Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [nebula-dev] NatTable EPP

> Sooo.. any consensus here ..

haha! 

Okay, here are my _true_ objections : when somebody offers a new widget, they are all excited and have major plans for everlasting support. The reality is that support declines with time and that eventually there is a widget that functions properly but is pretty much in maintenance mode, if even that.

This means that the current excited ones (all you who are reading this) have to take on the burden of the maintenance of that widget. Mik Kersten once told me, it is like giving somebody a puppy, all is cute and cuddly in the beginning but the beast gets bugs and must be taken care of. 

Now that we open up the possibility to start a subproject this sounds appealing to people. After all,  having your own sub-project sounds better then just be one-of-the-widgets, right? The reality probably is that it is just a major additional overhead and time waster. 

The definition of a subproject according to the eclipse development process [1]

"a place to innovate, test new ideas, grow functionality that may one day be moved into another Project, and develop new committers" 

Does the NatTable fit this? I'm not sure. I seems a full featured and stable widget and an excellent pearl in the nebula-proper realm. Does it need to fit the above? No, rules are there to be twisted.
 
Having said this, NatTable has it's community, a dedicated site (nattable.org) the svn repository seems active with almost daily commits and there are commits by several people. 

Do we want to enforce rules on creating sub-projects? If so, what are they?

Edwin and others, can you please comment on this?

Wayne, the link in [1] says that we formally cannot create a subproject since Nebula itself has not matured. Is this true, does this possibly block?


[1] http://www.eclipse.org/projects/dev_process/development_process_2011.php#4_9_Incubators

Regards,

Wim



On Mon, Nov 28, 2011 at 3:35 PM, Edwin Park <esp@xxxxxxxxxx> wrote:
Sooo.. any consensus here on how to move forward? Wim, Nicolas - are you ok with moving forward as a subproject or do you still have strong objections?

Edwin



On Wed, Nov 23, 2011 at 2:25 AM, Tom Schindl <tom.schindl@xxxxxxxxxxxxxxx> wrote:
Hi,

I don't see the problem with creating a subproject(s). The main reason
is that NatTable can follow its normal release schedule and I could
envision others doing the same if they want too.

For me the Nebula-Project is getting a Container-Project for widgets
where people who don't want to deal with all the Eclipse Project stuff
bring their widgets to Nebula-Proper and Nebula-Incubation and others
create their own sub-project like Edwin feels more comfortable.

At least 3 times a year (similar to the Eclipse Release Train) all
Nebula-Projects are doing a release whereas subprojects can do releases
in between.

A release a very special thing within the Eclipse Project (e.g. we need
a IP-Log, provide a project plan, ...) so I disagree we can simply see a
release as a snapshot because if we want to get preceived serious we'd
need to have a maint-branch and one where we develop new features.

Tom

Am 23.11.11 02:17, schrieb Edwin Park:
> Hmm, interesting. Wim, Nicolas - I agree with what you are saying, but I
> also see it as something of a departure from how Nebula has been
> organized to date.
>
> In order to do what you say, I think there will need to be stronger
> boundaries between widgets in terms of source, branches, builds,
> artifacts, documentation, etc. The overhead of delineating widgets
> within the Nebula project then begins to approach the overhead of having
> a separate subproject to begin with...
>
> I'm not saying that it can't be done, but my feeling is that it will be
> much more difficult to reorganize all of Nebula than to just have
> NatTable as a separate subproject. And I'd love for it not to take
> another year for us to bring NatTable to Eclipse.  ^_^;;
>
> Edwin
>
>
> On Tue, Nov 22, 2011 at 5:05 PM, Wim Jongman <wim.jongman@xxxxxxxxx
> <mailto:wim.jongman@xxxxxxxxx>> wrote:
>
>     Hi,
>
>     Yes I agree with Nicolas. If the only way of getting you guys here
>     is through a subproject then I will vote +1 but I think that  we are
>     stronger if we join forces. We will still be responsible for our own
>     widgets but we can work together on the commons (build, website,
>     wiki, blogs, etc..)
>
>     Shall I continue to file the CQ while we discuss this?
>
>     Regards,
>
>     Wim
>
>     On 22 nov. 2011, at 12:36, Nicolas Richeton
>     <nicolas.richeton@xxxxxxxxx <mailto:nicolas.richeton@xxxxxxxxx>> wrote:
>
>>     Hi all,
>>
>>     I'm not sure that creating subprojects for a single widget would
>>     be a good thing. I understand the reasons why NatTable needs to be
>>     a subproject, but I think these are issues we have to solve for
>>     every widget :
>>
>>     * Being able to install independently :
>>     Each widget has his own plugins : widget, tests, feature, ...
>>      This won't change.
>>     Version numbers do not necessary have to be in sync for all Nebula
>>     widgets.
>>
>>     * Having its own build
>>     XViewer already have his own build because eclipse projects
>>     depends on it. This is something that can be done without being a
>>     subproject, but I still think this should be avoided. (If we look
>>     at an another projet : I believe SWT build fails if tests for one
>>     component fail). The key thing is : projects which depends on
>>     Nebula should not build against the trunk. They should build
>>     against an integration or a release branch/site which always
>>     builds. We have setup things in a way that make this work.
>>
>>     * Being able to release with its own schedule.
>>     We are looking for a 1.0 nebula release with several widgets, but
>>     we will always have to fix major/critial issues on a widget
>>     without waiting for the next major Nebula release. I believe we
>>     have to see a nebula release as a snapshot of the current stable
>>     widgets (for the annual eclipse release train) and keep an update
>>     site with intermediate/frequent releases.
>>
>>     It NatTable committers are ready to go fast and get out of
>>     incubation right away, I rather remove all widgets from the
>>     release project, keeping only NatTable and then add every stable
>>     widget one after one, than going thru the eclipse process for only
>>     one subproject.
>>
>>     That said, everything is possible :-)
>>     --
>>     Nicolas
>>
>>
>>
>>
>>
>>     Le 21 nov. 2011 à 23:26, Tom Schindl a écrit :
>>
>>>     Hi,
>>>
>>>     +1 for me on moving forward as a subproject to Nebula but Wim I'd
>>>     like
>>>     to hear Wims and Nicolas' input.
>>>
>>>     Tom
>>>
>>>     Am 21.11.11 23:05, schrieb Edwin Park:
>>>>     Hi all,
>>>>
>>>>     Once upon a time, there was a proposal to bring NatTable to
>>>>     Eclipse... :-)
>>>>
>>>>     It has taken some time, but we've gotten through all the IP
>>>>     attribution issues finally and I have the go ahead from Eclipse
>>>>     legal
>>>>     to push this forward again. In the meantime I know Nebula has been
>>>>     going through changes as well. Importantly, in order to move forward
>>>>     this needs to be pushed by the Nebula folks. Tom, would you still be
>>>>     the one to do this or does that fall to Wim now?
>>>>
>>>>     Also, I wanted to update everyone on the current status and sync
>>>>     up on
>>>>     the plan for moving forward to make sure we're all on the same page:
>>>>
>>>>     The original New Widget request for NatTable is here:
>>>>     https://bugs.eclipse.org/bugs/show_bug.cgi?id=328836
>>>>     The latest version of the code attached to this is NatTable
>>>>     2.2.0. The
>>>>     current production version is 2.3.0. No additional dependencies have
>>>>     been added in the interim, but there is new post-2.3.0 code in trunk
>>>>     that introduces a new dependency on Apache Poi for table export.
>>>>     This
>>>>     is packaged as a separate extension bundle however, so if there are
>>>>     issues with this we can always omit it from moving over to Eclipse.
>>>>
>>>>     The NatTable Eclipse Project Proposal is here:
>>>>     https://docs.google.com/document/d/1Krpj0ceaWqP-WndbR1Ba79gwhGCbUHNspw-j9naYPQc/edit?hl=en_US
>>>>     The only update that may need to be made for this is the tentative
>>>>     project schedule, which currently indicates code contribution and
>>>>     migration to Eclipse is Q4 2011..
>>>>
>>>>     The plan for moving NatTable to Nebula was to make it a Nebula
>>>>     subproject rather than incorporating it into the conglomerate Nebula
>>>>     project and build. Tom and I favored this approach because we wanted
>>>>     to preserve the ability for NatTable to maintain its own independent
>>>>     release schedule. NatTable has 7 active committers and a history of
>>>>     putting out regular releases over the past four years. I'm happy to
>>>>     take on the additional maintenance overhead of treating NatTable
>>>>     as a
>>>>     separate project since that's what it is now anyway.
>>>>
>>>>     Wayne needs confirmation from the Nebula project that we want to
>>>>     move
>>>>     forward with NatTable as a Nebula subproject, and then my
>>>>     understanding from legal is that we'd need someone from Nebula
>>>>     to kick
>>>>     off the CQ.
>>>>
>>>>     It's been over a year since we started this journey and I'm eager to
>>>>     see this through. :-)
>>>>
>>>>     Thanks,
>>>>     Edwin
>>>>     _______________________________________________
>>>>     nebula-dev mailing list
>>>>     nebula-dev@xxxxxxxxxxx <mailto:nebula-dev@xxxxxxxxxxx>
>>>>     https://dev.eclipse.org/mailman/listinfo/nebula-dev
>>>
>>>
>>>     --
>>>     B e s t S o l u t i o n . a t                        EDV
>>>     Systemhaus GmbH
>>>     ------------------------------------------------------------------------
>>>     tom schindl                 geschäftsführer/CEO
>>>     ------------------------------------------------------------------------
>>>     eduard-bodem-gasse 5-7/1   A-6020 innsbruck     fax      ++43 512
>>>     935833 <tel:%2B%2B43%20512%20935833>
>>>     http://www.BestSolution.at                      phone    ++43 512
>>>     935834 <tel:%2B%2B43%20512%20935834>
>>>     _______________________________________________
>>>     nebula-dev mailing list
>>>     nebula-dev@xxxxxxxxxxx <mailto:nebula-dev@xxxxxxxxxxx>
>>>     https://dev.eclipse.org/mailman/listinfo/nebula-dev
>>
>>     _______________________________________________
>>     nebula-dev mailing list
>>     nebula-dev@xxxxxxxxxxx <mailto:nebula-dev@xxxxxxxxxxx>
>>     https://dev.eclipse.org/mailman/listinfo/nebula-dev
>
>     _______________________________________________
>     nebula-dev mailing list
>     nebula-dev@xxxxxxxxxxx <mailto:nebula-dev@xxxxxxxxxxx>
>     https://dev.eclipse.org/mailman/listinfo/nebula-dev
>
>
>
>
> _______________________________________________
> nebula-dev mailing list
> nebula-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/nebula-dev


--
B e s t S o l u t i o n . a t                        EDV Systemhaus GmbH
------------------------------------------------------------------------
tom schindl                 geschäftsführer/CEO
------------------------------------------------------------------------
eduard-bodem-gasse 5-7/1   A-6020 innsbruck     fax      ++43 512 935833
http://www.BestSolution.at                      phone    ++43 512 935834
_______________________________________________
nebula-dev mailing list
nebula-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/nebula-dev


_______________________________________________
nebula-dev mailing list
nebula-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/nebula-dev



Back to the top