ode-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sathwik B P <sathwik...@gmail.com>
Subject Re: ODE, ODE 2.0, the Attic and the Incubator
Date Fri, 01 Feb 2019 09:54:26 GMT
Hi Guys,
Thanks for coming back.

Keeping aside Aron's great efforts on the proposal and undergoing POC. I
have few questions, I have some options under each question.

How do we plan to build a dev/user community around this Next Gen ODE
following the proposal after the POC stage?
a) Take it to incubator under Apache ODE brand name with it's own source
repo.
b) Direct import into a new source repo, without going to the incubator.
(probably with a single commiter)

What do we do with the ODE 1.x source repositories [ode-console.git,
ode-jacob.git, ode.git]?
a) Make them read only, no maintenance or releases hereafter.

What happens to the ode.apache.org website?
a) Add a header indicating the visitors that Next Gen ODE is under way and
existing website contents/downloads are no logner supported?

regards,
sathwik

On Fri, Feb 1, 2019 at 9:22 AM Aaron Anderson <aanderson@apache.org> wrote:

>  Thanks everyone for your positive feedback! Here are the milestone tasks
> I have been working towards to have a kernel of functionality to extend
> from:
> 1. Create a Maven build project that supports modular extensions -  done
> 2. Create a server process that starts and stops an embedded Ignite server
> and OpenWebBeans context - done
> 3. CLI interaction to run Ignite tasks - done
>
> 4. Configuration service including installing and uninstalling modules -
> currently working on this
> 5. Process life-cycle - building Ignite tasks and caches to manage process
> instances TBD
>
> 6.  Compiler - translate source into sequences of instructions
> (executable) - I have some code from a previous effort I intend to re-use
> TBD
>
> 7. Interpreter - load a executable and step through instructions, invoke
> instruction instance providing executable and Ignite context TBD
> 8. Add individual Tomcat and CXF modules for receiving and sending REST
> requests - TBD
>
> After this bare minimum functionality is available I would like to
> accomplish the following:
>
> 1. Create a partial SCXML extension
> 2. Create a partial extensible BPMN 2.0 extension
> Creating two different extensions should highlight shared service
> abstraction and extensiblity deficiencies.
>
> To complete the prototype I would like to add assembly/composite
> deployment functionality for managing and dynamically deploying source and
> binary artifacts to the platform such as process definitions, static web
> content, Java libraries, etc. This would involve local file caching and
> classloader management with a limited child CDI context. Given the
> complexity of this work I plan on working on this towards the end of the
> prototype phase.
> Based on my current progress and research of Ignite and OpenWebbeans I
> have a good idea of how to do the remaining work. Two aspects I still need
> to do more research on is what format the the executable and processes
> should be stored in within Ignite.
>
> For the executable format I am leaning towards a namespace aware XML
> format, perhaps compressed. Mapping XML QNames to Java annotations using
> CDI is straight forward and would also make it easy to extend. For process
> instances I am looking into using the native Ignite binary format but I
> need to perform some more testing to see how layering extended data into it
> would work.
> Please let me know if anyone has any questions or would like further
> clarification on the tasks outlined above. If it would be helpful I can
> check in all my latest code into my github repo or even create a branch in
> the new git repository so everyone can see what I have done so far.
>
> Regards,
> Aaron
>
>
>     On Wednesday, January 30, 2019, 11:20:34 PM CST, Antoine Toulme <
> antoine@toulme.name> wrote:
>
>  Hey Aaron,
>
> I have been a committer with little to no activity for a long time. That
> said, I think ODE is a unique project and has great potential - some of it
> unachieved.
>
> You had a long laundry list of items you’d like to see happen.
> Can you sort them in order of priority and decompose them enough others
> can pick them up with you?
>
> Just like Paul - costs me nothing to keep it going, so +1 here too.
>
> Cheers,
>
> Antoine
>
> > On Jan 30, 2019, at 9:04 PM, Paul Brown <paulrbrown@gmail.com> wrote:
> >
> > Aaron - it has been a long time. (In fact, ODE grew out of work at my
> > company back in 2002-2003... if memory serves.)
> >
> > That said, it costs me nothing to let things move forward and offer what
> > advice I can, but I expect to have a meager amount of time.
> >
> > So +1 to some new life if that’s the way you want to go!
> >
> > On Wed, Jan 30, 2019 at 19:14 Aaron Anderson <aanderson@apache.org>
> wrote:
> >
> >>
> >> Hi Rich,
> >>
> >> As an ODE committer I would love to collaborate and modernize the
> existing
> >> ODE project. I received several direct emails from ODE users who are
> also
> >> interested in working on a new incarnation. Over the past few weeks I
> have
> >> been prototyping Ignite configurations and tasks for use in a business
> >> process management platform and I am pleased with my progress. I am
> fully
> >> dedicated to using this platform in a couple of internal applications I
> >> will be working on over the next year and I would eagerly like to
> synergize
> >> with others through the Apache Way. My current goal is to build
> >> functionality that satisfies many of the features I previously outlined
> so
> >> that myself and others could have something extensible to work from. As
> I
> >> mentioned in my previous communication process automation is big
> business
> >> and augmenting the mature and feature rich Ignite platform could
> provide a
> >> convenient low cost alternative to the other proprietary solutions in
> the
> >> space.
> >>
> >>
> >> With all that said it is entirely understandable that the current ODE
> PMC
> >> members desire to retire the ODE project to the attic and move on. I
> >> believe several of the members were part of the original ODE software
> >> contribution and incubator process 14+ years ago. Over the last few
> years
> >> interest in BPEL has diminished and they probably have no personal or
> >> financial incentives to remain involved with the project even if it
> were to
> >> be revived.
> >>
> >>
> >> If anyone else is interested in rebooting the ODE project now is the
> time
> >> to speak up. In any event I plan to continue to share my work on my
> github
> >> repo.
> >>
> >>
> >> Regards,
> >>
> >>
> >> Aaron
> >>
> >> -------------------------------------------
> >> On Wed, 1/16/19, Rich Bowen <rbowen@apache.org> wrote:
> >>
> >> Subject: ODE, ODE 2.0, the Attic and the Incubator
> >> To: user@ode.apache.org
> >> Date: Wednesday, January 16, 2019, 3:00 PM
> >>
> >> (Oops. I mailed this to users@ rather than user@. Sorry for the
> >> crosspost, but I wanted to be sure as
> >> wide an audience as possible sees
> >> this.)
> >>
> >> Hi, ODE folks,
> >>
> >> Please forgive this intrusion by an
> >> outsider. I am aware that I lack
> >> both the history of your project, and
> >> the technical understanding of
> >> your software. However, I represent the
> >> Board of Directors, and we
> >> wanted to discuss something with you
> >> before you take your project to the
> >> Attic.
> >>
> >> In today's Board meeting, we discussed
> >> your desire to move to the Attic,
> >> while at the same time taking a new
> >> project, with the same name, and the
> >> same mission, albeit with different
> >> architecture, to the Incubator.
> >>
> >> We agreed to postpone your Attic
> >> proposal another month, in the hopes of
> >> ensuring that you're aware of all the
> >> options that are available to you.
> >>
> >> There is considerable precedent, at
> >> Apache, for projects completely
> >> reboot their code, while retaining the
> >> existing project name and
> >> community. The Apache HTTP server did
> >> this very early on. Axis also did
> >> this many years ago. The advantages of
> >> this approach are numerous:
> >>
> >> * You retain your brand equity (ie, the
> >> value that you've already
> >> established in your name)
> >> * You tell the user community that
> >> you're still solving the same
> >> problem, although you're solving it
> >> differently
> >> * You save the hassle of going through
> >> the Incubator for code that will
> >> be developed "in house"
> >> * You save the hassle of having to
> >> Attic your project
> >> * You eliminate a huge amount of user
> >> confusion. "That project is dead."
> >> "No, it's not, look over here." "Wait,
> >> two projects with the same name?
> >> Which one should I be using?" and so
> >> on.
> >>
> >> While the Board is not going to dictate
> >> which route you should take
> >> here, we strongly encourage you to use
> >> your existing structure -
> >> website, mailing lists, revision
> >> control, etc - to bootstrap the new
> >> proposal discussed at https://s.apache.org/rSlq  Leverage the existing
> >> expertise in your existing community.
> >> Build the new thing, and deprecate
> >> the old thing, while being completely
> >> transparent to your users that the
> >> new thing is a completely new thing,
> >> and isn't trying to be a drop-in
> >> replacement.
> >>
> >> We believe that doing the
> >> community/project "reboot" in place, rather
> >> than going via the Attic and Incubator,
> >> is to the benefit of everyone -
> >> you, your user community, the attic,
> >> the incubator, Infrastructure, and
> >> the Apache marketing team. And there
> >> are lots of people on the board,
> >> the members list, and the Community
> >> Development team, who would be glad
> >> to offer advice, assistance, and
> >> whatever support you need to make this
> >> happen.
> >>
> >> What do you think?
> >>
> >> --Rich, for the Board of Directors.
> >>
> >>
> >>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message