[buro9]

I still do this kind of thing, and literally thirty minutes ago logged back on to an old server to fix the scraping script to deal with some bad data on the origin website.

It's quite pleasing when it works, and a hell of a lot easier to deal with as you just read the web page and scrape away.

I think when web pages become more clearly marked up we'll start seeing less and less of the API mess we've got ourselves into. If the websites did what we would hope they would, there wouldn't be a great need for a RESTful aspect of them.