A long, long time ago I once validated my website (the old one) on what was the Bobby website. Even though at the time my site was valid XHTML I was presented with a whole slew of errors and that was the end of that. I thought, "Hell, no one reads it anyway, there's no need for accessibility".

Recently, It's been niggling me again and I thought I'd take a look to see how far off I was. The Bobby website has now change and is called Watchfire WebXACT.

As it turned out there wasn't too much to change. It only failed to pass Priority 2 and 3 automatic checks requiring the following:

  • The javascript event handlers for the comments had to be changed so they didn't rely on a mouse
  • I had to associate form controls with labels on the search form I have
  • I had to create link phrases that make sense
  • I had to make sure I didn't use the same link phrase on a page
  • Include default place holding characters in the search box
  • Separate adjacent links with more than whitespace

It turns out if I'd not bothered with comments things would have been better to start with - I wouldn't have the javascript to worry about for starters. As it was I had to add an onkeypress equivalent to the mouse one. The form controls one was the oddest since I'd never played with forms before and so just wasn't familiar with the elements. Creating link phrases that make sense and making sure I didn't use the same link phrase was a biggy: For example on any page I have the "Permanent Link" and "Comment" links used multiple times. I had to alter my Nanoblogger templates so these actually read "Permanent Link to e2006-08-01T21_59_02.txt" and "Comment on e2006-08-01T21_59_02.txt". Using the post ID was the only way I could guarantee not having the same link text on a page - using the post title didn't guarantee this. Unfortunately the links, although now satisfying the link phrase requirements, were ugly to look at so I used CSS to hide those bits for visual readers - which gets me past the automatic accessibility checks it seems, but I'm not really sure how correct that really is. The separating adjacent links thing only applied to one link I have that's hidden normally - it's there for non CSS browsers to skip to the navigation.

Unfortunately, In doing the accessibility changes I decided it was high time I started using a revision control system when modifying my web templates, etc, and so decided to learn and figure out Subversion. That took me a bit of time, as things do nowadays, which is why it's taken until now for this post

So my site (or most of it) now passes Priority 1, 2 and 2 automatic checkpoints. There are some warnings on the manual checkpoints, but I figure I'm off to a good enough start for now.

Oh, still to do are defining some access keys, although I've started with them; The basics are there: Fresh, Stale, Query. Although I can only do what I can in the confines of Nanoblogger - I've already tweaked bits of the code, I don't want to do too much more because maintaining it gets tricky then.