Over the past year I’ve struggled to get the Baptcare website to validate when tested against http://validator.w3.org/. My problem was the complexity of the IBM WebCM content management system that powers the Baptcare website and figuring out how to understand and edit its presentation templates, menus and navigation files.
But today I finally got the www.baptcare.org.au website to validate – and it’s only taken a whole year, hurrah!
If you run a website, why not go to http://validator.w3.org/ and test its pages (not just the home page) for HTML errors? If you find any, instruct your web designer to fix them, although if you didn’t originally insist on standards compliance they might well refuse. If you’re commissioning a new website, insist on 100% validation in the contract, and be clear that you’ll withhold payment until it’s achieved.
What types of errors needed fixing on the Baptcare site? Here are a few of the most common ones:
- An XHTML DOCTYPE was changed to a transitional HTML and that solved about 75% of the errors.
- Some tags ended in /> and some in just >. I changed them all to > to match the DOCTYPE.
- Tags were nested incorrectly. For example, there were dozens of <a> anchor tags that contained <h3> heading tags, like this <a><h3></h3></a>. It should have been the other way round.
- There were plenty of lonely, forgotten </p> tags sitting there on their own with no preceding <p> tag.
- Lots of ampersands & in URL links had to be changed to & in order to validate.
So do all Baptcare’s pages now validate 100%? Well, let’s be honest, no. There’s still one recurring error I haven’t got rid of yet, an empty <ul></ul> that the CMS automatically sticks in some pages and not others. That’s technically a validation error because an unordered list should contain one or more <li> list items, but let’s be reasonable… it’s not the end of the world!
And I’m ignoring the Locations page which currently has over 500 errors (originally it was 1,400 which is a quite impressive figure!) and which I intend to completely replace soon with a new, standards-compliant version incorporating the Google Maps API.
No related posts.
Congratulations! I went through that trauma myself at the start of this year. The Philanthropy Aus website (philanthropy.org.au) is all coded from scratch, so it was a bit easier for me to go in and fix things that weren’t validating, but still an epic task!
I found it really hard to know where to start with W3C too – also because I didn’t just want to validate the code, but make it more accessible on a practical level. Knowing that govt websites have their own enforced standards of accessibility, I dug around a bit and found the Victorian Government’s Accessibility Toolkit (http://www.egov.vic.gov.au/index.php?env=-innews/detail:m1496-1-1-8-s-0:n-319-1-0–) which was a great place to start for the kind of organisation we are, and much more accessible itself than the W3C homepage.
The Vic Govt’s toolkit of course linked back to the W3C’s Web Content Accessibility Guidelines (http://www.w3.org/TR/WCAG10/), which I found provided a lot more practical advice/guidance than I could have found on my own.
Overall I found that the whole process, though lengthy, was so invaluable in giving me these building-blocks of design that I’d overlooked by picking things up as I went along for many years (such as the “degrading gracefully” principles that I believe I’ve commented about before). It was very satisfying to do
Anyway! I think the only parts that don’t validate now on the Phil Aus site are the pages with embedded widgets, like Google Maps and Feedburner email subscriptions. It’s amazing how much mess goes into the code of CMS sites.
Emily
on September 5th, 2008
Emily,
Surprising how fun HTML coding can be isn’t it? Very satisfying when you figure it out.
Good point about widgets. Why can’t Google and other sites give us widgets that validate without us having to rewrite them?
And the VIC Gov guidelines are useful – but I had to read the lot recently and that wasn’t so fun.
admin
on September 5th, 2008