[Dune] Release plans for Dune 2.2?

Martin Nolte nolte at mathematik.uni-freiburg.de
Thu Mar 15 10:58:47 CET 2012


Hi Oli,

to me your statement looks like a contradiction. Either it should take too 
much time, it shouldn't be a show stopper, or we cannot have another release soon.

Best,

Martin

On 03/15/2012 10:43 AM, Oliver Sander wrote:
> Am 15.03.2012 09:10, schrieb Martin Nolte:
>> Hi Oli, hi Christoph,
>>
>> What about the life time of geometry objects? I thought this was
>> considered a showstopper for a 2.2 release ;-).
>>
> I still think it is. It should take to much time; I already had
> a look at UGGrid.
> best,
> Oliver
>
>
>> Anyway, I'm basically on your side unless additional work is required.
>>
>> Best,
>>
>> Martin
>>
>> On 03/15/2012 07:31 AM, Oliver Sander wrote:
>>> Hi,
>>> I am in favor of a quick 2.2 release. Christoph's reasons are valid
>>> (even though a bit more backporting to 2.1 would also help him),
>>> but I have another one: We plan to get a Dune release into the next
>>> Debian stable distribution. The next Debian feature freeze is
>>> scheduled to be around June, so if we have released Dune 2.2 until
>>> then it can be included. Otherwise we will be forced to upload 2.1.
>>>
>>> Best,
>>> Oliver
>>>
>>> Am 14.03.2012 08:39, schrieb Christoph Grüninger:
>>>> Hi Dune developers,
>>>> do you have any plans for a Dune 2.2 release?
>>>> I'd like to discuss a release before June (PDESoft / Dune Developer
>>>> Meeting / Dune User Meeting), after June, or even later like end of the
>>>> year or early next year.
>>>> Some arguments I was thinking about:
>>>>
>>>> - It is difficult to stay compatible to both Dune 2.1 and Dune 2.2-svn.
>>>> The warnings from dune-geometry drive me crazy. And I stumbled over a
>>>> problem when creating a tarball for a external dune module; it could not
>>>> find dune-grid because dune-geometry was not in the include path but the
>>>> dune-grid tests needs it.
>>>>
>>>> - The list of recent changes is not too bad and I'd like to have some of
>>>> the new features in projects relying only on stable releases. This
>>>> contains convenience features (DUNE_UNUSED, DUNE_DEPRECATED_MSG),
>>>> compatibility features (autoconf warnings, SuperLU 4.3), and the bunch
>>>> of little bug fixes (e.g. naming of VTK files, bug in array detection,
>>>> missing header in array.hh)
>>>>
>>>> - We must wait for open tasks like return geometries as real objects,
>>>> virtual refinement, general dynamic parameter interface for grids. Or we
>>>> reschedule them for dune 2.2+1.
>>>>
>>>> - We deprecated quite a lot of things. A later release would even
>>>> increase this number making the transition more work for users.
>>>>
>>>> - If we can release before the developer meeting, new features could be
>>>> directly implemented. Otherwise they either should be postponed or would
>>>> delay the release.
>>>>
>>>> When speaking about releasing. Would you mind to have a look on the wiki
>>>> page about the release manager's job description? Felix Albrecht started
>>>> the page and I added my own experiences:
>>>>
>>>> http://users.dune-project.org/projects/main-wiki/wiki/Guides_release_manager
>>>>
>>>>
>>>>
>>>> Bye
>>>> Christoph
>>>>
>>>
>>> _______________________________________________
>>> Dune mailing list
>>> Dune at dune-project.org
>>> http://lists.dune-project.org/mailman/listinfo/dune
>>
>
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune

-- 
Dr. Martin Nolte <nolte at mathematik.uni-freiburg.de>

Universität Freiburg                                   phone: +49-761-203-5630
Abteilung für angewandte Mathematik                    fax:   +49-761-203-5632
Hermann-Herder-Straße 10
79104 Freiburg, Germany




More information about the Dune mailing list