OK - given the qualification of intent based on the posts to this thread it seems to me that the PMC is being presented with a choice:
1. Move Fortress/ECM to Codehaus 2. Keep it here at Apache
So - looking at the alternatives in front of us.
Move Fortress/ECM to Codehaus -----------------------------
benefits - user's of ECM/Fortress style components get
potential viable option for continuation of
development; Avalon dev team can get back to
our core interests disadvantages - probability of delivery of a binary
compatibility adapter fast approaches zero but
this can be offset by providing equivalent
functionality; user priority in an Apache
maintained solution is out the window risks - basically this comes down to predicting if
an initiative will retain interest or if it
fades away - if it fades away the user's
are caught between a rock and a hard placeKeep it here at Apache ----------------------
benefits - binary migration will happen, existing
users of ECM/Fortress semantics will be assured
of a codebase that will be maintained under the
Apache umbrella disadvantages - development will be discontinued outside of
of initiative related to migration; a fork will
probably occur; stuff will continue to surface
that distracts the avalon dev community; support
overhead for ECM/Fortress content may be an
issue relative to the community interests risks - a fork happens, conflicts arise, community gets
draw into politics as opposed to technicsSummary -------
Just do it.
Potential benefits to users interested in migrating can be enabled by expanding merlin feature set (e.g. finder facility, etc.). Potential risk of on-going noise IMO overrides the risk to users of being stranded - simply because we can deliver equivalent functionality (its already real close). But one proviso - no formal endorsement - go ahead and do it - no need for permission - but in the meantime Avalon maintains a release here independently of your initiative. If your initiative is good - then fine and dandy - if not, Avalon provides the fall back solution.
I.e. don't ask the Avalon PMC to endorse good intentions. Instead - demonstrate to the user community and the Avalon PMC your ability to provide appropriate maintenance and support. In the meantime the Avalon dev community can move forward with a programme that addressed ECM/Fortress functional requirements for the user community that want to leverage an Apache solution.
Cheers, Stephen.
Leo Simons wrote:
Dear Avalon PMC,
we all know avalon has seen some rough times lately. In an attempt to end this, the new direction for avalon is towards a single, unified platform that will be based on the new container, avalon-merlin. It is well-understood that the Avalon PMC and a substantial part of the avalon community considers this the best choice for avalon, and we wish to respect that choice.
However, concerns have risen over the legacy of the older avalon products, in particular avalon-fortress, avalon-ecm, and their respective dependencies. We wish to ensure the continued vitality, stability and availability of these products. There are many commercial and non-commercial efforts based around these products, and the new unified platform offers uncertainties and risk we are not prepared to accept.
There has been discussion about, and invitations, to make that work within the scope of the avalon project itself, perhaps by partitioning the project into distinct subprojects, splitting avalon up into multiple new projects, or taking similar measures.
After many weeks of deliberation, we can only come to the conclusion that this is a way forward that is bound to fail. As has been shown over the last two years, avalon is not a project where such a setup can ever work productively. We feat that by going down this path, avalon as a community will suffer. This will negatively impact apache and all of avalon's users. We believe there's a better alternative.
We would like to propose that apache transfers maintenance of avalon-fortress and avalon-ecm over to us. We will endeavor to create a stable, solid, backwards- and forwards-compatible open source project where these containers and their community can prosper under the terms of the Apache License in peaceful co-existence with the avalon project.
We will briefly explain what we mean with "maintenance transfer": our suggestion is that the avalon project ceases development of these codebases, and transfers them to us, intact. The transfer should include branding (the "fortress" and "ecm" names) and package structure (the "org.apache.avalon.fortress" package and parts of the "org.apache.excalibur" and "org.apache.avalon.excalibur" packages). We believe such an integral transfer is the only way to ensure a continued smooth user experience.
Without getting into further details as to what such a transfer would entail right away or how it should be executed, we would like to give you and the rest of the avalon community the opportunity to consider the pros and cons of this move, for apache, for avalon, and for our users, and then let us know if you're willing to co-operate with us on this issue to the ultimate benefit of all parties involved.
sincerely,
Sasvata Chatterjee Peter Donald Berin Loritsch Leif Mortenson Michael Nash Peter Royal Leo Simons Mauro Talevi Carsten Ziegeler
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
--
|---------------------------------------| | Magic by Merlin | | Production by Avalon | | | | http://avalon.apache.org | |---------------------------------------|
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
