HTML5 Splits Into Two Standards

mikejuk writes "Until now the two standards bodies working on HTML5 (WHATWG and W3C) have cooperated. An announcement by WHATWG makes it clear that this is no longer true. WHATWG is going to work on a living standard for HTML which will continue to evolve as more technologies are added. W3C is going the traditional and much more time consuming route of creating a traditional standard which WHATWG refers to as a 'snapshot' of their living standard. Of course now being free of W3C's slower methods WHATWG can accelerate the pace of introducing new technologies to HTML5. Whatever happens, the future has just become more complicated — now you have to ask yourself 'Which HTML5?'"

Comments

Dumb idea.

[kingramon0]

So when browsers claim to be fully HTML5 compliant, will that even have any meaning anymore?

Re:Dumb idea.

[icebike]

It will create so many new jobs and so much specialist knowledge, it can't fail to improve the economy!

Ah, yes, that's it, they are trying to institutionalize the Broken Window Falacy.

Personally, I suspect the term "Living standard" is code for we don't want any standards we can't subvert, and we want the freedom to pack in as much
proprietary crap as we can and go after patent license fees down the road.

This can't help but lead to IE6 all over again.

Re:Dumb idea.

[Anonymous Brave Guy]

Is Microsoft, by chance, involved in WHATWG?

Not really. This sort of madness is driven by the same fools at places like Google and Mozilla who think pushing a new update every six weeks is a good idea.

One day, they will notice that most real developers on real projects can't and don't want to keep up with that kind of unstable foundation.

One day, they will notice that most users don't like being hassled every few weeks by their browser update mechanism or their UIs forever moving around in subtle (or occasionally not-so-subtle) ways.

One day, they will notice that the only people in the industry who actually like the rapid releases are people making cute demos on blogs, people at the aforementioned Google and Chrome who are comparing anatomical measurements, and people who want to be one of the above.

One day, they will acknowledge that their quality control processes are demonstrably not up to the job of supporting such rapid releases, and they do keep breaking things, and those things aren't always minor details you can get away with for another six weeks.

One day, they will acknowledge that quite a few of the minor things that break are actually their prototype implementations of whizzy new features, which means developers of real production projects can't use those whizzy new features even on browsers that support them, which entirely defeats the purpose of pushing out new features at such a breakneck pace in the first place.

Until then, most of the projects I work on will continue to recommend that our professional customers use IE, and IE will remain the only platform that we will contractually support, because unlike nonsense like "living standards", it is a reasonably stable platform that we can test against to a professional standard. And that matters a lot more to both us and our clients than supporting this week's proposal for a multi-resolution-friendly <img> tag that only works on Chrome cloud cuckoo channel.

Re:Dumb idea.

[Stiletto]

Six weeks is not really an unreasonably short release cadence, unless you work for a defense contractor or something.

Re:Dumb idea.

[tooyoung]

Six weeks is not really an unreasonably short release cadence

But it is an unreasonably short update cadence for the user. You have totally missed the parent's point - people don't want to be updating their software every six weeks. For the home user, this is an annoyance. For the SMB or enterprise, this is a nightmare.

Just because you can release new features every six weeks doesn't mean that you should. As the parent said, this seems to be more for the "gee-whiz" factor than anything else. That, or some well intentioned soul doesn't understand that flooding the user base with software updates doesn't really equate to a good experience.

Re:Dumb idea.

[Anonymous Brave Guy]

The thing is, if you take a step back and look at the facts objectively, recent versions of IE actually have a pretty good track record for quality (and security, for that matter). Sure, you can go back to the IE6 era and point plenty of fingers, but then again you have to remember that some "bugs" in IE6 are really behaviours that hadn't been effectively standardised yet when IE6 was released; it predates CSS 2.1 by several years, for example. And of course, IE6 is more than a decade old. Criticising Microsoft's track record for having bugs or security vulnerabilities in IE6 is like criticising Mozilla because Netscape 6 wasn't their finest hour or condemning Apple for having weak support for the Web in MacOS 9 before Safari even existed.

Meanwhile, if we're looking at the situation today rather than historically, Firefox and Chrome both have appalling quality control. Since the six-week-release era, they've broken basic rendering and they've broken popular new CSS3 features like rounded corners and shadows. They've broken major third party integrations with Flash and Java, and they've broken the new HTML5 shinies that are supposed to replace them like the <video> and <canvas> elements. In several cases, they have compromised their design or don't even respect the basic architecture of the Internet in their never-ending quest to squeeze every last millisecond of performance out of your system, which is fine right up to the point where their caching just plain gets it wrong and what you see isn't the content you were supposed to see, or their direct integration with plug-ins allows something to block their application UI thread and the whole damn browser locks up because someone's AJAX request blocked one of the tabs.

So as surprising as it may seem to those of us who have been around for a while, looking at the issue today, based on the empirical data we have for bugs and effort spent fixing/working around them on various projects I'm involved with, I do currently use IE as my benchmark for browser quality. Moreover, I am 100% confident that for the features we're actually using, even recent trends around CSS3 and HTML5, IE clearly has superior quality to either Firefox or Chrome. Of course it's always possible that the selection of projects I'm talking about has been exceptionally unlucky and hit a huge number of corner cases, and I certainly won't presume to speak for every other project I don't work on...

Re:Dumb idea.

[Xest]

It doesn't help that they don't even seem to know what's in each release themselves. Case in point, I loaded up Firefox yesterday and it asked me if I wanted to install a security and stability update, so I clicked yes and it installed... ...but if it's just a security and stability update, why the fuck has my user interface changed? Were the old back/forward and home buttons a security risk then? Thanks Mozilla, for lying to me about what was in the update.

Honestly, if they can't even tell what they're putting into each patch there's really little hope for the process.

...now you have to ask yourself 'Which HTML5?'"

[ClaraBow]

The one supported by by Webkit and Gecko?

I have mod points

[LilBlackKittie]

and I wanted to moderate this story down for its appalling failure to call W3C "W3C" two times out of three.

How can a standard be "living"?

[An Anonymous Coward]

"Living standard" is kind of an oxymoron. The whole point of having a standard is so that authors have something to target, and developers know what is necessary to be standards compliant. A constantly evolving standard creates a moving target, which I believe is actually counter-productive.

Re:How can a standard be "living"?

[colinrichardday]

There is no reason the HTML5 standard needs to change that often if it's well thought out in the first place .

I believe that I've detected a problem.

Slow down

[MS]

The whole world should slow down. Stick with a stable standard for a while. And relax.

Re:Slow down

[CanHasDIY]

The whole world should slow down. Stick with a stable standard for a while. And relax.

This is probably the deepest, most profound statement on the internet today, if you take the time to really drink it in.

"Living Standard"? There is no such thing.

[gweihir]

A standard is a standard. It is not a moving target. That is its whole point.

Other things that are mandatory for a standard:
    - simple (or as simple as possible)
    - clear
    - easy to implement

I think this just killed HTML5, because now it will become a complex monster that basically is never ever compatible with anything. Funny how history repeats itelf because people are too stupid to learn its lessons.

No shit

[Sycraft-fu]

And with HTML 5 it is bad enough already. The standard is so amazingly complex that none of the browsers seem to have the same idea of how to support it. Things that will work in one don't in another, or they work less well and so on.

My favourite example is the HTML 5 Angry Birds game. In Chrome, it's "recommended browser" (something that shouldn't ever be necessary) it runs fast, and full featured, but Chrome seems to 'asplode on it randomly. Firefox is stable with it, but no sound/music, just visuals. IE is stable and has sound, but runs a bit slower than the others, it can't maintain 60 fps. This is even given that they've done work to make it work on all platforms.

So how about let's fuck off with new HTML standards until we have non-fucked up 5 implementations in at least most of the browsers. Then maybe we can worry about something new.

What kind of rubbish desktop are you using?

[Viol8]

"Today we have phones like my Andriod as well as IPhones that give a much better browsing experience than my desktop?!"

Are you having a laugh? The browsing "experience" on a smartphone doesn't come anywhere close the what I have on my dual 22 inch desktop monitors. If you seriously think that can be replicated on some rinky dink 3 inch screen then you must have problems with your eyesight..

"On my computer it flickers unless I use IE 9"

Then your computer is a piece of junk. Go and buy one built in the 21st century.

"Why should the best experiences be only for phone based applets?"

Errm , you do realise that applets are programs, not web pages?

We've been here before

[mrbester]

"HTML5" is a marketing buzzword, just like "Web 2.0". HTML 5 is a loose coupling of emergent technologies which is in a constant state of flux as new shiny stuff is added by the competing browsers (Internet Explorer is not one of these). 'Twas ever thus that new things appeared hoping to be part of the standard - either by saturation or by conscious decision - before the standard is declared. This is nothing new.

Re:My first thought...

[Xest]

I don't think it's that simple, part the problem is browser manufacturers vs. everyone else. The fact is whilst mainstream browser manufacturers often seem like the only entities who should care about HTML, there's actually more to it than that.

The impact of HTML standards development has relevance to other developers too, think how many applications export to HTML, do not underestimate how many business systems scrape websites and import HTML. Think of all the people who have to author and develop with and for HTML.

Effectively WHATWG was a coup, it was a hijacking of the standards process by the browser manufacturers. Presumably they got tired of having to deal with everyone else having a say as they do in the W3C and just decided to try and go their own way. Their criticism of W3C was that it was slow in the creation of new web standards, but who exactly was behind the failure to implement many existing standards properly, and newer W3C standards at all which was in part a major factor in that? Er, the browser manufacturers.

I'm not at all convinced it's a good thing so far, the HTML5 process seems to have been a bit of a shambles and some important areas have been overlooked and grossly neglected in the new standard (e.g. accessibility).

Re:Back on topic, the editor of both docs wrote th

[martin-boundary]

Without a W3C "snapshot" standard, there's a greater chance that companies will pick and choose the pieces they want/like in the "living" standard, leading to greater incompatibilities for users.

Part of the reason we've had a good level of interop on the web in the last ten years is because HTML4 didn't evolve. We need to do the same with HTML5, have a document that can remain unchanged for ten years at least, so that the web as a whole can sync up to the same document.

Quite the opposite

[YA_Python_dev]

The web browser interoperability in the last few years (after IE6) is a product of the WHATWG standard, that started in 2004 (it wasn't called HTML back then). Just an example: HTML 4.01 doesn't specify a way to parse HTML that actually works and doesn't specify at all how to handle errors. The result is that every browser had a slightly different and incompatible parsing algorithm. Let me make this clear: no browser ever implemented HTML 4.01. Not a single one of them. Because HTML 4.01 was extremely buggy and unmaintained. It caused the IE6 era. The HTML5 draft on W3C is less buggy but still severely incomplete, stopping making major changes just means that all browsers vendors are completely ignoring the HTML5 from W3C and going instead for the HTML standard that's actively maintained and updated.