Web Standards – Why do so many smart people just not get it?

John Dowdell of Macromedia has discussed this post at si-blog on use of web standards. JD doesn’t seem to get it. JD, what’s not to understand? Simon’s talking about adopting a process whereby rendered HTML:

  • meets W3C standards for markup (preferably a flavor of XHTML)
  • uses CSS to separate content from appearance
  • meets WAI standards for accessibility
  • meets accepted levels of usability

Maybe Simon hasn’t made himself clear enough, but I certainly got what he was saying. The known, accepted and proven benefits of web standards compliance are manifold, and I don’t need to go into them here. Suffice it to say that in my crew (who are all techie CF coders), generating non-standards compliant HTML just doesn’t cut it.

Nor does it cut it for the designers we contract in to build web site skins for us. We have a very strict set of minimum skin design guidelines we impose upon them:

  • all code to validate as XHTML 1.0 Transitional or XHTML 1.0 Strict
  • all layout and design code to reside in CSS only, with CSS to validate
  • CSS’ to be prepared for screen and print
  • all designs to meet WAI Web Content Accessibility Guidelines 1.0 Priority 2 Checklist points

We impose the same rules upon ourselves, including using Sean Corfield’s ColdFusion MX 6.1 Coding Standards. It makes our development lives immensely easier when everybody’s working to the same set of rules, those rules make practical sense and the quality of our end product is higher.

To my mind web standards compliance is really all about three things – quality, consistency and cross-platform usability.

10 Replies to “Web Standards – Why do so many smart people just not get it?”

  1. So, when the other person said “supports web standards”, you read that as “build software which implements every document the W3C writes, as soon as it writes it”? Or…?

    (Sometimes I get the sense there’s a required philosophical stance to “support web standards” which trumps any actual thing to do, particularly in this case where the writer was talking about some underdefined workgroup rather than specifics such as a business site, intranet site, browser maker, etc.)

    It’s good that you laid out four functional requirements for your group to meet that label. This is more specific and inarguable than the unsubstantiated advantages of what I recall of the original article. Agreed…?

  2. Not at all.
    Here’s where I operate from… My job as a Project Manager/Team Leader in a large federal government department mean that all our developments must meet some extremely strict requirements for accessibility in particular. Much like the s508 requirements in the US. Meeting those needs while building sites that creative Marketing types want to “amaze” their audience with can be pretty onerous.
    I have found that the best way to meet those requirements is to stick to strongly standards-based development. Well structured HTML, CSS, good XML-like CF code, use of Mach-II and meeting or exceeding WAI Priority 2 on all rendered pages means that we avoid problems with exclusion of site audience members post-implementation. Plus, we have readable, tight, reusable code in a lot of instances.
    This is good. For example, on the issue of accessibility, failing to meet those standards cost the Sydney Olympics AU$50K/day of the games ($? maybe more…). They were fined for failing to achieve an accessibility standard required by Australian law for any and all web sites. It would only take someone on an accessibility crusade to create havoc in the Australian web community, ‘cos I guarantee greater than 99 per cent of the sites out there just don’t comply. Take a look at http://www.academyclub.com.au/ for example. Great site done by some guys I know fairly well, but not a hope of meeting any accessibility requirement. I often think it’d be an interesting exercise to get a group of users needing accessible websites together and motivated to write a few letters of complaint just to see the reaction…

  3. Hmm, so, do you actually do a functional defintion of your requirements, or just leave it at that top-level label “must support ‘web standards'” and hope that their meaning for that phrase aligns with yours?

    (I’m getting the feeling that this is, at root, one of those “why, *everyone* agrees on what X is” issues….)

  4. Absolutely. Functional definition of the requirements is a critical and early part of every project we do. Our clients come to us and say “we want a site that does blah”. We prepare a functional spec with them, laying down explictily in the spec that one of the requirements of any technical solution is meeting all points of our strict standards compliance checklist.
    As a government department, we can’t afford the luxury of skirting around these needs. We have legislative- and policy-based frameworks at a federal government level (as I understand, does the US) which potentially impose heavy penalites upon us for failure to comply.
    What it has all boiled down to over the past few years is that we have developed a strong skill base in web standards-based development and our sites get progressively easier to maintain; both at the HTML and CF level.

  5. I don’t get it. And I’d rather think that’s not because I’m not smart, so I liked your title (although you clearly meant JD).

    Might it be that supporting ‘Web Standards’ is not really the standard practice?

    if the ‘benefits of web standards compliance’ are ‘known, accepted and proven’, they should have been the de facto standards by now.

    Well, I guess if I were to design under imposed ‘heavy penalites for failure to comply’, I’d also go by the ‘Web Standards’.

    It seems to me more ‘Web Best Practices’ than ‘Web Standards’, with some benefits but also with a steep learning curve and hassle, clearly not suitable for everyone.

    And as coming from Flash background, I’d say that Flash is the de facto animation (and maybe RIA) web standard today, not SVG, and it works. I don’t see anything wrong with it. Does Flash fit anywhere in the ‘Web Standards’? Does ‘supporting Web Standards’ mean no Flash? (That’s a sincere question).

    Best regards,

  6. Burak,
    Web Standards now have enough history, and I think enough people know what the W3C is and what it does that the term and what it generally means is clear.
    Certainly, compliance with web standards isn’t yet the standard practice. However, I (and others) think that’s because too many people are still locked in the “just get it built” mindset. Frankly, I’d still develop according to standards, even if there wasn’t a requirement for me to do so. Doing so makes my job easier, not harder. The learning curve isn’t so steep, and declaring it hard is misleading, as it isn’t. It does however come with a cost in time and learning – but what’s so bad about learning a new skill?
    There are some great, and cheap, tools out there to help you. I’d be lost without HTML Validator and the Web Developer extension for Firefox.
    As for Flash, nothing wrong with it when implemented properly. We use Flash on several of our sites. Recent Flash tools allow you to code for accessibility (e.g. screen readers) and we have even implemented Flash with standards compliant HTML code. Here are some HOWTOs:

    Flash Satay at A List Apart

    Hope this was helpful!

  7. “Functional definition of the requirements is a critical and early part of every project we do.”

    Cool. That’s what I was seeking in the original article… there were statements like “‘Web standards’ will make things cheaper and faster” but no functional definition of the label itself.

  8. The problem that I inevitably run up against with Web Standards is that fully 50%+ of my users still use Netscape 4.79. The remainder use Internet Explorer 5.0. These are the corporate standards for us. Thus, learning web standards is actually detrimental to my career since XHTML compliant pages as often as not simpley fail to render at all in 4.79.

  9. Adam
    Being stuck in NS4/IE5-land is definitely going to make things difficult for you. These browsers, as you state, are at best problematic when dealing with standards-based code. Is there a valid, stated reason for using browsers so old?
    I fear those responsible for making the decision about browser platforms where you work are well behind the times, and thus well behind the 8-ball here. If you (or anyone you know) have any influence, I suggest you read this presentation and this site. Hopefully you’ll have a hand in convincing someone to make a progressive decision.
    I would also be encouraging management to take a look at the browser profile of their visitors. They would be well-advised to adopt one of the top five as a more up-to-date standard.

  10. Since all of our development is done for an intranet rather than a public internet site the developers are bound to developing for what our users are familiar with. Most of our users have been using Netscape 4.7 for 5+ years now, and have no interest at all in changing.

    Even before joining this company where 4.7 was a corporate standard, I worked for a web company where 90% of our hits were I.E. 5.5, 5% were on I.E. 6.0, 1% were Mozilla, and the rest were divided up among random other browsers.

    The point is that the installed user base is still not completely in support of web standards, and until it is, publishing only on web standards could be like shooting yourself in the foot.

Leave a Reply