You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by Nicola Ken Barozzi <ni...@apache.org> on 2003/11/07 10:15:14 UTC

[Possible Incubation] Apache Repo

- repository and board are CCed -
- please reply to general@incubator.apache.org -

There is a discussion going on over at repository@apache.org about the 
creation of an Apache jar and artifact repository. The discussion is 
being constructive and is bringing in people from diverse efforts: 
Maven, Ruper, Greebo, infrastructure@apache.org, and others.

Ruper and Greebo developers have explicitly stated their intentions of 
working on an implementation of it, and Maven developers are starting to 
join now, but I'm sure some will like to join the coding effort too. In 
fact the Maven repository format has been the starting point of the 
discussion.

Here is the info about this effort:
http://nagoya.apache.org/wiki/apachewiki.cgi?ASFRepository

While on the other list the discussion about the spec is progressing 
nicely, I'd like to start discussing about the Project effort, with the 
developers and code, as discussing without code gets futile fast.

* Who would like to Sponsor it? board? pmc@incubator?
* Mentors?
* How to start?

Proposal:
This is a new kind of incubation that I'd like to try. It's not an 
established project that wants to come to Apache, rather an effort that 
wants to try and see if it can work, similar to Geronimo but coming 
mainly from Apache committers.

Hence I propose that the Incubator PMC decides to start incubating the 
project as "Repo", giving it:

  - dev list (just for code; spec discussions at repository@apache.org)
  - site under incubator.apache.org/projects/repo to write down the spec
  - CVS module 'incubator-repo' open to all Apache committers that
    request it and to the initial external contributors.
    In the CVS module the developers can place their current codebases
    that they donate and we wait and see what happens.

Over at krysalis.org and here with Geronimo I have seen that when there 
is a bunch of people with the same itch and an open CVS space, things 
happen.

Shall we try it with Repo?

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
             - verba volant, scripta manent -
    (discussions get forgotten, just code remains)
---------------------------------------------------------------------


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by Nicola Ken Barozzi <ni...@apache.org>.
Geir Magnusson Jr. wrote:
> 
> On Friday, November 7, 2003, at 11:29 AM, Nicola Ken Barozzi wrote:
> 
>>
>> It has already been done there, at http://www.krysalis.org/ruper/ .
> 
> Wasn't it already done with jjar and whatever the thingy in Maven is 
> called?

Not really, it does more... and to some extent less (IIUC it does not 
yet use distributed repositories).

I know JJAR quite well, as I have been using it and proposing for quite 
some time, and I also had some patches ready, but then we decided to 
follow the Maven repository format not to duplicate repositories.

>> The developers of Ruper in the meantime have become Apache committers 
>> (two for Gump and one is coming in with JUDDI). The issue here is that 
>> Maven already has some code for this, and 
>> http://greebo.sourceforge.net/ is partecipating too.
>>
>> If you want to call it "Incubating Ruper and have others join" it's ok 
>> for me, as my initial idea is to use Ruper as a starting codebase, and 
>> have others add stuff, but I didn't want to make this a prerequisite 
>> for a Repo project.
> 
> Maybe I should kick a release of jjar out :)

Maybe you should seriously think of putting JJAR on the table and join 
both the repository@apache.org and the Repo efforts  :-)

So ok, the Repo project would start with Ruper, Greebo, Maven-fetch, and 
JJAR as starting points. I didn't count JJAR in as it has been still for 
so long.

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
             - verba volant, scripta manent -
    (discussions get forgotten, just code remains)
---------------------------------------------------------------------



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by Nicola Ken Barozzi <ni...@apache.org>.
Jason van Zyl wrote:
> On Fri, 2003-11-07 at 12:16, Nicola Ken Barozzi wrote:
> 
>>peter royal wrote:
>>
>>>On Nov 7, 2003, at 11:29 AM, Nicola Ken Barozzi wrote:
>>>
>>>>The developers of Ruper in the meantime have become Apache committers 
>>>>(two for Gump and one is coming in with JUDDI). The issue here is that 
>>>>Maven already has some code for this, and 
>>>>http://greebo.sourceforge.net/ is partecipating too.
>>>>
>>>>If you want to call it "Incubating Ruper and have others join" it's ok 
>>>>for me, as my initial idea is to use Ruper as a starting codebase, and 
>>>>have others add stuff, but I didn't want to make this a prerequisite 
>>>>for a Repo project.
>>>
>>>The initial email wasn't clear about what codebases would be part of 
>>>this initiative.
>>>
>>>So if Ruper is the codebase that would be attached, the initiative 
>>>should be re-titled as you suggest.
>>
>>>This really smells like an extension and abstraction of what Maven 
>>>provides and initiated. Why not do this effort under their umbrella?
>>
>>There is no need to be provocative, Peter.
>>
>>The Krysalis jar repository effort started well before we ever knew of 
>>the Maven one, by using Jakarta Commons JJAR. Actually the Maven one had 
>>not even stared at that time.)
> 
> I am not trying to stir the pot here, but you really can't make
> statements that are purely conjecture and not be held accountable. It's
> simply not fair to insinuate there is validity in your efforts simply
> because you think you started first.

Jason, I have read this mail, and I can say that finally this is the 
time to get things straight and finish all this miscommunication nonsense.

First of all, I don't think I started first. I don't even care who 
started first, as I believe that who thinks that others steal his ideas 
has little ideas to start with ;-)

> It doesn't matter who started first because different implementations
> are not a bad thing, but please Nicola get your facts straight. You used
> JJAR for your first implementation, yes? Then I'm not sure how it has
> come to pass that you don't know that JJAR was written as a direct
> result of my discussions with Geir and it was for the express purpose of
> using a repository of JARs and I didn't see you on the scene anywhere
> for quite some time after it's creation. Maven started in Alexandria and
> moved to Turbine which is where most people think it started.

Yup, I am aware of the above and AFAIK I can confirm that it's correct.

What I want to say is that at that time, we were not aware of this, and 
not aware of your Maven plans. All we could see is a JJAR project on 
Commons. This is in reply to Peter Royal's insinuation that we copied a 
feature of Maven instead of collaborating.

> Using a repository isn't a revelation, the idea has been in existence
> long before either of us came around. 

Yup.

> But that's not my point, the point
> is it neither matters who was first (because there's probably someone
> who tried this long before either of our efforts if we actually looked)
> and using that as a reason for the existence of a codebase is a
> pointless one. 

Yup.

> If you have another implementation of something akin to
> what Maven does then that's completely fine but let it stand on its
> merit and stop pointing your fingers at Maven. 

I'm not trying to point my fingers at Maven. If it comes out like this, 
it's unintentional, and please accept my apologies.

> If you want to incubate Ruper then say so. That's perfectly valid. It is
> also perfectly valid to have a competitor to aspects of the
> functionality of Maven. I have no problem with that, but stop calling it
> a grand community effort while intoning that Maven is devoid of a
> community persuing these same things that you are talking about in
> addition to dismissing the Maven developers a close minded bunch because
> we chose a path (i.e. not merging with your effort) that we considered
> was in Maven's best interest. 

Again, I'm not trying to do this, sorry if it seems so. On 
repository@apache.org I have tried to be clear on this, but because of 
past incomprehensions it's hard.

>>There have been multiple requests from us to merge efforts, but to no 
>>avail. 
> 
> You have cited this reason many, many times but what you originally
> asked for was entirely unreasonable. You asked for an entire merger of
> codebases and developers which the Maven committers were not prepared to
> accept. It was posed essentially as an ultimatum which stated that a
> flat out merger was the only acceptable solution. You decided not to
> participate and the situation was made worse when essentially implied
> the Maven developers were a bunch of thieves or more specifically that I
> took your idea and ran with it.

I never said that you took my idea, and never posed an ultimatum.

What I did see was massive duplication of efforts in Apache and 
elsewhere that I did not want to accept.

There is Gump, why POM?
There is Ant, why Jelly?
There is Centipede cents, why Maven plugins?

Simply that was the decision of the Maven developers, ok.
But please leave me the possibility to have a different opinion on this 
way of doing things.

http://www.jroller.com/page/nicolaken/20030228

>>I don't see why would an indipendent and cross-project repository effort 
>>and library have to be under Maven.
> 
> It certainly doesn't but you also can't ignore what Maven has done for
> the notion of a repository.

Of course.

> In any case I think that if you wish to incubate Ruper then I am +1.

Thanks. I hope that the results will be so good that Maven will be 
compelled to use it. I also hope that Maven developers will participate 
in the project.

Jason, thanks for your fair and clear comments. I hope that this can put 
  an end to our incomprehensions about what happened.

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
             - verba volant, scripta manent -
    (discussions get forgotten, just code remains)
---------------------------------------------------------------------



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by Jason van Zyl <jv...@maven.org>.
On Fri, 2003-11-07 at 12:16, Nicola Ken Barozzi wrote:
> peter royal wrote:
> > On Nov 7, 2003, at 11:29 AM, Nicola Ken Barozzi wrote:
> > 
> >> The developers of Ruper in the meantime have become Apache committers 
> >> (two for Gump and one is coming in with JUDDI). The issue here is that 
> >> Maven already has some code for this, and 
> >> http://greebo.sourceforge.net/ is partecipating too.
> >>
> >> If you want to call it "Incubating Ruper and have others join" it's ok 
> >> for me, as my initial idea is to use Ruper as a starting codebase, and 
> >> have others add stuff, but I didn't want to make this a prerequisite 
> >> for a Repo project.
> > 
> > The initial email wasn't clear about what codebases would be part of 
> > this initiative.
> > 
> > So if Ruper is the codebase that would be attached, the initiative 
> > should be re-titled as you suggest.
>  >
> > This really smells like an extension and abstraction of what Maven 
> > provides and initiated. Why not do this effort under their umbrella?
> 
> There is no need to be provocative, Peter.
> 
> The Krysalis jar repository effort started well before we ever knew of 
> the Maven one, by using Jakarta Commons JJAR. Actually the Maven one had 
> not even stared at that time.)

I am not trying to stir the pot here, but you really can't make
statements that are purely conjecture and not be held accountable. It's
simply not fair to insinuate there is validity in your efforts simply
because you think you started first.

It doesn't matter who started first because different implementations
are not a bad thing, but please Nicola get your facts straight. You used
JJAR for your first implementation, yes? Then I'm not sure how it has
come to pass that you don't know that JJAR was written as a direct
result of my discussions with Geir and it was for the express purpose of
using a repository of JARs and I didn't see you on the scene anywhere
for quite some time after it's creation. Maven started in Alexandria and
moved to Turbine which is where most people think it started.

Using a repository isn't a revelation, the idea has been in existence
long before either of us came around. But that's not my point, the point
is it neither matters who was first (because there's probably someone
who tried this long before either of our efforts if we actually looked)
and using that as a reason for the existence of a codebase is a
pointless one. If you have another implementation of something akin to
what Maven does then that's completely fine but let it stand on its
merit and stop pointing your fingers at Maven. 

If you want to incubate Ruper then say so. That's perfectly valid. It is
also perfectly valid to have a competitor to aspects of the
functionality of Maven. I have no problem with that, but stop calling it
a grand community effort while intoning that Maven is devoid of a
community persuing these same things that you are talking about in
addition to dismissing the Maven developers a close minded bunch because
we chose a path (i.e. not merging with your effort) that we considered
was in Maven's best interest. 

> There have been multiple requests from us to merge efforts, but to no 
> avail. 

You have cited this reason many, many times but what you originally
asked for was entirely unreasonable. You asked for an entire merger of
codebases and developers which the Maven committers were not prepared to
accept. It was posed essentially as an ultimatum which stated that a
flat out merger was the only acceptable solution. You decided not to
participate and the situation was made worse when essentially implied
the Maven developers were a bunch of thieves or more specifically that I
took your idea and ran with it.

> I don't see why would an indipendent and cross-project repository effort 
> and library have to be under Maven.

It certainly doesn't but you also can't ignore what Maven has done for
the notion of a repository.

In any case I think that if you wish to incubate Ruper then I am +1.

-- 
jvz.

Jason van Zyl
jason@zenplex.com
http://tambora.zenplex.org

In short, man creates for himself a new religion of a rational
and technical order to justify his work and to be justified in it.
  
  -- Jacques Ellul, The Technological Society


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by Jason van Zyl <jv...@maven.org>.
On Fri, 2003-11-07 at 13:15, Adam R. B. Jack wrote:
> > > This really smells like an extension and abstraction of what Maven
> > > provides and initiated. Why not do this effort under their umbrella?
> 
> I believe the repository (or plural) needs artefacts beyond what Java
> projects provide. 

Maven is certainly focused on Java, and this is where Maven will remain
focused. I believe the Java focus is what the repository will
predominantly be used for.

> I believe it was intended to contain all apache content,
> and not just Maven built artefacts. Maven's is a nice tight use case, and
> very valuable for that, but it is only one.

It is but I honestly don't see many tools being developed that will
actually be used outside the Java landscape. Python and Perl have their
own Maven-like mechanisms already and in the domain of native code this
is predominantly handled by package management systems for the various
platforms.

As witnessed by the submission of various candidate codebases: Ruper,
Greebo, JJAR. All Java-centric like Maven and I honestly don't think
anyone at Apache is going to try to complete with the likes of CPAN and
things like up2date or apt-get to nab libraries for C/C++ are likely not
going to be superceded by any repository scheme dreamed up here.

My projection would be that it would start as a universal effort but
support for other languages will fall to the wayside except for Java
because it is the only language which is in the embryonic stages of its
development of tools like CPAN.

Other than simply downloading, the use of the repository for languages
other than Java will be scant in my estimation. Simply because the tools
already exist for building, pulling down dependencies and all the rest
of it.

Anyone who is talking on the repository@apache list appears to be a Java
head. I've seen Justin pop in to answer some queries about httpd but the
initiative seems to be heavily biased towards Java already for reasons
stated above.


-- 
jvz.

Jason van Zyl
jason@zenplex.com
http://tambora.zenplex.org

In short, man creates for himself a new religion of a rational
and technical order to justify his work and to be justified in it.
  
  -- Jacques Ellul, The Technological Society


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


RE: [Possible Incubation] Apache Repo

Posted by Alex Karasulu <ao...@bellsouth.net>.
I completely agree with you Adam perhaps you email should also go to the
repository@apache.org list as well.  

> > > This really smells like an extension and abstraction of what Maven
> > > provides and initiated. Why not do this effort under their umbrella?
> 
> I believe the repository (or plural) needs artefacts beyond what Java
> projects provide. I believe it was intended to contain all apache content,
> and not just Maven built artefacts. Maven's is a nice tight use case, and
> very valuable for that, but it is only one.
> 
> Ant is another possible umbrella, and perhaps one with better coverage. At
> the end of the day the commonality is the repository, the specification,
> and
> tools for manipulating content. I believe this is sufficiently important
> to
> the whole community to warrant it's own project.
> 
> regards,
> 
> Adam
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org




---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by "Adam R. B. Jack" <aj...@trysybase.com>.
> > This really smells like an extension and abstraction of what Maven
> > provides and initiated. Why not do this effort under their umbrella?

I believe the repository (or plural) needs artefacts beyond what Java
projects provide. I believe it was intended to contain all apache content,
and not just Maven built artefacts. Maven's is a nice tight use case, and
very valuable for that, but it is only one.

Ant is another possible umbrella, and perhaps one with better coverage. At
the end of the day the commonality is the repository, the specification, and
tools for manipulating content. I believe this is sufficiently important to
the whole community to warrant it's own project.

regards,

Adam


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by Nicola Ken Barozzi <ni...@apache.org>.
peter royal wrote:
> On Nov 7, 2003, at 11:29 AM, Nicola Ken Barozzi wrote:
> 
>> The developers of Ruper in the meantime have become Apache committers 
>> (two for Gump and one is coming in with JUDDI). The issue here is that 
>> Maven already has some code for this, and 
>> http://greebo.sourceforge.net/ is partecipating too.
>>
>> If you want to call it "Incubating Ruper and have others join" it's ok 
>> for me, as my initial idea is to use Ruper as a starting codebase, and 
>> have others add stuff, but I didn't want to make this a prerequisite 
>> for a Repo project.
> 
> The initial email wasn't clear about what codebases would be part of 
> this initiative.
> 
> So if Ruper is the codebase that would be attached, the initiative 
> should be re-titled as you suggest.
 >
> This really smells like an extension and abstraction of what Maven 
> provides and initiated. Why not do this effort under their umbrella?

There is no need to be provocative, Peter.

The Krysalis jar repository effort started well before we ever knew of 
the Maven one, by using Jakarta Commons JJAR. Actually the Maven one had 
not even stared at that time.)

There have been multiple requests from us to merge efforts, but to no 
avail. In the end, to reduce duplication, we decided to follow the Maven 
repo layout, but since at that time Maven was in such a flux, and we are 
using Ant, we did our own tool, Ruper. Since then Ruper has grown, and 
Krysalis now has Ruper2 and Version, which are much more advanced.

I don't see why would an indipendent and cross-project repository effort 
and library have to be under Maven.

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
             - verba volant, scripta manent -
    (discussions get forgotten, just code remains)
---------------------------------------------------------------------



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


RE: [Possible Incubation] Apache Repo

Posted by Alex Karasulu <ao...@bellsouth.net>.
Guys,


I'm afraid to get in the middle of this but I just wanted to give my
perspective on it.  

> > If you want to call it "Incubating Ruper and have others join" it's ok
> > for me, as my initial idea is to use Ruper as a starting codebase, and
> > have others add stuff, but I didn't want to make this a prerequisite
> > for a Repo project.
> 
> This really smells like an extension and abstraction of what Maven
> provides and initiated. Why not do this effort under their umbrella?

Perhaps this could just be a Maven sub project, however because of 
the interest the concept of a repository has across the board, I think
it should be incubated to attract people from the many corners of Apache
and outside of Apache.  The incubator is a good place to attract people
and grow a community just around the repo concept.  The lower barrier to 
entry for committers makes it more hmmmm nurturing of a community.

I think the important thing is to end up with one de facto standard rather
than have duplicated code doing the same thing all over the place.  Use as
Jason suggested decorators to enable any specific uses of the repository
to stretch it for the various needs different groups may have.

In the end Apache should be the winner.  People should go to Apache to
get the library in one place.  You don't want to confuse users with
multiple implementations.  One at Avalon, another in Maven and so on.  
The incubator gives this initiative a clean start inviting all the 
players - it's a chance to make it a bit more open to people like myself.
Incubation will let me put my two cents in and I know you would 
appreciate that in terms of the openness the incubator opens to attract
new people.

Once the community is formed put it back into maven.  But keep it 
there.  Don't have a repo at Avalon and XYZ and so on.

Again I can't help but volunteer my 2 cents.

Sincerely,
Alex




---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by peter royal <pr...@apache.org>.
On Nov 7, 2003, at 11:29 AM, Nicola Ken Barozzi wrote:
> The developers of Ruper in the meantime have become Apache committers 
> (two for Gump and one is coming in with JUDDI). The issue here is that 
> Maven already has some code for this, and 
> http://greebo.sourceforge.net/ is partecipating too.
>
> If you want to call it "Incubating Ruper and have others join" it's ok 
> for me, as my initial idea is to use Ruper as a starting codebase, and 
> have others add stuff, but I didn't want to make this a prerequisite 
> for a Repo project.

The initial email wasn't clear about what codebases would be part of 
this initiative.

So if Ruper is the codebase that would be attached, the initiative 
should be re-titled as you suggest.

This really smells like an extension and abstraction of what Maven 
provides and initiated. Why not do this effort under their umbrella?
-pete


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by "Geir Magnusson Jr." <ge...@optonline.net>.
On Friday, November 7, 2003, at 11:29 AM, Nicola Ken Barozzi wrote:
>
> It has already been done there, at http://www.krysalis.org/ruper/ .

Wasn't it already done with jjar and whatever the thingy in Maven is 
called?

>
> The developers of Ruper in the meantime have become Apache committers 
> (two for Gump and one is coming in with JUDDI). The issue here is that 
> Maven already has some code for this, and 
> http://greebo.sourceforge.net/ is partecipating too.
>
> If you want to call it "Incubating Ruper and have others join" it's ok 
> for me, as my initial idea is to use Ruper as a starting codebase, and 
> have others add stuff, but I didn't want to make this a prerequisite 
> for a Repo project.

Maybe I should kick a release of jjar out :)


geir

-- 
Geir Magnusson Jr                                   203-247-1713(m)
geirm@optonline.net


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by Nicola Ken Barozzi <ni...@apache.org>.
peter royal wrote:
> On Nov 7, 2003, at 4:15 AM, Nicola Ken Barozzi wrote:
> 
>> Proposal:
>> This is a new kind of incubation that I'd like to try. It's not an 
>> established project that wants to come to Apache, rather an effort 
>> that wants to try and see if it can work, similar to Geronimo but 
>> coming mainly from Apache committers.
> 
> That seems contradictory to everything I thought the incubator was. It 
> was my understanding that the incubator was for existing projects that 
> wanted to move into the Apache umbrella? To be indoctrinated in the 
> 'Apache Way'.
> Even Geronimo came to the incubator with a codebase..

Well, there is already a codebase, more than one actually. It's written 
in the mail (you have snipped it in this reply). Think about how 
Geronimo started and you have a good idea of what I envision for this.

>> Hence I propose that the Incubator PMC decides to start incubating the 
>> project as "Repo", giving it:
>>
>>  - dev list (just for code; spec discussions at repository@apache.org)
>>  - site under incubator.apache.org/projects/repo to write down the spec
>>  - CVS module 'incubator-repo' open to all Apache committers that
>>    request it and to the initial external contributors.
>>    In the CVS module the developers can place their current codebases
>>    that they donate and we wait and see what happens.
> 
> What's the benefit of doing this in the incubator vs sf.net or java.net?

It has already been done there, at http://www.krysalis.org/ruper/ .

The developers of Ruper in the meantime have become Apache committers 
(two for Gump and one is coming in with JUDDI). The issue here is that 
Maven already has some code for this, and http://greebo.sourceforge.net/ 
is partecipating too.

If you want to call it "Incubating Ruper and have others join" it's ok 
for me, as my initial idea is to use Ruper as a starting codebase, and 
have others add stuff, but I didn't want to make this a prerequisite for 
a Repo project.

> If this goes through, how is this not the start of the incubator 
> becoming a 'sandbox' for ASF committers?

Unfortunatley no.

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
             - verba volant, scripta manent -
    (discussions get forgotten, just code remains)
---------------------------------------------------------------------



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by peter royal <pr...@apache.org>.
On Nov 7, 2003, at 4:15 AM, Nicola Ken Barozzi wrote:
> Proposal:
> This is a new kind of incubation that I'd like to try. It's not an 
> established project that wants to come to Apache, rather an effort 
> that wants to try and see if it can work, similar to Geronimo but 
> coming mainly from Apache committers.

That seems contradictory to everything I thought the incubator was. It 
was my understanding that the incubator was for existing projects that 
wanted to move into the Apache umbrella? To be indoctrinated in the 
'Apache Way'.

Even Geronimo came to the incubator with a codebase..

> Hence I propose that the Incubator PMC decides to start incubating the 
> project as "Repo", giving it:
>
>  - dev list (just for code; spec discussions at repository@apache.org)
>  - site under incubator.apache.org/projects/repo to write down the spec
>  - CVS module 'incubator-repo' open to all Apache committers that
>    request it and to the initial external contributors.
>    In the CVS module the developers can place their current codebases
>    that they donate and we wait and see what happens.

What's the benefit of doing this in the incubator vs sf.net or java.net?

If this goes through, how is this not the start of the incubator 
becoming a 'sandbox' for ASF committers?
-pete


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by Nicola Ken Barozzi <ni...@apache.org>.
Noel J. Bergman wrote:
>>There is a discussion going on over at repository@apache.org about the
>>creation of an Apache jar and artifact repository. The discussion is
>>being constructive and is bringing in people from diverse efforts:
>>Maven, Ruper, Greebo, infrastructure@apache.org, and others.
> 
>>* Who would like to Sponsor it? board? pmc@incubator?
> 
> An ASF Member (you) can sponsor it.  It doesn't take a PMC, unless that fell
> by the wayside somewhere along the way.

It needs a PMC that takes it under it's wings. In this case it seems it 
would be the Incubator PMC.

I can Mentor it, but it needs a PMC overseeing it and accepting it.

>>This is a new kind of incubation that I'd like to try. It's not an
>>established project that wants to come to Apache, rather an effort
>>that wants to try and see if it can work, similar to Geronimo but
>>coming mainly from Apache committers.
> 
> It doesn't seem different at all.  There are established codebases outside
> of the ASF, e.g., Ruper and Greebo, existing communities around them, and an
> ASF Member (you) interested in sponsoring them into the Incubator.
> 
> Given those criteria, it seems reasonable for it to follow the normal
> incubation process.  

Hmmm, the different thing is the fact that it's not a single community 
and codebase. Other than that, I tend to agree.

> There is no guarantee that it will succeed, but there
> isn't supposed to be one.

Yup, of course.

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
             - verba volant, scripta manent -
    (discussions get forgotten, just code remains)
---------------------------------------------------------------------



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by Nicola Ken Barozzi <ni...@apache.org>.
Greg Stein wrote:
> On Fri, Nov 07, 2003 at 03:18:05PM -0500, Noel J. Bergman wrote:
...
>>It doesn't seem different at all.  There are established codebases outside
>>of the ASF, e.g., Ruper and Greebo, existing communities around them, and an
>>ASF Member (you) interested in sponsoring them into the Incubator.
> 
> Not to mention that Maven is *already* an ASF project. And infrastructure
> is a pseudo-project, if you will :-)

Exactly! This is why I find this incubation a bit "different".

> I believe the "output" of the project would fall under the auspices of the
> infrastructure team. If it "feels big" and/or infrastructure is hinky
> about taking it on, *then* we could find a different home.

Yes, IMHO we should not decide the final destination of it. What we need 
though is some PMC that accepts to make incubation. I'll ask the 
Incubator PMC then.

-- 
Nicola Ken Barozzi                   nicolaken@apache.org
             - verba volant, scripta manent -
    (discussions get forgotten, just code remains)
---------------------------------------------------------------------



---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


Re: [Possible Incubation] Apache Repo

Posted by Greg Stein <gs...@lyra.org>.
On Fri, Nov 07, 2003 at 03:18:05PM -0500, Noel J. Bergman wrote:
> > There is a discussion going on over at repository@apache.org about the
> > creation of an Apache jar and artifact repository. The discussion is
> > being constructive and is bringing in people from diverse efforts:
> > Maven, Ruper, Greebo, infrastructure@apache.org, and others.
> 
> > * Who would like to Sponsor it? board? pmc@incubator?
> 
> An ASF Member (you) can sponsor it.  It doesn't take a PMC, unless that fell
> by the wayside somewhere along the way.
> 
> > This is a new kind of incubation that I'd like to try. It's not an
> > established project that wants to come to Apache, rather an effort
> > that wants to try and see if it can work, similar to Geronimo but
> > coming mainly from Apache committers.
> 
> It doesn't seem different at all.  There are established codebases outside
> of the ASF, e.g., Ruper and Greebo, existing communities around them, and an
> ASF Member (you) interested in sponsoring them into the Incubator.

Not to mention that Maven is *already* an ASF project. And infrastructure
is a pseudo-project, if you will :-)

I believe the "output" of the project would fall under the auspices of the
infrastructure team. If it "feels big" and/or infrastructure is hinky
about taking it on, *then* we could find a different home.

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org


RE: [Possible Incubation] Apache Repo

Posted by "Noel J. Bergman" <no...@devtech.com>.
> There is a discussion going on over at repository@apache.org about the
> creation of an Apache jar and artifact repository. The discussion is
> being constructive and is bringing in people from diverse efforts:
> Maven, Ruper, Greebo, infrastructure@apache.org, and others.

> * Who would like to Sponsor it? board? pmc@incubator?

An ASF Member (you) can sponsor it.  It doesn't take a PMC, unless that fell
by the wayside somewhere along the way.

> This is a new kind of incubation that I'd like to try. It's not an
> established project that wants to come to Apache, rather an effort
> that wants to try and see if it can work, similar to Geronimo but
> coming mainly from Apache committers.

It doesn't seem different at all.  There are established codebases outside
of the ASF, e.g., Ruper and Greebo, existing communities around them, and an
ASF Member (you) interested in sponsoring them into the Incubator.

Given those criteria, it seems reasonable for it to follow the normal
incubation process.  There is no guarantee that it will succeed, but there
isn't supposed to be one.

	--- Noel


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org