Having asked nicely for a WEP key from a housemate, I now have an internet connection in the evenings. I'm still not sure how much time I want to spend on a computer, but it might at least let me work on some free software outside of work. I've borrowed a huge book on PostgreSQL, so now have to find time to read it.

This week, I have mostly been in Mönchengladbach, at the German office of credativ GmbH. It's a completely different experience working here, compared to Rugby. For one, the building is a lot bigger - several stories, compared to the tiny room we have at the moment in England (soon to change, hopefully). I've been meeting a lot of people, and trying to remember their names. It's been a while since I was last in Germany - exactly seven years, in fact. This time round, I get to visit German pubs, and try Kölsch and Pils. I discovered yesterday that I share a birthday with another credativ employee, Bernd. ...

Oh Pete, I deliberately didn't use your name because I wasn't attacking you in particular; I was aiming at the completely misleading article whose pagerank you (and Digg) increased to the point where it achieves higher results in searches for "XDS drag and drop" than the XDS specification itself. Please don't take it personally. Of course, you still fail utterly for not providing any explanation of why you were so pleased to have this feature in your original post, and for not linking to, say, the GNOME 2.20 release notes, the relevant bug, or just writing something that won't feed anti-GNOME trolls for the next five years. But I'm glad to see that I've at least provoked the former, even if you also called me an ass at the same time. :) ...

Recently, people linked to a random news site claiming GNOME has added "XDS support" in 2.20. XDS is an extension to the XDND drag-and-drop protocol. This is very misleading. The headline reads as if GNOME has not had any drag-and-drop support until now, which is ridiculous. The Digg headline was even worse: "GNOME Gets Real Drag n' Drop Support with XDS". Bullshit. What actually happened is that Nautilus (the GNOME file manager) has received a patch to support this extension in the main window bit. It will probably help File Roller act more intuitively with respect to dragging files to Nautilus windows, and it is a good thing. It won't work with Nautilus's list view, yet. It is also possibly the least interesting of the new features in GNOME 2.20. It deserved one sentence tagged on to the File Roller bit of the release notes. I am not going to "scream with joy". ...

Today, I finally found a place to live in Rugby. It turned out that a landlord friend of my boss had a room to let, in a very convenient location. I'm within ten minutes' walking distance of the station, the town centre, and the (hopefully soon to be) new office. It's also quite cheap (university accommodation prices), with all utilities and bills except internet access included in the rent. Yay. ...

My first day at credativ went well. In the morning I updated some RHEL packages on a couple of test servers. I suspect the live systems will be next. In the afternoon I copied over some PostgreSQL configuration tuning for an upgrade from 7.x to 8.x. I'm getting up to speed with the various systems - there's even the threat of fixing bugs in Java code, in the future.

I just realised that my new job title is 'Technical Engineer', so will probably annoy Sadiq.

Late on Tuesday evening, I successfully got a branch of Choob to run without its 'Object DB'. Presently, it runs only a few of the more simple plugins - for instance, the 'Alias' plugin is not supported, so I spent a couple of minutes trying to work out why it wasn't replying to commands. With ODB gone, vast swathes of complicated parsing code can be removed from the bot's core. JJTree should no longer be needed to compile the bot. The next step is probably to consolidate the core database code using Hibernate, and then get the more complicated plugins to work again. ...

So, during a conversation at PyCon UK, Keith White noticed that he owned the same model (HP nx6325) laptop as me. He also informed me that his processor was 64-bit, and asserted that mine probably was as well. I was sceptical. It's a Mobile Sempron - last I checked, all Semprons were 32-bit. Of course, I don't keep up with hardware at all. Checking /proc/cpuinfo: tim@regulus:~$ cat /proc/cpuinfo <br /> processor : 0<br /> vendor_id : AuthenticAMD<br /> cpu family : 15<br /> model : 76<br /> model name : Mobile AMD Sempron(tm) Processor 3500+<br /> stepping : 2<br /> cpu MHz : 800.000<br /> cache size : 512 KB<br /> fdiv_bug : no<br /> hlt_bug : no<br /> f00f_bug : no<br /> coma_bug : no<br /> fpu : yes<br /> fpu_exception : yes<br /> cpuid level : 1<br /> wp : yes<br /> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext fxsr_opt rdtscp lm 3dnowext 3dnow up pni cx16 lahf_lm extapic cr8_legacy ts fid vid ttp tm stc<br /> bogomips : 1597.70<br /> clflush size : 64 ...

Of course, there are no packages for Jackfield yet. Rather than mess about with installing Jackfield, I badgered Stuart Langridge to do it for me. So, I now have a working shiny clock thing. Making packages for Jackfield requires solving a few problems, like gconf keys for showing Jackfield and multiple instance detection.