: HTML5: Is it approved or not? (and other questions)


hugh
2011-11-04, 01:01 PM
Got an email from Microsoft to view MSN Canada's new "HTML5" website.

When I tried to go there I was unable as the site would only let me view it with Internet Explorer.

So my questions are these:


Are Chrome and Opera HTML5 ready?
Has the WWW Consortium approved HTML5?
Can an true HTML5 site be browser specific? Isn't the whole idea behind HTML5 that it be virtually plug in free and browser agnostic?

hugh
2011-11-04, 01:03 PM
In the note to me MS says

This new MSN Video site features HTML 5 powered by IE9, making it an even more dynamic, rich experience. Hosted on the cloud, the site goes way beyond the ‘normal’ video browsing experience with features that include:

So is this really an HTML5 site if it can't be viewed with Chrome or Opera?

audacity
2011-11-04, 01:16 PM
HTML5 is a generic name given to a bunch of different unrelated things. There are a number of related technologies that are decidedly not HTML5 (Javascript and CSS3 for instance) that people use the term HTML5 to refer to.

So, often people say "HTML5" when they mean "latest web technologies".

Jake
2011-11-04, 01:22 PM
Are Chrome and Opera HTML5 ready?
I remember this.
http://www.digitalhome.ca/forum/showthread.php?t=128890

Gino Cerullo
2011-11-04, 05:27 PM
Got an email from Microsoft to view MSN Canada's new "HTML5" website.

When I tried to go there I was unable as the site would only let me view it with Internet Explorer.

Many HTML5 sites that the big companies are putting up are demo sites so one of the requirements to viewing them is that you do it with their browser of choice. Apple had HTML5 demo pages running a while ago that were similarly restricted to Safari. Once the standard has settled down and the bugs have been worked out they should be opened to the general public and all browsers. Some sites are already starting to use tags that where developed for the HTML5 standard.

So my questions are these:

Are Chrome and Opera HTML5 ready?
For the most part the latest versions of Chrome, Opera, Safari and IE are HTML5 ready although their may be some bugs that they are still working out.

Has the WWW Consortium approved HTML5?
No, not really. The new tags that make up HTML5 are for the most part ready for use now but some are still being fleshed out.

Can an true HTML5 site be browser specific? Isn't the whole idea behind HTML5 that it be virtually plug in free and browser agnostic?
Technically no, but that won't stop a site's developer from using the browser agent to restrict access if they wanted to.

The whole point of the WWW Consortium and HTML, CSS and all the other acronyms that are a part of web standards have always been to establish platform agnostic technologies so that all web sites functioned the same no matter which browser was used. For the most part, it's the developers who built sites using browser specific tags and extensions that caused many of the problems of the past. Microsoft didn't help the situation at all when they released IE6 that did not follow many web standards of the time and promoted the use of IE6 specific tags as a way to gain market share for its browser which in turn blocked access to IE6 specific sites from other browsers.

scampbell
2011-11-05, 07:51 AM
Can an true HTML5 site be browser specific?
It wouldn't surprise me if a "HTML 5 powered by IE9" site was.

hugh
2011-11-05, 09:58 AM
So, in a nutshell.

There are no HTML5 standards but there is lots of ideas on what should be considered HTML5

Saying your site is HTML5 is meaningless because there are no approved standards.

audacity
2011-11-05, 01:38 PM
No, in a strict sense the term HTML5 does have meaning - it's just that how the word is colloquially used is technically wrong.

I'm sure someone could make a "HTML5 validator" to test whether a site conforms to the HTML5 standards. In fact, this is what people are doing when it comes to actually implementing the HTML5 "standards". I use quotes because in some cases the specifications as defined by the W3C leave room for ambiguity and I remember seeing that Microsoft has been putting a lot of effort into writing (and releasing) unit tests (http://samples.msdn.microsoft.com/ietestcenter/) for each feature to try and put more clarity into the standard.

Really, if the W3C was doing a better job, when it writes their standards they really should be releasing a set of test cases to clarify the definition of that standard.

hugh
2011-11-05, 01:46 PM
Audacity, first you said

HTML5 is a generic name given to a bunch of different unrelated things

then you said

No, in a strict sense the term HTML5 does have meaning - it's just that how the word is colloquially used is technically wrong.

So which is it? A generic and inexact term to vaguely describe an assortment of technologies or something that has meaning?

My original belief was that HTML 5 was a "specification" that, if properly implemented, could render the same markup consistently. Clearly that is not the case! It seems like HTML 5 is like Web 2.0, something simply means, "stuff that is better than before but not a standard"

audacity
2011-11-05, 02:20 PM
It's confusing, I know. In the first quote I was talking about the term "HTML5" as it is colloquially used.

HTML is comprised of a series of standards that are exact and specific. HTML5 is a superset of those standards, a list of each standard and version of that standard. If you meet each of those standards, then I suppose you can say "HTML5 compliant".

The WHATWG - the people who pushed for the HTML5 direction over the XHTML direction have basically dropped the version number from HTML (http://www.sitepoint.com/html5-is-dead-long-live-html/) and are "improving" HTML on a feature-by-feature basis, and no longer using superset version numbers anymore. This aligns much closer to reality than the "big version number" approach that had been done heretofore with HTML 4.01, XHTML 1.0, or CSS3.

None of the browser makers have implemented all of the HTML features to a finished spec before working on the "next set" of features. For instance, according to the article I linked, CSS2.1 is complete, yet no browsers offer a full bug-free implementation. Yet, there are CSS3 features that are widely implemented and stable across all browsers, such as rounded corners.

Should a site developer step back and say "I'm going to stick with CSS2 until CSS2.1 is properly implemented by everyone - and using CSS3 features today is unthinkable!". No. They would (should?) look at the situation on a feature-by-feature basis. There are some CSS3 features that are widely implemented and are stable for all browsers. Use them if your site needs them and the features are widely supported by the browsers your customers use.

Still, we could technically create a CSS2.1 or CSS3 compliance test because those specs do strictly define what is required; and most browsers would fail that test.

And when people use the term "HTML5" they are often talking about recent features that I would call "post-HTML5".

So, yes. When most people say "HTML5" they are using the term like "Web 2.0" and include many features that aren't technically part of the HTML5 spec. Yes, we can technically create a "HTML5 validator" that aligns to the published specs that are part of the superset HTML5 version number, and all browsers would fail to achieve 100% on this test.

I'm sorry if I'm doing a poor job of explaining the situation. May I recommend Dive Into HTML (http://diveintohtml5.info/) - a online book which describes HTML5 feature-by-feature, but it has a bit of history there that explains how we got to where we are today.

JamesK
2011-11-05, 03:35 PM
Can an true HTML5 site be browser specific?

Any web site can check the browser type. Of course some browsers can spoof their ID too.

ScaryBob
2011-11-05, 03:38 PM
HTML5 is not expected to be finalized until 2014. In the meantime, recommended drafts exist, with a final proposal expected in 2012.

Unapproved or draft "standards" are used all the time, even in hardware devices. Usually, all it usually takes is a few software or firmware tweaks to meet the final standard. A good example of this was the wireless draft N standard, which was in use for many months. Some companies use software that meets the draft to find inconsistencies in implementations.

Sometimes, companies roll out non-standard software to try and force their own last minute changes into the final standard. (MS is probably the most egregious offender.) That causes problems for consumers and the entire industry since it means that software will not work or communicate properly with competing products.