I've finally built my new replacement blog. There are lots of things I have had to compromise on. One of the things that annoys me most is that Haloscan's JavaScript is not compatible with properly served XHTML. Though Haloscan claims that it's 'Valid XHTML', if you actually serve your pages as application/xhtml+xml, the JavaScript doesn't work. 
This is for a simple reason - it uses document.write.
I have an idea of how I can fix this, and it involves a really gnarly script which would go through the JavaScript and convert the data in to, say, an XML format and use that to replace the comment and TrackBack links using equally horrible regular expressions. That'll have to wait. For other XHTML wannabes who use Haloscan, I'll fix that up sometime soon as a Python script (probably).
We're getting there.
So, without further ado, I present my new blog. It doesn't look too good. Hopefully, I'll make it look nicer soon. If you are subscribed to this blog via RSS, the FeedBurner RSS will redirect in the next day or so when I'm happy with how it works.
I'm planning to open source this. The PHP scripts which run it are very hacky and need significant improvement before I can unleash them on the poor, unsuspecting public. 
