I'm going to try something different. (*insert knowing laughter here*) it seems the topic of "Why KDE" is coming up more and more often in my day-to-day life both on- and off-line. sometimes the actual topic is "Why open source" but i usually manage to steer it into "Why KDE" because KDE is something i'm rather interested in and it provides a great case study. through these conversations i've come across some interesting thoughts, and i'm going to record them here over the next little while, one per blog entry. here's #1.

Some time before KDE 3.0 David Jarvie, the author of KAlarm, asked me some questions about the KOrganizer alarm daemon. I answered that it would be a great thing, if KOrganizer and KAlarm could share the same daemon, because I thought that by eliminating redundancies development would become easier and we could use our development resources more efficient. So we imported KAlarm into the KDE CVS and David did a great job implementing the code needed for sharing the daemon. The result was that we had a new application and the shared daemon in KDE 3.0. So far so good...

Now that the first dust about the first beta is settled, my thinking of how to continue gets more and more in foreground.

Bug fixing continues at a nice rate, but there are still some show stopper bugs that were fine for beta1, but are not for whatever comes after:

So, finally KOrganizer's bug count got down from almost 95 bugs in July/August to 44 right now. The last few weeks I was really busy producing patches at quite a rate. But every time I tested a patch, I ran into the next bug, and then the next, and so on.

But I think, korganizer is really getting into shape now for the KDE 3.2 release.
As far as I'm concerned, all the important issues are resolved / fixed (I don't know about Cornelius, and he's the maintainer of korganizer, so this list is only "unofficial"):

This didn't seem to work first time, so I'll try again:

I did a major reorganisation of the code just before the freeze hit. This caused a lot of breakage (eg. the wrapper classes were borked, as was KPart support), fortunately things are working again now. One important change is the use of separate classes to handle proxies for QVariant types and opaque pointers, this makes the code a hell of a lot easier to understand. Now that the structural stuff is done, Ian Geiser's SQL bindings seem to be basically working (though the QObject wrapper class needs a couple of bugs squashing) so the only thing that's left actually broken is the QPainter wrapper class. All in all, this code is looking good for the 3.2 release.

It seems we're coming close to a new release. When? Dunno, just give time to time for bugfixing. Krecipes 0.4 has quite a few additional features, so we cannot hold the beast anymore on the servers. Someone will have to download it soon before it explodes!

okay so my old visor prism is starting to show its age... its big its bulky and after owning two newtons before that, its cumbersome to use... so im shopping.

the new palms are getting better at their input, but its still a 9 out of 10 on the most inane ways to put data into your pda... but damn, they are fast, and far cheeper than the linux and wince offerings...

i'm having a coding frenzy at home at the moment while waiting
for my adsl to be installed. amazing the amount of work you can
get done when irc, mail and even sane tv listings are all unavailable.

The skinny: two people have reported that KStars will not compile for them; both are using Redhat 9 and gcc 3.2.2. The error message is:

lx200classic.cpp:29: structure `eqNum' with uninitialized const members

Last night I did a lot of bug fixes and janitor work on bugs.kde.org and others were also working hard to get some stuff done before the string freeze that hits us as of today. Still we (#kde-pim) realized that we are not quite there, which is why we will most probably release Kontact as version 0.8 with the following features disabled: