Leo Sutic wrote:
On Mon, 12 Jul 2004 15:02:19 -0700, Niclas Hedhman <[EMAIL PROTECTED]> wrote:
Oops, I was actually referring to LSD, but I think you are part of the "baby steppers" :o) as well. I want to be a "baby stepper" when it is practical, and in this case it seems like the most logical thing.
Baby steps is probably the way to go, and a fork would allow you to take those baby steps unrestricted.
If you stay with 4.x, then each baby step has to be considered in the context of A4, Excalibur, Phoenix, etc. etc. etc.
Sure.
For example, is the inclusion of Avalon Meta in the specs a baby step or not? I'd say it is a baby step from a Merlin perspective. But not from a Phoenix or Fortress perspective.
I have in my mind a distinction between "Avalon" and "Avalon Framework", and it could be that this is the problem at hand. Avalon Framework will never include Meta, not even under a 'free-for-all-A5-fork'. Hence the Framework will have small refinements in the AF4.x path. And it will become AF5 with the deprecated stuff removed, but remain largely compatible.
<snip/>
I have carefully read your suggestions, and in principle we are in agreement. The only bit of difference is;
1. In my world AF4 exists. "Avalon 4" doesn't exist as a specification, but as an umbrella for Framework, ECM, Phoenix, Fortress and Merlin. So, is it not possible to agree that "Avalon 4" is now split, and we can still agree on the term "Avalon Framework 4" which will remain compatible as much as possible for component authors, and reasonably easy for container authors to evolve with.
2. Timing. I am not comfortable of making a release of Avalon 5, Avalon Planet, without being able to deliver. We need ramp-up period, and it will take a fair amount of time.
Cheers Niclas
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
