Spring Richclient Status & Future

[jh@schaubroeck.be]

Hi,

As you may know, we had planned to have a Spring Richclient 1.0 release
somewhere at the end of Januari. It's March already and no release has
been pushed out. Don't start to panic just yet, I'll let you in on the
details.

Somewhere in Januari I contacted the people of SpringSource to discuss
the position of Spring Richclient within the Spring universe. As we're
in different parts of the world, it takes some time to schedule a
meeting online and contact the right people, thus time was creeping past
the previously set release date. Our goal with this meetings was to not
only discuss the present and future of the project but also the
involvement of SpringSource. The outcome is the following.

Current codebase:

We have set a new target date for the release: March 17. All bugs should
be fixed while documentation will be provided as needed. We don't want
this to delay the release date anymore and interfere with our future
plans. If documentation is requested at some point, we'll provide this
on the developer blog. This release will have maintenance updates as
requested.

Future plans:

We've decided to start a new codebase and re-brand the project to Spring
Desktop. One of the reasons behind this is the possibility to update the
existing core techniques after some research. The current codebase core
is about two years old, quite stable but not state-of-the-art anymore.
Another reason is that this allows us to start with the modularity in
mind and review the existing codebase.

We're doing some research in several areas right now and plan to start
with the codebase right after the Spring Richclient 1.0 release. We'll
take advantage of this new start to shift to the SpringSource
infrastructure (svn) as well. The aim is to align Spring Desktop 1.0
(mind the name!) with the next big Spring release train somewhere in
October.

SpringSource will help us out by providing the infrastructure, being
involved in the defining of the project (scope/roadmap) and devoting
some time for research and coding. They've made it clear that they find
this project to be valuable and want to align it with the future Spring
release.

I hope this clears things up and removes any doubt about where this
project stands and is heading towards.

ps: if you're enchanted by these words don't hesitate to pull up your
sleeves and help us out by sending in patches/documentation or just by
providing us with your feedback at the forum
(http://forum.springframework.org/forumdisplay.php?f=28) or the
developers maillist (springframework-rcp-dev@lists.sourceforge.net).

Kind Regards,
Jan

[Shake]

Okay,

so what's about applications developed with/for Rich Client Platform? Will the new "desktop" thing be compatible (easy or at least possible path of migration).
Or is the new project a complete new thing?

So at the moment I "read" Spring RCP failed and will be dropped asap...


And now I have one big wish for the new project:
Please try to avoid those "magic" things where several beans implement interfaces that are automatically called by Spring. And don't call "Application.getInstance()" anywhere.
At the moment it not possible to run parts of the application without a started Application.
And therefore it is very hard/impossible to write small unit tests without several lines of Spring related code.

As rule of thumb I suggest that it should be possible to run a Spring Desktop application without Spring (when all the necessary wireing is done by hand).


Thanks for your work. Hope the new project will get some drive and release early

[jh@schaubroeck.be]

so what's about applications developed with/for Rich Client Platform? Will the new "desktop" thing be compatible (easy or at least possible path of migration).
Or is the new project a complete new thing?

You could look upon the Spring Desktop as a new project with a lot of influence from Spring Richclient. We don't want to have all the bagage of the current codebase and have our hands tied. This is the moment to fix things at the core where needed. So there will probably be a lot of changes, but I can't predict the future and put a number on how much of a change it will be.

As for the migration path, I'm using Spring RCP on a few projects as well so I do want to provide some kind of migration path.

So at the moment I "read" Spring RCP failed and will be dropped asap...

Failed? No surely not. The current codebase is quite stable in its own. We do acknowledge that it needs a big refactoring and therefor opted to rebrand the project instead of having a 2.0 release. This allows us to start with a clear view and as said without having our hands tied to the current codebase.

Spring Richclient will be maintained, releases will be deployed as needed.

And now I have one big wish for the new project:
Please try to avoid those "magic" things where several beans implement interfaces that are automatically called by Spring. And don't call "Application.getInstance()" anywhere.
At the moment it not possible to run parts of the application without a started Application.
And therefore it is very hard/impossible to write small unit tests without several lines of Spring related code.

As rule of thumb I suggest that it should be possible to run a Spring Desktop application without Spring (when all the necessary wireing is done by hand).

This already illustrates some of the hard-to-fix parts in the current codebase (eg modularization) and a reason for Spring Desktop. Any remarks/concerns are welcome and will be noted.

Thanks for your work. Hope the new project will get some drive and release early

That's what were aiming at, thanks,
Jan

[hipparco]

i agree with you, RCP works fine, but it's an alfa version and a mayor refactoring it's needed.


Is the time for a wish list for the new version???

- Modular, perhaps with OSGI
- Better suport for MDI.
- Something similar that eclipse perspectives.
- new Event layer.
- Simpler custom binders



this is a great new for RCP users

[Shake]

I don't wanna miss the start of the new "Desktop" project. Is there any mailing list up now? Any code repository?

[adalberts]

Hi all,

we have started using Spring including Spring RCP some time ago. However, there is a significant group of people in our company that push for a mainstream rich client platform, i.e. Eclipse or Netbeans RCP. I like Spring very much and would really like to use it consistently in all layers, without locking to much into the world of Eclipse or Netbeans.

Even though I fully understand that it would be asking too much to get a detailed picture about the newly born Spring Desktop, wouldn't it be possible to flesh out at least some vision, where the new project would be going, how it positions itself wrt to the aforementioned platforms.

Just some info to gain more buy in?

Thank you

a.

[-FoX-]

It seems that the release goal of "March 17" is overdue for more than a month now. Or perhaps you were talking about March 17 2009?

Are there any updates on this subject yet?

[Andrius Šabanas]

What do you mean by saying that release is overdue? Spring Rich Client 1.0 was released on March 17, 2008. AFAIK, no release dates for Spring Desktop were announced yet.

Andrius

[Andreas Senft]

According to Jira, Spring Desktop is scheduled for October 13.

Overdue is the maintenance release of Spring RCP (1.0.1) which has been scheduled for April 21.

[LievenDoclo]

Hey guys,

Here's a thought:

Why don't you dive into Spring RCP's internals, get familiar with it and offer to help out whenever and wherever you can? Currently, there aren't many people actively engaged in RCP's development.
I try to make time to answer as many questions as I can, it would be nice if more people did that . I'm hardly an expert, but I've worked a lot with RCP and that seems to be enough...

I do notice that there are a lot more people trying out RCP since it's release last month, but not a lot of people are contributing back to the community, which is a shame. Only the community can tell what in RCP and the future Desktop should be included to be what people want.

As for the 1.0.1 release, you'll have to check with Jan, but he's quite busy at the moment. If there are bugs which you really need fixed, put them in Jira or vote for one of the existing ones (or better yet, provide a patch).

PS: don't take this personal, this is a general observation.