Category Archives: XML Prague

Oxygen 11 and an Old Bug

Oxygen 11 is out and I just installed it on my laptop. It’s still the best XML IDE there is, by far, but I’m getting a bit annoyed by a bug that has persisted for a year now.

Long story short, but basically the tabs bar that keeps track on the open documents and indicates the current document is buggy and will not correctly focus on the current document if that document is too far right to be visible without scrolling, and a previous document, also too far right to be visible without scrolling, has just been closed. The bar and the actual visible document do not synch so while the document itself is open and editable, the tab is missing. Very annoying, to say the least, if you are working with a highly modularised stylesheet and need to have more than a handful documents open at the same time. Apparently the bug is in a third party component and until that component is updated or replaced, the bug will persist.

I can live with that bug, though, and the new version seems to have enough fun stuff to keep me busy for a while. XProc support, for example, is a welcome addition. I’ve been wanting to try it ever since Norman Walsh presented it at XML Prague, and now there should be no further excuses.

Linux on the Laptop

Following the unfortunate events surrounding my presentation at XML Prague (a fabulous event, by the way; you should have been there), I now run Debian GNU/Linux as my primary OS on my work laptop. There is a Windows XP partition, so far, but my plan is to use Xen and virtualisation, and run the Windows operating systems as Xen domains.

The laptop installation that failed contained my first attempts at virtualisation, by the way. Microsoft’s Virtual PC ran Windows 2003 Server and Cassis, the Document Management System that I’m part of developing at Condesign Operations Support, and was connected to my XP installation through a loopback adapter. In theory, this is a very nice setup since it is possible to simply run a complete image of an OS and the server setup as part of a demonstration and then reset it to its pre-demo state for the next show. In practice, however, Virtual PC does not deliver. The hardware it emulates is very limited and everything it does is rather slow. It was enough to wet my appetite, however (together with my friend Niklas’ obsession with Xen), so I decided to do it right, now that I had to wipe the old drive anyway.

My Debian installation does not yet run a Xen kernel, but I’ll keep you posted.

XML Prague Nerves

XML Prague is only a week away so I’ve been working on my slides. I’ve done stuff on resource naming (links identifying a resource by name rather than location), markup, linking systems (actually only one, XLink), publishing, “document trees” and more, but it all feels deceptively simple to me. Right now, the whitepaper’s most important (and only?) message, use an abstraction layer such as URNs to always identify whatever resources you are using, seems trivial, even. The question, therefore, is if I really am that brilliant at what I do (I’d like to think so, obviously) or if the whole thing is so simple that it doesn’t need pointing out, in which case I’m in trouble and people will fall asleep (or worse) at my presentation.

As I said, I do include some other stuff in the presentation, but in my mind it’s all a result of the basic premise.

Or maybe it’s just me being nervous. We’ll see.