A lot of people (here and
here, for example) are discussing XAML, an upcoming MS technology which many think is similar to XUL. In case people are confused, no-one has actually seen XAML yet; we only have rumours as to what it is.
The only thing we know is that it is a markup language for describing applications and possibly a few clues from a screenshot. In fact, I don't think we even know for sure that it's for describing user interfaces at all. In fact, if it can, it may be a 'side effect' of what it's actual purpose is.
I suspect the goals of XAML are quite different from the goals of XUL (apart from any MS evil conspiracy). For instance, XUL is designed to be a cross-platform UI language, akin to HTML, that uses existing Web standards such as CSS, XML, DOM, RDF and so on, and includes some interesting features such as overlays and XBL for extensibilty. In my opinion, the only missing things could be fixed up in time with a few good developers who can dedicate themselves to it.
One of the difficulties in promoting XUL and Mozilla is that many people have this misconception that XUL shouldn't be used because IE has more users. That's unfortunate. That confusion is created because people hear the name Mozilla, and immediately lump it in the 'browser' category, forever more being known only as an IE competitor, not as something more. There's really no reason to do this of course. A better way to think is to allow room for both Mozilla and IE. There really isn't any reason to think that just because a user uses IE for browsing that they shouldn't use a mail reader or an RSS reader or an IM client that doesn't use Mozilla technology such as XUL or the Gecko engine.
So, even if there are lots of competing XML UI languages (and there are many), there's no reason to think you need to use one or the other because one is more popular.
The RDF Interest Group logs point to a
RDF file containing US States meta-info. It lists capitals, regions and the neighbours of states. Upon looking at the data, I thought of an interesting challenge: write a script that could generate a map -- perhaps with SVG -- using only the list of regions and neighbours from the RDF. Obviously, the map would be completely wrong, but it would be interesting to see what the results would be. Better accuracy would require more data: land areas, border sizes, etc...