Submitted by At PDC ’09 Microsoft’s Steven Sinofsky, president of the Windows and Windows Live division, revealed the first details of the company’s next browser, Internet Explorer 9. Even though the new browser is still in an early development stage, the first few builds are being tested internally. It is poised to come with some fancy improvements – including HTML5 and CSS3 support.
The most interesting improvements is that Internet Explorer 9 will move graphics and text rendering to the graphics chip, using DirectX’s Direct2D. Sinofsky showed several demonstrations, one of which showed the difference between non-GPU and GPU rendering: Bing Maps did about 14fps without using the GPU, and up to 60fps while using the GPU.
Internet Explorer 9 will also come with a new JavaScript engine, as well as better standards support – including HTML5 and CSS3. This means that this new version of Internet Explorer finally gets the features other browser have had for a while.
Obviously, Internet Explorer 8 was released in March, so it might take a while for IE9 to be ready. In the meantime, Channel9 has a number of videos detailing the new features coming in Internet Explorer 9.
Forget the other stuff, how about spell check?
Not just spell check, but how about implementing a global spell-check service like some other systems?
“Obviously performance is an issue, and the fact that IE9 is crawling back from the brink is something that developers are taking note of. The proud disclosure that IE9 posted a 32% score on the Acid3 test score, elicited a tremendous groan…a bit like telling the audience that Chris Rock couldn’t make it but Carrot Top could.”
http://www.betanews.com/article/PDC-2009-Postkeynote-Day-2-What-are…
Seems like Microsoft spends a whole boatload of developers working on the wrong issues.
Web developers want to use most of what ACID3 suggests, and IE seems to be lacking AGAIN.
How is it, that Webkit AND Gecko managed to go from some 30 – 40 % score on the ACID3 Test to 100% within a year, and even the constantly manpower-starved KHTML team managed to come close to 100%, but all-powerful Microsoft continues to miss the boat almost completely?
The saying goes that one should never attribute to malice what can be adequately explained by stupidity, but years and years of basically ignored web developers, plus a new Microsoft only, non-standard format (Silverlight) makes it look like Microsoft’s sleazy attempts to hijack the internet communication protocols are not over yet.
http://ourlan.homelinux.net/qdig/?Qwd=./Exactly_correct&Qiv=name&Qi…
Ah, Silverlight… nice to see it still exists.
i am actually really looking forward to a final release of silverlight 4. the besta shows a lot of promise and i really like the technology in general.
http://www.neowin.net/news/live/09/11/18/silverlight-4-beta-availab…
one of my other things i really like is Yahoo’s YUI (version 3 especially). If you haven’t heard of it check it out. it’s BSD license as well.
http://developer.yahoo.com/yui/3/
Edited 2009-11-18 21:12 UTC
I probably would like Silverlight technology too, if it were an openly standardized, freely reimplementable platform.
But as a closed and patented technology, controlled by a company known for monopoly power abuse, it simply is not a viable way forward.
You know, Google Wave is new too, and the company behind it is as powerful as Microsoft, but Google made the APIs and the communications Protocols open, and even open-sources a client and a server for others to build on. So Google Wave will not lock-in anyone.
And just preceding the bit about requiring Silverlight to view the IE9 demos, this gem can be found…
To their credit, they didn’t say they were going to use them…
I was pleasantly suprised the video’s aren’t Silverlight only, but you could also download MP4, etc.
But they are not on that site, they are on this site:
http://channel9.msdn.com/tags/IE-9/
Yes, they are a bit hidden, but the links are their.
HTML 5 ? with video support ?
I don’t like them forcing me to use silverlight. We don’t need more buggus closed plugins.
I wish they stop making such hype without good results.
Yes, instead we should use loose web standards like HTML 5 video tag that won’t work on every browser since W3C is so f–king weak piece of shit organization that they can’t even demand using specific codec. Instead web devs can roll dice and decide which users won’t have access on content, good job W3C!
That codec issue is way overblown. The img tag doesn’t have fixed formats either. And for good reason. A couple years from now we’ll probably have new improved codecs and then you don’t want a video tag stuck on an old format.
True, but you don’t have to create 3 separate versions of the same image file just to be compatible with the three main platforms. The lack of a standard codec for the video tag means that Flash will remain the preferred video delivery method for most content producers/distributors.
If you want to deliver video over the web without requiring viewers to install additional software, there will be two main choices. One, use Flash and support all three major platforms with a single file. Or two, use the video tag which will (probably) require providing a WMV version for Windows users, a MOV/MP4 version for Mac users, and an OGV version for Linux/UNIX users/everyone else (which also means greater time/resources spent encoding the videos, greater server diskspace requirements, more work testing each version of the video, etc).
Regardless of my own feelings on the matter, it’s not hard to predict the typical response if I were to offer those options to a boss or paying clients.
Here’s hoping. At this point, I think the best (realistic) outcome would be for the MPEGLA to “pull a Unisys” with h.264 licensing (which would probably spur a large-scale migration to alternatives).
That hasn’t really been a problem with images (modern browsers can still render GIF files, for example).
If you provide ogv, it can be viewed with firefox and chrome regardless of the OS, right?
The others can well make do without viewing the video (or install a plugin if they insist on viewing it). This is especially true for websites that target technical audiences.
True. For that matter, most technical users would be more than happy with plain text link to download file and play it offline (speaking for myself, at least). However, for most content, technical users are a small portion of the audience.
But when providing content to a general, non-technical audience, there are different concerns – especially for commercial content producers & distributors. They want the largest possible/potential audience, so they want to minimize the amount of effort that’s required from users in order to view their content.
That is funny.
W3C didnt ‘decide’ aka ‘force’ a codec because it wasnt influenced by companies. The situation over there is NOT a ‘lets decide’ one. Its a matter of not having a perfect option. Actually w3c couldnt even decide because each browser may just do it the way it wants.
You have no idea what the f–k you are talking about. HTML is at least open. Want to get trapped in silverlight hell? Be my guest, im not following. I dont want a single company’s not-open technology on web.
Microsoft is trying to support html5 but its fanboys are not happy with it. lol
I was pleasantly suprised the video’s aren’t Silverlight only, but you could also download MP4, etc.
But they are not on that site, they are on this site:
http://channel9.msdn.com/tags/IE-9/
Yes, they are a bit hidden, but the links are their.
Edits in square brackets:
“Internet Explorer 9 To Get GPU Rendering, [partial, likely broken] CSS3, [partial, likely forked] HTML5 Support [and break a bunch of things people spent their own time on money building on top of to support standards with js libraries today, just like they did with IE8]
How many times does Microsoft get to play this game before people catch on. They have no strategic interest in contributing to an open platform. They will screw this up, and probably on purpose.
http://blogs.msdn.com/ie/archive/2009/11/18/an-early-look-at-ie9-fo…
No promise of SVG.
No promise of DOM2.
No promise of SMIL.
No promise of compliant ECMAscript (perhaps though there is a JIT compiler).
No promise of Theora codecs with the HTML5.
No promise of attempting to pass acid3 compliance tests.
Still ten years behind in some aspects.
Silverlight still required.
Yeah, because Microsoft promises are worth gold.
Microsoft’s promise to not follow standards is worth gold.
Even if they would, you’ll find an excuse to batch MS, because at the end, the problem is not MS but your twisted litle head.
And please don’t start with your link spamming to stories about “The Evil MS” Is not necessary.
Edited 2009-11-18 22:47 UTC
In contrast to yours, my post (the one to which you replied) noted actual facts.
In talking about IE9, Microsoft indeed made:
– No promise of SVG.
– No promise of DOM2.
– No promise of SMIL.
– No promise of compliant ECMAscript (perhaps though there is a JIT compiler).
– No promise of Theora codecs with the HTML5.
– No promise of attempting to pass acid3 compliance tests.
Ergo, their future plans for IE9 clearly do keep Microsoft’s browser 10 years behind.
Plain, simple, unadulterated facts.
Edited 2009-11-18 23:05 UTC
They didn’t say they wouldn’t implement those standar eather, right? those are “unadulterated” facts.
Edited 2009-11-18 23:12 UTC
When Microsoft were inundated with feature requests and bug reports from users complaining that their IE browser did not comply with web standards such as DOM2 and SVG, Microsoft simply marked them as “will not fix”.
They remain that way to this day.
And now you are acting like a cornered litle troll, time to ignore you.
Do you actually do any worthwhile web development? Anyone who does knows that on this particular note MS has definitely been the problem. Like any company, there are things they do well, and things they don’t. IE is definitely on the DON’T DO WELL side of the equation.
In saying that, if they do in fact fully support in a standards compliant manner the things they are saying they will kudos to them for taking onboard what web devs have been telling them for a long time.
i think by saying “a long time” you’re vastly understating things
Edited 2009-11-19 08:09 UTC
I would love to run a survey of some kind asking developers what they care more about, border-radius or SMIL.
I am a huge web developer geek, my blog is written using html 5 semantic elements and all the CSS3 that is implemented at the moment by anyone. I’ve got a bookshelf full of javascript books, and enjoy learning new web frameworks on the weekends. And I had never heard of SMIL until you started complaining about it.
Just out of curiosity, what are you wanting to do with it that you can’t do now because IE doesn’t support SMIL? From what I have read it looks like hypercard on the old macs, are you looking to blow the world away with another myst game or something?
I hadn’t either, but having just read about it, it looks like it is supported by WMP, Quicktime, Totem and Amarok, among other players, but is not natively supported by any web browser. It also looks like a fairly redundant technology in a web browser considering you can do the same things with CSS/DOM/Javascript.
SMIL compliance comprises two of the acid3 tests (subtests 75 and 76).
http://en.wikipedia.org/wiki/Acid3#Standards_tested
Browsers will not score 100 on acid3 tests without compliance to SMIL.
Safari, Opera, Epiphany and Google Chrome score 100/100.
http://en.wikipedia.org/wiki/Synchronized_Multimedia_Integration_La…
http://en.wikipedia.org/wiki/SVG_animation
SVG animation is the standards-compliant way of doing animations instead of Flash or Silverlight.
Edited 2009-11-19 00:46 UTC
How about pointing us at all of the wonderful websites that are currently based on SMIL … ?
Edited 2009-11-19 02:25 UTC
Chip on your shoulder much?
You are asking about chickens and we haven’t got an egg yet.
http://arstechnica.com/open-source/news/2009/01/smil-animation-and-…
Anyway, currently, a site that has SVG games or animation might well use SMIL, but it probably uses javascript instead.
Here is a site where you can get a tatse of what is possible, IF you have a compliant browser:
http://srufaculty.sru.edu/david.dailey/svg/SVGAnimations.htm
The rarity of such sites for real is very much the whole point though. SMIL and SVG are standards that were first recommended in basic form circa 2000, and more advanced form circa 2005.
We could have had the equivalent of flash (.swf) and silverlight interactive (non-video) functionality on all platforms without browser plugins way back in 2001, but it didn’t happen because a certain vendor refused to support SMIL, SVG, compliant javascript and DOM2.
Still, there is the odd place where you can go:
http://scriptdraw.com/
http://aviary.com/home
Flash and Silverlight plugins are awkward proprietary work-arounds for the failure to support those standards.
Flash and Silverlight plugins can these days be used by software suppliers to control which suppliers may, and which may not, produce products which can be used to browse the rich web. It can be used to control which architectures and platforms are, and are not, viable.
That is just wrong. It is a situation which desperately needs to be fixed.
Edited 2009-11-19 05:52 UTC
Indeed– the egg being browsers that actually support the feature that we are discussing.
Yes, those work in Chrome and Opera, but not Firefox.
The problem really seems to be the poor definition of SMIL. There are actual SMIL documents with the MIME type application/smil+xml. SMIL documents and embedded SMIL are not supported natively by any web browser.
What you seem to be referring to is SVG animation which is a slightly modified and extended version of the animation portion of the latest W3C SMIL recommendation. Webkit and Presto support SVG animation to various degrees.
SVG animation is only related to SMIL in that the same tag names, attributes and behaviors were chosen for the animation support in SVG. SMIL itself is a separate technology.
I think you’ll find that most people, including me, would agree that, all other things being equal, open solutions are best on the web.
The reality of the situation is that these solutions are not all equal. SVG, the audio/video tags, and SMIL are all fragmented standards that have to be clumsily tied together through Javascript and DOM. Not to mention the fact that they are severely lacking in performance.
Ignoring politics, the proprietary solutions win hands down.
Both Flash and Silverlight provide fully featured multimedia platforms. In the area of tools, Flash has excellent support and Silverlight isn’t very far behind. From the performance angle, Flash is a bit of a dog, but is slowly improving and the latest versions are finally taking advantage of hardware acceleration. Silverlight is the undisputed king of rendering performance among web tools and its HD video delivery is second to none.
If this is so vitally important to you, I’d suggest that you spend less time proselytizing on the web and instead, go apt-get yourself a copy of kdevelop and get cracking on a decent framework to tie all those open web technologies together, on tool support for SVG animation and on modifying both Cairo and Skia to use hardware acceleration to improve their horrid rendering performance.
The crown here is still up for grabs, but technology is going to win this one. Not politics.
Nah. Fanboys don’t write code. They just take to the forums, whine, and complain.
Edited 2009-11-19 10:09 UTC
Both Flash ans Silversight are hacks. They suck, plain and simple, and by design at that. They are not accessible, ON PURPOSE. It’s designed to present you with annoying adverts that can’t be easily skipped without another hack on top of it.
SVG is accessible and is not designed to sell expensive toolkits but to provide useful content. Reason #1 why SVG is not used that much is because IE doesn’t support it. And you can write all the tools and libraries you want, while IE does not support it, developers won’t use it.
I do web development and would like to use SVG but I always have to work around that because most of the web users are using IE.
Exactamundo.
Shudder!
http://www.google.com.au/search?q=define%3A+dystopia&ie=utf-8&o…
“dystopia on the Web” indeed.
http://wordnetweb.princeton.edu/perl/webwn?s=dystopia
The day is not far off when most web users will be using something other than IE. For many sites, that day has been here for some time already.
http://arstechnica.com/microsoft/news/2009/11/october-2009-browser-…
(Look at the Ars Technica browser share graph. 81% share of “something else”. For the sites that techies visit, apparently the techies are overwhelmingly using “something else”. Happily, it is the techies that drive the web).
All users won’t all be using the same “something else”, but all of the “something elses” they will be using will render SVG (and a great deal of other rich content) without a problem.
Edited 2009-11-19 11:47 UTC
Compared to AJAX both Flash and Silverlight are elegant solutions to providing a consistent experience for internet users.
The idea of working with AJAX again makes me nauseous. HTML and Javascript were never, ever meant to be used to that extent.
Silverlight actually has a good api, the problem though is that it doesn’t have the install rate of Flash. I know a lot of FOSS advocates here don’t like Silverlight but I haven’t seen much competition coming from Linuxland. If Silverlight had come out 5 years ago from Linuxland it would have made Flash a relic. The mythical FOSS army never delivered when it came to competing with Flash.
Reason #2 is that it is slow and excessively large. Whoever came up with using XML as the format for an already CPU intensive operation must have been out of their mind.
Actually parsing XML is not the most CPU intensive part and many parsers actually use the GPU to render SVG. There is also SVGZ, which is compressed SVG. AFAIK swf files are also compressed XML.
Which? I know it can be done because I’ve written code for my current hobby project which renders vector graphics on the GPU, but I don’t know of any renderers in common use that do that.
Maybe Opera does and Apple may have some accelerated implementation in Safari, but Firefox, Chrome/Chromium, Epiphany, Konqueror, and Arora all use software scanline rasterizers.
Load up some of those SVG animation examples that lemur linked earlier. You’ll get barely tolerable frame rates on things that a GPU could render as an afterthought.
http://sourceforge.net/projects/svgl/
That’s surprising. It looks like a fairly comprehensive project. They even took the same approach I did, using the SGI tesselator to triangulate paths. It’s a shame that it hasn’t seen any real development in over five years and that other projects like Cairo haven’t picked up on it.
So, in other words, you singled out Microsoft even though open source browsers don’t support SMIL at all — or support it very badly — and there isn’t any appreciable SMIL content even available. This is why I don’t take you seriously, lemur. This is just another example of your unreasonable bigotry. You don’t seem to get this: Just because a standards body ratifies something doesn’t make it significant or meaningful as a standard. What matters is ADOPTION. Look at HTML5. The whole industry has decided that it’s important, and the market is dragging Microsoft along. Not so with SMIL. It’s shelfware.
Edited 2009-11-19 09:52 UTC
SMIL won’t ever take off while IE does not support it. The problem is IE, it’s not hard to get. You keep accusing lemur2 of religious anti-Microsoft feeling, but you seem to be religiously defending Microsoft anyway. SMIL may not be the most important thing, but he has a point. He also cited SVG. Why can’t a browser support SVG in 2009? IE is the #1 reason why the web sucks today.
http://ourlan.homelinux.net/qdig/Exactly_correct/tick.svg
Edited 2009-11-19 11:25 UTC
Wrong, and HTML5 refutes your argument. IE doesn’t support it, but industry pressure — having other browsers support it — has basically forced MS to support it. I’m not a MS fanboy. I’m very realistic & honest about their limitations and problems. What I object to, though, is rank hypocrisy from people like lemur who criticize other platforms, even though his own house isn’t in order. When he stops trying to mislead people, there will be no need for conflict.
Whoah! You’re not an MS fanboy? You are as much an MS fanboy as Lemur2 is a Linux fanboy.
Regarding desktop issues today, one can be equally fanatical, but more realistic, by being a fanboy for the the MS camp, since MS clearly dominates in that area today.
Yeah. Me too. That’s why I am somewhat distrustful of either of you. But if I had to choose, I’d say I distrust Lemur2 a bit less. His heart is in the right place, at least.
Bullshit. Take a look at this thread. I’m not defending MS for falling behind on web standards. I’m not making excuses for them. I’m not asserting they’re better. All that I’ve done is point out lemur’s hypocrisy. He consistently asserts (paraphrasing) that proprietary is crap, open source is the best, etc, etc. There is no such thing as BETTER. Just different. And perhaps complementary, if you use them in a way that makes sense.
Look, I tend to use whatever makes the most sense, and I tend to ignore people who assert that other things are better, purely based on ideology. Like you.
I could care less about his heart. What I care more about is intellectual dishonesty, which he clearly exemplifies much of the time.
I’m taking far more than just this thread into account. I’ve taken note of both of your posting histories over the last year or so. I’ve tended to take a more active role in scolding Lemur. But you’ve been on the radar, as well.
I stand by my earlier observation.
Meh. You scold me, I scold you … gets us nowhere. No amount of pointless posturing makes anyone right.
The only “right” lies in what is best for the most people.
I note that the only counter to that last obeservation from both yourself and Tomcat is ad hominem attack. That is a sure sign of losing the argument for the parties that are using the ad hominem.
“Better” is in the eye of the beholder.
Silverlight is certainly better for Microsoft (an undeniable point which straight away debunks Tomcat’s claim that there is no such thing as better). With control over how users can browse rich web content, Microsoft would be in a position to dictate what platforms, devices and architectures can be used to view web content, and which cannot. Microsoft would be in a position to be the sole supplier of tools which can create rich web content. Microsoft would be in a position to let content providers be able to control the browsing experience. Microsoft would be in a position to be able to siphon off a significantly larger proportion of the “content delivery” revenue stream than they currently can.
However, as I point out, “better” is strictly defined by your point of view. From the point of view of the user of a web client, Silverlight and Flash are terrible. Abysmal.
From the point of view of a supplier of an alternative web browser platform/architecture/machine (say for example ARM or Nokia), Silverlight and Flash are terrible. Abysmal.
There are far, far more people in the group whose natural point of view (if they knew the facts) would be that Silverlight and Flash are terrible. People don’t want to be controlled. People don’t want to be dictated to. People don’t want to be ripped off. People want to be able to express themselves, and have their own choices.
Majority rules. Ergo, open standards are better, and Silverlight and Flash are terrible. Abysmal.
Please note this argument (unlike Tomcat’s) has precisely nothing to do with fanaticism. It is also utterly honest.
Edited 2009-11-19 23:25 UTC
Well, when it comes to the web browser, following standards is BETTER. supporting SVG is better than not, yes. It’s not even a matter of ideology, it is a technical matter of interoperability. The core function of a web browser it to interoperate. The web moves at the speed of the slowest browser, and that is IE. And supporting HTML 5 is the minimum, not supporting HTML would mean the browser wouldn’t even browse basic pages. All browsers do support HTML. Not supporting standards like SVG still makes it the worst browser of them all. At least browsers like lynx have a point and don’t support it for a reason.
Edited 2009-11-20 06:42 UTC
If I’m consistent, how can I be hypocritical?
I don’t think that word means what you think it means.
http://www.google.com.au/search?q=define%3A+hypocrisy&ie=utf-8&…
Actually, I’m certain it doesn’t mean whatever you think it might mean.
You really don’t get it, do you? Sigh!
It is always better for people if they have a choice. It is better for PEOPLE if IE, Firefox, Opera, Safari, Chrome and as many others as you want to all support SVG, DOM2, correct javascript etc, etc, than if IE doesn’t and the rest do. If IE alone supports Silverlight as well, then that would be good for people too. More choice.
It is not about what is better for me, or for you, or for Microsoft, but rather what is better for everyone.
If everyone’s only choice is Silverlight alone, and they must run IE under Windows in order to view the rich web content, then that clearly is a terrible outcome for almost everyone.
This quote from poster spiderman expresses it pretty well:
Well, when it comes to the web browser, following standards is BETTER. supporting SVG is better than not, yes. It’s not even a matter of ideology, it is a technical matter of interoperability. The core function of a web browser it to interoperate. The web moves at the speed of the slowest browser, and that is IE. And supporting HTML 5 is the minimum, not supporting HTML would mean the browser wouldn’t even browse basic pages. All browsers do support HTML. Not supporting standards like SVG still makes it the worst browser of them all.
Quote please. I’d like just one example of a position I have espoused that you believe is dishonest.
Put up or shut up.
Edited 2009-11-20 10:26 UTC
Deficiencies in IE go back way further than HTML5. They go back ten years further. Microsoft have utterly refused to support stable web standards in DOM2, SVG and correct javascript for ages.
There is no misdirection here whatsoever, these are perfectly straightforward facts.
This is a chicken and egg-problem. But if the browsers support it, the webdevelopers would start to use it. SVG and canvas is used in Google Maps (for non-IE browsers) for example.
It appears that Safari was able to pass acid3 with abysmal SMIL support (http://blog.codedread.com/archives/2008/03/26/webkit-nightly-not-sm…) so acid3 is in no way an indicator of proper SMIL compliance.
Just out of curiosity, I went to w3schools to try out some of their SMIL examples, and none of them worked on the latest versions of Firefox, Chrome or Opera.
Attempting to load a .smil file directly results in the various browsers either displaying the source or attempting to download the file.
Interestingly enough, that is the page I used for reference when making my original comment. Perhaps when digging through it, you missed this relevant piece:
Wikipedia could very well be wrong, but my experiences seem to prove otherwise.
Do you know of any SMIL examples that actually work in a web browser?
I’m waiting for lemur2 to post links to all of these supposedly great SMIL sites that we’re all missing out on.
The hate is strong with this one…
I think you need to use SMIL embedded in the page. Like SVG is often used.
The web has gotten more compatible in recent years. But the remaining blight has mostly to do with incompatible Javascript. As a desktop admin, I can tell you that if the page doesn’t look quite right, my users grumble a little. If it doesn’t work at all then they shout. And often that means resorting to IE under Crossover Office.
And it makes me want to Kill! Kill! Kill! 😉
By comparison I put SVG, CSS3, and SMIL all in roughly the same insignificant category.
Fair point.
However, when DOM2, SVG, CSS3, and SMIL are used in conjunction with compatible (and fast) ECMAscript, then web sites can deliver rich content to ANY browser on any client platform, and this can be done at least as well as other proprietary solutions such as Silverlight or Flash.
The bold bits are the vitally important bits.
IMO any requirement to have a particular platform or propietary software blob (including codecs) in order to view some types of web should be utterly shunned.
Edited 2009-11-19 00:33 UTC
Agreed. My main point is that the only person on the internet that seems to really care about SMIL is lemur2, us web developers care much more about CSS and Javascript.
ever ran acid3 through the w3c css validator?
Ever seen a requirement that a test suite must itself pass the things it tests for?
the acid tests aren’t really designed for standards compliance, they are more for corner cases and failovers. the w3c validator probably just isnt built to handle that many unrealistic scenarios at the same time.
well, how would you handle a negative value when only positives are alowed?
ignore the sign? take the nearest valid value? ignore the whole line? or render it contrary to the standart?
and now you expect the browser to guess what the autor things is right?
if you want to test correct rendering, give it something correct to render
A set of test cases which test only for behaviour with correct data input?
How much real testing have you ever done?
Keep hearing this response. No, the Acid3 test itself doesn’t pass validation – it’s not supposed to. The W3C validator is a test for the correctness of the content. The Acid tests are for the correctness of the browser, and that includes testing whether the browser handles invalid content correctly.
If you ask me, I’d say that the big mistake was in deciding to handle invalid content without flashing a big error and aborting in the web developer’s face. If you coded an ODF document in vi, in binary mode, by hand, and just guessed whenever you weren’t sure of the codes… didn’t bother to use delimiters properly, would you expect it to load succesfully? Display properly? Well… why should you expect deverse web browsers to “just figure it out” reliably? As standards go, web standards suffer from the problems of the nebulous RS-232 standard… to about the power of about 15. Maybe 20.
I’m mixed on that. As a developer, I’d very much prefer something that fails clearly and cleanly, instead of behaving unpredictably – that way potential problems are much easier to find.
As a user though, I’d really rather the browser try to deal with the fact that developers aren’t always as perfect as they should be…
If the authors had been forced to to write proper documents from the beginning, users wouldn’t have to care about sloppy documents, browser devs wouldn’t have to worry about sloppy documents (malformed documents could be categorically rejected as a security risk) and page authors wouldn’t have to worry about how various browsers deal with their and others’ sloppy documents. It would have been a win-win-win-win situation instead of the lose-lose-lose-lose mess that we have now. And have no good way of escaping from.
I share your ambivalence. What choice have we now? But the situation that we are in is such an unnecessary train wreck. And we *should* recognize it as being a train wreck, so that we can avoid the next one. Not that I’m particularly hopefull on that count.
Edited 2009-11-19 02:34 UTC
As a developer who has spent most of my time using systems languages and only rarely worked in web land, I certainly agree. Building a web page feels like a guessing game in a way that is very disconcerting to a C/C++ programmer.
But I have often wondered if the precedent of lax interpretation of HTML created by the earlier browsers provided a level of accessibility that contributed to the pervasiveness of the web today. I guess we’ll never know.
It’s by design.
The standard specifically says, if you don’t understand a part, please skip it. That means older browsers are able to view a lot of newer content.
If you have a browser that doesn’t understand SVG at all, you might be able to see the text embedded in the SVG. That’s by design.
No HTML was specifically created to allow older browsers to view new content and skip anything it doesn’t understand. That’s how it’s suppossed to work. The ACID3-test was specifically created not only to test new tech. but also to let all browsers reliable fail in the same way.
ODF is not binary, it’s a zip-file. Editing with vi would not make a lot of sense directory. Editing ODF-XML-files with vi works very well thank you very much. Just zip it afterwards and it’s possible to create valid ODF just fine.
Three words: Developers Developers Developers
Nice to see that Microsoft got the priorities again wrong, please the shiny things audience and deliver incomplete crap underneath and tell the developers that is all you got, live with it.
My personal guess is they have lost the developers community, in a developers centric site I have less than 10% IE overall and those are mostly hits from major corporations. Until they fix the basic issues like SVG support ecmascript compliance etc.. they wont get them back no matter how shiny the next version of IE is, period.
The only care about Windows developers, Windows developers, Windows developers.
The web is something standanding in the way of their desktop monopoly, because it makes the desktop you use irrelevant.
…damned if they don’t.
But whose fault is that? Or did I forget those five years of complete web stagnation and spyware/virus rape?
It was all just a dream…
Yeah. That all happened in Season 9 of “Dallas”.
You can see from those charts that IE has gone from not even comparable to other browsers, to about 70x slower, to only slightly slower the FF (which means of course about 3x slower then webkit) in 3 versions. That really is huge, I don’t think there is any vender that has made this amount of progress in this amount of time. Granted, they were about 15 years behind when they started going on this, but still, it is something to be happy about.
When IE7 becomes the standard minimum it means we get stuff like child selectors, attribute selectors, inline-block, min/max-width, and about 90% of css hacks. For people who aren’t living on the bleeding edge of web work, that totally changes the way you make web pages. IE8 should chop off another 9% of css hacks, let us use anything (even the stupid stuff) in CSS2.1 with reckless abandon, and pretty much bring what we can use up to about what firefox 3.0 supported.
IE9 will probably still be the least common denominator, but if they keep going at this pace I wouldn’t be surprised if IE10 brought them truly into the game.
Please Microsoft create something so people don’t use IE6 anymore. Don’t create a Vista+ IE9, but also let people update atleast XP.
People keep using it because of incompatibilities because of lazy developers. The only real way around it is adopting what Google has which is provide the webkit technology via a plugin.
They did – it’s called IE7. Anyone still using IE6 is doing so either because they’re stuck with some corporate web app that won’t work on anything newer, or because they don’t know or care about upgrading (XP machines won’t upgrade the browser automatically, afterall).
What about all those corporations that have lots and lots of Windows 2000 desktops ? Their is no IE7, IE8 or IE9 for Windows 2000. Not only that, but Microsoft says: we support Windows 2000.
I don’t know, Firefox-maybe? Companies have had five years to work on adoption.
If it cheers you up, Microsoft are going to stop supporting Windows 2000 completely in July 2010.
Hehe –
That should teach corporations not to trust Microsoft technology.
There are obviously some companies, which got locked into IE6 so thoroughly, that until today the cost of migrating to SOMETHING other seems to be bigger than putting up with the considerable cost of maintaining an outdated software stack.
Now I ask: What do corporations think about Silverlight? Is it a good idea betting the company’s IT infrastructure on something which most likely will be obsolete in 10 years, with no low-pain upgrade path available?
I hope to get proven wrong, but until today Microsoft managed to create lock-in’s that have been strong enough to hinder migration even from Microsoft technology to Microsoft technology. TCO if a format also contains migrating away from it some time in the future.
Fortunately for companies that need to still be running IE6, there is available a plugin for IE far, far better than Silverlight.
http://blog.chromium.org/2009/09/introducing-google-chrome-frame.ht…
Hopefully one can browse the modern web without having to escape from IE6.
Edited 2009-11-19 12:11 UTC
Why do you think this problem doesn’t apply to other commercially supported systems? Software gets old and due to dependencies, new stuff doesn’t always work on old versions. It’s not like the cost of maintaining the old system is so high, otherwise people would be upgrading more.
I have yet to see an internal silverlight adoptions in one of the corporations I have been in, my personal guess is, first of all they are pretty slow movers anyway, secondly most of them have been burned by the entire IE6 stack that they are even slower to adopt yet another Microsoft only solution (Java however is everwhere internally in those corps on the server stack, they love simply the multi platform capabilities and that they can scale their installations upwards infinitely)
http://www.theregister.co.uk/2009/11/20/silverlight_4_windows_bias/
Some things about Silverlight will work better, or only work at all, with a Windows client.
Who would have thought?
I’m shocked, truly shocked.
I bet no-one saw this coming.
(Where are the sarcasm tags when you need them?)
Edited 2009-11-21 07:28 UTC
Somehow, my heart fails to bleed when Silverlight Windows version suddenly gets one more feature than the Silverlight Mac version.
They are planning to make silverlight a new way of making windows applications (to replace windows.forms, IIRC). As a technology, it’s every bit as exciting than a new version of MFC would be – that is, unless you are a windows developer, you just don’t need to care.
This comment as well as Silverlight itself shows just how profound is the utter lack of understanding of what is meant by the word “standard” amongst Windows people.
http://www.w3.org/standards/webofdevices/independence
http://www.w3.org/standards/webdesign/
http://www.w3.org/standards/techs/deviceindependenceauthoring#w3c_a…
http://www.w3.org/TR/2004/REC-webarch-20041215/
http://tech.slashdot.org/story/09/11/20/2314220/New-Microsoft-Silve…
I like this quote:
“Silverlight is the Zune of application frameworks.”
Classic.
This one isn’t bad also:
“A cross platform framework having platform specific features is hardly superior.”
Edited 2009-11-21 11:11 UTC
Since when was Silverlight a standard?
And since when did we need to care about windows-and-mac-only technology?
Edited 2009-11-21 11:28 UTC
Since its purpose was announced as being a technology to deliver material over the web.
The public access web.
You know, similar to public television broadcasts … must follow standards. No cartel-members-equipment-only broadcasting allowed.
See … I told you Windows people just didn’t understand standards.
It pretty much failed at that (ajax, JITted javascript and html5 is currently eating the remnants of its lunch). Now, IIUC, it’s being pushed as a “normal” application development framework. It might work okay at that, and to satisfy that purpose MS needs to add certain “unsafe”/platform dependent features.
What makes me one of the “Windows people” again? I work on Linux at my day job, and use Linux pretty much exclusively at home.
Well, “Linux people” wouldn’t think of Silverlight as any kind of application development framework.
Sounds unlikely, because Silverlight indeed is one.
It isn’t a viable one for Linux. One is far better served using any of a number of alternatives which offer far better cross platform capabilities.
That’s pretty obvious.