Re: Status of the kdepim package?
Am Mittwoch, 28. November 2012 schrieb Julian:
> On 28/11/12 03:21, Martin Steigerwald wrote:
> > Am Montag, 26. November 2012 schrieb Julian:
> >> On 25/11/12 12:00, David Smith wrote:
> >>> On Nov 25, 2012 12:51 AM, "Lisandro Damián Nicanor Pérez Meyer"
> >>> <perezmeyer<mailto:perezmeyer@gmail.com>@
> >>> <mailto:perezmeyer@gmail.com>gmail.com<mailto:perezmeyer@gmail.com>
> >>> >
> >>>
> >>> wrote:
> >>>> Control: tag -1 upstream
> >>>> thanks
> >>>>
> >>>> Hello,
> >>>>
> >>>> the bug you reported is a KDE bug. Please report it via
> >>>> Help -> Report bug... menu item of the respective application
> >>>> or by visiting https<https://bugs.kde.org/>://
> >>>
> >>> <https://bugs.kde.org/>bugs.kde.org<https://bugs.kde.org/>/
> >>> <https://bugs.kde.org/> and following instructions
> >>>
> >>>> to report a new bug.
> >>>
> >>> Hello, I've tried forwarding a bug related to kmail in Debian
> >>> Wheezy to upstream using this method and it was rejected because
> >>> upstream said this version of kmail is no longer maintained. Do
> >>> you know if upstream might have made a mistake here or are kmail
> >>> bugs in Wheezy really unforwardable to upstream? ...or perhaps
> >>> the above method of reporting the bug for KDE upstream is putting
> >>> the wrong version?
> >>>
> >>> My bug report about a crash in Wheezy's kmail is here:
> >>> https://bugs.kde.org/show_bug.cgi?id=309532
> >>>
> >>> My debian bug report is here:
> >>> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=692275
> >>>
> >>> upstream closes this bug as unmaintained and won't look at it. The
> >>> bts tracker then marks it as fixed upstream. Which I think might
> >>> cause any kmail bugs, including rather important ones, to fall off
> >>> the radar.
> >>>
> >>> Thanks for your time.
> >>
> >> Hi, I'm not an official maintainer of the debian packages or
> >> kdepim/akonadi libraries and applications with dependencies on such.
> >> I will say this. I'm a big fan of KDE. Since KDE4 I've been using
> >> iceweasel.
> >> Not sure of the status of the kdepim debian package, still appears
> >> to be stuck at
> >> 4.4.x on all suites.
> >> Perhaps someone official would like to comment?
> >> Jules.
> >
> > Please look in the mailing list archives.
> >
> > Last official status was, thats its KDEPIM 4.4.x for Wheezy. Debian
> > Qt/KDE developers also shared their reasoning for this decision.
> >
> > I am not aware of any official released KDEPIM 2 packages for Debian
> > in any suite, not even in http://qt-kde.debian.net
> >
> > I think for stable it at least would need to be KDEPIM from KDE SC
> > 4.9.3 as thats there first version search and filters seem to work
> > nicely (most of the time). Possibly even 4.10 to catch further
> > improvements to Nepomuk and how Akonadi accesses it.
> >
> > And then this would need to be tested for quite a while.
[…]
> I'm sure there is some reasoning from way back, but its pretty obvious
> to me that kdepim/kmail/akonadi was a fail for squeeze and other
> distros.
I disagree here. I used KDEPIM 4.4 on Squeeze as well. It worked quite
nicely. AFAIR there isn´t even that much of a difference between using it
now and back then. And heck I use it with lots, – I really mean – of data:
martin@merkaba:~> du -sh ~/Mail
16G /home/martin/Mail
martin@merkaba:~> find ~/Mail -type f | wc -l
471690
And no its way more mails. I have more than one million archived mails in
mbox folders.
At the time I switch this to KDEPIM 2 I hope it will be able to grok it
nicely. At the moment I am not even sure it will be able to switch due to
low free space on /home. Only 8.7 GiB left at the moment and I think I
better make more free space on it before migration for the Akonadi
database.
> I'm quite aware of the bugs with this package so I wonder if keeping
> kdepim down @ 4.4.x everywhere is the best plan, particularly if
> the problem is that there isn't much to test.
Which bugs do you refer to *concretely*? Spreading FUD does not help
anyone.
> My biased against kdepim/akonadi/kmail is based purely on my
> experience with the squeeze 4.4.7 release. People mention
> a better experience with wheezy (4.4.11), if kdepim 4.4.x is
> not good enough for stable (squeeze) I really hope its good
> enough for wheezy.
Well Wheezy has newer KDEPIM, so first thing before speculating about the
state of KDEPIM for Wheezy would be actually trying it out.
Up to now I got the impression that you didn´t do anything of the
following:
1) Test KDEPIM 4.4.11 in Wheezy.
2) Compare with recent KDEPIM 2 in other distro (like Kubuntu) or self-
compiled
And do this with real data, i.e. a copy of your mail account.
Thus I get the impression that your suggestions are pure speculation.
I admit that I didn´t do option 2, so my impressions are speculation as
well. Somaybe a recent KDEPIM 2 is up to the task. But at least I did
option 1 and I know that it works well enough for me and I do read kdepim
and kdepim-users to get at least some impression about KDEPIM 2. Sure I´d
like the improvements in KMail 2, when its ready for Debian stable.
As I do not know whether it is and I did not test it, I am not going to
ask the Debian Qt/KDE team to change their decision.
Are you willing to test KDEPIM 4.4.11 in Wheezy and a self-compiled /
Kubuntu based KDEPIM 2 side by side and share your impressions to provide
the Debian Qt/KDE team with some real data?
Are you willing to work on KDEPIM 2 test packages?
> If upstream kicks it back down with "unsupported" Then maybe debian
> should do the same.
Unless I know it better by having tested both KDEPIM 4.4.11 and KDEPIM 2 I
will not ask the Debian Qt/KDE team to change or even review their
decision for Wheezy. Why?
They decided and put their efforts into ensuring that KDEPIM 4.4.11 for
Wheezy is a good experience. Their man power is limited. By reviewing the
decision they split their man power between two things. And besides that
there is another good reason:
Debian Wheezy is partly frozen.
The Debian Qt/KDE team would need a good case in order to get KDEPIM 2
included in Wheezy.
Are you willing to help by providing real data for such a good case? Yes,
thats actually some work to do.
Sometimes I would be willing to help, but I am pretty sure it needs at
least one or two days to setup KDEPIM 2 in parallel in a safe way for my
data. And even more effort to test and use it accordingly.
Heck, yes, maybe the decision for KDEPIM 2 will not be based completely on
such hard data but on hearsay a bit as well, due to being complex to test
for any use case. But I got the impression that some members of Debian
Qt/KDE actually tested KDEPIM 2 which is more than I did. Complaints in
kdepim and kdepim-users got quieter, but suggest to at least have KDE SC
4.9.3 for a quite good experience. Thus I think its even quite good to
wait for KDE 4.10 since it contains further Nepomuk/Akonadi feeds Nepomuk
related improvements.
And that said mixing KDE PIM from KDE SC 4.9.3 with KDE 4.8.4 might also
lead to some issues. Upgrading KDE SC 4.8.4 to KDE SC 4.9.3 would need
additionally testing.
Thus I hope for KDE SC 4.10 packages including KDEPIM 2 for testing, but I
am aware that unless there is an intent to change the decision for Wheezy
there is no need to hurry this. Cause Wheezy+1 is still some time away.
Thus I understand if Debian Qt/KDE developers concentrate on Wheezy as the
next stable for now.
And given that Debian switched to KDE 4 not before KDE SC 4.2 and from my
preliminary testing I can only but agree to that decision, I do get the
impression, that the Debian Qt/KDE team has quite a good estimate as to
when something is ready to let loose on Debian stable users.
--
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7
Reply to: