The first disconnect seems to be over whether the W3C “recommendations” are descriptive, like the IETF standards, or prescriptive. Historically, they’ve been the latter, so while criticizing them for not being like the IETF is probably true, it seems pointless. Their intent is not to be a dictionary, but rather a blueprint.

Another disconnect seems to be over specs themselves. If they need to be worded in a “programmatic” way, it’s because they are, after all, for programmers. Those who aren’t programmers should count themselves lucky they haven’t had to implement the vague generalities I’ve seen that pass for specs. Throwing tantrums and writing with passion both have their uses, and they can indeed be beautiful to watch and read. But they’re useless when the rubber hits the road and the code needs writ. (While it’s quite true you can write code with passion, there is no such thing as passionate code.) So I think we need to cut the spec writers a bit of slack when they answer a passionate discourse with, “that’s lovely, but WTF happens when these two selectors are used in combination?” They need hard facts, not passion, if they’re ever going to produce something that works.

Then there’s the most serious, in my opinion, disconnect. It seems similar to hubris, to me. The disconnect with reality over the power and importance of the W3C in the grand scheme. I’ve seen cries that the W3C should exert its power and put the hammer down on offenders, rejecting the smallest errors in anyone’s browser as rendering it unworthy of the name.

To that POV I have only one response. What do you think would happen if every HTML, XHTML, CSS, XML and releated spec simply ceased to exist today? Gone, in a puff of electronic smoke?

Personally, I think the answer is “Nothing at all.” I lived through the wars; I even coded pages for Netscape 0.9. It didn’t stop me from making websites, though sailors would blush if in hearing when I did. Do you honestly think that had the W3C refused to certify IE6 as a browser, that it would have had even a one percent effect on its share? Or that if it had certified Opera as the only real browser, that it would have increased its share by any noticeable amount? How much rent do you pay in that dream world?

Here in the web development community, we recognize the standards, er, recommendations, as something that makes our lives easier. But we’re it, folks. No one else cares (and the browser makers themselves only care because we’re a small but vocal fragment of their market). I think the W3C has been very wise in not taking a hard line, even in using the dreadful “recommendation” term. If you don’t have power, the first rule you have to live by is to not act as if you have it; you’ll only harm yourself if you do.

There is no possible entity that can dictate terms to the world of the web. My audience doesn’t care a fig about standards/recommendations. My clients only care because I can convince them it saves them money if I can build that way as opposed to other ways. It’s the money, not the standards, that attracts them. And I still lose clients to IE-specific designers every month (or more often).

Let’s say for the sake of illustration that the perfect CSS3 spec is created tonight, and every designer falls to the ground, awestruck over its perfection, and the effect is to get it ratified immediately. How many years will it take before we can use it? 3? 6? I had workarounds for CSS1 code breaks in IE for almost a decade. If I’m going to wait 6 years to be able to use it properly, why do I care about another few years to get it right (and possibly make it easier to implement in the browsers along the way)?

Is the system perfect? Good gosh no. But if we follow the suggestion that we ban the implementors from the spec development process (or simply dramatically cut back their input) we have to realize we lose a very crucial part of the mix. I know from experience that a small, often insignificant change in a spec can result in a major improvement (or degradation) in implementation time. I don’t want to lose that.

My best suggestion is to get more designer input into the process. That requires changes on both sides. The W3C has to open the process up a little more. I see no problem in keeping the basic process intact if that happens. But by the same token, designers have to stop going “ooh, ick. A spec.” They’ve got to be willing to try and speak another language, to at least meet the programmers halfway.

(Programmers are like the French. They generally harbor great disdain for those who don’t speak their language, but will fall over themselves to help someone who gives an honest try at speaking it. Perhaps there needs to be a cadre of “speakers to programmers” among the designers, as well as “speakers to designers” among the programmers; dedicated groups whose job it is to translate the ramblings of one into words and concepts the other can relate to.)

It’s not an easy road to walk. I don’t envy the W3C one bit, and remain amazed they make the progress they do.

Apparently no one took their own advice. There are use patterns that are XHTML yet littered with XML errors all over the web, written by professionals that know better.

No Jeff, I did not pick on you. I read your recent article on the W3C and what could change. Then I read Zeldmans article above. This was about the third or fouth piece I had read about what is wrong with the W3C. To me, you clean up your own house before you criticize others.

Web Standards have become a crutch. They have been applied by people who see then as a way to justify their own convention instead of working with the spec and standards as a path to growth. This defies any kind of a Agile approach to the problem and makes web designers and developers look like imperitive morons.

We just found a use case in ie that does not conform to CSS specs in IE. We will work through it make it work and produce the findings. I have a lot of respect for the ppl who program browsers and the ppl who write the spec. They can not begin to fathom every use test retro test for how HTML and CSS are applied. But they do a good job. They need our support.

Utlimately, change has to happen – but it shouldn’t be at the extreme end of damning Web Standards altogether (I’d suggest anyone who thinks along those lines is probably someone who never fully appreciated Web Standards in the first place!) – this is inline with Jeremy’s statement from his article: ‘I just don’t think we need to dump the baby out with the bathwater. I think we can avoid any water disposal related infanticide by just changing what needs to be changed.’

The W3C and their activities (or slow-pace of them) are cited as much of the cause of the lack of improvements, but it shouldn’t be the only consideration and sticking point.

Sure, there are calls for a revolution, an apparent uprising of sorts, but will this really manifest as a revolution at all in 2008? Or are we glancing at a possible continuation of 2007 with nothing changing – lots of talk on blogs and little action? We’ll have to wait to find out…

Happy New Year

But, your obsession with validation, which you share with a large percentage of those involved with the “standards movement,” is misguided, and I will take you up on the opportunity to point out why.

You are attacking me personally over the fact that a site I didn’t design or code (blueflavor.com) has four measly unencoded ampersands on it’s homepage. Four unencoded ampersands has your panties in a bunch? Seriously? And why? Did those unencoded ampersands somehow make you have a less-than-ideal experience on the site? Of course they didn’t. You would have never known there were errors if you didn’t validate the page.

At the same time, there’s site after site after site out there displaying hundreds — even thousands — of validation errors. There are developers using tables for layout, using tag soup code, and generally crapping all over the Internet. And yet, you come after me personally, because I work for a company who has four measly unencoded ampersands in a constantly-changing feature box on its homepage, but otherwise validates. C’mon. Is Blue Flavor really the enemy? Am I the enemy?

Of course not. You only chose to validate the page so that you could come here and prop yourself up as being better than me. Nothing feels better than pointing out the shortcomings of others, right? Well, congratulations. I hope your penis feels huge right now. You’ve really done the web a massive favor by pointing out four measly unencoded ampersands on an agency’s site. Job well-done.

I suggest: next time, target your passion at educating all those thousands of web developers out there that still haven’t got a clue about standards, rather than publicly attacking a company that has been a longtime standards advocate, but accidentally let a few completely insignificant errors slip through the cracks.

Quite frankly, it’s pedants like you that make me want to puke in my mouth a little every time I hear the phrase “standards movement” these days.

Don’t get me wrong, I’m all for validation. but statements like the following don’t (to me) make any sense:

How can you have insight or be of help when you first need to validate?

It depends on why things are not validating. It’s ever so easy to break HTML, most especially if you allow comments, or pull in data from external sources, or have to use a third-party back-end, or any number of reasons. Minor validation issues such as un-encoded ampersands etc are hardly deal-breakers. I find the idea that simply because a page doesn’t validate the author can not have any viable insight into other areas (or even into code itself) is a little far-fetched and a bit OTT. Sure – if that non-validating code is tag-soup era crap then I agree that their opinion could certainly be seen in a lot dimmer light than might otherwise be; but on most designer/developer blogs any validation errors are not of that nature. They are of the every day ‘no control over what the template engine spits out’, or ‘damn, missed an ampersand’ type errors.

And even with errors like that, why can’t the person have a valid view on how to fix the W3C’s poor communication? Or on how to get more designer input? Or on how to run a group of people efficiently? Etc. Validation is rather irrelevant to the vast majority of the problems outlined in this post.

When I’m in trouble in the jungle I don’t want to ignore the advice of a bush-craft expert because he didn’t tie his shoes with a reef knot today, if you catch my drift.

Validation Is not a minor point. To validate means to be up to code. To use the recommendations as the engineers intended. How can you have insight or be of help when you first need to validate? Further, how do you get a body of engineers that wrote the recommdations we use to notice us when we will not respect their work by validating. It is a measure of quality. Look at it like it IS a unit test you can do. Because it literally is.

They need to reorganize how standards are implemented at the OEM level and how they are developed as recommendations. Scrum and agile methods would be a boon to the process. The W3C has a functional structure. Agile and Scrum would bring a open air culture. Everything from heart stints to the MRI and Turbo Tax has been developed using Scrum. It could work here.

Everyone seems to have an idea as to how to fix this or that or what should be done. Yet, there doesn’t seem to be consensus. Don’t know if there even needs to be. But, if you can easily point out a problem, why not come to the table with an idea for the solution? Stomping your foot, folding your arms and holding your breathe isn’t going to solve anything.

I’ve seen on this blog an others the mention of leadership yet, no real definition of what the expectation is for this leadership? Leadership can be many things – like everything else in this community, everyone has a different idea as to what that is.

I would like to mention that I think most people in this community are good people. The design community has a lot of passionate people and of course, not everyone is going to agree. I don’t see that as a bad thing. I think it’s the way in which people disagree that reveals much about a person.

@ Don Ulrich – you mention Jeff Croft being a publicity seeker… like that’s a bad thing? He’s a web designer/Info Architect. It’s a cruel world out there. How would you suggest he get work if he doesn’t seek publicity? Seems like a no brainer to me.

While I agree that it’s a little silly having passionate discussions about standards on pages that don’t validate – I hardly think it matters relative to the discussions that have been going on recently. Also, look around and do not generalise too much – there are those of us ‘pissing and moaning’ that do validate. In fact there are some of us that (getting pedantic) send out XHTML 1.1 with and XML MIME to browsers that accept it. So please, no unhelpful generalisations.

Validation is a minor point, and having a page that doesn’t validate in no way means you’re in no position to have valid, helpful, or insightful thoughts on the myriad of problems with web standard production processes today.

I find it very interesting that most (but not all) of the people at the W3C have this tech-minded view of everything that’s been said recently. Help by commenting on the draft specification, fine – although not easy (specs are written for programmers to understand). Help by unit testing? Most designers have no clue what that even means – nor should they. Help by getting involved with the already established mailing lists etc – well, aren’t we all saying that those (poorly publicised) communication avenues are part of the problem?

The W3C peeps need to stop looking for technical help from designers. That isn’t what we can offer, and it isn’t what we are complaining about. A few also need to step back from the manuals and the programming and see the bigger picture – which is where we have our complaints, and is where we can help. Jeff Croft and co may write in a generalised manner and with some passion – but that’s the point. It amazes me how many of the people working to produce the specifications think in a progromatic way, and that if a problem or answer isn’t framed in a progromatic way – it’s just ignored, or in your case, looked upon as a worthless pointless bitch-fest. Thinking like that will not help anyone.

Ann V. from Opera suggested that if you want browser manufacturers to respond to new things be specific. Unit testing for new things is a lenghty process and takes time. If browser vendors get a focused response they can respond better. Maybe they need to start a wish list and create a dialog with joe average standards guy.
We don’t have an axe to grind just pixels to push. Fact is anyone of us have genuine quality input into the process.

Write the publicity seekers like Jeff Croft and ask him if can stop writing diatribe long enough to validate Blue Flavor and actually participate in valid standards. Bust any of the trend setters that do not validate and ask them if they know so much whay don’t their pages validate. Where is Mark Pilgrim when you really need him?