> -----Original Message-----
> From: Jonathan Hawkes [mailto:[EMAIL PROTECTED]
> Sent: Friday, March 05, 2004 12:17 PM
> To: Avalon Developers List
> Subject: Re: [avalon] State of Consensus
>
> > I don't have a position about Phoenix, but we don't have a replace for
> > Fortress. We don't have a lightweight container - Fortress is not
> > lightweight, but is the best we have now.
> >
> > Until Merlin is a decoupled container allowing differents uses of it, we
> > should not retire anything.
>
> I hesitantly agree. I want a single container implementation, but at the
> moment I would not touch Merlin with a 10' pole!
If we don't touch it, we can't improve it. :)
I think Niclas has a point:
> Just about everyone agrees Merlin is "bad code", so now let's get on with
> fixing it together, instead of picking on Stephen's effort. He is open to
> change, improvements and ways to do it better. But it doesn't work by
> saying "Merlin sucks, let's scrap it and start over (again)."
Also, note Stephen's proposal:
1. We move Fortress into sustainable retirement
2. We move Phoenix into terminal retirement
Fortress doesn't have to immediately disappear. No one is going to delete
it from CVS. But further effort could be focused on making Merlin more
ECM/Fortress compatible rather than working on Fortress itself. We already
have a nice Fortress 1.1 release that can support current users.
When this discussion starts we generally have the following arguments:
Dev #1: We really should unify efforts on our containers -- make
one solid extensible code base. It will rock!
Dev #2: Let's use Merlin. Let's drop Fortress and Phoenix and everyone
work on Merlin.
Dev #1: We can't drop Fortress until Merlin has support for ECM/Fortress
components
Dev #2: So why don't you help? Let's all work on Merlin and make it
cool.
Dev #1: Merlin's ugly. Let's start from scratch.
This continues on until someone gets upset and leaves.
The problem with ECM/Fortress support in Merlin is that it's a catch 22. No
one wants to say "we're going to drop ECM/Fortress in favor of Merlin" until
Merlin has ECM/Fortress support somehow. And no one wants to put time into
Merlin because it's complex and we already have Fortress. I'm guilty of
this one as much as anyone else. After some thought I'm finally feeling
that I need to dig into Merlin and improve it.
Okay, so perhaps we don't "officially" drop ECM/Fortress until Merlin has
proper support. But that doesn't mean we can't plan for it and work towards
it. That doesn't mean we can't tell our users, "Hey, Merlin is going to get
better. Fortress and Phoenix are going away." That doesn't mean we can't
have a goal and a roadmap which states that what is now "Merlin" will one
day be the "Avalon" container.
J. Aaron Farr
SONY ELECTRONICS
DDP-CIM
(724) 696-7653
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
