ws-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Eran Chinthaka Withana <eran.chinth...@gmail.com>
Subject Re: What's next for WS@Apache? Discuss.
Date Wed, 21 Apr 2010 00:49:02 GMT
Hi,

I'm still confused about this re-org. People started this re-org, saying
that the projects inside old-WS project was not manageable and no knows
whats going inside all the projects. It was also mentioned that, since
project X people also get to have a say in project Y (where X,Y are in WS),
it will be a mess.
Now why is this theory not applying to Axiom or XmlSchema projects. They are
totally independent projects, that can be used beyond Axis2. Also some one
concentrating only on XmlSchema or Axiom might not be interested in Axis2
project or its Karma. So why do you wanna put them inside Axis2 or new WS
project, once again making the same mistake we had earlier?

Axis2 depends on Axiom and XmlSchema, but not the other way round.

Dan, if new WS project should not be an umbrella project, then why do you
think we should put Axiom/XmlSchema inside it?

Thanks,
Eran Chinthaka


On Tue, Apr 20, 2010 at 8:34 PM, Daniel Kulp <dkulp@apache.org> wrote:

> On Tuesday 20 April 2010 7:54:27 pm Glen Daniels wrote:
> > Hi Sanjiva,
> >
> > Sorry for the late reply.
> >
> > On 4/17/2010 3:37 AM, Sanjiva Weerawarana wrote:
> > > Glen, your proposal amounts to having gone thru a lot of pain to have
> > > simply TLPed Axis. The entire board prerogative was to break up WS
> > > wasn't it?
> >
> > My interpretation was that it was more "WS is too big" than "WS should
> > entirely go away".  There are plenty of other projects at Apache that
> have
> > a reasonable number of subprojects, with related functions and
> development
> > communities.  I'm not yet convinced that there is sufficient community
> > around some of the individual subprojects to bootstrap a happy PMC, and
> > there also may be good reason to keep some of these things together.
> >
> > I think reducing WS from >17 subprojects in a two-level container down to
> a
> > single-level container for 6 common Web Service related components would
> be
> > a pretty good accomplishment... and nothing prevents that from being an
> > intermediate step towards spinning off further TLPs down the road.
> >
> > I'm not saying I'm -1 to more TLPs, by the way, just exploring my
> > off-the-cuff impressions.
>
> Just want to say I completely agree with everything Glen said.  Well put.
> It's also my understanding that getting WS down to a smaller, more
> concentrated project targeting shared WS technologies would be acceptable
> and
> really would no longer be an "umbrella" project.
>
> Dan
>
>
>
> > --Glen
> >
> > > I'd prefer to make a bunch of new TLPs - give people a chance to grow
> > > into new roles as well. All the ones you listed as "keep" should be
> TLPs
> > > IMO. There is no "minimum size" required to be a TLP.
> > >
> > > Sanjiva.
> > >
> > > On Thu, Apr 15, 2010 at 8:38 AM, Glen Daniels <glen@thoughtcraft.com
> > >
> > > <mailto:glen@thoughtcraft.com>> wrote:
> > >     Hey y'all,
> > >
> > >     So as per my earlier mail to the dev lists, we need to talk
> about...
> > >
> > >     * Which subprojects should be promoted to TLPs?
> > >     * Which subprojects should be migrated to the Attic?
> > >     * What should the structure look like for what remains?
> > >
> > >     Here's what we've got to work with.
> > >
> > >     1. Axiom (commons)
> > >     2. Neethi (commons)
> > >     3. XmlSchema (commons)
> > >     4. Tcpmon (commons)
> > >     5. Guththila (commons)
> > >     6. JaxMe
> > >     7. jUDDI
> > >     8. Scout
> > >     9. Muse
> > >     10.Woden
> > >     11.WSIF
> > >     12.WSS4J
> > >     13.XMLRPC
> > >
> > >     Let's talk Attic first.  It seems WSIF and Muse have been pretty
> much
> > >     inactive for some time now, so I'd propose we get the ball rolling
> > >     with proposals to Attic both of those.  JaxMe also seems ripe for
> > >     the Attic.
> > >
> > >     I think that it's pretty clear jUDDI and Scout should migrate
> > >     together to a
> > >     new TLP, Apache jUDDI, with Kurt as chair.  Thoughts?
> > >
> > >     Does anyone outside of Axis2/C use Guththila?  If not, I'd suggest
> > >     that migrate to Axis.
> > >
> > >     Is there enough activity on XMLRPC to keep it alive?  Jochen?
> > >
> > >     Personally, I'd probably prefer to leave Axiom, Neethi, XmlSchema,
> > >     and Tcpmon
> > >     as subprojects of WS.  I'd like to get rid of the "commons" layer,
> > >     though,
> > >     since I think the "new" WS project should itself be a set of
> > >     commonly useful
> > >     WS components.  I know other ideas have been discussed for the
> > >     commons stuff,
> > >     so let's start that discussion and get the various opinions out on
> > >     the table?
> > >
> > >     That leaves Woden and WSS4J.  Maybe Woden should stay in WS?  As
> for
> > >     WSS4J,
> > >     I'm not sure if it should stay or become a TLP.
> > >
> > >     So... what do you guys think?
> > >
> > >     Thanks,
> > >     --Glen
> > >
> > >     P.S.  Reminder - please fill out your sections of our board report
> > >     for this
> > >     month if appropriate!
>
> --
> Daniel Kulp
> dkulp@apache.org
> http://dankulp.com/blog
>

Mime
View raw message