> From: Stephen McConnell [mailto:[EMAIL PROTECTED]] 
> 
> Berin Loritsch wrote:
> 
> >Before we go too far down the holy grail crusader mentalities, lets 
> >keep in mind that neither ContainerKit nor Merlin's Meta 
> Model (which 
> >is what it is) has been voted on or accepted as the standard.  
> >Therefore NEITHER deserve the moniker Avalon Meta Model.  
> That cannot 
> >be given until we have decided on a standard.
> >
> ><rant>
> >The problem with excalibur as it stands is that there are too many 
> >forks of functionally equivalent things.  We have a fork of 
> >ContainerKit, a fork of Merlin, and a fork of the proposed 
> extension mechanism.
> >  
> >
> 
> I don't think this is correct.
> 
> * there is a fork of a part of containerkit dealing with meta-info 
> management and container independent meta management resources

But a fork nonetheless

> * there is no fork of Merlin that I am aware of

A fork is something that started in one repository and for some
reason is moved to another while the original code is still maintained
separately.  Therefore, in the purest sense Merlin 2 is a fork of
Merlin 1.

> * there isn't a fork of the proposed extension mechamism

Your version was a fork of the one in Fortress.  You developed it
independently without discussing with us why we made the decisions
we did.

> There are some actions to seperate the container indepednent 
> resources 
> in the Merlin project concerning the proposed lifecycle 
> management and 
> the meta info management into seperate packages that are 
> independent of 
> the Merlin package.This is not forking - this is just putting 
> thing in 
> place so that they are more easily re-used by other projects.
> 
> ></rant>
> >
> >Now, as the table that was in another thread showed, the ContainerKit
> >Meta Model and the Merlin 2 Meta Model are largely equivelant.
> >
> 
> Sorry for being padantic - but there are massive differences. 
>  The minor 
> differences your referring to concern the meta-info model (the type 
> level stuff).  Beyond that the containerkit resources proved 
> unworkable 
> in a dynamic environment.  Due to a general refusal by the author of 
> containerkit to address needs outside of his own area a fork 
> of a part 
> of containerkit was made - in order to leverage the usable 
> content.  The 
> Merlin Meta Model is something else - it includes the Merlin 
> container 
> specific content - it is part of the Merlin project.  It 
> leverages the 
> container neutral meta-info layer.

Then we need to address exactly how much of a system *should*
be made dynamic.  And forgive me for being pedantic, but shouldn't
we have just one meta model?  Merlin==Fortress==Phoenix?

I dislike too much stuff happening behind the scenes so that I
have no clue what is going on.


> >  The major distinction is that Merlin has support for Stage and 
> >Extension meta info.
> >I need some clarification on the Stage meta info--if it is 
> referring to
> >Stephen's first cut at forking the extension mechanism then 
> it needs to
> >die.  If not, then I need to know what it is talking about.
> >
> This is all documented.  Please refer to the following page and the 
> respective javadoc on the meta-info types under the 
> org.apache.excalibur.meta.info package.
> http://home.osm.net/doc/merlin/extensions.html

I'll look at the document, but can you just say 'Yes' or 'No' and
then provide the link for further clarification?  It'll make things
quicker.


> >Either way, it is important to note that *Extensions* are just that--
> >*Extensions*.  Therefore containers are not required to 
> implement them.
> >However, if they are implemented, they should be implemented in a
> >consistent
> >manner.  I would be happy with a meta model that validates that the
> >extension can be satisfied, and if the container does not support the
> >extension/stages then it needs to fail-fast and tell the 
> developer that
> >they cannot use that component in the container.  The 
> documentation for
> >extensions must be resilliantly clear that you cannot expect 
> components
> >that take advantage of extensions to work in absolutely 
> every container.
> >
> >Generally speaking, if ContainerKit does not provide 
> validation, I would
> >like to see it do so.  I would *really* like to see a 
> merging of Merlin/
> >ContainerKit meta info.  If there are any blockers to that happening
> >that are technical in nature (technical, not political) I 
> would like to 
> >hear it.  
> >
> 
> Technical:
> 
>   ContainerKit meta-data model cannot be used in a context where you
>   dynamically create associations between type profiles.  This is
>   brought about by the problem that the entire conterkit structures
>   are immutable.  This prevents dynamic construction and the potential
>   for dynamic replacement.

Would it be able to regenerate the meta-data model when we reload
configurations?

We can make ContainerKit allow dynamic reassignment.


>   ContainerKit attempts to define too much about how a 
> container should
>   be implemeted, resulting in a very large redundancy of classes.  The
>   only realistic approach to building a common container 
> utilities group
>   is the progressive defintion of areas of common concern, progressive
>   collaboration between parties with different solution, and from that
>   the establishment of sharable utilities.

Could you give me an example where the ContainerKit model artificially
constrains you?


> Operational:
> 
>   You need collaboration for something to be common - it 
> needs to build
>   based on an open attribute to community ideas.

True.  However, if the original author believes that something should be
immutable, it is worth our time to find out why.

> >Another thing that would be emensely helpful is a suite of test
> >cases for both ContainerKit and Merlin that prove their compatibility
> >and their claims.  Merlin has absolutely no testcases which makes me
> >nervous.  I haven't looked at the ContainerKit code, but if 
> it has the
> >same problem it needs to be rectified.
> >
> 
> Merlin does not include formal test-cases at this time.  It 
> does include
> several demonstrations and is constantly validated against a 
> large number
> of components here on my machine.  Please feel free to jump 
> in with test
> cases into the Merlin project - all contributions will be highly
> appreciated. In the meantime, my priorities are to make sure 
> that Merlin
> is complete and well documeted before launching into the test-case
> preparation.

I've got other things to do.  Formal testcases are very important,
and part of the proper validation of components and other pieces.
One of the things I have to do is add more formal testcases for
Fortress.



--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

Reply via email to