I don't think this so called test/benchmark has everything that is under the HTML5 umbrella. It lacks WebSQLDatabase, Geolocation, offline storage, CSS3(debatable) and many more.
So since this test is not complete, I call this article: false advertising.
CSS is developed by W3C, but it's a separate working group, and CSS is orthogonal to HTML5 (i.e. you can have 100% HTML5-conformant browser that does not support any CSS at all!)
> CSS3 is not HTML5 and that fact is not debatable at W3C.
HTML5 is like AJAX, it's just an umbrella term in reality. When you hear "AJAX library" you probably think of a library that does a lot more than just HTTP requests. Same goes for HTML, it's almost useless without some CSS.
> Geolocation has been removed from the main HTML5 spec.
InnerHTML wasn't in the spec originally, but everyone used/had it anyway because it was a good idea.
> WebSQL has been dropped in favour of IndexedDB (both are separate specs).
It was dropped only from Firefox, just like h.264, but that didn't stop anybody from using it. Some browsers have IndexedDB, some have WebSQL. IndexedDB was just a work in progress last time I checked, WebSQL already had existing implementations.
Yes, but we're talking about W3C documents using W3C terminology, not colloquial misnomers.
> InnerHTML wasn't in the spec originally, but everyone used/had it anyway because it was a good idea.
I was mistaken about Geolocation. It hasn't been removed (as many other similar features were), this one been developed as separate spec from the beginning.
Separation of HTML5 and other HTML5-ish specs is decided by way W3C works, internal politics, HTML group charter, and inter-dependencies between the feature and rest of HTML. It doesn't mean that something was or wasn't a good idea.
> It was dropped only from Firefox, just like h.264, but that didn't stop anybody from using it.
That's not related. W3C test suite covers W3C HTML5, not Firefox (and H.264 vs Firefox is a completely different problem).
WebSQL suffers from very real problem — it has only one DB implementation, which is SQLite. This means that sites will end up denending on that particular SQL dialect with all its quirks (e.g. sqlite storage is weakly, dynamically typed) and in long-term will require to be bug-compatible with SQLite.
I encourage everyone not to use it, even if it's available now, because we'll miss last chance to remove it.
(I love SQLite in my apps, but API for the web needs to be much smaller and tighter than that to withstand decades of people copying & pasting code they don't understand).
quote: "While we are glad that Microsoft is keen on using test suites to measure standards compliance, these particular tests on MSDN only cover a very limited spectrum of capabilities, and statements about browser compliance made here are frankly misleading.
The SVG-in-HTML tests cited by MSDN are known to be broken, do not conform to the HTML5 parsing rules or the DOM specification, and in fact make use of an IE bug. Microsoft's engineers know about this bug _and_ the fact that the test itself is wrong [1][2], so it is surprising that this test is touted as an important one in furthering interoperability between browsers. Firefox's beta builds, on the other hand, actually do have an HTML5 parser in place [3], and nightly builds [4] will pass legitimate tests for SVG-in-HTML parsing.
Additionally, some of the CSS tests listed on MSDN are inaccurate, and browsers that actually pass these tests are not compliant with the standard. There is an official CSS3-selectors test suite where we pass all the relevant pieces [5] and we think that for web developers, sites like caniuse.com [6] are much more useful when discussing HTML5, CSS3, and SVG support in browsers. Mozilla will continue to contribute to open and fair test suites that adhere to the specifications in question [7], and encourage other browser vendors to do so as well."
Funny indeed. I mean look, IE9 is definitely a BIG and most excellent step in the right direction. If only it was supported on Win XP. If we can get rid of IE8 and have the few top-notch browsers start going head-to-head with features and getting updated regularly, it would be a good day indeed.
The results table is also extremely misleading. There are many tests with no results for a particular browser, and in some cases the table gives a 0% score for no results in a category, and in other cases 100% for, say, two passes plus a bunch of "no results". At this point, you should probably treat the percentages as random numbers, not meaningful metrics of anything.
Get back to me when IE passes a significant battery of Firefox-submitted tests. W3C is owned by Microsoft at a time when Microsoft has lost all right to call themselves a browser vendor for any purpose other than to harm the browser market.
Get back to me when IE 9 runs on any of the thousands of machines in my company, where Windows Vista/7 is a non-starter at least for the next two years. Get back to me when it is able to run on any machine I have at home. It never will. With the possible exception of Safari, every other browser tested runs everywhere required, I think.
I think they will be served just fine by improving their own compliance where it counts, which as I interpret these and other tests, is already better than IE's.
Why go to great lengths to prop up such a sagging organization (w3c, Microsoft, take your pick)? It has never been beneficial in the past to do so. Microsoft has only historically materially supported browser standards at w3c when they were so severely behind they could not use their other monopolistic behaviors to destroy the browser market.
w3c has to value the contribution other companies besides Microsoft in their real plans rather than just being Microsoft's rescue vehicle one year and their sword in the backs of the competition the next when they suddenly lose interest in open standards again and shut everything down because w3c thinks if Microsoft no longer thinks it is important no one else should either.
In their behavior with respect to browser (and other standards), yes. Despite the supposed one company one vote model, generally every other company can be moving forward and if Microsoft signals thumbs down, it will come to a grinding halt.
If Microsoft stops browser development, the w3c takes that as a sure sign that browsers are no longer important to anyone, which is fine, as long as w3 doesn't value the other memberships.
whatwg was taken over by w3c as a desperate move on Microsoft's behalf and immediately put under a Microsoft chair despite their long history, and slowed down by years.
This is very good, I'm glad the team have done such a good job - but I don't think the IE team will manager to 'win back' developers unless IE9 can replace IE6/7 and 8.
Ideally IE9 needs to fix the problems that the older legacy browsers created. Because it can't be run on Windows XP, IE9 can only complement the mistakes that were previously made.
My suspicion is that part of the motivation for IE9 is encourage upgrades from XP. Otherwise, there isn't much of an economic basis for its development because unlike Chrome, IE doesn't send every keystroke in the address bar to a server for analysis.
Two important notes: current stable version of Chrome 7 so beta of IE9 should be compared with Chrome 8. Say it takes 6 months for IE9 to go gold, in that time Chrome stable will have had two more releases bringing it up to 9 which will probably contain a number of enhancements to close the gape if not over take IE9.
43 comments
[ 3.2 ms ] story [ 89.6 ms ] threadhttp://www.betanews.com/article/Microsoft-launches-Internet-...
So since this test is not complete, I call this article: false advertising.
Geolocation has been removed from the main HTML5 spec.
CSS3 is not HTML5 and that fact is not debatable at W3C.
However, W3C does not follow buzzwords. When W3C says "HTML5" it means http://www.w3.org/TR/html5/
CSS is developed by W3C, but it's a separate working group, and CSS is orthogonal to HTML5 (i.e. you can have 100% HTML5-conformant browser that does not support any CSS at all!)
HTML5 is like AJAX, it's just an umbrella term in reality. When you hear "AJAX library" you probably think of a library that does a lot more than just HTTP requests. Same goes for HTML, it's almost useless without some CSS.
> Geolocation has been removed from the main HTML5 spec.
InnerHTML wasn't in the spec originally, but everyone used/had it anyway because it was a good idea.
> WebSQL has been dropped in favour of IndexedDB (both are separate specs).
It was dropped only from Firefox, just like h.264, but that didn't stop anybody from using it. Some browsers have IndexedDB, some have WebSQL. IndexedDB was just a work in progress last time I checked, WebSQL already had existing implementations.
Yes, but we're talking about W3C documents using W3C terminology, not colloquial misnomers.
> InnerHTML wasn't in the spec originally, but everyone used/had it anyway because it was a good idea.
I was mistaken about Geolocation. It hasn't been removed (as many other similar features were), this one been developed as separate spec from the beginning.
Separation of HTML5 and other HTML5-ish specs is decided by way W3C works, internal politics, HTML group charter, and inter-dependencies between the feature and rest of HTML. It doesn't mean that something was or wasn't a good idea.
> It was dropped only from Firefox, just like h.264, but that didn't stop anybody from using it.
That's not related. W3C test suite covers W3C HTML5, not Firefox (and H.264 vs Firefox is a completely different problem).
WebSQL suffers from very real problem — it has only one DB implementation, which is SQLite. This means that sites will end up denending on that particular SQL dialect with all its quirks (e.g. sqlite storage is weakly, dynamically typed) and in long-term will require to be bug-compatible with SQLite.
I encourage everyone not to use it, even if it's available now, because we'll miss last chance to remove it.
(I love SQLite in my apps, but API for the web needs to be much smaller and tighter than that to withstand decades of people copying & pasting code they don't understand).
If you want tests to cover other areas then feel free to submit them, or encourage others to do so.
Also, I'm not sure how seriously I can take an HTML5 conformance report that doesn't actually validate correctly (<center> element!?):
http://validator.w3.org/check?uri=http://test.w3.org/html/te...
Historically nobody wanted to write test cases. I think Microsoft/W3C now discovered incentive to do so…
Unless someone is saying that the MS-submitted ones are incorrect, I fully endorse their submission.
quote: "While we are glad that Microsoft is keen on using test suites to measure standards compliance, these particular tests on MSDN only cover a very limited spectrum of capabilities, and statements about browser compliance made here are frankly misleading.
The SVG-in-HTML tests cited by MSDN are known to be broken, do not conform to the HTML5 parsing rules or the DOM specification, and in fact make use of an IE bug. Microsoft's engineers know about this bug _and_ the fact that the test itself is wrong [1][2], so it is surprising that this test is touted as an important one in furthering interoperability between browsers. Firefox's beta builds, on the other hand, actually do have an HTML5 parser in place [3], and nightly builds [4] will pass legitimate tests for SVG-in-HTML parsing.
Additionally, some of the CSS tests listed on MSDN are inaccurate, and browsers that actually pass these tests are not compliant with the standard. There is an official CSS3-selectors test suite where we pass all the relevant pieces [5] and we think that for web developers, sites like caniuse.com [6] are much more useful when discussing HTML5, CSS3, and SVG support in browsers. Mozilla will continue to contribute to open and fair test suites that adhere to the specifications in question [7], and encourage other browser vendors to do so as well."
http://ie.microsoft.com/testdrive/benchmarks/Acid3/Default.h...
See here and followups for discussion: http://lists.w3.org/Archives/Public/public-html-testsuite/20...
"IE CSS3 compliance continues to bring up the rear"
Get back to me when IE 9 runs on any of the thousands of machines in my company, where Windows Vista/7 is a non-starter at least for the next two years. Get back to me when it is able to run on any machine I have at home. It never will. With the possible exception of Safari, every other browser tested runs everywhere required, I think.
Why go to great lengths to prop up such a sagging organization (w3c, Microsoft, take your pick)? It has never been beneficial in the past to do so. Microsoft has only historically materially supported browser standards at w3c when they were so severely behind they could not use their other monopolistic behaviors to destroy the browser market.
w3c has to value the contribution other companies besides Microsoft in their real plans rather than just being Microsoft's rescue vehicle one year and their sword in the backs of the competition the next when they suddenly lose interest in open standards again and shut everything down because w3c thinks if Microsoft no longer thinks it is important no one else should either.
It is?
If Microsoft stops browser development, the w3c takes that as a sure sign that browsers are no longer important to anyone, which is fine, as long as w3 doesn't value the other memberships.
whatwg was taken over by w3c as a desperate move on Microsoft's behalf and immediately put under a Microsoft chair despite their long history, and slowed down by years.
Ideally IE9 needs to fix the problems that the older legacy browsers created. Because it can't be run on Windows XP, IE9 can only complement the mistakes that were previously made.
Microsoft rarely seems to think that altruism can align with good business sense, and I think it suffers because of this.
The charts are showing only the tests IE9 excel at, which is a fraction of the picture.
The tests also compare IE9 vs current stable releases. I want to see that comparison against beta releases.
Oh wait lets also write down what operating systems IE9 is compatible with:
Google Chrome: Windows XP, Windows Vista, Windows 7, Mac OSX, Linux
Firefox: Windows XP, Windows Vista, Windows 7, Mac OSX, *nix
Opera: Windows XP, Windows Vista, Windows 7, Mac OSX, Linux
IE9: Windows Vista, Windows 7
Oh look IE9 runs on only the latest MS operating system. They don't even support their own older OSes.
Lets not even get into the security aspects.
The latest version of Firefox for OS X requires at least version 10.4 (released 2005). Ditto for Opera.
Google Chrome requires at least 10.5 (released 2007), and doesn't support PPC (the transition to Intel was `completed' in 2007).
Windows Vista went to retail in 2007.
Firefox, Opera and Chrome all still work on Windows XP though.
Meanwhile what fraction is underserved by IE9 not working on Windows XP?
I have no objection to your numbers as long as we also compare apples to apples and quantify the impact.
If I hadn't checked myself into Compulsive Domain Buyers Anonymous already, I would grab ie9hacks.com and ie9bugs.com