Web Designer, Heal Thy Self | January 24, 2004

I'm really glad so many of you enjoyed my article on web accessibility. A few interesting points were raised in the comments that I think are worth expanding on.

Isofarro is correct that this site has a few accessibility issues of it's own. The text in the header is definitely a problem (as there is no text equivalent) and one I'll add to my growing todo list. I also agree that the main nav poses accessibility issues. The font I've chosen is a very small pixel font which a few people have mentioned is difficult to read. Also, because the nav items are rendered as images, you cannot change their size. However, unlike the first problem which was an oversight, this was actually a conscious decision.

This site was always intended to be a personal site. I never imagined it would get popular and the site was built with a limited audience in mind. Also, as it was a personal site, I wanted to have a bit of fun with the design and do things I wouldn't normally do on a commercial project. I'd previously use pixel fonts on my photo gallery (definitely not an accessible site) and wanted to carry the branding through to this site. At the time I did consider the accessibility issue, but decided in this instance, the aesthetic issue was more important. This is something many accessibility advocates would find hard to justify.

I believe you need a solid understanding of accessibility issues to build great websites. However I also believe that you need to balance them against other aspects such as usability, standards compliance, aesthetics etc. I'm not suggesting that you should ignore one for the other. The importance thing is to understand the issues and make conscious decisions based on all the contributing factors. At it's core, this is the essence of design. Understanding the problem at hand, all the contributing factors, and crafting something that's an elegant balance.

Accessibility, like usability and web standards, isn't a stick you use to beat people with, and it's also not an absolute. A site isn't either accessible or not (although some sites are pretty bad), it's a question of degree. When building commercial sites, I tend to aim for somewhere between AA and AAA compliance. In the case of this site, the lack of a textural equivalent for the header text means technically it wouldn't even get an A rating. However, apart from that minor oversight, this site stands up pretty well.

Building to a site to a set of accessibility guidelines is an indication of intent. It says that you have an understanding about the issues and you're doing something positive about them. These days, for most commercial sites I build, I suggest people have an accessibility statement outlining their policy and the guidelines they are aiming for (something that's also on the todo list for this site). The important thing to note is the word "intent". Very few sites get built that reach the guidelines 100% on each page, and fewer stay that way once they are live. An accessibility statement (and bobby rating) says "this is the level we're aiming for, but if you encounter any problems, let us know and we'll endeavour to fix them".

This brings me neatly on to my next topic, as the same holds true when building a site to web standards. I'm sure some people put up the little xhtml/css web standards buttons to say "ha ha, look at me, I'm totally compliant". However for most people it's simply an indication of intent. It says that you are aware of the issues and will aim to have your site validate to a set of standards. This is something I do on this site.

On the left hand side of my index page, under my blog roll and RSS feed, you'll see a couple of little validation buttons. Clicking on these links will take you to the W3C validation page which will attempt to validate this site. If you try it now, you'll see that the site doesn't currently validate. The site did validate when it was first built, and for a good few months after that. However over time a few errors have crept in. I added a couple of geo meta tags which I copied and pasted. In doing so I forgot that I needed to close the tags and this has caused a few errors. The rest of the errors are caused by character encoding, often from copying and pasting url's or writing posts in a word processor.

So does having these validation images/links make me a fraud. Am I trying to pull the wool over peoples eyes to make them think this site validates when it doesn't? Well, apart from the fact that there would be no benefit in me doing this, I'd be pretty dumb to add a validation link and not expect people to click it. The links are there as a demonstration of intent, to show transparency and hopefully, if errors creep in, people will be kind enough to let me know.

If you've read some of the article on this site, you'll realise that I have a strong interest in things such as usability, accessibility, design, process and web standards. Not being a single issue kind of a guy, I feel that all these competing pressures are of equal importance. Web design doesn't take place in a vacuum and requires a holistic approach to get right. There are a number of people who feel it's all or nothing. Either your site is 100% compliant or you're part of the problem. However I believe the right approach is to offer incentives to encourage people to build better websites, not beat them up if they don't as far as you'd like.

Posted at January 24, 2004 8:00 PM

Comments

Keith said on January 24, 2004 11:42 PM

I’m sure you can imagine I’m in complete agreement with you and I understand exactly where you’re coming from. Very well said!

Colly said on January 25, 2004 12:53 AM

Andy, my first comment here…

I have to agree with your flexible approach to both accessibility and standards, and I’m encouraged to read that others build with both sides of the brain.

I feel you simply cannot mess when building an expensive site for a client - you can’t promise full, uncompromised access for everyone on every device, and as you say, place the validation links up front, without being wholly committed to delivering, to trawling the site through the mucky hands of usability groups, and knowing yourself that the framework IS there, behind the glamorous facade that wins the brief and eventually gets the cheque signed.

Where it gets interesting, and we can begin to burn the bottom of the envelope, is within sites such as your blog, and the personal sites of many designers. On my personal arts site (well, personal in the way that Skillswap probably is to you), I suggest to the viewer that our design agency is committed to standards and accessibility, and invite users to validate the site. But I know it often doesn’t due to character encoding, umpteen link styles or my latest grand CSS experiment happening live.

So, I think we’re all for a little leniance outside of client briefs, as you suggest. It does no great harm to preach and fail all at once. Half the visitors to the arts site are web developers always keen to find our latest faults, and suggest workarounds and alternative approaches, thus keeping the site fresh, and flexible.

Without personal sites within which to push the ideas that visit us in the dead of night, I doubt any of us would have forced the time to push things the way we collectively are.

Joel said on January 25, 2004 9:08 AM

Someone once said: “I don’t practice what I preach because I’m not the sort of person I’m preaching to.”

I don’t believe you need to make excuses. Even Zeldman uses the excuse “it’s a personal site”. What’s the point?

If you believe things should be done a certain way, and then you don’t do them that way, people will criticise you for it. But what does it matter, after all, “it’s a personal site”.

This points out certain things. Why bother having validation buttons? Many sites validate that can’t be arsed to show off about it. That yours shows off about it and doesn’t validate is just amusing. If you didn’t have these buttons to start with, you wouldn’t have to worry about your site not validating, save from the point of view of professional pride. But at least you wouldn’t be lying about it. Perhaps you should have taken them off the moment you realised your site didn’t validate. Why make excuses?

I wouldn’t say you were a fraud, but you’re certainly making excuses. If the site doesn’t validate, take the buttons off or make it validate. Is this hard?

Matt said on January 25, 2004 10:50 AM

I think of this in a very practical way. If a client pays me to build them an accessible website then that’s my priority. If I then want to sell someone else an accessible website I know I’m not going to get very far if my own web design business website is not accessible. If I build accessible websites at work, then I would be doing myself a favour if I built accessible websites when I play, just to keep my hand in.

Andy Budd said on January 25, 2004 11:36 AM

Hi Joel, I do believe things should be done a certain way. However, whereas some people have a very dogmatic, black and white view of the world, I understand that good web design is a process, not an end point.

Accessibility and web standards aren’t all or nothing. It’s not about getting things right 100% of the time or simply not bothering. It’s about intent and good will.

If you read my posts about web standards and then validate my page, does the fact that my links have character encoding issues make my views on semantic markup and universal design any less valid?

I’ve been involved in enough mailing lists and web discussions to understand that people hate zealots. You can’t change peoples view by ramming strict doctrines down their throats, in the real world you need to be a little more flexible.

On the topic of making excuses, I’m sorry that you felt that’s what I was doing. What I was actually trying to do was put some of the decisions I made in context and explain the reasoning behind using the validation buttons. I personally thought I did a reasonable job, but obviously not.

So to quickly recap, on the subject of my nav bar, I wasn’t making the excuse that this was a personal site and the rules didn’t apply. I’ve built a site that in most cases is pretty accessible. However using my knowledge and understanding of accessibility issues, I made a judgement call on a single issue. There are people who would dogmatically follow the letter of the law, but I feel it’s important to be flexible and have the ability to make decisions based on more than one criteria.

On the subject of the XHTML/CSS buttons, I thought I’d explained that part of their reason was to help me keep the site valid. Until yesterday I didn’t realise the pages weren’t validating. The reason I found out was because somebody mentioned it in my comments. I would have preferred if they had emailed me to let me know, but they seemed to get so much joy pointing this out publicly, who am I to stop their fun?

So no, making the page validate isn’t that hard. Although I hope you’ll give me the benefit of a couple of days grace period in order to change &’s to &’s. In the meantime, I’m glad that you find so much humour in my XHYML/CSS buttons. Little things and all that.

Joel said on January 25, 2004 12:08 PM

I think anyone should be allowed a couple of day’s grace to validate their site if that is their intention Andy. The way I read what you wrote was that you were reinterpreting the whole notion of having such buttons as an indication of attitude towards web standards, that they didn’t have to indicate that the page actually validated. Glad to hear that’s not actually the case.

That said, I myself validate all my pages whenever I make a change and don’t leave little ampersands unencoded for longer than a few minutes. Nor do I rely on others to tell me my pages don’t validate. I’ve never left a non-validating page online, yet I myself don’t bother with validation buttons nor do I greatly evangelise on the topic. But I do think those who do bother with such buttons have an even greater responsibility to live up to what they signify. Don’t you Andy?

That said, mountain out of a molehill like most things. No, I don’t go around testing anyone’s validation buttons for a cheap thrill and a sense of superiority, usually looking at their layout is good enough for that.

Little things and all that. Agreed.

Andy Budd said on January 25, 2004 1:06 PM

Well, thanks very much for your input.

I’m glad that you’ve never left a non valid page online. You must be very proud. Infallibility is a rare thing.

Joel said on January 25, 2004 1:18 PM

Yeah, actually, fairly proud. Shouldn’t I be?

To me “web standards” is as much living up to my own standards as it is the standards of the W3C.

You can take the piss and belittle that, but this is only because you know I’m right Andy. But really there’s no need to be sarcastic, because you’d probably enjoy my company in the pub.

Andy Budd said on January 25, 2004 5:23 PM

Too me, web standards are a means to an end. They are not an end unto themselves. The reason I’m interested in web standards is their ability to make the job of web design easier, and make the web more accessible.

However CSS and web standard design has developed a bit of a bad reputation, partially due to the zealous and dogmatic approach of some of it’s supporters. I agree that standards are important, but they are only one part of a much more complicated mix.

A sure way of putting people off web standards is to adopt an all or nothing approach. So many times I’ve seen CSS novices post their sites for peer review, only to be shot down in flames because of a small and usually insignificant validation error. This is a sure way to put people off.

I think Zeldman’s approach in his “Designing with web standards book” is a good one. Rather than choose the “all or nothing” route, he suggests a middle way. A concept most people with an interest in eastern culture would be familiar with. So what if some of the examples in his book contain validation errors. It’s the intent that’s important.

There will always be those who take a very rigid view of web standards. I just don’t think it’s a particularly helpful approach to adopt.

Joel said on January 25, 2004 6:25 PM

I’ve never seen so much squirming over not living up to what one advocates.

Congratulations on fixing your ampersands and metatags. Wasn’t hard was it.

Andy Budd said on January 25, 2004 9:00 PM

Joel, this is starting to get rather dull. I enjoy having lively discussions about web design issues but this is going nowhere.

So let’s just call it a day shall we?

Seth said on January 26, 2004 1:49 AM

Andy,

You say that some people put up those “little xhtml/css web standards buttons “to flaunt the fact that their pages are valid (as if they shouldn’t be proud of it).

I’d hope so, as that is there purpose. They are not there to say “hey, I’m trying here.” I’m pretty sure that checkmark on the button is saying “Done,” not “Almost Done.”

From the w3c validator (and only once your page has actually validated is it even displayed):

“To show your readers that you have taken the care to create an interoperable Web page, you may display this icon on any page that validates.”

I sure hope that my bottle of aspririn that says it has certified isn’t just in the process of TRYING to get certified, but actually has been.

I hope that my automobile that says it has a certain crash rating isn’t just trying to get that crash rating, but actually has it.

No one will get killed by your web site, but the principle is the same. The organization that does the certifying has laid out certain rules and it is expected that people will follow them.

I’m not saying that practicing web standards is an all or nothing thing. But, I think the little buttons are.

Mike Stenhouse said on January 26, 2004 2:11 AM

Oooh, I thought that was going to get a little nasty for a minute there… I hate to do this but I think I actually agree with both Andy and Joel. Sitting squarely on the fence. In an ideal world we’d all have the time to make every page of our web sites validate. With CSS this is dead easy - it’s fairly static, but HTML changes so often. For most of us, personal web sites take second place to, well, pretty much everything else. Would you rather have Andy writing new content or fixing niggling validation issues? Ideally both, but the poor chap has to eat, sleep and get on with life!

Back to Andy’s original point… I think that the intent to have a valid and accessible site is enough, as long as it’s sincere. By that I mean: make design decisions with XHTML and WAI in mind and if you have to compromise either, make sure you know what you’re doing and why you’re doing it. How many commercial projects will allow you the time to figure out a compliant workaround to an obscure CSS bug when there’s a reasonable old-style HTML solution? Not many. Would your client rather sacrifice potential audience and use Flash Satay or play it safe and make sure everyone can get the fancy animation? Probably not. They may have the best of intentions but if AAA accessibility costs 200% more than AA, there just isn’t sufficient ROI. As developers/designers we just have to soldier on, doing our best to do things ‘right’.

Aleksandar Vacić said on January 27, 2004 3:28 PM

I can only agree on the comment about character encoding. My blog is iso-8859-1, but it seems that MT is not converting all as it should.

On Jan 7th (orthodox Christmass) I posted a painting from artist named Nebojša Đuranović. Well, now go and validate this entry’s page and you’ll see that character “š” is invalidating the page.