkrantz
On 3/15/06, Henri Yandell <[EMAIL PROTECTED]> wrote:
>...snip good stuff..
> Have TLPs and have each TLP's website be at geronimo.apache.org.
> Investigate federations. Even investigate sharing mailing lists.
>...snip good stuff..

Jakarta and XML have gone that 'federation' route with a bunch of
'friendly' TLPs.  HTTP Server may eventually spin off more
sub-projects to TLP status as well (such as mod_python).

The APR spin-off from HTTP Server was probably the first federation
(although it wasn't called that).  HTTP Server depends upon APR and
they have a large committer and PMCer overlap (but not total), but
from the Foundation/Board's perspective - they are independent TLPs
and are treated as such.  The same technical arguments seem to be made
for ActiveMQ and ServiceMix in their dependency relationship with
Geronimo.  -- justin


RE: ActiveMQ Graduation From Incubator

2006-03-15 Thread Noel J. Bergman
> Noel J. Bergman wrote:
> > Personally, I believe that ActiveMQ ought to be a TLP.

> Just to be clear, though, that's just a personal opinion

Which part of "Personally, I believe" wasn't clear?  ;-)

> > What makes a project with multiple codebases an umbrella
> > is a gray area.

> I've posted *my* first-pass definition of the term: a TLP that
> has no deliverable packages of its own, only from its subprojects.

So Geronimo would not be, DB would be?

> that's my working definition.

Not at all bad for a first cut.   Certainly thought provoking.

--- Noel


Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Henri Yandell
On 3/15/06, Rodent of Unusual Size <[EMAIL PROTECTED]> wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> Noel J. Bergman wrote:
> >
> > Personally, I believe that ActiveMQ ought to be a TLP.
>
> Just to be clear, though, that's just a personal opinion
> at this time, and in no way a 'dis is how t'ings is gonna
> be' statement.  Right? :-)
>
> > What makes a project with multiple codebases an umbrella is a gray
> > area.
>
> I've posted *my* first-pass definition of the term: a TLP that
> has no deliverable packages of its own, only from its subprojects.
> There's no 'Apache Jakarta' package, so it's an umbrella.
> There *is* an 'Apache HTTP Server' package, so that isn't.
> Subject to exceptions, clarifications, and further refinements,
> of course, but that's my working definition.

Ah, a subject dear to my heart. Here are some definitions I've been working to:

* An umbrella. A collection of communities under a single community.

* A disjoint umbrella.  The situation in which the subcommunities of
an umbrella have low overlap - this is synonymous with the negative
meaning of "Jakartization" and the thing the board are trying to avoid
occuring in more than one place.

* The ASF. A disjoint umbrella.

* Jakarta Commons. A non-disjoint umbrella - works well, though it's
always a struggle to avoid entropy's slow pull towards disjoint.

History:

Jakarta: by means of being a disjoint umbrella - started to mimic the
foundation it was meant to be a part of. This is a redundancy problem
- we don't need N groups of people being foundations at the ASF - the
idea is for the foundation to be as small as it can be and still
support the communities for which it exists.

No more umbrellas: thus umbrellas became passe. The Incubator was
created as a managed umbrella to be a proto-foundation; the ASF
remains the primary umbrella. All the problems inherent in Jakarta and
other umbrellas remain; but they are now being dealt with by one group
- the board.



Speaking as an Incubator PMC member, I am currently -1 towards
Geronimo ActiveMQ and +1 towards Apache ActiveMQ. This is because I've
not yet heard much about how Geronimo plans to avoid disjointedness.

I do really like the point that Geronimo has a common product - but
I'm not sure it's enough. Plus how will it affect the use of the
subprojects (because subprojects they would be) as stand alones?

---

Lastly, speaking as someone who has used up the last couple of years
of his Apache time dealing with the problem of being the bridge
between a disjoint umbrella, and a board/foundation which is not
organized for dealing with disjoint umbrellas, I recommend not
creating a large umbrella.

Have TLPs and have each TLP's website be at geronimo.apache.org.
Investigate federations. Even investigate sharing mailing lists.
However do not give yourself the pain of trying to funnel it all
through a single chair. TLPs are not firewalled from each other
(though it's not very obvious that that's the case); so investigate
what it is you really want to achieve and don't focus on the single
TLP part of things.

TLPs are more about delegated oversight from the members/board, than community.

Hen


Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Alan D. Cabrera




Noel J. Bergman wrote:

  Alan D. Cabrera wrote:

  
  
This is not a vote, but simply a discussion about the graduation of
ActiveMQ from the Incubator.

  
  
Personally, I do not consider ActiveMQ ready.  And I do believe that it
should be targeting TLP status.  It has its own community, is separately
releasable and useable in many projects, not just as part of a J2EE server,
and would do better as its own TLP.  To reiterate, these are my views.  The
Incubator PMC may share or differ in its collective view.

Keep in mind that I am not saying anything negative about ActiveMQ.  I like
the project.  I have had quite constructive discussions with members of the
project about possibly using the project.  It simply has a way to go before
it is ready as a TLP.  For that matter, as others have pointed out, it still
has some way to go in migrating infrastructure, of which JIRA is only one
issue, and is being addressed.

Generally speaking, I concur with the point made by others: new projects
should learn from the mistakes of others, not emulate them.

	


I only see infrastructure issues in your list of concerns that would
prevent the graduation of ActiveMQ.

You express an opinion that it should be a TLP but mention that it has
a long way to go before it's ready for that.  Can you enumerate what
remains, aside from the infrastructure issues, to be done to graduate
as a TLP?  If AMQ has less inspiring aspirations and was to initially
land as a sub-project, can you enumerate what remains to be done to
graduate?

IMO, aside from the infrastructure issues, AMQ is good to go as a
sub-project.  It should start there and if it's worthy enough, evolve
into a TLP.  I see no good reason for it to stay in the incubator at
this time.


Regards,
Alan






Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

James Strachan wrote:
> 
> What if folks involved in the project & on the Geronimo project don't  
> want it to be a TLP - at least not for a while yet? e.g. can't we  
> just use the Geronimo PMC until the time folks want/decide to start  
> to go TLP? Or is going TLP now mandatory?

Unless there is a really good reason for AMQ *not* to graduate
into Geronimo, that's what will happen.  At which point it
will be part of Geronimo and the Geronimo PMC applies.  While
it's still in the incubator, though.. no, AMQ *can't* use the
Geronimo PMC for oversight.  AMQ's PPMC can have zero to N of
the Geronimo PMC members on it, but it would also have some
from the incubator (at least the mentors -- currently just you)
and some committers from the podling (who would *not* be on the
Geronimo PMC).  Sponsoring TLPs can provide advice and
guidance to podlings, but not direction -- they're self-directing.
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBgRj5rNPMCpn3XdAQLnLQP/cZMXTLLaS+/wvVXwXqzseNwA8PqUziaJ
H3aRUgfzskZtYcZ0a6XNKQsIFtPbmtq7hC1Wpd3vQ6b7hYt4wmjkAsw7ZhgXH5yE
p1sI1WXJvG44Mr27mQdAVF/zuAnOFaAe4yThWyR9Jp4tb95qPY56MsuDkO0y4Cd9
zBfyXY7edfY=
=PKwQ
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Dain Sundstrom wrote:
> 
> Just to make sure this was allowed, before pitching it to the 
> communities, I asked a few of the Board members at Euro OS con and 
> they said it was possible.  I didn't want to get into a situation 
> where we do all of the work to uproot and move exiting communities 
> and then end up with just a bunch of separate TLPs because of some 
> unknown apache rules.  Now it seem like that is exactly what will 
> happen.

Why do you say that?  Noel has stated his current *personal* opinion,
and his *personal* opinion of board policy as

> Keep in mind that the ASF Board has established a pretty conscious
> decision for projects to go TLP, and to disband umbrella projects.

IMHO, a podling that was envisioned as joining an existing TLP
will end up doing so unless there are really good reasons for it
to go TLP instead.  The only time (again IMHO) that the collective
desires of the specific communities involved (in this case,
AMQ and Geronimo) would be overruled is if the determination
is made that such a merger would not make sense or would not
be in the best interests of the ASF.As an hypothetical
example, I very much doubt that a podling for a pattern
recognition filtering package would be allowed to join the HTTP
Server project, despite being originally intended for filtering
server logs.  A traffic-filtering module written to the httpd
API would be unlikely to go anywhere else.  An IRC server based
on the httpd framework might go either way.

It's not just rhe podling's and sponsor's communities that need
to be considered, although they're by far the biggest part.  It's
also the overall Apache community that needs to be factored in.
If going subproject is considered to be likely to be harmful to
the ASF, it won't happen.  Otherwise it probably will.

All IMHO.
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBf8a5rNPMCpn3XdAQL5VwP9E/nNpdNA3CHD4jjLrCqwSRy7dgXUjzQj
OCCujB5mm2H8LMVhNYgHSfRYlrQAAayXrizwTwUT5zZIOstbX9scUSGM5MCnHOg8
ZakW6O/+N41EaNkdWhBBdwPHVNl9PmTqbwb5sMW1Abo+RKtPEVUG1gnheoFyqd4M
6w82VLeqVi0=
=NVCO
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

James Strachan wrote:
> 
> What if folks involved in the project & on the Geronimo project don't  
> want it to be a TLP - at least not for a while yet? e.g. can't we  
> just use the Geronimo PMC until the time folks want/decide to start  
> to go TLP? Or is going TLP now mandatory?

No, it's not mandatory.  And (AFAIK) no-one's going to force it
TLP against the consensual wishes of its participants and sponsor
unless there's a bloody good reason.
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBf2JZrNPMCpn3XdAQK1+wQAtugkocu4mpiMjbSW1Trgby2tbGlme3rt
ZoHH6dj27vAGIEqwqP8K0elHhh7fezea0cItawPcXTHjwXKQMkbonv1XGdfdeAIZ
nrwIfyJAb3t3P0qR5O3oekEeazFu7odeoWf6vYKypg0JyBc1jPTRxEd6QB6yEvai
dxSP6AMPGzw=
=EygF
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-15 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Noel J. Bergman wrote:
> 
> Personally, I believe that ActiveMQ ought to be a TLP.

Just to be clear, though, that's just a personal opinion
at this time, and in no way a 'dis is how t'ings is gonna
be' statement.  Right? :-)

> What makes a project with multiple codebases an umbrella is a gray
> area.

I've posted *my* first-pass definition of the term: a TLP that
has no deliverable packages of its own, only from its subprojects.
There's no 'Apache Jakarta' package, so it's an umbrella.
There *is* an 'Apache HTTP Server' package, so that isn't.
Subject to exceptions, clarifications, and further refinements,
of course, but that's my working definition.
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBf065rNPMCpn3XdAQKD9wQA3KqhUOsk9WG2OAvFyGmZF+e9mbQTq9O/
aXOa4qVFWfNns/rc2NpPwO6jewSepneo+uVKMBk/5MCjczuwFZSmq9uMyE9LJ/Na
czsKVChEYYrHJnrDoKXoQqmaIxnDLBtYU/SEHggzWH8L9RuryFUXPV5/HGSCSLKu
EvPHDljpWJU=
=dSia
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Dain Sundstrom

Noel,

I see this as a big source of my frustration, and I hope we can do  
something about this.  From the perspective of anyone in an  
incubating project, you represent the incubator.  So when you express  
"just your opinion" it very difficult if not impossible for someone  
to see the distinction.  This is compounded by the fact that you get  
to vote on a lot of issues related to an incubating project, and  
given your ability to influence others, your opinion quickly becomes  
incubator policy or in the absence of policy, failed votes.


I would like to see the incubator encourage communities to make their  
own decisions via an open community oriented process based within the  
guidelines of the ASF.  I think there is a conflict of interest when  
those there to help incubate new projects and indoctrinate them in  
the apache way also pushing a personal agenda, and I think we should  
avoid this at all costs.


-dain

On Mar 14, 2006, at 10:26 PM, Noel J. Bergman wrote:


Dain Sundstrom wrote:


I agree it is important to have as much as possible on apache
hardware.  It was my understanding until I read this thread
here, that infrastructure was fine with leaving JIRAs for
imported projects hosted remotely until the JIRA had a better
import tool.


Please keep in mind that when I speak, Sam speaks, Ken speaks, Justin
speaks, etc., we are airing our own views.  In general, I would  
assume that
someone is speaking as an individual rather than a role, unless  
otherwise
indicated.  Greg Stein, for example, only uses his @apache.org  
address when
posting as the ASF Chairman.  If you see him posting from another e- 
mail

address, he is posting as just Greg.  In other cases, you might notice
someone specifically mention in the e-mail what hat they are  
wearing.  When
the Incubator PMC speaks, it does so with a consensus, which may or  
may not

coincide with the individual views of all of its members.

Individuals have differing priorities.  How the community makes  
decisions
and how it interacts with the rest of the ASF are high on my list.   
The ASF
is about individuals making a deliberate and concerted effort to  
say that
collaboration and consensus are the key principles.  Even when we  
argue

about something, that's an expression of that structure.  Getting this
across to people joining the ASF is an crucial part of what  
Incubation is

supposed to do, in my view.

--- Noel




Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Justin Erenkrantz
On 3/14/06, James Strachan <[EMAIL PROTECTED]> wrote:
> Just out of interest, who decides on if its going to be a TLP or
> Geronimo sub project & how is that decision made?

Only the Board can approve a new TLP.  If the Board does not approve a
podling as a TLP, the Incubator PMC is then responsible for
'releasing' that project into the oversight of another PMC.  The
Incubator PMC should only 'release' when it is confident that all of
the legal and community procedures and policies have been successfully
completed.  -- justin


Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread James Strachan

On 15 Mar 2006, at 03:54, Noel J. Bergman wrote:
Personally, I do not consider ActiveMQ ready.  And I do believe  
that it
should be targeting TLP status.  It has its own community, is  
separately
releasable and useable in many projects, not just as part of a J2EE  
server,
and would do better as its own TLP.  To reiterate, these are my  
views.  The

Incubator PMC may share or differ in its collective view.


Just out of interest, who decides on if its going to be a TLP or  
Geronimo sub project & how is that decision made?



Keep in mind that I am not saying anything negative about  
ActiveMQ.  I like

the project.


Great :)



I have had quite constructive discussions with members of the
project about possibly using the project.  It simply has a way to  
go before

it is ready as a TLP.


What if folks involved in the project & on the Geronimo project don't  
want it to be a TLP - at least not for a while yet? e.g. can't we  
just use the Geronimo PMC until the time folks want/decide to start  
to go TLP? Or is going TLP now mandatory?




For that matter, as others have pointed out, it still
has some way to go in migrating infrastructure, of which JIRA is  
only one

issue, and is being addressed.


What other issues are there?

James
---
http://radio.weblogs.com/0112098/



Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Dain Sundstrom

Noel,

Our goal when starting the incubation process of ActiveMQ, OpenEJB,  
ServiceMix, WADI, and XBean, was to consolidate the Geronimo  
community.  The vision was to have a single community focused on  
building a modular server architecture based on a single core.  The  
global deliverable would be Geronimo the J2EE server, but each of the  
sub projects would be deliverable as a standalone (basically the core  
with one plugin installed).  This is what we pitched to the external  
projects and what they agreed to.


Just to make sure this was allowed, before pitching it to the  
communities, I asked a few of the Board members at Euro OS con and  
they said it was possible.  I didn't want to get into a situation  
where we do all of the work to uproot and move exiting communities  
and then end up with just a bunch of separate TLPs because of some  
unknown apache rules.  Now it seem like that is exactly what will  
happen.


We (the communities) want to form a single community focused on this  
goal, are you saying that this is not possible anymore?If this no  
longer the case, I think we have an obligation to inform the  
incubating communities, so they can decide if they want to continue  
incubation and become an Apache TLP or go back to where they were.


-dain

On Mar 14, 2006, at 7:54 PM, Noel J. Bergman wrote:


Dain Sundstrom wrote:


When AMQ entered the incubator as a sponsored project from Geronimo,
the current understanding of incubator rules was that AMQ would
simply use the Geronimo pmc since the Geronimo pmc is expected to be
the home for the project.  Since then the incubator rules have been
rewritten several time and based on the emails I saw today


I'm not sure where you got the idea that this or any other policy was
changing from under ActiveMQ.  My comments, for example, regarding  
Yoko that
"this is not a Geronimo sub-project.  Incubator projects are just  
that:
Incubator projects whose final destination will be determined at  
graduation"
were reiteration, not new policy.  Sam and others have said pretty  
much the

same thing.

It has never, since the inception of the idea of a PPMC, been the  
case that
a project could use another PMC for its PPMC.  If there is a  
sponsoring PMC,

it is certainly welcome (and usually expected) to participate, but the
Incubator PMC has sole discretion and authority to bring new  
projects into

the ASF.

Consider Derby.  Derby had a PPMC.  The DB PMC was not in charge of  
Derby.
The Incubator PMC managed Derby until graduating it.  Derby went  
into DB

(although it would have been fine, IMO, if Derby had gone TLP).

Personally, I believe that ActiveMQ ought to be a TLP.  It is  
clearly the
case that although ActiveMQ can be used by Geronimo, it is an  
independently
usable, separately releasable project with its own community that  
happens to
have some overlap with parts of Geronimo.  The same is true of a  
number of
projects that Geronimo has sponsored (thank you :-)), and which  
ought to be

separate TLPs in my view.

Keep in mind that the ASF Board has established a pretty conscious  
decision

for projects to go TLP, and to disband umbrella projects.  The
"Jakartaization of" is not considered a compliment.  What makes a  
project

with multiple codebases an umbrella is a gray area.

--- Noel




RE: ActiveMQ Graduation From Incubator

2006-03-14 Thread Noel J. Bergman
Dain Sundstrom wrote:

> I agree it is important to have as much as possible on apache
> hardware.  It was my understanding until I read this thread
> here, that infrastructure was fine with leaving JIRAs for
> imported projects hosted remotely until the JIRA had a better
> import tool.

Please keep in mind that when I speak, Sam speaks, Ken speaks, Justin
speaks, etc., we are airing our own views.  In general, I would assume that
someone is speaking as an individual rather than a role, unless otherwise
indicated.  Greg Stein, for example, only uses his @apache.org address when
posting as the ASF Chairman.  If you see him posting from another e-mail
address, he is posting as just Greg.  In other cases, you might notice
someone specifically mention in the e-mail what hat they are wearing.  When
the Incubator PMC speaks, it does so with a consensus, which may or may not
coincide with the individual views of all of its members.

Individuals have differing priorities.  How the community makes decisions
and how it interacts with the rest of the ASF are high on my list.  The ASF
is about individuals making a deliberate and concerted effort to say that
collaboration and consensus are the key principles.  Even when we argue
about something, that's an expression of that structure.  Getting this
across to people joining the ASF is an crucial part of what Incubation is
supposed to do, in my view.

--- Noel



RE: ActiveMQ Graduation From Incubator

2006-03-14 Thread Noel J. Bergman
Alan D. Cabrera wrote:

> This is not a vote, but simply a discussion about the graduation of
> ActiveMQ from the Incubator.

And should have been on [email protected], not [EMAIL PROTECTED]
Whether or not to cross-post to a myriad of other lists is a separate
question of netiquette.  And the Geronimo PMC can address the use of
[EMAIL PROTECTED]  In the meantime, since this was cross-posted to at least two
public lists, I am replying to [EMAIL PROTECTED]

Now, on to the main topic:



Personally, I do not consider ActiveMQ ready.  And I do believe that it
should be targeting TLP status.  It has its own community, is separately
releasable and useable in many projects, not just as part of a J2EE server,
and would do better as its own TLP.  To reiterate, these are my views.  The
Incubator PMC may share or differ in its collective view.

Keep in mind that I am not saying anything negative about ActiveMQ.  I like
the project.  I have had quite constructive discussions with members of the
project about possibly using the project.  It simply has a way to go before
it is ready as a TLP.  For that matter, as others have pointed out, it still
has some way to go in migrating infrastructure, of which JIRA is only one
issue, and is being addressed.

Generally speaking, I concur with the point made by others: new projects
should learn from the mistakes of others, not emulate them.

--- Noel



RE: ActiveMQ Graduation From Incubator

2006-03-14 Thread Noel J. Bergman
Dain Sundstrom wrote:

> When AMQ entered the incubator as a sponsored project from Geronimo,
> the current understanding of incubator rules was that AMQ would
> simply use the Geronimo pmc since the Geronimo pmc is expected to be
> the home for the project.  Since then the incubator rules have been
> rewritten several time and based on the emails I saw today

I'm not sure where you got the idea that this or any other policy was
changing from under ActiveMQ.  My comments, for example, regarding Yoko that
"this is not a Geronimo sub-project.  Incubator projects are just that:
Incubator projects whose final destination will be determined at graduation"
were reiteration, not new policy.  Sam and others have said pretty much the
same thing.

It has never, since the inception of the idea of a PPMC, been the case that
a project could use another PMC for its PPMC.  If there is a sponsoring PMC,
it is certainly welcome (and usually expected) to participate, but the
Incubator PMC has sole discretion and authority to bring new projects into
the ASF.

Consider Derby.  Derby had a PPMC.  The DB PMC was not in charge of Derby.
The Incubator PMC managed Derby until graduating it.  Derby went into DB
(although it would have been fine, IMO, if Derby had gone TLP).

Personally, I believe that ActiveMQ ought to be a TLP.  It is clearly the
case that although ActiveMQ can be used by Geronimo, it is an independently
usable, separately releasable project with its own community that happens to
have some overlap with parts of Geronimo.  The same is true of a number of
projects that Geronimo has sponsored (thank you :-)), and which ought to be
separate TLPs in my view.

Keep in mind that the ASF Board has established a pretty conscious decision
for projects to go TLP, and to disband umbrella projects.  The
"Jakartaization of" is not considered a compliment.  What makes a project
with multiple codebases an umbrella is a gray area.

--- Noel



RE: ActiveMQ Graduation From Incubator

2006-03-14 Thread Noel J. Bergman
> > That's not actually a formal requirement though, correct?
> Not at this time.  We're still discussing what the SHOULD and MUST
> will be, as I mentioned in the fuller context of what you quoted.

Hmmm ... or perhaps I hadn't made it as clear as I thought I had.  I just
went back and found that particular e-mail.  Sorry.

To be clear, as noted above, we're still discussing SHOULD and MUST.  There
is no *requirement* today that a project have at least 3 Mentors.  I hope
that we will end up agreeing that it SHOULD, but with leeway to allow for
the PMC to apply human judgment.

--- Noel



RE: ActiveMQ Graduation From Incubator

2006-03-14 Thread Noel J. Bergman
Rodent of Unusual Size wrote:

> 'So close to graduation'?  Whence comes that?  I think that
> proximity is still very much up in the air, particularly
> given Noel's opinion that [...]

Keep in mind that is *my* opinion.  The Incubator PMC as a whole may or may
not agree.

For a guy who is seriously independent, I'll put it this way: *I* will fight
to ensure *our* *collective* decision making process.

--- Noel



RE: ActiveMQ Graduation From Incubator

2006-03-14 Thread Noel J. Bergman
Rodent of Unusual Size wrote:

> Noel J. Bergman wrote:
> > On the community side, we're still a bit shy of Mentors on ActiveMQ
> > (James is the only one, and we are looking for at least 3 per project)

> That's not actually a formal requirement though, correct?

Not at this time.  We're still discussing what the SHOULD and MUST will be,
as I mentioned in the fuller context of what you quoted.

--- Noel



Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Dain Sundstrom

On Mar 14, 2006, at 11:08 AM, Rodent of Unusual Size wrote:


-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Dain Sundstrom wrote:



AFAIR, that was never *my* understanding.  AFAIK, that has *never*
been the way the incubator has worked.  Every podling has supposed
to have had a PPMC.  If I'm wrong, please correct me; where did you
(and evidently others) read whatever it was that said a TLP PMC
could serve for a podling?


If you remember back, Geronimo started in the incubator before there
 was the concept of a PPMC.  Geronimo was the first project to get a
 PPMC because geronimo was target to be a top level project.  All of
 the other projects in the incubator at that time were targeted to be
 subprojects to it made most since to get those sub projects working
 with their sponsoring pmc and the incubator was there to make sure
we weren't ending up with umbrella subprojects, and instead projects
 that acted as a single whole.


So, basically, the idea that a sponsoring PMC could/should
direct a podling comes from the time of Geronimo's own incubation?


That was the way it was when we were incubated, and I was not aware  
of the change.



Option 1 is clearly not appropriate for a project that has an
existing community.  Option 2 is not appropriate for a project that
is supposed merging communities with another.


You disgree with the doctrine of 'we don't know where a
podling will go until graduation,' I take it.


I think a podling can change direction during incubation, but I think  
they do and should always have a target in mind.



Option 2 sets up a separate independent group, and once that is setup
it will be hard to merge.


I disagree.  There's nothing preventing the TLP PMC
members from getting on the PPMC.  And other podlings
have managed to merge with little or no pain.  Derby,
for example.


I think Derby has done a great job integrating into DB, but I would  
like to see even closer ties in the Geronimo project.



I think we need an incubation procedure that instead is designed to
setup and assure that the new incubating group is merging the target
communities and that incubation is only complete once continuous
whole.  This is exactly what the Geronimo incubation were suppose to
achieve.  In originally email I sent out on this and the
conversations I had with a some of the board members before the
email, I asked if we can "consolidate" our communities.  This is what
everyone was excited about and thought was possible in the incubator
and now I feel that the new incubation rules seem to be setup to
prevent exactly this


One of the purposes of the incubator is to normalise expectations.
'Indoctrinate,' if you like, newcomers in the Apache ethos.  A
group of people working on an external project, which comes
wholesale to Apache with that education being provided by the
accepting TLP, can lead to exactly the sort of problems we had
a few years ago.  So the rules aren't there to prevent the
consolidation of communities; they're there (in part) to limit
heresy. :-)


That makes since.

-dain


Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Dain Sundstrom wrote:
> 
>> AFAIR, that was never *my* understanding.  AFAIK, that has *never*
>> been the way the incubator has worked.  Every podling has supposed
>> to have had a PPMC.  If I'm wrong, please correct me; where did you
>> (and evidently others) read whatever it was that said a TLP PMC
>> could serve for a podling?
> 
> If you remember back, Geronimo started in the incubator before there
>  was the concept of a PPMC.  Geronimo was the first project to get a
>  PPMC because geronimo was target to be a top level project.  All of
>  the other projects in the incubator at that time were targeted to be
>  subprojects to it made most since to get those sub projects working
>  with their sponsoring pmc and the incubator was there to make sure
> we weren't ending up with umbrella subprojects, and instead projects
>  that acted as a single whole.

So, basically, the idea that a sponsoring PMC could/should
direct a podling comes from the time of Geronimo's own incubation?
> Option 1 is clearly not appropriate for a project that has an
> existing community.  Option 2 is not appropriate for a project that
> is supposed merging communities with another.

You disgree with the doctrine of 'we don't know where a
podling will go until graduation,' I take it.

> Option 2 sets up a separate independent group, and once that is setup
> it will be hard to merge.

I disagree.  There's nothing preventing the TLP PMC
members from getting on the PPMC.  And other podlings
have managed to merge with little or no pain.  Derby,
for example.

> I think we need an incubation procedure that instead is designed to
> setup and assure that the new incubating group is merging the target
> communities and that incubation is only complete once continuous
> whole.  This is exactly what the Geronimo incubation were suppose to
> achieve.  In originally email I sent out on this and the
> conversations I had with a some of the board members before the
> email, I asked if we can "consolidate" our communities.  This is what
> everyone was excited about and thought was possible in the incubator
> and now I feel that the new incubation rules seem to be setup to
> prevent exactly this

One of the purposes of the incubator is to normalise expectations.
'Indoctrinate,' if you like, newcomers in the Apache ethos.  A
group of people working on an external project, which comes
wholesale to Apache with that education being provided by the
accepting TLP, can lead to exactly the sort of problems we had
a few years ago.  So the rules aren't there to prevent the
consolidation of communities; they're there (in part) to limit
heresy. :-)
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBcUs5rNPMCpn3XdAQLA/AQAvkHNgfZ04zg4kdwMNQu7+b2GghWUu+nf
kHi8oCr9EAhI/LthNlX+BkrIk02Nrg6VbC+I0Gu5vwAB7D2/VnLeBKbwSAOYMfwp
CmOg9DPJ/lOsQQsD5fzb6T2hSp78foEdsLYwEItyVUPbRLhEwgC/Sv3ZQQbn5QPF
Do5jCKjEQhA=
=huxO
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Noel J. Bergman wrote:
> 
> On the community side, we're still a bit shy of Mentors on ActiveMQ (James
> is the only one, and we are looking for at least 3 per project)

That's not actually a formal requirement though, correct?
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBcHjJrNPMCpn3XdAQKo5AP+NlL85QgGEbDkEqXs2YJNT24AlqyYgOpz
lK2eZpZnzxJRcFO+CQxnhBk1jmnr7XvimR1ta2ume6M7UeI0LtedRLkApOU2n0Hu
M+3CMp62MuZqx1/eyh4642XWj+au0ycPFPn3wns/ZOv4+SUE7RDrksxyyYr1bqOn
rvK9O8mafjw=
=j9bx
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Dain Sundstrom

On Mar 14, 2006, at 5:48 AM, Rodent of Unusual Size wrote:


-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Dain Sundstrom wrote:


When AMQ entered the incubator as a sponsored project from Geronimo,
the current understanding of incubator rules was that AMQ would
simply use the Geronimo pmc since the Geronimo pmc is expected to be
the home for the project.


AFAIR, that was never *my* understanding.  AFAIK, that has
*never* been the way the incubator has worked.  Every podling
has supposed to have had a PPMC.  If I'm wrong, please correct
me; where did you (and evidently others) read whatever it
was that said a TLP PMC could serve for a podling?


If you remember back, Geronimo started in the incubator before there  
was the concept of a PPMC.  Geronimo was the first project to get a  
PPMC because geronimo was target to be a top level project.  All of  
the other projects in the incubator at that time were targeted to be  
subprojects to it made most since to get those sub projects working  
with their sponsoring pmc and the incubator was there to make sure we  
weren't ending up with umbrella subprojects, and instead projects  
that acted as a single whole.



If you ask me setting up a separate pmc for these projects is an
incrediably bad idea.  Our objective is to create a single community
between Geronimo, ActiveMQ, OpenEJB, ServiceMix and WADI.  Putting
these projects into separate boxes makes this very difficult.


I believe there are two options:

1. Submit the code through the IP-vetting process.  The people
   join the TLP dev mailing list and earn merit for commit over
   time like anyone else.
2. Submit the code+people for processing through the incubator
   as a podling.  The committers get commit access right away,
   but now it's both the code and the community that's being
   vetted.  And the eventual disposition of the podling is not
   a foregone conclusion.

There is no fast-track to commit access.


I don't want a fast-track to commit either (I have a long history of  
fighting that at Geronimo), but I believe we need a third option, in  
between the two you present.  Option 1 is clearly not appropriate for  
a project that has an existing community.  Option 2 is not  
appropriate for a project that is supposed merging communities with  
another.  Option 2 sets up a separate independent group, and once  
that is setup it will be hard to merge.  I think we need an  
incubation procedure that instead is designed to setup and assure  
that the new incubating group is merging the target communities and  
that incubation is only complete once continuous whole.  This is  
exactly what the Geronimo incubation were suppose to achieve.  In  
originally email I sent out on this and the conversations I had with  
a some of the board members before the email, I asked if we can  
"consolidate" our communities.  This is what everyone was excited  
about and thought was possible in the incubator and now I feel that  
the new incubation rules seem to be setup to prevent exactly this


-dain


Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Sam Ruby
Dain Sundstrom wrote:
> On Mar 13, 2006, at 5:05 PM, Rodent of Unusual Size wrote:
> 
>> Dain Sundstrom wrote:
>>
>>>
>>> I understand this concern and agree with the solution, but we should
>>> remember that AMQ entered the incubator before this was a rule, so I
>>> for one didn't think it appled to them, since they are so close to
>>> graduation.
>>
>>
>> 'So close to graduation'?  Whence comes that?  I think that
>> proximity is still very much up in the air, particularly given
>> Noel's opinion that
> 
> 
> If you read the email history, you will see that it was stated that  the
> new rules would only apply to "projects close to gradation".  So  I hope
> you can see my point, regardless of if you agree with that point.
> 
>>> .. and the ASF community building is only just getting started.  No
>>> PPMC, yet, for which we need more Mentors.
>>
>>
>> These have nothing to do with when it entered incubation; the
>> need for a PPMC has been there right along, and the 'ASF
>> community building' is a sine qua non.  (I have no opinion,
>> myself, about the degree of 'ASF community building' that
>> has occurred in ActiveMQ.)
> 
> 
> When AMQ entered the incubator as a sponsored project from Geronimo, 
> the current understanding of incubator rules was that AMQ would  simply
> use the Geronimo pmc since the Geronimo pmc is expected to be  the home
> for the project.  Since then the incubator rules have been  rewritten
> several time and based on the emails I saw today, the  current rules
> that Noel is promoting (3+ mentors) hasn't even be  approved by the
> incubator.  I personally find this incredibly  frustrating, so please
> take my comments with a grain of salt.
> 
> If you ask me setting up a separate pmc for these projects is an 
> incrediably bad idea.  Our objective is to create a single community 
> between Geronimo, ActiveMQ, OpenEJB, ServiceMix and WADI.  Putting 
> these projects into separate boxes makes this very difficult.
> 
> I would like to know, why have the incubator rules changed to, in my 
> opinion, force all projects TLP?   Maybe the incubator is the wrong 
> place to bring these types projects.  Is there another process to  bring
> in a project we plan on integrating?  If not, maybe the board  should
> consider setting something else up.

"If you love someone, set them free. If they come back they're yours; if
they don't they never were"

I firmly believe that the destination for a code base should be
determined at the EXIT of incubation.  If each and every one of these
ultimately ends up at Geronimo by general consent of all the parties
involved, then (by definition) everybody is happy.

What I am unconfortable with is codebases being proposed with a
precondition being placed on where they land.

A sponsor is needed to inject a bit of accountability into the process,
and to reduce the tendency towards the ASF becoming a sourceforge with
lots of abandoned projects.  But that pretty much is the extent of
sponsorship.

Every code base should be looked at with the possibility of being a TLP.
 And with the possibility of being incorporated within an existing project.

Saying "I want ActiveMQ at the ASF", and saying "I think ActiveMQ would
make a fine addition to Geronimo" are both reasonable things to say.
Saying "I want ActiveMQ at the ASF, but only if it is destined to be a
part of Geronimo" is not.

- Sam Ruby


Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Geir Magnusson Jr



Sam Ruby wrote:


What I am unconfortable with is codebases being proposed with a
precondition being placed on where they land.

A sponsor is needed to inject a bit of accountability into the process,
and to reduce the tendency towards the ASF becoming a sourceforge with
lots of abandoned projects.  But that pretty much is the extent of
sponsorship.


The only thing I'd like to add is that I feel that a sponsoring PMC 
should take interest in the mentoring and development of the project it 
sponsored...


geir


Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Dain Sundstrom wrote:
> 
> Official policy documents would be really nice, especially  
> considering they take a huge amount of time to develop and would  
> hopefully slow down the rate of change in the incubator.

Yup.  Policy still evolving, though, makes that a bit problematic
as you've noted.  Jean is doing a great job with what's there,
though.
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBbMIJrNPMCpn3XdAQL0rQP9GoAY+PK1nKmOUD2on/5JSDqP/C0ZdEEk
QF/O6gFbFgxl/1Q+8xNOdzFIQmePxQwTv4h5Lh44zShMfL9tH35P991MKR8aKkfR
AGm9IBx0plZPbA3iKF2BeC6hjIKJE17Pdgt0s961uoxQu4iSinAsh5safPInBziz
Q6KeQVt9yfY=
=ttW3
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-14 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Dain Sundstrom wrote:
> 
> When AMQ entered the incubator as a sponsored project from Geronimo,  
> the current understanding of incubator rules was that AMQ would  
> simply use the Geronimo pmc since the Geronimo pmc is expected to be  
> the home for the project.

AFAIR, that was never *my* understanding.  AFAIK, that has
*never* been the way the incubator has worked.  Every podling
has supposed to have had a PPMC.  If I'm wrong, please correct
me; where did you (and evidently others) read whatever it
was that said a TLP PMC could serve for a podling?

> If you ask me setting up a separate pmc for these projects is an  
> incrediably bad idea.  Our objective is to create a single community  
> between Geronimo, ActiveMQ, OpenEJB, ServiceMix and WADI.  Putting  
> these projects into separate boxes makes this very difficult.

I believe there are two options:

1. Submit the code through the IP-vetting process.  The people
   join the TLP dev mailing list and earn merit for commit over
   time like anyone else.
2. Submit the code+people for processing through the incubator
   as a podling.  The committers get commit access right away,
   but now it's both the code and the community that's being
   vetted.  And the eventual disposition of the podling is not
   a foregone conclusion.

There is no fast-track to commit access.

> I would like to know, why have the incubator rules changed to, in my  
> opinion, force all projects TLP?   Maybe the incubator is the wrong  
> place to bring these types projects.  Is there another process to  
> bring in a project we plan on integrating?  If not, maybe the board  
> should consider setting something else up.

See above.  Code can come in quickly; people cannot.
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBbJtZrNPMCpn3XdAQLy1wQAioqZdIRGZtBQCZbfHh7q4isKt6Wn1zPn
ZkLzMQloRlbBl+qN6Pmss9j/AAohak2nHmZ2NUO8yl75Is7koyb8p69ZI9ozXJfg
diki/C5pmVOeeXFqmWYqdgYtrgpjYGESFrME7dkfrvZkd82OPb1rhCkiBhWWuS1a
NQ7mlDm1Z88=
=eKpN
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread Dain Sundstrom

On Mar 13, 2006, at 5:31 PM, Dain Sundstrom wrote:


On Mar 13, 2006, at 5:05 PM, Rodent of Unusual Size wrote:


Dain Sundstrom wrote:


I understand this concern and agree with the solution, but we should
remember that AMQ entered the incubator before this was a rule, so I
for one didn't think it appled to them, since they are so close to
graduation.


'So close to graduation'?  Whence comes that?  I think that
proximity is still very much up in the air, particularly given
Noel's opinion that


If you read the email history, you will see that it was stated that  
the new rules would only apply to "projects close to gradation".   
So I hope you can see my point, regardless of if you agree with  
that point.


I just realized the size of the cross posting on this thread.  To be  
specific, I am referring to the proposed new rules thread on the  
[EMAIL PROTECTED] list (which isn't even on this cross post).  Of  
course, I can't point to a specific email that made this policy, but  
that was my understanding at the time.


Official policy documents would be really nice, especially  
considering they take a huge amount of time to develop and would  
hopefully slow down the rate of change in the incubator.


Sorry about that,

-dain


Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread Dain Sundstrom

On Mar 13, 2006, at 5:05 PM, Rodent of Unusual Size wrote:


Dain Sundstrom wrote:


I understand this concern and agree with the solution, but we should
remember that AMQ entered the incubator before this was a rule, so I
for one didn't think it appled to them, since they are so close to
graduation.


'So close to graduation'?  Whence comes that?  I think that
proximity is still very much up in the air, particularly given
Noel's opinion that


If you read the email history, you will see that it was stated that  
the new rules would only apply to "projects close to gradation".  So  
I hope you can see my point, regardless of if you agree with that point.



.. and the ASF community building is only just getting started.  No
PPMC, yet, for which we need more Mentors.


These have nothing to do with when it entered incubation; the
need for a PPMC has been there right along, and the 'ASF
community building' is a sine qua non.  (I have no opinion,
myself, about the degree of 'ASF community building' that
has occurred in ActiveMQ.)


When AMQ entered the incubator as a sponsored project from Geronimo,  
the current understanding of incubator rules was that AMQ would  
simply use the Geronimo pmc since the Geronimo pmc is expected to be  
the home for the project.  Since then the incubator rules have been  
rewritten several time and based on the emails I saw today, the  
current rules that Noel is promoting (3+ mentors) hasn't even be  
approved by the incubator.  I personally find this incredibly  
frustrating, so please take my comments with a grain of salt.


If you ask me setting up a separate pmc for these projects is an  
incrediably bad idea.  Our objective is to create a single community  
between Geronimo, ActiveMQ, OpenEJB, ServiceMix and WADI.  Putting  
these projects into separate boxes makes this very difficult.


I would like to know, why have the incubator rules changed to, in my  
opinion, force all projects TLP?   Maybe the incubator is the wrong  
place to bring these types projects.  Is there another process to  
bring in a project we plan on integrating?  If not, maybe the board  
should consider setting something else up.


-dain




Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread Rodent of Unusual Size
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Dain Sundstrom wrote:
> 
> I understand this concern and agree with the solution, but we should 
> remember that AMQ entered the incubator before this was a rule, so I 
> for one didn't think it appled to them, since they are so close to 
> graduation.

'So close to graduation'?  Whence comes that?  I think that
proximity is still very much up in the air, particularly given
Noel's opinion that

> .. and the ASF community building is only just getting started.  No 
> PPMC, yet, for which we need more Mentors.

These have nothing to do with when it entered incubation; the
need for a PPMC has been there right along, and the 'ASF
community building' is a sine qua non.  (I have no opinion,
myself, about the degree of 'ASF community building' that
has occurred in ActiveMQ.)
- --
#kenP-)}

Ken Coar, Sanagendamgagwedweinini  http://Ken.Coar.Org/
Author, developer, opinionist  http://Apache-Server.Com/

"Millennium hand and shrimp!"
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.4 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iQCVAwUBRBYW95rNPMCpn3XdAQJuJwP8Cv+Waz8q0lTJMeHM72nTNNzeyBBjczGf
y+2l6vKrY65ueZLXOOCEZ4lBEScfsYMaiZ/YuHBgd25Gq//SVqH6fnkRUK8V63JC
4ArlE4+ZKg/lvt1msCC2YeNiFZxG1nC1OB3iK4M+QncNDEJaDhFWxT7Vqz7vt1vY
iN5arjQ6Mzc=
=pzI1
-END PGP SIGNATURE-


Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread John Sisson

Dain Sundstrom wrote:

On Mar 13, 2006, at 8:16 AM, Noel J. Bergman wrote:


Alan,

Others are commenting on the infrastructure issues still on the plate 
for

ActiveMQ.  And, yes, migrating out of JIRA is a PITA.  Jeff is proposing
that we end up running lots of JIRA instances because we have to pull in
several sets of JIRA imports from atlassian, codehaus, and elsewhere.  I
have no idea when Atlassian will support single project imports from 
JIRA as

it does from Bugzilla.


I agree it is important to have as much as possible on apache 
hardware.  It was my understanding until I read this thread here, that 
infrastructure was fine with leaving JIRAs for imported projects 
hosted remotely until the JIRA had a better import tool.  With the new 
possibility that Jeff Turner laid for an import, I think we should 
work on having one big import for everything that is moving or has 
moved from codehaus.  This would let us grab the ActiveMQ, ServiceMix, 
OpenEJB, and XBean jiras.


Is ActiveMQ and ServiceMix's JIRAs running at codehaus, it doesn't 
appear to be (tracert showed a different network path compared with 
codehaus)?


As with the other issues here, this is a NEW possibility that I for 
one wasn't even aware existed until John brought it up.  I don't think 
we should hold this against the AMQ community.


I don't like holding up progress, but my concern is how much incentive 
the project would have to move JIRA after incubation and whether it 
would be better to do it now?  Also no date has been given by Atlassian 
for when JIRA will have a better import tool - we could be waiting a 
while.  The "Has the project migrated to our infrastructure?" item on 
the status page seems misleading/pointless if a number of the project's 
services are running on external systems.


Glad to see Hiram has started a thread on Infrastructure to discuss the 
migration.


John
On the community side, we're still a bit shy of Mentors on ActiveMQ 
(James
is the only one, and we are looking for at least 3 per project), and 
the ASF
community building is only just getting started.  No PPMC, yet, for 
which we

need more Mentors.

There has been some concern about growth rate within the Incubator.
Requiring 3+ Mentors to help oversee projects puts a natural, but 
scalable,
limit on our growth.  In the meantime, we have some growing pains, 
because

projects are already here, and lack the resources.


I understand this concern and agree with the solution, but we should 
remember that AMQ entered the incubator before this was a rule, so I 
for one didn't think it appled to them, since they are so close to 
graduation.  Anyway, I think we can easily get a few more people to be 
mentors.  I certainly will volunteer, but I don't think I qualify due 
to the member restriction.


-dain





Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread Dain Sundstrom

On Mar 13, 2006, at 8:16 AM, Noel J. Bergman wrote:


Alan,

Others are commenting on the infrastructure issues still on the  
plate for
ActiveMQ.  And, yes, migrating out of JIRA is a PITA.  Jeff is  
proposing
that we end up running lots of JIRA instances because we have to  
pull in
several sets of JIRA imports from atlassian, codehaus, and  
elsewhere.  I
have no idea when Atlassian will support single project imports  
from JIRA as

it does from Bugzilla.


I agree it is important to have as much as possible on apache  
hardware.  It was my understanding until I read this thread here,  
that infrastructure was fine with leaving JIRAs for imported projects  
hosted remotely until the JIRA had a better import tool.  With the  
new possibility that Jeff Turner laid for an import, I think we  
should work on having one big import for everything that is moving or  
has moved from codehaus.  This would let us grab the ActiveMQ,  
ServiceMix, OpenEJB, and XBean jiras.


As with the other issues here, this is a NEW possibility that I for  
one wasn't even aware existed until John brought it up.  I don't  
think we should hold this against the AMQ community.


On the community side, we're still a bit shy of Mentors on ActiveMQ  
(James
is the only one, and we are looking for at least 3 per project),  
and the ASF
community building is only just getting started.  No PPMC, yet, for  
which we

need more Mentors.

There has been some concern about growth rate within the Incubator.
Requiring 3+ Mentors to help oversee projects puts a natural, but  
scalable,
limit on our growth.  In the meantime, we have some growing pains,  
because

projects are already here, and lack the resources.


I understand this concern and agree with the solution, but we should  
remember that AMQ entered the incubator before this was a rule, so I  
for one didn't think it appled to them, since they are so close to  
graduation.  Anyway, I think we can easily get a few more people to  
be mentors.  I certainly will volunteer, but I don't think I qualify  
due to the member restriction.


-dain


Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread Hiram Chirino
I'll start at thread on infrastructure to discuss how best to get the
JIRA migration done.

Regards,
Hiram

On 3/13/06, Justin Erenkrantz <[EMAIL PROTECTED]> wrote:
> On 3/13/06, Hiram Chirino <[EMAIL PROTECTED]> wrote:
> > I agree that we should work on getting our JIRA in the ASF infrastructure!
> > But I don' think accomplishing that task should be gate the limit's a
> > poddling from graduation from the incubator.
>
> Well, I do.  At the end of the Incubation process, it must have all
> resources on ASF-standard resources or I will vote -1.
>
> > I did a quick check and it
> > seems even the Maven TLP's JIRA is not on Apache infrastructure. see:
> > http://maven.apache.org/issue-tracking.html
>
> I would not look at Maven as a good example of where the lines should
> be drawn.  -- justin
>


--
Regards,
Hiram


Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread James Strachan

On 13 Mar 2006, at 17:27, Justin Erenkrantz wrote:

On 3/13/06, Hiram Chirino <[EMAIL PROTECTED]> wrote:
I agree that we should work on getting our JIRA in the ASF  
infrastructure!

But I don' think accomplishing that task should be gate the limit's a
poddling from graduation from the incubator.


Well, I do.  At the end of the Incubation process, it must have all
resources on ASF-standard resources or I will vote -1.


Given the complexity of moving just a single project from a JIRA  
server into a different server, how about we create a brand new JIRA  
project on Apache's JIRA server for ActiveMQ and leave the old one  
around for legacy pre-Apache stuff?


James
---
http://radio.weblogs.com/0112098/



Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread Justin Erenkrantz
On 3/13/06, Hiram Chirino <[EMAIL PROTECTED]> wrote:
> I agree that we should work on getting our JIRA in the ASF infrastructure!
> But I don' think accomplishing that task should be gate the limit's a
> poddling from graduation from the incubator.

Well, I do.  At the end of the Incubation process, it must have all
resources on ASF-standard resources or I will vote -1.

> I did a quick check and it
> seems even the Maven TLP's JIRA is not on Apache infrastructure. see:
> http://maven.apache.org/issue-tracking.html

I would not look at Maven as a good example of where the lines should
be drawn.  -- justin


Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread Hiram Chirino
Hi Noel,

We've got a JIRA out there to create the PPMC lists:
https://issues.apache.org/jira/browse/INFRA-757

But perhaps it has fallen through the cracks.  Should I ping the
infrastructure mailing lists about this issue?  Any help with this
would be most appreciated!

Regards,
Hiram


On 3/13/06, Noel J. Bergman <[EMAIL PROTECTED]> wrote:
> Alan,
>
> Others are commenting on the infrastructure issues still on the plate for
> ActiveMQ.  And, yes, migrating out of JIRA is a PITA.  Jeff is proposing
> that we end up running lots of JIRA instances because we have to pull in
> several sets of JIRA imports from atlassian, codehaus, and elsewhere.  I
> have no idea when Atlassian will support single project imports from JIRA as
> it does from Bugzilla.
>
> On the community side, we're still a bit shy of Mentors on ActiveMQ (James
> is the only one, and we are looking for at least 3 per project), and the ASF
> community building is only just getting started.  No PPMC, yet, for which we
> need more Mentors.
>
> There has been some concern about growth rate within the Incubator.
> Requiring 3+ Mentors to help oversee projects puts a natural, but scalable,
> limit on our growth.  In the meantime, we have some growing pains, because
> projects are already here, and lack the resources.
>
> --- Noel
>
>

--
Regards,
Hiram


RE: ActiveMQ Graduation From Incubator

2006-03-13 Thread Noel J. Bergman
Alan,

Others are commenting on the infrastructure issues still on the plate for
ActiveMQ.  And, yes, migrating out of JIRA is a PITA.  Jeff is proposing
that we end up running lots of JIRA instances because we have to pull in
several sets of JIRA imports from atlassian, codehaus, and elsewhere.  I
have no idea when Atlassian will support single project imports from JIRA as
it does from Bugzilla.

On the community side, we're still a bit shy of Mentors on ActiveMQ (James
is the only one, and we are looking for at least 3 per project), and the ASF
community building is only just getting started.  No PPMC, yet, for which we
need more Mentors.

There has been some concern about growth rate within the Incubator.
Requiring 3+ Mentors to help oversee projects puts a natural, but scalable,
limit on our growth.  In the meantime, we have some growing pains, because
projects are already here, and lack the resources.

--- Noel



Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread Hiram Chirino
Hi John,On 3/13/06, John Sisson <[EMAIL PROTECTED]> wrote:
Alan D. Cabrera wrote:> This is not a vote, but simply a discussion about the graduation of> ActiveMQ from the Incubator. The status file is located here:>> 
https://svn.apache.org/repos/asf/incubator/activemq/trunk/STATUS>> We are proactively seeking feedback in the interest of graduation.>>> Regards,> Alan>In regards to the infrastructure items, IMHO JIRA is a critical service
required by the project and should be run on ASF infrastructure.I agree that we should work on getting our JIRA in the ASF infrastructure!  But I don' think accomplishing that task should be gate the limit's a poddling from graduation from the incubator.  I did a quick check and it seems even the Maven TLP's JIRA is not on Apache infrastructure. see:  
http://maven.apache.org/issue-tracking.html 
Some of the benefits of moving ActiveMQ's JIRA data to the ASF would be:* JIRA svn integration (ability to see svn changes associated with aJIRA issue)* Backups of JIRA data under ASF controlSeems like now would be a good time to do this, but currently JIRA does
not have the ability to import data for a single project from dataexported from another JIRA instance (http://jira.atlassian.com/browse/JRA-1604 ).
Agreed.  I would think this one of the biggest problems that the ASF infrastructure team has with JIRA.Regards,Hiram
Would it be practical to follow Jeff Turner's suggestion inhttps://issues.apache.org/jira/browse/INFRA-713 , e.g. downloading acopy of JIRA 'Standalone' (built-in hsql db), load the data into it,
delete everything not relating to ActiveMQ, and create an export andthen run ActiveMQ's JIRA in a separate JIRA instance?Once JIRA's import/export is enhanced we then could import the data intothe main ASF instance.
Thoughts?Regards,John-- Regards,Hiram


Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread Alan D. Cabrera

John Sisson wrote:

Alan D. Cabrera wrote:

This is not a vote, but simply a discussion about the graduation of
ActiveMQ from the Incubator. The status file is located here:

https://svn.apache.org/repos/asf/incubator/activemq/trunk/STATUS

We are proactively seeking feedback in the interest of graduation.


Regards,
Alan

In regards to the infrastructure items, IMHO JIRA is a critical 
service required by the project and should be run on ASF infrastructure.


Some of the benefits of moving ActiveMQ's JIRA data to the ASF would be:
* JIRA svn integration (ability to see svn changes associated with a 
JIRA issue)

* Backups of JIRA data under ASF control

Seems like now would be a good time to do this, but currently JIRA 
does not have the ability to import data for a single project from 
data exported from another JIRA instance ( 
http://jira.atlassian.com/browse/JRA-1604 ).


Would it be practical to follow Jeff Turner's suggestion in 
https://issues.apache.org/jira/browse/INFRA-713 , e.g. downloading a 
copy of JIRA 'Standalone' (built-in hsql db), load the data into it, 
delete everything not relating to ActiveMQ, and create an export and 
then run ActiveMQ's JIRA in a separate JIRA instance?


Once JIRA's import/export is enhanced we then could import the data 
into the main ASF instance.


Thoughts? 


Sounds great.  It seems that you have a good start on this.  Thanks for 
volunteering!


/me ducks out of the room...


Regards,
Alan





Re: ActiveMQ Graduation From Incubator

2006-03-13 Thread John Sisson

Alan D. Cabrera wrote:

This is not a vote, but simply a discussion about the graduation of
ActiveMQ from the Incubator. The status file is located here:

https://svn.apache.org/repos/asf/incubator/activemq/trunk/STATUS

We are proactively seeking feedback in the interest of graduation.


Regards,
Alan

In regards to the infrastructure items, IMHO JIRA is a critical service 
required by the project and should be run on ASF infrastructure.


Some of the benefits of moving ActiveMQ's JIRA data to the ASF would be:
* JIRA svn integration (ability to see svn changes associated with a 
JIRA issue)

* Backups of JIRA data under ASF control

Seems like now would be a good time to do this, but currently JIRA does 
not have the ability to import data for a single project from data 
exported from another JIRA instance ( 
http://jira.atlassian.com/browse/JRA-1604 ).


Would it be practical to follow Jeff Turner's suggestion in 
https://issues.apache.org/jira/browse/INFRA-713 , e.g. downloading a 
copy of JIRA 'Standalone' (built-in hsql db), load the data into it, 
delete everything not relating to ActiveMQ, and create an export and 
then run ActiveMQ's JIRA in a separate JIRA instance?


Once JIRA's import/export is enhanced we then could import the data into 
the main ASF instance.


Thoughts?

Regards,

John