Elliot Jay Stocks is a designer, speaker, and author. He is the Creative Director of Adobe Typekit, the founder of typography magazine 8 Faces, one half of Viewport Industries, and an electronic musician.

Misunderstanding web standards

Posted on 09 November 2006 8 comments

With every new website project undertaken both at work and at home, I find myself becoming more thorough about web standards, and as a knock-on effect, more frustrated at how poorly some designers are following standards-based practice. But there’s something else as well: Some people are misunderstanding what ‘web standards’ actually means. A conversation I had with a friend this morning highlighted some of the key issues of standards-based web design, as well as the misunderstanding surrounding them. To summarise, the main focus of my response was that said friend had mistakenly thought that I had claimed CSS was the key to accessibility. He cited a BBC programme that studied web accessibility and stated that CSS hadn’t come out on top.

My response:

Yes, it’s absolutely right that they should not mention CSS – it plays no part whatsoever in a site’s accessibility.

The prime thing above everything else you can do to make a site accessible is to have clean, semantic markup: structuring a document with proper hierarchy (using the appropriate h1s, h2s, lis, etc.), regardless of where elements will physically appear on the screen once the styling is introduced; giving semantically meaningful IDs and classes; and (something I’m really into at the moment) using things like dl / dt / dd elements where appropriate, instead of plain text with span wraps or – even worse – tables. This is something very few people do.

The upside of doing all this stuff, aside from it being the most important part of site accessibility hands down, is that it’s also the most important thing you can do to make a site Search Engine Optimised. So two birds with one stone!

But CSS plays no part in this whatsoever. CSS simply allows us the freedom we never used to have to write markup that is completely devoid of presentational elements. So the BBC were right.

However, I would argue that it’s unlikely someone is going to have an understanding of all of the above if they have no understanding of CSS, as they are effectively still in the Dark Ages world of font tags, tables, and markup littered with presentation-related code.

If there is a myth out there about CSS ‘future-proofing’ a site, I’m certainly not a believer of it: CSS demonstrates how changeable sites can be (CSS Zen Garden being the perfect example), so in fact the opposite is true: if there’s any ‘future-proofing’ to be done, it’s in the mark-up, not the style.


  1. Matthew


    25 April 2007 @ 12:18AM #

    I like what you are saying, but I find myself slightly cringing at some of the claims you’re making. A large part of accessibility (as noted in the w3c reccommendations) is having the content, structure and presentation layers separate. So, if you want your site to have any sort of presentaion, one could argue that it must have a stylesheet mate. Even though it’s an IF, it does raise the question if CSS has an important role in accessibility. So I don’t know if I totally agree with your statement that CSS plays NO part in accessibility. I do agree that semantic markup is the key here, but CSS does have it’s part. Ask yourself this, can this site separate its presentation, content and structural layers without CSS? Accessibility doesnt referr only to those with screen readers. I know there are a lot of ways to look at this, I just want to keep you thinking.

  2. Elliot Jay Stocks

    Elliot Jay Stocks

    26 April 2007 @ 12:54AM #

    Matthew, I know what you’re trying to say, and I appreciate you digging deeper into the debate; however, dig even further and you’ll see that CSS doesn’t have a role to play in accessibility.

    Why am I saying that again after what you’ve said?

    Because your scenario assumes that there has to be a CSS file. If you want an accessible website and you want it to have a design, yes, I agree that you’ll need a CSS file in order to separate the content from the presentation. But a website doesn’t require a CSS file to be accessible – it only requires a CSS file to be designed.

    I think this deeper crux of the matter is what I was getting at; I hope you can see why I stand by my argument.

  3. Erika


    16 May 2007 @ 11:26PM #

    I have to say, I am still one of those designers who has a hard time really truly understanding what “accessibility” means and an even harder time with CSS.

    I have taken the time to learn how to create a site with it, but find it frustrating that a) I have to jump through hoops to get it to play nice-nice with IE 6 AND 7 and, b) that a majority of my visitors are still using IE 6 or 7. Actually, last I checked it Firefox was 4th in browsers that viewed my site!

    I go to all these (what I assume are) beautiful websites built using CSS and they are all funky in my IE 7 browser.

    I am totally against forcing a user to download something to what may or may not be their computer in order to view my site properly. I think that’s just pretentious and not very smart.


  4. Elliot Jay Stocks

    Elliot Jay Stocks

    17 May 2007 @ 12:43AM #

    Erika, you’re completely right to be against anyone who forces a user to download a piece of software in order to view a website. However, I think you’ll find that the vast majority of desigers working with Web Standards do not adopt this approach. It flies in the face of accessibility.

    Many designers (like me) have simply set up procedures to ensure sites degrade gracefully in older (and less capable) browsers. It doesn’t mean they won’t work in IE; it simply means that users with more capable browsers will get a few extra treats. I’ve talked about this in quite a bit of detail here and here.

    I would urge you to persist with CSS-based layouts and building with Web Standards. It’s a tough hill to climb, but the benefits far outway the reasons against working that way. I read your post about the subject and I remember feeling the same way as you just before it all clicked into place. Although you may feel that CSS-based sites are excluding users, it’s far more true with table-based layouts and non Standards-compliant markup, which have very major accessibility problems.

  5. Matthew


    18 May 2007 @ 02:14AM #

    Elliot, let me quote myself really quick here.

    “So, if you want your site to have any sort of presentaion, one could argue that it must have a stylesheet mate.”

    I did put that if in there. Because you’re right, if you want a site to have no styling, it can be accessible and semantically correct without a stylesheet. But as far as I could tell , you were referring to real world examples where styling is part of the site. Because let’s face it, no one wants a site without styling. I never said a stylesheet is required, but I still disagree to say it has no part. In real world sites, there is styling. I consider that a role fulfilled by CSS.

    I think you would be safer saying that CSS is not required to have a semantic and accessible site. But once you add a design(like most clients will want to do), you NEED a stylesheet to keep it semantic and accessible. That, as far as I know, qualifies as a role in keeping things up to standards.

    Can you see what I’m saying here? CSS doesnt make things accessible, it helps to keep things accessible by giving us a way to separate the content and presentation. That is it’s role. But it is not required in a site.

    As a last thought I just want to say, I think we are somewhat agreeing here. My initial post was defending that CSS has a role to play. We both agree it is not required. But in a real world site, it has a role. And I think if you read my first post you will see that this was my stance. CSS has it’s part, but is not required.

  6. Erika


    18 May 2007 @ 08:18PM #

    Thanks Elliot!

    As far as designers who try and force their users to download more “CSS friendly” browsers, I have definitely seen quite a few designers who’ve adopted this approach. Maybe they aren’t the best examples of tolerant designers, but they’re out there. If I have time I will shoot you a few links.

    I’m confused as to how a table website would exclude anyone from viewing a website. I just have yet to see any proof of that although a lot of pro-CSS designers claim that. What exactly does that mean?

    I’ve also gone through some of the CSS site showcased on CSSBeauty in my IE browser to find that many of them don’t “degrade gracefully” as you put it. I can’t click on a menu because the “floats” aren’t configured properly. Whereas, I’ve never had a problem with table-based website.

    Don’t get me wrong, I will continue to work on it. I am just having a hard time really understanding how CSS is better for the end-user and the designer when I spend so much time trying to make it “degrade gracefully” and I’ve never had to do worry about that (or rather rarely) with a table-based site.

    Hit me up privately if you wish! I plan to suck as much knowledge out of you as possible so be prepared!

    - e

  7. Elliot Jay Stocks

    Elliot Jay Stocks

    20 May 2007 @ 11:15PM #

    @ Matthew: Yes, I’m happy to say that I do think we are actually agreeing with each other here! It seems a miscommunication on both our parts managed to disguise that fact. A very enjoyable discussion, nonetheless!

  8. Matthew


    24 May 2007 @ 04:50AM #

    Elliot, it was a pleasure. I shall take more time in the future to get my point accross more clearly.
    @erika – you said “I’m confused as to how a table website would exclude anyone from viewing a website.” Can I ask you something? If you were blind, how would you browse the internet? Answer: you would use an assistive application that would read the screen for you. Now here is the kicker, table based layouts are not only interpretted as big tables of data (instead of something more like a document) but table based layouts often dont have any sort of structure to them that is readable to screen readers.
    Here is a good link for you:
    I think its an old document(from 2000) but it is the basics. And it may qualify for the proof you wanted.

© 2005 – 2014 Elliot Jay Stocks. All rights reserved. Powered by Harmony and tracked by Gaug.es. To keep updated with new content, you might like to subscribe to my RSS feed.