Standards Are Not Just Stuff and Nonesense
While HTML5’s generals play with toy soldiers, designers and developers who just want the war to be over, get on with the fight by speaking about, writing about, teaching and using HTML5. Andy Clarke
One thing that all the cutting edge designers and boutique shops keep forgetting is that this wonderful thing we know as the ‘web’ (complete with HTML, CSS, JavaScript and all the newly emerging goodies of HTML5) is also used in the most mundane of ways, on web sites that simply need to communicate information to the end user without being a showcase of CSS mastery, Ajaxian trickery or the cutting edge coding examples of <canvas>, <video>, etc., etc.
No, these organizations are themselves large, lumbering beasts, lacking the agility, flexibility and luxury to push the envelope and explore the fringes. The poor folk who work there instead have to rely on and comply to STANDARDS, not specifications that shift like sand in the desert, but codified, nailed down, no-way-but-this-way STANDARDS. Academia, Governance, heck even large Corporate sites all have to meet Standards requirements to conform to a slew of trivial things like laws, shareholders, marketing and branding, and a whole raft of things that talented and exciting boutiques get to challenge. Work for a government agency? Good luck challenging anything my friend. Heck, in some areas of development ‘webmasters’ are union jobs, complete with scheduled morning and afternoon breaks, an hour for lunch, start at 8, finish at 5. I work for one such large organisation, and I know of what I speak – it might not be as bad as some other locations (and I do get to try out cool new stuff), but I’ve also met and commiserated with others even less fortunate than I more than once.
Yes, it’s sad; I hear you muttering, “I’d never do that, I do art”. Reality check #2 – not everyone gets to be Andy Clarke (or Dave Shea, or Dan Cederholm, Tantek Çelik, Wendy Chisholm, Aaron Gustafson, Jeremy Keith, Ethan Marcotte, Eric Meyer, Nicole Sullivan or Jeffrey Zeldman – the Superfriends), nope, instead they have J-O-B Jobs, and mortgages, groceries, two kids in high school and a retirement portfolio that is still too weak to allow retirement at age 55. For them, STANDARDS are the tool-set that they must use, and so Standards are important.
Standards are more than just specifications. Standards are translated into numerous languages, printed with actual paper & ink in books (which are sold in shops, stored in libraries and used as teaching materials in schools), combed over by legal folk, management folk, and other stake holders who might not know a CSS child-element from a JSON call, but they do know what ‘broken’ means, and for them, ‘broken’ can sometimes mean big trouble: law suits, lost business, bad press and on and on. There is no room in a STANDARD for “…fix-it-in-the-mix” mentality, for “…we’ll get to that in the next iteration”. Nope, standards are, by necessity, solid as the rock of Gibraltar, and often just as exciting. Specs? They can be exciting, inventive, invigorating and challenging. Specs alone however cannot be Standards, not until they’ve also traveled down the political road of poking and prodding from all the stakeholders, not just the artisans and technologists.
Believe it or not, the W3C is the best collection of people who understand these issues; they have representation from all those other stakeholders, yet those representatives can also speak to the technology that is driving this all, both the tried and true, but also the new and exciting. Other groups, perhaps looser in structure and smaller in size (I’m pointing at you WHAT WG) might appear to be faster, nimbler, and more ‘with it’, but that’s only because they lack the perspective and insights that the larger organization can contribute. As a vocal accessibility advocate, all I need do is point to <canvas>, and last years Bespin experiment – yes, cool, but yet totally inaccessible, because, guess what, the spec never accounted for accessibility; the drafters either forgot, ignored or at best deferred that important part of the puzzle – making that aspect of HTML5 out-of-bounds for anyone mandated to create accessible content; content creators like Government, Academia and increasingly, Big Business. The expertise inside of the W3C however is now working on this boring but critical part of the specification; the original spec authors long since moving on to other experimental bits like (oh, say) Microdata…
Anyway, to you, the edgy young professional who currently has the dream web job, not the mortgage paying J-O-B : take heart, this will work itself out. Saner voices will win the day, HTML5 will be the next Standard, and in 3 years time this will all be a past nightmare, just like the Browser Wars of the late ’90s. But at the same time remember, the W3C is not stalling anything, but rather they are ensuring that all this cool stuff can be used by everyone: end users, cutting edge boutique developers like Mr. Clarke, and our poor beleaguered civil servant, currently on his prescribed 15 minute break in the cafeteria (or out back in the loading dock, grabbing a quick smoke). And think twice about what STANDARDS really mean.
![[Photo: John Foliot]](http://john.foliot.ca/wp-images/foliot_twitter.jpg)
Hey! I'm John Foliot, and this is my personal blog. 
People born in the Year of the Pig are chivalrous and gallant. Whatever they do, they do with all their strength. For Boar Year people, there is no left or right and there is no retreat. They have tremendous fortitude and great honesty. They don't make many friends but they make them for life, and anyone having a Boar Year friend is fortunate for they are extremely loyal. They don't talk much but have a great thirst for knowledge. They study a great deal and are generally well informed.
January 13th, 2010 at 5:45 am
Note that the “drafters” didn’t design canvas. It was basically reverse-engineered from WebKit’s proprietary implementation.
January 13th, 2010 at 6:38 am
I used to work in a quasi-public sector organisation, which in 2004 paid $howmuch?!? for a tables-n-spacers website. Until 2005 the webteam had to use IE5 to test their work (downloading other browsers was a disciplinary offence).
Therefore, I hear you, mate, I hear you!
January 13th, 2010 at 7:15 am
Interesting perspective on innovation. I’m actually a professor who teaches search engine advertising and landing page design. Our standard is xhtml 1.0 and approaches that work cross-browser.
One remark. You’re obviously speaking from a production perspective in academia. In terms of producing students, they have to know about what’s at the edge. In terms of research, people need to be creating the edge.
January 13th, 2010 at 11:22 am
I bow to thee, oh, wise one.
January 13th, 2010 at 11:57 am
@Ms2ger – whether or not WHAT WG drafted or reverse engineered the canvas element is less the issue: those engineers did not account for requirements like accessibility, as that topic/item/requirement was outside of their thinking and scope. Engineers are but one voice at the table though…
@bruce – right! I was actually thinking of guys like you when I wrote this post; I was also recalling my days as a consultant to the Canadian Federal Government, and work I was involved with for a major Canadian Bank. I too have the scars of frustration.
@Bud Gibson – while I currently work in an academic environment, I’ve also had years of experience outside of Stanford. I’m all for teaching students (especially CS Students) to stretch for the next boundary; those students also must learn that the web is *NOT* just technology, it’s an entire ecosystem that has massive impact on our society. It’s way more than just the inane “cool, HTML5 is gonna kill off Flash…” comments that populate blogs and the twitter-sphere, and the bureaucracy that those pups rail against is a reality that they *MUST* acknowledge and accept. (When I was in my 20’s I thought I had the world by it’s tail too!)
January 13th, 2010 at 12:39 pm
Thanks for the very nice post. I long for the day when cool is no longer cool and we focus again on making sure things work for most, if not all, web users. Standards are important and like you I think the W3C generally do a pretty good job in dealing with these issues, although sometimes the language they use is somewhat impenetrable.
January 13th, 2010 at 1:26 pm
Per our previous conversation, I think there’s a lot of room between your point of view and what Andy wrote on Stuff and Nonsense. I don’t think that Andy is an artist and not right here in the trenches with everyone else. He works with clients, and large organizations, too.
Every project has different parameters, and sometimes we can adopt certain techniques, and sometimes we can’t. I don’t see why there can’t be a variety of offerings in the specs and I definitely do not believe that corporations should have an impact on what is offered, a say, certainly, but not remove things that may provide advancements. I am naive, and I think I will always be. A bit too Pollyanna optimistic, if I do say so myself. We need the best possible tools in our arsenal in order to deliver the best possible deliverables, within schedule and on-budget, that will also stand the test of time and be useful for a decade or so.
I’ve spent 15 years contracting in so many environments — design agencies, corporate publishing and financial institutions, etc. I understand the corporate environment with the proprietary software, but I also don’t think we should keep quiet and allow that incompetency to persist. Many employees have equity in the place they are employed. They are tied to the success of the business, whether it is the retirement funds, the daily job, or the resume attribution. If we don’t speak up and share info, we are not good at our work.
Andy’s role in the web design community is multi-fold. He’s one of us, running his own shop and working with his own clients. He is also an inspirational educator, holding the lantern on possible ways forward — no one needs to use all of his suggestions, but he explores them, shares them, (adds to his income), and inspires others to explore. There’s always a possibility that the techniques can be available for those with the modern browser, but elegantly degrade for other browsers. Certainly it would impact a brand if the degraded version was crap — but that’s our job as designers/developers to know what to do and when.
I’m disappointed to find that corporations are controlling the reins of the W3C and the standards projects. It’s frustrating enough trying to manage all the different proprietary issues out there, for example – video, without having to declare things a multitude of times for each of the different proprietary browser (CSS3 declarations x3 is annoying the piss out of me). And that’s just one thing.
Web standards should be about setting quality as the goal, without all this pandering to “the man.” “The man” should realize it is in his best interest for the code to be clean, and that the value in their product should be in the talent inherent in it, something that cannot be copied. Accessibility, scalability, and the ability to update quickly/easily/cheaply — those should be the benchmarks.
I hope I made some sense. I’m tired and my brain didn’t stop thinking about this last night, so I’m not certain that I’m clear. It’s a bit disappointing because at midnight last night I had all this passion and clarity, and now I’m just wiped.
January 13th, 2010 at 3:41 pm
Reality check #2 – not everyone gets to be Andy Clark [..], nope, instead they have J-O-B Jobs, and mortgages, groceries, two kids in high school and a retirement portfolio that is still too weak to allow retirement at age 55.
You so you know, My earning potential is linked only to the hours I can stay awake, with no guarantee of a salary, sick pay, holiday pay or other benefits. I have a mortgage, bills to pay, an eighteen-year old to pay through university for the next four years and no pension, or retirement plan at all, and no savings.
January 13th, 2010 at 10:52 pm
@Andy – This is not an issue about money, and it was not my intent to imply otherwise. Rather, my intent is to speak out for those of whom I spoke. I have no illusions in my mind that you are in some kind of glass tower, but neither do you commute to a nondescript government building every day to occupy your cube and care for the departmental website. Those people are affected by the machinations of Standards building as well as any of those more fortunate to live the ‘bigger’ dream that you have achieved. I respect your work, and your contributions to the larger community, and I have no beef with you.
But those that are inspired by you, and strive to follow in your footsteps must understand that our industry is not always glamorous, exciting and creative; sometimes you have to use the tool-set for more mundane but equally important work. They should care about standards building, not dismiss them as impediments to progress and creativity – collectively we are the guardians of this open source wonder that is the web. And I believe that this message should be heard as well. As the title of this blog entry notes, Standards are not just Stuff and Nonsense.
January 14th, 2010 at 4:26 am
It is my understanding that accessibility should be one of the considerations in web design, anyway. See the intro chapter of Andy’s Transcending CSS — you two probably have the same beef.
Corp workers mistakenly view non-corp workers as having achieved the “bigger” dream. It’s not like that, in fact. The difficulties of a non-corp career are multifold. It’s not what you may imagine. Boy, oh, boy, it isn’t… (ouch! from me from my recent experiences).
Accessibility should be a part of specs and always a thought in design processes. There are cases where it it is less applicable, though. Creativity is what may allow us to best provide for it, though.
Specifically, in HTML5, what are the beefs?
January 14th, 2010 at 11:21 am
Hi Jen,
I’ve also worked both sides of the street (being an independant consultant for approximately 7 years before landing at Stanford 3.5 years ago), and I’m not suggesting that one side is better than the other. I loved being my own boss, and was doing some significant work consulting to the Canadian Federal Government before moving to California, but I also expeienced the roller-coaster reality of the ‘financial’ part of being self-employed. I’d do it again, but am happy where I am now too.
I say ‘bigger dream’ in the context that Andy has achieved a level of success that many others aspire to; he’s earned it, I do not begrudge him of it. But by the same token, there *is* a difference between my fictional civil servant and “…a small web site design studio, …[that has] built an enviable worldwide reputation” (that BTW is a dream worth working towards)
I do take umbridge with your comment (colored as it were by my specialty and focus) that “There are cases where it (accessibility) is less applicable”. Never! Accessibility must be foundational, and it should be every users right to access your (or your client’s) content equitably – not identically, but certainly equitably! In that I will never back down.
The ‘beef’ (and I am still working on a larger reponse to you outside of this thread) is that there is currently a split that is growing wider every day between the vision and implementation of HTML5, and who, actually, decides what is the Standard, and what is being proposed as part of the Standard. There are currently 2 versions of the HTML5 specification – versions that differ in many ways (and both, it should be noted, not yet codified as being ’stable’, although aspects of both are supported in the majority of the browsers). Which one should you (will you) use?
And herein is my overall complaint with Andy’s blog post. While designers and developers today should keep abreast of emergent technologies, they should also be focused on the Standards process, not dismiss it as “…generals play(ing) with toy soldiers…” and advocating “…using HTML5, CSS3 or indeed any other tool that makes ours, our clients and lives of people who use the web sites and applications that we make better.” Which version of HTML5? W3C’s or WHAT WG’s? If you use <video> how will you provide captions? If you use <canvas> how will non-sighted users interact with your content? Rushing headlong into HTML5 because industry thought leaders are suggesting that it’s OK, and to ignore the Standards process in favor of immeadiate results and solutions is short-sighted, fool-hardy and ultimately damaging – not to mention it might also make your work of this month completely obsolete by December. Are you really willing to take that risk?
If you are prepared to see your work as temporal as a Sand Mandala, then by all means explore the richness that HTML5 is offering. However if what you want is content “…that will also stand the test of time and be useful for a decade or so” then tread lightly and with a serious ‘heads-up’ approach. And if you want both worlds, be concerned and involved in the Standards creation process – believe it or not it’s more than just generals and toy soldiers. It’s people like you and me too.
January 14th, 2010 at 10:19 pm
“I do take umbridge with your comment (colored as it were by my specialty and focus) that “There are cases where it (accessibility) is less applicable”. Never! Accessibility must be foundational, and it should be every users right to access your (or your client’s) content equitably…”
I have a specific case, a painter’s portfolio. There’s a point at which the accessibility is impacted by the subject matter of the site. That’s what I referred to. Titles and alt tags and tabindices, etc, aside… nothing would deliver the paintings to absolutely everyone.
January 15th, 2010 at 4:00 pm
Hi, John, from the land of beleaguered government web workers-
Yes, it’s as you say: we need nice, stable, open standards for any work we do. We are beholden to our taxpayers to ensure that our output is available to all, regardless of what browser they use and whether or not they also use screen readers or magnifiers or speech synthesis. We also need to do our work so that it doesn’t have to be redone all the time to use the Stupid Web Trick du jour, or to recover from having done so before. We have about 350,000 pieces of content (not including navigation pages!) on Mass.gov – whimsy is not very attractive.
But there’s more. We don’t use HTML just for “websites”. HTML is becoming the standard interface for our mission critical applications. Despite efforts to keep the presentation separate from the business logic, there is always some markup that gets muddled in with the business logic, where it is risky to make changes.
From an overall IT perspective, web pages are just another source of data, so we are interested in metadata standards that will work with them as well as with more structured applications.
Indeed, the impression that I get is that HTML5 is being crafted with the bespoke, stand-alone website in mind, and perhaps won’t speak to the other, more utilitarian uses of HTML.
If it doesn’t meet legal and practical needs, it’s unlikely to be adopted as a government standard, and what good is a specification that governments and large corporations can’t use?
April 6th, 2010 at 7:54 am
Standards are not just stuff and nonesense. You’re absolutely right by that. But where would the web be without widely used specifications that don’t meet all requirements that make it a standard.
Examples are:
* HTTP (http://tools.ietf.org/html/rfc2817 – status: proposed standard)
* URI (http://tools.ietf.org/html/rfc4395 – status: best current practice)
* CSS 2.1 (http://www.w3.org/TR/CSS2/ – status: candidate recommendation)
Another example is WAI-ARIA. This is a very important specification to ensure accessibility of rich internet applications. The now available version of specification is already implemented in the major browsers. But no HTML specification is available in which the attributes are defined that WAI-ARIA relies upon. That makes it rather complicated if you care about accessibility and be standards-compliant at the same time. The simple fact that you use the aria-required attribute on this page proves my point…
I’m with you that standards are really important. Sounds logical, since I work for the national government, have a J-O-B job, and mortgages, groceries, four kids ranging from 6 to 24 and a retirement portfolio that is still too weak to allow retirement at age 55.
But your story doesn’t convince me.
Why?
1. Because today’s web wouldn’t even exist when only standards were allowed to be used;
2. And because most claims for standards compliance are a farce. Most web pages don’t even validate against HTML specifications that are declared in the document itself, are open standards and haven’t changed for over a decade.
When it comes to web accessibility, the situation is even far worse. Since October 2008, I’ve examined hundreds of websites for Self Declarations of Conformity (SDoC’s) in relation with the Web Content Accessibility Guidelines. More precise: WCAG 1.0, priority 1 checkpoints; this is the bare minimum for web accessibility. In all cases, the claim could not be confirmed. Only claims proved to be trustworthy that were based on a test carried out by an independent third party.
I still believe that ‘the industry’ is able produce valid SDoC’s. But first the majoririty of web companies has to get its act together.
This is also the case for website owners, since they accept claims that are solely based on the _assumption_ that claims are true.
In other words: the current system of checks and balances simply doesn’t work. It’s not a technical problem, but a management problem that managers cannot make technical specialists responsible for.
In my opinion, only using standards sounds logical, but on the web it easily leads to numerous complications for website makers. Policy makers should be aware of that. Instead of requiring open standards for the web, they could also consider allowing open specifications that are widely implemented (open specification = all requirements of an open standard, except for the requirement that it is _approved_ as an open standard).
This is why I like the approach that thw W3C has taken in WCAG 2.0. Requirement of Success Criteron 4.1.1. is that web pages can be parsed. Page validation is a ’sufficient technique’ (www.w3.org/TR/UNDERSTANDING-WCAG20/ensure-compat-parses.html). This way, useful specifications such as WAI-ARIA can be applied, while maintaining the integrity of the markup language. It uses the property of browsers that elements and attributes are ignored that are not supported.
Accessibility issues (such as the element…) may occur, but those issues can be dealt with in WCAG 2.0 in a similar way as technologies such as Flash, preferably using a ‘layered construction technique’. That’s the term we use for ‘progressive enhancement’. It’s not about creating fallbacks for people with disabilities, but about creating enhancements for browsers that support it and thus create a richer experience.