I am a Web Designer

The accessibility laws are vague to say the least at the moment in the UK and although there are general guidelines for creating and publishing websites (and I’m happy to say that I adhere to them), they aren’t necessarily law… yet. Aside from making sure that your service is accessible to all users, there is no specific law that applies to websites. The grey area here is: what exactly constitutes a service. In my view it’s making sure that your business website contains information that can be viewed by all site visitors. You business website provides a service of informing visitors about your business. If you have a strictly personal blog/portofolio site then perhaps these accessibility concerns might be relaxed – but why would you want to exclude visitors from your site even if it’s just personal? I can appreciate the philosophy that because all web content is officially published, then that makes it subject to the same constraints as all other published material and thus, the same accessibility guidelines. This would make even personal website subject to accessibility law. In the UK this subject has been actively analyzed since the tail end of last year and hopefully the latest studies will result in clearer laws. You can read some comments at uk webstandards.org. In the meantime, I shall continue to create websites, personal and business that adhere to generally accepted accessibility principles on the web.

I’ve just read a decent post by Jonathon Snook about his displeasure at using CSS reset stylesheets. A reset stylsheet is a generic ‘cover-all-bases’ css file that strips all css properties that are passed into a web page by the browser. In theory, this creates a clean slate upon which to build your website so that it will be identical in each browser. But I don’t use them and don’t feel the need to do so. When I have used them I found that I spent most of my time ‘re-correcting’ the CSS resets and the eric myer level of resetting I simply found obtrusive and doesn’t fit in with my coding practice at all. In fact I reckon that it takes me more time to un-reset the css in that Eric Myer version than it does to target the specific elements I actually need to reset and simply write a basic reset style for each web site. I am quite comfortable with the subtly different ways that each browser renders its h1 element. Like Snook, I don’t have a problem with people using them, but I’m so glad that someone with a fairy high profile such as his has dared to suggest that they’re not for everyone.

UPDATE:
Following the Jonathan Snook post, Eric Myer posted his very eloquent response.

I occasionally potter down to the odd forum, happen across the odd tutorial and during the course of a typical days work I’ll research new ideas or even old ones from those at the top of my profession. They bang on about semantics in (x)html, about accessibility, about reaching as large an audience as possible without unduly effecting their enjoyment of the webpage. But so often I come across non-semantically named IDs and CLASSES in HTML written by these bastions of web standards. One recurring trend it to construct a multi-column layout with naming conventions such as right-col, middle-col, left-col. Now where’s the semantics in that? What does right-col tell me about the content in that section? Nothing at all – its meaningless. Ok, when designing a webpage its not always easy to know what kind of content is going to be placed in that section – especially when you’ve included a content management system whereby the client can add his/her own content later on. But right-col? This has nothing to do with any kind of content – it relates purely to its visual position in a fully styled webpage. A webpage must make sense without any styling whatsoever and its HTML should be written in the order it is intended to be read. A better naming convention might be simply col1, col2? Better, but this still suggests a styling/presentational concern. More semantic might be text-block1, text-block2 if the anticipated content is largely text. You get the idea.
There is one tutorial which show a method for creating a three column layout in css and I’m highlighting it here because it has worryingly been adopted my many web designers (even the mighty A List Apart has an example that sees the centre column appear ahead of the left column in the HTML) – but not only does it use meaningless naming conventions of rightcol, middlecol and leftcol, but it doesn’t even list them in the right order!! To achieve its three columns, the designer has use a combination of left floats and right floats so that leftcol is on the left, rightcol is on the right and the middlecol… you guessed it, is in the middle. However, when viewed without its styling, you will read the left column first (fine), then you’ll read the right column and then you’ll read the middle column – so it’s totally meaningless and impossible for screen-readers to present affectively or for those with older browsers, or mobile devices who simply don’t have styling turned on. I have written my own three column layout which functions beautifully in all browsers, degrades gracefully and can be read in the order it was written!! Perhaps when I finally organise my tutorial section, I’ll include it there.
So, to conclude. Make sure your HTML element IDs and CLASSES make sense and as best you can, describe the content and NOT its position on the page. While you’re at it – make sure the webpage makes just as much sense without styling as it does with.

Its the annual naked css day today, which is a day set aside for all standards advocates to remove styling from their website in order to show that their sites make just as much sense without the pretty colours and element positioning. Good idea me thinks – but glad its only for a day or the web would become very dull indeed wouldn’t it?

To know more about why styles are disabled on this website visit the

Annual CSS Naked Day website for more information.

Well, I’ve upgraded to 2.5 – and I am impressed with the admin interface – I don’t think I’ll need to supply my own anymore. As for everything else?? Well it’s a blog isn’t it! But I suspect its more efficient and much more secure than the previous version – and far more secure that some of the versions I had been running on various sites. All updated now, so rest easy!

I’ve just done a bit of browsing, as you do – and come across mon.itor.us a free website monitor. Like all uptime monitors it, it checks a url at certain intervals and if it finds a site to be down – it emails and tells you so. I have used similar systems in the past and they work reasonable well, with occasional false negatives. But many of those services are either restrictive (as in freewebmonitor.com: only allows three websites to be checked per account) or unreliable (monastic – many many false negatives). So here’s the new kid on the (my) block, Monitorus.

This a comprehensive looking FREE service with no limits on the number of url’s submitted. It will check port:80, it’ll PING your site. it has facility to check websites externally and internally (I’ve yet to work out what internal checking entails – but glad that it its offered).  All in all it looks promising. But,as they say (who are they? the wizards??), the proof of the pudding is in the eating – so I’ll wait and see over the next couple of months.

http://kilianvalkhof.com/ An interesting experiment to determine just how many CSS classes can be applied to a single element.

While I’m at it, Microsoft have released pretty comprehensive details of their CSS support in all IE browser versions upto IE8 Full release. Little or no CSS3 support it seems, which  is a shame. Perhaps IE9 (shhhhh, I didn’t tell you that’s what it’ll be called ) will show more support in this area.