Thanks Daryl...
I'll try to keep this short [note: I failed, but please read on] -- I
mostly have a bunch of questions, but just for everyone's info, here's a
summary of what's happened so far. Note that it doesn't include any
sort of plan for what's next.
- Several months ago we did some prototyping work in the Struts
sandbox in a project called Struts Ti. In it, we basically converted
our core Page Flow code to run on top of XWork (the action-dispatch
layer of WebWork) instead of Struts 1.x. I consider this to be *proof
of concept* only.
- The WebWork/Struts merger was announced. This produced clarity
(no pun intended :) ) around the idea of a request/action-based Struts
2.x line. The WebWork code is starting out in the Struts Ti sandbox
under a "Phase 1" directory.
- Our Page Flow port moved into a "Phase 2" directory in Struts Ti.
This doesn't mean that any of that code will actually make it into
Struts Action -- it's conceptually the kind of thing that would go into
the next phase of the project, but again, it's POC only.
We need to decide what our community thinks is the right thing to do,
and then I think we all need to be involved in the discussions of Struts
Action if that's where we think we're headed. I'll try to kick off as
much of the NetUI-relevant discussions as possible, but first there are
a few things to clarify over here in Beehive land.
Scott and Antony both responded positively (thanks for the feedback,
guys) on the original thread. Steve said he'd like to see this be an
all-or-nothing move: all of NetUI (including the tags) would go to
Struts. Beyond that, no one replied -- how do people feel about trying
to merge our technology into Struts?
If that's an easy question, read on. :)
My personal feeling on this is that if we merge NetUI with Struts in any
way, then we need to provide a good upgrade/compat path for our current
users. It's a given. Beyond that, though, I think it's likely that
there would be major changes, and also that not all of our *programming
model* would make it into the core of Struts Action. What we can offer
in ease-of-use is huge... but it's a radical enough leap from raw
Struts/WebWork that I'm guessing it will take much more
discussion/debate. This is just my sense for things -- I'm not speaking
with any knowledge of how this will all pan out. If it's true that
NetUI doesn't simply get dropped into Struts Action whole-hog...
- Do we bail out?
- Do we push for a Struts subproject that would add whatever doesn't
make it into the core?
- Do we keep Beehive's NetUI intact, but built on the new Struts Action?
I'd be for the middle option, but I'm interested in hearing everyone's
thoughts on it. ** It's important that we come to consensus on this
particular question, in order for us to move forward. **
Last question: if we do move towards Struts, do we hold off on adding
major new functionality in NetUI on the Beehive side? I think so, but
again, I want to hear what everyone thinks.
Finally, for the record I think that once we get on the same page
philosophically, we should be getting involved in the Struts community.
I will try to drive as much of that as possible, but I think that the
Struts lists (and the wiki) are the right place for us to have the
technical discussions (if we do decide we're in favor of merging NetUI
with Struts).
Thanks,
Rich
p.s. Please excuse my less-than-regular emails in the coming weeks. I
just started a new job and I'm trying to maintain my ability to sleep. :)
Daryl Olander wrote:
Now that Rich has been made a committer on Struts (Congrats Rich), I think
it's time that we start to discuss how NetUI (Page Flows) and Struts Action
2.0 are related. Rich can you start off the discussion with a summary of
the direction the page flow code is going (phase 2) and what you think the
direction between the two projects is?
Thanks
