We’re in trouble with web standards. It’s been many many years since they’ve been updated and the inklings of irrelevance are starting to drip all over. In order for designers and developers to keep using them, it’s time to move things forward.
Developing with web standards is now standard.
With web standards, we’ve come a long way.
As of today, it’s much cheaper and easier to hire a web-standards developer/designer. Code management is easier, and browser-compatibility is a diminishing problem. Frameworks have taken the load off standards-based development and we’re all watching new and interesting standards-based sites appear monthly.
The Web Standards Movement is losing ground
There are two groups that need to hear more outcry about future web-standards: browser-makers and the W3C.
Have you ever tried to do create opacity with CSS in all major browsers? They all allow it, but it’s not a standard. It is a browser-specific CSS rule. I’m going to repeat this, so everyone can digest it: it is a BROWSER-SPECIFIC CSS rule.
This is, to me, a sign that browsers are moving forward and standards are trailing in the wake. Which designer wouldn’t love opacity? It’s in the CSS3 specification, most of which is under “Working Draft”, meaning: Not complete. Yet it is a fundamental design attribute, and would make parts of development much easier.
The W3C, as has been stated by many blogs, is moving slowly. Slow for progress, and slug-like for the internet, stifling innovation. It’s time to put down the education role of current web standards and be more aggressive about moving the web forward. Progress needs to happen soon, or browser vendors will keep adding new rules to do interesting things and we’ll be forced, once again, to write browser-specific code for our clients.
Time for a Web-Forward Movement.
Our voices need to project towards our development environment (the browser) and the code we use (the W3C), and fostering innovation and commonality in that. In order for that to happen, the W3C needs to work more quickly on proposing and implementing. A vocal majority of people dependent on this environment and code can create change, and it has remained stagnant for too long. If we are forced to write browser-specific code again to achieve advancement on the web, then the last 7 years have been for naught.
It is time to move the web forward! So please, won’t you join me in my cause?
Great write-up Kenny! I am in … where do I sign-up?
I can totally share your sentiment that, “This is, to me, a sign that browsers are moving forward and standards are trailing in the wake.”
Thanks @Martin
The best way to sign up is to blog, twitter and write about it. I haven’t committed to making a website for it, but we’ll see if this article generates enough interest and conversation to merit something more. Until then, spread the word, it’s time to be more vocal about this stuff now!
I agree completely, and one thing I think we can do towards moving the web forward is — collectively, as industry leaders — use the hell out of the right browser-specific CSS rules. It’s been proven that if the community gets behind things, they will become standardized, whether they W3C likes it or not (see XMLHttpRequest, for example).
So which proprietary extensions are the right ones? I would say if something meets most of the following criteria then it’s something we should jump all over:
I realize suggesting we use the hell out of properties that aren’t part of any spec is not “the web standards way,” but I truly believe that using things like -webkit-border-radius or text-shadow (for example) causes absolutely no harm to anyone, and may actually help these useful tools to become standardized. If they sit untouched by leading designers, the W3C has no incentive to consider their inclusion as standards.
@Jeff
While I agree that in order for something to become standardized we should use it, I don’t think mass-use is necessary a case for standardization. In a lot of ways the “blink” tag should never have come along, correct?
However, I can see your point and I think it’s a solid one. I know on your personal site you just added the webkit-transition, but this, to me is a fault. That is controlling minute behavior, not presentation and should not be a CSS Rule. This is why the standard governing body of the W3C needs to move quicker to determine where those fall.
Unfortunately, I don’t trust the general development and designer public to use appropriate browser-based rules appropriately. I really believe that’s massive design-by-committee.
The point, that we both agree on, is that we’re at a giant stalemate, browser’s are coming out with new tools, and it’s slowly causing problems… again.
I can’t believe I have to put “again” after that statement.
I, for one, appreciate your call to action. I’m ready to do “it”, whatever “it” may be. Perhaps if you build your website, you can list all browser specific code that might be useful to a designer in an easy to use format (perhaps with which browsers have it implemented).
I can see that argument (that transitions are behavior versus presentation), and that’s why I didn’t mention -webkit-transition in my comment. It’s debatable whether or not it fits the criteria. I don’t think there’s much debate possible about things like opacity, shadows, and rounded corners, though.
I totally agree we’re at a stalemate, but I’m not sure I really agree browser makers adding new tools is causing problems. I do see the potential for such problems (as you note, it’s happened before), but so far, I haven’t seen any real world problems with the new-fangled CSS properties Mozilla and Apple have been adding. I think this is in large part due to the namespacing of them.
For me, it boils down to this: I wish the standards bodies would make this new stuff, get it standardized, and push it out in a timely fashion so the browser makers could implement it. There’s no doubt that is the ideal situation. However, I don’t believe that will ever happen. The W3C has proven to me that it’s incapable of working that way. So, in lieu of that, I’m happy the browser makers are making the new shiny, otherwise there would simply be no new shiny. Less-than-perfect is better-than-nothing, I guess. :)
Kenny, your thoughts sum up several snippets of conversation that I am hearing from friends and colleagues both in person and online. You’ve cleanly given us the WHAT?, so the question quickly moves to HOW? Justin has some nice ideas. I think without some sort of concerted effort, this becomes just another snippet of conversation speaking to the need.
@Justin & @Chad
I think those are excellent suggestions and I’m trying to put something together ASAP. However, until that time people need to talk about it.
The How, is a difficult problem to answer and one where I will contribute my own idea for a solution very very soon.
@kenny I looked about but couldn’t find your email to take this off-thread (which is probably a good thing for spam reasons). I’m willing to help with whatever you come up with. My relevant contact information should be visible from this post.
Time to kick http://www.webstandards.org/ back into action, I say.
I so wished it was true, but my experience is another one. Web standards are not state of the art at least here in europe. Knowledge emerges and grows, but there are large projects with disgusting code and even table layouts.
One of the problems is that the client has a hard time figuring out if it’s a high quality website, watching only the design aspects (must be shiny!). There are many unexpected web designers, who are not aware of basic things like doctype switching. There’s a lot to do, still.
But: That shouldn’t hinder us to go in new directions and do everything we can to move the web forward.
Browsers have to implement edge cases to explore the possibilities, the W3C can’t do that. CSS3 should accelerate the development of new standards because the modules can be implemented separately and become recommendation. Sadly, it seems that the CSS working group is not as agile as we all wish.
We shouldn’t get stuck because something only works in one browser or doesn’t work in one (“IE6 doesn’t support it, so we can’t use it” is still one of the most-heard complaints I get), and browsers shouldn’t restrict themselves because there’s no recommendation for a particular function.
I agree completely: but how do we kick this into a movement instead of yet another blog-sphere bitch-fest?
For talk to be effective it must be acted upon. There’s already been a lot of talk over this and almost no action.
“The W3C” isn’t the issue here. That’s like blaming the Capitol building for the lack of progress that goes on inside. The W3C is made up of hundreds of companies, including all the major browser vendors, and if they wanted to advance the state of the art, there was nothing the W3C as an organization did to stop them. The vendors just stopped talking. That’s why CSS3, the first drafts of which were written on stone tablets, has taken this long to get any traction.
If you want something to blame, blame MS for stopping at IE 6. The end of browser competition created the ideal environment for Mozilla, et al., to stop innovating, and they ended up focusing on stability (not that that’s a bad thing) and chrome, while not touching the rendering engine (and I’m sure they’d say we as developers wanted it that way).
Not that I’m complaining. It’s that stagnation that gave Flash the opportunity to turn itself into a legit business tool, which in turn is good for my stock options.
@Matt
I disagree. Although Microsoft is seemingly the browser-maker with commercial interests, I think that they have taken a lot of steps forward in the past few years. My point is not to point the finger at Microsoft, I don’t think they’re to blame. In fact, I think Microsoft’s recent moves and openness are welcome. They are not the problem child alone, but rather it is the entire W3C/Browser-Maker governing body that needs to be pushed.
As I said in my post, web standards are now standards. While some developers and designers may take different routes, that tends to happen whenever you have a large body of people working with the same material. It’s a preference more then a problem.
W3C is who develops the standards and it is out of respect for their work that they are included. If browser-makers want to get together and develop their own standards, I’m not ok with it, but at least we’ll have one governing body of standard syntax, selectors, rules etc.
…and Flash, I feel, has found it’s niche, for which I, as a user, enjoy. So you’re retirement owes me a thank you letter :)
Reading this post (which I very much agree with on the whole) and reading the comments, I’m wondering if a big part of the industry’s problem is ideological, more than anything else.
For example:
contrasted with:
Here’s the question on my mind: given the track record, can we really expect a governing body of any kind to be truly forward thinking and innovative? I don’t see it happening. Maybe there would be a way to promote designers/developers to push the edge and rework the governing body to at least respond faster?
I’ve always looked at the standards movement in a communist/capitalist way. The analogy is going to be grossly simplified but in general the big pushes in browser/feature innovation were going on during the more free market capitalist days when there wasn’t a large bureaucratic body, the web standards group, at the helm imposing their views on the process. No, I’m not saying that I’d like to go back to the free wheeling days before standards but it did allow for a lot of innovation and “new shiny”.
Standardizing was good, we’ve had “the great leap forward” but now we’re in the time where communism suffers. We’re just hanging out and not innovating. I’d like to see a bit of a balance.
All-in-all good thoughts and a nice discussion. I just hope something more comes of this chatter other then the chatter itself.
I work for a browser vendor, but I’m commenting here privately and not representing them.
Firstly, I wish I worked in the U.S.A. where everyone uses Standards. I was recently training some twenty-something developers, all of whom use tables and spacers and test in IE. I regularly see websites that are barely accessible to people with a disability here in the UK.
The problem with standards is that everyone wants something different from them. Designers want all the great visual effects that the community requested when I asked readers of the Web Standards Project to tell the CSS Working Group what they want. (http://www.webstandards.org/2008/01/18/tell-the-css-wg-what-you-want-from-css3/)
Implementors want standards that define tbe behaviour, syntax etc tightly enough that they can be implemented in an interoperable way. That takes time. For example, I was tech lead in a project in which we regularly wished that we could use -border-radius. We couldn’t because IE can’t handle it. Neither can Opera (my employer), because when we were developing our latest version, the Firefox and Safari implementations were not interoperble because of an ambiguity in the spec (see http://www.css3.info/border-radius-apple-vs-mozilla/). I personally think it’s bad for the web to have endless lists of vendor-specific CSS (-mozilla-border-radius, -o-border-radius, -webkit-border-radius and the like. Compare with the opacity property, which is interoperably implemented without vendor prefixes because of a tight spec).
Another example from the same project: we removed file-type icons that are added to the end of links via CSS background-image because IE7 couldn’t be relied upon to render them aesthetically if a link wrapped a line. That’s not because IE is naughty, but because the CSS spec says “The tiling and positioning of the background-image on inline elements is undefined in this specification”. A specification that doesn’t specify is inadequate, and more haste won’t solve that problem. In fact, I believe that the CSS working group have been going back over the specs and plugging these holes. That’s unglamorous, and it doesn’t get sexy transitions and animations out the door, but it does help an interoperable, predicatble web.
Specs take time because implementors, who are often competitors, have to be involved. There was some talk that implementors shouldn’t be allowed to write the specs, but that wouldn’t work. The xhtml 2 spec is, as far as I can tell, dead in the water because no-one wants to implement it, although it’sa beautiful spec. The community often ask for a CSS parent selector (I’ve argued for it myself) but it would never be implemented because it’s processor-heavy, and when the browser vendors are fighting on standards, speed and security, no-one will slow their browser down with a parent selector.
Also bear in mind the different perspectives about speed. HTML 5 is probably a decade away from implementation; hardly a breakneck speed, yet many people complain that it is being developed without consensus - and seeking consensus takes time, but without consensus there is no implementation.
So, I’m all for the web forward campaign. But I personally don’t believe that everyone doing their own thing takes the web forward. The browser wars, with the marquee tags and inconsistent DOMs and table layouts were fantastically innovative, but it caused a hell of a mess which I still see today. Do we want to go back there?
I think there are a lot of reasons to believe that mess wouldn’t happen again. CSS namespacing with vendor prefixes alone goes a long way towards ensuring that. And, everyone having lived through that hell once makes everyone less likely to let it happen again.
Everything you say is totally accurate, Bruce. But, in light of it, we basically have two choices:
Frankly, I’ve been using the same basic CSS properties for five or six years, now. I want some new toys, and I’m not prepared to wait a decade for them.
“Frankly, I’ve been using the same basic CSS properties for five or six years, now. I want some new toys, and I’m not prepared to wait a decade for them.”
oh, I absolutely agree- I want to use border-radius, text-shadow, @font-face and css animations (which I too think are legitimate in CSS - http://www.brucelawson.co.uk/2008/bling-and-the-separation-of-css-and-svg/ )
I’m certainly not saying that it’s perfect now — just cautioning against the unrealistic panacea that everything would be perfect by next tuesday if those pesky browser vendors were out of the speccing process.
@Bruce
When I wrote this article I wanted to make sure of two things:
1) This didn’t turn into a Microsoft Bash 2) That accountability was handled to all parties I see as decision makers
I don’t, however, believe that web developers or designers as a whole (as in everyone who has a blog, builds a website, etc) should make the decision, but rather that more designers and developers should join the consortium. Someone, however, needs to make a decision at the W3C and browser-makers need to implement that decision.
@Jeff
While I think namespacing is alright, it still smells to me like a wolf in sheeps clothing. That being said, I agree that as far as innovation goes, this is the best option currently.
Hi Kenny, you said, “more designers and developers should join the consortium”. Anyone can comment on the mailing lists, and I think that anyone can apply to be an invited expert and actually be part of the working group.
I don’t think there’s a list of invited experts anywhere, but I know Molly Holzschlag is an invited expert. Andy Clarke was (don’t know if he still is) but I assume that the rest of the CSS eleven are. That’s a heck of a lot of talent, co-driving the CSS bus.
I think it’s a great topic and I’ve got to join forces with Matt Wilcox when above he asks:
That question has been long standing. Many designers know and realize that there’s an obvious problem, but where to start working on the solution? I can’t provide a difinitive answer but to be brutally honest it doesn’t look like we’re getting much help from the W3C, although I’m sure work is being done.
Jeff Croft got it right with his ultimatum, and of course we’ll all choose option 2:
Although limited, we’re seeing more and more use of proprietary CSS, mostly as a result of frustrations similar to Jeff’s:
I think we’ll see more and more designers fall to the frustration, begin implementing as seen fit, and the community itself will become much more involved in developing new standards. The negative side effect of such a result would be disorganization, which relates to Bruce’s comment about joining the consortium. There are many organized mailing lists to follow and participate in. Hopefully we’ll see some more action on that front (from myself included).
“Frankly, I’ve been using the same basic CSS properties for five or six years, now. I want some new toys, and I’m not prepared to wait a decade for them.”
I’m with Jeff on this. I’m ready for some shiny new toys.
F*#k it I’m going back to using tables till you guys get this sorted out! Just kidding, thanks to anyone delving into pushing the technology forward. Great article, the action is in the discussion I guess. Who or what is the real Chief, as in too many Chiefs not enough indians? Where’s a good Chief when you need one then.