You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@avalon.apache.org by Stephen McConnell <mc...@apache.org> on 2002/11/29 00:18:55 UTC
[PROPOSAL] a clazz act
I would like to move forward with a joint initative between the Avalon
Dev. Team and the Jakarta Commons [clazz] guys to work up an
implementation of an Avalon meta model for component and service
defintions. I've already spoken to Stephen Colebourne and Dmitri
Plotnikov (committers on the [clazz] project) and there both keen to
work with us on this. From their point of view is about validation of
the [clazz] package against real requirements, and from our point of
view is about validation of the [clazz] package and delivery of a
standard Avalon meta solution.
To kick things off I have in mind the following:
* granting commit access to Stephen and Dmitri so they can work on a
sandbox project
* reciprical commit rights for concerned avalon committers to the
[clazz] project
so we can work on clazz (doc enhancement, unit test, debugging, etc)
Enabling co-development on a meta model implementation, leading to:
* validation and grounding of the [clazz] package based on our
requirements
* consolidation of current meta solutions under a single unified
object model
* capable of supporting mixed loading strategies
* capable of supporting model extension
Before moving forward on this, I wanted to get feedback from people
concerning the commit access exchange which doesn't follow the normal
model. However - there is plenty of evidence in the clazz project itself
that these guys know what they are doing. Secondly, this also means
voting here and there, so first of all I'll like get an idea of who here
are sufficiently interested in this to actually contribute.
Cheers, Steve.
--
Stephen J. McConnell
OSM SARL
digital products for a global economy
mailto:mcconnell@osm.net
http://www.osm.net
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Stephen McConnell <mc...@apache.org>.
Gary Shea wrote:
>Hi Steve --
>
>I have been studying the JMI/MOF stuff recently, just starting to get a
>feel for it. Seems to me clazz and JMI have a lot in common.
>
Yep.
>What are
>the distinguishing features that make clazz more interesting for this
>application?
>
Several observations:
(a) It is here and now and I don't see any problems in
dealing with JMI/MOF interoperability in the future
(b) The [clazz] package is a much more pragmatic approach
driven by implemetation requirements. JMI covers a lot
more space - basically a Java API to MOF - and MOF is a
much bigger animal.
(c) [clazz] offers a small and direct API
(d) [clazz] comes under Apache license
But I should make it clear that I'm not a MOF fan. Now considering that
most of guys on the JNI team know me - I'm sure this email will get to
one of more of them and they will ask for an explination - and the
answer will can summed up on one work - *heavy*, both technically and
politically.
:-)
Cheers, Steve.
>Regards,
>
> Gary
>
>
>
>>>>I would like to move forward with a joint initative between the Avalon
>>>>Dev. Team and the Jakarta Commons [clazz] guys to work up an
>>>>implementation of an Avalon meta model for component and service
>>>>defintions. I've already spoken to Stephen Colebourne and Dmitri
>>>>Plotnikov (committers on the [clazz] project) and there both keen to
>>>>work with us on this. From their point of view is about validation of
>>>>the [clazz] package against real requirements, and from our point of
>>>>view is about validation of the [clazz] package and delivery of a
>>>>standard Avalon meta solution.
>>>>
>>>>To kick things off I have in mind the following:
>>>>
>>>> * granting commit access to Stephen and Dmitri so they can work on
>>>>a sandbox project
>>>> * reciprical commit rights for concerned avalon committers to the
>>>>[clazz] project
>>>> so we can work on clazz (doc enhancement, unit test, debugging, etc)
>>>>
>>>>Enabling co-development on a meta model implementation, leading to:
>>>>
>>>> * validation and grounding of the [clazz] package based on our
>>>>requirements
>>>> * consolidation of current meta solutions under a single unified
>>>>object model
>>>> * capable of supporting mixed loading strategies
>>>> * capable of supporting model extension
>>>>
>>>>Before moving forward on this, I wanted to get feedback from people
>>>>concerning the commit access exchange which doesn't follow the normal
>>>>model. However - there is plenty of evidence in the clazz project
>>>>itself that these guys know what they are doing. Secondly, this also
>>>>means voting here and there, so first of all I'll like get an idea of
>>>>who here are sufficiently interested in this to actually contribute.
>>>>
>>>>Cheers, Steve.
>>>>
>>>>
>
>
>--
>To unsubscribe, e-mail: <ma...@jakarta.apache.org>
>For additional commands, e-mail: <ma...@jakarta.apache.org>
>
>
>
>
>
--
Stephen J. McConnell
OSM SARL
digital products for a global economy
mailto:mcconnell@osm.net
http://www.osm.net
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Gary Shea <sh...@gtsdesign.com>.
Hi Steve --
I have been studying the JMI/MOF stuff recently, just starting to get a
feel for it. Seems to me clazz and JMI have a lot in common. What are
the distinguishing features that make clazz more interesting for this
application?
Regards,
Gary
> > > I would like to move forward with a joint initative between the Avalon
> > > Dev. Team and the Jakarta Commons [clazz] guys to work up an
> > > implementation of an Avalon meta model for component and service
> > > defintions. I've already spoken to Stephen Colebourne and Dmitri
> > > Plotnikov (committers on the [clazz] project) and there both keen to
> > > work with us on this. From their point of view is about validation of
> > > the [clazz] package against real requirements, and from our point of
> > > view is about validation of the [clazz] package and delivery of a
> > > standard Avalon meta solution.
> > >
> > > To kick things off I have in mind the following:
> > >
> > > * granting commit access to Stephen and Dmitri so they can work on
> > > a sandbox project
> > > * reciprical commit rights for concerned avalon committers to the
> > > [clazz] project
> > > so we can work on clazz (doc enhancement, unit test, debugging, etc)
> > >
> > > Enabling co-development on a meta model implementation, leading to:
> > >
> > > * validation and grounding of the [clazz] package based on our
> > > requirements
> > > * consolidation of current meta solutions under a single unified
> > > object model
> > > * capable of supporting mixed loading strategies
> > > * capable of supporting model extension
> > >
> > > Before moving forward on this, I wanted to get feedback from people
> > > concerning the commit access exchange which doesn't follow the normal
> > > model. However - there is plenty of evidence in the clazz project
> > > itself that these guys know what they are doing. Secondly, this also
> > > means voting here and there, so first of all I'll like get an idea of
> > > who here are sufficiently interested in this to actually contribute.
> > >
> > > Cheers, Steve.
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
RE: [PROPOSAL] a clazz act
Posted by Leo Sutic <le...@inspireinfrastructure.com>.
> From: Leo Simons [mailto:leosimons@apache.org]
>
> However, I think there is little harm in giving these guys
> _cvs access_ to avalon-sandbox (as opposed to committer
> priviledges)
I don't want to separate committer *responsibilities* and
cvs access.
/LS
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Leo Simons <le...@apache.org>.
I agree with Paul.
However, I think there is little harm in giving these guys _cvs access_
to avalon-sandbox (as opposed to committer priviledges), so long as
everyone invloved understands that this is purely because of technical
convenience, and an avalon committer (ie Steve) takes responsibility
over the code these guys commit, just as if he had applied a patch. The
fact these guys are committers @ jakarta makes me trust them enough to
be sure they won't wreck cvs.
I don't think there's a reason to make a big deal out of this though; if
things don't work out I think there'd probably be little problem hosting
this development inside jakarta-commons-sandbox, or alternatively
working on a patch submission basis.
I'm not sure atm whether I'll have time or an itch to contribute;
haven't actually looked at clazz yet.
gotta fly,
- Leo
On Fri, 2002-11-29 at 08:03, Paul Hammant wrote:
> Stephen,
>
> I think we should be rigid with the six months of patched contributions
> before granting commit/vote rights to Avalon. It may appear rude to
> Stephen Colebourne and Dmitri Plotniko, but we have no idea who they
> are, their character etc. A normal get to know you period with a normal
> script leading up to invite is appropriate I think.
>
> - Paul
>
> >
> > I would like to move forward with a joint initative between the Avalon
> > Dev. Team and the Jakarta Commons [clazz] guys to work up an
> > implementation of an Avalon meta model for component and service
> > defintions. I've already spoken to Stephen Colebourne and Dmitri
> > Plotnikov (committers on the [clazz] project) and there both keen to
> > work with us on this. From their point of view is about validation of
> > the [clazz] package against real requirements, and from our point of
> > view is about validation of the [clazz] package and delivery of a
> > standard Avalon meta solution.
> >
> > To kick things off I have in mind the following:
> >
> > * granting commit access to Stephen and Dmitri so they can work on
> > a sandbox project
> > * reciprical commit rights for concerned avalon committers to the
> > [clazz] project
> > so we can work on clazz (doc enhancement, unit test, debugging, etc)
> >
> > Enabling co-development on a meta model implementation, leading to:
> >
> > * validation and grounding of the [clazz] package based on our
> > requirements
> > * consolidation of current meta solutions under a single unified
> > object model
> > * capable of supporting mixed loading strategies
> > * capable of supporting model extension
> >
> > Before moving forward on this, I wanted to get feedback from people
> > concerning the commit access exchange which doesn't follow the normal
> > model. However - there is plenty of evidence in the clazz project
> > itself that these guys know what they are doing. Secondly, this also
> > means voting here and there, so first of all I'll like get an idea of
> > who here are sufficiently interested in this to actually contribute.
> >
> > Cheers, Steve.
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Stephen McConnell <mc...@apache.org>.
Nicola Ken Barozzi wrote:
>
> Stephen McConnell wrote:
>
>>
>> Leo Sutic wrote:
>>
>>> What I'd like to know is if this will lead to a dependency
>>> on clazz for framework? I would be -1 on any such dependency,
>>> especially while clazz is still very much growing in scope and
>>> dependencies itself.
>>>
>>
>> My initial impression is that the dependencies we would be looking at
>> are basically at the core of the [clazz] package (not the reflection
>> or attributes stuff). This means basically the
>> org.apache.commons.clazz.* (i.e. none of the subpackages). Before
>> conconsidering the framework relationship I think there is some
>> validation/experimentation to be done - and yes, I agree that while
>> clazz is evolving and in sandbox - its not a candidate for the
>> framework. On the otherhand - I figure that working with the [clazz]
>> project now is more likely to ensure it meets are needs and that
>> [clazz] gets a shot in the arm though validation against real
>> requirements.
>
>
> Why can't we collaborate with them in Jakarta Commons Sandbox?
>
Yep - that works.
S
--
Stephen J. McConnell
OSM SARL
digital products for a global economy
mailto:mcconnell@osm.net
http://www.osm.net
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Nicola Ken Barozzi <ni...@apache.org>.
Stephen McConnell wrote:
>
> Leo Sutic wrote:
>
>> What I'd like to know is if this will lead to a dependency
>> on clazz for framework? I would be -1 on any such dependency,
>> especially while clazz is still very much growing in scope and
>> dependencies itself.
>>
>
> My initial impression is that the dependencies we would be looking at
> are basically at the core of the [clazz] package (not the reflection or
> attributes stuff). This means basically the org.apache.commons.clazz.*
> (i.e. none of the subpackages). Before conconsidering the framework
> relationship I think there is some validation/experimentation to be done
> - and yes, I agree that while clazz is evolving and in sandbox - its not
> a candidate for the framework. On the otherhand - I figure that working
> with the [clazz] project now is more likely to ensure it meets are needs
> and that [clazz] gets a shot in the arm though validation against real
> requirements.
Why can't we collaborate with them in Jakarta Commons Sandbox?
--
Nicola Ken Barozzi nicolaken@apache.org
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
---------------------------------------------------------------------
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Stephen McConnell <mc...@apache.org>.
Leo Sutic wrote:
>What I'd like to know is if this will lead to a dependency
>on clazz for framework? I would be -1 on any such dependency,
>especially while clazz is still very much growing in scope
>and dependencies itself.
>
My initial impression is that the dependencies we would be looking at
are basically at the core of the [clazz] package (not the reflection or
attributes stuff). This means basically the org.apache.commons.clazz.*
(i.e. none of the subpackages). Before conconsidering the framework
relationship I think there is some validation/experimentation to be done
- and yes, I agree that while clazz is evolving and in sandbox - its not
a candidate for the framework. On the otherhand - I figure that working
with the [clazz] project now is more likely to ensure it meets are needs
and that [clazz] gets a shot in the arm though validation against real
requirements.
Cheers, Steve.
>
>/LS
>
>
>
>--
>To unsubscribe, e-mail: <ma...@jakarta.apache.org>
>For additional commands, e-mail: <ma...@jakarta.apache.org>
>
>
>
>
>
--
Stephen J. McConnell
OSM SARL
digital products for a global economy
mailto:mcconnell@osm.net
http://www.osm.net
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
RE: [PROPOSAL] a clazz act
Posted by Paul Hammant <pa...@yahoo.com>.
Leo,
I am sure that 90% of us (me included) got in without the usual six-month of contributions. Look
at the number of people that actually commit versus _can_ commit. Sometimes it has worked in our
favour - Look at Nicola - he was only invited in to do xdocs++ work. Now he has stepped up to a
far more responsible role as PMC chair. In fact its a _greatly_ unthanked role for Nicola.
I am going to stick by my guns and say "no". Note, this is a vote, not a veto.
- Paul
--- Leo Sutic <le...@inspireinfrastructure.com> wrote: > > From: Paul Hammant
[mailto:Paul_Hammant@yahoo.com]
> >
> > Stephen,
> >
> > I think we should be rigid with the six months of patched
> > contributions
> > before granting commit/vote rights to Avalon.
>
> We haven't been rigid before (me & the guy writing the C#
> version of framework, for example). I don't think this is
> reason enough.
>
> What I'd like to know is if this will lead to a dependency
> on clazz for framework? I would be -1 on any such dependency,
> especially while clazz is still very much growing in scope
> and dependencies itself.
>
__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
RE: [PROPOSAL] a clazz act
Posted by Leo Sutic <le...@inspireinfrastructure.com>.
> From: Paul Hammant [mailto:Paul_Hammant@yahoo.com]
>
> Stephen,
>
> I think we should be rigid with the six months of patched
> contributions
> before granting commit/vote rights to Avalon.
We haven't been rigid before (me & the guy writing the C#
version of framework, for example). I don't think this is
reason enough.
What I'd like to know is if this will lead to a dependency
on clazz for framework? I would be -1 on any such dependency,
especially while clazz is still very much growing in scope
and dependencies itself.
/LS
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Paul Hammant <Pa...@yahoo.com>.
Stephen,
I think we should be rigid with the six months of patched contributions
before granting commit/vote rights to Avalon. It may appear rude to
Stephen Colebourne and Dmitri Plotniko, but we have no idea who they
are, their character etc. A normal get to know you period with a normal
script leading up to invite is appropriate I think.
- Paul
>
> I would like to move forward with a joint initative between the Avalon
> Dev. Team and the Jakarta Commons [clazz] guys to work up an
> implementation of an Avalon meta model for component and service
> defintions. I've already spoken to Stephen Colebourne and Dmitri
> Plotnikov (committers on the [clazz] project) and there both keen to
> work with us on this. From their point of view is about validation of
> the [clazz] package against real requirements, and from our point of
> view is about validation of the [clazz] package and delivery of a
> standard Avalon meta solution.
>
> To kick things off I have in mind the following:
>
> * granting commit access to Stephen and Dmitri so they can work on
> a sandbox project
> * reciprical commit rights for concerned avalon committers to the
> [clazz] project
> so we can work on clazz (doc enhancement, unit test, debugging, etc)
>
> Enabling co-development on a meta model implementation, leading to:
>
> * validation and grounding of the [clazz] package based on our
> requirements
> * consolidation of current meta solutions under a single unified
> object model
> * capable of supporting mixed loading strategies
> * capable of supporting model extension
>
> Before moving forward on this, I wanted to get feedback from people
> concerning the commit access exchange which doesn't follow the normal
> model. However - there is plenty of evidence in the clazz project
> itself that these guys know what they are doing. Secondly, this also
> means voting here and there, so first of all I'll like get an idea of
> who here are sufficiently interested in this to actually contribute.
>
> Cheers, Steve.
>
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
RE: [PROPOSAL] a clazz act
Posted by Berin Loritsch <bl...@citi-us.com>.
> From: Stephen McConnell [mailto:mcconnell@apache.org]
>
> That seems to leave me with three strategies:
>
> (A) Introduce a third Avalon meta model implementation based
> on [clazz] that while out of scope in terms of a framework
> solution in the short term - would aim to meet the original
> objectives of the proposal.
>
> (B) Evolving the existing meta package to follow the same
> object model as [clazz] - i.e. addition of a meta-meta
> layer equivalent to the core subset of [clazz] dealing
> with class, property and operation definitions. This
> would enable the conservative approach of potentially
> incorporating [clazz] classes at a later stage when
> and if [clazz] is out of commons sandbox.
>
> (C) Continue with what we have.
>
> Any thoughts on the three options above or other
> options/ideas I have not thought about?
Steve, let's look at work for the new Container. There is allot we
can do without binding ourselves too early in the process. We can
keep this in the back of our minds while we attack many of the other
parts of the new container. For one thing, are we starting with
Avalon 5 or not? If so there is allot to discuss as a community before
we can even think of *implementations*.
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Stephen McConnell <mc...@apache.org>.
Based on the feedback on this so far there are a couple of things I think I can summarise:
* doing anything sophisticated re commit privs isn't going
to fly so lets drop that idea
* there are issues concerning the sandbox status of [clazz]
that impact solutions we develop with the objective of
establishing a framework level solution
I've been thinking about the question of where this should happen - here or over on commons. My conclusion is that that it should happen here because this is something that concerns this community - after all, we are talking about a common meta model solution for avalon. On the otherhand I don't see much benefit in kicking something new off without at least of couple of people from Avalon jumping in as well.
That seems to leave me with three strategies:
(A) Introduce a third Avalon meta model implementation based
on [clazz] that while out of scope in terms of a framework
solution in the short term - would aim to meet the original
objectives of the proposal.
(B) Evolving the existing meta package to follow the same
object model as [clazz] - i.e. addition of a meta-meta
layer equivalent to the core subset of [clazz] dealing
with class, property and operation defintions. This
would enable the conservative approach of potentially
incorporating [clazz] classes at a later stage when
and if [clazz] is out of commons sandbox.
(C) Continue with what we have.
Any thoughts on the three options above or other options/ideas I have not thought about?
Cheers, Steve.
--
Stephen J. McConnell
OSM SARL
digital products for a global economy
mailto:mcconnell@osm.net
http://www.osm.net
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Berin Loritsch <bl...@apache.org>.
Peter Donald wrote:
>I would prefer that only people who have demonstrated a sustained interest in
>developing Avalon and working with existing committers were to be offered the
>privlidge.
>
>commons sandbox is open to everyone here - only need to ask to get access so I
>would recomend that further development happens over there.
>
>
>
+1
---------------------------------------------
Introducing NetZero Long Distance
1st month Free!
Sign up today at: www.netzerolongdistance.com
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Peter Donald <pe...@realityforge.org>.
I would prefer that only people who have demonstrated a sustained interest in
developing Avalon and working with existing committers were to be offered the
privlidge.
commons sandbox is open to everyone here - only need to ask to get access so I
would recomend that further development happens over there.
--
Cheers,
Peter Donald
*------------------------------------------------*
| The student who is never required to do what |
| he cannot do never does what he can do. |
| - John Stuart Mill |
*------------------------------------------------*
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
Re: [PROPOSAL] a clazz act
Posted by Stephen McConnell <mc...@apache.org>.
Leo Sutic wrote:
>Stephen,
>
>what does clazz provide that we want?
>
Primarily a meta-meta model.
That means is that someone writting a tool for [clazz] based meta model
does not need to be concerned with anything specific to Avalon. They
will write tools in terms of clazz properties, clazz operatioins,
attributes etc - and never need to have any specific knowlege about "our
domain". For example, tools such as descriptor and document generators,
navigators, conversion tools, ... that are written for clazz will work
with our model (bacause the model will use the same meta-meta model).
Secondly, the package provides a nice structure for integration of meta
information at a clazzloader level. The object model provides support
for locating a clazz, instance creation from the clazz and so on. With
this seperation between Java class and the [clazz] notion of a class, my
impression is that we will be able to achieve a lot of behind the scenes
simplification of "component and service management".
Cheers, Steve.
>
>
>/LS
>
>
>--
>To unsubscribe, e-mail: <ma...@jakarta.apache.org>
>For additional commands, e-mail: <ma...@jakarta.apache.org>
>
>
>
>
>
--
Stephen J. McConnell
OSM SARL
digital products for a global economy
mailto:mcconnell@osm.net
http://www.osm.net
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>
RE: [PROPOSAL] a clazz act
Posted by Leo Sutic <le...@inspireinfrastructure.com>.
Stephen,
what does clazz provide that we want?
/LS
--
To unsubscribe, e-mail: <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>