RE: A Merlin Top Level Project

Thanks for the very interesting history lesson.

I guess what I'm really looking for is what the responsibilities of Avalon
are vs. Merlin. Where are the boundries? What can Avalon do without Merlin,
and vice-versa.

I am assuming that Merlin, with the standalone approach, will be able to run
on top of other frameworks other than Avalon? If not, I don't see the point
is spinning it out.




> -----Original Message-----
> From: Niclas Hedhman [mailto:[email protected]]
> Sent: July 22, 2004 14:49
> To: Avalon framework users
> Subject: Re: A Merlin Top Level Project
>
>
> On Thursday 22 July 2004 13:14, David Leangen wrote:
>
> > Until now, I had thought that Avalon and Merlin necessarily worked
> > together. Can one actually use Avalon without a component container like
> > Merlin?
>
> Well, that depends on what you mean by "use". There are stuff in Avalon
> Framework that can be used completely without any container.
> But more importantly, there are other containers than Merlin.
> Basically, Avalon started out with Excalibur Component Manager
> (ECM) as the
> first container, and Cocoon being the first and most important
> "user". Later
> Phoenix was built in parallel to deal with Avalon in a
> 'stand-alone' fashion.
> We could call these the first generation containers.
> Fortress was initially developed as an interim solution for an
> easy migration
> path for ECM users.
> Merlin took some ideas from Phoenix, but also had the ambition to
> integrate
> all the approaches into a single container.
> The many people working on different containers have had a big problem to
> agree on exactly how such convergence of the 4 containers should
> look like.
>
> Five main things happened over a period of 1.5 - 2 years;
>
> * ECM was deprecated
>
> * Phoenix was deprecated, but forked into Loom and the developers went to
> codehaus.org
>
> * A 'Single Unified Platform' vote was cast and passed earlier this year.
> Merlin was considered the reference platform, and work started on
> filling the
> contract holes that existed.
>
> * Friction still remained, and both a "Excalibur TLP"
> (Fortress-based) and a
> Merlin TLP was proposed, but the higher ranking ASF folks thought
> that was
> too much, and said an Excalibur TLP was ok (and formed) but
> Merlin-proponents
> should continue with the 'Single Unified Platform' (SUP) that the
> community
> had decided on.
>
> * The recent site layout change exposed many subtle complimentary
> changes of
> Avalon in line (we thought) with the SUP, but the Excalibur
> people felt that
> this was completely unacceptable, and a massive debate rised at
> all kinds of
> levels, technical, personal, 'legal' and other. I.e. The active Avalon
> committers (esp. myself & Stephen) was not considered good
> guardians of the
> Avalon Legacy and have failed to co-operate with Excalibur folks.
> In my book,
> the old in-fighting just remained, but over two projects instead
> of within
> one.
>
> I have tried to keep this rendition of history as objective as
> possible, but
> please note that I have been part of this process, and therefor
> biased, some
> way or the other.
>
> > In any case, my opinion is that it's important to get the
> infighting over
> > with as quickly as possible.
>
> I think everyone are in agreement with this, and have been so for
> the last
> couple of years, but obviously unable to stop it.
>
> > A few healthy academic discussions are
> > necessary from time to time, but it sounds like things have degenerated
> > beyond that.
>
> Yes it has.
> I think Merlin is a great platform with enormous potential if nutured
> correctly. We just want to get on doing that, with the feedback
> of the users.
>
> Cheers
> Niclas
> --
>    +------//-------------------+
>   / http://www.bali.ac        /
>  / http://niclas.hedhman.org /
> +------//-------------------+
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]